51279 DISTORTIONS TO AGRICULTURAL INCENTIVES A GLOBAL PERSPECTIVE, 1955­2007 Edited by Kym Anderson DISTORTIONS TO AGRICULTURAL INCENTIVES DISTORTIONS TO AGRICULTURAL INCENTIVES A Global Perspective, 1955­2007 Kym Anderson Editor A COPUBLICATION OF PALGRAVE MACMILLAN AND THE WORLD BANK © 2009 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved 1 2 3 4 12 11 10 09 A copublication of The World Bank and Palgrave Macmillan. PALGRAVE MACMILLAN Palgrave Macmillan in the United Kingdom is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire, RG21 6XS. Palgrave Macmillan in the United States is a division of St. Martin's Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representa- tives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. This volume is a product of the staff of the International Bank for Reconstruction and Development /The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without per- mission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. ISBN: 978-0-8213-7665-2 (softcover) ISBN: 978-0-8213-7973-8 (hardcover) DOI: 10.1596/978-0-8213-7665-2 (softcover) DOI: 10.1596/978-0-8213-7973-8 (hardcover) eISBN: 978-0-8213-7666-9 Library of Congress Cataloging-in-Publication Data Distortions to agricultural incentives: a global perspective, 1955­2007 / edited by Kym Anderson. p. cm. Includes bibliographical references and index. ISBN 978-0-8213-7665-2 (pbk.)--ISBN 978-0-8213-7973-8 (hardback)--ISBN 978-0-8213-7666-9 (electronic) 1. Agricultural subsidies. 2. Agriculture and state. 3. International trade. I. Anderson, Kym. HD1415.D57 2009 338.1'8--dc22 2009011004 Cover design: Tomoko Hirata/World Bank. Cover photo: © Ray Witlin/World Bank Photo Library. Printed in the United States. Dedication To the memory of T. W. Schultz (1902­98), D. Gale Johnson (1916­2003), and Bruce L. Gardner (1942­2008), whose fine minds were sharpened at the University of Chicago and who contributed perhaps more than any other economists of the 20th century to our understanding of the need to reduce distortions to agricultural incentives around the world. OTHER TITLES IN THE SERIES Distortions to Agricultural Incentives in Africa, edited by Kym Anderson and William A. Masters, 2009. Distortions to Agricultural Incentives in Asia, edited by Kym Anderson and Will Martin, 2009. Distortions to Agricultural Incentives in Europe's Transition Economies, edited by Kym Anderson and Johan Swinnen, 2008. Distortions to Agricultural Incentives in Latin America, edited by Kym Anderson and Alberto Valdés, 2008. CONTENTS Foreword xxi Acknowledgments xxv Contributors xxix Abbreviations xxxiii Map: The 75 Focus Countries xxxvi PART I INTRODUCTION 1 1 Five Decades of Distortions to Agricultural Incentives 3 Kym Anderson PART II EVOLUTION OF DISTORTIONS IN ADVANCED ECONOMIES 65 2 Japan, Republic of Korea, and Taiwan, China 67 Masayoshi Honma and Yujiro Hayami 3 Western Europe 115 Tim Josling 4 United States and Canada 177 Bruce L. Gardner 5 Australia and New Zealand 221 Kym Anderson, Ralph Lattimore, Peter J. Lloyd, and Donald MacLaren PART III EVOLUTION OF DISTORTIONS IN EMERGING ECONOMIES 257 6 Eastern Europe and Central Asia 259 Kym Anderson and Johan Swinnen ix x Contents 7 Latin America and the Caribbean 289 Kym Anderson and Alberto Valdés 8 Sub-Saharan and North Africa 323 Kym Anderson and William A. Masters 9 China and Southeast Asia 359 Kym Anderson and Will Martin 10 India and Other South Asian Countries 389 Ashok Gulati and Garry Pursell PART IV GLOBAL MARKET AND WELFARE EFFECTS OF DISTORTIONS 417 11 Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 419 Peter J. Lloyd, Johanna L. Croser, and Kym Anderson 12 Global Distortions to Key Commodity Markets 459 Kym Anderson, Johanna L. Croser, Signe Nelgen, and Ernesto Valenzuela 13 General Equilibrium Effects of Price Distortions on Global Markets, Farm Incomes, and Welfare 505 Ernesto Valenzuela, Dominique van der Mensbrugghe, and Kym Anderson Appendix A: Methodology for Measuring Distortions to Agricultural Incentives 565 Kym Anderson, Marianne Kurzweil, Will Martin, Damiano Sandri, and Ernesto Valenzuela Appendix B: Global Distortions Database, 1955­2007 595 Kym Anderson and Ernesto Valenzuela Index 619 Contents xi Figures 1.1 Gross Subsidy Equivalents of Assistance to Farmers, over Time and by Region, 1955­2007 17 1.2 NRAs to Agriculture, by Regions, 1980­84 and 2000­04 19 1.3 NRAs to Exportable, Import-Competing, and All Covered Agricultural Products, High-Income and Developing Countries, 1955­2004 22 1.4 NRAs, Key Covered Products, High-Income and Developing Countries, 1980­84 and 2000­04 27 1.5 NRAs to Agricultural and Nonagricultural Tradable Products and RRA, All Focus Countries, 1955­2004 31 1.6 RRAs to Tradables, Asia, Africa, and Latin America, 1965­2004 32 1.7 RRAs to Agriculture, High-Income Countries, 1955­2007 36 1.8 Cross-Country Dispersion of NRAs and RRAs, 2000­04 37 1.9 Relationships between Real GDP Per Capita and RRA, All Focus Countries, 1955­2007 39 1.10 Relationship between RRA and the TBI for Agriculture, Focus Regions, 1980­84 and 2000­04 40 1.11 Welfare Reduction Indexes for Covered Tradable Farm Products, by Region, 1960­2007 47 1.12 Regional Shares of Global and Country Shares of Asian Welfare Reduction from Agricultural Policies, 1981­2004 49 1.13 Trade and Welfare Reduction Indexes for 28 Major Agricultural Products, 2000­04 50 1.14 NRAs for Japan, Republic of Korea, and China, and Date of Accession to GATT or WTO, 1955­2005 56 2.1 RRA to Agricultural Versus Nonagricultural Tradables, Japan, Korea, and Taiwan, China, 1955­2007 86 2.2 NRA to Rice, Japan, Korea, and Taiwan, China, 1955­2007 94 2.3 RRA to Agriculture and Real GDP Per Capita, Japan, Korea, and Taiwan, China, 1955­2004 101 2.4 RRA to Agriculture and Relative GDP Per Agricultural Worker, Japan, Korea, and Taiwan, China, 1955­2004 102 3.1 NRAs to Agriculture, EU-6 and Western European Average, 1956­1964 128 3.2 NRAs to Agriculture, EU-9 and Western European Average, 1965­1974 133 3.3 NRAs to Agriculture, EU-12 and Western European Average, 1975­1984 137 3.4 NRAs to Agriculture, EU-15 and Western European Average, 1985­1994 140 3.5 NRAs to Agriculture, EU and Western European Average, 1995­2007 144 xii Contents 3.6 NRAs to Agriculture with and without Decoupled Payments, Western Europe, 1956­2007 145 3.7 NRAs to Exportable, Import-Competing, and All Agricultural Industries, EEC/EU and Other Western European Countries, 1956­2007 146 4.1 Farm Household Income as a Percent of National Household Income, United States, 1930­2005 178 4.2 Expenditure on Commodity Programs and Payments to Farmers, United States, 1955­2005 187 4.3 CCC Commodity Program Outlays, United States, Fiscal Years 1980­2007 189 4.4 NRAs to Exportable, Import-Competing, and All Covered Agricultural Products, United States and Canada, 1955­2007 193 4.5 NRAs to All Agriculture without and with Decoupled Support, United States and Canada, 1955­2007 203 4.6 NRAs to All Nonagricultural Tradables, All Agricultural Tradable Industries, and RRA, United States and Canada, 1955­2007 205 5.1 Real GDP Per Capita in Australia, New Zealand, and Other High- Income Countries Relative to the United States, 1870­2004 222 5.2 NRAs to Exportable, Import-Competing, and All Covered Products, Australia and New Zealand, 1946­2007 232 5.3 NRAs to Manufacturing, All Nonagricultural Tradables, All Agricultural Tradable Industries, and RRA, Australia and New Zealand, 1946­2007 236 5.4 Real Agricultural Total/Multifactor Productivity Growth, Australia and New Zealand, 1927­2004 249 5.5 Real Farmland Prices, New Zealand, 1978­2004 250 6.1 NRAs to Agriculture, Eastern European Countries, 1992­2007 268 6.2 NRAs to Agriculture, Individual Eastern European Focus Countries, 2000­03 271 6.3 RRAs to Agriculture, Eastern European Focus Countries, 1992­95 and 2000­03 280 7.1 NRAs to Agriculture, Individual Latin American Countries and Unweighted Regional Average, 1980­84 and 2000­04 301 7.2 NRAs, by Product, Latin American Countries, 1980­84 and 2000­04 302 7.3 NRAs to Exportable, Import-Competing, and All Agricultural Products, Latin American Region, 1965­2004 302 7.4 GSEs of Assistance to Farmers, Latin American Countries, 1980­84 and 2000­04 307 7.5 NRAs to Agricultural and Nonagricultural Tradable Products and RRA, Latin American Region, 1965­2004 311 Contents xiii 7.6 Relationship between RRA and the TBI for Agriculture, Latin American Focus Countries, 1980­84 and 2000­04 312 8.1 NRAs to Agriculture, Individual African Focus Countries and Unweighted Regional Average, 1975­79 and 2000­04 335 8.2 NRAs, Key Covered Products, African Focus Countries, 1975­79 and 2000­04 338 8.3 NRAs to Exportable, Import-Competing, and All Agricultural Products, African Region, 1955­2004 340 8.4 GSEs of Assistance to Farmers, African Focus Countries, 1975­79 and 2000­04 343 8.5 NRAs to Agricultural and Nonagricultural Tradable Products and RRA, Africa Region, 1960­2004 348 8.6 Relationship between RRA and the TBI for Agriculture, African Focus Countries, 1975­79 and 2000­04 352 9.1 NRAs to All Agriculture and to Rice, China and Southeast Asian Countries, 1980­84 and 2000­04 365 9.2 NRAs to Exportable, Import-Competing, and All Agricultural Products, China and Southeast Asia, 1970­2004 369 9.3 NRAs to Agricultural and Nonagricultural Tradable Products and RRAs, China and Southeast Asia, 1970­2004 372 9.4 Rice NRA and International Rice Price, Southeast Asia, 1970­2005 378 9.5 NRA for Rice, Malaysia, 1960­2004 379 9.6 Relationship between RRA and the TBI for Agriculture, China and Southeast Asian Countries, 1980­84 and 2000­04 381 9.7 RRAs and Log of Real Per Capita GDP, Select Asian Countries, 1955­2005 383 10.1 Real Effective Exchange Rate Index, India, 1964­2007 397 10.2 Unweighted Average Tariffs on Imports of Agricultural and Nonagricultural Goods, India, 2002­06 398 10.3 NRAs to all Agricultural Tradable Industries, All Nonagricultural Tradables, and RRAs, India, 1965­2004 402 10.4 NRAs to all Agricultural Tradable Industries, All Nonagricultural Tradables, and RRAs, Pakistan, 1973­2005 402 10.5 NRAs to All Agricultural Tradable Industries, All Nonagricultural Tradables, and RRAs, Bangladesh, 1974­2004 403 10.6 NRAs to all Agricultural Tradable Industries, All Nonagricultural Tradables, and RRAs, Sri Lanka, 1955­2004 403 10.7 RRAs, South Asia, 1965­2004 409 11.1 NRAs to Farmers in High-Income and Developing Countries, for All Covered Farm Products, 1960­2007 432 11.2 CTEs Affecting Covered Farm Products in High-Income and Developing Countries, 1960­2007 432 xiv Contents 11.3 NRAs Affecting Covered Farm Products in Africa, Asia, Latin America, and European Transition Economies, 1960­2007 433 11.4 CTEs Affecting Covered Farm Products in Africa, Asia, Latin America, and European Transition Economies, 1960­2007 434 11.5 WRIs for Covered Tradable Farm Products, by Region, 1960­2007 436 11.6 TRIs for Covered Tradable Farm Products, by Region, 1960­2007 439 11.7 WRIs and TRIs for Covered Tradable Farm Products, by Country, 2000­04 441 11.8 TRIs and WRIs for Covered Tradable Farm Products, by Region, 1985­89 and 2000­04 451 11.9 Country Contributions to the Global TRI and WRI, 2000­04 452 11.10 WRI and Real Per Capita GDP, All 75 Countries, 1961­2004 453 11.11 NRA, TRI, and WRI for Covered Tradable Farm Products, World, 1960­2004 453 12.1 GSEs of Assistance to Farmers Globally, by Product, 1980­84 and 2000­04 465 12.2 NRAs, Key Covered Products, High-Income and Developing Countries, 1980­84 and 2000­04 469 12.3 NRAs, Rice, Milk, and Sugar, by Country, 2000­04 471 12.4 NRAs, Beef, Pig Meat, and Poultry, by Country, 2000­04 473 12.5 NRAs, Wheat, Maize, and Soybeans, by Country, 2000­04 475 12.6 NRAs, Cotton, Cocoa, and Coffee, by Country, 2000­04 476 12.7 Rice NRA and International Rice Price, South Asia, 1970­2005 480 12.8 NRAs for Rice and Per capita Income, 1955­2007 481 12.9 TRI and WRI for 12 Key Covered Products, 2000­04 491 12.10 Global TRI and WRI for Covered Tradable Farm Products, by Commodity, 1985­89 and 2000­04 492 12.11 Country Share of the Global Commodity-Specific TRI for Rice, Sugar, Beef, Cotton, and Milk, 2000­04 493 12.12 Country Share of the Global Commodity-Specific WRI for Rice, Sugar, Milk, Beef, and Cotton, 2000­04 494 A.1 A Distorted Domestic Market for Foreign Currency 569 A.2 Distorted Domestic Markets for Farm Products 589 B.1 Number of Countries for Which NRA and CTE Estimates Are Provided for 30 Key Farm Products 596 B.2 Shares of Global Production of 30 Key Farm Products Covered in NRA and CTE Estimates 597 Contents xv Tables 1.1 Key Economic and Trade Indicators of Focus Countries, by Region, 2000­04 15 1.2 Poverty in Africa, Asia, Latin America, and Europe's Transition Economies, 1981­2005 16 1.3 Growth of Real GDP and Exports, Focus Countries, 1980­2004 17 1.4 NRAs to Agriculture, Focus Countries, 1955­2007 20 1.5 NRAs to Agricultural Exportables, Import-Competing Products, and the TBI, Focus Regions, 1955­2007 23 1.6 Dispersion of NRAs across Covered Agricultural Products, Focus Regions, 1965­2007 26 1.7 NRAs to Agricultural Products Relative to Nonagricultural Industries, 1955­2007 29 1.8 NRAs to Agricultural and Nonagricultural Tradables, and the RRA, by Region, 1955­2007 33 1.9 GSEs of Assistance to Farmers, Total and Per Farm Worker, by Region, 1965­2007 41 1.10 CTEs of Policies Assisting Producers of Covered Farm Products, Percent and Per Capita, by Region, 1965­2007 44 1.11 Intensity of Public Agricultural R&D Investment, High-Income and Developing Countries, 1971­2004 57 1.A.1 Export Orientation, Import Dependence, and Self-Sufficiency in Primary Agricultural Production, Focus Countries, 1961­2004 59 2.1 Economic Growth and Structural Transformation in Japan, Korea, and Taiwan, China, 1955­2004 72 2.2 Changes in Agricultural Structure in Japan, Korea, and Taiwan, China, 1955­2004 75 2.3 NRAs to Selected Agricultural Products in Japan, Korea, and Taiwan, China, 1955­2007 87 2.4 NRAs to Agricultural Relative to Nonagricultural Industries in Japan, Korea, and Taiwan, China, 1955­2007 90 2.5 CTEs for Selected Agricultural Products in Japan, Korea, and Taiwan, China, 1955­2007 96 2.6 Changes in Direct Tax Burdens and the Allocations of National Government Subsidies to Agricultural and Nonagricultural Sectors, Japan, 1878­1937 105 2.7 Farm-Nonfarm Income Disparity in Japan's Economic Development, 1885­2000 106 3.1 NRAs to Agricultural Industries Relative to Nonagricultural Industries, EU-15 and EFTA-3, 1956­2007 147 3.2 NRAs to Covered Farm Products, EU, 1956­2007 149 3.3 NRAs to Covered Farm Products, Non-EU Western European Countries, 1956­2007 151 xvi Contents 3.4 NRAs to All Agriculture, Individual Western European Countries, 1956­2007 154 3.5 Gross Subsidy Equivalents of Assistance to Farmers, Total, Per Farm Worker and by Product, Western European Countries, 1956­2007 155 3.6 CTEs of Policies Assisting Farmers, Covered Products, Total and Per Capita and by Product, Western European Countries, 1956­2007 158 4.1 Customs Receipts as a Percentage of Value of Imports, United States, 1821­2000 184 4.2 NRAs to Covered Farm Products, United States and Canada, 1955­2007 194 4.3 NRAs to Agricultural Relative to Nonagricultural Industries, United States and Canada, 1955­2007 201 4.4 GSEs of Assistance to Farmers, by Product, Per Farm Worker and Total, United States and Canada, 1955­2007 206 4.5 CTEs of Policies Assisting Farmers, Covered Products, Total and Per Capita and by Product, United States and Canada, 1960­2005 209 4.6 PAC Disbursements during the Election Cycle, United States, November 2004­October 2006 213 5.1 NRAs to Covered Farm Products, Australia, 1946­2007 229 5.2 NRAs to Covered Farm Products, New Zealand, 1955­2007 231 5.3 NRAs to Agricultural Products Relative to Nonagricultural Industries, Australia and New Zealand, 1946­2007 234 6.1 Key Economic and Trade Indicators, Eastern European and CIS Countries, 2000­04 261 6.2 NRAs to Agriculture, Eastern European and CIS Focus Countries, 1992­2007 269 6.3 NRAs, Key Covered Farm Products, Eastern European and CIS Focus Countries, 1992­2005 272 6.4 TBI, Eastern European and CIS Focus Countries, 1992­2007 273 6.5 Dispersion of NRAs across Covered Agricultural Products, Eastern European and CIS Focus Countries, 1992­2007 274 6.6 Components of NRAs to Agriculture, Eastern Europe and CIS, 1961­2007 275 6.7 GSEs of Assistance to Farmers, Total and Per Farm Worker, Eastern European and CIS Focus Countries, 1992­2007 277 6.8 Percentage CTE of Policies Assisting Producers of Covered Farm Products, Eastern European and CIS Focus Countries, 1992­2007 279 7.1 Key Economic and Trade Indicators, Latin American Countries, 2000­04 291 7.2 NRAs to Agriculture, Latin American Countries, 1965­2004 300 7.3 NRAs, Key Covered Farm Products, Latin American Focus Countries, 1965­2004 303 Contents xvii 7.4 Dispersion of NRAs across Covered Agricultural Products within Latin American Focus Countries, 1965­2004 304 7.5 NRAs to Agricultural Relative to Nonagricultural Industries, Latin American Region, 1965­2004 305 7.6 GSEs of Assistance to Farmers, Total and Per Farm Worker, Latin American Countries, 1965­2004 308 7.7 GSEs of Policies Affecting Farmers in Latin America, by Product and Subsector, 1965­2004 309 7.8 Percentage CTE of Policies Affecting Covered Farm Products, Latin American Countries, 1965­2003 314 7.9 Value of CTE of Policies Affecting Covered Farm Products, Latin American Countries, 1965­2003 316 8.1 Key Economic and Trade Indicators, African Focus Countries, 2000­04 325 8.2 NRAs to Agriculture, African Focus Countries, 1955­2004 334 8.3 Dispersion of NRAs across Covered Agricultural Products, African Focus Countries, 1955­2004 337 8.4 NRAs, Key Covered Farm Products, All African Focus Countries, 1955­2004 339 8.5 NRAs to Agricultural Relative to Nonagricultural Industries, African Region, 1955­2004 341 8.6 GSEs of Assistance to Farmers, Total and Per Farm Worker, African Focus Countries, 1955­2004 345 8.7 Percentage CTE of Policies Assisting Producers of Covered Farm Products, African Focus Countries, 1961­2004 349 9.1 Changes in Poverty in Asia, 1981­2005 360 9.2 NRAs to Agriculture, China and Southeast Asia, 1960­2004 366 9.3 NRAs, by Covered Product, China and Southeast Asia, 1970­2005 367 9.4 Annual GSEs of Assistance to Farmers, Total and Per Farm Worker, Asian Economies, 1955­2004 370 9.5 NRAs to Agricultural and Nonagricultural Tradable Industries and RRAs, China and Southeast Asia, 1960­2004 373 9.6 CTE of Policies Assisting Producers of Covered Farm Products, China and Southeast Asia, 1970­2004 376 10.1 Shares of Agriculture in GDP and Employment, South Asian Countries, 1965­2004 391 10.2 Distribution of Fertilizer and Electricity Subsidies and Subsidy Rates for Key Crops, India, 2004 395 10.3 Trade Status of Farm Commodities, South Asian Countries, 1965­2005 405 10.4 NRAs, TBIs and Dispersion of Covered Farm Products, South Asia, 1965­2004 406 xviii Contents 11.1 NRAs for Africa, Asia, Latin America, European Transition Economies, and High-Income Countries, All Farm Products, 1960­2007 429 11.2 CTEs for Africa, Asia, Latin America, European Transition Economies, and High-Income Countries, All Covered Farm Products, 1960­2007 431 11.3 WRIs for Asia, Africa, Latin America, European Transition Economies, and High-Income Countries, All Covered Tradable Farm Products, 1960­2007 437 11.4 TRIs for Asia, Africa, Latin America, European Transition Economies, and High-Income Countries, All Covered Tradable Farm Products, 1960­2007 440 11.5 PDIs for Asia, Africa, Latin America, European Transition Economies, and High-Income Countries, All Covered Farm Products, 1960­2007 442 11.6 CDIs for Asia, Africa, Latin America, European Transition Economies, and High-Income Countries, All Covered Farm Products, 1960­2007 443 11.7 WRIs, by Country and Region, All Covered Tradable Farm Products, 1960­2007 444 11.8 TRIs, by Country and Region, All Covered Tradable Farm Products, 1960­2007 447 12.1 Coverage of Gross Value of Agricultural Production at Undistorted Prices for 12 Key Covered Products, 2000­04 461 12.2 Share of Global Gross Value of Agricultural Production for Key Covered Products, by Region, 2000­04 463 12.3 Share of Regional Gross Value of Agricultural Production for Major Covered Products, by Region, 2000­04 464 12.4 GSEs of Assistance to Farm Industries, by Focus Country Group, 1965­2007 466 12.5 NRAs, 12 Key Covered Farm Products, All Focus Countries, 1965­2004 470 12.6 CTEs of Policies Assisting Producers of Covered Farm Products, All Focus Countries, 1965­2007 478 12.7 Shares of Production Exported and of Consumption Imported for Major Covered Products, by Region, 2000­03 482 12.8 Share of Global Production of Seven Mostly Nontraded Staple Crops, by Region, 1995­2004 484 12.9 Average of Focus Developing Countries' Self-Sufficiency Ratios for Seven Mostly Nontraded Staple Crops, by Region, 1961­2005 485 12.10 Additional Contribution of Seven Noncovered Staples to Values of Agricultural Production and to Aggregate NRAs in Focus Developing Countries, by Region, 1966­2004 486 Contents xix 12.11 Global TRIs, by Commodity, 1965­2004 489 12.12 Global WRIs, by Commodity, 1965­2004 490 12.A.1 Summary of NRA Estimates by Major Product, Africa, Asia, and Latin America, 2000­04 496 13.1 Structure of Price Distortions in Global Goods Markets, 1980­84 and 2004 509 13.2 Economic Welfare Impact of Going Back to 1980­84 Policies, by Country/Region 516 13.3 Impact of Going Back to 1980­84 Policies on Indexes of Real Export and Import Prices, by Region 519 13.4 Terms-of-Trade Contribution to Real Income Changes from Going Back to 1980­84 Policies, by Region 520 13.5 Impact of Going Back to 1980­84 Policies on Shares of Global Output Exported, and Developing Country Shares of Global Output and Exports, by Product 521 13.6 Impact of Going Back to 1980­84 Policies on Agricultural and Food Output and Trade, by Country/Region 523 13.7 Impact of Going Back to 1980­84 Policies on Self-Sufficiency in Agricultural and Other Products, by Product and Region 526 13.8 Impact of Going Back to 1980­84 Policies on Shares of Production Exported and of Consumption Imported by the World, High-Income and Developing Countries 528 13.9 Impact of Going Back to 1980­84 Policies on Shares of Agricultural and Food Production Exported, by Country/Region 529 13.10 Impact of Going Back to 1980­84 Policies on Real International Product Prices 530 13.11 Impact of Going Back to 1980­84 Policies on Real Factor Prices, by Region 531 13.12 Impact of Going Back to 1980­84 Policies on Sectoral Value Added, Agricultural and All-Sector Policy Changes 532 13.13 Impact on Real Income of Full Liberalization of Global Merchandise Trade, by Country/Region, 2004 536 13.14 Regional and Sectoral Sources of Welfare Gains from Full Liberalization of Global Merchandise Trade, 2004 539 13.15 Impact of Full Global Liberalization on Shares of Global Output Exported, and Developing Country Shares of Global Output and Exports, by Product, 2004 541 13.16 Impacts of Full Global Trade Liberalization on Agricultural and Food Output and Trade, by Country/Region, 2004 542 13.17 Impact of Global Liberalization on Share of Agricultural and Food Production Exported by Country/Region, 2004 545 xx Contents 13.18 Impact of Global Liberalization on Self-Sufficiency in Agricultural and Other Products, by Region, 2004 546 13.19 Share of Production Exported and of Consumption Imported by World, High-Income, and Developing Countries, before and after Full Global Liberalization of All Merchandise Trade, by Product, 2004 547 13.20 Impact of Full Global Liberalization of Agricultural and All Goods Markets on Real International Product Prices, 2004 548 13.21 Impacts of Full Global Merchandise Trade Liberalization on Real Factor Prices, 2004 549 13.22 Effects of Full Global Liberalization of Agricultural and All Merchandise Trade on Sectoral Value Added, by Country and Region, 2004 550 13.A.1 Protection Structure in GTAP Version 7 Prerelease and in the Distortion Rates Drawn from the World Bank Project, 2004 558 B.1 Summary of NRA Coverage Statistics, World Bank Agricultural Distortions Project 597 B.2 Coverage of Gross Value of Global Agricultural Production at Undistorted Prices, for 30 Key Products and Four Product Groups, 2000­04 599 B.3 Project's Coverage of National Agricultural Production in Focus Countries at Undistorted Prices, Regional Averages, 1980­2004 600 B.4 Shares of Global Agricultural Production for 30 Major Covered Products, by Region, 2000­04 601 B.5 Share of Regional Agricultural Production for 30 Major Covered Products, by Region, 2000­04 603 B.6 Share of Global Agricultural and Processed Food Exports for 30 Major Covered Products, by Region, 2000­03 605 B.7 Share of Global Agricultural and Processed Food Imports for 30 Major Products, by Region, 2000­03 607 B.8 Shares of Production Exported and Consumption Imported, by Farm Product and Region, 2000­03 609 B.9 Per Capita Income, Focus Countries, 2005 611 B.10 Variables in the Global Agricultural Distortions Database, 1955­2007 612 B.11 Variables in the Supplementary Global Agricultural Trade and Welfare Reduction Indexes Database, 1955­2007 615 FOREWORD In his seminal 1973 book on World Agriculture in Disarray, Professor D. Gale Johnson despaired at the persistence of high agricultural protection in Organisa- tion for Economic Co-operation and Development (OECD) countries, the anti- agricultural and antitrade policies of developing countries, and the tendency for both sets of countries to insulate their domestic food market from international price fluctuations, thereby exacerbating price volatility for the rest of the world. Since the vast majority of the world's poorest households depend on farming for their livelihoods, this disarray not only was highly inefficient but also contributed to global inequality and poverty. Yet the situation worsened over the next dozen years, with agricultural protection in Europe, North America, and Japan peaking and international food prices plummeting in 1986, thanks in large measure to an agricultural export subsidy war between the United States and the European Community. The World Bank's World Development Report 1986 was devoted to the issue of agricultural protection. It urged reform via the General Agreement on Tariffs and Trade's Uruguay Round, which was launched in September of that year and which had agricultural trade-related policies high on its agenda. Simultaneously, under the direction of its chief economist at the time, Anne Krueger, the Bank under- took a research project aimed at measuring the extent to which 18 developing- country governments were pursuing antiagricultural policies. That Krueger, Schiff, and Valdés project was summarized in an article in the World Bank Economic Review in 1988 and detailed in a series of five books in 1991 and 1992. It revealed that, during 1960­84, most developing countries were reducing farm incomes not only by heavily taxing agricultural exports, but even more so by protecting manufacturers from import competition and overvaluing the national currency. xxi xxii Foreword However, from the 1980s, many low-income and some high-income countries began to reform their agricultural price and trade policies. Sometimes this reform was undertaken unilaterally, but some was also undertaken in response to international pressures such as Uruguay Round stipulations, commitments required for accession to the World Trade Organization, and structural adjust- ment loan conditionality by international financial institutions. Meanwhile, reforms in some middle-income economies (most noticeably the Republic of Korea) had "overshot," going from taxing their farmers to protecting them from import competition, which raised concerns that other emerging economies may follow suit and pursue the same agricultural protection growth path of more- advanced economies. Though the OECD Secretariat began to monitor its members' agricultural policies beginning in the late 1980s, there has been no systematic comparable monitoring of policy developments in developing countries. The World Bank launched a major research project in 2006 aimed at filling this void. The papers emerging from that project (see http://www.worldbank.org/agdistortions) have since been edited into a series of four regional books, which are summarized in this volume along with comparable studies of high-income countries' policies since the mid-1950s. By including 75 countries that together account for more than 90 percent of global gross domestic product (GDP), agricultural output, and population, the study provides a representative study of global developments in policies affecting farmer incentives over the past half century. Moreover, the proj- ect generated a global panel dataset of annual estimates of distortion by product and country. By making this dataset freely available to the public, the expectation is that further economic analysis of these critical questions will be stimulated. The present volume concludes by reporting results from a global economy- wide model aimed at addressing the following questions, among others: How much have reforms since the early 1980s improved net incomes of farmers in developing countries? What more could be achieved by removing the remaining distortions to agricultural incentives? The authors find that the economic welfare cost to the world of global distortions to goods trade fell by 58 percent between the early 1980s and 2007, and the cost to developing countries fell by 46 percent. That is, the world has gone about halfway toward liberalizing goods markets globally during the past quarter century. Developing countries have gained dis- proportionately from those reforms, and their farmers have gained far more than nonfarmers in those countries. Moreover, developing countries would benefit 50 percent more than high-income countries from completely freeing global markets for agricultural and other goods, and again their farmers would be the major beneficiaries. Of the prospective overall gain to developing countries, half Foreword xxiii would be due to agricultural policy reforms; such is the extent of global distor- tions remaining in agriculture relative to other goods markets. In turn, this suggests that developing countries have a huge stake in whether the Doha Develop- ment Agenda, especially its agricultural negotiations, are brought to a successful conclusion. Justin Yifu Lin Senior Vice President and Chief Economist The World Bank ACKNOWLEDGMENTS This book provides an overview of the evolution of distortions to agricultural incen- tives caused by price, trade, and exchange rate policies in a large sample of countries spanning the world. Following the introduction and summary chapter, it includes studies of four sets of high-income economies and five sets of emerging economies that together account for more than 90 percent of agricultural production and 95 per- cent of global GDP. The chapters are followed by two appendixes: one provides the methodology used to measure the nominal and relative rates of assistance to farmers and the taxes and subsidies on food consumption, while the other provides statistical information on the coverage of annual estimates of those rates of assistance. The authors of the five emerging-economy chapters are indebted to the other authors of the country case studies underlying those regional summaries. The country studies are reported in four companion volumes, published by the World Bank in 2008 and early 2009, which cover Africa (coedited by Kym Anderson and William A. Masters), Asia (coedited by Kym Anderson and Will Martin), Latin America (coedited by Kym Anderson and Alberto Valdés), and European transi- tion economies (coedited by Kym Anderson and Johan Swinnen). Staff of the World Bank's regional departments provided generous and insight- ful advice and assistance throughout the project, including participating in Bankwide seminars on the draft studies of each region. The World Bank's country directors also offered advice on the studied countries when clearing the working paper versions of each chapter. The authors of this book's chapters benefited from feedback provided by many participants at workshops and conferences in which draft papers were presented over the past two years. We also appreciate the insightful comments and questions from the book's external reviewers. All the country authors are extremely grateful to Ernesto Valenzuela and the team of very able research assistants he managed, including PhD students Johanna L. xxv xxvi Acknowledgments Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, and Damiano Sandri, all of whom helped compile the global mega-spreadsheet of dis- tortion estimates (accessible at http://www.worldbank.org/agdistortions, along with more than 70 working papers that have detailed appendixes not included in the published volumes). Our thanks extend to Johanna L. Croser and Marie Damania for assisting in the initial copyediting of many of the country chapters; to Janice Tuten, who handled the production and all the final copyediting of the manuscript; and to Stephen McGroarty, who supervised its publication. Both methodologically and in terms of previous estimates of distortions, we are indebted to the economists who have plowed this ground before us. They include the team of authors and editors who contributed to the seminal Krueger, Schiff, and Valdés volumes published in 1991­92; the team at the International Food Policy Research Institute (IFPRI) led by David Orden that generated a recent research report on producer support estimates and consumer support estimates (PSEs and CSEs) that covered four large Asian countries; and especially the team in the Trade and Agriculture Directorate of the Organisation for Economic Co- operation and Development (OECD), who have been generating PSEs and CSEs for high-income countries for more than two decades and for some transition and developing economies during the past four years. We are extremely grateful to OECD staff members for providing access to their files and for many useful com- ments on our work as it progressed. The direct contributions to the project by IFPRI (for the Ethiopian and Indian case studies and the South Asian overview), the OECD and the Food and Agricul- ture Organization of the United Nations (FAO, for the Ghana case study), and the United States Department of Agriculture (for the Russian case study) are also greatly appreciated. The OECD and FAO are jointly seeking funds from the Bill and Melinda Gates Foundation to continue the estimation of agricultural policy indicators for developing countries beyond the time series covered in the present study, beginning with a sample of African countries. Our thanks extend to the project's senior advisory board, whose members have provided sage advice and much encouragement throughout the planning and implementation stages of the project. The Board comprises Yujiro Hayami, Bernard Hoekman, Anne Krueger, John Nash, Johan Swinnen, Stefan Tangermann, Alberto Valdés, Alan Winters and, until his untimely death in March 2008, Bruce Gardner. For financial assistance, grateful thanks go to the Development Research Group of the World Bank and trust funds of the governments of the United Kingdom, The Netherlands. Japan, and Irelend. This combined support made it possible for the study to include countries from all regions of the world except the Acknowledgments xxvii Middle East (which accounts for less than 2 percent of global agricultural produc- tion). We also are extremely grateful to the Rockefeller Foundation for providing the opportunity for the authors of this book to gather to plan the project's final outputs at the Rockefeller Conference Center in beautiful Bellagio, Italy, Novem- ber 13­17, 2006. This book is dedicated to the memory of my mentors, T. W. (Ted) Schultz (1902­98) and D. Gale Johnson (1916­2003), and to Bruce Gardner (1942­2008), all of whose fine minds were sharpened at the University of Chicago's Department of Economics and who contributed directly and indirectly perhaps more than any other economists of the 20th century to our understanding of the need to reduce distortions to agricultural incentives around the world.1 Bruce Gardner, who was professionally active until shortly before his death, was a strong supporter of the present research project, both as a member of its senior advisory board and as author of the North America chapter in this volume. He was a gentleman as well as a scholar, and is sorely missed by friends and colleagues alike. Kym Anderson March 2009 1. T. W. Schultz's famous book, Transforming Traditional Agriculture (Yale University Press, 1964), helped the world understand that farmers, however poor, are efficient producers who respond to incentives and hence to government distortions to prices. The volume arising from a workshop he organized on resources, incentives, and agriculture for the American Academy of Arts and Sciences (Distortions of Agricultural Incentives, Indiana University Press, 1978), further clarified the wastefulness of government intervention in agricultural markets. The trade paper in that volume is by D. G. Johnson, who drew on his seminal book, World Agriculture in Disarray (Macmillan, 1973) which pointed to the inequity as well as inefficiency of farmers being taxed in poor countries and subsidized in rich countries. Both economists' contributions are celebrated in special features of journals: Schultz in the Review of Agricultural Econom- ics 28 (3), fall 2006; Johnson in Economic Development and Cultural Change 52 (3), April 2004. Bruce Gardner's seminal contributions include his paper "Causes of U.S. Farm Commodity Programs," in Journal of Political Economy 95 (2), April 1987, and his book American Agriculture in the Twentieth Century: How It Flourished and What It Cost (Harvard University Press, 2002). CONTRIBUTORS Kym Anderson is the George Gollin Professor of Economics at the University of Adelaide and a research fellow of the Centre for Economic Policy Research, London. During 2004­07, he was on an extended sabbatical as lead economist (trade policy) in the Development Research Group of the World Bank in Washington, DC. Johanna L. Croser has been a short-term consultant with this project and is a PhD and law student at the University of Adelaide, having completed her graduate eco- nomics coursework at the University of British Columbia in Vancouver, Canada. Bruce L. Gardner was a professor and the department chair of agricultural and resource economics at the University of Maryland in College Park, MD, until his untimely death in March 2008. He also served as an assistant secretary in the U.S. Department of Agriculture. Ashok Gulati is the Asian Director for the International Food Policy Research Institute in New Delhi, India. Prior to that, he headed the institute's Markets, Trade, and Institutions Division in Washington, DC. Yujiro Hayami is the chair of the graduate faculty of the Foundation for Advanced Studies on International Development and a visiting professor in the National Graduate Institute of Policy Studies, Tokyo. Masayoshi Honma is a professor of agricultural and resource economics at the University of Tokyo, where he focuses on farm policy issues. He is also a member of the board of trustees of the International Food Policy Research Institute in Washington, DC. xxix xxx Contributors Tim Josling is professor emeritus at the Freeman Spogli Institute for International Studies, and previously was a professor in the Food Research Institute at Stanford University. His research covers a wide range of agricultural trade policy areas, including protectionism. Marianne Kurzweil is a young professional at the African Development Bank in Tunis. During 2006­07, she was an extended-term consultant with this project in the Development Research Group at the World Bank in Washington, DC. Ralph Lattimore is a private consultant, but during much of the period of this project he was an economist with the Trade and Agriculture Directorate of the Organisation for Economic Co-operation and Development in Paris. Peter J. Lloyd is professor emeritus in and former dean of the Department of Economics at the University of Melbourne. Prior to that, he was a professorial fel- low in the Institute of Advanced Studies at the Australian National University, Canberra. Donald MacLaren is an associate professor in the Department of Economics at the University of Melbourne. His theoretical and empirical research focuses on agri- cultural trade policy issues, including the analysis of nontariff barriers to trade. Will Martin is research manager of the Rural Development Unit in the Develop- ment Research Group at the World Bank in Washington, DC. He specializes in trade and agricultural policy issues globally, but especially in Asia, and has written extensively on trade policies affecting developing countries. William A. Masters is a professor and the associate head of the Department of Agricultural Economics at Purdue University. He is currently coeditor of the jour- nal Agricultural Economics. Previously, he was a lecturer at the University of Zimbabwe (1988­90). Signe Nelgen was a short-term consultant with this project before becoming a PhD student at the University of Adelaide. Previously, she was associated with the Hamburg Institute of International Economics. Garry Pursell is a visiting fellow at the Australia South Asia Research Centre at Australian National University, after serving for many years in the South Asia Department of the World Bank in Washington, DC. Damiano Sandri is a PhD candidate in economics at the Johns Hopkins Univer- sity. During 2006­07, he was a short-term consultant with this project in the Development Research Group at the World Bank in Washington, DC. Contributors xxxi Johan Swinnen is a professor in the Department of Economics and director of LICOS Center for Institutions and Economic Performance at the Katholieke Uni- versiteit Leuven in Belgium, and a senior fellow of the Center for European Policy Studies in Brussels. Alberto Valdés is a research associate at Universidad Católica de Chile in Santiago. Previously, he was an adviser in the Agriculture Department of the World Bank and director of trade and food security at the International Food Policy Research Institute, both in Washington, DC. Ernesto Valenzuela is a lecturer and research fellow at the School of Economics and Centre for International Economic Studies at the University of Adelaide. During 2005­07, he was an extended-term consultant at the Development Research Group of the World Bank in Washington, DC. Dominique van der Mensbrugghe is the lead economist in the Development Prospects Group of the Development Economics Vice Presidency of the World Bank in Washington, DC, where he specializes in the global economywide modeling. ABBREVIATIONS AIDA Agricultural Income Disaster Assistance (of Canada) AMTA Agricultural Market Transition Act (of the United States) ASEAN Association of Southeast Asian Nations CAIS Canadian Agricultural Income Stabilization CAP Common Agricultural Policy (of the European Union) CCC Commodity Credit Corporation (of the United States) CDI consumer distortion index CEE Central and Eastern Europe CET common external tariff CFIP Canadian Farm Income Program CGE computable general equilibrium (model) cif cost, insurance, and freight CIS Commonwealth of Independent States (of the former Soviet Union) CMEA Council of Mutual Economic Assistance (of Eastern Europe and Central Asia) CMO common market organization CPI consumer price index CSE consumer support estimate (or earlier, consumer subsidy equivalent) CTE consumer tax equivalent CWB Canadian Wheat Board DDA Doha Development Agenda (of the WTO) EAEC Eurasian Economic Community EC European Community ECU European currency unit xxxiii xxxiv Abbreviations EEC European Economic Community EEP Export Enhancement Program (of the United States) EFTA European Free Trade Association ERA effective rate of assistance EU European Union FAO Food and Agriculture Organization FAS Foreign Agriculture Service (of the United States) fob free on board GATT General Agreement on Tariffs and Trade GDP gross domestic product GNP gross national product GRIP Gross Revenue Insurance Program (of Canada) GSE gross subsidy equivalent GTAP Global Trade Analysis Project IAC Industries Assistance Commission (of Australia) IFPRI International Food Policy Research Institute IMF International Monetary Fund ITC International Trade Commission Mercosur Mercado Común del Sur (Southern Common Market) MFP multifactor productivity NAFTA North American Free Trade Agreements NDP net domestic product NFF National Farmers Federation (of Australia) NISA Net Income Stabilization Account (of Canada) NPS non-product-specific NRA nominal rate of assistance NRP nominal rate of protection NSE net subsidy equivalent NTB nontariff barriers to trade NTM nontariff measure OECD Organisation for Economic Co-operation and Development PAC political action committee (in the United States) PDI producer distortion index PDS public distribution system PPP purchasing power parity PRSP Poverty Reduction Strategy Paper (of the World Bank) PSE producer support estimate (or earlier, producer subsidy equivalent) REER real effective exchange rate RRA relative rate of assistance Abbreviations xxxv SAL Structural adjustment loan (from the World Bank) SMAs statutory marketing authorities (in Australia) STE state trading enterprise TBI trade bias index TEC tax equivalent to consumers TFP total factor productivity TRI trade reduction index TRQ tariff rate quota UNCTAD United Nations Conference on Trade and Development URAA Uruguay Round Agreement on Agriculture USDA U.S. Department of Agriculture WDI World development indicators WRI Welfare reduction index WPI Wholesale price index WTO World Trade Organization Note: All dollar amounts are U.S. dollars (US$) unless otherwise indicated. The 75 Focus Countries (Shown in grey) IBRD 36062 September 2009 This map was produced by the Map Design Unit of the World Bank. The boundaries, colors, denominations, and any other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Part I Introduction 1 Five Decades of Distortions to Agricultural Incentives Kym Anderson "When you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely, in your thoughts, advanced to the stage of science." -- Sir William Thompson (1889, pp. 73­74) Every decade or two, food becomes newsworthy globally. Mostly it is about a price spike, either upward (hurting consumers, as in 1973 and 2008) or downward (hurting farmers in open economies, as in 1986), and most such price spikes are a consequence of major policy shifts, since local weather-induced supply shocks in a many-country trading world tend to offset each other. In 1986, for example, it was the food export subsidy war between Western Europe and North America that drove real international food prices to their lowest level since 1930. The price hikes of 1973 and 2008, by contrast, were partly a consequence of a unilateral pol- icy decisions by a single large player. In 1973, the Soviet Union departed from its policy of self-reliance and entered the international grain market in a significant way to offset a domestic shortfall (Morgan 1979). In 2008, the United States and the European Union (EU) decided to subsidize biofuel production and mandate targets for its use domestically. On both occasions, other governments imposed export restrictions to help insulate their consumers from the price rise, which 3 4 Distortions to Agricultural Incentives: A Global Perspective pushed international prices even higher and drove more exporting countries to follow suit. Policy thus contributes to market volatility, an undesirable situation because volatility around the long-run trend terms of trade slows economic growth (Williamson 2008). Yet trade policy measures are very blunt instruments for dealing with volatility, especially in the modern era of myriad financial instru- ments for risk management, and their beggar-thy-neighbor feature diminishes the international public good contribution of trade openness. Less newsworthy to the mass media, but probably far more important in its effect on the long-run growth and distribution of global welfare, are gradual policy developments in individual countries and their combined effect on other countries via the trend terms of trade in international markets.1 This study is about one such set of trade-related policy developments that, over the past half century, has had dramatic effects--in some ways negative, in others positive--on distortions to agricultural incentives and thus also to consumer prices for food. Given the impor- tance of farm and food prices for the world's poor, those policy-imposed distor- tions affect not only economic growth but also income inequality and poverty. The benefits from specialization in production and exchange have been recog- nized for millennia, yet governments have chosen to restrict international trade, including in agricultural goods. Sometimes, these restrictions are carried out via export taxes, to raise government revenue or to lower the price of food for domes- tic consumers. An early example was the tax on wine exports from Greece in the first century BC and from France and Germany in the dark ages (Johnson 1989). More commonly, trade restrictions take the form of import duties or bans, often as part of a broader foreign policy, as with the United Kingdom's imports of wine from France versus Portugal and Spain in the 1700s and 1800s (Nye 2007). The practice was so striking that wine was used as the example of British imports in the first treatise on the theory of comparative advantage (Ricardo 1817). For advanced economies, the most common reason for farm trade restrictions in the past two centuries has been to protect domestic producers from import competition as they come under competitive pressure to shed labor in the course of economic development. But in the process, those protective measures hurt not only domestic consumers and exporters of other products but also foreign producers and traders of farm products, and they reduce national and global eco- nomic welfare. For decades, agricultural protection and subsidies in high-income (and some middle-income) countries have been depressing international prices of farm products, which lowers the earnings of farmers and associated rural busi- nesses in developing countries. The Haberler (1958) report to the General Agree- ment on Tariffs and Trade (GATT) contracting parties forewarned that such distortions might worsen, and indeed they did between the 1950s and the early 1980s (Anderson and Hayami 1986), thereby adding to global inequality and Five Decades of Distortions to Agricultural Incentives 5 poverty because three-quarters of the world's poorest people depend directly or indirectly on agriculture for their main income (World Bank 2007).2 In addition to this external policy influence on rural poverty, the governments of many developing countries have directly taxed their farmers over the past half century. A well-known example is the taxing of plantation crop exports in postcolonial Africa (Bates 1981). At the same time, many developing countries have chosen also to pursue an import-substituting industrialization strategy, predominantly by restricting imports of manufactures, and to overvalue their currency. Together, those measures indirectly taxed producers of other tradable products in developing economies, predominantly farmers (Krueger, Schiff, and Valdés 1988, 1991). Thus, the price incentives facing farmers in many developing countries have been depressed by agricultural price and international trade poli- cies in both their own and other countries. This disarray in world agriculture, as Johnson (1991) described it in the title of his seminal book, means there has been overproduction of farm products in high- income countries and underproduction in more-needy developing countries. It also means there has been less international trade in farm products than would be the case under free trade, thereby thinning markets for these weather-dependent products and thus making them more volatile. Using a stochastic model of world food markets, Tyers and Anderson (1992) find that instability of international food prices in the early 1980s was three times greater than it would have been under free trade in those products. During the past quarter century, however, numerous countries have begun to reform their agricultural price and trade policies, a process that has raised the extent to which farm products are traded internationally (see table 1.A.1 in the annex of this chapter), but not nearly as fast as globalization has proceeded in the nonfarm sectors of the world's economies.3 A key purpose of the present study is to examine empirically the extent to which those reforms have reversed the above-mentioned policy developments of the previous three decades. True, empirical indicators of farm sector support (called producer support estimates, or PSEs) have been provided in a consistent way for 20 years by the secretariat of the Organisation for Economic Co-operation and Development (OECD) for its 30 member countries (OECD 2008a). However, there are no comprehensive time series rates of assistance to producers of nonagricultural goods to compare with the PSEs, nor are there PSE figures for advanced economies in earlier decades, which are of more immediate relevance if how the two groups of countries' poli- cies evolved during similar stages of development is to be seen. As for developing countries, almost no comparable time series estimates have been generated since the Krueger, Schiff, and Valdés (1988) study, which covered the 1960­85 period for just 17 developing countries.4 An exception is a new set of estimates of nominal 6 Distortions to Agricultural Incentives: A Global Perspective rates of protection for key farm products in China, India, Indonesia, and Vietnam since 1985 (Orden et al. 2007). The OECD (2006) also has released PSEs for Brazil, China, and South Africa, and for several Eastern European countries. The present study complements and extends those two efforts and the seminal Krueger, Schiff, and Valdés (1988) study. It builds on them by providing similar estimates for other significant (including many low-income) developing economies, by developing and estimating new, more comprehensive policy indicators, and by employing the calculated price wedges in a global economy-wide model to estimate the effects of recent and prospective policy developments. These estimates can be helpful, as evidenced in subsequent chapters in this book, in addressing such questions as: Where is there still a policy bias against agricultural production? To what extent has there been overshooting in the sense that some developing-country food producers are now being protected from import competition along the lines of the examples of earlier-industrializing Europe and Japan? What are the political economy forces behind the more- successful reformers, and how do they compare with those in less-successful countries where major distortions in agricultural incentives remain? Over the past two decades, how important have domestic political forces been in bringing about reform relative to international forces (for example, loan conditionality, rounds of multilateral trade negotiations within the GATT, regional integration agreements, accession to the World Trade Organization [WTO], and the globalization of supermarkets and other firms along the value chain) and compared with forces operating in earlier decades? What explains the pattern of distortions across industries and the choice of support or tax instruments within the agricultural sector of each country? What policy lessons and market implications may be drawn from these differing experiences with a view to ensuring better growth- enhancing and poverty-reducing outcomes--including less overshooting, which results in protectionist regimes--in still-distorted economies during their reforms in the future? The study is timely for at least three reasons. First, the WTO is in the midst of the Doha Round of multilateral trade negotiations, and agricultural policy reform is one of the most contentious issues in those talks. Indeed economy-wide model- ing suggests that as much as two-thirds of the global welfare gains from removing all merchandise trade restrictions and agricultural subsidies would come from reform of agricultural policies, even though agriculture accounts for less than 8 percent of world gross domestic product (GDP) and exports (Anderson, Martin, and van der Mensbrugghe 2006). Second, poorer countries and their development partners are striving to achieve the Millennium Development Goals by 2015, in particular the goals related to alleviation of hunger and poverty. And third, world food prices spiked in 2007 and 2008 to extremely high levels. Governments in Five Decades of Distortions to Agricultural Incentives 7 some developing countries, in their haste to deal with the inevitable protests from consumers, reacted in far from optimal ways. Examining the policy responses to this and other episodes of food prices spikes (most notably in 1973­74) is impor- tant in determining which responses work better than others. The present study includes 75 countries that together account for between 90 and 96 percent of the world's population, farmers, agricultural GDP, and total GDP. The sample countries also account for more than 85 percent of farm produc- tion and employment in Africa, Asia, Latin America, and the transition economies of Europe and Central Asia. Per capita income in the sample countries covers a broad spectrum, from some of the poorest countries (Ethiopia and Zimbabwe) to the richest.5 Nominal rates of assistance and consumer tax equivalents (NRAs and CTEs) are estimated for more than 70 products, with an average of almost a dozen per country. In aggregate, the coverage represents around 70 percent of the gross value of agricultural production in the focus countries, and just under two-thirds of global farm production valued at undistorted prices over the period covered. Not all countries had data for the entire 1955­2007 period; the average number of years covered is 41 per country.6 Of the world's 30 most valuable agricultural products, the NRAs cover 77 percent of global output, ranging from two-thirds for livestock, three-quarters for oilseeds and tropical crops, and five-sixths for grains and tubers. Those products represent an even higher share (85 percent) of global agricultural exports (see appendix B of this volume for details). Having such a comprehensive coverage of countries, products and years offers the prospect of obtaining a reliable picture of both long-term trends in policies, and annual fluctuations around those trends, for individual countries and com- modities as well as for country groups, regions, and the world as a whole. The results can also serve as inputs into explanations of why government policies evolved as they did, and can thereby contribute to policy dialogues because, as Stigler (1975, p. ix) wrote,"Until we understand why our society adopts its policies, we will be poorly equipped to give useful advice on how to change those policies." This chapter begins with a brief summary of the long history of national distortions to agricultural markets. It then outlines the methodology used to gen- erate annual indicators of the extent of government interventions in markets, details of which are provided in Anderson et al. (2008a, 2008b) and appendix A of this volume. A description of the economies being examined and their economic growth and structural changes over recent decades is then briefly presented as a preface to the main section of the chapter, in which the NRA and CTE estimates are summarized across regions and over the decades since the 1950s. These esti- mates are discussed in far more detail in the regional studies that follow, chap- ters 2­10. A summary of an additional set of indicators of agricultural price dis- tortions, presented in chapter 11, is based on the trade restrictiveness index first 8 Distortions to Agricultural Incentives: A Global Perspective developed by Anderson and Neary (2005). In chapter 12, the focus shifts from countries to commodities, and various distortion indicators are used to provide a sense of how distorted each of the key farm commodity markets is globally. Chapter 13 uses the study's NRA and CTE estimates to provide a new set of results from a global economy-wide model. It quantifies the impacts of reforms under- taken since the early 1980s, and of the policies still in place as of 2004, on global markets, net farm incomes, and welfare. That chapter concludes by drawing on the lessons learned to speculate on the prospects for further reducing the disarray in world agricultural markets. National Distortions to Farmer Incentives: The Long History, Briefly While much government intervention in agricultural trade over the centuries has been aimed at stabilizing domestic food prices and supplies, there has been a gen- eral tendency for poor agrarian economies to tax agriculture relative to other sec- tors. As nations industrialize, their policy regimes typically gradually change from negatively to positively assisting farmers relative to other producers, and from subsidizing to taxing food consumers. Consider the United Kingdom, the first country to have an industrial revolu- tion. Prior to that period--from the late 1100s to the 1660s--England used export taxes and licenses to prevent domestic food prices from rising excessively. Then, from 1660­90, a series of laws gradually raised food import duties (making imports prohibitive under most circumstances) and reduced export restrictions on grain. Under the Corn Laws of 1815, these provisions were made even more protective of British farmers. The famous repeal of the Corn Laws in the mid- 1840s is often said to have heralded a period of relatively unrestricted food trade for the United Kingdom, although even then protection was retained for another generation for breweries and distilleries--and hence grain producers--via restric- tions on imports of wine and spirits. According to Nye (2007), it was only after the passage of the 1860 Anglo-French Treaty of Commerce that Britain moved closer to free trade than France, and that other European countries began to open up. But agricultural protection returned in the 1930s, and steadily increased over the next five decades. Indeed, the period of opening up in the 19th century was quite short for some countries in Europe, and agricultural protection levels in those countries throughout the 20th century were somewhat higher on average than in Britain. Kindleberger (1975) describes how the 19th century free trade movements in Europe reflect the national economic, political, and sociological conditions of the time. Agricultural trade reform was less difficult for countries, such as the United Five Decades of Distortions to Agricultural Incentives 9 Kingdom, with overseas territories capable of providing the metropole with a ready supply of farm products. The fall in the price of grain imports from the United States in the 1870s and 1880s provided a challenge for all, however. Denmark coped relatively well, by moving more into livestock production to take advantage of cheaper grain. Italy coped by sending many of its citizens to the New World. Farmers in France and Germany successfully sought protection from imports, however, and so began the post-industrial-revolution growth of agricul- tural protectionism in densely populated countries. Meanwhile, tariffs on West European imports of manufactures were progressively reduced after the GATT came into force in the late 1940s, thereby encouraging agricultural production rel- ative to manufacturing production (Lindert 1991; Anderson 1995). Japan provides an even more striking example of the tendency to switch from taxing to increasingly assisting agriculture relative to other industries. Its industri- alization began later than in Europe, after the opening up of the economy follow- ing the Meiji Restoration in 1868. By 1900, Japan had switched from being a small net exporter of food to becoming increasingly dependent on imports of rice (its main staple food, which was responsible for more than half the value of domestic food production). This was followed by calls from farmers and their supporters for rice import controls. Their calls were matched by equally vigorous calls from manufacturing and commercial groups for unrestricted food trade, since the price of rice at that time was a major determinant of real wages in the nonfarm sector. The heated debates were not unlike those that led to the repeal of the Corn Laws in the United Kingdom six decades earlier. In Japan, however, protectionist forces triumphed, and a tariff was imposed on rice imports from 1904. That tariff then gradually increased over time, raising the domestic price of rice to more than 30 percent above the import price during World War I. Even when there were food riots because of shortages and high rice prices just after the war, the Japanese gov- ernment's response was not to reduce protection but instead to extend it to its colonies and to shift from a national to an imperial rice self-sufficiency policy. That involved accelerated investments in agricultural development in the Korean and Taiwanese colonies behind an ever-higher external tariff wall that by the latter 1930s had driven imperial rice prices to more than 60 percent above those in international markets (Anderson and Tyers 1992). After Japan lost its colonies at the end of World War II, its agricultural protection growth resumed and spread from rice to an ever-wider range of farm products. Other high-income countries, which were settled by Europeans relatively recently and are far less densely populated, have had a strong comparative advan- tage in farm products for most of their history following Caucasian settlement. They have thus felt less need to protect their farmers than Europe or Northeast Asia. Indeed, until the present decade, Australia and New Zealand, like developing 10 Distortions to Agricultural Incentives: A Global Perspective countries, had in place policies that discriminated against their farmers (Anderson, Lloyd, and MacLaren 2007). In the Republic of Korea and in Taiwan, China in the 1950s, as in many newly independent developing countries, an import-substituting industrialization strat- egy was initially adopted, which harmed agriculture. But in those two economies, unlike in most other developing countries, that policy was replaced in the early 1960s with a more neutral trade policy that resulted in very rapid export-oriented industrialization. This development strategy in those densely populated economies imposed competitive pressure on the farm sector, which, just as in Japan in earlier decades, prompted farmers to lobby (successfully, as it happened) for ever-higher levels of protection from import protection (Anderson and Hayami 1986). Many less-advanced and less rapidly growing developing countries not only adopted import-substituting industrialization strategies in the late 1950s and early 1960s (Little, Scitovsky, and Scott 1970; Balassa and associates 1971) but also imposed direct taxes on their exports of farm products. It was common in the 1950s and 1960s--and in some cases all the way to the 1980s--to also use dual or multiple exchange rates so as to indirectly tax both exporters and importers (Bhagwati 1978; Krueger 1978). This added to the antitrade bias of developing countries' trade policies. Certainly within the agricultural sector of each country, import-competing industries tended to enjoy more government support than those that were more competitive internationally (Krueger, Schiff, and Valdés 1988; Herrmann et al. 1992; Thiele 2004). The Krueger, Schiff, and Valdés study also reveals that, at least up to the mid-1980s, direct disincentives for farmers, such as agricultural export taxes, were less important than indirect disincentives, such as import protection for the manufacturing sector or overvalued exchange rates, both of which pulled resources away from agricultural industries producing tradable products. In short, historically, countries have tended to gradually shift from taxing to subsidizing agriculture relative to other sectors in the course of their economic development, although less so, and at a later stage of development, the stronger a country's comparative advantage in agriculture (Anderson and Hayami 1986; Lindert 1991). Hence, at any point in time, farmers in poor countries have tended to face depressed terms of trade relative to product prices in international mar- kets, while the opposite was true for farmers in rich countries (Anderson 1995). The exceptions were rich countries with an extreme comparative advantage in agriculture--Australia and New Zealand. While the economic policy history of developing countries has been docu- mented extensively in previous surveys, less well known is the extent to which many emerging economies have belatedly followed the example of Korea and Taiwan, China in abandoning import substitution and opening their economies. Five Decades of Distortions to Agricultural Incentives 11 Some countries (for example, Chile) started in the 1970s while others (for exam- ple, India) did not do so in a sustained way until the 1990s. Other countries have adopted a very gradual pace of reform, with occasional reversals, while others have moved rapidly to open markets. Still others have adopted the rhetoric of reform but in practice have done little to open their economies. To get a clear sense of the overall impact of these reform attempts, there is no substitute for empirical analysis that quantifies the nature and degree of market intervention by governments over time. Methodology for Measuring Price Distortions7 The main focus of the present study's methodology is on the government- imposed distortions that create a gap between domestic prices and what they would be under free markets. Since it is not possible to understand the character- istics of agricultural development with a sectoral view alone, the study not only estimates the effects of direct agricultural policy measures (including distortions in the foreign exchange market) but also generates estimates of distortions in nonagricultural sectors for comparative evaluation. Specifically, the NRA for each farm product is computed as the percentage by which government policies have raised gross returns to farmers above what they would be without the government's intervention (or lowered them, if the NRA is less than zero). Product-specific input subsidies are included in those NRA esti- mates. A weighted average NRA for all covered products is derived using the value of production at undistorted prices as weights (in contrast to the PSEs and con- sumer support estimates, or CSEs, computed by OECD [2008a], which are expressed as a percentage of the distorted price). To that NRA for covered prod- ucts is added a "guesstimate" of the NRA for noncovered products (on average around 30 pecent of the total) and an estimate of the NRA from non-product- specific (NPS) forms of assistance or taxation. Since the 1980s, some high-income governments have also provided "decoupled" assistance to farmers, though because that support in principle does not distort resource allocation, its NRA has been computed separately and is not included for direct comparison with the NRAs for other sectors or for developing countries.8 Each farm industry is classified as import-competing, a producer of exportables, or a producer of nontradables (with its status sometimes changing over the years) in order to generate the weighted average NRAs for the two different groups of covered tradable farm products for each year. A production-weighted average NRA for nonagricultural tradables is also generated, for comparison with that for agricul- tural tradables, via the calculation of a percentage relative rate of assistance 12 Distortions to Agricultural Incentives: A Global Perspective (RRA), defined as: RRA 100*[(100 NRAagt) (100 NRAnonagt) 1], t t where NRAag and NRAnonag are the percentage NRAs for the tradable parts of the agricultural (including noncovered) and nonagricultural sectors, respectively.9 Since the NRA cannot be less than 100 percent if producers are to earn anything, neither can the RRA (since the weighted average NRAnonagt is non-negative in all our country case studies). If the agricultural and nonagricultural sectors are equally assisted, the RRA is zero. This measure is useful in that if it is below (above) zero, it provides an internationally comparable indication of the extent to which a country's sectoral policy regime has an antiagricultural (proagricultural) bias. This approach is not well suited to analysis of the policies of former socialist economies in Europe and Asia prior to their reform era, as prices played only an accounting function and currency exchange rates were enormously distorted. The price comparison approach, however, provides a valuable a set of indicators for the reform era in these countries, as for other market economies, of distortions to incentives for farm production, consumption, and trade, and of the income trans- fers associated with interventions.10 In addition to the mean NRA, a measure of the dispersion or variability of the NRA estimates across the covered farm products also is generated for each econ- omy. The cost of government policy distortions to incentives in terms of resource misallocation tends to grow as the degree of substitution in production increases. In the case of agriculture, which involves the use of farmland that is sector-specific but transferable among farm activities, the greater the variation of NRAs across industries within the sector, the higher the welfare cost of those market interven- tions. A simple indicator of dispersion is the standard deviation of the covered industries' NRAs. While most of the focus is on agricultural producers, this study also considers the extent to which consumers are taxed or subsidized. To do so, a CTE is calcu- lated by comparing the price that consumers pay for their food and the interna- tional price of each food product at the border. Differences between the NRA and the CTE arise from distortions in the domestic economy that are caused by trans- fer policies and taxes and subsidies that cause the prices paid by consumers (adjusted to the farmgate level) to differ from those received by producers. In the absence of any other information, the CTE for each tradable farm product is assumed to be the same as the NRA from border distortions and the CTE for non- tradable farm products is assumed to be zero. To obtain dollar values of farmer assistance and consumer taxation, the coun- try authors' NRA estimates are multiplied by the gross value of production at undistorted prices to obtain an estimate in U.S. dollars of the direct gross subsidy equivalent (GSE) of assistance to farmers. These GSE values are calculated in con- stant dollars and are also expressed on per-farm-worker basis. Likewise, a value of the consumer transfer is derived from the CTE, by assuming consumption value is Five Decades of Distortions to Agricultural Incentives 13 the gross value of production at undistorted prices divided by the self-sufficiency ratio for each product (production divided by consumption, derived from national volume data or the Food and Agriculture Organization's [FAO's] com- modity balance sheets). These transfer values can be summed across products for a country, and across countries for any or all products, to obtain regional aggre- gate transfer estimates for the studied economies. Needless to say, there are numerous challenges in applying the above method- ology, especially in developing economies with poor-quality data. Ways to deal with the standard challenges are detailed in the second section of appendix A, while country-specific issues are discussed in the relevant subsequent chapters. The next section summarizes estimates that have emerged from aggregating the NRA and related estimates provided by the project's country case studies,11 prefaced by a brief review of the relative size, economic growth, and structural changes that have taken place in key regions of the world over recent decades. After the national country studies were completed and the global database of NRA and CTE estimates was assembled (see Anderson and Valenzuela 2008), it was possible to estimate partial equilibrium country and global commodity indexes of the trade and welfare restrictiveness of agricultural policies. Since those estimates are not part of the country studies reported in Parts II and III of this volume, the methodology for them is not laid out until chapter 11, following which is a summary of the index estimates. Empirical Estimates of National Distortions to Farmer Incentives For the purposes of the present study, the world economy is divided into high- income areas (Western Europe, the United States and Canada, Japan, and Aus- tralia and New Zealand),12 three developing country regions (Africa, Asia, and Latin America), and Europe's transition economies in the 1990s plus Turkey. (Turkey is included in this last group because it is in the same geographic region and, like others in that region, has been seeking EU accession, which necessarily has influenced the evolution of its agricultural price and trade policies.) North America and Europe (including the newly acceded eastern members of the EU) each account for almost one-third of the global economy, and the remaining one-third is shared almost equally by developing countries and the other high-income countries. When the focus turns to just agriculture, however, developing countries are responsible for slightly over half the value added globally, with Asia accounting for two-thirds of that lion's share. The developing countries' majority becomes stronger still in terms of global population and even more so in terms of farmers, almost three-quarters of whom are in Asian develop- ing countries. Hence the vast range of per capita incomes and agricultural land 14 Distortions to Agricultural Incentives: A Global Perspective per capita, and thus agricultural comparative advantages, across the country groups in table 1.1 and the strong concentration of poor people in Asia. The num- ber of poor people in Asia has diminished dramatically over the past quarter cen- tury, though, and even more as a percentage of Asia's population (unlike in Africa), but 60 percent of the world's population living on less than US$1 a day still lived in Asia in 2005 (down from 87 percent in 1981 and 76 percent in 1993-- see table 1.2). The decline in Asian poverty was associated with much faster eco- nomic growth and export-led industrialization in the region than in the rest of the world: since 1980, Asia's per capita GDP has grown at four times, and exports at nearly two times, the global average (table 1.3). The share of Asia's GDP that is exported is now one-third above that for the rest of the world and for Latin America, as summarized from the World Bank (2007) by Sandri, Valenzuela, and Anderson (2007). In 2000­04, just 12 percent of Asia's GDP came from agriculture, on average. That stands in contrast to the situation in Africa, where the share for our focus countries ranges from 20 to 40 percent, and in even starker contrast to Latin America and Europe's transition economies, where it is down to 6 percent (and to just 2 percent on average in high-income countries). The share of employment in agriculture remains very high in Asia, however, at just under 60 percent--the same as in Africa and three times the share in Latin America and Eastern Europe, although more farmers work part-time on their farms in Asia than in other devel- oping countries. By contrast, less than 4 percent of workers in high-income countries are still engaged in agriculture. Hence the much greater impact of own- country and rest-of-world distortions to agricultural incentives to welfare, inequality, and poverty in developing countries than in high-income countries. NRAs and GSEs to agriculture Perhaps the simplest measure for capturing the aggregate extent of distortions to agricultural prices globally is to examine the trend in the GSE of assistance (posi- tive or negative) to farmers. Figure 1.1a shows the GSE for five-year periods since 1960 for the world as a whole, for developing countries, and for high-income countries plus Europe's transition economies. The dark line suggests that, apart from the dip during the period of high world food prices in 1973­74, the net global GSE has been steadily rising over the past half century, especially when the decoupled assistance to farmers in high-income countries is included.13 The decomposition of those global transfers by country group in figure 1.1a reveals two distinct trends, each kinked partway through the period. On one hand, aggregate support for farmers in high-income countries rose steadily throughout the period from the 1950s to the early 1990s, before declining slightly over the 15 subsequent years--somewhat more rapidly at the end of the period, when Table 1.1. Key Economic and Trade Indicators of Focus Countries, by Region, 2000­04 National relative to world (world 100) Share (%) of world Agricultural Agricultural RCA,a trade Agricultural Agricultural GDP land agriculture specialization Population Total GDP GDP workers per capita per capita and food indexb Africa 10 1 6 11 14 148 -- -- Asia 51 10 37 73 20 34 80 0.03 Latin America 8 5 8 3 64 171 -- -- Europe and Central Asia 7 4 6 3 48 178 -- -- Western Europe 6 29 16 1 454 46 106 0.03 United States and Canada 5 33 11 0.3 636 186 119 0.08 Australia and New Zealand 0.4 2 2 0.1 405 2,454 354 0.62 Japan 2 13 5 0.2 610 5 12 0.84 All focus countries 90 96 91 92 -- -- -- -- Other (nonfocus) developing and transition economies 10 4 9 8 -- -- -- -- Source: Sandri, Valenzuela, and Anderson (2007), compiled mainly from World Bank 2007. a. RCA, the revealed comparative advantage index, is the share of agriculture and processed food in national exports as a ratio of that sector's share of global exports. b. Primary agricultural trade specialization index is net exports as a ratio of the sum of exports and imports of agricultural and processed food products (world average 0.0). 15 16 Distortions to Agricultural Incentives: A Global Perspective Table 1.2. Poverty in Africa, Asia, Latin America, and Europe's Transition Economies, 1981­2005 % of people living on less than US$1/day who are rural, 1981 1993 2005 2002 People living on less than $1/day (millions): Sub-Saharan Africa 157 247 299 69 East Asia and Pacific 948 600 180 85 of which China 730 444 106 90 South Asia 387 341 350 75 of which India 296 280 267 74 Latin America and the Caribbean 27 34 28 34 European transition economies 3 10 16 50 World 1,528 1,237 879 74 Asia's share of world 87 76 60 n.a. People living on less than $1/day (% of population): Sub-Saharan Africa 40 44 39 East Asia and Pacific 69 36 10 of which China 74 38 8 South Asia 42 29 24 of which India 42 31 24 Latin America and the Caribbean 7 7 5 European transition economies 1 2 3 World 42 27 16 Source: Chen and Ravallion (2008) and, for rural share, Ravallion, Chen, and Sangraula (2007). Note: n.a. not applicable. world food prices shot up. On the other hand, farmers in developing countries were increasingly taxed by price and trade policies from the early 1960s to the late 1970s and early 1980s, a trend that gradually reversed and, by the mid-1990s, resulted in positive aggregate assistance to farmers. Thus, the GSE contributions of the two groups to the global trend are offsetting in the 1980s but additive from the early 1990s to 2004. Over 2005­07, when food prices in international markets rose steeply, transfers to farmers in high-income countries fell back considerably (simi- lar to what happened in 1973­74). Though there are not sufficient estimates to show the change during those recent years for developing countries, their governments also responded by reducing or suspending import tariffs and temporarily restricting the export of food, so developing countries may have con- tributed to, rather than offset, the downward GSE trend in high-income countries. Five Decades of Distortions to Agricultural Incentives 17 Table 1.3. Growth of Real GDP and Exports, Focus Countries, 1980­2004 (at constant 2000 prices, percent per year, trend-based) Total GDP per Export Agriculture Industry Services GDP capita volume Africa -- -- -- -- -- -- Asia 3.1 8.6 7.5 7.1 5.5 11.2 Latin America -- -- -- 5.4 3.6 7.2 Europe and Central Asia -- -- -- -- -- -- Western Europe 0.8 1.4 2.6 2.3 2.0 5.5 United States and Canada 2.7 2.6 3.4 3.2 2.0 6.7 Australia and New Zealand 2.8 2.5 3.6 3.3 2.0 6.5 Japan 1.7 2.0 3.0 2.5 2.1 4.0 All focus countries World 2.0 2.5 3.2 3.0 1.4 6.1 Source: Sandri, Valenzuela, and Anderson (2007), compiled from World Bank 2007. Note: -- not available. Figure 1.1. Gross Subsidy Equivalents of Assistance to Farmers, over Time and by Region,a 1955­2007 a. Over time 300 constant 2000 US$ billions 200 100 0 100 200 59 64 69 4 9 4 9 4 9 4 7 ­7 ­7 ­8 ­8 ­9 ­9 ­0 ­0 ­ ­ ­ 55 60 70 75 80 85 90 95 00 05 65 19 19 19 19 19 19 19 19 20 20 19 net, global (decoupled payments are included in the higher, gray line) developing countries (no averages for periods 1955­59 and 2005­07) high-income countries and Europe's transition economies 18 Distortions to Agricultural Incentives: A Global Perspective Figure 1.1. (continued ) b. By region 130 90 constant 2000 US$ billions 50 10 0 30 70 110 Africa Australia Latin Europe North Japan Western Asia and America and America Europe New and the Central Zealand Caribbean Asia 1980­84 2000­04 c. Among high-income countries 100 90 80 70 60 percent 50 40 30 20 10 0 56 59 62 65 68 71 74 77 80 83 86 89 92 95 98 01 04 07 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 Australia and New Zealand Canada United States Japan Western European non-EU countries EU countries Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Five Decades of Distortions to Agricultural Incentives 19 In figure 1.1b, GSE contributions are further subdivided into four high-income regions and four developing-country regions. It is clear that Japan and Western Europe are the biggest contributors among high-income countries and that Asia is the biggest contributor by far to the developing country trends--in both subperi- ods. As will become clear, the latter result is largely driven by China and India, although growth in protection in Korea added to the overall increase in assistance in the region. Figure 1.1c shows a further disaggregation of the contribution of the various high-income countries over the full time series. It reveals that while the EU (the EU-15, ignoring the new East European members), Japan, and the United States were equally large contributors in 2005­07, the original six members of the European Economic Community were responsible for more than half of the aggregate support in high-income countries during the first decade of the Com- mon Agricultural Policy (CAP) (1963­72) and Japan and the United States for less than one-fifth each. Though the GSE estimates account for inflation, by being expressed in constant dollar terms, they do not take into account the fact that the agricultural sector is growing and at different rates across countries. That shortcoming is avoided by using the NRA, which is the extent of policy-induced price support expressed as a percentage of the value of agricultural output at undistorted prices. Figure 1.2 and table 1.4, which show the same eight country groups as figure 1.1b, Figure 1.2. NRAs to Agriculture,a by Regions, 1980­84 and 2000­04a 140 120 100 80 percent 60 40 20 0 20 40 Africa Asia Latin Europe Australia North Western Japan America and and America Europe and the Central New Caribbean Asia Zealand 1980­84 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. Includes noncovered products and NPS (but not decoupled) assistance. 20 Table 1.4. NRAs to Agriculture,a Focus Countries, 1955­2007c (percent) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa 14 8 11 15 13 8 1 9 6 7 -- Asia 27 27 25 25 24 21 9 2 8 12 -- Latin America 11 8 7 21 18 13 11 4 6 5 -- Europe and Central Asiab -- -- -- -- -- -- -- 10 18 18 25 Western Europe 44 57 68 46 56 74 82 64 44 37 18 United States and Canada 13 11 11 7 7 13 19 16 11 17 11 Australia and New Zealand 6 7 10 8 8 11 9 4 3 1 2 Japan 39 46 50 47 67 72 119 116 120 120 81 Developing countries 26 23 22 24 22 18 8 2 6 9 -- High-income countries 22 29 35 25 32 41 53 46 35 32 17 All focus countries (weighted average) 3 5 6 0 2 5 17 18 17 18 -- Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Note: -- not available. a. Weighted average for each country, including NPS assistance and authors' guesstimates for noncovered farm products (but not decoupled assistance), with weights based on gross value of agricultural production at undistorted prices. Estimates for China pre-1981 and India pre-1965 are based on the assumption that the NRA to agriculture in those years was the same as the average NRA estimates for those countries for 1981­84 and 1965­69, respectively, and that the gross value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. Developing-country and world aggregates are computed accordingly. b. Countries in Europe and Central Asia are not included in the high-income or developing-country aggregates. Five Decades of Distortions to Agricultural Incentives 21 reveal much more similarity among the regions' NRAs than the GSEs for the developing-country groups. They also reveal that Japan is now proportionately far more protective of its farmers than Western Europe. The bottom row of table 1.4 provides a similar trend to the GSE trend line: the global average NRA, which was close to zero up to the early 1980s, increased as developing countries began to phase out export taxation, and has since hovered at about 18 percent. The GSE and NRA numbers are based not only on estimates of assistance to covered products but also on NPS assistance and on guesstimates of assistance to the roughly 30 percent of the value of farm products that have not been included in the study's explicit price comparison exercise. Figure 1.3 summa- rizes the most robust NRA estimates for covered farm products, categorized each year as either exportable, import-competing, or nontradable. The weighted average trends across those covered products are very similar to the GSE trends both for high-income countries and, in a slightly more muted fashion, for developing countries. What is more striking about figure 1.3 is the marked difference in the levels of support to import-competing versus exportable covered products. Exportables in high-income countries have received relatively little support other than during the export subsidy war of the mid-1980s, while in developing countries they were increasingly taxed from the late 1950s until the 1980s. Taxation has been gradually phased out over the past two decades, although a little remained in 2004, for exam- ple in Argentina, and considerably more was added by various developing coun- tries in 2008 in response to concerns about the spike in international food prices. Importables, by contrast, have been assisted throughout the past five decades, and the long-run fitted trend line has almost the same slope for both sets of countries (but a lower intercept for developing countries). Two lessons can be drawn from this: first, there is a strong antitrade bias for agricultural goods in high-income and developing countries that has not diminished much over the past half century (in fact, it increased substantially in the 1980s); and second, growth in agricultural import protection appears to have accompanied global economic growth through the 1990s, and has slowed only slightly since then. These lessons hold even when the eight regions examined here are disaggregated, as shown in table 1.5, which also reports estimates of a trade bias index (TBI) for each region. Though the index confirms the broad global conclusion, it indicates a sizable decline in the antitrade bias in developing countries since the 1980s, especially in Asia. That antitrade bias means that the rates of assistance are not uniform across commodities, which indicates that the resources being used within the farm sector are not being put to their best use. The extent of that inefficiency--over and above that caused by too many or too few resources in aggregate in the sector--is crudely indicated by the standard deviation of NRAs among covered products in each 22 Distortions to Agricultural Incentives: A Global Perspective Figure 1.3. NRAs to Exportable, Import-Competing, and All Covered Agricultural Products,a High-Income and Developing Countries, 1955­2004 a. Developing countries 90 70 50 30 percent 10 10 30 50 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 b. High-income countries plus Europe's transition economies 90 70 50 30 percent 10 10 30 50 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 import-competing total exportables Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. Covered products only. The total also includes nontradables. focus country. This dispersion index, summarized for our eight regions in table 1.6, has fluctuated across time and varied between regions, but the global average has remained around 70 percent throughout the period, with no discernable trend. Table 1.5. NRAs to Agricultural Exportables, Import-Competing Products, and the TBI,a Focus Regions, 1955­2007 (percent) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa NRA, agricultural exportables -- 30.1 38.4 42.6 42.6 35.0 36.7 35.8 26.1 24.6 -- NRA, agricultural import- competing products -- 18.6 11.8 1.9 14.5 13.2 58.3 5.2 9.8 1.6 -- TBI -- 0.41 0.45 0.44 0.50 0.43 0.60 0.39 0.33 0.26 -- Latin America NRA, agricultural Exportables -- 20.4 12.8 27.0 25.2 27.1 25.0 10.5 3.5 4.6 -- NRA, agricultural import- competing products -- 26.3 8.7 2.8 1.1 13.6 5.1 19.4 12.5 20.6 -- TBI -- 0.37 0.20 0.25 0.26 0.36 0.29 0.25 0.14 0.21 -- South Asiac NRA agricultural exportables -- 37.5 37.2 30.0 36.1 27.9 20.6 15.8 12.0 6.2 -- NRA agricultural import- competing products -- 39.2 41.2 39.4 45.1 37.9 63.3 25.1 14.5 26.5 -- TBI -- 0.55 0.56 0.50 0.56 0.48 0.51 0.33 0.23 0.26 -- China and Southeast Asia c NRA, agricultural exportables -- 55.5 55.1 51.8 50.1 50.0 41.0 20.8 2.2 0.1 -- NRA, agricultural import- competing products -- 10.3 8.9 9.4 2.6 0.5 15.1 3.3 13.3 12.3 -- TBI -- 0.50 0.51 0.47 0.49 0.50 0.49 0.23 0.14 0.11 -- (Table continues on the following pages.) 23 24 Table 1.5. NRAs to Agricultural Exportables, Import-Competing Products, and the TBI,a Focus Regions, 1955­2007 (continued) (percent) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Japan, Korea, and Taiwan, China NRA, agricultural exportables 18.1 5.7 4.3 15.4 10.3 25.1 48.9 57.1 57.0 70.3 -- NRA, agricultural import- competing products 35.6 43.3 52.8 54.1 76.6 83.7 124.9 127.4 127.0 134.6 122.6 TBI 0.40 0.26 0.32 0.25 0.38 0.32 0.34 0.31 0.31 0.27 -- European transition economies NRA, agricultural exportables -- -- -- -- -- -- -- 3.2 1.0 1.0 15.2 NRA, agricultural import- competing products -- -- -- -- -- -- -- 32.5 35.4 35.7 32.3 TBI -- -- -- -- -- -- -- 0.27 0.27 0.27 0.13 Western Europe NRA, agricultural exportables 9.3 17.4 31.7 22.5 33.3 31.1 50.1 38.0 15.0 8.1 1.7 NRA, agricultural import- competing products 59.4 77.2 82.9 55.7 61.7 79.5 87.6 67.2 52.8 50.5 28.9 TBI 0.31 0.34 0.28 0.21 0.18 0.27 0.20 0.17 0.25 0.28 0.21 North America NRA, agricultural exportables 2.7 2.8 6.1 5.1 2.9 5.4 10.5 6.0 5.4 7.6 4.1 NRA, agricultural import- competing products 8.6 9.3 8.8 6.7 10.5 19.7 23.6 18.6 11.3 16.8 11.0 TBI 0.05 0.06 0.02 0.01 0.07 0.11 0.10 0.10 0.05 0.08 0.06 Australia and New Zealand NRA, agricultural exportables 3.8 4.7 6.6 5.8 5.5 7.6 6.5 3.6 2.2 0.2 0.2 NRA, agricultural import- competing products 7.9 8.3 9.3 11.7 8.7 8.4 6.5 3.8 2.0 2.0 1.5 TBI 0.04 0.03 0.02 0.05 0.03 0.01 0.00 0.00 0.00 0.02 0.01 Developing countriesc NRA, agricultural exportables -- 46.5 44.6 45.4 43.9 41.4 35.8 18.7 5.5 3.0 -- NRA, agricultural import- competing products -- 12.7 13.5 7.8 12.8 16.5 37.7 22.6 22.0 23.0 -- TBI -- 0.53 0.51 0.49 0.50 0.50 0.53 0.34 0.23 0.21 -- High-income countries NRA, agricultural exportables 4.2 7.4 13.5 10.3 11.3 12.1 22.3 15.9 8.1 6.9 2.9 NRA, agricultural import- competing products 31.2 45.9 50.2 36.5 47.4 58.1 71.4 62.4 53.9 50.7 30.8 TBI 0.21 0.26 0.24 0.19 0.24 0.29 0.29 0.29 0.30 0.29 0.21 World c NRA, agricultural exportables -- 23 20 23 25 24 17 7 1 0 -- NRA, agricultural import- competing products -- 35 37 27 34 38 57 43 38 36 -- TBI -- 0.43 0.42 0.39 0.44 0.45 0.47 0.35 0.28 0.26 -- Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Note: -- not available. a. NRAs for noncovered products are included here (unlike in figure 1.3). b. TBI (1 NRAagx 100)/(1 NRAagm/100) 1, where NRAagx and NRAagm are the weighted average percentage NRAs for the exportable and import-competing parts of the agricultural sector, with weights based on production valued at undistorted prices. TBIs shown here are calculated using the regional five-year averages of NRAagx and NRAagm. c. Estimates for China pre-1981 and India pre-1965 are based on the assumption that NRAs to agriculture in those years were the same as the average NRA estimates for those countries for 1981­84 and 1965­69, respectively, and that the gross value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. The developing-country and world averages are computed accordingly. 25 26 Table 1.6. Dispersion of NRAs across Covered Agricultural Products,a Focus Regions, 1965­2007 (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa 31 30 37 36 36 31 25 25 -- Asia 56 42 49 53 66 56 57 64 -- Latin America 49 44 52 52 44 42 32 40 -- Europe and Central Asia 34 33 41 26 39 56 39 45 44 Western Europe 119 85 112 98 122 86 69 74 64 United States and Canada 29 15 31 62 71 39 31 37 28 Australia and New Zealand 40 45 26 17 20 14 12 7 5 Japan 69 82 156 143 175 162 136 143 116 All focus countries (weighted average) 54 45 55 51 59 53 43 48 -- Product coverageb 68 70 71 73 73 72 71 68 70 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Note: -- not available. a. Dispersion for each region is a simple average of the country-level annual standard deviations around a weighted mean of NRAs per country across covered products each year. b. Share of gross value of total agricultural production at undistorted prices accounted for by covered products. Five Decades of Distortions to Agricultural Incentives 27 The NRAs for different covered products are not random, because their weighted averages across developing or high-income countries cover a wide spectrum. Fig- ure 1.4 shows that rice, sugar, and milk (the rice pudding ingredients) are by far the most assisted farm industries in both sets of countries, with beef and poultry next. For high-income countries, cotton has the next-highest NRA after beef and poultry, yet it has the lowest (most negative) NRA in developing countries. When the country authors' best guesstimates of the NRAs for the various non- covered farm products are aggregated across countries, their weighted average tends to be less than half that for covered products. This finding is not surprising, since many of them are nontraded staples or fruits and vegetables that receive lit- tle or no government attention--part of the reason they were not covered by authors in the first place. Their inclusion in the weighted average across all prod- ucts therefore dampens the positive or negative NRA average for a region's covered products. Though NPS and decoupled assistance to farmers added only a Figure 1.4. NRAs, Key Covered Products, High-Income and Developing Countries, 1980­84 and 2000­04 a. Developing countries b. High-income countries sugar rice 387 milk sugar rice milk poultry beef wheat poultry maize cotton pig meat pig meat coffee soybeans soybeans maize beef wheat coconuts barley cotton rapeseed 100 50 0 50 100 150 50 0 50 150 250 percent percent 1980­84 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. 28 Distortions to Agricultural Incentives: A Global Perspective little to the NRA for most developing countries, NPS added 2 or 3 percentage points until the early 1980s and has contributed about 5 points since then for high-income countries. If decoupled payments are included, the NRA for high- income farmers rises another 5 percentage points for the 1980s and early 1990s and as many as 10 points by during the present decade (table 1.7). Assistance to nonfarm sectors and RRAs The antiagricultural policy biases of the past were due to more than agricultural policies. Also important in developing countries, according to Krueger, Schiff, and Valdés (1988), has been border protection to the manufacturing sector--the dominant intervention in the tradables part of nonagricultural sectors. Unfortu- nately, it has not been possible with the resources available for this study to quan- tify the distortions to nonfarm tradable sectors as carefully as for agriculture. Authors typically have had to rely on applied trade taxes (for exports as well as imports) rather than being able to undertake price comparisons, and hence they usually do not capture quantitative restrictions on trade, which were important in earlier decades but decreasingly so in recent times. Nor do they capture distor- tions in the services sectors, some of which now produce tradables (or would in the absence of interventions preventing their emergence). As a result, the esti- mated NRAs for nonfarm importables are smaller and decline less rapidly than in fact was the case--and likewise for nonfarm exportables, except their NRAs in some cases would have been negative. Of those two elements of underestimation, the former bias certainly dominates, so the authors' estimate of the overall NRA for nonagricultural tradables should be considered a lower-bound estimate (more so in the past), so that its decline is less rapid than it should be.14 Despite these methodological limitations, the estimated NRAs for nonfarm tradables are very sizeable prior to the 1990s. For developing countries as a whole, the average NRA value has steadily declined throughout the past four to five decades, from around 45 percent in the 1960s to around 30 percent in the 1970s, 16 pecent in the 1980s, and less than 10 percent since the mid-1990s as policy reforms have spread (second last row of table 1.7). The decline in NRA has contributed to a decline in the estimated negative RRA for farmers: the weighted average RRA was worse than 50 percent up to the mid-1970s but improved to an average of 38 percent in the 1980s, 12 percent in the 1990s, and just above zero (1 percent) in 2000­04. The trend in RRAs and their two component NRAs for developing coun- tries is starkly illustrated in figure 1.5, where it is clear that the falling positive NRAs for nonfarm producers have contributed even more to the rise of the RRA than has the gradual disappearance of the negative NRAs for farmers. When trends are decomposed by region, it is clear that Asia has been the major contributor to this Table 1.7. NRAs to Agricultural Products Relative to Nonagricultural Industries, 1955­2007 (percent, weighted averages) a. Developing countriesd 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 1965­69 NRA, covered products 33.4 29.6 28.8 30.2 27.6 23.3 13.2 4.9 3.8 6.7 NRA, noncovered products 9.0 7.9 7.6 9.8 9.8 7.1 0.3 0.1 3.9 6.3 NRA, all agricultural products 27.1 24.0 23.1 24.9 23.1 19.1 9.8 3.5 3.9 6.6 Total agricultural NRA (including NPS)a 25.8 22.7 21.8 23.7 22.0 17.8 8.3 1.8 5.7 8.7 NRA, decoupled assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.2 0.3 Total agricultural NRA (including NPS decoupled assistance) 25.8 22.7 21.8 23.7 22.0 17.8 8.3 1.6 5.9 9.0 NRA, all agricultural tradablesa 27.9 25.4 25.3 28.2 25.7 20.9 10.3 2.2 6.2 8.9 NRA, nonagricultural tradables 56.9 43.1 45.0 30.6 27.3 18.8 14.0 12.7 9.0 5.7 RRAc 54.1 47.7 48.4 44.9 41.6 32.9 21.2 13.1 2.5 3.1 (Table continues on the following page.) 29 30 Table 1.7. NRAs to Agricultural Products Relative to Nonagricultural Industries, 1955­2007 (continued) (percent, weighted averages) b. High-income countries (including Europe and Central Asia) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 1965­69 1970­74 NRA, covered products 22.0 30.9 37.0 27.2 34.2 38.9 55.6 43.2 32.0 28.7 16.6 NRA, noncovered products 10.0 14.6 14.9 11.5 14.1 13.7 21.1 17.0 17.0 16.7 91.9 NRA, all agricultural products 18.5 26.2 30.7 22.8 28.6 31.8 45.6 35.6 27.5 25.0 12.6 Total agricultural NRA (including NPS)ba 22.3 28.8 32.4 23.7 30.2 36.3 49.6 40.0 31.3 29.4 16.5 NRA, decoupled assistance 0.0 0.0 0.0 0.0 0.6 4.7 5.8 5.5 8.1 9.6 9.9 Total agricultural NRA (including NPS decoupled assistance) 22.3 28.8 32.4 23.7 30.8 41.0 55.4 45.5 39.4 39.0 26.4 NRA, agricultural tradables 23.0 30.2 34.2 25.0 32.2 37.9 52.0 41.7 32.5 30.1 17.0 NRA, nonagricultural tradables 7.5 8.7 9.1 6.3 4.5 3.8 3.7 2.5 2.1 1.8 0.2 RRAc 14.3 19.7 23.0 17.6 26.5 32.8 46.6 38.2 29.8 27.8 17.2 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. NRAs including NPS assistance, that is, the assistance to all primary factors and intermediate inputs as a percentage of the total primary agricultural production valued at undistorted prices. b. TBI (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricul- tural sector. The regional average TBI is calculated from the regional averages of the NRAs for exportable and import-competing parts of the agricultural sector. c. RRA 100*[(100 NRAagt )/(100 NRAnonagt ) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. d. Estimates for the NRA and RRA for China pre-1981 and India pre-1965 are based on the assumption that the agricultural NRAs in those years were the same as the average NRA estimates for those countries for 1981­84 and 1965­69, respectively, and that the value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. Developing-country and world aggregates are computed accordingly. Five Decades of Distortions to Agricultural Incentives 31 Figure 1.5. NRAs to Agricultural and Nonagricultural Tradable Products and RRA,a All Focus Countries, 1955­2004 a. Developing countriesb 100 80 60 40 percent 20 0 20 40 60 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 b. High-income countries (not including those in Europe and Central Asia) 100 80 60 40 percent 20 0 20 40 60 9 4 9 4 9 4 9 4 9 4 ­5 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 55 60 65 70 75 80 85 90 95 00 19 19 19 19 19 19 19 19 19 20 NRA, nonagricultural NRA, agricultural RRA Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. The RRA is defined as 100*[(100 NRAag t ) (100 NRAnonagt ) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. b. Estimates for China pre-1981 and India pre-1965 are based on the assumption that the NRA to agriculture in those years was the same as the average NRA estimates for those countries for 1981­84 and 1965­69, respectively, and that the gross value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. 32 Distortions to Agricultural Incentives: A Global Perspective Figure 1.6. RRAs to Tradables,a Asia, Africa, and Latin America, 1965­2004 20 10 0 10 percent 20 30 40 50 60 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Africa Latin America and the Caribbean Asia Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. Five-year weighted averages with value of production at undistorted prices as weights. In Asia, estimates for China pre-1981 are based on the assumption that the NRA to agriculture and nonagricultural tradables, and hence the RRA in those earlier years, were the same as the average NRA estimates for China in 1981­89. dramatic reform (figure 1.6 and table 1.8). Within Asia, China and India have con- tributed most to that outcome (Anderson and Martin 2009b). Although average changes in NRAs and RRAs for high-income countries look minor by comparison with those for developing countries, they hide major differences across the high-income countries. Even figure 1.7 doesn't do justice to the striking case of Australia and New Zealand, where the phasing out of high manufacturing tariffs provided nearly as large a boost to farmer incentives as in developing countries as a group. There are two differences between the trend in those two farm-exporting countries and the trend in developing countries, how- ever. One is that much of the cut in Australia's manufacturing protection occurred in the decade before 1955­59, when the RRA was far more negative (not shown in figure 1.7 but see Anderson et al. 2009); the other is that Australian and New Zealand farmers enjoyed considerable direct assistance for much of the time manufacturing was protected, and that farm assistance was cut at the same time as was manufacturing protection, which kept the RRA from rising greatly starting in the 1970s (table 1.8). Despite the very considerable policy reforms of the past quarter century, there is still a long way to go before differences in rates of assistance across countries are Table 1.8. NRAs to Agricultural and Nonagricultural Tradables, and the RRA,a by Region, 1955­2007 (percent) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa NRA, agricultural -- 13.3 19.6 25.0 22.1 13.5 0.3 15.4 8.7 12.0 -- NRA, nonagricultural -- 3.7 2.7 1.5 5.7 1.6 9.2 2.7 2.0 7.3 -- RRA -- 15.2 21.4 26.0 25.9 13.1 8.3 17.1 10.4 18.0 -- Latin America NRA, agricultural -- 11.4 9.3 23.0 19.0 12.9 11.2 4.4 5.5 4.9 -- NRA, nonagricultural -- 26.9 31.3 27.8 23.3 18.5 16.8 7.3 6.6 5.4 -- RRA -- 30.2 30.9 39.8 34.2 26.6 24.0 2.7 1.0 0.5 -- South Asiab NRA, agricultural -- 4.1 4.4 9.7 7.7 1.8 47.1 0.2 2.4 12.7 -- NRA, nonagricultural -- 114.4 117.8 81.7 57.8 54.6 39.9 18.6 15.0 10.1 -- RRA -- 51.5 51.9 39.8 41.6 33.3 5.1 15.5 14.9 3.4 -- China and Southeast Asiab NRA, agricultural -- 43.6 42.6 40.1 35.7 34.5 27.8 12.0 4.9 7.1 -- NRA, nonagricultural -- 36.5 36.5 33.7 30.8 20.6 23.3 19.8 9.6 5.5 -- RRA -- 58.7 58.0 55.2 50.8 43.4 41.6 26.4 4.2 1.5 -- Japan, Korea, and Taiwan, China NRA, agricultural 30.1 39.9 48.8 51.3 75.5 78.8 124.3 129.9 130.5 138.1 126.1 NRA, nonagricultural 8.6 8.3 6.1 4.2 3.5 2.4 2.5 1.4 1.1 0.6 1.0 RRA 19.7 29.1 40.2 44.9 69.6 74.6 118.7 126.7 128.1 136.7 123.7 (Table continues on the following pages.) 33 34 Table 1.8. NRAs to Agricultural and Nonagricultural Tradables, and the RRA,a by Region, 1955­2007 (continued ) (percent) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 European transition economies NRA, agricultural -- -- -- -- -- -- -- 10.0 18.3 16.1 17.0 NRA, nonagricultural -- -- -- -- -- -- -- 9.8 5.5 4.6 2.7 RRA -- -- -- -- -- -- -- 0.1 12.2 11.0 13.9 Western Europe NRA, agricultural 43.8 57.0 67.5 45.7 56.3 74.4 82.0 63.4 43.6 36.8 18.5 NRA, nonagricultural 8.0 7.2 5.7 3.8 2.5 1.5 1.7 1.3 1.5 1.4 1.2 RRA 33.1 46.5 58.6 40.4 52.6 71.9 79.0 61.3 41.5 34.9 17.1 North America NRA, agricultural 12.5 10.5 10.9 7.5 7.6 13.8 20.2 16.1 11.4 17.3 11.2 NRA, nonagricultural 6.1 7.4 7.4 5.5 4.1 3.8 3.7 3.3 2.1 1.5 1.3 RRA 6.0 2.9 3.3 1.8 3.4 9.7 15.8 12.4 9.1 15.5 9.7 Australia and New Zealand NRA, agricultural 5.5 6.6 8.3 7.9 7.3 10.6 8.7 4.3 2.9 1.0 0.6 NRA, nonagricultural 20.0 21.5 24.0 19.7 14.3 13.5 10.3 6.4 3.4 2.4 2.4 RRA 12.1 12.2 12.6 9.9 6.1 2.6 1.5 2.0 0.5 1.4 1.8 Developing countriesb NRA, agricultural -- 24.0 27.3 31.9 25.5 21.0 15.6 3.9 4.0 7.4 -- NRA, nonagricultural -- 58.3 60.0 45.8 37.3 34.6 27.0 16.7 9.8 6.3 -- RRA -- 52.0 54.5 53.3 45.8 41.3 33.6 17.6 5.3 1.1 -- High-income countries NRA, agricultural 23.0 30.9 36.8 26.5 34.7 43.0 55.5 48.2 36.6 33.9 18.3 NRA, nonagricultural 7.5 8.5 7.7 5.4 3.6 3.4 3.2 2.5 1.7 1.3 0.7 RRA 14.3 20.6 27.1 19.9 30.1 38.3 50.6 44.6 34.3 32.1 19.2 World b NRA, agricultural -- 5.6 7.6 0.8 2.6 5.7 18.7 19.7 18.4 18.6 -- NRA, nonagricultural -- 19.0 20.5 16.1 13.7 10.0 9.8 7.6 6.0 4.0 -- RRA -- 11.3 10.7 13.2 9.8 3.6 8.1 11.3 11.8 14.0 -- Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Note: -- not available. a. The RRA is defined as 100*[(100 NRAag t ) (100 NRAnonagt ) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. b. Estimates for the RRA for China pre-1981 and India pre-1965 are based on the assumption that the agricultural NRAs in those years were the same as the average NRA estimates for those countries for 1981­84 and 1965­69, respectively, and that the value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. Developing-world and country aggregates are computed accordingly. 35 36 Distortions to Agricultural Incentives: A Global Perspective Figure 1.7. RRAs to Agriculture,a High-Income Countries, 1955­2007 200 150 100 percent 50 0 50 9 4 9 4 9 4 9 4 9 4 7 ­5 ­6 -6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 ­0 65 55 60 70 75 80 85 90 95 00 05 19 19 19 19 19 19 19 19 19 20 20 Non-EU Western Europe EU Japan/Rep. of Korea United states Canada Australia/New Zealand Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. The RRA is defined as 100*[(100 NRAagt ) (100 NRAnonagt ) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. removed. The spread of NRAs and RRAs for all our focus countries for 2000­04 is shown in figure 1.8. As economic growth continues, there are fewer countries below the zero lines in those ladders, reflecting an ongoing tendency for political economy forces to transform countries from taxing to subsidizing their farmers relative to producers of nonfarm goods as per capita incomes grow. For the whole panel dataset, that tendency over the 50-year period examined appears to be almost exactly the same for developing as for high-income countries, the only dif- ference being the average RRA starting point (see the country fixed effects regres- sion lines in figure 1.9).15 A final way of summarizing two of the reform indicators used above--the RRA and the TBI, which allow assessment of changes in the antiagricultural or proagri- cultural production and trade biases in policy regimes--is to map them in two- dimensional space. Figure 1.10 shows agriculture's TBI on the horizontal axis and the RRA on the vertical axis. An economy with no antiagricultural or proagricul- tural bias (RRA 0) and no antitrade or protrade bias within the farm sector (TBI 0) would be located at the intersection of the two axes in figure 1.10. Africa, Asia, and Latin America (the last shown in the figure as Latin America and Five Decades of Distortions to Agricultural Incentives 37 Figure 1.8. Cross-Country Dispersion of NRAs and RRAs, 2000­04 a. NRAa Norway Switzerland Iceland Korea, Rep. of Japan Slovenia Taiwan, China Romania Latvia EU-15 Lithuania Hungary Slovak Republic Colombia Czech Republic Canada Turkey Philippines Vietnam Estonia India United States Russian Federation Mozambique Indonesia Mexico Ecuador Poland Sri Lanka Kenya China Chile Brazil Bulgaria Bangladesh Dominican Republic New Zealand Malaysia Pakistan Madagascar Australia Uganda Mali South Africa Chad Burkina Faso Cameroon Thailand Benin Togo Ghana Nicaragua Kazakhstan Nigeria Egypt, Arab Rep. of Senegal Ethiopia Ukraine Sudan Tanzania Argentina Côte d'Ivoire Zambia Zimbabwe 50 0 50 100 150 200 percent 38 Distortions to Agricultural Incentives: A Global Perspective Figure 1.8. (continued) b. RRA Korea, Rep. of Norway Switzerland Iceland Japan Taiwan, China Slovenia Romania Lithuania Latvia EU-15 Colombia Hungary Canada Estonia Czech Republic Turkey Philippines United States India Slovak Republic Poland Indonesia Mexico Russian Federation Mozambique Ecuador Kenya Chile China Malaysia Vietnam Brazil New Zealand Dominican Republic Australia South Africa Bulgaria Nigeria Thailand Ghana Nicaragua Sri Lanka Kazakhstan Madagascar Pakistan Cameroon Ukraine Bangladesh Sudan Uganda Argentina Senegal Egypt, Arab Rep. of Tanzania Ethiopia Côte d'Ivoire Zambia Zimbabwe 100 50 0 50 100 150 200 percent Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Note: NRA for all agriculture products, including NPS. Five Decades of Distortions to Agricultural Incentives 39 Figure 1.9. Relationships between Real GDP Per Capita and RRA,a All Focus Countries, 1955­2007 400 300 200 RRA (%) 100 0 100 1 0 1 2 3 ln real GDP per capita high-income country RRA observations high-income country fitted values developing-country RRA observations developing-country fitted values Coefficient Standard error R2 Developing countries 0.26 0.02 0.17 High-income countries 0.28 0.03 0.14 Source: Author's derivation with country fixed effects, using data in Anderson and Valenzuela (2008) that are based on RRA estimates reported in the project's national country studies. the Caribbean) were all well to the southwest of that neutral point as of 1980­84, but by 2000­04 all had moved much closer to the vertical axis (meaning they had reduced their antitrade bias in agriculture), and all but Africa had become closer to the horizontal axis--although Asia is now above rather than below that axis, which means the countries in that region are assisting farmers relatively more than producers of other tradable products. While that can lead to just as much waste of resources as the earlier, antiagricultural, policy bias, it is only in Korea and in Taiwan, China that the 2000­04 RRA is well above zero (it is just 1 percent for China and 4 percent for Southeast Asia). Producer assistance and consumer taxation in value terms In conclusion, this survey returns to the dollar values mentioned at the outset. Table 1.9 shows the full time series, in aggregate and by per person engaged in 40 Distortions to Agricultural Incentives: A Global Perspective Figure 1.10. Relationship between RRA and the TBI for Agriculture, Focus Regions, 1980­84 and 2000­04 150 Japan 100 Japan Western Europe RRA (%) 50 Western Europe North America Europe and Central Asia Asia 0 Africa Africa Latin America Australia and Asia and the Caribbean New Zealand 50 0.6 0.5 0.4 0.3 0.2 0.1 0 trade bias index 1980­84 2000­04 Source: Author's derivation using data in Anderson and Valenzuela (2008) that are based on NRA and RRA estimates reported in the project's national country studies. agriculture. Two points are worth stressing from the latter estimates. One is that for high-income countries, when decoupled payments are included, assistance per farmer has continued to rise over time, even when expressed in constant 2000 U.S. dollars. By 2000­04, the annual transfer was double what it was in 1975­79, which, in turn, is more than double the transfer in 1965­69. At US$13,400 per farmer, that is a very large transfer to a group whose household incomes are not significantly below those of nonfarm workers when adjusted for differences between urban and rural costs of living.16 The second point to note from table 1.9b is that for half this period, farmers in developing countries were effectively taxed between US$0.30 and US$0.50 per day by price-distorting policies, at a time when many hundreds of millions of them were struggling to survive on less than US$1 a day per house- hold member. Thankfully, most of the export taxation that drove those transfers was dismantled, although some was resurrected in response to high international food prices in 2007 and 2008. The move from negative to positive transfers to developing country farmers since the mid-1990s is not necessarily a good thing, however, even if society believes farmers should be compensated for the cost of adjusting to the structural changes that accompany rapid economic growth-- there are almost always more efficient ways to raise welfare of poor people than via price and trade policies. Table 1.9. GSEs of Assistance to Farmers, Total and Per Farm Worker, by Region, 1965­2007 a. Total (constant 2000 US$ billions per year) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa 5 9 11 6 1 7 5 6 -- Asia 69 110 126 106 38 8 38 57 -- Latin America 2 13 13 14 10 4 5 4 -- Europe and Central Asia a 3 1 2 10 1 4 19 19 26 Western Europe 89 76 124 133 123 117 76 55 34 United States and Canada 19 16 19 31 36 30 20 30 32 Australia and New Zealand 1 1 1 1 1 1 0 0 0 Japan 16 18 37 33 53 59 53 42 22 All focus countries 46 21 30 64 165 201 208 202 -- Developing countries 79 133 152 136 48 6 58 75 -- High-income countries 125 112 182 199 214 207 150 127 80 High-income countries, including decoupledb 125 112 186 223 238 235 193 173 130 World (scaled)c 51 24 34 70 192 225 233 223 -- (Table continues on the following page.) 41 42 Table 1.9. GSEs of Assistance to Farmers, Total and Per Farm Worker, by Region, 1965­2007 (continued) b. Per person engaged in agriculture (constant 2000 US$ per year) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa 77 137 142 81 10 69 44 46 -- Asia 114 166 174 136 45 9 41 60 -- Latin America 79 410 408 407 305 132 177 143 -- Europe and Central Asia 123 54 101 443 45 141 541 569 888 Western Europe 4,259 4,433 8,163 10,097 10,889 12,397 9,617 8,427 5,141 United States and Canada 3,733 3,436 4,189 6,730 8,220 7,478 5,582 9,182 8,763 Australia and New Zealand 3,504 3,452 3,492 4,744 3,226 1,613 1,221 442 1,061 Japan 1,301 1,845 4,830 5,432 10,234 13,957 16,234 16,933 12,469 All focus countries 59 27 34 66 164 185 181 172 -- Developing countries 116 177 185 154 51 6 53 67 -- High-income countries 3,261 3,491 6,483 8,166 9,972 11,325 9,745 9,871 7,180 High-income countries, including decoupleda 3,261 3,491 6,625 9,151 11,091 12,857 12,539 13,447 11,681 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Note: -- not available. a. Data for Europe and Central Asia are only Turkey until 1991. Figures are from FAOSTAT, which may differ from national statistics. Estimates for the NRA and RRA for China pre-1981 and India pre-1965 are based on the assumption that the agricultural NRAs in those years were the same as the average NRA estimates for those countries for 1981­84 and 1965­69, respectively, and that the value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. Developing-country and world aggregates are computed accordingly. b. Decoupled payments to farmers are excluded from all rows except the final two. c. These figures assume that the NRA in nonfocus countries is the same as the average for the focus countries in each region (including decoupled payments in the case of other high-income countries), and that their share of the value of regional agricultural production at undistorted prices is the same as their average share of the region's agricultural GDP at distorted prices during 1990­2004. For the countries of North Africa (other than the Arab Republic of Egypt) and the Middle East, NRAs are assumed to be the same as Turkey's. Five Decades of Distortions to Agricultural Incentives 43 Since it is mostly border (trade) policies rather than domestic measures that are responsible for the nominal assistance provided to farmers, those border measures also confer a tax on consumers of farm products. The extent of those CTEs, shown in table 1.10 for all covered farm products, also differs from NRAs in that the consumption weights are not the same as production weights for tradable products. But in percentage terms, their order of magnitude is similar to the NRAs. Again, two points are noteworthy. One is that developing countries' price- distorting policies prior to the 1990s provided up to US$0.30 per capita per year to consumers, directly at the expense of farmers, who on average are far poorer than urban dwellers. The other is that the cost to consumers in high-income countries of farm support policies is miniscule relative to their incomes, amount- ing even at their peak in the most protected countries (other than Japan) to well under US$1 a day. It is thus not surprising that domestic opposition to these policies is weak. Additional Indicators of Agricultural Price Distortions The indicators used above to summarize trends in the extent of distortions within the agricultural sector of a country include the unweighted or weighted mean NRA of covered products, the standard deviation of covered product NRAs, the weighted mean NRA for exportable versus import-competing covered products, and the TBI of the agricultural sectors' covered plus noncovered trad- able products. The reason for reporting these various indicators of dispersion of NRAs--apart from their being informative in their own right--is that theory suggests the national economic welfare cost of government policy distortions to incentives in terms of resource misallocation tends to be greater, the greater the degree of substitution in production in response to changes in price. In the case of agriculture, which involves the use of farmland that is sector-specific but substitutable among farm activities, the greater the variation of NRAs across industries within the sector then the higher will be the welfare cost of those market interventions. While these various indicators of dispersion are useful, it would also be helpful to have a single indicator to capture the overall welfare or trade effect of each country's regime of agricultural price distortions in place at any time, and to trace its path over time and make cross-country comparisons. To that end, chapter 11 draws on a theoretical literature that has developed in recent years to provide such indicators for national price and trade policies that are well grounded in theory. They belong to the family of indexes first developed by Anderson and Neary (2005) under the catch-all name of trade restrictiveness indexes. 44 Table 1.10. CTEs of Policies Assisting Producers of Covered Farm Products, Percent and Per Capita, by Region, 1965­2007 a. Percent 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa 12 16 9 6 16 8 0 3 -- Asia 12 15 2 15 14 3 5 10 -- Latin America 7 18 13 12 10 13 6 8 -- Europe and Central Asiaa 17 6 7 28 1 2 9 17 12 Western Europe 74 49 59 70 65 49 37 32 17 United States and Canada 8 6 7 13 10 2 5 2 2 Australia and New Zealand 15 11 10 10 10 8 4 2 1 Japan 67 68 93 99 135 119 116 107 81 Developing countries 12 16 5 14 10 0 5 8 -- High-income countries 42 30 40 45 50 41 32 27 16 All focus countries 23 14 21 10 15 16 15 16 -- b. Per capita (constant 2000 US$ per year) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa 15 27 16 12 21 10 0 3 -- Asia 6 13 2 22 18 4 6 10 -- Latin America 4 24 22 31 16 34 11 12 -- Europe and Central Asiaa 20 9 9 41 1 13 15 30 20 Western Europe 240 202 306 276 247 207 138 103 50 United States and Canada 40 35 43 67 42 8 18 9 11 Australia and New Zealand 112 87 80 68 56 49 22 12 8 Japan 130 157 287 258 435 498 443 344 221 Developing countries 7 15 5 22 13 1 6 8 -- High-income countries 153 136 207 195 199 176 124 94 59 All focus countries 30 18 39 23 27 29 26 23 -- Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Note: -- not available. a. Europe and Central Asia data are for Turkey only until 1991. 45 46 Distortions to Agricultural Incentives: A Global Perspective To capture distortions imposed by each country's border and domestic policies on its economic welfare and its trade volume, Lloyd, Croser, and Anderson (2009a) define a welfare reduction index (WRI) and a trade reduction index (TRI) and estimate them for the project's focus countries since 1960, taking into account that for some covered products the NRA and CTE differ (because there are domestic measures in place in addition to or instead of trade measures). As their names suggest, these two new indexes respectively capture in a single indica- tor the (partial equilibrium) welfare- or trade-reducing effects of distortions to consumer and producer prices of covered farm products from all agricultural and food policy measures in place (while ignoring noncovered farm products and indirect effects of sectoral and trade policy measures directed at nonagricultural sectors). The WRI measure reflects the welfare cost of agricultural price-distorting policies better than the NRA or CTE because it recognizes that the cost of a government-imposed price distortion is related to the square of the price wedge. It thus captures the disproportionately higher welfare costs of peak levels of assis- tance or taxation, is larger than the mean, and is positive regardless of whether the government's agricultural policy favors or hurts farmers. The WRI and TRI meas- ures also have the advantage of providing a theoretically sound indicator of the welfare (or trade) effect in a single sectoral measure that is comparable across time and place. In this way, the WRI and TRI come close to what a computable general equilibrium (CGE) can provide in the way of estimates of the trade and welfare (and other) effects of the price distortions captured by the product NRA and CTE estimates, while having the advantage of providing an annual time series. The time series derived for this project, available as Anderson and Croser (2009), is detailed in chapters 11 and 12. The WRI estimates in figure 1.11 indicate an increasing tendency for covered products' policies to reduce welfare between the 1960s and the mid-1980s, and then to improve welfare in the 1990s (when the WRI declines). This pattern is generated by different policy regimes in the different country groups, though: in high-income countries, covered products were assisted throughout the years reported, although less so after the 1980s, whereas covered products in developing countries were disprotected until the most recent years. That is, the WRI has the desirable property of correctly identifying the welfare consequences that result from both positive and negative assistance regimes, because it captures the disper- sion of NRAs among covered products: the larger the variance in assistance levels, the greater the potential for resources to be used in activities that do not maximize economic welfare. One consequence of these tendencies is that the WRI values are much higher than the NRAs for high-income countries. Another is that the WRI for Africa spikes in the mid-1980s, in contrast to the NRA, which moves close to zero. The Five Decades of Distortions to Agricultural Incentives 47 Figure 1.11. Welfare Reduction Indexes for Covered Tradable Farm Products, by Region, 1960­2007 a. Africa, Asia, and Latin America 80 60 percent 40 20 0 4 9 4 9 4 9 4 9 4 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 60 65 70 75 80 85 90 95 00 19 19 19 19 19 19 19 19 20 Africa Asia Latin America b. Developing countries, high-income countries, and Europe's transition economies 100 80 60 percent 40 20 0 4 9 4 9 4 9 4 9 4 7 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 ­0 60 65 70 75 80 85 90 95 00 05 19 19 19 19 19 19 19 19 20 20 high-income countries Europe's transition economies developing countries Source: Authors' calculations based on NRAs and CTEs in Anderson and Valenzuela (2008). 48 Distortions to Agricultural Incentives: A Global Perspective reason is that while Africa was still taxing exportables, it moved (temporarily) from low to very high positive levels of protection for import-competing farm products (table 1.4). At the aggregate level, African farmers received almost no government assistance then (NRA close to zero), but the welfare cost of its mix- ture of agricultural policies as a whole was at its highest according to the WRI. A third consequence is that for developing countries, the average WRI in the years 1995 to 2004 is around 20 percent, even though the average NRA for covered products in those years is close to zero (see figure 1.3), again reflecting the high dispersion across product NRAs--particularly between exportables and import- competing goods--in each country.17 It is possible to use the value of production (at undistorted prices) to estimate the contribution to global welfare reduction by the various regions' policies affect- ing covered agricultural products. Figure 1.12a shows that, since 1993, high- income countries (plus Korea) have been responsible for around 60 percent of that global welfare reduction, developing Asia for about 25 percent, and Africa for nearly 10 percent. Asia's share has been about 40 percent for the whole period since 1981, but as developing Asia's contribution declined over the 1980s, the con- tribution of Japan and Korea grew. Figure 1.12b reveals that China's contribution has fallen substantially over the 1990s. Recall, though, that the WRI picks up the variance for these countries' product NRAs within the sector. Thus, even though China had an aggregate agricultural NRA of only 6 percent in the period 1994­2005, it had negative NRAs on cotton and rice and positive NRAs for other products, so it still had a moderately large WRI. For developing countries as a group, the trade restrictiveness of agricultural policy was roughly constant until the early 1990s. Thereafter it declined, especially for Asia and Latin America, according to the TRI estimates. For high-income countries, the TRI time path was similar, although the decline began a few years later. Aggregate results for developing countries are driven by the exportables subsector, which is being taxed, and the import-competing subsector, which is being protected (albeit by less than in high-income countries). For high-income countries, policies have supported both exporting and import-competing agricul- tural products. Even though these policies heavily favor the latter, assistance to exporters has offset the antitrade bias from the protection of import-competing producers to some degree in terms of impacts on those countries' aggregate volume of trade in farm products. Thus, up to the early 1990s, the TRI for high-income countries was below that for developing countries. To use again the example of Africa, the TRI peaked in 1985­89, when the NRA was closest to zero, correctly identifying the trade-reducing effect of positive protection to the import-competing farmers and disprotection to producers of exportables (figure 11.6 and table 11.4 of Lloyd, Croser, and Anderson 2009a). Figure 1.12. Regional Shares of Global and Country Shares of Asian Welfare Reductiona from Agricultural Policies, 1981­2004 a. Regional share of global welfare reduction 100 90 80 70 60 percent 50 40 30 20 10 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 19 9 19 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 20 9 20 0 20 1 02 20 3 04 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 9 0 0 0 19 20 Latin America Africa other Asia Japan and Korea other high-income countries Western Europe b. National share of welfare reduction in developing Asia (excluding Rep. of Korea) 100 90 80 70 60 percent 50 40 30 20 10 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 19 9 19 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 20 9 20 0 20 1 20 2 20 3 04 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 9 0 0 0 0 19 Vietnam Thailand Taiwan, China Sri lanka Philippines Pakistan Malaysia Indonesia India China Bangladesh Source: Author's derivation using data from Anderson and Croser (2009), based on estimates in Anderson and Valenzuela (2008). a. The value of welfare reduction is computed as the WRI multiplied by the value of production at undistorted prices valued at constant 2000 U.S. dollars. European transition economies are not included, nor are countries not included in the Anderson and Valenzuela sample (together, these make up less than 5 percent of global GDP). 49 50 Distortions to Agricultural Incentives: A Global Perspective Chapter 12 differs from the preceding chapters in that it focuses on commodi- ties rather than countries, showing to what extent global markets for some farm commodities are distorted relative to others. It also shows that the basic food staples of the poor in low-income countries are reasonably well covered in our database for African countries where they matter most, and that their noncoverage in other regions does not bias the aggregate NRA estimates for covered products. Chapter 12 also reports new partial equilibrium estimates of global commodity TRIs and WRIs, to parallel the country-based ones reported in chapter 11. A sum- mary of both indicators for 28 key farm commodities is provided in figure 1.13 (taken from Lloyd, Croser, and Anderson 2009b). It shows that the most distorted of all commodities in 2000­04, in terms of both their global welfare cost and their trade restrictiveness, are rice, sugar, milk, and beef--although cocoa trade is just as restricted as beef, because export restrictions are still prevalent in major supplying countries such as Côte d'Ivoire. Figure 1.13. Trade and Welfare Reduction Indexes for 28 Major Agricultural Products, 2000­04 140 120 100 80 60 percent 40 20 0 20 40 su ice b lk ou ou n so dn ry h s se um co me ba oa m y ee let m t t ra g mtea se at so e ds yb gg w ans su m at pa ow e co cof oil co fee ru uts w va l m r gr p tto f lm er ss r oo co ee p oa ea ga ca bbe rg ut rle nf aiz i n lt pe e he e a c il r n sa e l pi sh welfare reduction index trade reduction index Source: Lloyd, Croser, and Anderson (2009b). Five Decades of Distortions to Agricultural Incentives 51 Economy-Wide Effects of Past Reforms and Remaining Policies It is clear that there has been a great deal of change over the past quarter century in policy distortions to agricultural incentives throughout the world: the antiagri- cultural and antitrade biases of policies of many developing countries have been reduced; export subsidies of high-income countries have been cut; and some rein- strumentation toward less inefficient and less trade-distorting forms of support, particularly in Western Europe, has begun. However, protection from agricultural import competition has continued an upward trend in both rich and poor coun- tries, notwithstanding the Uruguay Round Agreement on Agriculture (URAA), which aimed to bind and reduce farm tariffs. What, then, have been the net economic effects of agricultural price and trade policy changes around the world since the early 1980s? And how do those effects on global markets, farm incomes, and economic welfare compare with the effects of price distortions still in place as of 2004? The final chapter by Valenzuela, van der Mensbrugghe, and Anderson (2009) uses a global economy-wide model to provide a combined retrospective and prospective analysis that seeks to assess how far the world has come--and how far it still has to go--in harmonizing world agricultural policy. It quantifies the impacts of both past reforms and current policies by comparing the effects of the project's NRA and CTE distortion esti- mates for the period 1980­84 with those of 2004. Several key findings from that economy-wide modeling study are worth emphasizing. First, the policy reforms from the early 1980s to the mid-2000s improved global economic welfare by US$233 billion per year, and removing the distortions remaining as of 2004 would add another US$168 billion per year. This change suggests that in terms of global welfare, the world had moved three-fifths of the way toward global free trade in goods over a quarter century. Second, developing economies benefited proportionately more than high- income economies (1.0 percent compared with 0.7 percent of national income) from those past policy reforms, and would gain nearly twice as much as high- income countries by completing the reform process (an average increase of 0.9 percent compared with 0.5 percent for high-income countries). Of the prospective welfare gains from global liberalization, 70 percent would come from agriculture and food policy reform. This is a striking result given that the shares of agriculture and food in global GDP and global merchandise trade are only 3 and 6 percent, respectively. The contribution of farm and food policy reform to the prospective welfare gain for just developing countries is even greater, at 72 percent. Third, the share of global farm production exported (excluding intra-EU trade) in 2004 was slightly smaller as a result of those reforms since 1980­84, because there were less farm export subsidies. Agriculture's 8 percent share in 52 Distortions to Agricultural Incentives: A Global Perspective 2004 contrasts with the 31 percent share for other primary products and the 25 percent for all other goods--a "thinness" that is an important contributor to the volatility of international prices for weather-dependent farm products. If the poli- cies distorting goods trade in 2004 were removed, the share of global production of farm products that is exported would rise from 8 to 13 percent, thereby reduc- ing the instability of prices and quantities of products traded. Fourth, developing countries' share of the world's primary agricultural exports rose from 43 to 55 percent, and its farm output share from 58 to 62 percent, because of those reforms, with increases in both in nearly all agricultural indus- tries except rice and sugar. Removing the remaining goods market distortions would boost export and output shares to 64 and 65 percent, respectively. Fifth, the average real price in international markets for agricultural and food products would have been 13 percent lower had policies not changed over the past quarter century. Evidently, the impact of the decline in RRA in high-income countries (including the cuts in farm export subsidies) on international food prices more than offset the opposite impact of the increase in RRA (including the cuts in agricultural export taxes) in developing countries over that period, and the net effect was higher food prices. By contrast, removing the distortions remaining as of 2004 is projected to raise the international price of agricultural and food products by less than 1 percent on average. This is contrary to earlier modeling results based on the Global Trade Analysis Project (GTAP) protections database. For example, Anderson, Martin, and van der Mensbrugghe (2006) estimated prices would rise by 3.1 percent, or for just primary agriculture, by 5.5 percent. The lesser impact in the new results presented here is due to the fact that export taxes in developing countries based on the above NRA estimates (most notably for Argentina) are included in the new database: removing those taxes offsets the international price-raising effect of eliminating import protection and farm sub- sidies elsewhere. Sixth, for developing countries as a group, net farm income (value added in agriculture) is estimated to be 4.9 percent higher than it would have been without the reforms of the past quarter century, which is more than ten times the propor- tional gain for nonagriculture. If policies remaining in 2004 were removed, net farm incomes in developing countries would rise a further 5.6 percent, compared with just 1.9 percent for nonagricultural value added. In addition, returns to unskilled workers in developing countries--the majority of whom work on farms--would rise more than returns to other productive factors from that liber- alization. Together, these findings suggest both inequality and poverty could be alleviated by such reform, given that three-quarters of the world's poor are in farm households in developing countries (Chen and Ravallion 2007). Five Decades of Distortions to Agricultural Incentives 53 Finally, removal of agricultural price-supporting policies in high-income countries would undoubtedly lead to painful reductions in income and wealth for farmers in those countries if the farmers were not compensated. It should be kept in mind, though, that the majority of farm household income in high-income countries comes from off-farm sources (OECD 2008b), and that the gainers in the rest of their societies could readily afford to compensate the losers from the benefits of freeing trade. Prospects for Further Reform It is not obvious how future agricultural policies might develop. A quick glance at the indicators could lead one to view developments from the early 1960s to the mid-1980s as an aberrant period of welfare-reducing policy divergence (negative and declining RRAs in low-income countries alongside positive and rising RRAs in most high-income countries) that has given way to welfare-improving and poverty-reducing reforms during which the two country groups' RRAs are con- verging. But on inspection of the NRAs for exporting and import-competing sub- sectors of agriculture (figure 1.3), it is clear that the convergence of NRAs to near zero is mainly with respect to the exporting subsector, while NRAs for import- competing farmers are positive and trending upward over time at the same rate in both developing and high-income countries--notwithstanding the URAA, which aimed to bind and reduce farm tariffs. True, applied tariffs have been lowered or suspended as a way of dealing with the international food price spike in 2008, but this, and the food export taxes or quantitative restrictions imposed that year by numerous food-exporting developing countries, may be in place only until inter- national prices return to trend (as happened after the price hike of 1973­74 and the price dip of 1986­87). Indications are very mixed as to why some countries have reformed their price- distorting agricultural and trade policies more than others in recent decades. Some reforming countries have acted unilaterally, apparently convinced that it is in their national interest to do so. China is the most dramatic and significant example of a developing country adjusting its agricultural and trade policy over the past three decades according to its national interest, while among high-income countries only Australia and New Zealand (and to a lesser extent the EU) are in that category. Others countries have reformed their policies partly to secure bigger and better loans from international financial institutions and then, having taken that first step, have continued the process, even if somewhat intermittently. India is one example, but there are numerous others in Africa and Latin America. Few countries have gone backwards in terms of increasing their antiagricultural bias, but Zimbabwe and perhaps Argentina qualify during the present decade--and 54 Distortions to Agricultural Incentives: A Global Perspective numerous others joined them in 2008, at least temporarily, in response to the sud- den spike in international food prices. Still other countries have reduced their agricultural subsidies and import barriers at least partly in response to the GATT's multilateral URAA, the EU being the most important example (encouraged by its desire for preferential trade agreements, including its recent eastward expansion). EU reforms suggest growth in agricultural protection can be slowed and even reversed if accompanied by reinstrumentation away from price supports to decoupled measures or more direct forms of farm income support. The starker examples of Australia and New Zealand show that one-off buyouts can bring faster and even complete reform.18 But in developing countries, where levels of agricultural protection are generally lower than in high-income countries, there are fewer signs of a slowdown in the upward trend in agricultural protection from import competition over the time period studied. Indeed, there are numerous signs that developing-country governments want to retain the option to raise agricultural NRAs in the future, particularly via import restrictions. One indicator is the high tariff bindings developing coun- tries committed themselves to following the Uruguay Round of the WTO: as of 2001, actual applied tariffs on agricultural products averaged less than half the corresponding bound tariffs for developing countries of 48 percent, and less than one-sixth in the case of least developed countries (Anderson and Martin 2006). Another indicator of reluctance to undertake agricultural trade reform is the unwillingness of many developing countries to agree to major cuts in bound agri- cultural tariffs in the WTO's ongoing Doha Round of multilateral trade negotia- tions. Indeed, many developing countries believe that high-income countries should commit to reducing their remaining farm tariffs and subsidies before developing countries should offer further reform commitments of their own. Yet modeling results reported in Valenzuela, van der Mensbrugghe, and Anderson (2009) suggest that if high-income countries alone were to liberalize their agricul- tural markets, such subglobal reform would provide less than two-thirds of the potential gains to developing countries that could result from global agricultural policy reform. In addition, current negotiations have brought to the fore a new proposal for agricultural protectionism in developing countries, based on the notion that agri- cultural protection is helpful and needed for food security, livelihood security, and rural development. This view has succeeded in bringing "special products" and a "special safeguard mechanism" into the multilateral trading system's agricultural negotiations, despite the fact that such policies, which would raise domestic food prices in developing countries, may worsen poverty and jeopardize the food secu- rity of the poor (Ivanic and Martin 2008). Five Decades of Distortions to Agricultural Incentives 55 For developing countries to wait for high-income country reform before liber- alizing their own farm trade is unwise as a poverty alleviating strategy, not least because the past history revealed in the NRAs summarized above suggests such reform will be, at best, slow in coming. In the United States, for example, the most recent two five-year farm bills were a step backwards from the previous regime, which at least sought to reinstrument protection toward less trade-distorting measures (Gardner 2009). Nor have the world's large number of new regional integration agreements of recent years been very successful in reducing farm pro- tection. Furthermore, for developing countries to postpone their own reform would be to forego a major opportunity to boost their own and (given the size and growth in South-South trade of late) their neighbors' economies. It would be dou- bly wasteful if, by being willing to commit to reform in that way, they are able to convince high-income countries to reciprocate by signing on to a more ambitious Doha Round, the potential global benefits of which are considerable.19 Developing countries that continue to open their domestic markets and prac- tice good macroeconomic governance will keep growing. Typically, growth will be more rapid in manufacturing and services activities than in agriculture, especially in the more densely populated countries where agricultural comparative advan- tage is likely to decline. Whether such economies become more dependent on imports of farm products depends, however, on what happens to their RRA. The first wave of Asian industrializers (Japan, and then Korea and Taiwan, China) chose to slow the growth of food import dependence by raising their NRAs for agriculture even as they were reducing their NRAs for nonfarm tradables, such that their RRAs became increasingly above the neutral zero level. A key question is: will later industrializers follow suit, given the past close association of RRAs with rising per capita income and falling agricultural comparative advantage? Figure 1.9 suggests developing countries' RRA trends of the past three decades have been on the same trajectory as the high-income countries prior to the 1990s, so unless new forces affect their policies, the governments of later industrializing economies may well follow suit. One new force is constraints on farm subsidies and protection policies of WTO member countries following the Uruguay Round. Earlier industrializers were not bound under GATT to keep down their agricultural protection. Had there been strict discipline on farm trade measures at the time Japan and Korea joined GATT in 1955 and 1967, respectively, their NRAs may have been halted at less than 20 percent (figure 1.14). At the time of China's accession to WTO in December 2001, its NRA was less than 5 percent according to this present study, or 7.3 per- cent for just import-competing agriculture. Its average bound import tariff com- mitment was about twice that (16 percent in 2005), but what matters most is China's out-of-quota bindings on the items whose imports are restricted by tariff 56 Distortions to Agricultural Incentives: A Global Perspective Figure 1.14. NRAs for Japan, Republic of Korea, and China, and Date of Accession to GATT or WTO, 1955­2005 200 150 Japan (1955 16.6%) 100 Korea (1967 7.4%) NRA (%) China 50 (2001 4.5%) 0 50 100 9 4 9 4 9 4 9 4 9 5 ­5 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 55 60 65 70 75 80 85 90 95 00 19 19 19 19 19 19 19 19 19 20 Japan Korea China Source: Based on estimates in Anderson and Valenzuela (2008). rate quotas. The latter tariff bindings as of 2005 were 65 percent for grains, 50 per- cent for sugar and 40 percent for cotton (Anderson, Martin, and Valenzuela forth- coming). Clearly, the legal commitments even China made on acceding to the WTO are a long way from current levels of support for its farmers, and so are unlikely to constrain the government very much in the next decade or so. The legal constraints on developing countries that joined the WTO earlier are even less constraining. For India, Pakistan, and Bangladesh, for example, estimated NRAs for agricultural importables in 2000­04 are 34, 4, and 6 percent, respectively, whereas the average bound tariffs on their agricultural imports are 114, 96, and 189 percent, respectively (WTO, ITC, and UNCTAD 2007). And like other devel- oping countries, these countries have high bindings on product-specific domestic support (10 percent), and another 10 percent for NPS assistance, for a total of 20 more percentage points of NRA (17 percent in China's case) that legally could come from domestic support measures--compared to 10 percent in India and less than 3 percent in the rest of South Asia. It is hoped developing countries will choose not to make use of the legal wiggle room they have allowed themselves in their WTO agreements to follow Japan, Korea, and Taiwan, China into high agricultural protection. A much more effi- cient and equitable strategy would be to instead treat agriculture in the same way they have been treating nonfarm tradable sectors. That would involve opening the sector to international competition, relying on more efficient domestic policy Five Decades of Distortions to Agricultural Incentives 57 measures for raising government revenue (for example, income and consumption or value-added taxes), and assisting farm families--including younger members seeking off-farm employment--via public investment in rural education and health, rural infrastructure, and agricultural research (Otsuka and Yamano 2006; Otsuka, Estudillo, and Sawada 2009). Historically, developing countries' expendi- ture on public agricultural research has amounted to the equivalent of less than 1 percent of the gross value of farm production (table 1.11), so it would not be dif- ficult to double that level of investment with a diversion of just a small amount of the price support currently provided to farmers in those developing countries that provide farm import protection or input subsidies. As for high-income countries, the above distortion estimates show that they have all lowered price supports for their farmers since the 1980s. In some coun- tries, those supports have been partly replaced by assistance that is at least some- what decoupled from production. If that trend continues at the pace of the past quarter century, and if there is no growth of agricultural protection in developing countries, most of the disarray in world food markets may be removed before the middle of this century. However, if the WTO's Doha Development Agenda col- lapses, and governments thereby find it more difficult to ward off agricultural protection lobbies, it is more likely that developing countries will follow the same agricultural protection path this century as that which was taken by high-income countries during the last century. One way to encourage developing countries to follow a more liberal policy path could be to extend the Integrated Framework's Diagnostic Trade Integration Study process to a broader range of low-income countries. That process, which provides action plans for policy and institutional reform and lists investment and technical assistance needs, could be expanded to include the "aid for trade reform" proposal that has been discussed in the context of the Doha Round (Hoekman 2005)--regardless of the fate of that round. Table 1.11. Intensity of Public Agricultural R&D Investment, High-Income and Developing Countries, 1971­2004 (expenditure as percentage of gross value of agricultural production at undistorted prices, unweighted average across available countries) 1970s 1980s 1990s 2000­04 All high-income countries 3.6 2.8 1.9 1.5 All developing countries 1.0 0.8 1.0 1.0 Asia 0.3 0.6 0.7 1.3 Latin America 0.2 0.4 0.6 0.6 Sub-Saharan Africa (not including South Africa) 1.2 1.0 0.9 0.7 Source: Anderson and Valenzuela (2008), based on R&D data from the Consultative Group on International Agricultural Research (CGIAR) Agricultural Science and Technology Indicators Website at http://www.asti.cgiar.org. 58 Distortions to Agricultural Incentives: A Global Perspective Areas for Further Research The fact that indications are mixed as to why some countries appear to have reformed their agricultural and trade policies more than others, and that it is therefore unclear as to how policies might develop in the future, should not be surprising. After all, the long history of globalization is full of episodes of sen- sible policy reforms that for all sorts of reasons get reversed (Findlay and O'Rourke 2007; North, Wallis, and Weingast 2009). If one is to better understand what might happen in the context of continuing economic growth and terms-of- trade volatility, more in-depth analysis of the political economy of past policy behavior is warranted. That is now possible thanks to the new panel set of distor- tion estimates reported here, and some early findings from such analyses will appear in Anderson (forthcoming). That collection includes a broad range of the- oretical and econometric analyses aimed at better understanding the political economy forces that generated the evolving pattern of intersectoral and intra- sectoral distortions to farmer and food consumer incentives over recent decades. To contribute further to policy debate, a second area requires further research-- economy-wide modeling of the impacts on agricultural markets, national eco- nomic welfare and income distribution of alternative policies. Such research is now easier to do well following the recent development of microsimulation add- ons to such CGE models. How the reform of current policies--both in the country under consideration and in the rest of the world--affects the extent of poverty and inequality is explored in a series of new country case studies in Anderson, Cockburn, and Martin (forthcoming), using global and national economy-wide models that are enhanced with detailed earning and spending information of numerous types of urban and rural households. The impact of rest-of-world policy on each coun- try's terms of trade is informed by the project's agricultural distortions database and generated using the same model as in this book's final chapter (Valenzuela, van der Mensbrugghe, and Anderson 2009). It is hoped that other analysts will make use of the project's agricultural distortions database to explore other policy scenarios, including a continuation of past agricultural protection growth for import-compet- ing farmers as an alternative counterfactual to further freeing of trade. A third area for further research is growth diagnostics. For example, how much of the long-term divergence in per capita incomes between current high- income countries on the one hand and developing and transition economies on the other can be explained by the faster long-term growth of agricultural NRAs and RRAs up to the mid-1980s in the former? How much can be explained by the domestic market-insulating fluctuations in NRAs around those long-term assis- tance trends, which have contributed to the volatility of the terms of trade for agricultural-exporting countries? The recent work of Williamson (2008) offers one suggestion as to how such research might proceed. Finally, how much of Five Decades of Distortions to Agricultural Incentives 59 growth since the early 1980s in individual developing countries can be attributed to the reduction in domestic agricultural disincentives? Ravallion and Chen (2007) show that the decline in the antiagricultural bias in farm price policies has contributed significantly to China's poverty reduction. Rural growth has been shown to be a key contributor to the reduction in poverty in India, too (Ravallion and Datt 1996). These types of studies can now be revisited using the more com- prehensive set of measures of the extent of changes in distortions to agricultural incentives summarized above. Annex Table 1.A.1. Export Orientation, Import Dependence, and Self-Sufficiency in Primary Agricultural Production, Focus Countries,a 1961­2004 (percent at undistorted prices) a. Exports as share of production 1961­64 1970­74 1980­84 1990­94 2000­04 Africa 19 17 12 7 8 Asia 5 4 4 6 5 Latin America 24 27 16 16 27 Western Europe 13 16 27 37 43 United States and Canada 14 14 20 20 21 Australia and New Zealand 41 35 44 43 48 Japan 1 2 1 0 1 All focus countries 11 11 13 16 16 Developing countries 8 8 7 8 8 High-income countries 14 15 22 26 29 b. Imports as share of apparent consumption 1961­64 1970­74 1980­84 1990­94 2000­04 Africa 2 2 5 4 4 Asia 4 4 8 16 14 Latin America 2 4 7 10 17 Western Europe 32 28 34 41 46 United States and Canada 4 4 5 9 12 Australia and New Zealand 3 2 3 5 6 Japan 23 24 24 26 27 All focus countries 11 10 12 19 18 Developing countries 3 4 8 14 13 High-income countries 18 16 20 25 27 (Table continues on the following page.) 60 Distortions to Agricultural Incentives: A Global Perspective Table 1.A.1 Export Orientation, Import Dependence, and Self- Sufficiency in Primary Agricultural Production, Focus Countries,a 1961­2004 (continued) (percent at undistorted prices) c. Self-sufficiency ratio 1961­64 1970­74 1980­84 1990­94 2000­04 Africa 120 117 107 104 105 Asia 102 100 96 89 91 Latin America 129 132 110 107 114 Western Europe 78 85 90 94 94 United States and Canada 111 112 119 114 111 Australia and New Zealand 165 151 174 170 183 Japan 78 78 77 74 74 All focus countries 100 101 101 96 98 Developing countries 105 104 99 93 95 High-income countries 96 98 103 101 102 Source: Compiled using the project's estimates of total agricultural production valued at undistorted prices and the FAO's total agricultural trade value data, in Anderson and Valenzuela (2008). a. Includes intra-EU trade. Benin, Burkina Faso, Chad, Iceland, Mali, Togo, and the countries of Europe and Central Asia are not included. Notes 1. Some of the more transformational policy developments happened quite promptly, such as the end of colonization around 1960; the creation of the Common Agricultural Policy (CAP) in Europe in 1962; the floating of exchange rates and associated liberalization, deregualtion, privatization, and democratization in the mid-1980s; the opening of China's economy from 1979; and the demise of the Soviet Union in 1991. 2. According to the FAOSTAT (http://www.fao.org), fewer than 15 million relatively wealthy farm- ers in developed countries, with an average of almost 80 hectares per worker, currently are being helped, at the expense of not only consumers and taxpayers in those rich countries but also the major- ity of the 1.3 billion relatively impoverished farmers and their large families in developing countries who, on average, must earn a living from just 2.5 hectares per worker. 3. In the two decades to 2000-04, the value of global exports as a share of GDP rose from 19 to 26 percent, even though most of GDP is nontradable governmental and other services (World Bank 2007, as summarized in Sandri, Valenzuela, and Anderson 2007). 4. A nine-year update for the Latin American countries in the Krueger, Schiff, and Valdés sample by the same country authors, and a comparable study of seven central and eastern European countries, contain estimates at least of direct agricultural distortions (see Valdés 1996, 2000). The Krueger, Schiff, and Valdés (1991) chapters on Ghana and Sri Lanka have protection estimates back to 1955, as does the study by Anderson and Hayami (1986) for Korea and Taiwan, China (and Japan, and much earlier in the case of rice). 5. See appendix B. The only countries not well represented in the sample are the many small ones and those in the Middle East. In total, however, the omitted countries account for less than 5 percent of the global economy. 6. By way of comparison, the seminal multicountry study of agricultural pricing policy by Krueger, Schiff, and Valdés (1991) covered an average of 23 years to the mid-1980s for 18 focus countries, which collectively accounted for 5­6 percent of global agricultural output. The producer and consumer sup- port estimates of the OECD (2008a) cover 22 years for 30 countries that account for just over one- quarter of the world's agricultural output valued at undistorted prices. Anderson (2009) compares the Five Decades of Distortions to Agricultural Incentives 61 direct and total rates of protection estimated by Krueger, Schiff, and Valdés for their sample of coun- tries with the comparable NRAs and RRAs from the present study for the same sample of countries and same years, and finds them to be almost identical. However, the indicators of antiagricultural and antitrade bias are about two-thirds greater when the fuller sample of 41 developing countries in the present study is used. This suggests that, with a fuller sample, Krueger, Schiff, and Valdés could have stressed the key policy implications of their study even more forcefully. 7. Only a brief summary of the methodology is provided here. For details see Anderson et al. (2008a, 2008b) or appendix A. 8. The extent to which a payment is production-neutral (the degree of decoupling) differs depending on the way it is administered and the expectations of the recipient (Thompson, Dewbre, and Martini 2007). No attempt here is made to evaluate the extent of the distortion that might still be present as poli- cies are decoupled from output price and quantity produced. Rather, values of OECD estimates under certain categories of support are included as decoupled. For the years 1979­85, there was just one cate- gory, called "direct payments;" from 1986, those payments are specified to comprise the OECD's items C (payments based on area planted/animal numbers), D (payments based on historical entitlements), F (payments based on input constraints), and G (payments based on overall farming income); and for 2005­07, those items are replaced by the similar but newly defined items C to E. This categorization for economic purposes (and that for NPS assistance) should not be confused with the legal allocation of domestic support measures into the WTO colored "boxes" in the context of international commitments. 9. Farmers are affected not just by prices of their own products but also by the incentives nonagri- cultural producers face. That is, it is relative prices and hence relative rates of government assistance that affect producer incentives. More than seventy years ago, Lerner (1936) presented his Symmetry Theorem, which proved that in a two-sector economy, an import tax has the same effect as an export tax. This carries over to a model that also includes a third sector producing only nontradables. 10. Data availability also affects the year from which NRAs can be computed. For Europe's transi- tion economies, that starting date is 1992, for Vietnam it is 1986, and for China it is 1981. 11. While the next section summarizes the findings in subsequent chapters, readers should be aware that the emerging economy chapters in Part III of this book are themselves summaries of the findings in a large number of developing-country case studies that are detailed in one of four com- panion volumes covering Africa (Anderson and Masters 2009), East and South Asia (Anderson and Martin 2009a), Latin America and the Caribbean (Anderson and Valdés 2008) and European and Central Asian transition economies (Anderson and Swinnen 2008). 12. Korea and Taiwan, China are are categorized here as developing rather than high-income because at the beginning of the 50-year period under study they were among the poorest economies in the world. 13. The GSE estimates for 1955­59 are smaller for high-income countries but less negative for developing countries, so the aggregate effect is an empirical issue. Unfortunately, our sample of devel- oping countries for that period is too small to provide a reliable estimate of the net effect. 14. As a reality check, compare this project's regional NRAs for nonagricultural tradables for the 1960s with the spot-year national NRAs from manufacturing import protection (in brackets) for the eight countries reported in Little, Scitovsky, and Scott (1970) and Balassa and associates (1971): 120 percent in South Asia (96 percent in Pakistan); 40 percent in East Asia (8 percent in Malaysia, 29­46 percent in the Philippines, and 30 percent in Taiwan, China); and 30 percent in Latin America (141 percent in Argentina, 89-99 percent in Brazil, 89 percent in Chile, and 20­22 percent in Mexico). 15. The R2 values improve (to 0.16 and 0.26, respectively) and the slope of each line steepens (coef- ficients become 0.36 and 0.32, respectively) if the years since 1990 for high-income countries and before 1985 for developing countries are ignored. 16. The drop to US$6,000 per farmer in 2005­07 will likely prove to be temporary if international food prices return to trend levels soon. 17. National WRIs are aggregated across countries using as weights an average of the value of con- sumption and production at undistorted prices; TRIs use the absolute difference between the values of production and consumption at undistorted prices as weights. This is unlike NRAs and RRAs (or CTEs), which use as weights just the value at undistorted prices of production (or consumption). Like 62 Distortions to Agricultural Incentives: A Global Perspective NRAs, RRAs, and CTEs, national and regional WRIs and TRIs for the five-year periods are unweighted averages of the annual indexes. 18. For a detailed analysis of the buyout option versus the slower and less complete cashout option (moving to direct payments), as well as the uncompensated gradual squeeze-out or sudden cutout options, see Orden and Diaz-Bonilla (2006). 19. On the size of those potential net benefits compared with those from other opportunities that could address the world's most important challenges as conceived by the Copenhagen Consensus project (whose expert panel ranked trade reform as having the second highest payoff among those dozens of opportunities), see http://www.copenhagenconsensus.org, including the prepublication ver- sion of the trade paper by Anderson and Winters (2009). References Anderson, J. E., and J. P. Neary. 2005. Measuring the Restrictiveness of International Trade Policy. Cambridge, MA: MIT Press. Anderson, K. 1995. "Lobbying Incentives and the Pattern of Protection in Rich and Poor Countries." Economic Development and Cultural Change 43 (2): 401­23. ______. 2009. "Krueger/Schiff/Valdés Revisited: Agricultural Price and Trade Policy Reform in Devel- oping Countries Since the 1980s." School of Economics Working Paper 2009-21, University of Adelaide. Forthcoming in Applied Economic Perspectives and Policy. ______, ed. Forthcoming. The Political Economy of Agricultural Price Distortions. Cambridge and New York: Cambridge University Press. Working paper versions of the chapters are available at http://www.worldbank.org/agdistortions. Anderson, K., J. Cockburn, and W. Martin, eds. Forthcoming. Agricultural Price Distortions, Inequality and Poverty. Washington, DC: World Bank. Working paper versions of the chapters are available at http://www.worldbank.org/agdistortions. Anderson, K., and J. Croser. 2009. National and Global Agricultural Trade and Welfare Reduction Indexes, 1955 to 2007. Supplementary database at http://www.worldbank.org/agdistortions. Anderson, K., and Y. Hayami. 1986. The Political Economy of Agricultural Protection: East Asia in Inter- national Perspective. London: Allen and Unwin. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC. ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson, K., P. J. Lloyd, and D. MacLaren. 2007. "Distortions to Agricultural Incentives in Australia since World War II." The Economic Record 83 (263): 461­82. Anderson, K., R. Lattimore, P. J. Lloyd, and D. MacLaren. 2009. "Australia and New Zealand." Chapter 5 in this volume. Anderson, K., and W. Martin, eds. 2006. Agricultural Trade Reform and the Doha Development Agenda, London: Palgrave Macmillan and Washington, DC: World Bank. ______. 2009a. Distortions to Agricultural Incentives in Asia, Washington, DC: World Bank. ______. 2009b. "China and Southeast Asia." Chapter 9 in this volume. Anderson, K., W. Martin, and E. Valenzuela. Forthcoming. "Long Run Implications of WTO Accession for Agriculture in China." In China's Agricultural Trade: Issues and Prospects, ed. C. Carter and I. Sheldon. St. Paul, MN: International Agricultural Trade Research Consortium. Anderson, K., W. Martin, and D. van der Mensbrugghe. 2006. "Market and Welfare Implications of Doha Reform Scenarios." In Agricultural Trade Reform and the Doha Development Agenda, ed. K. Anderson and W. Martin. London: Palgrave Macmillan and Washington, DC: World Bank. Anderson, K., and W. Masters, eds. 2009. Distortions to Agricultural Incentives in Africa, Washington, DC: World Bank. Five Decades of Distortions to Agricultural Incentives 63 Anderson, K., and J. Swinnen, eds. 2008. Distortions to Agricultural Incentives in Europe's Transition Economies. Washington, DC: World Bank. Anderson, K., and R. Tyers. 1992. "Japanese Rice Policy in the Interwar Period: Some Consequences of Imperial Self Sufficiency." Japan and the World Economy 4 (2): 103­27. Anderson, K., and A. Valdés, eds. 2008. Distortions to Agricultural Incentives in Latin America. Washing- ton, DC: World Bank. Anderson, K., and E. Valenzuela. 2008. Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007. Core database at http://www.worldbank.org/agdistortions. Anderson, K., and L. A. Winters. 2009. "The Challenge of Reducing International Trade and Migration Bar- riers." Policy Research Working Paper 4598, World Bank, Washington DC (also in Global Crises, Global Solutions (2nd edition), ed. B. Lomborg. Cambridge and New York: Cambridge University Press, 2009). Balassa, B., and associates. 1971. The Structure of Protection in Developing Countries. Baltimore: Johns Hopkins University Press. Bates, R. H. 1981. Markets and States in Tropical Africa: The Political Basis of Agricultural Policies. Berkeley: University of California Press. Bhagwati, J. N. 1978. Foreign Trade Regimes and Economic Development: Anatomy and Consequences of Exchange Control Regimes. Cambridge, MA: Ballinger. Chen, S., and M. Ravallion. 2007. "Absolute Poverty Measures for the Developing World, 1981­2004." Policy Research Working Paper 4211, World Bank, Washington, DC. ______. 2008. "The Developing World Is Poorer Than We Thought, But No Less Successful in the Fight Against Poverty." Policy Research Working Paper 4703, World Bank, Washington, DC. Findlay, R., and K. O'Rourke. 2007. Power and Plenty: Trade, War, and the World Economy in the Second Millennium. Princeton, NJ: Princeton University Press. Gardner, B. 2009. "United States and Canada." Chapter 4 in this volume. Haberler, G. 1958. Trends in International Trade: A Report by a Panel of Experts, Geneva: GATT. Herrmann, R., P. Schenck, R. Thiele, and M. Wiebelt. 1992. Discrimination Against Agriculture in Developing Countries? Tubingen: J. C. B. Mohr. Hoekman, B. 2005. "Making the WTO More Supportive of Development." Finance and Development 42: 14­18. Ivanic, M., and W. Martin. 2008. "Implications of Higher Global Food Prices for Poverty in Low- Income Countries." Agricultural Economics 39: 405­16. Johnson, D. G. 1991. World Agriculture in Disarray. London: St Martin's Press. Johnson, H. 1989. The Story of Wine. London: Mitchell Beasley. Kindleberger, C. P. 1975. "The Rise of Free Trade in Western Europe, 1820­1875." Journal of Economic History 35 (1): 20­55. Krueger, A. O. 1978. Foreign Trade Regimes and Economic Development: Liberalization Attempts and Consequences. Cambridge, MA: Ballinger. Krueger, A. O., M. Schiff, and A. Valdés. 1988. "Agricultural Incentives in Developing Countries: Measuring the Effect of Sectoral and Economy-wide Policies." World Bank Economic Review 2 (3): 255­72. ______. 1991. The Political Economy of Agricultural Pricing Policy, Volume 1: Latin America, Volume 2: Asia, and Volume 3: Africa and the Mediterranean. Baltimore: Johns Hopkins University Press for the World Bank. Lerner, A. 1936. "The Symmetry Between Import and Export Taxes." Economica 3 (11): 306­13. Lindert, P. 1991. "Historical Patterns of Agricultural Protection." In Agriculture and the State, ed. P. Timmer. Ithaca, NY: Cornell University Press. Little, I. M. D., T. Scitovsky, and M. Scott. 1970. Industry and Trade in Some Developing Countries: A Comparative Study. London: Oxford University Press. Lloyd, P. J., J. Croser, and K. Anderson. 2009a. "Welfare- and Trade-Based Indicators of National Dis- tortions." Chapter 11 in this volume. ______. 2009b. "Global Distortions to Key Commodity Markets." Chapter 12 in this volume. Morgan, D. 1979. Merchants of Grain. New York: Viking Press. North, D. C., J. J. Wallis, and B. R. Weingast. 2009. Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Cambridge and New York: Cambridge University Press. 64 Distortions to Agricultural Incentives: A Global Perspective Nye, J. V. C. 2007. War, Wine, and Taxes: The Political Economy of Anglo-French Trade, 1689­1900. Princeton, NJ: Princeton University Press. Orden, D., F. Cheng, H. Nguyen, U. Grote, M. Thomas, K. Mullen, and D. Sun. 2007. "Agricultural Pro- ducer Support Estimates for Developing Countries: Measurement Issues and Evidence from India, Indonesia, China and Vietnam." IFPRI Research Report 152, International Food Policy Research Institute, Washington DC. Orden, D., and E. Diaz-Bonilla. 2006. "Holograms and Ghosts: New and Old Ideas for Reforming Agri- cultural Policies." In Agricultural Trade Reform and the Doha Development Agenda, ed. K. Anderson and W. Martin. London: Palgrave Macmillan and Washington, DC: World Bank. OECD (Organisation for Economic Co-operation and Development). 2006. Agricultural Policies, Mar- kets and Trade in the Central and Eastern European Countries and the New Independent States: Mon- itoring and Outlook. Paris: OECD. ______. 2008a. Producer and Consumer Support Estimates, OECD Database, 1986­2007. http://www.oecd.org (online database for 1986­2007 estimates; OECD files for estimates using an earlier methodology for 1979­85). ______. 2008b. The Role of Farm Households and the Agro-Food Sector in the Economy of Rural Areas: Evidence and Policy Implications, Paris: OECD. Otsuka, K., J. P. Estudillo, and Y. Sawada, eds. 2009. Rural Poverty and Income Dynamics in Asia and Africa. London and New York: Routledge. Otsuka, K., and T. Yamano. 2006. "Introduction to the Special Issue on the Role of Nonfarm Income in Poverty Reduction: Evidence from Asia and East Africa." Agricultural Economics 35 (supplement): 373­97. Ravallion, M., and S. Chen. 2007. "China's (Uneven) Progress Against Poverty." Journal of Development Economics 82: 1­42. Ravallion, M., S. Chen, and P. Sangraula. 2007. "New Evidence on the Urbanization of Poverty." Policy Research Working Paper 4199, World Bank, Washington, DC. Ravallion, M., and G. Datt. 1996. "How Important to India's Poor is the Sectoral Composition of Eco- nomic Growth?" World Bank Economic Review 10 (1): 1­26. Ricardo, D. 1817. On the Principles of Political Economy and Taxation. London: John Murray. Sandri, D., E. Valenzuela, and K. Anderson. 2007. "Economic and Trade Indicators, 1960 to 2004." Agricultural Distortions Working Paper 02, World Bank, Washington, DC. http://www .worldbank.org/agdistortions. Stigler, G. J. 1975. The Citizen and the State: Essays on Regulation. Chicago: University of Chicago Press. Thiele, R. 2004. "The Bias Against Agriculture in Sub-Saharan Africa: Has It Survived 20 Years of Struc- tural Adjustment Programs?" Quarterly Journal of International Agriculture 42 (1): 5­20. Thompson, Sir William (Baron Kelvin of Largs). 1889. Popular Lectures and Addresses, Volume I. London: Macmillan. Thompson, W., J. Dewbre, and R. Martini. 2007. "Is OECD Farm Support Less Decoupled Than It Use to Be?" Paper presented at the Annual Meetings of the American Agricultural Economics Associa- tion, Portland, OR, 29­30 July. Tyers, R., and K. Anderson. 1992. Disarray in World Food Markets: A Quantitative Assessment. Cam- bridge and New York: Cambridge University Press. Valdés, A. 1996. "Surveillance of Agricultural Price and Trade Policy in Latin America During Major Policy Reforms." World Bank Discussion Paper 349, World Bank, Washington, DC. Valdés, A., ed. 2000. "Agricultural Support Policies in Transition Economies." World Bank Technical Paper 470, World Bank, Washington, DC. Valenzuela, E., D. van der Mensbrugghe, and K. Anderson. 2009. "General Equilibrium Effects of Price Distortions on Global Markets, Farm Incomes, and Welfare." Chapter 13 in this volume. Williamson, J. 2008. "Globalization and the Great Divergence: Terms of Trade Booms and Volatility in the Poor Periphery, 1782 to 1913." NBER Working Paper 13841, Cambridge, MA. World Bank. 2007. World Development Indicators. Washington, DC: World Bank. WTO, ITC, and UNCTAD (World Trade Organization, International Trade Commission, and United Nations Conference on Trade and Development). 2007. Tariff Profiles 2006. Geneva: WTO. Part II EVOLUTION OF DISTORTIONS IN ADVANCED ECONOMIES 2 Japan, republic of Korea, and Taiwan, China Masayoshi Honma and Yujiro Hayami* The story of agricultural policy in Northeast Asia over the past 50 years illustrates the dramatic changes that can occur in distortions to agricultural incentives faced by producers and consumers at different stages of economic development. In this study of Japan, the Republic of Korea, and the island of Taiwan, China, the degree of distortions for key agricultural products and for the agricultural sector as a whole over a period when these economies transitioned from low- or middle- income to high-income status (the period from 1955 to 2004 plus, in the case of Japan, some pre-World War II experience) is estimated, a period that encompasses the so-called East Asian economic miracle of dramatic industrial development. Theodore Schultz (1978) established that as economies advance from low- to high-income status, agricultural policies tend to shift from taxation of to subsi- dization of agriculture. In this regard, Japan, Korea, and Taiwan, China are clear examples. This chapter compares the policy evolution in these economies and provides information on the effect of policies and underlying economic condi- tions on changes in agricultural distortions. Our findings shed light on how agri- cultural distortions may change over different stages of economic development in other countries. To begin, a succinct summary of core characteristics of the region in terms of the nature of the three economies--including their resource endowments, which * The authors are grateful for helpful comments from workshop participants, for invaluable help with data compila- tion by Johanna Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, Damiano Sandri, and Ernesto Valenzuela. The working paper version of this chapter (Honma and Hayami 2008) contains additional back- ground material and appendix tables. This chapter draws on the introductory and country chapters in Anderson and Martin (2009), with data updated using Anderson and Valenzuela (2008). 67 68 Distortions to Agricultural Incentives: A Global Perspective to some degree determined the course of their modern economic growth and development--is given. The evolution of agricultural policies in the three economies is then reviewed, followed by a discussion of how to measure distor- tions to agricultural incentives using the methodology from Anderson et al. (2008a, 2008b), the focus of which is on nominal and relative rates of assistance (NRAs and RRAs). Implications of empirical findings of this study for policy reforms in the three economies are discussed in the final section, where lessons for other economies experiencing similar structural transformations in the course of their economic growth are also identified. The study finds that significant growth in agricultural protection began when Japan, Korea, and Taiwan, China entered the middle-income stage of economic development. Statistical observations are found to be consistent with the hypoth- esis that the success of rapid industrialization that advanced these economies beyond the middle-income stage resulted in a decline in agriculture's comparative advantage associated with the growing income disparity between farmers and employees in nonagricultural sectors. In several cases, demand from farmers for a reduction of farm-nonfarm income disparity materialized in the form of increased assistance to agriculture. This was manifest predominantly through rapid and sustained growth in border protection of agricultural products. Economic Development and Structural Change A government's choice of agricultural policies, particularly price-distorting poli- cies, is closely related to the process of economic development. As identified by Schultz (1978), there are two agricultural problems. One is the "food problem," which underlies policies commonly adopted in low-income countries that exploit or tax agriculture. Those policies contrast with the policies that protect or subsidize agriculture in many high-income countries seeking to solve "the farm problem." Schultz's hypothesis became an established paradigm among agricultural econo- mists, who supported it in several empirical studies (Anderson and Hayami 1986; Hayami 1988; Krueger, Schiff, and Valdés 1991). More recently, Hayami (2005) and Hayami and Godo (2004) have added the "disparity problem" as specific to mid- dle-income economies. They suggest it is important to study how distortions in agricultural incentives change as economies move through different stages of development. The most distinguishing characteristic of Japan, Korea, and Taiwan, China dur- ing the period examined here is their unusually rapid rates of economic growth and industrial development. In describing the so-called East Asian miracle, the World Bank (1993) depicted Japan as the front runner along the development spectrum, with Korea and Taiwan, China, together with Hong Kong, China and Japan, Republic of Korea, and Taiwan, China 69 Singapore, as the second group.1 Mainland China and the various countries of the Association of Southeast Asian Nations (ASEAN)2 are following behind, though with rapid rates of economic growth. Since the changing nature of distortions of agricultural incentives seems to be closely related to rate of economic growth and structural change, an overview is provided of the development of the three economies examined here and associated changes in economic and agricultural structures. Initial conditions and development strategies Weather and topographical conditions in Northeast Asia are characterized by reg- ular monsoon rain, together with mountainous, undulating land in which irriga- tion water is controlled relatively easily with efforts at the family and community levels. This makes the region well suited to rice production by small family farms organized into village communities. Indeed, the traditional agrarian structure involved smallholder farms of an average size of about one hectare that were pre- dominantly dependent on rice cultivation. It is important to recognize that, unlike in Southeast Asia, large agribusiness plantations dependent on hired labor were almost completely absent not only in Japan and Korea (temperate-zone coun- tries), but also in Taiwan, China, where tropical cash crops such as sugar and bananas comprised a significant share of agriculture. Prior to the land reforms that followed World War II, the rural community was stratified across landlords, land-owning cultivators, and landless tenants. Agricultural laborers subsisting on hired-labor wages were not a significant component of the rural population. Historically, there is a high degree of similarity in the agrarian structures of Japan, Korea, and Taiwan, China, due in part to the fact that Japan brought its institutions to its colonies--Taiwan, China starting in 1895 and Korea starting in 1910. In all of these places, the most fundamental land-related institution was the fee simple title granted to land owners through cadastral surveys, in return for their payment of land tax. Japanese efforts to develop the colonies concentrated on agriculture, and particularly on rice after Japan experienced a supply shortage after the Rice Riots in 1918. The promotion of rice production through agricul- tural research and extension systems, irrigation and drainage infrastructure, and import protection (see Anderson and Tyers 1992) was considered a major success from Japan's viewpoint in that rice imports from the two colonies increased from 5 to 20 percent of consumption in Japan between 1915 and 1935. The increased export of rice and other primary commodities and the correspon- ding inflow of manufactured commodities meant dependency of the Korean and Taiwanese economies on agriculture remained high. This tendency was especially pronounced in the southern part of Korea, as Japanese industrial development 70 Distortions to Agricultural Incentives: A Global Perspective efforts on the Korean Peninsula were concentrated in the north: the hydroelectric power of the Yalu River fed a complex of chemical industries larger than those that existed in Japan at the time. The heavy dependency on agriculture in the southern part of Korea was furthered by urban destruction during the Korean War (1950­ 53). Relative to what is now the Republic of Korea, Taiwan, China had more active commerce and industry, largely because the dominant cash-crop sector there required significant amounts of processing and marketing activities relative to subsistence crops such as rice and barley that were produced in Korea. In fact, the situation in Taiwan, China under colonial rule was somewhat akin to Japan between the mid-19th to the early 20th centuries, when commercial treaties imposed by Western countries deprived Japan of tariff autonomy. Consequently, Japan specialized in labor-intensive manufactured products based on farm- supplied materials such as silk reeling, tea processing, and cotton weaving. This scenario accorded with comparative advantage under virtual free trade, leading to a wide dispersion of small and medium industries in rural areas in Japan. The predominance of small-scale industry in Japan and Taiwan, China contrasts with the concentration of Korean industry in large-scale establishments in urban areas. The reasons Japan, Korea, and Taiwan, China came out as the forerunners in the East Asian miracle are numerous.3 Here, suffice it to say that their success was due to pertinent borrowing of technology from advanced economies. Gerschenkron (1962), for example, suggests that the later industrialization of an economy begins, the larger the scope for economic growth through technology borrowing. The question remains, however, as to why Japan, Korea, and Taiwan, China were particularly successful in technology borrowing among the many countries that began late starters. One commonly cited speculation is the fact that these three resource-poor countries were heavily endowed with cheap but relatively well-educated labor, a situation that made initial borrowing of labor- intensive technologies fairly efficient and smoothed the way for borrowing of capital- and knowledge-intensive technologies. Another reason is the great crises faced by these economies--Japan's defeat in World War II, the Korean War, and Taiwan, China's loss of the mainland to the communists--all of which compelled leaders to adopt policies that would lead to economic success for the sake of main- taining their legitimacy, instead of indulging in rent-seeking activities (Hayami and Godo 2005). Despite much similarity, there were also significant differences in the industri- alization strategy adopted by the three, especially between Korea and Taiwan, China. In Japan, despite policies aimed at promoting the development of capital- intensive industries after the recovery of tariff autonomy in 1911, small- and Japan, Republic of Korea, and Taiwan, China 71 medium-sized industries survived as a major component of the industrial sector, and many of them located in rural areas. In Taiwan, China, although the Nation- alist Party tightly controlled formal sectors, there was little government interven- tion in the activities of small- and medium-sized companies, which were able to grow through various marketing and financial linkages among themselves and with foreign firms. They became very internationally competitive (Ho 1979, 1982). In contrast, government control in Korea was stronger and more wide- ranging, especially under the military administration of Pak Chong-hui (1961­ 79). All formal credit issuance was channeled from nationalized banks to large industry, while foreign direct investment was tightly controlled (Cole and Park 1983; Amsden 1989). This strategy underlies the high concentration of industrial production in a small number of large enterprises in Korea. Economic growth and structural transformation Quantitative indicators of economic development in Japan, Korea, and Taiwan, China over the past five decades show similarities and differences, as summarized in table 2.1. The first three rows indicate real gross domestic product (GDP) per capita in constant 2000 prices at purchasing power parity (PPP), taken from Heston, Summers, and Aten (2006). In 1955, Japan's GDP per capita was more than US$3,000, whereas it was less than US$1,500 in Taiwan, China. Japan's strong economic growth thereafter pushed the figure to US$4,500 in 1960 and US$5,000 in 1961. In 1960, Korea and Taiwan, China had per capita GDP averages of around US$1,500; by 1978 in Taiwan, China, and 1983 in Korea, both countries were at US$5,000. Japan was the first, in 1970, to surpass US$10,000, a level that Taiwan, China and Korea reached in 1988 and 1991, respectively. Roughly speaking, in terms of growth of per capita GDP, Japan was ahead of Taiwan, China by about two decades and Taiwan, China was ahead of Korea by about half a decade, although these margins decreased over time. Though the criteria of classification are not universal, it is convenient to classify economic development in the three economies into four stages:4 · Low income (US$1,500 or less): pre-1950 for Japan and pre-1960 for Korea and Taiwan, China; · Lower middle income (US$1,500­5,000): 1950­60 for Japan and 1960­80 for Korea and Taiwan, China; · Upper middle income (US$5,000­10,000): 1960­70 for Japan and 1980­90 for Korea and Taiwan, China; and · High income (US$10,000 or more): post-1970 for Japan and post-1990 for Korea and Taiwan, China. 72 Distortions to Agricultural Incentives: A Global Perspective Table 2.1. Economic Growth and Structural Transformation in Japan, Korea, and Taiwan, China, 1955­2004 1955 1960 1970 1980 1990 2000 Real GDP per capita, Japan 3,128 4,509 11,391 15,520 21,703 23,971 constant 2000 $a Korea 1,429 1,458 2,552 4,497 9,593 15,702 Taiwan, China 1,241 1,444 2,846 5,963 11,248 19,184 Share of agriculture Japan 17.4 9.0 4.2 2.4 1.7 1.1 in GDP (%) Korea 46.9 39.1 29.2 16.2 8.9 4.9 Taiwan, China 28.9 28.2 15.3 7.5 4.0 2.0 Share of agricultural Japan 33.8 26.8 15.9 9.1 6.2 4.5 workers in Korea 79.7 60.2 49.1 37.1 18.1 10.0 economically active Taiwan, population (%)a China 53.6 50.2 36.7 19.5 12.8 8.9 Share of farm Japan 40.7 36.5 25.1 18.3 14 8.2 household Korea 61.9 58.2 44.7 28.4 15.5 8.6 population in total Taiwan, population (%) China 50.7 49.8 40.9 30.3 21.1 16.5 Agricultural GDP per Japan 51.5 33.6 26.4 26.4 27.4 24.4 worker/total GDP per Korea 58.8 65.0 59.5 43.7 49.2 49.0 worker (%) Taiwan, China 53.9 56.2 41.7 38.5 31.3 22.5 Sources: Heston, Summers, and Aten (2006); Japanese Ministry for Agriculture, Farming and Fisheries (JMAFF), Nogyo Hakusho Fuzoku Tokei-hyo (Statistical Appendix of Agricultural White Paper), various issues; government of Korea, Major Statistics of Agriculture, Forestry and Fisheries, various issues; government of Taiwan, China, Taiwan Agricultural Year Book, various issues. a. Shares of agriculture in GDP and labor force include forestry and fisheries. Changes in other indicators in table 2.1 are closely related to changes in per capita real GDP over the four stages. The GDP share of agriculture in Japan in 1955--the lower-middle-income stage--was 17 percent, substantially less than the share during the low-income stage for Korea and Taiwan, China (47 and 29 percent, respectively). By 1970 in Taiwan, China and 1980 in Korea, the shares had declined to similar levels to that of Japan at the lower-middle-income stage-- 15 percent and 16 percent, respectively. By the time Japan entered the high- income stage in 1970, its agricultural share of GDP was 4 percent, about the same as that of Korea and Taiwan, China in the 1990s, when they entered the high- income stage. Although Korea and Taiwan, China experienced similar changes in per capita real GDP over the four stages, significant differences can be observed in their economic structures. The GDP share of agriculture in 1955 in Korea was nearly 50 percent, whereas it was less than 30 percent in Taiwan, China. Similar differences Japan, Republic of Korea, and Taiwan, China 73 can be observed with respect to agriculture's share of the labor force, which was as high as 80 percent in Korea, versus less than 55 percent in Taiwan, China. In both countries, the share of agriculture in GDP declined significantly over time--to 3.8 percent in Korea and 1.7 percent in Taiwan, China by 2004--although the share in Korea was still double that of Taiwan, China at that point. Historical differences can be observed in agriculture's share of the labor force in Korea and Taiwan, China. Interestingly, however, the difference in the labor force share of agriculture disappeared by 2004, with about 7.7 percent of Korean employment in agriculture versus 7.5 percent in Taiwan, China. The relatively faster decline in the labor force share of agriculture in GDP in Korea reflects its concentration of industries in urban areas. In Taiwan, China, characterized by the wide dispersion of industries in rural areas, farmers increased their incomes from off-farm employment while continuing to be classified as farmers. In contrast, a much greater percentage of rural people in Korea had to quit farming and migrate to urban areas to obtain nonfarm employment. These differences are reflected in the much faster decrease in the farm household population as a percentage of the total population in Korea relative to Japan and Taiwan, China. The last rows in table 2.1 report the ratios of agricultural GDP per worker to total GDP per worker. This can be considered an indicator of the relative labor productivity of agriculture to total labor productivity of the whole economy. It may also be regarded as an indicator of the income gap between the agricultural sector and the whole economy. While the relative labor productivity of agricul- ture in nominal terms was not substantially different among the three economies in 1955, in Japan it declined sharply, from 52 percent in 1955 to nearly 25 percent in 1970 and thereafter. The ratio did not begin to decline in Korea and Taiwan, China until 1960. In Korea, it reached 44 percent in 1980 (from 65 percent in 1960) and stayed nearly constant thereafter. In Taiwan, China, the ratio contin- ued to decline, to 23 percent in 2000--a rate lower than that of Japan, reflecting the abundance of nonfarm employment opportunities for farmers in Taiwan, China. These measures should be interpreted with great care, however. At first glance, a faster decline in this ratio of agricultural labor productivity to total labor pro- ductivity in Taiwan, China relative to Korea appears to indicate more rapid growth in agricultural labor productivity in Korea. In fact, however, the faster decline ratio in Korea is caused by more rapid out-migration of farm labor to urban occupations in Korea relative to Taiwan, China. Thus, growth in the labor productivity of farmers engaging in agricultural activities relative to that of other workers would not have been slower and could have been even faster in Taiwan, China if the ratio was calculated using output per hour of labor instead of output per worker according to the official sectoral labor force classification. 74 Distortions to Agricultural Incentives: A Global Perspective Changes in the structure of agriculture How did the structure of agriculture in Japan, Korea, and Taiwan, China change in the course of their economic development? As of 2004, Japan's 2.9 million farm households accounted for 2.6 million workers engaged mainly in agricultural activities, an average of less than one person per household. Japan's Agricultural Census defines a farm household as one that either operates 0.1 hectare or more of farmland or produces annual sales of agricultural products of at least 150,000 yen (US$1,250 at the exchange rate of 120 yen/US$1). Thus, very small units of farm operation in which no full-time worker engages in farm production are classified as farm households. Indeed, full-time farm households having no family member engaged in agricultural employment accounted for 15 percent of total farm households in 2004. However, noncommercial farm households that operate less than 0.3 hectares of farmland or have annual sales of less than 500,000 yen accounted for 26 percent of total farm households. Moreover, part-time farm households whose income from nonagricultural sources exceeds agricultural income accounted for half of total farm households. Though the number of agricultural workers in Japan declined from 14 million in 1955 to 2.6 million in 2004, the number of farm households declined only from 6 million in 1955 to 2.9 million in 2004. Slow decreases in the number of farm households, together with decreases in agricultural land (from 6.1 to 4.7 million hectares between 1955 and 2004), resulted in a small increase in arable land per farm, from 1.01 hectares in 1955 to 1.61 hectares in 2004 (table 2.2). The average area of agricultural land per farm in Japan is very small by global standards-- Europe's are 20 to 45 times larger, and those of the United States are 125 times larger. The slow growth of small-sized operations has been a key constraint to the growth of agricultural productivity, resulting in a continual decline in the com- parative advantage of Japanese agriculture, particularly of land-intensive activi- ties, in the course of rapid industrial development.5 Table 2.2 shows that in 2004, 2.9 million farm households in Japan contained 9.4 million people, with an average family size of 3.2 persons. In the same year, 1.2 million farm households in Korea represented 3.4 million people and 1.8 mil- lion hectares of arable land, with average family and farm sizes of 2.8 persons and 1.5 hectares, respectively. In Taiwan, China, 0.72 million farm households con- tained 3.2 million people and 0.84 million hectares of arable land, while average family and farm sizes were 4.5 persons and 1.2 hectares, respectively. It is notable that the number of people in farm households in Korea declined at a much faster rate than in Japan and Taiwan, China as a result of faster decreases in both the number of farm households and the number of persons per household in Korea. These observations reflect the scarcity of nonfarm employment opportunities in Japan, Republic of Korea, and Taiwan, China 75 Table 2.2. Changes in Agricultural Structure in Japan, Korea, and Taiwan, China, 1955­2004 1955 1960 1970 1980 1990 2000 2004 Number of farm Japan 6,043 6,057 5,342 4,661 3,835 3,120 2,934 households Korea 2,218 2,350 2,483 2,155 1,768 1,383 1,240 (thousands) Taiwan, China 733 786 880 891 860 721 721 Population Japan 36,347 34,411 26,282 21,366 17,296 10,467 9,400 in farm Korea 13,300 14,559 14,422 10,827 6,661 4,031 3,415 households Taiwan, (thousands) China 4,603 5,373 5,997 5,389 4,289 3,669 3,225 Persons per farm Japan 6.01 5.68 4.92 4.58 4.51 3.35 3.20 household Korea 6.00 6.20 5.81 5.02 3.77 2.91 2.75 Taiwan, China 6.28 6.84 6.81 6.05 4.99 5.09 4.47 Arable land Japan 6,095 6,071 5,796 5,461 5,243 4,830 4,714 (thousands Korea 1,995 2,025 2,298 2,196 2,109 1,918 1,836 of hectares) Taiwan, China 873 869 905 907 890 852 836 Arable land Japan 1.01 1.00 1.08 1.17 1.37 1.55 1.61 per farm Korea 0.90 0.86 0.93 1.02 1.19 1.39 1.48 household Taiwan, (hectares) China 1.19 1.11 1.03 1.02 1.03 1.18 1.16 Share of Japan 70.7 49.5 31.9 17.0 13.8 13.1 14.3 agricultural Korea -- -- 75.8 65.2 56.8 47.2 39.3 income in total Taiwan, farm household China -- -- 48.7 24.8 20.1 17.6 22.0 income (%) Share of rice Japan -- 47.4 37.9 30.0 27.8 25.4 22.8 in value of Korea -- 59.3 37.3 34.1 36.9 32.9 27.6 agricultural Taiwan, production (%) China -- 36.5 25.7 19.8 12.1 9.6 7.1 Sources: JMAFF, Nogyo Hakusho Fuzoku Tokei-hyo (Statistical Appendix of Agricultural White Paper); government of Korea, Major Statistics of Agriculture, Forestry and Fisheries; government of Taiwan, China, Taiwan Agricultural Year Book; various issues. Note: -- not available. Korea's rural areas due to its urban-centered industrialization. Indeed, from 1970 to 2004, the share of agricultural income in the total income of farm households declined from 32 to 14 percent in Japan and 49 to 22 percent in Tai- wan, China, whereas in Korea it was 76 percent in 1970 and 39 percent in 2004. Japan has lost 23 percent of its arable land area over the past 50 years, falling from 6.1 million hectares in 1955 to 4.7 million hectares in 2004. The decrease in arable land under cultivation was thus a significant contributor to changes in farm 76 Distortions to Agricultural Incentives: A Global Perspective size in Japan in terms of arable land per farm household. Meanwhile, the arable land area in Korea deceased from 2 million hectares in 1955 to 1.8 million hectares in 2004, and arable land in Taiwan, China remained almost constant (0.87 million hectares in 1955 and 0.84 million hectares in 2004). Farm-size changes in Korea and Taiwan, China were almost exclusively the result of changes in the number of farm households. In Japan and Korea, average farm sizes increased slowly, from 1.0 and 0.9 hectares in 1955 to 1.6 and 1.5 hectares in 2004, respectively, whereas farm size in Taiwan, China remained almost constant during that period. The faster increase in farm size in Korea relative to Taiwan, China was the result, again, of faster out-migration of farm workers and their families to urban areas. The distinctly urban-centered industrialization of Korea is clearly reflected in its high share of agricultural income in total farm household income. In all three economies, the ratio decreased as off-farm employment for members of farm households increased. In Japan, the ratio decreased from 70 percent in 1955 to 32 percent in 1970 and further to 14 percent in 2004, corresponding to the shift from the lower-middle-income to upper-middle-income and finally high-income stage of development. In Taiwan, China, where the ratio of agricultural income to total farm income was already below 50 percent in 1970 when its economy was in the lower-middle-income stage, it dropped to 22 percent in 2004. In contrast, in Korea the ratio was 76 percent in 1970 and still nearly 40 percent in 2004, higher than in Taiwan, China but also higher than in Japan at comparable development stages. Major differences in the adjustment of agriculture to economic growth based on industrial development are also reflected in changes in the commod- ity mix of farm production. Traditionally, rice was the most important crop in all three economies, but its significance declined as per capita income increased. Changes in the relative importance of rice were different. From 1960 to 2004, the share of rice in the total value of agricultural production declined from 47 to 23 percent in Japan and from 59 to 28 percent in Korea. In contrast, the share of rice in Taiwan, China, which was less to begin with, at 37 percent, decreased rapidly, to 7 percent in 2004. The contrast reflects the fact that the agricultural sector of Taiwan, China traditionally depended less on rice because of both its greater opportunity to grow cash crops and its success in achieving greater agricultural diversification toward high-valued commodities such as vegetables, fruits, poultry, and pig meat more efficiently than Japan and Korea in response to the shift in domestic demand for more income-elastic commodities. Japan, Republic of Korea, and Taiwan, China 77 Evolution of Agricultural Policy after World War II Japan6 In the decade following World War II, Japan worked hard to recover from the devastation of the war. The primary emphasis of agricultural policy was on increasing domestic food production and delivering food equitably at low costs to consumers. To this end, the government invested heavily in agricultural research and extension and land infrastructure, while upholding many of the rigid controls on rice procurement from farmers and delivery to consumers established under the Food Control Laws enacted during the war. Immediately following World War II, land reform was carried out in accor- dance with the strong recommendations of occupying authorities. The urgent need to increase agricultural production through increased production incentives to cultivators was sufficiently strong to overcome the opposition of landlords to strengthening the rights of tenants through government control of rents and land prices. During the four years from 1947 to 1950, the government purchased 1.7 million hectares of farmland from landlords and transferred 1.9 million hectares, including state-owned land, to tenant farmers, which amounted to about 80 percent of the land under tenancy before the land reform. Although land reform resulted in a considerable change in the distribution of land ownership, the size distribution of operational holdings remained basically the same. As a result, the traditional agrarian structure of Japan, characterized by small-scale family farms with an average size of about l hectare, remained intact, despite the rise and the fall of landlordism (table 2.2). There is no doubt, how- ever, that the land reform promoted more equal asset and income distribution among farmers, and hence contributed to social stability in Japan's rural sector. But the fact that small-scale family farms continued to be the basic unit of agricultural production meant that land reform did not induce changes in the basic direction of technological developments. Though land reform contributed to an increase in standards of living and consumption levels, its contribution to capital formation and productivity growth in agriculture were not significant (Kawano 1969). As Japan set off on its "miraculous" economic growth path following the end of the Korean War in the 1950s, agriculture began to face serious adjustment prob- lems. The rate of growth in agricultural productivity, rapid by international stan- dards, was not rapid enough to keep up with growth in the industrial sector, and intersectoral terms of trade did not improve during the 1950s. This was partly because of the pressure of surplus agricultural commodities in the United States and other exporting countries, and partly because domestic demand for major 78 Distortions to Agricultural Incentives: A Global Perspective staple cereals (especially rice) approached saturation after the bumper crop of 1955. As a consequence, incomes and living standards of farm households lagged behind those of urban households during the 1950s. In 1961, real GDP per capita in Japan exceeded US$5,000 pushing Japan into the upper-middle-income stage of economic development. Correspondingly, the major goal of agricultural policy shifted from increased production of food sta- ples to reducing the rural-urban income gap. The need to assist farmers increased during the 1960s, as the rural-urban income gap progressively widened and the out-migration of agricultural labor accelerated. The difficulty of structural adjust- ment in agriculture as a result of the rapidly growing economy led to the enact- ment in 1961 of the Agricultural Basic Law, a national charter for agriculture. The law declared that it was the government's responsibility to raise agricultural pro- ductivity and thereby to close the gap in income and welfare between farm and nonfarm households. In order to raise agricultural productivity and improve farming efficiency, it was considered essential to increase the scale of farm operation by eliminating inefficient farm units and by promoting cooperative operations among remaining farms. Despite such efforts at structural adjustment, the rate of agricultural pro- ductivity growth was not increased sufficiently to prevent the rural-urban income gap from widening further. The Food Control System, which was originally designed to provide food security to consumers, thus became the chief instrument to protect farmers. Under the system, based on the 1942 Food Control Law, most food items were placed under direct government control. However, as the Japanese economy recovered from World War II, the number of items under control was reduced so that only rice remained under direct control after 1952. Initially, the whole marketing process of rice from producers to consumers was under direct control of the Japanese Food Agency and prices were regulated from the farmgate to the retail level, although the regulations were gradually relaxed. After Japan entered the upper-middle-income stage of development in the 1960s, the Food Control System became a powerful instrument for rice farmers, and they organized political lobbying to raise rice prices for government pur- chases. Their pressure resulted in a rice price determination formula in 1960 called the production cost and income compensation formula, established under the Food Control System. This formula was designed to reduce the gap between farm and nonfarm income and wages by raising rice prices. The goal appears to have been achieved: income per agricultural worker compared to income per worker in manufacturing improved after 1960 following a rapid rise in agricul- tural relative to manufacturing prices. The increase in the price of rice, which con- stituted about 40 percent of the total value of agricultural output before 1970, was a major factor in improving the domestic terms of trade for agriculture. The rise Japan, Republic of Korea, and Taiwan, China 79 in agricultural prices, together with increases in off-farm income, resulted in a marked reduction in the income per capita gap between agricultural and nonagri- cultural households. Protecting rice farmers through a price policy was possible in Japan because rice trade was completely controlled by the state-trading system. During the 1960s, the price of rice was raised not only far above the world price but also above the market equilibrium price under autarky. As an upper-middle-income country, Japan was able to let consumers and taxpayers shoulder the costs of agricultural protection. However, there was a limit on increasing agricultural protection through price policy. The high protected prices of rice resulted in an expansion of rice produc- tion in excess of consumption. The accumulated surplus of rice in government storage forced authorities to introduce controls on rice acreage in 1969 (they are still in place today). Further, the dramatic increase in income and wages of indus- trial workers after 1960 meant that the diet of Japanese people changed. Average consumption per industrial employee (deflated by the consumer price index) doubled over 1955­70, and again in the decade-and-a-half following that. Corre- spondingly, rice was no longer a major wage good for industrial workers. To cope with the increasing rice surplus, the Food Control System was revised. The direct control on rice distribution was relaxed by introducing nongovernment distribu- tion channels. Finally, in 1995, the Food Control Law was replaced by the Staple Food Law, whereby the role of government was limited to stock holding opera- tions for food security, although state trading of rice is maintained for interna- tional trade. Real GDP per capita in Japan exceeded US$10,000 in 1969, putting it over the high-income threshold. Demand for agricultural protection from the farm bloc increased. Japan's comparative advantage, though, continued to shift from agricul- ture to industry, while internal resistance to protectionism declined because the nonfarm population became affluent and, hence less resistant to shouldering the cost of agricultural protection in the form of high food prices or subsidies to farm producers. However, though internal resistance weakened, external pressure for liberalization of agricultural imports increased. Following the replacement of the General Agreement on Tariffs and Trade (GATT) with the World Trade Organization (WTO) in 1995, Japan was required to reform domestic agricultural policy. Under the WTO's Uruguay Round Agree- ment on Agriculture (URAA), in particular, Japan had to adjust agricultural poli- cies to be more consistent with the globalization of the economy. In 1995, Japan converted nontariff border measures to tariffs for 28 commodities. At the beginning of implementation, rice was exempted from tariffication in compensation of larger minimum access imports of rice, namely 4 percent of domestic consumption in 80 Distortions to Agricultural Incentives: A Global Perspective 1995, rising to 8 percent by 2000. In 1999, Japan adopted tarriffication for rice, leaving the minimum access imports at 7.2 percent of domestic consumption. The Basic Law on Food, Agriculture and Rural Areas, also enacted in 1999, was a replacement for the 1961 Agricultural Basic Law. Four years earlier, the Food Control Law had been abolished to liberalize the domestic rice market. The 1999 Basic Law obliged the government to draft a Basic Plan for Food, Agriculture and Rural Areas for the promotion of the comprehensive and systematic implementa- tion of policies on food, agriculture, and rural areas. The plan is intended to be redrafted every five years. Under the current Basic Plan, released in 2005, a key point of the new agricultural policy is to target government assistance to farmers who satisfy certain conditions, especially on minimum farm size. That is, it com- pels farmers who want to continue farming under government assistance to expand the size of their farm operation. Korea7 Before 1960, Korea was a low-income country, with per capita income below US$1,500. In the years just before 1960, the economy was struggling to recover from the Korean War. The agricultural policy adopted at this stage aimed to main- tain low domestic consumer prices for staple foods, notably rice and barley, as well as for fertilizer. The Grain Management Law, enacted in 1950, gave the govern- ment authority to regulate the price of staple foods. But government control was not very effective during the 1950s since the market share of government- controlled rice was less than 10 percent. The government, which was supposed to purchase grain directly from farmers, was unable to purchase sufficient amounts due to budgetary constraints and an inflation-driven spike in grain prices in the mid-1950s. As a creative solution, schemes to collect rice as an in-kind land tax and to barter fertilizer for rice were initiated. Though the former measure was successful, the latter was not because the implicit price of rice in the barter was lower than the market price. In the end, grain imports from the United States under Public Law 480, which accounted for 8 to 12 percent of total domestic grain during 1956­65, helped the Korean government hold down grain prices. In the 1960s, Korea launched wide-ranging policies to promote industrializa- tion under the development autocracy of Pak Chong-hui. Agricultural policies at this time were designed to keep the price of staple food crops low so as to uphold a low cost of living and stable wage rates for industrial workers, rather than main- taining adequate incomes for farmers. Government purchase prices for staple crops, which were considered necessary for the purpose of increasing industrial profits and capital formation, were below market prices. Over time, the Korean government's price intervention became more intense. The market share of Japan, Republic of Korea, and Taiwan, China 81 government-controlled rice was expanded to 20­25 percent during the 1960s; the increase was used mainly to maintain low domestic prices. These agriculture- taxing policies continued during the beginning of Korea's lower-middle-income stage of economic development. As Korea quickly advanced toward becoming an upper-middle-income coun- try, the direction of agricultural policy gradually moved toward supporting farm- ers. In the early 1970s, the buffer-stock operation for noncereal products was set in motion to counteract price declines. In addition to chemical fertilizers, pesticides and farm machineries were added to the list of subsidized inputs, alleviating the adverse impact of import protection to manufacturers of those inputs for farmers. Through the 1970s, the government's purchase prices for rice and barley were steadily raised with the aim of both increasing food production and reducing the urban-rural income gap. Although the government raised producer prices for sta- ple food grains, it did so without a comparable rise in the market prices of rice and barley in order to prevent the cost of living and the wage rate of industrial workers from rising. Likewise, it assisted livestock producers in part by using import quo- tas rather than tariffs to protect them from import competition, with the rent from those quotas being captured by the producer-managed meat import agency.8 The implementation of the two-price system, however, conflicted with the need to maintain financial and monetary stability. As the difference between the purchase and sale prices of rice and barley widened, the deficit of the grain man- agement fund increased. Since a large portion of this deficit was financed by long- term overdrafts from the Bank of Korea, the policy contributed heavily to infla- tionary pressure. Expansion of the government deficit due to the two-price policy became a serious constraint on the policy. Upon entering the upper-middle-income stage of development in the 1980s, the Korean government moved to reduce both tariff and nontariff protection for manufacturing industries. In contrast, agricultural policies toward protecting farmers were strengthened, and the producer prices of farm products were increased to levels far above border prices by means of quantitative import restric- tions on most agricultural commodities. Significant policy changes followed Korea's transition to a high-income coun- try in the early 1990s. Mostly, these changes were related to the URAA. According to the provisions, Korea's quantitative restrictions were converted to tariffs for all agricultural products except rice. But Korea retained the status of a developing country in the Uruguay Round negotiations, giving it special treatment in imple- menting commitments to reduce border protection. The agricultural products under tariffication were subject to a protection reduction commitment of 24 per- cent, on average, within 10 years, with a minimum cut of 10 percent. Tariff rates of Korean agricultural products were over 60 percent on average. Tariffs on products 82 Distortions to Agricultural Incentives: A Global Perspective considered particularly important in Korea were thus cut by the minimum rate of 10 percent. In addition, imports of many agricultural products were begun under the min- imum market access commitment. This commitment required that for all agricul- tural products, at least 3 percent of consumption must be purchased from over- seas in the first year and the import share must increase annually up to 5 percent of consumption within 10 years. Low tariff rates were applied to the in-quota vol- ume so as to guarantee easy market access from exporting countries. Key agricul- tural products such as rice, barley, oranges, red peppers, garlic, and onions began to be imported under the commitment. Rice, the most important agricultural product in Korea, was temporarily exempted from tariffication, as provided in annex 5.B of the URAA. Under the exception plan, rice was subject to an import quota, beginning with 1 percent of total consumption and gradually increasing to 4 percent in 2004, the final imple- mentation year. If Korean rice had not been exempted from tariffication, Korea would have complied with the standard market access commitment of 3­5 per- cent. The temporary exemption from tariffication expired in 2004, but Korea opted to continue invoking a rice exemption from tariffication for another 10 years, to 2014. Taiwan, China9 In the years after World War II, Taiwan, China suffered from high inflation rates, serious shortages of food and other necessities, and a heavy budget burden for its defense systems. The government gave the highest priority to economic stabiliza- tion, food production increases, and the repair of war damages. To alleviate the intense population pressure on limited land, it decided to grant incentives to farmers. Together with the land reform program implemented between 1949 and 1953, war-damaged irrigation and drainage facilities were repaired, fertilizers and other farm inputs were made available, and farmers' organizations were strengthened. During this recovery stage, the Sino-American Joint Commission on Rural Reconstruction (JCRR) was established in Nanking in 1948. That agency played an important role in the postwar rural reconstruction and development of Taiwan, China. From 1951 to 1965, the United States provided a total of US$1.5 billion in aid. Approximately one-third went to agriculture, which was used to build infrastructure and foster human resources for agriculture. Substantial imports of commodities financed by U.S. aid and increases in domestic production, especially of food, helped relieve demand pressures in Taiwan, China. Japan, Republic of Korea, and Taiwan, China 83 During its low-income stage of economic development (before 1960), agricul- tural policy in Taiwan, China was designed mainly to supply rice at low, stable prices to the nonfarm population. In those days, two important taxes were imposed on farmers: the farm land tax and the hidden rice tax. They were levied by means of compulsory rice purchases and the rice-fertilizer barter system. The compulsory purchase of paddy rice from landowners at official prices was another source of gov- ernment control over rice, such that all rice paddy land was subject to the paddy land tax plus the compulsory procurement of rice. The compulsory procurement was assessed on the basis of tax units determined by land productivity. The differ- ence between the government procurement prices and farmers' market prices con- stituted a hidden tax on paddy landowners, who were mostly farm operators after the implementation of land reform program. Though the hidden tax was gradually reduced as per capita income in Taiwan, China rose, it was not abolished until 1973. The government's rice collection by all of these methods during 1950­70 aver- aged 50 to 60 percent of the total amount of rice produced minus farmers' home consumption. By 1973, this share had declined to 20 percent. In subsequent years, it increased again because of the implementation of the guaranteed rice price policy. The total of this hidden rice tax was larger than the total income tax of Taiwan, China before 1963 and was more than twice the farm land tax before 1961, except in 1954. After 1961, when Taiwan, China moved into the lower-middle-income stage of economic development, the hidden rice tax decreased rapidly: the ratio of the hid- den rice tax to the total income tax was only 8.5 percent in 1971 (Kuo 1975). Agricultural policy geared to exploit agriculture for the sake of supporting industrial development (and military development) largely ended during the 1970s, when the shift to subsidizing agriculture began. This period was when Taiwan, China rapidly expanded its labor-intensive light industries in response to increases in export demand. Because several of these light industries, such as gar- ments and footwear, were located in rural areas, nonfarm incomes became increasingly important to farm households. Farmers in Taiwan, China were able to take advantage of employment in manufacturing without leaving home, and many of them engaged in nonfarm self-employed industrial activities during less- busy farm seasons. The need for farmers to rely on agricultural protection policies was therefore smaller than in Korea. It was 1978 when GDP per capita exceeded US$5,000 in Taiwan, China, push- ing it into the upper-middle-income stage of economic development. Still, to help equalize the income level of farm workers with that of the rapidly expanding industrial sector, the government offered loans and subsidies for promoting farm mechanization, which were designed to raise farmers' labor productivity. The expansion of rice production began to slow down in response to an increased emphasis on livestock and fishery products and high-value export crops. Increases 84 Distortions to Agricultural Incentives: A Global Perspective in industrial employment also drove up the cost of farm labor. Labor productivity in agriculture continued to lag behind that of the industrial sector, and the gap between farm and nonfarm per capita incomes was increased, especially for farm- ers who relied mainly on rice production. The problems faced by Taiwan, China agriculture were similar to those that many other industrial countries experienced at a comparable development stage, especially Japan in the early 1960s and Korea in the late 1970s. Per capita consumption of rice in Taiwan, China fell from 140 kilograms per year in 1968 to 74 kilograms in 1988. Correspondingly, an excessive stock of rice accu- mulated. In order to reduce production, farm extension workers encouraged farm- ers to plant other crops in rice fields, but their efforts were not successful because no economic incentive was provided. A six-year rice crop substitution plan inaugurated in 1984 gave direct subsidies of 1 metric ton of paddy rice per hectare to farmers who shifted their rice fields to corn or sorghum, or 1.5 metric tons of paddy rice per hectare to farmers who shifted to crops other than corn and sorghum. In addition, corn and sorghum were purchased by the government at guaranteed prices. Under the program, rice production declined to 1.84 million metric tons in 1988, 0.9 mil- lion metric tons less than the peak of 1976. In 1988, the paid-in-kind subsidy was changed to a cash payment to improve the efficiency of the program. Taiwan, China entered the high-income stage of development in 1988, when its real GDP per capita exceeded US$10,000. The most important change in agricul- tural policy in the years that followed was related to the economy's accession to the WTO on January 1, 2002. In line with its level of economic development, Taiwan, China agreed to bring its tariff rates to a level between those of Japan and Korea--as such, its average nominal tariff rate was reduced from 20 percent to 14 percent in the year following accession and gradually to 12.9 percent by 2004. The plan exempted 137 items covered under tariff rate quotas (TRQs). Of the 41 products that were under import quota restrictions before WTO accession, 18 were moved to tariffication after accession. Rice received a special exemption, and the remaining 22 items are now governed by the tariff rate quota regime. Similar to the arrangement in Korea, the special treatment of rice in Taiwan, China is based on the rules of annex 5 of the URAA. The quota of rice imports was set in 2002 at 8 percent of the average domestic consumption between 1990 and 1992 (144,720 tons of brown rice). By negotiation, this amount was divided into governmental and private import quotas. The government rice quota (65 percent of rice imports) was subject to the same treatment as rice purchased from local growers. The imported rice cannot be exported for food aid nor can it be used for animal feed. The remaining 35 percent was imported by private firms and allocated on first-come, first-served basis. For both private and government quo- tas, there is a ceiling on the price mark-up of NT$23.26 per kilogram for rice and Japan, Republic of Korea, and Taiwan, China 85 NT$25.59 for rice products when they are sold on the domestic market. If the sale of quota rice is slow, the price mark-up can be cut by NT$3 every two weeks. The mark-up reduction can be continued until all rice is sold. Measurement of Distortions to Agricultural Incentives The main focus of the study presented here is the gap between domestic prices and what they would be under free markets. Since it is not possible to understand the characteristics of agricultural development with a sectoral view alone, the project's methodology not only estimates the effects of direct agricultural policy measures (including any distortions in the foreign exchange market), but also generates estimates of distortions in nonagricultural sectors for comparative evaluation. Specifically, the study computes an NRA for farmers including an adjustment for direct interventions on inputs such as border protection on fertil- izers. It also generates an NRA for nonagricultural tradables, for comparison with that for agricultural tradables, via the calculation of an RRA (see Anderson et al. 2008a, 2008b). For Japan and Korea, an NRA is calculated for rice, wheat, barley, soybeans, beef, pig meat, poultry, eggs, and milk. For Taiwan, China, estimates are calculated for rice, wheat, beef, pig meat, poultry, and eggs. Domestic prices are converted to U.S. dollars using market foreign exchange rates except for 1955­64 in Korea and for 1955­61 in Taiwan, China, for which the shadow exchange rates estimated for Korea by Frank, Kim, and Westphal (1975) and for Taiwan, China by Scott (1979) are used in order to take into account the distortions to the foreign exchange market in early years. Aggregate NRAs on output for each county are calculated using weights based on domestic production of commodities valued at undistorted prices. In addition to the commodities above covered in this study, several other crops are included in the calculation of RRAs for Japan and Korea. These include apples, cabbage, cucumbers, grapes, mandarins, pears, spinach, strawberries, onions, and sugar for Japan and cabbage, red peppers, and garlic for Korea. The estimates for these products come from the Organisation for Economic Co-operation and Development's (OECD's) estimates of producer and consumer support estimates (PSEs and CSEs; see OECD 2008). Data for these crops are available only for years 1986 and later. Distortions of those crops prior to 1986 are assumed to be at the level of 20 percent in Japan and 90 percent in Korea of the NRAs for the available covered products. The percentage of agricultural output covered in this study is between 55 and 70 (valued at undistorted prices), though it is difficult to judge the levels of NRAs for the residual products. It is assumed to be made up of the following share trends (at distorted prices) between 1955 and the present: 50 to 80 percent for 86 Distortions to Agricultural Incentives: A Global Perspective import-competing products and 50 to 20 percent for nontradables in Japan and Korea. Distortions of the residual products are assumed to be zero for nontradables, and the same as that of the 11 (Japan) and 4 (Korea) OECD products for import- competing products. For Taiwan, China, distortions of all the noncovered residual products are assumed to be zero, as most of them are nontradable or exportable. To compute the RRA, the NRA for nonagricultural industries is first estimated. For the latter, weighted tariffs are available only for selected years for Japan, Korea, and Taiwan, China. Data are linearly interpolated for the years for which they are not available. For the early years, tariff rates are estimated as total tariff revenue divided by value of imports. Assuming the exportable industries receive no assis- tance, the weighted average tariff is multiplied by the share of import-competing industries in the value of all nonagricultural tradables. This procedure is likely to underestimate assistance to nonagricultural industries, especially in Korea, where subsidized credits targeting certain industries were the major form of assistance. The estimation results for nominal and relative rates of assistance (NRAs and RRAs) to selected commodities are summarized in five-year averages in tables 2.3 and 2.4 for Japan, Korea, and Taiwan, China.10 Annual movements of the RRA are shown in figure 2.1 to compare protection patterns of Japan, Korea, and Taiwan, China. Japan's RRA was 13 percent in 1955, when the country was still in the lower- middle-income stage of economic development. But it rose quickly as Japan Figure 2.1. RRA to Agricultural Versus Nonagricultural Tradables,a Japan, Korea, and Taiwan, China, 1955­2007 250 200 150 percent 100 50 0 50 100 1955 1959 1963 1967 1971 1975 1979 1983 1987 1991 1995 1999 2003 2007 Japan Taiwan, China Korea Source: Honma and Hayami (2008) a. RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. Table 2.3. NRAs to Selected Agricultural Products in Japan, Korea, and Taiwan, China, 1955­2007 (percent) a. Japan 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Import-competing products 53.7 66.5 79.9 77.8 110.8 111.8 153.1 149.3 147.3 146.5 124.3 Rice 72.5 91.0 122.9 164.9 210.8 267.2 591.6 656.2 535.4 607.0 362.8 Barley 35.7 38.7 16.5 10.1 63.1 88.6 203.6 141.7 129.0 121.5 197.6 Wheat 36.1 39.1 42.4 25.4 76.2 111.8 170.3 200.4 204.2 128.6 269.5 Beef 27.4 68.3 130.8 106.0 215.3 136.7 208.9 177.0 191.8 149.1 39.3 Pig meat 17.9 59.8 12.4 3.4 2.9 12.6 0.5 5.7 10.3 5.5 138.0 Poultry 33.1 42.8 33.5 36.7 31.1 16.8 17.6 25.1 41.4 74.0 11.7 Eggs 3.0 3.3 5.3 2.8 3.3 1.4 19.9 23.2 33.8 27.6 17.1 Milk 44.2 96.2 162.1 165.4 385.7 211.5 365.2 280.3 238.0 273.2 101.0 Apples -- -- -- -- -- -- 32.0 24.1 27.7 31.4 17.3 Cabbage -- -- -- -- -- -- 10.3 31.1 127.5 177.5 204.6 Cucumbers -- -- -- -- -- -- 57.1 17.4 29.8 43.2 31.1 Grapes -- -- -- -- -- -- 87.2 82.2 117.7 177.4 178.6 Mandarins -- -- -- -- -- -- 21.1 44.8 47.3 32.4 46.4 Pears -- -- -- -- -- -- 35.0 24.0 64.2 157.3 128.7 Spinach -- -- -- -- -- -- 89.2 138.0 236.7 134.4 32.3 Strawberries -- -- -- -- -- -- 11.0 25.1 26.5 16.8 7.2 Onions -- -- -- -- -- -- 55.3 80.8 144.4 284.2 294.9 Soybeans -- -- -- -- -- 410.7 259.4 21.3 42.2 67.2 68.5 Sugar -- -- -- -- -- 229.6 198.3 158.4 159.0 154.7 106.6 Exportables n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Total of covered products 53.7 66.5 79.9 77.8 110.8 111.8 153.1 149.3 147.3 146.5 107.4 from domestic measures 0.4 3.1 10.0 9.2 10.9 9.6 8.0 6.0 4.5 4.5 4.0 from border (import) measures 54.1 63.4 69.9 68.6 99.9 102.1 145.2 143.3 142.8 142.0 102.9 Dispersion of covered productsb 39.4 40.3 69.4 82.2 156.1 142.6 175.3 161.5 136.1 142.5 116.0 87 % coverage (at undistorted prices) 69 65 59 55 55 56 69 68 67 67 76 (Table continues on the following pages.) 88 Table 2.3. NRAs to Selected Agricultural Products in Japan, Korea, and Taiwan, China, 1955­2007 (continued) (percent) b. Republic of Korea 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Import-competing productsa 3.9 4.4 16.6 47.6 73.8 122.8 166.7 201.9 182.9 213.6 116.4 Rice 8.2 7.0 5.4 31.3 59.6 118.4 214.4 265.9 294.3 385.9 213.3 Barley 41.2 83.5 72.3 120.3 101.2 165.9 357.0 524.3 543.0 562.8 275.6 Wheat 43.0 26.7 11.2 0.4 26.5 92.2 144.4 216.0 122.8 135.4 -- Beef 38.8 34.4 64.9 73.9 162.6 163.2 126.2 200.8 159.9 167.8 182.3 Pig meat 15.2 21.7 158.7 204.1 202.9 169.1 124.7 149.3 116.2 134.4 103.1 Poultry 11.8 6.9 131.4 103.5 161.7 94.2 86.6 155.6 171.7 179.2 55.7 Eggs 27.1 24.7 23.0 0.1 7.5 14.9 19.4 28.0 26.6 54.3 31.6 Milk -- -- 173.3 108.8 189.0 179.8 185.2 203.7 140.7 149.8 137.0 Cabbage -- -- -- -- -- -- 30.0 30.0 29.1 27.6 27.0 Peppers -- -- -- -- -- -- 175.0 245.4 145.5 197.0 235.7 Soybeans 13.0 18.8 58.8 80.0 122.2 253.0 361.8 508.2 625.6 757.4 729.2 Garlic -- -- -- -- -- -- 250.3 288.8 213.3 122.6 128.1 Exportablesa n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Total of covered products 3.9 4.4 16.6 47.6 73.8 122.8 166.7 201.9 182.9 213.6 147.3 from domestic measures 0.2 0.4 0.9 4.2 7.1 5.3 5.5 5.9 6.1 5.2 4.4 from border (import) measures 3.7 4.7 15.7 43.4 66.7 117.5 161.2 196.0 176.9 208.5 143.0 Dispersion of covered productsb 34.1 40.5 85.0 82.5 89.0 80.1 114.8 164.2 200.1 225.4 206.0 % coverage (at undistorted prices) 48 57 67 65 65 61 60 57 51 46 55 c. Taiwan, China 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­02 2003­07 Exportablesa 23.5 7.5 5.7 20.7 13.4 35.9 89.5 161.4 167.6 203.1 -- Rice 29.6 6.6 17.9 9.4 7.6 32.5 103.3 161.4 167.6 203.1 -- Pig meatc 8.1 64.0 99.7 98.3 60.6 42.6 64.8 n.a. n.a. n.a. -- Import-competing productsa 33.0 5.3 21.7 26.7 32.5 49.1 55.4 93.6 126.3 160.0 -- Wheat 48.2 36.0 39.4 32.2 57.2 92.3 -- -- -- -- -- Beef 13.7 41.2 28.8 22.0 79.6 77.0 101.3 98.5 82.6 72.8 -- Pig meatc n.a. n.a. n.a. n.a. n.a. n.a. n.a. 107.1 131.3 173.2 -- Poultry 47.5 3.7 21.2 27.1 30.0 63.6 84.6 143.0 228.7 279.5 -- Eggsd n.a. n.a. n.a. n.a. n.a. 0.7 26.8 23.9 17.9 24.7 -- Nontradablesa 0.0 0.0 0.0 0.0 0.0 0.0 n.a. n.a. n.a. n.a. -- Eggsd 0.0 0.0 0.0 0.0 0.0 0.0 n.a. n.a. n.a. n.a. -- Total of covered productsa 23.2 7.2 6.2 20.0 14.0 35.1 76.1 109.5 134.0 167.8 -- Dispersion of covered productsb 33.4 35.3 47.5 40.5 40.5 34.5 56.9 66.1 86.9 106.4 -- % coverage (at undistorted prices) 53 49 49 48 50 42 35 34 35 36 -- Source: Anderson and Valenzuela (2008), based on authors' spreadsheet. Note: -- not available; n.a. not applicable. a. Weighted averages, with weights based on the unassisted value of production. b. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of NRAs of covered products. c. Pig meat changed trade status in 1989, from import-competing to exportable. The period average reported here corresponds to 1985­88 for the import-competing product, and 1989­94 for the exportable product. d. Eggs are assumed to be a nontradable product with zero distortions prior to 1983. 89 Table 2.4. NRAs to Agricultural Relative to Nonagricultural Industriese in Japan, Korea, and Taiwan, China, 90 1955­2007 (percent) a. Japan 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, covered products 53.7 66.5 79.9 77.8 110.8 111.8 153.1 149.3 147.3 146.5 107.4 NRA, noncovered products 5.3 6.5 7.7 8.0 12.1 13.0 23.7 26.7 42.0 50.5 18.6 NRA, all agricultural products (excluding NPS) 38.8 45.8 50.4 46.9 65.9 68.3 112.4 110.5 112.8 115.2 70.9 All importables 46.1 55.0 62.1 58.1 81.2 82.1 127.5 124.4 127.6 129.1 124.3 All exportables -- -- -- -- -- -- -- -- -- -- -- All nontradables 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 40.8 Trade bias index (TBI) 0.32 0.35 0.38 0.37 0.45 0.45 0.56 0.55 0.56 0.56 0.55 NRA, non-product-specific (NPS) assistance -- -- -- -- 4.8 4.0 6.4 5.8 6.8 5.1 3.3 Inputs -- -- -- -- 4.8 4.0 6.4 5.8 6.8 5.1 3.3 Other -- -- -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, all agricultural products (including NPS) 38.8 45.8 50.4 46.9 66.8 72.3 118.8 116.3 119.6 120.4 74.3 NRA, decoupled payments 0.0 0.0 0.0 0.0 3.6 15.1 7.1 2.5 2.7 4.6 5.0 NRA, all agricultural products (including NPS and decoupled assistance) 38.8 45.8 50.4 46.9 70.4 87.4 125.9 118.9 122.3 124.9 79.3 NRA, all agricultral tradables (including NPS) 46.1 55.0 62.1 58.1 87.6 86.1 133.8 130.2 134.4 134.2 127.7 NRA, all nonagricultural tradables 2.5 3.9 3.8 2.8 1.6 1.1 1.3 1.1 0.8 0.7 0.6 RRAd 42.5 49.1 56.2 53.7 84.6 84.0 130.9 127.6 132.4 132.7 126.4 b. Republic of Korea 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, covered products 3.9 4.4 16.6 47.6 73.8 122.8 166.7 201.9 182.9 213.6 147.3 NRA, noncovered products 1.7 0.2 7.0 15.3 25.3 37.4 64.3 88.0 74.6 71.7 49.3 NRA, all agricultural products (excluding NPS) 3.2 4.0 13.4 35.7 56.3 89.4 126.1 152.8 129.8 137.3 80.6 All importables 3.3 4.9 16.3 46.1 71.8 118.6 159.8 197.6 164.8 171.9 116.4 All exportables -- -- -- -- -- -- -- -- -- -- -- All nontradables 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 TBI 0.03 0.05 0.14 0.32 0.42 0.54 0.62 0.66 0.62 0.63 0.54 NRA, NPS assistance -- -- -- -- 0.6 0.7 2.2 7.1 7.3 4.4 3.9 Inputs -- -- -- -- 0.6 0.7 2.2 7.1 7.3 4.4 3.9 Other -- -- -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, all agricultural products (including NPS) 3.2 4.0 13.4 35.7 56.4 90.1 128.1 159.8 137.0 141.7 84.4 NRA, decoupled payments 0.0 0.0 0.0 0.0 0.1 0.4 0.5 5.3 2.7 7.2 12.3 NRA, all agricultural products (including NPS and decoupled assistance) 3.2 4.0 13.4 35.7 56.4 90.5 128.6 165.2 139.7 148.8 96.8 NRA, all agricultural tradables (including NPS) 3.3 4.9 16.3 46.1 71.9 119.3 161.7 204.7 171.9 176.3 120.3 NRA, all nonagricultural tradables 45.6 37.1 22.3 11.4 11.7 6.8 5.7 3.3 2.3 1.7 1.5 RRAd 32.6 21.4 4.8 30.5 54.0 105.4 147.8 195.0 165.8 171.6 117.0 (Table continues on the following page.) 91 Table 2.4. NRAs to Agricultural Relative to Nonagricultural Industriese in Japan, Korea, and Taiwan, China, 92 1955­2007 (continued) (percent) c. Taiwan, China 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­02 2003­07 NRA, covered productsa 23.2 7.2 6.2 20.0 14.0 35.1 76.1 109.5 134.0 167.8 -- NRA, noncovered products 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -- NRA, all agricultural productsa 11.8 3.5 3.0 9.2 7.0 14.6 26.4 37.2 45.5 60.0 -- NRA, NPS assistance -- -- -- -- -- -- -- --. -- -- -- Total agricultural NRA (including NPS)b 11.8 3.5 3.0 9.2 7.0 14.6 26.4 37.2 45.5 60.0 -- TBIc 0.15 0.05 0.02 0.12 0.05 0.15 0.27 0.11 0.02 0.00 -- NRA, all agricultural tradables 15.8 4.7 3.9 12.0 8.9 18.5 32.7 45.0 53.6 69.2 -- NRA, all nonagricultural tradables 8.8 9.3 8.8 7.5 7.0 5.2 4.5 2.6 1.8 1.1 -- RRAd 22.5 4.2 4.5 4.2 1.7 12.7 27.0 41.3 51.0 67.3 -- Source: Anderson and Valenzuela (2008), based on authors' spreadsheet. Note: -- = not available. a. NRAs including product-specific input subsidies. b. NRAs including product-specific input subsidies and non-product-specific (NPS) assistance. Total of assistance to primary factors and intermediate inputs divided by total value of primary agriculture production at undistorted prices (percent). c. TBI (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricul- tural sector. d. RRA is defined as 100*[(100 NRAagt ) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. Japan, Republic of Korea, and Taiwan, China 93 became an upper-middle-income country in the 1960s, to 30­40 percent. The RRAs of Korea and Taiwan, China, meanwhile, both in the low-income stage of development in the 1950s and the lower-middle-income stage in the 1960s, were at very low levels, involving negative rates for some years before the mid-1970s. Following Japan's entry into the high-income stage of development in the 1970s, it increased its RRA steadily, except during the global food crises in 1973­74. The RRA reached a peak in 1994, although this year followed a bad rice harvest (one- quarter below average). Japan's RRA was within the 100­150 percent range after the mid-1980s, except in 1994. In Korea, the rapid rise of agricultural assistance began in the late 1970s when the country moved from the lower- to the upper-middle-income stage of develop- ment. Taiwan, China followed Korea with an increase in RRA, though the differ- ence in the level of the RRA between the two economies continued to be significant during the upper-middle-income stage in both countries. It is interesting to see that Taiwan, China was behind Korea in terms of the RRA level until the mid- 1990s. After this time, however, Korea's RRA fell. After Korea and Taiwan, China entered the high-income stage of development in the 1990s, a relatively high RRA was maintained in both economies, albeit with some fluctuations. The wide fluctuations in RRA in the late 1990s were caused by the currency crises in East Asia that began in 1997. This resulted in a sharp decline in the RRA in Korea in 1997 and 1998. Sharp increases in the RRA in Taiwan, China in 1999 and 2000 were caused by shortages of livestock products because of the September 1999 earthquake and reduced production of pig meat resulting from the spread of foot-and-mouth disease in 1997. Although the paths of the RRAs were different during the middle- income stage of development, both Korea and Taiwan, China started at slightly negative protection levels in the low-income stage in the 1960s and reached very high RRAs (about 120 percent in Korea and 70 percent in Taiwan, China) by 2000. Movements of the NRAs for covered farm products were similar to those of the RRAs in all three economies until the late 1970s, when the growth of the NRAs for nonagricultural products was much faster, particularly in Taiwan, China. In con- trast to the path of the RRA, Taiwan, China kept pace with Korea in terms of the growth of the NRA for agriculture, albeit at the level of about 10 years behind that of Korea. Taiwan, China then caught up with Japan and Korea at a 150­180 per- cent NRA in the late 1990s. The three economies all maintained policies to protect covered agricultural products that were considered politically important and sen- sitive. However, the importance of these covered products declined over time with a smaller share of those commodities in the value of production. Thus, the growth in the RRA was less than that in the NRA because the RRA takes into account the nondistorted uncovered products whose share in value of production increased. Fluctuations in the RRA and NRA consist mainly of changes in the NRA of individual commodities and changes in the weight of each commodity. Because 94 Distortions to Agricultural Incentives: A Global Perspective Figure 2.2. NRA to Rice, Japan, Korea, and Taiwan, China, 1955­2007 900 700 500 percent 300 100 100 1955 1959 1963 1967 1971 1975 1979 1983 1987 1991 1995 1999 2003 2007 Japan Taiwan, China Korea Source: Honma and Hayami (2008) rice is the most important agricultural product in Japan, Korea, and Taiwan, China, its protection had a large influence on the RRA. A clear upward trend in the NRA of rice was present in all three economies (figure 2.2). In Japan, the NRA for rice was as high as nearly 100 percent in the 1960s, when Japan had already entered the upper-middle-income stage of development, whereas it was nearly zero in Korea and Taiwan, China in the lower-middle-income stage. From the late 1970s, when Korea and Taiwan, China approached the upper-middle-income stage, the NRA for rice began to rise sharply and continued to rise thereafter. The fastest increase in the NRA for rice occurred in Japan from the late 1970s. It peaked in the late-1980s. The rapid increase was caused, to a large extent, by a rapid appreciation of the Japanese yen relative to the U.S. dollar. Though the border price of rice declined sharply, there was no transmission to domestic market prices because of the control of rice imports by the government. In 1993, a temporary interruption in the peak of the rice NRA in Japan, caused by a bad harvest, resulted in a shortage of Japonica rice on world markets. This raised border prices, while domestic prices were kept relatively stable under the Food Control System. There- after, further increases in the NRA for rice were counteracted by yen depreciation and also by the acceptance of minimum access obligations in the URAA in 1995 and the later shift to tariffication in 1999. In Korea and Taiwan, China, the upward trend in the rice NRA continued after the 1970s. Such increases were a major factor underlying rapid increases in the Japan, Republic of Korea, and Taiwan, China 95 RRA in Korea during the upper-middle-income stage of development, because the weight of rice in agricultural production continued to be high. Korea's exemption from tariffication in the URAA allowed the NRA for rice to grow even under the implementation of WTO commitments. Similar to the NRA for covered products in figure 2.1, Taiwan, China followed Korea in the growth of the NRA for rice. Tai- wan, China mimicked the situation in Korea, though with a 5- to 8-year lag between the 1970s and the mid-1990s. In more recent years, the gap in NRAs for rice between Taiwan, China and Korea has increased, but the protection level of rice in the former appears to be maintaining a rising trend. Consumer tax equivalents on food Though support provided to farmers in Japan, Korea, and Taiwan, China has come mostly via food import restrictions, there have been additional schemes whereby crop producer prices were supported at levels above those charged to grain and soy- bean consumers (including feedmixers providing livestock producers with animal feedstuffs). Thus, the consumer tax equivalent (CTE) is below the NRA for some crop products. As a result, together with the different weights of various products in consumption as compared with production, the average NRA for covered products was approximately 50 percent above the CTE for both Japan and Korea in 2000­04 (compare tables 2.3 and 2.5). In this way, consumers were spared some of the implicit tax that otherwise would have been imposed on them had border measures alone been used to raise producer prices above international levels. Sources of Growth in Agricultural Protection Measures The experiences of Japan, Korea, and Taiwan, China are good examples of policy switching from exploitation to protection of agriculture when economies grow through industrial development. This shift is most clearly illustrated by the cases of Korea and Taiwan, China, whose agricultural protection levels were negative in the 1950s and the 1960s and began the rise sharply in the 1970s with the success of industrial development. The growth of agricultural protection in Japan, Korea, and Taiwan, China is empirically documented in Anderson, Hayami, and Honma (1986), which draws attention to three characteristics of the East Asian growth of agricultural protection, based on the nominal rates of protection for agricultural products, in comparison with other advanced economies: first, the rapid rise over time in pro- tection rates in the three economies in East Asia; second, the faster increase in agricultural protection in the three East Asian economies than in other industrial economies for the period of 1955 to 1980; and third, the fact that the highest level of agricultural protection reached by the three economies as of 1980 was rivaled 96 Table 2.5. CTEs for Selected Agricultural Products in Japan, Korea, and Taiwan, China, 1955­2007 (percent) a. Japan 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Apples -- -- -- -- -- -- 30.8 23.8 27.5 31.0 17.0 Barley 34.9 32.6 10.2 5.8 20.3 29.1 158.1 131.9 118.9 105.9 114.3 Beef 27.4 68.3 130.8 106.0 215.3 136.7 208.9 177.0 191.8 149.1 38.5 Cabbage -- -- -- -- -- -- 9.2 30.8 127.0 176.6 204.0 Cucumbers -- -- -- -- -- -- 56.4 17.2 29.7 42.9 30.9 Eggs 3.0 3.3 5.3 2.8 3.3 1.4 19.7 22.9 33.6 27.2 17.0 Grapes -- -- -- -- -- -- 85.3 81.6 117.2 176.3 177.8 Mandarins -- -- -- -- -- -- 20.0 44.5 47.0 32.0 46.0 Milk 44.2 96.2 162.1 165.4 385.7 211.5 365.2 280.3 238.0 273.2 93.5 Onions -- -- -- -- -- -- 54.0 80.4 143.9 282.9 294.0 Pears -- -- -- -- -- -- 35.0 24.0 64.2 157.3 103.4 Pig meat 17.9 59.8 12.4 3.4 2.9 12.6 0.5 5.7 10.3 5.5 138.0 Poultry 33.1 42.8 33.5 36.7 31.1 16.8 17.6 25.1 41.4 74.0 11.7 Rice 73.1 85.7 103.9 137.3 179.6 232.6 548.5 613.2 506.4 574.6 348.6 Soybeans -- -- -- -- -- 0.0 0.0 0.0 0.0 0.0 0.0 Spinach -- -- -- -- -- -- 89.2 138.0 236.7 134.4 32.3 Strawberries -- -- -- -- -- -- 10.0 24.8 26.2 16.4 7.0 Sugar -- -- -- -- -- 167.0 185.9 154.8 155.4 151.6 119.6 Wheat 35.9 33.0 27.4 14.1 26.7 37.2 136.1 108.4 73.4 68.7 73.6 All covered products 53.0 62.2 66.8 67.6 93.2 98.8 134.9 119.3 116.1 106.6 81.0 Import-competing 53.0 62.2 66.8 67.6 93.2 98.8 134.9 119.3 116.1 106.6 86.9 Exportables -- -- -- -- -- -- -- -- -- -- -- Dispersion, covered productsb 39.1 39.6 66.5 75.0 144.0 97.2 154.5 149.3 130.0 141.8 108.3 b. Republic of Korea 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Wheat 46.2 22.5 11.1 1.1 16.2 46.0 132.8 167.4 80.5 80.6 -- Barley 40.8 77.8 64.9 96.6 57.3 119.6 325.6 411.5 341.2 327.5 174.4 Rice 7.7 5.5 5.0 29.1 54.5 113.4 211.5 261.7 290.8 385.3 213.3 Beef 38.8 34.4 64.9 73.9 162.6 163.2 122.1 200.7 153.9 167.7 182.3 Pig meat 15.2 21.7 158.7 204.1 202.9 169.1 124.7 149.3 116.2 134.4 103.1 Poultry 11.8 6.9 131.4 103.5 161.7 94.2 86.6 155.6 171.7 179.2 55.7 Eggs 27.1 24.7 23.0 0.1 7.5 14.9 19.4 28.0 26.6 54.3 31.6 Milk -- -- 173.3 108.8 189.0 179.8 185.2 203.7 140.7 149.8 137.0 Cabbage -- -- -- -- -- -- 30.0 30.0 29.1 27.6 27.0 Peppers -- -- -- -- -- -- 175.0 245.4 145.5 197.0 235.7 Soybeans 19.8 8.2 51.6 63.2 95.2 245.4 112.2 75.5 63.6 66.8 91.9 Garlic -- -- -- -- -- -- 250.3 288.8 213.3 122.6 128.1 All covered products 5.0 5.4 14.5 39.7 63.9 114.3 148.5 176.4 144.9 154.1 135.1 Import-competing 5.0 5.4 14.5 39.7 63.9 114.3 148.5 176.4 144.9 154.1 115.7 Exportables -- -- -- -- -- -- -- -- -- -- -- Dispersion, covered productsb 34.7 37.6 85.3 81.1 92.3 82.3 95.1 118.1 107.3 116.2 81.7 (Table continues on the following pages.) 97 98 Table 2.5. CTEs for Selected Agricultural Products in Japan, Korea, and Taiwan, China, 1955­2007 (continued) (percent) c. Taiwan, China 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­02 2003­07 Rice 29.6 6.6 17.9 9.4 7.6 32.5 103.3 161.4 167.6 203.1 -- Wheat 38.3 16.4 29.6 14.6 1.6 0.3 -- -- -- -- -- Beef 13.7 41.2 28.8 22.0 79.6 77.0 101.3 98.5 82.6 72.8 -- Pig meat 8.1 64.0 99.7 98.3 60.6 42.6 76.5 103.9 131.3 173.2 -- Poultry 47.5 3.7 21.2 27.1 30.0 63.6 84.6 143.0 228.7 279.5 -- Eggs 0.0 0.0 0.0 0.0 0.0 0.7 26.8 23.9 17.9 24.7 -- All covered productsa 21.1 7.7 6.9 19.0 15.2 38.4 82.7 116.5 136.8 166.5 -- Import-competing 6.1 13.2 26.0 23.3 27.3 49.0 74.7 102.1 129.1 159.5 -- Exportables 23.7 7.0 5.2 19.1 13.7 36.2 89.6 161.4 167.6 203.1 -- Dispersion, covered productsb 33.3 35.9 47.2 40.6 40.2 32.2 34.1 56.2 87.1 106.0 -- Source: Anderson and Valenzuela (2008), based on authors' spreadsheet. Note: -- = data not available. a. Weighted averages, with weights based on the unassisted value of consumption, where consumption is derived using the value of production and self-sufficiency ratios (derived from the FAOSTAT Database) as production/consumption. b. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of CTEs of covered products. Japan, Republic of Korea, and Taiwan, China 99 only by Switzerland. Anderson, Hayami, and Honma also note that the growth of agricultural protection in these economies during the three decades to 1980 was exceptionally rapid compared with Western countries that started down the path of industrialization earlier. That is, East Asia was not exceptional in having increasing levels of agricultural protection, but it was exceptional in the speed at which it reached the world's highest level. Figures 2.1 and 2.2 indicate that the protection growth in terms of RRA and NRA continued at the same speed for about 20 years after the previous study period. The rapid growth of agricultural protection in industrializing economies can be explained largely by the shift in comparative advantage away from agriculture to industry as the result of successful industrialization. The decline in agriculture's comparative advantage increased the intersectoral resource adjustment costs that would have been shouldered by farmers if they were left to the competitive market mechanism. Instead, the increase in costs boosted farmers' demand for agricul- tural protection. This problem typically applies when industrial growth has been so rapid that intersectoral adjustments are not fast enough under free markets to prevent a widening rural-urban income disparity.11 The association between the rise in agricultural protection and the decline in agriculture's comparative advantage was tested in Honma and Hayami (1986) using multiple regression analysis and a pooled data set for 15 countries at six points in time ending in 1980. A strong correlation was found between the level of aggregate NRP12 and the index of agriculture's labor productivity relative to the total economy's labor productivity. Honma and Hayami conclude that the high level of agricultural protection in East Asia resulted not so much from factors unique to East Asia but mainly from factors common to all industrial countries. It should be noted, however, that there are differences in the process of intersectoral resource adjustment between Japan and the other two economies. In 1955, the first year of the analysis presented here, Japan was already in the middle-income stage of economic development and about to enter the so-called high-growth era character- ized by extremely rapid industrialization and a widening income gap between rural and urban households.13 Soon after shifting from the middle-income to the high- income stage of development, a process that took less than two decades, Japan increased its level of agricultural protection. Protection measures were raised to a level comparable with that of the European Economic Community during the 1960s. Meanwhile, Korea and Taiwan, China were still in the low-income stage of eco- nomic development in the 1950s. As both economies entered the middle-income stage in the 1960s, productivity growth in agriculture lagged behind that of non- agriculture as a result of successful industrialization. With delays in labor out- migration from farming, farmers' income levels tended to decline relative to those of nonfarmers. Nevertheless, because agriculture was such a large sector in terms 100 Distortions to Agricultural Incentives: A Global Perspective of both national income and labor force, it was impossible during the middle- income stage of development for governments to secure sufficient finance from nonagricultural sectors to raise support for farmers to the extent needed to close the income gap. Thus, despite the growing rural-urban income disparity, Korea and Taiwan, China retained low levels of agricultural protection until the late 1970s and early 1980s, respectively. The agricultural problem confronted by middle-income economies like Korea and Taiwan, China in the 1960s and 1970s has been called the "disparity problem" by Hayami (2005) and Hayami and Godo (2004), referring to the income dispar- ity between farm and nonfarm households. The problem is a lag in productivity growth in agriculture relative to nonagriculture, brought about by insufficient labor out-migration from farming in response to the successful industrialization that raised these economies to the middle-income stage. Farmers, who observe nonfarm workers' rapid escape from poverty, begin to realize how relatively poor they are, even if their income level did not decrease from the previous stage. The resulting dissatisfaction among farmers often becomes a significant source of social instability. Once an economy reaches the middle-income stage, that dissat- isfaction becomes a prime concern of policy makers, who might adopt agricul- tural protection measures to appease farmers and prevent the dissatisfaction from elevating into serious antigovernmental sentiment. Such protection may not be strong enough to close the income gap between farmers and urban workers until an economy graduates from the lower-middle- income stage, however. Because the share of agriculture in both national income and the labor force is still large, it is difficult to either raise sufficient revenue from the nonfarm sectors to close the growing farm-nonfarm income gap with direct support payments or to pass on the cost of agricultural protection to consumers by raising food import barriers, as increases in food prices erode real wages paid by the large number of small-scale enterprises that rely heavily on cheap labor. Faced with the disparity problem, policy makers in middle-income countries are forced to search for ways to protect farmers within the constraint of the food problem that is still binding because a large number of urban workers are still poor and thus still dedicate a large share of household expenditure to food. In the early 1990s, during which all three Northeast Asian economies were in the high-income stage, the decline in relative agricultural income (in terms of agricultural GDP per worker divided by total GDP per worker) stopped in Japan and Korea. In Taiwan, China, the relative agricultural income continued to decline until recently (table 2.1), despite the high level of agricultural protection. The rea- son relative agricultural income continued to decline in Taiwan, China was that the total economy's labor productivity increased more rapidly than agriculture's labor productivity even after 1990. Japan, Republic of Korea, and Taiwan, China 101 Figure 2.3. RRA to Agriculture and Real GDP Per Capita, Japan, Korea, and Taiwan, China, 1955­2004 250 200 150 RRA (%) 100 50 0 50 0 5,000 10,000 15,000 20,000 25,000 30,000 real GDP per capita at 2000 constant prices ($) Japan Korea Taiwan, China Source: Authors' computations. Agricultural protection in Korea rose faster and to a higher level during its upper-middle-income stage of development than in Taiwan, China or Japan during theirs. RRAs in Korea, for example, significantly exceeded those of Taiwan, China and Japan for the same levels of per capita income throughout their upper- middle-income stage (figure 2.3). The difference could reflect the different costs of intersectoral adjustment (corresponding to changes in comparative advantage) that farmers had to bear. In Korea, the shift of labor from agriculture to nonagricul- ture involved the migration of workers from rural to urban areas, whereas in Japan and Taiwan, China farmers increased their nonfarm activities while continuing to live in their home villages and towns and farming part-time. Correspondingly, both the pecuniary and psychological costs of intersectoral labor reallocation were much higher for farmers in Korea. In Japan, the decline in relative agricultural income ceased in the 1970s when the country reached the high-income mark. This was due to a deceleration in the growth of labor productivity in the total economy after reaching the high-income stage. The Korean experience after 1990, mean- while, is likely the result of fast increases in agricultural labor productivity resulting from the rapid out-migration of agricultural labor to urban activities (table 2.1). 102 Distortions to Agricultural Incentives: A Global Perspective Figure 2.4. RRA to Agriculture and Relative GDP Per Agricultural Worker, Japan, Korea, and Taiwan, China, 1955­2004 250 200 150 RRA (%) 100 50 0 50 0 10 20 30 40 50 60 70 ratio of agricultural GDP per worker to total GDP per worker (%) Japan Korea Taiwan, China Source: Authors' computations. The relationship between relative agricultural income and the RRA in Japan, Korea, and Taiwan, China for 1955­2004 is shown in figure 2.4.14 Except for Korea in 1990 and 2000, there is a negative correlation for all countries and all periods. The correlation is weak, however, when relative agricultural income is more than about 40 percent (a rate that corresponds to the low-income and lower-middle- income stages of economic development). Upon reaching the upper-middle-income stage of development in the 1980s, Korea and Taiwan, China strengthened their agricultural protection policies. This followed Japan's protection pattern in the 1960s, when the income gap was widening and protection measures were deemed necessary to close it. Under such circumstances, politicians were not able to resist pressure from the farm lobby to institute policies to prevent farmers' incomes from lagging behind those of non- farm workers. In the case of Korea, policy makers may have had a specific reason for strengthening agricultural protection, particularly at the farmgate level--the constant threat of communist aggression from the North. The imminent hostility prompted commercial and industrial interests to support farmers and thereby maintain political stability. If the income gap had been adequately dealt with during the middle-income stage of development, problems caused by agricultural protection in the upper- middle-income stage might have been avoided. Yet the disparity problem in the Japan, Republic of Korea, and Taiwan, China 103 middle-income stage has received relatively little attention in academic and policy debate, despite the fact that there are many economies attempting to reach the upper-middle-income and high-income stage through industrialization. The growing income disparity between farm and nonfarm populations could become a major source of social and political instability elsewhere in Asia, from ASEAN countries to China and eventually to South Asia, particularly India. In analysis by Honma and Hayami (1986), it was found that political power in the agricultural sector is maximized when the share of agriculture declines to 4 to 5 percent of GDP or 5 to 8 percent of the labor force. Japan has passed this range, and Korea recently entered this peak zone in terms of both GDP and labor force, as did Taiwan, China in terms of the labor share (having passed over this zone in 1990 in terms of GDP share). Political economy factors may well underlie the rise of agricultural protectionism in Korea at the high-income stage after 1990, as observed in terms of the NRA at the farmgate level despite no apparent further increase in its agricultural comparative disadvantage.15 Japan's Pre-World War II Experience The pattern of growth of agricultural protection in Japan, Korea, and Taiwan, China in the era of East Asian economic miracle, as outlined in the previous sec- tion, was consistent with the hypothesis that rapid growth in protection occurred when these economies were in the middle-income stage under the dictate of the "disparity problem" described earlier. Under the disparity problem, when farm- ers' income levels decline relative to those of nonfarmers, the economy is often characterized by a dual structure: a formal sector consisting of large, modern enterprises and government agencies, and an informal sector consisting of agri- culture and other small and medium enterprises. That was the case for Japan in the half-century before World War II, during which Japan advanced from the low- income to the middle-income stage of economic development. Japan set upon a modern economic growth path with the transformation in political structure that occurred under the Meiji restoration of 1868. In short, Japan went from a union of feudal fiefs under the hegemony of Tokugawa shogun (tycoon) to a modern nation state in the form of the constitutional monarchy under an emperor who was a symbol of national unification without actual ruling power. The immediate impetus for this political reorganization was the threat of colonization by Western powers that became obvious through the gunboat diplo- macy of the United States and the use of Admiral Perry's fleet. The national slogan of the Meiji state was to establish fukoku kyouhei (a wealthy nation and strong army) for the sake of preserving national independence. To achieve this goal, eco- nomic policies in Meiji Japan were aimed at the promotion of modern industries 104 Distortions to Agricultural Incentives: A Global Perspective to catch up to the economic power of Western nations. At the time, Japan did not have the freedom to set import and export duties above 5 percent ad valorem lev- els, as per commercial treaties signed by the Tokugawa tycoon with Western pow- ers in the mid-19th century. Thus, industrial promotion policy relied mainly on subsidies in areas such as the import of machines and factories and the purchase of their designs, the employment of engineers and skilled workers, and the collec- tion and dissemination of information on overseas technologies and markets. It was mainly through taxation of agriculture--through the newly established land tax system--that subsidies for industrial promotion and other modernization measures were financed. As the data in table 2.6 show, in the early Meiji period (before 1900), the agri- cultural sector shouldered about 90 percent of the total direct tax burden, which amounted to about 15 percent of agricultural GDP. Agriculture's share of the gov- ernment's subsidies at the time, however, amounted to less than one-quarter, or less than 5 percent of agricultural GDP. Evidently, at the beginning of its modern economic growth period, Japan promoted its modern sectors through the exploitation of the traditional sectors, a strategy commonly practiced by develop- ing economies upon becoming independent of colonial powers in the 1960s. Tax- ation and subsidization imbalanced between agriculture and nonagriculture was even greater than the data of table 2.6 reveal, as a disproportionately high share of the population educated at publicly financed schools were from nonfarm house- holds. Under strong promotion by the government, industrialization progressed rapidly in Japan, especially in labor-intensive manufacturing. Comparative advantage in this sector was unhampered owing to a virtual free trade scenario created by the absence of tariff autonomy in Japan. Changes in the position of agriculture in the total economy over the course of modern economic development under the Meiji restoration are summarized in table 2.7. The real GDP per capita series shows that the Japanese economy moved from the low-income to the middle-income stage by the beginning of the 20th cen- tury, with the share of agriculture in GDP at about 40 percent. This is roughly com- parable to the share of agriculture in GDP in Korea and Taiwan, China when they advanced to the middle-income stage. Meanwhile, the growth of labor productiv- ity in agriculture lagged behind that of industry, resulting in a continual decline in the ratio of labor productivity in agriculture to labor productivity in industry (table 2.7, column 6). This decline reflects successful industrial development, as in Korea and Taiwan, China during the era of East Asia's economic miracle. Neverthe- less, because the terms of trade did not improve for agriculture throughout the inter-war period (table 2.7, column 7), income per capita in farm households con- tinued to decline relative to that of nonfarm employees' households, parallel to the decline in agriculture/industry real productivity ratio (table 2.7, column 8). Japan, Republic of Korea, and Taiwan, China 105 Table 2.6. Changes in Direct Tax Burdens and the Allocations of National Government Subsidies to Agricultural and Nonagricultural Sectors, Japan, 1878­1937 a. Tax burdens Direct tax burdena Direct tax rateb Agriculture Nonagriculture Agriculture Nonagriculture (yen (% of (yen (% of millions) total) millions) total) (%) (%) 1878­82 63.6 91 6.3 9 -- -- 1888­92 58.5 86 9.8 14 14.9 2.0 1898­02 99.1 74 35.4 26 11.7 2.7 1908­12 153.4 54 132.2 46 11.2 5.5 1918­22 295.7 41 431.1 59 7.5 4.8 1928­32 205.5 33 421.3 67 8.1 3.8 1933­37 197.3 26 559.2 74 6.5 4.0 b. Subsidy allocations Subsidy receiptc Subsidy rated Agriculture Nonagriculture Agriculture Nonagriculture (yen (% of (yen (% of millions) total) millions) total) (%) (%) 1881 0 0 0.7 100 -- -- 1891 0 0 2.5 100 0 0.49 1901 0.4 2 18.7 98 0.05 1.41 1911 0.3 1 27.8 99 0.02 1.09 1921 0.6 1 51.8 99 0.02 0.55 1931 21.4 17 101.5 83 1.17 1.11 1934 28.3 28 71.0 72 1.14 0.58 Source: Tobata and Ohkawa (1956) for tax and subsidy data; Ohkawa and Shinohara (1979) for sectoral net domestic product (NDP) data. Note: -- not available. a. Includes both national tax and local rates. b. Direct tax burden divided by sectoral NDP. c. National government subsidies. d. Subsidy receipt divided by sectoral NDP. Growing dissatisfaction among farmers in Japan gave rise to strong political lobbying--organized by the politically powerful landlords--for reduced tax bur- dens and increased support to agriculture. The result was a significant reduction in the tax burden and a greater allocation of government subsidies to agriculture 106 Table 2.7. Farm-Nonfarm Income Disparity in Japan's Economic Development, 1885­2000 Agriculture/ GDP Share of Nominal Tariff Average Agriculture/ manufacturing Farm/nonfarm per capita agriculture rate of rate tariff rate, industry labor terms of household (2000 US$, in GDP protection on rice all products productivity trade Income ratio PPP)a (%)b for rice (%)c (%)d (%)e ratio (%)f (1885=100)g (%)h 1885 1,092 45 15 -- -- 75 100 76 1890 1,285 48 34 -- -- 67 115 87 1900 1,498 39 21 -- 3.7 49 102 52 1910 1,656 32 35 14 16.2 37 98 47 1920 2,154 30 16 10 10.7 50 99 48 1930 2,350 18 57 14 22.6 31 104 32 1935 2,693 18 134 41 23.8 24 136 38 1955 3,519 21 49 -- 3.5 55 163 77 1960 5,063 13 98 -- 6.5 39 169 70 1970 12,337 6 150 -- 6.9 25 303 94 1980 17,056 4 205 -- 2.5 25 342 116 1990 23,580 2 481 -- 2.7 26 379 115 2000 26,220 1 560 778 2.1 22 347 101 Source: See notes below. a. GDP per capita at PPP for 2000 from World Bank (2006), linked with the series from OECD (2003). b. The share of agriculture in nominal GDP for 1885­1935 and share in NNP for 1885­1935 are from Ohkawa and Shinohara (1979). Data for 1960­2000 are from World Bank (2006). c. Nominal rates of protection for rice for 1885­1960 are calculated by the difference between the domestic wholesale price of rice and the unit value of imported rice as percentage of the latter. For 1970­2000, it is calculated by the difference between the domestic wholesale price of rice and unit value of world rice imports multiplied by 1.18, expressed as a percentage of the latter. Data are from Kayo (1977) and Bank of Japan, Yearbook of Wholesale Price Indexes, various years for domestic wholesale price of rice, and Nihon Boeki Seiran, Toyo Keizai Shinposha, 1935, Yearbook of Japan Foreign Trade Statistics, Japan Tariff Association, and FAOSTAT, FAO for border prices. d. Tariffs for 1910, 1920, 1930 and 1935 are tariffs in 1908, 1918, 1928 and 1933, respectively, from Ohkawa, Shinohara and Umenura (1967). Tariff rate for 2000 is ad valorem tariff equivalent of specific duty, 341 yen per kilogram, which was reported to the WTO by the government of Japan. e. Tariffs for 1900, 1910, 1920, 1930, and 1935 are tariffs in 1898, 1908, 1918, 1928, and 1933, respectively, from Ohkawa, Shinohara, and Umenura (1967). Tariffs for 1955­ 2000 are average tariffs calculated by total tariff revenue as percentage of total import cif value in the Ministry of Finance, Monthly Report of Financial Statistics, various issues. f. The ratio of real GDP per worker in agriculture (including forestry and fishery) to real GDP per worker in industry (including mining). 1885­1970 from Hayami (1986). 1980­ 2000 values are extended from 1970 using real GDP and the numbers of employed persons from the Annual Reports of National Accounts. g. For 1985­1960, the ratio between the price index of agricultural products and the price index of manufacturing products in Ohkawa, Shinohara, and Umenura (1967). Values for 1970­80 are extended from 1960 using the Ministry of Agriculture, Forestry and Fishery's price index of agricultural products and the Bank of Japan's domestic corporate goods price index for manufacturing industry products. h. For 1885­1935, the ratio in household income per household member between farm and nonfarm households in Otsuki and Takamatsu (1982). Values for 1955­2000 are the ratio in per capita income between farm households and employees' households based on the Ministry of Agriculture's Farm Household Economy Survey and the Ministry of Internal Affairs' National Survey of Family Income and Expenditure. Farm households in 1990­2000 exclude noncommercial farm households. 107 108 Distortions to Agricultural Incentives: A Global Perspective in the first half of the 20th century. Before the beginning of the 20th century, land- lords were largely satisfied by the government's support to agricultural research and extension services and land infrastructure improvements such as irrigation and drainage systems, which proved to be highly effective in raising rice yields per hectare and thereby raising land prices and land rents for the benefits of landlords (Hayami and Yamada 1991). However, as rapid industrial development con- tributed to a continued decline in the comparative advantage in agriculture, landlords' demands began to shift toward border protection on agricultural com- modities, especially rice. Their strong lobbying achieved the implementation of a rice tariff at 15 percent ad valorem in the first year of the Russo-Japanese War (1904­05), a tariff that was justified as a way to raise government revenue to finance the war. Despite the initial intention to terminate the tariff at the end of the war, landed interests lobbied extensively to make it permanent in the form of a specific duty. Thereafter, the rice tariff became an issue of a major public contro- versy in Japan, similar to the controversy caused by the Corn Laws in the United Kingdom a century earlier and German grain tariffs half a century later. The imperial Agricultural Society, representing the landed interests, and the Tokyo Chamber of Commerce, representing the interests of manufacturers and traders of export commodities, lobbied strongly for opposite ends. The battle ended with a victory for the landed interests and the successful imposition of a specific duty on rice at 1 yen per 60 kilograms. This outcome contrasts with the victory of the bourgeoisie in the United Kingdom and the repeal of the Corn Laws in 1846, and resembles the situation in Germany in which tariff protection was installed on food grains (wheat and rye) in 1879 under Bismarck. But unlike the United Kingdom, which was able to estab- lish itself as the workshop of the world, the comparative advantage of industry was less certain in Germany, so that industrialists found it advantageous to seek protection on their products while approving some protection on agriculture. In addition, the rapid growth of the Social Democratic Party in Germany, a labor party initially based on orthodox Marxist doctrine, was considered a common menace by the Junkers (landed nobility in Prussia and Eastern Germany) and the industrialists in Western Germany. In fact, the simultaneous implementation of grain tariffs and iron and steel tariffs in Germany was the result of a united cam- paign of landlords in Eastern Germany and Western Germany (Gerschenkron 1943). This experience was repeated in other countries that came late to industrial- ization, such as France and Italy, in their attempts to match the United Kingdom in industrial strength (Kindleberger 1951). Japan's protectionist policies were similar: after tariff autonomy was recovered in 1911, Japanese industrialists actively lobbied for industrial protection, especially in heavy and chemical industries. They also Japan, Republic of Korea, and Taiwan, China 109 campaigned for reductions in tariffs on imports of industrial raw materials such as raw cotton and iron ore (Little, Scitovsky, and Scott 1970; Yamazawa 1984). As a result, Japan saw the emergence of tariff escalation, with lower rates applied to materials for industrial processing and higher rates applied to its imports of processed final products. Although agricultural production was raised by means of increases in food tariff rates, this was largely paralleled by increases in industrial tariff rates, which can be inferred from the movements in the average tariff rate for all products compared with the movements in just the rice tariff rate (table 2.7, columns 4 and 5). However, tariffs were largely exempt on the imports of raw materials for industrial production, so effective rates of industrial protec- tion were much higher than the nominal rates implied by the tariff rates. In par- ticular, a zero tariff on raw cotton was instrumental in making the cotton spinning industry the top foreign exchange earner in Japan and, at the same time, com- pletely eradicated domestic cotton farming. Although the rice tariff was raised successively from 14 percent ad valorem in 1910 to 41 percent in 1935, which with quantitative import restrictions increased the nominal rate of protection on rice from 21 percent in 1900 to 134 percent in 1935, improvement in the terms of trade for agriculture was slower than the decline in the agriculture/industry labor productivity ratio. This resulted in the continual decline of farmers' household income per capita relative to that of non- farmers' throughout the interwar period (table 2.7). Although agricultural protec- tion began to increase significantly during this period, it was evidently insufficient to counteract the loss of agriculture's comparative advantage owing to rapid industrial development. To undertake agricultural protection at a scale sufficient to close the widening income disparity, Japan waited until after World War II, when the Japanese economy advanced to the upper-middle-income and the high- income stages of development, so that nonagricultural sectors could bear the cost of agricultural protection. In the early 20th century, the share of food in house- hold consumption expenditure (the Engel coefficient) was higher than 60 percent. This implies that the elevation of food prices had a large effect on the cost of living and, hence, on the wage rate of workers, which caused serious damage to labor- intensive industries, which were then at the center of the Japanese economy. Indeed, as mentioned earlier, the fear that high food prices would damage indus- trial development was a major motive in the Japanese government's decision to launch rice development programs in Korea and Taiwan, China following the Rice Riots in 1918. These programs were launched despite opposition from landlords in Japan to policies fostering competition to Japanese rice farming. The importa- tion of Japonica rice free of tariffs from the two colonies became a major factor aggravating agricultural depression in Japan during the 1930s. 110 Distortions to Agricultural Incentives: A Global Perspective The situation changed dramatically after Japan advanced to the upper-middle- income stage of development in the 1960s. Although supports on agricultural product prices were raised rapidly, industrial wage rates were raised even faster, so that the Engel coefficient fell from 52 percent in 1955 to 31 percent in 1980 and further to 17 percent in 1995 (Hayami and Godo 2002). Meanwhile, the center of gravity in Japanese industry moved from labor-intensive manufacturing to capi- tal- and knowledge-intensive activities. In this environment, Japanese industrial- ists were able to tolerate increases in food prices so as to prevent farm-nonfarm income disparity from widening. Industrialists found it was to their advantage to support farmers and to keep them as allies against organized labor and left-wing activities under the Cold War regime (similar to the attitude of German industri- alists toward the grain tariff campaign a century earlier). The major surge of Japan's agricultural protectionism continued until it was counteracted by the seri- ous trade frictions with food-exporting countries, particularly the United States. What Have We Learned? This chapter's estimates of the NRAs for selected individual commodities and the RRA between agricultural and industry show that the growth of agricultural pro- tection in Northeast Asia, together with the decline of industrial protection rates, caused the RRA to rise in the three economies over the five postwar decades under investigation. The experience in Japan, Korea, and Taiwan, China can be explained by factors common to rapidly industrializing economies, especially the high cost of industrial adjustment shouldered by farm producers in the process of rapid industrial development. However, the agricultural protection level continued to grow even after 1980 in all the three economies despite apparently decreased need for agricultural support to prevent widening rural-urban income disparity. All three economies suffered a problem common in the high-income stage of economic development, notably a widening income gap between agricultural and nonagricultural sectors upon reaching middle-income status. If the income gap had been dealt with more appropriately at the middle-income stage, problems caused by agricultural protection in the high-income stage could have been signif- icantly less. All of this suggests that greater attention needs to be paid to the agricultural problem in the middle-income stage, the so-called "disparity problem." The chal- lenge at that stage of development is to find a compromise between the conflicting need, on one hand, to reduce the farm-nonfarm income gap, and on the other, to supply low-cost food to a large number of workers in urban areas when the gov- ernment's capacity to raise sufficient revenue from nonagricultural sectors is weak and when food import restrictions effectively tax net buyers of food. The Japan, Republic of Korea, and Taiwan, China 111 somewhat contrasting patterns of agricultural and industrial growth between Korea and Taiwan, China led to different solutions to that challenge, which may provide insights for some later-developing economies. Notes 1. On the dramatic transformation of Korea and Taiwan, China from slow, inward-looking eco- nomic growth to rapid, export-led economic growth, see, for example, Mason et al. (1980) and Tsiang (1980). 2. ASEAN countries include Brunei Darussalam, Cambodia, Indonesia, the Lao People's Demo- cratic Republic, Malaysia, Myanmar, the Philippines, Singapore, Thailand, and Vietnam. 3. For the perspective presented here on such a question, see Hayami (2005, section 8.4) and Aoki and Hayami (1998). 4. In terms of real GDP per capita in 2000 constant prices, for example, China exceeded US$1,500 in 1990 and US$5,000 in 2004, whereas Thailand passed the US$1,500 level in 1968 and the US$5,000 level in 1991. Among high-income countries, the United Kingdom and France exceeded US$10,000 per capita in 1960 and 1964, respectively, whereas the United States had sur- passed US$10,000 in 1950. 5. For an explanation as to why farm size has remained so small in Japan, see Godo (2007), who argues that farmers have far less interest in selling to neighboring farmers than in keeping their farm- land in the hope of making huge capital gains when the opportunity arises for their farmland to be converted to nonagricultural uses such as shopping centers. While in principle there are strict regula- tions on farmland use, in practice those laws can be manipulated to increase the probability of conver- sion, by pressuring local authorities and politicians. 6. This section draws heavily on Hayami (1988). 7. This section draws heavily on Moon and Kang (1989). 8. This drove a small wedge between the NRA for producers and the consumer tax equivalent for beef (Anderson 1986), similar to what occurred under the scheme operating in Japan in the 1970s. On why the government chose that scheme rather than a more efficient equally protective tariff plus a con- sumer subsidy funded by the tariff revenue, see Hayami (1979) and Anderson (1983). 9. This section draws heavily on Mao and Schive (1995). 10. The NRAs for commodities are different from those estimated by OECD. Major differences between the NRA study presented here and the OECD study for PSE are twofold: (1) domestic prices in the present study are wholesale prices, whereas the OECD uses farmgate prices for PSE and prices paid by consumers at the farmgate level for CSE; and (2) border prices in the calculations presented here are based on the study in Anderson and Hayami (1986), whereas the OECD uses a different set of reference prices. The fact that the producer price was often above the wholesale (consumer) price in the case of grains and soybeans in Japan and Korea is captured by setting the NRA equal to the measured CTE times the ratio NRA/CTE in Anderson, Hayami, et al. (1986) and Anderson (1989) for the period to 1985 and times the negative of the ratio PSE/CSE in OECD (2007) for the period from 1986. Most differences in NRA between the measures in the OECD and the present studies come from the differences in border prices. For example, the border price of rice is common for Japan and Korea as the world import unit value adjusted by a quality coefficient. But the OECD's border price of rice in Japan is based on the price of rice imports by Japan, while the price in Korea is China's export price of rice adjusted by transportation costs and, from 2001, average import prices of rice from China, Thailand, and the United States. The methodology in the present study makes the series of NRAs for rice more stable than that of the OECD for recent years. The discrepancy also explains the stability of the NRA for Korean rice presented here compared with the NRA in the OECD study. The border prices are also different for meat products. In the estimation of NRAs for beef, pork, and chicken, the OECD basically uses the meat data of Canada or the United States for border prices, while the study here uses Japan's import price for beef and unit values for pork and chicken (or Hong Kong, China import 112 Distortions to Agricultural Incentives: A Global Perspective prices for the 1950s). Our approach is preferred for estimating NRAs consistently for longer time peri- ods, particularly for the period when Korean imports were absent or negligible. Also, the approach here is preferred for comparing the NRAs of Korea and Taiwan, China for border prices. 11. For more on agricultural policies in the process of economic development, see Hayami (2005) and Hayami and Godo (2004). 12. Actual data used for the level of agricultural protection in the regression analysis are the nomi- nal protection coefficients (NPC 1 NRP 100). 13. Indeed, Japan's tariff protection for rice began in 1904; after 1918, it includes its colonies of Korea and Taiwan, China in what became an imperial rice self-sufficiency policy. An earlier set of esti- mates of the nominal rate of rice protection suggests it grew from 9 percent in 1903­07 to 21 percent in 1908­12, and to 27 percent in 1913­17. It then fell to an average of 13 percent in 1918­27 with the greater inflow of rice from the colonies before rising again to 26 percent in 1928­32, 45 percent in 1933-37, and 84 percent in 1938, according to Anderson, Hayami, and Honma (1986). See table 2.7 for new estimates for an even longer period. 14. The average RRA for 1955­59 is paired with agricultural GDP per worker relative to total GDP per worker in 1955 and so on. 15. The shares of Korean agriculture in GDP and the labor force were 3.8 percent and 7.7 percent, respectively, in 2004, while agriculture represented 1.7 percent of GDP and 7.5 percent of the labor force for the same year in Taiwan, China (table 2.1, rows 2 and 3). References Amsden, A. 1989. Asia's Next Giant: South Korea and Late Industrialization. New York: Oxford Univer- sity Press. Anderson, K. 1983. "The Peculiar Rationality of Beef Import Quotas in Japan." American Journal of Agricultural Economics 65 (1): 108­12, February. ------. 1986. "The Peculiar Rationality of Beef Import Quotas in Japan and Korea." In The Political Economy of Agricultural Protection: East Asia in International Perspective, ed. K. Anderson and Y. Hayami. Boston, London, and Sydney: Allen and Unwin. ------. 1989. "Korea: A Case of Agricultural Protection." In Food Price Policies in Asia, ed. T. Sicular. Ithaca: Cornell University Press. Anderson, K., Y. Hayami, et al. 1986. The Political Economy of Agricultural Protection: East Asia in Inter- national Perspective. Boston, London, and Sydney: Allen and Unwin. Anderson, K., Y. Hayami, and M. Honma. 1986. "The Growth of Agricultural Protection." In The Political Economy of Agricultural Protection: East Asia in International Perspective, ed. K. Anderson and Y. Hayami. Boston, London and Sydney: Allen and Unwin. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC (and appendix A of this volume). ------. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson, K., and W. Martin, eds. 2009. Distortions to Agricultural Incentives in Asia. World Bank, Washington, DC. Anderson, K., and R. Tyers. 1992. "Japanese Rice Policy in the Interwar Period: Some Consequences of Imperial Self Sufficiency." Japan and the World Economy 4 (2): 103­27. Anderson, K., and E. Valenzuela. 2008. Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007. Core database at http://www.worldbank.org/agdistortions. Aoki, M., and Y. Hayami, eds. 1998. The Institutional Foundation of East Asian Economic Development. London: Macmillan. Japan, Republic of Korea, and Taiwan, China 113 Cole, D. C., and Y. C. Park. 1983. Financial Development in Korea, 1945­1978. Cambridge, MA: Harvard University Press. Frank, C. R., K. S. Kim, and K. Westphal. 1975. Foreign Trade Regimes and Economic Development: South Korea. New York: Columbia University Press. Gerschenkron, A. 1943. Bread and Democracy in Germany. Berkeley: University of California Press. ------. 1962. Economic Backwardness in Historical Perspective. Cambridge, MA: Harvard University Press. Godo, Y. 2007. "The Puzzle of Small Farming in Japan." Pacific Economic Papers 365, Australia-Japan Research Centre, Australian National University, Canberra. http://www.crawford.anu .edu.au/pdf/pep/pep-365.pdf Hayami, Y. 1979. "Trade Benefits to All: A Design of the Beef Import Liberalization in Japan." American Journal of Agricultural Economics 62 (2): 342­47. ------. 1986. Nogyo Keizairon [Agricultural Economics]. Tokyo: Iwanami. ------. 1988. Japanese Agriculture under Siege: The Political Economy of Agricultural Policies. London: Macmillan. ------. 2005. "Emerging Agricultural Problem in High-Performing Economies in Asia." Presidential Address at the 5th Conference of the Asian Society of Agricultural Economists, Zahedan, Iran, August 29­31. Hayami, Y., and Y. Godo. 2002. Nogyo Keizairon [Agricultural Economics] Tokyo: Iwanami. ------. 2004. "The Three Agricultural Problems in the Disequilibrium of World Agriculture," Journal of Agriculture and Development 1: 3­16. ------. 2005. Development Economics: From the Poverty to the Wealth of Nations. London: Oxford University Press. Hayami, Y., and S. Yamada, eds. 1991. The Agricultural Development of Japan: A Century's Perspective. Tokyo: University of Tokyo Press. Heston, A., R. Summers, and B. Aten. 2006. "Penn World Table Version 6.2." Center for International Comparisons of Production, Income and Prices, University of Pennsylvania Ho, S. P. S. 1979. "Decentralized Industrialization and Rural Development in South Korea and Tai- wan." Economic Development and Cultural Change 28: 77­96. ------. 1982. "Economic Development and Rural Industry in South Korea and Taiwan." World Development 10: 973­90. Honma, M., and Y. Hayami. 1986. "Determinants of Agricultural Protection Levels: An Econometric Approach." In The Political Economy of Agricultural Protection: East Asia in International Perspec- tive, ed. K. Anderson and Y. Hayami et al. Boston, London, and Sydney: Allen and Unwin. ------. 2008. "Distortions to Agricultural Incentives in Japan, Korea and Taiwan, China." Agricultural Distortions Working Paper 35, World Bank, Washington, DC. Kawano, S. 1969. "Effects of the Land Reform on Consumption and Investment of Farmers." In Agriculture and Economic Growth: Japan's Experience, ed. K. Ohkawa, B. F. Johnston, and H. Kaneda, 374­97. Tokyo: University of Tokyo Press. Kayo, N., ed. 1977. Basic Statistics for Japanese Agriculture. Tokyo: Norin Tokei Kyokai. Kindleberger, C. P. 1951. "Group Behavior and International Trade." Journal of Political Economy 59: 30­46. Krueger, A. O., M. Schiff, and A. Valdés, eds. 1991. Political Economy of Agricultural Pricing Policies. Baltimore: Johns Hopkins University Press. Kuo, S. W. Y. 1975. "Effects of Land Reform, Agricultural Pricing Policy, and Economic Growth on Multiple Crop Diversification in Taiwan." Philippine Economic Journal 14 (1 and 2). Little, I. M. D., T. Scitovsky, and M. Scott. 1970. Industry and Trade in Some Developing Countries. London: Oxford University Press. Mao, Y.-K., ,and C. Schive. 1995. "Agricultural and Industrial Development in Taiwan." In Agriculture on the Road to Industrialization, ed. J. Mellor. Baltimore: Johns Hopkins University Press. Mason, E. S., M. J. Kim, D. H. Perkins, K. S. Kim, and D. C. Cole. 1980. The Economic and Social Modernization of the Republic of Korea. Cambridge, MA: Harvard University Press. 114 Distortions to Agricultural Incentives: A Global Perspective Moon, P. Y., and B. S. Kang. 1989. "Trade, Exchange Rate, and Agricultural Pricing Policies in the Republic of Korea." World Bank Comparative Studies on the Political Economy of Agricultural Pricing Policy, World Bank, Washington, DC. OECD (Organisation for Economic Co-operation and Development). 2003. The World Economy: Historical Statistics. Paris: OECD Development Centre. ------. 2008. Producer and Consumer Support Estimates, OECD Database 1986­2007. http://www .oecd.org Ohkawa, K., and M. Shinohara. 1979. Patterns of Japanese Economic Development: A Quantitative Appraisal. New Haven: Yale University Press. Ohkawa, K., M. Shinohara, and M. Umenura, eds. 1967. Long-Term Economic Statistics in Japan since 1868. Tokyo: Toyo Keizai Shimposha. Otsuki, T., and N. Takamatsu. 1982. On the Measurement of Income Inequality in Prewar Japan. Tokyo: International Development Center of Japan. Schultz, T. W., ed. 1978. Distortions of Agricultural Incentives. Bloomington: Indiana University Press. Scott, M. 1979. "Foreign Trade." In Economic Growth and Structural Change in Taiwan: The Postwar Experience of the Republic of China, ed. J. Galenson. Ithaca: Cornell University Press. Tobata, S., and K. Ohkawa, eds. 1956. Nihon no Keizai to Nogyo [Economy and Agriculture in Japan]. Tokyo: Iwanami. Tsiang, S. C. 1980. "Foreign Trade and Investment as Boosters for Take-Off: The Experience of Taiwan, China." In Export-Oriented Development Strategies, ed. V. Corbo, A. O. Krueger, and F. Ossa. Boulder, CO: Westview Press. World Bank. 1993. The East Asian Miracle: Economic Growth and Public Policy. London: Oxford University Press. ------. 2006. World Development Indicators 2006. Washington, DC: World Bank. Yamazawa, I. 1984. Nihin no Keizai Hatten to Kokusai Bungyo [Economic Development and Interna- tional Division of Labor in Japan]. Tokyo: Iwanami. 3 Western Europe Tim Josling* European agricultural policy, in particular the Common Agricultural Policy (CAP) of the European Union (EU), has long been a matter of international interest. Over- seas producers view agricultural policy in Europe as a major impediment to the opening up of international trade in farm products. Internally, the policy has been no less controversial, with several member states seeking to reform the CAP and others regarding it as a foundation for economic integration. The overall percep- tion, both within and outside the European Union (EU), is of a highly protective policy that shelters a high-cost agricultural sector from the winds of competition blowing from the Americas, Australia, New Zealand, and North Africa. Supporters of the CAP claim social and political benefits from such protection, arguing that reduction of border protection and cutting of domestic support would lead to the depopulation of rural Europe and the destruction of social stability. Detractors see a policy that encourages overproduction, with surpluses dumped on world markets and misallocation of scarce resources away from more profitable uses--in short, they view the CAP as a poster child for agricultural distortion. Some facts are undisputed. Western Europe has transformed from a place of devastated infrastructure and productive capacity in the aftermath of World War II to the home of some of the world's most sophisticated economies. Economic integration, the development of a single market within the EU, and a common approach toward external trade are widely agreed to have contributed *The author is grateful for excellent research support by Uli Kleinwechter and Teresa Rojas Lara of Humboldt University; for invaluable help with data compilation by Johanna Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, Damiano Sandri, and Ernesto Valenzuela; and for helpful comments from workshop participants and Stefan Tangermann. The working paper version of this chapter (Josling 2008a) contains additional background material and appendix tables. 115 116 Distortions to Agricultural Incentives: A Global Perspective to that success.1 Simultaneously, the agricultural sector has gone through a dra- matic transformation, from isolated national markets and traditional produc- tion methods to an integrated EU-wide market and technologically advanced farms. Transforming agriculture in this way within two generations is a remark- able feat. The transformation, though, has had positive and negative impacts. On one hand, it has encouraged investment in agriculture and sheltered producers from foreign competition. On the other hand, it has encouraged the production of high-cost commodities that have shrinking markets. This chapter focuses on agricultural policy developments in Western Europe and their consequent distortions to the economies within the region. It attempts to answer four related questions. · Has the level of protection, and the consequent distortion in resource use and consumer purchases, been increasing or decreasing over the five decades from 1955­2007? · How has the formation and subsequent enlargement of the EU influenced patterns and trends in agricultural protection? · What have been the main drivers of agricultural policy and to what extent have external influences impacted those forces? · What can one say about the future trends and protection levels in Western Europe? The chapter differs in some respects from the others in this book in several ways. First, Western Europe, besides encompassing a large number of countries, also contains different agricultural capacities, climatic conditions, economic structures, and political views. The differences among these countries are often as important as the similarities, and therefore generalizations are difficult to make. Inevitably, much of the focus is on the EU as an aggregate unit, rather than on each individual member country, though this misses some of the richness of the diversity of conditions. But from the viewpoint of the rest of the world, it is instructive to assess the aggregate impact of policies implemented in the EU and more broadly in Western Europe. Second, the process of economic integration in Western Europe has been more intense and comprehensive than in any other region, developed or developing. Though agriculture has been incorporated into the process of integration more fully than in most other regions, there are still differences in this regard within European countries. This means that the process of integration plays a much greater role in Western Europe than in other regions in explaining the pattern and trends in distortions. And as membership of the European Community (later the EU) grew, countries that previously had autonomous policies adopted the CAP. Western Europe 117 The number of agricultural policies in Western Europe has thus declined over the 50-year period considered by this study. Third, the external aspects of agricultural policy have played a more signifi- cant role in developing domestic policy in Western Europe than in most other regions. This is in spite of the fact that the CAP has often appeared resistant to pressures from abroad. The narrative of the past 50 years of agricultural policy in Western Europe is closely linked to the development of trade rules for agricul- tural products in the General Agreement on Tariffs and Trade (GATT) and the subsequent obligations undertaken as a result of the Uruguay Round and the resulting transformation of the GATT into the World Trade Organization (WTO). Other trade agreements, notably the obligations to former colonies through successive Lomé Agreements, have played a significant role in driving some commodity policies within the CAP. In the EU, the separation between domestic and foreign policy in the area of agriculture has always been blurred. Western European countries that are not members of the EU have been influ- enced by many of the same influences. Agriculture in Western Europe, 1955­2007: An Overview Agriculture in Western Europe enjoys a degree of diversity that reflects a wide vari- ety of soils and climatic conditions, the latter of which range from arid Mediter- ranean regions to the Arctic Circle. Superimposed on this natural diversity is the complexity of different social, economic, and political conditions in the 18 coun- tries that are the subject of this chapter.2 History has played a major part in creating this patchwork, particularly the different paths that countries took from feudal- ism to independent farming units and the inheritance laws that influenced the extent to which land ownership was transmitted from generation to generation. Average farm size varies considerably in the countries of Western Europe, in turn reflecting the relative political and social importance of landowners and small farmers. By the late 19th century, these various factors had determined the struc- ture of farming in the Western European region that is still visible today. The total utilized farm area in the 15 members of the EU in 2004 (EU-15) was 129 million hectares, distributed over 6.3 million holdings with an average size of 20.2 hectares.3 The sector employed 6.2 million persons, representing 3.8 percent of civilian employment in the EU (Eurostat 2006, table 2.0.1.2). Including the three countries not in the EU in 2004 (Iceland, Norway, and Switzerland), agriculture rep- resented 4.9 percent of the labor force, down from 30.0 percent in 1950. The value of output from these farms was 300 billion and gross value added was 155 billion (at distorted prices), or 1.6 percent of total gross domestic product (GDP) for those countries.4 118 Distortions to Agricultural Incentives: A Global Perspective Total Western European agricultural output has increased over the past 50 years by about 2.2 percent each year, a slower rate than for other sectors of the economy.5 As a result, the share of agriculture in GDP declined from 13.6 percent in 1955 to 2.9 percent in 2004. Only in Greece and Iceland is the share of agricul- tural output in GDP above 5 percent. Spain, Portugal, and Finland have agricul- tural sectors that contribute between 3 and 4 percent of GDP, while in Germany and the United Kingdom the share is now less than 1 percent. Participation in agricultural activities is highest in Greece and Portugal, with shares in excess of 12 percent of the active labor force. In Finland, Ireland, and Spain more than 5 percent of the labor force works in agriculture. By contrast, only about 1 percent of the United Kingdom's labor force participates in agricultural pursuits. In the immediate postwar period, 1949 to 1959, productivity growth in Western Europe's agricultural sector compared favorably with that in the manu- facturing, with output per worker in agriculture increasing by more than that in manufacturing in most countries. Productivity growth came from a combination of output increases as a result of mechanization and modernization, and the outflow of labor as other sectors absorbed rural workers. Though the productivity increase slowed somewhat in subsequent decades, it remained a key component in the development of the agricultural sector and the sector's role in postwar reconstruction. The strong farm productivity increase and accompanying farm labor force decrease is striking, and the decline in the share of agricultural workers in the labor force continues to the present day. The number of full-time equivalent "annual work units" employed in agriculture in the EU-15, as calculated by Eurostat, fell from 8.6 million to 5.9 million beteween 1991 and 2004 (Eurostat 2006). Structural change has also been rapid in European agricul- ture. The rate of consolidation of farms has risen over the past five decades, but the average size of farms still varies widely among countries, with the United Kingdom and Denmark having the largest farms, at 57 and 55 hectares per farm, respectively, and Italy and Greece having the smallest farms, at 7 and 5 hectares, respectively (Eurostat 2006). The countries of Western Europe also differ dramatically in terms of impor- tance of agriculture to international trade. In 2004, agricultural exports accounted for more than 10 percent of total merchandise exports in three countries: Denmark (18.7 percent), Greece (19.9 percent), and Ireland (11.6 percent). By contrast, they accounted for only 2.5 percent of German exports, 3.0 percent of Swedish exports, 2.8 percent of Finnish exports, and 4 percent of exports from the United Kingdom. Agricultural products, meanwhile, accounted for 6.1 percent of all imports to and 6.0 percent of all exports from the EU-15 as a whole. The EU-15 region ran a net deficit of 3.4 billion with respect to the rest of the world in food- stuffs and other agricultural products in 2004. Western Europe 119 In Western Europe, as elsewhere, agriculture has had to compete with nonfarm sectors for labor and capital. Growth rates of the manufacturing and service sec- tors therefore have been major influences on the economic health of the farming sector. In general, agriculture has provided an outflow of labor--both directly, as farmers and farmworkers become part of the industrial workforce (either by migration or by devoting more of their time each year to nonagricultural employ- ment), and indirectly, by offering less attractive employment to young people in rural areas. Though the outflow has been in progress for decades, the postwar period has been remarkable in the magnitude of the exodus. Obtaining capital within the agricultural sector, on the other hand, has been less problematic, as farmers typically have been able to raise funds in the financial markets, particu- larly through dedicated rural lenders and retained earnings. Though rates of return have not been high, nonpecuniary satisfaction and a lingering feeling on society's part that domestic agricultural activity enhances food security have encouraged rural investment. Sustained capital flows into the agricultural sector have resulted in significant transformation and modernization throughout Western Europe in the six decades since World War II. While the continent still has pockets of traditional farming, particularly in the south, in general, it now has a high level of technical expertise and moderate-sized farms. The transition has allowed Europe to become more internationally competitive in the second half of the postwar period compared to the first half. One major link between the agricultural and nonfarm sectors is through cur- rencies. Strong export performance by the industrial sector tends to appreciate the exchange rate and reduce the domestic cost of commodities, the prices of which are set in international markets. Thus, agriculture in countries with strong cur- rencies tends to be under pressure from reduced price levels as a result of export success in the nonfarm sector. For countries with weak currencies, exchange rate developments tend to raise the price levels of imports and exports, so the agricul- tural sector faces less competition from abroad--although governments may take action to lower domestic food prices. Foreign exchange market developments have had major impacts on the competitiveness of the agricultural sector in some countries, and have played a significant role in policy making in the EU. Agricultural Policy Prior to the Mid-1950s Current Western European agricultural policy reflects economic and social condi- tions of rural areas and political realities of the day, but many of the underlying factors are deeply rooted in experience and history. While some of this experience is shared among the countries of Western Europe, much of it is specific to the ways in which the countries concerned reacted to historical trends and events. 120 Distortions to Agricultural Incentives: A Global Perspective Many of these events were a product of the broad economic and political develop- ments in the 19th century, during which the pattern of land ownership was estab- lished and transportation and education systems extended into rural areas. But policy also influenced the reaction to these developments and led to significant disparities among neighboring countries. In addition to the social and political conditions that governed past agricul- tural policy in Western Europe, two other factors have been pervasive: the colonial experience of the countries concerned, and the reaction of those countries to the Industrial Revolution.6 The United Kingdom, with an extensive empire from which it could import both tropical and temperate agricultural products, was in a good position to benefit from trade. As the leader in both the technological revolution in agriculture and the industrialization of manufacturing processes, a policy of low-priced food played to the strengths of the economy. By contrast, Germany (and the numerous small states that preceded the creation of the unified Germany) had few overseas territories and lagged the United Kingdom in manu- facturing technology. As a result, German agriculture remained a protected sector. Political ideas reinforced these differences. In the United Kingdom, the free trade movement won widespread following by promising better living conditions for the urban workforce. Though landowners resisted, they lost ground over time to manufacturing interests. German intellectuals, however, pushed an infant indus- try strategy, which large landowners found to be in line with their own interests. By the end of the 19th century, significant differences between the United Kingdom and Germany had emerged in the prevailing economic paradigm and the agricultural policies that supported it. All of European agriculture was impacted by the growth of trade in temperate agricultural products from the New World in the 1870s, made possible by the opening up of new territory and falling rail and ocean transport costs. The intro- duction of refrigeration was important in that it began to make the transport of livestock products and grains from the Americas, Asia, and Australia profitable. In the case of the United Kingdom, the high tariffs that had been embodied in the Corn Laws had already been repealed in 1846. As a result, the political climate was such that manufacturing interests prevailed over agrarian pressures, and agricul- ture shrank in the face of overseas competition. Much of the cereals and meat came from colonies, and could be paid for in pounds, so import substitution was not a priority. In addition, the structure of farming in the United Kingdom was generally more able than in other countries to withstand the low prices.7 As a result, pressure for protection was less than in many other European countries. Other countries followed the lead of the United Kingdom and reluctantly accepted the benefits of cheap grain from the New World. Denmark stands out as the country that embraced the new relative price structure most completely, and Western Europe 121 the Netherlands reacted in a similar way. Livestock farming received a boost from the lower feed costs, in particular the rearing of cereal-fed livestock such as pigs and chickens. In addition, Denmark inherited an efficient farm structure from the early 19th century, and developed a cooperative system that fit well into the live- stock economy that flourished on the cheap grain of the 1890s.8 French agricul- tural markets, which were also relatively open to trade in the middle of the 19th century, increased protectionism sharply with the tariff of 1881, which imposed high duties on livestock imports (Tracy 1989). The level of protection peaked with the Meline tariff of 1892 and remained high until World War I. However, industrial tariffs were also increased over that time, masking the dis- tortive impact of agricultural protection. German farming benefited significantly from higher protection in the last quarter of the 19th century. As livestock from the United States and grain from Russia threatened to depress domestic prices, tariffs were introduced in 1879, ini- tially at a moderate level, though they increased over the next decade. Chancellor Otto von Bismarck was adamant that farm imports be controlled, and he presided over a bitter trade dispute with the United States over the sanitary con- ditions under which U.S. pork was produced for export to Germany (Snyder 1945). Upon Bismarck's ouster in 1890, Germany briefly returned to more open agricul- tural trade, albeit in the face of opposition by Prussian landowners (Tracy 1989). For the next 20 years, liberal and protectionist economic paradigms clashed, while disparate views on the desirability of industrialization kept the issue of agricultural polices at the political forefront. Such 19th-century differences between the "adjusters" and the "protectors" remained through the first half of the 20th century. The prime factors that played a role in the development of agri- culture during that period include the economic impact of World War I and the Great Depression. After World War I, the United Kingdom attempted to expand production by granting farmers subsidies to supplement their market earnings. These deficiency payments, introduced in 1917, were accompanied by a liberal import regime for farm products (except sugar). An attempt to introduce price guarantees in 1920 was repealed the following year. Though protection emerged in the 1930s as a result of depressed world prices, the effect was mitigated by imperial preferences that allowed agricultural products in from the Dominions (Australia, Canada, New Zealand, and South Africa) and the colonies. Domestic marketing became the focus of farm policy, and was institutionalized through the introduction of marketing boards in 1931, several of which remained in place until the 1970s. French agriculture was badly damaged during World War I, with both infra- structure and productive capacity destroyed. Agriculture did not get much assis- tance from trade protection in the 1920s, as protection in the nonfarm sector was 122 Distortions to Agricultural Incentives: A Global Perspective higher than that in agriculture. However, the reaction to the Great Depression was to introduce quotas on agricultural imports and intervene in the domestic market. State marketing was established through institutions such as the Office National Interprofessionel du Blé, founded in 1936. Germany, meanwhile, instituted the Ministry of Agriculture in the wake of the war, and the Weimar Republic attempted to take over responsibility for agricul- tural policy from the states. Rapid industrialization, however, reduced the signifi- cance of the farming sector, and low prices during the Depression took a heavy toll on farm incomes (Roesener 2000).9 The Third Reich attempted to capitalize on the decline of agriculture by promising state protection and higher social standing for the rural population, introducing policies designed to promote self-sufficiency and increase the control of the state over marketing and trade of agricultural products. Production, however, did not reach planned targets, and at the outbreak of World War II in 1939, the level of farm output was no higher than in 1935. Labor shortages and the need to keep urban prices down undermined attempts by the National Socialist German Workers' Party to return Germany to its rural past. Denmark, which remained neutral in World War I, expanded its agricultural sales to both the United Kingdom and to Germany in the postwar period (Tracy 1989), despite the increase in German tariffs in 1925. The Depression, however, hit Danish livestock production by 1931. Efforts to improve trade relations with the United Kingdom (to offset the preferences granted to competitors such as New Zealand) and with Germany were partly successful, and Denmark eventually had to compromise on its traditional liberal trade policy and introduce tariffs on grain. Later in the decade it began subsidizing producers of livestock products and restricting production by means of marketing quotas. Eventually, in 1938, grain imports were discouraged by compulsory mixing requirements for millers. Along with that of the Netherlands, Denmark's agricultural experiment of trading at world prices appeared to be at an end. The influence of all these political and economic trends in Western European agriculture in the early 20th century was disrupted by World War II. To an extent probably not experienced in any other region, World War II had a significant impact on the agricultural sector. Not only did the war itself cause havoc to infra- structure and destroy productive land, but the sector was drawn in to the war effort to provide food and industrial raw materials. The United Kingdom, with its vulnerability to blockades of imports, began to mobilize the civilian population to grow more food. In the postwar period, pro- duction rebounded rapidly to its pre-Depression levels, assisted by the introduc- tion of guaranteed prices in the 1947 Agriculture Act. The need for additional domestic production was premised in part on the chronic shortage of foreign exchange in the early postwar period, as exports failed to finance the imports Western Europe 123 needed for reconstruction and to service the debt that had been accumulated during the war. The fact that the pound was a global currency led policy makers to reject any efforts toward devaluation. Successive governments thus pursued a policy of high domestic prices as a way of discouraging agricultural imports. Occupied France, under the Vichy Regime, attempted to restore the country's agricultural "destiny."10 However, devasted infrastructure delayed the restoration of the sector in the immediate postwar period. France began a period of national planning, which included goals for the agricultural sector. The first agricultural plan (1948­52) called for an expansion of exports, while the second plan (1954­57) established a fund for market intervention. As economic integration became a reality, the opening up of markets in Europe to French farm products became an important goal. The salvation of rural France was to be in exporting products to the industrial heartland of Europe, assisted by some protection against overseas suppliers. Though Germany's agricultural sector maintained production levels through most of World War II, food shortages emerged in 1944. By the end of the war, nutritional deprivation was an acute problem and hunger relief became an inter- national issue. Wartime controls over trade were maintained and the new govern- ment in West Germany (established in 1949) encouraged domestic production by price incentives linked to production costs, with little regard to competitiveness. By the time Germany faced the prospect of opening its agricultural market to French and Dutch farm products, the predominance of inefficient small-scale German farms was a highly contentious political issue. Policy Distortions 1955­2004: A Chronology The current map of agricultural protection was drawn in the early postwar period, as the countries of Western Europe struggled to rebuild their economies and restore commercial and political relationships. The defining moment in the devel- opment of agricultural policy was undoubtedly the formation of the European Economic Community (EEC) in 1957. Indeed, no analysis of the distortions caused by agricultural policy in Western Europe can avoid a detailed examination of the development of the CAP and the enlargement of the EEC from the original six countries (Belgium, France, West Germany, Italy, Luxembourg, and the Netherlands) to the 15 that were members of the EU in 2004.11 The stages in the enlargement of the EEC/EU provide the backdrop to any timeline of the descrip- tion and tabulation of the changing pattern of direct and indirect distortions to farm incentives. Not only did each enlargement cause an examination of the EEC/EU policy toward agriculture, but it changed the reach of that policy by including more farms under its umbrella. Thus, the nature and magnitude of the 124 Distortions to Agricultural Incentives: A Global Perspective incentives faced by domestic producers and consumers of primary agricultural and processed food products in Western Europe were affected as much by the adoption of the CAP as through more traditional agricultural policy processes. The other side of this coin is that the number of countries in Western Europe that remain outside the EU has steadily declined.12 Three of these countries-- Norway, Iceland, and Switzerland--still have autonomous agricultural policy despite being members of the European Free Trade Association (EFTA) since 1960. Though these EFTA members represent a declining share of total agricul- ture in Western Europe, their policy decisions are of particular interest. While it would be too simplistic to say that they represent the "control group" in the exper- iment of designing and implementing a common policy for agriculture, they certainly offer important lessons. In this section, agricultural policy distortions are discussed in the context of each of the five decades from 1955 to 2004. The discussion relates this timeline to the evolution of the EU as it expanded membership and of the CAP as the dom- inant vehicle of support for Western European agriculture. Emphasis is placed on the levels of support and key policy prices in the prospective member countries relative to that of the EU as a whole. As each new group of countries gained access to the EU, so the geographical reach of the CAP changed. In turn, the CAP became the focus of much of the external pressure that faced the EU as it grew in significance in world trade. But the macroeconomic conditions, including infla- tion, exchange-rate fluctuations, and nonfarm growth in Western Europe, were a vital backdrop to the agricultural policy decisions and need to be considered in parallel with the more specifically agricultural aspects of the development of the CAP.13 1955­64: Agriculture in a period of rapid economic growth The period of postwar reconstruction was followed by a rapid expansion of economic activity during the 1950s and into the 1960s. This was the period of the German and Italian "miracles," as both these economies grew at rates far above those of the United Kingdom and France. In turn, this growth provided the demand for consumer goods that allowed neighboring countries to expand their own economies. Trade within the newly created EEC expanded rapidly as trade barriers in manufactured goods were removed over the period 1957­64. The formation of the EEC in 1957 set the scene for the integration of Western European agricultural markets.14 For this to happen, agricultural trade had to be included in the move toward the free flow of goods among the original six coun- tries.15 Though this was eventually incorporated, such an agreement on free inter- nal trade was possible only by erecting a protective border around the EEC to Western Europe 125 shelter agriculture from foreign competition. The development of a CAP was therefore the result of a compromise between those who wanted agriculture to be a full part of the free internal market and those who preferred a more interven- tionist system. A common market organization (CMO) was developed for each of the main commodities. Administered prices were related to a "target" price level for each commodity. Imports were allowed in at "threshold" price, calculated on the basis of the target price and transport costs. A "variable levy" was charged on the basis of the gap between offer prices on the world prices and the threshold price. Excess production could be taken off the market by national agencies at "intervention" prices--again, fixed relative to the target price.16 Prices were set each year by the Council of Ministers upon the recommendation of the European Commission. The period 1955­64 was one of policy initiatives for agricultural integration as well as the restoration of trade flows across Europe. National farm prices differed considerably across the six countries of the EEC and also between the six coun- tries and neighboring countries that had chosen not to join in the integration experiment. The path toward more integrated internal markets for agricultural goods proved rocky. Different farm structures, commodity balances, and histori- cal protection levels created a minefield for those advocating a free internal agricultural market and common prices. Eventually, in 1962, common rules for agricultural markets were agreed upon. A transition period was instituted until 1967, when all prices were supposed to have been harmonized. A look at selected Western European countries illustrates the wide differences in circumstances over this period, and illustrates the difficulties of achieving a common policy. West Germany, upon its creation in 1949, found itself with a structure of small farms because the most productive and largest farms were now in East Germany. Meanwhile, German industry was being encouraged to expand into other markets in the region and overseas, and its success added to the strains on agriculture. Tight controls over imports of cereals and the use of marketing boards to regulate the domestic market kept agricultural prices high, while pro- duction was encouraged from every small farm. The choice of a common price for wheat (and other grains) in 1961 was a major political issue in Germany: soft wheat prices were about US$110 per ton at that time, a level matched only by Italy among the six EEC members.17 In the end, the German government resisted calls for the common price to be set at a lower level, thus setting the stage for the devel- opment of surplus production in the EEC within the decade. Though France had farms larger than Germany's in the Paris Basin, it was hampered by the remnants of feudal strip farming in Normandy and Brittany and low productivity in the Massif Central and the Midi. Cereal farming in particular had recovered from the wartime disruption and by the 1960s had surpluses to 126 Distortions to Agricultural Incentives: A Global Perspective send to the deficit areas of Europe, including Germany. Soft wheat prices in France averaged only US$81 per ton at the start of the CAP transition period, and thus had to rise sharply to reach the agreed price levels in the young CAP. The Netherlands, along with Denmark, had an efficient farming structure based on milk, poultry, and eggs. While the Netherlands was already trading agricultural products with Germany, it sought increased market access. The fact that Denmark stayed out of the EEC, preferring to maintain its connection to the British market, gave Dutch farmers a welcome degree of preference. The Netherlands and Belgium, where the soft wheat price level in the early 1960s was between the high in Germany and the low in France, experienced some significant increases in the cost of animal feed in the movement to common prices. Italy shared with Germany some significant structural problems, with small farms dominating the southern part of the country and relatively high-cost cereal production in the center and the northern part. However, expansion of the live- stock sector in the north, based on imported grains, linked Italy's agricultural interests with those of the Netherlands. In fact, in the move to standardize prices, Italy was allowed to maintain imports of feedstuffs at a lower tariff than countries in Northern Europe. The Western European agricultural market was in effect split by the decision to push for a CAP for the six members of the EEC. The United Kingdom chose to stay outside the EEC, concerned about the element of "supranationality" intro- duced in the Treaty of Rome, and the CAP was negotiated without the input of the major Western European food import market. In any case, the EEC was not entirely in agreement over the prospect of the United Kingdom joining.18 The United Kingdom had its own troubles, with macroeconomic imbalances proving difficult to control. The balance of payments was chronically in deficit, and remained a problem until the devaluation of 1967. The 1957 Agriculture Act, which introduced deficiency payments as a way of maintaining high producer prices while keeping consumer prices close to world market levels, was an impor- tant aspect of U.K. foreign policy, because much of the imported food came from former colonies. Import patterns were so significant from a policy perspective that they contributed to the United Kingdom's decision not to join the EEC. However, in an attempt to show leadership among other countries that chose not to partici- pate in the EEC, the United Kingdom sponsored the EFTA in 1960. Seven coun- tries signed up to this "integration-lite" experiment, which differed from the EEC both in terms of its lack of a common tariff and supranational institutions and because it excluded agricultural (and fisheries) trade from its provisions. Facing the prospect of having to phase out preferential access from Dominions and former colonies, as well as the recurrent balance of payments problems and the fear of food-price-led inflation, led to significant debate on the costs of Western Europe 127 agricultural protection. Sparked by a paper by Nash (1955), there were several attempts to calculate the value of British farm output at world prices and compare the result with actual farm values. Several years later, Nash and Attwood (1961) repeated the same calculation using Danish prices, where distortions were notice- ably less, to value British production. McCrone (1962) elaborated these studies into a comparison of 13 countries in 1955­56. Howarth (1971) followed the same method and added an estimate for 1966. According to Howarth's estimates, agri- cultural protection levels increased markedly from 1956 to 1966, at a time when trade in nonagricultural goods was being liberalized.19 More evidence of the increase in protection was found in a study by Anderson and Hayami (1986). The level of protection (as measured by the nominal protec- tion coefficient) was calculated for eight Western European countries for the period 1955­1980. Switzerland stood out as having the highest level of protection, albeit a level that did not increase over the first decade of the period, 1955 to 1965. By contrast, estimated protection did increase in Italy, Sweden, Germany, and the Netherlands over this decade. In France and the United Kingdom, protection actually decreased in the decade up to 1965. The estimates made for the present study broadly confirm the conclusions of the Howarth and the Anderson and Hayami studies.20 Figure 3.1a shows the nom- inal rate of assistance (NRA), the percentage by which a product's domestic price exceeds the price at a country's border), including non-product-specific (NPS) support and aggregated over all commodities, for four of the original six members of the EEC.21 At the time when the Treaty of Rome was being discussed, assistance levels were modest (by later standards) and not too widely dispersed.22 The CAP was "launched" in 1962, with prices that reflected political compromise rather than economic foresight, and by the time the common price regime was in place European agriculture was operating on a price plateau for the major products that was well above world market levels. The corresponding calculations of the NRA for the EFTA countries (those that chose the path of less institutional integration and no free trade in agriculture) are shown in figure 3.1b. Austria, Denmark, and the United Kingdom had NRA levels in the same range as in the EEC. Portugal was in effect taxing its agricultural sec- tor by holding prices below their full market value--a stance that reflected both its own political structure and its level of development.23 Sweden increased its assis- tance to agriculture over the decade, and gave incentives to its agricultural sector to an even greater extent than the countries that formed the EEC. But the coun- tries with the highest level of support by far were Norway and Switzerland, where domestic prices were around twice as high as in other EFTA countries.24 The pattern for these countries is what one would expect from their experience in the prewar period. The United Kingdom, the Netherlands, and Denmark had 128 Distortions to Agricultural Incentives: A Global Perspective Figure 3.1. NRAs to Agriculture, EU-6 and Western European Average, 1956­64 a. EU-6 members 140 120 100 80 percent 60 40 20 0 20 1956 1957 1958 1959 1960 1961 1962 1963 1964 Germany Netherlands Western European average France Italy b. Other European countries 300 250 200 150 percent 100 50 0 50 1956 1957 1958 1959 1960 1961 1962 1963 1964 Switzerland Norway Sweden United Kingdom Denmark Austria Portugal Source: Anderson and Valenzuela (2008), based on authors' spreadsheet, which draws heavily on OECD (2008). Western Europe 129 lower levels of protection, reflecting their history of imports of grains, and France emerged in the decade leading up to World War II as an exporting country with lower production costs and less opportunities for protection at the border. Indeed, one of the main attractions of the EU for France was the opportunity it presented to sell French products abroad without having to bear the costs of sub- sidies. Switzerland, Norway, Sweden, and Germany have higher levels of protec- tion, as one might expect from predominantly importing countries attempting to generate acceptable incomes for farmers. The countries that made up the EEC increased their protection on average over the decade from 1956­64, explaining in large part the chorus of complaints from overseas suppliers about the protection- ist nature of the emerging CAP.25 If the United Kingdom stayed out of the EEC in part because the grouping was more protectionist in terms of agriculture than itself, the same could not be said of all the other EFTA members.26 Protection levels in Norway and Sweden, both of which joined the EFTA in 1965, were high, as they were in Finland (Gulbrandsen and Lindbeck 1973). In fact, the prospect of joining the EEC was seen as a threat to agriculture in these countries, as they would have had to reduce prices and face competition from the grains produced by France and the livestock produced by the Netherlands. Only Denmark looked favorably on the prospect of expanding its sales to the EEC countries, but in the end it chose to remain aligned with the United Kingdom, its traditional market for farm products. Another country that was among the "charter members" of the EFTA, reflecting its historic trade ties with the United Kingdom, was Portugal. The country had been under a dictatorship since 1928, when the military suspended democratic processes, and the economy was being run on a corporatist model with strong central control. Though Portugal's African colonies provided an income and Brazil remained a source of capital and a link with Latin America, Portugal was outside the mainstream of Western Europe and remained a relative backwater until its return to democracy in 1974 (Avillez, Finan, and Josling 1988; Corkhill 1995). Spain, also shunned by many European governments in the postwar years, was additionally hampered by a rigid corporatist economic policy. Innovation and social mobility were discouraged and central and southern landed interests domi- nated political life well into the 1960s. The Franco regime controlled wages, prices, and trade, and large state corporations were prefered over smaller urban enter- prises. Spain was excluded from the Marshall Plan in 1948, and it was not until 1958 that agricultural output regained the level of 1929, just prior to the Great Depression and before the Spanish Civil War. Farm structure, however, was better in Spain than in many parts of Western Europe, and by the 1970s investment in fruit and vegetable production had begun to increase output. The transition to democracy in starting in 1974 offered hope that accession to the EU was possible, 130 Distortions to Agricultural Incentives: A Global Perspective though economic growth stalled for much of the period from the death of Franco to accession to the EU (Lieberman 1995). For both Spain and Portugal, the oil price shocks of 1973 and 1975 had a major detrimental effect on their economies. Between the mid-1950s and 1964, Iceland was predominantly an agricultural and fisheries economy, with meager amouts of arable land and extensive imports of cereals. Dairy and sheep production was able to fulfill domestic and export needs of wool and meat. Though an agricultural development plan for 1951­60 tried to increase farm size (to ensure that every farm had at least ten hectares), it did not succeed (OECD 1967). A law passed in 1960 instituted fixed wholesale and retail prices for major food items and offered export subsidies to dispose of sur- pluses. Consumer subsidies were also used in this period. Under a revised agricul- tural plan of 1963­66, however, public investment in agriculture decreased as the burden of supporting high-cost production became apparent. In the 1950s, Greece was still a largely agricultural country, where cotton and tobacco represented a major portion of exports. Most temperate-zone foods were imported, and domestic production was hampered by low productivity and a fragmented farm structure that was in part the result of prewar land reform. Association with the EEC in 1961 promised to expand exports, and the govern- ment began to increase incentives for producers and institute minimum prices for domestic output (OECD 1967). But this program was carried out in the context of an economic strategy that supported manufacturing sector as the engine of growth (Pepelasis et al. 1980). 1965­74: Agriculture in a period of macroeconomic instability Price levels set under the CAP were at the high end of the range of existing prices in EEC members states, while strong upward convergence was evident over the period from 1962 to 1967. Price differences within the EEC reflected the gradient of prices from surplus to deficit areas, though this was modified by setting inter- vention prices in the surplus areas that did not always reflect transport costs. Achievement of a single support price in 1967, set in units of account with a value equal to the U.S. dollar, was a major political success, but the uniformity of the CAP was short-lived. Currency instability undermined the realization of an open internal market for agricultural products and a common price support scheme. The devaluation of the British pound sterling in 1967 heralded a period of financial instability in Europe. The revaluation of the Deutsche mark in 1968 and the devaluation of the French franc the same year caused havoc in the operation of the CAP, leading to the introduction of artificial "green" exchange rates for the conversion of prices decided in Brussels by the Council of Ministers into local Western Europe 131 currencies (Josling 1970; Josling and Harris 1976). The currency changes in 1968 also exposed the problems of administering regional price differentials. Earlier in the year, grain from France was flooded into the German market as a form of arbi- trage (the relative intervention prices were different from the relative values of the Deutsche mark and the French franc on money markets), causing serious storage problems. When the currency changes took place, compensation for what would have been price declines in Germany was granted by ad hoc tax relief for German farmers, and a system of border taxes and subsidies was introduced to offset the exchange rate changes. This process was extended to other EEC members by con- verting common prices through "green rates" that lagged the developments in the market rates. The border tax/subsidy regime, known as monetary compensatory amounts (MCAs), was imposed on both internal and international trade. The effect was to undermine the central concept of free trade within the EEC, to chal- lenge the notion of common price levels, and in essence give some control over price policy back to the member states (Heidhues et al. 1978). The level of protection in the United Kingdom remained low over the second half of the 1960s relative to major European countries (FAO 1973). Nevertheless, the protectionism debate heated up over the desirability of expansion of U.K. agriculture to help offset the balance of payments shortfall. Though the 1967 devaluation helped to correct a misalignment of the exchange rate, inflation reduced the competitive advantage of British exports. Agriculture was caught up in a wave of "buy British" sentiment in the late 1960s, and price supports were increased in the Annual Price Reviews. The debate on "import saving" exposed the economic costs of agricultural protection and led to a reevaluation of the role of the relatively small agricultural sector in the U.K. economy. The government moved to stabilize the costs of the "deficiency payment" program employed since 1947 by negotiating minimum import prices for cereals with exporting countries. Later, the policy of allowing relatively free import access was replaced by a system of variable levies, to generate some revenue from imports. But it was the prospect of entry into the EEC that dominated the agricultural policy debate in the United Kingdom in the late 1960s. The first enlargement of the EEC took place in 1973, with the accession of Denmark, Ireland, and the United Kingdom. The CAP, however, was a major prob- lem in the negotiations. On one hand, the United Kingdom was under pressure to accept the CAP as part of the "acquis communautaire," the accumulated regulations and directives of the existing EEC, and the enlargement process was supposed to focus merely on changing the language of regulations so as to reflect the new mem- bership. Any change in the policy itself was to be deferred until the new countries joined. On the other hand, at the political level, the United Kingdom was deter- mined to protect the preferential access of its former Dominions and colonies, 132 Distortions to Agricultural Incentives: A Global Perspective though this was unacceptable to the European suppliers for whom the prospect of free access to the United Kingdom, a large importer, was attractive. In the end, the compromise was to negotiate some assurances for traditional suppliers (New Zealand for specific quantities of meat and dairy goods, for example) and to incorporate the former colonies into the arrangements that had already been set up for those of Belgium, France, and the Netherlands. EEC negotiations with Ireland and Denmark were of a different nature. For those two countries, traditional exporters of livestock products to the United Kingdom, the opportunity to diversify to the European market was a welcome prospect. The higher farm support prices in the EEC were not a major hurdle, and tilted the balance of economic advantage in favor of accession. Norway also participated in the negotiations, and an agreement was reached among the governments. But when the question of Norwegian accession was put to a popular vote, the concerns of Norwegian farmers and fishermen prevailed. The vote went against membership, and the government withdrew from the tentative accession agreement. Internal debates about accession to the EU often have been focused on the impact of price changes, particularly for farmers and consumers, as EU prices for most farm products had been considerably higher during the 1960s than in the applicant countries (except Norway). But the aspiring members had agricultural support systems that were more in touch with the conditions on world markets. So the high world prices in 1973­75 for many commodities masked the full impact of the price increases expected from accession. As a result, the additional distortions due to the CAP were relatively small over the first two years of the EU-9. The countries that remained in the EFTA made hurried bilateral arrangements to preserve at least some of their access to the United Kingdom and Denmark. However, these agreements did not cover agricultural products, which had been excluded from EFTA trade liberalization. In the case of Portugal, some conces- sions were agreed: in 1972, a trade agreement was signed between the EEC and Portugal to allow for continued access of Portuguese exports into the bloc. But the domestic farm policies of Sweden, Switzerland, and Norway, along with Finland and Austria, continued to be determined largely on a national basis. These developments in rates of assistance are shown in figure 3.2. The NRA for the EU-6 (and the Western Europe average, heavily weighted by the EU countries) hovered at a little less than 80 percent over the decade 1965­74, plunging at the end of that period as world prices rose in 1973­74.27 The entry of the three new members did not cause this drop, even though it brought some new constraints to the development of the CAP. Meanwhile, the EFTA countries that chose to stay outside continued with their own policy trajectories (figure 3.2b). Switzerland and Norway kept domestic prices at more than twice the level of those on world Western Europe 133 Figure 3.2. NRAs to Agriculture, EU-9 and Western European Average, 1965­74 a. EU-9 members 100 90 80 70 60 percent 50 40 30 20 10 0 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 EU-6 Western European average Ireland United Kingdom Denmark b. Other Western European countries 300 250 200 150 percent 100 50 0 50 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 Switzerland Norway Sweden Finland Austria Portugal Source: Anderson and Valenzuela (2008), based on authors' spreadsheet, which draws heavily on OECD (2008). 134 Distortions to Agricultural Incentives: A Global Perspective markets, while Portugal continued to tax its farmers whatever the price levels on world markets. Rates of assistance to farmers in Sweden, Finland, and Austria were reduced somewhat in the early 1970s as the high world prices substituted for some of the protection given by policy interventions. 1975­84: Agriculture out of control The macroeconomic instability that followed the first global oil shock, in October 1973, had a significant impact on the level of policy prices set under the CAP. Farm input costs rose sharply and the real value of price supports declined. Politi- cians responded by increasing prices sharply to keep up with costs. In addition, the fear of worldwide food shortages made price restraint less attractive as a polit- ical argument. When commodity prices declined, the EU was left seriously uncompetitive in many temperate-zone agricultural products. The United Kingdom, which became a member of the EEC in 1973, was expected to assert a moderating influence on agricultural policy. Indeed, much of the opposition to the U.K. accession from countries such as France stemmed from the feeling that it would attempt to force its tradition of low market prices and high payments from the exchequer onto CAP policy. But these fears proved unfounded. The United Kingdom found a convenient way of keeping its own prices down through the medium of the "green money" system mentioned above, and turned its attention to limiting the budget cost of membership through a budget "rebate." The impact on agricultural distortions of the first enlargement of the EEC was therefore somewhat mixed. Protection levels went up in the United Kingdom and in Denmark and Ireland. But as a result of the transition arrangements, the United Kingdom was able for a time to avoid the impact of high EU prices. The accession period negiotiated for the United Kingdom (seven years) called for import subsi- dies paid by the EEC on farm products entering the United Kingdom. Coupled with the subsidies to offset the depreciation of the pound in the mid-1970s, the price of food rose by less that had been feared at the time of the entry debate. Denmark and Ireland made good use of their expanded opportunities for live- stock exports, selling to Europe and the United Kingdom. The green money system continued to add to the level of internal distortion in the late 1970s. In February 1979, a common price of 100 ECU28 by the Council of Ministers was equal to 110.8 ECU in Germany and 71.8 ECU in the United Kingdom. Thus, support prices were maintained in Germany at a level of 54 per- cent higher than that in the United Kingdom. Only Denmark eschewed the polit- ical convenience of masking exchange rates by the use of fictional green rates. The ECU system remained in place for another decade, undergoing modification as Western Europe 135 part of the single market of 1992, and was finally eliminated by the launch of the euro in 1999. Countries that remained outside the EFTA had the liberty of being able to run their agricultural policy without the need to comply with the CAP. EFTA had no direct impact on agriculture over this period. Switzerland maintained high prices in the late 1970s, with a support price for wheat of US$398 per metric ton, com- pared to US$200 in Germany and US$190 in the United Kingdom (Hallett 1981, based on numbers from the International Wheat Council). Any form of opening up of trade in grains between the EEC and Switzerland, therefore, would have been difficult to implement. Norway, which chose not to join the EEC in 1973, was able to pursue an autonomous policy based on the perceived need to retain population in the north- ern areas of the country. Protection levels were almost as high as in Switzerland, though Sweden resisted the temptation to farm the cold northern regions as a matter of national security. The experience of Finland fell somewhere in between that of its Scandinavian neighbors. Though it had an extensive area of high cost agriculture in the northern parts of the country, rural interests were not able to maintain such high levels of protection against imports as were granted in Norway. The EEC did welcome one new member over this period. Greece, long an asso- ciate member, was welcomed into full membership in 1981, following reestablish- ment of political freedoms. In agricultural terms, Greece posed no problems with respect to the temperate-zone products, as it was likely to increase its imports of them from the EEC as its own protection was withdrawn. But exports of Greek fruits and olives posed a problem for the region, and southern members argued (successfully) for extensive transition periods before opening up to Greek compe- tition. This set the scene for similar arrangements when Spain and Portugal fol- lowed Greece into the EEC five years later. In other regards, the experience of Spain and Portugal was different, as neither had negotiated associate membership status in the 1960s. Both countries had moved from dictatorships to democracy in the mid-1970s, but their agricultural policy was still focused on domestic concerns, including structural issues. Land reform, for example, was a serious concern in Portugal, following the 1974 revo- lution. The breakup of the large farms caused a significant drop in output of grains, a situation made worse by successive droughts. Prices, which were increased in an attempt to generate adequate income for the new class of small farmers created by land reform efforts, rose above those in Spain and the EEC. Dairy production was encouraged in the north and in the Azores. Though tomato processing became a significant export industry in the 1960s, other Mediterranean products were less advanced (Avillez, Finan, and Josling 1991). 136 Distortions to Agricultural Incentives: A Global Perspective By the mid-1980s, Portugal was not in a position to compete in Europe, and ended up needing a significant transition period to develop the institutional capacity to administer the CAP. In Spain, the period before accession also was one of economic stress and rela- tive stagnation. Cereal and livestock production lagged behind that of the rest of the continent, and marketing systems required modernization. Imports of feed grains and oilseeds, mostly from the United States, faced little in the way of trade barriers. Prices for other commodities--wheat, for example--did not differ greatly from those in the rest of Europe. But the fruit and vegetable sector in Spain, with support from the state, was undergoing a structural change--specifically, it had accelerated and wine and olive production. As a result, the prospect of substantial exports of these products loomed over the accession negotiations. The issue of the entry of Spanish farm products into the EEC took on particular political significance in Italy, France, and several countries in North Africa. Changes in the level of agricultural assistance in Western Europe over the period 1975­84 reflect these developments (figure 3.3). In the EU-9 as a whole, NRA levels increased steadily over the decade as a result of cost-based price deci- sions made under in an environment of rising inflation and lack of effective budget constraints. From a level of about 40 percent in 1975, the NRA increased to more than 80 percent in 1983. Overall support levels were very low for the applicant countries of Southern Europe, leading to political tensions over the adoption of the CAP by these countries. A transition period was needed both to cushion domestic consumers and to grant the producers in the EU-9 time to gear up for competition from Spain and Portugal. Of the EFTA countries, Switzerland, Norway, and Iceland continued to provide a high level of support to farmers, while Sweden, Finland, and Austria assisted their farmers at similar rates to those in the EEC. 1985­94: Agriculture as an international concern By the mid-1980s, the issue of domestic agricultural policy in Western Europe had become a central topic of concern in the multilateral trade system, for two rea- sons. First, high levels of border protection were retained to give a broad umbrella of protection against overseas competition under which the market orders could continue. Second, surpluses of cereals, meat, dairy products, and sugar, all of which other Organisation for Economic Co-operation and Development (OECD) countries produced for export, were growing. As a result, the CAP came under criticism abroad as a major cause of low world prices and at home for high sup- port costs and (at least in the United Kingdom) for high consumer prices (Tyers and Anderson 1992). Western Europe 137 Figure 3.3. NRAs to Agriculture, EU-12 and Western European Average, 1975­84 a. EU-12 members 120 100 80 60 percent 40 20 0 20 40 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 EU-9 Western European average Portugal Spain b. Other Western European countries 300 250 200 percent 150 100 50 0 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 Switzerland Norway Iceland Finland Sweden Austria Source: Anderson and Valenzuela (2008), based on authors' spreadsheet, which draws heavily on OECD (2008). 138 Distortions to Agricultural Incentives: A Global Perspective In 1984, the GATT began to discuss and the OECD to measure the extent of the distortions generated by domestic agricultural policy and by border instruments. The consensus was that for several markets, the impact of domestic policies spilled over to the world market in a way that caused a reaction by other countries, either to subsidize exports or to increase protection against imports. A common solution seemed to be the answer. If all domestic policies could be brought under control, and if the nature of border measures could be disciplined, then the situa- tion would be ameliorated. But the CAP was singled out as the policy that needed to change the most, and so the policy was firmly on the international agenda. Attempts to argue that U.S. support per farmer was significantly greater than in the EEC (as a result of larger farm size) garnered little sympathy. So long as the focus was on the impact on world markets, rather than the impact on domestic incomes, the CAP was subject to skepticism. Hence, the attitude of the European Union (EU)29 during the Uruguay Round was largely defensive. Eventually, in 1992, the MacSharry reforms forced the CAP to adjust from within and allowed the EU to agree to the negotiated strictures of the Uruguay Round Agreement on Agriculture (URAA) on export subsidies and domestic support payments. The 1992 CAP reforms marked a change in instrumentalities as well as price levels. In compensation for a price drop, farmers received a direct payment based on historical hectareage for cereals and oilseeds and a regional yield. The lower price for cereal-based animal feed led to an increase in the use of barley, wheat and corn, and helped to reduce cereal stocks. In addition, headage payments were introduced for beef and sheep, and the incentive to produce was reduced, as mar- ginal output increases were effectively sold at lower prices.30 The milk and sugar regimes, however, were not included in 1992 reforms. The adoption of the CAP reforms by the Southern countries (Portugal and Spain joined the EEC in 1986) posed few problems for the major crop and live- stock product producers, and did not raise the level of distortion overall. The main impact in those two countries was on the Mediterranean products--particularly olive oil, wine, citrus, and tobacco. For those products, the stimulus increased pressure on markets and hence calls for policy review. Several changes with respect to the position of the EFTA countries occurred between 1985 and 1994. They had been offered, and negotiated, an industrial free trade area (the European Economic Area, or EEA) with the EU. The EEA allowed free trade in manufactured goods and cooperation in regulatory issues: in effect, it extended previous bilateral agreements to include several aspects of trade that had been incorporated in the 1992 European single market. Though some quotas on agricultural goods were expanded, there was no progress toward the incorpora- tion of the rural sector in economic integration. The EFTA countries, which still did not have a say in EU decisions, were thus unable to influence regulations that Western Europe 139 would apply to them. Meanwhile, the politically neutral stance that previously prevented several EFTA countries from having close ties with the EEC became less important with the end of the Cold War. So, before the ink was dry on the EEA, Sweden made the decision to apply for full EU membership, and three other EFTA countries followed suit. Though the path toward EU accession was attractive to many countries of Western Europe, some ultimately did not make the journey. Norway, once again, chose to stay out. Switzerland also found that membership was problematic; even membership in the EEA was rejected in a referendum, particularly by the rural German-speaking Swiss cantons. In Austria and Finland, rural and agricultural concerns made negotiations complex, but both countries joined. A subsidy sys- tem, paid for in part by the new entrants themselves, was implemented to allow marginal farms in disadvantaged areas in both countries to continue to operate. Sweden faced a different dilemma: that it would have to reverse an agricultural policy reform that was generally in the direction that the EU wished to go. Until the late 1980s, Sweden maintained a policy much like that of other Nordic coun- tries. Suffering from overproduction, Sweden tried voluntary milk quotas and cereal set-asides, but to no avail. In June 1990, the parliament passed a bill that dramatically altered Swedish farm and food policy, abolishing agricultural sup- port and export refunds. In return, the government issued direct annual payments and an early retirement scheme to farmers. When export subsidies were reinstated upon Sweden's joining the EU, prices rose somewhat. One impact of joining the EU can be gauged by comparing cereal prices in the acceding countries and the EU. Finland, in particular, had to weigh the burden of reduced prices on the farm sector against the potential benefits of better manufac- turing-sector access to the rest of Europe. Nevertheless, the accession of Austria, Finland, and Sweden proceeded, a process smoothed by the fact that these coun- tries were able to pay for schemes that temporarily sheltered their farmers. These countries have not used their membership to press for higher agricultural prices at the EU level. The pattern of protectionism is illustrated in the calculated NRA for the EU and EFTA. Figure 3.4 shows the level of NRA for the EU-12 peaking at close to 100 percent in 1986, when world prices were at historical lows, and generally declining thereafter. By 1994, the NRA for the EU was down to 40 percent. Among the three countries that acceded in the 1985­94 period, Finland had the highest rate of assistance, and thus had the most adjustments to make upon accession. In the case of Austria, protection levels increased in the five years leading up to mem- bership, causing concern that the agricultural sector would be adversely effected after joining the EU. But the new EU members were not among the strongest sup- porters of agriculture in the EFTA countries, and so the tensions were manageable 140 Distortions to Agricultural Incentives: A Global Perspective Figure 3.4. NRAs to Agriculture, EU-15 and Western European Average, 1985­94 a. EU-15 members 180 160 140 120 percent 100 80 60 40 20 0 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 Sweden Finland Western European average EU-12 Austria b. Other Western European countries 500 450 400 350 300 percent 250 200 150 100 50 0 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 Iceland Switzerland Norway Source: Anderson and Valenzuela (2008), based on authors' spreadsheet, which draws heavily on OECD (2008). Western Europe 141 by means of transitional arrangements. As a result, there was very little impact on the EU from the enlargement from 12 to 15 members, in terms of either reducing or increasing the level of agricultural protection.31 The EFTA members that chose not to join the EU, meanwhile, maintained their high protection levels for agriculture. NRAs in Switzerland fell somewhat in 1989 but rebounded by 1990. Though there was a slight downward trend in Iceland and Norway over 1985­94, the rate of assistance remained well above the average for Western Europe.32 In any case, the bipolar nature of support in the region, as between nonmembers and members of the EU, was by that time firmly established. 1995­2007: Agriculture restrained The coming into force of the URAA in 1995 brought significant changes in the instrumentalities of the CAP. First, the agreement obliged the EU to convert its variable levy to a fixed tariff. Thus the threshold price from which the variable levy was calculated ceased to be the central determinant of protection levels. However, as a result of the careful choice of reference prices to calculate the tariff equivalent (known as "dirty tariffication") and the convenience of using an unweighted aver- age for tariff reductions, the impact on the price levels of sensitive goods such as sugar, beef, and dairy products was not great. For cereals, the creation of a special category of domestic support that was associated with output controls (the blue box) and hence deemed to be less distorting to trade meant that the new MacSharry payments were not required to be reduced. Moreover, a remnant of the "variable levy" was retained (at the insistence of the United States), because the tariff-inclusive import price was not allowed to rise above 150 percent of the intervention price. The time horizon of the URAA tariff and support reductions, 1995 through 2000, partially overlapped with that of the application of the MacSharry CAP reforms, which ran 1994­96. Though it became clear that though the 1992 CAP reform may have allowed the EU to agree to the URAA, the reform process would need to be continued if the WTO constraints were to be respected over the decade. Even more significant was the prospect of eastward expansion, where 10 countries already had an explicit promise of membership. While these countries were not large agricultural exporters, they did have significant farming populations. Farmers in the EU were focused on additional competition in certain commodi- ties, but the European Commission was concerned about the potential burden of paying subsidies to farmers in the prospective entrant countries. In either case, it was clear that the enlargement would have significant effects on the viability of the CAP. 142 Distortions to Agricultural Incentives: A Global Perspective In 1995, Franz Fischler took over as commisioneer for agriculture for the EU. He articulated a vision of a competitive, market-oriented (and simultaneously environmentally sustainable and socially acceptable) agricultural sector. The path chosen under his leadership was to continue down the path of the MacSharry reforms by reducing support prices and substituting direct payments tied by "cross-compliance" obligations to environmental goals. The Commission's propos- als for agricultural reform, included in the Agenda 2000 document, also dealt with budget and regional policy challenges facing the EU (Moyer and Josling 2002). The primary focus of Agenda 2000, though, was the enlargement of the EU to include the Central and Eastern European countries. Ten of these countries had already negotiated "Europe" agreements that would lead to bilateral free trade (including most of agriculture) over 10 years. As pressure mounted to fulfill the political pledge to allow for the reuniting of East and West Europe, concern about the impact on both agricultural markets in the EU and the financial cost of extending the CAP to the new members was also rising. Agenda 2000 was the Commission's suggestion for how the budget and agricultural challenges could be met. The agricultural reform component of Agenda 2000 was presented to the Council of (Agriculture) Ministers in 1997 and eventually endorsed with much of the Commission's ideas intact in 1999. Though the Agriculture Ministers' plan was modified by the European Council (a regular meeting of heads of state or government), the agreement that emerged was a major step in the evolution of the CAP.33 One important concession was won by those who favored even more reform of the CAP: there would be a midterm review of the effectiveness of the Agenda 2000 reforms in 2003 (the halfway point of the 2000­06 fiscal horizon). Fischler was able to convert what could have been a routine review of progress into a further step along the path set by the MacSharry and Agenda 2000 reforms by shepherd- ing through the Council of Ministers a proposal for consolidating the direct pay- ments associated with the compensation for price decreases in particular crops into the Single Farm Payment Scheme, a system that decreased links between farmers' production decisions and market prices on one hand and direct payments on the other (Swinnen 2008b). The 2003 reform package also began to tackle the difficult issue of reform of the dairy sector. Following on the heels of the 2003 reform, the European Commission pro- posed changes in the regimes for the Mediterranean crops--specifically, cotton, tobacco, rice, and hops. The thrust of reforms surrounding these products was to blend them with the Single Farm Payment Scheme while reducing distortions in the market. New proposals for fruits and vegetables were adopted in 2004, with subsidies for processing being converted into producer payments and those for Western Europe 143 the withdrawal of fresh produce from the market being chaneled through pro- ducer organizations. Wine and sugar reform proposals were introduced in 2005 and the sugar reform was agreed in 2006. This new regime aimed to eliminate the production quota system, paying compensation to the adversely impacted seg- ments of the wine market--producers, crushers and refiners, and overseas pro- ducers that rely on sales to the EU market--and cutting support prices. The wine reforms were adopted in 2008 with the aim of cutting capacity to meet shifting demand. The aggregated impact of these various policy changes is shown in figure 3.5. The EU-15 exhibited an NRA of around 40 percent following its absorbtion of three new members (and the introduction of the disciplines of the URAA) in 1995. By 2004, this had declined to approximately 30 percent. The reforms, combined with higher world prices, lowered the distortion to about 13 percent by 2007. The same trend is seen in the calculation of NRAs for the EFTA coun- tries that have retained their agricultural policy autonomy. For example, Norway and Switzerland have reduced distortions, as measured by the NRA, to about half the levels seen in the late 1990s. Iceland seems to be the EFTA coun- try where the tendency toward more moderate levels of assistance has been resisted most. Detailed Estimates of Nominal Assistance, 1955­2007 Having calculated the NRAs to inform the above description of the evolution of agricultural policies in Western Europe, it is possible to use the database to illus- trate several additional points. First, the NRA calculations need to be qualified by the change in the nature of the policy instruments adopted.34 In the latter 1980s, direct payments were limited to such products as olive oil and durum wheat, with market price support making up the bulk of assistance given to agriculture. From 1992, the total support (including direct payments and NPS support) and the market price support (limited to price support for individual commodities) began to diverge. By 2004, almost one-half of the 100 billion in assistance to agriculture came in the form of income payments and provision of services of an NPS nature (OECD 2006). There is no doubt that such payments benefit agriculture and keep resources that might otherwise leave in the sector, but the direct distortionary impact on commodity output, consumption, and trade is less, and arguably a much smaller source of the programs' economic cost.35 If the assistance from these somewhat decoupled measures had been included in the calculation of sup- port for Western European farmers, their NRAs would have declined much less after 1992 (see figure 3.6). 144 Distortions to Agricultural Incentives: A Global Perspective Figure 3.5. NRAs to Agriculture, EU and Western European Average, 1995­2007 a. EU-15 to 2004, EU-25 for 2005­06, and EU-27 for 2007 60 50 40 percent 30 20 10 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Western European average EU-15/25/27 b. Other Western European countries 250 200 150 percent 100 50 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Iceland Switzerland Norway EU-15 Source: Anderson and Valenzuela (2008), based on authors' spreadsheet, which draws heavily on OECD (2008). Western Europe 145 Figure 3.6. NRAs to Agriculture with and without Decoupled Payments, Western Europe, 1956­2007 120 100 80 percent 60 40 20 0 56 59 62 65 68 71 74 77 80 83 86 89 92 95 98 01 04 07 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 Western European average including Western European decoupled payments average Source: Anderson and Valenzuela (2008), based on authors' spreadsheet, which draws heavily on OECD (2008). Second, when products are classified as exportables, importables, and nontrad- ables, it becomes clear that exportables are assisted far less than import-competing farm industries in the case of the EU--though exporters are assisted only slightly less in the case of non-EU countries of the region (figure 3.7). Third, the rate of assistance to Western Europe's farmers dwarfs the small and declining NRAs for producers of nonagricultural tradable products, and so the relative rate of assistance (RRA) is almost the same as the NRA for agriculture (table 3.1). Fourth, the NRA varies greatly across the range of covered farm commodities (which account for more than 75 percent of the region's agricultural production when valued at undistorted prices). The extent of dispersion in NRAs for the EU, which increased up to the 1980s before starting to decline, has nevertheless contin- ued to increase in non-EU countries (see second-to-last row of tables 3.2 and 3.3). This suggests there is great scope to improve the efficiency of resource use within each country's farm sector. Fifth, leaving aside NPS and decoupled support, most of the assistance to farm- ers for covered products has come from border measures rather than domestic price supports (see near bottom of tables 3.2 and 3.3). This aspect of European Figure 3.7. NRAs to Exportable, Import-Competing, and Alla Agricultural Industries, EEC/EU and Other Western European Countries,a 1956­2007 a. EEC/EU 140 120 100 80 60 percent 40 20 0 20 40 60 56 59 62 65 68 71 74 77 80 83 86 89 92 95 98 01 04 07 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 import-competing total exportables b. Other Western European countries 300 250 200 percent 150 100 50 0 56 59 62 65 68 71 74 77 80 83 86 89 92 95 98 01 04 07 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 import-competing total exportables Source: Anderson and Valenzuela (2008), based on authors' spreadsheet, which draws heavily on OECD (2008). a. All NPS assistance is apportioned to tradables. The EU is the original 6 countries to 1972, 9 countries to 1985, 12 to 1994, 15 to 2004, 25 to 2006, and 27 thereafter. The other Western European countries in our study comprise the original 7 EFTA members (Austria, Denmark, Norway, Portugal, Sweden, Switzerland, and the United Kingdom) plus Finland, Iceland, Ireland, and Spain to 1970; then without Denmark, Ireland, and the United Kingdom from 1973 (when they joined the EEC-6); then without Portugal and Spain from 1986 (when they joined the EC); and without Austria, Finland, and Sweden from 1995 (when they joined the EU). 146 Table 3.1. NRAs to Agricultural Industries Relative to Nonagricultural Industries, EU-15 and EFTA-3, 1956­2007 a. EU-15 member countries (weighted average, percent) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, covered products 49.0 61.9 72.4 47.3 56.3 69.9 79.8 60.8 39.8 32.9 14.7 NRA, noncovered products 14.1 28.2 38.4 26.0 33.3 39.1 45.8 32.7 23.3 19.3 -- NRA, all agricultural products (excluding NPS) 39.8 53.6 64.4 42.5 51.3 62.7 72.6 54.6 36.2 30.0 11.2 All importables 56.0 74.3 80.2 52.9 58.8 76.5 84.0 63.8 50.4 48.2 28.6 All exportables 3.9 13.0 27.4 18.5 30.0 27.3 44.5 33.3 12.0 6.0 0.5 NRA, all agricultural products (including NPS) 39.8 53.6 64.4 42.5 53.3 71.0 76.8 59.2 40.4 34.2 15.4 NRA, decoupled payments 0.0 0.0 0.0 0.0 0.2 1.7 2.3 8.1 16.7 17.9 19.5 NRA, all agricultural products (including NPS and decoupled assistance) 39.8 53.6 64.4 42.5 53.5 72.6 79.1 67.3 57.0 52.1 35.0 NRA, all agricultural tradables (including NPS) 39.8 53.6 64.4 42.5 53.3 71.0 76.8 59.1 40.4 34.2 15.4 NRA, all nonagricultural tradables 8.2 7.4 5.7 3.8 2.5 1.4 1.7 1.3 1.4 1.4 1.1 RRAa 29.2 43.1 55.5 37.3 49.7 68.5 73.9 57.1 38.4 32.4 14.1 (Table continues on the following page.) 147 148 Table 3.1. NRAs to Agricultural Industries Relative to Nonagricultural Industries, EU-15 and EFTA-3, 1956­2007 (continued) b. Iceland, Norway, and Switzerland (weighted average, percent) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, covered products 256 257 255 252 256 232 339 295 204 171 99 NRA, noncovered products 169 172 173 179 166 169 150 111 78 64 141 NRA, all agricultural products (excluding NPS) 228 230 230 230 230 218 282 236 165 137 79 All importables 220 221 221 220 227 211 262 234 169 149 114 All exportables 247 249 249 255 239 240 334 242 159 119 70 NRA, all agricultural products (including NPS) 228 230 230 230 233 233 313 266 197 164 90 NRA, decoupled payments 0 0 0 0 11 56 31 32 48 68 68 NRA, all agricultural products (including NPS and decoupled assistance) 228 230 230 230 244 289 344 298 245 233 158 NRA, all agricultural tradables (including NPS) 228 230 230 230 233 233 313 266 197 164 90 NRA, all nonagricultural tradables 4.0 4.3 4.2 3.0 2.5 2.1 1.8 1.9 2.1 2.1 1.6 RRAa 216 216 217 220 225 226 306 259 191 159 87 Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. Note: -- not available. a. The RRA is defined as 100*[(100 NRAagt ) (100 NRAnonagt ) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. Table 3.2. NRAs to Covered Farm Products, EU,a 1956­2007 (percent) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Exportables 17 12 4 7 17 11 8 10 6 2 1 Barley n.a. n.a. n.a. 35 45 14 n.a. n.a. n.a. 1 0 Rapeseeds n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0 0 0 Rice 48 41 1 23 6 2 146 136 34 17 2 Tomatoes 9 16 n.a. n.a. n.a. n.a. n.a. 4 0 1 0 Wheat n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 16 4 0 Wine 16 13 4 4 4 11 1 11 4 2 1 Import-competing 60 90 103 68 70 93 93 70 56 55 28 Barley 39 59 66 n.a. n.a. n.a. 108 104 32 n.a. n.a. Beef 20 79 62 52 12 120 150 93 100 132 86 Eggs 28 18 20 1 23 17 22 10 9 1 0 Maize 9 24 55 37 56 43 90 89 31 25 19 Milk 259 301 314 269 431 335 291 122 87 61 19 Oats 3 34 46 20 45 0 37 47 47 23 8 Oilseeds 55 2 0 0 0 0 109 55 0 0 0 Pigmeat 28 75 139 113 88 111 24 13 25 28 17 Potatoes -- -- -- 43 90 48 16 16 13 10 10 Poultry 130 64 54 88 55 75 79 105 81 64 78 Sheepmeat 170 197 343 340 277 189 164 97 45 38 65 Soybeans 13 11 -- 0 0 0 121 61 0 0 0 Sugar 137 121 295 23 131 140 227 162 167 191 111 Wheat 11 52 67 5 2 38 84 75 n.a. n.a. n.a. 149 (Table continues on the following page.) 150 Table 3.2. NRAs to Covered Farm Products, EU,a 1956­2007 (continued ) (percent) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, all covered products 44 63 82 54 61 78 81 58 40 33 15 Domestic market support 0 0 0 0 0 0 3 2 0 0 0 Border market support 44 63 82 54 61 78 77 56 40 33 15 Dispersion of NRA of covered products 81 85 120 99 116 90 82 51 45 53 38 % Coverage at undistorted prices 74 77 78 80 79 78 78 78 79 79 74 Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. Note: -- not available; n.a. not applicable. a. Weighted averages, with weights based on the unassisted value of production. Dispersion is the simple five-year average of the annual standard deviation around the weighted mean. The EU is the original 6 countries to 1972, 9 countries to 1985, 12 to 1994, 15 to 2004, 25 to 2006, and 27 thereafter. Table 3.3. NRAs to Covered Farm Products, Non-EU Western European Countries,a 1956­2007 (percent) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Barley 36 39 29 18 34 5 223 219 186 105 76 Beef 65 81 29 59 130 186 161 166 174 225 149 Eggs 64 69 71 80 321 328 145 148 296 161 186 Maize 91 53 47 34 48 34 51 89 124 89 64 Milk 259 275 251 262 301 223 227 240 242 181 73 Oats 23 48 71 58 132 34 103 115 154 113 83 Oilseeds 28 18 14 8 7 14 58 126 334 260 287 Pig meat 51 63 74 38 12 45 71 82 180 172 116 Potatoes 130 144 167 59 63 63 38 35 -- -- -- Poultry 56 52 25 101 188 207 104 74 370 437 383 Rice 18 25 14 29 21 12 52 -- -- -- -- Sheep meat 110 81 78 64 75 80 168 170 101 68 51 Soybeans -- -- -- 28 25 30 13 -- -- -- -- Sugar 265 253 446 47 126 121 250 198 233 255 173 Tomatoes 0 0 0 5 4 4 4 -- -- -- -- Wheat 29 42 43 7 14 46 126 170 195 91 61 Wine 0 0 0 0 0 0 0 0 -- -- -- Wool -- -- -- -- 124 36 78 207 187 167 207 (Table continues on the following page.) 151 152 Table 3.3. NRAs to Covered Farm Products, Non-EU Western European Countries,a 1956­2007 (continued ) (percent) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Exportables 37 43 55 37 35 30 98 133 236 172 70 Import-competing 82 81 66 43 57 61 169 191 194 171 114 All covered 68 69 63 41 50 51 119 152 204 171 99 Domestic market support 0 0 0 0 2 12 -20 -17 7 12 11 Border market support 68 69 63 41 46 33 138 169 198 159 88 Dispersion of NRA of covered products 90 91 125 67 109 103 81 73 84 105 112 % Coverage at undistorted prices 73 73 74 75 74 75 86 75 69 69 80 Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. Note: -- not available. a. Weighted averages, with weights based on the unassisted value of production. Dispersion is the simple five-year average of the annual standard deviation around the weighted mean. Non-EU Western European countries in this study include the original seven EFTA members (Austria, Denmark, Norway, Portugal, Sweden, Switzerland, and the United Kingdom) plus Finland, Iceland, Ireland, and Spain to 1970; then without Denmark, Ireland, and the United Kingdom from 1973 (when they joined the EEC6); then without Portugal and Spain from 1986 (when they joined the EC); then without Austria, Finland, and Sweden from 1995 (when they joined the EU). Western Europe 153 policies is in sharp contrast to the situations in the United States, where domestic support plays a major role (Gardner 2009). Sixth, the average rate of assistance to farmers varies by country even among EU members (table 3.4). This is because the commodity composition of farm out- put varies across countries and over time, and hence so do the weights (based on value of production at undistorted prices) used to calculate the average. Sixth, the gross subsidy equivalent of the support provided to Western Europe's farmers varies hugely across countries (table 3.5a) and across commodities (table 3.5c). Milk, beef, and wheat are subsidized most in total. Expressed per per- son engaged in farming, subsidies are highest in Norway and Switzerland. Among EU members, Denmark, the Netherlands, and France have rates almost as high (table 3.5b). Note that the subsidy equivalent per EU farmer drops by two-thirds in 2005­07, much more than the 45 percent drop in the aggregate value over 2000­04, because the number of EU farmers has expanded dramatically with the latest eastern enlargement (even though, strictly speaking, the new member coun- tries are not in Western Europe). These figures are not actual cash transfers from the EU's budget in Brussels, but simply the value equivalent of the price support provided primarily via protection from non-EU imports. Finally, since most assistance to farmers comes from border measures, those same measures raise consumer prices of farm products. The extent of the con- sumer tax equivalent (CTE) is similar to that of the NRA for covered products in percentage terms (less so the further a country is from self-sufficiency), but vary considerably across countries when expressed on a per capita basis: after Iceland, Norway, and Switzerland, it is Ireland, Denmark, and the Netherlands where con- sumers are most harmed (table 3.6). Notice that the value of the implicit transfer from consumers is very small on a per capita basis, at about US$100 for EU coun- tries and US$300 for EFTA countries in 2000­04 (valued at 2000 US$). Since this is less than 0.5 percent of per capita income, it is not surprising that consumers see little benefit in joining together to lobby collectively against agricultural supports. Policy Trends and Turning Points since 1955 This section attempts an analytical explanation of the reasons behind the evolution of policy and policy-related distortions since the mid-1950s. It complements the historical/institutional narrative and NRA estimates of the two previous sections by emphasizing the political economy forces behind the long-run trends and turn- ing points, focusing on the CAP as the main policy for an increasing number of Western European countries (policy developments in the countries that opted to stay outside the EU show some similarity, however). Emphasizing the trends in policy allows for consideration of the changes in policy instruments over time and 154 Table 3.4. NRAs to All Agriculture,a Individual Western European Countries, 1956­2007 (percent) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Austria 25 33 36 12 17 17 49 71 45 41 19 Denmark 29 33 37 37 78 86 81 54 41 34 15 Finland 81 93 101 73 75 37 136 134 52 41 17 France 41 49 70 43 50 74 82 64 40 33 15 Germany 52 83 99 63 68 89 83 62 46 38 16 Iceland -- -- -- -- 229 252 346 289 173 144 153 Ireland 20 41 49 43 84 117 132 83 67 62 34 Italy 5 30 48 32 37 61 60 46 32 27 11 Netherlands 43 78 104 80 89 107 80 53 48 42 21 Norway 188 193 196 201 219 240 259 235 191 165 98 Portugal 6 4 7 9 24 15 41 37 28 26 15 Spain 13 14 12 2 4 1 53 44 31 26 14 Sweden 87 113 120 77 85 79 90 88 48 41 18 Switzerland 253 251 249 244 243 229 349 285 201 165 83 United Kingdom 65 60 43 32 68 83 95 70 47 40 20 Total Western Europe, weighted average 44 57 68 46 56 74 82 64 44 37 17 EU countriesb 40 54 64 43 53 71 77 59 40 34 15 Other Western Europeb 228 230 230 230 233 233 313 266 197 164 90 Source: Anderson and Valenzuela (2008) based on authors' spreadsheet. Note: -- not available. a. Weighted averages, with weights based on the unassisted value of production. Dispersion is the simple five-year average of the annual standard deviation around the weighted mean. b. The other Western European countries in our study comprise the original seven EFTA members (Austria, Denmark, Norway, Portugal, Sweden, Switzerland, and the United Kingdom) plus Finland, Iceland, Ireland, and Spain to 1970; then without Denmark, Ireland and the United Kingdom from 1973 (when they joined the EEC6); then without Portugal and Spain from 1986 (when they joined the EC); and without Austria, Finland, and Sweden from 1995 (when they joined the EU). Table 3.5. Gross Subsidy Equivalents of Assistance to Farmers, Total, Per Farm Worker and by Product, Western European Countries,a 1956­2007 a. Total (constant 2000 US$ millions) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Austria 422 566 651 192 462 440 8,184 3,108 1,715 1,338 541 Denmark 1,439 1,667 1,983 2,403 5,719 5660 4,133 3,804 2,565 1,920 888 Finland 1,219 1,416 1,820 1,733 2,185 932 3,644 3,685 1,000 730 287 France 9,866 14,437 22,418 17,393 26,423 31,314 26,608 25,399 15,707 11,012 5,225 Germany 8,735 18,158 25,895 21,893 32,312 33,297 24,905 21,113 13,869 10,125 3,975 Iceland -- -- -- -- 314 246 245 177 113 96 123 Ireland 254 572 710 1,235 3,744 4,026 3,553 3,421 2,646 2,035 837 Italy 651 6,596 11,772 8,903 13,826 17,067 13,585 13,061 8,728 6,257 2,646 Netherlands 1,812 3,778 5,634 6,153 9,735 10,432 7,083 6,695 5,228 3,593 1,483 Norway 1,432 1,497 1,710 1,924 2,898 2,876 3,061 3,437 2,566 1,836 1,208 Portugal 58 70 170 295 858 416 1,218 1,579 1,194 924 509 Spain 841 911 919 816 741 604 8,792 9,431 6,446 5,440 3,372 Sweden 2,573 3,323 3,832 3,050 4,229 3,363 3,180 3,096 1,499 1,047 413 Switzerland 3,148 3,486 3,981 5,130 5,599 5,817 6,491 6,541 4,438 3,014 1,945 United Kingdom 8,937 9,177 7,181 6,915 16,530 17,597 14,013 12,752 8,138 5,498 2,411 Total, Western Europe 41,271 65,653 88,676 76,403 124,095 132,880 128,695 117,300 75,850 54,866 31,173 EU membersa 21,065 42,969 65,718 54,342 108,415 119,393 103,890 97,256 68,734 49,920 27,618 Other Western Europea 20,206 22,685 22,957 22,061 15,680 13,487 24,805 20,044 7,116 4,946 3,555 (Table continues on the following pages.) 155 156 Table 3.5. Gross Subsidy Equivalents of Assistance to Farmers, Total, Per Farm Worker and by Product, Western European Countries,a 1956­2007 (continued ) b. Per person engaged in agriculture (constant 2000 US$) 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Austria 805 1,161 409 1,257 1,361 28,429 12,023 7,947 7,661 3,468 Denmark 5,240 6,627 10,339 26,813 30,680 24,401 25,217 20,676 18,950 9,865 Finland 2,820 3,853 4,267 6,547 3,340 15,411 18,224 6,160 5,659 2,540 France 4,285 6,571 6,223 11,736 16,887 17,304 20,111 15,437 13,678 7,412 Germany 4,303 6,894 7,235 11,707 13,875 13,252 14,324 11,891 11,069 4,938 Iceland -- -- -- 24,186 19,696 17,016 11,597 8,195 7,529 10,223 Ireland 1,752 2,169 5,026 14,801 17,647 17,261 18,759 15,662 12,680 5,363 Italy 1,445 2,514 2,350 4,373 6,288 5,825 6,731 5,632 5,176 2,524 Netherlands 10,230 15,336 18,827 30,384 33,320 22,665 22,195 19,561 15,434 6,928 Norway 5,992 7,945 10,489 17,122 18,461 21,634 26,848 22,641 18,274 12,988 Portugal 39 143 276 765 357 839 1,925 1,668 1,492 891 Spain 182 230 255 259 218 2,973 5,303 4,364 4,418 3,024 Sweden 8,365 10,773 10,105 15,409 13,520 14,501 15,848 8,999 7,422 3,200 Switzerland 13,128 16,193 23,432 27,951 30,661 33,093 34,182 25,954 19,938 13,796 United Kingdom 10,358 8,954 10,507 23,461 25,825 22,013 21,525 14,852 10,863 4,971 Total, Western Europe 2,957 4,258 4,433 8,164 10,089 11,392 12,425 9,612 8,369 2,471 EU membersa 3,323 5,348 5,501 11,090 14,295 10,193 11,504 9,051 7,935 2,256 Other Western Europea 2,395 2,689 2,999 2,908 2,799 22,448 20,325 23,927 18,664 13,316 c. By product (constant 2000 US$ millions) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Barley 1,649 2,866 3,627 2,072 5,456 1,707 6,532 6,361 1,621 120 50 Beef 2,529 7,192 7,950 7,914 6,252 13,145 19,075 16,169 12,144 9,877 6,177 Eggs 604 817 534 893 1,808 1,436 1,673 1,037 571 130 123 Maize 57 557 1,323 1,396 2,715 2,204 3,589 3,297 1,394 1,036 1,020 Milk 20,793 23,917 28,101 31,074 45,150 41,790 37,896 32,811 22,919 15,418 2,822 Oats 405 1,229 1,611 1,081 2,027 229 812 849 335 144 70 Oilseeds 83 23 27 29 84 66 2,833 1,899 44 32 41 Pig meat 3,936 8,899 14,823 14,400 15,941 15,699 7,011 4,732 5,175 5,401 4,051 Potatoes 2,653 2,804 2,841 2,518 5,846 2,802 1,127 1,221 912 465 735 Poultry 1,863 1,973 1,987 3,229 3,717 4,060 4,337 5,687 4,492 3,424 4,042 Rice 246 267 62 578 126 104 529 598 220 84 12 Sheep meat 1,194 1,562 1,997 3,195 5,274 4,064 3,163 2,581 1,432 1,087 1,458 Soybeans 0 0 -- 2 2 1 571 337 0 0 0 Sugar 2,354 2,433 4,151 518 4,772 4,087 5,091 4,807 4,229 3,232 1,351 Tomatoes 44 387 710 1,046 1,539 1,345 887 356 44 75 0 Wheat 1,706 4,943 6,559 1,229 304 6,465 9,849 9,114 2,039 546 126 Wine 1,630 998 568 681 664 847 399 1,707 721 240 267 Wool -- -- -- -- 26 2 18 16 11 8 9 Total, Western Europe 36,784 56,703 75,678 65,415 101,451 99,844 105,391 93,581 58,303 41,319 21,916 EU membersa 19,158 36,948 55,749 47,291 88,543 88,805 84,016 77,079 53,218 37,773 19,033 Other Western Europea 17,626 19,755 19,929 18,124 12,908 11,038 21,375 16,502 5,086 3,546 2,883 Source: Anderson and Valenzuela (2008) based on authors' spreadsheet. Note: -- not available. a. The EU is the original 6 countries to 1972, 9 countries to 1985, 12 to 1994, 15 to 2004, and 27 thereafter. The EFTA countries are the original 7 to 1970; then 8 (with Iceland) to 1972; then 6 from 1973 (when the United Kingdom and Denmark joined the EEC) to 1985; then 5 from 1986 (when Portugal joined the EC) to 1994; then 3 after Austria, 157 Finland, and Sweden joined the EU in 1995. For 2005­07, the "EU members" and "total Western Europe" aggregates also include the most recent 12 countries that joined the EU in May 2004 and January 2007. 158 Table 3.6. CTEs of Policies Assisting Farmers, Covered Products, Total and Per Capita and by Product, Western European Countries,a 1956­2007 a. Aggregate CTE by country (percent) 1956­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Austria 74 88 77 22 24 22 63 88 39 35 17 Denmark 42 44 39 41 95 79 70 48 38 32 12 Finland 113 123 132 91 93 45 202 201 46 40 17 France 49 52 75 50 54 76 61 45 34 30 15 Germany 65 95 107 67 67 82 59 41 35 30 13 Iceland -- -- -- -- 42 72 272 210 117 95 104 Ireland 25 38 45 43 125 126 115 73 56 46 20 Italy 5 33 52 35 37 58 46 35 29 26 10 Netherlands 52 89 118 82 96 111 83 56 47 38 18 Norway 272 282 285 276 230 70 93 116 114 111 85 Portugal 1 11 18 11 34 21 37 33 25 24 16 Spain 19 20 19 4 0 5 36 30 24 21 13 Sweden 104 106 134 76 108 90 124 109 46 44 23 Switzerland 275 271 261 253 223 97 245 215 143 144 83 United Kingdom 69 65 46 34 71 89 77 54 44 40 22 Total, Western Europe 53 65 74 49 59 70 65 49 37 33 17 EU membersa 42 62 80 54 60 78 59 42 34 30 15 Other Western Europea 72 70 62 39 51 37 119 141 133 131 84 b. Total CTE per capita (constant 2000 US$) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Austria 95 95 26 54 44 1026 354 145 108 46 Denmark 220 211 276 624 498 343 299 221 165 55 Finland 291 353 327 413 181 699 665 717 525 207 France 246 353 283 380 367 259 229 159 115 55 Germany 235 311 261 362 330 211 157 110 82 33 Iceland -- -- -- 1070 969 937 465 310 277 379 Ireland 94 114 252 792 696 495 435 299 221 65 Italy 125 218 165 226 243 178 162 117 87 33 Netherlands 228 322 290 439 408 271 245 188 129 48 Norway 319 367 446 593 333 374 447 326 241 226 Portugal 19 41 29 121 58 107 134 105 84 48 Spain 32 34 29 0 6 146 150 108 89 56 Sweden 256 346 254 457 311 303 284 138 119 52 Switzerland 708 704 843 790 627 820 792 486 354 266 United Kingdom 185 135 136 302 272 188 157 111 86 41 Total, Western Europe 162 205 198 306 276 226 207 138 103 46 EU membersa 161 232 240 339 319 209 178 128 96 41 Other Western Europea 165 156 119 193 128 293 488 422 310 254 Source: Anderson and Valenzuela (2008) based on authors' spreadsheet. Note: -- = not available. a. The EU is the original 6 countries to 1972, 9 countries to 1985, 12 to 1994, 15 to 2004, and 27 thereafter. The EFTA countries are the original 7 to 1970; then 8 (with Iceland) to 1972; then 6 from 1973 (when the United Kingdom and Denmark joined the EEC) to 1985; then 5 from 1986 (when Portugal joined the EC) to 1994; then 3 after Austria, Finland, and Sweden joined the EU in 1995. For 2005­07, the "EU members" and "total Western Europe" aggregates also include the most recent 12 countries that joined the 159 EU in May 2004 and January 2007. 160 Distortions to Agricultural Incentives: A Global Perspective the role of instrument changes in improving efficiency in meeting the various pol- icy objectives. Discontinuities in the trends can also come from the impact of changes in ideas among domestic political forces that impact farm policy in Western Europe, the introduction of international rules and institutions, and the impact of bilateral and other trade agreements. Long-run trends Several important trends can be identified as having shaped Western European agricultural policy in the postwar period: the production increases in European agriculture that have transformed the sector from a significant importer of temperate-zone agricultural products to a major exporter; the changing emphasis over time away from the production of undifferentiated farm commodities and toward quality and other attributes that allow for product differentiation; the increasingly globalized market that both provides raw materials and intermediate inputs for the European food system and offers expanding access to consumers in other countries; and the increasing concern with environmental and other objec- tives that are constantly changing the nature of the coalition of interests necessary for agricultural policy support. The reinstrumentation of agricultural policy has been a consequence of trends away from import protection that allowed the sector to have a sheltered environment for growth and toward direct payments that are supported as rewards for socially responsible farming practices. Intervention in the market and the dumping of surpluses abroad has largely given way to a regulatory approach that encourages quality foods that have a ready market and protection of locational and other indicators associated with quality attributes. The reinstrumentation itself has been largely responsible for the moderation of the levels of protection relative to nonagricultural sectors and for less consumer taxation mentioned above. Production increases Somewhat ironically, it is the strong growth of agricultural production in Western Europe that brought about the most intractable policy problems. Policies intro- duced in the EEC were premised on the fact that agriculture needed support because it could not compete with domestic industry for labor and capital or with production from overseas farms that were larger and perhaps less constrained by social and environmental factors. The instruments used were designed to insulate domestic producers from developments in world markets. World prices were seen as too volatile to act as a basis for stable domestic markets. When prices were propped up by imports entering over the protective wall, this system was at least manageable, despite its inefficiency. But when production outstripped demand, Western Europe 161 surpluses began to accumulate. New instruments had to be developed, while those that intended as occasional market supports undertook new functions. Pressures arising from production increases were first noticed in the cereals market. As a result of the initial decision on cereal prices, production of wheat, barley, and maize (corn) increased sharply. Use of grains for animal feed dropped rapidly, leading the search for alternative feeds to begin in earnest. One external factor played a significant role in the development of such alternatives: the level of protection of several nongrain feedstuffs was fixed at zero (or otherwise very low levels) in the Dillon Round of the GATT, which concluded in 1962. Feed com- pounders in Europe found that soybean meal (from imported soybeans) and cassava chips (from Thailand and Indonesia) made a cheaper feed for livestock than domestic corn and barley. The expanded imports that resulted exacerbated the domestic supply imbalances, adding to the growing stocks of cereals.36 Dairy prices had also been set high in the EEC, encouraging production and discouraging consumption. Structural change in the dairy sector added to the increase in production.37 Surpluses showed up early, and various mechanisms were suggested to get rid of the "butter mountains" that had resulted from inter- vention buying. Taxes on the use of nondairy products were suggested but not approved. Subsidies for the use of skimmed milk powder in animal feed were effective but did not discourage production. Modest price cuts (through corespon- sibility levies) also were too small to curb production. Finally, in 1981, quotas on milk production were introduced. They have been in place ever since, and have been an effective, if inefficient, way to restrain production. Beef production also expanded rapidly in the EEC, in part as a reflection of the increase in dairy production (most beef in Europe comes from the dairy herd) and in part a reaction to hill farming and other livestock subsidies. But in the 1970s, consumption of veal slumped as a result of health concerns related to the use of growth hormones, and in the 1990s beef consumption plummeted follow- ing the discovery of a probable link between Bovine Spongiform Encephalopathy (BSE) in cattle and a new variant of Creutzfeldt-Jakob disease (nvCJD) disease in humans. So the policy response had to focus on reassuring consumers rather than discouraging producers from overproduction. Another product where surpluses appeared was sugar, a crop that has a very long production history in the region (sugar beets were grown for 150 years as a rotation crop for Northern Europe and provided a valuable calorie source in times of interuption in trade).38 But both France and the United Kingdom had former colonies that depended on the European market to sell their sugar cane. When the United Kingdom acceded to the EEC in 1973, a guaranteed market was conceded for its overseas suppliers. So the EU has struggled with an import commitment alongside a growing export surplus. When other exporters challenged the legality 162 Distortions to Agricultural Incentives: A Global Perspective of the EU's policy of reexporting the imported sugar, a panel found that the EU had exceeded its export subsidy commitment under the WTO. The sugar policy has now been reformed, by reducing support prices and paying compensation to growers and beet crushers, so that it is less intrusive on world markets. In each of these cases, the main problem was that prices were rigid downwards, largely as a result of the decision-making system. Farm ministers were reluctant to return from the annual price negotiations in Brussels having agreed to price cuts. In some cases, financial control was also lacking, largely because the impact of price decisions on spending were not apparent to those (the farm ministers) mak- ing the decisions. As the budget cost of export subsidies was shared among all members (and the import levies pooled), the true (foreign exchange) cost to each country was the internal price. Under these circumstances, it was not surprising that price cuts were rare. Productivity growth put pressure on the costs of the pol- icy and on the external impact, but the lack of financial accountability and the dif- ficulty of getting the political support needed for a major policy change led to an environment of continual crisis and controversy. Comparisons with other high-income countries The predominant feature of agricultural policy in Western Europe over the past 50 years has been the high level of support (relative to other temperate-zone pro- ducers) given to most agricultural sectors. Though agricultural protection in East Asia rose from low levels in the 1950s to be well above that in Western Europe (Honma and Hayami 2009), protection in Australia, Canada, New Zealand, and the United States was generally at a much lower level (Gardner 2009; Anderson, Lattimore, Lloyd, and MacLaren 2009). Including agriculture in the experiment of economic integration was a bold move for the EEC. The economic rationale was similar to that in other sectors: the agricultural sector would gain from rationalization on a continent-wide basis and the price of agricultural products would be reduced. In turn, this created the hope that an integrated European agricultural system could compete with the Americas and the temperate-zone producers of the Southern Pacific. But this would have required earnest competition among EEC farmers and halting of support for the less efficient of them. In general, the level of protection in Western European agri- cultural policy over the past 50 years can be attributed to the reluctance of politi- cians (and to a large extent, of society in general) to expose the sector to true competition. Integration of agricultural markets has indeed taken place, but in an environment where inefficiencies were able to survive and production decisions did not reflect the realities of the market. In short, the distortions that have con- tinued in European agriculture have been the result of the political reluctance to encourage needed adjustments in the sector. The social costs of such adjustments Western Europe 163 played a dominant role in shaping crucial decisions to undertake agricultural support--support that absorbed resources that could have had greater benefits to society elsewhere. The process of outmigration from agriculture has undoubtedly been slowed by the incentive structure of the CAP. Thus the policy has been successful in its own terms, by mitigating the adjustment cost, but expensive in societal terms by delaying changes that would have been beneficial. The unwillingness of governments to subject their agricultural sectors to the rigors of the market was also evident in the decisions of those countries that chose not to join the EU. The need to lower support prices in Norway and Switzerland in the event of membership was a major factor in generating opposition to EU acces- sion in rural communities. These high-cost countries have been reluctant to expose their farmers to competition from the EU and to induce the structural changes and entrepreneural initiatives that offer the chance for competitiveness. Each has protected its agriculture behind high tariffs, thus blunting market signals and burdening the downstream sectors with higher costs. Environmental and quality objectives The five decades discussed above cover a period of a major transformation of agricultural policies not just in terms of the instruments used but in the objectives that lie behind government intervention in the agricultural sector. The focus of agricultural policy has over the period swung away from a narrow focus on the income level of farmers toward a raft of policy objectives ranging from the preser- vation of the coutryside and the provision of healthy food to rural development and animal welfare. Farm incomes have over time become more of a constraint on policy change than a rationale for the existence of policy. Those who benefit financially from the policy have been able to slow down the shift in emphasis (and budget allocation) but not stop it altogether. The need to include incentives in the improvement of farm structures was recognized in the 1960s, when the Mansholt Report emphasized the dangers of persuing all objectives through price support. The introduction of a structural program proved slow, as it was in competition with funds for market manage- ment. Similarly, the need for funds for rural development was realized in part as a way of helping agriculture adjust. Eventually, the "second pillar" of the CAP was constructed in the Agenda 2000 reform and supplied with enough funding to make it a significant aspect of policy. National funds are also now authorized, subject to constraints, to be shifted from price and income support to rural development programs (modulation) that meet other objectives. Second pillar spending currently accounts for about 19 percent of total CAP expenditure. Environmental objectives have also become important in shaping the CAP and policies in the EFTA countries. Originally, the thrust was to prevent the negative 164 Distortions to Agricultural Incentives: A Global Perspective impacts of intensive farming on the countryside (and adjacent urban areas) and on soil, air, and water pollution. Over time, this has evolved into an emphasis on the positive contribution of agriculture as the dominant activity in much of the Europe's more scenic rural areas. In the 1990s, environmental objectives became associated with the term "multifunctionality," reflecting the range of public goods provided by rural businesses. But other exporters pointed to the risk that this could become an excuse for a blank check to policy makers who were looking for reasons to continue income support in a more acceptable guise--thus perpetuat- ing distortions to resource allocation. The first evidence of major changes to the politics and political economy of the CAP came as a result of public concerns over food safety. Building on public reactions to reported instances of human exposure to hormones used in live- stock production, the EU banned the use of such substances in animal produc- tion. In addition to the international ramifications of this decision (U.S. and Canadian beef and beef products were banned), the impact on domestic policy was palpable, as it ushered in perhaps the most significant change in the percep- tion of the CAP, its link with food quality, and the production practices employed by farmers.39 Advocates of more emphasis on local foods and on more animal-friendly livestock practices joined with those who opposed the use of transgenic technology in foodstuffs to form a formidable new voice in agri- cultural policy. Alongside this trend, food safety and quality policies and regula- tions on animal welfare have become increasingly centralized. The policy has itself been redefined as one that emphasizes quality over quantity and the pro- motion of foods that reflect the cultural richness of rural Europe. Farmers spot- ting this trend have shifted their own production and marketing practices toward these new demands. While some argue that this market orientation reduces distortions, others might regard the policy-supported changes in demand as distortions for the consumers who cannot or do not wish to shift their eating habits in this way. Reinstrumentation The original policy instruments chosen by the EEC were intended to provide a stable price environment for the agricultural sector. The variable levy smoothed out the fluctations that would have otherwise been transmitted to the domestic market. The export subsidies (restitutions) allowed goods to be withdrawn from the domestic market and sold at lower world prices. Only dairy (after 1981) and sugar (strictly speaking, a "temporary" market order) were subject to supply con- trols. Some producer subsidies existed, including those for durum wheat and olive oil, but these were also backed up by tariff protection. Wine was regularly taken off the internal market for distillation, and producer groups could withdraw fruits Western Europe 165 and vegetables with assistance from the European Commission if prices fell to "crisis" levels. The changes that have come about in the instruments used in the CAP are as important in their impacts as the levels of protection themselves. Though these changes have not come easily, they have had a lasting impact on the shape of the CAP. The most significant modification of policy instruments came in 1992, with the MacSharry reforms.40 Direct payments in compensation for price declines added flexibility to the policy, as described earlier. But this transition also followed a pattern observed in other countries. The Unite States, for example, began to move from price supports to direct payments and to decouple those payments from production and prices during the 1980s (Moyer and Josling 1990). At the time of the Uruguay Round, it became apparent that one way to reduce overseas threats to the CAP was to make the instrumantalities more similar to those of trading partners. Since the mid-1980s, the gap between total producer support and that derived from market price support has widened, particularly after the 1992 reforms. Cur- rently, market price support accounts for just over one-half of total CAP support, compared to 90 percent 20 years ago. To the extent that the new instruments are significantly less trade distorting, they also are less distorting for the domestic economy. A similar trend is noticeable in Switzerland: the "decoupling" of pay- ments from output and price was introduced even though the absolute level of support is still greater in that country than in the EU. Though Norway has not increased its degree of decoupling, the absolute level was in any case lower as a result of subsidies aimed at keeping people in the northern regions of the country regardless of their levels of output. The decoupling of support from price and out- put has also allowed for a trend to attach payments to other policy objectives in these EFTA countries as it has in the EU. Turning points Though these trends have evolved since 1955, several turning points have been crucial in influencing the macroeconomic environment, which in turn affected the level of distortion in agricultural policies in Western Europe over the period. These include, for example, the increase in the price of oil in 1973 and changes in commodity markets and food prices. The oil price increase impacted the CAP in two respects. First, it increased costs to farmers in a way that was readily apparent and led to compensating price increases. Coming not long after the end of the transition period in 1967 (a time when policy makers were attempting to reduce price incentives) and the exchange rate changes of 1968 (when prices diverged again within the EU), 166 Distortions to Agricultural Incentives: A Global Perspective the pressure for price adjustment was difficut to resist. Moreover, the oil price increases coincided with the sharp rise in cereal and sugar prices on world mar- kets, thus fueling fears of longer-term shortages and masking the level of under- lying distortions that were present in the system. The opportunity to make use of the high price period to reduce the support prices was lost. Instead, support prices followed the world market upward, only to be left isolated when world prices declined again. The oil price hike also coincided with the accession of the United Kingdom to the EEC, a shift that was expected, based on the country's history of interna- tional openness, to bring a fresh look at European farm policy. Some countries, such as the Netherlands, welcomed this while others were prepared to resist (notably France but also Germany). However, the period of inflation and cur- rency movements that followed in the 1970s overwhelmed the possibility of pol- icy improvements and left the United Kingdom reluctantly accepting the CAP so long as the budget cost to the United Kingdom was constrained and it con- trolled the pace of adjustment of the "green" rate of exchange. This compromise lasted through the runup to monetary union, when policy prices in the EU became common again. Political changes also have had an impact on the CAP, and hence on the level of distortion in Western European agriculture.41 The fall of the Berlin Wall in 1989 and the subsequent unification of Germany offered an opportunity for the return of Central and Eastern European countries to the economic and political main- stream. Agreements were negotiated with Central and Eastern European countries that were deemed to be ready for accession, and agricultural products were included, albeit with some quantitative restrictions. All agricultural policy deci- sions since that time have had to be measured in light of their implications for the larger European market. This has acted as a disincentive to increase support levels in the existing EU. To an extent unmatched in most other developed regions, agricultural policy in Western Europe has been influenced by external pressures as well as by those of a more domestic nature. But a major turning point came in the mid-1980s as a direct result of this external pressure. Up to that point, decision making within the CAP always took place within the context of the state of world markets, which impacted budget costs in particular. But the Uruguay Round introduced, within the URAA, a legal framework that restricted the scope for domestic action. Through the negotiation and enforcement of tariff reductions and scheduled cuts in subsidies, the URAA also gave domestic policy makers a lever on the policies of other countries.42 Thus the CAP, along with the policies of other countries, was challenged after 1995, and policy changes that would have been mainly of internal interest became internationally significant.43 Western Europe 167 Prospects for Future Reform of Western European Policy The essence of the CAP is that it represents a bargain among member countries, as all countries in the EU maintain their own agricultural policy objectives even if they have ceded the ability to use many instruments to acheive those objec- tives. The CAP, then, is a result of the bargaining process within the EU as much as a coordinated and consistent European view of appropriate role for govern- ment in agriculture. Reform can come about as a result of changes in the terms of the bargain just as easily as a shift in the model of agriculture that underlies policy or the reactions to exogenous events. However, it is still possible to specu- late on the change in underlying attitudes to agriculture in the development of the CAP, and hence hazard a guess as to where those trends might be heading. In the light of the political economy constraints and incentives mentioned ear- lier, this section attempts to draw out the prospects for further policy reform in Western Europe. One issue is whether the consolidation of Western Europe in the EU, and by implication the extension of the CAP to all Western European agriculture, will be completed in the foreseeable future. This implies the entry of Norway and Switzerland to the EU. Thus far, Norway has resisted in part because its position as an oil and gas exporter has enabled it to have a high level of income as an "adjunct" to the EU market. Though investment could be stimulated by accession, it is not enough to convince the electorate to "share" its fish resources, subject its farmers to price decreases, and possibly force the modification of control and use of energy supplies. But the possible exhaustion of Norway's oil reserves and an autonomous reform of its agricultural policy (alongside a more acceptable fisheries policy in the EU) could change the situation. The level of agricultural protection, though, is still considerably higher in Norway than in the EU, and significant policy adjustments would be needed before Norway's agricultural sector could be assimilated into the CAP.44 In Switzerland, most aspects of policy, both industrial and agricultural, have been steadily converging with those of the EU. As a result, one would expect the economic aspects of accession to be less significant than the political aspects. Not only does Switzerland have a strong history of avoiding foreign entanglements, but its status as a participatory democracy poses problems for the transfer of sov- ereignty to the EU level. If a time comes when major sectors of the Swiss economy are harmed by exclusion from the EU, one might expect to see an application to join. But as in the case of Norway, the level of farm protection is still considerably higher than in the EU.45 An enlargement issue of a different kind is the prospect of the accession of Turkey to the EU, one that poses an economic and political dilemma for the EU. In this regard, the impact on the level of distortion in Western Europe may well be 168 Distortions to Agricultural Incentives: A Global Perspective significant. On the one hand, the geopolitical argument points to the benefits of offering membership to a strategically important country. On the other hand, Turkey has an income level of about one-third that of the average EU level, and its membership would add some 20 percent to the population of the bloc. The impact on the EU economy of Turkey's joining would thus be considerable (in a positive as well as a negative direction), with the challenge to agriculture being a major consideration. For these reasons, one might expect a long period of transition before Turkey is fully incorporated into the single EU market (see also Anderson and Swinnen 2008). More trade agreements are in the cards for the EU, as it consolidates its rela- tions with its neighbors in North Africa and the Middle East and extends its "neighborhood policy" to include Central Asia and the Caucasus. The current wave of trade agreements under negotiation also include arrangements between the EU and the countries of Latin America, through an EU-Mercosur trade pact. To avoid getting left behind relative to competitors in North America and East Asia, trade policy plans of the European Commission also include agreements with major Asian countries. Where they include improved market access for farm products, these agreements will tend to put downward pressure on agricultural distortions. The extent to which this pressure is reflected in real changes will depend on the political significance attributed to the free trade agreements them- selves. Domestic agricultural policy has proved resistant to many such trade policy developments, but not all. The Everything but Arms agreement that gave the least developed countries tariff- and quota-free access to the EU market included agricultural products, and led in turn to the reform of politically sensitive sugar policy in 2006. The CAP has always absorbed a major share of the EU budget, though that share, once above 75 percent, has now been reduced to about 45 percent. More countries are being included in the activities funded from the budget without cor- responding increases in the level of funding. While the prospect of substantially increased funds available for agricultural policy is slim, this in turn means that shifts in instruments toward the less distorting direct (decoupled) payments is increasingly difficult. More likely is the addition of funds for projects such as the development of biofuels (which could benefit agriculture in a more acceptable way, depending on how the perception of the environmental effects of such interventions evolve). The impact of the Uruguay Round on the CAP has been emphasized above. Negotiations surrounding the Doha Round, the WTO's follow-up to the Uruguay Round, have been underway since 2001. The Doha Round would cut tariffs, eliminate export subsidies, and reduce the scope for trade-distorting domestic support. Though the talks are presently on hold, there is still the possibility that Western Europe 169 they may be concluded by the end of 2009. If an agreement is eventually reached within the range of options on the table, the EU would phase out its export subsi- dies altogether by 2013 and significantly cut its ability to support prices. Though much of domestic policy is now sheltered in the green box (and thus not subject to reductions), the Doha Round would "lock in" those aspects of reform. Tariffs would be cut by about 60 percent, with some exceptions for the most sensitive products, such as dairy and beef. WTO dispute settlement panels have already found that the sugar and banana regimes were inconsistent with WTO obliga- tions: other countries will no doubt explore further challenges where they con- sider the CAP too distortive in world markets. One vital question with respect to the future of the CAP is whether the trend is toward a renationalization of the policy. There have been small moves in this direction, with flexibility built in to the direct payments. But the degree of decou- pling, for some commodities, is at the discretion of the member countries. How far this trend will go is a matter of political decisions well above that of the agri- cultural policy.46 So long as they do not influence competition, there is a logic to the renationalization of payments that are not related to farm output, and the possibility of an eventual division between the market mechanisms that would be controlled at the EU level and the payments for public goods that may revert to the national level. In sum, there could be a convergence of likely and desirable policy directions before 2020. A continuation of decoupling of support from price and output should over time significantly reduce distortions, though if such pay- ments are "recoupled" to other objectives there could be some economic cost. Trade agreements will reinforce this trend, as access to the EU market will be an important incentive for other countries to conclude such agreements. What, if any, are the policy lessons for developed, developing, and transition economies of the trends and turning points in Western Europe's agricultural poli- cies? In many ways, the EU is unique in having agreed at its founding on a system of internal free trade and a common policy for market management and external protection. In that regard, the lessons are unlikely to be directly applicable. But the development of the CAP does offer some general lessons. Among the most impor- tant of these has to do with the pitfalls of attempting to manage markets in such a way as to give farmers adequate incomes. Technical advances, demand shifts, and the natural responses by private actors to incentives offered by policy will under- mine the ability of policy administrators to achieve the required outcome. In general, the CAP is too inflexible to adapt to changes in the internal or external market, ministers of agriculture meeting periodically are unlikely to be able to adjust policy instruments and price levels in the face of such changes. So the pol- icy is always a step behind the market. In addition, the income streams generated by the policy ensure that there are always interests lined up against change. 170 Distortions to Agricultural Incentives: A Global Perspective The combination of inflexible decision making and a bias toward the status quo make it impossible for the CAP to keep its relevance over the decades.47 But external pressure through the GATT talks and the WTO, as well as a feeling that the CAP was no longer viable in its original form, have encouraged a series of changes that have shifted the nature of agricultural policy in Europe. The policy is still somewhat inflexible, and changes may be even harder now with 27 members, but the reduction of prices of most of the main products to near world market levels provides much more scope for creative policy making. The flexibility of direct payments, though it can be overused, gives hope for agricultural policy that is less distortive of world markets and less obtrusive of economic adjustments on the domestic front. The way in which this was accomplished should indeed be of interest to other countries. What Have We Learned? Three main conclusions emerge from this chapter. First, the level of distortion to the Western European economies resulting from direct and indirect policy inter- ventions in agriculture has declined in recent years. Protection of agriculture, highest in the 1960s, fell in the 1970s as world markets became tight and European agriculture appeared, temporarily, to be more competitive. But reaction to the period of high world prices and inflation at home led to a rapid increase in distor- tion between the mid-1970s and early 1980s. That the level of distortion in the EU remained high for several years attracted criticism from at home and abroad, prompting changes in policy that have led to steady decline and increased stability in the level since the mid-1980s. In the countries that remained outside the EU, distortions remain high and more variable, although in the last few years there have been signs of convergence with the EU. Second, the act of joining the EU has had a mixed impact on country levels of distortion, and hence on the total level of distortion in the region. In some cases the accession of new members has raised the protection level and in others it has decreased it. Protection levels in Western Europe increased as a result of the 1973 accession of Denmark, Ireland, and the United Kingdom. Agriculture in each of these countries expanded and their competitiveness with respect to world markets was eroded. The entry of Southern European countries in the 1980s also increased the overall level of protection by including countries such as Spain and Portugal, which had relatively low levels of assistance, in the CAP. But many of the countries that chose to stay outside the EEC in the 1960s were also those with high-cost farming sectors. The high levels of protection were not "imported"into the EU from countries such as Finland and Austria. Hence, the overall result of the incorporation of the many of the Nordic and Alpine countries has been to reduce Western Europe 171 overall protection in Western Europe somewhat when they eventually joined the EU. This is confirmed by the nature of the two major countries in Western Europe that have remained outside the EU (Switzerland and Norway), both of which have far higher levels of agricultural distortion than the EU. This suggests that their eventual accession could play some role in reducing distortions in these countries, even if not in the current EU countries. The accession of the 10 new EU members from Central and Eastern Europe posed--and continues to pose--a challenge for the CAP and the process of reform. Though producers in the EU expressed the fear that the competition from those countries would oversupply markets in the EU, particularly for products such as pig meat, the European Commission was more concerned with the implied com- mitment to distribute direct payments to the large number of farmers that were about to join the EU and come under the political umbrella of the CAP. As it hap- pened, prices rose steadily in the new entrants, but surplus production has not been a problem. Market disruption has been limited to isolated cases, and the east- ern half of Europe is rapidly becoming integrated with Western markets. The prospect of additional costs under the CAP proved a real stimulant for reform, as funds allocated for a EU containing 15 countries were to be spread over 27 coun- tries. So one can reasonably conclude that the latest EU enlargement has tightened the constraints already limting the CAP, even though it raised farm protection in Central and Eastern European countries (Anderson and Swinnen 2009). The third conclusion is that domestic policy operates in an international envi- ronment, even as domestic politicians proclaim their autonomy. The development of policy in Western Europe became an international issue in itself, and external pressures could not be ignored. This posed a major challenge for policy makers, who could have used the external pressure as a reason for positive change or as a negative force that calls for defenses that deflect its impact. For many years, the EU attempted the latter strategy, though since 1990 it has made significant use of bilateral and multilateral agreements to guide agriculture in a direction that makes it compatible with a more open trade system. So economic distortions have been reduced as compatibility issues with the international economic system have been resolved. The pace of such developments has been dictated by those who were gaining most from the unreformed policy, but in the end the political bene- fits that flowed from reforming the policy came to outweigh the costs of change. Though considerable distortion in Western European agriculture still exists, the path to reducing that distortion is opening up. Allowing farmers to respond to consumer demands not artificially stimulated by government regulation is the most desirable way to point agricultural policy. The rhetoric of the CAP has adopted this approach for the past five years, and the policy is slowly moving in that direction. 172 Distortions to Agricultural Incentives: A Global Perspective Notes 1. In a recent survey of European economic growth since 1950, Crafts and Toniolo (2008) conclude that incentive structures are a crucial explanation of comparative growth rates of the economies of Europe (east as well as west). 2. The countries of Western Europe, for the purposes of this chapter, include the 15 countries that were members of the EU (the EU-15) in early 2004 plus Iceland, Norway, and Switzerland. In terms of policy development, each country is considered to have abandoned its autonomous domestic farm policy when it joined the EU. Thus, 18 countries had independent policies at the start of the period, in 1955, but by 2004 the number of independent policies had dropped to four--the EU, Norway, Iceland, and Switzerland. The EU then expanded eastward to include eight Central European countries (plus Cyprus and Malta) in May 2004, and again in January 2007 to include also Bulgaria and Romania. Most of this chapter focuses on the period prior to the EU's expansion eastward, while the chapter by Anderson and Swinnen (2008) focuses on Eastern Europe and the former Soviet Union. 3. Comparable figures for the United States show a utilized agricultural area of 379 million hectares and 2.1 million farms, with an average size of 180 hectares. 4. The agricultural sector in the United States employs 0.7 percent of the civilian labor force and contributes 0.9 percent of GDP. 5. The economy of Western Europe as a whole expanded by 2.7 percent over 1955­2004. 6. For a discussion of the different reactions of the Western European countries to the mid- 19th-century period of relatively free trade, see Kindleberger (1975). Swinnen (2008a) emphasizes changes in social and political institutions (such as the expansion of voting rights, changes in landlord- tenant relationships, and the emergence of agribusiness organizations) that undoubtedly contributed to the policy environment. 7. The Enclosure Acts of the 18th century gave the United Kingdom a farm structure that allowed it to take advantage of the emerging technologies (mechanization) and farming practices of the period (Orwin 1949). 8. Output per hectare in Denmark more than doubled between 1880 and 1930, whereas the same measure stayed steady in the United Kingdom (Ingersent and Rayner 1999). Agriculture accounted for 45 percent of Denmark's GDP in 1880. 9. Rural incomes in Germany declined by almost 40 percent between 1929 and 1932. 10. Marshall Petain, in particular, had a vision of the agricultural destiny of France, reviving the notion of agricultural protection that had been promoted by Meline in the 19th century. 11. The enlargement of the EU to include 10 new members in May 2004 is discussed in the context of its influence on policy later in this chapter. Implications for the new members are dealt with in detail in Anderson and Swinnen (2009). 12. The abreviation EU is sometimes used even for the period before the transformation of the EEC into the EC and later into the EU. 13. See Josling (2008b) for a larger discussion on external influences on CAP reform. 14. The original six countries of the EEC were Belgium, France, Germany, Italy, Luxembourg, and the Netherlands. In economic terms, Belgium, Luxembourg, and the Netherlands had already agreed to an economic union (the Benelux Union) in 1948, and so were on the road to becoming a single unit. 15. The inclusion of agriculture in the free movement of goods within the EEC was at the insis- tence of the Dutch, who had struggled with the same issue in the formation of the Benelux Union. 16. This terminology was used for the cereal market regime. Some differences in instruments and nomenclature, however, were introduced in the other commodity-market organizations. 17. The world price over this period was about US$60 per ton. 18. The "non" from Charles de Gaulle in 1963 that ended the first set of talks about U.K. accession reflected a fear that the nature of the EEC, and in particular the CAP, would be compromised by the admission of a country that boasted of its "special relationship" with Washington. The United Kingdom was widely seen as a Trojan horse for U.S. policy interests, and its accession a sure recipe for continued pressure on the CAP. Western Europe 173 19. The United Kingdom appeared to be an exception to this trend, though Howarth's price sup- port measures did not reflect the full range of input subsidies that were introduced over those years. 20. The present study has a broader product coverage than the earlier attempts to measure protection. 21. Though independent data for the other two (much smaller) members, Belgium and Luxembourg, are not included in this study, it is likely that the distortions are similar in those two countries, as they have been in an economic union since 1922 (to which Luxembourg contributes just 4 percent of the union's population of less than 11 million). 22. Italy, however, had low rates of assistance through the 1950s. 23. Portugal was one of the countries studied in the study led by Kreuger, Schiff, and Valdés (1988). Its income level was below that of Brazil at the time, and the country had many of the features common to developing countries. 24. Price data for Switzerland and Norway for some products were not available prior to the latter 1970s so the NRAs for them were assumed to be similar to the earliest years for which data are avail- able. Hence the lesser degree of fluctuation in their lines in figures 3.2b and 3.3b. 25. The Kennedy Round of GATT talks (1963­68) took place over this time, and the main focus of the exporting countries was to constrain the protectionist tendencies of the CAP. In this they were largely unsuccessful. 26. The major political reason for the decision of Sweden and Norway not to join the EEC was the requirement for neutrality that each had enshrined in their constitution. Finland was even more con- strained, as it bordered on the Soviet Union, and was accordingly inhibited in foreign policy initiatives. Likewise, Austria was also constrained by the postwar treaties, and Switzerland by its determination not to join alliances or even multilateral organizations. Ireland stayed out both to keep in step with the United Kingdom, its major market, and to retain its neutrality. Ireland did not join EFTA in spite of the links with the United Kingdom, but it did retain some preferences into that market. 27. It may seem that the measure of assistance used in this study is unduly influenced by world price movements. But the NRA indicates the distortions at any particular level of world prices: if world prices change, then so do the distortions caused by policy. This phenomenon appeared again when food prices soared during 2005­08. 28. The European currency unit (ECU) was a basket of currencies used as a unit of account in the European Community prior to the launch of the euro in 1999. 29. The European Union was established in 1993 by the Maastricht Treaty. 30. Livestock protection was changed only incidentally, as tariffs and levies on pigs, poultry, and eggs had been tied to the levies on cereals. 31. Another member of EFTA, Iceland, did not participate in talks about joining the EU. The con- cern that that country would have to adhere to the Common Fisheries Policy has always been a major political hurdle. Lichtenstein, an independent country in a customs union with Switzerland, joined the EFTA-EU accord even though Switzerland did not. 32. Estimates for Iceland are included only from 1979, when comparable data became available from the OECD. 33. Among the other changes introduced in the Agenda 2000 reforms was the definition of a "second pillar" of the CAP to complement the market price and income support. This second pillar was to channel funds to rural development and be managed as a joint activity between the European Commission and the member states. An element of cofinancing was also introduced, as was the requirement to move some of the direct payments into this pillar (modulation). As spending on rural development has an ambiguous relationship with agricultural incentives (it is designed largely to offer attractive alternatives to farming), these second pillar funds are not be included here as distortionary payments. 34. The NRA was calculated with NPS payments and direct decoupled payments (both included and excluded) to help identify the importance of these policy changes. 35. The extent to which a payment is production-neutral (the degree of decoupling) differs depending on the way it is administered and the expectations of the recipient, so the categorization of 174 Distortions to Agricultural Incentives: A Global Perspective policies by the degree of production-neutrality is a hazardous endeavor. No attempt is made here to evaluate the extent of the distortion that might still be present from policies defined as decoupled from output price and quantity produced. Rather, the value the OECD estimates is included under certain categories of support as decoupled. For the years 1979­85, there was just one category, called "direct payments." From 1986, those payments are specified to comprise the OECD's items C (payments based on area planted/animal numbers), D (payments based on historical entitlements), F (payments based on input constraints), and G (payments based on overall farming income), all of which are defined in the OECD PSE-CSE Database. For 2005­07, those items were replaced by similar but newly defined items C to E. While the administration of the EU's direct payments policies varies among the member countries, data is insufficient to do more than assume the percentage points of NRA from decoupled payments is the same for all EU members. This categorization for economic purposes (and that for NPS assistance) should not be confused with the legal allocation of domestic support measures into the WTO's "colored boxes" in the context of international commitments. 36. In addition, the soybean oil produced along with the soybean meal was a major competitor for butter and for olive oil, increasing the surpluses of these crops. 37. Trade issues also played a role here, as the United Kingdom insisted on continued preferential access for New Zealand for butter as part of its accession arrangements. 38. Sugar beet production was developed and encouraged in the Napoleonic era as a food security strategy. 39. The emergence of this parallel justification for the CAP, as a guardian of a safe food supply, is described in Roderer-Rynning (2009). 40. A change in the oilseeds policy preceeded that in the cereals market, when a GATT panel ruled that payments to processors tied to the use domestic raw material were inconsistent with GATT obli- gations. This change allowed the EU to gain experience with a direct payment program. 41. The significance of EU enlargement in the Agenda 2000 and the 2003 Fischler reforms is explored in Swinnen (2008b). 42. Swinbank and Daugbjerg (2006) explore the links between the 2003 Fischler reforms and the WTO trade negotiations that began in 2001. 43. External influences on the CAP also included the many regional trade agreements concluded between the EU and other countries. In these agreements, domestic agriculture was usually sheltered from direct competition. But over time, the pressure to allow some market access became an additional stimulus to reform. 44. Much the same could be said for Iceland. 45. Due to the small economic size of Norway and Switzerland, however, the absolute value of the subsidies is much smaller than that of the EU. As a result, the international implications of such protection are less immediate. 46. In addition to the political nature of these decisions on funding the CAP, the institutional changes also have an impact. Thus, for instance, if the Lisbon Treaty were to be ratified, the European Parliament would have a right to codecision on major aspects of the CAP (though it would keep its consultative role for technical issues). See Roederer-Rynning (2009) for a fuller description of these changes. 47. The link between the CAP and the way in which decisions are made in the EU has been empha- sized by several writers. See Swinbank (1989) for an early discussion, and Pokrivcak, Crombez, and Swinnen (2006) for a more recent examination of the importance of decision-making institutions. References Anderson, K., and Y. Hayami. 1986. The Political Economy of Agricultural Protection. London and Sydney: Allen and Unwin. Anderson, K., R. Lattimore, P. Lloyd, and D. MacLaren. 2009. "Australia and New Zealand." Chapter 5 in this volume. Western Europe 175 Anderson, K., and J. Swinnen. 2009. "Eastern Europe and Central Asia." Chapter 6 in this volume. Anderson, K., and E. Valenzuela. 2008. Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007. Core database at http://www.worldbank.org/agdistortions. Avillez, F., T. Finan, and T. Josling. 1988. Trade, Exchange Rates, and Agricultural Pricing Policy in Portugal: The Political Economy of Agricultural Pricing Policy, World Bank Comparative Studies, Washington, DC: World Bank. ______. 1991. "Portugal." In The Political Economy of Agricultural Pricing Policy: Volume 3 Africa and the Mediterranean, ed. A. Krueger, M. Schiff, and A. Valdés, Baltimore: Johns Hopkins University Press. Corkhill, D. 1995. The Portuguese Economy Since 1974. Edinburgh: Edinburgh University Press. Crafts, N., and G. Toniolo. 2008. "European Economic Growth, 1950­2005: An Overview." CEPR Discussion Paper 6863, Centre for Economic Policy Research, London. Eurostat Database. 2006. http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home/ (accessed June 2006). FAO (Food and Agriculture Organization). 1973. "Agricultural Protection: Domestic Policy and International Trade." FAO Conference Document C 73/LIM/9. Rome: FAO. Gardner, B. 2009. "The United States and Canada." Chapter 4 in this volume. Gulbrandsen, O., and A. Lindbeck. 1973. The Economics of the Agricultural Sector. Uppsala: Almquist and Wicksell. Hallett, G. 1981. The Economics of Agricultural Policy. Oxford: Blackwell Press. Heidhues, T., T. Josling, C. Ritson, and S. Tangermann. 1978. "Common Prices and Europe's Farm Policy," Thames Essay 14, Trade Policy Research Centre, London. Honma, M., and Y. Hayami. 2009. "Japan, Republic of Korea, and Taiwan China." Chapter 2 in this volume. Howarth, R. W. 1971. Agricultural Support in Western Europe. London: Institute of Economic Affairs. Ingersent, K., and A. J. Rayner. 1999. Agricultural Policy in Western Europe and the United States. Cheltenham: Edward Elgar. Josling, T. 1970. "Exchange Rate Flexibility and the Common Agricultural Policy of the European Economic Community." Review of World Economics 104 (1): 57­95. ______. 2008a. "Distortions to Agricultural Incentives in Western Europe." Agricultural Distortions Working Paper 61, World Bank, Washinton, DC. ______. 2008b. "External Influences on CAP Reform: An Historical Perspective." In The Perfect Storm: The Political Economy of the Fischler Reforms of the Common Agricultural Policy, ed. J. F. M. Swinnen. Brussels: Centre for European Policy Studies Publications. Josling, T., and S. Harris. 1976. "Europe's Green Money." The Three Banks Review 109: 57­72. Kindleberger, C. P. 1975. "The Rise of Free Trade in Western Europe, 1820­1875." Journal of Economic History 35 (1): 20­55. Lieberman, S. 1995. Growth and Crisis in the Spanish Economy, 1940­93. London: Routledge. McCrone, G. 1962. The Economics of Subsidizing Agriculture. London: Allen and Unwin. Moyer, W., and T. Josling. 1990. Agricultural Policy Reform: Politics and Processes in the EC and USA. Hemel Hempstead, U.K.: Harvester-Wheatsheaf Publishers. ______. 2002. Agricultural Policy Reform: Politics and Process in the EU and the US in the 1990s. Aldershot, U.K.: Ashgate. Nash, E. F. 1955. "The Competitive Position of British Agriculture," Journal of Agricultural Economics 11 (3): 222­41. Nash, E. F., and E. A. Attwood. 1961. The Agricultural Policies of Britain and Denmark: A Study in Reciprocal Trade. London: Land Books. OECD (Organisation for Economic Co-operation and Development). 1967. Agricultural Policies in 1966. Paris: OECD. ______. 2006. Agricultural statistics database. http://www.oecd.org (accessed April 2007). ______. 2008. Producer and Consumer Support Estimates, OECD Database 1986­2007. http://www .oecd.org (online database for 1986­2007 estimates; OECD files for estimates using an earlier methodology for 1979­85). 176 Distortions to Agricultural Incentives: A Global Perspective Orwin, C. S. 1949. A History of English Farming. London: Thomas Nelson and Sons. Pepelasis, A., G. Yannopoulos, A. Mitos, and others. 1980. "The Tenth Member--Economic Analysis." Sussex European Papers 7, University of Sussex. Pokrivcak, J., C. Crombez, and J. F. M. Swinnen. 2006. "The Status Quo Bias and Reform of the Common Agricultural Policy: Impact of Voting Rules, the European Commission and External Changes." European Review of Agricultural Economics 33 (4): 562­90. Roederer-Rynning, C. 2009. "The Common Agricultural Policy." In Policy-Making in the EU, ed. H. Wallace, M. Pollak, and A. Young. Oxford: Oxford University Press. Roesener, W. 2000. "The History of German Agriculture." In Agriculture in Germany, ed. S. Tangermann. Frankfurt: Verlag. Snyder, L. L. 1945. "The American-German Pork Dispute, 1879­1891." Journal of Modern History 17 (1): 16­28. Swinbank, A. 1989. "The Common Agricultural Policy and the Politics of European Decision Making." Journal of Common Market Studies 27 (4): 303­22. Swinbank, A., and C. Daugbjerg. 2006. "The 2003 CAP Reform: Accomodating WTO Pressures." Comparative European Politics 4 (1): 47­64. Swinnen, J. F. M. 2008a. "The Political Economy of Agricultural Protection: Europe in the 19th and 20th Century." Paper prepared for the Project on Political Economy of Agricultural Distortions, World Bank, Washington, DC. ______. 2008b. "The Political Economy of the 2003 Reform of the Common Agricultural Policy." In The Perfect Storm: The Political Economy of the Fischler Reforms of the Common Agricultural Policy, ed. J. F. M. Swinnen,. Brussels: Centre for European Policy Studies Publications. Tracy, M. 1989. Government and Agriculture in Western Europe, 1880­1988. Hemel Hempstead, U.K.: Harvester Wheatsheaf. Tyers, R., and K. Anderson. 1992. Disarray in World Food Markets: A Quantitative Assessment. Cambridge and New York: Cambridge University Press. 4 United States and Canada Bruce L. Gardner* There is much in common between the agricultural sectors of the United States and Canada. The number of farms in North America has been falling for decades, a well-known trend that has given the public a sense that agriculture is a hard-pressed industry in decline. What is less well known is that the decline has greatly slowed over the past 20 years. During the 1950s and 1960s, the number of farms in the United States fell at the rate of 3 percent annually, declining by half in 20 years. Between 1995 and 2005, the number declined by only 4 percent for the whole decade (figure 4.1).1 To fully understand the trend, it is necessary to consider farms of different sizes separately. The number of farms with annual sales of more than US$500,000, for example, quadrupled between 1978 and 2005, from 18,000 to 79,000. The number of farms sales between US$100,000 and US$500,000 also increased during those years. The number of farms with sales of less than US$25,000 per year, however, has held quite steady. Farms with sales between US$25,000 and US$100,000 represent the largest portion of the total decline. Along with the decline in the number of farms, U.S. land dedicated to farming has also declined, at an annual rate of 0.4 percent between 1950 and 2005. Over those years, this has resulted in the loss of about 230 million acres of farmland. *Professor Gardner died in March 2008. The editing of this chapter (which does not include a discussion of the latest U.S. Farm Bill, but see Orden, Blandford, and Josling [2009] for more on the topic) was done as faithfully as possible given his original manuscript. The author was grateful for helpful comments from workshop participants, and for invaluable help with data compilation by Johanna Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, Damiano Sandri, and Ernesto Valenzuela. The working paper version of this chapter (Gardner 2008) con- tains additional background material and appendix tables. 177 178 Distortions to Agricultural Incentives: A Global Perspective Figure 4.1. Farm Household Income as a Percent of National Household Income, United States, 1930­2005 160 140 120 100 percent 80 60 40 20 0 30 35 40 45 50 55 60 65 70 75 80 85 90 95 00 05 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 Source: Gardner (2002, figure 3.12) and U.S. Department of Agriculture, Economic Research Service, Briefing Rooms (http://www.ers.usda.gov/Briefing/). Available cropland acreage, however, has remained relatively constant over this period, with 345 million acres harvested in 1950, versus 321 million acres in 2005. Considering the increase in irrigated and better-drained acreage, quality-adjusted cropland may even have increased slightly. Together with declining farm numbers, the acreage data imply an increase in the average size of farms. But as the decline in numbers has slowed significantly in recent decades, so has the growth in aver- age farm size. In 2005, the average U.S. farm had 444 acres (180 hectares); 30 years earlier, the average size was 431 acres (175 hectares). In Canada, the number of farms peaked in 1941. In the 40 years between 1961 and 2001, the number of farms declined by half, from 480,000 to 246,000. This is a rate of decline of 1.7 percent annually, and faster than the U.S. rate of 1.4 percent during the same period. Though Canada's land devoted to cultivated crops declined over the 1960­2000 period, it did so less rapidly than the number of farms: Canada had 111 hectares of cropland per farm in 1961, and 147 in 2001 (compared to about 75 hectares of cropland per U.S. farm in 2002). Yields and value of output per hectare of cropland were lower in Canada than in the United States, however, reflecting the generally cooler and drier climate of Canada. Total agricultural land declined much more slowly in Canada than in the United States--indeed, hardly at all, from 69.8 million hectares in 1961 to 67.5 million hectares in 2001. United States and Canada 179 Relative to the decline in the number of farms, the farm labor force in Canada and the United States continues to fall at a more rapid pace, mainly because there is less unpaid family labor. At the same time, the use of material inputs (fertilizers, fuels, and purchased feed additives) has doubled in the United States since 1950, leaving the aggregate input index calculated by the U.S. Department of Agriculture (USDA) remarkably constant, at 102 in both 1950 and 2000 (base 1996 100). But agricultural output has continued to grow at a steady clip, so total factor produc- tivity (TFP) growth has been an impressive 1.9 percent per year, with no evidence of slowdown in the trend in recent years despite energy price shocks, environmen- tal constraints, and concerns about exhaustion of TFP gains attributable to earlier breakthroughs in improved hybrid seeds and other innovations. So far, continued advances in genetics, livestock management, capital equipment, and economies of scale have kept the real cost of U.S. farm products on a pronounced downward path. These cost declines have been largely reflected in lower prices of farm prod- ucts and hence lower costs of raw materials for other foods. While the average price of farm products rose 40 percent between 1978 and 2005 (1.2 percent per year), the gross domestic product (GDP) deflator rose at an annual rate of 3.2 percent. Thus the real price of agricultural output fell by an average of 2 percent during this period--essentially the same as the rate of TFP growth. Overall farm size (in terms of total land and output per farm) has increased in Canada at a rate faster than in the United States since the 1950s. As of 2001, the average Canadian farm was 670 acres, compared to 440 acres for American farms. Larger farm size in Canada has not hindered output or productivity growth, though: real output per farm increased more rapidly than farm size, reflecting increased yields per hectare and a strong trend in total factor productivity growth, which is estimated to have increase 2.5 percent annually in recent decades (Furtan 2006). The fact that farm prices have fallen in real terms largely in parallel with cost decreases indicates that real incomes of farmers may not have benefited from farm productivity growth. Yet the incomes of people employed in agriculture have in fact grown, both in real dollar terms and relative to real incomes of the nonfarm population. From the 1930s to the 1960s (which includes the Great Depression and Dust Bowl years), farm households could be reasonably categorized as a low- income population. By the 1990s, farm and nonfarm household incomes had equalized. Since 2000, U.S. farm household incomes have been significantly higher than those of nonfarm households (figure 4.1). How has this income growth been achieved, given that the decline in real prices has equaled the decline in farmers' costs of production? The answer is the increas- ing importance of farm households' integration into the nonfarm economy, so that in recent years, off-farm income sources account for 85­90 percent of average 180 Distortions to Agricultural Incentives: A Global Perspective U.S. farm household incomes. These data suggest that income from farming itself may indeed be quite low. However, a full understanding of the farm income data requires consideration of differences between farms of different sizes. Numeri- cally, a majority of farms--87 percent in 2004--have less than US$10,000 annu- ally in sales. The costs of farming at this scale are such that these farms on average earned only US$1,020 from farming, and more than half are estimated to have losses from their farm enterprises. Nonetheless, the average household income of these farms is US$71,500, thanks to off-farm income. At the other end of the spec- trum, family farms classified by the USDA as commercial scale operations (those which have US$250,000 or more in sales) earned an average of US$145,300 from farming plus US$46,038 from off-farm sources, and these latter farms produce more than two-thirds of U.S. agricultural output.2 On a per-farm basis, net income from farming in Canada during 2001­05 was almost the same as in the United States, although net farm income in Canada, as in the United States, was highly variable from year to year.3 On average, Canadian farm households receive less off-farm income than in the United States. Agricultural Policies in the United States Legislative proposals in the United States to improve the economic situation of farmers through governmental intervention in commodity markets were first devel- oped conceptually as remedies for the precipitous fall in product prices in the after- math of World War I. After several failed attempts to enact farm support bills in the 1920s, the further decline in agriculture's situation with the onset of the Great Depression of 1929­32 led to legislative success in the landmark New Deal pro- grams beginning in 1933. These programs had the principal purposes of increasing the incomes of farm people (with an explicit goal stated in terms of "parity" of farm income with reference to a pre­World War I standard) and stabilizing farmers' rev- enues. The fateful choice was to attempt to achieve both these goals by means of supporting farmers' prices received for a subset of commodities ("basic" commodi- ties in U.S. law--initially, wheat, corn, cotton, rice, tobacco, pork, and milk). The four main feasible means of market intervention in pursuit of higher pro- ducer receipts had been implemented for key commodities and were already in use in the 1930s: production controls, government purchases of commodities for stockpiling at price support "loan rates" at which farmers could forfeit the basic commodities instead of making repayment of the loan plus interest, disposal of surplus stocks through distribution or subsidized sale for export or to domestic consumers, and direct payments to producers. The first direct payments, in 1933, were tied to farmers' idling of land or destruction of livestock, so from their incep- tion they were not classical production-inducing subsidies. United States and Canada 181 In 1936, the U.S. Supreme Court ruled that the federal government had no authority to administer land-idling acreage controls under New Deal farm legisla- tion, on the constitutional grounds "that powers not granted are prohibited. None to regulate agricultural production is given, and therefore legislation by Congress for that purpose is forbidden." Subsequently, the Court's alleged respect for precedent was not extended to this decision, and many later production control measures have passed constitutional muster.4 At the time, the result of the Court's decision was a merger of prior concerns about conservation with measures to remove acreage from commodity production. This was done in the Soil Conserva- tion and Domestic Allotment Act of 1936 principally by defining "soil-depleting" crops (the main basic commodities) and "soil-conserving" crops (grasses and legumes), and paying farmers to substitute the latter for the former. Trade policy was almost negligible in the 1930s' programs. The policy propos- als of the 1920s, in contrast, had given a central role to export promotion. The Smoot-Hawley Tariff Act, signed into law in June 1930 (despite a petition to Pres- ident Hoover signed by more than 1,000 economists asking that he veto the bill) protected manufacturing much more than agricultural products. It is notable, however, that the crossover Democratic votes needed to pass the bill in the Senate (the Republicans were then the main protectionists and the Democrats free traders) were lured by the desire to protect sugar (Louisiana), wool (Wyoming), and Florida fruit (Benedict 1953). By 1934, it was clear that even these elevated tariffs were insufficient to protect sugar, and the first Sugar Act added sugar to the list of basic commodities, authorizing import and production quotas. With respect to overall agricultural trade, by 1934 the combination of the Great Depres- sion and trade restrictions in the United States and elsewhere reduced both U.S. agricultural exports and imports to about US$600 million per year each, one- sixth of the levels of 1920. Experience during and after World War II made it apparent that export demand was capable of creating farm prosperity to an extent, and with far less cost and turmoil, than a decade of intensive effort by the federal government had been able to deliver in the 1930s. Under the Marshall Plan, U.S. exports of food- stuffs were 19 million tons annually in 1947­50, compared to 4 million tons in 1935­39. However, three aspects of the situation in the United States were obstacles to free-trade ideas in agriculture. First, the Marshall Plan and subsequent agricul- tural exports were in large part financed by subsidies rather than being bought abroad at world market prices. Second, U.S. commodity policy held some domes- tic commodity prices above world levels, so that import restrictions were vital to these policies (otherwise, the program would have to support the world price, not just the U.S. price). Third, the relatively few U.S. importable farm commodities-- notably sugar, dairy products, and some meats and fruits--had sufficient political 182 Distortions to Agricultural Incentives: A Global Perspective power to maintain protection via tariffs or quotas even when giving them up as part of a larger trade liberalization agreement would have been beneficial to the nation as a whole. Section 22 of the Agricultural Adjustment Act of 1933 required import quotas to be imposed if imports threatened the effectiveness of a price support program. This situation led the United States to join Europe in pressing for a waiver of agri- cultural products from General Agreement on Tariffs and Trade (GATT) members to reduce export subsidies or provide increased import access to their markets. Some experts argued vigorously for changing U.S. farm programs to make them compatible with liberal trade, notably in the "Brannan Plan" of 1948 and other proposals to replace production controls and price supports by payments to farmers and adjustment assistance (Johnson 1950). But U.S. policy did not turn to favor- ing the inclusion of agriculture in the GATT until the 1960s. By then, European farm policy had taken a decisively protectionist path that precluded significant agricultural trade liberalization. Beginning with the Agricultural Trade and Development Act of 1954 (P. L. 480), U.S. policy followed a path suggested by the Marshall Plan of using food aid to foreign countries as a mechanism for surplus disposal. During 1956­64, about one-fourth of U.S. agricultural exports were shipped under this program. Though P. L. 480 exports have had varying degrees of concessionary pricing, depending on the status of the importing country, overall there has been a substantial subsidy element in these exports. Since the 1970s, the program has shipped a fairly con- stant amount of just under US$1 billion in commodities annually. In addition, going back to 1935, government-provided export credit and guarantees of repay- ment to private sector lenders have been used to stimulate foreign demand for U.S. commodities. Exports have also been promoted through USDA grants to farm or commercial interests for the purpose of informational and sales efforts abroad. Revamped as the Marketing Assistance Program in 1990, spending under such programs still amounts to about US$100 million annually. More explicit export subsidies were paid through most of the post­World War II period, most notably in wheat, where their role was negotiated under the Inter- national Wheat Agreement starting in 1949.5 In the 1960s, more than 85 percent of U.S. wheat exports were assisted by subsidies. Then, during the worldwide commodities boom of the 1970s, it appeared that the era of export subsidies might be replaced by commodity scarcity, with trade policies restraining rather than subsidizing exports. The United States embargoed grain shipments to the Soviet Union and a few other countries for short periods during 1974­80. Com- modity scarcity proved temporary, however. The worldwide collapse in commod- ity prices of the 1980s provided the stimulus for export promotion programs. The European Community intensified its longstanding practice of export subsidies (Josling 2008). The United States began offering specific-destination export United States and Canada 183 subsidies in retaliation in the early 1980s, regularizing this approach in the Export Enhancement Program (EEP) established in 1983 under the Reagan Administra- tion. Where feasible, Canada met the subsidy competition through the pricing policies of the Canadian Wheat Board (CWB). The EEP, like the pre-1970s export subsidies, was first and foremost a wheat program. It began as low-price government sales of wheat stocks held by the Commodity Credit Corporation (CCC) as a result of the domestic price support programs to North Africa in 1983. The EEP was complicated, using a payment-in- kind approach. USDA would determine particular countries and commodities for which it believed export subsidies would be helpful in selling U.S. products. Exporters would then negotiate a deal with a foreign buyer at a price discounted from prevailing world trading prices. The exporter would then apply to USDA for a payment sufficient to make up the difference between the market price and the negotiated discount price. USDA, if it approved the sale, would give the exporter sufficient wheat from CCC stocks to cover the payment, called the export "bonus." By the late 1980s, the bonuses were totaling US$1 billion annually, with more than 80 percent of EEP commodities accounted for by wheat in 1985­89. The program was widened and generalized so that CCC wheat stocks could be used to subsidize exports of other commodities, and in 1990, when available CCC wheat stocks were exhausted, in-kind bonuses were replaced by cash. As the United States expanded its export subsidies in the 1980s, Europe and Canada, within their respective support structures, met the competition with increased export subsidies of their own. The Uruguay Round of GATT negotia- tions was a natural venue for mutual agreement to rein in this costly competition. After long and tortuous negotiations, the finalized Uruguay Round Agreement on Agriculture (URAA) contained disciplines that, together with strategic rethinking and changed grain market conditions, greatly reduced the role of export subsidies after the mid-1990s in both the United States and the European Union. Both have agreed such subsidies should be outlawed, which has allowed farm policy discus- sion in the Doha Round of the GATT's successor, the World Trade Organization (WTO), to focus primarily on market access (import protection via tariffs and quotas) and domestic price supports via subsidy payments to producers. Import protection measures A hundred years before the first domestic commodity support programs were enacted, international trade policy was already a hot political issue in the United States. Manufacturing interests of the North wanted protection and, after 1820, received it in the face of opposition from Southern agricultural interests, who bought imported manufactured producer and consumer goods and also linked their capacity to export some products, especially cotton, to the U.S. willingness to 184 Distortions to Agricultural Incentives: A Global Perspective import from Europe. The North, to succeed politically, needed an alliance with the West, which was in place until the Jackson presidency, when his vetoes of legisla- tion for Western improvements changed the balance of interests in favor of lower tariffs--ultimately, the "free trade tariff " legislation of 1846, which generated average tariff rates of 25 percent in 1850 (calculated from tariff receipts as a per- centage of the value of dutiable imports), as compared to 57 percent in 1830. Higher tariff protection resumed after the Civil War, especially on manufactured goods, with average rates of about 40 percent between 1870 and 1910. Through- out this period, there were tariffs on imported agricultural products such as wool and sugar, but rates averaged about five times as much for manufactured imports as for agricultural imports (Davis, Hughes, and McDougall 1965). Table 4.1 shows the evolution of U.S. protection as measured by customs duties as a percentage of the value of imports. This measure does not always provide a Table 4.1. Customs Receipts as a Percentage of Value of Imports, United States, 1821­2000 (percent) Manufactured Agricultural All products products merchandise 1821 47 47 43 1830 64 62 57 1840 21 00 18 1850 27 3 25 1860 19 2 18 1870 46 47 50 1880 39 6 29 1890 41 8 30 1900 40 15 28 1910 29 9 21 1920 7 2 6 1930 18 8 15 1940 15 7 13 1950 9 2 6 1960 -- -- 7 1970 -- -- 7 1980 -- -- 3 1990 -- -- 3 2000 -- -- 1.6 Sources: Davis, Hughes, and McDougall (1965) for manufactured and agricultural products; Carter (2006) for all merchandise. Note: -- not available. United States and Canada 185 good indicator of the trade effects of tariffs, however, notably because a tariff so high as to shut off all imports generates no customs duties and so counts the same as free trade in calculating the numerator of this measure. Irwin (2006) reviews this and other shortcomings of that table 4.1 indicator, and estimates a trade restrictiveness index that takes into account the effect of a tariff on imported quantities and the fact that the distortive effects of a tariff increase more than pro- portionately with the height of the tariff. His estimates indicate that for all mer- chandise trade, while customs duties as a percentage of all imports had fallen to 60 percent of the 1875 level in 1931, the TRI in 1931 remained at 97 pecent of its 1875 level--that is, the reduction in the crude measure greatly overstates the extent of liberalization. But between 1931 and 1960 the TRI fell faster than the crude measure. By 1960, the extent of overall trade liberalization between 1875 and 1960 was about the same for both measures. That is, by either measure, the restrictivenes of tariffs in the post­World War II period was about one-fourth the level of the late 19th century (Irwin 2006). Because the United States has always been a net exporter of agricultural goods, and imported manufactured goods are used directly and in the production of inputs used in farming, the tariff structure effectively taxed agriculture through- out the 19th century. The estimates of Irwin (2006) indicate that for the 1870­1900 period, an average 30 percent protection on imported goods (duties as a percentage of all import values including duty-free imports) generated a net subsidy to import-competing manufacturers of 15 percent and a net tax on agri- culture (and other exporters) of about 11 percent.6 This situation was reversed by increasing protection (through both import restrictions and domestic support) of agriculture after 1920. Import duties on wheat, maize, wool, sugar, and meat were raised sharply in "emergency" legislation of 1921 and the Tariff Act of 1922 when "the representatives of the agricultural states had committed themselves to a policy of high and even ruthless protection" (Taussig 1931, pp. 452­53). At the same time, imports of many manufactured products used in farming were made duty-free.7 After 1910, U.S. manufacturing became sufficiently export-oriented itself as not to be the strong political force for protection it once was, and after World War II protection of manufacturing steadily declined to the point that by 2000, import duties as a percentage of aggregate import value had declined to 1.6 percent (albeit with additional signficant protec- tion of politically sensitive sectors such as textiles through nontariff barriers). Tar- iffs on agricultural products were also reduced, though protection from imports increased due to quantitative restrictions. In short, between the broad periods of 1820­1900 and 1930­2000, there was a substantial turnaround in manufacturing as compared to agricultural import protection. However, it was already apparent in the 1920s that too much agricultural production was exported to make import 186 Distortions to Agricultural Incentives: A Global Perspective protection effective in alleviating farmers' losses in the post­World War I price plunge, and this changed the focus of farm policy in terms of other measures. Overall, protection of agricultural products relative to merchandise has increased over the past 50 years. Though rates of import protection of the agricul- tural and manufacturing sectors were equal in the late 1950s and 1960s, protec- tion of manufacturing has since fallen faster than assistance to agriculture. It should be noted, however, that comparing market-distorting agricultural protec- tion with merchandise tariff rates has become decreasingly relevant in capturing the main U.S. import protection elements in either agriculture or manufacturing, because in agriculture and in industries such as steel, automobiles, and textiles, the more important distortions of international trade have become quantitative restrictions, often in the form of "voluntary restraint" agreements between the United States and exporting countries. More economically relevant measures compare internal U.S. prices with international prices for the same goods, as dis- cussed below. Nonetheless, the picture remains essentially the same--that govern- mental action to assist agriculture is increasing in impact compared to action to protect manufacturing. Export subsidy measures An indicator of the role of export subsidies can be obtained from the value of sub- sidies paid per unit quantity of exports. In the United States, these have been most significant for wheat, as noted above. Subsidies were as high as US$1.3 billion for exported wheat in 1993 under the EEP. They were targeted to particular sales, however, and not available for all exports. This raised questions about their efficacy in actually increasing export quantities. By focusing subsidy funding on particular sales, the government's expenditures were made more effective at increasing those targeted sales. Subsidized wheat received payments as high as US$43 per ton in 1991. But at the same time, large sales to other importers, notably Japan, were not subsidized at all. Because of the targeted nature of the subsidies, the proper measurement of the subsidy, not to mention its effects on net exports, is difficult to nail down (as is the effects of the wheat export embargoes of the 1970s).8 The USDA budget provides information on the range of export promotion activities. In fiscal year9 2005, as in the several years prior, there were no export subsidies under the EEP or the more recent Dairy Export Enhancement Program. A total of US$2.2 billion was spent on USDA's Foreign Agriculture Service (FAS) programs, but US$1.7 billion of this was on food aid programs. Market development programs and export credit guaran- tees are export promotion activities that also have features making them similar to export subsidies. In fiscal 2005, FAS spent US$184 million on market development programs such as sending teams abroad to make the case for U.S. commodities United States and Canada 187 and informational campaigns in the United States and abroad. Export credit guar- antees covered US$2.6 billion in sales during fiscal 2005 (at a budgetary cost of US$137 million, which is a rough indicator of the subsidy element of the pro- gram). In response to a WTO dispute resolution panel in 2005, the United States has eliminated some high-risk countries from the guarantees and made other changes "intended to remove any long-term subsidy component of the program" (USDA 2006, p. 38). Unlike import restrictions, whose effects can be estimated by comparing internal and world reference prices because they are focused on a few commodities and create big effects, these export programs are spread so thinly across many commodities that it would not be credible to attribute to them any observed elevation in U.S. commodity prices relative to foreign prices for the same commodities. Rather, it makes more sense to treat such expenditures together with domestic programs as elements of overall non-product-specific (NPS) support, following Organisation for Economic Co-operation and Development (OECD) practice. Direct producer support through commodity programs Payments to producers have been a central element in U.S. agricultural policy since the first New Deal programs of 1933. Figure 4.2 shows payments in billions of U.S. Figure 4.2. Expenditure on Commodity Programs and Payments to Farmers, United States, 1955­2005 45 40 35 2000 US $ billions 30 25 20 15 10 5 0 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 spending level payments to farmers Source: Author's compilation from the Congressional Budget Office (CBO), Washington, DC. 188 Distortions to Agricultural Incentives: A Global Perspective dollars in real terms using the GDP deflator of 2000 100. For comparison, the figure also shows the total level of spending on "commodity stabilization and sup- port" as measured by the U.S. Office of Management and Budget. Spending levels are higher than payments prior to 1990 because much of the spending was for removing products from the market using stockpiling programs, purchases for sale to schools and other food assistance, or subsidized sales abroad. In recent years, spending has been overwhelmingly dominated by direct payments to producers.10 Summarizing the effects of U.S. direct support programs is difficult because of the variety of policy instruments used and their evolution toward being increas- ingly decoupled from production decisions over time. Throughout the 1950s, the price suppport loan rates at which farmers could forfeit basic commodities to the CCC remained a primary policy instrument, and were reinforced by payments for idling land. By the 1960s, lower loan rates and direct payments to reduce forfeitures at a supported price entered the policy arsenal. In the 1970s, direct payments were further institutionalized in the form of "target prices" and "deficiency payments" to farmers when the price of basic crops they produced fell below the target. By the 1980s, deficiency payments were being made on only 85 percent of a farmer's base acreage, and for a historical level of "program yields." Deficiency payments thus came to be somewhat decoupled from production decisions, although the specific basic crop for which payments were received had to be grown on the correspon- ding base acreage. Farmers also had to comply with annual acreage reduction requirements to be eligible for deficiency payments. Annual acreage reductions idled at an average of 46 million acres during 1983­88 as world agricultural prices collapsed in the mid-1980s, and then fell to much lower levels. Income support in the 1996 and 2002 Acts The most important change in support to farmers contained in the 1996 Farm Act was its Title 1, the Agricultural Market Transition Act (AMTA). With rising world prices for U.S. farm commodities in 1995 and 1996, AMTA replaced target prices, deficiency payments, and annual acreage set-asides for wheat, rice, feed grains, and cotton by a scheme of fixed "production flexibility" payments. The payments were based on amounts farmers received, or would have received if they had partici- pated, in the pre-1996 deficiency payment program. The amount of the payment was independent of prices and was fixed by each farm's production history. It could not be increased or decreased by changes in the farm's acreage or yield of the pro- gram crops, while the program allowed farmers to plant a wide range of crops on base acreage (the reasons for the "production flexibility" label). The production flexibility payments were in this sense further decoupled from both prices and pro- duction, although payments were not divorced from all production decisions in United States and Canada 189 that producers lost payments if they increased plantings of nonsupported fruits or vegetables or if they left farming completely. The aggregate of payments was sched- uled to decline from about US$6 billion annually in 1996 and 1997 to US$4 billion in 2002. AMTA payments constituted the bulk of projected commodity support in 1996, and the projections were for substantially less government spending on com- modity programs than had occurred before 1995, as figure 4.3 shows. The 2002 Farm Act Within a year after the 1996 Farm Act was introduced, commodity prices began to fall substantially. By 1998, as the Asian financial crisis and China's lack of an expected increase in imports weakened world demand and hence U.S. exports and prices, it became clear that prices were likely to remain at low levels for some time. These low prices triggered two policy responses, one automatically and one through Congressional response. The automatic response came through the "marketing loan" program. This program was introduced in 1990 to replace the former loan rate program of supported market prices through forfeitures of commodities at an established "loan rate" price. The forfeiture program created a market price floor at the loan-rate price, because farmers could always get that price through the loan program. To keep the program from resulting in the government accumulating Figure 4.3. CCC Commodity Program Outlays, United States, Fiscal Years 1980­2007 35 30 25 US$, millions 20 15 10 5 0 1980 1985 1990 1995 2000 2005 2010 CCC outlays 1996 projection 2006 revision 2002 projection Source: Author's compilation from the CBO, Washington, DC. 190 Distortions to Agricultural Incentives: A Global Perspective unwanted stocks of commodities, loan-rate prices were mostly kept below actual average prices after the 1960s. The marketing loan program precluded stock accu- mulation altogether by having the government not actually acquire grain, but instead by offering farmers a "loan deficiency payment" equal to the difference between the policy-determined loan-rate price (which varied from county to county) and the local county price. Because the loan-rate prices were set below average prices, hardly any loan deficiency payments were made during 1990­97. But by 1998, commodity prices fell sufficiently below loan-rate levels to trigger US$500 million in loan deficiency payments. Unlike the target-price-related defi- ciency payments paid on a historically fixed level of output, the loan deficiency payments are made on all of current output. In fiscal years 1999, 2000, and 2001, these payments rose to US$3.4 billion, US$6.4 billion, and US$5.3 billion, respec- tively, as all the grains, cotton, and rice experienced continuing historically low market price levels. The Congressional response to these low prices was to enact emergency "market loss assistance" programs in each of the years 1998­2001. These pro- grams provided an average of US$4.8 billion annually for those four years and added proportionally to each farmer's production flexibility payments (in most cases doubling them during 1999­2001). Together, the market loss assistance pro- gram and production flexibility payments were the main factors behind the huge expansion of commodity program spending during 1998­2001 as compared to the levels anticipated in 1996, shown in figure 4.3. Actual spending in those years averaged more than US$20 billion, rather than the US$6 billion that was expected. By 2001, it was clear that Congress was dissatisfied with the political wrangling involved with annual emergency legislation and that more permanent revision of the 1996 Farm Act would be legislated. The Farm Security and Rural Investment Act of 2002 provided for additional spending on farm programs of more than US$50 billion above the projected baseline spending if the 1996 Farm Act had been continued. This increase was possible because at that time the U.S. budget was--and was expected to continue to be--in surplus, and because farm com- modity producers had sufficient political power to defeat competitors for the funds available. This expanded commitment is the most significant--and to some the most shocking--aspect of the 2002 Farm Act.11 In addition to the issue of spending levels, the 2002 Act addressed structural issues in terms of the form and scope of farm subsidies. Some farm groups, mainly centered on wheat growing in the Great Plains, wanted to return to supply management, in order to reduce production and increase commodity prices. Environmental groups pushed to have a substantial part of the new spending allo- cated to conservation and environment-improving programs. As enacted, the 2002 legislation continued the production flexibility payments from 1996 United States and Canada 191 (renamed fixed direct payments). Though supply management was rejected, a new "countercyclical payment program" was added to provide payments that rise or fall inversely with market prices (although the quantity base for payments remained fixed, with an option for one-time updating to 2002, for each farmer). This amounted basically to a reinstitution of pre-1996 deficiency payments, but without annual acreage reduction set-aside requirements and with farmers retain- ing the planting flexibility introduced in 1996. In addition, earlier production quotas for peanuts were replaced with strengthened support programs similar to the programs for other crops, and a new program of direct support payments for milk was added. Conservation and environmental programs ended up with a sub- stantial share of the new spending, but not as much as the proponents of these programs had argued for. The Congressional Budget Office (CBO) estimated that the innovations of the 2002 Act would cost an average of US$8 billion over the 10 fiscal years 2002­11.12 Of this, US$4.5 billion was for direct payments to farmers under either the fixed payments or the mandated new countercyclical payment program. It was esti- mated that the Act would lead to 10-year spending increases of US$0.5 billion in marketing loans and loan deficiency payments, US$0.5 billion for the new peanut program, US$160 million for the new dairy program, and US$43 million for increasing support in the sugar program, partly offset by savings projected at US$26 million from tightening payment limitation slightly, for a total of a US$5.7 billion spending increase for all commodity programs. Meanwhile, new initiatives in the Conservation Reserve Program and the Wetlands Reserve Program (both long-term, paid land-idling programs that have enrolled a total of more than 30 million acres since 1990), as well as the Environmental Quality Improvement Program, the Farmland Protection Program, and a new Conservation Security Program were projected to cost US$1.3 billion annually over 10 years (FAPRI 2002). The expanded costs of these programs consisted mostly of payments made to farmers to encourage the use of soil con- servation or water quality improvement practices. Some of the practices reduce production, such as replacing cultivated crops near streams or lakes with "filter strips" of grass in the Conservation or Wetlands Reserve Program. Other "working lands" programs are roughly neutral with respect to production or even result in an increase in production, such as the Farmland Protection Program payments to farmers in exchange for their maintaining land in farming rather than selling it for commercial development. Total outlays in fiscal 2003 and 2004 were lower than projected because higher- than-expected commodity prices resulted in less spending than had been forecast on countercyclical payments and loan deficiency payments. But in fiscal 2005, these payments increased as commodity prices fell. 192 Distortions to Agricultural Incentives: A Global Perspective Overall summary of market-distorting support Table 4.2a shows the nominal rates of assistance (NRAs) for key U.S. agricultural products, which are partly based on the OECD's PSE-CSE Database for 1986­2007, and, for years back to 1955, on the author's use of the OECD's method. This measure of NRAs excludes the deficiency, production flexibility and fixed direct payments, and the market loss assistance and countercyclical pay- ments. These various support payments have been at the heart of U.S. farm poli- cies, as described earlier, but have become increasingly decoupled from produc- tion decisions and arguably can now be concluded to have minimal or at least small effects on production or trade.13 The main policies included in the NRAs in table 4.2a are import protection, most notably for sugar, but also for dairy prod- ucts and meats; export subsidies, most notably the EEP and earlier export subsi- dies for wheat, but also affecting rice and poultry in some years; and the various loan-related payments to producers that a producer can increase by increasing output, which are closest to a classical production subsidy. Plotting these data sep- arately for exported products and for imported products that compete with U.S. products shows somewhat higher rates of assistance for importables (figure 4.4a). Given the dominance of exportables in overall U.S. production, its line in figure 4.4a is close to the line for all covered products. NRAs vary greatly across U.S. agricultural commodities, as summarized in the dispersion measure shown in the second-to-last row of table 4.2a, and even more so if the other products that get negligible support are considered (the latter accounting in aggregate for one-third of the value of U.S. agricultural production at undistorted prices, see bottom row of table 4.2a). Gardner (2008, appendix table 4.4) describes support through various gov- ernment payments (including those excluded from table 4.2a, termed here as "decoupled") for the top 25 commodities, accounting for 91 percent of the U.S. farm value of production in 2004. Overall, commodities accounting for 42 per- cent of production received significant support, with payments amounting to 6.9 percent of the value of production in 2004. Yet 14 of the top 25 commodities, and a total of 58 percent of the value of production, received no significant support. Appendix table 4.5 of Gardner (2008) shows a broader picture of U.S. federal government activity in support of agriculture. In addition to the commodity programs, there are conservation programs, export programs, governmentally underwritten loan programs for farmers, crop insurance, research funding, and marketing and regulatory programs. These additional activities had a price tag of US$15 billion in 2005.14 The sum of US$34.1 billion for fiscal 2005 amounts to 14 percent of the market value of U.S. farm cash receipts for all crops and livestock. United States and Canada 193 Figure 4.4. NRAs to Exportable, Import-Competing, and All Covered Agricultural Products, United States and Canada,a 1955­2007 a. United States 70 60 50 40 percent 30 20 10 0 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 b. Canada 70 60 50 40 percent 30 20 10 0 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 import-competing products total exportables Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. a. Includes the NRAs shown in table 4.2 and what is termed here as NPS assistance, all of which is attributed to tradables. NPS assistance for 1986­2007 includes payments classified by OECD as based on input use (E1) and miscellaneous payments (H). The total line does not include what here is termed "decoupled" assistance. Table 4.2. NRAs to Covered Farm Products, United States and Canada,a 1955­2007 (percent) 194 a. United States 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Exportables 1.0 1.1 6.7 5.7 2.1 5.0 10.5 5.5 5.1 8.1 4.8 Cotton 5.8 0.2 65.2 43.2 9.2 15.6 32.6 24.6 27.8 70.0 77.0 Eggs -- -- -- -- 4.1 5.8 4.6 8.1 2.2 0.0 0.0 Maize 0.0 1.4 2.4 3.4 0.4 2.6 10.4 0.6 4.8 7.0 6.6 Poultry 0.0 0.0 0.0 0.0 0.0 0.0 7.0 1.3 0.3 0.0 0.0 Rice 0.0 0.0 0.0 0.0 2.6 15.6 38.8 20.1 7.8 52.7 2.2 Sorghum -- -- -- -- -- 37.3 15.0 0.2 4.8 5.5 7.2 Soybeans 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.1 6.6 12.4 0.6 Wheat 0.0 3.6 17.8 14.4 9.6 11.2 29.1 25.5 5.4 4.3 0.2 Import-competing products 8.8 9.4 9.4 6.6 9.8 21.1 25.2 19.9 9.5 17.5 8.9 Barley -- -- -- -- -- -- 61.9 59.6 4.8 5.3 3.2 Beef 0.0 0.0 2.0 2.0 2.0 2.0 1.3 1.0 0.0 0.1 0.0 Milk 20.2 23.6 21.0 17.0 25.4 61.8 96.9 59.9 78.8 66.5 24.2 Pig meat 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.0 0.0 0.0 0.0 Sheep meat -- -- -- -- -- 7.1 2.4 1.2 3.4 13.8 9.9 Sugar 50.4 65.9 134.3 18.2 40.3 120.2 158.3 78.8 96.1 115.6 47.6 Wool -- -- -- -- -- -- 1.2 0.9 0.9 16.1 28.2 Nontradables 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Potatoes 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 All covered products 4.7 4.8 7.8 5.9 5.1 11.1 16.2 11.2 6.2 11.6 6.3 Domestic market support 0.0 0.0 0.0 0.0 0.1 0.9 2.3 0.4 2.2 3.6 1.0 Border market support 4.7 4.8 7.8 5.9 5.0 10.2 13.9 10.8 4.1 8.0 5.3 Dispersion of NRA of covered products 17.3 23.1 46.6 15.7 14.6 39.8 55.3 27.6 31.4 38.4 23.3 % Coverage at undistorted prices 66 66 66 69 70 69 66 67 65 66 68 b. Canada 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Exportables 2.5 2.4 2.2 2.8 4.3 8.2 2.4 1.4 1.4 0.4 Barley 2.0 2.0 2.0 5.8 8.0 10.4 2.5 1.4 2.8 1.0 Beef 1.0 1.0 1.0 1.2 2.6 4.0 3.9 n.a. n.a. n.a. Peas 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Pig meat 2.0 2.0 2.0 2.0 2.0 6.3 5.1 3.8 2.0 0.4 Rapeseeds 0.0 0.0 0.0 0.0 0.0 9.6 0.5 0.1 0.1 0.0 Soybeans n.a. n.a. n.a. n.a. n.a. 2.8 0.4 0.5 1.2 0.3 Wheat 3.0 3.0 3.0 3.1 5.3 10.4 1.4 0.6 0.6 0.2 Import-competing products 15.7 12.0 12.2 32.3 47.2 53.1 44.0 41.3 49.1 46.9 Beef n.a. n.a. n.a. n.a. n.a. n.a. n.a. 2.0 1.8 0.7 Egg -- -- -- -- 41.1 30.1 33.1 27.7 10.6 64.7 Maize 4.9 4.7 3.5 2.3 2.0 12.8 2.5 3.3 8.7 2.9 Milk 34.4 34.4 34.4 162.8 307.2 314.2 182.6 109.4 125.9 94.4 Poultry 17.0 17.8 33.5 24.4 23.3 23.7 25.2 2.7 2.6 6.9 Soybeans 0.0 0.0 0.0 0.0 0.0 n.a. n.a. n.a. n.a. n.a. Sugar -- -- -- -- 32.5 79.5 31.0 17.7 -- -- Nontradables 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Potatoes 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 All covered products 8.7 7.2 7.0 16.0 22.5 28.6 20.8 12.3 13.5 11.4 Domestic market support 0.0 0.0 0.0 0.1 0.5 13.2 4.1 1.9 1.4 0.4 Border market support 8.7 7.2 7.0 15.9 22.1 15.4 16.6 10.4 12.1 11.0 Dispersion of NRA of covered products 11.1 11.0 13.8 47.5 84.5 87.2 50.1 31.1 35.9 32.5 % Coverage at undistorted prices 71 76 79 84 86 83 81 79 76 76 Source: Anderson and Valenzuela (2008), drawing on author's spreadsheet. Note: -- not available; n.a. not applicable. 195 a. The main U.S. policies included are import protection, export subsidies, and the various loan-rate-related price support and payments to producers. Excludes deficiency, production flexibility and fixed direct payments, and the market loss assistance and countercyclical payments (see text for discussion). Weighted averages use weights based on the unassisted value of production. Dispersion is the standard deviation shown as the simple five-year average of the annual standard deviation around the weighted mean. 196 Distortions to Agricultural Incentives: A Global Perspective Agricultural Policies in Canada Canada, which remained a British colony during the early period of trade policy examined here, did not introduce substantial tariff protection of manufacturing until the 1870s, and then at less than the U.S. levels (Fowke 1946). Agricultural protection became important earlier, notably with wheat import tariffs directly against U.S. exports (and U.S. duties levied on Canadian grain exports). The sig- nificance of particular policy steps has been debated, but a convincing case can be made that "The tariff (of 1843) was a true break in the old colonial system, and was brought about, not by capitalists seeking to establish new industries, but by pioneer farmers trying to exclude outsiders from the local markets" (Jones 1941, p. 537). Thus, while attempts to measure the relative protection of manufacturing and agriculture comparable to those cited above by Irwin for the United States for the 19th century are not available for Canada, it seems clear that agriculture was never implicitly taxed in Canada the way it was in the United States (or in Australia and New Zealand--see Anderson et al. 2009). With respect to the context of agricultural support, Canadian agriculture varies substantially from east to west in cropping patterns and size of farms, in ways roughly parallel to corresponding U.S. areas south of the border. In those parallel areas, there are many similarities, notably between the Prairie Provinces and the Northern Plains. However, the history of policy and current practices in the two countries are quite distinct in several respects. The Great Depression of the 1930s hit Canadian farmers as hard as it hit those in the United States. In both countries, the effect was particularly severe in the prairies. The Canadian government provided some debt relief through the Prairie Farm Rehabilitation Act of 1935, but Canada did not introduce an integrated set of programs or make substantial income transfers to farmers as the New Deal did in the United States. Indeed, it is arguable that at that time and still to the present day, while a series of ad hoc programs have been implemented, Canada has no compre- hensive farm policy comparable to the omnibus U.S. farm bills. This is attributable in part to Canada's comparably more decentralized political system, under which many policies are carried out at the provincial rather the federal level, notably in marketing, where provinces have set up marketing legislation for fruits and animal products going back to 1926 (see Schmitz, Furtan, and Baylis 2002). In Canada, the first major federal support similar to that of the United States was provided under the Agricultural Stabilization Act of 1958. This program guaranteed producers 80 percent of the average price over the previous 10 years, by means of payments from the federal government. Only small payments were made, and in 1975 the payment trigger was raised to 90 percent of a five-year moving average price and a mechanism was introduced for compensating farmers for rapid increases in cash costs of production, which "indicated the Canadian United States and Canada 197 Government's continuing commitment to ensuring short-term solvency in agri- culture" (OECD 1978, p. 29). Canada's agricultural policies for grains and oilseeds have followed a path sim- ilar in some respects to those of the United States, but with less reliance on crop supply management through acreage controls and more reliance on collective marketing, most notably through the CWB. The CWB has a legal monopoly on the sales of Canadian wheat and barley into foreign markets--all exports as well as sales of wheat to domestic millers of flour must be sold through the agency. Farmers then receive a pooled price depending on the receipts the CWB is able to earn from the exported and domestically sold commodities. The monopoly pow- ers of the CWB have been challenged in Canadian courts, much as the original U.S. programs of the 1930s were in U.S. courts, on constitutional grounds. In a case brought in 1994, a group of barley growers sued the CWB, arguing that it "breached the rights of individual farmers guaranteed under the Canadian Char- ter of Rights and Freedoms" (Schmitz and Furtan 2000, p. 145). The plaintiffs' plea was rejected, and following Canadian law, they had to pay court costs of the defendants--the government of Canada. The year 1995 was a watershed in that, under pressure because of federal budg- etary deficits and the URAA's disciplines on export subsidies, transportation sub- sidies under the Western Grain Transportation Act were ended. These had cost an average of about US$12 per ton of grain, and their elimination was estimated to have saved the federal government about US$400 million in 1995. With respect to income support for producers, Canada has undertaken far- reaching experimentation in its series of grain programs over the last 30 years. The Western Grains Stabilization Program (WGSP), initially enacted in 1976, made payments from a fund partly financed by growers when their aggregate cash receipts from grains fell below a five-year moving average. After accumulating large deficits without providing satisfactory income protection to producers, the WGSP was abandoned. The Farm Income Protection Act of 1991 marked an important turning point in Canada's approach to farm support in crop production, when it moved away from policies aimed at particular commodities and toward a whole-farm approach. The 1991 Act introduced the Gross Revenue Insurance Program (GRIP) and the Net Income Stabilization Account (NISA). These programs were tailored to each producer's situation, with the GRIP making crop-specific pay- ments to producers when their production multiplied by the marketwide average market price fell below that producer's established average yield multiplied by a "target" price. The program was packaged as insurance in that each producer paid a premium for this coverage (though approximately two-thirds of the premium cost was paid by a combination of provincial and federal funds). 198 Distortions to Agricultural Incentives: A Global Perspective The GRIP combined features of the U.S. deficiency payment and subsidized crop insurance programs, and its comprehensive approach has attractive features. But it proved to have too little political support from farmers to justify its budgetary costs in the belt-tightening environment of the mid-1990s, and the GRIP expired after 1995. NISA, a more broadly conceived (in that it avoided support of specific com- modities) and less costly program, continued for another decade. NISA is essentially a subsidized savings account into which producers may contribute 2 percent of the value of qualifying grain sales, to be matched by 1 percent each from provincial and federal governments. The producer may withdraw funds from the account if either annual farm operating income or family income falls below established triggers (see Huff [n.d.] and Gray and Smith [1997] for further discussion). In 1998, Canada introduced the Agricultural Income Disaster Assistance (AIDA) program. Under that program, funded 60 percent by federal and 40 per- cent by provincial governments, anyone who files income tax returns as a farmer qualifies for an indemnity payment if their gross returns fall below 70 percent of the similarly calculated returns over the average of the three preceding years (Schmitz, Furtan, and Baylis 2002). In 2001, the Canadian Farm Income Program (CFIP) replaced AIDA. The experimental nature of farm income support has been intensified since 2004 with the phasing out of NISA and AIDA, and replacement with new and ad hoc programs. The Canadian Agricultural Income Stabilization (CAIS) Program, introduced in fiscal 2003/04, combines insurance and income support features. It makes payments to producers when a farmer's "production margin" falls below the "reference margin" for the farm, calculated from previous years' experience. The reference margin is a measure of returns minus costs that counts fewer expense items than under previous programs, because "experience with previous farm programs such as the CFIP indicated that including a high number of allowable expenses often resulted in reference margins being low and in many cases negative. This often resulted in producers being ineligible for benefits" (Agriculture and Agri-Food Canada 2006). In response to the livestock losses resulting from the Bovine Spongiform Encephalopathy (BSE, "mad cow disease") crisis, which devastated the beef export business, and as a "bridge" to the CAIS Program, the Transitional Income Support Program was introduced in 2004 to assist both livestock and grain producers. This program, together with more liberal payouts under the CAIS, raised the costs of farm support in 2003­05 as compared to earlier years. In 2006, the government introduced payments under the Grains and Oilseeds Program that will further increase support. The 21st-century programs are organized and marketed under the Agricultural Policy Framework, which the government of Canada and the provinces and United States and Canada 199 territories agreed upon in late 2003 to coordinate agricultural policy in five areas: business risk management, environment, food safety and quality, innovation, and renewal (see Agriculture and Agri-Food Canada 2005). The latter two areas include what are traditionally described as rural development and research and extension programs. Business risk management covers CAIS and the more recent payment programs. All of the programs have notable differences in funding and delivery from province to province. Unlike for grains, Canada has maintained supply management programs with strong control measures for dairy, poultry, and eggs. This is in sharp contrast to the United States, where supply management in livestock is entirely absent.15 Canada's supply management history grew out of practices established by provincial mar- keting boards, which were well established by the 1960s (Schmitz, Furtan, and Baylis 2002). The Canadian Dairy Commission, created in 1966, introduced supply management ideas that culminated in market-share quotas, under which a farmer must have an established quota in order to sell milk. Import quotas (converted to tariff-rate quotas under the URAA) keep milk from entering Canada at the high established retail domestic price, while milk for processed dairy products is sold at lower prices so that it is competitive in world markets. Similar but less complex supply management programs exist for broilers (chickens), eggs, and turkeys. Overall summary of market-distorting support Table 4.2b shows the nominal rates of assistance (NRAs) for key Canadian agri- cultural products, which are partly based on the OECD's database of producer support estimates (PSEs) for 1979­2007, and, for years back to 1961, on the author's use of the OECD's method. The key products cover between 75 and 85 pecent of Canadian agricultural production. The average NRA for those covered products was around 8 percent up to the mid-1970s, rose to 28.6 pecent during the export price war period of 1985­89, and has since come back to around 12 percent. The dispersion in NRAs across the product range rose substantially up to the mid-1980s, though it has more than halved since then. As usual, the estimated NRA for importables is well above that of exportables according to the classification based on trade status presented here--and well above the average importables NRA for the United States. Agricultural Relative to Nonagricultural Support To create a more complete picture of the policy distortions to farmer incentives in North America, guesstimates of the NRAs for noncovered products (a weighted average across exportables, import-competing products, and nontradables) are 200 Distortions to Agricultural Incentives: A Global Perspective first provided. NPS support--that is, input subsidies and decoupled support, as measured by certain OECD PSE categories--is then considered over the past three decades. The NRA for tradable farm products (including NPS support but not decoupled payments) is also compared with the NRA for nonagricultural trad- ables by calculating a relative rate of assistance (RRA). These results are summa- rized in table 4.3. Input subsidies and other NPS assistance is of nontrivial importance to the overall NRA for agriculture in North America as compared with the rest of the world, adding between one-quarter and one-third to the sectoral NRA in the United States and only slightly less in Canada. Payments based on input use (OECD E1, as defined in the PSE-CSE Database) and miscellaneous payments (OECD H) are included in this category. Of even more significance than NPS assistance are decoupled payments, which are assumed to encourage production less than market price support policies. Following certain OECD categories, not only the various commodity income support payments described above as increasingly decoupled from prices and production over time, but also payments made for long-term acreage idling under programs such as the Conservation Reserve Program, are examined here, as well as subsidies for crop insurance and ad hoc annual disaster payments (see note to table 4.3 for the OECD categories included).16 As can be seen from table 4.3a and figure 4.5a, those decoupled payments nearly doubled the NRA for the United States in the mid-1980s and added about one-third in the 1990s and the present decade. For Canada, decoupled payments have been somewhat less important, although still nontrivial (table 4.3b and figure 4.5b). As pointed out, it is difficult to summarize the effects of U.S. support pro- grams because of the variety of policy instruments used. Although any estimate is conjectural, the set of most reasonable estimates indicate that in 1999­2005, the marketing loan program has increased U.S. output of grains and soybeans by about 2 percent, the direct payment program (including the 2002 Farm Act changes) by about 1 percent, and crop insurance subsidies by more than 1 per- cent, for a total effect of 4 to 5 percent more of these commodities being pro- duced than would have been in the absence of commodity support programs. The long-run consequences (on conservation, agricultural research, and technol- ogy adoption) of commodity support policies are quite a different matter not considered here. Looking at the levels of border protection and domestic support payments themselves, overall government support for farmers in the United States rose in the decade or so between the mid-1970s and mid-1980s, after falling over the two decades prior. Support has since fallen back slightly, even when decoupled pay- ments are included, although the level was nearly as high during the period of low Table 4.3. NRAs to Agricultural Relative to Nonagricultural Industries, United States and Canada,a 1955­2007 (percent) a. United States 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, covered products 4.7 4.8 7.8 5.9 5.1 11.1 16.2 11.2 6.2 11.6 6.3 NRA, noncovered products 8.0 7.7 6.0 5.0 4.7 6.6 10.9 8.4 8.5 9.6 5.4 NRA, all agriculture (excluding NPS) 5.8 5.8 7.2 5.6 5.0 9.7 14.4 10.2 7.0 10.9 5.2 All importables 8.6 8.9 8.5 6.1 8.4 16.9 20.9 16.4 9.1 15.0 8.7 All exportables 2.7 2.7 6.5 5.5 2.6 5.3 10.6 6.2 5.8 8.4 4.7 All nontradables 8.0 6.8 5.1 4.3 4.2 5.7 9.4 7.2 7.3 8.2 2.3 NPS 7.0 5.2 3.3 1.1 1.2 1.8 2.7 3.7 3.3 4.9 4.3 NRA, all agriculture (including NPS) 12.8 10.9 10.5 6.9 6.2 11.5 17.1 13.9 10.2 15.6 9.8 NRA, decoupled payments 0.0 0.0 0.0 0.0 0.1 5.2 9.3 5.9 6.3 8.4 5.6 NRA, all agriculture (including NPS and decoupled) 12.8 10.9 10.5 6.9 6.3 16.7 26.4 19.9 16.5 24.1 15.3 NRA, all agricultural tradables (including NPS) 12.5 10.8 10.8 7.0 6.2 12.0 18.0 14.4 10.4 16.5 9.9 NRA, all nonagricultural tradables 6.1 7.3 7.4 5.4 3.8 3.5 3.5 3.2 2.1 1.5 1.4 RRAb 6.0 3.2 3.2 1.5 2.2 8.2 13.9 10.8 8.1 14.7 8.4 (Table continues on the following page.) 201 Table 4.3. NRAs to Agricultural Relative to Nonagricultural Industries, United States and Canada,a 1955­2007 202 (continued ) (percent) b. Canada 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Covered products 8.7 7.2 7.0 16.0 22.5 28.6 20.8 12.3 13.5 11.4 Noncovered products 6.6 5.4 4.9 9.8 12.9 14.9 10.6 6.2 7.1 6.7 All agriculture (excluding NPS) 8.1 6.8 6.5 15.0 21.2 26.3 18.9 11.0 12.0 9.1 All importables 15.1 11.7 12.0 31.1 45.2 50.9 41.8 36.2 42.0 45.0 All exportables 4.3 3.5 3.0 4.6 6.3 10.1 4.8 2.9 3.2 0.9 All nontradables 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NPS -- 4.9 4.0 5.7 9.1 13.2 12.8 7.8 12.1 2.2 Inputs -- -- -- 0.5 0.8 6.9 5.6 2.4 1.8 2.1 Other -- 4.9 4.0 5.2 8.3 6.3 7.2 5.5 10.4 0.1 All agriculture (including NPS) 8.1 11.7 11.0 20.8 30.3 39.7 31.6 18.9 23.9 11.4 Decoupled payments 0.0 0.0 0.0 0.7 4.7 8.6 8.0 5.3 10.3 11.8 All agriculture (including NPS and decoupled) 8.1 11.7 11.0 21.5 35.0 48.3 39.6 24.3 34.2 23.2 All agricultural tradables (including NPS) 8.1 11.7 11.0 20.8 30.3 39.7 31.6 18.9 23.9 11.4 All nonag tradables 9.2 6.9 6.0 5.1 4.8 3.9 2.6 1.1 0.8 0.8 RRAb 1.0 4.5 4.7 14.9 24.3 34.5 28.3 17.6 22.9 10.5 Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet and OECD (2008 and earlier years). Note: -- not available. a. Decoupled support includes "direct payments" in the years 1979­85. From 1986 those payments are specified to comprise the OECD's items C (payments based on area planted/animal numbers), D (payments based on historical entitlements), F (payments based on input constraints), and G (payments based on overall farming income). And for 2005­07, those items replaced by similar but newly defined items C to E. The values of those payments are estimated by the OECD. See text for discussion of NPS support. b. RRA is defined as 100*[(100 + NRAagt)/(100 + NRAnonagt) ­ 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. United States and Canada 203 Figure 4.5. NRAs to All Agriculture without and with Decoupled Support,a United States and Canada, 1955­2007 a. United States 70 60 50 40 percent 30 20 10 0 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 00 03 06 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 b. Canada 70 60 50 40 percent 30 20 10 0 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 00 03 06 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 decoupled support excluded decoupled support included Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet and OECD (2008). a. Decoupled support includes "direct payments" in the years 1979­85. From 1986 those payments are specified to comprise the OECD's items C (payments based on area planted/animal numbers), D (payments based on historical entitlements), F (payments based on input constraints), and G (payments based on overall farming income). For 2005­07, those items replaced by similar but newly defined items C to E. The values of those payments are estimated by the OECD. NPS support is also from OECD estimates as defined in the note to figure 4.4 (see text for discussion). 204 Distortions to Agricultural Incentives: A Global Perspective world agricultural prices during 2000­04 as it was when prices were low in the late 1980s (see table 4.3). It dropped substantially after 2005 as international food prices spiked, but this does not represent a change in policy so much as the coun- tercyclical design of some U.S. payments and a reduced need to support farmers because of higher prices. Support for farmers in Canada too rose steeply in the decade or so from the mid-1970s, but has since fallen back even more than in the United States, especially when decoupled payments are taken into account. As in the United States, the NRA for Canada dropped substantially after the early 2000s as international food prices rose. By contrast, assistance to producers of nonagricultural tradable goods has been lower than for farmers throughout this period, and has declined more than for the farm sector. Hence, even leaving decoupled payments aside, the rate of assistance to producers of agricultural products relative to those producing nonagricultural goods is now considerably higher than in the early 1960s in the United States, other than in the years of historically high international prices, namely 1995 and 2006­07 (table 4.3a and figure 4.6a). The 2008 Farm Bill left the existing support programs in place and created others that could raise expenditures even if agricul- tural prices remain higher than they generally were during 1990­2006.17 Assis- tance to Canada's producers of nonagricultural tradable goods, too, has been lower than for farmers since the mid-1960s, and has declined more than for Canada's farm sector and more than in the United States. Though the RRA to the farmers became considerably higher in the 1980s than in earlier decades, in the 1990s it returned to levels similar to those in the United States, again with the low- est levels in the years of historically high international commodity prices, namely 1995 and 2007 (table 4.3b and figure 4.6b). When expressed in real (constant 2000) U.S. dollars rather than percentage price wedges, the decine in border and domestic support recently is much less evi- dent, because over time the value of the farm sector has grown and the number of farmers has declined. The peak real value of support in total dollars was higher in the latter 1980s than this decade, but when expressed on a per farmer basis it was even higher in 2000­04 (before international food prices rose) than in 1985­89. When the market price support component is expressed by product, the lion's share in recent decades has gone to dairy and sugar, which receive more border protection than cotton and maize (table 4.4). Consumer Tax Equivalents (CTEs) While much of the support for farmers worldwide comes from border meas- ures, in the case of North America large shares also come from NPS meaures and from payments that are decoupled somewhat from production. Hence, United States and Canada 205 Figure 4.6. NRAs to All Nonagricultural Tradables, All Agricul- tural Tradable Industries, and RRA,a United States and Canada, 1955­2007 a. United States 50 40 30 percent 20 10 0 10 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 b. Canada 50 40 30 percent 20 10 0 10 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 agricultural tradables nonagricultural tradables RRA Source: Anderson and Valenzuela (2008), drawing on author's spreadsheet. a. The RRA is defined as 100*[(100 NRAag t ) (100 NRAnonagt ) 1]. consumers in the region are not taxed to the same extent as producers are assisted--although taxpayers bear additional costs. Because of the differing net trade status of each product, the weighted average of ad valorem CTEs across all covered farm products also differs because consumption rather than production Table 4.4. GSEs of Assistance to Farmers, by Product, Per Farm Worker and Total, United States and Canada, 1955­2007 206 a. By covered product (constant 2000 US$ millions, excluding NPS and decoupled payments) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Barley -- 14 19 44 97 141 518 392 48 53 20 Beef 0 25 442 555 541 549 388 307 53 69 25 Cotton 458 19 1,573 1,541 593 792 1,170 1,030 1,094 1,692 2,442 Eggs -- -- -- -- 395 495 302 432 174 34 167 Maize 0 273 509 998 163 848 1,927 152 825 1,316 1,081 Milk 3,365 4,253 4,116 3,829 7,775 13,778 12,986 9,875 3,761 8,125 6,666 Peas -- 0 0 0 0 0 0 0 0 0 0 Pig meat 0 31 41 56 55 52 124 297 68 46 12 Potatoes -- 0 0 0 0 0 0 0 0 0 0 Poultry 0 105 127 248 222 197 791 363 68 38 90 Rapeseeds -- 0 0 0 0 0 68 4 1 2 0 Rice 0 0 0 0 54 212 261 219 76 344 33 Sheep meat -- -- -- -- 40 40 14 5 12 40 28 Sorghum -- -- -- -- 436 636 261 4 40 40 36 Soybeans 0 0 0 0 0 0 91 18 734 1,274 90 Sugar 515 594 995 476 711 1,015 1,173 800 759 824 552 Wheat 0 317 1,308 1,266 1,248 1,689 2,108 1,737 333 240 16 Wool -- -- -- -- -- -- 1 1 0 3 5 b. Per person engaged in agriculture (constant 2000 US$, including NPS but not decoupled payments) 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Total, North America 3,115 3,739 3,445 4,181 6,720 8,222 7,370 5,580 9,165 7,364 United States 3,383 3,886 3,431 3,728 6,303 7,710 6,831 5,185 8,795 7,279 Canada 1,590 2,914 3,524 6,478 8,857 11,412 11,379 8,591 12,056 8,034 c. Total (constant 2000 US$ millions) 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 North America, covered products 5,632 9,130 9,014 12,331 20,445 22,185 15,636 8,044 14,142 11,264 United States 4,708 8,115 7,731 9,029 16,177 18,088 12,781 6,195 12,227 9,109 Canada 924 1,015 1,283 3,302 4,268 4,096 2,856 1,850 1,915 2,156 North America, covered and noncovered products 9,886 12,539 12,345 15,387 25,132 28,745 20,788 13,195 19,816 12,985 United States 8,552 11,263 10,821 11,816 20,473 24,224 17,599 11,100 17,591 10,717 Canada 1,334 1,276 1,524 3,570 4,659 4,521 3,188 2,095 2,225 2,268 North America, NPS 7,658 6,192 3,399 4,077 5,822 6,863 8,580 6,999 10,228 10,166 United States 7,658 5,269 2,417 2,678 3,815 4,548 6,397 5,491 7,974 9,598 Canada 0 923 981 1,399 2,007 2,315 2,182 1,508 2,254 568 North America, including NPS 17,544 18,731 15,743 19,464 30,954 35,608 29,368 20,194 30,044 23,151 United States 16,210 16,533 13,238 14,494 24,288 28,772 23,997 16,590 25,564 20,315 Canada 1,334 2,199 2,505 4,970 6,666 6,836 5,371 3,603 4,480 2,836 North America, decoupled payments 0 0 0 529 11,199 16,979 11,526 11,012 15,598 14,795 United States 0 0 0 352 10,193 15,529 10,177 9,990 13,657 11,826 Canada 0 0 0 177 1,006 1,451 1,349 1,022 1,941 2,970 North America, NPS and decoupled payments 17,544 18,731 15,743 19,993 42,153 52,587 40,894 31,205 45,643 37,947 United States 16,210 16,533 13,238 14,846 34,481 44,301 34,174 26,580 39,222 32,141 Canada 1,334 2,199 2,505 5,147 7,672 8,286 6,720 4,625 6,421 5,806 207 Source: Anderson and Valenzuela (2008) based on author's spreadsheet. Note: -- not available. 208 Distortions to Agricultural Incentives: A Global Perspective weights are used. There are also direct consumer subsidies in the United States, notably through the food stamp program, and in some years for some products those direct subsidies more than offset the tax component of the trade measures used there to support producer prices, resulting in negative CTEs. The rises and falls in the degree of distortion on the consumption side of the market can be seen in table 4.5. Table 4.5b shows how trivial these transfers from or to food consumers in North America have been on a per capita basis. Even in Canada, they amount to only US$154 or less per capita per year. This, together with the free rider problem associated with collective action, helps explain why consumers in that region do not counterlobby farmers over farm support programs. It also means the trade and national economic welfare effects of U.S. programs are less than programs delivering the same NRA for farmers in other countries via trade measures that generate a CTE equal to the NRA. The Politics of United States and Canadian Policies Explaining the political forces behind United States and Canadian agricultural policies requires qualitative rather than quantitative analysis, but the contrasts between the political treatment of commodities within and between the two coun- tries suggests hypotheses in several areas: (a) historical legacies of commodity pro- ducer cooperation, (b) the importance of supply-chain participant cohesion and the link of that cohesion with technological change, (c) the role of budgetary pres- sures, and (d) the inherent weakness of opposition to agricultural support. There are economic and cultural factors behind each of these.18 The importance of a historical legacy of producer cooperation was emphasized by Olson (1985). He noted the traditional and continuing strength of the dairy industry and traced that back to the long-standing organization of producers into marketing cooperatives. Olson's underlying point is that lobbying is a voluntary collective activity of precisely the kind highly susceptible to free-rider problems, that this is in fact the chief hurdle to an interest group obtaining subsidies, and that cooperative organizations have already solved this problem sufficiently to permit effective lobbying. This hypothesis fits well with what would otherwise be perhaps the chief puzzle in Canadian as compared to U.S. agricultural support, namely why poultry has had a well entrenched support system via supply control and import protection in Canada but receives virtually nothing in the United States. As Schmitz, Furtan, and Baylis (2002) explain, supply control measures grew out of cooperative activity in those commodities by provincial producer organizations. But there were no such corresponding poultry organizations in the United States. Table 4.5. CTEs of Policies Assisting Farmers, Covered Products, Total and Per Capita and by Product, United States and Canada, 1960­2005 a. Aggregate CTE (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005 Total, North America 5.7 8.4 6.1 7.2 12.6 9.8 1.9 4.7 2.4 6.3 United States 5.4 8.4 5.9 5.8 11.0 7.5 0.4 6.8 4.2 8.5 Canada 10.1 8.2 7.8 19.1 27.3 31.0 24.6 14.1 15.6 14.1 b. CTE per capita (constant 2000 US$) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005 Total, North America 27 41 36 44 68 42 8 18 9 26 United States 26 40 34 35 58 32 2 26 16 35 Canada 50 47 54 127 154 134 95 54 54 56 (Table continues on the following pages.) 209 210 Table 4.5. CTEs of Policies Assisting Farmers, Covered Products, Total and Per Capita and by Product, United States and Canada, 1960­2005 (continued ) c. CTE by covered product (constant 2000 US$ millions), United States, 1960­2005 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005 Barley -- -- -- -- -- 89 99 100 95 89 Beef 0 426 547 531 475 776 1,529 1,806 1,813 1,976 Cotton 16 1,489 1,284 482 657 1,028 865 965 1,320 1,955 Eggs -- -- -- 20 9 79 54 308 409 519 Maize 246 435 822 95 566 1,320 2,947 3,257 3,702 4,374 Milk 3,398 3,304 3,056 5,140 10,134 9,039 5,777 1,738 4,264 1,005 Pig meat 0 0 0 0 0 1,142 1,748 2,253 2,362 3,122 Potatoes 0 0 0 0 0 0 0 0 0 0 Poultry 0 0 0 0 0 134 1,076 1,263 1,328 1,657 Rice 0 0 0 25 100 46 138 201 199 357 Sheepmeat -- -- -- 29 5 6 6 13 53 57 Sorghum -- -- -- 0 0 143 189 185 163 161 Soybeans 0 0 0 0 0 199 337 365 376 483 Sugar 1,011 1,683 791 965 1,051 1,421 1,018 926 992 593 Wheat 147 591 572 449 546 88 126 917 859 1,293 Wool -- -- -- -- -- 2 1 1 0 0 United States 4,819 7,929 7,073 7,696 13,525 7,835 378 7,010 4,677 10,422 d. CTE by covered product (constant 2000 US$ millions), Canada, 1961­2005 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005 Barley 11 15 29 68 86 17 16 0 0 0 Beef 26 41 52 54 108 70 54 6 5 0 Eggs -- -- -- 70 127 98 121 89 33 141 Maize 17 23 27 23 23 22 1 0 0 0 Milk 709 682 749 2,542 3,185 2,847 2,137 1,471 1,641 1,663 Peas 0 0 0 0 0 0 0 0 0 -- Pig meat 31 41 54 56 45 7 0 0 0 0 Potatoes 0 0 0 0 0 0 0 0 0 0 Poultry 105 127 245 227 203 271 361 30 31 30 Rapeseeds 0 0 0 0 0 22 1 1 1 0 Soybeans 0 0 0 0 0 8 1 2 5 6 Sugar -- -- -- 44 39 43 31 25 -- -- Wheat 28 25 37 38 55 152 42 3 6 0 Canada 927 954 1,194 3,122 3,871 3,557 2,767 1,622 1,710 1,841 Source: Anderson and Valenzuela (2008) based on author's spreadsheet. Note: -- = not available. 211 212 Distortions to Agricultural Incentives: A Global Perspective Of course the efficacy of lobbying depends not only on getting organized to make one's case, but also on the legislators' listening and acting favorably. It is almost axiomatic that legislators have an interest in listening to their electorate, who hire and fire them. But a legislator cannot act in accord with the requests of all his or her constituents. Why do agricultural commodity interests get heard well in so many cases? Traditionally, U.S. farmers were said to have a political advan- tage because of the structure of Congress, which has two Senators for every state, or one Senator for each 300,000 residents of the Dakotas, Wyoming, and Montana compared to one for each 20 million residents of California. However, recent efforts at reform (such as taking payments away from producers who have more than US$1 million in off-farm income) have been defeated in the House of Repre- sentatives (where each Representative equally has about 600,000 resident con- stituents) as soundly as in the Senate. The general principal of representation is that, whenever possible, a legislator provides constituents what they ask for. Complicating the efficacy of this system in the United States, however, is that constituents often request conflicting poli- cies. A strength of cotton growers is that they present a unified position of the supply chain, including growers, ginners, shippers, and millers, to the agriculture committees of Congress. This has led to policies such as the "step two" payments to cotton textile millers that compensate them for paying prices for cotton that exceed the low world prices that trigger payments to cotton growers. (This was stopped, however, to comply with a WTO ruling.) Similarly, a strength of corn growers is that they have powerful agribusiness allies who support ethanol subsi- dies, while a former weakness of the grain producers was their asking for acreage control measures that restricted raw material supplies for such agribusiness (for- mer because the 1996 Farm Act revoked the authority of the Secretary of Agricul- ture to administer annual acreage reduction programs that until the 1990s were a key feature of grains policy). Cross-commodity dispute is a related problem, aris- ing, for example, on the part of fruit and vegetable growers excluded from decou- pled payment programs. A political strength of sugar import restrictions is that high sugar prices have created a large market for corn-based sweeteners. Monetary contributions help in getting an interest group's case heard by politi- cal leaders. Though farmers are not notably profligate donors, agricultural political action committees (PACs) have been important in the United States. Table 4.6 summarizes PAC donations to politicians or candidates reported during the 2006 election cycle (November 2004­October 2006). These data are from the legally required reporting of PACs for donations of more than US$100,000 and do not capture all political spending. Though the US$6.7 million they donated may seem a large sum, there were 1,200 nonagricultural PACs that spent more than US$100,000 in the United States during this period, and their aggregate spending United States and Canada 213 Table 4.6. PAC Disbursements during the Election Cycle, United States, November 2004­October 2006 (US$ millions) Sugar Alliance 1.2 Farm Credit Council 1.0 Dairy 0.8 Texas Farm Bureau 0.5 Cotton Council 0.3 Sugar 0.3 Cattlemens Beef 0.3 Dairy 0.3 Broiler Chickens 0.3 FLA sugar 0.2 Indiana Farm Bureau 0.2 Sugar beets 0.2 Farmers' Group 0.2 Peanuts 0.2 Eggs 0.2 Michigan Farm Bureau 0.2 Farmer Coop 0.1 Sugar beets 0.1 Beef 0.1 Total 6.7 Source: Political Moneyline, http://www.tray.com. was US$800 million. The largest agricultural PAC, the Sugar Alliance, ranked 154th of the 1,200. Agriculture's share of the spending, US$6.7 million or just under 1 percent of the US$800 millon total, is about the same as agriculture's share of national GDP. Overall spending by PACs in the broad Standard Industrial Classifi- cation code for agriculture was US$15 million, but the majority of this was from agribusiness firms--notably Deere, Deans Foods, the International Dairy Foods Association, Cargill, Tysons, Archer-Daniel-Midlands, Conagra, and Heinz Foods. These companies have interests that are often the same as the commodity produc- ers' interests, but not always, and what the agribusiness firms focus on in their lob- bying is more typically their own specific issues of concern. In short, it does not appear likely that money is the source of farmers' exceptional political influence. Technical or other exogenous economic changes create diverse interests within a commodity group that can make policy making aimed at benefiting the group as a whole unattractive to legislators. This happened most notably in the case of a for- merly important commodity program that disappeared completely, the U.S. potato 214 Distortions to Agricultural Incentives: A Global Perspective price support program. During 1945­50, the potato program became highly con- tentious and eventually died because the Western potato growers were able to pro- duce profitably at prices lower than the prevailing support levels, and they saw their market potentially capped by restraints needed to control the costs of the support system (which at the time relied heavily on production control for the main sup- ported commodities). Similarly, the long-standing marketing order pricing sys- tems for California oranges and lemons was ended in the 1990s, mainly because of intraindustry disagreement about its operation, while differing interests of peanut and tobacco farmers with and without quotas to produce for the high-priced domestic market contributed to demise of these quota program earlier this decade. One might expect consumer interests to oppose farm legislation that would increase food prices, but this has only rarely surfaced as a significant political force (the notable case being U.S. grain export embargoes of the 1970s). This lack of opposition seems to be associated with longstanding positive feelings that the general public has about farmers and farming. A broad-based source of resistance to agricultural support that has been effective is budgetary pressure at times when fiscal discipline is perceived to be a high priority. In both the 1985 and 1990 farm legislation, reductions in payments were enacted so that farm bills could meet overall Congressional budget limits. The 1996 farm legislation was also designed to limit subsidy expenditures, but that budget discipline quickly broke down. As noted earlier, the Farm Security and Rural Investment Act of 2002 covering crops planted in 2002­07 increased spending well above baseline levels from con- tinuation of the legisltation it replaced. This Act was popular in Congress, having passed in the House of Representatives by a vote of 280 to 141 and in the Senate by 64 to 35. The Bush Administration did not raise serious objections, and the bill was signed into law with praise from farm-group representatives.19 However, small-farm and environmental advocacy groups were unhappy that amendments that would have imposed more stringent payment limits on large farms, redi- rected some commodity program payments to conservation and environmental programs, and imposed various regulatory restraints on agribusiness failed. Out- side the community of U.S. agricultural interests, the 2002 Act has been widely reviled, as mentioned earlier.20 Just after the Act was passed, three western provinces of Canada, along with a dozen Canadian farm groups, asked for Can$1.3 billion to offset the effects of the new U.S. payments. Since that time, Canada has raised its payments to producers too, as herein described. Prospects for Reform Both the United States and Canada have enacted agricultural support programs that have distorted their domestic commodity markets. Considering how similar the countries are in many respects, there are some notable differences between United States and Canada 215 the policy approaches in the two countries in the choice of policy instruments. But the similarities are more fundamental, and several of them bear on the pros- pects for reform: (a) strong political resilience of support for farmers, (b) weak political expression of consumers' interests, (c) trends toward less reliance on measures that seek to control prices in particular markets (via stockpiling, sup- ply management, and import tariffs), (d) trends toward more reliance on whole-farm income support (especially in Canada) and payments under com- modity programs that are delinked from current production (in the United States), (e) acceptance of multilateral liberalization of agricultural trade, but only with maintenance of protection for some producer interests, (f) modest increases in political influence of environmental protection, and (g) episodic political strength of taxpayer interests in cutting farm-support spending, under circumstances that give general budget reduction. As a result of these forces and trends, real spending on agricultural support has not diminished over time even as the share of the agricultural sector in the overall economy has diminished greatly, while at the same time both the United States and Canada have moved since the early 1990s in the direction of reduced directly market-distorting poli- cies. In the face of historically low commodity prices 1998­2002, the United States maintained policies that forestalled output reductions that these low prices would otherwise have induced. What options does the preceding summary suggest for reform? First, the best prospects remain international negotiations, most notably the Doha Round of the WTO, despite the failure of the negotiations to make much progress thus far. Nonetheless, it remains the case that the United States and Canada would sign on to such an agreement if it involved provisions that reform proponents could point to as offsetting gains in agricultural export markets. Second, there remain prospects that a combination of environmental and taxpayer interests could shift agricultural support spending toward public-good provision in ways that would be less market-distorting than current policies.21 Reforms in this direction were formulated and promoted by several organized coalitions of interests in preparation for the 2008 U.S. Farm Bill, mainly by environmentalists, farmland preservation- ists, and internationally oriented agribusiness groups. Third, while it is not in the cards now, it is possible that there will a resurrection of a general predisposition to economic liberalism, as occurred in Australia and New Zealand (Anderson et al. 2009). Not so long ago, this predisposition in North America served to limit the scope of market-distorting legislation considerably, and could again, as the recent legal debate in Canada on the CWB's authorities illustrates, despite its outcome. The ascent of free-market Republicans in the 1980s suggested this might be a real possibility for the United States. But the policy salience of this strain of opinion has been thwarted by that the party's embrace of muscular nationalism, military internationalism, and cultural conservatism. Similarly, once-prominent free-trade 216 Distortions to Agricultural Incentives: A Global Perspective advocacy by Democrats has been swamped by the wave of industrial protection- ism allied with antitrade prairie populism the party has undertaken in recent political campaigns. The prospects for each of the preceding openings to reform--a liberalizing WTO agreement, conservation and environmental pressures, and a resurgence of deregulatory policy dispositions--would be enhanced with the traditional econo- mist's recipe of compensating the losers from policy changes with nondistorting transfers. Though the direct payments under the U.S. 1996 and 2002 Farm Acts are a move in this direction, what is really required are one-time payments, or buyouts. U.S. policy has carried out three notable experiments in buyouts since 1980: in dairy, peanuts, and tobacco. The dairy buyout (Dairy Production Termi- nation Program) was implemented in 1986­87, at a time when milk surpluses were chronic, and was introduced following the calculation that having dairy farmers agree to sell their herds and leave dairying would reduce budgetary out- lays. Farmers made offers to the government stating the price per hundred pounds of milk producing capacity of their herds at which they would be willing to leave the business. Producers who participated had to sell all their female cattle and agree to remain out of dairy farming for five years, and attempts were made to ensure that cattle sold were slaughtered or exported and not sold to another dairy farm. The program is estimated to have reduced U.S. milk production by about 7 percent in the short term but to have had no long-term effect (Dixon, Susanto, and Berry 1991). The positive lesson of the program, however, is that buyouts can be successfully implemented by USDA as a means of getting farmers to participate in policy reform. Buyouts to permanently remove producers from certain commodity support programs have been undertaken for peanuts and tobacco since 2000. The 2002 Farm Act authorized payments to peanut producers to buy their production quo- tas, which had well-defined values under preexisting programs. Premiums over market values of quota had to be paid, but again the approach was proven feasible. Similarly, tobacco quotas and the entire tobacco price support program were ter- minated in 2005. However, whether buyouts would be feasible for commodities without production-limiting quotas or where program-created assets are not so precisely defined is questionable. An approximation to a buyout was, in the minds of some proponents of production flexibility, the contract payments in the 1996 Farm Act. But this legislation did not formally end the preexisting support struc- ture and, in fact, did not end the support programs. Rather, the approach enabled additional payments to farmers. While students of the prospects for this approach remain cautious (see Orden 2006 and Orden, Blandford, and Josling 2009), some- thing like it coupled with changes of approach as outlined above are the most likely feasible avenue to future reform. United States and Canada 217 Notes 1. This chapter's coverage of the United States draws to some extent on Gardner (2002). 2. There is one category of farms in the USDA classification, called "limited resource farms," that has low incomes both from farming and from off-farm sources. There were an estimated 199,000 such farms in 2004, or 10 percent of all farms; their average household income was US$7,700. Their average farm earnings was a loss of US$5,900. Apart from these farms, the other 90 percent of U.S. farm house- holds are doing well economically, either from farm or off-farm income sources. The data underlying this and subsequent information about farm household economics are developed by the Economic Research Service of the USDA, using the organization's annual Agricultural Resource Management Survey of about 10,000 farms. 3. For 2001­05, the annual net income per farm that yields the Can$11,000 average are, respec- tively: Can$11,000, Can$6,000, Can$11,100, Can$16,400, and Can$10,600 (Statistics Canada 2006). 4. The Court changed its view in decisions of Mulford v. Smith (1939) and Wickard v. Filburn (1942), which upheld, respectively, tobacco and wheat marketing controls enacted in legislation of 1938. The reasoning was that agricultural production affected by such programs influenced national markets and so could be legislated under constitutional powers to regulate interstate commerce. 5. An earlier wheat agreement, in 1933, created a schedule of quotas limiting the shipments of wheat by exporting countries, but this proto-OPEC broke down and the agreement was allowed to expire in 1935 (Gale and Zaglits 1949). Explicit export subsidies on Northwestern U.S. wheat were also implemented for a time early on in the New Deal (Nourse, Davis, and Black 1935). 6. Irwin's estimate of a net subsidy rate of 15 pecent is less than the tariff protection rate of 30 per- cent primarily because of the effect of the higher prices of import-competing goods on increasing the cost of production of nontraded goods. 7. It was such political successes of agriculture in tariff legislation that led H. L. Mencken to pillory farmers in one of his famous diatribes: "Has anyone heard of a farmer practising or advocating any political idea that was not absolutely self-seeking--that was not, in fact, deliberately designed to loot the rest of us to his gain? . . . There has never been a time, in good seasons or bad, when his hand were not itching for more . . . One might almost argue that the chief, and perhaps even only aim of legisla- tion in These States is to succor and secure the farmer" (Mencken 1958, pp. 158­60). 8. For estimates of effects of U.S. wheat export subsidies, see Gardner (1996). On the effects of grain export embargoes, see USDA (1986). 9. The U.S. government's fiscal year runs from October 1 of the previous calendar year to September 30 of the current calendar year. Thus, fiscal year 2005 represents the time period October 1, 2004 to September 30, 2005. 10. The spike in spending relative to payments in 2000 is an artifact of government budgetary spending being reported on a fiscal year basis while payments to farmers are reported on a calendar year basis. In 2000, an election year, Congress rushed to get payments to farmers that would normally have gone out after October 1 before October 1, resulting in payments that would normally have occurred in two fiscal years occurring in one calendar year. 11. A Washington Post editorial termed the bill "The Mother of All Pork." Business Week magazine opined: "It's a dreadful piece of legislation-bad for most farmers, bad for consumers, and horrendous for taxpayers" (May 7, 2002). The New York Times editorialized against the Act on several occasions, notably in a piece entitled "The Hypocrisy of Farm Subsidies" (December 1, 2002). 12. Ten-year projected spending was estimated in accordance with Congressional budgetary proce- dures, even though the Act only authorized programs for the six years 2002­07. The "baseline budget scoring" assumption is that those programs will be reauthorized to cover the 10-year period. 13. This argument is more difficult to make for the 1980s, when these payments were tied to signif- icant annual land-idling requirements. It also ignores wealth effects, insurance effects, and remaining production restrictions, as discussed in the appendix to Gardner (2008). Whether any of these effects are quantitatively important is an empirical question. While impossible to estimate with precision from the data available, the analysis to date suggests these effects are small. 218 Distortions to Agricultural Incentives: A Global Perspective 14. "Price tag" is a vague term and is used because the figures are not all derived from a consistent set of U.S. budgetary concepts. Most notably, the export credit guarantees are not the expenditures of the government on these guarantees; rather, they are the value of loans guaranteed. Unless there are defaults on these loans (funds borrowed by foreign importers to buy U.S. exports are not repaid to the U.S. lenders), the actual outlays on these programs is negligible. In fact, defaults are rare. In U.S. budg- etary parlance, the value of loans guaranteed is the "program level," as shown in the USDA budget summary. 15. In the original programs of the 1930s, there were U.S. supply control efforts in livestock. The only significant such program since 1955, however, is the Dairy Herd Buyout Program of the mid-1980s. 16. Neither the categories "NPS" nor "decoupled" presented here correspond directly to use of these terms in WTO domestic support notifications or as has occurred in dispute settlement argu- ments. To illustrate, the NPS for 1986-2007 herein includes diesel fuel tax exemptions not reported by the United States to the WTO under the URAA category of NPS, while it excludes crop and revenue insurance subsidies that are reported as WTO NPS support. The "decoupled" category herein includes not only the "decoupled income support" as defined in URAA annex 2 as WTO green box (which is how the United States reports its fixed direct payments); countercyclical payments that the United States reports as NPS are also included in this category. Subtleties in distinguishing between these cat- egories of payments are discussed in the appendix to Gardner (2008). Both classifications are subject to WTO dispute settlement litegation. The decoupled support herein also includes conservation pay- ments (both for long-term land-idling and working lands) that are reported to the WTO by the United States as green box and the crop and revenue insurance subsidies reported to the WTO as NPS sup- port. The use of these terms herein also differs from common classifications in U.S. farm bill discus- sions and budgets that, for example, often clearly separate commodity support from conservation expenditures. 17. See Orden, Blandford, and Josling (2009) for discussion of the 2008 U.S. Farm Act. 18. For more on the political economy of U.S. farm policy, see Gardner (2002) and Orden, Paarlberg, and Roe (1999). 19. As a reminder that U.S. presidents do not always accept what Congress delivers in support of agriculture, President Reagan in 1985 vetoed a farm bill on budgetary grounds, in the midst of the farm crisis of the 1980s. Congress could not muster the two-thirds majorities needed to override the veto, so the president's action was decisive. 20. It is also notable that the more market-oriented members of Congress, even if they represent agricultural constituencies, opposed the bill. Among the opponents were not only the House Republi- can leadership, but also members of the Agriculture Committee, such as Boehmer (Republican-Ohio) and Dooley (Democrat-California), who jointly wrote a Washington Post opinion piece entitled "This Terrible Farm Bill" (May 2, 2002). Similarly strong opposition was voiced in the Senate by Richard Lugar (Republican-Indiana), the senior Republican on the Senate's Agriculture Committee. 21. Even if these policies were more market-distorting, as they could possibly become, they would more likely be of the kind that would reduce agricultural output rather than increase it. References Agriculture and Agri-Food Canada. 2005. "Agricultural Policy Framework: Federal-Provincial-Territorial Programs." Program Planning, Integration, and Management Directorate, Ottawa. http://www .agr.gc.ca/progser/index_e.phtml. ______. 2006. "The Canadian Agricultural Income Stabilization (CAIS) Program." Program Planning, Integration, and Management Directorate, Ottawa. http://www.agr.gc.ca/caisprogram/factsheets/ faq_margins.html. Anderson, K., R. Lattimore, P. J. Lloyd, and D. MacLaren. 2009. "Australia and New Zealand." Chapter 5 in this volume. United States and Canada 219 Anderson, K., and E. Valenzuela. 2008. "Estimates of Global Distortions to Agricultural Incentives, 1955 to 2007." Core database at http://www.worldbank.org/agdistortions. Benedict, M. R. 1953. Farm Policies of the United States, 1790­1950. New York: Twentieth Century Fund. Carter, S. B., ed. 2006. Historical Statistics of the United States: Colonial Times to 1970, New York: Cambridge Uinversity Press. Davis, L., J. R. T. Hughes, and D. M. McDougall. 1965. American Economic History. Homewood, IL: Richard D. Irwin Inc. Dixon, B., D. Susanto, and C. R. Berry. 1991. "Supply Impact of the Milk Diversion and Dairy Termination Programs." American Journal of Agricultural Economics 73 (3): 633­40. FAPRI (Food and Agricultural Policy Research Institute). 2002. "Farm Security and Rural Investment Act of 2002: Preliminary FAPRI Analysis." Unpublished paper, FAPRI. Fowke, V. C. 1946. Canadian Agricultural Policy: The Historical Pattern. Toronto: University of Toronto Press. Furtan, W. H. 2006. "Canadian Agriculture and Food Policy." Saskatchewan Agrivision Corp., Saskatoon. Gale, E. G., and O. Zaglits. 1949. "Intergovernmental Agreements Approach to the Problem of Agricul- tural Surpluses." In Readings on Agricultural Policy, ed. O. Jesness, 306­27. Philadelphia: Blakiston. Gardner, B. L. 1996. "The Political Economy of U.S. Export Subsidies for Wheat." In The Political Economy of American Trade Policy, ed. A. Krueger, 291­331. Chicago: University of Chicago Press. ______. 2002. American Agriculture in the Twentieth Century: How It Flourished and What It Cost. Cambridge, MA: Harvard University Press. ______. 2008. "Distortions to Agricultural Incentives in the United States and Canada." Agricultural Distortions Working Paper 62, World Bank, Washington, DC. Gray, R., and V. Smith. 1997. "Harmonization and Convergence of Canadian and U.S. Grains and Oilseeds Policies." Policy Issues Paper 4, Montana State Univerity, Bozeman, Montana. Huff, H. B. n.d. "Changing Role of Public Policy in Canadian Agriculture." Unpublished paper. Irwin, D. A. 2006. "Historical Aspects of U.S. Trade Policy." NBER Reporter http://www.nber.org/ reporter/summer06/irwin.html. Johnson, D. G. 1950. Trade and Agriculture. New York: McGraw Hill. Jones, R. L. 1941. "The Canadian Agricultural Tariff of 1843." Canadian Journal of Economics and Political Science 7: 528­37. Josling, T. 2009. "Western Europe." Chapter 3 in this volume. Mencken, H. L. 1958. Prejudices: A Selection. New York: Vintage Books. Nourse, E., J. Davis, and J. Black. 1935. Three Years of the Agricultural Adjustment Administration. Wash- ington, DC: Brookings Institution. OECD (Organisation for Economic Co-operation and Development). 1978. Recent Developments in Canadian Agricultural Policy. Paris: OECD. OECD PSE-CSE Database (Producer and Consumer Support Estimates, OECD Database 1986­2007). Organisation for Economic Co-operation and Development. http://www.oecd.org/document/ 59/0,3343,en_2649_33727_39551355_1_1_1_1,00.html. Olson, M. 1985. "Space, Agriculture, and Organization." American Journal of Agricultural Economics 72 (3): 928­37. Orden, D. 2006. "Feasibility of Farm Program Buyouts." Presented at North American Agrifood Mar- ket Integration Workshop, Alberta, Canada, May 31­June 2. Orden, D., D. Blandford, and T. Josling. 2009. "Determinants of Farm Policies in the United States, 1996­2008." Agricultural Distortions Working Paper 89, World Bank, Washington, DC. http:// www.worldbank.org/agdistortions. Forthcoming in Political Economy of Distortions to Agricultural Incentives, ed. K. Anderson. Orden, D., R. Paarlberg, and T. Roe. 1999. Policy Reform in American Agriculture: Analysis and Progno- sis. Chicago: University of Chicago Press. 220 Distortions to Agricultural Incentives: A Global Perspective Schmitz, A., and H. Furtan. 2000. The Canadian Wheat Board. Regina, Saskatchewan: Canadian Plains Research Center. Schmitz, A., H. Furtan, and K. Baylis. 2002. Agricultural Policy, Agribusiness, and Rent-Seeking Behavior. Toronto: University of Toronto Press. Statistics Canada. 2006. CANSIM table. http://www40.statcan/101/cst01/prim10.htm. Taussig, F. W. 1931. The Tariff History of the United States. New York: G. P. Putnam's Sons. U.S. Department of Agriculture. 1986. "Embargoes, Surplus Disposal, and U.S. Agriculture." Economic Research Service, Agricultural Econonomics Report 564, U.S. Department of Agriculture, Washington, DC. ______. 2006. "Budget Summary and Annual Performance Plan." Office of Budget and Program Analysis, U.S. Department of Agriculture, Washington, DC. http://www.obpa.usda.gov/budsum/ 2007/fy07budsum.pdf. 5 Australia and New Zealand Kym Anderson, Ralph Lattimore, Peter J. Lloyd, and Donald MacLaren* Following its gold rush in the 1850s, Australia surpassed the United Kingdom in having the highest per capita income in the world, and New Zealand was not far behind. Both countries suffered from depressions in the 1890s before recovering their equal-first ranking just prior to World War I, though for the following seven decades their incomes lagged those of the United States and some other developed economies (figure 5.1).1 The middle of the 20th century brought a brief recovery in ranking thanks in part to the Korean War­induced boom in wool prices, when Australia and New Zealand were ranked equal second after the United States. But since then their rankings have continued to slide as the economies of Western Europe, Canada, and Japan grew faster than the United States, while Australia-- and especially New Zealand--grew slower (Anderson, Lloyd, and MacLaren 2007). By 2004, these two antipodean economies were ranked 25th and 37th respectively, according to the World Bank Atlas method of measuring gross national product (GNP) per capita (World Bank 2006). By that standard, at least, the long-term economic performance of both the Australian and New Zealand economies over the 100 years to the 1970s has to be described as relatively poor. During the past three decades, by contrast, these two small economies (with just 20 million and 4 million people, respectively) outperformed most other *The authors are grateful for research assistance from Johanna Croser, Ceren Erdogan, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, Damiano Sandri, and Ernesto Valenzuela, and for helpful comments from workshop participants and Rod Forbes, Robin Johnson, Jonathan Pincus, and Grant Scobie. The working paper version of this chapter (Anderson, Lloyd, and MacLaren 2007; Anderson et al. 2008) contains additional background material and appendix tables. 221 222 Distortions to Agricultural Incentives: A Global Perspective Figure 5.1. Real GDP Per Capita in Australia, New Zealand, and Other High-Income Countries Relative to the United States, 1870­2004 1.6 1.4 1 1.2 United States 1.0 0.8 0.6 0.4 18 5­ 4 18 0­ 9 18 5­ 4 18 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 19 5­ 4 19 0­ 9 20 5­ 4 00 99 4 7 7 8 7 8 8 9 8 9 9 0 9 0 0 1 0 1 1 2 1 2 2 3 2 3 3 4 3 4 4 5 4 5 5 6 5 6 6 7 6 7 7 8 7 8 8 9 8 9 9 ­0 18 0­ 7 18 Australia New Zealand other Western Europe Canada Nordic countries Source: Based on 1990 international Geary-Khamis dollars from Maddison (2003), shown relative to the United States, which is set as the numeraire at 1.00. "Nordics" include Denmark, Finland, Norway and Sweden; "other Western Europe" includes all countries with data from 1870, namely Belgium, France, Germany, Italy, Netherlands, Portugal, Spain, Switzerland, and the United Kingdom. high-income countries, with their per capita incomes growing half as fast again as the Organisation for Economic Co-operation and Development (OECD) average. The marked difference between recent and earlier relative performance in Australia and New Zealand is due to major economic policy reforms, in particular the belated opening of their economies, first to each other and then to the rest of the world. Having been more protectionist toward manufacturing than other OECD countries for most of the 20th century (Anderson and Garnaut 1987) and having not participated in the industrial trade policy reforms agreed to by other General Agreement on Tariffs and Trade (GATT) contracting parties in the first seven rounds of multilateral trade negotiations (1947 to 1979), Australia and New Zealand have since undergone a remarkable degree of opening up of their current and capital accounts. This liberalization reversed the downward trend in their trade as a share of gross domestic product (GDP), particularly for other OECD countries (including the United States, in which the share trebled over the past three decades) while accel- erating the downward trend in their average import tariff (Anderson et al. 2008). The fact that increased openness in Australia and New Zealand was accompanied Australia and New Zealand 223 by many domestic microeconomic and macroeconomic reforms, and that it coin- cided with a long period of rapid global economic growth that was stimulated by the information and communication technology revolution and by the opening up of other world economies (especially nearby resource-poor countries in East Asia) added to the scope for boosting gains from freeing their international trade and investment and floating their currencies. Another difference between the economies of Australia and New Zealand and most other OECD countries is that they are relatively well endowed with agricul- tural land per worker. Coupled with the development of valuable production and marketing technology, this endowment provides Australia and New Zealand with a significant comparative advantage in agricultural products. The advantage is especially strong for New Zealand, which, unlike Australia, does not contain an abundance of mineral and energy resources. Trade protectionism in these economies thus meant restrictions on imports of manufactured goods, making their trade policy regimes similar to those of developing countries. True, in some decades they also had periods of agricultural subsidies, but overall the trade policy regime in both countries has long involved an antiagricultural bias. The fact that those agricultural subsidies have been virtually eliminated over the past two decades also makes Australia and New Zealand an interesting political economy study, given the extreme difficulties other OECD countries have experienced in reforming their farm support programs (see Gardner 2009; Josling 2009; and Honma and Hayami 2009). This chapter examines the extent to which the antiagricultural bias in Australia and New Zealand has changed since World War II, identifying the forces that caused the policy evolution in each case. It begins by summarizing the structural changes that have accompanied economic growth in both countries since the 1950s. It then describes the emergence of manufacturing protection and some agricultural subsidies, and, after the 1970s, the dramatic dismantling of those inter- ventions. The analysis is conducted by compiling a new time series of nominal rates of protection for both agriculture and manufacturing, stretching back to the mid-1940s for Australia and to the mid-1950s for New Zealand. The reasons for these policy choices are also explored--first, the gradual growth in market inter- ventions to the 1970s and then their relatively rapid dismantling thereafter. The chapter concludes by discussing prospects for further policy reform and lessons for both other high-income countries and resource-rich developing economies. Growth and Structural Changes since 1950 The growth performance of Australia and New Zealand for most of the 20th cen- tury is in stark contrast with that since the late 1980s, when the two countries out- performed many other advanced economies in terms of GDP per capita growth 224 Distortions to Agricultural Incentives: A Global Perspective (World Bank 2006). The past two decades have also been a period of especially rapid total factor productivity (TFP) growth in Australia and New Zealand (Parham et al. 1999; Dowrick 2001; Statistics New Zealand 2006), mostly due to growth in employment and hours worked per worker (Card and Freeman 2004), in contrast to the United Kingdom, for example. Australia's annual TFP growth rate accelerated a full percentage point during the 1990s (Parham 2004), while New Zealand's rose four-fifths of a percentage point. Since that was not the expe- rience of other OECD countries, Parham asserts that domestic factors must explain a major part of the increase, an important one being the comparatively greater openness of the Australian economy to trade and investment.2 A more recent econometric study by Diewert and Lawrence (2006) demon- strates that productivity growth has been the dominant contributor to the growth in real welfare in Australia since 1960, with the terms of trade playing only a very minor role. Certainly prices in international markets for primary products relative to manufactures have been on a downward trend over the past century, but the decline has averaged less than 0.5 percent per year. But the difference between recent and earlier performance in Australia and New Zealand is also due very sub- stantially to economic policy reforms of the past three decades. The belated free- ing of markets in the two economies not only has arrested the decline in per capita income ranking in the two countries, but is having a remarkable influence on their patterns of production and trade. For these natural-resource-rich, relatively lightly populated economies,3 the most appropriate theory of comparative advantage is a blend of two core models developed in the 20th century: the Heckscher-Ohlin-Samuelson model, which assumes all factors of production are mobile between sectors, and the Ricardo- Viner model, which assumes some factors are sector-specific. Such a blend is pro- vided by Krueger (1977) and explored further by Deardorff (1984). They consider two tradable sectors each using intersectorally mobile labor plus one sector-specific factor (natural-resource capital or industrial capital). Assuming that labor exhibits diminishing marginal product in each sector, and that there are no serv- ices or nontradables and no policy distortions, then at a given set of international prices the real wage is determined by the aggregate per worker endowment of natural-resource and industrial capital. The commodity composition of a coun- try's trade--that is, the extent to which a country is a net exporter of primary or industrial products--is determined by its endowment of natural relative to indus- trial capital compared with that ratio for the rest of the world. Leamer (1987) develops this model further and relates it to paths of economic development. If the stock of natural resources is unchanged, rapid growth by one or more economies relative to others in the availability of industrial capital per worker would cause those economies to strengthen their comparative advantage Australia and New Zealand 225 in nonprimary products. However, a discovery of minerals or energy raw materials would strengthen that country's comparative advantage in mining and weaken its comparative advantage in farm and other goods, ceteris paribus. It would also boost national income and, hence, the demand for nontradables, which would cause mobile resources to move into the production of nontradables, further reducing farm and industrial production (Corden 1984). Domestic or foreign savings can be invested to enhance the stock and to improve the quality not only of industrial capital but also of labor or natural resources, and to provide capital to the nontradables sector. Any such increase in the net stock of produced capital per worker will put upward pressure on real wages. That will encourage, in all sectors, the use of more labor-saving techniques and the develop- ment and importation of new technologies that are less labor intensive. Which types of capital expand fastest in a free-market setting depends on their expected rates of return. The more densely populated, natural-resource-poor a country is, the greater the likelihood that the highest payoff would be in expand- ing capital stocks for nonprimary sectors. At early stages of development of such a country, when it has a relatively small stock of natural resources per worker, wages would be low and the country would have a comparative cost advantage in unskilled labor-intensive, standard-technology manufactures. As the stock of industrial capital grows, there would be a gradual move toward exporting more capital- and skill-intensive manufactures. Natural-resource-abundant economies such as Australia and New Zealand, however, would develop a comparative advan- tage in manufacturing at a late stage of development, and their industrial exports would be relatively capital intensive. In New Zealand, with lesser mineral and energy resources per worker and poorer climatic conditions for broadacre crop- ping than in Australia, agricultural comparative advantages would be stronger in aggregate but less focused on cereal and oilseed cropping than in Australia. The above theory of changing comparative advantages has been used successfully to explain the evolving pattern of exports of Australia and its Asian trading partners (Anderson and Garnaut 1980, 1987; Anderson and Smith 1981; Anderson 1998), and is also consistent with New Zealand's trade pattern. It can be used also to explain shocks to that evolutionary pattern, as with mining booms; and it is consistent with the larger shares of farm revenue from livestock and horticultural crops compared with grains, oilseeds, sugar, and cotton for New Zealand relative to Australia. But the evolving pattern of a country's production and trade specialization also depends on policy choices and their changes over time. In the case of Australia and New Zealand, their long history of industrial protectionism (reflected in the relatively high implicit tariffs) resulted in a smaller share of GDP traded than would be expected for economies of their size.4 Protectionism also ensured a big- ger manufacturing sector than would have emerged under free trade, which was 226 Distortions to Agricultural Incentives: A Global Perspective possible in their full-employment setting only at the expense of other sectors. In both countries, the proportion of GDP accounted for by manufacturing in the early 1960s were close to the OECD average of around 30 percent, even though Australia and New Zealand have always been lightly populated and so have weak comparative advantage in manufactures. The removal of the ban on key mineral raw material exports in the early 1960s and the tariff reforms of the 1970s and 1980s, however, corrected that distortion for Australia. Between 1960 and 2005, manufacturing's share of GDP fell much more rapidly for Australia than for the average OECD country, to just 11 percent, while the mining sector's share trebled (Anderson, Lloyd, and MacLaren 2007). The share of Australian exports accounted for by mining, meanwhile, more than trebled between the early 1960s and early 1980s, helped by the dramatic rises in energy raw material prices during the 1970s.5 Even though that trend lowered agriculture's relative contribution, the share of exports in the gross value of farm production increased considerably, from around 55 percent in the early 1960s to 75 percent after the mid-1970s (Anderson et al. 2008). Moreover, the growth in agricultural exports came from an increasing range of farm products as farmers diversified away from the traditional wheat and sheep enterprises to beef, cotton, sugar, dairy products, wine, and rapeseed. Natural-resource-based products were not the only exportables discouraged by protectionist measures in Australia and New Zealand, however. Export industries within the manufacturing sector, as well as services exports, were also affected. Together, those two sectors represented one-twelfth of Australia's exports in the early 1950s. By 1980, their contribution was barely above one-quarter, but by 1990 it had risen to one-third and by 2005 to 44 percent, or 22 percent for each sector, the first time either sector surpassed the 21 percent share of exports represented by agriculture. The transformation in New Zealand was similar, despite the absence of the mineral resources with which to create a mining boom. Furthermore, New Zealand was impacted much more than Australia by the coming into force from 1983 of the Australia New Zealand Closer Economic Relations Trade Agreement (ANZCERTA): its lower wages allowed it to rapidly expand exports of manufac- tures and services to the much bigger Australian economy under that preferential arrangement. Together with the virtual elimination of manufacturing protection, these forces brought to a halt the decline in agriculture's share of New Zealand's GDP, in fact raising it from a low of 7 percent in the late 1980s to 9 percent in the early 2000s--notwithstanding the abolition of nontrivial agricultural subsidies in the 1980s. Over that same period the share of food and agricultural products in New Zealand's exports have fallen somewhat, to the benefit of other manufactures and services. With the decade-long buildup and then removal in the mid-1980s of subsidies to agricultural exporting industries (see next section), agricultural and Australia and New Zealand 227 processed food as a percentage of the gross value of primary agricultural produc- tion expanded and contracted somewhat substantially, though it has since been stabilized despite the growth in exports of nonagricultural goods and services during the past two decades. Policy Evolution This section begins with a brief history of policies up to the early 1970s and then moves to a discussion of the changes in the next dozen years before reforms accel- erated in the mid-1980s. Since there were relatively few agricultural subsidies or farm import barriers (other than quarantine restrictions) through most of the past century, the story in both Australia and New Zealand is more about the indirect antiagricultural bias that resulted from the protection of manufacturing. Nonethe- less, policies that directly distorted various agricultural markets post­World War II are also examined.6 We know that it is relative prices, and hence relative rates of government assistance, that affect incentives. In a two-sector model an import tax has the same effect on the export sector as an export tax (the Lerner symmetry theorem); this carries over to a model that includes a third sector producing only nontradables (Vousden 1990). For that reason the average nominal rate of assistance (NRA) for the tradable parts of the agricultural sector is reported, based on NRA estimates for individual agricultural industries, along with the average NRA for the tradable parts of all nonagricultural sectors. The NRA is the equivalent of the percentage by which government policies have raised the producer price above what it would be without the government's intervention.7 With those two sectoral NRAs a relative rate of assistance (RRA) can be calculated. The RRA is defined as 100[(1 NRAagt 100) (1 NRAnonagt 100) 1], where NRAagt and NRAnonagt are the average percentage NRAs for the tradables parts of the agricultural and nona- gricultural sectors, respectively. Since the NRA must be greater than 100 percent if producers are to earn anything, so too must the RRA. The usefulness of this measure is that if it is below zero, it indicates the extent to which the policy regime has an anti- agricultural bias. The converse applies when RRA is positive. The cost of government policy distortions to incentives, in terms of resource misallocation, increases with the degree of substitution in production. In the case of agriculture, which involves the use of farmland that is sector-specific, the greater the variance of NRAs across industries within the sector, the higher the welfare cost of market interventions. A simple, if crude, index of that cost is the standard deviation of industry NRAs within agriculture. The weighted mean NRA for the sector (using the values of production at unassisted prices as weights), along with the standard deviation around that mean each year, is thus reported here. 228 Distortions to Agricultural Incentives: A Global Perspective Prior to the early 1970s The long history of industrial protectionism in Australia and New Zealand has its roots in the formation in 1901 of the Australian Federation, which New Zealand decided not to join, instead becoming separately independent of the United Kingdom. At that time, tariff revenue accounted for almost one-fifth of govern- ment revenue in both countries, a very high share for what at the time were the world's highest-income economies (the share of tariff revenue typically declines as per capita income rises) and twice that of the Nordic countries, for example, even though per capita incomes in those countries was barely half that in Australia and New Zealand (Anderson et al. 2008). Tariffs on manufactures rose steadily in the decades that followed (Lloyd 2008). They were supplemented by quantitative import restrictions first imposed in the late 1930s (as "wartime measures" in Australia and as industry protection in New Zealand), and then rose even further in the 1960s when they substituted for the import licences as the latter were removed (in 1960, with minor exceptions, in the case of Australia).8 This trend over the 1950s and 1960s contrasted strongly with what other high-income coun- tries were doing at that time, which was lowering tariffs on manufactures as part of multilateral trade negotiation under the GATT.9 By the early 1970s, the average manufacturing tariff in Australia and New Zealand exceeded that of any other OECD country (Anderson and Garnaut 1987). Meanwhile, the governments of Australia and New Zealand intervened in numerous markets for farm products, but the subsidies and protection they provided to agricultural industries was only a modest offset to the indirect disin- centives caused by manufacturing protection during this era. In the immediate post­World War II period, Australia's agricultural programs were directly taxing the farm sector. Most of that was removed by the end of the Korean War, however, at which time farmers enjoyed a boom in export prices that spurred the highest inflation in Australia since its gold-rush era of the 1850s. Farm assistance then rose gradually, such that by the end of the 1960s the NRA averaged 17 percent in Australia, whereas in New Zealand it averaged little more than 2 percent until the mid-1970s. The standard deviation of agricultural NRAs also rose during that period, indicating increasing misallocation of resources within the agricultural sectors of these two countries (tables 5.1 and 5.2). A striking feature of agricultural assistance in Australia and New Zealand at that time was that it applied to export industries as much as to import-competing ones (figure 5.2). Export industries in New Zealand, including wool, dairy prod- ucts, and meat, were assisted by so-called stabilization schemes (Sandry and Reynolds 1990). Similar schemes in Australia assisted wheat, manufactured dairy products, sugar, and dried vine fruit. The Australian schemes often also contained so-called home consumption price schemes, whereby domestic consumers were Table 5.1. NRAs to Covered Farm Products, Australia, 1946­2007 (percent, for fiscal years starting July 1) 1946­49 1950­54 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Exportables 7.5 0.9 6.4 7.0 10.0 7.6 3.6 4.6 5.6 4.8 3.0 0.0 0.0 Rice 3.2 1.1 11.4 15.0 14.8 22.0 20.4 15.2 10.6 2.5 2.3 1.7 1.9 Wheat 24.2 8.4 1.9 6.1 10.1 7.2 0.4 2.6 3.8 2.1 1.1 0.0 0.0 Barley 14.1 5.8 4.1 3.1 4.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Oats 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Grapes, total 10.5 4.5 5.6 9.7 18.7 39.7 19.2 21.3 18.3 13.3 4.9 0.0 0.0 Sugar 8.2 0.7 12.8 15.9 32.8 7.6 6.2 4.6 12.4 5.8 1.7 0.0 0.0 Cotton 0.8 2.0 26.7 52.1 73.9 53.4 17.6 4.4 2.0 0.0 0.0 0.0 0.0 Wool 0.0 0.0 0.0 0.0 0.0 6.0 1.4 1.0 1.0 5.4 0.7 0.0 0.0 Beef and veal 0.0 0.0 0.0 0.0 0.0 1.4 1.8 1.4 1.2 0.3 0.0 0.0 0.0 Mutton and lamb 0.0 0.0 0.0 0.0 0.4 1.6 1.8 1.4 1.8 0.9 0.0 0.0 0.0 Pig meat 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Milk 2.1 18.7 46.9 43.1 74.5 32.8 35.8 32.2 39.6 23.8 19.3 0.0 0.0 Apples -- -- 6.0 6.0 6.0 9.0 5.4 3.4 1.2 0.4 0.1 0.0 0.0 Sunflowers -- -- -- -- -- -- -- 5.6 1.4 0.0 0.0 0.0 0.0 Import- competing products 0.0 10.1 13.4 12.5 13.1 18.3 11.6 8.0 3.7 1.8 0.4 0.1 0.0 Maize 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Sorghum 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 2.0 0.0 0.0 0.0 0.0 Oilseeds 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Tobacco 0.0 34.2 51.0 46.9 51.3 250.0 122.2 56.4 37.6 48.5 19.8 0.0 0.0 229 Poultry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 (Table continues on the following page.) 230 Table 5.1. NRAs to Covered Farm Products, Australia, 1946­2007 (continued ) (percent, for fiscal years starting July 1) 1946­49 1950­54 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Bananas -- -- 0.0 0.0 0.0 0.0 0.0 4.8 1.0 0.1 0.0 0.0 0.0 Olives -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Oranges -- -- 25.0 25.0 25.0 25.8 32.8 38.2 13.0 2.7 0.7 0.6 0.6 Soybeans -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 Nontradables 1.2 12.6 31.4 41.6 78.1 25.3 19.5 24.2 12.2 1.8 0.2 0.0 0.0 Eggs 1.7 14.7 43.7 61.8 141.2 35.0 26.0 35.8 18.4 3.4 0.4 0.0 0.0 Potatoes 0.0 8.0 8.0 8.0 8.0 7.2 7.2 8.0 3.2 0.0 0.0 0.0 0.0 Weighted average of covered products 7.0 1.8 7.8 8.5 12.3 8.8 4.6 5.4 5.7 4.4 2.6 0.0 0.0 Domestic market support -- -- 1.4 1.6 2.5 1.0 0.8 0.9 0.5 0.4 0.2 0.0 0.0 Border market support -- -- 6.5 7.0 9.8 7.8 3.9 4.6 5.2 4.0 2.5 0.0 0.0 Dispersion of covered products 7.4 12.2 17.3 21.3 36.1 54.0 27.9 16.9 12.2 10.9 5.8 0.4 0.4 % Coverage (at undistorted prices) 91 84 82 86 87 84 85 86 75 82 80 80 80 Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. Note: -- not available. a. Weighted averages, with weights based on the unassisted value of production (actual back to 1966, and the average for 1966­69 for earlier years). Dispersion of NRAs is their standard deviation of the simple five-year average of the annual standard deviation around the weighted mean. Table 5.2. NRAs to Covered Farm Products, New Zealand, 1955­2007 (percent) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Exportables 0.1 0.1 0.2 2.8 13.1 18.8 11.8 1.2 0.8 0.9 0.6 Barley -- -- -- -- -- 10.6 1.8 0.0 0.0 0.0 0.0 Beef 0.1 0.1 0.3 5.0 10.0 15.6 11.0 1.4 1.0 1.0 0.3 Coarse grains 4.0 4.0 4.0 4.0 4.0 4.0 2.2 0.4 0.0 0.0 0.0 Other fruits and vegetables 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Maize -- -- -- -- -- 12.0 2.2 0.0 0.0 0.0 0.0 Milk 0.2 0.2 0.2 1.0 16.0 18.0 11.6 1.4 1.0 1.0 0.3 Oats -- -- -- -- -- 9.2 2.4 0.0 0.0 0.0 0.0 Other crops 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Sheep meat 0.1 0.1 0.3 5.0 19.0 32.9 27.8 1.8 1.0 1.0 0.3 Wool 0.0 0.0 0.0 5.0 11.0 19.0 10.2 1.4 1.0 1.0 0.3 Import-competing 28.3 28.3 28.8 32.0 27.0 28.1 41.1 22.5 19.6 15.8 8.9 Eggs 59.0 59.0 59.0 59.0 59.0 59.0 80.2 38.2 50.4 36.4 25.0 Pig meat 2.0 2.0 2.8 5.4 19.9 10.3 2.8 0.0 0.0 0.2 0.3 Poultry 31.0 31.0 31.0 31.0 31.0 31.0 61.6 57.0 40.8 34.6 28.3 Wheat 11.0 11.0 11.0 11.0 11.0 11.0 6.6 0.6 0.0 0.0 0.7 Mixed trade status Grapes 120.0 120.0 134.0 106.0 53.8 23.2 23.8 5.0 5.0 5.0 5.0 All covered 1.8 1.8 1.9 5.0 14.3 20.0 14.9 2.9 2.1 2.0 1.3 Domestic market support 0.0 0.0 0.0 0.0 0.0 0.2 2.0 0.0 0.0 0.0 0.0 Border market support 1.8 1.8 1.9 5.0 14.3 19.8 12.9 2.9 2.1 2.0 1.3 Dispersion of NRA of covered products 40.7 40.7 44.5 36.3 23.2 17.4 27.1 17.5 17.3 14.0 9.8 % Coverage at undistorted prices 100 100 100 100 100 100 100 100 100 100 80 Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. 231 Note: -- not available. a. Weighted averages, with weights based on the unassisted value of production (actual back to 1966, and the average for 1966­69 for earlier years). Dispersion of NRAs is their standard deviation of the simple five-year average of the annual standard deviation around the weighted mean. 232 Distortions to Agricultural Incentives: A Global Perspective Figure 5.2. NRAs to Exportable, Import-Competing, and Alla Covered Products, Australia and New Zealand, 1946­2007 a. Australia 25 20 15 10 percent 5 0 5 10 15 20 06 01 46 51 56 61 66 71 76 81 86 91 96 20 20 19 19 19 19 19 19 19 19 19 19 19 b. New Zealand 45 40 35 30 percent 25 20 15 10 5 0 9 4 9 4 9 4 9 4 9 4 7 ­5 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 ­0 55 60 65 70 75 80 85 90 95 00 05 19 19 19 19 19 19 19 19 19 20 20 total import-competing products exportables Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. a. The total NRA can be higher than both the exportable and importable averages because nontradables assistance is also included. forced to pay more than the export price (Sieper 1982; Edwards 2006). These schemes, which required the pooling of domestic and export returns, could only be implemented with the support of the Australian state governments.10 All other policy measures combined--fertilizer subsidies, income tax incentives, rural credit measures, subsidies to agricultural research and extension, and public Australia and New Zealand 233 investment in land and water development and rural infrastructure--added the equivalent of no more than 2 percent to Australian farmers' gross income as of the early 1970s (table 5.3). The net effect of farm and nonfarm policies on agricultural incentives is sum- marized in table 5.3 and illustrated in figure 5.3. For Australia, the negative effect on incentives from agricultural policies in the 1940s was trivial compared with that from nonagricultural ones, mostly import protection for manufacturers. Together, those policies effectively reduced farmers' gross returns by more than 20 percent. Price stabilization and other agricultural policies gradually provided more direct assistance to Australian farmers over the 1950s and 1960s, when manu- facturing protection remained steady, so that degree of overall taxation fell from 27 percent to just under 10 percent by 1965­69, as measured by the RRA. Of that decline, about two-thirds is due to changes in nonagricultural policies and only one-third to changes in direct assistance to farmers. New Zealand farmers, mean- while, were effectively taxed an average of more than 20 percent until 1972­73. The degree of antiagricultural bias in the two countries to the early 1970s was thus very similar to that of many developing countries in those decades. Meanwhile, by the late 1960s, home consumption price schemes imposed tax equivalents of over 100 percent on Australian consumers of butter, cheese, sugar, and eggs. These dis- tortions are unusual in that the imposition they imposed on consumers or users of exportables was greater than the price impact on producers--something that does not arise from standard trade barriers. Reforms of the early 1970s Disenchantment with interventionist trade and related economic policies gradually spread in the 1960s. At the 1968 National Development Conference in New Zealand, policy makers nearly agreed to simultaneously reduce import protection and tariff compensation to agriculture. But it was not until the 1970s that tariff reductions began in both countries. In Australia, a 25 percent across- the-board cut in July 1973, preceded by some minor cuts in 1970­71, started the tariff reform process. It was accelerated in the 1980s and continued through the 1990s. As a result, the average NRA to Australian manufacturing fell from 23 to 3 percent, and the effective rate from 36 to 5 percent over those three decades. In the 1990s alone, both the mean and the standard deviation of Australia's import tariffs on goods halved, bringing the average tariff for manufactures down to 4.2 percent in 1999.11 Currently, the only manufacturers with significant tariff protection are motor vehicles and parts, textiles, clothing, and footwear. Excluding those products, the average effective rate of assistance to Australian manufactur- ing is just 3 percent (Productivity Commission 2000a).12 Table 5.3. NRAs to Agricultural Products Relative to Nonagricultural Industries, Australia and New Zealand, 234 1946­2007 (percent, for fiscal years starting July 1) a. Australia 1946­ 1950­ 1955­ 1960­ 1965­ 1970­ 1975­ 1980­ 1985­ 1990­ 1995­ 2000­ 2005­ 49 54 59 64 69 74 79 84 89 94 99 04 07 NRA, covered products 7.0 1.8 7.8 8.5 12.3 8.8 4.6 5.4 5.7 4.4 2.6 0.0 0.0 NRA, noncovered products 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, all agriculture (excluding NPS) 6.4 1.5 6.3 7.4 10.7 7.6 3.9 4.6 4.4 3.7 2.1 0.0 0.0 All importables -- -- 3.8 4.5 4.9 7.4 4.9 3.5 1.2 0.8 0.2 0.0 0.1 All exportables -- -- 5.7 6.4 9.3 7.0 3.3 4.2 4.8 4.3 2.6 0.0 0.0 All nontradables -- -- 31.4 41.6 78.1 25.3 19.5 24.2 12.2 1.8 0.2 0.0 0.0 Trade bias index -- -- 0.02 0.02 0.04 0.00 0.02 0.01 0.04 0.03 0.02 0.00 0.00 NPS 0.6 1.2 1.8 2.1 2.0 2.1 1.4 1.1 0.9 0.7 0.8 0.5 2.2 Inputs -- -- -- -- -- -- 1.2 0.7 2.0 3.0 3.3 2.8 2.8 Other -- -- -- -- -- -- 1.4 1.1 1.1 2.3 2.5 2.3 0.6 NRA, all agriculture (including NPS) 5.7 2.7 8.1 9.5 12.7 9.4 5.3 5.8 5.3 4.4 2.9 0.5 2.2 NRA, decoupled assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 1.4 1.2 1.2 2.3 2.6 NRA, all agricultural products (including NPS and decoupled assistance) 5.7 2.7 8.1 9.5 12.7 9.4 5.4 6.6 6.6 5.6 4.1 2.8 4.8 NRA, all agricultural tradables (including NPS) 5.9 2.3 7.3 8.4 10.9 8.9 4.9 5.2 5.1 4.5 3.0 0.5 0.2 NRA, all nonagricultural tradables 28.0 23.5 19.6 20.7 20.7 16.8 12.0 11.1 8.2 5.3 2.6 2.0 2.0 RRA 26.3 17.1 10.3 10.2 8.2 6.8 6.4 5.3 2.9 0.7 0.4 1.5 1.8 b. New Zealand 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, covered products 1.8 1.8 1.9 5.0 14.3 20.0 14.9 2.9 2.1 2.0 1.3 NRA, noncovered products n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. -- NRA, all agricultural products (excluding NPS) 1.8 1.8 1.9 5.0 14.3 20.0 14.9 2.9 2.1 2.0 1.2 All importables 28.3 28.3 28.8 32.0 27.0 29.6 42.1 24.3 20.7 17.6 9.8 All exportables 0.1 0.1 0.2 2.8 13.1 18.9 12.0 1.2 0.8 0.9 0.6 All nontradables -- -- -- -- -- -- -- -- -- -- -- Trade bias index 0.22 0.22 0.22 0.22 0.11 0.08 0.21 0.18 0.16 0.13 0.07 NRA, NPS -- -- -- -- -- -- 5.5 0.5 0.6 0.4 0.3 Inputs -- -- -- -- -- -- 4.6 0.5 0.6 0.4 0.3 Other -- -- -- -- -- 8.9 0.9 0.0 0.0 0.0 0.0 NRA, all agriculture (including NPS) 1.8 1.8 1.9 5.0 14.3 28.9 20.9 3.4 2.6 2.4 1.4 NRA, decoupled assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 NRA, all agricultural products (including NPS and decoupled assistance) 1.8 1.8 1.9 5.0 14.3 28.9 20.9 3.5 2.6 2.4 1.5 NRA, all agricultral tradables (including NPS) 1.8 1.8 1.9 5.0 14.3 28.9 20.9 3.4 2.6 2.4 1.4 NRA, all nonagricultural tradables 21.3 24.0 34.3 30.0 21.7 20.3 16.6 10.8 6.5 3.8 3.3 RRA 16.1 17.8 24.1 19.0 6.0 7.1 3.5 6.7 3.6 1.3 1.8 Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. Note: -- not available; n.a. not applicable. a. Before including Non-Product Specific (NPS) assistance. 235 b. Total of assistance to primary factors and intermediate inputs divided by the total value of primary agriculture production at undistorted prices (percent). c. RRA 100[(1 NRAagt 100) (1 NRAnonagt 100) 1], where NRAagt and NRAnonagt are the average percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. 236 Distortions to Agricultural Incentives: A Global Perspective Figure 5.3. NRAs to Manufacturing, All Nonagricultural Tradables, All Agricultural Tradable Industries, and RRA,a Australia and New Zealand, 1946­2007 a. Australia 50 40 30 20 10 percent 0 10 20 30 40 50 01 06 46 51 56 61 66 71 76 81 86 91 96 20 20 19 19 19 19 19 19 19 19 19 19 19 nonagricultural tradables agricultural tradables RRA b. New Zealand 60 50 40 30 percent 20 10 0 10 20 30 9 4 9 4 9 4 9 4 9 4 7 ­5 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 ­0 55 60 65 70 75 80 85 90 95 00 05 19 19 19 19 19 19 19 19 19 20 20 manufacturing nonagricultural tradables agricultural tradables RRA Source: Anderson and Valenzuela (2008), drawing on authors' spreadsheet. a. The RRA is defined as 100 [(100 NRAagt ) (100 NRAnonagt) 1]. Australia and New Zealand 237 Australia's agricultural subsidies and regulatory interventions also have been close to eliminated over those 35 years. The average NRA to the farm sector has fallen from 16 percent in the early 1970s to less than 2 percent at present, and the standard deviation has fallen from almost 60 percent to less than 1 percent. The process was piecemeal13 and gradual, often involving a series of partial steps, but it was persistent, beginning in 1972 with milk and then two to four years later with cotton and tobacco. It took another decade, though, for supports for rice and eggs to begin to be dismantled, and almost a further decade for assistance to the grape industry to be cut. The process also experienced a significant reversion--the introduction of the reserve price scheme for wool in 1973--that took until the early 1990s to unravel.14 There were just two farm groups still benefiting signifi- cantly from government programs in the latter 1990s: tobacco and milk produc- ers, each with an NRA of more than 20 percent in 1997. Deregulation of tobacco marketing arrangements began in 1995 and was completed in 2000, bringing effective assistance to tobacco growing down from 30 to 2 percent over that period. As from July 1, 2000, the remaining impediments to a free domestic mar- ket in fluid milk began to be dismantled, for which an untied, one-off compensa- tion grant (providing a total of around US$1.5 billion, or US$110,000 per dairy farm) has been paid, funded by a consumer levy at the retail level over 2000­08. Contrary to some pessimistic forecasts at the time, dairy output has continued to increase as production from those leaving the industry (17 percent over the first three years) has been more than compensated by output and productivity improvements on remaining farms (Harris 2005a, 2005b). In New Zealand, the antiagricultural bias declined in the 1970s as the result of a modest reduction in manufacturing protection and a big boost to beef, sheep, and dairy farmer assistance. In these two steps, the effective taxation of agriculture (the negative of the RRA) fell from 22 to 14 percent, and then to less than 7 per- cent. It fell further over the next dozen years as agricultural assistance increased slightly and manufacturing protection continued to fall slightly, then rose a little from the late 1980s as assistance to farmers fell faster than that to manufacturing, and finally fell to just 1 percent as the last of the interventions were removed from the late 1990s (figure 5.3). Thus, distortionary government assistance to both manufacturing and agricul- ture, and hence the overall antiagricultural bias, has now all but disappeared in both Australia and New Zealand after being in place for more than seven decades.15 Over the past 35 years, it was nonagricultural policy that overwhelm- ingly contributed to the improvement of farmers' incentives in Australia and New Zealand--in fact, the decline in manufacturing protection was sufficiently large enough to more than offset the decline in direct assistance to farmers from the late 1960s. Farmers also have benefited from the fact that service sectors too have not 238 Distortions to Agricultural Incentives: A Global Perspective been spared reform in these countries. Banking; post and telecommunications; ports; higher education; health; and rail, air, and sea transport have been opened up. In addition, there has been progressive outsourcing of many government serv- ices; and substantial reforms to competition policy and practice, including priva- tization and the corporatization and demonopolization of numerous government enterprises, are well advanced.16 Moreover, by 1985 both currencies were floating and foreign investment flows began to be freed up. That complemented financial sector reform and contributed to foreign direct investment, equity and foreign currency transactions growing at several times the pace of GDP. Even the previ- ously highly unionized labor markets have undergone considerable reform. Households have gained substantially from these widespread reforms, including consumers of food who for most of the past two decades have faced tax equiva- lents of well below 10 percent on their food purchases (compared with the OECD average of between 23 and 36 percent over that period--see OECD 2008). Reasons behind the Policy Evolution In the early postwar years, successive Australia and New Zealand governments sought to expand the output of the agricultural sector in order to improve the bal- ance of trade position in the context of a fixed exchange rate regime. There were numerous factors in common, but since some were country-specific, it is better to consider the two countries separately. Australia In the Great Depression years of the 1930s, and even more so in the two decades following World War II, agricultural policy instruments were introduced in profu- sion, including price stabilization schemes implemented through buffer funds, guaranteed prices and deficiency payments, home-consumption price schemes for wheat, dairy, sugar, and fruits and vegetables; input subsidies on fuel, fertilizers and interest rates; import subsidies; tax concessions; publicly funded research and development; publicly funded extension services; land development schemes; and the provision of rural infrastructure (Godden 1997; Mauldon 1990). For some commodities--for example, sugar and wheat--there were national- level statutory marketing authorities (SMAs) or, in the language of the GATT and, later, the World Trade Organization (WTO), state trading enterprises (STEs). This instrument was also used by state governments across a wide range of commodi- ties. It often took the form of a producer-controlled board, with compulsory monopsony or monopoly powers in order to effect "orderly marketing" and sometimes used in conjunction with marketing quotas. Frequently, the SMAs Australia and New Zealand 239 were linked with regulations that required the shipment and storage of grain to be handled by state-owned enterprises. This form of intervention by Commonwealth and state governments began in the 1920s and became central to marketing policy (Mauldon 1990).17, 18 In 1973, the Australian Labor Party won office and, as a political party with few roots in rural areas, set about establishing a review of the objectives and instru- ments of agricultural policy in order to define a set of principles that would underpin intervention. The result was the so-called Green Paper (Harris et al. 1974). The approach adopted by the authors was that of applied welfare economics: market failure provides a reason for government intervention to be potentially beneficial through improving economic efficiency. Some years later, Sieper (1982) argued that the public choice or private interest explanations fit better with reality, and particularly with the choice of policy instruments, than those provided by the public interest approach. Although the insights provided by that set of theories were not entirely new, Sieper's use of them provided fresh and powerful insights into the Byzantine world of Australian agricultural policy as it existed up to the late 1970s. In 1974, the Industries Assistance Commission (IAC) was established from the then-existing Tariff Board. The IAC was designed to act as a forum for public dis- cussion of economic policy by undertaking reviews and evaluation of economy- wide government intervention and regulation. During the 1970s and 1980s, it produced reports on wheat, sugar, dairy, dried vine fruits, rural adjustment, and income stabilization (Edwards 1987). This development signalled the govern- ment's intention that policies should be evaluated from the viewpoint of the public interest rather than, or perhaps as well as, the private interests of farm groups and regions. In other words, the year 1973 marked a watershed: applied welfare economic analysis would be used to analyse the effects of various policy instruments and the recommendations of that analysis would be treated objec- tively, even if not always implemented in full (or, in a few cases, at all). In 1982, the Balderstone Report set out the agricultural policy options for the 1980s (Balderstone 1982). The basic thrust was that of reduced protection, not just in agriculture but throughout the entire economy. The authors supported the position that the agricultural sector should receive compensation on the grounds of equity (but not efficiency as discussed in the 1970s; see Martin [1990]) for hav- ing to pay tariffs on imported inputs. In the report, the authors adopted an incre- mentalist approach that supported the continuation of many of the then-existing instruments--for example, input subsidies and price discrimination between domestic and export markets. They also suggested that the objective of price stabilization should be achieved through price underwriting linked to market conditions rather than through existing instruments such as buffer funds. 240 Distortions to Agricultural Incentives: A Global Perspective In 1988, the Commonwealth government issued a policy statement stating that the agricultural policy objectives were to be, among other things, enhancing pro- ductive capacity through balanced management of natural resources, development of human skills and improved research; developing a more productive industry structure through an economywide approach to policy reform with lower and more balanced assistance; responding to the international environment through better marketing of agricultural produce and reductions in international trade distortions; and offering positive assistance for structural adjustment (see Godden 1997). With this, it was obvious that the policy objectives of the earlier postwar years had been substantially altered. The Australian Wheat Board lost its monopoly in the domestic food wheat market in 1989, as had been recommended by the IAC in 1984, but it retained (and continues to retain) its export monopoly. In 2000, the single export desk for wheat was reviewed under the National Competition Policy and, in the absence of strong evidence one way or the other, the government accepted the recommenda- tion that the export monopoly be retained until 2010. Reviews under National Competition Policy of the single export desk for rice and for sugar also recom- mended their continuation, based on a national interest test. In 1991, a review of the statutory arrangements for the marketing of agricul- tural products (Industry Commission 1991) was published. The investigation focused on the economic effects of these arrangements, in particular the effects on efficient resource allocation. The Commission concluded that although many of the arrangements were beneficial to producers and to society generally, there were instances where this was not so. The SMAs did not operate in the public interest when they had powers to require producers to participate (that is, the compulsory acquisition of product), when potential producers were excluded from entry (for example, when there are nontradable quotas in place), or where prices were raised for users and consumers (for example, price discrimination between the domestic and export markets). In these instances, it was concluded that efficiency of resource use was not achieved.19 The outcome of this review has been the disap- pearance of most, but not all, SMAs. From its beginnings in the latter 1970s, the leadership of the National Farmers Federation (NFF, and before it the Grains Council) adopted an economy-wide approach to economic issues. The activities of the NFF also encouraged the devel- opment of a free-market faction in the Liberal Party in the late 1970s (Anderson and Garnaut 1987). That approach was to be the hallmark of much of the rigor- ous computable general equilibrium analysis conducted by the IAC over the decade to 1989 on matters both agricultural and more general. Early in that period, the government had convinced the farm sector of the need for rationaliza- tion of the sector. As a quid pro quo, the government would introduce measures Australia and New Zealand 241 to improve efficiency in the sectors supplying inputs to agriculture as well as the infrastructure required for distribution--for example, rail, storage, and the wharves. However, some commodity groups within the NFF complained that the leadership of the organization had turned its back on protecting its constituents from harsh economic conditions (for example, drought and the vagaries of inter- national markets) and from the deregulatory zeal of the government (including through one of its main sources of advice on agricultural matters, namely, the IAC/IC). The objectives and the instruments of agricultural policy that existed from the early postwar years until the late 1980s reflected not only the supply of assistance to producers through government from consumers (users) and taxpayers but also the demand for assistance from producer groups. As shown by Anderson (1978), the interaction between the two sides of the political market generated a wide dispersion of support from the late 1960s to the mid-1970s. His findings are con- sistent with the interest group theory of political economy, or public choice theory: the land-intensive subsectors such as cereals and livestock were found to be lightly supported, whereas the labor-intensive subsectors such as milk produc- tion and horticulture were heavily supported. Martin (1990) outlined several models in the literature on public choice theory and evaluated the extent to which each was relevant to the developments in agricultural policy that occurred in the 1980s. In assessing the role of the IAC, he concluded that "the establishment of the IAC also appears to have had a remarkably strong influence on the policy process, and to have been associated with and contributed to the change in ideas about industry assistance measures" (p. 198). MacLaren (1992) argues, in a comparison of the processes of agricultural policy reforms in Australia and the European Community (EC), that the IAC proved significant in the reform of Australian agricultural policy because it altered the political economy of policy making. No other industrial country has a body with the functions of the IAC, and the lack of such a body may help to explain why reductions in agricultural protectionism in other industrial countries, with the exception of New Zealand, have been very dif- ficult to achieve. Over the past 25 years, much of the intervention in both agriculture and manu- facturing in Australia has disappeared. In agriculture, the focus of government policy on incomes and prices has given way to concerns about the quality of the resources used by the agricultural sector and the efficiency with which they are used. During the same period, the principal farm lobby groups have acquiesced to the view that markets can perform better than governments in the provision of an economically healthy sector. Subsectors that traditionally were heavily supported have accepted the benefits of reduced market-based intervention. In 2000, even the dairy sector accepted total deregulation of the prices of both market and 242 Distortions to Agricultural Incentives: A Global Perspective manufacturing milk. The subsequent free interstate trade in market milk led to better exploitation of regional comparative advantages. The bribe paid to secure acceptance by all regions was an adjustment assistance scheme financed by con- sumers who will continue to pay a high price for milk during the transition period of eight years.20 This acceptance of reduced government involvement in the mar- kets for agricultural products contrasts markedly with the situation in most other OECD countries, and in many developing countries, as confirmed by the contin- uing difficulties in defining modalities for domestic support and market access in the Doha Round. New Zealand Well into the 1960s, New Zealand continued an isolationist economic policy (import selection through licensing) that had been introduced just after the Great Depression. Import selection effectively prohibited imports of competing goods by using a system of import licensing underpinned with high tariffs. Import pro- hibitions remained in place for selected food products, the most notable of which was for margarine, which could be purchased only with a doctor's certificate! Price supports, coupled with import licensing, increasingly protected the domes- tic wheat industry. At the other extreme, export prohibitions on coarse grains tended to lower the price of feed wheat, barley, and oats to poultry and pork pro- ducers and other users. For a short period around 1955, the import licensing regime was liberalized (Rayner and Lattimore 1991), but a balance of payment crisis in 1957 (falling dairy product prices) caused a policy reversal. In this strong import-substitution environment, exports were heavily concentrated in meat, wool, and dairy products--the industries in which New Zealand had its strongest comparative advantage and to which the government provided the least assistance (table 5.2). Until the mid-1950s, these "pastoral" exports represented nearly 95 percent of total merchandise exports from New Zealand. At that point, nonagricultural exports began to grow, so that by 1969 agriculture's share of merchandise exports fell to 84 percent, in part due to the advent of the highly selective New Zealand­Australia Free Trade Agreement that began in 1966. The New Zealand economy suffered a series of domestic and international eco- nomic shocks over the seven-year period starting in 1967 (Dalziel and Lattimore 2004). In 1967, export earnings were severely reduced by a fall in international wool prices, leading to the exchange rate being devalued. Furthermore, New Zealand was beginning to import inflation via its fixed exchange rate policy and the expanding U.S. fiscal deficit surrounding the Vietnam War. Domestically, an increase in the rate of inflation led to a breakdown in the national wage-setting Australia and New Zealand 243 mechanism, resulting over the next few years in large wage increases. The govern- ment then began increasing farm input subsidies to counteract the drop in income. Another terms-of-trade shock came in 1972­73, when meat prices rose to high levels as part of a general agricultural commodity price boom. Though New Zealand revalued its exchange rate to counteract the macroeconomic effects, the United Kingdom joined the European Economic Community (EEC) in 1973, reducing New Zealand's longstanding preferential access to this very profitable agricultural export market.21 The first oil crisis came close on the heels of these occurrences, causing an additional major adverse terms-of-trade shock. New Zealand's economic policies changed rather dramatically as a result. In an attempt to maintain consumption in the face of drastic falls in the terms of trade, a com- mand economy approach was adopted, under which industries, prices, interest rates, wages, and foreign exchange rates were more heavily regulated. High infla- tion and large fiscal and current account deficits ensued. Foreign borrowing increased significantly during this period, and was used in part to finance major energy projects in an attempt to reduce New Zealand's reliance on imported fuel. In order to finance this debt, agricultural output subsidies on exportables were used for the first time in an attempt to generate additional export earnings. Farm development subsidies also were increased substantially. The sheep and wool industries were the favored targets because they were the largest agricultural and export sectors, but export subsidies were also provided for some nontraditional export products. The result was a large increase in the NRA offered to the exportable sector and a closing of the historically large gap between rates of assistance to the import-competing and export sectors for about a decade (figure 5.2b). The NRA on outputs in exportable primary agricul- ture rose to 19 percent over 1980­84 (table 5.2). During the reform period from the mid-1980s, output assistance to exportable primary agriculture was abruptly removed and the rate of assistance fell to almost zero in less than a decade. The sharp rises within exportable primary agriculture in the early 1980s included 24 percent for beef in 1982 and 59 percent for mutton and lamb production in 1984. Output assistance was nearly as high for raw milk and wool production. No significant output subsidies applied to the other exportable subsectors, such as fruit and vegetables and deer production, nor to import-competing subsectors such as grains and oilseeds. Output assistance for other animal products appears to be the result of high nontariff phytosanitary barriers on pig meat (during the 1980s), poultry, and eggs. The NRAs for favored exportables (sheep, beef, wool, and dairy) all reached a peak in the 1975­84 period, although not in the same years because the deficiency payments were tailored to the different world price levels for each year. Most of 244 Distortions to Agricultural Incentives: A Global Perspective the importable NRAs were zero because New Zealand does not produce sugar, cotton, soybeans, or rice. The first agricultural adjustments occurred in 1981 in wheat, tobacco, and meat processing. Tobacco growers were offered a production quota buyout of NZ$7,000 per hectare. These quotas resulted from local-content regulations, which prescribed that half of tobacco leaf used for processing by the two firms (Wills and Rothmans) had to be grown in New Zealand. Around two-thirds of the producers accepted the grant and relinquished their production rights. The remaining producers continued in production. Wills ceased operations in New Zealand in 1987 but Rothmans continued until 1994, paying NZ$6.52 per kilo- gram for dried leaf grown in New Zealand even though the import price (and over-quota price) was around NZ$4 per kilogram. The NRA on tobacco at this time was, accordingly, 63 percent. In 1983, New Zealand entered into a new and much more comprehensive free trade agreement with Australia (ANZCERTA). This arrangement was the first vehicle, after 1955, used to begin the process of removing the 1938 import licens- ing system and reducing import tariffs. In the same year, under some pressure from the United States, it was announced that the agricultural output subsidies would be abolished. Wide-ranging macroeconomic and microeconomic reforms were introduced in 1984. They included monetary and fiscal policy reform and reductions in industry regulation and support. The process of removing almost all agricultural subsidies was begun by abolishing agricultural output subsidies and phasing out agricultural input subsidies--a process that was completed by 1990. Policies affecting nontradables also were important, as they promoted greater macroeco- nomic stability and resource mobility, both of which increased the effectiveness of changes in trade policies. The government did not remove the export monopoly rights of agricultural marketing boards over this period, and while it did signal in a general way that such action was on the agenda, it nonetheless introduced a new marketing board with monopoly export rights (for kiwifruit). Furthermore, wine- grape growers were given a subsidy to replace old varieties with newer ones. Import licenses for Australian goods were increasingly tendered and then abol- ished once tender premia had fallen to low levels. This was followed by the removal of all other import licenses; the last ones (on textiles, clothing, and footwear) were removed in the early 1990s. Tariffs were progressively reduced from 1986 until 2000. Further tariff reductions were then halted, as import tariffs had fallen to around the average of those in other high-income countries. Monetary policy in New Zealand was reformed in 1989, followed by liberaliza- tion of labor market legislation and fiscal policy constraints. These initiatives com- pleted the liberalization of the capital account begun earlier with the conversion to Australia and New Zealand 245 a floating exchange rate regime; full currency convertibility; and the removal of interest rate, price, and wage controls. In this environment, the New Zealand macroeconomy gradually stabilized over the period from 1984­91. The notable exception was unemployment, which continued to rise, to around 11 percent by 1991 (Evans et al. 1996). The unemployment rate did not fall to less than 4 per- cent, where it had been in 1984, until 2003. From 1990, the major change to agricultural policy was the removal of export monopoly rights administered by some agricultural marketing boards (for exam- ple dairy, meat, apples, and pears) and, in some cases such as dairy, the abolition of the marketing board itself. This process was complicated by the existence of bilateral import quotas--especially for exports to the European Union, United States, Canada, and Japan. In the case of dairy quotas, the import quota rights were given to the three export dairy companies. The largest of the three, Fonterra, subsequently bought those rights from the smaller ones. In the case of meat, a reg- ulatory body was retained within Meat New Zealand (the marketing and research organization that developed from the old New Zealand Meat Producers Board) to allocate bilateral quota use rights. Under pressure from trading partners in the Doha Round, Fonterra's quota rights are to be removed and allocated in an alter- native fashion that is yet to be decided. By 1990, government assistance to New Zealand agriculture was limited to biosecurity, adverse events relief, research and development subsidies, export market development assistance, and product quality control programs. The direct rate of assistance to exportable agriculture had fallen to around 1 percent. The consumer tax equivalent (CTE) for exportable food products was down to a similarly low level, and comprised mainly the import-export parity differential that can be gained by large export firms in dairy, meat, and forestry products. The CTE for importable food products is somewhat higher because there are still import tariffs on some of these import-competing items, although typically they are still within the range of 5­7 percent. The exceptions are poultry and eggs, imports of which are strictly controlled on animal health grounds. Accordingly, domestic prices of these items are significantly higher than world market prices. Because the OECD considers these phytosanitary restrictions nontariff barriers, they are analyzed here as elements of nominal assistance to outputs (even though part of that protection may be correcting for a market failure). By 2005, the rate of assistance to manufactured food products was in single digits. However, the traditional pattern of tariff escalation remains, in that tariffs on unrefined exportable products and noncompeting imported food components tend to be zero while higher value-added foods have tariffs in the 5­7 percent range. In the nonfood manufacturing sector, NRAs on output have fallen from 24 and 42 percent (on exportables and importables, respectively) in 1982 to 4 percent 246 Distortions to Agricultural Incentives: A Global Perspective in 2009. Following the removal of the last import licensing program in 1993, tar- iffs were reduced continuously until 2000, when the most favored nation (MFN) tariff reduction programs were halted. One reason for this policy change was to create bargaining chips in the pursuit of bilateral and plurilateral free trade agree- ments that New Zealand embarked on vigorously--most recently with China, the United States, and ASEAN (Association of Southeast Asian Nations) Plus Three. There were some larger, one-off tariff reductions over the period, though: early in the reforms the tariff on computers was reduced from 40 percent to zero, and in 1997 the last remaining MFN tariff on new cars (25 percent) was abolished. Prospects for Further Policy Reform Notwithstanding the huge amount of agricultural trade policy reform over the past two decades, plenty of policy issues remain on the table, predominantly in the natural resource and environmental areas. Three are illustrated here. The first is food and agricultural import restrictions for the protection of plants, animals, and human health, though the economic protection from import competition that this provides farmers has not been fully captured in the Produc- tivity Commission's NRA estimates for Australia, especially for horticultural products. While some of that protection may well be warranted on externality grounds, some (such as a complete ban on imports of certain fruits from all coun- tries) may be excessive from a national welfare viewpoint. The Australian govern- ment, mainly in response to pressure from other WTO members seeking greater market access, is slowly examining whether various measures are excessively restrictive. Typically, consumer costs are not included in such assessments, nor are any of the cheaper ways of reducing costs associated with the importation of disease (James and Anderson 1998). New Zealand, in particular, would be a beneficiary of a more liberal quarantine policy regime in Australia. Second, neither Australia nor New Zealand has so far allowed the commercial growing of genetically modified (GM) varieties of farm products, with the sole exceptions of cotton and carnations in Australia. While the Office of the Gene Technology Regulator in Australia has approved the commercial growing of GM canola, the state governments have placed moratoria on plantings. GM food can be sold only if strict labelling standards are adhered to (FSANZ 2007). These restric- tions may or may not be in economies' and consumers' interests, depending on their impact on market access abroad for Australian and New Zealand farm prod- ucts and on human health and the environment at home (Anderson and Jackson 2005), but emotion has played a larger role in formulating these policies than has sound technical and economic analysis, especially in the growing of GM crops. Third, environmental policy has become a very important issue in agriculture in Australia and New Zealand (Vitalis 2005). Water policy, for example, which was Australia and New Zealand 247 already becoming a major economic and political issue, was brought to a head in Australia during 2006 following the country's worst drought on record. Though there has been substantial reform in water policy reform in Australia in recent years, much remains to be done to make the most of this resource, particularly in rural areas, where the majority of water is used and proposals for reform and sev- eral national enquiries are underway (see, for example, Productivity Commission 2006). More efficient pricing of water may lead to substantial reallocation of resources within the agricultural sector, with possible declines in Australian pro- duction of cotton, rice, and milk as horticultural industries (and urban areas) bid away water from farmers.22 The remaining major frontier for policy reform that would boost farm incomes in Australia and New Zealand is the dismantling of agricultural subsidies and import protection abroad. A successful conclusion to the Doha Round of negotiations in the WTO provides the greatest promise for achieving that out- come. According to global Linkage Model results reported in Anderson, Martin, and van der Mensbrugghe (2006), removing all merchandise trade barriers and agricultural subsidies globally would raise agricultural value added or net farm incomes in Australia and New Zealand by 26 percent. Even using the Global Trade Analysis Project (GTAP) Model, in which supply response elasticities are lower than in Linkage, a similar study by Anderson and Valenzuela (2007) reports that agricul- tural value added in Australia and New Zealand would rise by 15 percent, compared with a rise of just 2 percent in value added by nonagricultural industries. Policy Lessons for Other Economies By way of conclusion, two lessons are worth emphasizing from the Australia and New Zealand experiences. For agricultural-subsidizing countries, these case stud- ies show that removing even the largest and longest-lasting farm subsidies is pos- sible. Even where that was done by providing generous adjustment assistance, support was time-bound. In the Australian case, adjustment assistance was able to be financed simply by delaying the rewards to domestic consumers and thereby creating a gap between the producer and consumer price, rather than through outlays from (and hence resistance by) the treasury.23 For developing countries still effectively taxing their agricultural sectors, these two case studies offer hope that good policy analysis and advisory institutions will be able to alter the political economy sufficiently to remove that taxation. More than that, the Australia and New Zealand cases illustrate the growth dividend that can come from reforming such distortionary policies. Having now dismantled virtually all their import protection and agricultural subsidy policy distortions, and having undertaken major domestic macroeconomic and microeconomic reforms over more than two decades, the outcomes of those undertakings are 248 Distortions to Agricultural Incentives: A Global Perspective beginning to be reaped. The impact on overall living standards was mentioned at the outset, but an indicator within the agricultural sector is the acceleration it has given to farm productivity growth. It needs to be borne in mind that farmers in Australia and New Zealand have not been immune from the standard problem facing small farms that requires them to "get big or get out" as the economy develops. It is true that farm sizes in Australia and New Zealand were large relative to those in most other market economies in the early post­World War II years, and that the "wool boom" of the early 1950s provided massive incomes for woolgrowers. Nonetheless, as wages grew in other sectors of the two economies, the need to adjust was felt strongly.24 Adjust- ment manifested itself in the same way as in other countries--that is, with farmers funding agricultural research and adapting and adopting the new technologies it generated as appropriate, and with the number of farms and farmers declining steadily to lower the labor intensity of the sector even as output expanded.25 Within that context, the removal of the antiagricultural policy bias over the past 30 years has substantially boosted, albeit with a not-unexpected delay, the rate of growth of farm productivity in Australia and New Zealand. Figure 5.4 shows that in New Zealand, TFP growth slowed during the dozen or so years of high agricultural subsidies, and only increased to its previous rate after those sub- sidies, along with manufacturing protection, were removed. This pattern has been reflected in farmland prices: after initially halving when the subsidy cuts were announced in the early 1980s, they more than recovered in real terms by the turn of the century as farmers profitably adjusted to the new deregulated, level-playing- field domestic economic environment (figure 5.5). In Australia, farm multifactor productivity (MFP) growth increased following the international price hikes in 1973­74, but then plateaued during the next decade until the reforms from the mid-1980s began to have an effect (figure 5.4). In the 1983­93 period, farm MFP grew at just 1.4 percent per year, low in com- parison to the 4.1 percent growth rate seen in 1993­2000 (Productivity Commis- sion 2005). Similar results are reported in Parham (2004): less than 1.5 percent during 1974­88, then 2.6 percent in 1988­93, and 4.3 percent in 1993­98.26 His estimates show that even that earlier rate of 1.5 percent compares favorably with that for the other sectors of Australia's economy, the MFP of which was well below 1 percent during 1973­93 (and only 1.8 percent in 1993­98). Clearly, farmers are capable not only of surviving without subsidies, but of becoming more productive after their removal--and without a noticeably faster rate of decline in the total number of farmers or farms than occurs with normal economic growth.27 The cases of Australia and New Zealand, in fact, contradict a popular view in the economic growth literature that natural resource abundance (including a comparative advantage in agriculture) is a curse rather than a blessing.28 Australia and New Zealand 249 Figure 5.4. Real Agricultural Total/Multifactor Productivity Growth, Australia and New Zealand, 1927­2004 a. Australia 140 120 100 100 index, 2002­03 80 60 40 20 0 5 7 9 1 3 5 7 9 1 3 5 7 9 1 3 ­7 ­7 ­7 ­8 ­8 ­8 ­8 ­8 ­9 ­9 ­9 ­9 ­9 ­0 ­0 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 hours worked capital input output multifactor productivity growth b. New Zealand 1,000 900 TFP: 4.4% 800 high subsidies 100 700 600 index, 1927 500 TFP: 2.7% 400 TFP: 3.4% TFP: 3.6% 300 200 100 0 27 31 35 39 43 47 51 55 59 63 67 71 75 79 83 87 91 95 99 03 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 Sources: Productivity Commission (2005) and Lattimore (2006). Three important aspects of the agricultural reform success in Australia and New Zealand need to be underlined. One is that it helps if assistance to nonagri- cultural sectors is cut at the same time. In Australia and New Zealand, those other microeconomic and macroeconomic reforms made it easier for farmers to adjust and raise their productivity. This is relevant for the many developing countries that still have industrial protection and behind-the-border restrictions on domestic 250 Distortions to Agricultural Incentives: A Global Perspective Figure 5.5. Real Farmland Prices, New Zealand, 1978­2004 18 1999 $NZ per hectare, thousands 16 14 12 10 8 6 4 2 0 19 8 19 9 19 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 19 9 19 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 20 9 20 0 20 1 20 2 20 03 04 7 7 8 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 9 0 0 0 19 arable dairy sheep Source: Lattimore (2006). market activities. For other high-income countries, where manufacturing protec- tion rates are already low and the macroeconomy is well managed, the reduction of high agricultural supports would be more painful unless coupled with adjust- ment assistance measures. Both sets of countries may benefit by following Aus- tralia's lead in the adoption of rural research and development corporations to manage research investments funded by a levy on farmers matched dollar for dol- lar by a grant from the federal government. Introduced in 1989, this innovative funding model, together with a similar model for the generic promotion of Australia's farm products,29 arguably has contributed significantly to the increased international competitiveness of Australian agriculture (see CIE 2003; Productiv- ity Commission 2007). Second, adjustment to agricultural assistance cuts is easier when heterogeneity within the agricultural sector is greater. This can take at least two forms. Hetero- geneity in rates of industry assistance within the sector, as captured by the standard deviation of farm industry NRAs shown near the bottom of tables 5.2 and 5.3, ensures that as high rates of assistance are lowered, resources will find it profitable to move to lightly assisted farm industries. Such transfers are far easier than shifting mobile resources to other sectors, and involve much less reduction in the value of sector-specific assets, particularly farmland. For Australia, the standard deviation was very high when the cuts in farm assistance began in the early 1970s, whereas for a country such as Norway, the support levels are very similar across farm indus- tries, allowing much less scope for adjustment within the sector. Australia and New Zealand 251 The other form of heterogeneity has to do with firm efficiency within each industry. If some farmers are much more efficient than others, then a cut in assis- tance to agriculture typically leads to the less efficient being bought out by the more efficient (who often are also more innovative and export-oriented). The productivity effects of that tendency could be substantial (Melitz 2003; Baldwin 2005; and Long, Raff, and Stahler 2007). In the case of the Australian dairy indus- try, it indeed seems to have been powerful (Harris 2005a). Productivity effects may also help explain the rapid rise in exports relative to domestic sales for a number of Australia's agricultural industries since the 1980s, including beef, canola, wine, and cotton. Four other indicators of the increasing globalization of agriculture in these two countries, two of which also indicate growth in intrasec- toral trade, are: an increasing share of consumption being imported even though the self-sufficiency indicator is rising, and a slightly declining ratio of net exports to exports plus imports of agricultural products (Anderson et al. 2008). The third important aspect of the agricultural reform success in Australia and New Zealand is the scope and quality of public institutions. Both countries had strong institutional arrangements that facilitated the development of policy advice, provided policy coordination, and administered the policy changes. The legal, accounting, and media systems ensured relatively transparent and informed policy debate, and assisted in resolving a host of firm-level complications that arose dur- ing policy implementation. The lesson for countries wishing to replicate the reform experience of Australia and New Zealand, especially in a developing-country con- text, is to not underestimate the importance of strong, pro-market, transparent, corruption-free institutions. Looking ahead in Australia and New Zealand, mean- while, getting domestic policies right in the food safety, natural resource (especially water), and environmental areas, along with securing tariff and subsidy cuts abroad from the WTO's Doha Development Agenda, have the potential to yield even further productivity growth for farmers. But those are stories for the future. Notes 1. Two recent analyses by economic historians point to successful exploitation of natural resources as the key explanation of Australia's relatively high per capita income prior to World War I (Broadberry and Irwin 2007; McLean 2007). 2. However, there is now evidence that the rate of growth of productivity in Australia has since slipped behind that of several OECD countries (Dolman, Parham, and Zheng 2007). 3. New Zealand has around five times the global average of both agricultural land per capita and arable land per worker, and Australia has around 25 times as much (Sandri, Valenzuela, and Anderson 2006). Of course the quality of farmland and associated water, rainfall, sunshine, etc. also matter, but even adjusting for these leave Australia and New Zealand are relatively very well endowed in agricul- tural resources per worker. 4. In terms of population, Australia is somewhat smaller than Argentina and Canada but similar in terms of arable land, while New Zealand is similar in population, arable land, and other agricultural attributes to the Nordic countries. But the antipodean location of Australia and New Zealand 252 Distortions to Agricultural Incentives: A Global Perspective compared with those other countries leads one to expect them to have traded less (and be specialized in more storable and less bulky exports) than these comparator countries, at least prior to East Asia's trade-led growth takeoff. 5. Mining was also an important export earner for Australia in the latter half of the 19th century, but due almost entirely to gold. Gold's share of total exports, 49 percent in 1861, fell to about one-sixth in the 1980s, returning to 28 percent in 1900. During 1961­90, wool and gold together accounted for almost three-quarters of all exports (Butlin 1962). 6. The portion of what follows that is focused on Australia draws on Anderson, Lloyd, and MacLaren (2007). 7. It is thus a generalization of the nominal rate of border protection due, for example, to an import tariff, which is the percentage by which the domestic price is raised above the import unit value. 8. The first major tariffs for the Australian Federation were imposed in 1907. According to the indexes constructed by Carmody (1952), by the 1920s the decade average of the general tariff on Australia's imports of items other than food beverages and tobacco was double that 1907 level, and by the 1930s it averaged 60 percent higher than in the 1920s. The Vernon Committee (Committee of Enquiry 1965) reports averages for tariffs above 12.5 percent for the period from 1938 to 1963: though they dipped somewhat in the late 1940s/early 1950s when import licences became the binding constraint, by the early 1960s they were back to the level of the late 1930s. The annual average level of protection since World War II is indicated by carefully constructed customs duty rates (Anderson, Lloyd, and MacLaren 2007). 9. The GATT came into effect in 1948. Even though Australia and New Zealand were founding signatories to that agreement, they both chose not to join the commitments to cut manufacturing tariffs--out of frustration with the unwillingness of other GATT contracting parties to commit to lowering their agricultural protection rates (Arndt 1965; Snape 1984; and Capling 2001). 10. That pooling was inefficient in at least two senses: it led to excessive volumes of production because producers received the average rather than the marginal price; and because there was little differentiation in terms of quality and variety, producers were discouraged from seeking out niche markets by differentiating their products. Additional stabilization schemes were implemented by indi- vidual states, such as for fresh milk and eggs, and these led to different incentives in the various states. These were possible because the states agreed not to trade across state borders, in contravention of Section 92 of the Constitution, which says there shall be no barriers to interstate trade. 11. This rate is still higher than that for other OECD countries (World Bank 2006). And WTO- bound tariffs average more than twice the applied rates. However, Australia uses nontariff import barriers less frequently than other OECD countries, apart perhaps from antidumping duties (Productivity Commission 2000a, 2000b, 2004). 12. Tariffs on motor vehicle imports fell from 40 to 15 percent over the 1990s and were cut again to 10 percent in 2005. For clothing, the decline over the 1990s was from 55 to 25 percent, and for footwear from 45 to 15 percent, with cuts to 17.5 and 10 percent in 2005, respectively. Further cuts, to as little as 5 percent, are scheduled for 2010 (Productivity Commission 2000a). 13. By contrast, the reductions in manufacturing protection were more systematic: the 1973 across-the-board tariff cut, a tariff review program begun in 1971 by the Tariff Board and subse- quently conducted by its successors (the Industries Assistance Commission, Industry Commission, and Productivity Commission), and the preannounced phased reductions in tariffs on textiles, cloth- ing, footwear, and motor vehicles and parts from 1988. 14. This stabilization scheme operated conservatively for 15 years until the government transferred the power to set the reserve price to growers in 1987. Growers promptly raised that reserve price, which operated on the world market, by 71 percent. Predictably, this encouraged growers to expand wool production and international buyers to reduce purchases (since the Australian Wool Corporation could then stockpile wool and thereby save the buyer the cost of storage). The scheme collapsed in 1991, following which the Australian Wool Corporation had to dispose of 4.75 million bales, at expense to the government and, even more, to woolgrowers (Richardson 2001). 15. Effective assistance to the mining sector is still slightly negative, as it was two decades ago (Industry Commission 1992; Productivity Commission 2004), although that will be less so when the government eases the current quantitative restrictions on exports of uranium and its derivatives. Australia and New Zealand 253 16. In addition, a comprehensive program of review of government regulations at all levels in Australia has been under way since the mid-1990s, with the aim of reducing/removing regulations that unjustifiably impede economic activities (Productivity Commission 2000b). For an early assessment of Australia's domestic microeconomic reforms, see Forsyth (1992, 2000). All Productivity Commis- sion reports on the myriad reforms are downloadable at http://www.pc.gov.au. Recent research on barriers to trade in a wide range of services in almost 40 countries found that services markets in Australia, relative to those in the other countries in the study, are now ranked as either very liberal (banking, distribution services, telecoms, engineering, and professional services) or moderately restrictive (other professional services and maritime services)--see Productivity Commission (2000c). 17. Even as recently as the late 1980s, there were some 10 Commonwealth and more than 50 state SMAs (Piggott 1990). 18. Agricultural SMAs were exempt from the regulation of competition under the Trade Practices Act 1974, although later on they were subject to assessment under the National Competition Policy. 19. In the course of its investigations, the Commission located more than 100 marketing arrange- ments at the state and territory level, though not all of which were SMAs. 20. The evidence suggests that the retail price of milk fell after deregulation as the supermarkets were able to exercise buying power with the milk processors, which, in turn, were able to exert down- ward pressure on the farmgate price that was no longer supported by state governments. 21. Unlike Australia, however, New Zealand continued to receive some nonreciprocal preferential access to the EEC market. 22. The impact of past underpricing of water for agriculture on farm returns has not been incor- porated in the NRA estimates reported in this paper. 23. For detailed descriptions of Australia's adjustment and adjustment assistance programs, see Harris (2005b). 24. In the first half of the 1950s, farm incomes averaged more than 20 percent above nonfarm self- employed incomes and more than twice those of male wage and salary earners. In the next dozen years, though, farm incomes fell to 5 percent below nonfarm self-employed incomes and to just 30 percent above those of male wage and salary earners (McKay 1967). The gap between farm and nonfarm incomes and spending power grew especially rapidly in between 1963 and 1973 (Glau 1971). 25. Australia's dairy industry saw several changes over the 25 years to 2004: the number of dairy farmers more than halved (from 22,000 to 10,000), the average herd size increased 2.5 times (from 85 to 210) average annual yield per cow nearly doubled (from 2,850 to 4,900 liters) and average annual milk production per farm quadrupled, from 0.25 to 1.05 million liters (Productivity Commission 2005; Harris 2005a). In addition, the dairy processing industry has become significantly more produc- tive (Balcombe, Doucouliagos, and Fraser 2007). Overall, the number of farmers in Australia has fallen by more than one-third over the past half-century (ABARE 2007). 26. See also the results since the mid-1970s in Fleming (2007), who shows that MFP growth in Australia's farms enabled producers to cope with the fact that, over the past three decades, the prices they paid for their inputs grew 1.6 percent more per year than the prices they received for their products. 27. Adjustment has been sharper within individual industries, of course, especially those that faced dramatic cuts in subsidies such as Australia's dairy industry (see previous footnote). 28. On this literature in general and its applicability to Latin America, see Lederman and Maloney (2006). See also Anderson (1998). 29. On the basic economics of these supply and demand shifting investments in the presence of spillovers, see, for example, Levin and Reiss (1988). References ABARE (Australian Bureau of Agricultural and Resource Economics). 2007. Australian Commodities Statistics 2006. http://www.abare.gov.au. Anderson, J. E. 1995. "Tariff Index Theory." Review of International Economics 3 (2): 156­73. 254 Distortions to Agricultural Incentives: A Global Perspective Anderson, K. 1978. "On Why Rates of Assistance Differ between Australia's Rural Industries." Australian Journal of Agricultural Economics 22 (2): 99­114. Anderson, K. 1998. "Are Resource-Abundant Economies Disadvantaged?" Australian Journal of Agricultural and Resource Economics 42 (1): 1­23. Anderson, K., and R. Garnaut. 1980. "ASEAN Export Specialisation and the Evolution of Comparative Advantage in the Western Pacific Region." In ASEAN in a Changing Pacific and World Economy, ed. R. Garnaut, 374­412. Canberra: ANU Press. ______. 1987. Australian Protectionism: Extent, Causes and Effects. Boston, London, and Sydney: Allen and Unwin. Anderson, K., and L. A. Jackson. 2005. "GM Crop Technology and Trade Restraints: Economic Impli- cations for Australia and New Zealand." Australian Journal of Agricultural and Resource Economics 49 (3): 263­81. Anderson, K., R. Lattimore, P. Lloyd, and D. MacLaren. 2008. "Distortions to Agricultural Incentives in Australia and New Zealand." Agricultural Distortions Working Paper 09, World Bank, Washington, DC. Anderson, K., P. J. Lloyd, and D. MacLaren. 2007. "Distortions to Agricultural Incentives in Australia Since World War II." The Economic Record 83 (263): 461­82. Anderson, K., W. Martin, and D. van der Mensbrugghe. 2006. "Distortions to World Trade: Impacts on Agricultural Markets and Incomes." Review of Agricultural Economics 28 (2): 168­94. Anderson, K., and B. Smith. 1981. "Changing Economic Relations between Asian ADCs and Resource- Exporting Developed Countries." In Trade and Growth in the Advanced Developing Countries, ed. W. Hong and L. Krause, 293­338. Seoul: Korea Development Institute Press. Anderson, K., and E. Valenzuela. 2007. "Do Global Trade Distortions Still Harm Developing Country Farmers?" Review of World Economics/Weltwirtschaftliches Archiv 143 (1): 108­39. ______. 2008. Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007. Core database at http://www.worldbank.org/agdistortions. Arndt, H. W. 1965. "Australia ­ Developed, Developing or Midway?" The Economic Record 41 (95): 418­40. Balcombe, K., H. Doucouliagos, and I. Fraser. 2007. "Input Use, Output Mix and Industry Deregulation: An Analysis of the Australian Dairy Manufacturing Industry." Australian Journal of Agricultural and Resource Economics 51 (2): 121­35. Balderstone, J. S., et al. 1982. "Agricultural Policy: Issues and Options for the 1980s." Working Group Report to the Minister for Primary Industry, Australian Government Publishing Service, Canberra. Baldwin, R. E. 2005. "Heterogeneous Firms and Trade: Testable and Untestable Properties of the Melitz Model." NBER Working Paper 11471, National Bureau of Economic Research, Cambridge, MA. Broadberry, S., and D. A. Irwin. 2007. "Lost Exceptionalism? Comparative Income and Productivity in Australia and the UK, 1861­1948." The Economic Record 83 (262): 262­86. Butlin, N. G. 1962. Australian Domestic Product, Investment and Foreign Borrowing, 1861­1938/39. Cambridge, U.K.: Cambridge University Press. Capling, A. 2001. Australia and the Global Trade System: From Havana to Seattle. Cambridge, U.K.: Cambridge University Press. Card, D., and R. B. Freeman. 2004. "What Have Two Decades of British Economic Reform Delivered?" In Seeking a Premier Economy: The Economic Effects of British Economic Reforms, 1980­2000, ed. D. Card, R. Blundell, and R. B. Freeman, 9­20. Chicago: University of Chicago Press. Carmody, A. T. 1952. "The Level of the Australian Tariff: A Study in Method," Yorkshire Bulletin of Economic and Social Research 4: 51­65, January. CIE (Centre for International Economics). 2003. The Rural Research and Development Corporations: A Case Study of Innovation, Canberra: Centre for International Economics. Committee of Enquiry (The Vernon Committee). 1965. Report of the Committee of Enquiry. Melbourne: Wilke and Co. Corden, W. M. 1984. "Booming Sector and Dutch Disease Economics: Survey and Consolidation." Oxford Economic Papers 36 (3): 359­80. Australia and New Zealand 255 Dalziel, P., and R. Lattimore. 2004. The New Zealand Macroeconomy: Striving for Sustainable Growth with Equity. Melbourne: Oxford University Press. Deardorff, A. V. 1984. "An Exposition and Exploration of Krueger's Trade Model." Canadian Journal of Economics 5 (4): 731­46. Diewert, W. E., and D. Lawrence. 2006. "Measuring the Contributions of Productivity and Terms of Trade to Australia's Economic Welfare." Consultancy Report, Productivity Commission, Canberra. Dolman, B., D. Parham, and S. Zheng. 2007. "Can Australia Match US Productivity Performance?" Staff Working Paper, Productivity Commission, Canberra. Dowrick, S. 2001. "The Australian Productivity Miracle." In Creating an Environment for Australia's Growth, ed. J. Nieuwenhuysen, P. J. Lloyd and M. Mead, 33­49. Cambridge, U.K. and Sydney: Cambridge University Press. Edwards, G. W. 1987. "Agricultural Policy Debate: A Survey." The Economic Record 63: 129­43. ______. 2006. "Fifty Years of Agricultural Policy." Presented at the 50th Annual Conference of the Australian of the Australian Agriculture and Resource Economics Society, Sydney, 8­10 February. Evans, L., A. Grimes, B. Wilkinson, and D. Teece. 1996. "Economic Reform in New Zealand: The Pursuit of Efficiency." Journal of Economic Literature 34 (4): 1856­1902. Fleming, E. 2007. "Use of the Single Factorial Terms of Trade to Analyse Agricultural Production." Australian Journal of Agricultural and Resource Economics 51 (2): 113­19. FSANZ (Food Standards Australia New Zealand). 2007. "Assessing the Safety of New Foods and Technologies." FSANZ, http://www.foodstandards.gov.au/foodmatters/gmfoods/ assessingthesafetyof387.cfm. Forsyth, P., ed. 1992. Microeconomic Reform in Australia. London and Sydney: Allen and Unwin. ______. 2000. "Microeconomic Policies and Structural Change." In The Australian Economy in the 1990s, ed. D. Gruen and S. Shrestha. Sydney: Reserve Bank of Australia. Gardner, B. 2009. "The United States and Canada." Chapter 4 in this volume. Glau, T. E. 1971. "The Cost­Price Squeeze on Australia Farm Income." Australian Journal of Agricultural Economics 15 (1): 1­19. Godden, D. 1997. Agricultural and Resource Policy: Principles and Practice. Melbourne: Oxford University Press. Harris, D. 2005a. Industry Adjustment to Policy Reform: A Case Study of the Australian Dairy Industry, RIRDC Publication 05/110, Rural Industries Research and Development Corporation, Canberra. ______. 2005b. Rural Industry Adjustment to Trade Related Policy Reform, RIRDC Publication 05/173, Rural Industries Research and Development Corporation, Canberra. Harris, S. F., J. C. Crawford, F. H. Gruen, and N. Honan, 1974. The Principles of Rural Policy in Australia: A Discussion Paper, Canberra: Australian Government Publishing Service. Honma, M., and Y. Hayami. 2009. "Japan, Republic of Korea and Taiwan, China." Chapter 2 in this volume. Industry Commission. 1991. "Statutory Marketing Arrangements for Primary Products." Report 10, AGPS, Canberra. Industry Commission. 1992. Annual Report. Canberra: AGPS. James, S., and K. Anderson. 1998. "On the Need for More Economic Assessment of Quarantine/SPS Policies." Australian Journal of Agricultural and Resource Economics 42 (4): 525­44. Josling, T. 2009. "Western Europe." Chapter 3 in this volume. Krueger, A. O. 1977. "Growth, Distortions and Patterns of Trade Among Many Countries." Princeton Studies in International Finance Section 40: 1­45. Lattimore, R. 2006. "Farm Policy Reform Dividends." Paper presented to the North American Agrifood Market Integration Consortium Meetings, Calgary, 1­2 June. Leamer, E. E. 1987. "Paths of Development in the Three-Factor, n-Good General Equilibrium Model." Journal of Political Economy 95 (5): 961­99. Lederman, D., and W. Maloney, eds. 2006. Natural Resources: Neither Curse nor Destiny. Stanford, CA: Stanford University Press for the World Bank. Levin, R. C., and P. C. Reiss. 1988. "Cost-Reducing and Demand-Creating R&D With Spillovers." Rand Journal of Economics 19 (4): 538­56. 256 Distortions to Agricultural Incentives: A Global Perspective Lloyd, P. 2008. "100 Years of Tariff Protection in Australia." Australian Economic History Review 48 (2): 99­145. Long, N. V., H. Raff, and F. Stahler. 2007. "The Effects of Trade Liberalization on Productivity and Welfare: The Role of Firm Heterogeniety, R&D and Market Structure." Economics Discussion Paper 0710, University of Otago, Dunedin. MacLaren, D. 1992. "The Political Economy of Agricultural Policy Reform in the European Commu- nity and Australia." Journal of Agricultural Economics 43 (3): 424­39. Maddison, A. 2003. The World Economy: Historical Statistics. Paris: OECD Development Centre. Martin, W. 1990. "Public Choice Theory and Australian Agricultural Policy Reform." Australian Journal of Agricultural Economics 34 (2): 189­211. Mauldon, R. 1990. "Price Policy." In Agriculture in the Australian Economy, ed. D. B. Williams, 310­28. Sydney: Sydney University Press. McKay, D. H. 1967. "The Small-Farm Problem in Australia." Australian Journal of Agricultural Economics 11 (2): 33­52. McLean, I. 2007. "Why Was Australia So Rich?" Explorations in Economic History 44 (4): 635­56. Melitz, M. 2003. "The Impact of Trade on Intra-industry Reallocations and Aggregate Industry Productivity." Econometrica 71 (6): 1695­725. OECD. 2008. OECD PSE-CSE Database (Producer and Consumer Support Estimates, OECD Database 1986­2007). Organisation for Economic Co-operation and Development. http://www.oecd.org/ document/59/0,3343,en_2649_33727_39551355_1_1_1_1,00.html. Parham, D. 2004. "Sources of Australia's Productivity Revival." The Economic Record 80 (249): 239­57. Parham, D., T. Cobbold, R. Dolamore, and P. Roberts. 1999. Microeconomic Reforms and Australian Productivity: Exploring the Links. Canberra: Productivity Commission. Piggott, R. 1990. "Agricultural Marketing." In Agriculture in the Australian Economy, ed. D. B. Williams, 287­309. Sydney: Sydney University Press, 1990. Productivity Commission. 2000a. Review of Australia's General Tariff Arrangements. Canberra: Produc- tivity Commission. ______. 2000b. Regulation and its Review 1999­00. Canberra: Productivity Commission. ______. 2000c. Trade and Assistance Review 1999­2000. Canberra: Productivity Commission. ______. 2004. Trade and Assistance Review 2003­04. Canberra: Productivity Commission. ______. 2005. "Trends in Australian Agriculture." Research Paper, Productivity Commission, Canberra. ______. 2006. Rural Water Use and the Environment: The Role of Market Mechanisms. Canberra: Productivity Commission. ______. 2007. "Public Support for Science and Innovation." Research Report, Canberra. Rayner, A. and R. Lattimore. 1991. "New Zealand." In Liberalising Foreign Trade, ed. D. Papageorgiou, M. Michaely, and A. Choksi. Cambridge, MA: Basil Blackwell. Richardson, B. 2001. "The Politics and Economics of Wool Marketing, 1950­2000," Australian Journal of Agricultural and Resource Economics 45 (1): 95­115. Sandri, D., E. Valenzuela, and K. Anderson. 2006. "Compendium of National Economic and Trade Indicators by Region, 1960 to 2004." Agricultural Distortions Working Paper 01, World Bank, Washington, DC. http://www.worldbank.org/agdistortions Sandry, R. and R. Reynolds, eds. 1990. Farming without Subsidies: New Zealand's Recent Experience. Wellington: GP Books. Sieper, E. 1982. Rationalising Rustic Regulation. Sydney: Centre for Independent Studies. Statistics New Zealand. 2006. Productivity Statistics 1988­2005. Wellington: Statistics New Zealand. Snape, R. H. 1984. "Australia's Relations with GATT." The Economic Record 60 (168): 16­27. Vitalis, V. 2005. "The New Zealand Experience." In Subsidy Reform and Sustainable Development: Economic, Environmental and Social Aspects, ed. Lori Ridgeway and others. Paris: OECD. Vousden, N. 1990. The Economics of Trade Protection. Cambridge, U.K. and New York: Cambridge University Press. World Bank. 2006. World Development Indicators. Washington, DC: World Bank. Part III EVOLUTION OF DISTORTIONS IN EMERGING ECONOMIES 6 Eastern Europe and Central Asia Kym Anderson and Johan Swinnen* In a recent survey of European economic growth since 1950, Crafts and Toniolo (2008) conclude that incentive structures have been a crucial influence on compara- tive growth rates of the economies of Eastern and Western Europe. Predating that, a 2006 report on trade performance and policies in Eastern Europe and Central Asia included as one of its key recommendations the need to reduce the mean and vari- ance of the tariff equivalents of trade barriers, in particular the need to reduce uni- laterally governments' antiexport bias, especially in countries exporting primary products (Broadman 2006). To carry out such reform in Europe's transition economies efficiently and effectively--and to see how recent policies align with those of the European Union (EU)--requires better information on the extent of reform during the past two decades and of current policy influences on incentives within and between sectors. Immediately prior to their transition to market economies, policies in the region greatly distorted producer and consumer incen- tives, especially for agricultural products. Those distortions have been reduced substantially in several countries, but large variations remain across the region and distortions appear to be increasing again in some countries. Now is thus an oppor- tune time to examine how policies affecting agriculture are evolving in the region, particularly as part of the adjustment to EU accession for 10 of the countries. *The authors are grateful for data assistance and very helpful discussions with Organisation for Economic Co- operation and Development (OECD) Secretariat staff, especially Olga Melyukhina; for the distortions estimates provided by authors of the focus country case studies; and for assistance with spreadsheets by Johanna Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, Damiano Sandri, and Ernesto Valenzuela. The working paper version of this chapter (Anderson and Swinnen 2009) contains additional background material and appendix tables. This chapter draws on the introductory and country chapters in Anderson and Swinnen (2008), with data updated using Anderson and Valenzuela (2008). 259 260 Distortions to Agricultural Incentives: A Global Perspective To assist that process, the study presented here assesses the changing landscape of agricultural protection or taxation patterns in Europe and Central Asia. It is based on a sample of 18 countries, including 11 Central and Eastern European (CEE) countries--the 10 new EU members (Bulgaria and Romania joined in January 2007 and the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, the Slovak Republic, and Slovenia joined in May 2004), plus Turkey--and seven Commonwealth of Independent States (CIS) countries (Kazakhstan, the Kyrgyz Republic, the Russian Federation, Tajikistan, Turkmenistan, Ukraine, and Uzbekistan). Together, these 18 countries accounted in 2000­04 for 89 percent of the region's agricultural value added, 91 percent of its population, and 95 percent of total gross domestic product (GDP). Some key characteristics of those economies are shown in table 6.1, drawn from the detailed compendium of indi- cators provided in the Sandri, Valenzuela, and Anderson (2007). Analyses of polit- ically feasible agricultural subsidy and trade policy reforms, and of policy options for coping with structural changes such as the recent boom in energy raw material prices that has intersectoral Dutch disease effects, need to be based on a clear understanding of the recent and current extent of policy interventions and the politico-economic forces behind their evolution. This study thus also seeks to bet- ter understand the political economy of distortions to agricultural incentives in Europe and Central Asia. With that better understanding, the study's third purpose is to explore prospects for further reducing distortions to agricultural incentives and their implications for agricultural competitiveness and trade of the different countries in the region, including those that have recently joined the EU. The great diversity within the group of European and Central Asian countries-- in terms of relative resource endowments and comparative advantages; stages of development and transition; agricultural and trade policy regimes; and member- ships of the EU, World Trade Organization (WTO), OECD, and regional trading agreements--make the set of countries chosen a rich sample for comparative study. Turkey and the CEE countries that are now EU members differ substantially from the rest of the former Union of Soviet Socialist Republics (USSR) that are now members of the CIS in that they have higher per capita income (three-quar- ters of the global average, compared with one-third for the CIS) and higher popu- lation density (half the global average land per worker and 70 percent of the global average agricultural land per capita, compared with 3.4 and 2.5 times, respectively, for the CIS). Growth and Structural Changes during Transition Before examining policy changes, it is helpful to review the economic growth and intersectoral changes that have taken place in Europe's various transition economies over the past two decades. The initial years of transition from central Table 6.1. Key Economic and Trade Indicators, Eastern European and CIS Countries, 2000­04 National relative to Percent of world: world (world 100) Agricultural RCA,a Trade Total Agricultural GDP land agriculture specialization Gini Population GDP GDP per capita per capita and food indexb Povertyc indexd Slovenia 0.03 0.07 0.04 216 32 52 0.68 0 -- Czech Republic 0.16 0.22 0.19 135 52 61 0.44 0 26 Hungary 0.16 0.20 0.14 122 72 90 0.40 0 27 Estonia 0.02 0.02 0.03 102 78 199 0.38 1 36 Poland 0.62 0.57 0.47 93 57 105 0.39 0 34 Slovak Republic 0.09 0.07 0.09 92 57 57 0.50 0 -- Lithuania 0.06 0.04 0.08 80 125 176 0.21 1 36 Latvia 0.04 0.03 0.03 76 132 364 0.51 0 38 Turkey 1.12 0.62 1.97 55 70 131 0.09 3 44 Romania 0.35 0.15 0.49 41 84 74 0.06 1 31 Bulgaria 0.13 0.05 0.15 39 86 143 0.37 0 29 CEE sample 2.75 2.05 3.67 74 70 98 0.09 1 37 Russian Federation 2.34 1.10 1.58 47 186 53 0.46 0 40 Kazakhstan 0.24 0.08 0.18 33 1737 76 -- 1 34 Ukraine 0.78 0.13 0.46 17 107 112 -- 0 28 Turkmenistan 0.07 0.01 0.06 18 881 92 -- 5 41 Uzbekistan 0.41 0.03 0.27 8 134 -- -- 0 37 Kyrgyz Republic 0.08 0.00 0.05 6 268 390 -- 0 30 Tajikistan 0.10 0.00 0.03 4 85 192 -- 7 33 CIS sample 4.02 1.37 2.62 34 270 -- 0.02 0 37 Other CEE/Central Asia 0.64 0.19 0.61 29 82 166 0.41 1 -- All CEE/Central Asia 7.43 3.60 6.90 48 179 -- 0.06 0 37 Source: Sandri, Valenzuela, and Anderson (2007), compiled from World Bank (2007). Note: -- not available. 261 a. RCA is the index of revealed comparative advantage share of agriculture and processed food in national exports as a ratio of that sector's share of global exports b. Primary agriculture trade specialization index (X M) (X M), 2000­02 (world average 0). c. Percentage of population living on US$1/day, from Chen and Ravallion (2007). d. Gini index for the most recent year during 2000­04, from Chen and Ravallion (2007). 262 Distortions to Agricultural Incentives: A Global Perspective planning to a more market-based economy saw production fall in the majority of sectors, before it recovered at varying rates from the mid-1990s. Real GDP for the Eastern Europe and Central Asia region as a whole fell by almost 6 percent per year during 1990­94. The decline for the CEE sample was only 0.6 percent, while for the CIS sample it was 11 percent and for the residual nonstudied countries of the CIS 12 percent. By contrast, annual GDP growth over 1995­2004 averaged 2.7 percent: the CIS sample was slowest (2.2 percent), the CEE sample somewhat higher at 3.2 percent, and the residual enjoyed 5.1 percent. Within those economies, agricultural value added measured at constant prices appears to have declined less rapidly than nonagricultural GDP in the early years of transition, but also to have grown less rapidly in the subsequent decade. The domestic terms of trade (the prices of their outputs relative to the prices of pur- chased inputs) apparently fell even more for farmers than for nonfarmers, how- ever, because agriculture's share of GDP measured in current prices declined even in the early transition period. Unlike in the central planning period, this did not allow faster industrialization but rather an expansion in the services sector, which increased from less than half the economy prior to 1993 to two-thirds by 2004. The halving of agriculture's share of GDP in Europe and Central Asia between 1992 and 2004 was accompanied by only a one-quarter decline in agriculture's share of employment, according to Food and Agriculture Organization (FAO) sta- tistics, which are not always consistent with national data because of definitional differences. In all three subgroups of countries, the share of the latter averaged three times the former by 2004, or five times in the case of the CEE-8 countries that joined the EU in 2004. This suggests much lower labor productivity on farms than in other employment. The share of farm and food products in total merchandise exports also has fallen, by as much as half in some countries in the region. When expressed as a ratio of that share for the world as a whole (an agricultural-revealed comparative advantage index), it becomes evident that most countries of the region have lost comparative advantage in farm products over the transition period. That index varies greatly across the region though, from a low of less than 0.5 for mineral-rich Russia and densely populated Slovenia to more than 3 for Latvia and the Kyrgyz Republic. The region as a whole, however, has become more open as a consequence of moving from a planned to a market economy, notwithstanding the continuation of numerous barriers to trade. A common indicator is the value of goods and services expressed as a percentage of GDP. For most countries, that percentage is now above the average for Western European countries (37 percent in 2004), with several countries approaching 60 percent by 2004.1 With this as background, the evolution of policy under Communism is briefly reviewed, followed by an examination of how sectoral and trade policies have Eastern Europe and Central Asia 263 changed in Europe and Central Asia in response to, or as contributors to, the macroeconomic and structural changes. Quantifying the Distortions to Agricultural Incentives The main focus of the present study's methodology is government-imposed dis- tortions that create a gap between domestic prices and what they would be under free markets. Since it is not possible to understand the characteristics of agricul- tural development with a sectoral view alone, the project's methodology not only estimates the effects of direct agricultural policy measures (including distortions in the foreign exchange market), but also generates estimates of distortions in nonagricultural sectors for comparative evaluation. Specifically, nominal rates of assistance (NRAs) for farmers are computed, including any input subsidies and non-product-specific (NPS) forms of assistance or taxation. It also generates a production-weighted average NRA for nonagricultural tradables, for comparison with that for agricultural tradables via the calculation of a relative rate of assis- tance (RRA; see Anderson et al. 2008a, 2008b). This approach is not well suited to analysis of policies of planned economies prior to their reform era, as prices then played only an accounting function and currency exchange rates were enormously distorted. During their reform era, however, starting in 1992, the price compari- son approach provides as valuable a set of indicators for European and Central Asian countries as for other market economies of distortions to incentives for farm production, consumption, and trade, and of the income transfers associated with interventions. While most of the focus is on agricultural producers, the extent to which consumers are taxed or subsidized is also considered. To do so, a consumer tax equivalent (CTE) is calculated by comparing the price consumers pay for their food and the international price of each food product at the border. Differences between the NRA and the CTE arise from distortions in the domestic economy that are caused by transfer policies and taxes or subsidies that cause the prices paid by consumers (adjusted to the farmgate level) to differ from those received by producers. To obtain dollar values of farmer assistance and consumer taxation, estimates of NRAs are multiplied by the gross value of production at undistorted prices to obtain an estimate in current U.S. dollars of the direct gross subsidy equivalent (GSE) of assistance to farmers. These GSE values are calculated in constant dollars, and are also expressed on a per-farm-worker basis. They (and their equivalent on the consumption side) can be added up across products for a country, and across countries for any or all products, to get regional aggregate transfer estimates for the studied economies. 264 Distortions to Agricultural Incentives: A Global Perspective To keep the task manageable, the sample of countries for which empirical esti- mates are provided below is limited to the 10 Central and Eastern European coun- tries that joined the EU in 2004 or 2007 plus Turkey and the three biggest CIS economies (Kazakhstan, Russia, and Ukraine).2 Nonquantitative policy assess- ments are undertaken for the other economies of Central Asia. Reliable price data are available only from 1992 to 2005 or 2007 or, in the case of Kazakhstan, for just 2000­04. The worst of the exchange rate distortions in the formerly planned economies were removed in the early 1990s, prior to the start of the period under study here. Since there were no reliable indicators of any remaining secondary market price for foreign currency, this study follows the practice of the OECD Secretariat in using official exchange rates in making price comparisons.3 Additionally, it is not possible to estimate the sectoral assistance equivalent of soft credits provided for some farms (and other enterprises). Throughout, there is no attempt to assess whether some interventions may have been warranted on national economic wel- fare grounds because of the presence of externalities, or failures in the markets for such things as land and credit, or policy failures such as underinvestment in public infrastructure in rural areas.4 The Communist era Incentives for agricultural producers and food consumers were massively dis- torted under Communist central planning, which was imposed from the 1920s until 1991 in the former Soviet Union and from the 1950s until the fall of the Berlin Wall in 1989 in Central and Eastern Europe. Distortions included collective farm property rights; centrally controlled organization of production allocation, processing, input provision, and marketing; price setting unrelated to demand- supply conditions (leading to rationing); and state-controlled trading and exchange rate systems. Land and farms were put under central planning and, in most countries (with the exception of Poland and former Yugoslavia), farming was forcibly organized in collective and state farms. In the former Soviet Union, this collectivization process and the associated forced migration (and worse) of many landowners and farmers contributed to massive incidence of hunger and death before World War II. From Lenin to Stalin and through most of the Khrushchev regime, agriculture was heavily taxed, and capital was drained from an impoverished countryside to finance urban industrial growth (Ellman 1988).5 All of this changed at the end of the Khrushchev regime, particularly under Brezhnev, when the leadership of the USSR decided to increase agricultural pro- duction, with a strong emphasis on livestock, a policy also followed by many of the Eastern European countries of the Soviet bloc (Liefert and Swinnen 2002). From Eastern Europe and Central Asia 265 the mid-1950s onward, and especially in the 1970s and 1980s, large amounts of support and investment were directed to agriculture. By 1980, almost 30 percent of total Soviet investment was going into agriculture (Gray 1990). At the same time, consumer prices were set low and producer prices high, with the gap covered by direct subsidies to processing and trading companies or by soft budget con- straints. Consequently, from 1970 to 1990, livestock herds and output in these countries grew by between 40 and 60 percent. The rise in feed requirements for the growing herds stimulated the crop sector. In the late 1980s, the average annual output of feed grain in Hungary and Poland increased by half and one-quarter, respectively, compared with output in the late 1960s. In the USSR, livestock feed requirements were so high that the country also became a substantial importer of feed commodities. By 1990, per capita consumption of livestock products and foodstuffs in gen- eral compared favorably with many OECD countries, even though per capita incomes in Central and Eastern Europe were much lower than the OECD average. This "achievement" came at a cost: large state subsidies, to both producers and consumers, were necessary to maintain the high levels of production and con- sumption. For example, by the end of the 1980s, direct budgetary subsidies to the agriculture and food economy were about 10 percent of GDP in the USSR and between 5 and 10 percent of GDP in most CEE countries. The bulk of these subsi- dies went to the livestock sector.6 Calculating the net transfers to farmers and to consumers under the Commu- nist regime is very difficult because of the large number of distortions caused by the state regulation of prices, production, consumption, exchange rates, and mar- keting organizations, along with the indirect nature of some of the subsidies. While it is generally true that producers of farm products were strongly subsidized by price setting toward the end of the Communist regime (in sharp contrast to the 1930s, when farmers were highly discriminated against), the complexity of the distortions led sometimes to offsetting effects. For example, while agricultural producers in the latter 1980s were supported by high output prices and low input prices, at the same time overvalued exchange rates effectively taxed agricultural (and other) exporters. Correcting for this overvaluation leads to significantly lower protection indicators. In addition, agriculture was not alone in being subsi- dized, as most (heavy) industry was also subsidized or at least protected from import competition. Available fragments of empirical evidence indicate that, in aggregate and real terms, there was substantial net subsidization of agriculture rel- ative to all other sectors as a group, although much more so for livestock produc- ers than for grain and oilseed farmers. This might suggest food consumers were taxed substantially, but under the central planning system wholesalers were told to sell their food to retailers below their production costs, for which they received 266 Distortions to Agricultural Incentives: A Global Perspective state subsidies. Overvalued exchange rates, which effectively taxed exports and subsidized imports, also lowered domestic consumer prices of tradable products. However, by restricting foreign imports and regulating trade, the Communist regime prevented its consumers from accessing higher-quality food products. Huffman and Johnson (2004) estimate that these welfare losses were equivalent to 50 percent to 75 percent of the direct subsidy benefits of consumers under the Communist regime. The reform era After 1989, the CEE-8 countries (those Central and Eastern European countries that joined the EU in 2004: the Czech Republic, Estonia, Hungary, Latvia, Lithua- nia, Poland, the Slovak Republic, and Slovenia) moved first and most rapidly toward market-based systems. Reforms in the Balkan countries, such as Romania and Bulgaria, were initially half-hearted and involved many inconsistencies during most of the 1990s, with government interventions continuing to heavily distort incentives. In the large CIS countries (Kazakhstan, Russia, and Ukraine), govern- ments continued important controls of the agricultural economy through inter- ventions such as regional trade controls, input supply controls, and the continuation of soft budget constraints. While the Kyrgyz Republic liberalized relatively quickly, other Central Asian countries moved more slowly, and some have undertaken far less reform and liberalization. Major controls are still in place in countries such as Turkmenistan and Uzbekistan. International trade was strongly regulated under the centrally planned system. Communist countries were integrated in the Council of Mutual Economic Assistance (CMEA) system, a planned, intercountry trading regime that traded mainly with other Communist countries. (One could think of the CMEA as the international version of a domestic centrally planned economy.) The CEE countries were less integrated than the Soviet republics, but still a large part of their trade volume went through the CMEA system. When the CMEA system collapsed in the early 1990s with the liberalization of the macroeconomy and of trade policies, impor- tant changes in trade and financial flows resulted. Trade liberalization reinforced the reallocation of production activities caused by the abolishment of central planning. Traditional international production allocations were no longer possi- ble when trade had to be financed by hard currencies and when inputs were accounted for at real costs. It also allowed the importation of high-quality Western agricultural products that previously had been restricted. At the same time, the liberalization of exchange rates removed discrimination against the sectors pro- ducing tradables. Trade liberalization led to a major international reorganization of production activities. Initially this had a very negative impact on the region's producers, as the Eastern Europe and Central Asia 267 traditional export markets dwindled due to a lack of hard currency and because Western countries remained closed to the region's agricultural exports. At the same time, the reduction of import constraints opened regional markets to imports from the West. Combined, these forces caused a worsening of the region's agricultural trade balance in the first half of the 1990s. Later on, however, agri- food trade intensified and growing exports (also to Western markets) contributed to the region's recovery. An important development was the shift from centrally imposed extreme specialization (for example, dairy production in the Baltics and cotton production in Central Asia) to more diversified production systems and less dependence on single commodities in those countries. Trade effects were only part of the international effects in the agri-food sys- tems. Possibly even more important was the massive inflow of foreign direct investment to food processing industries, which contributed to a major restruc- turing and to improvements in food quality and productivity enhancements and investments in agriculture (Dries and Swinnen 2004). Most recently, the wave of foreign investments in the retail sector caused further restructurings of the agri- food system, with important implications for both producers and consumers (Dries, Reardon, and Swinnen 2004). Progress in market reforms is not always correlated with the extent of distor- tions. On the one hand, Slovenia, which was a front runner in liberalization and developing a market economy, has a very high level of farm producer support that in 2004 was well above the average EU-15 rate. On the other hand, much slower reformers such as Bulgaria, Kazakhstan, and Ukraine have much lower--even negative--NRAs. Turkey, which has never been under Communist rule but nonetheless had a highly state-controlled food system (including price regulations and state processing companies), especially prior to the 1990s, maintained among one of the highest levels of support within Europe and Central Asia during 2004­ 05, despite the fact that there was a major policy reform after 2000, including a shift in assistance from market price support toward direct payments. NRAs to agriculture during transition As domestic markets and trade and currency exchange regimes were liberalized in the early 1990s in Europe and Central Asia, farm output declined dramatically. This was mainly the result of nominal input prices increasing much more strongly than output prices. Industrial output declined by a similar order of magnitude, while the services sector, which had been severely constrained under the Commu- nist system (at least as a stand-alone set of activities distinct from state-owned farm and industrial enterprises), grew rapidly after transition began. Beginning in the early 1990s, many trade and price distortions were removed throughout the region. Price, exchange rate, and trade policies were all liberalized; 268 Distortions to Agricultural Incentives: A Global Perspective Figure 6.1. NRAs to Agriculture, Eastern European Countries, 1992­2007 60 40 20 percent 0 20 40 60 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 19 19 19 19 19 19 19 19 20 20 20 20 20 20 20 20 Turkey CEE-10 Russian Federation Ukraine Source: Anderson and Valenzuela (2008), updated from estimates reported in Anderson and Swinnen (2009). subsidies were cut; hard budget constraints were introduced; property rights were privatized; and production decisions were shifted to companies and households. One consequence was that, on average, support to agriculture fell to very low lev- els in the early 1990s (as it did also for industrial production). Between 1992 and 1995, nominal assistance to agriculture averaged just 12 percent in the CEE-10 and was less than zero in Bulgaria and the three Baltic nations--as it was in Russia and Ukraine. By contrast, in Turkey, where nominal assistance averaged just 5 per- cent during 1986­89, the NRA rose to an average of 15 percent in 1992­95 and 25 percent in 1996­99 (figure 6.1 and table 6.2). Changes in policy in the region, and hence in rates of agricultural assistance, have not been smooth. Rather, they have been characterized by stop-go phases and sometimes even reversals of previous reforms, as is apparent from figure 6.1. Despite that heterogeneity of experiences, one can identify a couple of general phases in the policy changes. Following its initial collapse, support to agriculture increased during the mid- 1990s in some countries. In CEE, this was driven by the explicit introduction of new support policies, while in Russia it reflected primarily exchange rate develop- ments, which, in the presence of institutional constraints, constrained the pass- through of border prices to farmgate prices and pushed assistance rates up to high levels. Eastern Europe and Central Asia 269 Table 6.2. NRAs to Agriculture,a Eastern European and CIS Focus Countries, 1992­2007 (percent) 1992­95 1996­99 2000­03 2004­07d Bulgaria 19 11 0 7 Czech Republic 20 19 27 24 Estonia 14 20 20 23 Hungary 19 18 34 20 Latvia 15 30 36 28 Lithuania 19 29 32 29 Poland 10 24 7 32 Romania 22 29 53 50 Slovak Republic 28 26 30 21 Slovenia 64 79 80 31 CEE-10 12 22 24 31 Turkey 15 25 22 30 Russian Federation 8 25 13 19 Ukraine 21 1 11 2 Kazakhstan -- -- 0 5 All focus countries: Unweighted averageb 6 24 25 22 Weighted averagea 3 22 16 25 Dispersionc 26 21 26 14 Source: Anderson and Valenzuela (2008), updated from estimates reported in Anderson and Swinnen (2008). Note: -- not available. a. Weighted average for each country, including product-specific output and input distortions and NPS assistance as well as authors' guesstimates for noncovered farm products, with weights based on gross value of agricultural production at undistorted prices. b. The unweighted average is the simple average across the 14 countries of their national NRA (production-weighted) average NRAs. c. Dispersion is a simple four-year average of the annual standard deviation around a weighted mean of the national agricultural sector NRAs. d. Final column refers just to 2004­05 for Russia and Ukraine and 2004 for Kazakhstan; and the CEE values assume the NRA for each product is the same as for the EU-25 in 2004­06 and for EU-27 in 2007, such that the differences across CEE countries are due to differing national product weights. The increase in support started first in Central Europe, where, after the radical liberalization in the early 1990s, political pressures induced governments to rein- troduce a series of measures. NRAs increased from close to zero in 1992 to 20 to 30 percent (or more) in the second half of the 1990s, but then they stabilized in the lead-up to EU accession in 2004. Between 2000 and 2003, the average rate of assis- tance to agriculture in the CEE-10 was just under 25 percent (figure 6.1), slightly less than half the rate of assistance (including from programs somewhat decou- 270 Distortions to Agricultural Incentives: A Global Perspective pled from production) provided to farmers in the EU-15 at that time (see Josling 2008). Further East, two economic changes in the late 1990s had major impacts on agricultural incentives. First, the Russian financial crisis and the associated deval- uation of the ruble, in the presence of imperfect pass-through, caused a strong decline in the estimated rates of assistance to agriculture. This macroeconomic correction brought down estimated assistance rates dramatically. Second, in the 2000s, the hike in world energy and mineral prices and general economic growth improved many CIS governments' budgetary situations. The improvement in the budgetary situation, in turn, has increased the level of support to agriculture. In Russia, for example, the government announced that agriculture would be a pri- ority area for additional funding in 2005. Not all the additional funding is to go to subsidies, as some governments have plans to spend considerably on infrastruc- ture and quality upgrading in agriculture. Rural incomes have improved because of better (and timely) payments of farm workers' wages and pensions to farm and rural workers, and because of improved rural services. The combination of all these developments led to a somewhat lower weighted average NRA for agriculture in the region as a whole in the four-year period from 2000 to 2004 than in the period immediately before: 16 percent during 2000­03, compared with 22 percent in 1996­99 (table 6.2). In Russia, the level of average support fell even more (from 25 to 13 percent). However, supports rose again in 2004 and 2005, including in countries that have since joined the EU, before drop- ping again as international food prices rose in 2007. Meanwhile, the NRA moved closer to zero in Ukraine in 2005, but is probably still very negative in the rest of Central Asia. There is thus a very wide dispersion in average NRAs across coun- tries in the region, from very high levels in the highest-income country (Slovenia) to continuing negative levels in the poorest countries of Bulgaria, Kazakhstan, and Ukraine (figure 6.2). There are major differences in distortions across commodities, too. In the 1980s, virtually all commodities were supported, albeit some more than others. Now, some commodities are now taxed in the CIS (table 6.3). For example, by 2000­03, sugar, poultry, and milk were the most highly protected commodities in the CEE-10, while grains, beef, and pork were the least assisted. Meanwhile, in Russia and Ukraine, the range is even more extreme, from high positive assistance to livestock and sugar to high negative assistance to the production of the key feed inputs into livestock (coarse grains and oilseeds). Sunflower seeds, Russia's most commonly produced and traded oilseed, was in fact the only consistently exported commodity through the transition period. The case of Kazakhstan in 2004 was even starker, where import-competing producers were highly assisted while exporting industries had to endure negative assistance, such that even though the average NRA was close to zero, a strong antiagricultural trade bias prevailed. Eastern Europe and Central Asia 271 Figure 6.2. NRAs to Agriculture, Individual Eastern European Focus Countries, 2000­03 100 80 60 percent 40 20 0 20 ia ia Hu ia y a Re ic ic ey de a n Bu d ia an e ar ni Fe oni in tio n l bl en an tv ar ub rk st la ua ra ng pu La lg ra kh m Tu ov t Po Cz Rep Uk Es th Ro za Sl Li Ka ak h ec n ov ia Sl ss Ru Source: Anderson and Valenzuela (2008), updated from estimates reported in Anderson and Swinnen (2009). Government intervention and controls are especially important in a few key commodities within each country, often because of (real or imagined) food secu- rity concerns or the need to raise government revenue to meet other priorities. This is the case for grains and oilseeds in Bulgaria, Russia, and Ukraine, both those used for human consumption and those that support (via low feed input prices) the production of livestock products. It has also been the case for cotton in Tajik- istan, Turkmenistan, and Uzbekistan, where heavy taxation distorts incentives for producers--although open or porous borders make the taxing of cotton exports difficult when tax rates vary across countries in the subregion. The trade bias index (TBI) reported in table 6.4 is one way of capturing the diversity of assistance rates across farm commodities. The more negative that index, the greater the gap between assistance to import-competing farm indus- tries and assistance (or in some cases effective taxation) of export industries. Table 6.4 suggests that the antitrade bias has been a persistent feature of agricultural policies in the region throughout the transition period. Indeed, it has been worse in recent years than it was a decade earlier. A more comprehensive way to measure the extent of variance of rates across time is to calculate the standard deviation of NRAs for the covered products. These too have remained persistently high. On average, they have been higher in recent years than in the early stages of transition (table 6.5). 272 Distortions to Agricultural Incentives: A Global Perspective Table 6.3. NRAs, Key Covered Farm Products,a Eastern European and CIS Focus Countries, 1992­2005 (percent) 1992­95 1996­99 2000­03 2004­07 Wheat 6 13 2 9 Barley 1 16 5 6 Oats 11 7 27 4 Rye 0 14 10 2 Maize 16 3 16 21 Rapeseeds 8 18 1 0 Sunflower 13 13 13 4 Soybeans 45 0 9 4 Cotton 45 47 31 29 Sugar 23 80 73 91 Potatoes 25 25 60 57 Beef 16 20 36 53 Sheep meat 10 10 3 15 Pig meat 8 16 12 32 Poultry 26 43 52 75 Eggs 16 48 2 25 Milk 6 43 25 26 All covered products 2 19 13 22 Domestic market support 1 1 1 2 Border market support 4 15 11 19 Dispersion of product NRAs 21 29 29 31 Product coverageb 62 63 61 62 Source: Anderson and Valenzuela (2008), updated from estimates reported in Anderson and Swinnen (2009). a. Region's weighted average for each product and for all covered products, with weights based on gross value of agricultural production at undistorted prices. b. Dispersion is the standard deviation shown as the simple four-year average of the annual standard deviation around the weighted mean. The total amount of support is an imperfect indicator of distortions to incen- tives, since different trade, price, and subsidy instruments have different distortion effects. Most support to agriculture in the region was, and despite the reforms, still is provided via highly distortive and hence inefficient policy instruments. Under Communism, output price distortions were complemented by heavy distortions in input prices, in particular low fertilizer and energy prices and subsidized irriga- tion, while in the 1990s the majority of farm support in the CEE countries was provided by keeping output prices above border prices (see near the bottom of table 6.3). However, the share of support from those measures has declined over Eastern Europe and Central Asia 273 Table 6.4. TBI,a Eastern European and CIS Focus Countries, 1992­2007 (percent) 1992­95 1996­99 2000­03 2004­07b Bulgaria 0.02 0.17 0.18 0.30 Czech Republic 0.05 0.10 0.23 0.16 Estonia 0.21 0.16 0.01 0.24 Hungary 0.14 0.12 0.11 0.05 Latvia 0.35 0.18 0.15 0.22 Lithuania 0.50 0.32 0.19 0.07 Poland 0.19 0.19 0.24 0.26 Romania 0.19 0.28 0.40 0.23 Slovak Republic 0.03 0.09 0.05 0.06 Slovenia 0.26 0.40 0.38 0.18 CEE-10 0.15 0.16 0.23 0.02 Turkey 0.32 0.46 0.32 0.19 Russian Federation 0.11 0.31 0.34 0.24 Ukraine 0.12 0.25 0.21 0.42 Kazakhstan -- -- 0.01 0.32 All focus countries 0.15 0.16 0.23 0.02 Source: Anderson and Valenzuela (2008), updated from estimates reported in Anderson and Swinnen (2009). Note: -- not available. a. TBI [(1 NRAagx 100) (1 NRAagm 100) 1], where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector. b. Final column refers just to 2004­05 for Russia and Ukraine and 2004 for Kazakhstan. the past decade, consistent with developments within the EU-15. These policy changes are reflected in the composition of the assistance that farms have received. Under the Communist system, price support and output subsidies were the main component of agricultural assistance in the CEE countries, accounting for more than 80 percent of their NRA. After the reforms in the early 1990s, the share of market support and output subsidies declined substantially, falling below half. Since then, it has grown again, to around half of the NRA. Other important com- ponents of the NRAs of CEE countries and Turkey were input subsidies, direct payments, and other NPS subsidies, plus some decoupled payments in the most recent years (table 6.6).7 In the CIS countries, those payments include soft loans and debt forgiveness, which continue to play an important role. While fiscal con- straints for most of the 1990s limited the government's ability to support farms by this means, the budgetary situation changed in the 2000s as earnings from mineral and energy exports grew and these earnings have become a more important source of government assistance to farmers. 274 Distortions to Agricultural Incentives: A Global Perspective Table 6.5. Dispersion of NRAs across Covered Agricultural Products,a Eastern European and CIS Focus Countries, 1992­2007 (percent) 1992­95 1996­99 2000­03 2004­07c Bulgaria 18 21 25 48 Czech Republic 27 28 23 53 Estonia 24 28 20 45 Hungary 34 41 62 49 Latvia 42 40 44 58 Lithuania 47 47 53 54 Poland 31 28 27 53 Romania 48 52 59 69 Slovak Republic 25 27 25 49 Slovenia 50 42 39 57 CEE-10 35 35 38 54 Turkey 62 65 53 69 Russian Federation 37 33 40 40 Ukraine 66 48 37 32 Kazakhstan -- -- 28 39 All focus countriesb 39 38 38 45 Source: Anderson and Valenzuela (2008), updated from estimates reported in Anderson and Swinnen (2009). Note: -- not available. a. Dispersion for each country is a simple four-year average of the annual standard deviation around a weighted mean of NRAs across covered products. b. Unweighted average, that is, the simple average across the 14 countries of their four-year simple average dispersion measures. c. Final column refers just to 2004­05 for Russia and Ukraine and 2004 for Kazakhstan. The GSE of the assistance to farmers, expressed in constant 2000 dollar terms, shows that Turkey has been the largest supporter between the early 1990s and mid-2000s, though Russia is rapidly catching up to Turkey. Romania and Poland are the next largest aggregate supporters. The region as a whole provides support of more than US$24 billion per year, compared with just US$3 billion in the early years of transition (table 6.7a). When expressed on a per farmer basis, the range is huge. In 2000­03, for example, it ranged from negative amounts (­US$300) in Ukraine and Kazakhstan to an average of US$980 in the CEE countries, US$430 in Russia, more than US$2,200 in Hungary and Romania, and a huge amount in high-income Slovenia (table 6.7b). This compares with US$8,400 per farmer in the EU-15 in 2000­04 (Josling 2008). Slovenia's support has come down signifi- cantly since its accession to the EU, however, to an average of just under Eastern Europe and Central Asia 275 Table 6.6. Components of NRAs to Agriculture, Eastern Europe and CIS, 1961­2007 (percent) a. CEE-10 1992­94 1995­99 2000­04 2005­07 NRA, covered products 10.4 16.4 25.5 17.7 NRA, noncovered products 10.6 17.3 26.9 24.7 NRA, all agriculture (excluding NPS) 10.5 16.7 26.1 21.3 All importables 19.1 32.1 45.2 31.7 All exportables 2.4 7.0 10.9 12.0 NRA, NPS assistance 1.9 2.5 2.5 4.4 NRA, all agriculture (including NPS assistance) 12.4 19.2 28.6 25.7 NRA, decoupled payments 0.6 0.8 2.9 12.1 NRA, all agriculture (including NPS and decoupled assistance) 13.0 20.1 31.5 37.8 NRA, all agricultural tradables (including NPS) 12.9 19.5 26.4 15.9 NRA, all nonagricultural tradables 5.7 4.9 4.4 4.6 RRA 6.7 14.1 21.1 10.7 b. Russian Federation and Ukraine 1992­94 1995­99 2000­04 2005 NRA, covered products 23.7 8.6 4.2 11.0 NRA, noncovered products 23.9 11.3 6.2 14.1 NRA, all agriculture (excluding NPS) 23.7 9.4 4.9 12.0 All importables 25.4 20.3 24.7 22.9 All exportables 21.5 13.4 15.3 2.0 NRA, NPS assistance 6.3 4.8 2.7 0.0 NRA, all agriculture (including NPS assistance) 17.4 14.2 7.5 12.0 NRA, decoupled payments 2.6 0.6 0.0 0.0 NRA, all agriculture (including NPS and decoupled assistance) 14.8 14.8 7.6 12.0 NRA, all agricultural tradables (including NPS) 17.4 14.2 5.9 6.3 NRA, all nonagricultural tradables 4.9 9.0 8.1 7.3 RRA 21.5 4.8 2.0 0.9 (Table continues on the following page.) 276 Table 6.6. Components of NRAs to Agriculture, Eastern Europe and CIS, 1961­2007 (continued ) (percent) c. Turkey 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, covered products 18.8 17.7 6.9 8.1 29.5 4.0 19.7 21.2 20.0 29.4 NRA, noncovered products 18.8 17.7 6.9 8.1 29.5 4.0 19.7 21.2 20.0 27.8 NRA, all agriculture (excluding NPS) 18.8 17.7 6.9 8.1 29.5 4.0 19.7 21.2 20.0 28.8 All importables 10.8 9.6 5.8 19.7 19.6 28.5 60.2 80.5 54.0 45.0 All exportables 29.9 28.4 18.0 23.3 35.5 8.1 2.5 1.9 3.5 21.9 NRA, NPS assistance 0.1 0.3 1.9 0.6 0.3 0.0 0.0 0.0 3.3 1.5 NRA, all agriculture (including NPS assistance) 18.9 17.9 5.0 7.5 29.2 4.0 19.7 21.2 23.2 30.3 NRA, decoupled payments 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 3.7 2.9 NRA, all agriculture (including NPS and decoupled assistance) 18.9 17.9 5.0 7.5 29.1 4.1 19.7 21.2 26.9 33.2 NRA, all agricultural tradables (including NPS) 18.9 17.9 5.0 7.5 29.2 4.0 19.7 21.2 20.4 23.5 NRA, all nonagricultural tradables 60.5 140.8 49.6 55.7 32.8 20.5 10.0 2.3 0.9 0.5 RRA 46.5 64.0 35.9 35.6 46.6 13.6 8.8 18.6 19.3 23.0 Source: Anderson and Valenzuela (2008), updated from estimates reported in Anderson and Swinnen (2009). Eastern Europe and Central Asia 277 Table 6.7. GSEs of Assistance to Farmers, Total and Per Farm Worker, Eastern European and CIS Focus Countries,a 1992­2007 a. Total (constant 2000 US$ millions per year) 1992­95 1996­99 2000­03 2004­07b Bulgaria 671 381 17 197 Czech Republic 784 632 711 689 Estonia 73 82 74 90 Hungary 856 768 1,205 920 Latvia 208 167 195 179 Lithuania 332 414 395 361 Poland 1,378 3,106 857 4,314 Romania 1,921 2,064 3,332 4,073 Slovak Republic 421 338 309 301 Slovenia 431 483 381 143 CEE-10 4,509 7,674 7,441 11,265 Turkey 4,671 8,033 6,070 10,525 Russian Federation 1,486 7,394 3394 3,100 Ukraine 4,461 70 1157 182 Kazakhstan -- -- 34 69 All focus countries 3,234 23,032 15,715 24,778 b. Per person engaged in agriculturec (constant 2000 US$ per year) 1992­95 1996­99 2000­03 2004­07 Bulgaria 1,429 1,075 65 1,010 Czech Republic 1,423 1,255 1,581 1,762 Estonia 678 898 931 1,267 Hungary 1,335 1,372 2,494 2,253 Latvia 1,038 993 1,333 1,393 Lithuania 1,113 1,693 1,932 2,123 Poland 283 683 204 1,118 Romania 879 1,135 2,202 3,311 Slovak Republic 1,393 1,199 1,197 1,281 Slovenia 10,781 18,225 22,105 14,254 CEE-10 466 893 977 1,682 Turkey 344 566 414 702 Russian Federation 152 842 431 439 Ukraine 956 17 333 60 Kazakhstan -- -- 27 59 All focus countries 86 647 451 752 Source: Anderson and Valenzuela (2008), based on NRA estimates updated from Anderson and Swinnen (2009) and data on the number of farmers from U.N. Food and Agriculture Organization Statistical Database (FAOSTAT). Note: -- not available. a. GSEs include assistance to nontradables and NPS assistance. The number of farmers in these countries is difficult to obtain on a consistent basis. The FAOSTAT numbers may be subject to error. For example, Slovenia's may be understated in FAOSTAT, in which case its GSE per farmer is overestimated. b. Final period refers to 2004­05 for Russia and Ukraine and just 2004 for Kazakhstan. 278 Distortions to Agricultural Incentives: A Global Perspective US$14,000 per farmer in 2005­07. For the EU-acceding countries, per farmer assistance over the next few years is likely to move closer to the EU average. Since most of the support for farmers previously came through price-support measures, most notably import restrictions, these have the effect of raising con- sumer prices by a similar degree when calculated at the farmgate level. That means that prior to the mid-1990s, policies in all but Turkey and Slovenia imposed the equivalent of low or negative taxes on food consumers (CTEs), but thereafter the CTEs have become positive. The region's weighted average CTE in 2000­03 was 17 percent (table 6.8), compared with nearly twice that in the EU-15. CTEs in Romania and Slovenia have been well above that EU average this decade, and so presumably will fall during those countries' transition to the EU's Common Agri- cultural Policy (CAP), especially given the EU's policy reinstrumentation toward more direct farm income supports that do not raise consumer prices of food. Assistance to agriculture relative to other tradable sectors Import tariffs on primary agricultural commodities in Europe and Central Asia are on average twice as high as average tariffs in industry, but only half as high as tariffs on processed food. This is true both for the CEE and CIS countries. It suggests that while the region's farmers receive more tariff protection from competition abroad than do nonagricultural producers, food processors may be effectively protected despite having to pay more than world prices for primary farm products. Import-competing producers are only part of each sector, however. When support for producers of exports in each sector is also taken into account, an overall NRA for all nonagricultural tradable industries can be used, together with the average NRA for agricultural tradable industries, to calculate the RRA. In so far as the NRAs for nonfarm industries are positive, the RRA is lower than the NRA for agriculture. But in most cases, the nonagricultural NRA is very low. Thus the overall NRA for tradable primary agriculture in the region during 2000­03 is estimated to have averaged more than three times higher than for producers of nonagricultural tradables (15 as com- pared with 5 percent), so the RRA averaged 10 percent. Only in three countries-- Bulgaria, Kazakhstan, and Ukraine--has agricultural production been assisted less than nonagricultural tradables (RRA 0) during the present decade. And in virtu- ally all countries for which there is a time series, the RRA is higher at the end of the sample period than in the first few years of transition (figure 6.3). Forces behind Transitional Policy Choices Several political economy stylized facts widely observed in market economies-- for reasons explained in, for example, Anderson and Hayami (1986), Anderson (1995), Swinnen (1994), and de Gorter and Swinnen (2002)--are also found in Eastern Europe and Central Asia 279 Table 6.8. Percentage CTE of Policies Assisting Producers of Covered Farm Products,a Eastern European and CIS Focus Countries, 1992­2007 (percent, at primary product level ) 1992­95 1996­99 2000­03 2004­07e Bulgaria 20 10 3 7 Czech Republic 23 19 22 20 Estonia 15 12 9 20 Hungary 18 15 22 16 Latvia 2 28 32 32 Lithuania 20 21 20 29 Poland 4 2 18 25 Romania 6 16 39 29 Slovak Republic 13 15 16 17 Slovenia 48 58 45 24 CEE-10 2 11 24 23 Turkey 10 20 16 9 Russian Federation 37 13 16 24 Ukraine 25 0 3 3 Kazakhstan -- -- 4 16 All focus countries: Unweighted average 1 16 19 20 Weighted averageb 13 13 16 14 Dispersion of national CTEsc 27 17 14 10 Dispersion of region's product CTEsd 35 37 47 53 Source: Anderson and Valenzuela (2008) updated from estimates reported in Anderson and Swinnen (2009). Note: -- = not available. a. Assumes the consumer tax equivalent (CTE) is the same as the NRA derived from trade measures (that is, not including any input taxes/subsidies or domestic producer price subsidies/taxes). b. Weights are consumption valued at undistorted prices, where consumption (from FAO) is production plus imports net of exports plus change in stocks of the covered products. c. Simple four-year average of the annual standard deviation around a weighted mean of the regional average CTE across the covered products. d. Simple four-year average of the annual standard deviation around a weighted mean of the national average CTE for covered products. e. Final column refers to 2004­05 for Russia and Ukraine and just 2004 for Kazakhstan; while CEE values assume the CTE for each product is the same as for the EU-25 in 2004­06 and for EU-27 in 2007, such that the differences across CEE countries is due to differing national consumption weights. Europe and Central Asia's transition economies. Specifically, for this region, as elsewhere, farmer assistance tends to be higher in higher-income countries, and in countries with weaker comparative advantage in agriculture. Hence, it is likely that similar politico-economic interactions and mechanisms are at work 280 Distortions to Agricultural Incentives: A Global Perspective Figure 6.3. RRAs to Agriculture,a Eastern European Focus Countries, 1992­95 and 2000­03 Slovenia Romania Latvia Lithuania Hungary Czech Republic Estonia Turkey Regional average Slovak Republic Poland Russian Federation Bulgaria Kazakhstan Ukraine 40 20 0 20 40 60 80 percent 1992­95 2000­03 Source: Anderson and Valenzuela (2008), updated from estimates reported in Anderson and Swinnen (2009). a. RRA is defined as 100*[(100 NRAagt ) (100 NRAnonagt ) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respec- tively. For several countries no estimates are available for the 1992­95 period. in this region as in other parts of the world. However, those correlations are becoming weaker over time among the CEE countries. Taking on the EU's CAP is part of the explanation, but there are also other forces, both domestic and international, that underlie the political economy of agricultural policies in the CEE and CIS countries. Several of those key forces are discussed in the following subsections. Causes of rent extraction Traditionally, heavy negative government intervention in the form of depressed incentives tends to be concentrated on commodities that have the potential to provide export tax revenue for the government. This is especially the case in the cotton sectors of Tajikistan, Turkmenistan, and Uzbekistan. There, as in a number Eastern Europe and Central Asia 281 of African countries (see Anderson and Masters 2009), the government controls the cotton chain so as to extract rents, thereby depressing farmers' prices and pro- duction incentives. There is a clear division in Central Asia between the roughly neutral policy toward cotton in Kazakhstan and the Kyrgyz Republic (where cotton previously accounted for a relatively modest share of total exports) on the one hand, and on the other the extensive taxation and extraction of rents from cotton in Tajikistan, Turkmenistan, and Uzbekistan (countries where cotton traditionally was a very important export tax resource).8 In Turkmenistan and Uzbekistan, governments use state monopoly powers over marketing to transfer substantial resources out of agriculture. Most of the transfers in Uzbekistan appear to go to general govern- ment revenue, whereas in Turkmenistan much is wasted (for example, in ineffi- cient cotton mills, which have negative value added) or accrues to secret accounts under the president's personal control. Moreover, recently, potentially important reforms have been introduced in Uzbekistan to reduce some of the distortions to farm incentives, while almost none have taken place in Turkmenistan. In Tajikistan, the rent distribution is more opaque, but equally detrimental to farms, as a coalition between the government and a monopolistic private trading company has caused depressed prices and incentives for farmers. Not surprisingly, cotton farmers have responded sharply to these incentive distortions, both in area and output: with rapid growth in Kazakhstan and the Kyrgyz Republic, and with declines or stagnation in the other countries. The grain and oilseed export sectors of Bulgaria, Ukraine, and the grain sur- plus regions of Russia are similarly characterized by heavy government regulation and interventions. In traditional grain-exporting countries such as Bulgaria and Ukraine, for historic and psychological reasons, the grain sector has dispropor- tionate political significance. For example, in the mid-1990s in Bulgaria, ministers of agriculture were regularly forced to resign following reports of grain shortfalls or unregulated exports threatening the local grain supply. In Ukraine, ad hoc grain market interventions continued in recent years. Opportunities for rent seeking from distorted policies inhibit policy reform, as the few who benefit disproportionately from existing distortions lobby strongly for their continuation. This applies to various policies, such as cotton regulations in Central Asia, grain trade regulations in Bulgaria, Russia, and Ukraine and water policies in Central Asia. But it also applies to several policies in countries in which benefits go a specific group of farms. For example, the continuation of soft budget constraints in the large CIS countries, along with the failure of governments to enforce bankruptcies and strong land rights, disproportionately benefit large farming companies, while smaller family farms are often hurt by these policies. In Turkey, agricultural parastatal companies and marketing cooperatives benefit 282 Distortions to Agricultural Incentives: A Global Perspective from "farm support" and are major lobbyists in favor of market regulations and assistance packages. Sometimes, specific political, regional, or ethnic coalitions also play a role. In Kazakhstan, for example, many residents of the rich northern grain regions were Russian and German. After independence, power shifted to Kazakh nationals, lim- iting the Russian and German groups' influence in government and causing many to emigrate. Another recent example is Bulgaria, where the government's resist- ance to privatizing tobacco processing companies and its decision to allocate a dis- proportionate amount of subsidies to tobacco growers are due to the fact that the Turkish minority in Bulgaria is particularly active in the sector and traditionally has held key positions in the Ministry of Agriculture. Causes of increases in support during transition Increases in agricultural support in the CEE countries in the second half of the 1990s and more recently in CIS countries are the result of the interaction of domestic political forces with international events. The increase in farmer assis- tance in CEE countries was likely caused by "normal" domestic internal pressures that are brought to bear in a contestable political environment, which results in an increase in agricultural protectionism as per capita income increases and agricul- tural comparative advantage declines. Protectionism between the mid-1990s and mid-2000s was also a case of reversing somewhat the overshooting in reform dur- ing the first few years of transition. Overlaying these trends is the EU accession process, which encouraged CEE governments to target the levels of support expected in the EU by the end of the phase-in period of accession, so as to maximize the transfer of benefits from Brussels. However, it appears that in the years leading up to accession, the process had more impact on the introduction of new support instruments than on the overall level of support, probably because the full cost of that support had to be borne by the national economy prior to EU accession (Swinnen 2002). Another factor contributing to the increase in agricultural support in the mid- 1990s was the improvement in the government's budgetary situation, which allowed more subsidies to be given to farmers than was possible in the early years of transition. This factor has played a role throughout Europe and Central Asia, in particular in Russia and some of its neighbors, where recovery from the 1998 financial crisis has been aided by windfall gains from the dramatic rise in the export prices of energy raw materials. This factor has been stronger in countries where governments have more access to mineral resources, such as in Russia (oil and gas), Kazakhstan (oil), and Turkmenistan (gas). Eastern Europe and Central Asia 283 Crises and political change General political and economic crises have played an important role in inducing changes in agricultural distortions. The most obvious example is the fall of the Communist regime and the disintegration of the Soviet Union--and of the cen- tral directives coming from Moscow. However, even later there are several exam- ples in which general crises have triggered changes. Most often, the policy reforms came only after new elections induced a change in government, reflecting changed electoral preferences. For example, in Romania and Bulgaria, important progress in the removal of distortions and market reforms occurred only in the late 1990s, after electoral change brought reform-minded governments to power. In Bulgaria, significant progress was brought about by the financial crisis in 1996. Important reform progress was made in Ukraine in the years after the 1999 election in which the large farm lobby fell out with President Kuchma, who consequently introduced a series of important reforms that farmers had previously successfully opposed. Democratic political change, however, is not a sufficient condition in itself for better agricultural policies. For example, political revolutions in Ukraine and the Kyrgyz Republic (the "Orange Revolution" and the "Tulip Revolution," respec- tively) have not contributed to better agricultural policy. In fact, agricultural dis- tortions seem to have increased under the postrevolution Ukrainian government, while in the Kyrgyz Republic political change has mostly resulted in more market instability, while relatively little distortion remains in agriculture. Impact of international agreements EU accession, both prospective and actual, has had obvious and profound influ- ences on policy choices. The CEE countries that joined in May 2004 and January 2007 have raised domestic agricultural and food prices toward EU-15 levels (on average, since for some prices came down). An important part of the EU farm subsidies is now in the form of direct payments. Though CEE farms receive con- siderably less of these subsidies than do EU-15 farmers, they will gradually increase to reach EU-15 levels by 2010. Another important difference is that sub- sidies in the EU-15 will be given on a per farm basis (single farm payments) earlier than in CEE countries. CEE countries also have been induced also to undertake major regulatory improvements to stimulate their markets, including private investments in the food chain and public rural infrastructure investments. Likewise, their trade poli- cies have changed to allow free access for all products from other EU-27 member countries and, in most cases, also freer access for nonagricultural products from 284 Distortions to Agricultural Incentives: A Global Perspective non-EU countries (the latter because the common external tariff typically was lower than that previously applied in acceding countries). The EU accession process, however, has not caused a major increase in food prices in the CEE coun- tries. Two reasons behind this are the increased competition in consumer markets in the CEE countries with the full opening of agri-food markets to imports, and the massive inflow of foreign direct investment in the retail sector. Impacts of other international agreements (including WTO accessions) have varied. The Czech Republic, Hungary, Poland, Romania, the Slovak Republic, Slovenia, and Turkey have been members of the WTO since its creation in 1995. Albania, Armenia, Bulgaria, Estonia, Georgia, the Kyrgyz Republic, Latvia, Lithua- nia, and Ukraine joined the WTO later, while Russia and Kazakhstan are still negotiating their WTO accession. WTO accession has not strongly disciplined European and Central Asian coun- tries that were founding members in 1995 (Bacchetta and Drabek 2002). For those that had to negotiate their entry in the latter 1990s, constraints on introducing or maintaining distortions are more serious. And for large countries in the region still in the process of negotiating their accession, notably Kazakhstan and Russia, the WTO has been even tougher in its demands. Whether those demands will prove an agricultural trade-liberalizing force remains to be seen, but at least they will provide a ceiling on the extent to which agricultural protection and subsidies may be raised in the future. For the CEE countries, the most important WTO impact has been indirect: in anticipation of eastward enlargement, the EU was forced to introduce major changes to its CAP, which in turn has affected postaccession agricultural distor- tions in the CEE countries. A further and somewhat erratic influence has been regional trading arrange- ments among the European and Central Asian countries. These include the Eurasian Economic Community (EAEC), the Central European Free Trade Area, and the Baltic Free Trade Area. However, the impact of these agreements on reducing agricultural policy distortions has been generally limited, because the agreements include many exceptions for agricultural and food products, espe- cially for so-called sensitive products, which make up a substantial share of pro- duction. Moreover, Central Asian countries such as Kazakhstan and the Kyrgyz Republic have been reluctant to join the EAEC because the organization would impose Russia's trade and customs preferences on them. Influence of international institutions The role of other international institutions was very important at the start of tran- sition in Europe and Central Asia, as it provided policy reform guidance in all Eastern Europe and Central Asia 285 these countries. However, in more recent years, the influence of international institutions has declined. For countries joining the EU, policy advice from Brus- sels was perceived as more relevant. This is especially, but not only, the case for the EU accession countries. For countries aspiring to join the EU (such as most of the Balkan countries and Ukraine), or for those that view the accession countries as models for their own development strategies, policy advice from Brussels is taken seriously. Another reason is that in many of the countries of Southeast Europe and the CIS, improved macroeconomic situations have made them less beholden to international financial institutions requiring reforms as a condition for providing loans or financial assistance. Prospects for Further Reducing Distortions Clearly, there have been major reductions in distortions to agricultural incentives in Europe and Central Asia over the past two decades. In many countries in the region, average protection levels are now relatively low. However, there is still sub- stantial room for further reduction of distortions to agricultural incentives. This could be done through various means: overall reductions in support, shifting sup- port to less-distortive policy instruments, and focusing budgetary expenditures on public good investments (in infrastructure and in institutions that reduce trade costs) rather than on farm subsidies, shifting from a quantity-based to a quality-based policy paradigm, and so forth. In terms of further reductions in policy distortions, some of the most distortive cases concern taxation of agriculture, most notably the control and rent extrac- tion in the cotton sectors in some Central Asian countries. Removing those distor- tions would allow a substantial improvement in incentives to domestic producers. Though some progress has been made in recent years, much more can be done. Countries for which EU accession is unlikely to happen in the medium term, such as Turkey, Ukraine, and several of the Balkan countries, should focus their policy attention in the near term on efficiency improvements in both their policies and their agricultural economies. This is consistent with the objective of EU acces- sion, since the EU itself has moved in recent years to more decoupled farm sup- port and is demanding that member countries move in that direction and improve the efficiency of their farms and food companies.9 The same policy framework should be promoted in countries further east, including in those that are likely to spend more funds on agriculture in the com- ing years as their fiscal situation improves. Increased funding should be focused on upgrading infrastructure, on quality and efficiency of the agri-food system, and on the introduction or improvements of a variety of institutions necessary to support rural markets. In several of the poorer and larger CIS countries, institu- 286 Distortions to Agricultural Incentives: A Global Perspective tional and infrastructure problems, as well as corruption, remain major con- straints to trade and thereby distort farm incentives. Competition and antitrust policy are important related areas for policy atten- tion. In supply chains where farms have to sell their products to trading, process- ing, and retailing companies, the ability to choose freely between companies is of crucial importance in creating better conditions for farms. This applies across the region, where monopoly buyers (state-owned or private) push down prices and contract conditions, although the source of anticompetitive behavior and policy details are likely to differ, for example, between the increasing dominance of large retail chains in Central Europe versus some of the government-controlled cotton chains in Central Asia. Despite constraining political economy forces, there are prospects for further reducing distortions to agricultural incentives in the foreseeable future. Accession of the CEE countries to the EU has increased their levels of farm assistance, though they now face more competition within the enlarged EU. While reducing CEE farm assistance in the future will not happen without reductions in EU pro- tection levels, some reforms are currently underway in the EU (for example, the cut in EU sugar price support and the shift from per hectare payments to single farm payments). However, slow and intermittent progress in the WTO's Doha Round reduces the pressure for further reforms. Meanwhile, in the mineral- and energy-rich CIS countries, the rise in export earnings reduces budgetary con- straints on governments inclined to give assistance to farmers as national incomes grow. CIS regional trade policies that affect markets, largely ad hoc and nontrans- parent, are also important distortions. Eliminating these policy interventions would require fundamental reforms of Russia's political system, including a transforma- tion of attitudes and behaviors involving governance that Russian accession to the WTO is unlikely to alter in the medium term. Notes 1. This is a strong feature of Asia's economies in transition as well. For a comparison of the Asian and European transition experiences, see Swinnen and Rozelle (2006). 2. The only country from this region that was part of the Krueger, Schiff, and Valdés (1991) study was Turkey, for the period 1961 to 1983. However, a follow-up study subsequently undertaken for a few economies in transition has been made available in a World Bank technical report (Valdés 2000). 3. For a detailed discussion of how misalignments in the exchange rate affect the measurement and interpretation of NRAs in Russia during its transition period, see Liefert and Liefert (2008). 4. In addition to the present study's estimation of price wedges, the latter would require a sophisti- cated economic model of the national economy with some quantification of the difference between private and social marginal costs. Such welfare analysis is thus beyond the scope of the present study. 5. The dramatic implications-including millions of peasants dying of starvation-are documented in sobering detail in Conquest (1986). Eastern Europe and Central Asia 287 6. For an assessment of the support to farmers in the 1980s, see Cook, Liefert, and Koopman (1991). 7. Water price regulations and subsidies are important policy instruments in the irrigated regions of Central Asia, but it was not possible in this study to estimate their impact on NRAs. Energy policies are still used to assist various sectors, for example in Russia, but since they do not favor agriculture in particular, and are becoming less important, they too have been omitted from the NRA estimates. 8. Price and trade data were not sufficiently reliable to allow NRA calculations, but Pomfret (2008a, 2008b) and Christensen and Pomfret (2008) provide considerable informal information supporting the claims above. 9. From this perspective, it is important to point to the importance of other reforms, such as macroeconomic and regulatory reforms to stimulate food industry investment, labor market reforms to enhance off-farm employment opportunities, and credit reforms to stimulate access to rural credit. References Anderson, K. 1995. "Lobbying Incentives and the Pattern of Protection in Rich and Poor Countries." Economic Development and Cultural Change 43 (2): 401­23. Anderson, K., Y. Hayami, et al. 1986. The Political Economy of Agricultural Protection: East Asia in Inter- national Perspective. Boston, London, and Sydney: Allen and Unwin. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC (and appendix A of this volume). ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson, K., and W. Masters, eds. 2009. Distortions to Agricultural Incentives in Africa. Washington, DC: World Bank. Anderson, K., and J. Swinnen, eds. 2008. Distortions to Agricultural Incentives in Europe's Transition Economies. Washington, DC: World Bank. ______. 2009. "Distortions to Agricultural Incentives in Eastern Europe and Central Asia." Agricul- tural Distortions Working Paper 63, World Bank, Washington, DC. Anderson, K., and E. Valenzuela. 2008. "Estimates of Global Distortions to Agricultural Incentives, 1955 to 2007." World Bank, Washington, DC. http://www.worldbank.org/agdistortions. Bacchetta, M., and Z. Drabek. 2002. "Effects of WTO Accession on Policy-Making in Sovereign States: Preliminary Lessons from the Recent Experience of Transition Countries." Staff Working Paper DERD-2002-02, World Trade Organization, Geneva. Broadman, H. G., ed. 2006. From Disintegration to Reintegration: Europe and Central Asia in Interna- tional Trade. Washington, DC: World Bank. Chen, S., and M. Ravallion. 2007. "Absolute Poverty Measures for the Developing World, 1981­2004." Policy Research Working Paper 4211, World Bank, Washington, DC. Christensen, G., and R. Pomfret. 2008. "The Kyrgyz Republic." In Distortions to Agricultural Incentives in Europe's Transition Economies, ed. K. Anderson and J. Swinnen. Washington, DC: World Bank. Conquest, R. 1986. The Harvest of Sorrow. London: Oxford University Press. Cook, E. C., W. M. Liefert, and R. B. Koopman. 1991. "Government Intervention in Soviet Agriculture: Estimates of Consumer and Producer Subsidy Equivalents." Staff Report AGES 9146, Economic Research Service, U.S. Department of Agriculture, Washington, DC. Crafts, N., and G. Toniolo. 2008. "European Economic Growth, 1950­2005: An Overview." CEPR Discussion Paper 6863, Centre for Economic Policy Research, London. de Gorter, H., and J. Swinnen. 2002. "Political Economy of Agricultural Policies." In volume 2 of Hand- book of Agricultural Economics, ed. B. Gardner and G. Rausser. Amsterdam: Elsevier Science. Dries, L., and J. Swinnen. 2004. "Foreign Direct Investment, Vertical Integration and Local Suppliers: Evidence from the Polish Dairy Sector." World Development 32 (9): 1525­44. 288 Distortions to Agricultural Incentives: A Global Perspective Dries, L., T. Reardon, and J. Swinnen. 2004. "The Rapid Rise of Supermarkets in Central and Eastern Europe: Implications for the Agrifood Sector and Rural Development." Development Policy Review 22 (5): 525­56. Ellman, M. 1988. "Contract Brigades and Normless Teams in Soviet Agriculture." In Socialist Agricul- ture in Transition: Organizational Response to Failing Performance, ed. J. C. Brada and K. E. Wädekin, 23­33. Boulder, CO: Westview Press. Gray, K. R., ed. 1990. Soviet Agriculture: Comparative Perspectives. Ames, IA: Iowa State University Press. Huffman, S. K., and S. R. Johnson. 2004. "Impacts of Economic Reform in Poland: Incidence and Wel- fare Changes within a Consistent Framework." Review of Economics and Statistics 86 (2): 626­36. Josling, T. 2008. "Distortions to Agricultural Incentives in the Western Europe." Agricultural Distor- tions Working Paper 61, World Bank, Washington, DC. http://www.worldbank.org/agdistortions. Krueger, A. O., M. Schiff, and A. Valdés. 1991. The Political Economy of Agricultural Pricing Policy, Volume 3: Africa and the Mediterranean. Baltimore: Johns Hopkins University Press for the World Bank. Liefert, W., and O. Liefert. 2008. "The Russian Federation." In Distortions to Agricultural Incentives in Europe's Transition Economies, ed. K. Anderson and J. Swinnen. Washington, DC: World Bank. Liefert, W., and J. Swinnen. 2002. "Changes in Agricultural Markets in Transition Economies." Agricul- tural Economic Report AER806, Economic Research Service, U.S. Department of Agriculture, Washington, DC. Pomfret, R. 2008a. "Kazakhstan." In Distortions to Agricultural Incentives in Europe's Transition Economies, ed. K. Anderson and J. Swinnen. Washington, DC: World Bank. ______. 2008b. "Tajikistan, Turkmenistan and Uzbekistan." In Distortions to Agricultural Incentives in Europe's Transition Economies, ed. K. Anderson and J. Swinnen. Washington, DC: World Bank. Sandri, D., E. Valenzuela, and K. Anderson. 2007. "Economic and Trade Indicators for the Europe's Transition Economies, 1990 to 2004." Agricultural Distortions Working Paper 22, World Bank, Washington, DC. http://www.worldbank.org/agdistortions. Swinnen, J. 1994. "A Positive Theory of Agricultural Protection." American Journal of Agricultural Eco- nomics 76 (1): 1­14. ______. 2002. "Transition and Integration in Europe: Implications for Agricultural and Food Markets, Policy and Trade Agreements." The World Economy 25 (4): 481­501. Swinnen, J., and S. Rozelle. 2006. From Marx and Mao to the Market: The Economics and Politics of Agri- cultural Transition. New York: Oxford University Press. Valdés, A., ed. 2000. "Agricultural Support Policies in Transition Economies." World Bank Technical Paper 470, World Bank, Washington, DC. World Bank. 2007. World Development Indicators Database. World Bank. http://go.worldbank.org/ B53SONGPA0 (accessed June 2008). 7 Latin America and the Caribbean Kym Anderson and Alberto Valdés* While the vast majority of the world's poorest households depend on farming for their livelihoods, poverty tends to be less focused on rural areas in Latin America than in Africa or South Asia, as Latin American countries generally have higher levels of development, larger nonfarm sectors in relation to their overall economies, more extensive urbanization, and a greater concentration of land ownership. Nonetheless, poverty is sufficiently prevalent in numerous parts of Latin America and the Caribbean that it continues to be a concern. In the past, farm earnings in the region have often been depressed by the pro-urban, antiagri- cultural bias of government policies. True, progress has been made over the past two decades in reducing the policy bias, but many trade-reducing price distor- tions remain between sectors, as well as within the agricultural sectors of most Latin American countries. This study of Latin America is based on a sample of eight countries, including the big four economies of Argentina, Brazil, Chile, and Mexico; Colombia and Ecuador, two of the poorest tropical South American countries; the Dominican Republic, the largest Caribbean economy; and Nicaragua, the poorest country in Central America. Together, in 2000­04, these countries accounted for 78 percent of the region's population, 80 percent of the region's agricultural value added, and 84 percent of the total gross domestic product (GDP) of Latin America. *The authors acknowledge invaluable help with data management from Ernesto Valenzuela; data compilation by Johanna Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, and Damiano Sandri; and help- ful comments from workshop participants. The working paper version of this chapter (Anderson and Valdés 2008b) contains additional background material and appendix tables. This chapter draws on the introductory and country chapters in Anderson and Valdés (2008a), with data updated using Anderson and Valenzuela (2008). 289 290 Distortions to Agricultural Incentives: A Global Perspective The key characteristics of these economies--which account for only 4.5 per- cent of worldwide GDP, but 7.7 percent of agricultural value added and more than 10 percent of agricultural and food exports--are shown in table 7.1. The table reveals the considerable diversity within the region in terms of stages of develop- ment, relative resource endowments, comparative advantages and, hence, trade specialization and the incidence of poverty and income inequality. This means that these countries represent a rich sample for comparative study. By compari- son, Nicaragua's per capita income is only one-seventh the global average, while the incomes of Colombia and Ecuador are one-third of this average. By contrast, the per capita income in Argentina and Chile averages just one-eighth below and in Mexico one-eighth above the global average. Only Argentina, Brazil, and Nicaragua are well above the global average in endowments of agricultural land per capita; the Dominican Republic and Ecuador are well below this average; and Chile, Colombia, and Mexico are a little less than one-third above the average. Income inequality is high throughout Latin America and the Caribbean compared with the rest of the world--the Gini coefficient is near or above 0.5 and averages 0.52. This is well above the Gini coefficient for Africa and Asia. Likewise, the Gini coefficient for land distribution is high in Latin America: 0.58 for Chile and more than 0.7 in Argentina, Brazil, Ecuador, and Nicaragua, compared with an average of less than 0.5 in Asia (World Bank 2007). Even so, there is comparatively little absolute poverty except in the poorest tropical parts of the region. Though it relies on nearly twice as much agricultural land per capita as the rest of the world, Latin American agriculture is characterized by concentrated land ownership and a structure of production whereby medium and large commercial farms contribute the bulk of agricultural output. It is also a region with a high degree of urbanization. These features are important in understanding the forces behind agricultural policies. So, too, is the fact that, until a few years ago, most countries in the region were experiencing a high degree of macroeconomic insta- bility and high inflation. The manipulation of food prices for urban consumers in an attempt to reduce inflation was (and, in Argentina, still is) a dominant feature driving farm pricing policy. Most Latin American countries have gone undergone major economy-wide policy reforms that began, for some countries, in the 1980s (or the 1970s for Chile) and, for others, in the mid-1990s. Reforms centered on macroeconomic stabilization, trade liberalization, deregulation, and some privatization of state agencies. There was a considerable reassessment of the role of government in guiding economic development. Agricultural policies were an integral part of this reform process, although not the principle motivation of the reforms. This chapter begins with a brief summary of economic growth and structural changes in the region since the 1960s and of agricultural and other economic Table 7.1. Key Economic and Trade Indicators, Latin American Countries, 2000­04 National relative to Share (%) of world: world (world 100) Gini Agricultural index Agricultural RCA,a trade for per Total Agricultural GDP land agriculture specialization Poverty capita Population GDP GDP per capita per capita and food indexb incidencec incomed Latin American focus countries 6.49 4.49 7.73 69 178 219 0.42 7 52 Argentina 0.61 0.54 1.04 89 426 541 0.85 5 51 Brazil 2.88 1.54 3.38 54 184 355 0.66 8 57 Chile 0.25 0.22 0.24 86 120 386 0.63 2 55 Colombia 0.70 0.24 0.77 35 132 264 0.25 7 59 Dominican Republic 0.14 0.06 0.18 41 54 474 0.29 3 52 Ecuador 0.20 0.07 0.16 33 80 487 0.59 16 44 Mexico 1.62 1.82 1.89 112 133 64 0.17 7 46 Nicaragua 0.08 0.01 0.06 14 169 952 0.26 44 43 Other Latin American countries 1.84 0.84 2.05 46 148 -- -- -- -- Caribbean 0.20 0.07 0.13 36 23 -- -- -- -- Central America 0.52 0.21 0.78 41 55 504 0.26 -- -- South America 1.12 0.56 1.13 50 213 157 0.16 13 -- All Latin American countries 8.33 5.33 9.78 64 171 -- -- -- -- Source: Sandri, Valenzuela and Anderson (2007), compiled mainly from World Bank (2008). Note: -- not available. a. RCA revealed comparative advantage. The RCA index is the share of agriculture and processed food in national exports as a ratio of that sector's share of global exports. b. Primary agricultural trade specialization index is net exports as a ratio of the sum of exports and imports of agricultural and processed food products (world average 0.0). 291 c. Percentage of the population living on less than US$1 per day. d. The poverty incidence and Gini index are for the most recent year available between 2000 and 2004, except for Ecuador where they refer to 1998. The weighted averages for the focus countries use population as the basis for weights. 292 Distortions to Agricultural Incentives: A Global Perspective policies as they affected agriculture before and after the reforms of the mid-1980s to mid-1990s. It then summarizes estimates of the nominal rate of assistance (NRA) and the relative rate of assistance (RRA) to farmers delivered by national farm and nonfarm policies over the past several decades, as well as the impact of these policies on the consumer prices of farm products. Both farmer assistance and consumer taxation tend to be negative in periods when there is an antiagri- cultural, pro-urban consumer bias in a country's policy regime. The final sections describe lessons learned and draw key policy implications for the region. Growth and Structural Changes1 Since 1980, real GDP in Latin America has grown at an average annual rate of 5.4 percent, or 3.6 percent per capita. These rates are somewhat above the averages of other developing countries of 4.1 percent total and 2.3 percent per capita, but somewhat below Asia's averages of 7.1 percent total and 5.5 percent per capita. The region's comparative growth performance was much less rosy in the 1960s and 1970s, however, before it moved away from an import-substitution industri- alization regime. Among the focus countries in our study, Chile and Mexico have been the star performers since 1980, while Ecuador and Nicaragua have been the slowest growers. Nicaragua's civil conflict set the country back in the 1980s, but its economy grew twice as fast as Ecuador's in the 1990s. Though the industrial sector has grown much more slowly than overall GDP over the past 25 years, agriculture has grown even more slowly, at barely half the rate of the rest of the economy, while the service sector has taken the lead. Among our sample countries, the economies of Chile and Mexico have been among the most rapidly growing, and Argentina's and Ecuador's the most slowly growing, apart from Nicaragua. As a result of strong growth in service activities during the past two decades, the share of services in GDP has risen from barely one-half to two-thirds, while agriculture's share fell from 9 to 6 percent, on average, in our sample economies. The relative decline of agriculture has been slowest in Argentina, Brazil, and Nicaragua and the most rapid in oil-exporting Ecuador and Mexico, but also in Chile. In 2000­04, agriculture's share of GDP ranged from 4 percent in Chile and Mexico to twice that in Brazil and Ecuador, three times that in Colombia and the Dominican Republic, and more than four times that in Nicaragua. The share of overall employment accounted for by farming activities in Latin America and Caribbean countries has fallen somewhat more slowly than agricul- ture's GDP share, according to statistics in the FAOSTAT Database of the Food and Agriculture Organization of the United Nations (which, because of defini- tional differences, is not always consistent with databases produced by countries). Latin America and the Caribbean 293 In general, the share remains at much higher level than the GDP share, implying relatively low and slow-growing labor productivity on farms. The most rapid decline has occurred in Brazil, where the employment share in agriculture has fallen from one-half to less than one-sixth during the past 40 years. Agriculture's average share in exports has also declined, by about one-third each decade since the late 1960s. The only exception is Chile, where the share has risen dramatically, from one-eighth to one-third. Chile contrasts markedly with the other rapidly growing economy in our sample, Mexico, where the share of farm products in all goods exports has fallen from 58 percent to only 6 percent. Additionally, the decline in relative importance of farm exports has been more rapid in Latin America than in the rest of the world: the index of the revealed comparative advantage of Latin America in these products (defined as the share of agriculture and processed food in national exports as a ratio of the share of such products in worldwide merchandise exports) has fallen by about one-third since the 1960s, as has the region's index of trade specialization (defined as net exports as a ratio of the sum of the imports and exports of agricultural and processed food products). There has been a marked upturn in these two indexes during the past decade, however, not only in Chile but in several other reforming Latin American countries, including Argentina and Brazil. The indexes are now at high levels in all countries in the sample apart from Mexico, which is also the only country in the sample with a revealed comparative disadvantage in agriculture. Finally, before examining the region's policy reforms, it is important to note the increases in export orientation. A common indicator is the value of goods and services expressed as a percentage of GDP. Since the early 1990s, this indicator has roughly doubled in the three largest economies (Argentina, Brazil, and Mexico) but changed little in the other countries in our sample, apart from Chile, where it rose a few years earlier. Another indicator is the share of primary agricultural pro- duction that is exported. This share has jumped dramatically in the past 20 years, including in Mexico, where it is now more than 30 percent as a result of sharply increased specialization within the sector following the agricultural and trade policy reforms begun in anticipation of the North American Free Trade Agreement (NAFTA), which came into effect in 1994. Note, however, that import dependence has also grown as a consequence of trade specialization. Indeed, 17 of the region's 21 countries on which data are available are net food importers (de Ferranti et al. 2005). Only Argentina was a net exporter of cereals during 2003­05, even though all eight countries in our sample (except Mexico) are more than 100 percent self- sufficient in agricultural products in aggregate and even though the share of these countries in global exports of agriculture and food jumped from 6.8 to 9.6 percent between 1990­94 and 2000­04. 294 Distortions to Agricultural Incentives: A Global Perspective The Evolution of Agricultural and Trade Policies Like most other regions, Latin American countries encompass a diverse range of policies, political structures, and institutions, though there has been, to some extent, a common evolution in the ideology motivating economic policies, begin- ning in the 1960s. Prior to the reforms of the mid-1980s and early 1990s Until approximately the mid-1980s, agricultural price interventions in the region were largely a by-product of a development strategy based on a claim that the best way to grow the economy was to adopt a protectionist policy to encourage import-substitution industrialization. This policy also raised budgetary resources in the form of import tax revenue, which was supplemented in some countries (such as Argentina) through agricultural export taxes. Both of these approaches harmed the region's most competitive farmers and were offset only slightly by farm credit and fertilizer subsidies. Between the 1950s and the 1980s, there were concerns about high rates of infla- tion, especially where urban populations had strong political influence. Policy makers were under pressure to avoid large increases in food prices, which would potentially impact wage rates and thereby (according to theory prevailing at the time) accelerate inflation through the so-called cost-push effect. In addition to fiscal and inflation objectives that made farm export taxes attractive, there was, in the 1950s and 1960s, a widespread belief among the region's policy makers and followers of the structuralist school associated with Prebisch (1950, 1959, 1964)--notwithstanding the seminal book by Schultz (1964)--that the efficiency losses generated through the extraction of rents in agriculture were low and that the main impact would be to reduce land rents and land values. Argentina is a prime example of a case in which the view persisted that farmers in Latin America were unresponsive to price incentives. While the belief in this unresponsiveness has now largely disappeared, a few countries-- Argentina is one--still tax agricultural exports to generate fiscal revenues and lower consumer food prices. An empirical study of agricultural pricing policies led by Krueger, Schiff, and Valdés (1991) included five Latin American countries for the period 1960­84. Its main findings are fourfold. First, over the period examined and for the farm prod- ucts selected, the direct interventions affecting importables were positive, on aver- age, while the direct interventions on exportables were negative. Second, aggregating over all selected products, one sees that the net effect was negative, indicating that the direct tax on exportables dominated the protection on importables. Third, the Latin America and the Caribbean 295 rate of indirect taxation on agriculture (because of industrial protection policies and the overvaluation of the real exchange rate) was large and dominated the rate of direct taxation. And fourth, direct price policies stabilized agricultural prices relative to world prices, while indirect policies contributed little, if at all, to food price stability. The study found that direct protection for agricultural importables averaged 13 percent, while for exportables, it amounted to ­6 percent. The indi- rect taxation rate in the region averaged 21 percent so that the total taxation rate (direct and indirect) averaged 28 percent. The highest direct taxation was found in Argentina and the Dominican Republic (about 18 percent). As a percent of agri- cultural GDP, net income transfers out of agriculture (direct and indirect) reached 84 percent in Argentina, 56 percent in Chile, 43 percent in the Dominican Republic, and 42 percent in Colombia. Economic reforms from the mid-1980s to early 1990s By the 1980s, there was disillusionment with the results of the import-substitution strategy and wider acceptance of theoretical developments regarding the causes of inflation and macroeconomic instability in general. During the 1980s and early 1990s, a macroeconomic framework designed for open economies gradually dis- placed the closed economy approach in most Latin American countries. Govern- ments introduced economy-wide reforms with special emphasis on macroeco- nomic stabilization, deregulation, unilateral trade liberalization, and privatization. The goal of the reformers was to create a better climate for productivity and private investment in all economic sectors, including agriculture. In most Latin American countries, the major change in trade policy was the partial or total removal of most quantitative restrictions on imports and exports, the elimination of export taxes, and a program of gradual reduction in the levels of import tariffs. This yielded incentives to move resources from import-competing to export- oriented sectors, including in agriculture, which enhanced competitiveness and led to greater integration with the world economy. By the mid-1990s, the exchange rate was recognized as the most important "price" affecting the agricultural economy. At the outset of the reforms, it was expected that trade liberalization and the reduction of the fiscal deficit would lead to a depreciation of the real exchange rate (Krueger, Schiff, and Valdés 1988). Yet the reforms were followed by a significant appreciation of the currency that was associated with the opening of the capital account, greater inward foreign invest- ment, and a major increase in domestic real interest rates. Reforms in the service sector also played a critical role. Deregulation and privatization had a major impact on the availability in the marketplace of the more-reliable and lower-cost services used in agriculture, such as ports, airlines, and shipping transport. 296 Distortions to Agricultural Incentives: A Global Perspective The timing of reforms differed somewhat across countries, however. Colombia, for example, became a more open economy through export promotion beginning in 1967; it adopted a more ambitious liberalization of trade in 1990 and then went into a policy reform reversal beginning in 1992. In Chile, the controlled markets of 1950 to 1974 were followed by radical economic reforms toward trade liberaliza- tion, deregulation, and privatization between 1978 and 1982, before a second phase of reforms beginning in 1984. Mexico, meanwhile, introduced strong policy changes starting in the mid-1980s, before the signing of NAFTA. The changes involved more openness, deregulation, and privatization; a reduction in credit subsidies; and major changes in the role of government in the marketing of farm products. A wide variety of policy instruments have been applied to influence agricul- tural prices, even during the postreform period. Colombia, for example, has had minimum support prices, in addition to import tariffs, price compensation schemes, procurement agreements, a monopoly on grain imports by a govern- ment agency, export licenses and subsidies, and safeguards on imports. Moreover, until 1990, all imports of inputs were subject to prior import licenses. Then, in 1995, tariffs and tariff surcharges associated with price bands on more than 100 products were introduced. Mexico is another leader in interventions, includ- ing in the transition from highly government-controlled markets before the mid- 1980s to more market-oriented policies. Its policies include price support pro- grams (before the mid-1980s and in conjunction with state trading), credit and input subsidies, and direct income payments to farmers (ProCampo). Argentina, however, has undertaken simpler interventions. Agricultural exportables that are also wage goods have been subjected to export taxes, complemented by export bans in some years. The return to sizeable export taxes in late 2001 and their sub- sequent rises has been controversial, with the most recent rises leading to pro- longed protests by farmers in urban areas in mid-2008. Estimates of Latin American Policy Indicators The net effect of these various interventions on farmer and consumer incentives are quantified using the common methodology (Anderson et al. 2008a, 2008b) that has been adopted by the authors of this volume and the four preceding regional volumes. After a brief word on methodology, a summary of results follows.2 Methodology The NRA is defined as the percentage by which government policies have raised gross returns to producers above what they would be without the government's Latin America and the Caribbean 297 intervention (or lowered them, if the NRA is below zero). If a trade measure is the sole source of government intervention, the measured NRA will also be the con- sumer tax equivalent (CTE) rate at that same point in the value chain. The NRAs are based on estimates of assistance to individual industries at farmgate. The tar- geted degree of coverage of the products for which agricultural NRA estimates are generated is 70 percent of the gross value of farm production at undistorted prices. The authors of the country case studies also provide guesstimates of the NRAs for noncovered farm products. For countries with non-product-specific (NPS) agricultural subsidies or taxes, such net subsidies are then added to product-specific assistance to obtain NRAs for total agriculture and also for trad- able agriculture for use in generating an RRA. Farmers are affected not only by the prices of their own outputs, but also (albeit indirectly, because of the changes to factor market prices and the exchange rate) by the incentives nonagricultural producers face. In other words, both absolute and relative prices--and hence, relative rates of government assistance-- affect producer incentives. The direction of the economy-wide effect of distor- tions to agricultural incentives may be captured by the extent to which the trad- able parts of agricultural production are assisted or taxed relative to producers of other tradables. By generating estimates of the average NRA for nonagricultural tradables, it is then possible to calculate an RRA, which is defined in percentage terms as: RRA 100[(1 NRAagt 100) (1 NRAnonagt 100) 1], where t t NRAag and NRAnonag are the weighted average percentage NRAs for the trad- able parts of the agricultural and nonagricultural sectors, respectively. Since the NRA cannot be less than 100 percent if producers are to earn anything, neither can the RRA. If both sectors are assisted equally, the RRA is zero. Although this measure cannot fully capture the ultimate impacts on resource allocations to various sectors including nontradables (a computable general equilibrium model is needed for that), it is nonetheless useful in comparing policy biases across time and countries. If the RRA is below (above) zero, it indicates that a country's trade policy regime has an antiagricultural or proagricultural bias. In calculating the NRA for producers of agricultural and nonagricultural trad- ables, the methodology seeks to include distortions generated by dual or multiple exchange rates. Such direct interventions in the market for foreign currency were common in Latin America in the 1970s and 1980s, but not since the reforms. However, some authors of the Latin American country studies have had difficulty finding an appropriate estimate of the extent of this distortion, so the impact on NRAs has been included only for the Dominican Republic, Ecuador, and Nicaragua. Its exclusion for the other five countries means the estimated (typi- cally) positive NRAs for importables and (typically) negative NRAs for exporta- bles are smaller than they should be for these countries. In cases where the NRA 298 Distortions to Agricultural Incentives: A Global Perspective for importables dominates that for exportables, this omission would lead to an underestimate of the average (positive) NRA for such tradables sectors. This applies to nonagricultural sectors for all the countries studied in this chapter. In the most common cases in earlier decades, where the estimated NRA for importa- bles is dominated by a negative NRA for exportables in the farm sector, the esti- mate of the sectoral average NRA for agriculture would be less negative than it should be, and, hence, so would the RRA estimate.3 To obtain the values for farmer assistance and consumer taxation, the NRA estimates of the country authors have been multiplied by the gross value of pro- duction at undistorted prices to obtain an estimate in constant U.S. dollars of the direct gross subsidy equivalent (GSE) of assistance to farmers. This GSE value is then added up across products for each country and then across countries for any or all products to obtain regional aggregate transfer estimates for the countries being examined. An aggregate estimate for the rest of the region is obtained by assuming that the weighted average NRA for the countries not within the study is the same as the weighted average NRA for the countries within the study and that the share of each country in the region's gross value of farm production at undistorted prices is the same each year as the share of the country in the region's agricultural GDP measured at distorted prices. These GSE values are also expressed on a per farm worker basis. To obtain comparable value estimates of the consumer transfer, the CTE esti- mate at the point at which a product is first traded is multiplied by consumption (obtained from the FAO SUA-FBS Database), valued at undistorted prices, to obtain an estimate in constant U.S. dollars of the tax equivalent to consumers of primary farm products. These, too, are summed across products for a country and across countries for any or all products to obtain regional aggregate transfer esti- mates for the countries under study. Estimates of NRAs in agriculture On average (whether simple or weighted), agricultural price and trade policies in Latin America reduced farmer earnings throughout the postwar period right through to the 1980s. The extent of these (when expressed as a nominal tax equiv- alent) peaked at more than 20 percent in the 1970s but still averaged close to 10 percent in the late 1980s. The only countries in our sample that received posi- tive assistance from farm policies during that period were Chile (from the late 1970s, but only to a minor extent), Colombia, and Mexico. Argentina, Brazil, the Dominican Republic, and Ecuador each had negative rates of assistance that aver- aged well above 20 percent for at least one five-year subperiod, and, apart from the Dominican Republic, each had a negative average NRA even in the 1990s, as did Latin America and the Caribbean 299 Nicaragua. However, by the mid-1990s, Brazil and the Dominican Republic joined Chile and Colombia in that they had positive average NRAs. Meanwhile, Mexico had raised its assistance considerably before engaging in reform following negoti- ations to join the World Trade Organization and the NAFTA, while Argentina had all but eliminated its discrimination against its exporters in the 1990s, only to reinstate explicit export taxes again in late 2001 when it abandoned its fixed exchange rate with the U.S. dollar and nominally devalued by two-thirds. The NRAag for the region in the 1990s and the first half of the present decade averaged slightly under 5 percent (table 7.2). Its switch from negative to positive occurred in 1992. The effect of the policy reforms on NRAs over the past two decades is illus- trated in figure 7.1. For all countries except Chile, the national average NRA was less negative or more positive in 2000­04 than in 1980­84. This is true, too, for the majority of the commodity NRAs for the region, although assistance for several commodities (such as milk and poultry) was cut. This pattern may be seen in figure 7.2 and table 7.3, which also illustrate the diversity of the region's average rates across commodities. There is also a great deal of diversity across commodities within each country's farm sector. The extent of this diversity (as measured by the standard deviation) diminished, on average, by only about one-quarter during 1990­2004 compared with the prereform period of 1965­89. This is evident in table 7.4. The table reports the standard deviation of NRAs for covered products, which account for more than two-thirds of the value of agricultural production. This means there is still a great deal that may be gained in terms of improved resource reallocation within the agricultural sector if differences in rates of assistance for different industries are reduced. One striking feature of the pattern of farm price distortions in Latin America and the Caribbean as a whole is the strong antitrade bias. This is shown for agri- culture's import-competing and export subsectors in the region in figure 7.3 and for each country in table 7.5 (along with a trade bias index, or TBI). These esti- mates reveal that there has been little decrease in the bias over the past four decades, except in Brazil. Indeed, the average NRA for exportable farm products has been negative throughout virtually the whole period analyzed in all countries other than Chile (plus Brazil in the 1990s and Colombia since 2000), while the regional average NRA for import-competing farm industries has increased from near zero in the 1970s to 20 percent or more in the period since 1990 (with Chile again an exception with its NRA for import-competing industries falling to near zero). That is, despite the lower taxation of farm export industries, the region's antitrade bias has persisted because the average NRA for import-competing farm products has been rising recently in several of the countries being examined. 300 Table 7.2. NRAs to Agriculture,a Latin American Countries, 1965­2004 (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Argentina 22.7 22.9 20.4 19.3 15.8 7.0 4.0 14.9 Brazilc 6.1 27.3 23.3 25.7 21.1 11.3 8.0 4.1 Chile 16.2 12.0 4.5 7.2 13.0 7.9 8.2 5.8 Colombia 4.7 14.8 13.0 5.0 0.2 8.2 13.2 25.9 Dominican Republic 5.0 17.5 21.2 30.7 36.4 1.0 9.2 2.5 Ecuadorc 9.6 22.4 15.0 5.9 1.0 5.3 2.0 10.1 Mexico -- -- -- 2.9 3.0 30.8 4.2 11.6 Nicaraguac -- -- -- -- -- 3.2 11.3 4.2 Latin American focus countries Unweighted averageb 2.8 15.5 14.5 7.7 8.3 2.3 3.2 4.9 Weighted averagea 7.2 21.0 18.0 12.5 10.9 4.2 5.5 4.8 Dispersion of individual country average NRAsd 13.8 15.4 10.8 17.4 17.1 13.5 8.6 11.9 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). Note: -- not available. a. Weighted average for each country, including product-specific input distortions and NPS assistance as well as authors' guesstimates for noncovered farm products, with weights based on gross value of agricultural production at undistorted prices. b. The unweighted average is the simple average across the eight countries of their national NRA (weighted) averages. c. Ecuador and Brazil in the 1965­69 column refer to 1966­69 data; and Nicaragua 1990­94 column to 1991­94 data. d. Dispersion of average NRAs across countries is a simple five-year average of the annual standard deviation around a weighted mean of the national agricultural sector NRA each year. Latin America and the Caribbean 301 Figure 7.1. NRAs to Agriculture, Individual Latin American Countriesa and Unweighted Regional Average, 1980­84 and 2000­04 40 20 percent 0 20 40 a o r ile av a il bl n ua a do az bi d ric in ic pu ica Ch ag e) N c ex nt m Br ua te e i in gh Am ar lo ge M Ec om ic Co Ar ei in Re D nw at (u L 1980­84 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). a. There are no estimates for Nicaragua for 1980­84. The contributions to the overall NRA for agriculture for the region as a whole provided by covered products, noncovered products, and NPS assistance are sum- marized in table 7.5. NPS assistance has added only one or two percentage points during the past four decades. Input price distortions have also contributed little, on average, to the overall regional NRA in agriculture, reducing the negative value slightly in the 1980s and adding slightly to the positive value during the past decade or so. In Chile, input distortions have reduced the positive NRA in the farm sector because of protectionist policies that have raised the price of imported or import-competing farm inputs. This has also been the case in Argentina since the early 1990s and, to a smaller extent, in Colombia since the 1960s. There is little in the way of domestic producer subsidies or taxes, on average, in the region; the main exception is positive support measures in Mexico and slightly negative support measures in Argentina. The dollar value of the positive or negative assistance to farmers arising from agricultural price and trade policies has been nontrivial. The antiagricultural bias Figure 7.2. NRAs, by Product, Latin American Countries, 1980­84 and 2000­04 milk rice sugar poultry cotton pig meat coffee wheat beef maize cocoa soybeans eggs 80 60 40 20 0 20 40 60 80 100 120 percent, weighteda average across countries 1980­84 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). a. Weights based on gross value of agricultural production at undistorted prices--that is, each NRA (by country, by product) is weighted by the country's value of production of that commodity in a given year. Products with less than 1 percent of the gross value of regional production are excluded. These include: apples, cassava, cocoa, garlic, onions, palm oil, peanuts, and sesame. Figure 7.3. NRAs to Exportable, Import-Competing, and Alla Agricultural Products, Latin American Region, 1965­2004 30 percent, weighted averages 20 across eight countries 10 0 10 20 30 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 import-competing total exportables Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). Table 7.3. NRAs, Key Covered Farm Products, Latin American Focus Countries,a 1965­2004 (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Rice 27 5 10 5 8 12 26 34 Wheat 2 15 11 6 6 18 3 2 Maize 10 8 15 5 13 0 4 3 Other grains 4 3 4 6 2 0 14 11 Soybeans 3 5 15 11 21 10 4 10 Other oilseeds 4 3 15 21 23 11 16 21 Sugar 17 61 46 54 43 20 7 27 Cotton 7 2 14 16 23 12 6 11 Coffee 27 26 32 42 29 1 9 3 Cocoa 6 16 13 4 14 16 12 7 Fruits and vegetables 12 22 31 5 33 16 24 20 Beef 23 21 11 10 4 2 5 1 Pig meat 6 14 13 19 20 6 3 4 Poultry 110 144 108 33 23 23 8 19 Eggs -- -- -- 0 6 2 16 16 Milk 2 7 19 104 70 45 29 45 All covered products 13.0 25.1 19.6 14.6 14.3 0.9 0.8 2.7 Sources: Anderson and Valenzuela (2008) based on estimates reported in Anderson and Valdés (2008a). Note: -- not available. 303 304 Table 7.4. Dispersion of NRAs across Covered Agricultural Productsa within Latin American Focus Countries, 1965­2004b (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Argentina 18.5 17.8 19.9 15.7 12.1 7.1 9.4 12.6 Brazil 28.1 37.2 41.0 35.9 25.5 27.4 8.5 7.6 Chile 33.0 37.2 30.4 17.0 26.1 16.5 14.7 13.3 Colombia 34.8 21.2 29.9 42.5 34.1 27.2 31.0 46.0 Dominican Republic 86.5 64.0 89.3 83.0 102.3 137.1 92.6 132.8 Ecuador 99.0 88.6 104.8 106.2 48.5 18.8 27.9 29.6 Mexico -- -- -- 71.9 60.1 57.7 30.6 41.1 Nicaragua -- -- -- -- -- 40.1 35.7 27.7 Latin American focus countries Unweighted averagec 50.0 44.3 52.5 53.2 44.1 41.5 31.3 38.8 Product coveraged 54 65 68 71 68 65 69 70 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). Note: -- not available. a. Dispersion for each country is a simple five-year average of the annual standard deviation around a weighted mean of NRAs across covered products each year. b. The 1965­69 column for Ecuador and Brazil refers to 1966­69 data; the 1990­94 Nicaragua column to 1991­94 data. c. The unweighted average is the simple average across the eight countries of their five-year simple average dispersion measures. d. Share of gross value of total agricultural production at undistorted prices accounted for by covered products in the region. Table 7.5. NRAs to Agricultural Relative to Nonagricultural Industries, Latin American Region, 1965­2004 a. Unweighted averages for eight focus countries (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 a NRA, covered products 9.1 21.8 17.0 8.8 8.9 1.0 1.1 4.4 NRA, noncovered products 0.5 9.2 10.0 6.5 7.5 1.4 0.9 0.4 NRA, all agricultural productsa 5.4 17.0 15.0 8.3 9.3 0.4 0.7 2.7 Total agricultural NRA (including NPS)b 2.8 15.5 14.5 7.7 8.3 2.3 3.2 4.9 TBIc 0.22 0.18 0.31 0.41 0.33 0.26 0.25 0.26 NRA, all agricultural tradablesb 6.0 19.0 16.4 7.2 8.2 2.6 3.5 5.7 NRA, all nonagricultural tradables 16.8 20.6 15.6 14.3 13.4 7.7 7.3 6.5 RRAd 19.5 32.9 27.7 18.8 19.1 4.8 3.5 0.8 b. Weighted averages for eight focus countries (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 NRA, covered productsa 13.0 25.1 19.6 14.6 14.3 0.9 0.8 2.7 NRA, noncovered products 3.3 15.5 15.0 10.9 13.1 0.7 3.8 2.1 NRA, all agricultural productsa 8.6 21.7 18.1 13.6 14.0 0.8 1.7 2.5 Total agricultural NRA (including NPS)b 7.2 21.0 18.0 12.5 10.9 4.2 5.5 4.8 TBIc 0.20 0.25 0.26 0.36 0.29 0.25 0.14 0.21 NRA, all agricultural tradablesb 9.3 23.0 19.0 12.9 11.2 4.4 5.5 4.9 NRA, all nonagricultural tradables 15.9 27.8 23.3 18.5 16.8 7.3 6.6 5.5 RRAd 21.4 39.8 34.2 26.6 24.0 2.7 1.0 0.6 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). a. NRAs including product-specific input subsidies. b. NRAs including NPS assistance, that is, the assistance to all primary factors and intermediate inputs as a percentage of the total primary agricultural production valued at undistorted prices. c. TBI (1 NRAagx 100)/(1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricul- 305 tural sector. The regional average TBI is calculated from the regional averages of the NRAs for exportable and import-competing parts of the agricultural sector. d. RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. 306 Distortions to Agricultural Incentives: A Global Perspective peaked for the region in the 1975­84 period at nearly US$17 billion per year in constant 2000 dollar terms, assuming that the Latin American countries not under study had the same NRAs as the countries under study, keeping aside the case of Mexico (see the bottom row of table 7.6a). This is equivalent to a gross tax of almost US$400 for each person engaged in agriculture. Around 60 percent of this US$17 billion arose because of policies in Brazil. Thanks to the reforms of the past two decades, this taxation has gradually disappeared in all the countries under study except Argentina and Nicaragua. However, the reform does not mean that there is now no intervention. Rather, the old policy has been replaced by pos- itive assistance to farmers in the remaining six countries. This assistance has aver- aged US$6 billion per year, or around US$140 per farmworker, over the 1995­ 2004 period. The US$140 is small compared with per capita income for the region (about 4 percent), but it ranges from more than US$450 in Colombia (one-quar- ter of that country's per capita GDP in 2000­04) to US$1,700 in Argentina (negative one-third of that country's per capita GDP). The extent of this dramatic transformation in the region as a whole over the past two decades is illustrated in figure 7.4 for the individual countries and for key products. Table 7.7 reveals that, as in most other regions of the world, the lion's share of assistance goes to milk, sugar, and rice. Assistance to nonfarm sectors and RRAs The antiagricultural policy bias of the past was caused not merely by agricultural policies. The significant reduction in border protection for the manufacturing sector and the indirect impact of this on the drop in the price of nontradables after the initiation of the reforms, together with the deregulation and privatization of services, have also been important in the changes in the incentives affecting intersectorally mobile resources. The reduction in assistance to nonfarm tradable sectors has been as responsible for the expansion in agricultural exports since the early 1990s as the reduction in direct taxation on these agricultural exports. Quantifying this distortion in nonfarm tradable sectors as accurately as the quantification of the distortion in agriculture has not been possible. Our authors have had to rely on applied trade taxes (for exports, as well as imports) rather than undertaking price comparisons for nonfarm goods, and, hence, they have not cap- tured the quantitative restrictions on trade that were important in earlier decades but that have been less important recently.4 Nor have they captured distortions in the services sectors; many of these sectors now produce tradables (or would do so in the absence of interventions preventing the emergence of this production). As a result, the NRAs for nonfarm importables are underestimated, and the decline indicated is less rapid than the decline that actually occurred; the situation is Latin America and the Caribbean 307 Figure 7.4. GSEs of Assistance to Farmers, Latin American Countries, 1980­84 and 2000­04 a. Total per country 4,000 2,000 constant 2000 US$ millions 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 a ua bl n ile r zil a o do in bi ic pu ica a Ch ag ic ex nt m Br ua Re in ar lo ge M Ec om ic Co Ar N D b. Total per product milk poultry sugar rice pig meat cotton coffee wheat cocoa other grains other oilseeds beef eggs maize fruits and vegetables soybeans 0 0 0 00 0 0 00 0 0 00 00 00 00 00 00 00 0 0 0 1, 2, 3, 6, 5, 4, 3, 2, 1, constant 2000 US$ millions 1980­84 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). Table 7.6. GSEs of Assistance to Farmers, Total and Per Farm Worker, Latin American Countries,a 1965­2004 308 a. Total (constant 2000 US$ millions) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Argentina 1,699 2,630 2,466 2,850 1,533 738 595 2,473 Brazil 790 7,905 8,141 12,724 9,142 3,578 3,101 1,509 Chile 482 378 167 267 394 380 465 294 Colombia 358 1,555 1,719 583 5 905 1,562 1,835 Dominican Republic 61 457 603 694 561 22 150 39 Ecuador 192 477 453 121 23 132 68 324 Mexico -- -- 389 1,581 762 7,426 984 2,805 Nicaragua -- -- -- -- -- 32 140 54 Latin American focus countries 2,496 12,647 13,604 13,716 10,098 4,210 5,459 4,279 All Latin American countriesa 3,082 15,613 16,794 16,933 12,467 5,197 6,740 5,283 b. Per person engaged in agriculture (constant 2000 US$) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Argentina 1,094 1,776 1,727 2,030 1,054 498 404 1,693 Brazil 51 482 475 736 561 240 224 118 Chile 650 515 216 324 442 401 478 299 Colombia 119 483 483 153 1 244 419 496 Dominican Republic 88 641 859 1,003 803 33 238 66 Ecuador 199 475 446 114 20 108 54 260 Mexico -- -- 51 194 90 867 115 329 Nicaragua -- -- -- -- -- 81 351 137 Latin American focus countries 85 411 417 408 305 132 177 144 All Latin American countriesa 81 390 396 386 283 119 156 124 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). Note: -- not available. a. Assumes the rate of assistance in nonfocus countries is the same as the average for the focus Latin American countries excluding Mexico, and that their share of the value of Latin American and Caribbean (excluding Mexican) agricultural production at undistorted prices is the same as their average share of the region's agricultural GDP at distorted prices during 1990­2004, which was 23 percent. Farmer numbers are from FAOSTAT and may differ from national statistics. Table 7.7. GSEs of Policies Affecting Farmers in Latin America, by Product and Subsector, 1965­2004 a. By product (at undistorted farmgate prices, US$ millions) Rice Wheat Maize Other grains Soybean Other oilseeds Sugar Cotton 1965­69 24 17 92 0 1 0 8 19 1970­74 40 216 162 1 55 0 1,829 8 1975­79 230 91 475 56 436 81 1,619 159 1980­84 55 116 396 53 428 110 3,260 156 1985­89 55 65 707 10 1,533 151 1,980 380 1990­94 201 395 17 5 386 92 988 158 1995­99 569 79 373 151 279 256 233 36 2000­04 614 30 307 113 1,371 241 970 78 Fruits and All covered Cocoa Coffee vegetables Beef Pig meat Poultry Eggs Milk products 1965­69 1 127 19 289 1 10 -- 2 516 1970­74 8 169 41 440 4 15 -- 29 2,987 1975­79 32 815 163 404 53 116 51 236 4,131 1980­84 8 3,014 165 1,027 565 423 14 1,603 7,003 1985­89 17 1,738 623 327 504 344 66 944 6,716 1990­94 14 30 610 188 93 533 19 1,471 661 1995­99 10 536 977 704 110 378 225 1,393 476 2000­04 7 76 750 264 111 1,048 285 1,915 1,504 (Table continues on the following page.) 309 310 Table 7.7. GSEs of Policies Affecting Farmers in Latin America, by Product and Subsector, 1965­2004 (continued) b. By subsector (at undistorted farmgate prices, US$ billions) Total GSE, all direct assistance to farmersa GSE for just covered GSE for just non- Import- farm productsb covered farm products Total Exportables competing Nontradables 1965­69 0.5 0.1 0.6 0.7 0.1 0.0 1970­74 3.0 1.1 4.0 3.9 0.2 0.0 1975­79 4.0 1.5 5.5 5.5 0.0 0.0 1980­84 7.0 2.2 8.5 12.1 2.9 0.0 1985­89 6.7 3.1 7.5 10.7 0.9 0.0 1990­94 0.7 0.4 3.8 4.6 5.7 0.0 1995­99 0.5 1.2 5.3 2.3 3.9 0.0 2000­04 1.5 0.6 4.3 3.3 5.4 0.0 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). Note: -- not available. a. GSEs including assistance to nontradables and NPS assistance. b. GSEs including product-specific input subsidies. Latin America and the Caribbean 311 similar for nonfarm exportables, except that the actual NRAs would have been negative in most cases. Of these two elements of underestimation, the former bias probably dominated. Thus, the author estimations of the overall NRA for nonagricultural tradables should be considered a lower-bound estimate; this is especially true as one looks back in time, so that the decline indicated in the NRA estimates is less rapid than it actually would have been.5 Despite these methodological limitations, the estimated NRAs for nonfarm tradables prior to the 1990s are sizeable. For Latin America as a whole, the average value of the NRAs for nonfarm tradables has steadily declined throughout the past four decades as policy reforms have spread. This has therefore contributed to a decline in the estimated RRA among farmers. Thus, the RRA has fallen from below 30 percent in the 1970s to an average of 1 percent in 2000­04 (see table 7.5), and this appears (in figure 7.5) to have been caused as much by falling positive NRAs among nonfarm producers as by falling negative NRAs among farmers. The extent of the change in RRAs among individual countries over the past two decades is striking, particularly in the case of Brazil and the Dominican Republic (where negative RRAs virtually disappeared) and of Colombia (where RRAs switched from negative to positive). In figure 7.6, this is depicted by Figure 7.5. NRAs to Agricultural and Nonagricultural Tradable Products and RRA,a Latin American Region, 1965­2004 40 30 percent, weighted averages across eight countries 20 10 0 10 20 30 40 50 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Anderson and Valenzuela (2008) based on estimates reported in Anderson and Valdés (2008a). a. The RRA is defined as 100*[(100 NRAagt ) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. 312 Distortions to Agricultural Incentives: A Global Perspective Figure 7.6. Relationship between RRA and the TBI for Agriculture, Latin American Focus Countries, 1980­84a and 2000­04 a. 1980­84 0.3 0.2 0.1 Ecuador Chile 0 Mexico RRA Nicaragua 0.1 Colombia 0.2 Argentina 0.3 Dominican Republic 0.4 Brazil 0.5 0.6 0.5 0.4 0.3 0.2 0.1 0 TBI b. 2000­04 0.3 Colombia 0.2 0.1 Mexico Chile Ecuador 0 Dominican Republic Brazil RRA Nicaragua 0.1 Argentina 0.2 0.3 0.4 0.5 0.6 0.5 0.4 0.3 0.2 0.1 0 TBI Sources: Anderson and Valenzuela (2008) based on estimates reported in Anderson and Valdés (2008a). a. Nicaragua data refer to 1991­94 in the first period. Latin America and the Caribbean 313 countries being closer to the horizontal line in the middle of the figure (where RRA 0) in 2000­04 than in 1980­84. That figure also shows some movement to the right by countries over that period, indicating the extent to which their anti- trade bias within the farm sector has diminished. Were countries to have elimi- nated both their antiagricultural and antitrade policy biases, they would be located on the right-hand crossover of the RRA 0 and TBI 0 axes. Unfortu- nately, only Chile and Brazil were close to that point by 2004. The CTEs of agricultural policies The extent to which farm policies impact the retail consumer price of food and the price of livestock feedstuffs depends on a wide range of factors, including the degree of processing undertaken and the extent of competition along the value chain. Here, an attempt is made only to examine the importance of the impact of policies on the buyer's price at the level where the farm product is first traded internationally and, hence, where price comparisons are made (for example, for wheat, raw sugar, or beef).6 To obtain weights to make it possible to sum up across commodities and countries, the volume of apparent consumption is calculated simply as production plus net imports and then valued at undistorted prices. If there were no farm input distortions and no domestic output price distor- tions, such that the NRA was entirely the result of border measures such as an import or export tax, then the CTE would equal the NRA for each covered product. Because domestic distortions are relatively minor in Latin America and because the NRA tends to be positive for import-competing products and nega- tive for exportables (until recently), this is the case for the CTE as well. The weighted average CTE for the region has thus been negative for most of the period, averaging around 15 percent until the 1990s and marginally above zero thereafter (table 7.8a). Though the variance across products is somewhat less now than before the reforms of the past two decades, it is still considerable (table 7.8b). In proportional terms, the current transfers from consumers are largest in Colombia and Ecuador, but in dollar terms they are also large in Mexico. At its peak in the 1980s, the transfer from producers to consumers in the region amounted to US$7 billion per year at the producer level for the products covered in this project, whereas, in the present decade, the average transfer occurs from consumers to producers, while the total reaches around US$6 billion per year (table 7.9a). Among the covered products, the biggest transfers are for milk, poultry, sugar, and rice (table 7.9b). Even if one were to take account also of the assistance for non- covered products, the total per capita transfer from consumers in recent years would amount to less than US$15. 314 Table 7.8. Percentage CTE of Policies Affecting Covered Farm Products,a Latin American Countries, 1965­2003 (percent, at primary product level ) a. Aggregate CTEs, by country 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­03 Argentina 27.6 27.2 25.2 23.4 16.6 5.7 0.0 9.1 Brazil 2.1 25.4 19.8 25.8 26.5 23.1 2.1 1.3 Chile 7.1 1.5 2.8 9.0 23.8 18.1 14.2 10.7 Colombia 7.2 13.4 5.3 27.4 20.8 16.2 33.9 49.7 Dominican Republic 12.9 7.1 7.7 27.8 31.4 7.8 16.6 3.5 Ecuador 10.5 25.7 3.9 35.0 17.4 3.3 4.6 18.5 Mexico -- -- -- 1.3 0.8 22.3 1.9 9.9 Nicaragua -- -- -- -- -- 10.5 10.6 9.0 Latin American focus countries: Unweighted average 0.8 16.2 8.8 1.0 1.7 4.8 9.5 11.4 Weighted averageb 4.7 22.1 16.2 13.4 12.3 2.7 1.4 5.1 Dispersion of national CTEsc 15.5 13.4 14.5 29.2 26.0 17.4 15.0 18.8 b. Regional CTEs, by product 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­03 Rice 30 8 10 0 6 6 19 30 Wheat 17 0 32 19 8 22 8 13 Maize 9 4 13 11 14 4 8 4 Other grains 0 0 6 6 5 3 15 14 Soybean 4 5 15 13 19 10 5 9 Other oilseeds 0 0 24 22 22 10 8 17 Sugar 28 60 44 54 41 18 8 27 Cotton 6 1 14 24 23 23 7 7 Coffee 25 26 32 52 34 7 10 4 Cocoa 6 16 13 4 16 16 12 7 Fruits and vegetables 8 10 12 1 30 16 22 17 Beef 27 23 14 11 6 11 4 1 Pig meat 6 14 14 26 26 3 3 4 Poultry 110 132 98 26 18 17 7 21 Egg -- -- 10 0 6 2 16 17 Milk 5 3 18 70 54 38 28 44 Latin American focus countries: Weighted averageb 4.7 22.1 16.2 13.4 12.3 2.7 1.4 5.1 Dispersion of regional product CTEsd 35.2 46.4 34.6 30.4 23.5 16.3 13.8 18.6 Source: Anderson and Valenzuela (2008), based on estimates reported in Anderson and Valdés (2008a). Note: -- not available. a. Assumes the CTE is the same as the NRA derived from trade measures (that is, not including any input taxes/subsidies or domestic producer price subsidies/taxes). b. Weights are consumption valued at undistorted prices, where consumption (from FAO) is production plus imports net of exports plus change in stocks of the covered products. 315 c. Simple five-year average of the annual standard deviation around a weighted mean of the national average CTE. d. Simple five-year average of the annual standard deviation around a weighted mean of the regional average CTE for the covered products shown above. 316 Table 7.9. Value of CTE of Policies Affecting Covered Farm Products, Latin American Countries, 1965­2003 (constant 2000 US$ millions, at primary product level) a. Aggregate CTEs, by country 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­03 Argentina 993 1,367 1,442 1,696 903 321 3 748 Brazil 18 3,097 3,657 7,420 5,849 5,548 133 43 Chile 45 214 71 176 308 318 303 180 Colombia 208 566 4 1,204 640 622 1218 1160 Dominican Republic 45 24 27 46 93 85 96 44 Ecuador 104 276 20 309 134 42 75 350 Mexico -- -- -- 1,358 685 16,619 2712 4,965 Nicaragua -- -- -- -- -- 22 10 20 Latin American focus countries 871 5,545 5,038 8,831 5,078 11,755 4,276 6,013 All Latin American countriesa 1,054 6,846 6,219 10,902 6,269 14,507 5,279 5,938 b. Regional CTEs, by productb 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­03 Rice 116 79 538 371 145 156 563 535 Wheat 260 337 1,085 1,088 65 120 7 27 Maize 272 262 1,012 1,324 1,360 695 528 543 Other grains 1 3 117 128 44 99 11 28 Soybean 4 184 1,057 906 1,151 1,035 240 460 Other oilseeds 0 1 150 157 152 51 74 73 Sugar 29 3,320 2,540 3,892 2,009 9,666 2,092 287 Cotton 61 12 356 444 327 317 67 56 Coffee 101 121 300 1,581 512 56 105 21 Cocoa 0 3 7 2 3 2 1 1 Fruits and vegetables 20 41 193 136 83 731 46 806 Beef 924 1,186 923 2,424 344 268 671 115 Pig meat 4 14 167 1,507 439 26 22 309 Poultry 44 49 231 603 303 791 462 1,231 Egg -- -- 106 3 10 39 0 0 Milk 66 35 533 2,337 881 1,157 1,110 2,682 Latin American focus countries: 871 5,548 5,616 8,831 5,078 11,755 4,276 6,013 Source: Anderson and Valenzuela (2008) based on estimates reported in Anderson and Valdés (2008a). a. Assumes the rate of assistance to covered products in nonfocus countries is the same as the average for the focus Latin American countries excluding Mexico, and that their share of the value of Latin American and Caribbean (excluding Mexican) agricultural production at undistorted prices is the same as their average share of the region's agricultural GDP at distorted prices during 1990­2004, which was 23 percent. These dollar amounts do not include noncovered farm products, which amount to almost one- third of agricultural output (see last row of table 7.4), nor any mark-up that might be applied along the value chain. b. Mexico is included in the five-year product averages for 1975­79: thus, the Latin American country total is higher in absolute number than the Latin American country total 317 in part a, which excludes Mexico in this period. 318 Distortions to Agricultural Incentives: A Global Perspective What Have We Learned? The most salient feature of price and trade policies in the Latin American region since the 1960s is the major economic reforms, including significant trade liberal- ization, in most countries during the late 1980s and early 1990s. Overall levels of nonagricultural protection have declined considerably, most significantly in the industrial sector, and there have been reforms in the service sector (deregulation and privatization). Both changes have improved the competitiveness of the agri- cultural sector. More specifically, several features of the Latin American experi- ence of the past 40 or more years are worth highlighting. First, the region has seen a gradual movement away from the taxation of farm- ers relative to nonagricultural producers since the 1970s and the emergence of positive assistance for agriculture since the early 1990s. The gradual fall in the esti- mated (negative) RRA for the region, from as high as 40 percent in the early 1970s to less than 2 percent in the past decade, though not dissimilar to trends in Africa and Asia, has been nonetheless dramatic. Instead of being effectively taxed nearly US$17 billion per year, as occurred in the 1980s (or US$400 per per- son working in agriculture), farmers in Latin America now enjoy support of more than US$5 billion per year, or nearly US$125 per person employed on farms. An exception is Argentina, where there was a reversal of policy reform that involved a step back to direct export taxation in late 2001, though this has to be seen in the context of the massive devaluation in Argentina at the time, when the country abandoned exchange rate parity with the U.S. dollar. Thanks to the devaluation, Argentina continued to contribute to the rapid growth of Latin America's share in the global exports of farm products that was stimulated by the gradual elimina- tion of antiagricultural policies. Second, dispersion in average NRAs and RRAs across Latin America for farm- ers has not diminished much despite the reforms in all countries. This means there is still significant scope for reducing distortions in the region's use of resources in agriculture. This finding also indicates that political economy forces are at work in each country and that these are not changing greatly relative to the situation in other countries over time. Third, the dispersion in NRAs among farmers within each Latin American country examined here has also not diminished much. This result means there is still scope for reducing distortions in resource use within agriculture even in countries with an average NRAag and an RRA close to zero. As in other regions, the products in Latin America showing the highest rates of distortion and GSE values are rice, sugar, and milk. Fourth, the strong antitrade bias in assistance rates within the farm sector remains in place. In the 1970s, the NRA for import-competing farm industries Latin America and the Caribbean 319 averaged close to zero in the region. Since then, it has increased to an average of around 20 percent, while the NRA for agricultural exportables has become less negative. The fact that the average NRAs for import-competing and exportable agricultural industries have risen almost in parallel means that the antitrade bias has not fallen much. This may be understandable from a political economy view- point, but it nonetheless means that resources are not being allocated efficiently within the farm sector and--because openness tends to promote economic growth--that total factor productivity growth in agriculture is slower than it would be if the remaining interventions were removed. Fifth, the most important instruments of farm assistance or taxation continue to be trade-restrictive measures. Domestic taxes and subsidies on farm inputs and outputs and NPS have made only minor contributions to the estimates of NRAs for Latin America. Sixth, because the agricultural taxation or assistance is mostly due to trade measures, movements in the CTE closely replicate changes in farm support or tax- ation, which means that, before the reforms, food prices were kept artificially low (though in recent years, food prices have been above international levels, on aver- age). It also means there is considerable variation in CTEs across products and across countries in the region. CTEs are highest for milk, rice, and sugar, but are negative, on average, for maize, beef, and soybeans. The current level of taxation on food consumers in the region as a whole is small, though, amounting to less than US$15 per capita per year. And seventh, the decline in negative RRAs has been caused as much by cuts in protection in nonagricultural sectors as by reforms in agricultural policies. This underscores the fact that the reductions in distortions in agricultural incentives in the region have been part of a series of economy-wide reform programs and have not been caused merely by farm policy reforms. Poverty and Policy Implications The assistance trends surveyed in this chapter are, in one sense, encouraging for economic policy advisors: the long period of encouraging import substitu- tion in the industrial sector and of taxing primary exports, which so heavily discriminated against the agricultural sector in Latin America, has been largely relegated to history. However, as the above summary of the findings makes clear, this does not mean that policies are no longer distorting agricultural incentives. If Latin America follows the policy path chosen by more advanced economies--increasing agricultural assistance as per capita incomes rise-- there may be even more distortion in the future. This suggests that vigilance 320 Distortions to Agricultural Incentives: A Global Perspective will be needed among economic policy advisors in the years to come. Mean- while, the opposite policy problem remains in Argentina, where explicit export taxation was reintroduced in late 2001 and has been increased a number of times in the years since. Neither taxes on agricultural imports to reduce import competition for the benefit of poor farmers, nor taxes on agricultural exports to lower the cost of food for the urban poor, is the most efficient way to reduce poverty (Winters, McCulloch, and McKay 2004). Poverty-reducing objectives are laudable, but trade policy instruments are almost never the best way to achieve them. On the contrary, food trade taxes have the potential to worsen poverty, depending on the earning and spending patterns of poor households and on the alternative tax-raising instru- ments available. Far more preferable would be microeconomic reforms to miti- gate the deep-seated structural problems affecting the competitiveness of factor and goods markets. This is because the reforms have accentuated the differences between commercially oriented farmers and farmers who are less prepared to take advantage of the economic liberalization. Although countries have adopted vari- ous policies to mitigate the human costs of economic adjustment (especially since the mid-1990s), adverse effects on rural poverty and traditional agriculture were often left behind (Spoor 2000; Valdés and Foster 2007). Though many Latin American countries have implemented social safety net programs, such as direct income transfers and conditional cash transfers, to aid all poor people (including families in agriculture), the challenge for the years ahead is to improve the cover- age and effectiveness of poverty alleviation programs. Beyond reducing poverty, such programs are important because they contribute to investing in human cap- ital and act as a form of compensation to reduce the political obstacles to further economic reforms. Notes 1. The economic indicators quoted in this section are from the first nine tables in the appendix to Anderson and Valdés (2008a), based predominately on data compiled from the World Bank's 2008 World Development Indicators Database and the FAOSTAT Database by Sandri, Valenzuela, and Anderson (2007). 2. Annual estimates and additional details may be found in the appendix to Anderson and Valdés (2008a). 3. Other reasons for exchange rate misalignment are discussed in some country studies, but they are not quantified. Several country studies document the significant instability of real exchange rates, which has important influences on the relative profitability of tradable versus nontradable products. Furthermore, some countries, Brazil in particular, tend to have a highly unstable nominal exchange rate because of short-term speculative trading and the effect of political uncertainties on producer incentives. For the purposes of this project and the reasons given in Anderson et al. (2008a, 2008b), however, these these tendencies are not considered policy distortions. Latin America and the Caribbean 321 4. The distortions in the prices of the inputs in the production of nonfarm goods have also been ignored, again in contrast to the treatment of price distortions in estimating agricultural NRAs. 5. This bias is accentuated in cases in which distortions to exchange rates are not included, as noted in the methodology section. Exchange rate distortions have been included only in the studies on the Dominican Republic, Ecuador, and Nicaragua, and these economies are too small for their inclusion to affect noticeably the weighted average NRAs and RRAs for the region as a whole. 6. The consumer tax at the retail level is probably smaller in percentage terms but larger in value terms, because of the addition of marketing margins in the processing, distribution, and retail parts of the value chain. References Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measuring Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, Development Research Group, World Bank, Washington, DC (and appendix A of this volume). ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson, K., and A. Valdés. 2008a. "Distortions to Agricultural Incentives in Latin America and the Caribbean." Agricultural Distortions Working Paper 70, World Bank, Washington, DC. ______, eds. 2008b. Distortions to Agricultural Incentives in Latin America, Washington, DC: World Bank. Anderson, K., and E. Valenzuela. 2008. Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007. Core database at http://www.worldbank.org/agdistortions. de Ferranti, D., G. Perry, W. Foster, D. Lederman, and A. Valdés. 2005. Beyond the City: The Rural Contribution to Development. Washington, DC: World Bank. FAOSTAT Database. Food and Agriculture Organization of the United Nations. http://faostat.fao.org/ default.aspx (accessed December 2007). FAO SUA-FBS Database (Supply Utilization Accounts and Food Balance Sheets Database, FAOSTAT). Food and Agriculture Organization of the United Nations. http://faostat.fao.org/site/354/ default.aspx (accessed December 2007). Krueger, A. O., M. Schiff, and A. Valdés. 1988. "Agricultural Incentives in Developing Countries: Measuring the Effect of Sectoral and Economywide Policies." World Bank Economic Review 2 (3): 255­72. _______. 1991. Latin America. Volume 1 of The Political Economy of Agricultural Pricing Policy. Baltimore: Johns Hopkins University Press; Washington, DC: World Bank. Prebisch, R. 1950. "The Economic Development of Latin America and its Principal Problems." Document E/CN.12/89/Rev.1, United Nations, New York, NY. _______. 1959. "Commercial Policy in Underdeveloped Countries." American Economic Review, Papers and Proceedings 49 (2): 251­73. _______. 1964. "Towards a New Trade Policy for Development: Report by the Secretary-General of the United Nations Conference on Trade and Development." Document E/CONF.46/3, United Nations, New York, NY. Sandri, D., E. Valenzuela, and K. Anderson. 2007. "Economic and Trade Indicators for Latin America, 1960 to 2004." Agricultural Distortions Working Paper 19, World Bank, Washington, DC. Schultz, T. W. 1964. Transforming Traditional Agriculture. New Haven, CT: Yale University Press. Spoor, M. 2000. "Two Decades of Adjustment and Agricultural Development in Latin America and the Caribbean." LC/L.1352-P/I, Economic Reforms Series 56, United Nations Economic Commission for Latin America and the Caribbean, Santiago, Chile. Valdés, A., and W. Foster. 2007. "The Breadth of Policy Reforms and the Potential Gains from Agricul- tural Trade Liberalization: An Ex Post Look at Three Latin American Countries." In Key Issues for a 322 Distortions to Agricultural Incentives: A Global Perspective Pro-Development Outcome of the Doha Round. Volume 1 of Reforming Agricultural Trade for Devel- oping Countries, ed. Alex F. McCalla and John Nash, 244­96. Washington, DC: World Bank. Winters, L. A., N. McCulloch, and A. McKay. 2004. "Trade Liberalization and Poverty: The Evidence So Far." Journal of Economic Literature 42 (1): 72­115. World Bank. 2007. World Development Report 2008: Agriculture for Development, Washington, DC: World Bank. ______. 2008. World Development Indicators Database. World Bank. http://go.worldbank.org/ B53SONGPA0 (accessed December 2007). 8 Sub-Saharan and North Africa Kym Anderson and William A. Masters* In the 1960s and 1970s, many African governments had macroeconomic, sectoral, and trade policies that increasingly favored urban households at the expense of farm households, and the production of importable goods at the expense of exportables (Krueger, Schiff, and Valdés 1988, 1991). Similar biases were also prevalent elsewhere, but rarely to the same extent as in Africa. The magnitude of pro-urban (antiagricultural) and also pro-self-sufficiency (antitrade) interven- tions matters greatly for economic development, because agriculture is the main employer for the poor and is often a key export sector. Changes in these biases could help explain Africa's development experience, including the continent's slow pace of poverty alleviation and economic growth especially in the 1970s and 1980s, and its subsequent recovery since then. Much progress has been made in recent years in reducing the antiagricultural and antitrade biases of policy in Africa, and these changes have been associated with faster economic growth and poverty alleviation. Many price distortions remain, however, and with 60 percent of Sub-Saharan Africa's workforce still employed in agriculture and more than 80 percent of the region's poorest house- holds depending directly or indirectly on farming for their livelihoods (World Bank 2007; Chen and Ravallion 2007), agricultural and trade policies are still key influences on the pace and direction of change in Africa. *The authors are grateful for distortions estimates provided by authors of the focus country case studies, and for assistance with spreadsheets by Johanna Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, Damiano Sandri, and Ernesto Valenzuela. The working paper version of this chapter (Anderson and Masters 2008) contains additional background material and appendix tables. This chapter draws on the introductory and country chapters in Anderson and Masters (2009), with data updated using Anderson and Valenzuela (2008). 323 324 Distortions to Agricultural Incentives: A Global Perspective This chapter summarizes a set of case studies measuring distortions within and across countries over time. No attempt is made here to summarize the volumi- nous literature on policy and economic growth in Africa, the most recent major continental study being Ndulu et al. (2008). This chapter also makes no attempt to summarize the literature dealing with public investment or economic growth strategies more broadly, which was addressed recently by Spence et al. (2008). The goal is more narrowly defined--simply, to compare quantitative indicators of the past with recent agricultural price policies. Including Africa in this global study is crucial for several reasons. First, the con- tinent is home to many of the world's poorest people. In 2006, the region accounted for less than 2 percent of global gross domestic product (GDP) and exports and just 4 percent of agricultural GDP, but it also accounted for 12 percent of the world's farmers, 16 percent of agricultural land, and 28 percent of those liv- ing on less than US$1 a day (World Bank 2008). Second, it is the region where out- put and income growth has been slowest over the past half century, especially on a per capita basis. And third, it is where sectoral and macroeconomic (including exchange rate) policies have been among the most heavily interventionist, damp- ening the contribution of market incentives to growth. There is thus much to be learned from examining the policy history of the region, and there is great poten- tial for poverty alleviation if market-friendly, growth-enhancing policies are adopted and the recent large increase in development assistance funds is used wisely to complement and strengthen market forces. The African part of this study is based on a sample of 21 countries. It includes the Arab Republic of Egypt, the most populous and poorest country in North Africa, plus five countries of East Africa (Ethiopia, Kenya, Sudan, Tanzania, and Uganda), five countries in Southern Africa (Madagascar, Mozambique, South Africa, Zambia, and Zimbabwe), five large economies in West Africa (Cameroon, Côte d'Ivoire, Ghana, Nigeria, and Senegal), and five smaller economies of West and Central Africa for which cotton is a crucial export (Benin, Burkina Faso, Chad, Mali, and Togo), and for which price distortions are estimated for just cot- ton and four nontraded food staples. In 2000­04, these economies (aside from Egypt) together accounted for around 90 percent of the agricultural value added, farm households, total population, and total GDP of Sub-Saharan Africa. Esti- mates of distortions are provided for as many years and products as data permit, amounting to an average of 43 years and nine crop or livestock products per coun- try. The covered products account for more than two-thirds of the value of most countries' agricultural production. The 21 focus economies in Africa examined here accounted for only 1.3 per- cent of worldwide GDP but 11 percent of the world's farmers in 2000­04. These and related shares are detailed in table 8.1, which reveals the considerable diversity Table 8.1. Key Economic and Trade Indicators, African Focus Countries, 2000­04 Share (%) of world: National relative to world (world 100) Agricultural RCAa Total Agricultural GDP land agriculture Gini Population GDP GDP per capita per capita and food TSIb Povertyc indexd Benin 0.12 0.01 0.09 7 55 1,034 -- 31 39 Burkina Faso 0.19 0.01 0.09 5 111 953 -- 29 40 Cameroon 0.25 0.03 0.38 13 74 445 -- 15 45 Chad 0.14 0.01 0.07 5 695 -- -- -- -- Côte d'Ivoire 0.28 0.04 0.21 12 139 722 -- 18 48 Egypt, Arab Rep. of 1.13 0.26 1.11 23 6 175 -- 2 34 Ethiopia 1.08 0.02 0.23 2 58 958 -- 12 30 Ghana 0.33 0.02 0.2 6 88 748 -- 17 41 Kenya 0.52 0.04 0.29 8 103 636 -- 12 43 Madagascar 0.28 0.01 0.1 5 202 670 -- 63 47 Mali 0.2 0.01 0.1 5 353 624 -- 39 40 Mozambique 0.3 0.01 0.08 4 324 359 0.03 30 47 Nigeria 1.98 0.15 1.09 8 73 3 -- 71 44 Senegal 0.17 0.02 0.09 10 94 444 -- 13 41 (Table continues on the following page.) 325 326 Table 8.1. Key Economic and Trade Indicators, African Focus Countries, 2000­04 (continued ) Share (%) of world: National relative to world (world 100) Agricultural RCAa Total Agricultural GDP land agriculture Gini Population GDP GDP per capita per capita and food TSIb Povertyc indexd South Africa 0.73 0.42 0.39 59 275 134 0.52 9 58 Sudan 0.55 0.05 0.5 8 490 209 -- -- -- Tanzania 0.58 0.03 0.33 5 166 800 0.73 56 35 Togo 0.09 0 0.05 5 80 407 -- -- -- Uganda 0.42 0.02 0.15 4 60 938 0.8 83 46 Zambia 0.18 0.01 0.07 7 398 194 0.35 60 51 Zimbabwe 0.21 0.04 0.14 18 200 602 0.83 62 50 African focus countries 9.73 1.21 5.74 13 145 -- -- -- -- All Sub-Saharan Africa 9.37 0.98 4.93 10 164 -- 0.55 41 -- All North Africa 2.34 0.70 2.81 30 84 -- -0.78 -- -- All Africa 11.71 1.67 7.74 14 148 -- 0.20 32 -- Source: Sandri, Valenzuela, and Anderson (2007), compiled mainly from World Bank 2008. Note: -- not available. a. Revealed comparative advantage share of agriculture and processed food in national exports as a ratio of that sector's share of global exports. b. Primary agriculture trade specialization index (TSI) (X M) (X M), 2000­02 (world average 0). c. Percentage of population living on US$1/day, from Chen and Ravallion (2007). d. Gini index for the most recent year available between 2000 and 2004, from World Bank 2008. Sub-Saharan and North Africa 327 within the region in terms of stages of economic development, resource endow- ments, trade specialization, poverty incidence, and income inequality. The coun- tries are also very diverse in political and social development terms, and thus offer important opportunities for comparative study. Averages presented here all include South Africa, where per capita national income is more than four times larger than the other focus countries, but where income inequality is also among the highest in the world. The extent of poverty decline in Sub-Saharan Africa has been disappointing relative to other developing-country regions. Over the 1981­2004 period, the number of Sub-Saharan African people living on less than US$1 per day (in 1993 purchasing power parity terms) grew from 168 million to 298 million. As a percent of the population, the number of people in extreme poverty rose to 47 percent in 1990, then stabilized and eventually declined to 41 percent by 2004, marginally below the 42 percent level of 1981. More than two-thirds of that decline in poverty incidence over the past decade or so has been in rural areas, while most of the rest is explained by the rural poor moving to urban centers where their incomes may rise above the dollar-a-day threshold but many remain very poor. The African experience contrasts strongly with that of Asia, where even in South Asia the proportion of the population living on less than US$1 a day has fallen from one-half to less than one-third (Chen and Ravallion 2007). Policy choices have played an important role in observed rates of economic growth, structural change, and poverty alleviation in Africa. Many countries had increasingly severe antiagricultural and antitrade biases in the 1960s and 1970s, which contributed to farmers' poverty, especially in the 1970s. Subsequent reforms varied widely in terms of starting date, speed, and extent of policy change. The switch to policies that are less biased against farmers and trade began in some countries by the late 1970s but in many others only in the 1980s or even later. The transition is ongoing, often with periods of stalling and even reversal, the most notable recent example being Zimbabwe. Agricultural price distortions are not the only target of policy reform of course, but they are a key aspect of economic policy in most African countries. This chapter begins with a brief summary of economic growth and structural changes in the region since the 1950s and of agricultural and other economic pol- icy developments as they affected the farm sector at the time of and in various stages after independence from colonial powers. The chapter then summarizes estimates of the nominal rate of assistance (NRA) and the relative rate of assis- tance (RRA) to farmers delivered by national farm and nonfarm policies over the past several decades, as well as the impact of these policies on the consumer prices of farm products, using the project's methodology (Anderson et al. 2008a, 2008b). The final sections point to what has been learned and draw out implications of the 328 Distortions to Agricultural Incentives: A Global Perspective findings, including for poverty and inequality and for possible future directions of policies affecting agricultural incentives in Africa. Growth and Structural Change1 Between 1980 and 2004, per capita GDP for the 21 focus countries in Africa grew at just 0.7 percent per year. This was half the global average of 1.4 percent and a small fraction of Asia's 5.5 percent, so per capita incomes in Africa have fallen well below the income levels of other countries, especially those in Asia. The difference is due mainly to nonfarm growth, since agricultural GDP growth per capita was about the same in Africa as in other regions (0.6 percent in Africa compared with 0.5 percent for the world as a whole). The aggregate, long-term experience, though, masks large variation over time and across countries. Most notably, Africa experienced a sharp decline in agricul- tural output per capita from the early 1970s through the late 1980s, followed thereafter by a decline in total national income per capita. Both indicators stabi- lized and then shifted upward in the 1990s. More recently, during 2000­06, per capita GDP growth averaged 4.7 percent in Sub-Saharan Africa compared with 3 percent for the world as a whole (World Bank 2007). Trends in GDP are closely linked to changes in Africa's export volumes since the early 1960s. Exports grew at relatively slow rates in Africa compared with the global average of 6.1 percent, causing the region's share of global exports to halve. However, as African economies have gradually opened up, the share of exports in GDP has reversed its decline and began rising in several countries. Though African economies are slowly recovering from their decline during the 1970s and 1980s, only a few countries have achieved substantial restructuring away from agriculture and toward other activities. About one-quarter of the focus countries have seen their shares of agriculture in GDP actually increase over the entire 1965­69 to 2000­04 period. In nearly three-quarters of the focus countries, agriculture's share of GDP is more than 25 percent; it is more than 40 percent in Cameroon, Chad, and Ethiopia. The share of overall employment accounted for by farming activities has fallen in all focus countries but generally remains above 50 percent, which is much higher than farmers' share of GDP. These data under- score the relatively low incomes of farm households, and hence the continued importance of agricultural prices for social welfare. Agriculture is particularly important as a source of exports in Africa, account- ing for more than 70 percent of merchandise exports in Benin, Burkina Faso, Ethiopia, Tanzania, and Uganda during 2000­04. Agriculture's share of merchan- dise exports has actually risen in three of the focus African countries (Benin, Zambia, and Zimbabwe), though it has declined elsewhere partly because of rises Sub-Saharan and North Africa 329 in other primary exports, such as petroleum in Sudan, and partly because of growth in exports of manufactured goods, for example in Kenya, Madagascar, and Senegal. Such nonfarm exports have grown even faster in other regions, however, so the index of revealed agricultural comparative advantage (defined as the share of agriculture and processed food in national exports as a ratio of the share of such products in worldwide merchandise exports) has risen in most of the focus countries. The exceptions are Nigeria and Sudan, which have newly exploited mineral or energy deposits. While most African countries have an increasing level of revealed comparative advantage in agricultural exports, there is also rising domestic demand for farm output. During colonial times, production was heavily export oriented. At the start of the independence period in 1961­64, the total value of agricultural output was about 120 percent of consumption; that ratio has since declined to around 105 percent. The share of farm production that is exported has fallen from nearly 20 percent to just 8 percent, and the share of imports in domestic consumption of farm products has doubled, from 2 to 4 percent. The Evolution of Agricultural Trade Policies The trends in growth and development described above are closely linked to eco- nomic policies pursued by African governments. Before independence, most of Africa had been ruled since the 19th century by foreign powers whose explicit objective was to control trade, both for political reasons and to extract revenue. Interventions were typically managed through licensed monopolies, marketing boards, and restrictions on Africans' labor mobility, property ownership, and market participation. The few countries not ruled by a foreign power were con- trolled by a local aristocracy or immigrant minority whose economic policies were similarly repressive, as in Ethiopia, South Africa, and Zimbabwe. Majority rule came to Africa much later than to any other major region of the world, and it arrived in the 1960s at a time when central planning was widely seen as a promising strategy for economic development. The newly independent, elected governments typically kept the marketing boards and other instruments for intervention that had been developed by previous administrations, simply expanding their mandate to cover more people and larger regions of the country. Their stated goals were to be more inclusive and serve a larger fraction of the African population than the exclusive licensing and limited mandates of colonial institutions. The new governments also adopted new criteria for public employ- ment, with staffing priorities that reflected electoral politics instead of colonial interests. Both changes led to large increases in the public payroll and fiscal expen- diture. These steps were often underwritten by foreign donors, including the 330 Distortions to Agricultural Incentives: A Global Perspective former colonial powers plus other industrialized countries and oil exporters. Project aid and budget support grew rapidly, especially in the 1970s when loans were available at zero or negative real interest rates. These capital inflows covered growing fiscal deficits, current-account imbalances, and increasingly overvalued foreign exchange rates. In general, inflation was kept low, as governments chose to ration credit and foreign exchange rather than expand the money supply, although a few countries, such as Ghana and Zimbabwe, have experienced hyperinflation. African governments' use of externally funded, state-controlled development strategies seemed promising at the time. Many countries around the world were adopting similar approaches. Western aid to support economic interventions also helped counter the growing influence of the Soviet Union, which had supported African liberation movements against colonialism. In retrospect, it is evident that communist powers helped Africans pursue political freedoms they denied to their own people, while lenders and donors of aid helped Africans maintain economic controls they would never have tolerated at home. The net result was a substantial rise in the degree of African governments' economic intervention during the 1960s and 1970s, from the severe but targeted controls of colonial administrations to the more generalized attempts at state-led development of independently elected governments. The growth of African government intervention during the 1960s and 1970s had two major consequences. First, it fueled political instability, offering the incentives and the means for incumbents and their rivals to seize power and exploit government institutions. Some elected leaders were overthrown by force, while others became increasingly despotic and only a few allowed peaceful transi- tions. Political opportunism among both elected and self-appointed leaders com- pounded the second consequence of economic intervention, which was to weaken market institutions, distort economic incentives, and slow the pace of poverty alleviation. It is not known how fast African economies might have grown under different economic policies in the 1960s and 1970s, but the nature and extent of historical interventions was clearly associated with some degree of reduced growth and worsening poverty.2 In the 1980s, African governments faced mounting pressures for public-sector reform, triggered by a sudden rise in world real interest rates, and combined with global recession that worsened Africa's terms of trade. Domestic political con- cerns intensified, and governments found it increasingly difficult to finance the growing fiscal deficits associated with intervention. Lenders of last resort such as the World Bank and International Monetary Fund (IMF) made their aid condi- tional on devaluation, deregulation, privatization, and retrenchment. The Wash- ington-based institutions used similar criteria for their clients throughout the Sub-Saharan and North Africa 331 developing world, following the "Washington Consensus" reform agenda described by Williamson (1990). Trade policy reforms in the 1980s and 1990s were heavily influenced by struc- tural adjustment programs sponsored by the World Bank and the IMF. Loan con- ditions were widely debated and often blamed for the economic stresses that accompanied them, but the actual implementation of reforms was typically slow and often subject to reversal or offsetting policy changes. Senegal, for example, took out the first World Bank structural adjustment loan (SAL) in 1980 and received its last such loan in 1992 before switching to other instruments, such as a private sector adjustment credit received in 2004. The last African loan to have the term "structural adjustment" in its title was made to Mali in 2005.3 By then, the focus of World Bank­IMF conditionality had shifted to national Poverty Reduc- tion Strategy Papers (PRSPs), a mechanism designed to involve a broader range of stakeholders and a wider variety of government activities than had been involved in the SALs. The process was initiated in 1999, and as of mid-2008 a total of 33 African countries had some sort of PRSP on record with the World Bank and the IMF.4 Of the 21 focus economies, the only countries without one are Egypt, Sudan, and Zimbabwe. Africa is a large and diverse continent, divided into more than 50 sovereign nations with widely varying circumstances. Some of the smaller countries have had very distinctive policy and growth experiences. Botswana, Lesotho, and Swaziland, for example, had no choice but to maintain free trade in a customs union with their much richer and more powerful neighbor, South Africa. This enforced openness probably facilitated convergence toward South Africa's income level, helping them achieve Africa's fastest rates of poverty alleviation through the 1970s and 1980s. Other small countries, such as Cape Verde and Mauritius, experienced rapid economic growth due in part to high levels of migration, remittances, and capital flows. Africa's larger countries, including all of the 21 focus economies, have had relatively interventionist governments and slow poverty alleviation in this period, followed by reform and a degree of recov- ery. Studies of African economies customarily emphasize the diversity among countries, an extremely important point. There are also striking patterns across countries, as found in previous studies such as Ndulu et al. (2008). The new data presented below reveal both diversity and clear trends in policy choices. Measuring Rates of Assistance and Taxation The magnitude of government interventions affecting farmers and food con- sumers is quantified here using the common methodology (Anderson et al. 2008a, 2008b) that has been adopted by the authors of this volume and the four preceding 332 Distortions to Agricultural Incentives: A Global Perspective regional volumes in this series. After a brief description of that methodology, a summary of results follows.5 Methodology The NRA is defined as the percentage by which government policies have raised gross returns to farmers above what they would be without the government's inter- vention. Similarly, the consumer tax equivalent (CTE) is the percentage by which policies have raised prices paid by consumers of agricultural outputs.6 Negative NRA and CTE values imply net taxation of farmers or net subsidies to consumers. The NRA and CTE will be identical if the sole source of government intervention is a trade measure and the two are measured at the same point in the value chain, but in general there will also be some domestic producer or consumer taxes or subsi- dies to differentiate them. The NRAs are based on estimates of assistance to indi- vidual industries at the farmgate. The targeted degree of coverage of the products for which agricultural NRA estimates are generated is 70 percent of the gross value of farm production at undistorted prices. The authors of the country case studies also provide guesstimates of the NRAs for noncovered farm products. For coun- tries with non-product-specific (NPS) agricultural subsidies or taxes, such net sub- sidies are added to product-specific assistance to obtain NRAs for total agriculture and also for tradable agriculture for use in generating an RRA. Farmers are affected not only by the prices of their own outputs, but also (albeit indirectly, because of the changes to factor market prices and the exchange rate) by the incentives nonagricultural producers face. In other words, both absolute but relative prices--and hence, relative rates of government assistance-- affect producer incentives. If one assumes that there are no distortions in the mar- kets for nontradables and that the value shares of agricultural and nonagricultural nontradable products remain constant, the economy-wide effect of distortions to agricultural incentives may be captured by the extent to which the tradable parts of agricultural production are assisted or taxed relative to producers of other tradables (Vousden 1990, following Lerner 1936). By generating estimates of the average NRA for nonagricultural tradables, it is then possible to calculate an RRA, which is defined in percentage terms as: RRA 100[(1 NRAag t 100) (1 NRAnonag t 100) 1], where NRAag t and NRAnonag t are the weighted average percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. Since the NRA cannot be less than 100 percent if producers are to earn anything, neither can the RRA. If both these sectors are assisted equally, the RRA is zero. This measure is useful in that, if it is below (or above) zero, it provides an internationally comparable indication of the extent to which a country's policy regime has an antiagricultural or proagricultural bias. Sub-Saharan and North Africa 333 In calculating the NRA for producers of agricultural and nonagricultural trad- ables, this methodology seeks to include distortions generated by dual or multiple exchange rates. These have been important in many African countries, particu- larly during the 1970s and 1980s, making their estimated (typically) positive NRAs for importables and (typically) negative NRAs for exportables larger than they otherwise would have been. Dollar values of farmer assistance and consumer taxation are obtained by mul- tiplying the NRA estimates by the gross value of production at undistorted prices, to obtain an estimate in U.S. dollars of the direct gross subsidy equivalent (GSE) of assistance to farmers. The GSEs are then summed across products for a country and across countries for any or all products to get regional aggregate transfer esti- mates for the studied economies. GSE values are calculated in constant dollars, and are also expressed on a per-farm-worker basis. To obtain comparable dollar value estimates of the consumer transfer, the CTE estimate at the point at which a product is first traded is multiplied by consump- tion (obtained from the Food and Agriculture Organization's [FAO's] supply and utilization database) valued at undistorted prices to obtain an estimate in constant U.S. dollars of the tax equivalent to consumers (TEC) of primary farm products. This too is summed across products for a country, and across countries for any or all products, to get regional aggregate transfer estimates for the covered farm products of the African focus countries. Estimates of NRAs in agriculture Agricultural price, trade, and exchange rate policies have reduced the earnings of African farmers quite substantially, especially prior to the 1980s.7 The average rate of taxation on all agricultural production, as measured by the weighted average NRA, was less than 8 percent at the time many African countries achieved inde- pendence in the early 1960s, and then almost doubled to a peak around 15 percent in the 1970s, as interventions became more severe (table 8.2). Reforms have since reduced the average extent of taxation to below its level of the early 1960s, includ- ing a brief period in the late 1980s when a combination of policy reforms and low international commodity prices brought the weighted average NRA to near zero. Such averages hide considerable diversity within the region, particularly in South Africa, where net protection of farmers rose during the 1970s and early 1980s and declined thereafter, opposite of the trend in the Africa-wide average. A visual impression of the variation across countries and the extent of reforms between 1975­79 and 2000­04 is provided in figure 8.1, showing clearly the major reduction in taxation rates facing farmers in countries such as Cameroon, Ghana, Madagascar, Senegal, and Tanzania. That figure also shows the transition from 334 Table 8.2. NRAs to Agriculture,a African Focus Countries, 1955­2004c (percent) Region 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Cameroon West -- 2.9 6.0 7.4 14.4 11.2 2.4 1.1 1.3 0.1 Côte d'Ivoire West -- 23.5 29.3 28.1 30.8 32.2 24.3 19.5 20.0 24.5 Egypt, Arab Rep. of North 23.2 33.9 37.7 37.5 15.9 9.2 56.6 6.1 4.0 6.1 Ethiopia East -- -- -- -- -- 17.5 22.3 24.4 17.8 11.2 Ghana West 4.4 9.0 19.8 14.9 25.6 21.2 6.3 1.7 3.0 1.4 Kenya East 26.6 23.0 9.7 11.8 1.7 18.6 10.5 5.8 2.4 9.3 Madagascar South 0.2 5.9 11.1 13.5 27.1 38.8 18.2 5.4 2.9 1.0 Mozambique South -- -- -- -- 34.5 25.2 32.0 2.7 3.9 12.4 Nigeria West -- 20.7 11.9 6.7 6.3 9.4 8.2 3.9 0.4 5.4 Senegal West -- 9.3 7.2 22.4 22.7 20.5 4.7 5.6 6.1 7.5 South Africa South -- 4.1 9.4 0.7 3.8 22.9 11.7 10.8 5.7 0.1 Sudan East 11.7 20.4 31.8 43.4 24.3 29.3 35.4 47.8 24.5 11.9 Tanzania East -- -- -- -- 41.8 56.3 45.3 25.2 23.2 12.4 Uganda East -- 1.8 3.1 7.8 17.6 6.2 6.8 0.6 0.5 0.4 Zambia South -- -- 22.4 15.8 37.3 2.7 58.9 30.8 28.6 28.5 Zimbabwe South 16.9 27.2 25.5 26.0 28.6 24.0 24.1 24.9 20.8 38.7 African focus countries: Unweighted averageb 0.3 7.8 12.5 12.9 15.5 13.7 8.9 8.7 6.6 6.0 Weighted averagea 13.6 7.7 11.3 14.7 12.7 7.9 1.0 8.9 5.7 7.3 Dispersion of individual country NRAsc 20.8 13.4 15.1 14.3 17.1 21.2 29.5 16.1 12.3 13.5 Source: Anderson and Valenzuela (2008) based on estimates reported in the appendix and in Anderson and Masters (2008). Note: -- not available. a. Weighted average for each country, including product-specific output and input distortions and NPS assistance as well as authors' guesstimates for noncovered farm products, with weights based on gross value of agricultural production at undistorted prices. Cameroon, Côte d'Ivoire, Nigeria, Senegal, Uganda, and Zambia data under 1960­64 are for 1961­64; Tanzania data under 1975­79 are for 1976­79; and Ethiopia data under 1980­84 are for 1981­84. b. The unweighted average is the simple average across the 16 countries of their national NRA (weighted) average NRAs. c. Dispersion is a simple five-year average of the annual standard deviation around a weighted mean of the national agricultural sector NRAs each year. Sub-Saharan and North Africa 335 Figure 8.1. NRAs to Agriculture, Individual African Focus Countries and Unweighted Regional Average, 1975­79 and 2000­ 04a 20 0 percent 20 40 60 e a r da n a na ia ca ny ric qu oo er ha an as Af Ke ig bi er G Ug ag N am m h ad ut Ca oz So M M 20 0 percent 20 40 60 l ia an ia re a e ga p. b bi bw op an oi Re Ara d of ne m Iv Su nz ba hi Za d' Se t, Et Ta m yp te Zi Eg Cô 1975­79 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). a. Ethiopia data for the first period refer to 1981­84, as 1975­79 data are unavailable. 336 Distortions to Agricultural Incentives: A Global Perspective taxation to support of farmers in Mozambique and Kenya, as well as the transition from slight support to taxation in Nigeria, and the continuing heavy degree of tax- ation in Côte d'Ivoire, Zambia, and Zimbabwe. One important type of variation in distortions is the within-country disper- sion of product NRAs, as measured in table 8.3 by the standard deviation around the weighted mean NRA for covered agricultural products in each period. This dispersion was highest in 1985­89, when many reforms were only partly com- pleted, but even after the recent reforms it is no lower than it was at the beginning of the period. The dispersion of NRAs within African countries is an important target for reform, whatever the level of average NRA. Variation among products has a somewhat similar pattern across countries. Figure 8.2 shows the pattern of dispersion in the regionwide average NRA among the key farm commodities in the late 1970s and a quarter century later. As in other regions of the world, assistance is among the highest for sugar and milk, and is most negative for tropical cash crops such as coffee, cotton, cocoa, and tobacco. The dispersion over a wider range of products and the full time period is summa- rized in table 8.4. A third type of variation is cross-country diversity of national average NRAs. This is evident from the bottom of table 8.2: NRA averages for the agricultural sector became more similar between the latter 1950s and the early 1970s, then less similar until the latter 1980s, and then more similar again, so that by 2000­04 this type of dispersion was back to what it had been in the early 1960s. The fourth important type of variation is differential treatment of import- competing and exportable products, in a way that often favors self-sufficiency. The extent of antitrade bias is shown, in figure 8.3, as the gap between average NRAs for import-competing and exportable products. This gap grew from the 1950s through to the 1980s. It has since narrowed again, due mainly to changes in taxa- tion of exportables, but the gap is still sizeable. This is summarized in the trade bias index (TBI) reported for Africa as a whole in the middle row of table 8.5. Decomposing the NRA into components reveals a subtle but important influ- ence on the aggregate average. Since the late 1970s, the share of tradable farm products that are exportables has fallen from two-thirds to just over one-half (from 67 to 54 percent). Many governments tax trade in both directions, with neg- ative NRAs for exportables and positive NRAs for importables, so the changing composition of African agriculture from exportable to importable helps drive the aggregate NRA toward zero. This compositional effect adds to the changes within the exportables and import-competing subsectors illustrated in figure 8.3. In the African context, product-specific input price distortions contributed very little to the sectoral NRA estimates, and in many cases the case-study authors reported no values at all. Interventions in domestic markets also contributed relatively little. Table 8.3. Dispersion of NRAs across Covered Agricultural Products,a African Focus Countries, 1955­2004 (percent) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Cameroon -- 13.5 18.0 21.8 29.0 20.6 17.2 16.1 13.0 7.5 Côte d'Ivoire -- 25.1 28.0 33.1 46.2 33.3 33.1 26.2 23.4 33.1 Egypt, Arab Rep. of 21.9 14.7 17.1 21.3 32.2 31.9 89.6 33.0 28.7 22.1 Ethiopia -- -- -- -- -- 26.4 28.2 28.0 29.1 23.6 Ghana 9.8 17.2 29.9 29.0 47.9 69.6 56.3 26.2 17.2 25.5 Kenya 33.2 26.0 30.7 20.5 26.5 22.3 23.6 23.4 24.7 25.6 Madagascar -- 31.3 24.7 24.6 37.5 39.2 42.0 39.1 30.3 22.5 Mozambique -- -- -- -- 34.8 36.0 40.3 28.6 33.4 37.9 Nigeria -- 112.9 95.4 94.2 89.9 92.0 94.4 83.2 72.7 53.2 Senegal -- 20.3 16.1 33.5 44.5 38.2 58.8 67.1 14.3 18.6 South Africa 25.7 17.9 19.1 25.3 31.6 42.7 35.0 31.8 20.3 20.3 Sudan 34.2 34.9 34.1 36.2 40.0 31.7 54.4 75.3 41.2 63.2 Tanzania -- -- -- -- 38.6 39.1 41.3 46.5 47.3 51.9 Uganda -- 7.8 11.6 28.5 47.0 39.3 40.5 7.8 6.6 6.9 Zambia -- 14.5 29.6 26.6 36.1 34.8 35.4 39.2 36.1 38.1 Zimbabwe 74.6 71.0 47.3 36.9 27.7 28.1 24.4 25.2 25.3 33.9 African focus countries: Unweighted averageb 33.2 31.3 30.9 33.2 40.6 39.1 44.7 37.3 29.0 30.2 Product coveragec 68 73 72 72 70 67 66 66 66 68 Source: Anderson and Valenzuela (2008) based on estimates reported in the appendix and in Anderson and Masters (2008). Note: -- not available. a. Dispersion for each country is a simple five-year average of the annual standard deviation around a weighted mean of NRAs across covered products each year. Cameroon, Côte d'Ivoire, Nigeria, Senegal, Uganda, and Zambia data under 1960­64 are for 1961­64; Tanzania data under 1975­79 are for 1976­79; and Ethiopia data under 1980­84 are for 1981­84. 337 b. The unweighted average is the simple average across the 16 countries of their five-year simple average dispersion measures. c. Share of gross value of total agricultural production, valued at undistorted prices, accounted for by covered products. 338 Distortions to Agricultural Incentives: A Global Perspective Figure 8.2. NRAs, Key Covered Products, African Focus Countries, 1975­79 and 2000­04 sugar sorghum milk poultry bananas plantains wheat millet cassava yams sunflower maize rice coffee palm oil vanilla tea sheep meat beans beef cocoa sesame groundnuts cotton soybeans tobacco 100 50 0 50 percent, weighteda average across 21 countries 1975­79 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). a. Weights based on gross value of agricultural production at undistorted prices, with each NRA (by country, by product) weighted by the country's value of production of that commodity in a given year. Table 8.4. NRAs, Key Covered Farm Products, All African Focus Countries,a 1955­2004 (percent) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Bananas -- 2 4 0 2 1 1 3 5 1 Beans -- 6 2 3 39 53 66 25 24 25 Beef 13 21 29 37 4 11 23 38 1 26 Cassava 0 0 0 0 1 2 1 1 3 3 Cocoa 14 27 54 48 60 52 36 35 32 36 Coffee 11 27 36 44 62 53 42 37 21 12 Cotton 16 41 53 54 49 43 31 54 38 46 Groundnuts 29 27 38 51 46 44 17 30 36 40 Maize 4 12 3 7 12 1 38 8 2 5 Milk 35 22 32 42 1 22 67 27 8 15 Millet 77 19 6 4 1 1 0 1 3 2 Palm oil -- 25 31 44 17 25 12 108 41 13 Plantains 0 0 0 0 0 0 0 0 0 0 Poultry -- 13 13 16 24 18 3 6 13 3 Rice 62 38 39 22 14 14 29 0 8 5 Sesame seeds 40 53 64 65 68 60 48 48 50 38 Sheep meat 12 14 18 22 21 20 37 49 45 21 Sorghum 35 62 87 49 28 17 41 37 23 21 Soybeans -- -- 14 30 43 43 40 53 50 54 Sugar 22 6 11 24 11 1 42 2 7 44 Sunflower seeds -- 15 17 6 7 16 7 6 6 4 Tea 3 9 7 20 30 34 29 40 28 16 Tobacco -- 42 38 45 54 47 48 38 34 63 Vanilla -- 62 53 39 57 76 85 78 28 13 Wheat 13 27 13 6 12 5 19 4 1 1 Yams 0 0 0 0 1 1 0 1 4 3 339 All covered products 19.9 13.0 17.8 22.1 20.3 12.1 0.9 12.4 6.6 8.9 Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). 340 Distortions to Agricultural Incentives: A Global Perspective Figure 8.3. NRAs to Exportable, Import-Competing, and Alla Agricultural Products, African Region, 1955­2004 80 percent, weighted averages 60 across 16 countries 40 20 0 20 40 60 9 4 9 4 9 4 9 4 9 4 ­5 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 55 60 65 70 75 80 85 90 95 00 19 19 19 19 19 19 19 19 19 20 exportables total import-competing Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). a. The total NRA can be above or below the exportable and importable averages because assistance to nontradables and NPS assistance is also included. Most of the region's measured NRA is due to border measures, which are largely trade taxes, quantitative trade restrictions, and the operations of parastatal trad- ing companies. In absolute terms, the total value of taxes on farming has been substantial. Africa's antiagricultural bias in NRA terms peaked in the late 1970s, but the sector has since grown, so that in constant 2000 U.S. dollars the total value of annual transfers from farmers has risen from around US$2 billion in the early 1960s (tak- ing account of the fact that NRAs were available for only four-fifths as much agri- cultural production then as from 1980) to US$10 billion in the 1970s, and back to around US$6 billion in the 1980s (ignoring the mid-1980s period when inter- national prices were at record lows), 1990s, and 2000­04 (see bottom row of table 8.6a). The distribution across countries is shown in figure 8.4a, where it is clear that the major transfers in recent years have been from farmers in Ethiopia and Sudan in the East, Zimbabwe in the South, and Côte d'Ivoire and Nigeria in the West. What is also clear from that figure is how much decline there has been since the latter 1970s in such transfers, particularly in Egypt and Tanzania but also in many smaller African economies. For Africa as a whole, the latest estimate is equivalent to a gross tax of US$40 per year for each person engaged in agriculture, Table 8.5. NRAs to Agricultural Relative to Nonagricultural Industries, African Region, 1955­2004 (percent) a. Unweighted averages 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 NRA, covered products 0.0 14.5 19.3 20.2 24.8 20.5 11.6 13.3 9.1 8.9 NRA, noncovered products 0.6 1.0 0.4 0.8 1.3 1.5 3.8 3.5 3.0 2.9 NRA, all agricultural products 1.8 10.0 14.2 14.7 17.0 15.4 10.1 10.7 7.1 6.5 Total agricultural NRA (including NPS)b 0.3 7.8 12.5 12.9 15.5 13.7 8.9 8.7 6.6 6.0 TBIc 0.11 0.35 0.40 0.33 0.41 0.34 0.41 0.24 0.19 0.21 Assistance to just tradables: All agricultural tradablesb 3.1 10.9 19.7 20.6 26.2 21.5 13.9 13.9 9.3 9.4 All nonagricultural tradables 18.8 13.1 12.6 23.5 27.0 27.3 23.0 18.8 15.2 14.5 RRAa 13.2 21.2 28.7 35.5 41.8 38.2 29.7 27.5 21.2 20.9 Memo, ignoring exchange rate distortions: Total agricultural NRA 7.0 6.1 8.4 13.0 13.6 13.1 7.6 9.8 8.5 8.6 TBI, all agricultural products 0.00 0.16 0.13 0.03 0.11 0.29 0.45 0.03 0.03 1.31 RRAa 8.3 17.1 21.5 27.8 31.3 28.7 18.8 23.8 20.7 19.6 (Table continues on the following page.) 341 342 Table 8.5. NRAs to Agricultural Relative to Nonagricultural Industries, African Region, 1955­2004 (continued ) (percent) b. Weighted averages 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 NRA, covered products 19.9 13.0 17.8 22.1 20.3 12.1 0.9 12.4 6.6 8.9 NRA, noncovered products 0.5 3.6 1.8 0.2 0.3 3.3 7.6 4.8 5.1 5.2 NRA, all agricultural products 14.0 8.4 12.2 15.6 13.8 9.5 2.0 10.0 6.1 7.7 Total agricultural NRA (including NPS)b 13.6 7.7 11.3 14.7 12.7 7.9 1.0 8.9 5.7 7.3 TBIc 0.00 0.41 0.45 0.44 0.50 0.43 0.60 0.39 0.33 0.26 Assistance to just tradables: NRA, all agricultural tradablesb 24.1 13.3 19.6 25.0 22.1 13.5 0.3 15.4 8.7 12.0 NRA, all nonagricultural tradables 19.5 3.7 2.7 1.5 5.7 1.6 9.2 2.7 2.0 7.3 RRAa 36.5 15.2 21.4 26.0 25.9 13.1 8.3 17.1 10.4 18.0 Memo, ignoring exchange rate distortions: Total agricultural NRA 10.3 5.2 7.3 11.6 8.9 3.7 5.6 6.7 5.6 6.2 TBI, all agricultural products 0.03 0.14 0.17 0.16 0.29 0.05 0.26 0.01 0.30 0.20 RRAa 26.7 9.7 13.4 17.7 17.0 2.7 5.9 12.7 11.8 16.1 Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). a. RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. b. NRAs including NPS assistance, that is, the assistance to all primary factors and intermediate inputs as a percentage of the total primary agricultural production valued at undistorted prices. c. TBI (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector. The regional average TBI is calculated from the regional averages of the NRAs for exportable and import-competing parts of the agricultural sector. Sub-Saharan and North Africa 343 Figure 8.4. GSEs of Assistance to Farmers, African Focus Countries,a 1975­79 and 2000­04 a. Total per country 1.5 1.0 constant 2000 US$ billions 0.5 0 0.5 1.0 1.5 2.0 2.5 a e da a r n na ca ny ric qu oo ha an as Af Ke bi er G Ug ag am m h ad ut Ca oz So M M 1.5 1.0 constant 2000 US$ billions 0.5 0 0.5 1.0 1.5 2.0 2.5 l a ia p. ab e re ia ia n ga bi bw da an er op oi Re Ar of ne m ig Iv Su nz ba hi Za t, d' Se N Et Ta yp m te Zi Eg Cô 1975­79 2000­04 (continued) 344 Distortions to Agricultural Incentives: A Global Perspective b. Total per product sugar milk maize sorghum poultry millet wheat coffee palm oil tea rice beans yams cassava tobacco sheep meat groundnuts cotton cocoa beef 3,500 2,500 1,500 500 0 500 1,500 constant 2000 US$ billions 1975­79 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). down from more than three times that amount in the 1970s (bottom row of table 8.6b), but still larger than government investment or foreign aid targeted to agriculture (Masters 2008). As shown in table 8.7 and figure 8.4b, the burden of taxation was imposed mainly through the three major export cash crops (cocoa, coffee, and cotton), plus groundnuts, beef, rice, and sugar in the 1970s. Three decades later those cash crops are still the main source of transfer from agricul- ture, while sugar and milk have become positively assisted. In summary, the level and dispersion of agricultural NRAs confirm that there has been substantial reform toward less distortion of incentives. However, they also suggest that there are still many opportunities for policy changes that would Table 8.6. GSEs of Assistance to Farmers, Total and Per Farm Worker, African Focus Countries,a 1955­2004 a. Total (constant 2000 US$ millions) 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Benin -- -- -- 8 4 5 3 13 17 4 Burkina Faso -- -- -- 5 11 12 5 10 13 0 Cameroon -- 83 174 263 636 274 48 33 39 4 Chad -- -- -- 20 25 15 2 7 8 1 Côte d'Ivoire -- 406 603 742 2,223 1,535 1,047 752 878 911 Egypt, Arab Rep. of 1,561 2,472 3,348 4,153 2,046 1,204 5,348 582 354 571 Ethiopia -- -- -- -- -- 1,863 2,392 2,188 2,096 1,113 Ghana 103 188 350 334 727 404 91 28 78 34 Kenya 137 162 75 134 157 408 168 77 35 140 Madagascar 2 84 185 358 555 579 239 73 39 10 Mali -- -- -- 12 28 22 11 18 31 2 Mozambique -- -- -- -- 280 198 120 20 51 55 Nigeria -- 2,193 1,176 867 986 2,198 1,402 794 96 1,034 Senegal -- 76 54 234 377 220 45 37 31 42 South Africa -- 186 500 300 330 2,067 853 841 456 14 Sudan 344 686 1,200 2547 1,861 2,373 2,984 3,633 1,848 1,210 Tanzania -- -- -- -- 1,525 1,062 665 322 576 330 Togo -- -- -- 1 2 6 4 7 7 3 Uganda -- 36 64 199 462 144 111 12 18 14 Zambia -- -- 149 112 388 31 396 178 197 158 Zimbabwe 39 347 305 475 779 602 533 536 467 851 African focus countries 1,829 1,838 4,682 9,030 10,770 6,691 834 6,817 5314 6031 345 (Table continues on the following page.) 346 Table 8.6. GSEs of Assistance to Farmers, Total and Per Farm Worker, African Focus Countries,a 1955­2004 (continued ) b. Per person engaged in agriculture (constant 2000 US$) 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Benin -- -- 8 4 4 2 9 11 3 Burkina Faso -- -- 2 3 3 1 2 3 0 Cameroon 35 71 102 241 99 16 10 11 1 Chad -- -- 12 14 7 1 3 3 0 Côte d'Ivoire 275 368 402 1,072 644 382 250 280 292 Egypt, Arab Rep. of 363 459 535 250 144 672 75 43 67 Ethiopia -- -- -- -- -- -- 107 94 45 Ghana 86 149 130 248 120 23 6 15 6 Kenya 41 17 27 27 -- -- 8 3 11 Madagascar 34 67 116 162 151 56 15 7 2 Mali -- -- 4 9 6 3 5 7 0 Mozambique -- -- -- 53 34 21 3 7 7 Nigeria 174 86 60 69 153 96 54 6 68 Senegal 55 35 137 196 103 19 14 11 13 South Africa 75 197 122 156 1,097 442 440 250 8 Sudan 176 292 574 381 432 482 539 255 156 Tanzania -- -- -- 196 121 65 27 43 22 Togo -- -- 2 3 7 4 7 7 2 Uganda 10 15 42 88 24 16 2 2 2 Zambia -- 106 71 215 15 164 65 67 52 Zimbabwe 225 180 249 363 244 182 161 132 237 African focus countries 29 68 120 134 77 9 55 39 41 Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). Note: -- not available. a. Cameroon, Côte d'Ivoire, Nigeria, Senegal, Uganda, and Zambia data under 1960­64 are for 1961­64; Tanzania data under 1975­79 are for 1976­79; and Ethiopia data under 1980­84 are for 1981­84. Farmer numbers are from U.N. Food and Agriculture Organization Statistics Database (FAOSTAT) which may differ from national statistics. Sub-Saharan and North Africa 347 be both pro-poor and progrowth, for raising income for low-income farmers, and for improving resource allocation within and between countries. Assistance to nonfarm sectors and RRAs The antifarm policy biases of the past were due not just to agricultural policies, but also to policies affecting mobile resources engaged in other sectors. For exam- ple, to the extent that protection to manufacturing also has declined over time, the relative burden on agriculture has diminished even more than the agricultural NRA suggests. This study aims to capture the intersectoral effects through using the NRA on nonagricultural products to generate the RRA between farm and nonfarm activi- ties. The case studies are focused mainly on agricultural policy, and their NRAs for the nonfarm sector typically are measured simply using data on applied trade taxes rather than price comparisons. As a result, unlike for farm NRAs, the esti- mated nonfarm NRAs usually do not include the effects of quantitative trade restrictions that were significant in earlier decades but have been relaxed in recent times. The nonfarm NRAs also do not capture distortions in the services sectors, some of which now produce tradables or use resources that are mobile between sectors. It is not possible, therefore, to be confident that the estimated NRAs for nonfarm activities are smaller and decline less rapidly than in fact was the case, and that the RRA estimates understate the past level of antifarm bias. Even though the estimates of the NRA for nonfarm tradables should be consid- ered lower-bound estimates, they turn out to be quite large. Their unweighted average among the African focus countries rose from around 12 percent in the 1960s to 27 percent during 1975­84 before declining to around 15 percent during the most recent decade or so. As a result, the unweighted RRA is lower and dips even more (to 42 percent) in the middle of the studied period than does the NRA for agriculture, before returning at the end of the period to around the 20 percent it was in the early 1960s (figure 8.5). CTEs of agricultural policies If there were no farm input distortions and no domestic output price distortions, so that the NRA is entirely the result of border measures such as an import or export tax or restriction, and there were no domestic consumption taxes or subsi- dies in place, the CTE would equal the NRA for each covered product. But such domestic distortions are nevertheless present in several African countries. Also, the value of consumption weights used in obtaining the CTEs are quite different from the value of production weights used for obtaining weighted average NRAs 348 Distortions to Agricultural Incentives: A Global Perspective Figure 8.5. NRAs to Agricultural and Nonagricultural Tradable Products and RRA,a Africa Region, 1960­2004 30 percent, weighted averages 20 across 16 countriesa 10 0 10 20 30 40 4 9 4 9 4 9 4 9 4 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 60 65 70 75 80 85 90 95 00 19 19 19 19 19 19 19 19 20 NRA nonagriculture NRA agriculture RRA Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). a. The RRA is defined as 100*[(100 NRAagt ) (100 NRAnonagt ) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respec- tively. The five small cotton-exporting countries of West and Central Africa are not included here. (both measured at undistorted prices). Hence, the average CTEs are quite diferent from the average NRAs for numerous countries, particularly those exporting cash crops in order to import staple foods. This can be seen by comparing the country and product CTEs in table 8.7 with the corresponding NRAs in table 8.2. Nonethe- less, the weighted average CTE for the region has moved much like the NRA: start- ing at around 10 percent at the time of independence, falling to 17 percent (that is, a 17 percent consumer subsidy equivalent) by the early 1970s, and then gradually lessening and eventually reaching close to zero (with a blip in the latter 1980s when Egypt overshot in its reform efforts to reduce the suppression of domestic food prices just when the international price of food fell to record low levels). The variance in national CTEs within countries also rose before the reforms and fell after the latter 1980s (see table 8.7 including the bottom row). In dollar terms, the subsidies to consumers of farm products in Africa are largest in Sudan and Ethiopia, while the tax on consumers historically has been largest in Nigeria and South Africa. Prior to its reforms in the 1980s, Egypt was also a huge subsidizer of food consumers. The transfer, on average, from produc- ers to consumers in the region amounted in 2000­04 to around US$1.7 billion per Table 8.7. Percentage CTE of Policies Assisting Producers of Covered Farm Products,a African Focus Countries,d 1961­2004 (percent, at primary product level) 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Benin -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Burkina Faso -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Cameroon 0.4 0.7 1.3 3.7 3.7 1.1 0.4 0.2 0.0 Chad -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Côte d'Ivoire 9.4 20.1 8.4 3.8 10.8 3.9 4.6 4.3 3.8 Egypt, Arab Rep. of 47.1 49.5 49.6 20.8 12.3 109.5 2.7 13.9 2.8 Ethiopia -- -- -- -- 15.2 17.6 20.3 12.1 10.0 Ghana 2.1 4.4 2.5 4.6 1.7 10.2 4.0 0.8 2.8 Kenya 26.1 21.3 12.8 20.7 26.0 14.8 14.6 12.0 18.7 Madagascar 15.9 22.1 19.2 26.2 42.4 13.4 1.2 1.9 4.0 Mali -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Mozambique -- -- -- 50.5 39.6 53.4 3.6 5.5 31.1 Nigeria 31.2 23.1 14.0 9.0 4.3 15.2 5.6 7.4 0.9 Senegal 10.8 10.3 30.2 25.2 18.3 32.0 31.9 6.0 7.0 (Table continues on the following page.) 349 350 Table 8.7. Percentage CTE of Policies Assisting Producers of Covered Farm Products,a African Focus Countries,d 1961­2004 (continued ) (percent, at primary product level) 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 South Africa 4.0 10.2 0.2 6.7 29.8 14.7 8.6 6.6 0.6 Sudan 15.2 28.9 41.8 16.8 24.2 30.1 47.7 21.2 5.2 Tanzania -- -- -- 42.0 53.7 41.3 17.5 23.1 8.8 Togo -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Uganda 1.0 1.8 1.1 1.3 1.0 0.9 0.3 1.7 1.3 Zambia 26.7 38.5 46.3 54.3 20.8 68.0 54.4 30.5 31.3 Zimbabwe 28.7 35.4 40.1 53.7 39.4 37.1 42.4 36.6 63.7 African focus countries: Unweighted average 7.4 12.1 13.3 12.7 10.4 3.3 7.6 4.2 3.6 Weighted averageb 7.8 11.8 16.6 8.7 6.1 15.5 8.2 0.5 3.2 Dispersion of national CTEsc 21.3 22.8 19.8 22.7 21.6 40.6 19.9 13.9 17.9 Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). Note: -- not available. a. Assumes the CTE is the same as the NRA derived from trade measures (that is, not including any input taxes/subsidies or domestic producer price subsidies/taxes). b. Weights are consumption valued at undistorted prices, where consumption (from FAO) is production plus imports net of exports plus change in stocks of the covered products. c. Simple five-year average of the annual standard deviation around a weighted mean of the national average CTE. d. Cameroon, Côte d'Ivoire, Nigeria, Senegal, Uganda, and Zambia data under 1960­64 are for 1961­64; Tanzania data under 1975­79 are for 1976­79; and Ethiopia data under 1980­84 are for 1981­84. Sub-Saharan and North Africa 351 year, which is only one-third (when expressed in 2000 U.S. dollars) the annual average transfer in the 1970s. Among the covered products, the diversity in meas- ures across the continent means that there are no obvious stand-out products, unlike in other regions, where the biggest transfers are from consumers to produc- ers of milk, rice, and sugar. The link between antifarm and antitrade policies A visual picture of the overall finding that distortions have been reduced substan- tially since the 1970s is provided in figure 8.6. That figure shows values of agricul- ture's TBI on the horizontal axis, and its RRA on the vertical axis. An economy with no antiagricultural bias (RRA 0) and no antitrade bias within the farm sector (TBI 0) would be located at the intersection of the two axes in the upper right corner. In 1975­79, South Africa was the only economy anywhere near that point, and most other Sub-Saharan African economies were far to the southwest of it. In 2000­04, by contrast, Kenya and Nigeria were also close to that neutrality point, and all the other countries shown were far closer than they were in the 1970s. This is not to say there are few distortions left within the agricultural sector, though, because RRA and TBI values in the range of 20 to 40 and 0.2 to 0.4, respectively, are not small and because there is still a wide dispersion of product NRAs in most countries' agricultural sectors. Note also from figure 8.6 that the 2000­04 values fit roughly along a 45-degree line, as the tax burden on agriculture in these countries consists primarily of taxes on trade. International spillovers and multilateral agreements Though the distortion estimates take each country's border prices as given, in reality each country's policies do have some small effect on other countries' prices. An import restriction that raises domestic prices will lower prices else- where, while an export tax that lowers domestic prices will raise them elsewhere. In addition, attempts by one country to stabilize its domestic prices over time will reduce the stability of international prices. As a result, each country's openness to trade contributes to an international public good, offering other countries more favorable and often more stable border prices. This is a classic collective action problem, calling for a multilateral agreement to lock in freer trade policies. Collective action to stabilize world prices is precisely what was sought during the General Agreement on Tariffs and Trade's (GATT's) Uruguay Round Agree- ment on Agriculture, via tariff bindings and disciplines on administered domestic prices. Tariff bindings can reduce the extent of spillovers by restricting the range over which tariffs can increase in response to low prices. World Trade Organization 352 Distortions to Agricultural Incentives: A Global Perspective Figure 8.6. Relationship between RRA and the TBI for Agriculture, African Focus Countries, 1975­79 and 2000­04 a. 1975­79 Nigeria 20 10 South Africa 0 10 RRA (percent) 20 Sudan 30 Senegal Egypt, Arab Rep. of 40 Ghana Madagascar 50 Côte d'Ivoire 60 Zambia 70 Tanzania Zimbabwe Uganda Mozambique 80 1.0 0.8 0.6 0.4 0.2 0 0.2 TBI b. 2000­04 20 10 Mozambique Kenya 0 Ghana Nigeria South Africa 10 Madagascar RRA (percent) Sudan Uganda 20 Senegal Tanzania Egypt, Arab Rep. of 30 Côte d'Ivoire 40 Zambia 50 60 70 80 1.0 0.8 0.6 0.4 0.2 0 0.2 TBI Source: Anderson and Valenzuela (2008), based on estimates reported in the appendix and in Anderson and Masters (2008). Sub-Saharan and North Africa 353 (WTO) bindings, however, are so far above applied import tariffs that this disci- pline on food-importing members in years of low international prices is very weak. The most recent stage of the Doha Round of WTO-sponsored multilateral trade negotiations broke down in mid-2008 because many developing countries were calling for policy space in the form of a special safeguard mechanism that would have allowed even more scope for limiting imports, something richer members of the organization, including the United States, were not willing to sanction in a new agreement. Moreover, there is no corresponding GATT/WTO discipline on food export restrictions, which, as 2008 starkly revealed, can be problematic in years of high international prices. Africa's share of world trade is so small that its policies contribute relatively lit- tle to the collective action problem described earlier, except to the extent that African governments have sided with such countries as Indonesia and India in demanding special safeguards and thereby delayed or prevented the emergence of a new WTO agreement. As the victim rather than perpetrator of international agricultural policy spillovers, however, Africa could benefit greatly from a more effective system of multilateral trade rules. International trade agreements may also help African governments undertake reforms that would not otherwise be possible, allowing them to make commitments and assemble coalitions that could not otherwise be sustained. The details of WTO and other international agree- ments are outside the scope of this book, but generally the results presented here regarding national policies suggest that multilateral agreements can help each government deliver more favorable market conditions for agricultural develop- ment, at the very least, by limiting the rise of import restrictions in other coun- tries. In addition, following the imposition by numerous food-exporting develop- ing countries in 2008 of export restrictions that harmed food importers, perhaps WTO members may eventually agree to limit export restrictions as well. What Have We Learned? Each of the case studies presented in this volume provides detailed insights into Africa's wide variety of country experiences. Aggregating their results to charac- terize all of Africa necessarily obscures as much as it reveals. Making generaliza- tions is sometimes useful, however, if only to allow comparison with other regions, and to detect common trends that cannot be seen in individual cases. Sev- eral principal findings for the African countries considered in this study are pre- sented below. First, African governments have removed much of their earlier antifarm and antitrade policy biases, which had worsened in the late 1960s and 1970s, primarily through increased taxation of exportable products. Reforms of the 1980s and 354 Distortions to Agricultural Incentives: A Global Perspective 1990s reversed that trend, and average rates of agricultural taxation are now back to or below the levels of the early 1960s. Most of this gain has come from reduced taxation of farm exports. Second, the substantial distortions that remain continue to impose a large tax burden on Africa's poor. In constant 2000 U.S. dollar terms, transfers paid by farmers in the 21 focus countries peaked in the late 1970s, at more than US$10 bil- lion per year, or US$134 per farm worker. In 2000­04, the burden of taxation aver- aged US$6 billion per year, or US$41 per person working in agriculture. Even this lower amount is appreciably larger than public investment or foreign aid into the sector. The continuing taxation in Africa contrasts with the situation in both Asia and Latin America, where the average agricultural NRAs and RRAs reached zero by the early 21st century, although like Africa, those regions still have a wide dis- persion of NRAs across products and countries. Third, African farmers have become less taxed, in part because of the changing trade orientation of African agriculture. Reduced taxation of farmers has occurred partly because of a decline in the share of output that is exportable and a corresponding rise in the share that is import-competing. The rate of protection from imports for these products has fluctuated but remains positive. This helps only the few farmers who are net sellers of the protected products, however, and does so in a way that is less efficient and less equitable than many other possible interventions. Fourth, trade restrictions continue to be Africa's most important instruments of agricultural intervention. Domestic taxes and subsidies on farm inputs and outputs, along with NPS assistance, are a small share of total distortions to farmer incentives. As a result, policy incidence on consumers tends to mirror the inci- dence on producers, with fiscal expenditures playing a much smaller role than in more-affluent regions. And fifth, differences in NRAs and RRAs across commodities and countries are still substantial. Dispersion rates, as measured by the standard deviation in NRAs and RRAs across commodies and countries, rose and then fell over time. Looking forward, whatever the overall level of taxation or assistance, moving toward more uniform rates within the farm sector and between countries within the region could still yield substantial increases in efficiency of resource use. Where to from Here? While it is expected that the policy choices of African governments will continue to vary, it is hoped that the overall trend toward reform will continue. Despite dif- ficult conditions, many African governments will continue to reduce taxation of agricultural exports, improve market institutions, and invest in rural public goods. Sub-Saharan and North Africa 355 In that case, producers will continue to respond in ways that generate faster eco- nomic growth and sustained poverty alleviation. That has been the pattern in other regions, and African countries have shown their willingness and ability to begin these changes. Our hopes are tempered by experience, however, including the experience of agricultural policy transition in other regions. A fundamental concern in agricul- tural policy over time as economies progress toward becoming middle-income, though, is overshooting. In response to rural poverty and inequality, many coun- tries start protecting agriculture soon after they stop taxing it. This imposes large costs on consumers, and slows national economic growth. Countries that lock in relatively efficient and equitable policies as soon as they are attained can therefore enjoy a high payoff relative to those that allow farm support policies to become increasingly costly over time. In particular, policies that raise the prices of staple foods impose serious costs on the urban poor and on rural net buyers of these products, as has been demonstrated by recent increases in their prices for other reasons (Ivanic and Martin 2008). Rural-urban poverty gaps can be addressed in far more efficient ways than by subsidizing production or raising food prices. For example, rural poverty can be and has been alleviated in parts of Africa and Asia by increasing the mobility of some members of farm households who work full- or part-time off the farm and repatriate part of their higher earnings back to those remaining on the farm (Otsuka and Yamano 2006; World Bank 2007). Concerted government interven- tions through targeted social policy measures can also be an efficient and effective way to reduce gaps between rural and urban incomes and raise national incomes overall (Winters, McCulloch, and McKay 2004). Efficient ways of assisting the left- behind groups of poor (nonfarm as well as farm) households include public investment measures that have high social payoffs, such as basic education and health, rural infrastructure, and agricultural research and extension. Notes 1. The economic indicators quoted in this section are from the first 10 tables in the appendix, based predominately on the compilation of data from the World Bank (2008) and the United Nations' FAOSTAT Database by Sandri, Valenzuela, and Anderson (2007). 2. Recent studies attempting to measure the magnitude of various constraints on growth have addressed direct effects of exogenous factors, such as unfavorable demographic conditions and trans- port opportunities (Bloom and Sachs 1998), unfavorable temperature conditions and economic scale (Masters and McMillan 2001), and declining rainfall during the mid-1960s through the late 1980s (Barrios, Bertinelli, and Strobl forthcoming), as well as select variables such as institutions (Rodrik, Subramanian, and Trebbi 2004), policies (Glaeser et al. 2004), and inequality (Easterly 2007). A synthesis approach allows for simultaneous determination of government choices and economic outcomes, in models that link exogenous conditions to an equilibrium level of tax rates and public investment, which, in turn, drives growth (for example, McMillan and Masters 2003). 356 Distortions to Agricultural Incentives: A Global Perspective 3. A detailed listing of World Bank projects is available at http://go.worldbank.org/0FRO32VEI0. 4. A detailed listing of countries' PRSPs is available at http://www.imf.org/external/np/prsp/ prsp.asp. 5. Annual estimates and additional details may be found in the appendix. 6. A major difficulty in very poor countries (especially in very poor, landlocked countries) with high internal and border costs of trading, and with large price ranges during each year for farm prod- ucts because of high storage costs, is to obtain domestic and border prices at the same point in the marketing chain before calculating the NRA and CTE. The issue is discussed in general in appendix A, and the ways it has been dealt with in the country case studies are documented in detail in the various chapters in Anderson and Masters (2009). 7. Recall that the sample covers around 90 percent of Sub-Saharan Africa's economy. For North Africa, the sample includes only Egypt, which accounts for almost half the population of North Africa but only 37 percent of its GDP. References Anderson, K., and W. A. Masters. 2008. "Distortions to Agricultural Incentives in Sub-Saharan and North Africa." Agricultural Distortions Working Paper 64, World Bank, Washington, DC. ______, eds. 2009. Distortions to Agricultural Incentives in Africa. Washington, DC: World Bank. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC (and appendix A of this volume). ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson, K., and E. Valenzuela. 2008. Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007. Core database available at http://www.worldbank.org/agdistortions. Barrios, S., L. Bertinelli, and E. Strobl. Forthcoming. "Trends in Rainfall and Economic Growth in Africa: A Neglected Cause of the African Growth Tragedy." Review of Economics and Statistics. Bloom, D. E., and J. D. Sachs. 1998. "Geography, Demography, and Economic Growth in Africa." Brookings Papers on Economic Activity 29 (2): 207­96. Chen, S., and M. Ravallion. 2007. "Absolute Poverty Measures for the Developing World, 1981­2004." Policy Research Working Paper 4211, World Bank, Washington, DC. Easterly, W. 2007. "Inequality Does Cause Underdevelopment: Insights from a New Instrument." Journal of Development Economics 84 (2): 755­76. Glaeser, E. L., R. La Porta, F. Lopez-de-Silanes, and A. Shleifer. 2004. "Do Institutions Cause Growth?" Journal of Economic Growth 9 (3): 271­303. Ivanic, M. and W. Martin. 2008. "Implications of Higher Global Food Prices for Poverty in Low- Income Countries." Policy Research Working Paper 4594, World Bank, Washington, DC. Krueger, A. O., M. Schiff, and A. Valdés. 1988. "Measuring the Impact of Sector-specific and Economy- wide Policies on Agricultural Incentives in LDCs." World Bank Economic Review 2 (3): 255­72. ______. 1991. The Political Economy of Agricultural Pricing Policy, Volume 3: Africa and the Mediter- ranean. Baltimore: Johns Hopkins University Press for the World Bank. Lerner, A. 1936. "The Symmetry between Import and Export Taxes." Economica 3 (11): 306­13. Masters, W. A. 2008. "Beyond the Food Crisis: Trade, Aid and Innovation in African Agriculture." African Technology Development Forum 5 (1): 3­13. Masters, W. A., and M. S. McMillan. 2001. "Climate and Scale in Economic Growth." Journal of Economic Growth 6 (3): 167­86. McMillan, M. S., and W. A. Masters. 2003. "An African Growth Trap: Production Technology and the Time-Consistency of Agricultural Taxation, R&D and Investment." Review of Development Economics 7 (2): 179­91. Sub-Saharan and North Africa 357 Ndulu, B., S. A. O'Connell, R. H. Bates, P. Collier, C. C. Soludo, J.-P. Azam, A. K.Fosu, J. W. Gunning, and D. Njinkeu. 2008. The Political Economy of Economic Growth in Africa, 1960­2000. Cambridge and New York: Cambridge University Press. Otsuka, K., and T. Yamano. 2006. "Introduction to the Special Issue on the Role of Nonfarm Income in Poverty Reduction: Evidence from Asia and East Africa." Agricultural Economics 35 (supplement): 373­97. Rodrik, D., A. Subramanian, and F. Trebbi. 2004. "Institutions Rule: The Primacy of Institutions Over Geography and Integration in Economic Development." Journal of Economic Growth 9 (2): 131­ 65. Sandri, D., E. Valenzuela, and K. Anderson. 2007. "Economic and Trade Indicators for Africa." Agricul- tural Distortions Working Paper 21, World Bank, Washington, DC. http://www.worldbank.org/ agdistortions. Spence, M., et al. 2008. The Growth Report: Strategies for Sustained Growth and Inclusive Development. Washington, DC: World Bank. Vousden, N. 1990. The Economics of Trade Protection. Cambridge: Cambridge University Press. Williamson, J. 1990. "What Washington Means by Policy Reform." In Latin American Adjustment: How Much Has Happened?, ed. J. Williamson. Washington, DC: Institute for International Economics. Winters, L. A., N. McCulloch, and A. McKay. 2004. "Trade Liberalization and Poverty: The Empirical Evidence." Journal of Economic Literature 62 (1): 72­115. World Bank. 2007. World Development Report 2008: Agriculture for Development. Washington, DC: World Bank. ______. 2008. World Development Indicators. Washington, DC: World Bank. WTO, ITC, and UNCTAD (World Trade Organization, International Trade Commission, and United Nations Conference on Trade and Development). 2007. World Tariff Profiles 2006. Geneva: WTO. 9 China and Southeast Asia Kym Anderson and Will Martin* In the past, farm earnings in China and Southeast Asia have often been depressed by pro-urban, antiagricultural biases of government policies in developing coun- tries. Much progress has been made since the 1980s in reducing that policy bias, however, especially in China, where the changes have been transformational. Nonetheless, many trade-reducing price distortions remain within the agricultural sector, and some countries have moved from taxing to protecting their farmers, which involves a different but still inefficient use of national resources as compared to removing all price distortions. The study presented in this chapter examines China (excluding Hong Kong, Macao, and Taiwan and herein referred to simply as China) and the five large Southeast Asian economies of Indonesia, Malaysia, the Philippines, Thailand, and Vietnam. Estimates of distortions are provided for as many years as data permit over the past five decades, and for an average of eight crop and livestock products per economy, which in aggregate amounts to about 70 percent of the gross value of agricultural production in those economies.1 There is considerable diversity within Southeast Asia in terms of stage of devel- opment, relative resource endowments, and comparative advantage--and hence trade specialization, incidence of poverty, and income inequality. These economies thus provide a rich sample for comparative study. Per capita income in *This chapter draws on the introductory and country chapters in Anderson and Martin (2009), with data updated using Anderson and Valenzuela (2008). The authors are grateful for the distortions estimates provided by authors of the focus country case studies, and for assistance with spreadsheets by Johanna Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, Damiano Sandri, and Ernesto Valenzuela. The working paper version of this chapter (Anderson and Martin 2008) contains additional background material and appendix tables. 359 360 Distortions to Agricultural Incentives: A Global Perspective Vietnam is barely one-twelfth the global average, while in Indonesia and the Philippines it is around one-sixth, in China more than one-quarter, in Thailand more than one-third, and in Malaysia around three-quarters. In terms of endow- ments of agricultural land per capita, the Philippines is the least well endowed of the group, but even Indonesia has only about one-quarter of the global average endowment, while Malaysia and Thailand have about two-fifths and China one- half.2 That is, none of these Asian economies is relatively well endowed with crop or pasture land. This might suggest they would have a low comparative advantage in agricultural goods, were it not for two facts: these economies are at varying stages of industrial development, and the quality of and institutional arrange- ments/entitlement to their land and water vary greatly. As a result, there is a wide range in their comparative advantages and trade specialization. The share of agri- cultural and food products in the country's merchandise trade is about 60 percent of the global average share for China and the Philippines, while it is well above 100 percent for the other Southeast Asian countries examined here. The recent decline in poverty in the region is historically unprecedented: the number of Asian people living on less than US$1 per day (2005 purchasing power parity) has more than halved since 1981, with most of that decline in East Asia (especially China). That absolute decline represents a drop from 69 to 10 percent of the population (table 9.1). During the 10 years to 2002, no less than three- quarters of that decline in the proportion of Asia's poor occurred in rural areas, and another one-sixth was due to a movement out of poverty by rural people migrating to better opportunities in urban areas (Chen and Ravallion 2007, 2008).3 Policy developments have made nontrivial contributions to the growth, struc- tural changes, and poverty alleviation observed in East Asia over the past five decades. The transformational move away from planning and state-owned Table 9.1. Changes in Poverty in Asia, 1981­2005 1981 1987 1993 1999 2005 Number of people, millions China 730 412 444 302 106 Other East Asia 217 186 156 122 106 India 296 285 280 270 267 Other South Asia 91 99 61 89 84 Total, Asia 1,334 982 941 783 563 Percent of population East Asia 69 39 36 24 10 South Asia 42 37 29 27 24 Source: Chen and Ravallion (2008). China and Southeast Asia 361 enterprises to greater dependence on markets and private entrepreneurship has had a particularly dramatic effect in China and Vietnam since the 1980s. Also important has been the move in market economies away from import-substituting industrialization toward export-oriented development strategies. Agricultural policies were not the only--or even the main--target of these reforms, but they were an integral part of the process. This chapter begins with a brief summary of economic growth and structural changes in the Southeast Asia and China since the 1950s and of agricultural and other economic policies as they affected agriculture before and after various reforms--and in several cases, the fundamental regime changes--of the past half century.4 It then summarizes new estimates of the nominal rate of assistance (NRA) and the relative rate of assistance (RRA) to farmers delivered by national farm and nonfarm policies over the past several decades (depending on data avail- ability), and of those policies' impacts on consumer prices of farm products. Both farmer assistance and consumer taxation is negative in periods where there is an antiagricultural, pro-urban consumer bias in a country's policy regime. The final sections summarize what has been learned and draw out implications of the find- ings, including for poverty and inequality and for possible future directions of policies affecting agricultural incentives in this part of Asia. Growth and Structural Changes5 The most striking economic characteristic of East Asia's developing economies is their generally fast rate of economic growth and industrial development over the past three decades. The recent report of the Commission on Growth and Develop- ment (Spence 2008) noted that 13 of the world's economies have had sustained growth of real per capita income of more than 7 percent for at least 25 consecutive years since World War II, and that nine of those are in East Asia.6 Between 1980 and 2004, East Asia's per capita gross domestic product (GDP) grew at 6.3 percent per year, compared to a global average of just 4 percent and a South Asian average of 3.4 percent. Industrial growth in that period was more than three times the world's average of 2.5 percent; even agricultural growth was well above the world average of 2.0 percent per year except in Malaysia and the Philippines. As a conse- quence, per capita incomes of most East Asian economies have been converging rapidly--albeit from a low base--on those of rich countries, while other develop- ing and transition economies have, on average, been slipping further away from the U.S. level. A key driver of the rapid growth and industrialization of East Asia has been the decision by many countries in the region to become more open and switch away from an import-substituting development strategy to one that is export oriented. 362 Distortions to Agricultural Incentives: A Global Perspective That change occurred at different times in the focus countries examined here, fol- lowing the experience of Taiwan, China and Korea in the 1960s. China began to make the transition in the late 1970s, and Vietnam in the mid-1980s. As a result, export volumes grew at double-digit rates, and the share of exports in GDP rose steadily for the region, more than doubling in the 30 years to 2004. The East Asian region's share of global exports of nonfood manufactures has quadrupled since 1990, thanks especially to China's industrialization. In 2006, China accounted for 11 percent of the world's manufacturing exports, compared with less than 1 per- cent in 1990: a 20-fold increase in current U.S. dollar terms. The increase for Southeast Asia has been five-fold, contributing to the region's growing share of global manufacturing exports since 1990. With export-led industrial growth has come the dramatic restructuring of Asia's economies away from agriculture and toward not only manufacturing but also service activities. The farm sector's share of GDP in East Asia is now less than 30 percent of what it was in the latter 1960s. The biggest changes are in China and Indonesia, where agriculture's share of GDP dropped from more than 40 and 50 percent, respectively, in the 1960s to 13 percent in 2005. The Philippines, the slowest-growing of the economies examined here, has also been the slowest to move away from agriculture since the 1960s. The shares of overall employment accounted for by farming activities have fallen somewhat more slowly than agriculture's GDP shares, according to statistics in the FAOSTAT Database of the Food and Agriculture Organization (FAO) of the United Nations (which, because of definitional differences, are not always consis- tent with national databases). These shares remain much higher than the GDP shares, implying relatively low labor productivity on farms. Malaysia has seen a major fall, from 57 to 18 percent of the workforce, but elsewhere in the region the share of the labor force still in farming is between two-fifths and two-thirds. These shares would be somewhat less in full-time equivalent terms if part-time, off-farm work activities were accounted for more carefully (see, for example, Otsuka and Yamano 2006; Otsuka, Estudillo, and Sawada 2009), but nonetheless they under- score the fact that incentives faced by farmers affect the well-being of the majority of Chinese and Southeast Asian households. Agriculture's share of merchandise exports has declined even more dramati- cally than its GDP share in the past four decades, as the share of nonprimary goods has grown. Among the focus countries, only in Vietnam do more than one- sixth of exports come from farms. The declining relative importance of farm exports has been much more rapid in East Asia than in the rest of the world: the index of the revealed agricultural comparative advantage (defined as the share of agriculture and processed food in national exports as a ratio of the share of such products in worldwide merchandise exports) has fallen since the 1980s by about China and Southeast Asia 363 two-thirds for the region. So too has the index of agricultural trade specialization (defined as net exports divided by the sum of the imports and exports of agricul- tural and processed food products). That index, which ranges from 1 to 1, has become less and less positive since the 1970s, and indeed has become negative in the case of China and the Philippines. The apparent decline in agricultural comparative advantage is evident in the self-sufficiency data for primary farm products. The share of farm production exported has declined steadily for Malaysia and the Philippines but has been offset by increases in China, Thailand, and Vietnam. Since the 1980s, the share of imports in domestic consumption of farm products has also grown, though in China's case largely because of the need to import more cotton to supply its booming exports of textiles and clothing. As will become clear below, increasing dependence on imports of farm products in East Asia has occurred despite reduc- tions in the taxation of agricultural exports and increases in incentives provided to farmers via government policy reforms. Quantifying the Distortions to Agricultural Incentives The main focus of the present study's methodology is on government-imposed distortions that create a gap between domestic prices and what prices would be under free markets. Since it is not possible to understand the characteristics of agricultural development with a sectoral view alone, the project's methodology both estimates the effects of direct agricultural policy measures (including distor- tions in the foreign exchange market) and generates estimates of distortions in nonagricultural sectors for comparative evaluation. Specifically, NRAs for farmers, including any input subsidies and non-product-specific (NPS) forms of assistance or taxation, are computed. A production-weighted average NRA for nonagricultural tradables is also generated, for comparison with that for agricul- tural tradables via the calculation of an RRA (see Anderson et al. 2008a, 2008b). This approach is not well suited to analysis of China's and Vietnam's policies prior to their reform era, because prices then played only an accounting function and currency exchange rates were enormously distorted. For the reform era in these countries, however, the price comparison approach provides as valuable a set of indicators as for other market economies of distortions to incentives for farm production, consumption, and trade, and of the income transfers associated with interventions. While most of the focus of this study is on agricultural producers, the extent to which consumers are taxed or subsidized is also considered. To do so, a consumer tax equivalent (CTE) is calculated by comparing the price consumers pay for their food within a country and the international price of each food product at the 364 Distortions to Agricultural Incentives: A Global Perspective border. Differences between the NRA and the CTE arise from distortions in the domestic economy that are caused by transfer policies and taxes or subsidies that cause the prices paid by consumers (adjusted to the farmgate level) to differ from those received by producers. To obtain dollar values of farmer assistance and consumer taxation, the coun- try authors' NRA estimates are multiplied by the gross value of production at undistorted prices to obtain an estimate in U.S. dollars of the direct gross sub- sidy equivalent (GSE) of assistance to farmers. These GSE values are calculated in constant U.S. dollars, and are also expressed on a per-farm-worker basis. They (and their equivalent on the consumption side) can be added up across products for a country, and across countries for any or all products, to get regional aggre- gate transfer estimates for the studied economies. Throughout, it has not been possible to assess whether some interventions may have been warranted on national economic welfare grounds because of the presence of externalities, failures in land and credit markets, and policy failures such as underinvestment in public infrastructure in rural areas and interventions by local or provincial governments.7 NRAs to agriculture Prior to the 1980s, agricultural price policies, together with trade and exchange rate policies, almost always reduced farmers' earnings in China and Southeast Asia. The only exceptions were the Philippines in the latter 1960s and Indonesia in the latter 1970s. That explicit or implicit taxation declined from the early 1980s, however, and from the mid-1990s in China and 2000 in Southeast Asia the average NRA shifted so that it became slightly positive. The average hides consid- erable diversity within the region, however. In the Philippines, average NRAs became positive starting in the 1980s and have averaged close to 20 percent since then, while at the other extreme, average NRAs in China and Vietnam were heav- ily negative until the mid-1990s. Meanwhile, NRAs in Indonesia, Malaysia, and Thailand averaged much closer to zero from the early 1980s (table 9.2). A visual representation of both the differences across countries and the rise in average NRAs is clear in figure 9.1, where 2000­04 is compared with 1980­84 for all agricultural products and for rice (by far the region's most important food). The trends are less obvious when looking at the commodity NRAs for the region. Table 9.3, which compares the Southeast Asian averages with those for China, illustrates the diversity of the region's average NRAs across farm com- modities. As is true for other regions of the world, assistance is highest for sugar, but it is also high for maize, and for milk in China and rice and poultry in parts of Southeast Asia. This, together with NRA estimates for the higher-income China and Southeast Asia 365 Figure 9.1. NRAs to All Agriculture and to Rice, China and Southeast Asian Countries, 1980­84a and 2000­04 a. All agriculture 40 20 0 percent 20 40 60 Philippines Vietnam Indonesia China Malaysia Thailand b. Rice 100 50 percent 0 50 100 Malaysia Philippines Vietnam Indonesia China Thailand 1980­84 2000­04 Source: Anderson and Valenzuela (2008), drawn from NRA estimates reported in Anderson and Martin (2009). a. Data for Vietnam are for 1985­89 because 1980­84 estimates are not available. 366 Table 9.2. NRAs to Agriculture,a China and Southeast Asia, 1960­2004 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 b China -- -- -- -- 45.2 35.5 14.3 6.6 5.9 Southeast Asiac -- -- 8.8 0.0 4.6 0.4 4.2 0.0 11.1 Indonesia -- -- 2.6 9.3 9.2 1.7 6.6 8.6 12.0 Malaysia 7.2 7.5 9.0 13.0 4.6 1.3 2.3 0.2 1.2 Philippines 5.3 14.4 5.1 7.1 1.0 18.7 18.5 32.9 22.0 Thailand -- -- 20.3 14.0 2.0 6.2 5.7 1.7 0.2 Vietnamb -- -- -- -- -- 13.9 25.4 0.6 21.2 Source: Anderson and Valenzuela (2008), drawn from estimates reported in Anderson and Martin (2009). Note: -- not available. a. Weighted averages for each country, including product-specific input distortions and NPS assistance as well as authors' guesstimates for noncovered farm products, with weights based on gross value of agricultural production at undistorted prices. b. 1980­84 data for China are for 1981­84; 1985­89 data for Vietnam are for 1986­89. c. Weighted average for the five countries below, with weights based on gross value of agricultural production at undistorted prices. Table 9.3. NRAs, by Covered Product, China and Southeast Asia, 1970­2005 (percent) 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 a China Rice -- -- 56 34 30 7 7 Maize -- -- 35 16 25 5 13 Wheat -- -- 2 22 11 30 4 Sugar -- -- 44 45 12 27 29 Cotton -- -- 34 35 26 4 1 Milk -- -- 129 58 4 18 25 Pig meat -- -- 79 49 15 0 0 Poultry -- -- 25 27 3 0 0 All covered productsb -- -- 51 41 19 2 1 Dispersion of NRAsc -- -- 74 52 21 18 15 Southeast Asia Rice 22 1 1 0 7 1 21 Maize 2 13 14 26 28 28 20 Sugar 1 11 48 19 11 24 49 Coconut 7 1 11 20 34 23 10 Palm oil 15 14 1 2 2 9 3 Rubber 5 17 18 13 16 5 4 Pig meat 0 3 41 9 1 12 6 Poultry 6 48 70 36 24 38 39 All covered productsb 9 0 5 0 4 0 11 Dispersion of NRAsc 24 33 46 48 42 40 40 Source: Anderson and Valenzuela (2008), drawn from estimates reported in Anderson and Martin (2009). Note: -- not available. 367 a. The first and last columns of NRAs for China refer to 1981­84 and 2000­05, respectively. b. Weighted average across all covered products (including some not shown above), with weights based on the unassisted value of production. c. Simple five-year average of the annual standard deviation around a weighted mean of the national NRAs each year. 368 Distortions to Agricultural Incentives: A Global Perspective countries of Northeast Asia (Honma and Hayami 2009), suggests the production of these products within East Asia may be far from optimally allocated from the viewpoint of efficient resource use. There is a great deal of NRA diversity not only across countries in their average NRA but also across commodities within each Asian economy's farm sector. The extent of the latter type of diversity (as measured by the standard deviation) has declined markedly in China but has been greater in Southeast Asia over the past 25 years than in the two decades prior to that (see the dispersion indicators in the middle and bottom of table 9.3). It means that there is still much that could be gained from improved resource reallocation within the agricultural sector of Asian economies were intranational differences in rates of assistance to be reduced. A striking feature of the distortion pattern within the farm sector is its strong antitrade bias. This is evident in figure 9.2, which depicts the average NRAs for agriculture's import-competing and export subsectors for the region: the former average is almost always positive and its trend is upward-sloping, whereas the NRA average for exportables is almost always negative, although much less so in China from the mid-1980s, and it has been close to zero since the late 1990s. Though the gap between the NRAs for those two subsectors has not diminished greatly since the 1960s for the region as a whole, it clearly has narrowed for China. The rise in the average NRAs since the 1980s is too large to be explained just by economies becoming more import-dependent as they lose their comparative advantage in farm products with industrialization: the proportion of tradable farm production that is produced by the export subsector has not declined very much for most Asian economies. Nor can the main motive for altered interven- tion be solely to reduce distortions, for otherwise there would not have been over- shooting in going from negative to positive NRAs for some products (maize in China, for example, and rice in Indonesia), nor would there have been an increase in the average NRA for import-competing farmers. The U.S. dollar value equivalent of the positive or negative assistance to farmers due to agricultural price and trade policies has been nontrivial, but is dominated by China, where it was a tax of more than US$100 billion per year (in constant 2000 U.S. dollars) in the early 1980s but has become a subsidy of around US$15 billion in the past decade (table 9.4a). In Southeast Asia, too, the reforms do not mean there is no intervention now. Rather, their annual transfers changed from being a net negative for farmers of US$0.7 billion in 1985­89 to a net posi- tive of US$8 billion in 2000­04. In recent years, those totals represent transfers to farmers of around US$30 per farm worker in China and US$70 in Southeast Asia (table 9.4b). China and Southeast Asia 369 Figure 9.2. NRAs to Exportable, Import-Competing, and Alla Agricultural Products, China and Southeast Asia, 1970­2004 a. China 40 20 0 percent 20 40 60 80 4 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 19 9 19 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 20 9 20 0 20 1 20 2 2 3 00 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 9 0 0 0 0 19 b. Weighted averages across five Southeast Asian countries 60 50 40 30 percent 20 10 0 10 20 30 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 import-competing total exportables Source: Anderson and Valenzuela (2008), drawn from NRA estimates reported in Anderson and Martin (2009). 370 Table 9.4. Annual GSEs of Assistance to Farmers, Total and Per Farm Worker, Asian Economies,a 1955­2004 a. Total (constant 2000 US$ millions, using the U.S. GDP deflator) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Chinab -- -- -- -- 118,224 75,780 28,381 15,667 15,644 Indonesia -- -- 848 3,783 4,131 785 2,729 4,101 4,286 Malaysia 250 246 547 1,097 456 75 156 3 100 Philippines 225 735 1,082 903 299 1,399 1,850 3,832 1,951 Thailand -- -- 2,434 2,148 324 645 719 260 14 Vietnamb -- -- -- -- -- 726 1,815 18 1,602 b. Per person engaged in agriculture (constant 2000 US$, using the U.S. GDP deflator)a 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 b China -- -- -- -- 280 163 57 31 31 Indonesia -- -- 27 113 113 19 60 86 86 Malaysia 135 126 267 515 213 36 79 2 56 Philippines 33 99 132 99 30 132 163 318 155 Thailand -- -- 163 130 18 34 36 13 1 Vietnamb -- -- -- -- -- 33 73 1 57 Source: Anderson and Valenzuela (2008), drawn from NRA estimates reported in Anderson and Martin (2009). Note: -- not available. a. GSEs including assistance to nontradables and non-product-specific assistance. Farmer numbers are from FAOSTAT which may differ from national statistics. b. 1980­84 data for China are for 1981­84; 1985­89 data for Vietnam are for 1986­89. China and Southeast Asia 371 Assistance to nonfarm sectors and RRAs What matters for the incentives to produce agricultural goods is not just the NRA for agricultural products alone, but this rate of assistance relative to that for other traded goods. The antiagricultural policy biases of the past were due not just to agricultural policies, but to important changes in incentives affecting intersec- torally mobile resources. The latter has resulted in significant reductions in border protection to the manufacturing sector, which in turn has been the dominant intervention in the tradables part of nonagricultural sectors. That reduction in assistance to producers of nonfarm tradables has been at least as responsible for the improvement in farmer incentives as the reduction in direct taxation of agri- cultural industries. It has not been possible to quantify the distortions to nonfarm tradable sectors as carefully as for agricultural tradables. Past studies typically have had to rely on applied trade taxes (for exports as well as imports), plus some adjustments for exchange rate distortions and quantitative restrictions, rather than undertaking comprehensive price comparisons. Hence they usually do not capture fully the quantitative restrictions on trade that were important in earlier decades, though decreasingly so in recent times. Nor do they capture distortions in the services sec- tors, some of which now produce tradables (or would in the absence of interven- tions preventing their emergence). As a result, the estimated NRAs for nonfarm importables are smaller and decline less rapidly than in fact was the case--while those for nonfarm exportables have in some cases been negative. Of those two ele- ments of underestimation, the former bias certainly dominates, so the authors' estimate of the overall NRA for nonagricultural tradables should be considered a lower-bound estimate, and more so in the past than presently, so that its decline appears to have been less rapid than it has been in reality.8 Despite these methodological limitations, the estimated NRAs for nonfarm tradables were very sizeable prior to the 1990s. For the region as a whole, the aver- age NRA value has steadily declined since the 1980s as policy reforms have spread. This shift has contributed to a decline in the estimated negative RRA for farmers: the RRA in Southeast Asia in the early 1970s averaged 25 pecent and in China as recently as the early 1980s more than 50 percent, whereas now it averages slightly more than zero (figure 9.3). Thailand is the only country in the region with a negative RRA this decade (table 9.5). CTEs of agricultural policies The extent to which farm policies impact the retail consumer price of food and the price of livestock feedstuffs depends on a wide range of factors, including the degree of processing undertaken and the extent of competition along the value 372 Distortions to Agricultural Incentives: A Global Perspective Figure 9.3. NRAs to Agricultural and Nonagricultural Tradable Products and RRAs,a China and Southeast Asia, 1970­2004 a. China 60 40 percent, five-year averages 20 0 20 40 60 80 1981­84 1985­89 1990­94 1995­99 2000­04 b. Weighted averages across five Southeast Asian countries 30 20 percent, five-year averages 10 0 10 20 30 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Anderson and Valenzuela (2008), drawn from NRA estimates reported in Anderson and Martin (2009). a. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. Table 9.5. NRAs to Agricultural and Nonagricultural Tradable Industries and RRAs,a China and Southeast Asia, 1960­2004 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Chinab NRA, agricultural tradables 45.2 45.2 45.2 45.2 45.2 35.5 14.3 6.6 5.9 NRA, nonagricultural tradables 41.6 41.6 41.6 41.6 41.6 28.3 24.9 9.9 5.0 RRA 60.5 60.5 60.5 60.5 60.5 49.9 31.1 3.0 0.9 Southeast Asia NRA, agricultural tradables 5.8 5.6 10.2 0.1 4.9 0.9 4.7 0.0 12.1 NRA, nonagricultural tradables 11.5 15.4 20.2 22.0 21.1 18.0 11.5 8.2 8.1 RRA 15.5 8.5 25.3 18.0 13.4 16.1 14.5 7.7 3.7 Indonesia NRA, agricultural tradables -- -- 3.8 10.4 10.5 1.9 7.5 9.7 13.9 NRA, nonagricultural tradables -- -- 27.7 27.7 27.7 26.5 17.6 10.6 8.1 RRA -- -- 24.7 13.6 13.5 22.5 21.3 18.3 5.4 Malaysia NRA, agricultural tradables 7.6 7.9 9.4 13.7 4.9 1.4 2.6 0.2 1.5 NRA, nonagricultural tradables 7.4 7.0 7.1 6.5 5.2 3.9 2.8 2.0 0.9 RRA 14.0 13.9 15.5 18.9 9.6 2.4 0.3 2.2 0.6 (Table continues on the following page.) 373 374 Table 9.5. NRAs to Agricultural and Nonagricultural Tradable Industries and RRAs,a China and Southeast Asia, 1960­2004 (continued) (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Philippines NRA, agricultural tradables 1.7 14.3 6.0 7.2 4.0 15.8 16.7 35.7 23.5 NRA, nonagricultural tradables 19.0 20.3 16.3 16.3 12.9 11.0 9.9 8.6 6.4 RRA 17.4 5.0 19.8 20.3 14.9 4.3 6.1 24.9 15.9 Thailand NRA, agricultural tradables -- -- 23.1 15.9 2.3 6.9 6.4 1.8 0.2 NRA, nonagricultural tradables -- -- 16.1 16.0 14.2 11.1 10.0 8.9 7.8 RRA -- -- 33.7 27.5 14.4 16.3 14.9 6.5 7.4 Vietnamb NRA, agricultural tradables -- -- -- -- -- 15.9 26.4 0.0 20.7 NRA, nonagricultural tradables -- -- -- -- -- 4.3 11.2 1.5 20.8 RRA -- -- -- -- -- 19.2 17.4 1.3 0.0 Source: Anderson and Valenzuela (2008), drawn from NRA and RRA estimates reported in Anderson and Martin (2009). Note: -- not available. a. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. b. 1980­84 data for China are for 1981­84; 1985­89 data for Vietnam are for 1986­89. China and Southeast Asia 375 chain. This study, like the one undertaken by the Organisation for Economic Co- operation and Development (OECD, 2007), therefore attempts only to ask how much impact policies have on the buyer's price at the point on the value chain where the farm product is first traded internationally and hence where compar- isons are made between domestic and international prices (for example, for milled rice, raw sugar, or beef). To obtain weights to make it possible to sum across com- modities and countries, if they were not supplied from national sources, con- sumption data was obtained either directly from the FAO food balance sheets or, in the case of minor products, indirectly by using FAO values for trade data and assuming the undistorted value of consumption is production valued at undis- torted prices plus imports minus exports. If there were no farm input distortions and no domestic output price distor- tions, so that the NRA was entirely the result of border measures such as an import or export tax or restrictions, and there were no domestic consumption taxes or subsidies, the CTE would equal the NRA for each covered product. But such domestic distortions are present in several Asian economies. In China, for example, producer prices were held below consumer prices for several important crop products at least until the early 1990s: producers of food staples were taxed more than consumers were subsidized, even taking into account the "iron rice bowl" system under which urban consumers were able to purchase foodstuffs at low prices. Also, because of international trade, the weights used to aggregate product distortion rates on the consumption side differ from those on the production side of the market. Hence, the aggregate CTE differs somewhat from the aggregate NRA for each economy, as can be seen by comparing the CTEs in table 9.6a with the NRAs in table 9.2 for the 1980s. The CTE was negative until the mid-1990s for China, Thailand, and Vietnam but above zero thereafter. In total dollar terms, current transfers from consumers are clearly largest in China (table 9.6b), but on a per capita basis they are almost as large in the Philip- pines and Indonesia, at between US$22 and US$34 per capita (table 9.6c). They have been minor in Malysia and Thailand in recent years. In 2000­04, they were nontrivial as a percentage of income in Vietnam, though the introduction of export restrictions on rice in 2008 would have reduced this consumer taxation. The role of agricultural policies in stabilizing domestic prices An often-stated objective of food policies in Asia, as elsewhere, is to reduce fluctu- ations in domestic food prices and in the quantities available for consumption. For no product is that more the case than for rice, for which fluctuations in trade barriers are frequently used as a buffer against domestic or international shocks, rather than using trade as a source of cheaper imports or an opportunity for 376 Table 9.6. CTE of Policies Assisting Producers of Covered Farm Products,a China and Southeast Asia, 1970­2004 a. Percent CTE (at primary product level) 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 China -- -- 38.7 35.8 14.2 0.4 0.2 Indonesia 9.0 6.4 8.4 4.3 6.7 11.2 18.3 Malaysia 3.6 18.1 18.1 28.8 15.7 2.8 6.1 Philippines 4.5 7.4 3.1 23.7 22.3 40.2 30.6 Thailand 27.3 19.6 5.7 6.1 6.8 3.1 2.3 Vietnam -- -- -- 11.5 24.3 1.0 19.3 b. Aggregate CTE (constant 2000 US$ million at primary product level) 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 China -- -- 62,859 33,988 923 58,257 44,497 Indonesia 1,676 2,147 3,378 500 884 2,524 4,849 Malaysia 2 163 196 208 169 43 67 Philippines 890 467 96 1,808 2,059 4,178 2,509 Thailand 1,552 1,253 347 229 344 168 83 Vietnam -- -- -- 36 939 320 991 c. CTE per capita (constant 2000 US$ at primary product level) 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 China -- -- 60.9 30.6 0.8 46.6 34.2 Indonesia 13.3 15.3 21.6 2.9 4.7 12.4 22.3 Malaysia 0.2 12.7 13.5 12.6 9.0 2.0 2.8 Philippines 23.0 10.5 1.9 31.7 32.2 58.6 31.9 Thailand 40.5 28.9 7.2 4.4 6.2 2.8 1.3 Vietnam -- -- -- 0.6 13.6 4.3 12.3 Source: Anderson and Valenzuela (2008), derived from national NRA estimates reported in Anderson and Martin (2009). Note: -- not available. a. Assumes the CTE is the same as the NRA derived from trade measures (that is, not including any input taxes or subsidies or domestic producer price subsidies or taxes). Consumption values are production values at undistorted prices divided by the self-sufficiency ratios derived using FAO commodity balance sheets. 1985­89 data for Vietnam is 1986­89. The GDP deflator is used to bring current U.S. dollars to the level in the year 2000. 377 378 Distortions to Agricultural Incentives: A Global Perspective greater export earnings. Since Asia produces and consumes four-fifths of the world's rice (compared with about one-third of the world's wheat and maize), this market-insulating behavior of Asian policy makers means that even by 2000­04, only 6.9 percent of global rice production was traded internationally9 (compared with 14 and 24 percent for maize and wheat). Nominal rates of protection for rice have been above trend in years of low international prices and below trend in years when international prices for rice are high. The effect of a thin market and price insulation has been much more volatile for international prices for rice than for other grains. Figure 9.4 confirms that volatility. Even when averaging over all the focus coun- tries in Southeast Asia, the negative correlation between the rice NRAs and the international rice price is very high, at 0.59. Moreover, that volatility is evident whether the NRA trend is upward or downward. A clear illustration of the latter point is provided by Malaysia, where policy was reformed during its financial crisis years of 1985­87. Even though the growth in rice protection was reversed (fig- ure 9.5), insulation of prices around the trend level of protection continued. This begger-thy-neighbor dimension of each economy's food policy dramati- cally reduces the international public good role that trade between nations can play in bringing stability to the world's food markets. The more some countries insulate their domestic markets, the more they export their volatility to the inter- national market, and the greater the resulting volatility in that marketplace. This, Figure 9.4. Rice NRA and International Rice Price, Southeast Asia, 1970­2005 600 30 500 20 international price, US$ 10 400 NRA (%) 0 300 10 200 20 100 30 0 40 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 20 NRA, southeast Asia (right axis) international price (left axis) Source: Authors' compilation based on data in Anderson and Valenzuela (2008). Note: Correlation coefficient is 0.59. China and Southeast Asia 379 Figure 9.5. NRA for Rice, Malaysia, 1960­2004 280 240 200 160 y 6.3532x 12489 y 6.3075x 12681 percent 120 80 40 0 40 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Source: Authors' compilation based on data in Anderson and Valenzuela (2008). in turn, creates a perceived need for other countries to insulate. In most cases, volatility is exported through changes in import tariffs; but export taxes and export controls are sometimes also used by exporting countries. When NRAs in enough countries are adjusted in this way to changes in international prices, changes in world prices are exacerbated, so that even larger changes in NRAs are needed to achieve any given target level of stability in domestic prices--a classic collective action problem. A multilateral agreement to desist is thus needed. That is precisely what was sought during the Uruguay Round Agreement on Agriculture of the General Agreement on Tariffs and Trade (GATT) and through mechanisms such as tariff bindings and disciplines on administered domestic prices and export subsidies. Tariff bindings can reduce the extent of the problem by restricting the range over which tariffs can increase in response to low prices. To date, however, the bindings are so far above applied import tariffs that this discipline on food-importing members in years of low international prices is very weak. Moreover, there is no corresponding GATT or World Trade Organization (WTO) discipline on food export restrictions, which--as 2008 has starkly revealed--can be extremely prob- lematic in years of high international prices. What Have We Learned? One of the most salient features of price and trade policies in Southeast Asia and China since the 1980s is the spate of major economic reforms, including significant trade liberalization. A key feature has been a reduction in the taxation 380 Distortions to Agricultural Incentives: A Global Perspective of exportable agriculture. Another has been an upward trend in protection to import-competing agriculture in the region--any liberalization in this subsector has been outweighed by increases in protection of other products. Overall, levels of nonagricultural protection have declined considerably, which has improved the competitiveness of the agricultural sector in many economies, especially in China and Vietnam. These trends are captured in figure 9.6, which shows agriculture's trade bias index (TBI) on the horizontal axis and the RRA on the vertical axis. An econ- omy with no antiagricultural bias (RRA 0) and no antitrade bias within the farm sector (TBI 0) would be located at the intersection of the two zero lines in figure 9.6. Though China and all the focus countries of Southeast Asia were to the southwest of that neutral point as of 1980­84, by 2000­04 all but Indonesia and the Philippines had become closer to the vertical axis (meaning they had reduced their antitrade bias in agriculture). All had shifted up also, except for the Philippines, which had become closer to the horizontal axis--although some are now above rather than below that axis, which means they are assisting farmers relative to producers of other tradable products, a scenario that can lead to just as much waste of resources as the earlier, antiagricultural, policy bias. More specifically, the several features of the region's experience of the past four decades are worth highlighting by way of summarizing the key findings of this regional study. First, China and the focus countries in Southeast Asia have experienced both a gradual movement away from taxing farmers relative to nonagricultural produc- ers and, over the past decade, the emergence of slightly positive assistance on aver- age for farmers. The gradual fall in the estimated (negative) RRA for the region has been not dissimilar to but is more dramatic than the trends in other develop- ing countries. Instead of being efffectively taxed more than US$100 billion per year as in the early 1980s (more than US$200 per person working in agriculture), farmers in the region now receive support of more than US$30 per person employed on farms in China and US$70 in Southeast Asia. Second, the dispersion in NRAs to farmers has diminished in China but has increased in Southeast Asian countries on average. This result means there is still scope for reducing distortions in resource use within agriculture, even in coun- tries with an average NRA for agriculture and an RRA close to zero. Third, the antitrade bias in assistance rates within the farm sectors of Southeast Asia remains in place. The NRA for import-competing farm industries has increased over the decades studied, while the negative NRA for agricultural exportables has been reduced in absolute value. The fact that the average NRAs for import-competing and exportable agricultural industries have risen roughly in parallel in Southeast Asia (though not in China) means that those countries' China and Southeast Asia 381 Figure 9.6. Relationship between RRA and the TBI for Agriculture, China and Southeast Asian Countries, 1980­84 and 2000­04 a. 1980­84a 1.7 1.2 Korea 0.7 RRA Taiwan, China 0.2 0 Thailand Malaysia Indonesia Philippines 0.3 Bangladesh India Pakistan Sri Lanka China 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.1 0.2 TBI b. 2000­04 1.7 Korea 1.2 0.7 Taiwan, China RRA Philippines 0.2 India Indonesia China 0 Vietnam Malaysia Thailand Bangladesh Pakistan Sri Lanka 0.3 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.1 0.2 TBI Source: Authors' compilation based on data in Anderson and Valenzuela (2008). a. Data for Vietnam are for 1985­89 because 1980­84 estimates are not available. 382 Distortions to Agricultural Incentives: A Global Perspective antitrade bias within agriculture has not fallen much from the high levels of the past. This may be understandable from a political economy viewpoint, but it nonetheless means that resources are not allocated efficiently within the farm sec- tor and, since openness tends to promote economic growth (Spence et al. 2008), that total factor productivity growth in Southeast Asian agriculture is slower than it would be if remaining interventions were removed. Fourth, movements in CTEs closely replicate changes in farm support and tax- ation, because agricultural taxation and assistance are mostly due to trade meas- ures, although consumer subsidies in China prior to 1994 were an important exception. This broad pattern means that whereas food price reforms were kept artificially low, in recent years they have been above international levels on average in the region (although only trivially thus far for Malaysia and Thailand). It also means there is considerable variation in CTEs across countries in the region. The current level of taxation of food consumers for the region as a whole is rising. In 2000­04, it amounted to more than US$30 per capita per year in China and the Philippines. Fifth, the decline in negative RRAs has been due as much to cuts in protection for nonagricultural sectors as to reforms of agricultural policies. This underscores the fact that reductions in distortions to agricultural incentives in the region have been part of a series of economy-wide reform programs and not just due to farm policy reforms. Sixth, food policies continue to seek to reduce fluctuations in domestic food prices and in the quantities available for consumption via fluctuations in barriers to trade. This begger-thy-neighbor dimension of each economy's food policy reduces the international public good role that trade between nations can play in bringing stability to the world's food markets tremendously. This is especially the case for rice, for several reasons: because it is the main staple food in Asia, because countries in the region heavily intervene to insulate their markets from interna- tional price developments, and because Asia accounts for five-sixths of the global market for rice. Where to from Here? Provided they remain open and continue to free up domestic markets and prac- tice good macroeconomic governance, East Asia's developing economies will keep growing rapidly in the foreseeable future, and growth there will be more rapid in manufacturing and service activities than in agriculture. In the more densely pop- ulated economies of the region, that growth will be accompanied by rapid increases in per capita incomes of low-skilled workers where labor-intensive exports boom. Agricultural comparative advantage is thus likely to decline in such China and Southeast Asia 383 economies. Whether these economies become more dependent on imports of farm products depends, however, on what happens to the RRA. The first wave of Asian industrializers (Japan, and then Korea and Taiwan, China) chose to slow the growth of food import dependence by raising their NRA for agriculture even as they were bringing down their NRAs for nonfarm tradables, such that their RRAs became increasingly above the neutral zero level (Honma and Hayami 2009). A key question is: will later industrializers follow suit, given the past close associa- tion of RRAs with rising per capita income and falling agricultural comparative advantage? When the RRAs for Japan, Korea, and Taiwan, China are mapped against real per capita income, it is possible to superimpose the RRAs for low-income economies on that same graph to see how they are tracking relative to the first industrializers. Figure 9.7 does that for China (and India), showing that its RRA trend of the past 25 years is on the same trajectory as wealthier Northeast Asian countries. A glance at figure 9.3 suggests the same is true for Southeast Asia. One reason one might expect different government behavior now, compared to several decades ago, is because countries that industrialized early were not bound under GATT to restrict their agricultural protection. Had there been strict disci- pline on farm trade measures at the time Japan and Korea joined GATT in 1955 and 1967, respectively, their NRAs may have been halted at less than 20 percent. At the time of China's accession to the WTO in December 2001, its NRA was less Figure 9.7. RRAs and Log of Real Per Capita GDP, Select Asian Countries, 1955a­2005 200 Korea 150 Japan 100 RRA (percent) Taiwan, China 50 0 India 50 China 100 7 8 9 10 11 in real GDP per capita Source: Based on estimates in Anderson and Valenzuela (2008). a. Data begin in 1981 for China and in 1965 for India. 384 Distortions to Agricultural Incentives: A Global Perspective than 5 percent, or 7.3 percent for just import-competing agricultural products. By 2005, China's average bound import tariff commitment was about twice that (16 percent), but what matters most is China's out-of-quota bindings on the items whose imports are restricted by tariff rate quotas. The latter tariff bindings as of 2005 were 65 percent for grains, 50 percent for sugar and 40 percent for cotton (WTO, ITC, and UNCTAD 2007). China also has bindings on farm product-spe- cific domestic supports of 8.5 percent, and can provide another 8.5 percent as NPS assistance if it so wishes--a total 17 percent NRA from domestic support measures alone, in addition to what is available through out-of-quota tariff protection. Clearly, the legal commitments China made on acceding to the WTO are a long way from current levels of domestic and border support for its farmers, and so are unlikely to constrain the government very much in the next decade or so.10 The legal constraints on Asia's developing countries that joined the WTO earlier (except for Korea) are even less constraining. One can only hope that China and Southeast Asia will not make use of the legal wiggle room they have allowed them- selves in their WTO bindings and thereby follow Japan, Korea, and Taiwan, China into high agricultural protection.11 A much more efficient and equitable strategy would be for these countries to treat agriculture in the same way they have been treating nonfarm tradable sectors. It might be argued that such a laissez faire strategy is likely to increase rural- urban inequality and poverty, thereby generating social unrest. On the other hand, policies that lead to high prices for staple foods, in particular, involve poten- tially serious risks for the urban and rural poor who are net buyers of food in developing countries, as has been demonstrated by concerns about the recent increases in prices of these goods (Ivanic and Martin 2008). Available evidence suggests that problems of rural-urban poverty gaps have been alleviated in parts of Asia by some of the more mobile members of farm households finding full- or part-time work off the farm and repatriating part of their higher earnings back to their farm households (Otsuka and Yamano 2006; World Bank 2007). Concerted government intervention through social policy measures is hugely important both in reducing the gaps between rural and urban incomes (Hayami 2007) and in raising national incomes overall (Winters, McCulloch, and McKay 2004). Effi- cient ways of assisting any left-behind groups of poor (nonfarm as well as farm) households include public investment measures in areas that have high social payoffs--basic education, health, and rural infrastructure--as well as in agricul- tural research and development. What do the above lessons and implications suggest developing country policymakers should do when confronted, as in recent years, with a sharp upward movement in international food prices? In the past, as illustrated for rice in figures 9.4 and 9.5, many governments have simply either increased their export China and Southeast Asia 385 restrictions or lowered their import restrictions on food staples for the duration of the spike. Projections by international agencies in 2008 suggest that prices could remain high for the foreseeable future, and that growth in net food imports by rapidly industrializing economies of Asia is one of the significant contributors.12 The sudden fall in food prices in the latter half of 2008 does not necessarily mean those long-run projections are invalid, especially if large developing countries return to rapid growth once the current recession passes. Over the past two or more decades, China and India have steadily raised their RRAs, which previously had been sufficient to keep both countries very close to self-sufficiency in primary agricultural products. In terms of all agricultural and processed food trade, though, China became a net importer for the first time in 2000­04 (Sandri, Valenzuela, and Anderson 2007).13 Should the countries exam- ined here choose to keep their RRAs at current (close to zero) levels, their import dependence in agriculture could well increase over time. If so, other developing countries might well reconsider their current position in the WTO's Doha Round of trade negotiations. By agreeing to lower substantially their bound tariffs and subsidies on agricultural products over the next decade, the region's governments might well be able to extract greater concessions from high-income countries without having to reduce their actual applied RRAs for the foreseeable future should international food prices to return to the historically high levels of the middle half of the present decade. Notes 1. The time series and commodity coverage greatly exceeds that of earlier studies. Krueger, Schiff, and Valdés (1991) analyze Malaysia, the Philippines, and Thailand but for only three or four crops from 1960 to 1984; Orden et al. (2007) provide producer support estimates for China, Indonesia, and Vietnam for the period since 1985; and the OECD has begun examining China (OECD 2005). A com- mon finding of these earlier studies is that the average nominal rate of assistance to farmers is higher in higher-income settings and where agricultural comparative advantage is weaker, and that it is much higher for the import-competing subsector than for exporters of farm products in each economy. 2. In terms of overall (as compared to just crop and pasture) land endowment per capita, China is only one-third of the global average. 3. Income inequality has risen a little over the past two decades but is still low throughout much of the region compared with the rest of the world: as of 2004, the Gini coefficient is between 0.40 and 0.49 for Malaysia, the Philippines, and Thailand. Likewise, the Gini coefficient for land distribution is rela- tively low, at just 0.41 for China, less than 0.50 in Indonesia and Thailand, and 0.5 in Vietnam. This implies reasonably even distributions of land compared with, say, Latin America, where the Gini coef- ficient for land distribution is above 0.7 for major countries such as Argentina and Brazil (World Bank 2007). 4. Apart from the regime changes that occurred during this period, such as the move from social- ism to market economy in China and Vietnam, the region saw the end of colonization between the late 1940s and late 1950s: Indonesia from the Netherlands in 1949, Indochina from France in 1954, and Malaya from Britain in 1957. 5. The rest of this chapter draws in part on Anderson (2008). 386 Distortions to Agricultural Incentives: A Global Perspective 6. The nine economies in East Asia are China, Japan, Hong Kong (China), Indonesia, Korea, Malaysia, Singapore, Taiwan (China), and Thailand. Brazil is the only other large economy in the set, and the other three are Botswana, Malta, and Oman. 7. In addition to the present study's estimation of price wedges, the latter would require a sophisti- cated economic model of the national economy with some quantification of the difference between private and social marginal costs. Such welfare analysis is thus beyond the scope of the present study. 8. This bias is accentuated in cases in which distortions to exchange rates are not included, as noted above in the methodology section. Exchange rate distortions were included in the studies for China, Malaysia, and Vietnam. Their impact was greatest in China, where it made the RRA more negative, to the extent of about 2 percenatage points in the 1970s, 6 percentage points in the 1980s, and 3 points in the 1990s (Huang et al. 2009). 9. This was up from the pre-1990s half-decade global shares, which are all less than 4.5 percent (for example, 4.1 percent in 1985­89), and is greater than the Asian share of just 5.7 percent in 2000­04, according to the project's database (Anderson and Valenzuela 2008). 10. For more on this point, see Anderson, Martin, and Valenzuela (forthcoming). 11. Indications in the ongoing Doha Round of multilateral trade negotiations at the WTO are not encouraging. The Group of 33 developing countries, led by Indonesia but strongly supported by India and the Philippines, among others, is arguing for additional "special and differential treatment" for developing countries in the form of exemptions from agricultural tariff cuts for so-called "special products," and for a special safeguard mechanism that would allow such countries to impose even higher than bound tariffs in years of likely import surges. 12. According to the World Bank's May 2008 commodity forecast, by 2020, grain prices in real terms will still be 10 percent above 2006 levels, which in turn were 20 percent above the average for 2001­05. The International Food Policy Research Institute (Von Braun 2007) and the OECD and FAO (2008) similarly expect food prices to remain high well into next decade and beyond. 13. This change for China was largely due to increases in imports of cotton needed to supply China's surging production of textiles and clothing for export. References Anderson, K. 2008. "Distorted Agricultural Incentives and Economic Development: Asia's Experi- ence." The World Economy 32 (3): 351­84. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC (and appendix A of this volume). ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson K., and W. Martin. 2008. "Distortions to Agricultural Incentives in China and Southeast Asia." Agricultural Distortions Working Paper 69, World Bank, Washington, DC. ______, eds. 2009. Distortions to Agricultural Incentives in Asia. Washington, DC: World Bank. Anderson, K., W. Martin, and E. Valenzuela. Forthcoming. "Long Run Implications of WTO Accession for Agriculture in China." In China's Agricultural Trade: Issues and Prospects, ed. C. Carter and I. Sheldon. St. Paul, MN: International Agricultural Trade Research Consortium. Anderson, K., and E. Valenzuela. 2008. Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007. Core database at http://www.worldbank.org/agdistortions. Chen, S., and M. Ravallion. 2007. "Absolute Poverty Measures for the Developing World, 1981­2004." Policy Research Working Paper 4211, World Bank, Washington, DC. ______. 2008. "The Developing World Is Poorer Than We Thought, But No Less Successful in the Fight Against Poverty." Policy Research Working Paper 4703, World Bank, Washington, DC. Hayami, Y. 2007. "An Emerging Agricultural Problem in High-Performing Asian Economies." Policy Research Working Paper 4312, World Bank, Washington, DC. China and Southeast Asia 387 Honma, M., and Y. Hayami. 2009. "Distortions to Agricultural Incentives in Japan, Korea, and Taiwan, China." Chapter 2 in this volume. Huang, J., S. Rozelle, W. Martin, and Y. Liu. 2007. "Distortions to Agricultural Incentives in China." Chapter 3 in this volume. Ivanic, M., and W. Martin. 2008. "Implications of Higher Global Food Prices for Poverty in Low- Income Countries." Policy Research Working Paper 4594, World Bank, Washington, DC. Krueger, A., M. Schiff, and A. Valdés. 1991. The Political Economy of Agricultural Pricing Policy, Vol- ume 2: Asia. Baltimore: Johns Hopkins University Press for the World Bank. OECD (Organisation for Economic Co-operation and Development). 2005. Review of Agricultural Policies: China. Paris: OECD. ______. 2007. Agricultural Policies in OECD Countries: Monitoring and Evaluation 2007. Paris: OECD. OECD and FAO (Food and Agriculture Organization). 2008. OECD-FAO Agricultural Outlook 2008­ 2017. Paris: OECD. Orden, D., F. Cheng, H. Nguyen, U. Grote, M. Thomas, K. Mullen, and D. Sun. 2007. "Agricultural Pro- ducer Support Estimates for Developing Countries: Measurement Issues and Evidence from India, Indonesia, China, and Vietnam." IFPRI Research Report 152, International Food Policy Research Institute, Washington, DC. Otsuka, K., J. P. Estudillo, and Y. Sawada, eds. 2009. Rural Poverty and Income Dynamics in Asia and Africa. London and New York: Routledge. Otsuka, K., and T. Yamano. 2006. "Introduction to the Special Issue on the Role of Nonfarm Income in Poverty Reduction: Evidence from Asia and East Africa." Agricultural Economics 35 (supplement): 373­97. Sandri, D., E. Valenzuela, and K. Anderson. 2007. "Economic and Trade Indicators for Asia." Agricul- tural Distortions Working Paper 20, World Bank, Washington, DC. http://www.worldbank.org/ agdistortions. Spence, M., et al. 2008. The Growth Report: Strategies For Sustained Growth and Inclusive Development. Washington, DC: World Bank. Von Braun, J. 2007. "The World Food Situation: New Driving Forces and Required Actions." Food Pol- icy Report, International Food Policy Research Institute, Washington, DC. Winters, L. A., N. McCulloch, and A. McKay. 2004. "Trade Liberalization and Poverty: The Empirical Evidence." Journal of Economic Literature 62 (1): 72­115. World Bank. 2007. World Development Report 2008: Agriculture for Development. Washington, DC: World Bank. WTO, ITC, and UNCTAD (World Trade Organization, International Trade Commission, United Nations Conference on Trade and Development). 2007. World Tariff Profiles 2006. Geneva: World Trade Organization. 10 India and Other South Asian countries Ashok Gulati and Garry Pursell* Distortions to price incentives for agriculture from trade, exchange rate, and domestic policies in place in South Asian countries are sometimes significant. The analysis here presents the effects of such policies on several countries in the region--Bangladesh, India, Pakistan, and Sri Lanka--centering on India, which accounts for around four-fifths of South Asia's population, gross domestic prod- uct (GDP), and agricultural GDP.1 The principal focus is on the level of and trends in distortions for agriculture2 and how these have changed over time rela- tive to distortions for nonagricultural traded sectors (principally manufacturing). Previous studies have established that in India, Pakistan, and Sri Lanka, policies strongly favored manufacturing over the principal agricultural crops, although the extent of antiagricultural bias diminished considerably between the 1970s and 1995 (Pursell 1999).3 These new country studies, in addition to extending the ear- lier estimates up to 2005 and back to 1965, provide long-term estimates of distor- tions to relative agricultural incentives in Bangladesh for the first time.4 The new studies also broaden the coverage of previous research by including estimates for * The authors are grateful for the distortion estimates provided by authors of the focus-country case studies, and for assistance with spreadsheets by Johanna Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, Damiano Sandri, and Ernesto Valenzuela. The working paper version of this chapter (Gulati and Pursell 2008) con- tains additional background material and appendix tables. This chapter is a synthesis of the authors' paper on India (Pursell, Gulati, and Gupta 2009) and three other South Asian country studies, namely Ahmed et al. (2009); Bandara and Jayasuriya (2009); and Dorosh and Salam (2009). The analytical narratives are available in Anderson and Martin (2009), and the detailed estimates have been integrated into the project's global database (Anderson and Valenzuela 2008). An attempt is made to summarize the findings in a coherent manner rather than provide a critique of the find- ings and of quantitative estimates in the individual country studies. 389 390 Distortions to Agricultural Incentives: A Global Perspective fresh fruits and vegetables in India and for dairy in India and Pakistan. Both of these sectors account for a large share of the rural economy in South Asia as meas- ured by their contribution to GDP. The chapter is organized into six sections. The first summarizes how agricul- ture has steadily declined as a percentage of GDP and trade in South Asian coun- tries over the past four decades. This is followed by a discussion of how agriculture nevertheless still accounts for more than half of South Asian employment, that the share of food in household budgets remains very high, and that on both counts agricultural policies are highly sensitive politically. The third section describes how South Asian countries' trade policies have evolved and interacted with exchange rate changes, especially during the period of the steady and, in the end, massive devaluation of the Indian rupee between 1984 and 1992. The fourth sec- tion presents quantitative evidence on the long-term evolution of nominal rates of assistance (NRAs) to various agricultural subsectors (including via input subsi- dies) and to agriculture as a whole. This leads into a discussion of trends in incen- tives for farmers relative to incentives for producers of nonagricultural tradables (mainly manufactures). Finally, the sixth section examines the political economy forces that are likely to influence the direction of future policies in the region, including the possibility that a strong pro-agricultural bias may emerge along the lines followed by more advanced, densely populated countries in East Asia and elsewhere. Agriculture and the South Asian Economies At 77 percent of the regional total as of 2000, India accounts for by far the largest proportion of agricultural activity in South Asia, while Pakistan represents 11 per- cent, Bangladesh 8 percent, and Sri Lanka 2.2 percent, and Nepal 1.5 percent. Agriculture is more specialized in the smaller countries. Pakistan, for example, specializes in wheat and Bangladesh in rice, but even so their production is much less than India's. The share of agriculture in the South Asian economies has steadily declined over time (table 10.1). Most of the decline has been in crop agri- culture and horticulture. At independence, these two sectors represented more than half of GDP in India; by 2003, they had dropped to 15 percent of GDP. The livestock sectors in India and Pakistan, by contrast, have grown faster than the rest of the rural primary sector. In India, livestock represents about a quarter of agri- cultural GDP, and in Pakistan close to half (while having a much smaller role in the rural economies of Bangladesh and Sri Lanka). Fisheries and forestry are not included in the agricultural aggregates or the quantitative analysis of this project. They have relatively low GDP shares in India, Pakistan, and Sri Lanka. However, given the importance of fishing in Bangladesh India and Other South Asian Countries 391 Table 10.1. Shares of Agriculture in GDP and Employment, South Asian Countries, 1965­2004 (percent) 1965­69 1975­79 1985­89 2000­04 Share of GDP India 44 36 29 21 Pakistan 35 29 24 22 Bangladesh -- 55 31 22 Sri Lanka 29 28 24 17 South Asia (four-country sample) 43 36 29 21 Share of employment India 73 70 66 59 Pakistan 65 64 55 46 Bangladesh -- 76 67 54 Sri Lanka 55 53 49 45 South Asia (four-country sample) 74 70 65 57 Source: Sandri, Valenzuela, and Anderson (2007). Compiled from the World Bank's World Development Indicators and U.N. Food and Agriculture Organization Statistics Database (FAOSTAT). Note: -- not available. a. Aggregates for South Asia exclude Nepal, Bhutan, and the Maldives, which together account for about 1.5 percent of South Asian agricultural GDP. Separate data not available for Bangladesh before it separated from Pakistan in 1971. (it comprised 4.9 percent of total GDP and almost one quarter of more broadly defined agricultural GDP in 2004), caution is in order in making generalizations from the analysis below for Bangladesh's broader rural economy. Agriculture and trade For India, international trade in agricultural products (including livestock, fish, and forest products) has always been tiny in relation to the size of India's agricul- tural sector. In 2003/04, imports of these products were only 0.6 percent of the value of sectoral production (2.4 percent with edible oil imports included) and exports were 5.7 percent of production. In earlier periods, agricultural imports and exports represented major portions of India's trade (in 1961, for example, they were 27 percent and 44 percent of total merchandise imports and exports, respectively), but each still only accounted for just above 3 percent of agricultural production. In the late 1960s, agricultural imports began to be replaced by domestic 392 Distortions to Agricultural Incentives: A Global Perspective production, and since then they have constituted very small shares of total imports, even in years when essential products such as wheat and sugar had to be imported due to poor harvests. Over time, the share of agricultural products in India's total merchandise exports has also declined, in recent years to around 10 percent. However, the products exported are quite diverse. They include fish and fish preparations, oil cakes, cashew kernels, tea, coffee, tobacco, spices, fruit and vegetables, pulses, bas- mati rice, and, periodically, large quantities of sugar and common rice (more than 4 million tons of rice were exported in 2004, for example). Since the late 1980s, manufactures have typically accounted for 70­80 percent of India's total mer- chandise exports, compared to between 40 and 50 percent during the 1950s and 1960s. Indian service exports have also grown rapidly in recent years, with soft- ware, other information technology, and services outsourcing representing the most dynamic components. Together, exports from these activities were about US$20 billion in 2004, reached US$50 billion in 2008, and were increasing at a rate of 20­25 percent annually. Thus far, there has been no similar development in other South Asian countries. Agricultural trade has also declined in Bangladesh, Pakistan, and Sri Lanka, both in relation to total agricultural production and as a share of total exports and imports. On the import side, as in India, the decline was mainly a consequence of policies aimed at foodgrain self-sufficiency, especially through green revolution technologies. Currently, Pakistan exports rice and during normal seasons is self- sufficient in wheat. Similarly, Bangladesh is self-sufficient in rice during normal seasons and imports small quantities of wheat. Sri Lanka is normally self-sufficient in rice, though this is an outcome of substantial imports of wheat, which (as flour) has inceased over time as a share of household consumption. The outcome of these changes is that in Pakistan, primary commodities (including cotton) are currently around 11 percent of total imports and 11 per- cent of total exports, compared with 26 percent of total imports and 39 percent of total exports in 1973 (Hamid, Nabi, and Nasim 1990). In Sri Lanka, agricultural products (mainly tea, rubber, and coconut products) represented more than 90 percent of total exports during the 1960s and 1970s, but by 2005 their share had declined to 18 percent; the remainder was industrial exports, almost entirely ready-made garments. Over the same period, agricultural and processed food imports declined from almost half of total imports to around 11 percent: most imports are now intermediate manufactured materials (especially textiles for the ready-made garment exporters), machinery and equipment, and manufactured consumer goods. In Bangladesh, the pattern is somewhat different. Ready-made garments now dominate total exports. Agricultural exports (at present almost entirely jute and India and Other South Asian Countries 393 shrimp), meanwhile, are usually about 7­8 percent of total exports, compared with 40 percent (mainly jute and tea) in the mid-1970s. However, Bangladesh imports a wide range of agricultural products--wheat, cotton, sugar, edible oils, dairy products, spices, oil seeds, tobacco, and periodically rice. According to official trade statistics, imports of these products varied between about 12 and 18 percent of total imports between 2001 and 2005, a share that declined only slightly compared to the mid-1990s. But the share of agricultural products in Bangladesh's total imports would be considerably higher than this if an allowance were made for the very substantial unrecorded imports from India, especially of sugar and cattle. The largest consistent quasi-agricultural imports in South Asia are edible oils. In India, these expanded rapidly during the 1970s and early 1980s, triggering a major government program during the 1980s to substitute domestically produced oils for the imports. Edible oil imports declined for some time under the program, but despite very high tariffs, import growth resumed during the 1990s and the 2000s (the tariff on palm oil, for example, was 80 percent in 2006, though it was brought down to 5 percent in 2008 in the wake of surging global prices). During 2000­04, India's imports of edible oils ran about US$2.5 billion annually and accounted for about 40 percent of domestic consumption. Edible oils (mainly palm oil from Malaysia and Indonesia) are also major imports in Bangladesh, Nepal, Pakistan, and Sri Lanka. Tariffs are high in these countries, but not nearly as high as in India,5 a scenario that creates strong incentives for edible oils to be smuggled into India from neighboring countries, particularly Bangladesh and Nepal. Employment, food expenditure, and food safety nets While agriculture's contribution to GDP in India has declined by almost two- thirds since independence in 1947, in 2003, it still accounted for 58 percent of national employment. Bangladesh, Pakistan, and Sri Lanka experienced a similar decline in agricultural employment over the past 40 years but, as in India, at a much slower rate than the decline in agriculture's share in GDP (table 10.1).6 The share of food expenditure in South Asian household budgets, meanwhile, remains very high, although it is slowly declining. As of 2003, rural and urban households in India dedicated 54 percent and 42 percent, respectively, of total per capita consumption expenditure to food. The share of food in the budgets of the poorest 10 percent of households was significantly higher, around 62 percent in rural areas and 58 percent in urban areas. With such high shares of family budgets, it is not surprising that food prices and food availability are highly sen- sitive politically. 394 Distortions to Agricultural Incentives: A Global Perspective Since independence, the Indian government has worked to permanently elimi- nate the incidence of catastrophic famines that occurred during the colonial period and ensure basic foods to its citizens at affordable prices. In pursuit of these objectives, the government intervened in food grain markets through the Food Corporation of India (FCI) and established the present public distribution system (PDS), which had its roots in the rationing system introduced during pre- war years, in 1958. The current PDS sells basic foods at subsidized prices through a large number (currently about 460,000) of "fair price" shops. In June 1997, the system was modified to targeted PDS by distinguishing "below poverty line" and "above poverty line" buyers, with the former eligible for especially low prices and the latter eligible to buy at prices only slightly below free market prices. In 2004, the total central government food subsidy was estimated to be Rs 258 billion (about US$5.7 billion, or 0.83 percent of GDP), defined as the excess of the FCI's total pro- curement, handling and distribution costs over the subsidized sales value. In Bangladesh, Pakistan, and Sri Lanka, food policies similar to those in India have now been phased out. In Pakistan, ration shops were abolished in the late 1980s, although the government continues to subsidize wheat to private flour mills to keep flour prices low for consumers. The system of public grain distribu- tion (rice and wheat) that Bangladesh inherited upon independence, which became especially important following a 1974 famine, continued during the 1980s but was gradually phased out starting in the early 1990s. Currently, the principal safety net for the poor in Bangladesh consists of food-for-work and food trans- fers, mainly of wheat. Postindependence Sri Lanka also operated subsidy schemes for food grains distributed through ration shops. During the 1950s and 1960s, this mainly consisted of imported rice, but later, as in Bangladesh, wheat became the principal vehicle for the subsidies, as it could be purchased more cheaply than rice on world markets. The subsidies were very large (some rice was distributed free) and had high budgetary costs, especially during the 1970s world price boom. In part due to the budgetary expense, the subsidies were reduced after 1977 and the rice ration scheme was abolished in 1979. Since then, the main objective of policy has been to keep consumer prices stable around "reasonable" price levels, balanc- ing this objective against maintaining protection for paddy rice farmers. Input subsidies For the four South Asian countries, self-sufficiency in the production of food grains is a major policy objective, and green revolution development packages developed in the pursuit of self-sufficiency have comprised an array of govern- ment initiatives and subsidies for farmers. In India, for example, the largest of India and Other South Asian Countries 395 Table 10.2. Distribution of Fertilizer and Electricity Subsidies and Subsidy Rates for Key Crops, India, 2004a Subsidy as percent of gross value of production at: Percent of Domestic Reference total subsidy price price Rice 37.1 18.3 15.4 Wheat 35.2 27.2 18.9 Maize 2.1 9.3 8.8 Sorghum 1.8 12.4 11.5 Chickpeas 2.7 11.4 10.9 Groundnuts 2.7 8.4 5.7 Rape/mustard seeds 4.5 11.7 18.0 Soybeans 1.1 4.3 2.9 Sunflower seeds 0.5 8.8 8.6 Sugar 6.6 12.5 15.4 Cotton 5.6 13.2 12.3 Total: 11 crops 100.0 16.9 13.0 Source: Pursell, Gulati, and Gupta (2009). a. The total value of the input subsidies for these 11 crops this year was US$7.8 billion (US$1.9 billion for fertilizer and US$5.9 billion for electricity). these subsidies are for electricity and fertilizers (Gulati and Narayanan 2003). Estimates of the totals of these two subsidies across 11 crops in 2004 are summa- rized in table 10.2. By 2008, as global prices of fertilizers rose sharply, the fertilizer subsidy took on gigantic proportions, touching almost US$25 billion. The subsidy did, however, allow domestic prices to remain largely unchanged. In Pakistan, about 80 percent of cropped area is irrigated. However, except for urea, most input subsidies were phased out or substantially reduced during the 1990s. Bangladesh inherited upon independence a complex system by which fer- tilizers, pesticides, seeds, and irrigation (tubewell) equipment were supplied to farmers at subsidized prices by monopoly government organizations. With the important exception of the fertilizer, however, these subsidies and their accom- panying controls were withdrawn in the late 1970s. Since then, the principal farm input subsidy has been for urea: subsidies for nonnitrogenous fertilizers, all of which are imported, were abolished in 1991 but reintroduced in 2005. In Sri Lanka, a canal irrigation system built by the government starting in 1979 supplies water at subsidized prices. Sri Lanka also provides subsidies for fertiliz- ers, seeds, extension, and research, the largest amount of which goes to fertilizer subsidies. 396 Distortions to Agricultural Incentives: A Global Perspective Trade and exchange rate policies: India For 45 years after independence, India followed restrictive trade policies. During the 1950s and 1960s, agricultural exports were heavily controlled. When exports were allowed, they were subject to high export taxes, for example on jute and jute products, oilcakes, cotton, tea, and black pepper. Nearly all imports were either subject to discretionary import licensing or were "canalized" by monopoly gov- ernment trading organizations. Import licensing was regularly tightened in response to the steadily worsening foreign exchange situation, and tariffs were increased and reached very high levels by early 1966. The period's policies were characterized by a marked antiagricultural bias, which probably increased along with the rising overvaluation of the exchange rate and the countermeasures for industry that concentrated on providing higher incentives to manufacturers, both in the domestic market and in export markets, while attempting to keep agricul- tural prices low and stable. In June 1966, the devaluation of the rupee was accompanied by a brief liberal- ization episode during which import licensing was relaxed, tariffs were cut, and export subsidies were abolished or reduced. However, the import licensing system remained intact and by 1968, most of the liberalizing initiatives had been reversed and tight import and domestic controls reinstated. This remained the situation until the end of the 1970s, when a new phase of very slow partial liberalization commenced. During these years, inflation was steadily reduced, and by 1980 the real effective exchange rate (REER) had declined by 46 percent (figure 10.1). A balance of payments crisis was averted in 1980 and 1981 and the real value of the rupee maintained with the help of an International Monetary Fund loan, but from about April 1985 a new policy commenced under which the currency was steadily devalued in real terms. This trend continued uninterrupted for the next six years, almost on a monthly basis, until a sharp devaluation was imposed in July 1991. This was followed by more than a year of further depreciations, until September 1992. Over the whole period between 1985 and 1992, the rupee was devalued in real terms by around 145 percent (figure 10.1). The devaluations of the rupee radically changed, among other things, the envi- ronment for India's trade policies. They also had important repurcussions for manufacturing, the first sector to be exposed to liberalization policies, and in making the trade liberalization program started in 1991 quite painless. In addi- tion, many Indian manufacturing firms that previously felt vulnerable to import competition now found, following the correction of the earlier exchange rate overvaluation, they could not only easily compete with foreign manufacturers' imports but could outcompete foreign manufacturers in export markets. Com- bined with sweeping new domestic deregulation of manufacturing that accompa- nied the 1991 trade policy reform program, this created new momentum in the India and Other South Asian Countries 397 Figure 10.1. Real Effective Exchange Rate Index, India, 1964­2007 300 1980 = 100; an increase is a devaluation 250 200 150 100 50 0 19 4 19 6 19 8 70 19 2 19 4 19 6 19 8 19 0 19 2 19 4 19 6 19 8 19 0 19 2 19 4 19 6 20 8 20 0 20 2 20 4 06 6 6 6 7 7 7 7 8 8 8 8 8 9 9 9 9 9 0 0 0 19 19 Source: Authors' calculations based on official data. manufacturing sector in terms of investment, productivity improvements, and output expansion. A four-year program of tariff reductions also commenced: this brought tariffs down to levels far below the extremely high or prohibitive rates (averaging over 100 percent) of the 1980s, even though tariffs were still very high by international standards. But both domestic and trade policies affecting the rural sector were basically untouched by the 1991 reforms. Since 1992, the exchange rate has been managed by regular adjustments of the nominal rates (figure 10.1). However, as of the mid 1990s--five years after the 1991 reforms--about two-thirds of tradable GDP was still protected by some type of explicit nontariff barrier: about 36 percent of manufacturing, 84 percent of agriculture, and 40 percent of mining. During the second half of the 1990s, this began to change, in large measure as a response to international pressures linked to agreements and negotiations associated with the conclusion of the Uruguay Round of the General Agreement on Tariffs and Trade. Starting in 1998, India's general import licensing system was gradually dismantled, and on April 1, 2001, the last 715 of 2,714 tariff lines (including nearly all the agricultural tariff lines) were removed and the system itself was abolished. After the lifting of the import licensing controls, existing tariffs proved more than adequate to keep out competing imports, both manufactured and agricul- tural. At the same time, manufactured exports entered a new phase of very rapid expansion (20­25 percent annually during 2001­08), a trend that was bolstered by 398 Distortions to Agricultural Incentives: A Global Perspective similarly fast growth of services exports. Together with increased capital inflows, these developments created a strong balance of payments and historically high foreign exchange reserves (more than US$300 billion by August 2008), and were accompanied by fast economic growth (almost 9 percent per year during 2005­08). In response to the new confidence that these changes created, drastic reduc- tions in industrial tariffs were begun in April 2003. Over the next four years, the average industrial tariff was reduced by approximately two-thirds, from over 33 percent to about 12 percent (figure 10.2). A further tariff reduction was launched in the 2007 budget. All told, as measured by average ad valorem indus- trial tariffs, these cuts allowed India to shift from being one of the world's most protected countries to one of the world's lowest-protection countries. The tariff reduction programs of the 2000s, however, did not include agriculture and processed foods. In 2006, unweighted average tariffs protecting these sectors (Harmonized System 01­24) were about 40 percent (figure 10.2), almost four times the level of India's average industrial tariffs and among the highest in the world. This high level of formal agricultural protection, combined with high input subsidies, allows considerable scope for the past antiagricultural bias of the system to move to a pro-agricutural policy bias. The situation on the agricultural Figure 10.2. Unweighted Average Tariffs on Imports of Agricultural and Nonagricultural Goods, India, 2002­06 50 45 40 35 30 percent 25 20 15 10 5 0 2002 2003 2004 2005 2006 all tariff lines agriculture harmonized system 1­24 nonagriculture harmonized system 25­99 Source: Authors' computation based on data in Goyal (2007­08 and earlier years). Note: Tariffs for 2002 and 2003 include para-tariffs that were abolished in 2004. Averages are of ad valorem tariffs only and do not account for specific components of compound duties. India and Other South Asian Countries 399 tariff front, however, changed dramatically in 2007­08, when global commodity prices shot up: India lowered tariffs on edible oils, grains, and other food com- modities and imposed export controls on rice, wheat, and corn. Trade and exchange rate policies: Pakistan, Bangladesh, and Sri Lanka Because of their common history as British colonies, it is not surprising that there are many similarities between the trade and exchange rate policies of Bangladesh, India, Pakistan, and Sri Lanka. Soon after independence, Pakistan adopted an import substitution strategy that protected manufacturing firms against imports behind tariff and nontariff barriers. Sri Lanka followed similar highly interven- tionist trade policies. When it seceeded from Pakistan in 1971, Bangladesh gave special emphasis to the import substitution policies it inherited, since up to that point few manufacturing activities had been established in its territory. Like India, Pakistan and Sri Lanka fixed their nominal exchange rates during the 1960s but inflation rapidly appreciated their real exchange rates. Both countries attempted to manage the resulting current account pressures through schemes that raised the rupee price of foreign exchange for manufactured exports and for remittances from nationals working in foreign countries. However, the "traditional" agricul- tural exports that accounted for the largest share of total export earnings (about 90 percent in Sri Lanka) were subject not only to the overvalued official rate, but also to export controls and high export taxes. In Pakistan, trade policies strongly discriminated against rice and cotton and, prior to 1971, against jute and tea, which were produced in East Pakistan. In Sri Lanka, policies discriminated against the principal export crops--tea, rubber, and coconut products. In Pakistan, a big devaluation of the rupee in 1971/72--by about half in relation to the U.S. dollar--reduced growing foreign exchange distortions for some time. But during the rest of the 1970s, domestic inflation continued to appreciate the real exchange rate and erode export competitiveness. This changed in 1981, when a long series of nominal devalutions that consistently exceeded the excess of Pak- istan's domestic inflation over average inflation in its principal trading partners commenced. Real devaluation (as measured by Pakistan's REER index), which occurred in nearly every year between 1981 and 2004 and totalled about 137 per- cent over the 23 years, was particularly steep between 1984 and 1990 (about 67 percent).7 As in India, this long-term devaluation underpinned sweeping trade liberalization reforms during the 1990s that removed most quantitative restric- tions and drastically cut tariffs. However, Pakistan has retained its "positive list" system (a by-product of its difficult political relations with India) which prohibits imports from India of products not on the list. Combined with rules (enforced 400 Distortions to Agricultural Incentives: A Global Perspective by both countries) against trade across the land border and bureaucratic obstacles to trade with Pakistan on the Indian side, bilateral trade--especially of agricultural products--has been reduced to very low levels compared with its potential. During Bangladesh's early independence years after 1971, the taka exchange rate initially reflected the simultaneous devaluation of the Pakistani rupee. But the general political, social, and economic disruption caused by the war surrounding the secession was exacerbated by drought and floods and led to high inflation. Despite continuing use of quantitative import restrictions and high tariffs, this created serious problems for the tradable sectors, prompting the government to respond in 1975 with a very large taka devaluation, estimated at about 66 percent in real terms (Rahman 1994). The taka continued to be devalued until 1980, in real terms by about another 16 percent. Since 1980, in marked contrast to India and Pakistan, Bangladesh has managed its exchange rate so as to keep the REER index quite stable around a slowly devaluing long-term trend of just under 1 per- cent per year. During the 1980s, this was sufficient to ensure that the secondary rate did not diverge substantially from the official rate, and made it possible to gradually merge the two rates and unify them in January 1992. Bangladesh's action was also sufficient to support a sweeeping trade liberalization program that started slowly in the second half of the 1980s and accelerated between 1991 and 1996 with drastic tariff reductions and the removal of most quantitative restric- tions. This was possible principally because of the expansion of foreign exchange earnings from workers' remittances and garment exports. However, further trade liberalization stalled and went into reverse starting in 1997, with increasing use of para-tariffs on top of customs duties and tariff escalation to protect import sub- stitution manufacturing and a number of agricultural products. In Sri Lanka, from the late 1950s to 1977, under the government's import sub- stitution strategy, exchange controls, tariffs and nontariff barriers, and a formal dual exchange rate system after 1968, supported the official exchange rate at increasingly overvalued levels. This was followed by a major trade policy liberal- ization in 1977 that included a nominal devaluation of about 75 percent, abolition of the dual exchange rate system, substantial reductions in quantitative restric- tions, and tariff protection of manufacturing and of some agricultural industries. Export taxes on plantation crops were reduced during the 1980s and eventually removed in 1992. A second wave of trade liberalization started in 1991/92 and continued--albeit erratically and with some backtracking--during the 1990s and into the 2000s. As in Bangladesh, this was made possible by the rapid and sus- tained expansion of garment exports, which now account for about 60 percent of total exports, compared with less than 10 percent before the 1977 reforms. Cur- rently, Sri Lanka has an open trade policy regime without nontariff barriers and India and Other South Asian Countries 401 with moderate tariff-based protection of agriculture approximately equivalent to the tariffs protecting the import-competing sectors of manufacturing. Within agriculture, the most intervention-prone products are rice and several other import-competing food products, notably potatoes, onions, and chilies. Input subsidies for agriculture have also continued, especially very substantial fertilizer subsidies. Regional agricultural trade A by-product of India's highly protective agricultural trade policies is that trade in primary and processed agricultural and livestock products between India and its South Asian neighbors has been badly hindered.8 Ironically, the biggest hindrance very likely has been to Indian exports to these countries rather than to Indian imports from the region. However, in Bangladesh and Sri Lanka, it also reflects a realistic assessment that agricultural free trade with India would generate more agricultural imports from India than exports to India, in the process threatening the viability of some of these countries' more highly protected agricultural indus- tries, such as sugar, various fresh fruits and vegetables, and a wide range of processed foods in Bangladesh; and rice, potatoes, onions, and possibly dairy products in Sri Lanka. Bilateral trade between India and Pakistan is also hostage to the two countries' difficult political relationship, as reflected in Pakistan's "positive list" of products that can be legally imported from India. This list includes almost no agricultural products, and rules enforced in both countries (with a few minor exceptions) do not allow trade over the land border.9 Long-Run Trends in Assistance to Agriculture The main focus of the present study's methodology is on government-imposed distortions that create a gap between actual domestic prices and what they would be under free markets. Since it is not possible to fully comprehend the character- istics of agricultural development with a sectoral view alone, the project's methodology estimates both the effects of direct agricultural policy measures (including distortions in the foreign exchange market) and distortions in non- agricultural sectors for comparative evaluation. Specifically, NRAs are computed for farmers including any input subsidies and non-product-specific (NPS) forms of assistance or taxation. A production-weighted average NRA for nonagricultural tradables, for comparison with that for agricultural tradables via the calculation of a relative rate of assistance (RRA; see Anderson et al. 2008a, 2008b), is also gen- erated. Figures 10.3 to 10.6 show long-run trends in aggregate NRAs for agricul- ture, aggregate NRAs for manufacturing (plus mining, in the case of India), and Figure 10.3. NRAs to all Agricultural Tradable Industries, All Nonagricultural Tradables, and RRAs,a India, 1965­2004 120 80 40 percent 0 40 80 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 01 03 05 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Pursell, Gulati, and Gupta (2009). a. The RRA is defined as 100*[(100 NRAagt ) (100 NRAnonagt ) 1], where NRAag t and NRAnonag t are the percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respec- tively. The 1965­69 manufacturing and mining NRAs are guesstimates. Figure 10.4. NRAs to all Agricultural Tradable Industries, All Nonagricultural Tradables, and RRAs, Pakistan, 1973­2005 70 50 30 10 percent 10 30 50 70 73 75 77 79 81 83 85 87 89 91 93 95 97 99 01 03 05 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Dorosh and Salam (2009). a. The RRA is defined in the note to figure 10.3. 402 Figure 10.5. NRAs to All Agricultural Tradable Industries, All Nonagricultural Tradables, and RRAs, Bangladesh, 1974­2004 100 80 60 40 percent 20 0 20 40 60 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Ahmed, Bakht, Dorosh, and Shahabuddin (2009). a. The RRA is defined in the note to figure 10.3. Figure 10.6. NRAs to all Agricultural Tradable Industries, All Nonagricultural Tradables, and RRAs, Sri Lanka, 1955­2004 200 160 120 80 percent 40 0 40 80 19 5 19 7 19 5 19 7 59 19 1 19 3 69 19 1 19 3 19 5 19 7 19 9 19 1 19 3 19 5 19 7 19 9 19 1 19 3 19 5 19 7 20 9 20 1 20 3 05 5 5 6 6 6 6 7 7 7 7 7 8 8 8 8 8 9 9 9 9 9 0 0 19 19 19 NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Bandara and Jayasuriya (2009). a. The RRA is defined in the note to figure 10.3. 403 404 Distortions to Agricultural Incentives: A Global Perspective aggregate RRAs over 1965­2005 for the four South Asian countries included in the study. NRAs Table 10.3 shows the agricultural sector products for which there are long-term NRA estimates and that are included in the country aggregations. At undistorted prices, on average during 2000­04, these products accounted for about 70 percent of the total value of Indian agricultural production. Following this project's methodology, the total includes all crops and livestock production but excludes fisheries and forestry. At the end of the four-year period, fruit and vegetables accounted for about one-third of the value of production in the Indian sample, raw milk for about one-fifth, paddy rice for 17 percent, wheat for 9 percent, and the other nine crops (sorghum, maize, pulses, four types of oilseeds, sugar cane, and seed cotton) for the remaining 20 percent. In the 1960s and early 1970s, wheat and rice represented more than half the sample and fruit and vegetables about a quarter. Because the agricultural sectors in Bangladesh, Pakistan, and Sri Lanka are much less diversified than India's, a small number of products account for a large share of total agricultural production. Long-term NRA estimates examined here include seven products that accounted for 72 percent, 71 percent, and 64 percent, respectively, of the total value of production of their agricultural and livestock sectors during 2000­05. In Pakistan, the dominant product is wheat, followed by milk--together, these are generally just under half of total agricultural production and two-thirds of Pakistan's aggregate NRA time series. In Bangladesh, the domi- nant product is rice, which is generally 55­60 percent of the value of agricultural production and represents three-quarters to four-fifths of Bangladesh's aggregate NRA time series. The other five products in the Bangladesh sample (wheat, sugar cane, fresh vegetables [represented by potatoes], jute, and tea) together are only around 15 percent of agricultural production and have about a 20 percent weight in the aggregate NRA series.10 Table 10.3 also indicates the tradable status of the covered products. Over the 40 years analyzed for India, four of the 13 products were importable or nontrad- able; two were either exportable or nontradable; and the remaining seven fluctu- ated between importable, exportable, and nontraded status. However, the country studies for Bangladesh, Pakistan, and Sri Lanka treated all the products included in their long-run NRA samples as either importable or exportable over the whole period, with the exception of potatoes in Bangladesh (representing fresh vegeta- bles), which is classified as nontradable in every year. These differences in the treatment of tradable status in the case studies--and therefore in the way NRAs are estimated--complicate comparisons of NRAs among the four countries.11 India and Other South Asian Countries 405 Table 10.3. Trade Status of Farm Commodities, South Asian Countries,a 1965­2005 India Pakistan Bangladesh Sri Lanka Common rice/paddy X/NT X M M Basmati rice/paddy X Wheat M/X/NT M M Maize M/X/NT M Sorghum M/X/NT Pulses (chickpeas) M/NT Groundnuts M/X/NT Rape/mustard seeds M/NT Soya beans M/X/NT Sunflower seeds M/NT Sugar/sugar cane M/X/NT M M Cotton lint/seed cotton M/X/NT X Jute X Tea X X Rubber X Coconut/coconut products X Chilies M Potatoes NT M Onions M 6 fresh fruits and 7 vegetables X/NT Processed and raw milk M/NT M % coverage, 2000­04 or 2000­05 70 72 71 64 Source: Authors' compilation, based on the country case study spreadsheets. a. M, X and NT indicate whether the product was classified as importable (M), exportable (X), or as a nontraded tradable (NT). In Pakistan, Bangladesh, and Sri Lanka all the covered products were classi- fied as M or X for the entire period (except for potatoes in Bangladesh, which were classified NT and asumed to represent all vegetables). The India study recognized that tradable status can change from year to year depending on the location of domestic prices with respect to import and export parity prices, so for example M/X/NT means that wheat was importable in some years, exportable in some years, and nontraded in others. The Bangladesh and Sri Lanka country studies do not allow for the port and domestic handling costs of imports and exports and compare domestic prices directly with cif or fob prices. Blank cells mean that a long term NRA series is not available for that product. The coverage percentages are the share of the total value of the output of the covered products in the country's total agricultural output, both measured at undistorted prices. The total value of agricultural output is at farm level and excludes fisheries and forestry activities. Including fisheries in total agricultural production would substantially reduce the Bangladesh coverage percentage. In addition to output price distortions, the aggregate long-term NRA series in figures 10.3 to 10.6 also include the output price equivalent of input subsidies expressed as a percentage of the undistorted price. In India, these are the sum of fertilizer and electricity subsidies, for which estimates have been available since 1984. They have been allocated to the various crops in the manner summarized in table 10.4. The Pakistan and Bangladesh NRA estimates include just fertilizer 406 Table 10.4. NRAs, TBIs and Dispersion of Covered Farm Products,a South Asia, 1965­2004 (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 India NRA, importables 41 53 74 59 82 38 23 34 NRA, exportables 30 22 36 28 7 15 12 6 NRA, all agricultural products 0 0 6 2 25 2 1 16 TBIb 0.51 0.50 0.63 0.55 0.48 0.38 0.28 0.29 Dispersionc (13 products) 43 18 7 19 49 47 13 24 Pakistan NRA, importables 45 19 4 2 5 8 2 3 NRA, exportables 35 20 33 29 32 17 4 7 Total agricultural NRA 15 7 8 6 4 7 2 1 TBIb 0.55 0.27 0.31 0.28 0.35 0.10 0.02 0.09 Dispersionc (7 products) 106 75 43 50 65 32 28 40 Bangladesh NRA, importables -- 21 7 2 24 0 8 6 NRA, exportables -- 29 35 26 32 33 10 33 Total agricultural NRA -- 21 3 4 17 2 8 4 TBIb -- 0.10 0.30 0.23 0.45 0.33 0.00 0.37 Dispersionc (6 products) -- 52 71 68 191 78 68 101 Sri Lanka NRA, importables 6 9 4 1 2 22 32 13 NRA, exportables 39 41 45 31 21 24 2 6 Total agricultural NRA 25 16 26 14 10 1 12 9 TBIb 0.35 0.45 0.43 0.31 0.18 0.38 0.25 0.05 Dispersionc (7 products) 21 32 25 23 22 25 21 13 Source: From Anderson and Valenzuela (2008), based on the country case study spreadsheets. Note: -- not available. a. The Pakistan statistics in the last column are the averages for 2000­05. The statistics for Bangladesh in the 1970­74 column are for 1974 only. The NRA for total agriculture include guesstimates of the average NRA of the noncovered section of the agricultural sector (which are not based on explicit price comparisons). Because of differences in the way importables and exportables are defined the NRAs and TBIs for India cannot be compared with the TBIs for the other countries (see notes to table 10.3). b. The trade bias index, TBI (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector. c. Dispersion of the NRAs of the covered products (including products classified as nontraded) is the simple five-year average of the annual standard deviation around the weighted mean NRA of all covered products (in other words, importables, exportables and nontradeds). 407 408 Distortions to Agricultural Incentives: A Global Perspective subsidies. Sri Lanka's estimates, however, could not be quantified and allocated to the covered crops. There are a number of striking features of the aggregate NRA time series. First, in Bangladesh, India, and Pakistan, they average close to or just slightly below zero over the whole period, albeit with large fluctuations. Sri Lanka is different, with negative average NRAs up to 1993. The zero or negative NRAs in each of the four countries are despite pervasive nontariff barriers to imports--especially during the earlier years--and high tariffs continuing into the 2000s. Second, in Bangladesh, India, and Pakistan, there is practically no trend in the agricultural NRAs over the four decades surveyed, while in Sri Lanka there is a slowly increasing trend from about the late 1970s. This is despite a long-term, very substantial decline in world agricultural prices during the same period. Third, in all four countries during most periods, NRAs for importables have been positive, while NRAs for exportables generally have been negative, so that the trade bias indexes (TBIs) for exportables over the period have mostly been nega- tive and quite high (table 10.3). This is consistent with the policies followed until the early 1990s, whereby export earnings from traditional agricultural exports were exchanged at overvalued official exchange rates and also subject to export taxes, and only nontraditional exports (in practice manufactures) received more favor- able exchange rates or were eligible for export subsidies of various kinds. Fourth, in all four countries the NRAs fluctuate widely around their trend val- ues. These fluctuations are mainly due to large gyrations in international prices combined with largely successful efforts by South Asian governments to stabilize domestic prices. In India, for example, the NRAs were lowest in 1974 when inter- national agricultural prices were at record highs, and highest around 1987 when (in real terms) international agricultural prices were at record lows. For most of the past four decades, as in 2007­08, export restrictions have generated an implicit export tax that has varied substantially as international prices have moved up. Fifth, the dispersion of NRAs across the covered products is quite wide in each of the countries (table 10.3). It is highest in Bangladesh and Pakistan, mainly due to the contrast between the high protection of sugar cane in both countries and (pre-1990) of milk in Pakistan, and also the export taxes and hence negative NRAs of these countries' exportables. In India, NRA dispersion is affected by high pro- tection of sugar cane, rape and mustard seeds, and (until the mid-1990s) dairy (milk). In Sri Lanka, the extent of the dispersion is mainly due to high protection of import substitution potatoes, onions, and chilies contrasted with the taxation of agricultural exports. No obvious long-term trend in dispersion is apparent in Bangladesh and India, but since about 1990 previously very high NRA dispersion seems to have come down in Pakistan, and was markedly lower in Sri Lanka dur- ing 2000­04 than in previous years. India and Other South Asian Countries 409 Sixth, the contribution of covered products from fertilizer and electricity sub- sidies to the NRA has steadily increased in India. During 2000­04, subsidies contributed almost 10 percentage points to an average NRA of 16 percent. The Pakistan country study estimates that after 1990, the fertilizer subsidy added about 3 percentage points to the NRAs for wheat, paddy rice, cotton, and sugar cane, compared to about 7 percentage points before 1990 when non-nitrogenous fertilizers were also subsidized. The estimated contribution to NRAs in Bangladesh appears to be considerably lower, however, with only about 1 or 2 per- centage points added to the wheat and potato NRAs. RRAs The incentives facing farmers depend not only on the agricultural NRAs but also on how trade and other price-distorting policies affect incentives facing producers in other tradable sectors. In order to see how relative incentives have evolved in the four countries, figures 10.3 to 10.6 compare the NRAs for agriculture (the middle line in each graph) with estimates of NRAs for nonagricultural tradables (the top line). Comparing these gives the RRA, or the percentage difference between the agricultural NRA and the nonagricultural NRA,12 shown (except for a few years in India) as the bottom line in each graph. Five-year averages of each country's RRAs are shown in figure 10.7. Figure 10.7. RRAs, South Asia, 1965­2004 20 10 0 10 20 percent 30 40 50 60 70 80 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 India Pakistan Bangladesh Sri Lanka Source: From Anderson and Valenzuela (2008), based on the country case study spreadsheets. a. The RRA is defined in the note to figure 10.3. 410 Distortions to Agricultural Incentives: A Global Perspective The nonagricultural NRA series for India is a weighted average of estimated NRAs for manufacturing and mining. In Pakistan, it covers all of nonagriculture tradables (mainly manufacturing), while in Bangladesh and Sri Lanka just manufac- turing. The services sectors of the four countries are assumed to be nontradable. Because estimating a long-term assistance rate series for a country's manufacturing and other tradable sectors would be a major research task in itself, the reliability of the estimates for these four countries depends on the availability of prior empiri- cal research and the plausibility of the shortcuts used. In this regard, a major problem in South Asia is that owing to their often prohibitively high levels and to pervasive quantitative trade restrictions in the past, tariffs are not reliable guides to protection levels, especially for manufacturing. Though the India case study was able to use the results of previous detailed research on trends in implicit man- ufacturing protection (Pursell, Kishor, and Gupta 2007) combined with new esti- mates for India's mining sector, estimates for the other three countries are more problematic, particularly those for Bangladesh and Sri Lanka, which rely on infer- ences from changing tariff levels plus the assumption that the NRAs of manufac- tured exports were zero over the entire period. Consequently, the nonagricultural NRA series and hence the RRAs are at best rough approximations that could change with further research. Despite these shortcomings, the India and Pakistan nonagricultural NRA series are broadly consistent with the known history of these countries' trade policies, including periodic devaluations, long-term trends in their real exchange rates, and trade policy liberalization episodes. In India, in particular, the downward trend of India's nonagricultural NRA from very high levels in the 1960s was associated with the long-term real rupee devaluation between 1966 and 1979 and between 1985 and 1992, and the trade policy liberalizations that began in the 1980s and continued into the 2000s (figure 10.1). The Pakistan series (figure 10.4) reflects the sharp devaluation in 1971 that more than doubled rupee border prices and correspondingly cut estimated implicit protection, the long-term real devaluation that started in 1981, and the trade liberalization reforms of the 1990s. Nonagricul- tural NRAs trended down in Bangladesh (from 1994) and in Sri Lanka (from 1986), consistent with the timing of import policy liberalization and both coun- tries' rapidly growing garment export sectors. However, in other respects, the con- nections between these series and real exchange rate changes and trade policy developments in the two countries is less apparent and could benefit from further research.13 Subject to these caveats, the South Asian RRAs tell some interesting stories. First, for all four countries, they are negative in all years except the last four in India, indicating that incentives for these countries' rural sectors have been less than incentives for their nonagricultural tradable sectors. Second, the RRAs India and Other South Asian Countries 411 confirm earlier research indicating that up to the mid-1980s there were very high antiagricultural biases in the RRAs (here estimated at between minus 40 to minus 60 percent) for India, Pakistan, and Sri Lanka, and expand this result to include Bangladesh, where, according to these estimates, there was a similar very high antiagricultural bias during the first half of the 1970s and a negative but less marked antiagricultural bias into the mid-1980s. Third, in India, Pakistan, and Sri Lanka, there has been a clear, long-term decline in antiagricultural bias: by the first five years of the 2000s, the RRA indicator for India had turned slightly posi- tive (about +10 percent), while those for Pakistan and Sri Lanka RRA were only mildly negative (about 12 percent on average). There are no equivalent long-run trends in the Bangladesh series. According to these estimates, the antiagricultural bias was still very high during the 1990s (between 20 and 36 percent), but this too came down between 1999 and 2004, to an average of about 16 percent. Fourth, in India and Pakistan, the long-run downward trends in nonagricultural NRAs--especially falling manufacturing protection rates--have been by far the main force squeezing out the antiagricultural bias. As discussed above, over the 41 years of the study there was no clear long-term upward trend in these coun- tries' agricultural NRAs. However, the pattern was different in Sri Lanka, where the long-term increasing (less negative) trend of the RRAs resulted from both declining manufacturing NRAs and increasing (less negative) agricultural NRAs. Finally, the absence of any clear long-term trend in Bangladesh's consistently neg- ative RRAs is the result of approximately trendless NRAs for both agriculture and manufacturing. The latter is a somewhat unexpected result in view of the fairly comprehensive trade policy liberalization that occurred during the late 1980s and early 1990s, and the rapid expansion of Bangladesh's garment export industry. Where Are South Asian Agricultural Policies Heading? Between 2000 and 2005, RRA indicators in South Asia indicate that, with the exception of Bangladesh, past antiagricultural discrimination created by trade and other policies evolved to approximate neutrality between the agricultural and nonagricultural traded sectors. What is the likely direction of future poli- cies in the region? Since what matters is relative protection or assistance, the answer to this question depends on the probable direction of the trade and trade-related policies affecting these countries' nonagricultural (especially man- ufacturing but also mining and, increasingly, internationally traded services) and agricultural sectors. Regarding nonagricultural trade policies, protection levels in the manufactur- ing and mining sectors in India are now constrained by low tariffs. Even in the few industries that are still protected by high tariffs--such as textile fabrics, garments, 412 Distortions to Agricultural Incentives: A Global Perspective and automotive assembly--growing exports and domestic competition suggest that it is unlikely prices will rise much above world prices in the foreseeable future. India's large, low-cost export-oriented services sector is highly competitive internationally, a scenario that is also not likely to change in the short term. In Bangladesh, Pakistan, and Sri Lanka, while nearly all manufacturing quantitative restrictions have been removed, tariffs protecting import-substituting manufac- turing are on average higher than in India. This is especially the case in Bangladesh, where, since about 1997, the increasing use of para-tariffs on top of customs duties and steep tariff escalation is providing very high nominal and effective protection for many import-substitution industries. However, these countries have large, rapidly growing export industries, notably textiles and cloth- ing in Pakistan and clothing (ready-made garments) in Bangladesh and Sri Lanka. These export-oriented industries account for much larger shares of total manu- facturing GDP than do India's manufactured exports--about 40 percent in Pakistan, and probably a quarter to a third in Bangladesh and Sri Lanka. As long as the export expansion of these industries continues and is not slowed down or blocked by restrictive import policies, especially in developed countries, it seems unlikely that overall manufacturing protection will increase in the future, at least in Pakistan and Sri Lanka. The likely future direction of average manufacturing sector NRAs in Bangladesh is less certain, however, in view of the protectionist import-substitution policies that have been in place for the past 10 years. If South Asia's nonagricultural trade policies do not become more protective in the future, the trajectory of relative assistance for agriculture will depend princi- pally on the future path of agricultural protection and subsidy policies. The polit- ical economy surrounding this issue is complex, with some forces and arguments making it likely that a protectionist path for agriculture will be followed, and oth- ers constraining this kind of development. Politically, important considerations that favor protection over open trade policies are the very high share of employment in the South Asian rural sector, the desire to insulate farmers from the large price fluctuations that occur in world agricultural markets, and the feeling that each country should be self-sufficient (or nearly self-sufficient) in the production of basic foods and other agricultural commodities. The surge in global food prices in 2007­08, during which export bans of staple food products were instituted by some countries, served to reinforce this feeling of self-sufficiency. In India, the argument for self-sufficiency, and thereby protection, is further reinforced by the widely shared belief that Indian demand is too large to rely on world markets for supplies in the event of serious crop failures or other disruptions to supplies. In Sri Lanka, agricultural protection has an ethnic dimension, as past agricultural trade liberalization mostly affected Tamil farming areas and protection and subsidies favored Sinhalese farmers. India and Other South Asian Countries 413 In India, the political economy forces enumerated above are the basic reasons for the exclusion of agriculture and the food processing industries from trade lib- eralization reforms in 1991, for the fixing of very high tariff bindings (the major- ity at 100 percent or 150 percent) during the Uruguay Round negotiations, and for leaving agriculture out of the unilateral tariff reduction program that began in 2003. Most of the Uruguay Round Agreement on Agriculture tariff bindings of Pakistan (100 percent) and Bangladesh (200 percent) are also prohibitive or almost prohibitive: only Sri Lanka's bindings (all at 50 percent) seem to envisage limiting increases in applied tariffs to levels at which imports might be possible. But average applied agricultural tariffs in Pakistan and Sri Lanka (23 percent and 28 percent, respectively, in 2003) are well below the average in India (40 percent in 2005). However, because of the very high share of food in South Asian family budgets, there are strong pressures to keep agricultural prices low. For many years, this was a major objective of agricultural policies and was compatible with expanding pro- duction and increasing national self-sufficiency, largely resulting from successful adoption of green revolution technologies in crop agriculture. There were further benefits to low-income households from subsidized rice and wheat supplied through the PDS system in India and equivalent schemes in Bangladesh, Pakistan, and Sri Lanka. There are no organized groups in South Asia representing the interests of food consumers, like those that represent farmers and food processors. Nevertheless, in all the South Asian countries politicians and bureaucrats are aware of and sensi- tive to the consumer interest in food prices, especially mass staple products. In this way, consumer interests remain important counter-forces to producer lobbies pressing for higher agricultural prices. However, it appears that consumer inter- ests are unlikely to provide much resistance to increasing agricultural protection if domestic and external conditions create strong producer pressures in that direc- tion. Medium- or long-term scenarios favoring increasing protection could include the following elements: domestic production of major crops such as rice and wheat falling behind domestic demand, resulting in pressure for price increases or increases in input subsidies to maintain self-sufficiency; falling world prices but (in real terms) stable or even slowly increasing domestic prices; real exchange rate appreciation reducing national currency border prices while domestic prices remain about the same or slowly increase; and no or limited progress in reducing agricultural protection in developed countries in World Trade Organization (WTO) negotiations. Scenarios that might reduce or slow down pressures for increased agricultural protection include a long-term trend of increasing and more stable world prices. In this regard, the medium- and long-term outcomes of the surge in world 414 Distortions to Agricultural Incentives: A Global Perspective agricultural commodity prices during 2007 and 2008 will be very important. If world prices stabilize at significantly higher levels than past averages, South Asian countries are likely to reduce tariffs on agricultural products, or even impose export bans (as India has done on common rice, wheat, and corn), thereby rein- troducing substantial antiagriculural bias into their incentive systems. Other sce- narios that might work against increased agricultural sector protection include yield and other productivity increases--especially productivity increases in trans- port, storage, and marketing--or successful WTO negotiations on the reduction of developed-country protection and subsidies (especially export subsidies and domestic support) leading to greater willingness in South Asia to consider more open agricultural trade policies. Notes 1. These four countries account for more than 98 percent of South Asian agricultural GDP. The region's smaller economies of Nepal, Bhutan, and the Maldives were not included in the project, nor was Afghanistan. 2. Here and in other places in this chapter, the term "agriculture" is used broadly to include not only crop agriculture (including horticulture) and livestock activities- the focus of the global Agricul- tural Distortions Project--but also inland and ocean fisheries and forestry activities. In the tables and figures reporting nominal rate of assistance (NRA) estimates and in most of the rest of the text, the word "agriculture" means crop agriculture, horticulture, and livestock activities. Whether the broad or narrower meaning of the term is intended should be apparent from the context. 3. Unless otherwise indicated, national fiscal years are measured as in the following examples: India 1997 fiscal 1997/98 (April 1, 1997­March 31, 1998); Pakistan and Bangladesh 1997 fiscal 1996/97 (July 1, 1996­June 30, 1997); and Sri Lanka 1997 fiscal 1997 (January 1, 1997­December 31, 1997). 4. Bangladesh became independent from Pakistan in 1971 but separate data on the Bangladesh agricultural sector were not available until 1974. Before 1971, limited data and NRA estimates on agri- culture in the then East Pakistan are included in the Pakistan country study. In particular, there is no information on rice production, NRAs, or policies in East Pakistan. 5. Indian palm oil and other edible oil tariffs were drastically reduced in 2008 in order to stabilize domestic prices during the global spike in commodity prices. 6. According to these statistics, the overall agricultural employment share of the four countries during 2000-04 was 57 percent, but somewhat surprisingly, compared to India, the agricultural employment share was lower in Bangladesh (54 percent) and much lower (46 percent) in Pakistan. These contrasts between the South Asian countries' agricultural employment rates are very likely to a large extent the result of differences in the design of national employment surveys. 7. For a discussion of real exchange rate changes in South Asia after 1980, see World Bank (2004). 8. Two exceptions are Bangladesh's rice imports from India and India's imports of raw jute from Bangladesh, both of which are subject to low tariffs. The bilateral rice trade was interrupted in 2008, however, as a result of export restrictions India imposed on rice in order to insulate its domestic mar- ket from sharp increases in world rice prices. 9. For a discussion of the constraints on and potential of Indian-Pakistan bilateral trade see Naqvi and Schuler (2007). 10. Generalizations from this product sample to Bangladesh's rural sector as a whole need to be qualified, owing to the importance of fisheries (omitted from this study) in Bangladesh. 11. Another difference is that in the India and Pakistan case studies, these price comparisons start with estimated free on board (fob) and cost, insurance, and freight (cif) prices at the border, which are India and Other South Asian Countries 415 adjusted for port costs and domestic transport costs and margins and give import or export reference prices. However, the NRAs reported in the Bangladesh and Sri Lanka studies compare domestic prices directly with cif or fob prices without accounting for port or domestic handling costs and margins. 12. The RRA is defined as 100*[(100 NRAag) (100 NRAnonag) 1]. 13. For example, the 1977 devaluation and trade policy reforms are not reflected in the Sri Lanka series, while the Bangladesh NRA series increases during the trade policy reform period in Bangladesh under which quantitative restrictions were removed and tariffs were reduced during the late 1980s and the first half of the 1990s. References Ahmed, N., Z. Bakht, P. A. Dorosh, and Q. Shahabuddin. 2009. "Bangladesh." In Distortions to Agricul- tural Incentives in Asia, ed. K. Anderson and W. Martin, 305­37. Washington, DC: World Bank. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC (and appendix A of this volume). ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson, K., and W. Martin, eds. 2009. Distortions to Agricultural Incentives in Asia. Washington, DC: World Bank. Anderson, K., and E. Valenzuela. 2008. Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007. Core database at http://www.worldbank.org/agdistortions. Bandara, J., and S. Jayasuriya. 2009. "Sri Lanka." In Distortions to Agricultural Incentives in Asia, ed. K. Anderson and W. Martin, 409­40. Washington, DC: World Bank. Dorosh, P. A., and A. Salam. 2009. "Pakistan." In Distortions to Agricultural Incentives in Asia, ed. K. Anderson and W. Martin, 379­407. Washington, DC: World Bank. Goyal, A. 2007­08 and earlier years. BIG's Easy Reference Customs Tariff. New Delhi: Academy of Busi- ness Studies. Gulati, A., and G. Pursell. 2008. "Distortions to Agricultural Incentives in India and Other South Asia." Agricultural Distortions Working Paper 63, Washington, DC, World Bank. Gulati, A., and S. Narayanan. 2003. Subsidy Syndrome in Indian Agriculture. New Delhi: Oxford Uni- versity Press. Hamid, N., I. Nabi, and A. Nasim. 1990. Trade, Exchange Rate and Agricultural Pricing Policies in Pakistan. Washington, DC: World Bank. Naqvi, Z., and P. Schuler, eds. 2007. The Challenges and Potential of Pakistan-India Trade. Washington, DC: World Bank. Pursell, G. 1999. "Some Aspects of the Liberalization of South Asian Agricultural Policies: How Can the WTO Help?" In Implications of the Uruguay Round for South Asia: The Case of Agriculture, ed. B. Blarel, G. Pursell, and A. Valdés, New Delhi: Allied Publishers for the World Bank. Pursell, G., A. Gulati, and K. Gupta. 2009. "India." In Distortions to Agricultural Incentives in Asia, ed. K. Anderson and W. Martin, 339­77. Washington, DC: World Bank. Pursell, G., N. Kishor, and K. Gupta. 2007. "Manufacturing Protection in India Since Independence." Aus- tralia South Asia Research Centre Working Paper 2007/7, Australian National University, Canberra. Rahman, S. H. 1994. "The Impact of Trade and Exchange Rate Policies on Economic Incentives in Bangladesh Agriculture." Working Paper on Food Policy in Bangladesh 8, International Food Policy Research Institute , Washington, DC. Sandri, D., E. Valenzuela, and K. Anderson. 2007. "Economic and Trade Indicators for Asia." Agricul- tural Distortions Working Paper 20, World Bank, Washington, DC. http://www.worldbank.org/ agdistortions. World Bank. 2004. "Trade Policies in South Asia: An Overview." Report 29949. World Bank, Washington, DC. Part IV global market and welfare effects of distortions 11 Welfare-Based and Trade-Based Indicators of National Agricultural Distortions Peter J. Lloyd, Johanna L. Croser, and Kym Anderson* The methodology outlined in Anderson et al. (2008a, 2008b) and appendix A of this volume provides a number of ways to indicate the extent of distortions within the agricultural sector of a country (as distinct from between agriculture and other sectors, for which the relative rate of assistance indicator is used). They include the unweighted or weighted mean nominal rate of assistance (NRA) of covered products, the standard deviation of covered product NRAs, the weighted mean NRA for exportable versus import-competing covered products, and the trade bias index (TBI) defined as [(1 NRAagx 100) (1 NRAagm 100) 1], where NRAagx and NRAagm are the weighted average percentage NRAs for the exportable and import-competing parts, respectively, of the agricultural sectors' covered plus noncovered products. The reason for reporting the latter indicators of dispersion in addition to the means--apart from being informative in their own right--is that theory suggests the national economic welfare cost of govern- ment policy distortions to incentives in terms of resource misallocation tends to be greater when the degree of substitution in production is greater (Lloyd 1974). In the case of agriculture, which involves the use of farmland that is sector- specific but very transferable among farm activities, the greater the variation of * The authors are grateful for the NRA estimates provided by authors of country studies and for invaluable help with data compilation and manipulation by Esteban Jara, Marianne Kurzweil, Signe Nelgen, and Ernesto Valenzuela. This chapter draws heavily on Lloyd, Croser, and Anderson (2009a, 2009b). 419 420 Distortions to Agricultural Incentives: A Global Perspective NRAs across industries within the sector, the higher the welfare cost of those mar- ket interventions. While those various indicators of dispersion are useful, it would also be helpful to have a single indicator to capture the overall welfare or trade effect of each country's agricultural price distortion regime in place at any time. To that end, a theoretical literature has developed in recent years. This literature seeks to over- come aggregation problems across different intervention measures and across the product range by using a theoretically sound aggregation procedure that answers precise questions regarding the welfare and trade distortions imposed by each country's price and trade policies. The literature has developed considerably over the past two decades, particularly with the theoretical advances by Anderson and Neary (summarized in and extended beyond their 2005 book) and the theoretical simplifications by Feenstra (1995). Notwithstanding these advances, few series of consistently estimated indexes have yet been estimated across time and even fewer across countries. A prominent exception is Kee, Nicita, and Olarreaga (2008, 2009), which estimates a series for developing and high-income countries, though it provides estimates for only a snapshot ìn time (the early 2000s). Other country-specific studies include an application to Mexican agriculture in the late 1980s (Anderson, Bannister, and Neary 1995) and a long time series for U.S. trade policy (Irwin forthcoming). The purpose of this chapter is to provide estimates of indexes that are compa- rable across the focus countries and over the time period of the present study of global distortions to agricultural incentives. The estimates presented below make a significant contribution to the empirical literature in terms of welfare reduction indexes (WRIs) and trade reduction indexes (TRIs), as they provide the first panel set of consistent indexes for the agricultural sector for both developing and high- income countries. It is a global panel dataset that contains comparable estimates of annual NRAs and consumer tax equivalents (CTEs)1 for a wide range of agri- cultural products (around a dozen per country) over the past half century for 75 countries that together account for nine-tenths of the world's population and agricultural production and 96 percent of global gross domestic product (GDP).2 The indexes estimated in this chapter are well grounded in theory: they belong to the family of indexes first developed by Anderson and Neary (2005) under the catch-all name of trade restrictiveness indexes. To date, members of that family of indexes sometimes have been distinguished by using various adjectives; at the same time, others have used the trade restrictiveness index term for measures that have a different theoretical backing (for example, the one used by the Interna- tional Monetary Fund [IMF]--see Allen 2005). To avoid confusion, more precise descriptors of the distortions imposed by each country's border and domestic policies on its economic welfare and its trade volume are coined here: TRI and Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 421 WRI. The WRI is computed from subindexes that herein are called the producer distortion index (PDI) and the consumer distortion index (CDI). The PDI and CDI are needed if any product's NRA and CTE differ--that is, whenever there are domestic subsidies or taxes on production or consumption in addition to border measures, as there so often are for staple foods and other farm products. Thus the indexes estimated here capture the welfare-reducing and trade- reducing effects of all policies directly affecting consumer and producer prices of farm products from all agricultural and food policy measures in place.3 On the production side, by calculating the percentages by which domestic prices exceed border prices, the NRA estimates include assistance provided by all tariff and non- tariff trade measures, plus any domestic price support measures, plus an adjust- ment for the output-price equivalent of direct interventions on farm inputs. Where multiple exchange rates operate, an estimate of the import or export tax equivalents of that distortion are included as well (see Anderson et al. 2008a and appendix A of this volume). On the consumption side, CTE measures--also expressed as ad valorem rates--estimate the extent to which consumers are taxed or subsidized by various agricultural, social welfare, trade, and exchange rate pol- icy measures. Like NRAs, the range of measures included in the CTE estimates is wide, including domestic consumer and border taxes and subsidies, and quantita- tive measures, so as to fully capture the wedge between the price that consumers pay for each commodity and the international price at the border adjusted to account for marketing margins, quality differences, and the like. The TRI (or WRI) has the advantage of providing a theoretically sound partial equilibrium indicator of the trade (or welfare) effect in a single sectoral measure that is comparable across time and place. In this way, the TRI and WRI get some- what closer to what a computable general equilibrium (CGE) can provide in the way of estimates of the trade and welfare (and other) effects of the price distor- tions captured by the product NRA and CTE estimates--and have the advantage of being able to indicate trends over time, which a comparative static CGE model can do only if it is calibrated to a series of past years rather than to just one partic- ular year.4 The TRI (or WRI) is defined as the ad valorem trade tax rate which, if applied uniformly across all tradable agricultural commodities in a country, would generate the same reduction in trade (or economic welfare loss) as the actual cross-product structure of NRAs and CTEs for that country.5 Because it is a mean of order two, the WRI measure better reflects the true par- tial equilibrium welfare cost of agricultural price-distorting policies than the NRA or CTE. In particular, the WRI captures the disproportionately higher welfare costs of peak levels of assistance or taxation. Also, the WRI and TRI measures overcome aggregation problems when there are different NRAs for subsectors within agriculture. For example, if policies affecting the import-competing and 422 Distortions to Agricultural Incentives: A Global Perspective exporting subsectors had offsetting effects on farmer incentives, the aggregate NRA estimate may be close to zero even though the welfare- and trade-restricting consequences are considerable. Anderson et al. (2008a, 2008b) deal with that by estimating separate NRAs and CTEs for each product and then for the import- competing and exporting (and nontradable) product subgroups, and by using those subsector means to calculate their TBI. The WRI and TRI are a more suc- cinct and accurate method of summarizing that information. Defining the WRI and TRI The initial theoretical work on international trade distortions by Anderson and Neary, leading to their 2005 book, sought to derive a general equilibrium measure of the welfare-reducing effects of trade restrictions in a country's import-competing sector. They called this a trade restrictiveness index. The work was important in that it solved the problem of how to aggregate assistance across commodities in a theoretically meaningful way. Anderson and Neary (2005) solved the problem for a small, open economy in which imports are restricted by tariffs and nontariff measures (NTMs). They then provided variants of the trade restrictiveness index, including one based not on a welfare criterion but instead on an import volume criterion (the mercantilist trade restrictiveness index). In this chapter, a more gen- eral version of each of the Anderson and Neary indexes is developed for situations where, in addition to import measures, there are also export measures and possi- bly also direct domestic producer and consumer price distortions.6 The two indexes presented here are developed first for agriculture's import-competing subsector and then for its exporting subsector. The import-competing subsector The theory presented here considers an individual country, assuming it has a small, open economy in which all markets are competitive. However, the market for an import good may be distorted by a tariff and other nontariff border meas- ures or by behind-the-border measures such as domestic subsidies and price controls. First, the measure of the effect of a country's distortions on its import volume, the TRI, is examined. The TRI is defined as the uniform tariff rate which, if applied to all goods in the place of all actual tariffs and NTMs and other price distortions, would result in the same reduction in the volume of imports as the actual distortions. Consider the market for one good, good i, which is distorted by a combination of measures that distort the consumer and producer prices. For the producers of the good, the distorted domestic producer price, pP, is related to the world price, i Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 423 p*, by the relation pP p*(1 si ), where si is the rate of distortion of the producer i i i price in percentage terms. For consumers of the good, the distorted domestic con- sumer price, pC, is related to the world price by the relation pC p*(1 ri ), where i i i ri is the rate of distortion of the consumer price in percentage terms. In general, ri si. Using these relations, the change in imports in the market for good i is given by: Mi p*dxi i p*dyi i (11.1) p* dxi dpCri i 2 i p* dyi dpPsi, i 2 i (11.2) where the demand and the supply for good i, xi and yi , are functions of domestic price alone: xi xi(pC ) and yi yi (p P ), respectively. The neglect of cross-price i i effects, among other things, makes the analysis partial equilibrium. Strictly speaking, this result holds only for small distortions. In reality, rates of distortion are not small. If, however, it is assumed that the demand and supply functions are linear, then the reduction in imports is given by equation 11.2 with the slopes of the demand and supply curves in price­quantity space (dxi dpC and i dyi dpP, respectively) constant. i If the functions are not linear, this expression provides an approximation to the loss. With n importable goods subject to different levels of distortions, the aggregate reduction in imports, in the absence of cross-price effects in all markets, is given by: n n Mi p*2dxi dpCri i i p*2dyi dpPsi. i i (11.3) i 1 i 1 Setting the result equal to the reduction in imports from a uniform tariff results in: n n n p*2dxi dpCri i i p*2dyi dpPsi i i p*2dmi dpiT. i (11.4a) i 1 i 1 i 1 Solving for T results in: T {Ra Sb}, (11.4b) n where R riui with u1 p*2dxi dpC^ i i p*2dxi dpC, i i (11.4c) i 1 i n S sivi with vi p*2dyi dpP^ i i p*2dyi dpP, i i (11.4d) i 1 i and a p*2dxi dpC^ i i p*2dmi dpi i i i b p*2dyi dpP^ i i p*2dmi dpi. i (11.4e) i i 424 Distortions to Agricultural Incentives: A Global Perspective The TRI is best regarded as a true index of average distortion rates. More precisely, what is held constant is the value of imports in constant prices. R and S are indexes of average consumer and producer price distortions, respectively. They are arithmetic means. In the empirical section of this study, these are referred to as the NRA and the CTE. Evidently, T can be written as a weighted average of the level of distortions of consumer and producer prices. An important advantage of using this decomposi- tion of the index into producer and consumer effects is that it correctly treats the effects of NTMs and domestic distortions. It also allows one to deal with, and ana- lyze, the production and consumption sides of the economy separately.7 In equations 11.4c and 11.4d, the weights for each commodity are proportional to the marginal response of domestic production (or consumption) to changes in international free-trade prices. These weights can be written as functions of the domestic price elasticities at the free trade points of supply ( *) (demand [ *]) and i i the value of domestic production (consumption) at undistorted prices: n ui *(p*x*)^ i i i *(p*x *) i i i i n (11.5) vi *(p*y *)^ i i i *(p*y*). i i i i If, further, domestic price elasticities of supply (demand) are assumed to be equal across commodities, the elasticities in the numerator and denominator cancel, allowing R (S) by aggregating the change in consumer (producer) prices across commodities, using as weights the share of each commodity's domestic value of consumption (production) at undistorted prices. Estimating T in equation 11.4b also requires an assumption about the weights a and b (equation 11.4e). The weight a (b) is proportional to the ratio of the marginal response of domestic demand (supply) to a price change relative to the marginal response of imports to a price change. If the domestic demand and sup- ply curves have the same slope, then a b 0.5. As a special case, if ri si for all i--that is, if tariff rates are the only distortion, equation 11.4b reduces to a much simpler form: n n T tiwi wi *(p*m*)^ i i i *(p*m*). i i i (11.6) i 1 i Here, ti is the ad valorem tariff rate, which is equal to the rate of distortion of both consumer and producer prices, and * is the elasticity of import demand at the i free trade point. T is the mean of the tariff rates. This case can be used to obtain an alternative expression for the general case. But one must be careful, as this alterna- tive form requires computing an import-equivalent tariff rate for each tariff item when there is some distortion other than an ad valorem tariff. (The appendix to Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 425 Lloyd, Croser, and Anderson [2009a] derives the import-equivalent tariff and the alternative expression.) The measure of the effect of a country's distortions on its welfare, the WRI, is examined next. The derivation follows the same steps as the derivation of the TRI, leading to a simple comparison of the two indexes. The distortions in the market for good i create a welfare loss, Li. This loss is given by the sum of the change in producer plus consumer surplus net of the tar- iff revenue. This loss of producer and consumer surplus is given simply by: 1 2 P Li 2 (p*si) dyi dpi i (p*ri)2 dxi dpC , i i (11.7) where demand and the supply for good i are again functions of domestic price alone. Strictly speaking, this result too holds only for small distortions. For larger distortions, welfare losses are defined by the triangular-shaped areas under the demand and supply curves for the good. These areas can be obtained by integra- tion. On the assumption that the demand and supply functions are linear, the welfare loss is given by equation 11.7 with dyi dpi and dxi dpi being constant. If the functions are not linear, this expression provides an approximation to the loss. In the special case where ri si ti, the expression reduces to 1 2 Li 2 (p*ti) dxi dpi . i (11.8) Equation 11.8 yields the fundamental result that the loss from a tariff is pro- portional to the square of the tariff rate. This holds because the tariff rate determines both the price adjustment and the quantity response to this adjust- ment.8 If ri si, as is frequently true in agricultural markets, the expression in equation 11.7 yields the result that the consumer and the producer losses are each proportional to the square of the rate of distortion of the consumer or producer price, respectively. With n importable goods subject to different levels of distortions, the aggregate welfare loss, in the absence of cross-price effects in all markets, is given by: n n 1 L 2 (p*si)2dyi dpP i i (p*ri)2dxi dpC . i i (11.9) i 1 i 1 The uniform tariff rate that generates an aggregate deadweight loss identical to that of the differentiated set of tariffs is determined by the following equation: n n n (p*si)2dyi dpP i i (p*ri)2dxi dpC i i (p*W)2 dmi dpi, i (11.10) i 1 i 1 i 1 426 Distortions to Agricultural Incentives: A Global Perspective where W is the uniform tariff, which, if applied to all goods in the place of all actual tariffs and NTMs and other distortions, would result in the same aggregate loss of welfare as the actual distortions. Solving for W results in: W R 2a S 2b 1 2 , (11.11a) n 1 2 where R r2ui i with ui p*2 dxi dpC^ i i p*2 dxi dpC, i i (11.11b) i 1 i n 1 2 S s2vi i with vi p*2 dyi dpP^ i i p*2 dyi dpP, i i (11.11c) i 1 i and a p*2dxi dpC^ i i p*2 dmi dpi i i i b p* dyi dpP^ i 2 i p*2 dmi dpi. i (11.11d) i i W is the desired WRI. R and S are measures of the average levels of consumer and producer price distortions, respectively. They are means of order two. In the empirical section, R and S are referred to as the PDI and the CDI to distinguish them from the arithmetic mean forms, the NRA and CTE. Evidently, W can be written as an appropriately weighted average of the level of distortions of consumer and producer prices. It too is a mean of order two. As with the index T, the production and consumption sides of the economy for W can be dealt with and analyzed separately. Comparing the expression for the WRI in equation 11.11a with that for the TRI in equation 11.4b, it is clear that the weights in the construction of R , S , and W are the same as the weights for R, S and T. The only difference in the expressions for R , S and W is that, in the case of the TRI, one constructs arithmetic means (which are the means of order one) whereas, in the case of the WRI, one constructs means of order two.9 This difference is all due to the fact that the losses of import volume in each market are all proportional to the distortion rate whereas the losses of welfare are proportional to the squares of the distortions rates (compare equation 11.2 with equation 11.7). The tariff rate enters only once in the determination of the import loss, whereas the tariff rate enters twice in the determination of the welfare loss, once in the base and once in the height of the familiar dead-weight loss triangle of the supply­demand diagram for a small open economy. In the special case where ri si ti for all i, equation 11.11 reduces to a much simpler form: n 1 2 W (t)2wi i , where (11.12) i n n wi *(p*m*)^ i i i *(p*m*). i i i i Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 427 Further, if it is assumed that the elasticities of import demand are all equal, the weights are the share of imports of each good in total imports. This case can be used to obtain an alternative expression of the general case of the WRI. This is done in the appendix to Lloyd, Croser, and Anderson (2009a). Adding the exportables subsector The WRI and TRI indexes can each be extended to include the exportables sub- sector. For exportables, an export subsidy reduces welfare in the same way as an import tax in the import-competing sector, but it also increases trade, whereas the tariff reduces trade. It is necessary to keep track of import and export price distor- tions separately, for both producers and consumers, for the purpose of estimating the full WRIs and TRIs. In essence, this extension is done by extending the com- modity set and keeping separate track of the subsets of import-competing and exportable goods. The WRI for the whole tradables sector can be written as an expansion of equation 11.11 where goods 1 to n are import-competing products and goods n 1 to z are exportables: 2 W {(R M PM RX2 PX)a 2 (S M CM SX2 1 2 CX)b} , (11.13a) z n z n yipi yipi xipi xipi i n 1 i 1 i n 1 i 1 where PX z , PM z , CX z , CM z . (11.13b) yipi yipi xipi xipi i 1 i 1 i 1 i 1 For both import-competing and exportable subsectors in this study, producers and consumers continue to be aggregated separately, where the weights for each sub- sector are the share of the subsector's value of production (consumption) in the total value of production (consumption). Producer and consumer distortions are aggregated in the last step with the assumption that the aggregate demand and sup- ply curves have the same slope (that is, a b 0.5). The resulting measure can be regarded as the import tax or export subsidy, which, if applied uniformly, would give the same loss of welfare as the combinations of measures distorting consumer and producer prices in the import-competing and exportable subsectors. The TRI can be similarly decomposed as follows: T (RM PM RX PX)a (SM CM SX CX)b, (11.14) where , a, and b are as already defined; RM and SM are R and S from equa- tions 11.4c and 11.4d; and z z RX ri ui ; SX si vi . (11.15) i 1 n i 1 n 428 Distortions to Agricultural Incentives: A Global Perspective The aggregates in equation 11.15 are the weighted average levels of distortions to consumer and producer prices in the exportables subsector, respectively, with weights ui and vi given in equations 11.4c and 11.4d. Importantly, distortions to the exportables subsector enter equation 11.15 as negative values. This is because while a lowering of ri (the distortion of the consumer price of good i) or si (the distortion of the producer price of good i) in the import-competing subsector reduces the reduction index, a lowering of ri or si in the exportables subsector increases it. These extensions of the TRI and the WRI have precisely the same properties as the indexes for the import-competing sector. The World Bank's Agricultural Distortions Project Database The database generated by the World Bank's Distortions to Agricultural Incentives Project (Anderson and Valenzuela 2008) contains around 30,000 consistent esti- mates of annual NRAs to the agricultural sector and the same number of CTEs for a total of 75 countries between 1955 and 2007. Country coverage in the 1950s is much less than from 1960, however, so the series of index estimates presented here begins in that latter year; NRA and CTE estimates, meanwhile, are available for 2005­07 only for high-income and European transition economies (tables 11.1 and 11.2). The series contains data at the commodity level for a subset of agricul- tural products (called covered products) that account for around 70 percent of total agricultural production in each country examined. Aggregate NRAs and CTEs for various sectors and subsectors (including import-competing and exporting subsectors) are estimated using, respectively, the values of production and consumption at undistorted prices as weights.10 The range of policy measures included in the NRA estimates in the Distortions to Agricultural Incentives database is wide. By calculating domestic-to-border price ratios, the estimates include assistance provided by all tariff and nontariff trade measures, plus any domestic price support measures (positive or negative), plus an adjustment for the output-price equivalent of direct interventions on inputs. Where multiple exchange rates operate, an estimate of the import or export tax equivalents of that distortion are included as well. The range of measures included in the CTE estimates include both domestic consumer taxes and subsidies and trade and exchange rate policies, all of which drive a wedge between the price that consumers pay for each commodity and the international price at the border. The most aggregated summaries of NRA and CTE estimates for covered products for developing and high-income countries are provided in figures 11.1 and 11.2. Figure 11.1 supports two widely held views: that developing-country Table 11.1. NRAsa for Africa, Asia, Latin America, European Transition Economies, and High-Income Countries, All Farm Products, 1960­2007 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, covered import-competing products Africa 12 4 7 8 8 65 2 7 3 -- Asia 4 34 26 31 21 45 28 28 35 -- Latin America 20 3 4 2 10 4 17 9 19 -- All developing countries 11 26 17 23 17 39 22 22 28 -- European transition economies -- -- -- -- -- -- 31 34 34 30 High-income countries 54 59 42 56 70 84 73 64 60 31 World 48 50 37 46 46 66 51 43 44 -- NRA, covered exportables Africa 31 39 44 45 36 36 39 26 28 -- Asia 13 26 20 25 44 39 19 4 0 -- Latin America 23 17 30 26 27 24 9 3 4 -- All developing countries 25 29 29 30 40 37 19 5 3 -- European transition economies -- -- -- -- -- -- 4 1 0 15 High-income countries 4 10 8 7 8 17 13 6 5 3 World 2 4 7 11 24 21 8 1 0 -- (Table continues on the following page.) 429 Table 11.1. NRAsa for Africa, Asia, Latin America, European Transition Economies, and High-Income Countries, 430 All Farm Products, 1960­2007 (continued) (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 NRA, all covered farm productsb Africa 13 18 22 20 12 1 12 7 9 -- Asia 3 3 0 0 21 15 5 6 10 -- Latin America 13 13 25 20 15 14 1 1 3 -- All developing countries 9 5 9 8 20 13 5 4 7 -- European transition economies -- -- -- -- -- -- 7 15 15 21 High-income countries 32 39 29 36 43 58 49 36 32 16 World 24 24 15 18 6 16 18 16 16 -- NRA, all agriculturec Africa 8 11 15 13 8 1 9 6 7 -- Asiad 27 25 25 24 21 9 2 8 12 -- Latin America 8 7 21 18 13 11 4 5 5 -- All developing countries 23 22 24 22 18 8 2 6 9 -- European transition economies -- -- -- -- -- -- 10 18 18 25 High-income countries 29 35 25 32 41 53 46 35 32 17 World 22 21 13 15 8 17 18 17 18 -- Source: Anderson and Valenzuela (2008). Note: -- not available. a. Weighted using the value of production at undistorted prices. b. Includes nontradables. c. Includes covered and noncovered products. d. Estimates for China pre-1981 and India pre-1965 are based on the assumption that the NRAs to agriculture in those years were the same as the average NRA estimates for those economies for 1981­84 and 1965­69, and that the gross value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. This NRA assumption is conservative in the sense that for both countries, the average NRA was probably even lower in earlier years. Table 11.2. CTEsa for Africa, Asia, Latin America, European Transition Economies, and High-Income Countries, All Covered Farm Products, 1960­2007 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Import-competing products Africa 7 0 8 7 3 76 5 9 5 -- Asia 1 14 8 24 24 44 32 27 35 -- Latin America 23 11 0 8 4 1 28 11 18 -- All developing countries 6 11 4 18 17 39 29 22 27 -- European transition economies -- -- -- -- -- -- 12 21 31 30 High-income countries 53 56 41 54 65 66 57 55 50 30 World 46 44 32 43 43 55 41 38 39 -- Exportable products Africa 29 36 42 34 28 31 38 20 24 -- Asia 3 38 29 32 42 40 20 5 0 -- Latin America 25 14 25 24 27 21 12 1 0 -- All developing countries 23 36 33 30 38 37 20 5 1 -- European transition economies -- -- -- -- -- -- 6 4 2 1 High-income countries 4 11 9 9 6 11 8 2 3 0 World 0 8 9 11 24 24 11 4 2 -- All covered farm productsb Africa 8 12 16 9 6 16 8 0 3 -- Asia 0 12 15 2 15 14 3 5 10 -- Latin America 7 7 18 13 12 10 13 6 8 -- All developing countries 5 12 16 5 14 10 0 5 8 -- European transition economies -- -- -- -- -- -- 2 9 17 11 High-income countries 35 42 30 40 45 49 41 32 27 16 World 28 23 14 21 10 15 16 15 16 -- Source: Anderson and Valenzuela (2008). 431 Note: -- not available. a. Weighted using the value of consumption at undistorted prices. b. Includes nontradables. 432 Distortions to Agricultural Incentives: A Global Perspective Figure 11.1. NRAs to Farmers in High-Income and Developing Countries, for All Covered Farm Products, 1960­2007 60 40 percenta 20 0 20 1960­64 1970­74 1980­84 1990­94 2000­04 high-income countries high-income countries (including developing countries European transition economies) Source: Anderson and Valenzuela (2008). a. Percent is averaged using weights based on the gross value of agricultural production at undistorted prices. Figure 11.2. CTEs Affecting Covered Farm Products in High-Income and Developing Countries, 1960­2007 60 40 percent 20 0 20 1960­64 1970­74 1980­84 1990­94 2000­04 high-income countries high-income countries (including developing countries European transition economies) Source: Anderson and Valenzuela (2008). Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 433 governments had in place agricultural policies that effectively taxed their farmers through to the 1980s, and that the extent of those disincentives has lessened since then. The extent of taxation was of the order of more than 15 percent from the early 1960s to the mid-1980s. Since then, it has not only diminished but, on aver- age, has become slightly positive. Figure 11.1 also supports the view that the growth of agricultural protection in high-income countries has been in progress since the 1950s and began to reverse only after the 1980s (at which time there was a reinstrumentation toward other forms of support--not included here--that are somewhat decoupled from production). It is clear from figure 11.2 that con- sumers have experienced changes similar to producers in recent years. In develop- ing countries, taxation was negative (that is, consumer subsidization was positive) for most of the last 50 years, though less so since the 1990s. In high-income coun- tries, the implicit taxation of consumers from agricultural support rose until the early 1990s but has since fallen. Figures 11.3 and 11.4 show trends in NRAs and CTEs, respectively, for Euro- pean transition economies and the three developing-country regions of Africa, Figure 11.3. NRAs Affecting Covered Farm Products in Africa, Asia, Latin America, and European Transition Economies, 1960­2007 20 0 percent 20 40 1960­64 1970­74 1980­84 1990­94 2000­04 European transition economics Latin America Africa Asia Source: Anderson and Valenzuela (2008). Note: Estimates for China pre-1981 and India pre-1965 are based on the assumption that the NRAs to covered products in those years were the same as the average NRA estimates for those economies for 1981­84 and 1965­69, and that the gross value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. This NRA assumption is conservative in the sense that for both countries, the average NRA was probably even lower in earlier years. 434 Distortions to Agricultural Incentives: A Global Perspective Figure 11.4. CTEs Affecting Covered Farm Products in Africa, Asia, Latin America, and European Transition Economies, 1960­2007 30 10 percent 0 10 30 1960­64 1970­74 1980­84 1990­94 2000­04 Latin America Africa European transition economies Asia Source: Anderson and Valenzuela (2008). Note: Estimates for China pre-1981 and India pre-1965 are based on the assumption that the NRAs to covered products in those years were the same as the average NRA estimates for those economies for 1981­84 and 1965­69, and that the gross value of production in those missing years is that which gives the same average share of value of production in total world production in 1981­84 and 1965­69, respectively. This NRA assumption is conservative in the sense that for both countries, the average NRA was probably even lower in earlier years. Asia, and Latin America. On the production side, Africa is where there has been the least tendency to reduce taxation of farmers and subsidization of consumers of covered farm products. Indeed, its average NRA has been negative in all five- year periods except in the mid-1980s, when international prices of farm products reached an all-time low in real terms. By contrast, for both Asia and Latin America, NRAs crossed over from negative to positive after the 1980s. And in European transition economies, nominal assistance to farmers has trended upward following their initial shock in the early 1990s. For consumers in all four regions, agricultural policies have almost always involved consumer subsidization. Since the 1980s, however, food consumer subsidization in Asia, Latin America, and European transition economies has gradually disappeared and been replaced by a small degree of taxation. Within the farm sector of all developing regions, assistance to the import- competing subsector is typically well above that for the export sector (Lloyd, Croser, and Anderson 2009a and appendix tables B.1 to B.4 in this volume), meaning there is an antitrade bias in the structure of distortions. In the case of Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 435 developing countries, where the former NRA is positive and the latter negative, the two tend to offset each other such that the overall sectoral NRA is close to zero. Such a sectoral average can be misleading as an indication of the aggregate extent of price distortion within the sector. It can also be misleading when compared across countries that have varying degrees of dispersion in their NRAs for farm products (see Anderson et al. 2009). Measuring the WRIs and TRIs Table 11.3 reports WRIs for agricultural import-competing products, exportables, and all covered tradable products from 1960 to 2007 for the five main regions stud- ied here and for the world as a whole.11 The WRI results for covered products show a similar pattern over the five regions: there is constant or increasing tendency for policies to reduce welfare from the 1960s to the mid-1980s, but thereafter the oppo- site occurs in all regions, as can be seen from figure 11.5. This pattern is generated by different policy regimes in different regions. In high-income countries, agriculture was assisted throughout the period, peaking in the 1980s and falling thereafter. By contrast, in developing countries, agriculture was disprotected until the mid-1980s, and only thereafter did taxation of developing country farmers decline to the point that they received positive assistance by the turn of the century. The first point to note, then, is that the WRI has the desirable property of correctly reflecting the wel- fare consequences that result from both positive and negative assistance regimes for the agricultural sector. Note also that the WRI for high-income countries is above that for developing countries throughout the five decades shown in figure 11.5b. A second point to note is that the WRI provides a better indicator of the welfare cost of distortions than the average level of assistance or taxation in the Distortions to Agricultural Incentives database (NRA and CTE). Although the latter make a sig- nificant contribution in their own right (for example, as inputs into global com- modity or economy-wide models), they can be misleading as indicators of the extent of the welfare cost of assistance, due to the inclusion in the WRI of the "power of two." That is, a weighted arithmetic mean does not fully reflect the welfare effects of agricultural distortions because the dispersion of that support or taxation across products has been ignored. By contrast, the WRI captures the higher welfare costs of high and peak levels of assistance or taxation. A good example of this is the WRI for high-income countries. For high-income countries, the WRI series in figure 11.5 is higher than the NRA series in figure 11.1, owing to its capturing of the dispersion of the NRA. That is, the WRI reflects the disparity issue discussed in Lloyd (1974): the larger the variance in assistance levels, the greater the potential for resources to be used in activities that do not maximize economic welfare. A third point to note is that the WRI and its two components (PDI and CDI, reported in tables 11.5 and 11.6) reflect the true welfare cost of agricultural policies 436 Distortions to Agricultural Incentives: A Global Perspective Figure 11.5. WRIs for Covered Tradable Farm Products, by Region, 1960­2007 a. Africa, Asia, and Latin America 80 60 percent 40 20 0 4 9 4 9 4 9 4 9 4 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 60 65 70 75 80 85 90 95 00 19 19 19 19 19 19 19 19 20 Africa Latin America Asia b. Developing countries, high-income countries, and European transition economies 100 80 60 percent 40 20 0 4 9 74 9 84 9 4 99 04 7 ­6 ­6 ­7 ­8 ­9 ­0 ­ ­ ­ ­ 60 65 70 75 80 85 90 95 00 05 19 19 19 19 19 19 19 19 20 20 high-income countries European transition economies developing countries Source: Anderson and Croser (2009), based on NRAs and CTEs in Anderson and Valenzuela (2008). Table 11.3. WRIs for Asia, Africa, Latin America, European Transition Economies, and High-Income Countries,a All Covered Tradable Farm Products, 1960­2007 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Import-competing products Africa 59 52 53 47 51 98 43 32 30 -- Asia 36 45 46 50 48 62 48 44 48 -- Latin America 54 34 27 37 47 40 46 26 32 -- All developing countries 47 45 45 47 48 62 48 40 43 -- European transition economies -- -- -- -- -- -- 60 44 45 43 High-income countries 77 85 69 99 106 123 102 91 87 50 World 72 75 64 84 81 100 78 65 65 -- Exportable products Africa 37 44 48 49 48 55 58 41 40 -- Asia 24 43 34 34 48 45 24 10 7 -- Latin America 28 22 36 32 36 33 29 12 15 -- All developing countries 31 39 38 36 46 44 27 11 10 -- European transition economies -- -- -- -- -- -- 37 33 31 42 High-income countries 11 19 15 12 11 25 22 11 11 10 World 15 26 25 24 34 39 26 13 12 -- All covered farm tradables -- Africa 51 51 52 49 50 80 52 37 36 -- Asia 32 45 44 45 50 51 33 23 21 -- Latin America 37 26 36 35 42 37 39 18 22 -- All developing countries 41 43 44 43 48 51 36 23 22 -- European transition economies -- -- -- -- -- -- 47 40 40 44 High-income countries 55 66 54 73 77 95 77 60 58 33 World 53 59 51 62 61 70 54 39 38 -- Source: Authors' calculations based on product NRAs and CTEs in Anderson and Valenzuela (2008). 437 Note: -- not available. a. Regional aggregates are weighted using the average of the value of production and the value of consumption at undistorted prices. 438 Distortions to Agricultural Incentives: A Global Perspective when they have offsetting components, unlike the arithmetic mean measures of assistance, the NRA and CTE. This can be seen most clearly in the case of Africa, which, in the latter half of 1980s, was still taxing exportables but had moved (temporarily) from low to very high positive levels of protection for import- competing farm products (table 11.1). Figure 11.3 indicates that in 1985­89, the weighted average NRA for import-competing and exporting farmers in Africa was close to zero. However, figure 11.5a shows that the WRI for Africa peaked in this time period. Thus, while at the aggregate level African farmers received almost no government assistance during the late 1980s, the welfare cost of the mixture of agricultural programs as a whole was at its highest. For developing countries as a group, the trade restrictiveness of agricultural policy was slightly increasing until the early 1990s. Thereafter, it declined, especially for Africa and Asia, according to the TRI estimates for the five main regions and for different subsectors (figure 11.6 and table 11.4). For high-income countries, the TRI time path was similar. The aggregate results for developing countries are driven by the exportables subsector, which is being taxed, and the import-competing subsec- tor, which is being protected (albeit by less than in high-income countries--see tables 11.1 and 11.4). For high-income countries, policies support both exporting and import-competing agricultural products and, even though they favor the latter much more heavily (figure 11.1), the assistance to exporters somewhat offsets the antitrade bias from the protection of import-competing producers in terms of impacts on those countries' aggregate volume of trade in farm products. This is reflected in a much smaller TRI for high-income countries in the third as compared with the first row for high-income countries in table 11.4. Like the WRI, the TRI correctly aggregates the restrictiveness of subsector poli- cies that are masked in aggregate NRA and CTE measures, because they offset one another. Again using the example of Africa in 1985­89, when the NRA was closest to zero, the TRI peaks at this time in a way that correctly identifies the trade- reducing effect of positive protection to the import-competing subsector and dis- protection to the exportables subsector. For completeness, PDI and CDI estimates (tables 11.5 and 11.6) and the national WRI and TRI estimates (tables 11.7 and 11.8) are also included. The PDI and CDI estimates are not identical, but their similarity reflects the fact that most of the distortions to agricultural incentives, as compiled in Anderson and Valen- zuela (2008), are due to price distortions at national borders rather than to domestic measures. Even so, it is important to keep the PDI and CDI separate because they can be very different for some products. Likewise, the country detail in tables 11.7 and 11.8 reveals that considerable differences within each region are concealed in the regional aggregates reported in earlier tables and figures. Those differences are illustrated clearly for 2000­04 in figure 11.7, where individual Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 439 Figure 11.6. TRIs for Covered Tradable Farm Products, by Region, 1960­2007 a. Africa, Asia, and Latin America 60 40 percent 20 0 4 9 4 9 4 9 4 9 4 ­6 ­6 ­7 ­7 ­8 ­8 ­9 ­9 ­0 60 65 70 75 80 85 90 95 00 19 19 19 19 19 19 19 19 20 Latin America Africa Asia b. Developing countries, high-income countries, and European transition economies 70 50 percent 30 10 0 4 9 74 79 84 89 4 9 04 07 ­6 ­6 ­9 ­9 ­ ­ ­ ­ ­ ­ 60 65 70 75 80 85 90 95 00 05 19 19 19 19 19 19 19 19 20 20 high-income countries developing countries European transition economies Source: Anderson and Croser (2009), based on NRAs and CTEs in Anderson and Valenzuela (2008). Table 11.4. TRIs for Asia, Africa, Latin America, European Transition Economies, and High-Income Countries,a 440 All Covered Tradable Farm Products, 1960­2007 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Import-competing products Africa 9 2 7 7 5 71 4 8 4 -- Asia 3 24 17 27 22 45 31 28 36 -- Latin America 22 8 2 5 7 2 23 10 19 -- All developing countries 8 19 11 21 17 39 26 22 28 -- European transition economies -- -- -- -- -- -- 22 28 33 30 High-income countries 51 56 40 54 68 75 66 60 56 31 World 45 46 33 44 45 61 46 41 42 -- Exportable products Africa 30 38 43 39 32 33 38 23 26 -- Asia 9 32 24 28 42 40 20 4 0 -- Latin America 24 15 28 24 26 22 10 1 2 -- All developing countries 23 31 30 29 39 37 20 5 2 -- European transition economies -- -- -- -- -- -- 5 2 2 9 High-income countries 3 10 8 7 7 14 11 2 1 2 World 2 6 8 11 24 22 10 3 1 -- All covered farm tradables Africa 21 22 21 26 18 50 18 14 14 -- Asia 7 29 27 28 35 41 23 12 11 -- Latin America 24 14 21 18 19 14 17 5 8 -- All developing countries 17 26 24 26 31 38 22 11 11 -- European transition economies -- -- -- -- -- -- 8 14 14 6 High-income countries 30 33 23 32 40 45 39 33 29 15 World 27 30 23 30 34 41 28 20 18 -- Source: Authors' calculations based on product NRAs and CTEs in Anderson and Valenzuela (2008). Note: -- not available. a. Regional aggregates are weighted using the average of the value of production and consumption at undistorted prices. Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 441 Figure 11.7. WRIs and TRIs for Covered Tradable Farm Products, by Country, 2000­04 Australia Brazil TRI WRI China New Zealand Uganda Cameroon Thailand Chile Madagascar Argentina Senegal South Africa Egypt, Arab Rep. of Bulgaria United States Ukraine Kenya Nicaragua Indonesia India Pakistan Sri Lanka Ghana Estonia Bangladesh Ecuador Mexico Russian Federation Malaysia Vietnam Poland Slovak Republic Czech Republic Spain Côte d'Ivoire Canada Portugal Philippines Zambia Sudan Italy Ethiopia Turkey Tanzania Denmark France Hungary Germany Austria Netherlands Mozambique Colombia Nigeria United Kingdom Finland Dominican Republic Sweden Romania Latvia Lithuania Ireland Slovenia Zimbabwe Norway Switzerland Iceland Taiwan, China Japan Korea, Rep. of 50 0 50 100 150 200 250 percent Source: Anderson and Croser (2009), based on NRAs and CTEs in Anderson and Valenzuela (2008). 442 Table 11.5. PDIs for Asia, Africa, Latin America, European Transition Economies, and High-Income Countries,a All Covered Farm Products, 1960­2007 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Import-competing products Africa 60 54 53 48 52 93 43 32 31 -- Asia 39 53 52 52 48 60 46 43 46 -- Latin America 53 32 26 34 50 44 43 25 37 -- All developing countries 49 51 50 49 49 61 45 39 43 -- European transition economies -- -- -- -- -- -- 58 49 48 43 High-income countries 77 85 71 101 108 130 106 90 87 47 Exportable products Africa 38 45 49 52 50 53 56 39 39 -- Asia 24 37 29 31 49 44 24 9 7 -- Latin America 27 22 38 33 36 34 29 13 16 -- All developing countries 31 36 37 36 47 44 27 11 10 -- European transition economies -- -- -- -- -- -- 38 34 31 37 High-income countries 10 18 15 11 11 24 19 9 10 9 All covered farm productsb Africa 44 46 45 46 42 55 39 28 26 -- Asia 32 41 37 41 48 49 32 22 20 -- Latin America 29 24 36 34 42 39 35 18 23 -- All developing countries 45 48 49 48 51 55 36 24 23 -- European transition economies -- -- -- -- -- -- 46 43 40 40 High-income countries 52 62 51 69 73 95 76 56 54 31 Source: Authors' calculations based on product NRAs and CTEs in Anderson and Valenzuela (2008). Note: -- not available. a. Regional aggregates are weighted using the value of production at undistorted prices. b. Includes nontradables. Table 11.6. CDIs for Asia, Africa, Latin America, European Transition Economies, and High-Income Countries,a All Covered Farm Products, 1960­2007 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Import-competing products Africa 58 51 52 46 50 101 43 32 29 -- Asia 31 32 35 48 48 62 49 44 48 -- Latin America 55 35 27 39 43 34 45 25 24 -- All developing countries 43 36 37 46 47 62 48 39 41 -- European transition economies -- -- -- -- -- -- 61 38 42 43 High-income countries 77 84 68 96 101 114 94 88 84 50 Exportable products Africa 36 43 47 45 46 58 62 42 41 -- Asia 28 50 39 37 47 46 24 10 6 -- Latin America 30 21 34 31 36 32 29 9 12 -- All developing countries 33 43 41 37 45 45 27 11 9 -- European transition economies -- -- -- -- -- -- 36 32 30 47 High-income countries 10 19 15 12 10 25 22 11 11 10 All covered farm productsb Africa 44 43 44 39 40 62 39 27 26 -- Asia 29 39 38 42 46 49 33 24 21 -- Latin America 33 26 34 35 40 34 40 18 19 -- All developing countries 48 43 44 48 49 58 38 26 24 -- European transition economies -- -- -- -- -- -- 48 37 39 48 High-income countries 57 68 54 75 77 91 74 62 58 36 Source: Authors' calculations based on product NRAs and CTEs in Anderson and Valenzuela (2008). 443 Note: -- not available. a. Regional aggregates are weighted using the value of consumption at undistorted prices. b. Includes nontradables. Table 11.7. WRIs, by Country and Regiona, All Covered Tradable Farm Products, 1960­2007 444 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa 51 51 52 49 50 80 52 37 36 -- Cameroon 29 37 42 54 38 23 20 18 11 -- Côte d'Ivoire 35 47 45 48 44 39 37 32 41 -- Egypt, Arab Rep. of 49 53 54 40 46 134 32 29 21 -- Ethiopia -- -- -- -- 44 56 58 52 47 -- Ghana 24 44 40 62 89 75 39 21 30 -- Kenya 39 39 29 19 31 27 36 22 26 -- Madagascar 26 28 26 45 58 46 30 16 15 -- Mozambique -- -- -- 72 65 75 33 31 56 -- Nigeria 148 129 121 105 102 127 94 75 58 -- Senegal 19 18 44 46 41 60 66 12 19 -- South Africa 20 18 25 34 48 39 31 22 20 -- Sudan 35 40 51 40 40 65 79 42 44 -- Tanzania -- -- -- 71 72 68 62 54 50 -- Uganda 11 16 44 83 58 60 11 10 10 -- Zambia 26 38 48 59 32 70 59 40 43 -- Zimbabwe 41 45 50 56 46 42 47 40 72 -- Asia 32 45 44 45 50 51 33 23 21 -- Bangladesh -- -- 30 41 29 49 29 25 31 -- China -- -- -- -- 55 48 25 12 8 -- India 37 46 49 61 54 87 31 22 27 -- Indonesia -- -- 18 22 31 21 24 28 27 -- Korea, Rep. of 45 43 69 86 130 176 211 194 228 -- Malaysia 14 12 10 31 57 95 71 31 34 -- Pakistan 44 71 75 37 39 46 31 24 29 -- Philippines 18 36 30 21 33 46 32 51 42 -- Sri Lanka 32 28 29 37 26 29 39 35 30 -- Taiwan, China 30 46 52 35 43 85 124 155 190 -- Thailand -- -- 30 24 22 18 16 19 12 -- Vietnam -- -- -- -- -- 22 30 24 37 -- Latin America 37 26 36 35 42 37 39 18 22 -- Argentina 32 30 28 27 24 19 10 8 17 -- Brazil -- 16 43 36 42 39 34 8 7 -- Chile 53 27 28 28 16 34 23 18 13 -- Colombia 28 23 22 26 40 25 25 35 58 -- Dominican Republic 78 42 44 46 50 55 89 48 59 -- Ecuador -- 37 48 59 71 44 20 24 32 -- Mexico -- -- -- 43 48 42 54 30 33 -- Nicaragua -- -- -- -- -- -- 29 31 26 -- All developing countries 41 43 44 43 48 51 36 23 22 -- European transition economies -- -- -- -- -- -- 47 40 40 45 Bulgaria -- -- -- -- -- -- 28 26 22 29 Czech Republic -- -- -- -- -- -- 39 30 39 33 Estonia -- -- -- -- -- -- 28 27 29 31 Hungary -- -- -- -- -- -- 35 34 51 31 Latvia -- -- -- -- -- -- 54 47 66 31 Lithuania -- -- -- -- -- -- 54 52 67 31 Poland -- -- -- -- -- -- 27 27 36 33 Romania -- -- -- -- -- -- 36 44 65 51 Russia -- -- -- -- -- -- 46 34 33 -- Slovak Republic -- -- -- -- -- -- 31 30 37 32 Slovenia -- -- -- -- -- -- 60 72 69 45 Turkey 21 36 35 41 38 38 50 58 50 59 445 Ukraine -- -- -- -- -- -- 39 33 25 -- (Table continues on the following page.) 446 Table 11.7. WRIs, by Country and Regiona, All Covered Tradable Farm Products, 1960­2007 (continued ) (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 High-income countries 55 66 54 73 77 95 77 60 58 33 Australia 20 31 28 18 13 21 21 9 4 2 Austria 92 93 39 43 39 82 106 60 56 33 Canada 16 15 15 50 81 90 59 37 42 35 Denmark 82 84 93 157 139 121 71 55 50 26 Finland 129 138 108 129 69 204 204 65 58 31 France 93 118 94 118 124 115 74 55 51 32 Germany 142 146 109 133 134 117 73 58 52 28 Iceland -- -- -- 188 193 365 299 201 180 194 Ireland 66 99 97 187 179 169 93 74 69 44 Italy 89 90 73 88 99 93 63 49 47 23 Japan 74 94 106 155 150 248 240 210 213 163 Netherlands 137 159 129 170 164 132 76 64 56 33 New Zealand 11 12 14 20 24 28 13 10 9 7 Norway -- -- -- 292 198 256 229 174 164 117 Portugal 22 29 31 57 30 70 56 43 42 30 Spain 35 53 29 38 40 80 59 44 41 27 Sweden 149 184 137 204 163 139 122 64 61 35 Switzerland -- -- -- 219 190 347 284 195 172 108 United Kingdom 147 142 115 140 135 128 81 62 58 37 United States 13 20 12 12 26 35 22 19 25 16 Source: Authors' calculations based on product NRAs and CTEs in Anderson and Valenzuela (2008). Note: -- not available. a. Regional aggregates are weighted using the average of the value of production and the value of consumption at undistorted prices. Table 11.8. TRIs, by Country and Regiona, All Covered Tradable Farm Products, 1960­2007 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Africa 21 22 21 26 18 50 18 14 14 -- Cameroon 27 34 38 49 35 9 8 8 3 -- Côte d'Ivoire 17 16 37 50 28 31 27 27 39 -- Egypt, Arab Rep. of 5 2 2 15 8 95 12 17 6 -- Ethiopia -- -- -- -- 41 54 56 49 36 -- Ghana 3 13 18 42 59 66 32 11 25 -- Kenya 27 21 6 3 7 25 9 10 12 -- Madagascar 21 17 15 7 1 29 10 6 11 -- Mozambique -- -- -- 31 6 16 3 19 44 -- Nigeria 112 102 94 64 50 78 25 17 7 -- Senegal 19 13 38 45 35 36 36 8 16 -- South Africa 1 4 9 2 4 14 9 1 2 -- Sudan 29 28 29 29 23 56 40 18 31 -- Tanzania -- -- -- 24 22 42 41 22 30 -- Uganda 8 14 38 85 59 61 10 7 6 -- Zambia 21 1 1 36 12 46 28 7 29 -- Zimbabwe 35 39 43 51 29 37 19 10 12 -- Asia 7 29 27 28 35 41 23 12 11 -- Bangladesh -- -- 13 9 1 24 1 8 6 -- China -- -- -- -- 44 44 19 4 1 -- India 21 36 42 47 38 70 26 18 22 -- Indonesia -- -- 1 9 14 5 2 1 19 -- Korea, Rep. of 5 16 44 69 119 158 189 164 184 -- Malaysia 12 4 8 19 18 21 14 5 5 -- Pakistan 7 42 19 3 4 12 3 2 4 -- Philippines 4 2 1 0 3 16 18 39 27 -- 447 Sri Lanka 26 17 20 20 13 5 23 17 4 -- (Table continues on the following page.) Table 11.8. TRIs, by Country and Regiona, All Covered Tradable Farm Products, 1960­2007 (continued ) 448 (percent) 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Taiwan, China 6 3 16 8 19 25 37 67 96 -- Thailand -- -- 25 19 13 11 9 6 1 -- Vietnam -- -- -- -- -- 12 28 6 11 -- Latin America 24 14 21 18 19 14 17 5 8 -- Argentina 30 27 28 25 23 18 7 3 13 -- Brazil -- 12 28 19 20 13 11 0 0 -- Chile 9 7 15 4 8 24 17 14 8 -- Colombia 14 5 8 8 18 11 5 12 13 -- Dominican Republic 60 25 21 27 37 34 57 30 37 -- Ecuador -- 12 15 34 45 26 3 7 16 -- Mexico -- -- -- 12 16 13 26 8 17 -- Nicaragua -- -- -- -- -- -- 11 22 18 -- All developing countries 17 26 24 26 31 38 22 11 11 -- European transition economies -- -- -- -- -- -- 8 14 14 6 Bulgaria -- -- -- -- -- -- 11 10 6 12 Czech Republic -- -- -- -- -- -- 20 1 10 1 Estonia -- -- -- -- -- -- 2 16 4 6 Hungary -- -- -- -- -- -- 6 12 20 12 Latvia -- -- -- -- -- -- 32 22 21 11 Lithuania -- -- -- -- -- -- 36 14 6 3 Poland -- -- -- -- -- -- 15 7 9 17 Romania -- -- -- -- -- -- 8 20 41 31 Russia -- -- -- -- -- -- 31 16 22 -- Slovak Republic -- -- -- -- -- -- 2 7 5 0 Slovenia -- -- -- -- -- -- 8 17 24 12 Turkey 4 3 10 22 9 13 16 23 17 8 Ukraine -- -- -- -- -- -- 20 11 14 -- High-income countries 30 33 23 32 40 45 39 33 29 15 Australia 7 11 6 3 4 7 7 3 1 0 Austria 69 66 19 24 23 22 11 41 38 17 Canada 8 6 6 15 22 25 22 13 14 14 Denmark 35 35 3 61 70 75 51 37 32 12 Finland 39 17 7 10 18 106 153 50 41 16 France 58 73 44 47 69 72 51 32 29 13 Germany 98 112 66 64 81 73 52 39 33 14 Iceland -- -- -- 130 151 33 10 35 38 45 Ireland 4 12 7 96 117 128 81 63 55 26 Italy 45 48 33 34 52 49 31 25 23 8 Japan 64 73 73 102 105 144 134 132 127 106 Netherlands 89 120 86 96 110 84 55 48 40 17 New Zealand 2 2 2 8 11 1 2 2 1 0 Norway -- -- -- 31 32 152 195 155 140 88 Portugal 10 15 13 32 20 33 24 21 21 12 Spain 21 18 1 2 3 42 30 23 21 11 Sweden 46 41 42 51 49 71 59 47 42 18 Switzerland -- -- -- 197 168 81 44 17 14 37 United Kingdom 70 49 36 64 82 89 65 44 39 22 United States 4 2 1 4 7 7 6 2 4 1 Source: Authors' calculations based on product NRAs and CTEs in Anderson and Valenzuela (2008). Note: -- not available. a. Regional aggregates are weighted using the average of the value of production and consumption at undistorted prices. 449 450 Distortions to Agricultural Incentives: A Global Perspective country TRIs and WRIs are shown. That figure reveals the extremely high indexes for the most agricultural-protecting countries in the world, namely the three European Free Trade Area (EFTA) members (Iceland, Norway, and Switzerland) and the three advanced economies of Northeast Asia (Japan, Republic of Korea, and Taiwan, China). Notice also from figure 11.7 that while the WRI is always pos- itive, the TRI can be negative--in fact, the TRI is slightly negative for a few coun- tries, because of export or import subsidies. A useful way of summarizing the regional estimates is provided in figure 11.8, which shows their movement since the late 1980s, when most of the indexes peaked. The indexes suggest that agricultural policies were not reducing either the trade or welfare of a region if the region was located at the zero point of both axes--that is, in the bottom left corner of the diagram (the "sweet spot"). While almost no region is near that point, virtually all regions have moved toward it since 1985­89, and substantially so for the outliers, most notably Africa, but also considerably for Asia, the largest developing country region, and for the European Union. The biggest contributors to the global reduction in trade from farm policies are (in order) Japan, Korea, India, France, and Germany, while the biggest contribu- tors to the global reduction in welfare from farm policies are (again in order) Japan, the United States, Korea, China, and France (figure 11.9). Over the entire period between 1961 and 2004, the WRI has tended to be higher the higher is a country's real GDP per capita (figure 11.10). There has also been a negative correlation between the TRI and a trade specialization index defined as the ratio of net exports to the total value of exports plus imports of agriculture and food--and an even stronger negative correlation between the WRI and that trade specialization index. That is, agricultural-exporting countries tend to have lower measures of both the TRI and the WRI, while import-compet- ing countries tend to have more welfare- and trade-reducing policies in place. What can be said about agricultural distortions in the world as a whole? The fact that NRAs for high-income and developing countries diverged in opposite ways away from zero in the first half of the period under study, and then con- verged toward zero in the most recent quarter-century, means that their weighted average NRA traced out a fairly flat trend. By contrast, figure 11.11 shows the WRI and TRI for the world as a whole each tracing out a hill-shaped path and thus pro- viding less misleading indicators of the evolving disarray in world agricultural markets. Figure 11.11 also suggests that the global welfare cost of distortions was much higher than the NRA indicates, though more so in earlier decades than in the current one. The same is true also of the trade restrictiveness of farm policies globally, although much less at the beginning and end of the period studied than in the 1970s and 1980s. Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 451 Figure 11.8. TRIs and WRIs for Covered Tradable Farm Products, by Region, 1985­89 and 2000­04 a. Developing and transition economies 80 Africa 60 Europe and Central Asia Asia WRI (%) Europe and Central Asia 40 Africa Latin America Latin America 20 Asia 0 0 10 20 30 40 50 60 TRI (%) b. High-income countries 250 EFTA Japan EFTA Japan 200 150 WRI (%) EU 100 Australia and New Zealand EU 50 North America North America 0 Australia and New Zealand 25 0 25 50 75 100 125 150 TRI (%) 1985­89 2000­04 Source: Derived by the authors using data from Anderson and Croser (2009). Note: For Europe and Central Asia in the top panel, data for 1985­89 is actually for 1992­94. Figure 11.9. Country Contributions to the Global TRI and WRI, 2000­04 a. TRIa Japan Korea, Rep. of India France Germany United States United Kingdom Italy Russian Federation Indonesia Mexico Spain Turkey Netherlands Sudan Taiwan, China Philippines Romania Canada Argentina China 0 5 10 15 20 25 30 35 percent share, based on US$ values at undistorted pricesb b. WRIc Japan United States Korea, Rep. of China France Germany India Italy Turkey Spain United Kingdom Mexico Russian Federation Indonesia Canada Taiwan, China Netherlands Sudan Philippines Romania Switzerland 0 5 10 15 20 25 30 percent share, based on US$ values at undistorted pricesb Source: Derived from data in Anderson and Croser (2009). a. The global TRI in current U.S. dollars is obtained by multiplying the global TRI by the average of the value of global production and consumption at undistorted prices. Each country contribution is computed as the country-level TRI multiplied by the country-level average value of production and consumption at undistorted prices, as a share of the global aggregate TRI multiplied by the global average value of production and consumption at undistorted prices. b. The sum of all country contributions (which are necessarily all positive for the WRI) is 100. Country contributions of less than 1 percent are omitted from the figures. c. The global WRI in current U.S. dollars is obtained by multiplying the global WRI by the average of the value of global production and consumption at undistorted prices. Each country contribution is computed as the country-level WRI multiplied by the country-level average of the value of production and consumption at undistorted prices, as a share of the global aggregate WRI multiplied by the global average value of production and consumption at undistorted prices. 452 Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 453 Figure 11.10. WRI and Real Per Capita GDP, All 75 Countries, 1961­2004a 400 WRI (percent, five-year averages) 300 200 100 0 6 7 8 9 10 11 ln real GDP per capita WRI obs fitted values Source: Derived by the authors using data from Anderson and Croser (2009). a. The fitted regression line is WRI 105 19.8 lnGDPPC, Adj R2 0.14, n 498 ( 5.6) (9.1) Figure 11.11. NRA, TRI, and WRI for Covered Tradable Farm Products, World, 1960­2004 80 70 60 50 percent 40 30 20 10 0 4 69 4 79 84 89 94 99 4 ­6 ­7 ­0 ­ ­ ­ ­ ­ ­ 60 65 70 75 80 85 90 95 00 19 19 19 19 19 19 19 19 20 WRI TRI NRA Source: Anderson and Croser (2009), based on NRAs and CTEs in Anderson and Valenzuela (2008). 454 Distortions to Agricultural Incentives: A Global Perspective Finally, how do these estimates of partial equilibrium indicators of trade and welfare reduction compare with those generated by a global general equilibrium model? Even though there are numerous reasons for not expecting them to be the same, such a comparison can be a check on the orders of magnitude at least. The final chapter of this volume provides one such set of modeling results. It uses the economy-wide Linkage Model and the present project's NRAs and CTEs to examine what the trade, welfare, and other effects would be of removing all distor- tions to goods markets globally as of 2004. According to that model, global trade in all primary and lightly processed agricultural products would be US$154 billion higher, and global welfare would be US$168 billion higher, or US$101 billion if just agricultural and food policies were liberalized (Valenzuela, van der Mensbrugghe, and Anderson 2009, tables 13.14 and 13.16). This compares with the global TRI and WRI of US$138 billion and US$282 billion, respectively, for 2000­04 for just the 75 focus countries examined here and for just farm products (calculated as shown in the notes to figure 11.9). The welfare result from the Linkage Model is smaller than the WRI number, despite the model's broader coverage of products and countries, because it takes into account the general equilibrium effects of other (including nonagricultural) distortions at home and also distortions abroad insofar as they affect international prices, whereas the global WRI is obtained sim- ply by summing the WRIs of each country. A better comparison would have been with a set of model scenarios where only farm policies were liberalized in only one of the 75 countries at a time, but that would require 75 simulations; it remains an area for further research. Conclusion This chapter provides a panel set of index estimates that is well grounded in trade theory and that takes into account various forms of agricultural price and trade taxes and subsidies. The panel set covers 75 countries over the past half century. It provides a very useful supplement to the various indicators of the mean and vari- ance of aggregate NRAs and CTEs and the TBI used in previous chapters in this volume, especially from the viewpoint of the likely economic welfare or trade impacts of a country's structure of assistance to and taxation of agricultural indus- tries and food consumers. The indexes presented here can thus serve as inputs into cross-county studies of the impact over time of agricultural distortions on growth, poverty, unemployment, and so forth. They also are important supple- ments to the NRA and CTE in improving understanding of the long-term history of food and agricultural price and trade policies. That is especially true in seeking an index of global distortions when developing and high-income countries' NRAs or CTEs tend to offset each other. The new indexes suggest the world was not very Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 455 much less distorted by 2004 than it was in the 1960 (although it certainly was compared with the latter 1980s), and that the level of distortion is far higher than that suggested by the global average NRA or CTE. There would be high returns to further research in this area. The above esti- mates are based on the assumption that domestic price elasticities of supply (demand) are equal across commodities within a country. The estimates could thus be refined by relaxing the assumption that the price elasticities of demand (or supply) across goods are equal. This would entail a move to "marginal welfare weights," instead of production and consumption share weights when estimating the PDI and CDI, respectively. Kee, Nicita, and Olarreaga (2009) provide a methodology for estimating elasticities that could be adapted to the Distortions to Agricultural Incentives database. Finally, the above equations could also be developed so that estimates of the distortions of consumer and producer prices for a particular commodity in indi- vidual countries are aggregated across countries in order to obtain partial equilib- rium indexes of the reduction in world trade and economic welfare for any chosen global commodity market. The first attempt to do that is presented in the next chapter of this volume (Anderson et al. 2009). Notes 1. The NRA and CTE measures are related to the well-known producer and consumer subsidy equivalent (PSE and CSE) measures estimated by the OECD (2008). Their main conceptual difference is that the NRA and CTE are expressed as a percentage of the undistorted price, whereas the PSE and CSE are expressed as a percentage of the distorted price. The NRA and CTE values are identical if the only government interventions are at a country's border (such as a tariff on imports). In the case of agriculture, however, there are typically domestic production or consumption taxes or subsidies also in place, so the NRAs and CTEs differ. 2. See Anderson and Valenzuela (2008). Within each region, the shares of agricultural value added that the studied countries represented in 1990­2004 at distorted prices are 90 percent for high-income countries, 92 percent for European transition economies (including Turkey and Central Asia), and 86 percent for developing countries (76 percent in Africa, 94 percent in Asia, 81 percent in Latin America and the Caribbean, and 0 percent in the Middle East), and hence 88 percent globally. 3. Throughout this chapter, indirect effects of sectoral and trade policy measures directed at non- agricultural sectors are ignored. Standard assumptions in basic trade theory are adopted, notably that there are no divergences between private and social marginal costs and that benefits that might arise from externalities, market failures, and any other behind-the-border policies not represented in the analysis here, including such things as underinvestment in public goods. 4. For a set of CGE estimates of the welfare, trade, and various other economic effects of the poli- cies captured in the Agricultural Distortions database, see Valenzuela, van der Mensbrugghe, and Anderson (2009). 5. In addition, indexes can be defined as the ad valorem trade tax rate, which, if applied uniformly across countries for a particular product, would generate the same global reduction in trade in that product (or global economic welfare loss) as the actual cross-country structure of NRAs and CTEs for that tradable commodity. See chapter 12 of this volume for such an analysis. 456 Distortions to Agricultural Incentives: A Global Perspective 6. This chapter builds on chapter 12 of Anderson and Neary (2005), which is devoted to a consid- eration of how to deal with domestic price distortions. 7. MacLaren and Lloyd (2008) analyse the production side of the Australian agricultural sector with a production distortion index (although they use the word "assistance" rather than "distortion"). This is the uniform production subsidy that gives the same deadweight production loss as the actual differentiated structure of assistance, and so is exactly equal to the production component derived above. Here, a similar uniform consumption tax component (call it a consumption distortion index) is added, and seeks a TRI that gives the same trade-reducing effect as the sum of the actual trade effects on the two sides of the market. Likewise, a WRI is generated that gives the same deadweight welfare loss as the sum of the actual welfare losses on both sides of the market. 8. This insight is usually attributed to Harberger (1959). In fact, it was discovered by Dupuit (1844), more than 100 years before Harberger, while analysing the welfare loss resulting from commodity taxa- tion. In his words, "the loss of utility increases as the square of the tax" (Dupuit 1844, p. 281). Dupuit's contribution to consumer surplus and welfare analysis is considered in Humphrey (1992). 9. Expressions for measures of trade restriction and welfare reduction in Anderson and Neary (2005) use the same weights. 10. Estimates of the NRA for total agricultural production in studied countries are obtained by making guesstimates of the rates of assistance for the remaining 30 percent of agricultural production. Those guesstimates are not used in the present study, but their impact can be seen by comparing the third and fourth sets of rows of NRAs in Table 11.1. 11. National WRIs and TRIs are aggregated across countries using as weights an average of the value of consumption and production at undistorted prices. National and regional WRIs and TRIs for the five-year periods are unweighted averages of the annual indexes. References Allen, M. 2005. "Review of the IMF's Trade Restrictiveness Index." Policy Development and Review Department, International Monetary Fund, Washington, DC. Anderson, J. E., G. J. Bannister, and J. P. Neary. 1995. "Domestic Distortions and International Trade." International Economic Review 36 (1): 139­57. Anderson, J. E., and J. P. Neary. 2005. Measuring the Restrictiveness of International Trade Policy. Cam- bridge, MA: MIT Press. Anderson, K., and J. Croser. 2009. National and Global Agricultural Trade and Welfare Reduction Indexes, 1955 to 2007. Supplementary database at http://www.worldbank.org/agdistortions. Anderson, K., J. Croser, S. Nelgen, and E. Valenzuela. 2009. "Global Distortions to Key Commodity Markets." Chapter 12 in this volume. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC (and appendix A of this volume). ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson, K., and E. Valenzuela. 2008. "Estimates of Global Distortions to Agricultural Incentives, 1955 to 2007." Core database at http://www.worldbank.org/agdistortions. Dupuit, J. 1844. "On the Measurement of Utility of Public Works." Annals des Ponts et Chaussees, 2nd series, Volume 8. Feenstra, R. C. 1995. "Estimating the Effects of Trade Policy." In Handbook of International Economics, ed. G. N. Grossman and K. Rogoff. Amsterdam: Elsevier. Harberger, A. C. 1959. "Using the Resources at Hand More Effectively." American Economic Review 49 (2): 134­46. Humphrey, T. M. 1992. "Marshallian Cross Diagrams and Their Uses before Alfred Marshall: The Ori- gins of Demand and Supply Geometry." Economic Review (March/April): 3­23. Welfare-Based and Trade-Based Indicators of National Agricultural Distortions 457 Irwin, D. A. Forthcoming. The Battle over Protection: A History of US Trade Policy. Kee, H. L., A. Nicita, and M. Olearraga. 2008. "Import Demand Elasticities and Trade Distortions." Review of Economics and Statistics 90 (4): 666­82. ______. 2009. "Estimating Trade Restrictiveness Indexes." Economic Journal 119 (534): 172­99. Lloyd, P. J. 1974. "A More General Theory of Price Distortions in an Open Economy." Journal of Inter- national Economics 4 (4): 365­86. Lloyd, P. J., J. L. Croser, and K. Anderson. 2009a. "Welfare- and Trade-based Indicators of National Distortions." Agricultural Distortions Working Paper 72, World Bank, Washington, DC. ______. 2009b. "Global Distortions to Agricultural Markets: New Indicators of Trade and Welfare Impacts, 1960 to 2007." Policy Research Working Paper 4865, World Bank, Washington, DC. MacLaren, D., and Lloyd, P. J. 2008. "Measuring Assistance to the Agricultural Industry in Australia Using a Production Assistance Index." Paper presented at the 52nd Annual Conference of the Aus- tralian Agricultural and Resource Economics Society, Canberra, 5­8 February. OECD. 2008. OECD PSE-CSE Database (Producer and Consumer Support Estimates, OECD Database 1986­2007). Organisation for Economic Co-operation and Development. http://www.oecd.org/ document/55/0,3343,en_2649_33727_36956855_1_1_1_1,00.html. Valenzuela, E., D. van der Mensbrugghe, and K. Anderson. 2009. "General Equilibrium Effects of Price Distortions on Global Markets, Farm Incomes, and Welfare." Chapter 13 in this volume. 12 Global Distortions to Key Commodity Markets Kym Anderson, Johanna L. Croser, Signe Nelgen, and Ernesto Valenzuela* The regional books that preceded and provided the developing country details for the present one,1 like the regional summary chapters in Parts II and III of this book, are all country focused. While they include commodity details for their par- ticular country or region, they are not able to provide an overview for developing countries or high-income countries as a group, or for the world as a whole. Nor is there space in the opening chapter of this book to provide a detailed commodity- focused summary. This chapter seeks to fill that gap. The chapter begins by describing the overall project's coverage of 30 major commodities and their importance in regional and global agricultural production and trade (more details are in appendix B of this volume). It then summarizes the nominal rates of assistance (NRAs) and consumer tax equivalents (CTEs) for 12 key covered products, together with their gross subsidy and tax equivalents in constant dollars. The policies generating positive or negative NRAs and CTEs are often an attempt by a government not only to raise or lower the trend level of domestic producer or consumer prices relative to those in international markets, but also to reduce price and quantity volatility in the domestic market for key farm products. Given that this issue became headline news again when international food prices spiked in 2007 and 2008, the issue of whether domestic prices of key products *The authors are grateful for the estimates of distortions provided by the authors of country studies compiled by Anderson and Valenzuela (2008) and Anderson and Croser (2009), and for help with data compilation and manipu- lation by Esteban Jara and Marianne Kurzweil. 459 460 Distortions to Agricultural Incentives: A Global Perspective have in fact been more stable than prices in international markets over the past 50 years is examined. There is space to discuss this issue only briefly here; however, the Agricultural Distortions Project database provides much more scope for in- depth study of the role of policies in influencing market volatility. This chapter also examines seven largely nontraded food staples that are nonetheless important food items for poor people in low-income countries. Even though these commodities are only a small share of global production and exports of farm products, they can be crucial to the food security of large seg- ments of developing-country populations. The Agricultural Distortions Project database lends itself to placing the policies affecting (or ignoring) those products in a broader perspective. Finally, this chapter provides another new perspective on the project's data- base. It seeks to shed light on how relatively distorted various commodity markets are from the viewpoint of global trade or welfare restrictiveness. This analysis draws on the theory outlined in the previous chapter, but switches the focus from countries to products. True, a global model of each commodity market (or a global economy-wide computable general equilibrium [CGE] model) calibrated for a particular year of interest could provide such insights for that year: the NRA and CTE estimates for that product could be inserted into such a model to gener- ate partial (or general) equilibrium estimates of the global trade and welfare effects of those distortionary policies. However, global models do not exist for many commodities, and global CGE models such as the one used by Valenzuela, van der Mensbrugghe, and Anderson (2009) in the next chapter typically must aggregate many of the smaller commodities into groups to keep the model tractable. Moreover, such models are calibrated to a particular year and so are incapable of providing a long time series of estimates of the global trade and welfare effects of distortionary policies affecting particular commodity markets. The global trade reduction indexes (TRIs) and welfare reduction indexes (WRIs) used here are calculated for each of 12 key agricultural commodities for each year over the past half century, based on NRA and CTE estimates for the project's sam- ple of 75 countries. These two new indexes provide for each product the ad val- orem trade tax rate at which, if applied uniformly to that commodity in every country, would generate the same partial equilibrium reduction in trade or economic welfare as the actual structure across countries of NRAs and CTEs for that tradable commodity. NRA and CTE Coverage of Key Farm Products The Agricultural Distortions Project involves the estimation of annual NRAs and CTEs over the past five decades for 75 focus countries. In aggregate, the coverage Global Distortions to Key Commodity Markets 461 represents around 70 percent of the gross value of agricultural production in those focus countries and just under two-thirds of global farm production valued at undistorted prices over the period covered. The study has generated about 30,000 NRA and CTE estimates covering more than 70 different products, with an average of 11 products per country. Not all countries had data for the entire 1955­2007 period; the average number of years covered is 41 per country (see table 12.A.1 in the annex.). The NRAs estimated by the project cover more than three-quarters of global output of the 30 most valuable agricultural products in terms of their share of global farm production, and as much as five-sixths for grains and tubers. This chapter focuses on the 12 main commodities: three meats plus milk (worth 55 percent of global output of these 12 products), three grains plus soybeans (37 percent of this group's output), and three tropical products plus sugar (just 8 percent of global output of these 12 products, but of the most significance to agriculture in developing countries). All but one-seventh of global production of these 12 products is covered by the project's NRAs and CTEs (table 12.1). That Table 12.1. Coverage of Gross Value of Agricultural Production at Undistorted Prices for 12 Key Covered Products, 2000­04 (percent) NRA coverage (%) of Product's share of product's global global production value value of production of 12 key products Grains and oilseeds 93 37 Rice 92 13 Wheat 89 10 Maize 94 9 Soybeans 96 5 Tropical crops 80 8 Sugar 87 3 Cotton 82 3 Coconuts 60 1 Coffee 75 1 Livestock products 82 55 Milk 83 15 Beef 69 14 Pig meat 91 16 Poultry 81 10 All above products 86 100 Source: Authors' calculations based on the Anderson and Valenzuela (2008) database and FAO commodity balance and production data. 462 Distortions to Agricultural Incentives: A Global Perspective coverage is spread relatively equally across the five regions examined, as shown in table 12.2. Though each region's share of the 12 key products varies considerably, as a group, NRA coverage of those 12 products ranges from one-third of agricul- tural production in the focus countries of Africa to one-half in Latin America and high-income countries (table 12.3). Taken together, these coverage statistics sug- gest the NRAs can be considered very representative of the regional and global agricultural economies. GSEs and NRAs, by Product The gross subsidy equivalents (GSEs) of the NRAs for the 12 key products are summarized in figure 12.1 for developing and high-income countries sepa- rately, as well as for all of the study's focus countries. These estimates are obtained by multiplying the NRA by the value of production of each product at undistorted prices for each country, and summing across countries. The prod- ucts attracting the largest subsidies in 2000­04 were the rice pudding ingredi- ents of rice, milk, and sugar, plus beef, with milk dominating by far (and even more so two decades earlier). For some countries, the GSE for a particular product is negative, which offsets positive assistance in other countries. Rice is a case in point: it received positive assistance in both developing and high-income countries in 2000­04, but in 1980­84 developing countries in aggregate taxed rice production more than high-income countries subsidized it. Figure 12.1 also shows that assistance fell for the majority of those 12 prod- ucts in high-income countries over the past quarter century, becoming less positive, whereas in developing countries it rose, becoming less negative. Hence, for all focus countries as a group, the picture is mixed: sugar was more assisted in 2000­04 than in 1980­84; wheat, beef, and especially milk became less assisted over those years; and rice, maize, and pig meat have moved from being taxed in aggregate to being subsidized. Coffee, coconuts, and cotton are all less taxed now than in the early 1980s. The complete time series of the GSEs is summarized in table 12.4, from 1965 to 2004. Throughout that period, livestock assistance dominated crop assistance globally, and by a huge margin before the mid-1980s, when crop assistance was negative in many years. Typically, developing countries have not taxed milk so that, unlike for meat in earlier years, milk is not an offset to the positive assistance in high-income countries. Developing countries switched from negative to posi- tive assistance during the 1980s for wheat and sugar, and during the 1990s for rice, maize, and meat, while assistance for the three tropical crops of coconuts, coffee, and cotton remained slightly negative into the 2000s. Table 12.2. Share of Global Gross Value of Agricultural Production for Key Covered Products, by Region,a 2000­04 (percent) Eastern Europe High-income All focus Africa Asia Latin America and CIS countries countries countries Residual World Grains and oilseeds 11 39 5 6 23 84 16 100 Rice 3 81 2 0 5 92 8 100 Wheat 6 32 4 14 33 89 11 100 Maize 11 26 13 5 40 94 6 100 Soybeans 0 15 37 0 43 96 4 100 Tropical crops 10 36 12 5 11 74 26 100 Sugar 5 43 17 6 16 87 13 100 Cotton 11 30 5 14 22 82 18 100 Coconuts 0 60 0 0 0 60 40 100 Coffee 11 12 52 0 0 75 25 100 Livestock products 3 21 6 7 36 72 28 100 Milk 3 21 4 12 43 83 17 100 Beef 6 1 16 5 41 69 31 100 Pig meat 0 49 3 6 34 91 9 100 Poultry 2 27 9 5 38 81 19 100 Source: Authors' calculations based on the Anderson and Valenzuela (2008) database and FAO commodity balance and production data. a. The group averages refer to 30 key products, and in total there are more than 70 products covered by the project, even though only 12 are shown separately here. 463 464 Table 12.3. Share of Regional Gross Value of Agricultural Production for Major Covered Products,a by Region, 2000­04 (percent) Eastern Europe High-income Africa Asia Latin America and CIS countries countries Grains and oilseeds 16 23 13 12 16 Rice 2.8 13.6 1.9 0.1 1.0 Wheat 4.7 4.6 3.0 10.2 5.6 Maize 8.4 3.3 8.3 2.7 6.2 Soybeans 0.0 1.1 13.3 0.0 3.6 Tropical crops 4.3 3.8 8.0 3.3 1.6 Sugar 1.2 1.9 3.8 1.3 0.8 Cotton 2.3 1.0 0.9 2.1 0.9 Coconuts -- 0.8 -- -- -- Coffee 0.8 0.1 3.3 -- -- Livestock products 12 19 28 24 33 Milk 3.5 4.5 4.4 11.8 11.1 Beef 6.8 0.2 14.7 4.6 9.1 Pig meat -- 10.6 3.0 6.6 8.9 Poultry 1.6 3.6 5.9 2.9 6.2 Total of above 12 products 32 45 49 39 51 All covered products 68 66 70 61 72 Noncovered products 32 34 30 39 28 All agricultural products 100 100 100 100 100 Source: Authors' calculations based on the Anderson and Valenzuela (2008) database and FAO commodity balance and production data. Note: -- not available. a. The product group averages refer to 30 key products, and in total there are more than 70 products covered by the project, even though only 12 are shown separately here. Figure 12.1. GSEs of Assistance to Farmers Globally, by Product, 1980­84 and 2000­04 a. Developing countries b. High-income countries c. World rice milk milk milk rice rice beef sugar beef sugar poultry pig meat poultry wheat poultry pig meat pig meat sugar maize maize maize wheat coffee soybeans soybeans rapeseeds coconuts wheat barley soybeans cotton cotton cotton barley coffee beef rapeseeds coconuts 50,000 0 50,000 100,000 50,000 0 50,000 100,000 50,000 0 50,000 100,000 constant 2000 US$ millions constant 2000 US$ millions constant 2000 US$ millions 1980­84 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. 465 466 Table 12.4. GSEs of Assistance to Farm Industries, by Focus Country Group,a 1965­2007 a. All focus countries (constant 2000 US$ per year) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Grains and oilseeds 12,911 2,700 11,442 10.561 44.083 38.345 37.908 36,655 Rice 2,009 3,700 9,851 15.006 23.352 23.491 24.633 28,189 Wheat 8,365 881 974 7.010 16.094 12.663 7.993 3,489 Maize 2,477 1,995 1,300 2.259 5.507 1.770 3.347 4,124 Soybeans 61 114 682 307 869 422 1.935 853 Tropical crops 7,053 7,289 6,757 6.521 2.092 3.468 5.716 10,683 Sugar 8,287 6,247 547 3.134 7.816 7.211 8.958 10,750 Cotton 94 927 2,008 2.230 1.422 2.151 1.297 228 Coconuts 110 543 256 841 841 1.117 1.017 273 Coffee 1,030 1,425 5,040 6,584 3.462 476 928 21 Livestock products 61,368 66,214 105,824 77,798 97.486 94.166 86.491 76,757 Milk 35,581 39,518 72,029 73,126 73.973 59.982 46.208 43,974 Beef 7,350 7,364 10,554 17,018 27.272 21.052 19.998 13,986 Pig meat 15,792 15,132 17,550 19,655 9.729 3.382 9.874 8.927 Poultry 2,644 4,201 5,691 7,309 5.969 9.750 10.411 9,869 Total of above 12 products 81,332 56,225 110,509 60,716 143.661 135.979 130.116 124.095 b. Developing countries (constant 2000 US$ per year) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Grains and oilseeds 6,971 15,204 14,214 38,079 3,883 7,934 9,494 15,235 Rice 9,278 14,804 11,835 32,706 5,435 7,230 2,513 12,245 Wheat 1,574 108 115 453 3,191 1,361 4,720 2,297 Maize 673 396 1,579 4,611 28 2,092 1,106 1,222 Soybeans 61 112 685 308 1,611 27 1,155 529 Tropical crops 331 8,680 12,538 12,092 5,003 3,181 503 4,222 Sugar 2,980 6,249 4,819 1,982 889 440 2,273 5,103 Cotton 1,509 462 2,424 2,684 1,590 2,028 832 586 Coconuts 110 543 256 841 841 1,117 1,017 273 Coffee 1,030 1,425 5,040 6,584 3,462 476 928 21 Livestock products 755 1,814 11,494 22,745 2,035 7,066 14,248 12,930 Milk 263 55 9,639 11,242 13,198 6,443 5,610 8,684 Beef 1,914 2,793 307 298 1,583 608 1,926 965 Pig meat 671 883 1,352 36,180 16,910 1,207 3,323 2,125 Poultry 225 41 810 2,491 94 2,438 3,388 3,085 Total of above 12 products 7,396 25,699 15,258 72,916 10,921 4,049 23,238 32,387 (Table continues on the following page.) 467 468 Table 12.4. GSEs of Assistance to Farm Industries, by Focus Country Group,a 1965­2007 (continued) c. High-income countries (constant 2000 US$ per year) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07 Grains and oilseeds 19,889 12,523 25,672 27,495 47,967 46,277 28,416 21,431 15,359 Rice 11,291 11,128 21,696 17,685 28,791 30,717 22,126 15,955 11,431 Wheat 6,795 991 1,095 7,459 12,894 11,300 3,272 1,192 1,351 Maize 1,804 2,387 2,879 2,349 5,542 3,862 2,243 2,905 2,408 Soybeans 0 2 2 1 739 397 776 1,379 169 Tropical crops 10,304 3,555 11,135 6,948 15,900 14,940 8,037 6,595 5,120 Sugar 5,301 1 5,373 5,117 6,914 6,775 6,682 5,645 2,819 Cotton 1,414 1,390 416 455 168 119 467 816 1,992 Barley 3,563 2,135 5,324 1,359 7,154 7,175 1,858 129 307 Rapeseeds 26 29 22 17 1,664 1,110 36 6 2 Livestock products 62,126 68,044 94,370 100,534 99,476 87,122 72,259 63,791 34,486 Milk 35,312 39,462 62,441 61,852 60,757 53,568 40,626 35,260 13,117 Beef 9,277 10,171 10,854 17,324 25,661 21,648 18,062 14,953 8,519 Pig meat 15,121 14,249 16,196 16,534 7,176 4,590 6,543 6,802 7,206 Poultry 2,416 4,162 4,879 4,824 5,882 7,316 7,027 6,777 5,643 Total of above 12 products 92,319 84,122 131,176 134,976 163,343 148,339 108,712 91,817 54,964 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. Does not include NPS or decoupled assistance, nor assistance provided by nonfocus countries. Global Distortions to Key Commodity Markets 469 Figure 12.2. NRAs, Key Covered Products, High-Income and Developing Countriesa, 1980­84 and 2000­04 a. Developing countries b. High-income countries sugar rice 387 milk sugar rice milk poultry beef wheat poultry maize cotton pig meat pig meat coffee soybeans soybeans maize beef wheat coconuts barley cotton rapeseeds 100 0 100 100 0 100 200 300 percent percent 1980­84 2000­04 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. Product NRAs are averages of country NRAs weighted by the value of production at undistorted prices. These GSE values are the combined effect of output values reflected in tables 12.1 to 12.3 and NRAs shown in figure 12.2. Figure 12.2 confirms that the three rice pudding ingredients share the dubious honor of being the most assisted products in percentage terms in both developing and high-income countries. The time series since 1965 is shown in table 12.5 for all focus countries, where it is evi- dent that it is not just the value of the livestock sector but also its high NRAs that contribute to its dominance in GSE terms, with milk the stand-out assisted prod- uct. The distribution across countries of NRAs for these individual products is considered below, beginning with the most assisted. Rice, milk, and sugar The first thing that is striking about figure 12.3 is that virtually all countries for which NRAs have been estimated for rice, milk, and sugar assisted these three 470 Table 12.5. NRAs, 12 Key Covered Farm Products,a All Focus Countries, 1965­2004 (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 b Grains and oilseeds 11 6 5 3 21 16 14 17 Rice 6 11 12 10 26 25 23 39 Wheat 22 7 2 9 32 23 12 6 Maize 8 5 2 3 12 3 6 7 Soybeans 1 0 2 1 2 1 7 4 Tropical cropsb 34 5 9 8 4 7 11 27 Sugar 157 4 9 15 39 28 39 60 Cotton 0 9 9 12 8 10 6 3 Coconuts 24 8 3 11 19 34 22 8 Coffee 31 33 43 43 31 8 10 0 Livestock productsb 46 39 50 30 42 35 30 27 Milk 97 91 140 138 151 85 62 53 Beef 14 12 13 25 43 29 31 23 Pig meat 47 36 31 16 11 4 10 10 Poultry 20 26 26 29 21 26 20 19 Total of above 12 products 29 18 21 10 28 24 21 23 All covered productsb 24 15 18 6 16 18 16 16 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. The group averages refer to 30 key products, and in total there are more than 70 products covered by the project, even though only 12 are shown separately here b. Weighted averages using value of production at undistorted prices. Figure 12.3. NRAs, Rice, Milk, and Sugar, by Country, 2000­04 a. Rice b. Milk c. Sugar Japan Iceland Switzerland Korea, Rep. of Japan Bangladesh Norway EU-15 Taiwan, China Romania Turkey Switzerland Hungary Dominican Republic Korea, Rep. of Lithuania Colombia Canada Vietnam Romania Japan Malaysia Latvia United States Colombia Slovenia Sudan Philippines Turkey Hungary United States Nicaragua Mexico Ecuador Slovenia United States Colombia Mexico Global Distortions to Key Commodity Markets Slovak Republic Tanzania Ghana EU-15 Mozambique Côte d'Ivoire Czech Republic Pakistan Mozambique Mexico Turkey Poland Vietnam Poland Philippines India Estonia Slovak Republic Indonesia India Czech Republic EU-15 Sudan Russia Russia Indonesia Uganda South Africa Tanzania China Bulgaria Brazil Pakistan Nicaragua Nigeria Latvia India Bulgaria Kenya Madagascar Ukraine Ecuador Senegal Chile Chile China Sri Lanka Lithuania Bangladesh Uganda Nicaragua Egypt, Arab Rep. of Australia New Zealand Ecuador China Argentina Thailand Thailand Australia Kazakhstan Pakistan Dominican Republic Kazakhstan Brazil Zambia Egypt, Arab Rep. of Australia Egypt, Arab Rep. of Ukraine Madagascar 471 100 0 100 200 300 400 500 100 0 100 200 300 0 100 200 300 percent percent percent Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. 472 Distortions to Agricultural Incentives: A Global Perspective industries in 2000­04. The only exceptions are Ukraine and the Arab Republic of Egypt for milk and, for rice, Egypt, Zambia, Pakistan, and (very slightly) Thailand and China. Second, in virtually no country in the project's sample does the gov- ernment not intervene in the market for these three products. The third striking feature of figure 12.3 is the huge rates of assistance for these products in some countries, with the peak rate for each product exceeding 200 percent in 2000­04. This is far higher than the peak NRAs for the other products in the sample of 12, with the exception of poultry (and beef in Norway). These three features, espe- cially the third one, suggest there are characteristics these industries have in common that influence the political economy of support for them. One thing milk and sugar share with poultry is the need for immediate processing of the raw farm product before it is saleable to consumers, but that is also true of other prod- ucts such as cotton. The definitive political economy paper on these products has yet to be written, but perhaps the availability of the project's NRA database will catalyze such an analysis. With high protection for the 12 products examined here in virtually all mar- kets, their international price has been depressed perhaps more than that for most other farm products. This has been very harmful for the main unsubsidized exporters of these products, notably Thailand for rice, New Zealand for milk products, and Brazil and Australia for sugar. Beef, pig meat, and poultry Meat industries too tend to be mostly assisted, except for beef, where there are sev- eral countries still taxing its production (often implicitly, for example via export taxes or restrictions). The highest NRAs by far for these meats are in Norway, Switzerland, and Northeast Asia. Since there are almost no pig and poultry indus- tries with negative assistance, and since high-income countries dominate the upper part of figure 12.4, this suggests that those countries are overproducing these pro- tein-rich foods. And since all but low-income countries produce these products using intensive feeding of grains and oilseed meals, those policies are also putting upward pressure on the prices of those crop products--except perhaps in countries where assistance to intensive livestock industries is just to compensate for the pro- tection-induced high domestic price for feed grains and oilseeds. Wheat, maize, and soybeans These three temperate crop products are grown more in high-income countries than in any of the other four regions examined in the Agricultural Distortions Proj- ect (table 12.2). But where they are grown in developing countries (predominantly Figure 12.4. NRAs, Beef, Pig Meat, and Poultry, by Country, 2000­04 a. Beef b. Pig meat c. Poultry Norway Switzerland Switzerland 503 Switzerland Taiwan, China Iceland 498 Korea, Rep. of Norway Norway 341 Japan Taiwan, China 280 Turkey Korea, Rep. of Korea, Rep. of EU-15 Iceland Romania Iceland Kazakhstan Latvia Slovenia Ecuador Indonesia Taiwan, China Latvia Lithuania Romania Lithuania Slovenia Russia Japan Ecuador Slovak Republic Slovenia Russia Bulgaria EU-15 Kazakhstan Russia Ukraine Philippines Romania Estonia Chile Hungary Slovak Republic Brazil Eu-15 Bulgaria Canada Czech Republic Hungary New Zealand Egypt, Arab Rep. of Estonia Czech Republic United States Vietnam Philippines Australia Japan Turkey Ukraine Mexico Ukraine Czech Republic New Zealand Bulgaria Poland Argentina Mexico Mexico Nicaragua Colombia Canada Ecuador Latvia Brazil Thailand Lithuania New Zealand South Africa South Africa United States Canada Slovakia China Brazil Hungary Vietnam Australia Dominican Republic Nicaragua Estonia Thailand United States Sudan Philippines China Poland Poland Australia 0 100 200 0 100 200 0 100 200 300 473 100 100 100 percent percent percent Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. 474 Distortions to Agricultural Incentives: A Global Perspective in the Southern Hemisphere), the NRAs tend to be negative. There are also lots of countries where these products' NRAs are close to zero, and the peak NRAs are in countries that grow very little of them (again, mostly Norway, Switzerland, and Northeast Asia--see figure 12.5). Thus, the contribution of wheat, maize, and soybean subsidies to global farm subsidies is quite modest even though these prod- ucts account for a large share of the world's farm production (one-quarter of the 12 key products examined here, according to table 12.1). Coconut, cocoa, coffee, and cotton Coconut NRAs are not shown in figure 12.6 because there are only three countries in the sample for which they were estimated. In each case, the NRA by 2000 was slightly positive: 10 percent in Indonesia, 14 percent in the Philippines, and 17 percent in Sri Lanka for 2000­04. But in the last few decades of the 20th cen- tury, this tree crop's exports were taxed heavily. Cocoa NRAs are available for six countries, all on the equator. In all cases, pro- duction is still discouraged (negative NRAs), most heavily so in the major produc- ing country of Côte d'Ivoire. Coffee is now produced in many countries, and the extent of government intervention in this product varies from heavy taxation (again in Côte d'Ivoire, and almost as much as for cocoa) to slight assistance in the case of Brazil and Colombia in 2000­04. More than any of the key products examined here--and in sharp contrast to rice, milk, and sugar--cotton is simultaneously taxed heavily in developing coun- tries and subsidized heavily in high-income countries. The United States had an NRA of 70 percent for cotton in 2000­04,2 while several African countries have NRAs of around 70 percent--and several more were equally taxing of this industry in earlier decades (Baffes 2009). It also seems that some Central Asian countries still tax cotton producers substantially, although the data are not suffi- ciently robust to be able to estimate NRAs with confidence (Pomfret 2008). This wide diversity of NRAs means that freeing of cotton markets globally would lead to a big relocation of production from supporting countries, most notably the United States, to many poor countries, especially those in Africa and Central Asia that are currently underpricing raw cotton to growers. A recent study using an economy-wide model of the global economy (Anderson and Valenzuela 2007) strongly supports that inference. CTEs, by Product The CTEs of government intervention in nine of the key products examined here are fairly similar to the NRAs in percentage terms (compare tables 12.5 and Figure 12.5. NRAs, Wheat, Maize, and Soybeans, by Country, 2000­04 a. Wheat b. Maize c. Soybeans Korea, Rep. of Switzerland 757 Norway Nigeria Korea, Rep. of Japan Philippines Tanzania Romania Japan Slovenia Ecuador Switzerland Turkey Thailand Mexico Ghana Kenya Madagascar China Romania EU-15 India Colombia United States Turkey Slovenia Zambia Nicaragua Ecuador Sudan Egypt, Arab Rep. of Colombia India Romania Poland China Lithuania Indonesia Mozambique Colombia Chile Estonia Canada United States India South Africa Bulgaria Hungary Canada United States Chile EU-15 Kenya Poland Indonesia China Australia Egypt, Arab Rep. of New Zealand Slovak Republic EU-15 Cameroon Canada Uganda Brazil Thailand Australia New Zealand Brazil Australia Tanzania Brazil Latvia Russian Federation Bangladesh Mexico Mexico Bulgaria Hungary Czech Republic Ethopia Zambia Kazakhstan Ukraine Ethiopia South Africa Argentina Russian Federation Slovakia Pakistan Pakistan Nicaragua Argentina Argentina Ukraine Zambia Zimbabwe Zimbabwe Zimbabwe 475 100 0 100 200 100 0 100 100 0 100 200 percent percent percent Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. 476 Figure 12.6. NRAs, Cotton, Cocoa, and Coffee, by Country, 2000­04 a. Cotton b. Cocoa c. Coffee United States 70 Colombia Sudan India Brazil Brazil Indonesia Pakistan Colombia Ecuador Cameroon Mali Tanzania Australia Uganda Uganda Burkina Faso Cameroon Malaysia China Kenya Mozambique Ecuador Chad Ethiopia Benin Cameroon Vietnam Senegal Togo Nicaragua Nigeria Côte d'Ivoire Turkey Dominican Republic Madagascar Egypt, Arab Rep. of Mexico Zambia Ghana Zimbabwe Madagascar Tanzania Côte d'Ivoire Côte d'Ivoire Nigeria 100 50 0 50 100 50 0 100 50 0 50 percent percent percent Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. Global Distortions to Key Commodity Markets 477 12.6a), because intervention is mostly at the border and such measures affect producer and consumer prices equally. (The three tropical cash crops are not shown in table 12.6 because they are grown almost exclusively for export once they are lightly processed.) In cases where only border measures are used, the sign of the dollar equivalent of those taxes is the same as that of the GSE. However, the magnitude differs because these are heavily traded products and so countries dif- fer in the extent to which they are net importers or exporters of each one. Table 12.6b shows that of the meats, only pig meat consumption has been sub- sidized on a global basis. That global trend can be mostly attributed to China, and was phased out by the mid-1990s. Among the grain crops, rice consumption was taxed in aggregate in the focus countries only up to the mid-1980s, soybeans were taxed on net only in the 1970s, and maize consumption was taxed in aggregate only from the early 1990s. The Effects of Intervention on Price Variability Many governments intervene in commodity markets not only to alter the trend level of prices using long-term subsidies or taxes on farmers or food consumers, but also to attempt to reduce price and quantity volatility in the domestic market for key farm products. The justification sometimes given for such intervention in poor countries is that credit markets are underdeveloped or inefficient because of local monopoly lenders, so low-income consumers and producers have difficulty smoothing their consumption over time as prices fluctuate. There and in higher- income countries, the motive for intervention may be partly viewed also as a form of income insurance (Thompson et al. 2004), although it needs to be kept in mind that stabilizing prices is not the same as stabilizing incomes of the target house- holds. It is also true that to achieve price stability through altering trade barriers is extraordinarily difficult. Indeed, more than 60 years ago, Hayek (1945) warned that such intervention is likely to lead to government failure that could reduce welfare more than the cost of the market failure it seeks to overcome, given the high cost of the information needed to do it well. There is a huge analytical literature on the economics of price stabilization. Its innate connection with trade policy was highlighted by Johnson (1975) follow- ing the upward spike in world food prices in 1973­74. His analysis of grain prices suggested that if free trade in grain was in place in 1975, prices would be so much less variable--because trade could mitigate local supply variability--that only negligible quantities of carryover or storage would be profitable. A subsequent study of global food trade provided complementary results: using a stochastic model of world markets for grains, livestock products, and sugar, Tyers and Anderson (1992) found that instability of international food prices in the early 478 Table 12.6. CTEs of Policies Assisting Producers of Covered Farm Products, All Focus Countries, 1965­2007 a. Percent 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07a Cropsb 10 3 4 5 19 17 13 16 22 Rice 14 11 4 1 24 25 22 38 137 Wheat 19 2 3 12 27 16 6 2 5 Maize 11 7 8 2 5 3 2 2 3 Soybeans 1 3 1 3 1 0 7 4 8 Sugar 175 1 13 19 40 42 44 63 79 Livestock productsb 46 39 50 32 41 29 27 25 19 Milk 98 89 137 130 140 69 54 46 23 Beef 16 14 16 25 47 30 36 31 21 Pig meat 47 35 30 12 10 0 7 8 19 Poultry 23 28 27 28 18 21 18 19 16 Total of above 12 productsb 28 16 25 17 30 23 20 21 21 b. Aggregate value (constant 2000 US$ per year) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­07a Crops 13,090 20,549 10,751 12,591 46,803 49,139 38,465 40,780 34,167 Rice 7,405 15,615 4,303 511 21,994 23,714 22,973 27,390 15,748 Wheat 7,020 2,812 1,113 7,545 14,412 9,339 3,720 1,420 2,418 Maize 3,399 2,396 3,992 1,218 2,249 1,755 1,487 998 1,465 Soybeans 60 452 323 703 11 123 2,557 1,518 2,280 Sugar 10,016 4,065 1,667 3,636 8,159 17,718 10,702 11,450 12,255 Livestock products 62,252 66,748 106,150 80,323 96,353 83,687 82,670 76,759 41,500 Milk 34,929 38,158 70,180 69,282 69,593 52,186 41,196 40,069 13,019 Beef 8,622 8,945 12,604 17,432 30,110 23,143 24,990 18,906 12,589 Pig meat 15,702 15,119 17,544 13,400 8,648 82 6,971 7,487 9,676 Poultry 2,998 4,526 5,822 7,009 5,297 8,440 9,513 10,298 6,217 Total of above 12 products 75,342 46,200 116,901 92,914 143,156 132,826 121,135 117,539 75,667 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. a. The estimates for the period 2005­07 refer to only high-income-country policies. b. Weighted averages based on the value of consumption at undistorted prices. 479 480 Distortions to Agricultural Incentives: A Global Perspective Figure 12.7. Rice NRA and International Rice Price, South Asia, 1970­2005 600 30 20 500 10 0 400 10 percent US$ 300 20 30 200 40 50 100 60 0 70 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 20 NRA, South Asia (right axis) international price (left axis) Source: Authors' compilation based on data in Anderson and Valenzuela (2008). Note: Correlation coefficient is 0.75. 1980s was three times greater than it would have been under free trade in those products. Many countries vary their trade taxes and hence NRAs inversely with interna- tional prices, particularly for staple foods. Rice is perhaps the most obvious exam- ple. Anderson and Martin illustrate for Southeast Asia in chapter 9 in this volume, and figure 12.7 illustrates for South Asia, where the domestic rice NRA moves in the opposite direction to the world rice price with a high correlation coefficient of 0.75 (which compares with 0.59 for Southeast Asia for the same period). This desire to stabilize does not seem to be diminishing, even though the trend rate of NRA for rice is rising as incomes grow (see figure 12.8). One consequence of such domestic market-stabilizing activities by govern- ments is that the international market for food staples is "thinned." As shown in table 12.7, food staples are traded much less than tropical products by developing countries--and the numbers in the high-income countries column of that table would be much lower too had intra-European Union (EU) trade been excluded from the data. For example, only 6.9 percent of global rice production was traded internationally in 2000­03, compared with 14 and 24 percent for maize and wheat, and prior to the 1990s the global share of rice traded was less than 4.5 percent. If this market "thinning" matters most in low-income countries where con- sumption smoothing through time is most unaffordable for poor households, the Global Distortions to Key Commodity Markets 481 Figure 12.8. NRAs for Rice and Per capita Income, 1955­2007 800 600 NRA (percent) 400 200 0 200 6 7 8 9 10 11 in real GDP per capita developing (including Taiwan, China and Korea) R2 0.27 high-income countries and Europe and Central Asia R2 0.26 total R2 0.25 Source: Anderson and Valenzuela (2008), based on estimates reported in the project's national country studies. question arises as to how successful governments in that group of countries have been in keeping domestic price volatility below volatility in international markets. A recent attempt to test that, using the prices that generated the NRAs from this project, found that government intervention in low-income countries, on average, had destabilized prices relative to the international marketplace (Masters and Garcia 2009), apparently vindicating Hayek's concern cited above, and contrary to the general conclusion reached by Schiff and Valdés (1992) using data up to the mid-1980s. That is, policies continue to seek to reduce fluctuations in domestic food prices and in the quantities available for consumption via fluctuations in barriers to trade. This beggar-thy-neighbor dimension of each national economy's food policy reduces the international public good role that trade between nations can play in bringing stability to the world's food markets. The more some coun- tries insulate their domestic markets, the more other countries perceive a need to do likewise, exacerbating the effect on world prices so that even larger changes in NRAs are desired--a classic collective action problem, and one that was illus- trated yet again in 2007­08, when the imposition of export restrictions in key exporting countries in late 2007 and early 2008 contributed to the sharp increases 482 Distortions to Agricultural Incentives: A Global Perspective Table 12.7. Shares of Production Exported (X/Q) and of Consumption Imported (M/C) for Major Covered Products,a by Region, 2000­03 (percent) Latin Eastern America Europe High- and the and Central income Africa Asia Caribbean Asia coutnries Grains X/Q 2 7 11 13 29 M/C 17 9 22 11 27 Rice X/Q 6 6 1 2 32 M/C 28 1 12 59 15 Wheat X/Q 4 3 46 13 49 M/C 44 4 51 6 27 Maize X/Q 4 8 15 10 20 M/C 14 5 14 9 5 Tropical X/Q 52 38 45 32 47 crops M/C 13 18 12 42 42 Sugar X/Q 27 12 40 20 31 M/C 20 9 4 46 25 Cottonb X/Q 29 1 5 2 31 M/C 2 4 9 21 3 Coconuts X/Q -- 9 -- -- -- M/C -- -- -- -- -- Coffee X/Q 77 78 73 -- -- M/C 2 3 5 -- -- Livestock X/Q 1 4 10 7 20 M/C 8 6 5 9 14 Pig meat X/Q -- 1 12 6 21 M/C -- 2 8 12 20 Milk X/Q 0 0 5 2 7 M/C 1 1 5 1 3 Beef X/Q 1 2 10 6 23 M/C 9 48 5 14 20 Poultry X/Q 1 10 16 7 19 M/C 9 11 5 28 10 Total of above 12 X/Q 16 22 19 11 24 products M/C 13 14 12 11 19 Source: Authors' calculations based on Anderson and Valenzuela (2008) and FAO commodity balance and production data. a. The group averages refer to 30 key products, and in total there are more than 70 products covered by the project, even though only 12 are shown separately here. These data include intra-EU trade, which, if excluded, would have lowered substantially the numbers in the last column. b. Excluding data for the five cotton-exporting countries of Benin, Burkina Faso, Chad, Mali, and Togo. Global Distortions to Key Commodity Markets 483 in world prices in the first half of 2008. This is an area requiring considerably more analysis of past government behavior, and for which the current project's Agricultural Distortions Project database is well suited, but space and time limita- tions preclude it from being included in this volume. What about Nontraded Food Staples of Low-Income Countries? The NRA coverage of the 12 key traded products discussed above represents only one-third of agricultural production in Africa, compared with one-half of farm output in Latin America and high-income countries (table 12.3). Part of the rea- son for the difference is that in low-income countries where rural infrastructure is weak, trade costs are relatively high and so a larger proportion of food production focuses on products that tend to be mostly not traded internationally. They include root crops such as cassava, potatoes, sweet potatoes, and yams; grains such as millet; and fruits such as bananas and plantains. Apart from potatoes, these crops are almost exclusively grown in developing countries with hot climates, yet with different degrees of specialization across regions (table 12.8). Apart from bananas, these products are traded very little across borders, even with neighbor- ing countries (table 12.9). Yet in terms of calories and protein, bananas and plan- tains account for two-thirds of the fruit intake in Sub-Saharan Africa, and they, along with the four tubers and millet, account for one-quarter of all Sub-Saharan African food intake, according to Food and Agriculture Organization (FAO) food balance sheets in recent years. How much difference would it make if these products had been more fully included in the Agricultural Distortions Project database? Despite their impor- tance as a source of calories and protein, their share of the global value of produc- tion is quite low, because of their low prices. In aggregate, those seven products account for just 5 percent of the global value of farm output valued at domestic producer prices. In Africa, though, they account for a bit more than one-fifth of the regional value of agricultural production. For that reason, authors of the pro- ject's African studies typically include them in their sample of covered products. This can be seen from table 12.10, where the fourth row shows that these products accounted in 1995­2004 for one-third (22 percentage points) of the 67 percent coverage ratio for the African sample. The second set of rows in table 12.10 shows that had all developing countries included all seven of these products in their cov- ered sample, it would have raised their coverage by no more than 6 percentage points. Those rows also show that their inclusion in the African studies was nearly complete. 484 Table 12.8. Share of Global Production of Seven Mostly Nontraded Staple Crops, by Region, 1995­2004 (percent) High-income countries and Latin All developing Eastern Europe All focus Africa Asia America countries and CIS countries countries Residual World Cassava 37 28 14 79 0 79 19 100 Potato 2 29 4 35 52 87 13 100 Sweet potato 5 89 1 95 1 96 4 100 Yams 88 0 1 89 0 89 11 100 Millet 31 45 0 76 5 81 19 100 Banana 6 48 26 80 1 81 19 100 Plantain 56 2 13 70 0 70 30 100 Total of above seven crops 23 41 11 75 11 86 14 100 Source: Authors' calculations based on FAO (2008) volume of production data. Global Distortions to Key Commodity Markets 485 Table 12.9. Average of Focus Developing Countries' Self- Sufficiency Ratios for Seven Mostly Nontraded Staple Crops, by Region, 1961­2005 (regional production divided by regional production plus net imports) 1961­69 1970­79 1980­89 1990­99 2000­05 Africa Cassava 1.00 1.00 1.00 1.00 1.00 Potatoes 1.02 1.03 1.02 1.03 1.02 Sweet potatoes 1.00 1.00 1.00 1.00 1.00 Yams 1.00 1.00 1.00 1.00 1.00 Millet 1.00 0.99 1.00 1.00 1.00 Bananas 1.23 1.13 1.05 1.09 1.13 Plantains 1.00 1.00 1.00 1.00 1.00 Asia Cassava 1.04 1.10 1.18 1.13 1.04 Potatoes 1.00 1.00 1.00 1.00 1.00 Sweet potatoes 1.00 1.00 1.00 1.00 1.00 Yams 1.00 1.00 1.00 0.98 0.87 Millet 1.00 1.00 1.00 1.00 1.00 Bananas 1.04 1.08 1.06 1.04 1.04 Plantains 1.00 1.00 1.00 1.00 1.00 Latin America Cassava 1.00 1.00 1.00 1.00 1.00 Potatoes 1.00 0.99 1.00 0.99 0.99 Sweet potatoes 1.00 1.00 1.01 1.01 1.01 Yams 1.00 1.02 1.04 1.02 1.01 Millet 4.21 2.21 2.15 1.97 1.03 Bananas 1.23 1.21 1.25 1.48 1.52 Plantains 1.00 1.00 1.00 1.03 1.06 Source: Authors' calculations based on FAO (2008) volume of production and trade data. Together, those two facts suggest that the fuller inclusion of those seven staples in the covered product set would not have altered the developing countries' average NRA for covered products very much. To test that assertion more formally, the NRA for covered products, shown in the third set of rows in table 12.10, was recalculated to include any of the seven staples that were missing, assuming the NRAs for those missing staples were zero (since the nature of the market for these products is such that they attract very little government intervention). The final set of rows in table 12.10 show that such inclusion would bring the NRA average for covered products only very slightly closer to zero (for example, from 5.3 to 4.9 per- cent for developing countries as a group in 1995­2004). Moreover, their partial 486 Distortions to Agricultural Incentives: A Global Perspective Table 12.10. Additional Contribution of Seven Noncovered Staplesa to Values of Agricultural Production (VOP) and to Aggregate NRAs in Focus Developing Countries, by Region, 1966­2004 (percent, at undistorted prices) Latin All developing Africa Asia America countries Covered products' share of regional VOP (with seven staples' share in brackets) 1966­74 71 (16.5) 63 (1.3) 58 (1.2) 64 (2.7) 1975­84 69 (18.2) 71 (1.1) 69 (0.4) 71 (3.5) 1985­94 67 (18.6) 76 (1.3) 66 (0.5) 73 (4.0) 1995­2004 67 (22.1) 69 (2.6) 69 (0.9) 69 (5.8) Noncovered seven staples' share of regional VOP 1966­74 1.7 4.6 7.2 4.5 1975­84 2.4 5.9 8.3 5.8 1985­94 2.4 5.4 10.1 5.8 1995­2004 2.7 5.7 10.3 6.0 Covered products' weighted average NRA 1966­74 20.1 0.2 19.8 8.1 1975­84 16.2 10.7 17.1 13.8 1985­94 5.8 9.9 6.7 9.1 1995­2004 7.7 8.3 1.8 5.3 Covered plus seven noncovered weighted average NRAb 1966­74 19.6 0.2 17.6 7.6 1975­84 15.7 9.9 15.3 12.8 1985­94 5.6 9.2 5.8 8.4 1995­2004 7.4 7.7 1.6 4.9 Source: Authors' calculations based on FAO (2008) volume of production data and on NRAs from the project's national country studies as summarized in Anderson and Valenzuela (2008). a. The staples considered here are bananas, cassava, millet, plantains, potatoes, sweet potatoes, and yams. The undistorted prices for these products are assumed to be the domestic producer prices. b. Assumes the NRA and CTE for each of the seven noncovered staples is zero. omission from the covered set makes no difference to the NRA average for all agri- culture (including noncovered products), since in most cases the guesstimated NRA for noncovered, nontradable farm products in developing countries was zero anyway. Global Distortions to Key Commodity Markets 487 Global Commodity TRIs and WRIs3 Certainly, global models can estimate the trade and welfare effects of commodity policies, but such models typically are calibrated to a particular year and so are incapable of providing a long time series of estimates of the global effects of dis- tortionary policies affecting particular commodity markets. Drawing on the the- ory outlined in the previous chapter, but with a focus on products rather than countries, the analysis presented in this final section of the chapter provides time series estimates for a pair of partial equilibrium indexes that give more insights than just NRAs or CTEs can provide into the likely impact of policies in restrict- ing global trade in particular products and in reducing the contribution each product's market can make to global economic welfare. These two new indexes can provide for each product the ad valorem trade tax rate, which, if applied uniformly to that commodity in every country, would gen- erate the same partial equilibrium reduction in trade or economic welfare as the actual structure across countries of NRAs and CTEs for that tradable commod- ity. If one is willing to assume that the domestic price elasticities of supply are equal across countries for a particular commodity, and likewise for the domestic price elasticities of demand for that commodity (as indeed many global com- modity models do, for lack of country-specific econometric estimates), then there is no need to know the size of those elasticities in order to estimate the two new indexes. As in the previous chapter, these indicators are called the TRI and the WRI. One feature of both the WRI and the TRI is that they use a country's share of world production and consumption to determine the global welfare and trade effects of price-distorting policies. And an important feature of the WRI is that it takes into account the fact that the welfare effect of a policy such as an import tariff is related to the square of the tariff rate, which is particularly important in global commodity markets with a wide dispersion of NRAs across countries. The theoretical literature that identifies ways to measure the welfare- and trade-reducing effects of international trade policy in scalar index numbers stems from the theoretical advances by Anderson and Neary (summarized in and extended beyond their 2005 book) and the partial equilibrium simplifications by Feenstra (1995). Notwithstanding these advances, to the knowledge of the authors of this chapter, no long time series of indexes had been estimated across countries for individual commodities until Lloyd, Croser, and Anderson (2009) showed that the required theory is a straightforward variation of that summarized for coun- tries in chapter 11 of this volume. They have applied the theory to estimate the indexes for 28 of the commodity markets included in the Agricultural Distortions Project database, summing across countries for each product, in contrast to 488 Distortions to Agricultural Incentives: A Global Perspective chapter 11, where the summation is across products for each country. Here, results are summarized for just the 12 key global commodities that are the focus of this chapter, for each year since 1965, based on NRA and CTE estimates for the project's sample of 75 countries. Table 12.11 reports the time series of estimated global TRIs for each of the 12 agricultural commodities and for the three groups of commodities (grains and oilseeds, tropical crops, and livestock products). Generally, those TRIs are some- what above the NRAs reported in table 12.5, especially for tropical products for which the trade-reducing effects of import taxes of some high-income countries are reinforced by the export taxes of some lower-income countries. By contrast, for a few products, the global average TRI is less than the NRA, reflecting the fact that export subsidies have been in place for some higher-income countries or import subsidies for some lower-income countries. As indicated in the country studies presented in earlier chapters, the most trade-distorted products are sugar, milk, and rice. Among the grains, it is rice trade that has been taxed most since the 1970s, while among the oilseeds and trop- ical crops it is sesame and sugar trade that are taxed most. Maize and soybean trade has been taxed least among those crops shown, and at very low rates com- pared with livestock products, especially milk. Note, however, that the extent of distortions to trade has diminished more for livestock products than for crops since the 1980s, when agricultural price and trade reforms began to be imple- mented in numerous countries. Table 12.12 similarly reports the global WRI estimates. These are substantially above the NRAs, with five-year averages across the 12 commodities between 1965 and 2004 in the range of 55­85 percent, compared with the 6­24 percent range for the comparable NRA averages. This greater size is partly because the welfare cost is proportional to the square of the NRA, and partly because some NRAs are neg- ative and so offset positive NRAs in the process of averaging them whereas the welfare cost of those negative and positive NRAs are additive. Figure 12.9 shows that the most distorted among the 12 commodities in 2000­04 in terms of both global welfare cost and trade restrictiveness are rice, sugar, milk, and beef. A useful way of summarizing the WRI and TRI estimates for particular prod- ucts is provided in figure 12.10, which shows their movement since many of the indexes peaked in the late 1980s. The indexes would suggest policies for a particu- lar commodity market were not reducing either trade or welfare if the product were located at the zero point of both axes, that is, in the bottom left corner of the diagram (the "sweet spot"). Nearly all of the farm commodities shown have moved toward that spot since 1985­89, and very substantially so for the outliers-- namely milk and coffee--but also considerably so for wheat and maize. Table 12.11. Global TRIs, by Commodity, 1965­2004 (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Grains and oilseeds 18 15 15 20 19 18 15 9 Rice 50 58 42 41 58 53 32 43 Wheat 15 1 1 9 28 20 11 4 Maize 8 4 9 3 9 10 2 3 Soybeans 1 0 6 8 11 8 6 6 Tropical Crops 34 26 36 42 32 31 19 9 Sugar 143 27 40 47 56 44 41 55 Cotton 2 13 14 1 13 4 9 4 Coconuts 24 8 3 12 21 35 23 9 Coffee 30 31 37 46 33 13 12 2 Livestock products 53 37 48 53 49 37 23 25 Milk 83 79 133 131 125 63 53 45 Beef 20 17 18 32 47 32 33 32 Pig meat 37 28 25 47 25 11 9 8 Poultry 22 29 26 24 27 27 18 18 All of above 29 21 23 30 30 29 19 15 Source: Derived from estimates in Anderson and Croser (2009), based on NRA and CTE estimates in Anderson and Valenzuela (2008). 489 490 Table 12.12. Global WRIs, by Commodity, 1965­2004 (percent) 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Grains and oilseeds 44 42 47 49 89 82 61 58 Rice 65 86 75 75 150 152 116 141 Wheat 45 36 30 30 59 47 29 20 Maize 29 23 29 30 48 29 21 20 Soybeans 6 10 16 28 31 27 24 25 Tropical Crops 97 48 47 49 63 58 52 59 Sugar 224 58 68 72 99 76 77 87 Cotton 46 47 32 29 39 38 34 45 Coconuts 24 12 14 19 24 38 27 12 Coffee 32 35 44 50 38 31 22 15 Livestock products 83 76 91 89 88 68 54 52 Milk 161 149 218 182 191 111 83 73 Beef 43 42 47 66 93 76 72 68 Pig meat 79 66 59 70 42 33 27 28 Poultry 43 54 48 50 48 54 46 45 All of above 67 58 65 66 86 73 57 55 Source: Derived from estimates in Anderson and Croser (2009), based on NRA and CTE estimates in Anderson and Valenzuela (2008). Global Distortions to Key Commodity Markets 491 Figure 12.9. TRI and WRI for 12 Key Covered Products, 2000­04 160 140 120 100 percent 80 60 40 20 0 20 e r ilk ef try n t ns ze at e ts ea ga ric ffe tto nu be he m ai ea ul m su co m co co w yb po g co so pi WRI TRI Source: Derived from estimates in Anderson and Croser (2009), based on NRA and CTE estimates in Anderson and Valenzuela (2008). The countries that contribute most to the global TRI are shown in fig- ure 12.11 for the five most distorted products. These shares are related not only to the size of the index but also to the contribution of the country to global trade in that product. In the case of sugar, milk, and beef, many countries are highly protective of their domestic producers and so the contributions are rela- tively evenly spread across the countries. By contrast, rice trade restrictions are due mostly to a few Asian countries, notably, India, Japan, and Taiwan, China. Cotton trade distortions are even more concentrated, with subsidies in the United States the main contributor. Similar summary information for the country contributions to the global commodity WRIs are presented in figure 12.12. In this case, Japan is prominent in reducing world welfare in the markets of not just rice but also milk and beef. For cotton, distortionary policies not only in the United States but also in Turkey and several other large developing countries are dominant contributors, where it is the size of the country in global cotton production/consumption 492 Distortions to Agricultural Incentives: A Global Perspective Figure 12.10. Global TRI and WRI for Covered Tradable Farm Products, by Commodity, 1985­89 and 2000­04 a. Beef, milk, rice, sugar, and wheat 200 milk 150 rice rice WRI (percent) 100 sugar beef sugar beef milk wheat 50 wheat 0 0 25 50 75 100 125 150 TRI (percent) b. Coconut, coffee, cotton, maize, and pig meat 50 maize cotton pig meat 40 cotton coffee WRI (percent) 30 pig meat coconuts 20 maize coffee coconuts 10 0 10 0 10 20 30 40 TRI (percent) 1985­89 2000­04 Source: Derived from estimates in Anderson and Croser (2009), based on NRA and CTE estimates in Anderson and Valenzuela (2008). Figure 12.11. Country Share of the Global Commodity-Specific TRI for Rice, Sugar, Beef, Cotton, and Milk, 2000­04 a. Sugar b. Milk c. Rice (51 countries global ) (46 countries global (36 countries global TRI 54.8 TRI 44.5) TRI 42.9) India Japan Indonesia India United States United States India Japan Colombia Germany Germany Japan Switzerland Taiwan France France China Colombia Vietnam Lithuania Philippines Canada United Korea, Russia Kingdom Rep. of Pakistan Mexico Korea, Rep. of China Italy United Kingdom Italy Netherlands United Turkey Ukraine States 0 5 10 20 0 20 0 20 40 percent percent percent d. Beef e. Cotton (47 countries global TRI 32.0) (19 countries global TRI 4.1) Mexico United States Japan France China Canada Italy Pakistan Germany Brazil Brazil Korea, Rep. of India Slovenia Ukraine Côte d'Ivoire Czech Republic Spain Zambia Turkey Russian Federation Zimbabwe Ukraine Tanzania United States Colombia Egypt, Arab Rep. of Argentina Sudan Nigeria Nicaragua Poland Turkey 50 0 50 100 1,000 0 1,000 percent percent Source: Derived from estimates in Anderson and Croser (2009), based on NRA and CTE estimates in Anderson and Valenzuela (2008). Note: The decomposition over the five-year period can be greater than or less than 100, even though the decomposition sums to 100 in any one year. Here, the five-year averages are scaled so that the decompositions sum to 100. Focus countries have been omitted where the decomposition share has an absolute value of less than 2. 494 Figure 12.12. Country Share of the Global Commodity-Specific WRI for Rice, Sugar, Milk, Beef, and Cotton, 2000­04 a. Rice b. Sugar c. Milk d. Beef e. Cotton (51 countries global (46 countries global (36 countries global (47 countries global (19 countries global WRI 140.9) WRI 86.7) WRI 72.8) WRI 68.1) WRI 44.7) Taiwan, United States Japan Japan United States China Germany Sudan France United Japan France Colombia States Turkey Korea, Rep. of Japan Vietnam Switzerland Italy Lithuania Nigeria Korea, United Kingdom Germany Rep. of Canada United Italy Kingdom China United Indonesia Poland India States India Turkey Egypt, Arab Spain China Germany Spain Rep. of Bangladesh Pakistan Mexico India France Zimbabwe Turkey Slovenia 0 20 40 0 5 10 0 25 50 0 20 40 0 25 50 percent percent percent percent percent Source: Derived from estimates in Anderson and Croser (2009), based on NRA and CTE estimates in Anderson and Valenzuela (2008). Note: The decomposition over the five-year period can be greater than or less than 100, even though the decomposition sums to 100 in any one year. Here, the five-year averages are scaled so that the decompositions sum to 100. Focus countries have been omitted where the decomposition share has a value of less than 2. Global Distortions to Key Commodity Markets 495 that interacts with the percentage WRI to determine the aggregate contribution of each nation. In short, this application of these two additions to the family of so-called trade restrictiveness indexes provides very different indicators of dis- tortions to global agricultural markets than the NRAs and CTEs (and even more so than the OECD's producer and consumer support estimates, which are expressed as a percentage of distorted rather than undistorted prices and so are smaller than their NRA and CTE counterparts). More specifically, the TRI offers a much truer indication of the world trade effects of government inter- ventions in the markets for traded products, by properly accommodating trade subsidies alongside trade taxes, while the WRI offers a much truer indication of the global welfare effects of government interventions in the markets for traded products by also properly taking into account the fact that the welfare cost of a price distortion is proportional to the square of the tax or subsidy rate. These two indexes have been calculated with the help of a number of simpli- fying assumptions, most notably that each country is small and that its price elasticity of supply (demand) for a particular product is the same as that for every other country, that cross-price elasticities are zero, and that at the aggre- gate level the slopes of the demand and supply curve are equal. However, this is what trade negotiators typically assume when they attempt to calculate the trade effects of market access concessions they are considering exchanging. It is also what commonly would be assumed when calculating, for the arbitrator of a trade dispute settlement case, the magnitude of the trade damage from a viola- tion of commitments under a trade agreement. Models of the global market for particular farm products often have to make such assumptions too, for want of reliable or agreed econometric estimates of those elasticities for each country. Moreover, these indexes have the advantage over formal supply and demand models in that they can be expressed in time series form, thereby revealing trends and fluctuations over long periods, rather than just providing a snapshot at a point in time. 496 Distortions to Agricultural Incentives: A Global Perspective Annex Table 12.A.1 Summary of NRA Estimates by Major Product, Africa, Asia, and Latin America, 2000­04 a. Africa Weighted Number of average Gross value Product countries NRA, % of productiona Countries b Apples 1 0.3 0.15 South Africa Bananas 1 1.1 0.08 Cameroon Beans 3 25.1 0.49 Mozambique, Tanzania, Uganda Beef 3 26.0 5.89 Egypt, South Africa, Sudan Camel meat 1 87.7 0.10 Sudan Cashews 2 9.9 0.06 Mozambique, Tanzania Cassava 13 2.6 8.45 Benin, Burkina Faso, Cameroon, Chad, Côte d'Ivoire, Ghana, Madagascar, Mali, Mozambique, Nigeria, Tanzania, Togo, Uganda Chat 1 39.5 0.07 Ethiopia Cloves 1 18.7 0.05 Madagascar Cocoa 5 35.8 2.59 Cameroon, Côte d'Ivoire, Ghana, Madagascar, Nigeria Coffee 7 12.0 0.70 Cameroon, Côte d'Ivoire, Ethiopia, Kenya, Madagascar, Tanzania, Uganda Cotton 16 46.1 1.94 Benin, Burkina Faso, Cameroon, Côte d'Ivoire, Chad, Egypt, Mali, Mozambique, Nigeria, Senegal, Sudan, Tanzania, Togo, Uganda, Zambia, Zimbabwe Fruits and 1 0.0 0.14 Kenya vegetablesb Grapesb 1 7.4 0.21 South Africa Global Distortions to Key Commodity Markets 497 Weighted Number of average Gross value Product countries NRA, % of productiona Countries Groundnuts 8 40.3 1.72 Ghana, Mozambique, Nigeria, Senegal, Sudan, Uganda, Zambia, Zimbabwe Gum arabic 1 67.1 0.02 Sudan Hides and skins 1 48.4 0.03 Ethiopia Maize 13 5.4 7.24 Cameroon, Egypt, Ethiopia, Ghana, Kenya, Madagascar, Mozambique, Nigeria, South Africa, Tanzania, Uganda, Zambia, Zimbabwe Milk 2 14.6 2.99 Egypt, Sudan Millet 13 2.3 1.79 Benin, Burkina Faso, Cameroon, Chad, Mali, Mozambique, Nigeria, Senegal, Sudan, Tanzania, Togo, Uganda, Zambia Oilseeds 1 39.4 0.08 Ethiopia Orangesb 1 8.4 0.23 South Africa Roots and tubers 1 0.0 0.38 Cameroon Palm oil 1 12.6 0.73 Nigeria Peppers 1 10.2 0.00 Madagascar Plantains 5 0.1 1.93 Cameroon, Côte d'Ivoire, Ghana, Tanzania, Uganda Potatoes 2 0.0 0.07 Mozambique, Tanzania Poultry 1 2.7 1.36 South Africa Pulses 1 20.4 0.16 Ethiopia Pyrethrum 1 47.7 0.00 Tanzania Rice 10 5.5 2.45 Côte d'Ivoire, Egypt, Ghana, Madagascar, Mozambique, Nigeria, Sudan, Tanzania, Uganda, Zambia (Table continues on the following page.) 498 Distortions to Agricultural Incentives: A Global Perspective Table 12.A.1 Summary of NRA Estimates by Major Product, Africa, Asia, and Latin America, 2000­04 (continued) a. Africa Weighted Number of average Gross value Product countries NRA, % of productiona Countries Sesame 1 38.1 0.20 Sudan Sheep meat 2 21.4 1.57 South Africa, Sudan Sisal 1 0.0 0.01 Tanzania Sorghum 13 20.7 2.13 Benin, Burkina Faso, Cameroon, Chad, Mali, Mozambique, Nigeria, Sudan, Tanzania, Togo, Uganda, Zambia, Zimbabwe Soybeans 2 54.2 0.04 Zambia, Zimbabwe Sugar 8 43.7 1.03 Egypt, Kenya, Madagascar, Mozambique, South Africa, Sudan, Tanzania, Uganda Sunflower 3 3.5 0.15 South Africa, Zambia, Zimbabwe Sweet potatoes 4 0.2 0.34 Madagascar, Mozambique, Tanzania, Uganda Tea 3 16.4 0.58 Kenya, Tanzania, Uganda Teff 1 7.1 0.37 Ethiopia Tobacco 4 63.0 0.51 Mozambique, Tanzania, Zambia, Zimbabwe Vanilla 1 12.8 0.06 Madagascar Wheat 8 1.1 4.03 Egypt, Ethiopia, Kenya, South Africa, Sudan, Tanzania, Zambia, Zimbabwe Yams 12 3.3 5.73 Benin, Burkina Faso, Chad, Côte d'Ivoire, Ghana, Mali, Nigeria, Togo All covered products 21 8.9 58.8 Global Distortions to Key Commodity Markets 499 b. Asia Weighted Number of average Gross value Product countries NRA, % of productiona Countries Bananas 1 0.0 0.47 Philippines Barley 1 562.8 0.04 Rep. of Korea Beef 3 85.2 1.00 Rep. of Korea, Philippines, and Taiwan, China Cabbages 1 27.6 0.39 Rep. of Korea Cassava 1 10.0 0.42 Thailand Chickpeas 1 18.7 1.43 India Chilies 1 67.2 0.03 Sri Lanka Cocoa 1 0.0 0.02 Malaysia Coconuts 3 7.9 3.80 Indonesia, Malaysia, Philippines, Sri Lanka Coffee 2 1.7 0.68 Indonesia, Vietnam Cotton 3 5.1 4.79 China, India, Pakistan Eggs 2 51.3 0.64 Rep. of Korea and Taiwan, China Fruits and vegetables 1 8.9 23.10 India Fruits 1 0.0 9.23 China Garlic 1 122.6 0.26 Rep. of Korea Groundnuts 1 12.9 1.79 India Jute 1 38.7 0.18 Bangladesh Maize 6 12.6 16.30 China, India, Indonesia, Pakistan, Philippines, Thailand Milk 4 31.6 22.00 China, India, Rep. or Korea, Pakistan, and Taiwan, China Onions 1 53.4 0.02 Sri Lanka Palm oil 3 2.6 6.66 Indonesia, Malaysia, Thailand Peppers 1 197.0 0.28 Rep. of Korea (Table continues on the following page.) 500 Distortions to Agricultural Incentives: A Global Perspective Table 12.A.1 Summary of NRA Estimates by Major Product, Africa, Asia, and Latin America, 2000­04 (continued) b. Asia Weighted Number of average Gross value Product countries NRA, % of productiona Countries Pig meat 6 4.2 52.10 China, Rep. of Korea, Philippines, Taiwan in China, Vietnam Potatoes 2 6.2 0.44 Brunei Darussalam, Sri Lanka Poultry 7 12.2 17.50 China, Rep. of Korea, Philippines, Taiwan in China, Vietnam Rapeseeds 1 64.8 1.09 India Rice 12 18.5 67.00 Brunei Darussalam, China, India, Indonesia, Rep. of Korea, Malaysia, Pakistan, Philippines, Sri Lanka, Taiwan in China, Thailand, Vietnam Rubber 5 3.9 4.47 Indonesia, Malaysia, Sri Lanka, Thailand, Vietnam Sorghum 1 15.7 0.83 Indonesia Soybeans 5 16.9 5.22 China, India, Indonesia, Rep. of Korea, Thailand Sugar 8 43.1 9.18 China, India, Indonesia, Pakistan, Philippines, Thailand, Vietnam Sunflower 1 14.6 0.26 India Tea 3 7.5 0.56 Brunei Darussalam, Indonesia, Sri Lanka Vegetables 1 0.0 49.90 China Wheat 6 10.7 22.50 Brunei Darussalam, China, India, Rep. of Korea, Pakistan, and Taiwan, China All covered products 12 10.4 324.6 Global Distortions to Key Commodity Markets 501 c. Latin America Weighted Number of average Gross value Product countries NRA, % of productiona Countries Apples 1 0.2 0.15 Chile Bananas 2 24.3 0.69 Dominican Republic, Ecuador Barley 1 6.8 0.18 Mexico Beans 3 3.3 0.88 Dominican Republic, Mexico, Nicaragua Beef 7 1.3 14.30 Argentina, Brazil, Chile, Colombia, Ecuador, Mexico, Nicaragua Cassava 1 0.0 0.02 Dominican Republic Cocoa 1 6.7 0.08 Ecuador Coffee 6 3.3 3.20 Brazil, Colombia, Dominican Republic, Ecuador, Mexico, Nicaragua Cotton 2 10.7 0.86 Brazil, Colombia Eggs 1 15.7 1.84 Mexico Garlic 1 361.9 0.00 Dominican Republic Grapes 1 0.4 0.20 Chile Groundnuts 1 34.5 0.04 Nicaragua Maize 7 3.1 8.07 Argentina, Brazil, Chile, Colombia, Ecuador, Mexico, Nicaragua Milk 6 45.3 4.26 Argentina, Chile, Colombia,Ecuador, Mexico, Nicaragua Onions 1 74.0 0.01 Dominican Republic Palm oil 1 47.4 0.14 Colombia Pig meat 3 4.5 2.93 Brazil, Ecuador, Mexico Poultry 5 18.8 5.78 Brazil, Dominican Republic, Ecuador, Mexico, Nicaragua (Table continues on the following page.) 502 Distortions to Agricultural Incentives: A Global Perspective Table 12.A.1 Summary of NRA Estimates by Major Product, Africa, Asia, and Latin America, 2000­04 (continued) c. Latin America Weighted Number of average Gross value Product countries NRA, % of productiona Countries Rice 6 33.7 1.87 Brazil, Colombia, Dominican Republic, Ecuador, Mexico, Nicaragua Sesame 1 40.5 0.01 Nicaragua Sorghum 3 10.3 0.87 Colombia, Mexico, Nicaragua Soybeans 6 9.9 13.00 Argentina, Brazil, Colombia, Ecuador, Mexico, Nicaragua Sugar 7 26.5 3.71 Brazil, Chile, Colombia, Dominican Republic, Ecuador, Mexico, Nicaragua Sunflower 1 31.9 0.91 Argentina Tomato 2 37.0 1.68 Dominican Republic, Mexico Wheat 5 2.0 2.91 Argentina, Brazil, Chile, Colombia, Mexico All covered products8 2.7 68.6 Source: Drawn from estimates in Anderson and Valenzuela (2008). a. Annual average gross value of covered production at undistorted prices (US$ billions). b. Even though apples, fruit and vegetables, grapes, and oranges are covered only by one country, the weighted and simple averages differ because traded and nontraded products are treated separately. Notes 1. Those regional books cover Africa (Anderson and Masters 2009), Asia (Anderson and Martin 2009), Latin America (Anderson and Valdés (2008), and European transition economies (Anderson and Swinnen 2008). 2. Unfortunately, the European Union does not appear in figure 12.5 because a time series of its cotton NRA was not measured (cotton is a relatively small crop produced only in southern Europe). Nor has the OECD measured a producer support estimate (PSE) for cotton. However, independent estimates for recent years indicate that EU growers receive an even higher NRA than the 70 percent received by U.S. growers in 2000­04 (see Anderson and Valenzuela 2007). 3. This section draws heavily on Lloyd, Croser, and Anderson (2009), who develop the theory summarized here and provide index estimates for a much larger sample of products than reported below. Global Distortions to Key Commodity Markets 503 References Anderson, J. E., and J. P. Neary. 2005. Measuring the Restrictiveness of International Trade Policy. Cambridge, MA: MIT Press. Anderson, K., and J. Croser. 2009. National and Global Agricultural Trade and Welfare Reduction Indexes, 1955 to 2007. Supplementary database at http://www.worldbank.org/agdistortions. Anderson, K., and W. Martin, eds. 2009. Distortions to Agricultural Incentives in Asia. Washington, DC: World Bank. Anderson, K., and W. Masters, eds. 2009. Distortions to Agricultural Incentives in Africa. Washington, DC: World Bank. Anderson, K. and J. Swinnen, eds. 2008. Distortions to Agricultural Incentives in Europe's Transition Economics. Washington, DC: World Bank. Anderson, K., and A.Valdés, eds. 2008. Distortions to Agricultural Incentives in Latin America.Washington, DC: World Bank. Anderson, K. and E. Valenzuela. 2007. "The World Trade Organization's Doha Cotton Initiative: A Tale of Two Issues." The World Economy 30 (8): 1281­1304. ______. 2008. "Estimates of Global Distortions to Agricultural Incentives, 1955 to 2007." World Bank, Washington DC. Core database at http://www.worldbank.org/agdistortions. Baffes, J. 2009. "Benin, Burkina Faso, Chad, Mali, and Togo." In Distortions to Agricultural Incentives in Africa, ed. K. Anderson and W. Masters. Washington, DC: World Bank. FAO. 2008. FAOSTAT data, accessed from http://www.fao.org in October. Feenstra, R. C. 1995. "Estimating the Effects of Trade Policy." In Handbook of International Economics, Volume 3, ed. G. N. Grossman and K. Rogoff. Amsterdam: Elsevier. Hayek, F. A. 1945. "On the Use of Information in Society." American Economic Review 35 (4): 519­30. Johnson, D. G. 1975. "World Agriculture, Commodity Policy, and Price Variability." American Journal of Agricultural Economics 57 (5): 823­28. Lloyd, P. J., J. L. Croser, and K. Anderson. 2009. "How Do Agricultural Policy Restrictions to Global Trade and Welfare Differ Across Commodities?" Policy Research Working Paper 4864, World Bank, Washington, DC. Masters, W. A., and A. F. Garcia. 2009. "Agricultural Price Distortion and Stabilization." Agricultural Distortions Working Paper 86, World Bank, Washington, DC. Pomfret, R. 2008. "Tajikistan, Turkmenistan, and Uzbekistan." In Distortions to Agricultural Incentives in Europe's Transition Economics, ed. K. Anderson and J. Swinnen. Washington, DC: World Bank. Schiff, M., and A. Valdés. 1992. The Political Economy of Agricultural Pricing Policy, Volume 4: A Synthe- sis of the Economics in Developing Countries. Baltimore: Johns Hopkins University Press for the World Bank. Thompson, S. R., P. M. Schmitz, N. Iwai, and B. K. Goodwin. 2004. "The Real Rate of Protection: The Income Insurance Effects of Agricultural Policy." Applied Economics 36: 1­8. Tyers, R., and K. Anderson. 1992. Disarray in World Food Markets: A Quantitative Assessment. Cambridge and New York: Cambridge University Press. Valenzuela, E., D. van der Mensbrugghe, and K. Anderson. 2009. "General Equilibrium Effects of Price Distortions on Global Markets, Farm Incomes, and Welfare." Chapter 13 in this volume. 13 General Equilibrium Effects of Price Distortions on Global Markets, Farm Incomes, and Welfare Ernesto Valenzuela, Dominique van der Mensbrugghe, and Kym Anderson* There has been a great deal of change over the past quarter century in policy dis- tortions to agricultural incentives throughout the world: the antiagricultural and antitrade biases of policies of many developing countries have been reduced, export subsidies of high-income countries have been cut, and some reinstrumen- tation toward less inefficient and less trade-distorting forms of support, particu- larly in Western Europe, has begun. However, applied rates of protection from agricultural import competition have continued to increase in both rich and poor countries, notwithstanding the Uruguay Round Agreement on Agriculture, which aimed to bind and reduce farm tariffs. This chapter analyzes the net economic effects of agricultural price and trade policy changes around the world since the early 1980s, and compares those estimates with projections of how global markets, farm incomes and economic welfare as of 2004 would change if remaining policy distortions were removed. That is, this combined retrospective and prospective analysis seeks to assess how *The authors are grateful for the distortions estimates provided by authors of the focus country case studies, and for assistance with spreadsheets by Johanna Croser, Marianne Kurzweil, and Signe Nelgen. The working paper version of this chapter (Valenzuela, van der Mensbrugghe, and Anderson 2008) contains additional appendix tables. 505 506 Distortions to Agricultural Incentives: A Global Perspective far the world has come, and how far it still has to go, in removing the disarray in world agriculture that was so vividly portrayed in D. Gale Johnson's seminal 1973 study of the issue--and which seemed to have worsened by the time he revised that book in the late 1980s (Johnson 1991). To quantify the impacts of both past reforms and current policies, the distor- tions in the prerelease of version 7 of the GTAP (Global Trade Analysis Project) global protection database are amended here by replacing applied tariffs with dis- tortion rates that reproduce those estimated in the database, using domestic-to- border price comparisons, by authors of the developing-country case studies in the World Bank Agricultural Distortions Project. Likewise, a set of distortions for the period 1980­84 is generated, again aiming to reproduce trend distortion rates in the country case studies.1 Those two sets of distortion estimates suggest that while average distortions to incentives facing developing-country farmers gener- ally were much less in 2004 than in the early 1980s, nonetheless there remains a considerable range of rates across commodities and countries, including a strong antitrade bias in agricultural policies for many countries. Furthermore, nonagri- cultural protectionism is still rife in some developing countries, and agricultural price supports in some high-income countries remain high. Among other things, the present analysis is able to address questions such as: To what extent have government trade and domestic agricultural policies reduced dis- tortionary effects on agricultural markets and farm incomes since the early 1980s? Are policies as of 2004 still reducing farmer rewards in developing countries and thereby prolonging inequality across countries in farm household incomes? Are they depressing value added in primary agriculture more than in other sectors in developing countries, and earnings of unskilled workers more than of owners of other factors of production, thereby potentially contributing to inequality and poverty within those developing countries? With farm incomes well below nonfarm incomes in most developing countries, and with agriculture in those countries intensive in the use of unskilled labor, past or prospective rises in agricultural rela- tive to nonagricultural value added and in wages for the unskilled relative to skilled wages and capital earnings indicate a likely reduction in inequality and poverty. To provide answers to these and related questions, the amended GTAP distor- tion database is used in a global computable general equilibrium (CGE) model-- the Linkage model--to assess how agricultural markets, factor prices, and value added in agriculture versus nonfarm sectors would differ if (a) 1980­84 distortion rates were still in place, and (b) all price and trade policies that distort markets for farm and nonfarm goods as of 2004 were removed. It is important to include nonagricultural trade policies in the reform experiment because, as shown in the seminal study by Krueger, Schiff, and Valdés (1988), they were at least as harmful to developing-country farmers as were those countries' agricultural policies. General Equilibrium Effects of Price Distortions on Global Markets 507 The text below first presents results for the key countries and regions of the world and for the world as a whole, beginning with national economic welfare. Results for numerous individual countries are also presented.2 While no one anticipates a move to completely free markets globally in the near future, compar- ison with the 1980­84 results provides a sense of perspective on what is still to come relative to what has already happened in terms of policy changes over the past quarter century. The prospective analysis also serves as a benchmark to sug- gest what is at stake in terms of further reforms either unilaterally or via World Trade Organization (WTO) rounds of multilateral trade negotiations. At the same time, by showing how different the trade patterns of various countries would be without distortions, it also provides a better indication of agricultural compara- tive advantages in different parts of the world than is available by looking at actual trade and self-sufficiency indicators in the current distortion-ridden situation. The chapter begins with an examination of the extent of price distortions in 2004 and 1980­84 as calibrated in Valenzuela and Anderson (2008), with an emphasis mainly on import tariffs in the case of nonfarm products and, in the case of agricul- ture, on production and export taxes and subsidies. This is followed by a description of the Linkage model of the global economy to be used to analyze the consequences of removing those distortions. The key results of the two sets of simulations are then presented, beginning with the retrospective experiment. The bottom-line results for net farm incomes from both experiments are presented for global liberalization of all merchandise, as well as for reform of just agricultural policies. After comparing these results with earlier ones generated using the version 6 of the GTAP protection database,3 the paper concludes by highlighting the main messages. Key Distortions in Global Markets Border measures traditionally have been the main means by which governments distort prices in their domestic markets for products, with the relative prices of the various tradables affected by trade taxes-cum-subsidies. Product-specific domestic output and farm input subsidies have played a more limited role, in part because of their much greater overt cost to the treasury. In principle, services trade and foreign investment distortions also could distort incentives in the agricultural and industrial sectors, but they are ignored here because much controversy still surrounds their measurement and how they should be modeled.4 To quantify the impacts of both past reforms and current policies, the Alter- tax procedure (Malcolm 1998) is used to amend the distortions in the prerelease of version 7 of the GTAP global protection database.5 The amendments are mainly for developing countries. Following Anderson and Valenzuela (2007b), cotton distortions in the United States are also altered to better reflect policies 508 Distortions to Agricultural Incentives: A Global Perspective there. To simplify the discussion below, European transition economies (in which Turkey is included) are treated as one of the world's developing country regions, along with Africa, Asia, and Latin America.6 The latest prerelease of version 7 of the GTAP database includes estimates of bilateral tariffs and export subsidies and of domestic supports as of 2004 for more than 100 countries and country groups spanning the world. As with version 6 of the GTAP dataset (which relates to 2001), the protection data come from a joint CEPII (Centre d'Etudes Prospectives et d'Informations Internationales)/ ITC (International Trade Centre) project known as MAcMaps. MAcMaps is a detailed database on bilateral import protection at the Harmonized System 6 tar- iff line level that integrates trade preferences, specific and compound tariffs, and a partial evaluation of nontariff barriers such as tariff rate quotas.7 Version 7 of the GTAP database for 2004 has lower tariffs than the previous database for 2001 because of major reforms such as completing the implementation of the Uruguay Round Agreement on Agriculture and unilateral reforms including those result- ing from WTO accession negotiations by China and other recently acceding countries. As mentioned above, in the case of agriculture in developing countries, the distortion levels in that database have been replaced with an alternative set for numerous developing countries, based on nominal rate of assistance (NRA) estimates for 2004 that have come from the present World Bank project (Ander- son and Valenzuela 2008; Valenzuela and Anderson 2008). The sectoral averages of these amended values are shown in table 13.1 for 2004, and also for 1980­84. In the case of amendments to the import tariffs on individual farm products for any particular developing country, the bilateral tariff structure in version 7 of the GTAP database is preserved by simply lowering or raising the bilateral tariffs by the same proportion the average tariff on each product for 2004 is amended.8 According to this amended dataset, the weighted average applied tariff for agri- culture and lightly processed food in 2004 was 21.8 percent for developing coun- tries and 22.3 percent for high-income countries, while for nonfarm goods it was 7.5 percent for developing countries and just 1.2 percent for high-income coun- tries. Though export subsidies for farm products for a few high-income regions and export taxes in a few developing countries were still in place in 2004, these measures are generally small in their impact compared with tariffs, as are produc- tion subsidies and taxes.9 In 1980­84, however, developing countries had an aver- age agricultural export tax of 11 percent, while high-income countries had an average farm export subsidy of 21 percent. The average agricultural import tariff was lower for developing countries (16 percent) in 1980­84 than for high-income countries (26 percent), as opposed to the situation in 2004, when the two groups Table 13.1. Structure of Price Distortions in Global Goods Markets,a 1980­84 and 2004 (percent) 1980­84 2004 Primary Agriculture and Other Primary Agriculture and Other agriculture lightly processed food goods agriculture lightly processed food goods Domestic Export Domestic Export support subsidy Tariff Tariff support subsidy Tariff Tariff Africa 0.3 2.5 17.0 12.6 0.8 0.1 20.4 11.2 Egypt, Arab Rep. of 0.4 6.6 5.8 22.5 0.0 0.0 5.0 13.5 Madagascar 0.0 4.4 3.4 13.9 0.0 4.4 3.4 2.7 Mozambique 0.2 53.8 5.5 29.7 0.2 0.0 14.5 10.9 Nigeria 2.6 22.5 22.2 0.1 0.1 0.0 76.1 17.2 Senegal 0.0 6.7 5.7 8.7 0.0 1.1 6.2 8.9 South Africa 0.0 5.6 18.1 5.8 0.0 0.0 10.2 6.5 Tanzania 5.5 29.4 7.9 97.5 0.3 0.0 11.8 13.7 Uganda 0.9 14.1 16.1 50.1 0.0 2.6 9.2 5.5 Zambia 0.8 21.4 6.0 26.0 0.8 0.0 7.0 9.0 Zimbabwe 4.1 31.0 7.0 49.0 3.2 0.0 8.9 15.4 Rest of Africa 1.2 4.2 19.0 13.4 1.2 0.3 19.0 13.4 East and South Asia 2.4 16.7 19.5 34.6 2.4 0.6 29.6 8.1 China 8.5 38.7 6.9 46.1 0.0 0.2 6.5 7.1 Korea, Rep. of 2.8 0.0 106.7 7.6 0.0 0.0 319.4 5.9 Taiwan, China 0.4 21.7 98.0 5.7 0.4 0.0 84.2 3.9 Indonesia 0.2 3.1 27.9 28.0 0.0 1.6 7.3 4.9 Malaysia 4.8 0.2 3.1 5.2 0.0 0.2 5.0 5.9 Philippines 4.7 0.1 18.3 14.0 4.7 0.0 7.1 3.4 509 Thailand 0.2 0.7 29.8 19.1 0.2 0.0 26.2 12.9 (Table continues on the following pages.) 510 Table 13.1. Structure of Price Distortions in Global Goods Markets,a 1980­84 and 2004 (continued) (percent) 1980­84 2004 Primary Agriculture and Other Primary Agriculture and Other agriculture lightly processed food goods agriculture lightly processed food goods Domestic Export Domestic Export support subsidy Tariff Tariff support subsidy Tariff Tariff Vietnam 3.6 0.5 21.5 18.5 3.6 0.5 21.5 18.5 Bangladesh 1.0 2.2 10.6 26.7 1.0 0.0 9.9 22.5 India 4.9 8.8 8.9 86.2 10.1 2.5 2.9 20.8 Pakistan 0.7 2.7 15.0 53.3 0.0 0.2 19.4 18.5 Sri Lanka 1.1 14.1 22.1 53.1 0.6 0.3 23.8 5.8 Rest of East and South Asia 0.7 0.0 4.3 2.7 0.7 0.0 4.3 2.7 Latin America 3.8 9.6 9.8 15.7 0.2 1.4 7.2 6.7 Argentina 0.0 20.9 0.0 15.8 0.0 14.8 0.0 5.8 Brazil 5.0 17.1 3.2 33.4 0.0 0.0 4.8 8.9 Chile 3.0 0.0 4.8 6.2 0.0 0.0 2.4 1.8 Colombia 0.6 1.0 21.7 22.8 0.0 0.0 21.6 9.8 Ecuador 0.0 13.7 28.6 10.3 0.0 0.0 13.4 10.4 Mexico 14.3 9.6 19.1 6.8 1.2 0.0 6.2 3.4 Nicaragua 0.0 2.8 10.9 3.9 0.0 2.8 9.6 3.9 Rest of Latin America 1.7 0.3 9.9 9.9 1.7 0.3 9.9 9.9 Eastern Europe and Central Asia 0.8 2.6 13.8 9.6 0.8 0.3 15.9 4.8 Baltic states 3.4 0.0 8.2 0.9 3.4 0.0 8.2 0.9 Bulgaria 0.6 0.0 14.8 11.5 0.6 0.0 14.8 11.5 Czech Republic 0.6 0.0 3.0 0.5 0.6 0.0 3.0 0.5 Hungary 3.1 0.0 6.2 0.5 3.1 0.0 6.2 0.5 Poland 0.4 0.0 6.2 0.8 0.4 0.0 6.2 0.8 Romania 1.3 0.0 18.0 9.8 1.3 0.0 18.0 9.8 Slovak Republic 0.0 0.0 5.2 0.4 0.0 0.0 5.2 0.4 Slovenia 0.0 0.0 7.8 0.4 0.0 0.0 7.8 0.4 Russian Federation 1.7 0.9 18.9 7.4 1.7 0.9 18.9 7.4 Kazakhstan 0.9 0.0 3.4 2.7 0.9 0.0 3.4 2.7 Turkey 0.8 14.3 20.4 43.9 0.8 0.0 33.3 3.1 Rest of Eastern Europe and Central Asia 1.1 0.9 9.7 5.7 1.1 0.9 9.7 5.7 High-income countries 6.6 20.9 24.0 2.4 2.6 7.2 22.3 1.2 Australia 0.5 7.0 6.7 8.9 0.0 0.0 0.5 3.3 Canada 3.0 7.0 42.6 5.1 1.6 3.6 18.9 1.4 EU-15 1.2 28.6 13.3 2.0 1.2 12.8 6.9 0.7 Japan 13.1 0.0 120.9 0.9 2.0 0.0 151.7 1.7 New Zealand 5.3 15.4 1.7 18.0 0.0 0.2 0.7 3.3 Rest of Western Europe 101.7 54.0 59.5 4.0 2.6 13.4 53.9 2.2 United States 3.3 14.1 6.5 2.9 5.2 0.6 6.1 1.3 Developing countries 0.6 11.0 16.4 25.6 1.4 0.0 21.8 7.5 Africa 0.3 2.5 17.0 12.6 0.8 0.1 20.4 11.2 East Asia 5.6 21.5 24.3 29.6 0.3 0.0 41.6 6.7 South Asia 3.5 7.1 10.7 72.6 7.2 1.7 6.9 20.2 Latin America 3.8 9.6 9.8 15.7 0.2 1.4 7.2 6.7 Middle East 12.4 0.0 7.5 5.7 12.4 0.0 7.5 5.7 Eastern Europe and Central Asia 0.8 2.6 13.8 9.6 0.8 0.3 15.9 4.8 World total 2.3 4.7 20.1 10.1 1.9 3.5 22.1 3.3 Source: Authors' calculations based on Anderson and Valenzuela (2008). 511 a. Using value of production at undistorted prices as weights. 512 Distortions to Agricultural Incentives: A Global Perspective of countries had equivalent average tariffs of 22 percent. In addition, tariffs on nonagricultural imports were more than three times higher in 1980­84 than in 2004 for developing countries (table 13.1). The averages obscure large variations across countries and commodities. Of particular note are the agricultural tariffs. The rise in the average of those tariffs for developing countries since 1980­84 is representative for Africa and Europe's transition economies, but it fell in Latin America and in much of Asia--although the rise in the Republic of Korea was so dramatic as to cause the Asian average to increase by one-third (table 13.1). On their own, the averages are not necessarily good indicators of overall dis- tortions to farmers' incentives, although it certainly helps to see relative rates of assistance (RRAs) to agriculture versus nonagricultural goods (which is why RRAs are emphasized in the preceding chapters in this volume). Also of impor- tance is the composition of each country's trade. Three examples serve to illus- trate the point. First, if high-income countries' tariffs on temperate farm products are at a near-prohibitive level but are zero on tropical products such as coffee beans, those countries' import-weighted average agricultural tariff could be quite low even though agricultural value added in those rich countries had been enhanced substantially. A second illustration is the case of a developing country with a strong agricultural comparative advantage in all but one small farming industry, and with high tariffs to stave off import competition for that industry and for all manufacturing industries. Overall, agricultural value added would be depressed by that structure of protection, yet the import-weighted average tariff protection for agriculture would be high and possibly above that for manufac- tures. A third example is where the nonagricultural primary sector receives a sim- ilar level of import protection as the farm sector and less than the manufacturing sector but is much more export-focused than agriculture. In the latter case, trade reform may cause that other primary sector to expand at the expense not only of manufacturing but also of farming. Despite the use of production rather than trade weights to obtain sectoral averages rates of distortion in table 13.1, and even though the ratio of agricultural to other goods' tariffs for 2004 in that table is well above unity for many of the regions shown, it is not possible to say from those distortion rates alone whether developing-country policies have overshot in terms of moving away from an antiagricultural bias. Equally, it is not possible to know how the benefits of removal of agricultural tariffs in the protective coun- tries would be distributed among the various agricultural-exporting countries. What is needed to address such issues is a global general equilibrium model to estimate the net effects of all sectors' distortions in all countries on the various nations' agricultural markets and net farm incomes, to which the analysis below now turns. General Equilibrium Effects of Price Distortions on Global Markets 513 The Linkage Model of the Global Economy The model used for this analysis is the World Bank's global CGE model, known as Linkage (van der Mensbrugghe 2005). For most of this decade, it has formed the basis for the World Bank's standard long-term projections of the world economy and for much of its trade (and more recently migration) policy analysis (for example, World Bank 2002, 2004, 2005, 2006). It is a relatively straightforward CGE model but with some characteristics that distinguish it from other compara- tive static models such as the GTAP model (described in Hertel 1997). Factor stocks are fixed in Linkage, which means in the case of labor that the extent of unemployment (if any) in the baseline remains unchanged. Producers minimize costs subject to constant returns to scale in production technology, consumers maximize utility, and all markets--including for labor--are cleared with flexible prices. There are three types of production structures. Crop sectors reflect the substitution possibilities between extensive and intensive farming; livestock sec- tors reflect the substitution possibilities between pasture and intensive feeding; and all other sectors reflect standard capital and labor substitution. There are two types of labor, skilled and unskilled, and the total employment of each is assumed to be fixed (meaning no change in their unemployment levels). There is a single representative household per modeled region, allocating income to consumption using the extended linear expenditure system. Trade is modeled using a nested Armington structure in which aggregate import demand is the outcome of allo- cating domestic absorption between domestic goods and aggregate imports, and then aggregate import demand is allocated across source countries to determine the bilateral trade flows.10 Government fiscal balances are fixed in U.S. dollar terms in Linkage, with the fiscal objective being met by changing the level of lump sum taxes on households. This implies that losses of tariff revenues are replaced by higher direct taxes on households. The current account balance also is fixed. Given that other external financial flows are fixed, this implies that ex ante changes to the trade balance are reflected in ex post changes to the real exchange rate. For example, if import tariffs are reduced, the propensity to import increases and additional imports are financed by increasing export revenues. The latter typically is achieved by a depre- ciation of the real exchange rate. Finally, investment is driven by savings. With fixed public and foreign saving, investment comes from changes in the savings behavior of households and from changes in the unit cost of investment. The model solves only for relative prices, with the numeraire, or price anchor, being the export price index of manufactured exports from high-income countries. This price is fixed at unity in the base year. A virtue of beginning with the latest GTAP database (prerelease 5 of version 7) is that it includes bilateral tariffs that capture not only reciprocal but also 514 Distortions to Agricultural Incentives: A Global Perspective nonreciprocal preferential trade agreements, the latter of which provide low- income exporters duty-free access to protected high-income country markets. This allows the fact that future reform may cause a decline in the international terms of trade for developing countries that enjoy preferential access to agricul- tural and other markets of high-income countries (in addition to those that are net food importers because their comparative advantage is in other sectors, such as labor-intensive manufacturing. The version of the Linkage model used for this study is based on an aggrega- tion involving 24 sectors and 52 regions spanning the world (see table 13.A.1 in the annex). There is an emphasis on agriculture and food, which together com- prise half of those 24 sectors. Consistent with the rest of the present project, and with WTO practice, Korea and Taiwan, China are included in the "developing country" category.11 Because the results below are comparative static results, they do not include the (often much larger) dynamic gains that result from an acceleration in investment due to the reduction in tariffs on industrial goods lowering the cost of investment. Nor do they account for the structural changes that lead to different comparative advantages over time (which could have a notable impact on the terms-of-trade effect, but it is not possible to say in what direction). And because this version of the Linkage model assumes perfect competition and constant returns to scale, it captures none of the benefits of freeing markets that came from accelerated productivity growth, scale economies, and the creation of new markets (extensifi- cation versus intensification). There is also a dampening effect on estimates of welfare gains because of product and regional aggregation, which hides many of the differences across products in NRAs and consumer tax equivalents. The results therefore should be treated as providing much lower-bound estimates of the net economic welfare benefits from policy reform.12 Retrospective Analysis: How Different Would 2004 Have Been if the Agricultural and Merchandise Trade Policy Changes Globally since 1980­84 Had Not Happened? To simultaneously examine the effects of high-income country liberalization of agricultural policies and the reduction in the antiagricultural and antifarm trade policy biases in developing countries with and without liberalization of nonagri- cultural trade policies (to sense the relative contribution of the latter to the overall result for net farm incomes), a set of scenarios is created below in which 1980­84 distortion rates replace those for 2004. Global and national economic welfare effects are examined first, followed by changes in the terms of trade, adjustments General Equilibrium Effects of Price Distortions on Global Markets 515 to quantities produced and traded, effects on factor rewards, and percentage changes to agricultural value added (net farm income) relative to value added in the rest of the economy. Global and national economic welfare The Linkage model and the distortions database presented here provide a baseline projection of the world economy in 2004. This is first compared with a simulation in which all agricultural domestic and border subsidies and taxes plus import tar- iffs on other merchandise are replaced with the distortion structures of 1980­84, as summarized in table 13.1. The model presented here suggests that the reforms over the intervening two decades have improved global welfare by US$233 billion per year. (Keep in mind that a negative sign in this experiment has the same mean- ing as a positive sign in the next experiment, since both sets of simulations use 2004 policies as their baseline.) The distribution across regions of that change in economic welfare (or equivalent variation in income), reported in table 13.2, sug- gests two-thirds of those dollars accrued to high-income countries. However, as a share of national income, developing countries gained more, with an average increase of 1.0 percent compared with 0.7 percent for high-income countries. The results vary across developing countries, ranging from slight losses in a few cases to large proportional increases in cases such as China, Mozambique, and Nigeria. The second column of numbers and those in parentheses in table 13.2 show the amount of welfare gain due to changes in the international terms of trade for each country. For developing countries as a group, the terms of trade effect is adverse, while the opposite is the case for high-income countries. Nonetheless, even though that terms of trade change reduced their gains from improved effi- ciency of domestic resource use, developing country economies have benefited proportionately more than high-income economies from the policy reforms of the past quarter century. Decomposing the contribution to welfare of changes in national terms of trade To understand the contribution to welfare of the changes in international terms of trade shown in table 13.2, it is necessary to first examine the changes in import and export prices for farm and other products. For developing countries as a group, their terms of trade have worsened because of these reforms for two sets of reasons: for nonagricultural goods, export prices have been lowered by 0.4 per- cent while import prices have hardly been affected; and for farm products, reduced export prices (0.6 percent) have been compounded by 16 percent higher prices for their agricultural and food imports. The net effect is a deterioration of 516 Distortions to Agricultural Incentives: A Global Perspective Table 13.2. Economic Welfare Impact of Going Back to 1980­84 Policies, by Country/Region Change in Total real income due Total real income gain just to change income gain per year in terms of trade relative to 2004 (US$ billions) (US$ billions) benchmark (%)a North and Sub- Saharan Africa 2.8 1.8 0.5 (0.3) Egypt, Arab Rep. of 0.1 0.1 0.2 (0.1) Madagascar 0.0 0.1 1.1 (2.0) Mozambique 0.2 0.4 3.7 (7.2) Nigeria 0.2 0.9 0.5 ( 2.0) Senegal 0.0 0.0 1.2 ( 1.2) South Africa 2.6 1.4 1.8 ( 0.9) Tanzania 0.1 0.8 1.1 (8.6) Uganda 0.1 0.4 2.3 (6.3) Zambia 0.0 0.1 0.8 (2.8) Zimbabwe 0.2 0.3 8.7 (12.3) Rest of Africa 0.7 1.9 0.3 (0.7) East and South Asia 72.3 26.4 2.1 (0.8) China 46.4 29.1 3.5 (2.2) Korea, Rep. of 6.9 6.9 1.4 ( 1.4) Taiwan, China 2.9 3.1 1.2 ( 1.2) Indonesia 3.5 1.3 1.6 (0.6) Malaysia 0.5 0.5 0.6 ( 0.6) Philippines 0.1 1.4 0.2 (2.1) Thailand 1.5 0.5 0.6 ( 0.2) Vietnam 0.3 0.2 0.8 ( 0.5) Bangladesh 0.0 0.0 0.1 (0.0) India 8.7 5.6 1.7 (1.1) Pakistan 1.2 0.6 1.6 (0.7) Sri Lanka 0.9 0.4 5.8 (2.2) Rest of East and South Asia 0.7 0.9 0.5 ( 0.6) Latin America 7.1 13.7 0.4 (0.8) Argentina 1.7 0.1 1.4 (0.1) Brazil 5.3 6.8 1.2 (1.6) Chile 0.1 0.7 0.1 (1.0) Colombia 2.5 2.5 3.5 (3.5) Ecuador 0.6 0.3 2.5 (1.2) Mexico 2.6 3.6 0.5 (0.7) Nicaragua 0.0 0.0 0.6 (0.0) Rest of Latin America 0.5 0.2 0.1 ( 0.1) General Equilibrium Effects of Price Distortions on Global Markets 517 Table 13.2. Economic Welfare Impact of Going Back to 1980­84 Policies, by Country/Region (continued) Change in Total real income due Total real income gain just to change income gain per year in terms of trade relative to 2004 (US$ billions) (US$ billions) benchmark (%)a Eastern Europe and Central Asia 6.5 7.1 0.5 (0.6) Baltic states 0.3 0.2 1.1 (0.7) Bulgaria 0.4 0.3 3.4 (2.6) Czech Republic 0.4 0.3 0.6 (0.4) Hungary 0.3 0.3 0.4 (0.4) Poland 0.4 0.3 0.3 (0.2) Romania 0.8 0.5 2.2 (1.4) Slovak Republic 0.1 0.1 0.5 (0.5) Slovenia 0.0 0.0 0.1 (0.0) Russian Federation 3.7 1.5 0.8 (0.3) Kazakhstan 0.0 0.0 0.0 (0.0) Turkey 4.4 1.6 2.0 (0.7) Rest of Eastern Europe and Central Asia 4.4 2.1 4.2 (2.0) High-income countries 159.9 50.8 0.7 ( 0.2) Australia 2.4 0.2 0.5 (0.0) Canada 4.6 1.6 0.7 (0.2) EU-15 63.0 10.3 0.8 ( 0.1) Japan 14.6 20.3 0.5 ( 0.6) New Zealand 2.5 0.6 3.6 ( 0.8) Rest of Western Europe 59.7 15.9 12.1 ( 3.2) United States 10.8 2.6 0.1 (0.0) Hong Kong, China and Singapore 2.2 2.6 1.8 ( 2.1) Developing countries 73.1 49.3 1.0 (0.7) North Africa 0.6 0.1 0.3 (0.0) Sub-Saharan Africa 3.4 1.7 1.0 (0.5) East Asia 61.5 19.9 2.2 (0.7) South Asia 10.8 6.5 1.7 (1.0) Latin America 7.1 13.7 0.4 (0.8) Middle East 2.6 0.4 0.5 (0.1) Eastern Europe and Central Asia 6.5 7.1 0.5 (0.6) High-income countries 159.9 50.8 0.7 ( 0.2) World total 233.0 1.5 0.8 (0.0) Source: Authors' World Bank Linkage model simulations. a. Numbers in parentheses refer to that due to terms of trade effects. 518 Distortions to Agricultural Incentives: A Global Perspective 1.7 percent in their terms of trade. By contrast, high-income countries enjoyed an improvement of 0.8 percent in their terms of trade as a result of the policy changes, partly from nonfarm products but mostly from farm products, where the improvement in their export prices more than offset the higher prices of their imports (table 13.3). The contributions of those four elements to national economic welfare can be seen in table 13.4. Overall, the terms of trade effect for developing countries diminishes the welfare gains from reform by US$49 billion, bringing it down to US$73 billion per year. Of that, two-fifths is from the decline in agricultural export prices (in part because of less taxation and hence larger volumes of those exports now), another two-fifths from the decline in prices of nonfarm exports, and one-fifth from the rise in prices of their food imports (partly because of reduced assistance to farmers in high-income countries). For high-income coun- tries, however, the reduction in agricultural tariffs and subsidies and the conse- quent rise in international food prices helps farm exporters more than it hurts import-competing farmers in this group, and the improvement in welfare from lower prices of nonfarm imports more than offsets the loss due to lower prices for their nonfarm exports. Quantities produced and traded The retrospective results suggest that, as a result of the reforms of the past two decades, developing countries' aggregate shares of global output and exports of textiles and apparel have grown by about 3 percentage points while the shares for other nonfarm products have changed by no more than one percentage point. Their shares in agricultural and food markets, however, have changed more: their share of the world's primary agricultural exports has risen from 43 to 55 percent and the output share from 58 to 62 percent, and even their shares of processed foods have risen by one percentage point. The increases have occurred in nearly all agricultural industries--the exceptions being rice and sugar, where the growth in protectionism in high-income countries has been greatest. The share of global production of farm products that is exported (excluding intra-European Union [EU] trade) is slightly smaller as a result of the reforms, in contrast to the 5 percentage-point rise for textiles and clothing and the 3-point rise for other manufactures. Agriculture's 8 percent share in 2004 remains in stark contrast to the 31 percent share for other primary products and around 25 per- cent for all other goods, and this "thinness" is an important contributor to the volatility of international prices for these weather-dependent farm products (first columns of table 13.5). The fact that the past two decades of reform have not made agricultural products more traded globally is illuminating. The findings Table 13.3. Impact of Going Back to 1980­84 Policies on Indexes of Reala Export and Import Prices, by Region (percent) Percent change in export prices Percent change in import prices Percent change in terms of trade Agriculture Agriculture Agriculture and light Manufacturing and light Manufacturing and light Manufacturing processing and services Total processing and services Total processing and services Total Developing countries 0.6 0.4 0.4 15.7 0.0 1.3 16.3 0.4 1.7 North Africa 1.6 0.5 0.6 9.2 0.5 0.9 10.7 0.1 1.5 Sub-Saharan Africa 12.8 0.5 2.1 18.7 0.3 1.8 5.9 0.8 0.3 East Asia 1.8 0.2 0.2 7.1 0.6 1.1 5.4 0.8 1.3 South Asia 6.6 5.5 5.6 6.8 0.1 0.8 13.4 5.5 6.4 Latin America 5.4 0.5 1.1 12.1 0.4 0.5 17.5 0.2 1.6 Middle East 2.2 0.6 0.7 25.6 0.3 2.6 23.4 0.9 1.9 Eastern Europe and Central Asia 3.9 0.4 0.6 28.6 0.5 1.8 32.5 0.1 2.4 High-income countries 24.0 0.2 1.7 17.9 0.4 0.8 6.2 0.2 0.8 World total 17.6 0.3 1.0 17.0 0.3 1.0 0.6 0.0 0.0 Source: Authors' World Bank Linkage model simulations. a. Relative to the numeraire which in this version of the Linkage model is the price of high-income countries' exports of manufactures. 519 520 Table 13.4. Terms-of-Trade Contribution to Real Income Changes from Going Back to 1980­84 Policies, by Region (2004 US$ billions) Exports Importsa Total impact Agriculture Agriculture Net terms and light Manufacturing and light Manufacturing of trade Net real processing and services processing and services impact income gain Developing countries 20,567 20,323 11,907 3,458 49,340 73,150 North Africa 157 390 10 452 85 598 Sub-Saharan Africa 2,359 504 560 696 1,719 3,399 East Asia and Pacific 5,395 7,948 877 5,684 19,905 61,550 South Asia 1,441 6,218 818 365 6,476 10,761 Latin America 9,562 5,296 1,681 2,867 13,671 7,125 Middle East 256 1,560 3,545 1,352 376 2,584 Eastern Europe and Central Asia 1,909 2,536 6,072 3,409 7,108 6,504 High-income countries 65,099 12,403 32,558 30,710 50,847 159,880 World total 44,532 32,726 44,465 34,167 1,508 233,030 Source: Authors' World Bank Linkage model simulations. Table 13.5. Impact of Going Back to 1980­84 Policies on Shares of Global Output Exported, and Developing- Country Shares of Global Output and Exports,a by Product (percent) Share of global output Developing countries' Developing countries' exporteda share of global output share of global exportsa 2004 benchmark 1980­84 2004 benchmark 1980­84 2004 benchmark 1980­84 Paddy rice 1 2 81 85 56 21 Wheat 16 19 67 56 25 10 Other grains 11 14 55 45 35 17 Oilseeds 21 23 69 60 54 34 Plant-based fibers 25 44 74 72 50 72 Vegetables and fruits 9 8 72 69 69 56 Other crops 14 12 49 45 75 62 Cattle, sheep, goats 2 2 43 41 56 53 Other livestock 4 6 65 57 43 41 Wool 13 14 82 80 16 14 Beef and sheep meat 7 6 27 26 31 24 Other meat products 7 10 32 21 42 2 Vegetable oils and fats 20 19 52 49 80 73 Dairy products 5 8 29 31 28 54 Processed rice 5 6 76 79 85 60 Refined sugar 8 22 52 69 78 95 Other food, beverages, and tobacco 9 7 35 35 50 54 Other primary products 31 30 64 65 76 78 Textile and wearing apparel 28 23 53 50 74 71 Other manufacturing 24 21 32 33 43 42 Services 3 3 20 20 31 31 Agriculture and food 8 9 46 44 54 48 Primary agriculture 8 9 62 58 55 43 Processed foods 8 9 37 36 52 51 521 Source: Authors' World Bank Linkage model simulations. a. Excluding intra-EU trade. 522 Distortions to Agricultural Incentives: A Global Perspective summarized elsewhere in this volume show that the reforms in developing and high-income countries over the past two decades have reduced the antitrade bias in the agricultural trade of developing countries but increased that bias in high- income countries (thanks in part to the cut in their export subsidies). According to this Linkage model result, the latter slightly more than offsets the former in terms of their aggregate impact on the global share of farm production that is traded. The impacts on agricultural and food output and trade for various countries and regions suggest farm trade would have been two-thirds larger in real value terms had the past two decades of reform not occurred (last row of table 13.6). On the export side, that is almost entirely due to high-income countries, whose exports would have been more than twice as large had they not lowered their export subsidies and developing countries not lowered their export taxes. The global value of agricultural and food output, however, is virtually unchanged (just 3.6 percent less). This suggests that, in aggregate, the reform-induced output decline of high-income countries (11 percent) more than fully offset the reform- induced output expansion of developing countries (3 percent). Note that the big economies of China and South Asia, as well as Thailand and most of Latin Amer- ica, all enjoyed increased farm output because of the past quarter century's reforms. Note also what happens to agricultural imports: in real value terms, developing countries as a group would have had to import 50 percent more farm products in 2004 had the reforms of the past two decades not taken place, while high-income countries would have had to import nearly 80 percent more (last column of table 13.6). Combined with the export effects, that means the food and agricultural self-sufficiency ratio would have been very slightly lower in develop- ing countries and slightly higher in high-income countries (table 13.7). The extent of this reform on the tradability of different products is shown in table 13.8. Sugar, milk products, and cotton would have been exported more from developing to high-income countries had the latter group's assistance to those industries not grown over the past two decades. The net consequences of these impacts on the share of farm production exported by regions are shown in table 13.9. For developing countries as a group, there is no change for its 9.5 percent share, while for high-income countries the share is reduced by 3 percentage points by their export subsidy cuts and other reforms, to 13 percent (including intra-EU trade). Effects on product prices How do different agricultural and manufacturing goods' average prices in inter- national markets change with liberalization of distortionary agricultural policies Table 13.6. Impact of Going Back to 1980­84 Policies on Agricultural and Food Output and Trade, by Country/Region 2004 US$ billions Change relative to 2004 baseline (%) Output Exports Imports Output Exports Imports North and Sub-Saharan Africa 5.0 3.9 8.0 2.6 18.9 37.0 Egypt, Arab Rep. of 0.4 0.2 0.2 2.6 13.9 6.8 Madagascar 0.0 0.0 0.0 0.7 12.1 22.6 Mozambique 1.2 1.1 0.2 75.9 676.8 80.5 Nigeria 0.2 0.5 0.5 1.1 115.5 33.3 Senegal 0.2 0.2 0.1 9.6 125.0 9.8 South Africa 8.4 6.4 0.6 30.8 192.1 32.3 Tanzania 0.3 0.2 0.1 5.0 42.6 42.8 Uganda 0.0 0.1 0.0 1.0 21.1 14.9 Zambia 0.3 0.3 0.1 18.6 81.7 105.8 Zimbabwe 0.8 0.7 0.1 49.0 77.5 40.9 Rest of Africa 2.8 2.8 6.0 2.5 22.1 46.8 East and South Asia 42.4 3.9 29.9 4.6 8.2 44.4 China 44.4 5.8 23.6 12.6 57.8 96.3 Korea, Rep. of 6.3 0.6 1.1 10.4 112.5 13.1 Taiwan, China 1.8 2.1 0.5 8.7 440.3 12.8 Indonesia 0.5 0.6 0.4 0.8 7.8 9.1 Malaysia 1.8 1.7 1.1 9.8 22.0 28.3 Philippines 0.3 0.0 0.4 0.9 0.1 16.2 Thailand 1.1 0.3 0.7 2.0 4.5 27.4 523 (Table continues on the following pages.) 524 Table 13.6. Impact of Going Back to 1980­84 Policies on Agricultural and Food Output and Trade, by Country/Region (continued) 2004 US$ billions Change relative to 2004 baseline (%) Output Exports Imports Output Exports Imports Vietnam 1.3 1.3 0.3 8.7 60.6 24.6 Bangladesh 0.4 0.1 0.1 1.5 39.4 4.8 India 7.3 2.6 1.3 3.3 38.2 22.7 Pakistan 1.3 0.6 0.2 3.0 49.7 9.5 Sri Lanka 0.1 0.5 0.0 1.9 62.4 1.1 Rest of East and South Asia 0.5 0.3 1.3 2.4 13.4 25.8 Latin America and the Caribbean 22.5 13.8 6.5 6.9 20.6 26.8 Argentina 6.4 5.8 0.1 19.9 36.7 27.8 Brazil 18.4 12.4 0.7 18.2 48.5 30.7 Chile 1.1 0.2 0.1 11.0 7.8 12.7 Colombia 10.3 9.0 1.4 48.6 292.6 110.4 Ecuador 1.4 1.6 0.1 15.6 69.6 12.7 Mexico 1.5 2.9 1.2 2.3 54.0 12.6 Nicaragua 0.0 0.1 0.0 2.8 26.1 16.8 Rest of Latin America 3.9 0.0 2.9 4.6 0.2 32.2 Eastern Europe and Central Asia 10.3 11.6 23.9 2.6 53.4 91.6 Baltic states 0.8 0.2 1.0 11.3 18.6 75.3 Bulgaria 4.3 1.9 0.4 6.8 266.2 71.2 Czech Republic 1.1 0.0 1.0 6.3 2.5 57.0 Hungary 1.5 0.0 1.1 10.9 0.3 88.8 Poland 3.3 0.1 1.8 7.3 3.9 64.0 Romania 0.1 0.7 1.4 0.5 101.6 101.6 Slovak Republic 0.4 0.0 0.3 5.1 0.3 53.6 Slovenia 0.1 0.1 0.2 2.4 31.8 40.2 Russian Federation 7.8 0.5 11.2 8.0 28.5 126.3 Kazakhstan 0.6 0.7 0.1 5.0 75.7 33.6 Turkey 2.6 1.0 2.0 4.2 25.2 62.8 Rest of Eastern Europe and Central Asia 8.7 3.5 6.7 148.9 94.9 High-income countriesa 195.8 256.1 183.7 11.0 110.8 78.3 Australia 1.2 0.0 1.3 2.0 0.2 80.9 Canada 6.5 9.1 3.3 9.8 61.6 40.0 EU-15a 123.1 165.5 123.9 13.7 124.6 77.7 Japan 6.2 1.0 8.0 3.6 230.5 32.9 New Zealand 3.4 2.0 0.4 14.9 23.8 64.2 Rest of Western Europe 74.7 69.1 30.1 125.3 1849.8 409.0 United States 4.4 9.4 15.9 0.9 17.5 55.9 Hong Kong, China and Singapore 0.1 0.0 0.8 2.5 38.9 17.9 Developing countries 62.8 8.1 80.5 3.2 4.9 50.3 North Africa 0.4 1.2 2.1 0.7 35.2 21.4 Sub-Saharan Africa 5.5 2.7 5.9 4.3 15.5 50.0 East Asia 34.0 0.1 28.4 5.4 0.2 51.2 South Asia 8.4 3.8 1.4 2.8 41.2 12.3 Latin America 22.5 13.8 6.5 6.9 20.6 26.8 Middle East 7.3 10.3 12.2 7.1 154.2 58.6 Eastern Europe and Central Asia 10.3 11.6 23.9 2.6 53.4 91.6 High-income countries 195.8 256.1 183.7 11.0 110.8 78.3 World totala 133.0 264.2 264.2 3.6 66.9 66.9 Source: Authors' World Bank Linkage model simulations. 525 a. Excluding intra-EU trade. 526 Table 13.7. Impact of Going Back to 1980­84 Policies on Self-Sufficiencya in Agricultural and Other Products, by Product and Region (percent) Eastern High-income Developing Latin Europe and countries countries Africa America East Asia South Asia Central Asia 2004 1980­ 2004 1980­ 2004 1980­ 2004 1980­ 2004 1980­ 2004 1980­ 2004 1980­ bench- 84 bench- 84 bench- 84 bench- 84 bench- 84 bench- 84 bench- 84 mark policies mark policies mark policies mark policies mark policies mark policies mark policies Paddy rice 101 108 100 99 97 96 93 79 100 100 101 100 95 94 Wheat 141 190 88 79 67 61 80 59 68 43 100 98 102 96 Other grains 108 124 94 88 94 97 98 100 88 76 103 103 103 95 Oilseeds 104 124 97 88 104 102 140 110 66 64 100 99 106 103 Plant-based fibers 161 168 88 83 177 141 94 78 54 11 93 91 104 243 Vegetables and fruits 90 94 105 103 108 107 153 134 102 100 99 100 99 98 Other crops 90 94 113 107 138 115 143 122 110 115 104 103 90 91 Cattle, sheep, goats 100 100 100 100 101 101 102 101 98 97 100 100 102 103 Other livestock 101 98 100 102 101 103 101 104 99 99 100 100 99 106 Wool 161 167 92 91 103 103 103 102 78 74 96 95 96 96 Beef and sheep meat 101 106 97 95 96 94 108 104 83 86 126 110 95 89 Other meat products 100 149 100 82 92 74 121 83 101 92 96 91 96 74 Vegetable oils and fats 95 100 103 100 69 86 141 115 115 116 78 73 93 92 Dairy products 103 101 94 102 76 80 97 100 78 77 99 99 102 109 Processed rice 99 104 100 99 69 72 94 89 104 104 104 100 92 94 Refined sugar 98 69 102 130 95 173 131 239 98 103 96 92 98 96 Other food, beverages, and tobacco 99 99 103 103 101 100 108 107 105 107 106 104 100 100 Other primary products 76 74 122 123 180 181 148 152 84 85 75 88 115 116 Textile and wearing apparel 81 86 123 119 98 104 104 106 144 134 144 129 101 105 Other manufacturing 101 100 98 99 77 77 96 97 106 107 90 95 95 94 Services 101 101 101 100 101 101 100 100 101 101 100 99 101 101 Agriculture and food 100 104 101 100 100 100 112 109 100 97 100 99 99 99 Agriculture 99 104 100 98 104 101 115 106 96 92 100 99 100 103 Processed foods 100 104 101 101 94 98 110 110 104 104 100 97 99 96 Source: Authors' World Bank Linkage model simulations. a. Self-sufficiency is defined as domestic production as a percentage of domestic consumption measured in value terms at free on board (fob) prices. 527 Table 13.8. Impact of Going Back to 1980­84 Policies on Shares of Production Exported and of Consumption 528 Imported by the World, High-Income and Developing Countries (percent) Share of production exported Share of consumption imported a High-income countries Developing countries High-income countriesa Developing countries 2004 2004 2004 2004 benchmark 1980­84 benchmark 1980­84 benchmark 1980­84 benchmark 1980­84 Paddy rice 3 10 1 0 2 4 1 2 Wheat 37 40 6 3 11 10 17 22 Other grains 15 21 7 5 9 13 11 16 Oilseeds 31 37 16 13 26 22 16 21 Plant-based fibers 50 45 17 44 18 18 26 59 Vegetables and fruits 10 12 9 7 18 16 4 4 Other crops 7 8 21 16 16 13 11 10 Cattle, sheep, goats 1 1 2 2 2 2 2 2 Other livestock 6 8 3 4 6 9 3 3 Wool 60 59 2 2 35 36 10 11 Beef and sheep meat 6 6 7 5 5 4 10 11 Other meat products 6 13 9 1 6 10 8 18 Vegetable oils and fats 8 10 31 29 12 11 26 27 Dairy products 5 6 4 14 2 7 10 12 Processed rice 3 11 5 5 4 9 5 5 Refined sugar 4 4 12 30 5 33 10 13 Other food, beverages, and tobacco 7 5 12 11 8 6 9 8 Other primary products 20 18 37 36 38 39 22 20 Textile and wearing apparel 15 13 39 33 30 25 23 19 Other manufacturing 20 18 32 27 19 18 32 27 Services 3 3 5 5 2 2 5 5 Agriculture and food 7 8 9 10 8 8 8 10 Agriculture 9 12 7 7 10 10 7 8 Processed foods 6 7 12 12 7 7 10 11 Source: Authors' Linkage model simulations. a. Excluding intra-EU trade. General Equilibrium Effects of Price Distortions on Global Markets 529 Table 13.9. Impact of Going Back to 1980­84 Policies on Shares of Agricultural and Food Production Exported, by Country/Region (percent) 2004 benchmark 1980­84 Developing countries 9.5 9.5 North Africa 6.3 7.9 Sub-Saharan Africa 13.8 13.5 East Asia 8.4 7.7 South Asia 3.7 2.4 Latin America 18.1 16.3 Middle East 7.4 14.2 Eastern Europe and Central Asia 6.8 9.1 High-income countries 13.0 15.9 World totala 11.4 13.1 World total (excluding intra-EU trade) 8.1 8.7 Source: Authors' World Bank Linkage model simulations. a. Including intra-EU trade. and other protection? The answer depends not only on changes in the NRAs but also on the relative size of each subsector and the different degrees of responsive- ness of inputs to changes in relative output prices, and on the method used to weigh different countries' price changes. According to the Linkage model, with its default elasticities and (Paasche) weighting methods, the average real price in international markets would have been 13 percent lower for agricultural and food products had policies not changed over the past two decades (table 13.10). Inter- national prices for farm goods are higher now despite the substantial reduction in the antiagricultural policy bias in developing countries since the early 1980s: the effect of that is evidently more than offset by the reduction in agricultural tariffs and subsidies in high-income countries. Effects on factor rewards The relatively small percentage changes in net national economic welfare, reported in table 13.2, hide the fact that redistribution of welfare among groups within each country following trade reform can be much larger. This is clear from the impacts on real rewards to labor, capital, and land that are reported in table 13.11, where factor rewards are expressed in real terms by deflating by the aggregate consumer price index (CPI). Those results suggest that reform has 530 Distortions to Agricultural Incentives: A Global Perspective Table 13.10. Impact of Going Back to 1980­84 Policies on Real International Product Prices (percentage, change relative to 2004 baseline) Paddy rice 11.6 Wheat 15.4 Other grains 27.5 Oilseeds 8.6 Sugar cane and beet 0.5 Plant-based fibers 0.8 Vegetables and fruits 2.8 Other crops 2.6 Cattle, sheep, goats 0.5 Other livestock 2.0 Raw milk 0.4 Wool 1.9 Beef and sheep meat 15.0 Other meat products 45.5 Vegetable oils and fats 1.4 Dairy products 8.5 Processed rice 0.6 Refined sugar 2.5 Other food, beverages, and tobacco 0.1 Textile and wearing apparel 1.4 Other manufacturing 0.3 Merchandise trade 1.2 Agriculture and food 12.6 Primary agriculture 5.9 Agriculture and lightly processed food 17.6 Source: Authors' World Bank Linkage model simulations. Note: Model numeraire is the export price index of high-income countries' manufactured exports. raised the food price index by half a point while lowering by one-fifth of a point the overall CPI index. Unskilled workers in developing countries, according to these results, are better off from reform than skilled workers or capital owners; if they are also agricultural landowners, they have gained from increased rewards for that fac- tor too. For high-income countries, consistent with standard trade theory, skilled workers gained at the expense of unskilled workers and agricultural land rents have halved over what they would have been. Those European and Northeast Asian farmers renting agricultural land would have benefited from the large fall in farm rental costs, more or less offsetting the fall in prices for their output, while earnings of landowners in those countries would decline. Their loss is relative to General Equilibrium Effects of Price Distortions on Global Markets 531 Table 13.11. Impact of Going Back to 1980­84 Policies on Real Factor Prices,a by Region (percent, relative to 2004 baseline) Capitalb Landb Unskilled Skilled user user Aggregate Food wages wages cost cost CPI CPI Developing countries 2.1 1.7 1.5 4.1 1.0 0.4 North Africa 0.3 0.1 0.2 1.1 0.3 0.7 Sub-Saharan Africa 0.1 0.6 1.2 1.5 1.4 3.1 East Asia 4.5 3.7 3.4 6.2 0.7 1.9 South Asia 4.1 4.7 1.7 6.6 5.4 4.7 Latin America 0.0 0.1 0.2 8.1 2.2 0.2 Middle East 0.6 0.7 0.2 4.3 1.2 3.9 Eastern Europe and Central Asia 0.2 0.1 0.2 4.1 0.2 1.6 High-income countries 0.4 0.7 0.4 102.1 0.1 1.2 World total 0.1 0.9 0.7 21.1 0.2 0.5 Source: Authors' World Bank Linkage model simulations. a. Nominal factor prices are deflated by national aggregate consumer price index (CPI) in columns 5 and 6. b. The user cost of capital and land represents the subsidy inclusive of rental cost. the no-reform baseline, which ignores the fact that such farm landowners have long enjoyed protection-inflated returns, in some cases for decades prior to the 1980s. Effects on sectoral value added Of crucial interest in terms of these policies' impact on inequality and poverty is how they have affected value added in agriculture--or, in other words, net farm income. For poverty, it matters how much that indicator changes in absolute terms in low-income countries (given that three quarters of the world's poor are farmers in developing countries), while for within-country inequality how much the indicator changes relative to value added in nonfarm sectors also matters. These results are reported in table 13.12. The results show that for developing countries as a group, value added in agri- culture is 4.9 percent higher than it would have been without reform over the past two decades, compared with just 0.4 percent for nonagriculture. A similar-sized improvement has occurred in high-income countries for nonagriculture, but there net farm incomes would have been 36 percent higher without the global Table 13.12. Impact of Going Back to 1980­84 Policies on Sectoral Value Added, Agricultural and All-Sector Policy Changes 532 (percent, relative to 2004 baseline) US$ billions Percent Agricultural policies All sectors' policies Agricultural policies All sectors' policies Agricultural Nonagricultural Agricultural Nonagricultural Agricultural Nonagricultural Agricultural Nonagricultural GDP GDP GDP GDP GDP GDP GDP GDP North and Sub- Saharan Africa 0.9 0.2 2.3 0.6 0.9 0.0 2.2 0.1 Egypt, Arab Rep. of 0.0 0.7 0.1 0.0 0.1 1.1 0.8 0.1 Madagascar 0.0 0.1 0.1 0.0 3.4 3.1 7.1 1.0 Mozambique 0.3 0.0 0.3 0.4 22.7 0.1 24.8 10.0 Nigeria 1.2 0.8 0.6 0.8 9.3 1.7 4.7 1.6 Senegal 0.0 0.0 0.1 0.0 1.1 0.8 9.0 1.0 South Africa 0.1 0.1 1.6 1.6 0.7 0.1 20.3 0.8 Tanzania 0.0 0.1 0.1 0.3 0.3 1.3 1.3 4.2 Uganda 0.1 0.1 0.1 0.1 2.9 1.6 1.9 2.1 Zambia 0.0 0.0 0.3 0.0 0.6 0.6 28.2 0.3 Zimbabwe 0.2 0.2 0.3 0.1 38.9 4.9 62.7 3.8 Rest of Africa 0.0 1.4 3.1 1.2 0.1 0.5 4.9 0.4 East and South Asia 2.0 100.7 27.1 65.2 0.5 2.9 6.4 1.9 China 9.4 37.5 27.0 29.7 5.7 3.0 16.3 2.4 Korea, Rep. of 3.2 31.3 1.2 24.9 15.1 5.4 5.4 4.3 Taiwan, China 0.5 10.1 0.8 12.2 9.9 3.7 17.6 4.4 Indonesia 0.2 2.7 0.4 3.2 0.8 1.2 1.4 1.5 Malaysia 0.1 4.0 0.3 0.4 2.0 3.8 12.9 0.3 Philippines 1.9 1.0 0.6 1.9 15.6 1.7 4.6 3.3 Thailand 3.0 7.3 0.0 6.3 14.3 2.8 0.2 2.4 Vietnam 1.2 4.5 1.0 1.2 18.8 15.6 15.8 4.2 Bangladesh 0.3 2.1 0.3 0.9 3.8 4.4 3.8 1.9 India 10.6 1.3 2.7 16.1 8.3 0.3 2.1 3.5 Pakistan 0.1 0.2 0.5 0.1 0.5 0.2 2.8 0.1 Sri Lanka 0.3 1.3 0.6 1.0 7.1 9.6 12.8 7.4 Rest of East and South Asia 0.7 4.3 0.3 1.4 11.2 2.7 5.2 0.9 Latin America 40.7 34.6 10.8 40.2 37.0 2.3 9.8 2.7 Argentina 10.9 15.1 2.7 14.3 103.5 13.8 25.5 13.1 Brazil 13.0 21.3 7.6 8.0 42.6 4.2 24.9 1.6 Chile 0.2 0.7 0.1 1.0 5.5 0.9 1.8 1.3 Colombia 5.0 1.2 1.3 12.1 53.5 1.5 13.6 15.3 Ecuador 2.9 1.7 0.8 0.5 126.0 6.7 35.4 1.9 Mexico 0.1 3.4 0.9 6.3 0.3 1.0 4.0 1.8 Nicaragua 0.0 0.1 0.0 0.0 2.4 2.3 5.1 0.4 Rest of Latin America 8.6 2.1 0.0 0.9 28.7 0.6 0.0 0.2 Eastern Europe and Central Asia 6.2 4.4 1.7 0.9 5.2 0.3 1.5 0.1 Baltic states 0.1 0.2 0.0 0.3 8.9 0.5 0.5 0.9 Bulgaria 0.4 0.1 0.4 0.6 5.6 0.3 5.6 3.4 Czech Republic 0.7 0.3 0.0 0.8 20.9 0.3 1.1 0.8 Hungary 0.7 0.1 0.3 0.8 17.9 0.1 8.3 0.9 Poland 2.5 1.7 0.3 1.0 22.6 0.9 2.5 0.5 Romania 0.5 0.3 0.1 1.1 5.8 0.5 1.6 1.9 Slovak Republic 0.1 0.1 0.0 0.3 13.5 0.4 0.3 0.8 Slovenia 0.0 0.1 0.1 0.1 11.1 0.4 11.8 0.4 Russian Federation 2.3 1.3 0.8 5.6 6.6 0.3 2.3 1.2 Kazakhstan 0.5 0.5 0.1 0.1 23.0 1.2 5.0 0.4 Turkey 1.5 0.9 3.1 0.4 4.7 0.4 9.5 0.2 Rest of Eastern Europe and Central Asia 1.5 2.1 5.0 0.8 11.1 1.8 37.8 0.7 533 (Table continues on the following page.) Table 13.12. Impact of Going Back to 1980­84 Policies on Sectoral Value Added, Agricultural and All-Sector 534 Policy Changes (continued) (percent, relative to 2004 baseline) US$ billions Percent Agricultural policies All sectors' policies Agricultural policies All sectors' policies Agricultural Nonagricultural Agricultural Nonagricultural Agricultural Nonagricultural Agricultural Nonagricultural GDP GDP GDP GDP GDP GDP GDP GDP High-income countries 58.5 28.6 144.2 143.1 14.7 0.1 36.2 0.5 Australia 2.7 11.7 0.2 0.5 13.7 2.1 1.2 0.1 Canada 0.7 4.6 5.9 2.8 5.3 0.5 45.7 0.3 EU­15 47.4 45.9 36.6 14.7 25.4 0.4 19.6 0.1 Japan 7.6 93.2 7.3 149.3 16.8 2.3 16.1 3.7 New Zealand 2.7 4.4 0.5 0.8 57.2 5.4 9.8 0.9 Rest of Western Europe 3.6 8.4 88.3 25.8 25.8 1.3 631.3 4.0 United States 6.0 25.2 5.3 17.6 5.3 0.2 4.6 0.2 Hong Kong, China and Singapore 0.0 3.4 0.1 3.4 2.2 2.1 10.3 2.1 Developing countries 44.4 145.6 38.8 32.1 5.6 1.9 4.9 0.4 North Africa 0.3 1.8 0.1 0.7 1.1 0.8 0.3 0.3 Sub-Saharan Africa 0.6 2.0 2.2 1.3 0.8 0.5 3.1 0.3 East Asia 12.6 102.8 23.6 81.4 4.7 3.5 8.9 2.8 South Asia 10.7 2.1 3.5 16.2 6.7 0.3 2.2 2.7 Latin America 40.7 34.6 10.8 40.2 37.0 2.3 9.8 2.7 Middle East 8.9 6.1 0.4 5.7 25.4 0.9 1.1 0.8 Eastern Europe and Central Asia 6.2 4.4 1.7 0.9 5.2 0.3 1.5 0.1 World total 14.2 174.2 105.4 175.2 1.2 0.5 8.8 0.5 Source: Authors' World Bank Linkage model simulations. General Equilibrium Effects of Price Distortions on Global Markets 535 reforms. For East Asia and Latin America, the gain to farmers is twice as much, for South Asia and North Africa it is less than half as much, and for Sub-Saharan Africa the gain is just above the developing-country average (although farmers in South Africa and Nigeria are made substantially worse off). However, among the countries listed in Africa, net farm incomes would increase substantially only in Mozambique, Zambia, and Zimbabwe, while for the continent as a whole they would fall very slightly (by less than 1 percent). Part of the reason for this result is that nonagricultural primary sectors--in which numerous African countries have a strong comparative advantage--have expanded (raising Africa's self-sufficiency in that sector from 182 to 191 percent), and that in turn has boosted nontradables production and employment. Net farm incomes are also estimated to have fallen in Bangladesh and Vietnam, but in those countries earnings in textiles and cloth- ing expanded. Prospective Effects of Global Removal of 2004 Price-Distorting Policies In the light of the above assessment of partial reform over the past two decades, the potential results from removing the remaining policies as of 2004 are exam- ined. In this case, the scenarios involve full global liberalization of both agricul- tural policies and nonagricultural goods trade policies. Global and national economic welfare Beginning with the baseline projection of the world economy in 2004, all agricul- tural subsidies and taxes plus import tariffs on other merchandise, as summarized in table 13.1,13 are removed globally. The Linkage model results presented here suggest that would lead to a global gain of US$168 billion per year (table 13.13), compared with the above estimate of US$233 billion per year from the partial reform since the early 1980s (table 13.2), and suggesting that in a global economic welfare sense, by 2004 the world had moved three-fifths of the way toward global free trade in goods. As a share of national income, developing countries would gain nearly twice as much as high-income gains by completing that reform process (an average increase of 0.9 percent compared with 0.5 percent for high- income countries). The results vary widely across developing countries, ranging from slight losses in the case of some South Asian and Sub-Saharan African coun- tries that would suffer exceptionally large adverse terms of trade changes (and thus be worthy candidates for "aid-for-trade" assistance as envisaged as part of the WTO's Doha Development Agenda) to 8 percent increases in the case of Ecuador (whose main export item, bananas, is now heavily discriminated against in the EU 536 Distortions to Agricultural Incentives: A Global Perspective Table 13.13. Impact on Real Income of Full Liberalization of Global Merchandise Trade, by Country/Region, 2004 (relative to 2004 baseline) Change in Total real income due just Total real income gain to change in income gain per year terms of trade (% of (US$ billions) (US$ billions) baseline)a North and Sub- Saharan Africa 0.9 6.0 0.2 ( 1.1) Egypt, Arab Rep. of 0.2 0.6 0.3 ( 0.9) Madagascar 0.0 0.0 0.9 ( 1.2) Mozambique 0.1 0.1 2.4 ( 2.0) Nigeria 0.3 0.6 0.7 ( 1.3) Senegal 0.0 0.1 2.3 ( 4.0) South Africa 0.2 0.7 0.1 ( 0.5) Tanzania 0.0 0.0 0.5 ( 0.4) Uganda 0.0 0.0 0.6 ( 0.1) Zambia 0.0 0.0 0.1 ( 0.3) Zimbabwe 0.1 0.0 3.4 (0.5) Rest of Africa 0.5 3.8 0.2 ( 1.5) East and South Asia 29.7 4.9 0.9 ( 0.1) China 3.3 0.5 0.2 (0.0) Korea, Rep. of 14.0 0.2 2.8 (0.0) Taiwan, China 1.0 0.0 0.4 (0.0) Indonesia 0.5 0.0 0.2 (0.0) Malaysia 4.2 1.0 4.7 ( 1.1) Philippines 0.0 0.5 0.1 ( 0.7) Thailand 3.3 0.1 1.4 ( 0.1) Vietnam 1.9 0.9 5.3 ( 2.5) Bangladesh 0.2 0.8 0.4 ( 1.7) India 0.8 2.9 0.2 ( 0.6) Pakistan 0.1 0.6 0.2 ( 0.8) Sri Lanka 0.8 0.5 5.1 (3.1) Rest of East and South Asia 1.9 0.8 1.4 (0.5) Latin America 15.8 2.5 1.0 (0.2) Argentina 3.2 0.7 2.6 ( 0.6) Brazil 6.8 5.6 1.6 (1.3) Chile 0.3 0.2 0.4 (0.3) Colombia 2.2 0.7 3.1 (1.0) Ecuador 2.0 1.1 8.2 (4.4) Mexico 0.7 3.4 0.1 ( 0.6) Nicaragua 0.0 0.0 1.3 (0.4) Rest of Latin America 2.0 1.0 0.5 ( 0.3) General Equilibrium Effects of Price Distortions on Global Markets 537 Table 13.13. Impact on Real Income of Full Liberalization of Global Merchandise Trade, by Country/Region, 2004 (continued) Change in Total real income due just Total real income gain to change in income gain per year terms of trade (% of (US$ billions) (US$ billions) baseline)a Eastern Europe and Central Asia 14.2 3.6 1.2 ( 0.3) Baltic states 0.5 0.1 1.8 (0.3) Bulgaria 0.2 0.2 1.4 ( 1.4) Czech Republic 1.0 0.1 1.4 ( 0.2) Hungary 0.4 0.1 0.6 ( 0.1) Poland 2.0 0.1 1.2 (0.1) Romania 0.1 0.7 0.3 ( 1.9) Slovak Republic 0.7 0.1 2.3 (0.4) Slovenia 0.3 0.1 1.5 (0.3) Russian Federation 5.4 3.1 1.2 ( 0.7) Kazakhstan 0.4 0.2 1.1 (0.6) Turkey 1.3 0.5 0.6 ( 0.2) Rest of Eastern Europe and Central Asia 2.2 0.5 2.1 (0.4) High-income countries 102.8 11.3 0.5 (0.1) Australia 2.4 1.9 0.5 (0.4) Canada 0.6 1.2 0.1 ( 0.2) EU-15 56.8 3.8 0.7 (0.0) Japan 23.1 10.4 0.7 (0.3) New Zealand 2.2 1.8 3.2 (2.6) Rest of Western Europe 13.1 0.1 2.7 (0.0) United States 2.8 0.9 0.0 (0.0) Hong Kong, China and Singapore 1.7 1.4 1.4 (1.1) Developing countries 64.9 12.2 0.9 ( 0.2) North Africa 0.9 2.8 0.5 ( 1.5) Sub-Saharan Africa 0.0 3.2 0.0 ( 0.9) East Asia 30.1 1.0 1.1 (0.0) South Asia 0.4 3.9 0.1 ( 0.6) Latin America 15.8 2.5 1.0 (0.2) Middle East 4.2 0.2 0.8 (0.0) Eastern Europe and Central Asia 14.2 3.6 1.2 ( 0.3) World total 167.7 1.0 0.6 (0.0) Source: Authors' World Bank Linkage model simulations. a. Numbers in parentheses refer to that due to terms of trade effects. 538 Distortions to Agricultural Incentives: A Global Perspective market, where former colonies (via the now-expired Cotonou Agreement) and least developed countries previously enjoyed preferential duty-free access). The second column of numbers and those in parentheses in table 13.13 show the amount of that welfare gain due to changes in the international terms of trade for each country. For developing countries as a group, the terms-of-trade effect is slightly negative, while for high-income countries the effect is slightly positive. Regional and sectoral distribution of welfare effects One way to decompose the real income gains from full removal of price distortions globally, so as to better understand the sources for each region, is to assess the impacts of developing country liberalization versus high-income country liberaliza- tion in different economic sectors. These results are provided in table 13.14. They suggest global liberalization of agriculture and food markets would contribute 70 percent of the total global gains from merchandise reform. This is similar to the 63 percent found for 2015 by Anderson, Martin, and van der Mensbrugghe (2006b) using version 6 of the GTAP database anchored on 2001 estimates of dis- tortions. This robust result is remarkable given the low shares of agriculture and food in global gross domestic product (GDP) and global merchandise trade (less than 7 percent). For developing countries, the importance of agricultural policies is even greater, at 72 percent (compared with 69 percent for high-income countries--see row 7 of table 13.14). Five-sevenths of the global gains from removing agricultural policies are accounted for by the farm policies of high-income countries (columns 3 and 6 of table 13.14). Those policies also account for almost one-quarter of the overall gains from trade reform to developing countries, meanwhile, developing-country policies are responsible for 40 percent of high-income countries' gains (columns 1 and 3 of table 13.14). If only high-income countries were to liberalize their agri- cultural markets--as some countries have suggested in the WTO's ongoing Doha Round of trade negotiations--they would provide less than one-third of the potential gains to developing countries from global farm policy reform. Quantities produced and traded The full liberalization results suggest there would be little change in developing countries' aggregate shares of global output and exports of nonfarm products other than for textiles and apparel. Their shares in agricultural and processed food markets, however, change noticeably: the export share rises from 54 to 64 percent and the output share rises from 46 to 50 percent. More significantly, the increases occur in nearly all agricultural and food industries. As a result, the share of global Table 13.14. Regional and Sectoral Sources of Welfare Gains from Full Liberalization of Global Merchandise Trade, 2004 (relative to 2004 baseline) Gains,a by region (US$ billion) Regional gain (%) Developing High-income World Developing High-income World countries countriesb countries countriesb Developing countries liberalize: Agriculture and light processing 31.8 3.9 35.6 48.6 3.8 21.3 Manufacturing and services 5.6 36.7 42.3 8.6 35.8 25.2 Total 37.4 40.6 77.9 57.2 39.6 46.5 High-income countries liberalize: Agriculture and light processing 15.1 66.4 81.6 23.2 64.9 48.6 Manufacturing and services 12.8 4.6 8.2 19.6 4.5 4.9 Total 28.0 61.8 89.8 42.8 60.4 53.5 All countries liberalize: Agriculture and light processing 46.9 70.3 117.2 71.8 68.7 69.9 Manufacturing and services 18.4 32.1 50.5 28.2 31.3 30.1 Total 65.3 102.3 167.7 100.0 100.0 100.0 Source: Authors' World Bank Linkage model simulations. a. Small interaction effects are distributed proportionately; dollar values do not add exactly, due to rounding. b. Includes European transition economies, excludes Rep. of Korea and Taiwan (which are included in `developing countries'). 539 540 Distortions to Agricultural Incentives: A Global Perspective production of farm products that is exported rises dramatically for many indus- tries and, for the sector as a whole, increases from 8 to 13 percent excluding intra- EU trade (table 13.15). That "thickening" of international food markets would have a substantial dampening effect on the instability of prices and quantities traded in those markets. The impact of full trade reform on agricultural and food output and trade is shown for each country or region in table 13.16, where it is clear that global farm trade is enhanced by more than one-third, whereas the global value of output is virtually unchanged, dropping just 2.6 percent. This suggests that, in aggregate, the proagricultural policies of high-income countries are not quite fully offset by the policies of developing countries--whereas the antitrade biases in policies of both groups of countries reinforce each other. The increase in exports of those goods from developing countries would be a huge US$163 billion per year. Granted, Latin America accounts for nearly half of that increase, but all develop- ing regions' exports expand. This means developing countries' share of produc- tion exported would be much higher with full trade reform. In fact, table 13.17 shows exports would increase for almost all developing countries, rising in aggre- gate from 10 to 17 percent. Also of interest is what happens to agricultural imports: developing countries as a group would see farm imports growing less than exports under full trade lib- eralization (table 13.16). That means their food and agricultural self-sufficiency ratios would rise, although in aggregate only slightly. For high-income countries, that ratio would fall 5 percentage points (slightly less if Eastern Europe is included), while in East Asia and Africa it would rise 2­3 points, for South Asia it would be unchanged, and for Latin America it would jump from 112 to 126 per- cent (table 13.18). As already mentioned, full trade liberalization also substantially increases the share of agricultural and food production that is exported globally, thereby "thickening" international markets, which would dampen international food price fluctuations and thereby reduce concerns about vulnerability to import dependence. The extent of this global public good aspect of agricultural trade reform can be sensed for different products from the results reported in tables 13.19. Highly protected sugar and milk, as well as grains and oilseeds, are espe- cially noteworthy. Also noteworthy from that table is the extent to which the developing-country shares of output exported rise for certain products. The share of their grain production that is exported would double, for example, and for meat it would more than double, while for sugar it would rise nearly four-fold. Cotton (plant-based fibers) too would become more of the domain of developing countries. Table 13.15. Impact of Full Global Liberalization on Shares of Global Output Exported, and Developing Country Shares of Global Output and Exports,a by Product, 2004 (percent) Share of global Developing countries' Developing countries' output exporteda share of global output share of global exportsa 2004 Full global 2004 Full global 2004 Full global benchmark liberalization benchmark liberalization benchmark liberalization Paddy rice 1 2 81 82 56 42 Wheat 16 22 67 71 25 39 Other grains 11 15 55 57 35 56 Oilseeds 21 28 69 74 54 68 Plant-based fibers 25 25 74 83 50 79 Vegetables and fruits 9 15 72 77 69 80 Other crops 14 17 49 49 75 62 Cattle, sheep, goats 2 2 43 48 56 59 Other livestock 4 4 65 67 43 46 Wool 13 14 82 81 16 18 Beef and sheep meat 7 21 27 41 31 68 Other meat products 7 12 32 34 42 45 Vegetable oils and fats 20 30 52 58 80 84 Dairy products 5 11 29 33 28 41 Processed rice 5 7 76 79 85 87 Refined sugar 8 42 52 85 78 90 Other food, beverages, and tobacco 9 12 35 36 50 59 Other primary products 31 33 64 63 76 76 Textile and wearing apparel 28 35 53 57 74 77 Other manufacturing 24 26 32 31 43 43 Services 3 3 20 20 31 30 Agriculture and food 8 13 46 50 54 64 Agriculture 8 11 62 65 55 64 541 Processed foods 8 14 37 40 52 63 Source: Authors' World Bank Linkage model simulations. a. Excluding intra-EU trade. 542 Table 13.16. Impacts of Full Global Trade Liberalization on Agricultural and Food Output and Trade, by Country/Region, 2004 (relative to 2004 baseline) US$ billions Change relative to baseline (%) Output Exports Imports Output Exports Imports North and Sub-Saharan Africa 13.8 20.5 10.0 7.2 99.1 46.0 Egypt, Arab Rep. of 0.4 0.5 0.1 2.2 39.2 4.2 Madagascar 0.0 0.0 0.0 0.4 2.7 4.3 Mozambique 0.9 1.0 0.1 52.3 597.1 33.3 Nigeria 0.5 0.4 0.7 2.9 92.8 43.1 Senegal 0.0 0.0 0.0 1.9 35.0 0.3 South Africa 0.7 0.9 0.8 2.4 26.7 42.9 Tanzania 0.0 0.2 0.1 0.7 28.5 31.2 Uganda 0.0 0.0 0.0 0.6 1.3 1.5 Zambia 0.1 0.1 0.0 5.2 22.3 35.9 Zimbabwe 0.4 0.3 0.1 25.7 38.0 39.2 Rest of Africa 12.0 17.0 8.3 10.5 133.1 64.3 East and South Asia 25.0 39.5 24.7 2.7 83.4 36.7 China 6.2 7.7 6.7 1.7 76.5 27.5 Korea, Rep. of 1.0 1.0 6.2 1.7 194.1 75.0 Taiwan, China 1.9 0.3 1.5 9.1 62.8 35.5 Indonesia 1.1 1.6 1.0 1.8 21.6 21.5 Malaysia 1.6 1.3 0.7 8.9 17.0 17.8 Philippines 1.1 1.9 0.8 3.5 120.5 35.0 Thailand 9.5 8.3 1.9 17.4 133.0 78.1 Vietnam 0.5 1.1 0.6 3.3 54.0 55.6 Bangladesh 0.6 0.4 0.8 2.4 261.2 38.3 India 1.1 9.0 1.4 0.5 131.2 24.2 Pakistan 0.6 0.5 1.0 1.3 45.0 43.0 Sri Lanka 0.1 0.1 0.6 1.2 18.2 69.3 Rest of East and South Asia 8.0 6.4 1.4 41.5 266.1 29.5 Latin America 87.2 71.5 7.2 26.8 106.4 29.8 Argentina 12.2 15.1 0.3 37.8 95.6 81.8 Brazil 45.8 25.7 2.1 45.3 100.7 94.8 Chile 0.5 0.4 0.2 4.7 11.3 15.8 Colombia 3.1 4.9 1.1 14.6 161.4 81.7 Ecuador 4.2 4.6 0.3 46.1 198.7 71.8 Mexico 0.3 0.3 0.4 0.4 5.8 4.3 Nicaragua 0.0 0.1 0.0 2.9 21.6 19.4 Rest of Latin America 21.6 20.4 2.8 25.7 175.9 30.4 Eastern Europe and Central Asia 10.4 17.4 20.3 2.6 79.7 77.6 Baltic states 1.2 0.1 0.4 16.9 15.5 30.9 Bulgaria 4.2 2.6 0.6 6.6 366.5 118.1 Czech Republic 2.2 0.1 0.7 12.0 10.9 40.5 Hungary 0.9 0.4 0.8 6.0 17.1 66.6 Poland 1.7 2.5 2.5 3.9 80.7 88.8 Romania 0.2 1.3 1.1 1.0 190.5 78.3 Slovak Republic 0.9 0.1 0.4 11.3 12.0 64.1 Slovenia 0.6 0.1 0.2 17.1 54.1 26.2 Russian Federation 12.9 3.2 8.8 13.1 179.4 98.9 Kazakhstan 1.5 1.4 0.0 11.8 142.9 11.6 Turkey 2.0 2.3 2.9 3.1 61.5 92.1 Rest of Eastern Europe and 543 Central Asia 3.0 4.1 2.0 7.7 71.3 53.4 (Table continues on the following page.) Table 13.16. Impacts of Full Global Trade Liberalization on Agricultural and Food Output and Trade, 544 by Country/Region, 2004 (continued) (relative to 2004 baseline) US$ billions Change relative to baseline (%) Output Exports Imports Output Exports Imports High-income countries 233.2 9.2 89.8 13.1 4.0 38.3 Australia 12.0 7.0 0.2 19.8 41.2 11.1 Canada 1.6 3.6 2.7 2.4 24.1 32.8 EU-15 190.9 38.8 50.9 21.2 29.2 31.9 Japan 39.1 0.4 16.8 22.9 87.7 69.1 New Zealand 10.6 6.4 0.2 46.6 74.3 27.1 Rest of Western Europe 11.6 11.7 9.8 19.4 312.0 132.7 United States 12.8 0.6 9.3 2.6 1.1 32.4 Hong Kong, China and Singapore 0.1 0.0 0.1 2.1 6.3 1.6 Developing countries 137.6 163.6 64.6 7.1 100.0 40.4 North Africa 11.4 13.3 6.1 17.3 377.2 62.5 Sub-Saharan Africa 2.5 7.2 3.8 1.9 41.9 32.3 East Asia 25.1 29.5 20.8 4.0 77.4 37.4 South Asia 0.1 10.0 3.9 0.0 108.3 33.2 Latin America 87.2 71.5 7.2 26.8 106.4 29.8 Middle East 22.0 14.8 2.5 21.5 222.7 12.1 Eastern Europe and Central Asia 10.4 17.4 20.3 2.6 79.7 77.6 World total 95.7 154.4 154.4 2.6 39.1 39.1 Source: Authors' World Bank Linkage model simulations. General Equilibrium Effects of Price Distortions on Global Markets 545 Table 13.17. Impact of Global Liberalization on Share of Agricultural and Food Production Exported by Country/Region, 2004 (percent) 2004 Full global benchmark data liberalization Developing countries 9.5 16.9 North Africa 6.3 20.6 Sub-Saharan Africa 13.8 19.3 East Asia 8.4 15.1 South Asia 3.7 7.5 Latin America 18.1 28.2 Middle East 7.4 17.2 Eastern Europe and Central Asia 6.8 11.1 High-income countries 13.0 14.1 World total 11.4 15.4 Source: Authors' World Bank Linkage model simulations. Effects on product and factor prices How do different agricultural prices in international markets change with liberal- ization of distortionary agricultural policies and other protection? The average real international prices of agricultural and lightly processed food products would be only 1.3 percent higher in the absence of all merchandise trade distortions, or 2.0 percent if just agricultural policies were liberalized (table 13.20). The net effects of present distortions especially dampen the international prices of beef, milk, rice, and cotton. But they are propping up the prices of some other products, because export taxes are still in place in some developing countries, most notably Argentina. Because of the impacts on real rewards to labor, capital, and land, redistribu- tion of welfare among groups within each country following trade reform can be much larger than the aggregate change. Those effects are reported in table 13.21, where factor rewards are deflated by the overall consumer price index. As shown in the table, food prices would fall more than that overall CPI. Consistent with trade theory, those results suggest that unskilled workers in developing countries--the majority of whom work on farms--would benefit most from reform, followed by skilled workers, and then capital owners. Returns to immo- bile agricultural land also rises in developing countries, but by less than for more mobile factors. Insofar as unskilled workers spend a higher share of their income Table 13.18. Impact of Global Liberalization on Self-Sufficiencya in Agricultural and Other Products, by Region, 2004 High-income Developing North and Sub- Latin Eastern Europe countries countries Saharan Africa America East Asia South Asia and Central Asia 2004 Global 2004 Global 2004 Global 2004 Global 2004 Global 2004 Global 2004 Global bench- liberali- bench- liberali- bench- liberali- bench- liberali- bench- liberali- bench- liberali- bench- liberali- mark zation mark zation mark zation mark zation mark zation mark zation mark zation Paddy rice 101 105 100 99 97 96 93 72 100 101 101 101 95 92 Wheat 141 140 88 89 67 46 80 98 68 65 100 98 102 117 Other grains 108 102 94 98 94 91 98 119 88 81 103 105 103 113 Oilseeds 104 92 97 103 104 130 140 167 66 51 100 101 106 115 Plant-based fibers 161 112 88 97 177 265 94 107 54 58 93 95 104 118 Vegetables and fruits 90 78 105 109 108 103 153 221 102 104 99 98 99 92 Other crops 90 91 113 110 138 138 143 133 110 104 104 104 90 88 Cattle, sheep, goats 100 100 100 100 101 99 102 102 98 97 100 100 102 102 Other livestock 101 101 100 100 101 100 101 100 99 99 100 100 99 98 Wool 161 180 92 91 103 104 103 102 78 75 96 93 96 99 Beef and sheep meat 101 85 97 134 96 102 108 183 83 77 126 652 95 85 Other meat products 100 99 100 103 92 85 121 143 101 103 96 95 96 93 Vegetable oils and fats 95 85 103 114 69 191 141 143 115 116 78 66 93 96 Dairy products 103 100 94 101 76 79 97 102 78 78 99 99 102 104 Processed rice 99 95 100 101 69 63 94 85 104 108 104 104 92 87 Refined sugar 98 41 102 133 95 100 131 227 98 196 96 91 98 70 Other food, beverages, 99 97 103 105 101 100 108 112 105 113 106 94 100 98 and tobacco Other primary products 76 76 122 122 180 189 148 155 84 82 75 69 115 116 Textile and wearing apparel 81 76 123 128 98 91 104 91 144 155 144 153 101 95 Other manufacturing 101 102 98 96 77 74 96 91 106 105 90 89 95 95 Services 101 101 101 101 101 102 100 100 101 100 100 101 101 101 Agriculture and food 100 95 101 105 100 103 112 126 100 102 100 100 99 98 Agriculture 99 96 100 102 104 103 115 126 96 95 100 100 100 101 Processed foods 100 95 101 108 94 103 110 126 104 111 100 101 99 96 Source: Authors' World Bank Linkage model simulations. a. Self-sufficiency is defined as domestic production as a percentage of domestic consumption measured in value terms at fob prices. Table 13.19. Share of Production Exported and of Consumption Imported by World, High-Income, and Developing Countries, before and after Full Global Liberalization of All Merchandise Trade, by Product, 2004 (percent) Share of production exported Share of consumption imported a High-income countries Developing countries High-income countriesa Developing countries 2004 Global 2004 Global 2004 Global 2004 Global benchmark liberalization benchmark liberalization benchmark liberalization benchmark liberalization Paddy rice 3 7 1 1 2 3 1 2 Wheat 37 47 6 12 11 25 17 21 Other grains 15 16 7 15 9 14 11 15 Oilseeds 31 34 16 25 26 36 16 22 Plant-based fibers 50 31 17 24 18 22 26 25 Vegetables and fruits 10 13 9 15 18 30 4 7 Other crops 7 13 21 22 16 20 11 14 Cattle, sheep, goats 1 2 2 2 2 2 2 2 Other livestock 6 7 3 3 6 6 3 3 Wool 60 62 2 3 35 31 10 12 Beef and sheep meat 6 11 7 35 5 24 10 13 Other meat products 6 10 9 16 6 12 8 14 Vegetable oils and fats 8 11 31 43 12 24 26 34 Dairy products 5 10 4 14 2 10 10 14 Processed rice 3 4 5 8 4 9 5 7 Refined sugar 4 30 12 44 5 66 10 25 Other food, beverages, and tobacco 7 8 12 20 8 10 9 16 Other primary products 20 21 37 39 38 39 22 24 Textile and wearing apparel 15 19 39 48 30 37 23 31 Other manufacturing 20 21 32 36 19 20 32 38 Services 3 3 5 4 2 2 5 5 Agriculture and food 7 9 9 17 8 13 8 12 Agriculture 9 11 7 11 10 15 7 9 Processed foods 6 9 12 23 7 13 10 16 Source: Authors' Linkage model simulations. a. Excluding intra-EU trade. 548 Distortions to Agricultural Incentives: A Global Perspective Table 13.20. Impact of Full Global Liberalization of Agricultural and All Goods Markets on Real International Product Prices, 2004 (percent, relative to 2004 baseline) Agricultural All goods policies sectors policies Paddy rice 6.9 6.6 Wheat 1.8 1.4 Other grains 2.6 2.7 Oilseeds 2.2 2.4 Sugar cane and beet 1.1 2.0 Plant-based fibers 4.7 2.9 Vegetables and fruits 2.4 1.8 Other crops 1.7 1.0 Cattle, sheep, goats 0.2 1.1 Other livestock 1.2 2.1 Raw milk 0.7 0.2 Wool 3.5 3.3 Beef and sheep meat 5.6 4.6 Other meat products 1.3 0.6 Vegetable oils and fats 1.4 1.9 Dairy products 4.6 3.8 Processed rice 2.8 2.9 Refined sugar 2.5 1.3 Other food, beverages, and tobacco 1.7 1.3 Textile and wearing apparel 0.3 1.2 Other manufacturing 0.2 0.2 Merchandise trade 0.3 0.2 Agriculture and food 0.8 0.3 Agriculture 1.5 0.9 Agriculture and light processing 2.0 1.3 Source: Authors' World Bank Linkage model simulations. Note: Model numeraire is the export price index of high-income countries' manufactured exports. on food than the average citizen, these results understate the extent of their gain. These results suggest both inequality and poverty could be alleviated by such reform. Effects on sectoral value added Of crucial interest in terms of these policies' impact on inequality and poverty is how they affect value added in agriculture, in other words net farm income. General Equilibrium Effects of Price Distortions on Global Markets 549 Table 13.21. Impacts of Full Global Merchandise Trade Liberalization on Real Factor Prices,a 2004 (percent, relative to 2004 baseline) Unskilled Skilled Capitalb Landb Aggregate Food wages wages user cost user cost CPI CPI Developing countries 3.5 3.0 2.9 1.6 0.9 2.8 North Africa 7.0 7.7 5.3 0.5 5.2 7.2 Sub-Saharan Africa 3.2 3.2 3.8 0.2 3.8 4.9 East Asia 4.0 3.4 3.3 1.9 0.1 2.7 South Asia 0.6 2.3 1.2 6.2 1.6 0.3 Latin America 4.5 1.4 1.9 21.1 1.2 3.2 Middle East 8.3 2.9 4.7 43.8 3.3 10.5 Eastern Europe and Central Asia 1.7 3.2 2.6 4.5 2.3 4.5 High-income countries 0.2 1.0 0.5 17.9 0.6 3.6 World total 0.9 1.3 1.2 3.1 0.7 3.2 Source: Authors' World Bank Linkage model simulations. a. Nominal factor prices are deflated by national aggregate consumer price index (CPI) in columns 5 and 6. b. The user cost of capital and land represents the subsidy inclusive rental cost. These results for full global reform are reported in the first four columns of table 13.22. The results show that for developing countries as a group, value added in agri- culture rises by 5.6 percent, compared with 1.9 percent for nonagriculture, follow- ing full global reform of all merchandise trade. Latin America is where net farm income expands most, averaging 37 percent but exceeding 100 percent for Argentina and Ecuador and 40­50 percent for Brazil and Colombia. It also expands considerably in East Asia, and more than nonagricultural value added-- including in China. However, among the countries listed in Africa, net farm incomes would increase substantially only in Mozambique, Zambia, and Zimbabwe, while for the continent as a whole they would fall very slightly (by less than 1 percent). Partly that is because nonagricultural primary sectors, in which numerous African countries have a strong comparative advantage, would expand (raising Africa's self-sufficiency in that sector from 180 to 189 percent; see table 13.18). In turn, that expansion would boost production and employment of nontradable goods and services. Net farm incomes are estimated to fall also in South Asia (by 7 percent), but textiles and clothing incomes in that region would expand (raising self-sufficiency from 144 to 153 percent). In India, where the skilled-to-unskilled wage differential rises, incomes in skill-intensive goods and services production would also expand. 550 Table 13.22. Effects of Full Global Liberalization of Agricultural and All Merchandise Trade on Sectoral Value Added, by Country and Region, 2004 (relative to 2004 baseline) US$ billions Percent Agricultural policies All sectors' policies Agricultural policies All sectors' policies Non- Non- Non- Non- Agricultural agricultural Agricultural agricultural Agricultural agricultural Agricultural agricultural GDP GDP GDP GDP GDP GDP GDP GDP North and Sub- Saharan Africa 0.1 5.1 0.9 0.2 0.1 0.8 0.9 0.0 Egypt, Arab Rep. of 0.1 0.2 0.0 0.7 1.3 0.4 0.1 1.1 Madagascar 0.0 0.0 0.0 0.1 3.2 0.1 3.4 3.1 Mozambique 0.3 0.0 0.3 0.0 23.6 0.6 22.7 0.1 Nigeria 0.6 0.2 1.2 0.8 4.8 0.5 9.3 1.7 Senegal 0.0 0.0 0.0 0.0 1.5 0.8 1.1 0.8 South Africa 0.2 0.7 0.1 0.1 2.7 0.4 0.7 0.1 Tanzania 0.0 0.0 0.0 0.1 0.6 0.3 0.3 1.3 Uganda 0.1 0.0 0.1 0.1 1.6 0.4 2.9 1.6 Zambia 0.0 0.0 0.0 0.0 0.7 0.5 0.6 0.6 Zimbabwe 0.1 0.0 0.2 0.2 24.2 0.8 38.9 4.9 Rest of Africa 0.5 3.9 0.0 1.4 0.7 1.4 0.1 0.5 East and South Asia 1.4 24.4 2.0 100.7 0.3 0.7 0.5 2.9 China 4.6 2.5 9.4 37.5 2.8 0.2 5.7 3.0 Korea, Rep. of 4.0 7.2 3.2 31.3 18.7 1.2 15.1 5.4 Taiwan, China 0.5 0.8 0.5 10.1 11.3 0.3 9.9 3.7 Indonesia 0.3 1.1 0.2 2.7 1.1 0.5 0.8 1.2 Malaysia 0.2 0.9 0.1 4.0 6.3 0.8 2.0 3.8 Philippines 1.7 0.3 1.9 1.0 13.8 0.5 15.6 1.7 Thailand 2.9 2.7 3.0 7.3 14.0 1.0 14.3 2.8 Vietnam 1.4 0.0 1.2 4.5 22.8 0.0 18.8 15.6 Bangladesh 0.2 0.4 0.3 2.1 2.6 0.9 3.8 4.4 India 7.8 6.3 10.6 1.3 6.1 1.4 8.3 0.3 Pakistan 0.2 0.1 0.1 0.2 1.0 0.1 0.5 0.2 Sri Lanka 0.0 0.0 0.3 1.3 0.0 0.1 7.1 9.6 Rest of East and South Asia 0.6 2.3 0.7 4.3 9.6 1.4 11.2 2.7 Latin America 40.0 42.2 40.7 34.6 36.3 2.8 37.0 2.3 Argentina 12.4 8.1 10.9 15.1 116.8 7.4 103.5 13.8 Brazil 12.2 22.7 13.0 21.3 40.1 4.4 42.6 4.2 Chile 0.2 0.3 0.2 0.7 5.0 0.3 5.5 0.9 Colombia 5.0 2.1 5.0 1.2 53.5 2.7 53.5 1.5 Ecuador 2.6 2.9 2.9 1.7 113.1 11.4 126.0 6.7 Mexico 0.2 0.6 0.1 3.4 1.0 0.2 0.3 1.0 Nicaragua 0.0 0.0 0.0 0.1 3.0 1.4 2.4 2.3 Rest of Latin America 7.9 5.5 8.6 2.1 26.3 1.5 28.7 0.6 Eastern Europe and Central Asia 5.2 4.4 6.2 4.4 4.4 0.3 5.2 0.3 Baltic states 0.1 0.1 0.1 0.2 7.5 0.3 8.9 0.5 Bulgaria 0.3 0.1 0.4 0.1 5.1 0.4 5.6 0.3 Czech Republic 0.7 0.4 0.7 0.3 19.2 0.4 20.9 0.3 Hungary 0.7 0.3 0.7 0.1 16.8 0.4 17.9 0.1 (Table continues on the following pages.) 551 552 Table 13.22. Effects of Full Global Liberalization of Agricultural and All Merchandise Trade on Sectoral Value Added, by Country and Region, 2004 (continued) (relative to 2004 baseline) US$ billions Percent Agricultural policies All sectors' policies Agricultural policies All sectors' policies Non- Non- Non- Non- Agricultural agricultural Agricultural agricultural Agricultural agricultural Agricultural agricultural GDP GDP GDP GDP GDP GDP GDP GDP Poland 2.4 2.1 2.5 1.7 21.8 1.1 22.6 0.9 Romania 0.3 0.2 0.5 0.3 3.7 0.4 5.8 0.5 Slovak Republic 0.1 0.1 0.1 0.1 11.8 0.2 13.5 0.4 Slovenia 0.0 0.1 0.0 0.1 9.2 0.4 11.1 0.4 Russian Federation 2.2 0.7 2.3 1.3 6.3 0.2 6.6 0.3 Kazakhstan 0.5 0.4 0.5 0.5 23.1 1.1 23.0 1.2 Turkey 1.0 0.9 1.5 0.9 3.2 0.4 4.7 0.4 Rest of Eastern Europe and Central Asia 1.5 0.5 1.5 2.1 11.1 0.4 11.1 1.8 High-income countries 55.1 61.9 58.5 28.6 13.8 0.2 14.7 0.1 Australia 2.2 8.4 2.7 11.7 10.9 1.5 13.7 2.1 Canada 0.4 2.5 0.7 4.6 3.4 0.3 5.3 0.5 EU-15 42.9 16.7 47.4 45.9 23.0 0.2 25.4 0.4 Japan 7.6 4.5 7.6 93.2 16.7 0.1 16.8 2.3 New Zealand 2.7 4.1 2.7 4.4 57.7 5.0 57.2 5.4 Rest of Western Europe 3.6 6.5 3.6 8.4 25.8 1.0 25.8 1.3 United States 6.4 18.6 6.0 25.2 5.7 0.2 5.3 0.2 Hong Kong, China and Singapore 0.0 0.6 0.0 3.4 3.7 0.4 2.2 2.1 Developing countries 42.7 79.5 44.4 145.6 5.4 1.0 5.6 1.9 North Africa 0.1 3.9 0.3 1.8 0.4 1.8 1.1 0.8 Sub-Saharan Africa 0.2 1.2 0.6 2.0 0.3 0.3 0.8 0.5 East Asia 6.8 17.7 12.6 102.8 2.6 0.6 4.7 3.5 South Asia 8.2 6.7 10.7 2.1 5.1 1.1 6.7 0.3 Latin America 40.0 42.2 40.7 34.6 36.3 2.8 37.0 2.3 Middle East 9.2 3.3 8.9 6.1 26.3 0.5 25.4 0.9 Eastern Europe and Central Asia 5.2 4.4 6.2 4.4 4.4 0.3 5.2 0.3 World total 12.4 141.4 14.2 174.2 1.0 0.4 1.2 0.5 Source: Authors' World Bank Linkage model simulations. 553 554 Distortions to Agricultural Incentives: A Global Perspective Comparison with previous results based on version 6 of the GTAP protection data Results from global CGE models can differ for myriad reasons, even when the same model is used (van der Mensbrugghe 2006; Valenzuela, Anderson, and Hertel 2008). The full liberalization results studied in this section use the same model and prospective global reform experiment as in Anderson, Martin, and van der Mensbrugghe (2006a, 2006b), differing in just two important respects. One is that the present results refer to the world economy with its distortions as of 2004, whereas the earlier exercise took 2001 as its base, projected the world forward to 2005 to include some key policy reforms over those four years, and then projected another decade to 2015 (using the dynamic version of the Linkage model), by which time developing countries are a larger share of the modeled global econ- omy. The other difference is the use in the present exercise of the new 2004 agri- cultural distortions database for developing countries. As can be seen in the first column of table 13.A.1 in the annex, the estimates of assistance via trade taxes for agricultural and lightly processed food are somewhat lower for the developing- country regions in the new 2004 data set based on price comparisons than is the case in the GTAP model based predominantly on import tariffs. In particular, some agricultural export taxes are included in the revised distortions to incen- tives, particularly in Argentina, where they average 21 percent for agricultural products. With these differences, modeling a move to global free trade would gen- erate less import growth and more export growth in farm products for developing countries, but to differing degrees in different regions. In these new results, developing countries' share of global agricultural and food output rises from 46 to 50 percent, and their share of exports from 54 to 64 percent (compared with rises from 54 to 56 percent and from 51 to 55 percent, respectively, in the Anderson, Martin, and van der Mensbrugghe [2006a] study). The overall value of agricultural output in developing countries rises 7 percent in these new results, compared with just 2 percent previously. A major part of that difference is due to the simulated removal of Argentina's export taxes, which causes Argentina's farm output to jump. Argentinean export taxes are also partly responsible for a smaller set of impacts of liberalization on international prices of farm products. For primary agricultural prices in aggregate, the impact reported in table 13.20, 0.9 percent, compares with an average of 5.5 percent in Anderson, Martin, and van der Mensbrugghe 2006a, table 13.15. These differences affect the new economic welfare results for South Asia and Sub-Saharan Africa especially. Previously, both regions were estimated to have gained from global reform, whereas the new estimates suggest they may lose slightly (0.1 percent). South Asian farmers lose less in the current results than General Equilibrium Effects of Price Distortions on Global Markets 555 previously, however, because the new distortion estimates involve a lower level of agricultural protection in India's baseline data. As with all modeling, the results depend on the assumptions made in structur- ing the model. Of particular relevance here is that several assumptions create a downward bias on the estimates of gains from trade. They include constant (rather than increasing) returns to scale, no productivity effects of reform (for example, of the sort stressed by Melitz 2003), and no possibility for new markets to be created following reform. In addition, there is always the issue of product and regional aggregation: the less disaggregated the specification of the world economy, the smaller the estimated benefits from reform. Conclusions By way of summing up, several key findings from this study are worth emphasiz- ing, beginning with those from the retrospective experiment of comparing what conditions in 2004 would have been had the agricultural price and trade policy reforms over the previous quarter century not taken place: · without those policy reforms, global welfare in 2004 would have been lower by US$233 billion per year; · even though it depressed their terms of trade, developing economies benefited proportionately more than high-income economies (1.0 percent, compared with 0.7 percent of national income) from those policy reforms; · developing countries' share of the world's primary agricultural exports rose from 43 to 55 percent, and its farm output share from 58 to 62 percent, because of those reforms, with rises in nearly all agricultural industries except rice and sugar, where the growth in protectionism in high-income countries has been greatest; · the share of global farm production exported (excluding intra-EU trade) in 2004 was slightly smaller as a result of those reforms since 1980­84, because of less farm export subsidies, so agriculture's 8 percent share in 2004 stands in stark contrast to the 31 percent share for other primary products and the 25 percent for all other goods--a "thinness" that is an important contributor to the volatility of international prices for weather-dependent farm products; · the average real price in international markets would have been 13 percent lower for agricultural and food products had policies not changed over the past quarter century; · for developing countries as a group, value added in agriculture is 4.9 percent higher than it would have been without those reforms, and more than 10 times the proportional gain of just 0.4 percent for nonagriculture. 556 Distortions to Agricultural Incentives: A Global Perspective The findings from the second experiment, aimed at understanding the effects of the trade restrictions remaining in place as of 2004, are equally stark: · the global welfare cost of the policies remaining in 2004 is US$168 billion per year, which, when compared with the estimated gain of US$233 billion annu- ally from the partial reform since the early 1980s, suggests that in a global wel- fare sense the world had moved three-fifths of the way toward global free trade in goods between the early 1980s and the mid-2000s; · as a share of national income, developing countries would gain nearly twice as much as high-income countries by completing the reform process (an average increase of 0.9 percent, compared with 0.5 percent for high-income countries, but as high as 8 percent for banana-exporting Ecuador); · of those prospective welfare gains from global liberalization, 60 percent would come from agriculture and food policy reform--a striking result given that the shares of agriculture and food in global GDP and global merchandise trade are less than 9 percent; · the contribution of agricultural policy reform to the prospective welfare gain for developing countries is even greater, at 83 percent; · if only high-income countries were to liberalize their agricultural markets--as some countries have suggested in the WTO's ongoing Doha Round--that would provide less than two-thirds of the potential gains to developing coun- tries from global agricultural policy reform; · with full goods trade liberalization, the share of global production of farm products that is exported would rise from 8 to 13 percent, excluding intra-EU trade, thereby "thickening" international food markets and reducing the insta- bility of prices and quantities traded in those markets; · unskilled workers in developing countries--the majority of whom work on farms--would benefit most from reform, followed by skilled workers and then capital owners; and · net farm incomes in developing countries would rise by 5.6 percent, compared with 1.9 percent for nonagricultural value added, even more than the esti- mated gain from the partial reforms of the past quarter century. Together, those last two findings suggest both inequality and poverty could be alleviated by such reform, given that three-quarters of the world's poor are farmers in developing countries (Chen and Ravallion 2007). To get a more pre- cise sense of whether that is so, and to explore the extent to which it is own- country as distinct from rest-of world policies that are doing the harm, requires country case studies using national economy-wide models that are General Equilibrium Effects of Price Distortions on Global Markets 557 enhanced with detailed earning and spending information of numerous types of urban and rural households. One set of such studies for a dozen or so coun- tries across three continents is forthcoming in Anderson, Cockburn, and Martin (forthcoming). As for farmers in high-income countries, removal of agricultural price- supporting policies would undoubtedly lead to painful reductions in income and wealth if farmers are not compensated, though compensation by the gainers in other sectors could readily afford to compensate them from the benefits of freeing trade. The distortion estimates in table 13.1 show that all high-income countries have lowered the price supports for their farmers since the 1980s. In some coun- tries, that has been partly replaced by assistance that is somewhat decoupled from production (not shown in the table). If that trend continues at the pace of the past quarter century, and if there is no growth of agricultural protection in developing countries, then before the middle of this century most of the disarray in world food markets that so worried D. Gale Johnson through much of his academic life may be removed. However, if the WTO's Doha Development Agenda collapses, and governments thereby find it more difficult to ward off agricultural protection lobbies, there is the possibility of the recent progress being reversed in high- income and of emerging economies--even those as poor as India--following the same agricultural protection path this century as the one Anderson and Hayami (1986) show was taken by high-income countries in the last century. 558 Annex Table 13.A.1. Protection Structurea in GTAP Version 7 Prerelease and in the Distortion Rates Drawn from the World Bank Project, 2004 (percent) GTAP database version 7 prerelease Amended rates Primary Agriculture and Other Primary Agriculture and Other agriculture lightly processed food goods Agriculture lightly processed food goods Domestic Export Domestic Export support subsidy Tariff Tariff support subsidy Tariff Tariff Eastern Europe and Central Asia Russian Federation 1.7 0.1 7.5 7.4 1.7 0.9 18.9 7.4 Kazakhstan 0.9 0.0 2.9 2.7 0.9 0.0 3.4 2.7 Kyrgyzstan 1.0 0.1 3.1 5.0 1.0 0.1 3.8 5.0 Turkey 0.8 0.0 29.0 3.1 0.8 0.0 33.3 3.1 Rest of Eastern Europe and Central Asia 1.1 0.0 9.8 5.7 1.1 0.9 9.9 5.7 Bulgaria 0.6 0.0 17.0 11.5 0.6 0.0 14.8 11.5 Czech Republic 0.6 10.2 3.1 0.5 0.6 0.0 3.0 0.5 Estonia 0.0 9.7 6.2 0.9 0.0 0.0 5.0 0.9 Hungary 3.1 9.7 6.6 0.5 3.1 0.0 6.2 0.5 Latvia 13.1 9.9 3.7 0.9 13.3 0.0 3.3 0.9 Lithuania 0.5 9.4 13.1 1.0 0.5 0.0 12.1 1.0 Poland 0.4 8.3 6.1 0.8 0.4 0.0 6.2 0.8 Romania 1.3 0.0 19.8 9.8 1.3 0.0 18.0 9.8 Slovak Republic 0.0 10.4 5.5 0.4 0.0 0.0 5.2 0.4 Slovenia 0.0 10.5 6.3 0.4 0.0 0.0 7.8 0.4 East and South Asia Korea, Rep. of 0.0 0.0 172.7 5.9 0.0 0.0 319.4 5.9 Taiwan, China 0.4 0.0 77.4 3.9 0.4 0.0 84.2 3.9 China 0.0 0.0 12.6 7.1 0.0 0.2 6.5 7.1 Indonesia 0.0 0.0 6.4 4.9 0.0 1.6 7.3 4.9 Malaysia 0.0 0.0 2.4 5.9 0.0 0.2 5.0 5.9 Philippines 4.7 0.0 20.0 3.4 4.7 0.0 7.1 3.4 Thailand 0.2 0.0 22.1 12.9 0.2 0.0 26.2 12.9 Vietnam 3.6 0.0 15.5 18.5 3.6 0.5 21.5 18.5 Bangladesh 1.0 0.0 16.3 22.5 1.0 0.0 9.9 22.5 India 3.9 0.0 29.8 20.9 10.1 2.5 2.9 20.8 Pakistan 0.0 0.0 10.8 18.5 0.0 0.2 19.4 18.5 Sri Lanka 0.6 0.2 24.3 5.8 0.6 0.3 23.8 5.8 Rest of South Asia 0.5 0.0 5.0 15.6 0.5 0.0 6.9 15.6 Rest of East Asia 0.7 0.0 2.8 2.3 0.7 0.0 3.2 2.3 Middle East 12.4 0.0 9.0 5.7 12.4 0.0 7.5 5.7 North and Sub-Saharan Africa Egypt, Arab Rep. of 0.0 0.0 4.0 13.5 0.0 0.0 5.0 13.5 Morocco 0.0 0.3 33.3 20.0 0.0 0.4 28.4 20.0 Rest of North Africa 3.9 0.5 24.9 13.1 3.9 1.3 30.7 13.1 Madagascar 0.0 0.0 3.9 2.7 0.0 4.4 3.4 2.7 Mozambique 0.2 0.0 12.5 10.9 0.2 0.0 14.5 10.9 Nigeria 0.1 0.0 74.0 17.2 0.1 0.0 76.1 17.2 Senegal 0.0 0.0 8.4 8.9 0.0 1.1 6.2 8.9 South Africa 0.0 0.0 9.7 6.5 0.0 0.0 10.2 6.5 Tanzania 0.3 0.0 11.6 13.7 0.3 0.0 11.8 13.7 Uganda 0.0 0.0 9.5 5.5 0.0 2.6 9.2 5.5 Zambia 0.8 0.0 5.6 9.0 0.8 0.0 7.0 9.0 Zimbabwe 3.2 0.0 13.6 15.4 3.2 0.0 8.9 15.4 Rest of Western and 559 Central Africa 0.2 0.0 10.5 8.9 0.2 0.0 10.8 8.9 Rest of Africa 0.4 0.0 10.4 14.1 0.4 0.0 10.6 14.1 Table 13.A.1. Protection Structurea in GTAP Version 7 Prerelease and in the Distortion Rates Drawn from the 560 World Bank Project, 2004 (continued) (percent) GTAP database version 7 prerelease Amended rates Primary Agriculture and Other Primary Agriculture and Other agriculture lightly processed food goods Agriculture lightly processed food goods Domestic Export Domestic Export support subsidy Tariff Tariff support subsidy Tariff Tariff Latin America and the Caribbean Argentina 4.9 0.0 2.9 5.7 0.0 14.8 0.0 5.8 Brazil 0.0 0.0 4.5 8.9 0.0 0.0 4.8 8.9 Chile 1.7 0.0 1.3 1.8 0.0 0.0 2.4 1.8 Colombia 0.0 0.0 12.9 9.8 0.0 0.0 21.6 9.8 Ecuador 0.0 0.0 6.8 10.4 0.0 0.0 13.4 10.4 Mexico 1.3 0.0 8.6 3.4 1.2 0.0 6.2 3.4 Nicaragua 0.0 0.0 8.0 3.9 0.0 2.8 9.6 3.9 Rest of Latin America and the Caribbean 1.7 0.6 9.8 9.9 1.7 0.3 9.9 9.9 High-income countries Australia 0.0 0.0 0.7 3.3 0.0 0.0 0.5 3.3 New Zealand 0.0 0.0 2.8 3.3 0.0 0.2 0.7 3.3 EU-15 1.0 10.8 7.1 0.7 1.2 12.8 6.9 0.7 Rest of Western Europe 2.6 8.6 52.9 2.2 2.6 13.4 53.9 2.2 Canada 1.6 2.0 23.1 1.4 1.6 3.6 18.9 1.4 United States 4.0 0.5 2.5 1.3 5.2 0.6 6.1 1.3 Japan 2.0 0.0 141.1 1.7 2.0 0.0 151.7 1.7 Other high-income countries 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Source: Valenzuela and Anderson (2009), drawing on the authors' calculations from the GTAP version 7 prerelease and estimates drawn from Anderson and Valenzuela (2008). a. Using value of production at undistorted prices as weights. General Equilibrium Effects of Price Distortions on Global Markets 561 Notes 1. That distortions database is documented fully in Anderson and Valenzuela (2008) and is based on the methodology described in Anderson et al. (2008a, 2008b). Its implications for 1980­84 and 2004 for the countries and sectors in the GTAP model have been included in a recent research memo- randum for users of the GTAP and other databases of the global economy (Valenzuela and Anderson 2008). 2. Krueger, Schiff, and Valdés (1988), like Jensen, Robinson, and Tarp (2002), focus on effects of just own-country policies, the first using partial equilibrium and the second using national general equilibrium models. On the relationship between those two methodologies, see Bautista, Robinson, Wobst, and Tarp (2001). 3. Some of the questions raised here were addressed (but using the version 6 of the GTAP protec- tion database) by Anderson, Martin, and van der Mensbrugghe (2006a), who use the same Linkage model as in the present analysis, and by Anderson and Valenzuela (2007a) using the GTAP-AGR model. 4. This is reflected in the results emerging from attempts to include services distortions in trade reform modeling, which have led to widely differing results. Compare, for example, Brown, Deardorff, and Stern (2003); Francois, van Meijl, and van Tongeren (2005); and Hertel and Keeney (2006). 5. Version 7 is fully documented in Narayanan and Walmsley (2008), or go to http://www.gtap.org. 6. No new distortion estimates are available for countries in the Middle East, so in what follows little attention is given to this small and relatively affluent part of the global agricultural economy. 7. More information on the MAcMaps database is available in Bouët et al. (2008) and at http://www.cepii.fr/anglaisgraph/bdd/macmap.htm. For details of its incorporation into the GTAP version 7 dataset, see Narayanan and Walmsley (2008). 8. The bilateral tariff structure for 1980­84 is also preserved, so as to capture the changes only in average distortions for each product and not changes in the bilateral preferential contributions to those averages. 9. Using the GTAP version 6 database for 2001, Anderson, Martin, and Valenzuela (2006) find that agricultural production and export subsidies together contributed just 7 percent of the global welfare cost of agricultural protection. 10. The size of the Armington elasticities matters, see Zhang (2006, 2008) and Valenzuela, Ander- son, and Hertel (2008). The Linkage model assumes larger values than some other models because it is seeking to estimate long-run consequences of liberalization. An example of the difference this can make to the results is detailed in table 12A.2 in Anderson and Martin (2006). 11. The more affluent economies of Hong Kong, China and Singapore are in the World Bank's high-income category, but since they have close to free trade policies and almost no farm production anyway, their influence on the results is not noticeable. 12. In addition, the model does not include any divergences between private and social marginal costs and benefits that might arise from externalities, market failures, and other behind-the-border policies not represented in the amended GTAP protection database. These omissions could affect the welfare estimates in either direction. Also missing are any costs of adjustment to reform, since Linkage is not a dynamic model. 13. The only other policy change is the removal of export taxes on nonfarm products in Argentina This is done because they were introduced at the same time (end-2001) and for the same reason (for the government to gain popular support from the urban poor) as were the country's export taxes on farm products. References Anderson, K., J. Cockburn, and W. Martin, eds. Forthcoming. Agricultural Price Distortions, Inequality and Poverty. Anderson, K., and Y. Hayami, eds. 1986. The Political Economy of Agricultural Protection: East Asia in International Perspective. Boston, London, and Sydney: Allen and Unwin. 562 Distortions to Agricultural Incentives: A Global Perspective Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC (and appendix A of this volume). ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Anderson, K., and W. Martin, 2006. Agricultural Trade Reform and the Doha Development Agenda. Lon- don: Palgrave Macmillan and Washington, DC: World Bank. Anderson, K., W. Martin, and D. van der Mensbrugghe. 2006a. "Distortions to World Trade: Impacts on Agricultural Markets and Incomes." Review of Agricultural Economics 28 (2): 168­94. ______. 2006b. "Market and Welfare Implications of the Doha Reform Scenarios." In Agricultural Trade Reform and the Doha Development Agenda, ed. K. Anderson and W. Martin. London: Palgrave Macmillan and Washington, DC: World Bank. Anderson, K., W. Martin, and E. Valenzuela. 2006. "The Relative Importance of Global Agricultural Subsidies and Market Access." World Trade Review 5 (3): 357­76. Anderson, K., and E. Valenzuela. 2007a. "Do Global Trade Distortions Still Harm Developing Country Farmers?" Review of World Economics 143 (1): 108­39. ______. 2007b. "The World Trade Organization's Doha Cotton Initiative: A Tale of Two Issues." The World Economy 30 (8): 1281­304. ______. 2008. "Estimates of Global Distortions to Agricultural Incentives, 1955 to 2007." World Bank, Washington DC, available at http://www.worldbank.org/agdistortions. Bautista, R. M., S. Robinson, P. Wobst, and F. Tarp. 2001. "Policy Bias and Agriculture: Partial and General Equilibrium Measures." Review of Development Economics 5: 89­104. Bouët, A., Y. Decreux, L. Fontagné, S. Jean, and D. Laborde. 2008. "Assessing Applied Protection Across the World." Review of International Economics 16 (5): 850­63. Brown, D. K., A. V. Deardorff, and R. M. Stern. 2003. "Multilateral, Regional and Bilateral Trade-Policy Options for the United States and Japan." The World Economy 26: 803­28. Chen, S., and M. Ravallion. 2007. "Absolute Poverty Measures for the Developing World, 1981­2004." Policy Research Working Paper 4211, World Bank, Washington, DC. Francois, J. F., H. van Meijl, and F. van Tongeren. 2005. "Trade Liberalization in the Doha Development Round." Economic Policy 20: 349­91. Hertel, T., ed. 1997. Global Trade Analysis: Modeling and Applications. Cambridge and New York: Cam- bridge University Press. Hertel, T. W., and R. Keeney. 2006. "What's at Stake: The Relative Importance of Import Barriers, Export Subsidies and Domestic Support." In Agricultural Trade Reform and the Doha Development Agenda, ed. K. Anderson and W. Martin. London: Palgrave Macmillan and Washington, DC: World Bank. Jensen, H. T., S. Robinson, and F. Tarp. 2002. "General Equilibrium Measures of Agricultural Policy Bias in Fifteen Developing Countries." TMD Discussion Paper 105, International Food Policy Research Institute, Washington, DC. Johnson, D. G. 1991. World Agriculture in Disarray (revised edition). London: St. Martin's Press. Krueger, A. O., M. Schiff, and A. Valdés. 1988. "Agricultural Incentives in Developing Countries: Meas- uring the Effect of Sectoral and Economy-wide Policies." World Bank Economic Review 2 (3): 255­72. Malcolm, G. 1998. "Adjusting Tax Rates in the GTAP Data Base." GTAP Technical Paper 12, Purdue University, West Lafayette, IN. https://www.gtap.agecon.purdue.edu/resources/download/580.pdf. Melitz, M. 2003. "The Impact of Trade on Intra-industry Reallocations and Aggregate Industry Pro- ductivity." Econometrica 71 (6): 1692­725. Narayanan, G., and T. L. Walmsley, eds. 2008. Global Trade, Assistance, and Production: The GTAP 7 Data Base. West Lafayette, IN: Center for Global Trade Analysis, Purdue University. http://www.gtap.org. Valenzuela, E., and K. Anderson. 2008. "Alternative Agricultural Price Distortions for CGE Analysis of Developing Countries, 2004 and 1980­84." Research Memorandum 13, Center for Global Trade Analysis, Purdue University, West Lafayette, IN. https://www.gtap.agecon.purdue.edu/resources/ res_display.asp?RecordID=2925. General Equilibrium Effects of Price Distortions on Global Markets 563 Valenzuela, E., K. Anderson, and T. Hertel. 2008. "Impacts of Trade Reform: Sensitivity of Model Results to Key Assumptions." International Economics and Economic Policy 4 (4): 395­420. Valenzuela, E., D. van der Mensbrugghe, and K. Anderson. 2008. "General Equilibrium Effects of Price Distortions on Global Markets, Farm Incomes and Welfare." Agricultural Distortions Working Paper 73, World Bank, Washington, DC. van der Mensbrugghe, D. 2005. "Linkage Technical Reference Document: Version 6.0." Unpublished paper, World Bank, Washington, DC. http://www.worldbank.org/prospects/linkagemodel. ______. 2006. "Estimating the Benefits: Why Numbers Change." In Trade, Doha and Development: A Window into the Issues, ed. R. Newfarmer, 59­75. Washington, DC: World Bank. World Bank. 2002. Global Economic Prospects and the Developing Countries, 2002: Making Trade Work for the Poor. Washington, DC: World Bank. ______. 2004. Global Economic Prospects: Realizing the Development Promise of the Doha Agenda. Washington, DC: World Bank. ______. 2005. Global Economic Prospects: Trade, Regionalism, and Development. Washington, DC: World Bank. ______. 2006. Global Economic Prospects: Economic Implications of Remittances and Migration. Washington, DC: World Bank. Zhang, X. G. 2006. "Armington Elasticities and Terms of Trade Effects in Global CGE Models." Staff Working Paper, Productivity Commission, Melbourne. ______. 2008. "The Armington General Equilibrium Model: Properties, Implications and Alterna- tives." Staff Working Paper, Productivity Commission, Melbourne. Appendix A Methodology for Measuring Distortions to Agricultural Incentives Kym Anderson, Marianne Kurzweil, Will Martin, Damiano Sandri, and Ernesto Valenzuela* This appendix outlines the methodological issues associated with the task of measuring the impact of government policies on incentives faced by farmers and food consumers. The focus is on border and domestic measures that are due exclusively to governments' actions, and as such can be altered by a political deci- sion and have an immediate effect on consumer choices, producer resource allo- cation, and net farm incomes. Most commonly, these measures are in the form of import or export taxes, subsidies, and quantitative restrictions, supplemented by domestic taxes or subsidies for farm outputs or inputs and consumer subsidies for food staples. The incentives faced by farmers are affected not only by direct pro- tection or taxation of primary agricultural industries but also indirectly via poli- cies assisting nonagricultural industries, since the latter can have an offsetting effect by drawing resources away from farming. *The authors are thankful for invaluable comments from many project participants, including Bruce Gardner, Tim Josling, Will Masters, Alan Matthews, Johan Swinnen, Alberto Valdés, and Alex Winter-Nelson, plus Ibrahim Elbadawi. The text following was first circulated as Anderson et al. (2008a), and much of the material was first pub- lished as part of Anderson et al. (2008b). 565 566 Distortions to Agricultural Incentives: A Global Perspective What Theory Suggests Should Be Measured Three key purposes of the distortion estimates generated by this project are: · To provide a long annual time series of indicators showing the extent to which price incentives faced by farmers and food consumers have been distorted directly and indirectly by own-government policies in all major developing, transition, and high-income countries, and hence for the world as a whole (taking international prices as given); · Insofar as there are also forms of assistance provided to farmers that are decou- pled from their production decisions, to show the contribution of this addi- tional assistance as a transfer to farm households; and · At least for the most recent period, to attribute the price distortion estimates for each farm product to specific border or domestic policy measures, so that they can serve as inputs into various types of partial and general equilibrium economic models for estimating the effects of those various policies on indica- tors such as national and international agricultural markets; farm value added; income inequality; poverty; and national, regional, and global welfare. The first objective, of providing a long time series for a wide range of countries at different stages of development and hence with different complexities and quali- ties of data, requires that the indicators be simple. Simple indicators also make subsequent updates of the data easier. The third purpose, of making indicators useful for modelers seeking to distinguish market and household welfare effects, requires that distortion estimates also be provided for at least lightly processed foods.1 The Bhagwati (1971) and Corden (1997) concept of a market policy distortion as something that governments impose to create a gap between the marginal social return to a seller and the marginal social cost to a buyer in a transaction is followed in this project. Such a distortion creates an economic cost to society that can be estimated using welfare-measuring techniques such as those pioneered by Harberger (1971). As Harberger notes, this focus allows for great simplification in evaluating the marginal costs of a set of distortions: changes in economic costs can be evaluated taking into account the changes in volumes directly affected by such distortions, ignoring all other changes in prices. In the absence of diver- gences such as externalities, the measure of a distortion is the gap between the price paid and the price received, irrespective of whether the level of these prices is affected by the distortion. Other developments that change incentives facing producers and consumers can include flow-on consequences of the distortion, though these should not be Methodology for Measuring Distortions to Agricultural Incentives 567 confused with the direct price distortion that this project aims to estimate. If, for instance, a country is a major player in world trade for a given commodity, impo- sition of an export tax may raise the price in international markets, reducing the adverse impact of the distortion on producers in the taxing country. Another flow-on consequence is the effect of trade distortions on the real exchange rate, which is the price of traded goods relative to nontraded goods. Neither of these flow-on effects are of immediate concern, however, because if the direct distor- tions are accurately estimated, they can be incorporated as price wedges into an appropriate country or global economy-wide computable general equilibrium (CGE) model, which in turn will capture the full general equilibrium impacts (inclusive of real exchange rate effects) of the various direct distortions to pro- ducer and consumer prices. Importantly, the total effect of distortions on the agricultural sector will depend not just on the size of the direct agricultural policy measures, but also on the magnitude of distortions generated by direct policy measures altering incen- tives in nonagricultural sectors. It is relative prices, and hence relative rates of gov- ernment assistance, that affect producers' incentives. In a two-sector model, an import tax has the same effect on the export sector as an export tax (see the Lerner [1936] Symmetry Theorem). This carries over to a model that has many sectors, and is unaffected if there is imperfect competition domestically or internationally or if some of those sectors produce only nontradables (Vousden 1990). The Sym- metry Theorem is therefore also relevant for considering distortions within the agricultural sector. In particular, if import-competing farm industries are pro- tected, for example via import tariffs, this has similar effects on incentives to pro- duce exportables as does an explicit tax on agricultural exports; if both measures are in place, this is a double imposition on farm exporters. Direct agricultural distortions Consider a small, open, perfectly competitive national economy with many firms producing a homogeneous farm product with just primary factors. In the absence of externalities, processing, producer-to-consumer wholesale and retail marketing margins, exchange rate distortions, and domestic and international trading costs, that country would maximize national economic welfare by allowing both the domestic farm product price and the consumer price of that product to equal E times P, where E is the domestic currency price of foreign exchange and P is the foreign currency price of this identical product in the international market. That is, any government-imposed diversion from that equality, in the absence of any market failures or externalities, would be welfare reducing for that small economy. 568 Distortions to Agricultural Incentives: A Global Perspective Price-distorting trade measures at the national border The most common trade distortion is an ad valorem tax on competing imports (usually called a tariff), tm. Such a tariff on imports is the equivalent of a produc- tion subsidy and a consumption tax at rate tm. If that tariff on the imported pri- mary agricultural product is the only distortion, its effect on producer incentives can be measured as the nominal rate of assistance (NRA) to farm output con- ferred by border price support (NRABS), which is the unit value of production at the distorted price less its value at the undistorted free market price expressed as a fraction of the undistorted price:2 E P(1 tm) E P NRABS tm (A.1) E P The effect of that import tariff on consumer incentives in this simple economy is to generate a consumer tax equivalent (CTE) on the agricultural product for final consumers: CTE tm (A.2) The effects of an import subsidy are identical to those in equations A.1 and A.2 for an import tax, but tm in that case would have a negative value. Governments sometimes also intervene with an export subsidy sx (or an export tax, in which case sx would be negative). If that were the only intervention, NRABS CTE sx , (A.3) If any of these trade taxes or subsidies were specific rather than ad valorem (for example, $y/kilogram rather than z percent), the ad valorem equivalent can be calculated using slight modifications of equations A.1, A.2, and A.3. Domestic producer and consumer price-distorting measures Governments sometimes intervene with a direct production subsidy for farmers, sf (or production tax, in which case sf is negative, including via informal taxes in kind by local and provincial governments). In that case, if this is the only distor- tion present, the effect on producer incentives can be measured as the NRA to farm output conferred by domestic price support (NRADS), which is the same as above except sf replaces tm or sx, though the CTE in that case is zero. Similarly, if the government imposes only a consumption tax, cc , on this product (or con- sumption subsidy, in which case cc is negative), the CTE is as above except cc replaces tm or sx, but the NRADS in that case is zero. The combination of domestic measures and border price support provides the following total rate of assistance to output, NRAo, and total CTE: NRAo NRABS NRADS (A.4a) CTE NRABS ct (A.4b) Methodology for Measuring Distortions to Agricultural Incentives 569 What if the exchange rate system also distorts prices? When a multitier foreign exchange rate regime is in place, another policy-induced price wedge exists. A simple two-tier exchange rate system creates a gap between the price received by all exporters and the price paid by all importers for foreign currency, changing both the exchange rate received by exporters and that paid by importers from the equilibrium rate E that would prevail without this distortion in the domestic market for foreign currency (Bhagwati 1978). Exchange rate overvaluation of the type considered here requires controls by the government on current account transfers. A common requirement is that exporters surrender their foreign currency earnings to the central bank for exchange to local currency at a low official rate. This is equivalent to a tax on exports to the extent that the official rate is below what the exchange rate would be in a market without government intervention. That implicit tax on exporters reduces the incentive to export, and hence reduces the supply of foreign currency flowing into the country. With less foreign currency, demanders are willing to bid up the purchase price. That provides a potential rent for the government, which can be realized by auctioning off the limited supply of foreign currency extracted from exporters or creating a legal secondary market. Either mechanism will create a gap between the official and parallel rates. Such a dual exchange rate system is depicted in figure A.1, in which is it assumed that the overall domestic price level is fixed, perhaps by holding the money supply constant (Dervis, de Melo, and Robinson 1981). The supply of for- eign exchange is given by the upward sloping schedule, Sfx , and demand by Dfx , where the official exchange rate facing exporters is E0 and the secondary market rate facing importers is Em. At the low rate, E0, only Qs units of foreign currency Figure A.1. A Distorted Domestic Market for Foreign Currency Sfx local currency per unit of foreign currency Em Em E Ex E0 Dfx QS Q S QE quantity Source: Martin (1993). See also Dervis, de Melo, and Robinson (1981); Kiguel and O'Connell (1995, 1997); and Shatz and Tarr (2000). 570 Distortions to Agricultural Incentives: A Global Perspective are available domestically, instead of the equilibrium volume QE that would result if exporters were able to exchange at the "equilibrium rate" E units of local cur- rency per unit of foreign currency.3 The gap between the official and the second- ary market exchange rates is an indication of the magnitude of the tax imposed on trade by the two-tier exchange rate: relative to the equilibrium rate E, the price of importables is raised by em E, which is equal to (Em E), while the price of exportables is reduced by ex E, which is equal to (E E0), where em and ex are the fractions by which the two-tier exchange rate system raises the domestic price of the importable and lowers the domestic price of the exportable, respectively. The estimated division of the total foreign exchange distortion between an implicit export tax, ex, and an implicit import tax, em, will depend on the esti- mated elasticities of supply of exports and of demand for imports.4 If the demand and supply curves in figure A.1 had the same slope, em ex and (em ex) is the secondary market premium or proportional rent extracted by the government or its agents.5 When a government chooses to allocate limited foreign currency to different groups of importers at different rates, it is pursuing a multiple exchange rate sys- tem. Some lucky importers may even be able to purchase foreign currency at the low official rate. The more currency that is allocated and sold to demanders whose marginal valuation is below Em, the greater the unsatisfied excess demand at Em and hence the stronger the incentive for an illegal (or "black") market to form, and for more scrupulous exporters to lobby the government to legalize the sec- ondary market for foreign exchange and to allow exporters to retain some fraction of their exchange rate earnings for sale in the secondary market. Providing such a right to exporters to retain and sell a portion of foreign exchange receipts increases their incentives to export, thereby reducing the shortage of foreign exchange and hence the secondary market exchange rate (Tarr 1990). In terms of figure A.1, the available supply increases from Qs to Q s , bringing down the sec- ondary rate from Em to E m such that the weighted average of the official rate and E m received by exporters is E x (the weights being the retention rate r and (1 r)). Again, if the demand and supply curves in figure A.1 had the same slope, then the implicit export and import taxes resulting from this regime would be each equal to half the secondary market premium. In the absence of a secondary market and with multiple rates for importers below Em and for exporters below E 0, a black market often emerges. The more the government sells its foreign currency to demanders whose marginal valuation is below Em and the more active is the government in catching and punishing exporters selling in that illegal market, the more the rate for buyers will be above E. If the black market is allowed to operate "frictionlessly," there would be no for- eign currency sales to the government at the official rate and the black market rate Methodology for Measuring Distortions to Agricultural Incentives 571 would fall to the equilibrium rate E. So even though in the latter case the observed premium would be positive (equal to the proportion by which E is above the nominal official rate E0), there would be no distortion. For present purposes, since the black market is not likely to be completely frictionless, it can be thought of as similar to the system involving a retention scheme. In terms of figure A.1, E m would be the black market rate for a proportion of sales and the weighted average of that and E0 would be the exporters' return. Calculating E x in this case (and hence being able to estimate the implicit export and import taxes associated with this regime) by using the same approach as in the case with no illegal market thus requires knowing not only E0 and the black market premium but also guessing the proportion, r, of sales in that black market. In short, where a country has distortions in its domestic market for foreign cur- rency, the exchange rate relevant for calculating the NRAo or CTE for a particular tradable product depends, in the case of a dual exchange rate system, on whether the product is an importable or an exportable, while in the case of multiple exchange rates it depends on the specific rate that applies to that product each year. What about real exchange rate changes? A change in the real exchange rate alters equally the prices of exportables and importables relative to the prices of nontradable goods and services. Such a change can arise for many reasons, including changes in the availability of capital inflows, macroeconomic policy adjustments, or changes in the international terms of trade. When the economy receives a windfall--such as a greater inflow of foreign exchange from remittances or foreign aid or a commodity boom--the community moves to a higher indifference curve (Collier and Gunning 1998). While net imports of tradables can change in response to this inflow of foreign exchange, the domestic supply of and demand for nontradables must balance. The equilibrating mechanism is the price of nontradables. The price of nontradables rises to bring forth the needed increase in the supply of nontradables, and to reduce the demand for these products to bring it into line with supply (Salter 1959). While this type of change in the real exchange rate affects the incentive to pro- duce tradables, it is quite different from distortions in the market for foreign cur- rency analyzed above, in two respects. First, this real exchange rate appreciation reduces the incentive to produce importables and exportables to the same degree. In contrast with the multiple-tier exchange rate case, that appreciation does not generate any change in the prices of exportables relative to importables. Second, most such changes do not involve direct economic distortions of the type measur- able using tools such as producer or consumer surplus. If the government, or the private sector, chooses to borrow more from abroad to increase domestic spend- ing, this may raise the real exchange rate, but such an outcome is not obviously a 572 Distortions to Agricultural Incentives: A Global Perspective distortion. Moreover, symmetric treatment of any such "overvaluation" during periods of high foreign borrowing would require taking into account exchange rate "undervaluation" during periods of low foreign borrowing or repayment of foreign debt. For these reasons, the Krueger, Schiff, and Valdés (1988) and Orden et al. (2007) methodology of including deviations of real exchange rates from benchmark values is not followed in this project, unless these deviations arise from direct exchange rate distortions such as multiple-tier exchange rates.6 What if trade costs are sufficiently high for the product to be not traded internationally? Suppose the transport costs of trading are sufficient to make it unprofitable for a product to be traded internationally, such that the domestic price fluctuates over time within the band created by the cost, insurance, and freight (cif) import price and the free on board (fob) export price. Any trade policy measure (tm or sx) or the product-specific exchange rate distortion (for example, em or ex) then becomes redundant. In that case, in the absence of other distortions, NRAo 0 and CTE 0. However, in the presence of any domestic producer or consumer tax or subsidy (sf or tc), the domestic prices faced by both producers and con- sumers will be affected. The extent of the impact depends on the price elasticities of domestic demand and supply for the nontradable (the standard closed- economy tax incidence issue). To give a specific example, suppose just a production tax is imposed on farmers producing a particular nontradable, so sf 0 and tc 0. In that case: sf NRADS (A.5) 1 and sf CTE , (A.6) 1 where is the price elasticity of supply and is the (negative of the) price elastic- ity of demand.7 What if farm production involves not just primary factors but also intermediate inputs? Where intermediate inputs are used in farm production, any taxes or subsidies on their production, consumption, or trade will alter farm value added and thereby also affect farmer incentives. Sometimes a government will have directly offsetting measures in place, such as a domestic subsidy for fertilizer use by farmers, while simultaneously maintaining a tariff on fertilizer imports. In other situations, there Methodology for Measuring Distortions to Agricultural Incentives 573 will be farm input subsidies but an export tax on the final product.8 In principle, all these distortions could be brought together to calculate an effective rate of direct assistance (ERA) to farm value added. The nominal rate of direct assistance to farm output, NRAo, is a component of that the ERA, as is the sum of the nomi- nal rates of direct assistance to all farm inputs, or NRAi. In principle, all three rates can be positive or negative. ERAs were not estimated in this project because to do so requires knowing each product's value added share of output. Such data are not available for most developing countries even every few years, let alone for every year in the time series. In most developing countries, distortions to farm inputs are also very small compared to distortions to farm output prices, and purchased inputs are a small fraction of the value of output. But where there are significant distortions to input costs, the ad valorem equivalent of various distortions is accounted for by sum- ming each input's NRA times its input-output coefficient to obtain the combined NRAi, and then adding that to the farm industry's nominal rate of direct assis- tance to farm output, NRAo , to get the total NRA to farm production, called sim- ply NRA: 9 NRA NRAo NRAi. (A.7) What about post-farmgate costs? If a state trading corporation charges excessively for its marketing services, thereby lowering the farmgate price of a product (for example, as a way of raising government revenue in place of an explicit tax), the extent of that excess should be treated as if it is a tax. Some farm products, including some that are not internationally traded, are inputs into a processing industry that may also be subject to government inter- ventions. In that case, the effect of those interventions on the price received by farmers for the primary product also needs to be taken into account. Before explaining how, it is helpful to first to review the role that the value chain's mar- keting and distribution margins can play in the calculation of distortions to pri- mary agricultural activities, so as to ensure nondistortionary price wedges are not inadvertently included in any distortion calculation. Nondistortionary price wedges Thus far, it has been assumed there are no divergences between farmer, processor and wholesaler, consumer, and border prices other than because of subsidies or taxes on production, consumption, trade, or foreign currency. In practice, of course, this is not the case, and these costly value chain activities need to be explic- itly recognized and netted out when using comparisons of domestic and border 574 Distortions to Agricultural Incentives: A Global Perspective prices to derive estimates of government policy induced distortions.10 Such recog- nition also offers the opportunity to compare the size of the NRA with wedges associated with things such as trade and processing costs (used in trade facilita- tion and value chain analyses, respectively). It may also expose short-term situa- tions where profits of importers or exporters are amplified by less-than-complete adjustment by agents in the domestic value chain. Domestic trading costs Trading costs can be nontrivial both intranationally and internationally, especially in developing countries with poorly developed infrastructure.11 For example, domestic trading costs are involved in getting farm products to a port or to a domestic wholesaler (assuming the latter are at the international border-- otherwise another set of domestic transport costs needs to be added to obtain a relevant price comparison). Suppose domestic transport costs are equal to the fraction Tf of the price received by the farmer. Processor and wholesaler costs Domestic processing costs and wholesale and retail distribution margins can rep- resent a large share of the final retail price. Indeed, Reardon and Timmer (2009) argue that they are becoming an increasingly important part of the value chain in developing countries as consumers desire that ever more postfarm processing and services be added to their farm products, aided by the supermarket revolution's contribution to globalization.12 Here, the increases in the consumer price due to the processing and wholesaling activities are denoted as mp and mu, respectively, above the farmgate price plus domestic trade cost (or just mu above the price of the imported processed product, if the processing must be done prior to the prod- uct being internationally tradable) in the absence of market imperfections or gov- ernment distortions along the value chain. International trading costs International trading costs are not an issue in distortion calculations if the inter- national price used is the cif import unit value for an importable or the fob export unit value for an exportable. But they are relevant if there is no trade (because of, say, a prohibitive trade tax on the product) or if border prices are unrepresentative of actual prices (because of low trade volumes, for example). In those instances, it is recommended that an international indicator price series (such as from the World Bank or International Monetary Fund) be used to account for interna- tional trading costs (ocean or air freight, insurance, etc.).13 Tm denotes the pro- portion by which the domestic price of the import-competing product is raised above what it otherwise would be at the country's border, or equivalently that the price abroad of the exported product is greater by a fraction Tx of the fob price. Methodology for Measuring Distortions to Agricultural Incentives 575 Product quality and variety differences The quality of a product traded internationally is usually considered to be differ- ent from that of the domestically sold substitute, with consumers typically having a home-country bias.14 When appropriate, the domestic price should be deflated (inflated) by the extent to which the good imported is deemed by domestic con- sumers to be inferior (superior) in quality to the domestic product.15 Thus, qm is denoted as the deflating fraction to adjust for product quality and variety differ- ences in the case of importables. Similarly, for exported goods, and especially if an international indicator price is used in lieu of the fob export unit value (for exam- ple, when exports are close to zero and unrepresentative), the international price must be deflated (inflated) by the extent to which the good is deemed by foreign consumers to be inferior (superior) in quality relative to the indicator good. Here, qx denotes the deflating fraction to adjust for product quality and variety differ- ences in the case of exportables. Net effect of nondistortionary influences With all these influences, and so long as the product is still traded internationally, the relationship between the domestic farmers' price and the international price in the absence of government-imposed price and trade policies becomes the fol- lowing for an importable: Pf (1 Tf)(1 mp)(1 qm) E P . (A.8) 1 Tm For an exportable, it becomes: Pf (1 Tf)(1 mp)(1 Tx) E P . (A.9) 1 qx In both of these cases, the urban consumer price is above the producer price to the following extent: Pc = Pf (1 Tf)(1 mp)(1 mu), (A.10) where Pf is the farmgate price. Impact of distortions to food processing on agricultural NRAs Some farm products that are not internationally traded in their primary form (for example, raw milk and cane sugar) are tradable once lightly processed, and the downstream processing industry may also be subject to government interven- tions. In that case, the effect of the latter interventions on the price received by farmers for the primary product also needs to be taken into account, and that pri- mary product should be classified as tradable. 576 Distortions to Agricultural Incentives: A Global Perspective In the past, some analysts assumed any protection to processors is fully passed back to primary agriculture (as may be the case with a farmer-owned cooperative processing plant, for example). Such a situation effectively raises the farmers' price by the rise in the processors' price divided by the proportional contribution of the primary product to the value of the processed product. Another equally extreme but opposite assumption is zero pass-through by the processor back down the value chain to the farmer. That is likely to be the case if the raw material can be sourced internationally, but seems unlikely if the primary product is non- tradable and there is a positive price elasticity of farm supply (since an assisted processor would want to expand). A more neutral assumption is proportional pass-through by the processor either down the value chain to farmers and their transporters or up the value chain to consumers. This last assumption is equivalent to an equal sharing of the benefits along the value chain, which is more likely to be the case the more equally market power is spread among the players in the chain. This trio of examples illustrates the importance both of separating the primary and processed activities for the purpose of calculating agricultural assistance rates, and of being explicit about the extent of pass-through that is occurring in practice, and hence its consequences for the NRAs in both primary agricultural and processing activities.16 The above examples involving processors also can be generalized to any partic- ipants in the value chain. In particular, state trading enterprises and parastatal marketing boards may well intervene significantly, especially if they have been granted monopoly status by the government. Such domestic institutions may explain the econometrically estimated low degree of transmission of price changes at a border to farmgate domestic prices--even after significant reform of more-explicit price and trade policies (see Baffes and Gardner 2003, Warr 2008, and the references cited therein). Where reform also involves freeing up previously controlled parts of the marketing chain, the reduced marketing margin can pro- vide a benchmark against which to compare the prereform margin (as in Uganda from the mid-1990s, see Matthews, Claquin, and Opolot 2009). The mean and standard deviation of agricultural NRAs Only when a weighted average NRA for covered products for each country is gen- erated can the NRA for noncovered products be summed to obtain the NRA for all agriculture. When it comes to averaging across countries, here, each polity is an observation of interest, so a simple average is meaningful for the purpose of polit- ical economy analysis. But if one wants a sense of how distorted agriculture is in a whole region, a weighted average is needed. The weighted average NRA for cov- ered primary agriculture can be generated by multiplying each primary industry's Methodology for Measuring Distortions to Agricultural Incentives 577 value share of production (valued at the farmgate equivalent undistorted prices) by its corresponding NRA and adding across industries.17 The overall sectoral rate, denoted as NRAag, can be obtained by adding the actual or assumed infor- mation for the noncovered commodities and, where it exists, the aggregate value of non-product-specific (NPS) assistance to agriculture. A weighted average can be similarly generated for the tradables part of agriculture--including for industries producing products such as milk and sugar that require only light processing before they can be traded--by assuming all NPS assistance goes to tradables. That is denoted as NRAag t. In addition to the mean, it is important to provide a measure of the dispersion or variability of the NRA estimates across the covered products. The cost of gov- ernment policy distortions to incentives in terms of resource misallocation tend to be greater the greater the degree of substitution in production (Lloyd 1974). In the case of agriculture, which involves the use of farmland that is sector-specific but transferable among farm activities, greater variation of NRAs across indus- tries within the sector will result in higher welfare cost of those market interven- tions. A simple indicator of dispersion is the standard deviation of industry NRAs within agriculture.18 Trade bias in agricultural assistance A trade bias index (TBI) also is needed, to indicate the changing extent to which a country's policy regime has an antitrade bias within the agricultural sector. This is important because, as mentioned above, the Lerner (1936) Symmetry Theorem demonstrates that a tariff assisting import-competing farm industries has the same effect on farmers' incentives as if there were a tax on agricultural exports; if both measures are in place, this is a double imposition on farm exports. The higher the NRA to import-competing agricultural production (NRAagm) relative to that for exportable farm activities (NRAagx), the more incentive producers in that subsector will have to bid for mobile resources that would otherwise have been employed in export agriculture, other things equal. Once each farm industry is classified as import-competing, or a producer of exportables, or as producer of nontradables (though the status sometimes changes over the years--see the next section), it is possible to generate for each year the weighted average NRAs for each of the two groups of tradable farm industries. The weighted NRAs can then be used to generate an agricultural TBI, defined as: c 1 NRAagx TBI 1d, (A.11) 1 NRAagm where NRAagm and NRAagx are the average NRAs for the import-competing and exportable parts of the agricultural sector (their weighted average being NRAag t ). 578 Distortions to Agricultural Incentives: A Global Perspective This index has a value of zero when the import-competing and export subsectors are equally assisted, and its lower bound approaches 1 in the most extreme case of an antitrade policy bias. Indirect agricultural assistance and taxation via nonagricultural distortions In addition to direct assistance to or taxation of farmers, the Lerner (1936) Sym- metry Theorem further demonstrates that incentives are also affected indirectly by government assistance to nonagricultural production in an economy. The higher the NRA to nonagricultural production (NRAnonag), the more incentive producers in other sectors have to bid up the value of mobile resources that would otherwise have been employed in agriculture, other things equal. If NRAag is below NRAnonag, one might expect there to be fewer resources in agriculture than under free-market conditions in the country, notwithstanding any positive direct assistance to farmers, and, conversely, if NRAag NRAnonag. A weighted average can be generated for the tradables part of nonagriculture too, called NRAnonag t. One of the most important negative effects on farmers is protection from import competition for industrialists. Tariffs are part of that, but so too--espe- cially in past decades--are nontariff barriers to imports. Other primary sectors (fishing, forestry and minerals, and energy raw material extraction) on average tend to be subject to less direct distortions than either agriculture or manufactur- ing, but there are important exceptions. One example is a ban on logging, but if such a ban is for genuine natural resource conservation reasons it should be ignored. Another example is a resource rent tax on minerals. Unlike an export tax or quantitative restriction on exports of such raw materials (which are clearly dis- tortive and would need to be included in the NRA for mining), a resource rent tax, like a land tax, can be fairly benign in terms of resource reallocation (see Garnaut and Clunies-Ross 1983) and so can be ignored. The largest part of most economies is the services sector. It produces mostly nontradables, many of them by the public sector. Distortions in services markets have proven to be extraordinarily difficult to measure, and no systematic esti- mates across countries are available even for a recent period, let alone over time. The only feasible way forward in generating time series estimates of NRAnonag for this project is to assume all services are nontradable and that they, along with other nonagricultural nontradables, face no distortions. All the other nonagricul- tural products can be separated into exportables and import-competing products for estimating correctly their weighted average NRAs, ideally using production valued at border prices as weights (although in practice most country study authors had to use gross domestic product [GDP] shares). Methodology for Measuring Distortions to Agricultural Incentives 579 As already mentioned in the previous section on agriculture, foreign exchange rate misalignment relative to what fundamentals suggest is the value of a country's currency are ignored. This is because a real appreciation of the general foreign exchange rate uniformly lowers the price of all tradables relative to the price of nontradables, and conversely for a real devaluation. If a change in the exchange rate is caused by aid or foreign investment inflows, the excess of tradables con- sumption over tradables production leads to a new equilibrium. Certainly, such a new inflow of funds would reduce incentives for farmers producing tradable products, but this is not a welfare-reducing policy distortion. Thus, it is only the exchange rate distortions due to a dual or multiple exchange rate system that need to be included in the calculation of the NRAs for the exportable and import- competing parts of the nonagricultural sector, and hence of NRAnonag t, in the same way as discussed above for their inclusion in the calculation of NRAag t. Assistance to agricultural relative to nonagricultural production Given the calculation of NRAag t and NRAnonag t as above, it is then possible to calculate a relative rate of assistance, or RRA, defined as: NRAagt c 1 RRA 1d (A.12) 1 NRAnonagt Since an NRA cannot be less than 1 if producers are to earn anything, neither can the RRA. This measure is a useful indicator for providing international comparisons over time of the extent to which a country's policy regime has an antiagricultural or proagricultural bias. How Theory Is Put into Practice in This Study Making the above theory operational in the real world, where data are often scarce, especially over a long time period, is as much an art as a science.19 Thank- fully, the project did not have to start from scratch in many countries. NRAs are available from as early as 1955 in some cases, and at least from the mid-1960s or early or mid-1980s for the 18 countries included in the Krueger, Schiff, and Valdés (1988, 1991) and Anderson and Hayami (1986) studies. Much has been done to provide detailed estimates since 1986 of direct distortions to farmer (though not food processing) incentives in the high-income countries that are now members of the OECD, and (since the early or mid-1990s) in select European transition economies and Brazil, China, and South Africa (OECD 2007). In addition, at least for direct distortions, the Krueger, Schiff, and Valdés measures have been updated 580 Distortions to Agricultural Incentives: A Global Perspective to the mid-1990s for some Latin American countries (Valdés 1996) and also pro- vided for some European transition countries (Valdés 2000); a new set of esti- mates of simplified PSEs for a few key farm products for China, India, Indonesia, and Vietnam since 1985 is also now available from International Food Policy Research Institute (IFPRI; see Orden et al. 2007). Each of these studies uses varia- tions on the above methodology, but the basic price data at least, as well as the narratives attached to those estimates, are invaluable springboards for the present study.20 Time period coverage For European transition economies, it is difficult to obtain meaningful data prior to 1992. Estimates are not very meaningful before the 1980s for China and Vietnam for the same reason. For all other countries, the target start date is 1955, especially if that includes some preindependence years, as they show what a difference inde- pendence made, although for numerous developing countries the data simply are not available. The target finish date is 2004, though 2005 data are included where available. In most cases, the most recent few years offer the highest quality data. Farm product coverage Agricultural commodity coverage includes all major food items (rice, wheat, maize or other grains, soybeans or other temperate oilseeds, palm or other tropi- cal oils, sugar, beef, sheep or goat meat, pork, chicken and eggs, and milk) plus other key country-specific farm products (for example, other staples, tea, coffee or other tree crop products, tobacco, cotton, wine, and wool). Globally, as of 2001 (according to the Global Trade Analysis Project (GTAP) database, see Dimaranan 2006), one-third of the value added in all agriculture and food industries is highly processed food, beverages, and tobacco, which is handled in the same cursory way as for nonagricultural products. Fruit and vegetables are another one-sixth, while the remainder of products constitute the other half. Of that other half, meats are one-third, grains and oilseeds are almost another one-third, dairy products are one-sixth, and sugar, cotton, and other crops account for just over one-fifth. When high-income countries are excluded, those shares change quite a bit: highly processed food, beverages and tobacco are only half as important, fruits and veg- etables are somewhat more important and, when those two groups (which together account for 41 percent of the total) are excluded, the residual is equally divided between three groups: meats, grains and oilseeds, and other crops and dairy products. By focusing on all major grain, oilseed, and livestock products plus any key horticultural and other crop products, the coverage reaches the target Methodology for Measuring Distortions to Agricultural Incentives 581 of 70 percent of most countries' value added in agriculture and lightly processed foods. Priority is given to the most-distorted industries because the residual will have not only a low weight but also a low degree of distortion. In terms of household food expenditure, if highly processed food, beverages, and tobacco are excluded, fruits and vegetables account for almost one-quarter of spending in developing countries. When fruits and vegetables are also excluded, three groups each account for almost 30 percent of expenditure: pig and poultry products, red meat and dairy products, and grains and oilseed products. All other crops account for the remaining one-eighth. So from the consumer tax viewpoint, the desired product coverage is the same as suggested above from a production viewpoint. Each product is explicitly identified as import-competing, exporting, or non- tradable. For many products, that categorization changes over time, in some cases moving monotonically through those three categories and, in others, fluctuating in and out of nontradability. Hence, an indication of a product's net trade status is given for each year rather than just one categorization for the whole time series. For large-area countries with high internal and coastal shipping costs, some regions within the country may be exporting abroad even while other regions are net importers from other countries. In such cases, it is necessary to estimate sepa- rate NRAs for each region and then generate a national weighted average. Farm input coverage The range of input subsidies considered in any particular country study will depend on the degree of distortions in that country's input markets. In addition to fertilizer, other large input subsidies are likely to be electric or diesel power, pesti- cides, and credit (including occasional large-scale debt forgiveness, as in Brazil and Russia, although determining how that is spread beyond the year of forgive- ness is problematic).21 There are also distortions to water, but the task of measur- ing water subsidies is especially controversial and complex, so they are not included in the NRA calculations presented here (just as the OECD ignores them in its PSE calculations).22 Similarly, distortions to land and labor markets are excluded, apart from qualitative discussion in the analytical narrative of some country case studies. Trade costs For the international trading costs tm and Tx, the fob-cif gap in key bilateral trad- ing in the product in years when the product was traded in significant quantities is used. Both international and domestic trading costs are a function of both the 582 Distortions to Agricultural Incentives: A Global Perspective quality of hard infrastructure (roads, railways, and ports) and soft infrastructure (business regulations and customs clearance procedures at state and national bor- ders), each of which can be affected by government actions. But since it is difficult to allocate those costs between items that are avoidable and those that are unavoidable, measuring the aggregate size of the distortions involved in a compa- rable way for a range of countries is beyond the scope of this study.23 Classifying farm products as import-competing, exportable, or nontradable The criteria used in classifying farm products--import-competing (M), exportable (X), or nontradable (H)--are not straightforward. Apart from the complications raised above about whether a product is nontraded simply because of trade taxes or nontariff barriers, there will be cases where trade is minimal, or where the trade status is reversed because of the policy distortions, or the industry is characterized by significant imports and exports. A judgment must be made for each sector, for each year, as to whether the product should be classified as M, X, or H. In the case of the two tradable classifications, the choice of classification will determine which exchange rate distortion to use. If trade is minimal for trade cost rather than trade policy reasons, then it is classified as nontradable if the share of production exported and the share of consumption imported are each less than 2.5 percent--except in cases (for example, rice for China) where it is clearly an exportable year after year even though the self-sufficiency rate is rarely above 101 percent. Otherwise, where the share of production exported is substantially above (below) the share of consumption imported, the sector is classified as exportable (importable). In cases where the trade status has been reversed because of the policy distor- tion (for example, an export subsidy, in combination with a prohibitive import tariff, is sufficiently large as to encourage production enough to generate an export surplus), that product should be given the classification of the trade status that would prevail without the intervention (that is, import-competing). The same methodology applies where tariff preferences reverse a country's trade status for a product. Many countries enjoy preferential access for their exports into pro- tected markets of other countries. In some cases, these are bilateral or plurilateral free-trade agreements or customs unions. In other cases, they are unilaterally offered by high-income countries to developing countries under schemes such as the Generalized System of Preferences, the Cotonou Agreement (the now-expired agreement between European Union [EU] member countries and their former colonies), and the EU's Everything but Arms Agreement with least developed countries. In the few extreme cases where these preferences are such that they (in Methodology for Measuring Distortions to Agricultural Incentives 583 combination with a prohibitive import tariff) cause the developing country to become an exporter of a product that would otherwise be import-competing (for example, sugar in the Philippines), the product is nonetheless classified as import- competing, since it is this developing country's import-restrictive policy that allows the domestic price of the product to equal what is earned in exporting the product to the preference-providing country. In the case in which there are significant exports and imports of a product in a given year, closer scrutiny is required. If, for example, there are high credit or stor- age costs domestically, a product may be exported immediately following harvest but imported later in the year to satisfy consumers out of season. The product would be considered an exportable for the purposes of calculating the NRA, because even if there are policies restricting out-of-season imports (which would affect the CTE calculation), they would not increase that year's earlier production in the presence of high credit or storage costs. If trade and exchange rate distortions are sufficiently large to choke off inter- national trade in a product, then they contribute to the NRA and CTE only to the extent needed to drive trade in that product to zero: any trade taxes larger than that have an element of redundancy. Where there are trade policy distortions with no trade passing over them (that is, they are prohibitive), there may still be policy effects that need to be measured, though they will differ from those implied above. One example is where a prohibitive tariff that is high enough to force the price of imported goods above the autarky price results in no imports. In that case, the NRA would be less than that prohibitive tariff rate. Another common example is where there is an import tariff but the world price is high enough that the country is freely exporting this product. In that case, the domestic price would be determined by the world price less export trade costs and the import tariff would be irrelevant: there would be no distortion despite the presence of the import tariff measure. Similar conditions apply to exportable goods, where a prohibitive export tax may create a distortion equal to less than the tax rate. In this case, the distortion wedge would be equal to the difference between the autarky price and the world price less export trade costs. Or, if the country were freely importing the good, the export tax would be irrelevant and there would be no distortion despite the pres- ence of the export tax measure. The choice of international price to be compared with domestic prices therefore is not based solely on the actual trading status of the country (Byerlee and Morris 1993). Moreover, different prices may be needed for different regions of a large country that simultaneously export and import because internal (including coastal shipping) trading costs are so high relative to international trading costs (Koester 1986). In that case, the value of production is split according to those regions' production shares. If the only intervention in this 584 Distortions to Agricultural Incentives: A Global Perspective sector is a tariff on imports, the tariff rate is the NRA estimate for the import- competing part and zero is the NRA for the other part of the sector. Those differ- ent NRAs would be included in the weighted average calculations of the NRAs for the import-competing, exportable, and all tradables subsectors of agriculture used in the calculation of the TBI and RRA. Transmission of assistance and taxation along the agricultural value chain A crucial aspect of the NRA calculation for agricultural products is how any pol- icy measure beyond the farmgate is transmitted back to farmers and forward to consumers. Only a few parameters and exogenous variables are needed to obtain meaningful estimates of an individual agricultural product's NRA and CTE. Specifically, to take account of pass-through of distortions along the value chain, the following parameters are identified (although the default is equi- proportionate pass-through): · the extent to which any distortion to a primary farm product at the whole- f, sale level is passed back to farmers; and · , the extent to which any distortion to the downstream processed product is passed back to wholesalers of a primary farm product that is nontradable. CTE of the farm product Many farm products are processed and often used as an ingredient in further manufacturing of a food product before purchase by the final consumer (for example, wheat is ground to flour and then mixed with other ingredients before being baked and often sliced and packaged for sale as bread). Other farm products are used as inputs into different farm activities--again, often after some process- ing (for example, soybeans are crushed and the meal is mixed with maize or other feedgrains for use as animal feed, while the oil is sold for cooking). Because of these many and varied value chain paths, and because in practice it is difficult any- way to determine the extent to which a change in the primary farm product would be passed along any of those value chains, the OECD expresses its CSE simply as the level at which a product is first traded (for example, as wheat or soybean or beef). That practice is adopted here too for generating a consistent set of estimates across countries of the CTE for chapter 1 (even though authors of some individ- ual country studies report CTEs that they may have estimated in a more sophisti- cated way further along the value chain). In the absence of any domestic production or consumption taxes or subsidies directly affecting that product, the Methodology for Measuring Distortions to Agricultural Incentives 585 CTE at the point at which a product is first traded will be the same as the NRAo (and recall that the NRAo in that case also equals the NRA if NRAi is zero). Key required information A template spreadsheet has been designed to aid the management of individual country information and ensure a consistent comparison across regions and periods. The precise ways in which parameters and exogenous variables entered each country spreadsheet to endogenously generate the NRAs and CTEs are detailed in Anderson et al. (2008a). Most are straightforward, the main exception being the treatment of exchange rate distortions described below. The key exogenous variables needed are agricultural quantities produced and consumed (or imported and exported if a proxy for consumption is to be produc- tion plus net imports); wholesale and border prices of primary and lightly processed agricultural goods (and, where relevant, a quality adjustment to match border prices); agricultural input and output domestic subsidies and taxes (the default is zero); if there are distorted farm input markets, the input's share in the value of farm output at border prices (and, if there are only farmgate rather than wholesale prices for a primary good, the proportion of the farmgate value in the value at the wholesale level at border price); final food consumer domestic subsi- dies or taxes (the default is zero); and the official exchange rate (and, where preva- lent, the parallel exchange rate and the share of currency going through that secondary or illegal market, plus the product-specific exchange rate if a multiple exchange rate system is in place). Exchange rate distortions The treatment of exchange rate distortions in this project is worth spelling out since it differs from the method used by Krueger, Schiff, and Valdés (1988, 1991). If there are no exchange rate distortions, the official exchange rate is used. However, in the presence of a parallel market rate (which could be the black market rate if no legal secondary market exists), this is reported along with an estimate of the proportion of foreign currency that is actually sold by exporters at the parallel market rate. This proportion would be the formal retention rate where a formal dual exchange regime is in place, or otherwise a guesstimate of the pro- portion traded on the black market (premiums for which are provided by Easterly 2006 and International Currency Analysis 1993). The spreadsheet then computes an estimate for the equilibrium exchange rate for the economy, or the rate at which international prices are converted into local currency, to compute each NRA. 586 Distortions to Agricultural Incentives: A Global Perspective Relevant exchange rates for importers and exporters are then computed endogenously. If they are distorted away from the official exchange rate, the rele- vant exchange rates for importers and exporters are, respectively, the discounted parallel market rate and the weighted average of the official exchange rate and the discounted parallel rate according to the proportion of the exporter's currency that is sold on the parallel market. However, if a multiple exchange rate system is in place and that system provides for a specific rate for a product that differs from the general rates automatically calculated as above, then the automatically com- puted relevant exchange rate is replaced by the industry-specific rate. Guesstimates of NRAs for the noncovered agricultural products In calculating the weighted average rates of assistance for a sector or subsector, NRAs have to be guesstimated for the noncovered agricultural products for which price comparisons are not calculated (comprising 30 percent or so of the value of all farm production). In its PSE work, the OECD assumes the not-measured part has the same market price support as the average of the measured part. Another default is to assume the rates are zero. Orden et al. (2007) show that these two alternatives produce significantly different results for India, and so it is preferable to make informed judgments for the import-competing, exporting, and nontrad- able parts of the residual group of farm products. An average applied import tar- iff is often the best guess for only the import-competing products in that set if there is no evidence of explicit production, consumption, or export taxes or sub- sidies. Even though such a guess will miss nontariff trade barriers affecting these residual products, the bias will be small if their weight is small. NPS assistance to agriculture If there are NPS forms of agricultural subsidies or taxes in addition to product- specific ones, these are included in the NRAag in the same way (as a percentage of the total value of production) as done for these types of interventions in the cal- culation by the OECD (2007) of its total support estimate. Here, it is assumed that assistance goes only to tradables and is allocated as between importables and exportables on a pro-rata basis. No attempt is made to estimate the discouraging effects of underinvestment in rural infrastructure and underdevelopment of pertinent institutions. Also impor- tant is the structure of that expenditure within the rural sector. Though such expenditure may well be a nontrivial part of the distortions to agricultural incen- tives, unfortunately it is not captured in the above measures of distortions. Methodology for Measuring Distortions to Agricultural Incentives 587 In some high-income countries, governments also assist farm households with payments that are purported to be decoupled from production incentives. An example is the single farm payment in the EU. These payments are shown sepa- rately as part of NRAag because even though they are not intended to impact production, typically they do alter producer incentives somewhat. Where these payments are significant, their ad valorem equivalent is shown in charts when dis- cussing assistance to farmers as a social group, so as to be able to compare their order of magnitude with support from measures that more directly alter produc- tion incentives. Assistance to nonagricultural sectors If the nonagricultural sectors are assisted only via import tariffs on manufactures or export taxes on minerals, it is a relatively easy task to estimate a weighted aver- age NRAnonag once the shares of import-competing, exporting, and nontradables production are determined. In practice, however, there are also nontariff trade measures to consider among the measures affecting tradables (Dee and Ferrantino 2005; OECD 2005); most economies have myriad regulations affecting their many service industries. Those regulations can be very complex (see Findlay and Warren 2001). Since most of the outputs of service industries (including the public sector) are nontradable, the default in this study is to assume their average rate of government assistance--along with that of nontradable nonagricultural goods--is zero. The task of estimating the NRAnonag is then reduced to obtaining NRAs for only producers of import-competing and export-oriented nonagricul- tural goods, plus their shares of the undistorted value of production of nonagricul- tural tradables, in order to obtain the weighted average NRAnonag t for entering into the RRA calculation. Use of percentages in the chapters In order to simplify the presentation in the chapters, the NRAo, NRAi, NRA, CTE, and RRA are expressed as percentages rather than proportions. Dollar values of farmer assistance and consumer taxation Country authors' estimates of NRAs are multiplied by the gross value of produc- tion at undistorted prices to obtain an estimate in current U.S. dollars of the direct gross subsidy equivalent (GSE) of assistance to farmers. The GSEs can then simply be added up across products for a country and across countries for any or all prod- ucts to obtain regional aggregate transfer estimates for the studied countries. To get 588 Distortions to Agricultural Incentives: A Global Perspective an aggregate estimate for the rest of the region, the weighted average NRA for non- studied countries is assumed to be the same as the weighted average NRA for the studied countries, and the nonstudied countries' share of the region's gross value of farm production at undistorted prices each year is assumed to be the same as its share of the region's agricultural GDP measured at distorted prices. Just as the NRA (the percentage distortion to the gross price of farm products) is used to generate the GSE of assistance to farmers, so the RRA (the percentage distortion to the relative price of farm products as a group) can be used to gener- ate a net subsidy equivalent (NSE) of aggregate assistance to farmers. The same inflation technique as for GSE is used to obtain a regional aggregate NSE estimate that includes nonstudied countries. To obtain comparable dollar value estimates of the consumer transfer, the CTE estimate at the point at which a product is first traded is multiplied by the gross value of consumption at undistorted prices (proxied by production at undistorted prices plus net imports) to obtain an estimate in current U.S. dollars of the tax equivalent to consumers (TEC) of primary farm products. This, too, can be added up across products for a country and across countries for any or all products to obtain regional aggregate transfer estimates for the studied countries. An aggre- gate estimate for noncovered products in the studied countries, or for the region's or world's nonstudied countries, is not attempted in the project. The GSE and TEC dollar values can be illustrated in a supply-demand diagram for a distorted domestic market for a farm product of a small trading economy (see figure A.2). In the case of an import-competing product subjected to an import tariff tm , a production subsidy sf, and a consumption tax cc , the GSE is the rectangle abcd and the TEC is the rectangle ahfg. The GSE estimate is an over- statement to the extent of triangle cdj and the TEC estimate is an understatement to the extent of triangle efg, where those triangles are smaller the more price- inelastic are the supply and demand curves S and D, respectively. In the case of an exportable product subjected to an export tax tx , the GSE is the negative of the rectangle kruv and the TEC is the negative of the rectangle nquv. Methodology for Measuring Distortions to Agricultural Incentives 589 Figure A.2. Distorted Domestic Markets for Farm Products a. An import-competing product subjected to an import tariff tm plus a production subsidy sf and a consumption tax cc price S pm(1 tm)(1 sf) b c h f pm(1 tm)(1 cc) pm(1 tm) j g pm a d e D quantity b. An exportable product subjected to an export tax tx S px v k n px(1 t) u x q r price D quantity Source: Authors' derivation. Notes 1. Although it is not an explicit objective of the project, providing comparable estimates of distor- tions to lightly processed food industries in addition to primary agricultural industries at the farmgate level and to food consumers at the retail level could illuminate trade and processing costs that con- tribute to price gaps at different points in the value chain. 2. The NRA thus differs from the producer support estimate (PSE) as calculated by the Organisa- tion for Economic Co-operation and Development (OECD), in that the PSE is expressed as a fraction of the distorted value. It is thus tm (1 tm), and so for a positive tm it is smaller than the NRA and is necessarily less than 100 percent. 3. "Equilibrium" in the sense of what would prevail without this distortion in the domestic market for foreign currency. In the diagram, and in the discussion that follows, the equilibrium exchange rate E exactly balances the supply and demand for foreign currency. Taken literally, this implies a zero bal- ance on the current account. The approach here can readily be generalized to accommodate exogenous capital flows and transfers, which would shift the location of QE. With constant-elasticity supply and demand curves, all the results would carry through, and any exogenous change in those capital flows or transfers would imply a shift in the Dfx or Sfx curves. 590 Distortions to Agricultural Incentives: A Global Perspective 4. From the viewpoint of wanting to use the NRAo and CTE estimates later as parameters in a CGE model, it does not matter what assumptions are made here about these elasticities, as the CGE model's results for real variables will not be affected. What matters for real impacts is the magnitude of the total distortion, not its allocation between an export tax and an import tax: the traditional inci- dence result from tax theory that also applies to trade taxes (Lerner 1936). For an excellent general equilibrium treatment using an early version of the World Bank's 1-2-3 Model, see de Melo and Robinson (1989). There, a distinction is made between traded and nontraded goods (using the Armington [1969] assumption of differentiation between products sold on domestic markets and those sold on international markets), in contrast to the distinction between tradable and nontradable products made below in the text. 5. Note that this same type of adjustment could be made where the government forces exporters to surrender all foreign currency earnings to the domestic commercial banking system and importers to buy all needed foreign currency from the banking system and where that system is allowed by regu- lation to charge excessive fees. This apparently occurs in, for example, Brazil, where the spread is reportedly 12 percent. If actual costs in a nondistorted competitive system are only 2 percent (as they are in the less-distorted Chilean economy), the difference of 10 points could be treated as the equiva- lent of a 5 percent export tax and a 5 percent import tax applying to all tradables (but, as with non tar- iff barriers, there would be no government tariff revenue but rather rent, in this case accruing to com- mercial banks rather than to the central bank). This is an illustration of the point made by Rajan and Zingales (2004) of the power of financial market reform in expanding opportunities. 6. Results from a multicountry research project with a macro policy focus are reported in Little et al. (1993). 7. As in the two-tier exchange rate case, the elasticities are used merely to identify the incidence of these measures: as long as both NRAo and the CTE are included in any economic model used to assess the impact of the production tax, the real impacts will depend only on the magnitude of the total dis- tortion, sf , not on the estimated NRA and CTE. 8. On this general phenomenon of offsetting distortions for outputs and inputs (and even direct payments or taxes), see Rausser (1982). 9. Bear in mind that a fertilizer plant or livestock feedmix plant may enjoy import tariff protection that raises the domestic price of fertilizer or feedmix to farmers by more than any consumption sub- sidy (as had been the case for fertilizer in the Republic of Korea--see Anderson 1983), in which case the net contribution of this set of input distortions to the total NRA for agriculture would be negative. 10. That is not to say there is no interest in comparisons across countries or over time in, say, the farmgate price as a proportion of the fob export price, which summarizes the extent to which the pro- ducer price is depressed by the sum of internal transport, processing, and marketing costs plus effects such as explicit or implicit production or export taxes. Prominent users of that proportion, which can be less than half in low-income countries even where there is little or no processing, include Bates (1981) and Binswanger and Scandizzo (1983). Users need to be aware, though, that this ratio under- states the extent of farmer assistance (that is, it understates the rate of protection or overstates the rate of disprotection to farmers), possibly by a large margin. 11. On the basic economics of trading costs, see Limao and Venables (2001), Venables and Limao (2002), and Venables (2004). Such costs are affected by factors such as infrastructure within the country, at the border (ports and airports) and, in the case of landlocked countries, in transit countries. Also relevant are international freight etc. costs, and their impact on both the aggregate volume and product structure of international trade. See also the survey by Anderson and van Wincoop (2004), where it is reported that the tax equivalent of trading costs are estimated to be more than 170 percent in high-income countries and even higher in developing and transition economies, especially those that are small, poor, and remote. Trade facilitation, through lowering those trading costs (for example, streamlining customs clearance pro- cedures), can be the result of not only technological changes but also of government policy choices such as restrictions on which ships can be used in bilateral trade. For example, Fink, Mattoo, and Neagu (2002) estimate that the policy contribution to costs of shipping goods from developing countries to the United States is greater than the border import barriers. For more general background on imperfect competition in services markets, including cartelized international shipping, see Francois and Wooten (2001, 2006). Methodology for Measuring Distortions to Agricultural Incentives 591 12. The costs of processing and of wholesale/retail distribution, as well as domestic trading costs, change over time not only because of technological advances but also following policy changes. For example, government investment in rural infrastructure can lower trading costs. Reardon and Timmer (2007) argue that the global supermarket revolution is in part driven by the opening of domestic markets following the relaxation of government restrictions on foreign direct investment since the 1980s. These types of government policies are not included in the present project's measurement of distortions. 13. Trading costs may be unrelated to the product price (that is, specific rather than ad valorem), in which case the formulas should be adjusted accordingly (for example, if Tf is in dollars per ton). If this were the case with internationally traded costs, the domestic price of importables (exportables) would change less (more) than proportionately to P. The ad valorem assumption is preferable to the specific one in situ- ations where international price and exchange rate changes are less than fully passed though the domestic value chain to the farmer and consumer because of incomplete market integration caused, for example, by poor infrastructure or weak institutions. Ideally, in such cases one would estimate econometrically the extent to which the price transmission elasticity is below unity and use it to calculate the margin each year. Trading costs include storage costs that would be incurred to hold domestic products until the same time in the season when international trade takes place. Any subsidies or taxes on these or any other trad- ing costs should be included in the distortion calculus. On the importance of these domestic trading costs in low-income countries, see the case studies of Madagascar (Moser, Barrett, and Minten 2005), Rwanda (Diop, Brenton, and Asarkaya 2005), and Bangladesh (Balkht, Koolwal, and Khandker 2006). 14. On how and why the quality and variety of traded goods vary by country of origin, see Hummels and Klenow (2005). 15. It is assumed that the quality difference arises because one good provides more effective units of services than another, so that the relative price is a constant proportion of the value of the first good. When products are simply differentiated, without such a quality dimension (as in Armington 1969), there will be no fixed relationship between the two prices. 16. As with the incidence of the exchange rate distortion discussed above, from the viewpoint of wanting to use the NRA and CTE estimates later as parameters in a CGE model, the assumptions made here about the extent of pass-though along the value chain may not greatly affect the model's results for real variables such as prices, output, and value added. 17. Corden (1971) proposed that free-trade volume be used as weights, but since they are not observable (and an economy-wide model is needed to estimate them), the common practice is to com- promise by using actual distorted volumes but undistorted unit values or, equivalently, distorted values divided by (1 NRA). If estimates of own-price and cross-price elasticities of demand and supply are available, a partial equilibrium estimate of the quantity at undistorted values could be generated, but if those estimated elasticities are unreliable this may introduce more error than it seeks to correct. 18. The mean and standard deviation could be captured by a single measure, namely, the trade restrictiveness index developed by Anderson and Neary (2005). Calculating the TRI requires knowing the elasticity of import demand or, if the NRA and CTE are not identical, the price elasticities of domestic demand and supply for each product. However, by making the simplifying assumption that supply and demand elasticities are identical, it is a much simpler matter to calculate welfare-reducing and trade- reducing indexes, as shown in chapter 11 of this volume (Lloyd, Croser, and Anderson 2009). 19. In addition to the methodologies of Krueger, Schiff, and Valdés (1988, 1991) and the OECD (2007a) for estimating agricultural distortion and producer support indicators, see the recent review of methodologies of other previous studies by Josling and Valdés (2004). 20. Also of great help, particularly for trade and exchange rate distortions, are trade policy studies including the various multicountry studies such as the one summarized in Bhagwati (1978) and Krueger (1978) and more recent ones summarized in Bevan, Collier, and Gunning (1989); Michaely, Papageorgiou, and Choksi (1991); Bates and Krueger (1993); and Rodrik (2003). 21. For an analysis of agricultural input subsidies in India, see Gulati and Narayanan (2003). 22. On the difficulties in even defining property rights in water markets, even in advanced economies, see Donohew (2009) and Young and McColl (2009). 23. That these costs vary hugely across countries, and often dwarf trade taxes, is now clearly estab- lished. See, for example, World Bank (2006a, 2006b) and also http://www.doingbusiness.org and the 592 Distortions to Agricultural Incentives: A Global Perspective governance and anticorruption indicators at http://info.worldbank.org/governance. Also now avail- able is a database on information and communications cost indicators for 144 countries, at http://www.worldbank.org/ic4d. In some settings, trading cost-induced price bands due to missing or imperfect markets in rural areas cause poor farmers to forego cash crop production in order to ensure enough food production for survival (de Janvry, Fafchamps, and Sadoulet 1991; Fafchamps 1992). This contributes to a low level of supply responsiveness of poor producers to international price changes for those cash crops. References Anderson, J., and P. Neary. 2005. Measuring the Restrictiveness of International Trade Policy. Cambridge, MA: MIT Press. Anderson, J., and E. van Wincoop. 2004. "Trade Costs." Journal of Economic Literature 42 (3): 691­751. Anderson, K. 1983. "Fertilizer Policy in Korea." Journal of Rural Development 6 (1): 43­57. Anderson, K., and Y. Hayami, eds. 1986. The Political Economy of Agricultural Protection: East Asia in International Perspective. London: Allen and Unwin. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008a. "Methodology for Measur- ing Distortions to Agricultural Incentives." Agricultural Distortions Working Paper 02, World Bank, Washington, DC. ______. 2008b. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 1­30. Armington, P. 1969. "A Theory of Demand for Products Distinguished by Place of Production." IMF Staff Papers 16 (1): 159­78. Baffes, J., and B. Gardner. 2003. "The Transmission of World Commodity Prices to Domestic Markets Under Policy Reforms in Developing Countries." Journal of Policy Modeling 6 (3): 159­80. Balkht, Z., G. B. Koolwal, and S. R. Khandker. 2006. "The Poverty Impact of Rural Roads: Evidence from Bangladesh." Policy Research Working Paper 3875, World Bank, Washington, DC. Bates, R. H. 1981. Markets and States in Tropical Africa. Berkeley, CA: University of California Press. Bates, R. H., and A. O. Krueger, eds. 1993. Political and Economic Interactions in Economic Policy Reform: Evidence from Eight Countries. Oxford: Blackwell. Bevan, D., P. Collier, and J. W. Gunning. 1989. Peasants and Governments: An Economic Analysis. Oxford: Clarendon Press. Bhagwati, J. N. 1971. "The Generalized Theory of Distortions and Welfare." In Trade, Balance of Payments and Growth, ed. J .N. Bhagwati, et al. Amsterdam: North-Holland. ______. 1978. Foreign Trade Regimes and Economic Development: Anatomy and Consequences of Exchange Control Regimes. Cambridge, MA: Ballinger. Binswanger, H., and P. L. Scandizzo. 1983. "Patterns of Agricultural Protection." Report ARU 15, Agricultural and Rural Development Department, World Bank, Washington, DC. Byerlee, D., and M. Morris. 1993. "Calculating Levels of Protection: Is it Always Appropriate to Use World Reference Prices Based on Current Trading Status?" World Development 21 (5): 805­15. Collier, P., and J. W. Gunning, with associates. 1998. Trade Shocks in Developing Countries: Theory and Evidence. Oxford: Clarendon Press. Corden, W. M. 1971. The Theory of Protection. Oxford: Clarendon Press. ______. 1997. Trade Policy and Economic Welfare (second edition). Oxford: Clarendon Press. Dee, P., and M. Ferrantino, eds. 2005. Quantitative Methods for Assessing the Effects of Non-Tariff Measures and Trade Facilitation. London: World Scientific. de Janvry, A., M. Fafchamps, and E. Sadoulet. 1991. "Peasant Household Behaviour with Missing Markets: Some Paradoxes Explained." Economic Journal 101(409): 1400­17. de Melo, J., and S. Robinson. 1989. "Product Differentiation and the Treatment of Foreign Trade in Computable General Equilibrium Models of Small Economies." Journal of International Economics 27: 47­67. Methodology for Measuring Distortions to Agricultural Incentives 593 Dervis, K., J. de Melo, and S. Robinson. 1981. "A General Equilibrium Analysis of Foreign Exchange Shortages in a Developing Country." Economic Journal 91: 891­906. Dimaranan, B. V., ed. 2006. Global Trade, Assistance, and Production: The GTAP 6 Data Base. West Lafayette, IN: Center for Global Trade Analysis, Purdue University. http://www.gtap.org. Diop, N., P. Brenton, and Y. Asarkaya. 2005. "Trade Costs, Export Development and Poverty in Rwanda." Policy Research Working Paper 3784, World Bank, Washington, DC. Donahew, Z. 2009. "Property Rights and Western United States Water Markets." Australian Journal of Agricultural and Resource Economics 53 (1): 85­103. Easterly, W. 2006. Global Development Network Growth Database. New York University Development Research Institute, New York, NY. http://www.nyu.edu/fas/institute/dri/globaldevelopmentnet- workgrowthdatabase.html. Fafchamps, M. 1992. "Cash Crop Production, Food Price Volatility, and Rural Market Integration in the Third World." American Journal of Agricultural Economics 74 (1): 90­99. Findlay, C., and T. Warren. 2001. Impediments to Trade in Services: Measurement and Policy Implica- tions, London and New York: Routledge. Fink, C., A. Mattoo, and I. Neagu. 2002. "Trade in International Maritime Services: How Much Does Policy Matter?" World Bank Economic Review 16 (1): 81­108. Francois, J., and I. Wooton. 2001. "Trade in International Transport Services: The Role of Competi- tion." Review of International Economics 9 (2): 249­61. ______. 2006. "Market Structure and Market Access." Unpublished paper, Erasmus University, Rotterdam. Garnaut, R., and A. Clunies-Ross. 1983. The Taxation of Mineral Rents. London: Oxford University Press. Gulati, A., and S. Narayanan. 2003. The Subsidy Syndrome in Indian Agriculture. New Delhi: Oxford University Press. Harberger, A. 1971. "Three Basic Postulates for Applied Welfare Economics: An Interpretative Essay." Journal of Economic Literature 9 (3): 785­97. Hummels, D. L., and P. J. Klenow. 2005. "The Variety and Quality of a Nation's Exports." American Economic Review 95 (3): 704­23. International Currency Analysis. 1993 and earlier years. World Currency Yearbook (formerly Pick's Currency Yearbook). Brooklyn, NY: International Currency Analysis, Inc. Josling, T. E., and A. Valdés. 2004. "Agricultural Policy Indicators." ESA Working Paper 04-04, Food and Agriculture Organization, Rome. Kiguel, M., and S. A. O'Connell. 1995. "Parallel Exchange Rates in Developing Countries." World Bank Research Observer 10 (1): 21­52. ______, eds. 1997. Parallel Exchange Rates in Developing Countries. London: Macmillan and New York: St Martin's Press. Koester, U. 1986. "Regional Cooperation to Improve Food Security in Southern and Eastern African Countries." International Food Policy Research Institute (IFPRI) Research Report 53, IFPRI, Washington, DC. Krueger, A. O. 1978. Foreign Trade Regimes and Economic Development: Liberalization Attempts and Consequences. Cambridge, MA: Ballinger. Krueger, A. O., M. Schiff, and A. Valdés. 1988. "Agricultural Incentives in Developing Countries: Measuring the Effect of Sectoral and Economy-wide Policies." World Bank Economic Review 2 (3): 255­72. ______. 1991. The Political Economy of Agricultural Pricing Policy, Volume 1: Latin America, Volume 2: Asia, and Volume 3: Africa and the Mediterranean. Baltimore: Johns Hopkins University Press for the World Bank. Lerner, A. 1936. "The Symmetry Between Import and Export Taxes." Economica 3(11): 306­13. Limao, N., and A. J. Venables. 2001. "Infrastructure, Geographical Disadvantage, Transport Costs, and Trade." World Bank Economic Review 15 (3): 451­79. Little, I. M. D., R. N. Cooper, W. M. Corden, and S. Rajapatirana, eds. 1993. Boom, Crisis and Adjust- ment: The Macroeconomic Experience of Developing Countries. London: Oxford University Press. 594 Distortions to Agricultural Incentives: A Global Perspective Lloyd, P. J. 1974. "A More General Theory of Price Distortions in an Open Economy." Journal of International Economics 4 (4): 365­86. Lloyd, P. J., J. Croser, and K. Anderson. 2009. "Welfare-Based and Trade-Based Indicators of National Agricultural Distortions." Chapter 11 in this volume. Martin, W. 1993. "Modeling the Post-Reform Chinese Economy." Journal of Policy Modeling 15 (5&6): 545­79. Matthews, A., P. Claquin, and J. Opolot. 2009. "Uganda." Ch. 12 in K. Anderson and W. A. Masters (eds.), Distortions to Agricultural Incentives in Africa, Washington, DC: World Bank. Michaely, M., D. Papageorgiou, and A. Choksi, eds. 1991. Liberalizing Foreign Trade, 7: Lessons from Experience in the Developing World. Oxford: Basil Blackwell. Moser, C., C. B. Barrett, and B. Minten. 2005. "Missed Opportunities and Missing Markets: Spatio- Temporal Arbitrage of Rice in Madagascar." Unpublished paper, Colgate University, Hamilton, NY. OECD (Organisation for Economic Co-operation and Development). 2005. Looking Beyond Tariffs: The Role of Non-Tariff Barriers in World Trade. Paris: OECD. ______. 2007. OECD PSE-CSE Database (Producer and Consumer Support Estimates, OECD Data- base 1986­2007). OECD http://www.oecd.org/document/59/0,3343,en_2649_33727_39551355 _1_1_1_1,00.html. Orden, D., F. Cheng, H. Nguyen, U. Grote, M. Thomas, K. Mullen, and D. Sun. 2007. "Agricultural Pro- ducer Support Estimates for Developing Countries: Measurement Issues and Evidence from India, Indonesia, China and Vietnam." International Food Policy Research Institute (IFPRI) Research Report 152, Washington, DC. Rajan, R., and Zingales. 2004. Saving Capitalism from the Capitalists: Unleashing the Power of Financial Markets to Create Wealth and Spread Opportunity. Princeton, NJ: Princeton University Press. Rausser, G. C. 1982. "Political Economic Markets: PERTs and PESTs in Food and Agriculture." American Journal of Agricultural Economics 64 (5): 821­33. Reardon, T., and C. P. Timmer. 2007. "Transformation of Markets for Agricultural Output in Develop- ing Countries Since 1950: How Has Thinking Changed?" In Handbook of Agricultural Economics, Volume 3A, ed. R. E. Evenson, P. Pingali, and T. P. Schultz. Amsterdam: Elsevier. Rodrik, D., ed. 2003. In Search of Prosperity: Analytical Narratives on Economic Growth. Princeton: Princeton University Press. Salter, W. E. G. 1959. "Internal and External Balance: The Role of Price and Expenditure Effects." Economic Record 35 (71): 226­38. Shatz, H. J., and D. Tarr. 2000. "Exchange Rate Overvaluation and Trade Protection: Lessons from Experience." World Bank Policy Research Working Paper 2289, World Bank, Washington, DC. Tarr, D. 1990. "Second-Best Foreign Exchange Policy in the Presence of Domestic Price Controls and Export Subsidies." World Bank Economic Review 4 (2): 175­93. Valdés, A. 1996. " Surveillance of Agricultural Price and Trade Policy in Latin America During Major Policy Reforms." World Bank Discussion Paper 349, World Bank, Washington, DC. ______, ed. 2000. "Agricultural Support Policies in Transition Economies." World Bank Technical Paper 470, World Bank, Washington, DC. Venables, A. J. 2004. "Small, Remote and Poor?" World Trade Review 3 (3): 453­57. Venables, A. J., and N. Limao. 2002. "Geographical Disadvantage: A Heckscher-Ohlin-von Thunen Model of International Specialization." Journal of International Economics 58 (2): 239­63. Vousden, N. 1990. The Economics of Trade Protection. Cambridge: Cambridge University Press. Young, M. D., and J. C. McColl. 2009. "Double Trouble: The Importance of Accounting For and Defin- ing Water Entitlements Consistent with Hydrological Integrity." Australian Journal of Agricultural and Resource Economics 53 (1): 19­35. Warr, P. G. 2008. "The Transmission of Import Prices to Domestic Prices: An Application to Indonesia." Applied Economics Letters 15 (7): 499­503. World Bank. 2006a. Assessing World Bank Support for Trade, 1987­2004: An IEG Evaluation. Independ- ent Evaluation Group, World Bank, Washington, DC. ______. 2006b. Information and Communications for Development 2006: Trends and Policies. Washington, DC: World Bank. Appendix B Global Distortions Database, 1955­2007 Kym Anderson and Ernesto Valenzuela* The core global database developed as an integral part of the World Bank's Agricultural Distortions Project was a huge team effort involving more than 90 consultants, who prepared national case studies and associated spreadsheets, plus a small team of doctoral students (listed above), who served as research assistants in Washington, DC, and Adelaide, Australia. The core database is publicly avail- able at http://www.worldbank.org/agdistortions (Anderson and Valenzuela 2008). The case studies involve a total of 75 countries that together account for 92 per- cent of the world's population and agricultural gross domestic product (GDP) and 95 percent of total GDP. The same set of countries also accounts for more than 85 percent of farm production and employment in each of Africa, Asia, Latin America, and the transition economies of Europe and Central Asia. The only coun- tries not well represented in the sample are those in the Middle East and the many small ones that together account for less than 5 percent of the global economy. Nominal rates of assistance (NRAs) and consumer tax equivalents (CTEs) are estimated for more than 70 products, with an average of 11 products per country. In aggregate, the coverage represents around 70 percent of the gross value of agri- cultural production in the focus countries, and just under two-thirds of global farm production valued at undistorted prices over the period covered. Not all countries had data for all of the entire 1955­2007 period; the average number of years covered is 41 per country.1 Of the 30 most valuable agricultural products, the NRAs presented here cover 77 percent of global output, ranging from two- thirds for livestock, three-quarters for oilseeds and tropical crops, and five-sixths *The compilation of this database would not have been possible without the dedicated assistance of Johanna L. Croser, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Francesca de Nicola, and Damiano Sandri. 595 596 Distortions to Agricultural Incentives: A Global Perspective for grains and tubers. Those products represent an even higher share (85 percent) of global agricultural exports. Tables B.1­B.7 provide further details of the above coverage statistics. The tables that follow provide trade propensities of the different regions for the 30 major products (table B.8), per capita incomes (in purchasing power parity) of the study's 75 focus economies (table B.9), and a list of the key variables included in the core database (table B.10). Figure B.1 shows the number of countries for which NRA and CTE estimates are produced for each of the 30 key farm products, while figure B.2 shows the share of global production of each of those 30 key farm products that is covered by NRA and CTE estimates. The core database is the source of all the indicators of distortions to agricul- tural incentives in the four regional books2 that are summarized in part III of the present volume, as well as in the high-income country studies in part II of this volume. The database has also been used to provide new estimates of distortions to producer and consumer prices of farm products for use by national and global sectoral or economy-wide models (Valenzuela and Anderson 2008), one applica- tion of which is reported in chapter 13 of this volume. Furthermore, the core database of NRAs and CTEs has been the main source for a supplementary database of partial equilibrium indexes of the trade and welfare reductions due to government interventions in agricultural markets in the 75 focus countries (Anderson and Croser 2009). The theory of those trade and welfare reduction indexes is developed in chapter 11 of this volume (see also Figure B.1. Number of Countries for Which NRA and CTE Estimates Are Provided for 30 Key Farm Products 55 50 45 40 35 number 30 25 20 15 10 5 0 su t m r ze ilk po ef g y t e sh ba s p y ta t co es n yb ts lo s rg rs pe m co s ca e gr m va nd let co ts ts a ea po ea ga g nf an ed pi ultr ee rle ric ffe tto so oa co nu nu so we be he to ra hu m ai eg a ou il m m su e se ss w Source: Drawn from estimates in Anderson and Valenzuela (2008). Global Distortions Database, 1955­2007 597 Figure B.2. Shares of Global Production of 30 Key Farm Products Covered in NRA and CTE Estimates 100 90 80 70 60 percent 50 40 30 20 10 0 m s ze g e w at co at su n r ba ilk p ley lo y co rs rg e m se f s co at ca uts a nd s ta ts sh m s p et t pe e ea ga n ed ou egg e nf ltr av pi ric so ffe tto po nu e ra be e he co o ee ill hu to m ai ea r w n m m su ou ss yb so gr Source: Drawn from estimates in Anderson and Valenzuela (2008). Lloyd, Croser, and Anderson 2009a, 2009b), and a global summary of the indexes themselves is provided for countries in chapter 11 and for commodities in chap- ter 12. A list of the key variables included in that supplementary database of agri- cultural trade and welfare reduction indexes is shown in table B.11. Table B.1. Summary of NRA Coverage Statistics, World Bank Agricultural Distortions Project 2000­04 % of global: Number and size of countries Number Population Agricultural GDP Africa 21 11 7 Asia 12 51 37 Latin America 8 7 8 Subtotal, all developing countries 41 69 52 European transition economies 14 7 7 High-income countries 20 14 33 Total 75 92 92 (Table continues on the following page.) 598 Distortions to Agricultural Incentives: A Global Perspective Table B.1. Summary of NRA Coverage Statistics, World Bank Agricultural Distortions Project (continued) Number of years covered Maximum Average per country Africa 51 43 Asia 53 42 Latin America 51 39 Subtotal, all developing countries 53 43 European transition economies 47 17 High-income countries 53 52 Total 51 41 Number of products covered Maximum Average per country Africa 44 8 Asia 35 8 Latin America 27 10 Subtotal, all developing countries 59 9 European transition economies 25 12 High-income countries 39 15 Total 74 11 Total number of NRA estimates (years and products) Total Average per country Africa 7,318 348 Asia 3,546 296 Latin America 2,881 360 Subtotal, all developing countries 13,745 335 European transition economies 2,847 203 High-income countries 13,377 669 Total, focus countries 29,969 400 Share of global agricultural production of 30 key covered products, 2000­04 Percent Africa 5 Asia 28 Latin America 7 Subtotal, all developing countries 41 European transition economies 6 High-income countries 29 Total, focus countries 77 Source: Anderson and Valenzuela (2008). Table B.2. Coverage of Gross Value of Global Agricultural Production at Undistorted Prices, for 30 Key Products and Four Product Groups, 2000­04 Product's share of Coverage of product's global gross value of global gross value of production of 30 key production, % products, % Grains and tubers 85 31.4 Rice 92 9.5 Wheat 89 8.0 Maize 94 7.2 Cassava 41 2.4 Barley 83 1.6 Sorghum 70 0.7 Yams 77 0.8 Millet 29 0.6 Oats 61 0.3 Oilseeds 78 8.1 Soybeans 96 4.0 Groundnuts 34 1.2 Palm oil 90 1.0 Rapeseeds 64 1.0 Sunflower seeds 77 0.6 Sesame seeds 9 0.3 Tropical crops 74 7.6 Sugar 87 2.5 Cotton 88 1.9 Coconut 60 0.9 Coffee 75 0.7 Rubber 74 0.7 Tea 21 0.6 Cocoa 71 0.4 Chickpeas 61 0.3 Livestock products 72 52.9 Pig meat 91 12.4 Milk 83 12.0 Beef 69 10.5 Poultry 81 7.6 Eggs 35 6.5 Sheep meat 27 3.3 Wool 42 0.6 All above products 77 100.0 Source: Authors' calculations from Agricultural Distortions Project database. Note: Fruit and vegetables, which account for around 23 percent of global agricultural gross value of production in 2001 (according to GTAP database--see Dimaranan 2007) are not included in this table, even though several fruits and vegetables for selected countries are included in the analysis. 599 600 Distortions to Agricultural Incentives: A Global Perspective Table B.3. Project's Coverage of National Agricultural Production in Focus Countries at Undistorted Prices, Regional Averages, 1980­2004 (percent) 1980­84 1990­94 2000­04 Africa 67 66 68 Asia 75 73 66 Latin America 65 69 70 Subtotal, focus developing countries 72 72 67 European transition economies 62 61 63 High-income countries 73 73 72 Total, all focus countries 72 71 68 Source: Anderson and Valenzuela (2008). Table B.4. Shares of Global Agricultural Production for 30 Major Covered Products, by Region, 2000­04 Regional shares (%) of global gross value of agricultural production Product share Covered products in focus countries (%) of global European High- All gross value of Latin transition income focus production of Africa Asia America economies countries countries Residual World 30 products Grains and tubers 11 40 5 6 23 85 15 100 31 Rice 3 81 2 0 5 92 8 100 9 Wheat 6 32 4 14 33 89 11 100 8 Maize 11 26 13 5 40 94 6 100 7 Cassava 39 2 0 0 0 41 59 100 2 Barley 0 0 1 25 57 83 17 100 2 Sorghum 26 13 14 0 17 70 30 100 1 Yams 77 0 0 0 0 77 23 100 1 Millet 29 0 0 0 0 29 71 100 1 Oats 0 0 0 28 33 61 39 100 0 Oilseeds 4 22 20 4 28 78 22 100 8 Soybeans 0 15 37 0 43 96 4 100 4 Groundnuts 17 17 0 0 0 34 66 100 1 Palm oil 8 81 2 0 0 90 10 100 1 Rapeseeds 0 12 0 4 47 64 36 100 1 Sunflower seeds 3 5 16 39 14 77 23 100 1 Sesame 8 0 0 0 0 9 91 100 0 601 (Table continues on the following page.) 602 Table B.4. Shares of Global Agricultural Production for 30 Major Covered Products, by Region, 2000­04 (continued) Regional shares (%) of global gross value of agricultural production Product share Covered products in focus countries (%) of global European High- All gross value of Latin transition income focus production of Africa Asia America economies countries countries Residual World 30 products Tropical crops 10 36 12 5 11 74 26 100 8 Sugar 5 43 17 6 16 87 13 100 2 Cotton 17 30 5 14 22 88 12 100 2 Coconut 0 60 0 0 0 60 40 100 1 Coffee 11 12 52 0 0 75 25 100 1 Rubber 0 74 0 0 0 74 26 100 1 Tea 11 10 0 0 0 21 79 100 1 Cocoa 68 0 3 0 0 71 29 100 0 Chickpeas 0 61 0 0 0 61 39 100 0 Livestock products 3 21 6 7 36 72 28 100 53 Pig meat 0 49 3 6 34 91 9 100 12 Milk 3 21 4 12 43 83 17 100 12 Beef 6 1 16 5 41 69 31 100 10 Poultry 2 27 9 5 38 81 19 100 8 Eggs 0 1 3 7 23 35 65 100 6 Sheep meat 5 0 0 3 18 27 73 100 3 Wool 0 0 0 0 42 42 58 100 1 Total of above 30 products 6 28 7 6 29 77 23 100 100 Sources: Authors' calculations based on Food and Agriculture Organization (FAO) commodity balance to get volume of production data, and the Agricultural Distortions Project database to convert this to value of production at undistorted prices so we can add up across commodities. Table B.5. Share of Regional Agricultural Production for 30 Major Covered Products, by Region, 2000­04 Covered product shares (%) of regional gross value of agricultural production European High- Latin transition income All focus Africa Asia America economies countries countries Grains and tubers 37.3 21.8 14.3 16.8 15.2 19.6 Rice 2.8 13.6 1.9 0.1 1.0 6.3 Wheat 4.7 4.6 3.0 10.2 5.6 5.3 Maize 8.4 3.3 8.3 2.7 6.2 5.0 Cassava 9.8 0.1 0.0 -- -- 0.7 Barley -- 0.0 0.2 3.2 2.0 1.0 Sorghum 2.5 0.2 0.9 -- 0.3 0.4 Yams 7.0 -- -- -- -- 0.5 Millet 2.1 -- -- -- -- 0.1 Oats -- -- -- 0.6 0.2 0.1 Oilseeds 3.3 3.1 14.5 2.2 4.8 4.5 Soybeans 0.0 1.1 13.3 0.0 3.6 2.8 Groundnuts 2.0 0.4 0.0 -- -- 0.3 Palm oil 0.9 1.4 0.1 -- -- 0.6 Rapeseeds -- 0.2 -- 0.2 1.0 0.4 Sunflower seeds 0.2 0.1 0.9 2.0 0.2 0.4 Sesame seeds 0.2 -- 0.0 -- -- 0.0 (Table continues on the following page.) 603 604 Table B.5. Share of Regional Agricultural Production for 30 Major Covered Products, by Region, 2000­04 (continued) Covered product shares (%) of regional gross value of agricultural production European High- Latin transition income All focus Africa Asia America economies countries countries Tropical crops 7.9 5.1 8.1 3.5 1.7 4.2 Sugar 1.2 1.9 3.8 1.3 0.8 1.6 Cotton 2.3 1.0 0.9 2.1 0.9 1.1 Coconut -- 0.8 -- -- -- 0.3 Coffee 0.8 0.1 3.3 -- -- 0.4 Rubber -- 0.9 -- -- -- 0.4 Tea 0.7 0.1 -- -- -- 0.1 Cocoa 3.0 0.0 0.1 -- -- 0.2 Chickpeas -- 0.3 -- -- -- 0.1 Livestock products 13.7 19.0 29.9 30.5 40.4 27.9 Pig meat -- 10.6 3.0 6.6 8.9 8.3 Milk 3.5 4.5 4.4 11.8 11.1 7.3 Beef 6.8 0.2 14.7 4.6 9.1 5.3 Poultry 1.6 3.6 5.9 2.9 6.2 4.5 Eggs -- 0.1 1.9 3.7 3.2 1.7 Sheep meat 1.8 -- -- 0.9 1.3 0.7 Wool -- -- -- -- 0.6 0.2 Total of above 30 62.2 48.8 66.7 52.9 62.2 56.2 All covered 68.2 66.1 70.3 60.9 72.4 68.3 Noncovered products 31.8 33.9 29.7 39.1 27.6 31.7 All agricultural products 100.0 100.0 100.0 100.0 100.0 100.0 Source: Authors' calculations based on Agricultural Distortions Project database. Note: -- not available. Table B.6. Share of Global Agricultural and Processed Food Exports for 30 Major Covered Products, by Region, 2000­03 Share (%) of global agricultural and food exports Focus countries Product's share (%) of global High- European All gross value of Latin income transition focus agriculture and Africa Asia America countries economies countries Residual food exports Grains and tubers 0.9 16.0 7.4 60.4 5.2 89.9 10.1 8.0 Rice 2.0 53.9 0.3 27.2 0.0 83.4 16.6 1.5 Wheat 0.3 3.2 8.1 75.9 7.5 95.0 5.0 3.4 Maize 1.4 12.6 14.7 66.5 3.0 98.3 1.7 2.2 Cassava 0.1 72.6 0.0 72.8 27.2 0.1 Barley 0.0 0.0 75.0 17.2 92.2 7.8 0.6 Sorghum 0.4 0.1 0.0 0.4 0.9 99.1 0.2 Yams 21.5 21.5 78.5 0.0 Millet 3.2 3.2 96.8 0.0 Oats 45.9 0.8 46.7 53.3 0.0 Oilseeds 0.6 32.4 19.7 29.2 3.1 85.1 14.9 3.8 Soybeans 0.0 3.6 45.4 45.0 0.0 94.1 5.9 1.5 Groundnuts 1.0 8.3 3.4 12.7 87.3 0.1 Palm oil 0.1 84.4 0.6 85.0 15.0 1.4 Rapeseeds 0.1 87.5 3.4 91.0 9.0 0.4 Sunflower seeds 0.5 0.1 21.8 22.9 33.9 79.1 20.9 0.3 Sesame seeds 19.0 0.5 19.5 80.5 0.1 605 (Table continues on the following page.) 606 Table B.6. Share of Global Agricultural and Processed Food Exports for 30 Major Covered Products, by Region, 2000­03 (continued) Share (%) of global agricultural and food exports Focus countries Product's share (%) of global High- European All gross value of Latin income transition focus agriculture and Africa Asia America countries economies countries Residual food exports Tropical crops 14.9 18.6 14.7 18.4 1.3 70.9 29.1 5.8 Sugar 3.2 11.4 23.3 30.9 3.2 71.9 28.1 2.2 Cotton 23.6 4.2 1.7 46.7 0.8 76.9 23.1 0.8 Coconuts 62.2 62.2 37.8 0.0 Coffee 9.8 14.4 34.5 58.7 41.3 0.9 Rubber 93.3 93.3 6.7 0.5 Tea 15.6 27.9 43.6 56.4 0.6 Cocoa 69.1 0.6 2.3 71.9 28.1 0.7 Chickpeas 0.2 0.2 99.8 0.1 Livestock products 0.0 3.3 5.6 75.1 3.8 87.9 12.1 16.7 Pig meat 3.0 3.8 79.5 3.3 89.6 10.4 3.4 Milk 0.0 0.4 1.6 81.0 4.9 87.9 12.1 6.3 Beef 0.1 0.0 11.1 77.4 2.3 90.9 9.1 3.5 Poultry 0.1 16.6 13.0 54.6 5.0 89.3 10.7 2.5 Eggs 0.0 0.1 0.3 5.1 5.5 94.5 0.2 Sheep meat 0.0 89.8 0.2 90.0 10.0 0.5 Wool 79.6 79.6 20.4 0.2 Total of above 30 products 2.8 12.1 9.1 57.0 3.7 84.7 15.3 34.2 All agricultural and processed food exports 2.4 10.8 10.0 60.5 4.0 87.9 12.1 100.0 Source: Authors' calculations based on FAO trade data. Table B.7. Share of Global Agricultural and Processed Food Imports for 30 Major Products, by Region, 2000­ 03a Share (%) of global agricultural and food imports for each region Focus countries Product's share (%) of global High- European All gross value of Latin income transition focus agricultural and Africa Asia America countries economies countries Residual food imports Grains and tubers 6.9 7.4 8.9 21.3 2.9 47.4 52.6 9.0 Rice 10.6 12.3 4.6 19.5 1.3 48.3 51.7 1.6 Wheat 7.8 6.8 9.5 24.9 3.4 52.5 47.5 3.8 Maize 6.2 7.6 10.3 19.6 3.4 47.0 53.0 2.5 Cassava 0.0 0.0 0.0 0.0 100.0 0.1 Barley 0.6 0.5 27.9 3.5 32.5 67.5 0.6 Sorghum 0.9 50.2 0.0 51.0 49.0 0.2 Yams 0.0 100.0 0.0 Millet 0.0 0.0 100.0 0.0 Oats 14.9 1.3 16.1 83.9 0.1 Oilseeds 0.9 11.1 2.7 23.5 0.9 39.1 60.9 4.2 Soybeans 0.0 25.9 6.7 43.4 0.2 76.2 23.8 1.7 Groundnuts 0.5 0.0 0.0 0.5 99.5 0.1 Palm oil 2.3 1.4 0.0 3.7 96.3 1.5 Rapeseeds 28.0 0.3 28.3 71.7 0.5 Sunflower seeds 1.1 3.2 0.2 40.3 10.9 55.7 44.3 0.3 Sesame seeds 0.0 0.0 0.0 100.0 0.1 607 (Table continues on the following page.) Table B.7. Share of Global Agricultural and Processed Food Imports for 30 Major Products, by Region, 2000­ 03a (continued) 608 Share (%) of global agricultural and food imports for each region Focus countries Product's share (%) of global High- European All gross value of Latin income transition focus agricultural and Africa Asia America countries economies countries Residual food imports Tropical crops 1.0 6.5 0.8 11.1 6.0 25.5 74.5 6.1 Sugar 2.5 8.0 1.0 28.8 12.9 53.2 46.8 2.3 Cotton 0.4 16.2 3.1 0.7 8.0 28.4 71.6 0.8 Coconut 0.2 0.2 99.8 0.0 Coffee 0.0 0.1 0.2 0.3 99.7 0.9 Rubber 6.7 6.7 93.3 0.5 Tea 0.2 0.4 0.6 99.4 0.6 Cocoa 0.0 4.3 0.0 4.3 95.7 0.8 Chickpeas 23.5 23.5 76.5 0.1 Livestock products 0.5 3.3 3.7 62.3 4.5 74.4 25.6 16.8 Pig meat 2.2 2.5 78.7 5.6 89.0 11.0 3.6 Milk 0.6 2.3 3.0 57.2 3.2 66.3 33.7 6.3 Beef 1.2 5.8 7.6 68.8 3.8 87.2 12.8 3.4 Poultry 0.4 5.7 3.0 55.9 8.7 73.6 26.4 2.4 Eggs 0.1 2.1 0.4 4.2 6.8 93.2 0.2 Sheep meat 0.5 52.8 0.0 53.3 46.7 0.6 Wool 1.3 1.3 98.7 0.2 Total of above 30 products 2.3 5.8 4.4 39.0 3.9 55.3 44.7 36.0 All agricultural and processed food imports 2.4 11.1 4.4 62.2 5.8 85.9 14.1 100.0 Source: Authors' calculations based on FAO trade data. a. The overall coverage of imports is less than that of exports because a major food-importing region, the Middle East and North Africa, is represented in the project only by the Arab Republic of Egypt. Global Distortions Database, 1955­2007 609 Table B.8. Shares of Production Exported (X/Q) and Consumption Imported (M/C), by Farm Product and Region, 2000­03 (percent) European High- Latin transition income Africa Asia America economies countries World Grains X/Q 2 7 11 13 29 16 M/C 17 9 22 11 27 8 Rice X/Q 6 6 1 2 32 7 M/C 28 1 12 59 15 3 Wheat X/Q 4 3 46 13 49 24 M/C 44 4 51 6 27 15 Maize X/Q 4 8 15 10 20 15 M/C 14 5 14 9 5 7 Cassava X/Q 0 75 1 16 M/C 0 0 0 0 Barley X/Q 1 0 13 29 23 M/C 44 24 5 14 11 Sorghum X/Q 0 0 0 45 15 M/C 0 0 42 0 12 Millet X/Q 0 0 M/C 0 0 Oats X/Q 0 17 9 M/C 1 7 3 Oilseeds X/Q 4 32 35 25 30 33 M/C 2 51 27 6 62 23 Soybeans X/Q 10 2 32 15 38 31 M/C 5 49 12 41 23 26 Groundnuts X/Q 0 3 87 2 M/C 0 0 1 0 Palm oil X/Q 1 85 18 80 M/C 15 11 2 11 Sunflower seeds X/Q 0 0 6 17 18 13 M/C 0 0 0 4 34 9 Sesame seeds X/Q 78 160 78 M/C 0 0 0 Tropical X/Q 52 38 45 32 47 24 crops M/C 13 18 12 42 42 13 Sugar X/Q 27 12 40 20 31 25 M/C 20 9 4 46 25 18 (Table continues on the following page.) 610 Distortions to Agricultural Incentives: A Global Perspective Table B.8. Shares of Production Exported (X/Q) and Consumption Imported (M/C), by Farm Product and Region, 2000­03 (continued) (percent) European High- Latin transition income Africa Asia America economies countries World a Cotton X/Q 29 1 5 2 31 11 M/C 2 4 9 21 3 5 Coconuts X/Q 9 9 M/C 0 0 Coffee X/Q 77 78 73 75 M/C 2 3 5 4 Rubber X/Q 96 96 M/C 71 71 Tea X/Q 88 78 82 M/C 22 6 9 Cocoa X/Q 88 448 87 94 M/C 1 2434 14 49 Livestock X/Q 1 4 10 7 20 7 M/C 8 6 5 9 14 5 Pig meat X/Q 1 12 6 21 9 M/C 2 8 12 20 9 Milk X/Q 0 0 5 2 7 4 M/C 1 1 5 1 3 2 Beef X/Q 1 2 10 6 23 17 M/C 9 48 5 14 20 15 Poultry X/Q 1 10 16 7 19 15 M/C 9 11 5 28 10 11 Eggs X/Q 0 0 1 6 4 M/C 1 1 1 6 4 Sheep meat X/Q 2 2 35 26 M/C 7 0 22 16 Wool X/Q 43 43 M/C 3 3 Total of above X/Q 16 22 19 11 24 30 products M/C 13 14 12 11 19 Source: Authors' calculations based on FAO commodity balances data. a. Excluding data for the five cotton-exporting countries of Benin, Burkina Faso, Chad, Mali, and Togo. Table B.9. Per Capita Income, Focus Countries, 2005 (US$ thousands, Purchasing Power Parity) Africa European transition economies Benin 1.2 Bulgaria 9.3 Burkina Faso 1.1 Czech Republic 20.3 Cameroon 2.0 Estonia 16.5 Chad 1.5 Hungary 17.0 Côte d'Ivoire 1.6 Kazakhstan 8.7 Egypt, Arab Rep. of 4.6 Latvia 13.2 Ethiopia 0.6 Lithuania 14.1 Ghana 1.2 Poland 13.5 Kenya 1.4 Romania 9.4 Madagascar 0.8 Russian Federation 11.9 Mali 1.0 Slovak Republic 15.9 Mozambique 0.7 Slovenia 22.5 Nigeria 1.5 Turkey 7.8 Senegal 1.5 Ukraine 5.6 South Africa 22.5 High-income countries Sudan 1.7 Australia 34.1 Tanzania 0.9 Austria 34.1 Togo 0.7 Canada 35.0 Uganda 0.8 Denmark 33.6 Zambia 1.2 Finland 30.5 Zimbabwe 0.2 France 30.6 Asia Germany 30.4 Bangladesh 1.1 Iceland 35.5 China 4.1 Ireland 37.9 India 2.2 Italy 27.8 Indonesia 3.2 Japan 30.3 Korea, Rep. of 21.3 Netherlands 34.5 Malaysia 11.7 New Zealand 24.6 Pakistan 2.2 Norway 47.5 Philippines 3.0 Portugal 20.0 Sri Lanka 3.4 Spain 27.1 Taiwan, China 26.1 Sweden 32.0 Thailand 7.1 Switzerland 35.2 Vietnam 2.1 United Kingdom 31.4 Latin America United States 41.8 Argentina 10.8 Brazil 8.5 Chile 12.2 Colombia 5.9 Dominican Republic -- Ecuador 6.7 Mexico 11.4 Nicaragua -- Source: International Comparisons Project (World Bank 2008, table 1a). Note: -- not available. 611 612 Distortions to Agricultural Incentives: A Global Perspective Table B.10. Variables in the Global Agricultural Distortions Database, 1955­2007 Variable Description ccode Country code country Country year Year product Product (the word "general" refers to aggregate information) trade_status Product trade status nra Nominal rate of assistance (NRA), by product NRA NRAbms NRAdms NRAinputs, where NRAoutput NRAbms NRAdms nra_o NRAo, nominal rate of assistance to output, by product nra_i NRAi, nominal rate of assistance to input, by product nra_bms NRAbms, nominal rate of assistance to output conferred by border market price support, by product nra_bms_x NRAbms (exportable product), nominal rate of assistance to output conferred by border market price support, by product nra_bms_M NRAbms (importable product), nominal rate of assistance to output conferred by border market price support, by product nra_dms NRAdms, nominal rate of assistance to output conferred by domestic price support, by product nra_covt NRA_covered_products, value of production-weighted average of covered products NRA to covered products can be decomposed into: NRA_COVT = NRA_cov_bms NRA_cov_dms NRA_cov_inputs, where NRA_covered_output NRA_cov_bms NRA_cov_dms nra_cov_i NRA to inputs, value of production-weighted average of covered products nra_cov_o NRA to output, value of production-weighted average of covered products nra_cov_dms NRA to output conferred by domestic market price support, value of production-weighted average of covered products nra_cov_bms NRA to output conferred by border market price support, value of production-weighted average of covered products nra_bms_covm NRA to output conferred by border market price support, value of production-weighted average of covered products, importables nra_bms_covx NRA to output conferred by border market price support, value of production-weighted average of covered products, exportables nra_covh NRA, covered products, value of production-weighted average, nontradables nra_covm NRA, covered products, value of production-weighted average, importables nra_covx NRA, covered products, value of production-weighted average, exportables nra_covt_simpleave NRA, simple average of covered products Global Distortions Database, 1955­2007 613 Table B.10. Variables in the Global Agricultural Distortions Database, 1955­2007 (continued) Variable Description nra_ncm NRA, noncovered products, value of production-weighted average, importables nra_ncx NRA, noncovered products, value of production-weighted average, exportables nra_nch NRA, noncovered products, value of production-weighted average, nontradables nra_nct NRA, noncovered products (total), value of production- weighted average nps Non-product-specific assistance (NPSA), in U.S. dollars NRA_TOTAL accounts for covered and noncovered products nra_tott NRA, all (primary) agriculture, total for covered and noncov- ered and NPSA, value of production-weighted average. nra_totp NRA, all (primary) agriculture, total excluding NPSA nra_totm NRA, all (primary) agriculture, value of production-weighted average, importables nra_totx NRA, all (primary) agriculture, value of production-weighted average, exportables nra_toth NRA, all (primary) agriculture, value of production-weighted average, nontradables nra_agtrad NRA, tradables-only in (primary) agriculture, value of production-weighted average nra_nonagtrad NRA, nonagricultural sectors, tradables rra RRA, relative rate of assistance nra_totd NRA, total agriculture, including NPSA and decoupled payment (in HI countries) nra_agtrad_decpay NRA, agricultural tradables, including decoupled support in HI countries decpay Decoupled paymets (% of total value of production) nps_input Non-product-specific payments (only) to inputs, relevant to HI countries rra_decpay RRA with decoupled payments included, HI countries er_econ Exchange rate, estimated equilibrium, used in calculations q Volume of production, MT pd Domestic producer farmgate price, LC/MT er_prod (if available) product specific exchange rate vop_prod Value of production (at undistorted farm price), by product, U.S. dollars vop_covt Value of production, total covered products (at undistorted farm price), US$ vop_covh Value of production, covered products, nontradables (at undistorted farm price), US$ vop_covm Value of production, covered products, importables (at undistorted farm price), US$ (Table continues on the following page.) 614 Distortions to Agricultural Incentives: A Global Perspective Table B.10. Variables in the Global Agricultural Distortions Database, 1955­2007 (continued) Variable Description vop_covx Value of production, covered products, exportables (at undistorted farm price), US$ vop_nch Value of production, noncovered products, nontradables (at undistorted farm price), US$ vop_ncm Value of production, noncovered products, importables (at undistorted farm price), US$ vop_ncx Value of production, noncovered products, exportables (at undistorted farm price), US$ vop_nct Value of production, noncovered products, total (at undistorted farm price), US$ vop_tot Value of production, total agriculture (at undistorted farm price), US$ cte CTE, by product, US$ cte_covh CTE, all covered products, value of consumption-weighted average, nontradables cte_covm CTE, all covered products, value of consumption-weighted average, importables cte_covx CTE, all covered products, value of consumption-weighted average, exportables cte_covt CTE, total for all covered products, value of consumption- weighted average cte_covt_ave CTE, total for all covered products, simple average voc_prod Value of consumption (at undistorted farm price), by product, US$ voc_covh Value of consumption, covered products, nontradables (at undistorted farm price), US$ voc_covm Value of consumption, covered products, importables (at undistorted farm price), US$ voc_covx Value of consumption, covered products, exportables (at undistorted farm price), US$ voc_covt Value of consumption, total covered products (at undistorted farm price), US$ Derived variables to estimate trade flows ssr Self-sufficiency ratio, from FAO ( Q/C (volume)) or from OECD ( Q/C (value at undistorted prices)) shrimp Share of imports in consumption, FAOSTAT shrexp Share of exports in production, FAOSTAT ac_prod Apparent consumption (volume) pop_agreconact Population total economically active in agricultural (from FAOSTAT) pop_agric Population, agricultural (FAOSTAT) pop_nonagric Population, nonagricultural (FAOSTAT) pop_rural Population, rural (FAOSTAT) pop_total Population, total (FAOSTAT) pop_urban Population, urban (FAOSTAT) pop_toteconact Population, total economically active (FAOSTAT) Source: Anderson and Valenzuela (2008). Global Distortions Database, 1955­2007 615 Table B.11. Variables in the Supplementary Global Agricultural Trade and Welfare Reduction Indexes Database, 1955­2007 Variable Description country Country, including regional aggregates ccode Country code year Year, 1955 to 2007 period Indicator for five-year periods prod2 Product, including product aggregates trstat2 Trade status by product (X exportable, M import- competing) nra NRA, percent cte CTE, percent vop_prod Value of production, at undistorted prices in current US$ voc_prod Value of consumption, at undistorted prices in current US$ tri_covt Trade reduction index (TRI), country-level, all covered tradable products tri_covm TRI, country-level, covered import-competing products tri_covx TRI, country-level, covered exportable products tri_p_covt TRI, country-level, production side of the economy, all covered tradable products tri_c_covt TRI, country-level, consumption side of the economy, all covered tradable products tri_p_covm TRI, country-level, production side of the economy, covered import-competing products tri_c_covm TRI, country-level, consumption side of the economy, covered import-competing products tri_p_covx TRI, country-level, production side of the economy, covered exportable products tri_c_covx TRI, country-level, consumption side of the economy, covered exportable products cc_to_TRI_ctry Commodity-contribution to the country-specific TRI, percent tri_comm_covt TRI, commodity-specific, all covered tradable products tri_comm_covm TRI, commodity-specific, covered import-competing products tri_comm_covx TRI, commodity-specific, covered exportable products tri_comm_p_covt TRI, commodity-specific, production side of the economy, all covered tradable products tri_comm_c_covt TRI, commodity-specific, consumption side of the economy, all covered tradable products tri_comm_p_covm TRI, commodity-specific, production side of the economy, covered import-competing products tri_comm_c_covm TRI, commodity-specific, consumption side of the economy, covered import-competing products (Table continues on the following page.) 616 Distortions to Agricultural Incentives: A Global Perspective Table B.11. Variables in the Supplementary Global Agricultural Trade and Welfare Reduction Indexes Database, 1955­2007 (continued) Variable Description tri_comm_p_covx TRI, commodity-specific, production side of the economy, covered exportable products tri_comm_c_covx TRI, commodity-specific, consumption side of the economy, covered exportable products cc_to_TRI_comm Country-contribution to the commodity-specific TRI, percent wri_covt Welfare reduction index (WRI), country-level, all covered tradable products wri_covm WRI, country-level, covered import-competing products wri_covx WRI, country-level, covered exportable products wri_p_covt WRI, country-level, production side of the economy, all covered tradable products wri_c_covt WRI, country-level, consumption side of the economy, all covered tradable products wri_p_covm WRI, country-level, production side of the economy, covered import-competing products wri_c_covm WRI, country-level, consumption side of the economy, covered import-competing products wri_p_covx WRI, country-level, production side of the economy, covered exportable products wri_c_covx WRI, country-level, consumption side of the economy, covered exportable products cc_to_WRI_ctry Commodity-contribution to the country-specific WRI, percent wri_comm_covt WRI, commodity-specific, all covered tradable products wri_comm_covm WRI, commodity-specific, covered import competing products wri_comm_covx WRI, commodity-specific, covered exportable products wri_comm_p_covt WRI, commodity-specific, production side of the economy, all covered tradable products wri_comm_c_covt WRI, commodity-specific, consumption side of the economy, all covered tradable products wri_comm_p_covm WRI, commodity-specific, production side of the economy, covered import-competing products wri_comm_c_covm WRI, commodity-specific, consumption side of the economy, covered import-competing products wri_comm_p_covx WRI, commodity-specific, production side of the economy, covered exportable products wri_comm_c_covx WRI, commodity-specific, consumption side of the economy, covered exportable products cc_to_WRI_comm Country contribution to the commodity-specific WRI, percent Source: Anderson and Croser (2009). Global Distortions Database, 1955­2007 617 Notes 1. By way of comparison, the seminal multicountry study of agricultural pricing policy by Krueger, Schiff, and Valdés (1991) covers an average of 23 years to the mid-1980s for its 18 focus countries, which together accounted for 5­6 percent of global agricultural output; producer and consumer sup- port estimates in OECD (2008) cover 22 years for 30 countries that account for just over one-quarter of the world's agricultural output valued at undistorted prices. 2. The project's regional books cover Africa (Anderson and Masters 2009), Asia (Anderson and Martin 2009), Latin America (Anderson and Valdés 2008), and European transition economies (Anderson and Swinnen 2008). References Anderson, K., and J. Croser. 2009. National and Global Agricultural Trade and Welfare Reduction Indexes, 1955 to 2007. Supplementary database at http://www.worldbank.org/agdistortions. Anderson, K., and W. Martin, eds. 2009. Distortions to Agricultural Incentives in Asia. Washington, DC: World Bank. Anderson, K., and W. A. Masters, eds. 2009. Distortions to Agricultural Incentives in Africa. Washington, DC: World Bank. Anderson, K., and J. Swinnen, eds. 2008. Distortions to Agricultural Incentives in Europe's Transition Economies. Washington, DC: World Bank. Anderson, K., and A. Valdés, eds. 2008. Distortions to Agricultural Incentives in Latin America. Washington, DC: World Bank. Anderson, K., and E. Valenzuela. 2008. "Estimates of Global Distortions to Agricultural Incentives, 1955 to 2007." World Bank, Washington, DC. http://www.worldbank.org/agdistortions. Dimaranan, B. D. 2007. Global Trade, Assistance and Protection: The GTAP 6 Data Base. West Lafayette: Center for Global Trade Analysis, Purdue University. Krueger, A. O., M. Schiff, and A. Valdés. 1991. The Political Economy of Agricultural Pricing Policy, Volume 1: Latin America, Volume 2: Asia, and Volume 3: Africa and the Mediterranean. Baltimore: Johns Hopkins University Press for the World Bank. Lloyd, P. J., J. L. Croser, and K. Anderson. 2009a. "Global Distortions to Agricultural Markets: New Indicators of Trade and Welfare Impacts, 1960 to 2007." Policy Research Working Paper 4865, World Bank, Washington, DC. ______. 2009b. "How Do Agricultural Policy Restrictions to Global Trade and Welfare Differ Across Commodities?" Policy Research Working Paper 4864, World Bank, Washington, DC. OECD (Organisation for Economic Co-operation and Development). 2008 and earlier years. Agricul- tural Policies in OECD Countries: Monitoring and Evaluation. Paris: OECD. Valenzuela, E., and K. Anderson. 2008. "Alternative Agricultural Price Distortions for CGE Analysis of Developing Countries, 2004 and 1980­84." Research Memorandum 13, Center for Global Trade Analysis, Purdue University, West Lafayette, IN. https://www.gtap.agecon.purdue.edu/ resources/res_display.asp?RecordID=2925. World Bank. 2008. International Comparisons Project PPP Tables for 2005. Washington, DC: World Bank. Index Figures, notes, and tables are indicated by f, n, and t, respectively. A CTEs and, 347­48 Africa, 323­57. See also specific countries diversity across commodities and in Agricultural Distortions Project database, cultures, 354 429­31t, 433­34f, 595 empirical estimates, 19f, 20t, 23t, 26t antiagricultural and antitrade bias/TBI, 36, as measurement methodology, 332­33 39, 327, 340, 347, 351­53, 352f, 353­54 nonfarm sector, 341­42t, 347 conclusions regarding, 353­54 in poor, landlocked countries, 356n6 CTEs for, 44­45t, 332­33, 347­51, 349­50t, plantation export crops, taxation of, 5 356n6, 431t, 434f poverty in, 327, 354, 355 empirical estimates of distortion for, 14, RRAs, 332, 341­42t, 347, 348f, 351, 352f, 354 15­17t, 32f, 33f tax and tariff policies in, 336, 340­44, evolution of trade policies in, 329­31 353­54 exchange rates, 324, 330, 332­33, 341­42t WRIs and TRIs for, 46­48, 47f, 436f, 437t, exports from, 328­29, 354 438, 439f, 440t, 442­49t, 450, 451f, 455n2 future developments in, 53, 354­55 Agenda 2000 reforms of CAP, 142, 163, GDP in, 324, 325­26t, 328, 356n7 173n33, 174n41 general equilibrium effects in, 508, 509t, 512, Agricultural Adjustment Act of 1933 (U.S.), 182 516t, 519­20t, 523t, 529t, 532t, 535, 536t, Agricultural Distortions Project database, 542t, 545t, 549t, 550t, 554, 559t 595­617 growth and structural change in, 328­29 commodities in. See commodities GSE values for, 18f, 41­42t, 333, 343f, 345­46f CTE estimates, 428­35, 431t, 432f, 434f, key economic indicators, 324­27, 325­26t 595­96, 596­97f liberalization of agricultural trade in, 59­60t derived shares in global production, 601­2 measuring distortion in, 331­33 export shares, 605­6t nonfarm sector, 341­42t, 347 gross value of key commodities at undistorted nontraded food staples, 482t, 483­86, 484­86t prices, 599­600t NPS forms of assistance or taxation in, 332, NRA estimates, 428­35, 429­30t, 432f, 433f, 334t, 340f, 341­42t, 353 595­96, 596­97f, 597­98t NRAs for per capita income of focus countries, 611t in Agricultural Distortions Project production exported and consumption database, 429­30t, 433f imported, shares in, 609­10t agriculture, 333­47, 334t, 335f, 337t, 338f, regional coverage of, 595 339t, 340f, 341­42t, 348f regional production shares, 603­4t by commodity, 338f, 339t, 496­98t total shares in global production, 607­8t 619 620 Index Agricultural Distortions Project database South Asia and India; Southeast Asia and (continued ) China; specific countries variables, 612­16t in Agricultural Distortions Project database, WRIs and TRIs compared, 428­35, 429­30t, 429­31t, 433­34f, 595 432f, 433f antitrade bias, decline in, 21 Agricultural Income Disaster Assistance (AIDA) CTEs for, 44­45t, 431t, 434f program (Canada), 198 East Asian economic miracle, 67­70, 103­4 Agricultural Market Transition Act (AMTA), empirical estimates of distortion for, 13­14, U.S., 188 15­17t, 32f, 33t agricultural products. See commodities general equilibrium effects in, 508, 509­10t, Agricultural Stabilization Act of 1958 512, 516t, 519­20t, 522, 523­24t, 529t, (Canada), 196 532­33t, 535, 536t, 542­43t, 545t, 549t, Agricultural Trade and Development Act of 1954 550­51t, 554­55, 559t (U.S.), 182 GSE values for, 18f, 41­42t Ahmed, N., 389n liberalization of agricultural trade in, 28­32, AIDA (Agricultural Income Disaster Assistance) 59­60t program (Canada), 198 nontraded food staples, 482t, 484­86t AMTA (Agricultural Market Transition Act), NRAs for, 19f, 20t, 23­24t, 26t, 429­30t, 433f, U.S., 188 499­500t Anderson, J. E., 8, 43, 420, 422, 487 TBI, 36, 39 Anderson, Kym, 3, 5, 46, 51, 52, 54, 67n, 68, 95, WRIs and TRIs for, 46­48, 47f, 49f, 436f, 437t, 115n, 127, 221, 241, 259, 260, 278, 289, 438, 439f, 440t, 442­49t, 450, 451f, 455n2 323, 359, 389n, 419, 422, 427, 438, 459, Asian financial crisis (1998), 189 460, 477, 480, 487, 505, 507, 508, 538, Association of Southeast Asian Nations 554, 565, 579, 595 (ASEAN), 69, 111n2 Anglo-French Treaty of Commerce (1860), 8 Attwood, E. A, 127 antiagricultural trade bias/antitrade bias. See Australia and New Zealand, 221­56 trade bias index compared to other OECD countries, 221­23 ANZCERTA (Australia New Zealand Closer CTEs for, 44­45t, 245 Economic Relations Trade Agreement), in Agricultural Distortions Project database, 226, 244 431t, 432f Arab Republic of Egypt. See Egypt, Arab economic growth and structural changes, 221, Republic of 223­27 Argentina. See also Latin America and the empirical estimates of distortion for, 13, Caribbean 15­17t, 32, 34t CTEs, 314t, 316t exchange rates, 238, 242­43, 245 evolution of agricultural policy in, 294, farm MFP growth in, 248, 249f, 253n27 295, 296 farmland prices, subsidies affecting, 248, 250f general equilibrium effects in, 510t, 516t, future developments in, 53, 54 524t, 533t, 536t, 543t, 545, 549, 551t, 554, GDP in, 222f, 223, 225­26 560t, 561n13 general equilibrium effects in, 511t, 517t, growth and structural changes in, 292, 293 525t, 534t, 537t, 544t, 552t, 560t GSE values, 307f, 308t GSE values for, 18f, 41­42t income inequality in, 385n3 historical distortions of agricultural market key commodities, 471f, 473f, 475f, in, 9­10 493f, 501­2t key commodities key economic indicators, 290, 291t Australia, 471f, 472, 473f, 475f, 476f liberalization of agricultural trade in, 53 New Zealand, 471f, 472, 473f, 475f NRAs for, 21, 298, 299, 300t, 301, 301f, 304t, 306 liberalization of agricultural trade in, 59­60t, poverty in, 320 233, 237, 239­42, 244­46 return to export taxation in, 318 nonfarm sectors, 224­26, 234­35t, 236f RRA/TBI, 312f nontraded food staples, 482t, 484t WRIs and TRIs for, 441f, 445t, 448t, 452f NRAs for, 227 Armington elasticities, 513, 561n10 before early 1970s, 229, 233 ASEAN (Association of Southeast Asian reforms of early 1970s, 233, 237 Nations), 69, 111n2 in Agricultural Distortions Project Asia. See also Eastern Europe and Central Asia; database, 429­30t, 432f Northeast Asian advanced economies; covered commodities, 229­31t, 232f Index 621 empirical estimates, 19f, 20t, 24t, 26t TBI, 406­7t, 408 evolution of agricultural policy and, trade in agricultural products in, 391­93 243­44, 245, 246, 250 trade status of farm commodities in, nonfarm products, 234­35t, 236f 404, 405t PDIs, 456n7 WRIs and TRIs for, 49f, 441f, 444t, 447t policy interventions and outcomes Baylis, K., 208 before early 1970s, 228­33 beef. See livestock reforms of early 1970s, 233­38 Belgium, 126, 132, 172n14­15, 173n21. See also antiagricultural trade bias, 223, 227, 233, Western Europe 237, 248 Benelux Union, 172n14­15, 173n21 conclusions regarding, 247­51 Benin, 325t, 328, 345­46t, 349t, 476f, 496­98t further policy reform prospects in, Bhagwati, J. N., 566 246­47 Bhutan, 390, 393 historical overview of, 227­38 biofuel production, food price rise related reasons for evolution of, 238­46 to subsidization of (2007­2008), 3, 6­7 RRA, 227, 233, 234­35t, 236f, 237 Bismarck, Otto von, 108, 121 RRAs for, 32 Botswana, 331 tax and tariff policies in, 222, 225, 226, 228, Bovine Spongiform Encephalopathy (BSE), 233, 242, 245­46 161, 198 Western European agriculture and, 115, 120, Brannan Plan (1948, U.S.), 182 121, 122, 132, 162, 174n37, 243 Brazil. See also Latin America and the Caribbean WRIs and TRIs for, 436f, 437t, 439f, 440t, CTEs, 314t, 316t 441f, 442­49t, 451f, 456n7 exchange rate, 320n3 Australia New Zealand Closer Economic general equilibrium effects in, 510t, 516t, Relations Trade Agreement 524t, 533t, 536t, 543t, 549, 551t, 560t (ANZCERTA), 226, 244 growth and structural changes in, 292, 293 Austria. See also Western Europe GSE values, 307f, 308t CTEs, 158­59t income inequality in, 385n3 current chronology of policy distortions key commodities, 471f, 472, 473f, 475f, 476f, (1955­2007) 493f, 501­2t 1955­1964, 127, 128f key economic indicators, 290, 291t 1965­1974, 132, 133f, 134 measurement of distortions in, 579 1975­1984, 136, 137f NRAs, 298, 299, 300t, 301f, 304t, 306 1985­1994, 139, 140f Portugal and, 129, 173n23 EEC, decision not to join, 173n26 PSEs for, 6 GSE values, 155­56t RRA/TBI, 311, 312f, 313 NRAs in, 154t WRIs and TRIs for, 441f, 445t, 448t WRIs and TRIs for, 441f, 446t, 449t Brunei Darussalam, 500t BSE (Bovine Spongiform Encephalopathy), B 161, 198 Balderstone Report (1982; Australia), 239 Bulgaria, 172n2. See also Central and Eastern Balkan countries, 264, 285 European (CEE) countries; Eastern Baltic Free Trade Area, 284 Europe and Central Asia Baltic states. See Estonia; Latvia; Lithuania antiexport bias/TBI, 273t Bandara, J., 389n CTEs, 279t Bangladesh, 389­415. See also Asia; South Asia general equilibrium effects in, 510t, 517t, and India 524t, 533t, 537t, 543t, 551t, 558t exchange rates in, 390, 399­401, 408, 410, GSE values, 277t 413, 414n7 key commodities, 471f, 473f, 475f GDP in, 389­91, 391t, 393, 394, 397, key economic indicators in, 261t 412, 414n1 NRAs during transition period (from early general equilibrium effects in, 510t, 516t, 1990s), 268, 269t, 270, 271f, 274t 524t, 532t, 535, 536t, 542t, 551t, 559t political economy of policy choices in, 281, key commodities, 471f, 475f, 494f, 499t 282, 283, 284 NRAs for, 56, 401­11, 402­3f, 406­7t reform era (after 1989) in, 266, 267 policy interventions and outcomes RRAs, 278, 280f current practices, 393­401 WRIs and TRIs for, 441f, 445t, 448t future policies, 411­14 Burkina Faso, 476f, 496­98t 622 Index C RRA/TBI, 312f, 313 CAIS (Canadian Agricultural Income WRIs and TRIs for, 441f, 445t, 448t Stabilization) Program, 198­99 China, 359­87 Cameroon. See also Africa conclusions regarding, 379­82 CTEs, 349­50t CTEs, 363­64, 371, 375, 376­77t, 382 growth and structural change in, 328 domestic price stabilization, 375­79 GSE values, 343f, 345­46t East Asian economic miracle, as part of, 69 key commodities, 475f, 476f, 496­98t economic growth and structural change in, key economic indicators, 325t 361­63 NRAs for, 333, 334t, 335f, 337t exchange rates, 286n8, 363, 364, 371 WRIs and TRIs for, 441f, 444t, 447t future developments in, 53, 382­85 Canada. See North America GDP in, 361­62, 370t, 377t, 383f Canadian Agricultural Income Stabilization general equilibrium effects in, 508, 509t, 515, (CAIS) Program, 198­99 516t, 522, 523t, 532t, 536t, 542t, 549, CAP. See Common Agricultural Policy 550t, 559t (CAP), EU GSE values for, 19, 364, 370t Cape Verde, 331 income inequality in, 385n3 Caribbean countries. See Dominican Republic; key commodities, 471f, 472, 473f, 475f, 476f, Latin America and the Caribbean 477, 493­94f, 499­500t cattle. See livestock liberalization of agricultural trade in, 32, 59, CDI (consumer distortion index), WRI 379­80 computed from, 421, 435­38, market economy, move to, 385n4 443t, 455 measurement of distortions in, 579 Central and Eastern European (CEE) countries. nominal rates of protection for key See also Bulgaria; Czech Republic; commodities in, 6 Eastern Europe and Central Asia; nonfarm sectors, 371, 372f, 373­74t Estonia; Hungary; Latvia; Lithuania; North American market and, 189 Poland; Romania; Slovak Republic; NRAs for. See under Southeast Asia and China Slovenia; Turkey opening of economy in, 60n1 antiexport bias/TBI, 273t poverty in, 360t compared to CIS countries, 260 poverty reduction in, 59 EU, accession to, 142, 167­68, 171, 172n2, PSEs for, 6, 580 259, 260, 282, 283, 285, 286 RRAs, 39, 363, 371, 372f, 373­74t, 380­83, GDP in, 262, 265 381f, 383f, 385, 386n8 measuring distortions to agricultural TBI, 380­82, 381f incentives in, 264 time periods for which data are NRAs during transition period (from early available, 580 1990s), 268­70, 272­73, 274­75t WRIs and TRIs for, 48, 49f, 441f, 444t, 447t, political economy of policy choices in, 280, 450, 452f 282­84 WTO, accession to, 55­56, 56f, 384, 508 reform era (after 1989) in, 266 CIS. See Commonwealth of Independent States Central European Free Trade Area, 284 (CIS) countries cereals. See corn; grains; rice; wheat classification of products as import-competing, CGE (computable general equilibrium), 46, 421, exportable, or nontradable, 582­84 455n4, 506 CMEA (Council of Mutual Economic Chad, 325t, 328, 345­46t, 349t, 476f, 496­98t Assistance), 266 Chen, S., 59 collective action problem, 208, 351­53 Chile. See also Latin America and the Caribbean Colombia. See also Latin America and the CTEs, 314t, 316t Caribbean evolution of agricultural policy in, 295, 296 CTEs, 313, 314t, 316t general equilibrium effects in, 510t, 516t, evolution of agricultural policy in, 295, 296 524t, 533t, 536t, 543t, 551t, 560t general equilibrium effects in, 510t, 516t, growth and structural changes in, 292, 293 524t, 533t, 536t, 543t, 549, 551t, 560t GSE values, 307f, 308t growth and structural changes in, 292 key commodities, 471f, 473f, 475f, 501­2t GSE values, 307f, 308t key economic indicators, 290, 291t key commodities, 471f, 473f, 474, 475f, 476f, liberalization of agricultural trade in, 11 493­94f, 501­2t NRAs, 298, 299, 300t, 301, 301f, 304t, 306 key economic indicators, 290, 291t Index 623 NRAs, 298, 299, 300t, 301, 301f, 304t, 306 total shares in global production, 607­8t RRA/TBI, 311, 312f WRIs and TRIs, 487­95, 489­90t, 491­94f WRIs and TRIs for, 441f, 445t, 448t Common Agricultural Policy (CAP), EU, 115 commodities, 459­503. See also corn; cotton; 1955­1964, 124, 125, 126, 129 grains; livestock; milk and dairy; 1965­1974, 130, 132 oilseeds; potatoes; rice; sugar; wheat 1975­1984, 134­36 CTEs, 475­77 1985­1994, 136, 138 Agricultural Distortions Project database 1995­2007, 141­42 estimates, 596­97f Agenda 2000 reforms, 142, 163, empirical estimates, 27f 173n33, 174n41 global value of production, percentage conclusions regarding, 170­71 coverage of, 460­62 Eastern Europe and Central Asia measurement methodology, 584­85 affected by, 280 time series, 478­79t EFTA versus, 135 derived shares in global production, 601­2 Fischler reforms, 142, 174n41­42 export shares, 605­6t future developments, 168­70 exportable versus import-competing MacSharry reforms (1994­96), 141, 142, 165 products, 21, 22f, 582­84 as major transformational policy, 60n1, 124 general global equilibrium effects other policies affecting, 117 product prices, 530t, 548t policy trends and turning points, 153, 163­66 quantities produced and traded, 521t, 522, widespread adoption of, 116 528t, 540, 541t, 546­47t Common Fisheries Policy, EU, 173n31 gross value at undistorted prices, 599­600t Commonwealth of Independent States (CIS) GSE values, 462­69, 465f, 466­68t, 469f countries. See also Eastern Europe and measurement methodology for. See under Central Asia; Kazakhstan; Kyrgyz measurement methodology Republic; Russian Federation; Tajikistan; nontraded food staples of low-income Turkmenistan; Ukraine; Uzbekistan countries, 482t, 483­86, 484­86t compared to CEE countries, 260 NRAs, 460­62 GDP in, 262 Africa, 338f, 339t, 496­98t measuring distortions to agricultural Agricultural Distortions Project database incentives in, 264 estimates, 596­97f NRAs during transition period (from early Asia, 367t, 369f, 499­500t 1990s), 270, 273 Australia and New Zealand, 229­31t, 232f political economy of policy choices in, 280, coconut, cocoa, cotton, and coffee, 469f, 281, 282, 285 470t, 474, 476f reform era (after 1989) in, 266 Eastern Europe and Central Asia, 272t Communist era, Eastern Europe and Central empirical estimates, 27f Asia during, 264­66 global value of production, percentage computable general equilibrium (CGE), 46, 421, coverage of, 461­62, 461t, 463t, 464t 455n4, 506 GSE values and, 462­69, 465f, 466­68t, 469f consumer distortion index (CDI), WRI guesstimates for noncovered products, 586 computed from, 421, 435­38, 443t, 455 key covered products, 469f, 470t consumer support estimates (CSEs), 11, 60n6, Latin America and the Caribbean, 302f, 85, 455n1 303­4t, 501­2t consumer tax equivalents (CTEs), 7, 12­13. See livestock, 40t, 469f, 472, 473f also measurement methodologies, and rice, milk, and sugar, 469­72, 469f, 470t, under specific countries and regions 471f in Agricultural Distortions Project database, time series for, 469, 470t 428­35, 431t, 432f, 434f, 595­96, wheat, maize, and soybeans, 469f, 596­97f 470t, 472­74, 475f by commodity. See under commodities peanuts, in North America, 216 empirical estimates of distortion using, 43, price variability/stability, effects of interven- 44­45t tion on, 477­83, 480­81f macroeconomic modeling study, 51 production exported and consumption WRIs and TRIs compared, 46, 420­22, 424, imported, shares in, 482t, 483, 609­10t 438, 454­55 regional production shares, 603­4t Copenhagen Consensus project, 62n19 tobacco, 180, 214, 216, 237, 244 Corden, W. M., 566 624 Index corn Cyprus, 172n2 CTEs for, 478­79t Czech Republic. See also Central and Eastern derived shares of global production, 601t European (CEE) countries; Eastern export shares, 605t Europe and Central Asia global value of production, NRA percentage antiexport bias/TBI, 273t coverage of, 461t, 463t, 464t CTEs, 279t gross value of global production at general equilibrium effects in, 510t, 517t, undistorted prices, 599t 524t, 533t, 537t, 543t, 551t, 558t GSE values, 465f, 466­68t, 469f GSE values, 277t in North America, 180, 185, 212 key commodities, 471f, 473f, 475f, 494f NRAs for, 469f, 472­74, 475f key economic indicators in, 261t production export and consumption NRAs during transition period (from early imported, shares in, 482t, 609t 1990s), 269t, 271f, 274t regional production shares, 603t political economy of policy choices in, 284 total shares in global production, 607t reform era (after 1989) in, 266 U.K. Corn Laws (1815­1846), 8, 9, 108, 120 RRAs, 280f WRIs and TRIs, 488, 489­90t, 491f WRIs and TRIs for, 441f, 445t, 448t Côte d'Ivoire. See also Africa cocoa trade in, 50 D CTEs, 349­50t dairy products. See milk and dairy GSE values, 343f, 345­46t de Gaulle, Charles, 172n18 key commodities, 471f, 474, 476f, 493f, de Gorter, H., 278 496­98t de Nicola, Francesca, 67n, 115n, 177n, 221n, key economic indicators, 325t 259n, 289n, 323n, 359n, 389n, 595n NRAs for, 334t, 335f, 336, 337t, 340 Deardorff, A. V., 224 RRA/TBI, 352f Denmark. See also Western Europe WRIs and TRIs for, 441f, 444t, 447t CTEs, 153, 158­59t Cotonou Agreement, 582 current chronology of policy distortions cotton (1955­2007) in Africa, 324 1955­1964, 126, 127, 128f, 129 derived shares of global production, 602t 1965­1974, 131, 132, 133f in Eastern Europe and Central Asia, 271, 281 1975­1984, 134 export shares, 605t EU, accession to, 170 general global equilibrium effects, 522, 540 exports, contribution of agriculture to, 118 global value of production, NRA percentage GSE values, 153, 155­56t coverage of, 461t, 463t, 464t historical distortions of agricultural market gross value of global production at in, 9, 120­22 undistorted prices, 599t NRAs in, 154t GSE values, 465f, 466­68t, 469f productivity in, 172n8 in North America, 212 size of farms in, 118 NRAs for, 469f, 470t, 474, 476f WRIs and TRIs for, 441f, 446t, 449t production export and consumption Depression, 121­22, 129, 179, 180, 181, 196, imported, shares in, 482t, 609t 238, 242 regional production shares, 604t Diewert, W. E., 224 in Southeast Asia and China, 363, 384, 386n13 disparity problem in middle-income total shares in global production, 608t economies, 68, 100 in Western Europe, 502n2 distortions to agricultural incentives, 3­64. WRIs and TRIs, 489­90t, 491­94f See also specific countries and regions Council of Mutual Economic Assistance by commodity, 459­503. See also commodities (CMEA), 266 countries included in study, 7 Crafts, N., 259 empirical estimates of, 13­50. See also Creutzfeldt-Jakob (nvCJD) disease, 161 empirical estimates of distortions Croser, Johanna L., 46, 67n, 115n, 177n, 221n, to agricultural incentives 259n, 289n, 323n, 359n, 389n, 419, 427, future developments, 53­57, 56f, 57t 459, 487, 505n, 595n future research areas, 58­59 CSEs (consumer support estimates), 11, 60n6, general global equilibrium effects of, 51­53, 85, 455n1 505­63. See also general global CTEs. See consumer tax equivalents equilibrium effects Index 625 historical background, 4, 8­11 empirical estimates of distortion for, 13, importance of studying, 6­8 15­17t, 34t, 60n4 macroeconomic effects of past reforms EU and remaining policies, 51­53 accession to, 142, 167­68, 171, 172n2, 259, measurement methodology, 5­8, 11­13, 260, 282, 283, 285, 286 565­94. See also measurement CAP, effects of, 280 methodology trade agreements with, 168 policy-based nature of, 3­6 exchange rates, 263­68, 286n3 reforms reversing. See liberalization of future reductions in distortions, 285­86 agricultural trade GDP in, 260, 261t, 262, 265 single-indicator index, value of, 420 general equilibrium effects in, 508, 510­11t, World Bank project, 595­617. See also 512, 517t, 519­20t, 533t, 537t, 543t, 545t, Agricultural Distortions Project 549t, 551­52t, 558t database growth and structural changes in, 260­63 Doha Round GSE values for, 18f, 41­42t, 263, 274, 277t African agricultural policy and, 353 international agreements and institutions, Australia/New Zealand agricultural policy impact of, 283­85 and, 242 key economic indicators, 261t Eastern European and Central Asian agricul- measuring distortions to agricultural tural policy and, 286 incentives in, 263­64 general global equilibrium effects and, 535, nontraded food staples, 482t, 484t 538, 557 NPS forms of assistance or taxation in, 263, importance of agricultural policy reform to, 6, 273, 275t, 276t 54, 55, 57 NRAs for North American agricultural policy and, in Agricultural Distortions Project 183, 215 database, 429­30t, 433f Southeast Asian and Chinese agricultural empirical estimates, 19f, 20t, 24t, 26t policy and, 385, 387n11 key covered commodities, 272t Western European agricultural policy and, measuring distortions by, 263 168­69 during transition period (from early domestic trading costs, measuring, 574 1990s), 267­78, 268f, 269t, 271­72f, Dominican Republic. See also Latin America and 274­76t the Caribbean policy interventions and outcomes CTEs, 314t, 316t during Communist era, 264­66 evolution of agricultural policy in, 295 political economy of, 278­85 exchange rate distortions, 321n5 during reform era (after 1989), 266­67 growth and structural changes in, 292 during transition period (from early GSE values, 307f, 308t 1990s), 267­78, 268f, 269t, 271­72f, key commodities, 471f, 473f, 476f, 501­2t 274­76t key economic indicators, 290, 291t PSEs for, 6 NRAs, 297, 298, 299, 300t, 301f, 304t rent extraction in, 280­82 RRA/TBI, 311, 312f RRAs, 263, 275t, 276t, 278, 280f WRIs and TRIs for, 441f, 445t, 448t WRIs and TRIs for, 47f, 436f, 437t, 439f, 440t, Dorosh, P. A., 389n 442­49t, 451f, 455n2 Dupuit, J., 456n8 WTO accession in, 260, 284 Dutch disease, 260 economic welfare prospective analysis of, 535­38, 536­37t, 539t E retrospective analysis of, 515­18, 516­17t, EAEC (Eurasian Economic Community), 284 519­20t East Asia. See Asia; Northeast Asian advanced economy-wide equilibrium effects. See general economies; Southeast Asia and China global equilibrium effects East Asian economic miracle, 67­70, 103­4 Ecuador. See also Latin America and Eastern Europe and Central Asia, 259­88. See the Caribbean also specific countries CTEs, 313, 314t, 316t in Agricultural Distortions Project database, exchange rate distortions, 321n5 429­31t, 433­34f, 595 general equilibrium effects in, 510t, 516t, antiexport bias/TBI, 259, 270­71, 273f 524t, 533t, 535, 536t, 543t, 549, 551t, CTEs for, 44­45t, 263, 278, 279t, 431t, 434f 556, 560t 626 Index Ecuador (continued ) WRIs and TRIs for, 441f, 445t, 448t growth and structural changes in, 292 WTO accession, 284 GSE values, 307f, 308t Ethiopia. See also Africa key commodities, 471f, 473f, 475f, 476f, CTEs, 348, 349­50t 501­2t evolution of trade policies in, 329 key economic indicators, 290, 291t growth and structural change in, 328 NRAs, 297, 298, 300t, 301f, 304t GSE values, 343f, 345­46t RRA/TBI, 312f key economic indicators, 325t WRIs and TRIs for, 441f, 445t, 448t NRAs for, 334t, 335f, 337t, 340 EFTA. See European Free Trade Association WRIs and TRIs for, 441f, 444t, 447t Egypt, Arab Republic of. See also Africa Eurasian Economic Community (EAEC), 284 CTEs, 348, 349t European Free Trade Association (EFTA) evolution of trade policies in, 331 1955­1964, 126, 127, 129 general equilibrium effects in, 509t, 516t, 1965­1974, 132 523t, 532t, 536t, 542t, 550t, 559t 1975­1984, 135, 136 GSE values, 343f, 345­46t 1985­1994, 138­41 key commodities, 471f, 472, 473f, 475f, 476f, 1995­2007, 143 493­94f, 496­98t autonomous agricultural policies of key economic indicators, 325t members, 124 NRAs for, 334t, 335f, 337t, 340 members of, 146f RRA/TBI, 352f nonfarm sectors, 147­48t WRIs and TRIs for, 441f, 444t, 447t NRA analysis and, 147­48t, 153 Elbadawi, Ibrahim, 565n policy interventions and outcomes, empirical estimates of distortions to agricultural 163, 165 incentives, 13­50 WRIs and TRIs for, 450, 451f CTEs, 43, 44­45t European transition economies. See Eastern GSE values, 14­21, 17­18f, 41­42t Europe and Central Asia nonfarm sectors, 28­32, 29­30t, 31f, 33­35t European Union. See Western Europe NPS assistance, 21, 27­28, 29­30t Everything but Arms agreement, 168, 582 NRAs, 7, 11­13, 19­39. See also nominal rates exchange rates of assistance Africa, 324, 330, 332­33, 341­42t producer assistance and consumer taxation in Australia and New Zealand, 238, 242­43, 245 value terms, 40­43, 41­42t, 44­45t distortion of, 10, 12 by region, 13­14, 15­17t dual/multiple, 10 RRAs, 28­39, 31­32f, 33­35t, 36­40f Eastern Europe and Central Asia, TBI, 21, 23­25t, 36, 40f, 43 263­68, 286n3 trade restrictiveness indexes, 43­50. See also floating of, 60n1 welfare reduction and trade reduction Latin America and the Caribbean, 295, 297, indexes 299, 318, 320n3, 321n5 environmental objectives measurement methodology, 569­72, 569f, in Australia and New Zealand, 246­47 585­86 in North America, 191 Northeast Asian advanced economies, 74, 85 in Western Europe, 163­64 real exchange rate changes, 571­72 Erdogan, Ceren, 221n in South Asia and India, 390, 396­401, 398f, Estonia. See also Central and Eastern European 408, 410, 413, 414n7 (CEE) countries; Eastern Europe and in Southeast Asia and China, 286n8, 363, Central Asia 364, 371 antiexport bias/TBI, 273t Western Europe, 119, 124, 130, 131, 134, 165 CTEs, 279t exports general equilibrium effects in, 510t, 517t, from Africa, 328­29, 354 524t, 533t, 537t, 543t, 551t global production shares by commodity, GSE values, 277t 605­6t key commodities, 471f, 473f, 475f import-competing versus exportable key economic indicators in, 261t products, 21, 22f, 582­84 NRAs during transition period (from early from Latin America and the Caribbean, 293 1990s), 269t, 271f, 274t as share of production (1961­2004), 59t reform era (after 1989) in, 266 South Asia and India, agricultural trade in, RRAs, 280f 391­93 Index 627 U.S. subsidies for, 181­83, 186­87 free trade. See liberalization of agricultural trade WRI and TRI for exportables subsector, Furtan, H., 208 427­28 G F Gardner, Bruce L., 177, 192, 565n factor rewards GDP. See gross domestic product prospective analysis of, 545­48, 549t General Agreement on Tariffs and Trade retrospective analysis of, 529­31, 531t (GATT). See also Uruguay Round FAO (Food and Agriculture Organization), 13, Agreement on Agriculture 262, 292, 333, 362, 373, 483 discussion of agricultural distortions begun Farm Act of 1996 (U.S.), 188, 190, 212, 216 by, 138 Farm Act of 2002 (U.S.), 189­91, 216 domestic price stabilization efforts and, Farm Income Protection Act of 1991 379, 383 (Canada), 197 Haberler report (1958), 4 farm multifactor productivity (MFP) growth, Kennedy Round, 173n25 248, 249f, 253n27 lax treatment of agricultural protections farm products. See commodities under, 55, 56f Farm Security and Rural Investment Act of 2002 liberalization of agricultural trade via, 9 (U.S.), 190, 214 North American agricultural policy and, 182 FCI (Food Corporation of India), 391 Western European agricultural policy and, Feenstra, R. C., 420, 487 117, 170 Finland. See also Western Europe WTO replacing, 79, 117 CTEs, 158­59t general global equilibrium effects, 51­53, current chronology of policy distortions 505­63 (1955­2007) by commodity 1955­1964, 129 product prices, 530t, 548t 1965­1974, 132, 133f, 134 quantities produced and traded, 521t, 522, 1975­1984, 135, 136, 137f 528t, 540, 541t, 546­47t 1985­1994, 139, 140f comparison of results with different GTAP EEC, decision not to join, 173n26 data, 554­55 GDP, share of agricultural output in, 118 conclusions regarding, 555­57 GSE values, 155­56t data used for, 506 NRAs in, 154t economic welfare WRIs and TRIs for, 441f, 446t, 449t prospective analysis of, 535­38, Fischler, Franz, 142, 174n41­42 536­37t, 539t Food and Agriculture Organization (FAO), 13, retrospective analysis of, 515­18, 516­17t, 262, 292, 333, 362, 373, 483 519­20t Food Corporation of India (FCI), 391 factor rewards food processing prospective analysis of, 545­48, 549t distortions, measurement methodology, retrospective analysis of, 529­31, 531t 575­76 GTAP database, 52, 506­8, 513, 538, 554­55, nondistortionary trade costs of, 574 558­60t, 561n1, 561n3, 561n7, Forbes, Rod, 221n 561n9, 561n12 France. See also Western Europe key distortions in global markets, 507­12, CTEs, 158­59t 509­11t current chronology of policy distortions Linkage model, 247, 454, 506, 513­14 (1955­2007) NRAs and, 508, 514, 529 1955­1964, 124­29, 128f, 172n10 product prices 1965­1974, 131, 132 prospective analysis of, 545­48, 546t 1975­1984, 134, 136 retrospective analysis of, 522­29, 530t GSE values, 153, 155­56t prospective analysis of, 535­55 historical distortions of agricultural market economic welfare, 535­38, 536­37t, 539t in, 4, 8­9, 108, 121­23 product and factor prices, 545­48, key commodities in EU-15, 471f, 473f, 475f, 546t, 549t 493­94f quantities produced and traded, 538­40, NRAs in, 154t 541­47t policy trends and turning points, 161, 166 sectoral value added, 548, 550­53t WRIs and TRIs for, 441f, 446t, 449t, 450, 452f quantities produced and traded 628 Index general global equilibrium effects (continued ) grains. See also corn; rice; wheat prospective analysis of, 538­40, 541­47t in Australia and New Zealand, 242 retrospective analysis of, 518­22, 521t, CTEs for, 478­79t 523­29t derived shares of global production, 601t retrospective analysis of, 514­35 in Eastern Europe and Central Asia, 281 economic welfare, 515­18, 516­17t, export shares, 605t 519­20t general global equilibrium effects, 521t, 528t, factor rewards, 529­31, 531t 530t, 540, 541t, 546­48t product prices, 522­29, 530t global value of production, NRA percentage quantities produced and traded, 518­22, coverage of, 461t, 463t, 464t 521t, 523­29t gross value of global production at sectoral value added, 531­35, 532­34t undistorted prices, 599t RRAs and, 512 GSE values, 465f, 466­68t, 469f sectoral value added NRAs for, 469­74, 469f, 470t, 471f, 475f prospective analysis of, 548, 550­53t production export and consumption retrospective analysis of, 531­35, imported, shares in, 482t, 609t 532­34t regional production shares, 603t WRIs and TRIs compared, 454 total shares in global production, 607t Generalized System of Preferences, 582 in Western Europe, 161 genetically modified (GM) foods, 246 WRIs and TRIs, 488, 489­90t, 491­94f Georgia (nation), 284 WTO commodity forecast, 386n12 Germany. See also Western Europe Great Depression, 121­22, 129, 179, 180, 181, CTEs, 158­59t 196, 238, 242 current chronology of policy distortions Greece, 4, 118, 130, 135. See also Western Europe (1955­2007) GRIP (Gross Revenue Insurance Program), 1955­1964, 125­27, 128f, 129, 172n9 Canada, 197­98 1965­1974, 131 gross domestic product (GDP) 1975­1984, 134, 135 in Africa, 324, 325­26t, 328, 356n7 exports, contribution of agriculture to, 118 in Australia and New Zealand, 222f, 223, GDP, share of agricultural output in, 118 225­26 GSE values, 155­56t in Eastern Europe and Central Asia, 260, 261t, historical distortions of agricultural market 262, 265 in, 4, 9, 108, 109, 120­23 in high-income countries, 222f key commodities in EU-15, 471f, 473f, 475f, in Latin America and the Caribbean, 289, 290, 493­94f 291t, 292­93, 295, 298, 306 NRAs in, 154t in Northeast Asian advanced economies, 63, policy trends and turning points, 166 71­73, 72t, 78, 79, 84, 100­103, 101­2f, WRIs and TRIs for, 441f, 446t, 449t, 104, 105t, 111n4, 112n15 450, 452f in Southeast Asia and China, 361­62, 370t, Ghana. See also Africa 377t, 383f CTEs, 349t in Western Europe, 117­18, 172n8 empirical estimates of distortion for, 60n4 WRI and, 450, 453f evolution of trade policies in, 330 Gross Revenue Insurance Program (GRIP), GSE values, 343f, 345­46t Canada, 197­98 key commodities, 471f, 476f, 496­98t gross subsidy equivalent (GSE) values, 12 key economic indicators, 325t for Africa, 18f, 41­42t, 333, 343f, 345­46f NRAs for, 333, 334t, 335f, 337t for Australia and New Zealand, 18f, RRA/TBI, 352f 41­42t WRIs and TRIs for, 441f, 444t, 447t by commodity, 462­69, 465f, 466­68t, 469f Global Distortions Database. See Agricultural for Eastern Europe and Central Asia, 18f, Distortions Project database 41­42t, 263, 274, 277t global equilibrium effects. See general global empirical estimates of distortion, 14­21, equilibrium effects 17­18f, 41­42t Global Trade Analysis Project (GTAP), 52, for Latin America and the Caribbean, 18f, 506­8, 513, 538, 554­55, 558­60t, 561n1, 41­42t, 298, 307f, 308­10t, 318 561n3, 561n7, 561n9, 561n12 for North America, 18f, 19, 41­42t, 204, GM (genetically modified) foods, 246 206­7t Godo, Y., 68, 100 for Western Europe, 153, 155­57t Index 629 GTAP (Global Trade Analysis Project), 52, imports 506­8, 513, 538, 554­55, 558­60t, 561n1, exportable versus import-competing 561n3, 561n7, 561n9, 561n12 products, 21, 22f, 582­84 Gulati, Ashok, 389 to Latin America and the Caribbean, 295 Gupta, K., 389n as share of apparent consumption (1961­2004), 59t H South Asia and India, agricultural trade in, Haberler report (1958) to GATT, 4 391­93 Harberger, A., 456n8, 566 tariffs and subsidies on, measuring, 568 Hayami, Yujiro, 67, 68, 99, 100, 103, WRI and TRI for import-competing 127, 278, 579 subsector, 422­27 Hayek, F. A., 477, 481 India, 389­415 Heckscher-Ohlin-Samuelson model, 224 at Doha Round, 386n11 heterogeneity in agricultural sector, 250­51 exchange rates in, 390, 396­99, 397f Hong Kong, China, 68­69, 517t, 525t, 534t, 537t, GDP in, 389­91, 391t, 393, 394, 397, 544t, 553t, 561n11 412, 414n1 Honma, Masayoshi, 67, 68, 95, 99, 100, 103 general equilibrium effects in, 510t, 516t, Howarth, R. W., 127 524t, 533t, 536t, 543t, 549, 551t, 555, Hungary. See also Central and Eastern European 557, 559t (CEE) countries; Eastern Europe and GSE values for, 19 Central Asia input subsidies, 394­95, 395t, 409, 591n21 antiexport bias/TBI, 273t key commodities, 471f, 475f, 476f, 491, in Communist era, 265 493­94f, 499­500t CTEs, 279t liberalization of agricultural trade in, 11, 32 general equilibrium effects in, 510t, 517t, nominal rates of protection for key 524t, 533t, 537t, 543t, 551t, 558t commodities in, 6 GSE values, 277t NRAs for, 56, 401­11, 402­3f, 406­7t key commodities, 471f, 473f, 475f policy interventions and outcomes key economic indicators in, 261t current practices, 393­401 NRAs during transition period (from early future policies, 411­14 1990s), 269t, 271f, 274t poverty in, 59, 360t reform era (after 1989) in, 266 PSEs for, 580 RRAs, 280f RRAs, 381f, 383f, 385 WRIs and TRIs for, 441f, 445t, 448t tax and tariff policies, 397­99, 398f WTO accession, 284 TBI, 381f, 406­7t, 408 trade in agricultural products in, 391­93 I trade status of farm commodities in, 404, 405t Iceland. See also Western Europe WRIs and TRIs for, 49f, 441f, 444t, 447t, autonomous agricultural policy of, 124, 172n2 450, 452f Common Fisheries Policy, EU, 173n31 Indonesia CTEs, 158­59t conclusions regarding, 380 current chronology of policy distortions CTEs, 363­64, 371, 375, 376­77t, 382 (1955­2007) at Doha Round, 386n11 1955­1964, 124, 130 domestic price stabilization, 375­79 1975­1984, 136, 137f economic growth and structural change in, 1985­1994, 140f, 141 361­63 1995­2007, 143, 144f GDP in, 282f, 361­62, 370t, 377t GDP, share of agricultural output in, 118 general equilibrium effects in, 509t, 516t, GSE values, 155­56t 523t, 532t, 536t, 542t, 551t, 559t high levels of agricultural protection GSE values, 364, 370t in, 174n44 income inequality in, 385n3 key commodities, 471f, 473f key commodities, 471f, 473f, 474, 475f, 476f, nonfarm sectors, 148t 493­94f, 499­500t NRAs in, 148t, 153, 155t nominal rates of protection for key WRIs and TRIs for, 441f, 446t, 449t, 450 commodities in, 6 IFPRI (International Food Policy Research nonfarm sectors, 371, 372f, 373­74t Institute), 580 NRAs for. See under Southeast Asia and China IMF (International Monetary Fund), 330­31, 420 PSEs for, 580 630 Index Indonesia (continued ) key commodities, 471f, 473f, 475f, 491, RRAs, 363, 371, 372f, 373­74t, 380­83, 381f, 493­94f 383f, 385, 386n8 liberalization of agricultural trade in, 59­60t TBI, 380­82, 381f NRAs in, 19f, 20t, 21, 24t, 26t, 55, 56f, 87t, 90t, WRIs and TRIs for, 49f, 441f, 444t, 94f, 383 447t, 452f protection of agriculture in, 384 input subsidies RRAs, 86f, 101­2f, 383f intermediate inputs, 572­73 WRIs and TRIs for, 48, 436f, 437t, 439f, 440t, measurement methodology, 580 441f, 442­49t, 450, 451f, 452f in South Asia and India, 394­95, 395t, WWII, prior to, 103­10, 105­7t 409, 591n21 Jara, Esteban, 67n, 115n, 177n, 221n, 259n, 289n, Integrated Framework's Diagnostic Trade 323n, 359n, 389n, 419, 459n, 595n Integration Study, 57 Jayasuriya, S., 389n International Food Policy Research Institute JCRR (Sino-American Joint Commission (IFPRI), 580 on Rural Reconstruction), 82 International Monetary Fund (IMF), Johnson, D. Gale, 5, 477, 506, 557 330­31, 420 Johnson, Robin, 221n International Wheat Agreement (1949), 182 Josling, Tim, 115, 565n Ireland. See also Western Europe CTEs, 153, 158­59t K current chronology of policy distortions Kazakhstan. See also Commonwealth of (1955­2007), 131, 132, 133f, 134 Independent States (CIS) countries; EEC, decision not to join, 173n26 Eastern Europe and Central Asia EU, accession to, 170 antiexport bias/TBI, 273t exports, contribution of agriculture to, 118 CTEs, 279t GSE values, 155­56t general equilibrium effects in, 511t, 517t, NRAs in, 153, 154t 525t, 533t, 537t, 543t, 552t, 558t WRIs and TRIs for, 441f, 446t, 449t GSE values, 277t Irwin, D. A., 196 key commodities, 471f, 473f, 475f Italy. See also Western Europe key economic indicators in, 261t CTEs, 158­59t measuring distortions to agricultural current chronology of policy distortions incentives in, 264 (1955­2007) NRAs during transition period (from early 1955­1964, 124­27, 128f, 173n22 1990s), 269t, 270, 271f, 274t 1975­1984, 136 political economy of policy choices in, 281, GSE values, 155­56t 282, 284 historical distortions of agricultural market reform era (after 1989) in, 266, 267 in, 108 RRAs, 278, 280f key commodities in EU-15, 471f, 473f, 475f, Kee, H. L., 420, 455 493­94f Kennedy Round, GATT, 173n25 NRAs in, 154t, 173n22 Kenya. See also Africa size of farms in, 118 CTEs, 348, 349t, 351 WRIs and TRIs for, 441f, 446t, 449t, 452f growth and structural change in, 329 GSE values, 343f, 345­46t J key commodities, 471f, 475f, 476f, 496­98t Japan. See also Northeast Asian advanced key economic indicators, 325t economies NRAs for, 334t, 335f, 336, 337t, 340 CTEs for, 44­45t, 96t RRA/TBI, 352f empirical estimates of distortion for, 13, WRIs and TRIs for, 441f, 444t, 447t 15­17t, 33t Kleinwechter, Uli, 115n evolution of agricultural policy following Korea, Republic of. See also Northeast Asian WWII, 77­80 advanced economies GATT, accession to, 383 CTEs for, 97t general equilibrium effects in, 511t, 517t, economic growth and structural 525t, 534t, 537t, 544t, 552t, 560t change in, 362 GSE values for, 18f, 19, 41­42t empirical estimates of distortion for, 60n4 historical distortions of agricultural evolution of agricultural policy following market in, 9 WWII, 80­82 Index 631 fertilizers, tariff protections for, 590n9 in Agricultural Distortions Project GATT, accession to, 383 database, 429­30t, 433f general equilibrium effects in, 509t, 512, 514, in agriculture, 298­306, 300t, 301f, 302f, 516t, 523t, 532t, 536t, 542t, 550t, 559t 303­5t, 311f, 318 GSE values for, 19 by commodity, 302f, 303­4t, 501­2t historical distortions of agricultural market CTEs and, 313 in, 10 empirical estimates, 19f, 20t, 23t, 26t key commodities, 449­500t, 471f, 473f, 475f, as methodology, 296­98 481f, 493­94f nonfarm sectors, 305t, 306­11, 311f NRAs in, 55, 56f, 88t, 91t, 94f poverty in, 319­20 protections for agriculture in, 384 during reform period (late 1980s­1990s), RRA in, 39, 86f, 101­2f, 381f, 383f 295­96 TBI in, 381f RRAs, 297, 305t, 306­13, 311f, 312f, 318, 319, WRIs and TRIs for, 48, 441f, 444t, 447t, 321n5 450, 452f tax and tariff policies, 294­96, 306, 318­19 Krueger, A. O., 5­6, 10, 28, 60­61n6, 60n4, 224, WRIs and TRIs for, 47f, 48, 436f, 437t, 439f, 294, 506, 579­80, 585, 617n1 440t, 442­49t, 451f, 455n2 Kurzweil, Marianne, 67n, 115n, 177n, 221n, Lattimore, Ralph, 221 259n, 289n, 323n, 359n, 389n, 419, 459n, Latvia. See also Central and Eastern European 505n, 565, 595n (CEE) countries; Eastern Europe and Kyrgyz Republic, 261t, 262, 266, 267, 281, Central Asia 283, 284 agricultural exports from, 262 antiexport bias/TBI, 273t L CTEs, 279t Latin America and the Caribbean, 289­322. See general equilibrium effects in, 510t, 517t, also specific countries 524t, 533t, 537t, 543t, 551t in Agricultural Distortions Project database, GSE values, 277t 429­31t, 433­34f, 595 key commodities, 471f, 473f, 475f antitrade bias/TBI, 36, 299, 302f, 312f, 313, key economic indicators in, 261t 318­19 NRAs during transition period (from early conclusions regarding, 318­19 1990s), 269t, 271f, 274t CTEs for, 44­45t, 297, 298, 313, 314­17t, 319, reform era (after 1989) in, 266 431t, 434f RRAs, 280f empirical estimates of distortion for, 14, WRIs and TRIs for, 441f, 445t, 448t 15­17t, 32f, 33t, 60n4 WTO accession, 284 EU, trade agreements with, 168 Lawrence, E., 224 evolution of policies in, 294­96 Leamer, E. E., 224 exchange rates, 295, 297, 299, 318, 320n3, Lerner (1936) Symmetry Theorem, 567, 577, 321n5 578, 590n4 exports, role of, 293 Lesotho, 331 future developments in, 53 liberalization of agricultural trade, 5­6, GDP in, 289, 290, 291t, 292­93, 295, 59­60t 298, 306 in Australia and New Zealand, 5­6, 59­60t general equilibrium effects in, 508, 510t, 512, in Eastern Europe and Central Asia 516t, 519­20t, 522, 533t, 535, 536t, 540, during reform era (after 1989), 266­67 543t, 545t, 549t, 551t, 560t during transition period (from early growth and structural changes in, 292­93 1990s), 267­78, 268f, 269t, 271­72f, GSE values for, 18f, 41­42t, 298, 307f, 274­76t 308­10t, 318 by emerging economies, 10­11 imports, 295 future developments, 53­57, 56f, 57t income inequality in, 385n3 GATT leading to, 9 key economic indicators, 290, 291t general global equilibrium effects, 51­53, liberalization of agricultural trade in, 59­60t 506­63. See also general global measuring distortion in, 296­98 equilibrium effects NAFTA, 293, 296, 299 in Latin America and the Caribbean, nonfarm sectors, 305t, 306­11, 311f 295­96 nontraded food staples, 482t, 484­86t North American prospects for, 214­17 NRAs for in Northeast Asian advanced economies 632 Index liberalization of agricultural trade (continued ) Lomé Agreements, 117 economic development and structural Luxembourg, 172n14­15, 173n21 change, 74­76, 75t growth of economies, structural changes M related to, 71­73, 72t MacLaren, Donald, 221, 241, 456n7 poverty reduction and reduction in MAcMaps, 508, 561n7 antiagricultural bias, 59 macroeconomic effects. See general global South Asia and India, prospects in, equilibrium effects 412­14 MacSharry CAP reforms (1994­96), 141, in Southeast Asia and China, 379­80 142, 165 in Western Europe, 59­60t mad cow disease, 161, 198 Lichtenstein, 173n31 Madagascar. See also Africa Linkage model, 247, 454, 506, 513­14. See also CTEs, 349t general global equilibrium effects general equilibrium effects in, 509t, 516t, Lisbon Treaty, 174n46 523t, 532t, 536t, 542t, 550t, 559t Lithuania. See also Central and Eastern growth and structural change in, 329 European (CEE) countries; Eastern GSE values, 343f, 345­46t Europe and Central Asia key commodities, 471f, 475f, 476f, 496­98t antiexport bias/TBI, 273t key economic indicators, 325t CTEs, 279t NRAs for, 333, 334t, 335f, 337t general equilibrium effects in, 510t, 517t, RRA/TBI, 352f 524t, 533t, 537t, 543t, 551t WRIs and TRIs for, 441f, 444t, 447t GSE values, 277t maize. See corn key commodities, 471f, 473f, 475f, Malaysia. See also Asia; Southeast Asia and China 493­94f conclusions regarding, 382 key economic indicators in, 261t CTEs, 363­64, 371, 375, 376­77t, 382 NRAs during transition period (from early domestic price stabilization, 375­79 1990s), 269t, 271f, 274t economic growth and structural change in, reform era (after 1989) in, 266 361­63 RRAs, 280f GDP in, 282f, 361­62, 370t, 377t WRIs and TRIs for, 441f, 445t, 448t general equilibrium effects in, 509t, 516t, WTO accession, 284 523t, 532t, 536t, 542t, 551t, 559t livestock GSE values, 364, 370t in Australia and New Zealand, 225, 226, 237, income inequality in, 385n3 241, 243, 245, 252n14 key commodities, 471f, 476f, 499­500t CTEs for, 478­79t nonfarm sectors, 371, 372f, 373­74t derived shares of global production, 602t NRAs for. See under Southeast Asia in Eastern Europe and Central Asia, 265, and China 270, 271 RRAs, 363, 371, 372f, 373­74t, 380­83, 381f, export shares, 605t 383f, 385, 386n8 general global equilibrium effects, 521t, 528t, TBI, 380­82, 381f 530t, 541t, 546­48t WRIs and TRIs for, 49f, 441f, 444t, 447t global value of production, NRA percentage Mali, 325t, 331, 345­46t, 349t, 476f, 496­98t coverage of, 461t, 463t, 464t Malta, 172n2 gross value of global production at Mansholt Report (1960), 163 undistorted prices, 599t marginal welfare weights, 455 GSE values, 465f, 466­68t, 469f Marshall Plan, 129, 181, 182 in North America, 179, 180, 198­99 Martin, Will, 52, 67n, 241, 359, 389n, 480, 538, in Northeast Asian advanced economies, 81, 554, 565 83, 93, 95 Masters, William A., 323, 565n NRAs for, 40t, 469f, 472, 473f Matthews, Alan, 565n production export and consumption Mauritius, 331 imported, shares in, 482t, 609t McCrone, G., 127 regional production shares, 604t MDGs (Millennium Development Goals), 6 total shares in global production, 608t measurement methodology, 5­8, 11­13, 565­94. in Western Europe, 120­22, 161, 164 See also consumer support estimates; WRIs and TRIs, 488, 489­90t, 491­94f consumer tax equivalents; nominal rates Lloyd, Peter J., 46, 221, 419, 427, 456n7, 487 of assistance; producer support Index 633 estimates; welfare reduction and trade global value of production, NRA percentage reduction indexes coverage of, 461t, 463t, 464t commodities gross value of global production at classification, 582­84 undistorted prices, 599t coverage, 580­81 GSE values, 465f, 466­68t, 469f CTE of, 584­85 in North America, 191, 199, 216 NRA guesstimates for noncovered NRAs for, 469­72, 469f, 470t, 471f products, 586 production export and consumption quality and variety of, 574 imported, shares in, 482t, 609t direct agricultural distortions, 567 total shares in global production, 608t direct production tariffs and subsidies, 568 in Western Europe, 161, 164 dollar values, use of, 587­88, 589f WRIs and TRIs, 488, 489­90t, 491­94f exchange rates, 569­72, 569f, 585­86 Millennium Development Goals (MDGs), 6 food processing distortions, 575­76 Mozambique. See also Africa import tariffs and subsidies, 568 CTEs, 349t indirect distortions, 578­79 general equilibrium effects in, 509t, 515, intermediate inputs, 572­73 516t, 523t, 532t, 535, 536t, 542t, 549, key parameters and exogenous variables, 585 550t, 559t nonfarm sector distortions, 578­79, 587 GSE values, 343f, 345­46t NPS distortions, 577, 586­87 key commodities, 471f, 475f, 476f, 496­98t percentages, use of, 587 key economic indicators, 325t TBI, 577­78. See also trade bias index NRAs for, 334t, 335f, 336, 337t theory of, 566­67 RRA/TBI, 352f from theory to practice, 579­80 WRIs and TRIs for, 441f, 444t, 447t time periods for which data are available, 580 trade costs. See trade costs, measuring N transmission of distortion along value NAFTA (North American Free Trade chain, 584 Agreement), 293, 296, 299 Meline tariff (1892; France), 121, 172n10 Nash, E. A., 127 Melyukhina, Olga, 259n National Farmers Federation (NFF), Australia, Mencken, H. L., 217n7 240­41 methodology for measuring distortions. See Ndulu, B., 324, 331 measurement methodology Neary, J. P., 8, 43, 420, 422, 487 Mexico. See also Latin America and the Nelgen, Signe, 67n, 115n, 177n, 221n, 259n, Caribbean 289n, 323n, 359n, 389n, 419, 459, CTEs, 313, 314t, 316t, 317t 505n, 595n evolution of agricultural policy in, 296 Nepal, 390, 393 general equilibrium effects in, 510t, 516t, Net Income Stabilization Account (NISA), 524t, 533t, 536t, 543t, 551t, 560t Canada, 197­98 growth and structural changes in, 292, 293 Netherlands, 166. See also Western Europe GSE values, 307f, 308t Benelux Union, 172n14­15, 173n21 key commodities, 471f, 473f, 475f, 476f, CTEs, 153, 158­59t 493­84f, 501­2t current chronology of policy distortions key economic indicators, 290, 291t (1955­2007) NRAs, 298, 299, 300t, 301, 301f, 304t, 306 1955­1964, 126­27, 128f, 129 RRA/TBI, 312f 1965­1974, 132 WRIs and TRIs, 420, 441f, 445t, 448t, 452f GSE values, 153, 155­56t MFP (farm multifactor productivity) growth, historical distortions of agricultural market 248, 249f, 253n27 in, 121 Middle East and North Africa. See Africa key commodities in EU-15, 471f, 473f, milk and dairy 475f, 493f in Australia and New Zealand, 237, 241­43, NRAs in, 154t 245, 247, 253n20, 253n25 policy trends and turning points, 166 CTEs for, 478­79t WRIs and TRIs for, 441f, 446t, 449t, 452f derived shares of global production, 602t New Deal, 180, 181, 187, 196, 217n5 export shares, 605t New Zealand. See Australia and New Zealand general global equilibrium effects, 521t, 528t, New Zealand­Australia Free Trade Agreement 530t, 540, 541t, 546­48t (1966), 242 634 Index NFF (National Farmers Federation), Australia, in Eastern Europe and Central Asia, 263, 273, 240­41 275t, 276t Nicaragua. See also Latin America and empirical estimates using, 21, 27­28, 29­30t the Caribbean measurement methodology, 577, 586­87 CTEs, 314t, 316t in North America, 187, 200, 201­2t, 204, exchange rate distortions, 321n5 206­7t, 218n16 general equilibrium effects in, 510t, 516t, OECD measurement of, 11, 187 524t, 533t, 536t, 543t, 551t, 560t nondistortionary price wedges, 573­75 growth and structural changes in, 292 nonfarm sectors GSE values, 307f, 308t in Africa, 341­42t, 347 key commodities, 471f, 473f, 475f, 476f, 493f, in Australia and New Zealand, 224­26, 501­2t 234­35t, 236f key economic indicators, 290, 291t in Latin America and the Caribbean, 305t, NRAs, 297, 299, 300t, 301f, 304t, 306 306­11, 311f RRA/TBI, 312f measurement methodology, 578­79, 587 WRIs and TRIs for, 441f, 445t, 448t in North America, 199­204, 201­2t, 203f, 205f Nicita, A., 420, 455 NRAs and, empirical estimates of distortion Nigeria. See also Africa using, 28­32, 29­30t, 31f, 33­35t CTEs, 348, 349­50t, 351 South Asia and India, 402­3f, 410, 411­12 general equilibrium effects in, 509t, 515, 516t, in Southeast Asia and China, 371, 372f, 523t, 532t, 535, 536t, 542t, 550t, 559t 373­74t growth and structural change in, 329 in Western Europe, 119, 145, 147­48t GSE values, 343f, 345­46t nontariff measures (NTMs), WRIs and TRIs, key commodities, 471f, 475f, 476f, 493­94f, 422, 424, 426 496­98t nontraded food staples of low-income countries, key economic indicators, 325t 482t, 483­86, 484­86t NRAs for, 334t, 335f, 336, 337t, 340 North Africa. See Africa RRA/TBI, 352f North America (Canada and United States), WRIs and TRIs for, 441f, 444t, 447t 177­220 NISA (Net Income Stabilization Account), biofuel production, subsidization of (2008), 3 Canada, 197­98 Canadian agricultural policy, 196­99 nominal rates of assistance (NRAs), 7, 11­13, comparison of farms and farm income in U.S. 19­39. See also measurement method- and Canada, 177­80, 178f ologies, and under specific countries CTEs for, 44­45t, 204­8, 209­11t, 431t, 432f and regions empirical estimates of distortion for, 13, in Agricultural Distortions Project database, 15­17t, 34t 428­35, 429­30t, 432f, 433f, 595­96, EU view of U.K. as Trojan horse for policy 596­97f, 597­98t interests of U.S., 172n18 by commodity. See under commodities food subsidy war with Western Europe dispersion index, 21­22, 26t (1986), 3 future developments regarding, 53­56 future developments in, 55 general global equilibrium effects and, 508, general equilibrium effects in, 507, 511t, 517t, 514, 529 525t, 534t, 537t, 544t, 552­53t, 560t guesstimates for noncovered agricultural GSE values for, 18f, 19, 41­42t, 204, 206­7t products, 586 key commodities import-competing versus exportable Canada, 471f, 473f, 475f, 493­94f products, 21, 22f United States, 471f, 473f, 474, 475f, 476f, macroeconomic modeling study, 51 491, 493­94f mean and standard deviations, 576­77 liberalization of agricultural trade in, 59­60t, nonfarm sectors, 28­32, 29­30t, 31f, 33­35t 214­17 NPS assistance, 21, 27­28, 29­30t nonfarm sectors, 199­204, 201­2t, 203f, 205f PSEs compared, 589n2 nontraded food staples, 482t, 484t by region, 19­21, 19f, 20t NPS support in, 187, 200, 201­2t, 204, value of using NRA for, 43 206­7t, 218n16 WRIs and TRIs compared, 46, 420­22, 424, NRAs for 438, 450, 453f, 454­55 in Agricultural Distortions Project non-product-specific (NPS) distortions database, 429­30t, 432f in Africa, 332, 334t, 340f, 341­42t, 353 Canada, 193f, 195f, 199 Index 635 empirical estimates, 19f, 20t, 24t, 26t 1965­1974, 132, 133f nonfarm sectors, 200­204, 201­2t, 1975­1984, 135, 136, 137f 203f, 205f 1985­1994, 139, 140f, 141 U.S., 192, 193f, 194f 1995­2007, 143, 144f political economy of agricultural policies in, EEC, decision not to join, 173n26 208­14, 213t EU, future entry into, 167, 171 reinstrumentation in, 165 GSE values, 153, 155­56t RRAs, 200­204, 201­2t, 203f, 205f homogeneity of farm sector in, 250 U.S. agricultural policy, 180­92 key commodities, 471f, 472, 473f, 474, 475f direct subsidies to producers, 187­89, nonfarm sectors, 148t 187f, 189f NRAs in, 148t, 153, 154t, 173n24 export subsidies, 181­83, 186­87 policy trends and turning points in, 163, 165 historical overview of, 180­83 WRIs and TRIs for, 441f, 446t, 449t, 450 import protections, 183­86, 184t NPS. See non-product-specific (NPS) distortions NRAs, 192, 193f, 194f NRAs. See nominal rates of assistance Western European agriculture and, 115, 121, NTMs (nontariff measures), WRIs and TRIs, 162, 164 422, 424, 426 WRIs and TRIs for, 436f, 437t, 439f, 440t, nvCJD (Creutzfeldt-Jakob) disease, 161 441f, 442­49t, 450, 451f, 452f North American Free Trade Agreement O (NAFTA), 293, 296, 299 OECD. See Organisation for Economic Northeast Asian advanced economies. See also Co-operation and Development Japan; Korea, Republic of; Taiwan, China oil price increases (1973), CAP affected by, conclusions drawn from, 110­11 165­66 CTEs for, 95, 96­98t oilseeds East Asian economic miracle of, 67­70, 103­4 derived shares of global production, 601t economic development and structural edible oil imports in South Asia and India, 393 change in export shares, 605t agricultural structure, changes in, general global equilibrium effects, 521t, 528t, 74­76, 75t 530t, 540, 541t, 546­48t growth of economies, structural changes global value of production, NRA percentage related to, 71­73, 72t coverage of, 461t, 463t, 464t initial conditions and development gross value of global production at strategies, 69­71 undistorted prices, 599t exchange rates, 74, 85 GSE values, 465f, 466­68t, 469f GDP in, 63, 71­73, 72t, 78, 79, 84, 100­103, production export and consumption 101­2f, 104, 105t, 111n4, 112n15 imported, shares in, 609t growth of agricultural protection in, 95­103 regional production shares, 603t key commodities, 474 sunflower seeds from Russian Federation, 270 major crises faced by, 70 total shares in global production, 607t measurement methodologies, 85­95 WRIs and TRIs, 488, 489­90t nontraded food staples, 482t, 484t Olarreaga, M., 420, 455 NRAs for, 68, 85­86, 87­92t, 93­95, 94f, 99, Olson, M., 208 111­12n10 Orange Revolution, Ukraine, 283 rice, 69, 76­85, 87­89t, 94­95, 94f, 109, Orden, D., 586 112n13 Organisation for Economic Co-operation and RRAs in, 68, 85­95, 86f, 99, 101­2f, 112n14 Development (OECD) WWII agricultural distortions, measuring, 5­6, 11, evolution of agricultural policy 60n6, 61n8, 85, 111­12n10, 138, 174n35, following, 77­85 199­200, 264, 455n1, 579, 586, 589n2 Japan prior to, 103­10, 105­7t Australia and New Zealand compared to other Norway. See also Western Europe OECD countries, 221­23 absolute value of subsidies in, 174n45 CAP and, 136, 138 autonomous agricultural policy of, 124, 172n2 Communist era Eastern Europe and Central CTEs in, 153, 158­59t Asia compared to countries of, 265 current chronology of policy distortions Eastern European and Central Asian accession (1955­2007) to, 260 1955­1964, 127­29, 128f NPS support, measuring, 11, 187 636 Index P GSE values, 277t Pak Chong-hui, 71, 80 key commodities, 471f, 473f, 475f, 493­94f Pakistan, 389­415. See also Asia; South Asia key economic indicators in, 261t and India measuring distortions to agricultural exchange rates in, 390, 399­401, 408, 410, incentives in, 264 413, 414n7 NRAs during transition period (from early GDP in, 389­91, 391t, 393, 394, 397, 1990s), 269t, 271f, 274t 412, 414n1 reform era (after 1989) in, 266 general equilibrium effects in, 510t, 516t, RRAs, 280f 524t, 533t, 536t, 543t, 551t, 559t WRIs and TRIs for, 441f, 445t, 448t key commodities, 471f, 472, 475f, 476f, WTO accession, 284 493­94f, 499­500t policy interventions and outcomes, 3­6. NRAs for, 56, 401­11, 402­3f, 406­7t See also distortions to agricultural policy interventions and outcomes incentives, and under specific countries current practices, 393­401 and regions future policies, 411­14 political economy TBI, 406­7t, 408 CAP affected by, 166 trade in agricultural products in, 391­93 Eastern Europe and Central Asia, policy trade status of farm commodities in, choices in, 278­85 404, 405t EEC, countries not joining, 173n26 WRIs and TRIs for, 49f, 441f, 445t, 447t North American agricultural policies, politics PDI (producer distortion index), 421, 435­38, of, 208­14, 213t 442t, 455, 456n7 Portugal. See also Western Europe peanuts, in North America, 216 CTEs, 158­59t Petain, Marshall, 172n10 current chronology of policy distortions Philippines. See also Asia; Southeast Asia (1955­2007) and China 1955­1964, 127­30, 128f, 173n23 conclusions regarding, 380, 382 1965­1974, 132­34, 133f CTEs, 363­64, 371, 375, 376­77t, 382 1975­1984, 135­36, 137f at Doha Round, 386n11 1985­1994, 138 domestic price stabilization, 375­79 EU, accession to, 170 economic growth and structural change in, GDP, share of agricultural output in, 118 361­63 GSE values, 155­56t GDP in, 282f, 361­62, 370t, 377t NRAs in, 154t general equilibrium effects in, 509t, 516t, wine exports to U.K. in 17th-18th cs., 4 523t, 532t, 536t, 542t, 551t, 559t WRIs and TRIs for, 441f, 446t, 449t GSE values, 364, 370t potatoes income inequality in, 385n3 nontraded food staples of low-income key commodities, 471f, 473f, 474, 493f, countries, 482t, 483­86, 484­86t 499­500t in North America, 213­14 nonfarm sectors, 371, 372f, 373­74t in South Asia and India, 401, 404, NRAs for. See under Southeast Asia 408, 409 and China poultry. See livestock RRAs, 363, 371, 372f, 373­74t, 380­83, 381f, poverty 383f, 385, 386n8 in Africa, 327, 354, 355 TBI, 380­82, 381f in Latin America and the Caribbean, 319­20 WRIs and TRIs for, 49f, 441f, 445t, reduction in antiagricultural bias and 447t, 452f reduction in, 59 pigs. See livestock in Southeast Asia and China, 360t Pincus, Jonathan, 221n Poverty Reduction Strategy Papers (PRSPs), 331 Poland. See also Central and Eastern European Prebisch, R., 294 (CEE) countries; Eastern Europe and price wedges, 6, 46, 204, 286n4, 386n7 Central Asia price wedges, nondistortionary, 573­75 antiexport bias/TBI, 273t prices of products in Communist era, 265 prospective analysis of, 545­48, 546t CTEs, 279t retrospective analysis of, 522­29, 530t general equilibrium effects in, 511t, 517t, producer distortion index (PDI), 421, 435­38, 524t, 533t, 537t, 543t, 552t, 558t 442t, 455, 456n7 Index 637 producer support estimates (PSEs) total shares in global production, 607t NRAs compared, 589n2 WRIs and TRIs, 488, 489­90t, 491­94f OECD, 5­6, 11, 60n6, 61n8, 85, 111­12n10, Rice Riots (1918; Japan), 69 199­200, 455n1, 586, 589n2 Rojas Lara, Teresa, 115n for South and Southeast Asia, 580 Romania, 172n2. See also Central and Eastern productivity European (CEE) countries; Eastern in Australia and New Zealand, 224 Europe and Central Asia farm MFP growth, 248, 249f, 253n27 antiexport bias/TBI, 273t TFP, 179, 224, 248, 249f CTEs, 279t in Western Europe, 118, 160­62 general equilibrium effects in, 511t, 517t, products. See commodities 524t, 533t, 537t, 543t, 552t, 558t PRSPs (Poverty Reduction Strategy Papers), 331 GSE values, 277t PSEs. See producer support estimates key commodities, 471f, 473f, 475f Pursell, Garry, 389 key economic indicators in, 261t NRAs during transition period (from early R 1990s), 269t, 271f, 274t R&D (research and development) in political economy of policy choices in, agriculture, 57t 283, 284 Ravallion, M., 59 reform era (after 1989) in, 266 reform of agricultural market distortions. See RRAs, 278, 280f liberalization of agricultural trade WRIs and TRIs for, 441f, 445t, 448t, 452f reinstrumentation, in Western Europe, 164­65 Rome, Treaty of, 126, 127 relative rates of assistance (RRAs), 11­12. See RRAs. See relative rates of assistance also under specific countries and regions Russian Federation. See also Commonwealth of empirical estimates of distortions using, Independent States (CIS) countries; 28­39, 31­32f, 33­35t, 36­40f Eastern Europe and Central Asia future developments in, 53, 55 agricultural exports from, 262 general global equilibrium effects and, 512 antiexport bias/TBI, 273t macroeconomic modeling study, 52 CTEs, 279t rent extraction in Eastern Europe and Central general equilibrium effects in, 511t, 517t, Asia, 280­82 525t, 533t, 537t, 543t, 552t, 558t Republic of Korea. See Korea, Republic of GSE values, 277t research and development (R&D) in key commodities, 471f, 473f, 475f, 493f agriculture, 57t key economic indicators in, 261t revealed comparative advantage, 15t, 262, measuring distortions to agricultural 293, 329 incentives in, 264 Ricardo-Viner model, 224 NRAs during transition period (from early rice 1990s), 268f, 269t, 270, 271f, 274t, 275t, CTEs for, 478­79t 286n4, 287n7 derived shares of global production, 601t political economy of policy choices in, 281, export shares, 605t 282, 284 general global equilibrium effects, 521t, 528t, reform era (after 1989) in, 266, 267 530t, 541t, 546­48t RRAs, 280f global value of production, NRA percentage WRIs and TRIs for, 441f, 445t, 448t, 452f coverage of, 461t, 463t, 464t WTO, accession to, 284, 286 gross value of global production at Russo-Japanese War (1904­05), 108 undistorted prices, 599t GSE values, 465f, 466­68t, 469f S in Northeast Asian advanced economies, 69, Salam, A., 389n 76­85, 87­89t, 94­95, 94f, 109, 112n13 SALs (structural adjustment loans), 331 NRAs for, 469­72, 469f, 470t, 471f Sandri, Damiano, 67n, 115n, 177n, 221n, price variability/stability, effects of interven- 259n, 260, 289n, 323n, 359n, 389n, tion on, 480­81f 565, 595n production export and consumption Schiff, M., 5­6, 10, 28, 60­61n6, 60n4, 294, 481, imported, shares in, 482t, 609t 506, 579­80, 585, 617n1 regional production shares, 603t Schmitz, A., 208 in Southeast Asia and China, 364, 365f, 375, Schultz, Theodore, 67, 68, 294 378­79f, 382, 384 Scobie, Grant, 221n 638 Index sectoral value added evolution of trade policies in, 329, 331 prospective analysis of, 548, 550­53t general equilibrium effects in, 509t, 516t, retrospective analysis of, 531­35, 532­34t 523t, 532t, 535, 536t, 542t, 550t, 559t Senegal. See also Africa GSE values, 343f, 345­46t CTEs, 349­50t income diversity of, 327 evolution of trade policies in, 331 key commodities, 471f, 473f, 475f, 496­98t general equilibrium effects in, 509t, 516t, key economic indicators, 326t 523t, 532t, 536t, 542t, 550t, 559t measurement of distortions in, 579 growth and structural change in, 329 NRAs for, 333, 334t, 335f, 336, 337t GSE values, 343f, 345­46t PSEs for, 6 key commodities, 471f, 476f, 496­97t RRA/TBI, 352f key economic indicators, 325t Western European agriculture and, 121 NRAs for, 333, 334t, 335f, 337t WRIs and TRIs for, 441f, 444t, 447t RRA/TBI, 352f South Asia and India, 389­415. See also Asia; WRIs and TRIs for, 441f, 444t, 447t Bangladesh; India; Pakistan; Sri Lanka Sieper, E., 239 edible oil imports, 393 Singapore, 69, 517t, 525t, 534t, 537t, 544t, exchange rates in, 390, 396­401, 398f, 408, 553t, 561n11 410, 413, 414n7 Sino-American Joint Commission on Rural food safety nets in, 393­94 Reconstruction (JCRR), 82 GDP in, 389­91, 391t, 393, 394, 397, Slovak Republic. See also Central and Eastern 412, 414n1 European (CEE) countries; Eastern input subsidies, 394­95, 395t, 409, 591n21 Europe and Central Asia nonfarm sectors, 402­3f, 410, 411­12 antiexport bias/TBI, 273t NRAs for, 401­11, 402­3f, 406­7t CTEs, 279t policy interventions and outcomes general equilibrium effects in, 511t, 517t, current policies, 393­401 525t, 533t, 537t, 543t, 552t, 558t future of, 411­14 GSE values, 277t PSEs for, 580 key commodities, 471f, 473f, 475f regional agricultural trade in, 401 key economic indicators in, 261t RRAs, 401, 402­3f, 406­7t, 409­11, 409f NRAs during transition period (from early tax and tariff policies in, 397­401, 398f 1990s), 269t, 271f, 274t TBI, 406­7t, 408 reform era (after 1989) in, 266 trade in agricultural products in, 391­93 RRAs, 280f trade status of farm commodities in, 404, 405t WRIs and TRIs for, 441f, 445t, 448t South Korea. See Korea, Republic of WTO accession, 284 Southeast Asia and China, 359­87. See also Asia; Slovenia. See also Central and Eastern European China; Indonesia; Malaysia; Philippines; (CEE) countries; Eastern Europe and Thailand; Vietnam Central Asia colonialism, end of, 385n4 agricultural exports from, 262 conclusions regarding, 379­82 antiexport bias/TBI, 273t CTEs, 363­64, 371, 375, 376­77t, 382 CTEs, 278, 279t domestic price stabilization, 375­79 general equilibrium effects in, 511t, 517t, economic growth and structural change in, 525t, 533t, 537t, 543t, 552t, 558t 361­63 GSE values, 277t exchange rates, 286n8, 363, 364, 371 key commodities, 471f, 473f, 475f, 493­94f future developments in, 382­85 key economic indicators in, 261t GDP in, 361­62, 370t, 377t, 383f NRAs during transition period (from early GSE values, 364, 370t 1990s), 269t, 270, 271f, 274t income inequality in, 385n3 reform era (after 1989) in, 266, 267 liberalization of agricultural trade in, RRAs, 280f 379­80 WRIs and TRIs for, 441f, 445t, 448t nominal rates of protection for key WTO accession, 284 commodities in, 6 SMAs (statutory marketing authorities) in nonfarm sectors, 371, 372f, 373­74t Australia, 238, 240, 253n18 NRAs for, 363­71 Smoot-Hawley Tariff Act (U.S.), 181 agriculture, 364­68, 365f, 366­67t, 369f, South Africa. See also Africa 372f, 373­74t, 378­79f CTEs, 348, 350t, 351 by commodity, 367t, 369f Index 639 conclusions and future developments statutory marketing authorities (SMAs) in regarding, 380, 383, 384 Australia, 238, 240, 253n18 CTEs and, 375 Stigler, G. J., 7 nonfarm sectors, 371, 372f, 373­74t structural adjustment loans (SALs), 331 rice NRAs, 365f, 378­79f sub-Saharan Africa. See Africa poverty in, 360t subsidies, generally PSEs for, 580 in Africa, 348, 354 RRAs, 363, 371, 372f, 373­74t, 380­83, 381f, in Australia and New Zealand, 223, 226, 228, 383f, 385, 386n8 238, 243­45, 247, 248 tax and tariff policies, 379­80 in Communist era Eastern Europe and TBI, 380­82, 381f Central Asia, 265­66 WRIs and TRIs for, 49f. See also under specific historical shift from taxation to countries protection, 8­10 Soviet Union for inputs. See input subsidies Communist era, Eastern Europe and Central measurement methodology for. See Asia during, 264­66 measurement methodology demise of, 60n1 in Northeast Asian advanced economies, Finland's decision not to join EEC and, 95­103 173n26 South Asia and India, input subsidies in, international grain market, entry into 394­95, 395t, 409 (1973), 3 transmission of distortion along value soybeans. See grains chain, 584 Spain. See also Western Europe in U.S. CTEs, 158­59t direct subsidies to producers, 187­89, current chronology of policy distortions 187f, 189f (1955­2007) export subsidies, 181­83, 186­87 1955­1964, 129­30 Sudan. See also Africa 1975­1984, 135­36, 137f CTEs, 348, 350t 1985­1994, 138 evolution of trade policies in, 331 EU, accession to, 170 growth and structural change in, 329 GDP, share of agricultural output in, 118 GSE values, 343f, 345­46t GSE values, 155­56t key commodities, 471f, 473f, 475f, 493­94f, key commodities in EU-15, 471f, 473f, 475f, 496­98t 493­94f key economic indicators, 326t NRAs in, 154t NRAs for, 334t, 335f, 337t, 340 wine exports to U.K. in 17th­18th cs., 4 RRA/TBI, 352f WRIs and TRIs for, 441f, 446t, 449t, 452f WRIs and TRIs for, 441f, 444t, 447t, 452f Spanish Civil War, 129 sugar Spence, M., 324 CTEs for, 478­79t Sri Lanka, 389­415. See also Asia; South Asia derived shares of global production, 602t and India export shares, 605t empirical estimates of distortion for, 60n4 general global equilibrium effects, 521t, 522, exchange rates in, 390, 399­401, 408, 410, 528t, 530t, 540, 541t, 546­48t 413, 414n7 global value of production, NRA percentage GDP in, 389­91, 391t, 393, 394, 397, coverage of, 461t, 463t, 464t 412, 414n1 gross value of global production at general equilibrium effects in, 510t, 516t, undistorted prices, 599t 524t, 533t, 536t, 543t, 551t, 559t GSE values, 465f, 466­68t, 469f key commodities, 471f, 474, 499­500t in North America, 181, 184, 185, 191, 192, NRAs for, 401­11, 402­3f, 406­7t 204, 212 policy interventions and outcomes NRAs for, 469­72, 469f, 470t, 471f current practices, 393­401 production export and consumption future policies, 411­14 imported, shares in, 482t, 609t TBI, 406­7t, 408 regional production shares, 604t trade in agricultural products in, 391­93 total shares in global production, 608t trade status of farm commodities in, in Western Europe, 161­62, 164 404, 405t WRIs and TRIs, 488, 489­90t, WRIs and TRIs for, 49f, 441f, 445t, 447t 491­94f 640 Index sunflower seeds, from Russian Federation, 270. RRA in, 39, 86f, 101­2f, 381f, 383f See also oilseeds TBI in, 381f Swaziland, 331 WRIs and TRIs for, 49f, 441f, 445t, 448t, 452f Sweden. See also Western Europe Tajikistan, 261t, 271, 280, 281 CTEs, 158­59t Tangermann, Stefan, 115n current chronology of policy distortions Tanzania. See also Africa (1955­2007) CTEs, 350t 1955­1964, 127­29, 128f general equilibrium effects in, 509t, 516t, 1965­1974, 132, 133f, 134 523t, 532t, 536t, 542t, 550t, 559t 1975­1984, 135, 136, 137f growth and structural change in, 328 1985­1994, 139, 140f GSE values, 343f, 345­46t EEC, decision not to join, 173n26 key commodities, 471f, 475f, 476f, 493f, exports, contribution of agriculture to, 118 496­98t GSE values, 155­56t key economic indicators, 326t NRAs in, 154t NRAs for, 333, 334t, 335f, 337t, 340 WRIs and TRIs for, 441f, 446t, 449t RRA/TBI, 352f Swinnen, Johan, 259, 278, 565n WRIs and TRIs for, 441f, 444t, 447t Switzerland. See also Western Europe tax and tariff policies absolute value of subsidies in, 174n45 in Africa, 336, 340­44, 353­54 autonomous agricultural policy of, 124, 172n2 in Australia and New Zealand, 222, 225, 226, CTEs in, 153, 158­59t 228, 233, 242, 245­46 current chronology of policy distortions in Communist era Eastern Europe and (1955­2007) Central Asia, 265­66 1955­1964, 127­29, 128f historical shift to protection/subsidization 1965­1974, 132, 133f from, 8­10 1975­1984, 135, 136, 137f in Latin America and the Caribbean, 294­96, 1985­1994, 139, 140f, 141 306, 318­19 1995­2007, 143, 144f measurement methodology for. See EEC, decision not to join, 173n26 measurement methodology EU, future entry into, 167, 171 Northeast Asian advanced economies' move GSE values, 153, 155­56t away from, 95­103 key commodities, 471f, 472, 473f, 474, 475f, in pre-WWII Japan, 108­9 493­94f in South Asia and India, 397­401, 398f Lichtenstein, 173n31 in Southeast Asia and China, 379­80 nonfarm sectors, 148t transmission of distortion along value Northeast Asian advanced economies chain, 584 compared to, 99 U.S. import protections, 183­86, 184t NRAs in, 148t, 154t, 173n24 in Western Europe, 121, 122 policy trends and turning points in, 163, 165 WRIs and TRIs, tariffs and NTMs, 422, WRIs and TRIs for, 441f, 446t, 449t, 450, 452f 424, 426 TBI. See trade bias index T TFP (total factor productivity), 179, 224, Taiwan, China. See also Northeast Asian 248, 249f advanced economies Thailand. See also Asia; Southeast Asia CTEs for, 98t and China economic growth and structural conclusions regarding, 382 change in, 362 CTEs, 363­64, 371, 375, 376­77t, 382 empirical estimates of distortion for, 60n4 domestic price stabilization, 375­79 evolution of agricultural policy following economic growth and structural change in, WWII, 82­85 361­63 general equilibrium effects in, 509t, 514, 516t, GDP in, 282f, 361­62, 370t, 377t 523t, 532t, 536t, 542t, 550t, 559t general equilibrium effects in, 509t, 516t, 522, historical distortions of agricultural market 523t, 532t, 536t, 542t, 551t, 559t in, 10 GSE values, 364, 370t key commodities, 449­500t, 471f, 473f, 475f, income inequality in, 385n3 481f, 491, 493­94f key commodities, 471f, 472, 473f, 475f, NRAs in, 55, 56f, 89t, 92t, 94f 499­500t protections for agriculture in, 384 nonfarm sectors, 371, 372f, 373­74t Index 641 NRAs for. See under Southeast Asia and China EU, accession to, 167­68, 285 RRAs, 363, 371, 372f, 373­74t, 380­83, 381f, future directions for, 285 383f, 385, 386n8 general equilibrium effects in, 508, 511t, 517t, TBI, 380­82, 381f 525t, 533t, 537t, 543t, 552t, 558t WRIs and TRIs for, 49f, 441f, 445t, 448t GSE values, 277t Thompson, Sir William, 3 key commodities, 471f, 473f, 475f, 476f, 491, tobacco, 180, 214, 216, 237, 244 493­94f Togo, 326t, 345­46t, 350t, 476f, 496­98t key economic indicators in, 261t Toniolo, G., 259 measuring distortions to agricultural total factor productivity (TFP), 224, 248, 249f incentives in, 264 trade bias index (TBI). See also under specific NRAs during transition period (from early countries 1990s), 268f, 269t, 271, 274t, 276t Africa, antiagricultural and antitrade bias in, political economy of policy choices 36, 39, 327, 340, 347, 351­53, 352f, in, 281, 284 353­54 reform era (after 1989) in, 267 Australia and New Zealand, antiagricultural RRAs, 280f bias in, 223, 227, 233, 237, 248 WRIs and TRIs for, 441f, 445t, 448t, Eastern Europe and Central Asia, antiexport 452f, 455n2 bias in, 259, 270­71, 273f Turkmenistan, 261t, 266, 271, 280, 281 empirical estimates, 21, 23­25t, 36, 40f, 43 Tyers, R., 5, 477 Global Distortion Database indicating antitrade bias, 434 U Latin American and the Caribbean, antitrade Uganda. See also Africa bias in, 36, 299, 302f, 312f, 313, 318­19 CTEs, 350t measurement methodologies, 577­78 food processing distortions in, 576 South Asia and India, 406­7t, 408 general equilibrium effects in, 509t, 516t, Southeast Asia and China, 380­82, 381f 523t, 532t, 536t, 542t, 550t, 559t trade costs, measuring, 581­82 growth and structural change in, 328 changes in costs, 591n12 GSE values, 343f, 345­46t domestic, 574 key commodities, 471f, 475f, 476f, 496­98t factors affecting, 590n11 key economic indicators, 326t international, 574 NRAs for, 334t, 335f, 337t post-farmgate, 573 RRA/TBI, 352f processing costs, 574 WRIs and TRIs for, 441f, 444t, 447t product price, relationship to, 591n13 Ukraine. See also Commonwealth of prohibitive, 572 Independent States (CIS) countries; wholesale costs, 574 Eastern Europe and Central Asia wide variations in costs, 591­92n23 antiexport bias/TBI, 273t trade liberalization. See liberalization of agricul- CTEs, 279t tural trade EU, accession to, 285 trade reduction indexes (TRIs). See welfare future directions for, 285 reduction and trade reduction indexes GSE values, 277t trade restrictions on agricultural products. See key commodities, 471f, 472, 473f, 475f, 493f distortions to agricultural incentives; tax key economic indicators in, 261t and tariff policies measuring distortions to agricultural trade restrictiveness indexes, 43­50, 419­22. See incentives in, 264 also welfare reduction and trade NRAs during transition period (from early reduction indexes 1990s), 268f, 269t, 270, 271f, 274t, 275t Treaty of Rome, 126, 127 political economy of policy choices in, 281, TRIs. See welfare reduction and trade reduction 283, 284, 285 indexes reform era (after 1989) in, 266, 267 Tulip Revolution, Kyrgyz Republic, 283 RRAs, 278, 280f Turkey. See also Central and Eastern European WRIs and TRIs for, 441f, 445t, 448t (CEE) countries; Eastern Europe and United Kingdom. See also Western Europe Central Asia Corn Laws (1815­1846), 8, 9, 108, 120 antiexport bias/TBI, 273t CTEs, 158­59t CTEs, 278, 279t current chronology of policy distortions empirical estimates for, 13 (1955­2007) 642 Index United Kingdom (continued) GSE values, 364, 370t 1955­1964, 124, 126­29, 128f, 172n18, income inequality in, 385n3 173n19 key commodities, 471f, 473f, 476f, 493­94f, 1965­1974, 131­32, 133f 499­500t 1975­1984, 134­36 market economy, move to, 385n4 1985­1994, 136 nominal rates of protection for key 1995­2007, 141 commodities in, 6 EEC, accession to, 166, 170, 172n18, 243 nonfarm sectors, 371, 372f, 373­74t EU view of U.K. as Trojan horse for policy NRAs for. See under Southeast Asia and China interests of U.S., 172n18 PSEs for, 580 exports, contribution of agriculture to, 118 RRAs, 363, 371, 372f, 373­74t, 380­83, 381f, GDP, share of agricultural output in, 118 383f, 385, 386n8 GSE values, 155­56t TBI, 380­82, 381f historical distortions of agricultural market time periods for which data are available, 580 in, 4, 8­9, 108, 120­23, 172n7, 172n8 WRIs and TRIs for, 49f, 441f, 445t, 448t Ireland and, 173n26 key commodities in EU-15, 471f, 473f, 475f, W 493­94f Washington Consensus, 331 NRAs in, 154t Weimar Republic (Germany), 122 productivity in, 172n8 welfare reduction and trade reduction indexes size of farms in, 118 (WRIs and TRIs), 46­50 sugar production and policies in, 161 Agricultural Distortions Project database, WRIs and TRIs for, 441f, 446t, 449t, 452f NRA and CTE estimates, 428­35, United States. See North America 429­31t, 432­34f Uruguay Round Agreement on Agriculture by commodity, 487­95, 489­90t, 491­94f (URAA) conclusions regarding, 454­55 collective action as purpose of, 351 for covered tradable agricultural products, future developments stemming from, 53­55 by region, 46­48, 47f increased rates of protection despite, 505 defining, 422­28 macroeconomic effects of distortions and, 51 for exportables subsector, 427­28 North American agricultural policy and, 183 GDP and WRI, 450, 453f Northeast Asian advanced economies and, 79, general global equilibrium model 81, 84 compared, 454 South Asian and India affected by, 397, 413 for high-income countries, 436f, 437t, 439f, Southeast Asia and China, domestic price 440t, 441f, 442­49t stabilization efforts in, 379 for import-competing subsector, 422­27 Western European agricultural policy and, for major agricultural products, 50f 117, 138, 141, 143, 165, 166, 168 marginal welfare weights, 455 Uzbekistan, 261t, 266, 271, 280, 281 NRAs and CTEs compared, 46, 420­22, 424, 438, 450, 453f, 454­55 V PDI and CDI, WRI computed from, 421, Valdés, Alberto, 5­6, 10, 28, 60­61n6, 60n4, 435­38, 442­43t, 455 115n, 289, 294, 481, 506, 565n, 579­80, regional shares of Asian welfare reduction, 585, 617n1 48, 49f Valenzuela, Ernesto, 51, 54, 67n, 115n, 177n, results for, 435­54 221n, 259n, 260, 289n, 323n, 359n, 389n, PDI and CDI, 435­38, 442­43t 419, 438, 459, 460, 505, 507, 508, 565, 595 TRIs, 438, 439f, 440t, 441f, 447­49t van der Mensbrugghe, Dominique, 51, 52, 54, WRIs, 41f, 435­38, 436f, 437t, 444­46t 460, 505, 538, 554 tariffs and NTMs, 422, 424, 426 Vietnam. See also Asia; Southeast Asia and China Western Europe, 115­76. See also specific conclusions regarding, 380 countries CTEs, 363­64, 371, 375, 376­77t, 382 biofuel production, subsidization of (2008), 3 domestic price stabilization, 375­79 CAP. See Common Agricultural Policy economic growth and structural change in, (CAP), EU 361­63 conclusions regarding, 170­71 GDP in, 361­62, 370t, 377t, 383f CTEs for, 44­45t, 153, 158­59t, 431t, 432f general equilibrium effects in, 510t, 516t, current chronology of policy distortions 524t, 532t, 535, 536t, 542t, 551t, 559t (1955­2007), 123­43 Index 643 1955­1964, 124­30, 128f derived shares of global production, 601t 1965­1974, 130­34, 133t export shares, 605t 1975­1984, 134­36, 137f general global equilibrium effects, 521t, 528t, 1985­1994, 136­41, 140f 530t, 541t, 546­48t 1995­2007, 141­43, 144f global value of production, NRA percentage EEC/EU, formation of (1957/1993), coverage of, 461t, 463t, 464t 123­24, 173n29 gross value of global production at Eastern Europe and EU. See under Eastern undistorted prices, 599t Europe and Central Asia GSE values, 465f, 466­68t, 469f empirical estimates of distortion for, 13, in North America, 180, 182­83, 185, 186, 192, 15­17t, 34t 196, 197, 217n5 exchange rates, 119, 124, 130, 131, 134, 165 NRAs for, 469f, 472­74, 475f exports, contribution of agriculture to, 118 production export and consumption food subsidy war with North America imported, shares in, 482t, 609t (1986), 3 regional production shares, 603t future developments in, 53, 54, 167­70 total shares in global production, 607t GDP in, 117­18, 172n8 WRIs and TRIs, 488, 489­90t, general equilibrium effects in, 511t, 517t, 491­92f 525t, 534t, 537t, 544t, 552t, 560t Williamson, J., 58 GSE values, 18f, 19, 41­42t, 153, 155­57t Winter-Nelson, Alex, 565n historical distortions of agricultural incentives World Bank, 331, 595. See also Agricultural (before mid-1950s), 4, 8­9, 117­23 Distortions Project database key commodities in EU-15, 471f, 473f, 475f, World Trade Organization (WTO). See also 493­94f Uruguay Round Agreement on liberalization of agricultural trade in, 59­60t Agriculture nonfarm sectors in, 119, 145, 147­48t China's accession to, 55­56, 56f, nontraded food staples, 482t, 484t 384, 508 Northeast Asian advanced economies collective action problem and, 351­53 compared, 99 Doha Round. See Doha Round NRAs for Eastern European and Central Asian accession 1955­1964, 127, 128f to, 260, 284 1965­1974, 132, 133f GATT, replacing, 79, 117, 183 1975­1984, 136, 137f Latin American and the Caribbean accession 1985­1994, 139, 140f, 141 to, 299 1995­2007, 143, 144f North American agricultural policy and, 183, in Agricultural Distortions Project 187, 215, 216 database, 429­30t, 432f South Asia and India affected by, 413­14 analysis of, 143­53, 145­46f, 147­52t, 154t Southeast Asian and Chinese domestic empirical estimates of distortion, 19f, 20t, price stabilization efforts and, 379, 21, 24t, 26t 383­84 policy interventions and outcomes in, 153­66 Western European agricultural policy and, compared to other high-income countries, 117, 162, 168­70 162­63 World War I, 9, 121­22, 180 environmental objectives, 163­64 World War II long-run trends, 160 Japan prior to, 103­10, 105­7t macroeconomic effects, 165­66 North America and, 181 productivity issues, 118, 160­62 Northeast Asian advanced economies' reinstrumentation, 164­65 evolution of agricultural policy productivity increases in, 118, 160­62 following, 77­85 Rome, Treaty of, 126, 127 Western Europe's economic transformation tax and tariff policies in, 121, 122 following, 115, 122­23 URAA and, 117, 138, 141, 143, 165, WRIs. See welfare reduction and trade reduction 166, 168 indexes WRIs and TRIs for, 436f, 437t, 439f, 440t, WTO. See World Trade Organization 441f, 442­49t, 450, 451f wheat in Australia and New Zealand, 240, 242 Y CTEs for, 478­79t Yugoslavia, former, 264, 285 644 Index Z Zimbabwe. See also Africa Zambia. See also Africa CTEs, 350t CTEs, 350t evolution of trade policies in, 329, 330, 331 Zambia (continued) general equilibrium effects in, 509t, 516t, 523t, general equilibrium effects in, 509t, 516t, 532t, 535, 536t, 542t, 549, 550t, 559t 523t, 532t, 535, 536t, 542t, 549, growth and structural change in, 328 550t, 559t GSE values, 343f, 345­46t growth and structural change in, 328 key commodities, 475f, 476f, 493­94f, GSE values, 343f, 345­46t 496­98t key commodities, 471f, 472, 475f, 476f, 493f, key economic indicators, 326t 497­98t liberalization of agricultural trade in, 53 key economic indicators, 326t NRAs for, 334t, 335f, 336, 337t, 340 NRAs for, 334t, 335f, 336, 337t reversals of reform in, 327 RRA/TBI, 352f RRA/TBI, 352f WRIs and TRIs for, 441f, 444t, 447t WRIs and TRIs for, 441f, 444t, 447t ECO-AUDIT Environmental Benefits Statement The World Bank is committed to preserving Saved: endangered forests and natural resources. The · 21 trees Office of the Publisher has chosen to print · 7 million British Distortions to Agricultural Incentives: A Global thermal units of Perspective, 1955­2007 on recycled paper with total energy 30 percent postconsumer fiber in accordance · 1,971 pounds of net with the recommended standards for paper greenhouse gases usage set by the Green Press Initiative, a nonprofit · 9,493 gallons of program supporting publishers in using fiber that waste water is not sourced from endangered forests. For more · 576 pounds of information, visit www.greenpressinitiative.org. solid waste The international agricultural economy is bewilderingly complex; an up-to-date study of economic policies and their effects has been sorely needed. Kym Anderson has done a remarkable job of leading such a huge ambitious study. This volume provides the synthesis and will serve as the standard reference for years to come, documenting the impacts of current policies toward agriculture and analyzing changes over the past several decades. With increasing concerns about global food supplies, it should be required reading for academics and policy makers seeking to understand the current structure of the international food economy and considering policy alternatives. Anne O. Krueger Professor, Johns Hopkins University former First Deputy Managing Director, International Monetary Fund On the basis of a very large body of new empirical evidence from case studies covering 75 countries, Kym Anderson provides the most comprehensive analytical assessment ever done of the world's agricultural trade policies and the changes during the past 50 years. This book presents very strong and timely arguments for revitalizing the WTO trade negotiations to promote global economic growth and poverty alleviation and to help achieve the Millennium Development Goals. It is a must-read for scholars and students of agricultural development and international trade and others who wish to better understand the costs of the current antitrade bias in both developing and developed countries. Per Pinstrup-Andersen Professor, Cornell University and Copenhagen University former Director-General, International Food Policy Research Institute World Food Prize Laureate This book provides absolutely unique information on agricultural policies around the world, in an internationally comparable manner. It will serve a whole generation of analysts and policy makers with a benchmark against which to assess the pursuit of future policies in world agriculture. Stefan Tangermann Emeritus Professor, University of Göttingen former OECD Director for Trade and Agriculture Food and agriculture are key contributors to global welfare, and no sector is more prone to government intervention. Kym Anderson and his collaborators have measured that intervention over half a century and shown that it is still alive. Data and quantification are the necessary precursors to effective research and to policy reform, so this excellent book provides a major impetus to rationalizing international trade policy in agriculture. It is essential reading for all policy makers and students who care about feeding the world. L. Alan Winters Professor, University of Sussex Chief Economist, UK Department for International Development ISBN 978-0-8213-7665-2 Printed in the United States SKU 17665