Agriculture AND WTO THE CREATING A TRADING SYSTEM FOR DEVELOPMENT Editors Merlinda D. Ingco and John D. Nash Agriculture and the WTO Creating a Trading System for Development Agriculture and the WTO Creating a Trading System for Development Merlinda D. Ingco and John D. Nash Editors A copublication of the World Bank and Oxford University Press © 2004 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 07 06 05 04 The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the Board of 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 judgment 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 work is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The World Bank encourages dissemination of its work and will normally grant permission 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, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Photo credits: Ship/Grain image--© Paul A. Souders/Corbis; Tomator/conveyor belt image--Willard R. Culver/NATIONAL GEOGRAPHIC IMAGE COLLECTION/Getty Images. ISBN 0-8213-5485-X Library of Congress Cataloging-in-Publication data has been applied for. Contents List of Boxes vii List of Tables viii List of Figures xi Foreword xiii Acronyms and Abbreviations xv 1. WHAT'S AT STAKE? DEVELOPING-COUNTRY INTERESTS IN THE DOHA DEVELOPMENT ROUND 1 Merlinda D. Ingco and John D. Nash 2. TRADE AGREEMENTS: ACHIEVEMENTS AND ISSUES AHEAD 23 Merlinda D. Ingco and John Croome 3. EXPORT COMPETITION POLICIES 43 Harry de Gorter, Lilian Ruiz, and Merlinda D. Ingco 4. MARKET ACCESS: ECONOMICS AND THE EFFECTS OF POLICY INSTRUMENTS 63 Harry de Gorter, Merlinda D. Ingco, and Laura Ignacio 5. QUOTA ADMINISTRATION METHODS: ECONOMICS AND EFFECTS WITH TRADE LIBERALIZATION 95 Harry de Gorter and Jana Hranaiova 6. DOMESTIC SUPPORT: ECONOMICS AND POLICY INSTRUMENTS 119 Harry de Gorter, Merlinda D. Ingco, and Laura Ignacio 7. THE DISTRIBUTIONAL EFFECTS OF AGRICULTURAL POLICY REFORMS 149 Harry de Gorter, Merlinda D. Ingco, and Cameron Short 8. THE"MULTIFUNCTIONALITY" OF AGRICULTURE AND ITS IMPLICATIONS FOR POLICY 167 David Vanzetti and Els Wynen 9. FOOD SECURITY AND AGRICULTURAL TRADE POLICY REFORM 179 Merlinda D. Ingco, Donald Mitchell, and John D. Nash 10. MANAGING POTENTIAL ADVERSE IMPACTS OF AGRICULTURAL TRADE LIBERALIZATION 193 William Foster and Alberto Valdés v vi Agriculture and the WTO 11. THE SANITARY AND PHYTOSANITARY AGREEMENT, FOOD SAFETY POLICIES, AND PRODUCT ATTRIBUTES 215 Simonetta Zarrilli with Irene Musselli 12. AGRICULTURAL BIOTECHNOLOGY: A PRIMER FOR POLICYMAKERS 235 Donald J. MacKenzie and Morven A. McLean 13. GLOBAL INTELLECTUAL PROPERTY RIGHTS: A NEW FACTOR IN FARMING 253 Geoff Tansey 14. RULES AND OPTIONS FOR SPECIAL AND DIFFERENTIAL TREATMENT 269 Constantine Michalopoulos 15. SPECIAL TRADE ARRANGEMENTS TO IMPROVE MARKET ACCESS 291 Helen Freeman APPENDIX A: OECD POLICY EVALUATION MATRIX AND TRENDS IN POLICY FOR VARIOUS COMMODITIES 317 Cameron Short and Harry de Gorter APPENDIX B: THE AGREEMENT ON AGRICULTURE 349 INDEX 367 list of boxes Box 1.1 The Importance of Agriculture to Developing Countries 3 Box 1.2 The Human Face of Policy Incoherence 9 Box 1.3 Reforming Inefficient Support Systems: Recent Experience in Two Developing Countries 19 Box 1.4 Cambodia Rice: Challenges to Integration 20 Box 2.1 The Price of Multilateral Negotiations 24 Box 2.2 Uruguay Round Agreement on Agriculture: Important Documents 27 Box 3.1 Two Examples of Consumer-Only-Financed Export Subsidy Schemes 56 Box 4.1 Transparency and Tariffication 70 Box 4.2 More on Tariff Quotas 80 Box 5.1 Tariff Quotas: Categories of Principal Administration Methods 99 Box 5.2 Tariff Quotas: Categories of Additional Conditions 100 Box 5.3 Tariff Quotas: Economics of First-Come, First-Served (FCFS) 114 Box 5.4 Tariff Quotas: Factors Affecting the Impacts on Trade with an STE in Importing Country 115 Box 6.1 The Peace Clause and Domestic Subsidies 131 Box 8.1 Aspects of Valuation 172 Box 9.1 Food Security Indicators 182 Box 10.1 Farming without Subsidies: The Experience of New Zealand 206 Box 11.1 The Joint FAO/WHO Codex Alimentarius Commission (CAC) 218 Box 11.2 The Office International des Epizooties (OIE) 219 Box 11.3 The International Plant Protection Convention (IPPC) 220 Box 12.1 The Global Trade Effects of China Bt Cotton 238 Box 13.1 The U.K. Commission on Intellectual Property Rights' Recommendations on Agriculture and Genetic Resources 267 Box 15.1 Trade Agreements: AFTA to SADC 293 Box 15.2 A Brief History of the GSP Schemes 294 Box 15.3 The U.S. Generalized System of Preferences (GSP) and Argentina's Economic Crisis, 2001­02 298 Box 15.4 U.S. Generalized System of Preferences (GSP): Criteria and Conditions 299 Box 15.5 U.S. Generalized System of Preferences (GSP): Annual Timetable 300 Box 15.6 Parties to Regional Trade or Preferential EU Trade Agreements in Force as of April 2002 303 Box 15.7 The Americas: Trade Diversion--Displacement from Markets 311 Box 15.8 Chile: Strategy for Growth--Access or Diversion? 312 vii list of Tables Table 1.1 Rural-Urban Poverty Gap 4 Table 1.2 Agricultural Protection in Developing Countries, 1984­87 and 1994­98 11 Table 1.3 Applied Tariff Rates in Developing and Industrial Countries, 1994­98 (Percent) 12 Table 1.4 World Trade Simulation Model Product Categories 14 Table 1.5 Frequency of NTBs in Developing and Industrial Countries, 1994­98 (Percent) 15 Table 1.6 Export Growth Rates (Constant 1995 US$) 16 Table 1.7 Shares of Developing and Industrialized Countries in World Exports (Percent) 16 Table 1.8 Gains from Removing All Trade Barriers in Agriculture and Food Globally, Post-Uruguay Round, 2005 (in 1997 US$ Billions) 17 Table 3.1 Percentage of Total Export Volume Receiving Export Subsidies 44 Table 3.2 Total Export Subsidy Commitments 44 Table 3.3 Percentage Use of Value Commitments by Country 46 Table 3.4 Percentage Allocation of Total Value Commitments by Commodity 47 Table 3.5 Percentage Use of the Total Value Commitments Allocated to Each Commodity Group 48 Table 3.6 Percentage Use of the Total Volume Commitments Allocated to Each Commodity Group 49 Table 3.7 Countries Using over 90 Percent of Value Commitments 50 Table 3.8 Countries Using over 90 Percent of Volume Commitments 51 Table 3.9 Value Front-Loading 52 Table 3.10 Volume Front-Loading 52 Table 3.11 Export Subsidy Equivalents (ESEs) (Percent) 54 Table 4.1 Empirical Estimates of Transfers Due to Policies in World Agriculture, US$ Millions (1999­2001 average) 72 Table 4.2 Empirical Estimates of Transfers Due to Policies by Commodity in World Agriculture, US$ Millions (1999­2001 average) 73 Table 4.3 Examples of Tariff Peaks and Dispersion in Agriculture 75 Table 4.4 Tariff Escalation (Weighted Average MFN Applied Tariffs in Percentage) in the Quad Markets (U.S., EU, Japan, and Canada) 77 Table 4.5 Special Safeguards Tabled in the URAA 78 Table 4.6 Analyzing Tariff Equivalents Using the Swiss Formula (maximum tariff of 25%) 84 Table 4.7 World Production and Value of Production and Tariff-Quota-Related Protection for Grains and Oilseeds in the Six Regions of the PEM Model 88 Table 4.8 Import Quotas in PEM 88 Table 4.9 Tariffs Applicable to Tariff Quota Commodities 89 Table 4.10 Tariff Rates (US$/mt) 90 Table 4.11 Impact of Tariff Quota Border Protection (US$/mt) 91 Table 5.1 Tariff Quotas by Product Category 97 Table 5.2 Number of Tariff Quotas by Member Country 98 Table 5.3 Tariff Quotas by Principal Administration Method, 1995­2001 98 Table 5.4 Tariff Quotas: Simple Average Fill Rates, 1995­2001 107 Table 5.5 Distribution of Fill Rates, 1995­2000 108 viii List of Tables ix Table 6.1 Agricultural Support in OECD Countries, US$ Millions (1999­2001 Average) 123 Table 6.2 Domestic Producer Support Relative to Total and Trade-Distorting Support in World Agriculture, US$ Millions (1999­2001 average) 124 Table 6.3 Empirical Measures of Domestic Support by Commodity in World Agriculture, US$ Millions (1999­2001 Average) 124 Table 6.4 Composition of Domestic Support by Country, 1995­1998 (percent) 125 Table 6.5 Use of Total Aggregate Measurement of Support (AMS) Commitments by Member, 1995­2001 (percent) 126 Table 6.6 AMS Commitments Versus Actual 127 Table 6.7 Evolution of Aggregate Measure of Support (AMS) and Producer Support Estimate (PSE), US$ Billions 129 Table 6.8 Total Green Box Expenditures by Category, US$ Millions, 1995­98 133 Table 6.9 de minimis Support 134 Table 6.10 2001 Baseline Support by Commodity Region and Type (US$/tonne) 136 Table 6.11 Effect of Removing All Border Protection on Effective Producer Price and Production 137 Table 6.12 Effect on Effective Producer Price and Production of Removing All Direct Payments Support 137 Table 6.13 Effect of Trade Liberalization (US$ Millions) 138 Table 7.1 Liberalization Effects on Wheat Price, Production, and Exports by Region, Base Levels, and Percentage Change 153 Table 7.2 Liberalization Effects on Coarse Grains Price, Production, and Exports by Region, Base Levels, and Percentage Change 154 Table 7.3 Liberalization Effects on Oilseeds Price, Production, and Exports by Region, Base Levels, and Percentage Change 155 Table 7.4 Liberalization Effects on Rice Price, Production, and Exports by Region, Base Levels, and Percentage Change 155 Table 7.5 Welfare Effects of Liberalization by Region (US$ Millions) 156 Table 7.6 Distribution of the Effects of Global Liberalization on Mexico: Base Income Levels and Percentage Change 157 Table 7.7 Welfare Effects of Liberalization by Region (US$ Millions) 158 Table 7.8 Distribution of the Effects of Liberalization on Mexican Base Income Levels and Percentage Change 159 Table 7.9 Impacts of Hypothetical Changes in Maize Policies in Mexico (US$ Millions) 160 Table 9.1 Average Cereal Prices (US$/ton) 185 Table 10.1 Coefficients of Variation of World Prices of Selected Commodities 198 Table 10.2 Descriptive Statistics for Selected Commodities 199 Table 10.3 Decomposition of Real Producer Price for Wheat in Transition Economies 201 Table 10.4 Decomposition of Producer Price Changes: Argentina 202 Table 10.5 Decomposition of Producer Price Changes: Chile 203 Table 10.6 Decomposition of Producer Price Changes: Columbia 204 Table 12.1 Labeling Requirements for Genetically Modified Foods 242 Table 12.2 Online Resources 245 Table 12.3 The Cartagena Protocol on Biosafety 246 Table 13.1 Online Resources 254 Table 15.1 U.S. GSP Beneficiaries, 1999 (Independent Countriesa) 296 Table 15.2 Leading Sources of U.S. GSP Imports, 2000 297 Table 15.3 Leading Product Groups Imported by the U.S. Duty-Free under GSP, 2000 297 x Agriculture and the WTO Table 15.4 U.S. Imports from CBERA Countries, Total and Under-Selected Import Programs, MFN-Free, GSP, and CBERA 298 Table 15.5 U.S.-Sub-Saharan Africa Trade: Major U.S. Import Suppliers under the Generalized System of Preferences, and the African Growth and Opportunity Act (YTD Jan.­Jun, AGOA-Eligible Countries Only in US$ Millions) 302 Table 15.6 EU: Applied MFN Tariffs by HS Chapters 01­22, 2002 (percent and U.S.$ Billions) 304 Table 15.7 Effective Benefits of Quad Countries: Generalized System of Preferences for Least-Developed Countries, Late 1990s 308 Table 15.8 Preference Margins for Selected Groups of Agricultural Products Exported from the AACP to the EU, Lome Preferences and GSP Preferences Compared, 1999 EU Tariffs (1997 Trade Data) 309 Table 15.9 Preference Margins for Protocol Beef and Sugar Exported from Individual AACP Countries to the European Union under Lome Provisions, 1999 EU Tariffs (1997 Trade Data) 310 Table 15.10 Preference Margins for Selected Groups of Agricultural Products Exported from the AACP to the European Union under Lome Provisions, Hypothetical EU Tariffs after the Next WTO Round (1997 Trade Data) 313 Table A.1 OECD Support for Grains and Oilseeds 2001 (in US$ Millions) 318 Table A.2 Type of Support as a Percentage (%) of Total Support for a Commodity 320 Table A.3 Support to a Specific Commodity as a Percentage (%) of Total Type of Support 322 Table A.4 Support as a Percentage (%) for Grains and Oilseeds Distributed among Specific Commodities 324 Table A.5 OECD-6 as a Percentage (%) of OECD Total for Selected Commodities and Support Types 325 Table A.6 Policy Instruments for Grains and Oilseeds in Mexico (1980­present) 334 Table A.7 Model Outline 337 Table A.8 Expenditure by Decile, 2000 in Million Pesos (Per Capita in Parenthesis, US$) 343 Table A.9 Allocation of Household Survey Expenditure Categories 343 Table A.10 Maize Production Systems 344 Table A.11 Expenditures Per Capita by Decile, 2001 (US$) 344 Table A.12 Expenditure Parameters and Elasticities for the Base Period, 2001 345 Table A.13 Maize Production Coefficients 345 list of Figures Figure 1.1 Trade Has Propelled Growth, 1965­2000 6 Figure 1.2 Tariffs and Trade Growth 7 Figure 3.1 Baseline Allocation of the Value of Export Subsidies 46 Figure 3.A.1 Excess Supply Curve Shifts Outward from ES0 to ES1 61 Figure 3.A.2 Excess Supply Curve Shifts Inward from ES0 to ES2 61 Figure 4.1 Evolution of Border versus Total Support in OECD Agriculture (US$ Billions) 65 Figure 4.2 OECD: Grains and Oilseeds 66 Figure 4.3 OECD: Other Commodities 66 Figure 4.4 Average Agricultural Bound Tariffs by Region 2001 (in percent) 67 Figure 4.5 Applied versus Bound Tariffs in Developing Countries (Tariff Binding Overhang) 67 Figure 4.6 Applied versus Bound Tariffs in Developed Countries (Tariff Binding Overhang) 68 Figure 4.7 Average Bound Tariffs by Commodity, 2001 69 Figure 4.8 Nominal Protection Coefficient by OECD Country 71 Figure 4.9 Nominal Protection Coefficient by Commodity (OECD) 71 Figure 4.10 Water in the Tariff--Applied Tariffs versus Tariff Equivalent of Binding TRQs for Selected OECD Countries (%) 74 Figure 4.11 Water in the Tariff--Applied Tariffs versus Tariff Equivalent of Binding TRQs for Selected OECD Commodities (%) 74 Figure 4.12 The Three TRQ Regimes 79 Figure 4.13 Definitions of Alternative Tariffs 81 Figure 4.14 Out-of-Quota Imports with Quota Underfill and Out-of-Quota Imports 81 Figure 5.1 Share of Output under Tariff Quotas 109 Figure 6.1 Trends in Domestic Support 122 Figure 6.2 Trends in AMS versus PSE (1986­2001), US$ Billions 130 Figure 9.1 Cereal Import Costs per Ton 185 Figure 9.2 World Grain Stocks and Price, 1990­2000 186 Figure 10.1 Real Price of Selected Commodities (1960­97, US$ as of July 1997) 199 Figure 10.2 Price Band of Sugar in Chile 208 Figure 10.3 Price Band of Edible Oil in Chile 208 Figure 12.1 Global Area of Transgenic Crops 237 Figure A.1 OECD--Grains and Oilseeds 326 Figure A.2 OECD--Other Commodities 326 Figure A.3 Canada--Grains and Oilseeds 327 Figure A.4 Canada--Other Commodities 327 Figure A.5 United States--Grains and Oilseeds 328 Figure A.6 United States--Other Commodities 328 Figure A.7 European Union--Grains and Oilseeds 329 Figure A.8 European Union--Oilseeds 329 Figure A.9 European Union--Rice 330 Figure A.10 European Union--Other Commodities 330 Figure A.11 Japan--Grains and Oilseeds 331 Figure A.12 Japan--Other Commodities 331 Figure A.13 Switzerland--Grains and Oilseeds 332 xi xii Agriculture and the WTO Figure A.14 Mexico--Grains and Oilseeds 332 Figure A.15 Mexico--Rice 333 Figure A.16 Mexico--Other Commodities 333 Figure A.17 Production and Input Supply Responses to Increased Commodity Price 339 foreword There are misconceptions that trade liberalization all, one that crafts the world trading system in such is a done deal. In fact, we are a long way away from a way that developing countries are given both free world trade, particularly in areas of interest to strong incentives and better opportunities to use developing countries. The Doha negotiations mark trade integration more actively as a growth lever. To the first time that developing country interests have help these countries to take full advantage of global been placed at the center of a multilateral round of market opportunities, the World Bank has also trade negotiations. Those interests include agricul- stepped up efforts to make trade a more central ture, implementation of textile agreements, the use part of the policy dialogue in its operations. As one of antidumping, and the nature of special and dif- example, comprehensive trade diagnostic studies ferential treatment for developing countries, have been launched in 20 low-income countries in including the extent to which these countries the last two years, and about 40 countries--both should undertake more substantive commitments low and middle income--have been targeted for to liberalize their own trade regimes. But these are stepping up the Bank's trade-related operational some of the hardest areas for countries to address activity. The Bank took advantage of the oppor- and, unfortunately, from the very beginning key tunity of the Cancun Ministerial meeting to negotiating deadlines on most of these issues were announce a special Trade Assistance Program missed. As a result, substantial gaps remain designed to support progress on the Doha Devel- between the developing and the industrial coun- opment Agenda and to increase assistance to coun- tries, and in none of the areas is this more true than tries that take on development-promoting trade in agriculture. If any further proof of this were reforms. This will make needed resources available needed, it became painfully clear at the Cancún for countries implementing new trade reform pro- Ministerial meetings. grams associated with commitments they may The World Bank has been actively advocating the make in the Doha Agenda. These resources could reduction of trade-distorting payments and subsi- be used to expand activities that up-grade competi- dies to farmers and exporters in industrialized tiveness over the long term, such as training of countries, as well as the reduction of non-tariff bar- workers and reforms of trade-related institutions. riers to trade and tariffs by both developed and Such loans can be accelerated and, depending on developing countries, which act as a tax on develop- country circumstances, could be additional to ment. We have also been supporting developing existing country lending levels. countries in analyzing the outcomes of the Uruguay This volume is an integral part of the Bank's Round and preparing for the Doha negotiations, advocacy, research, and capacity-building program, through a program of research, technical assistance, with messages aimed at audiences in both develop- and capacity building in order to help ensure a pro- ing and industrialized countries. Its messages build development outcome of the negotiations. on those of two previous publications of the World We have stressed the importance of trade inte- Bank: The Uruguay Round and the Developing gration as an instrument for achieving the Mil- Countries (1996), and more recently, Development, lennium Development Goals (MDGs); though Trade, and the WTO: a Handbook (2002). The latter increased aid is essential to meeting the MDGS, the publication covers most areas of interest to devel- gains from trade integration are estimated to be far oping countries in the Doha negotiations, with the larger than any contemplated increase in aid flows. exception of agriculture. The current volume is In our view a pro-development outcome is, above thus its complement. Other forthcoming publica- xiii xiv Agriculture and the WTO tions will focus on topics in the agricultural trade essential) that the Doha process will continue to negotiations that are of interest to specific regions move forward and contribute to the achievement of of the developing world. long-term poverty reduction objectives. Given the Funding for the research that is the basis for this central role of agriculture in the Doha process, we publication, regional copublications, and work- hope this book will contribute to its ultimate success. shops was provided primarily through the United Kingdom's Department for Foreign and Interna- Kevin M. Cleaver tional Development (DFID). Additional resources Director were contributed by the World Bank­Netherlands Agriculture and Rural Development Partnership Program (BNPP). We gratefully acknowledge their support. Uri B. Dadush Although the outcome of the Cancún Ministerial Director was not what we had hoped, it seems likely (indeed International Trade Department acronyms and abbreviations AACP African ACP countries CGSB Canadian General Standards Board ABARE Australian Bureau of Agricultural CIF Cost, insurance, and freight and Resource Economics CMOB Common Market Organization for ABS Access and benefit sharing Bananas ACP African, Caribbean and Pacific CNL Competitive Need-Limitation Group of States CONASUPO National Company of Popular AFTA ASEAN Free Trade Area Subsistence (Mexico) AGOA African Growth and Opportunity COP Conference of the Parties Act CRP Conservaton Reserve Program AGST Agriculture Supporting Tables CSE Consumer Subsidy Equivalent AIA Advance Informed Agreement CTA Committee on Trade in Agriculture AMS Aggregate Measurement of Support (GATT) APEC Asia Pacific Economic Co-operation CTD Committee on Trade and Forum Development (WTO) ASERCA Support Services for Agricultural DS Domestic Support Marketing (Mexico) EBA Everything But Arms Trade ATPA Andean Trade Preference Act and Agreement Drug Eradication Act EC European Community or European BSE Bovine Spongiform Commission Encephalopathy EMS Equivalent Measurement of CAAS Chinese Academy of Agriculture Support Sciences EPA Economic Partnership Agreement CAC Codex Alimentarius Commission ES Export subsidies CAP Common Agricultural Policy ESE Export subsidy equivalent CARICOM Caribbean Community FAC Food Aid Convention CBD Convention on Biological Diversity FAO Food and Agriculture Organization CBERA Caribbean Basin Economic Recov- of the United Nations ery Act FAOSTAT Food and Agriculture Organization CBTPA Caribbean Basin Trade Partnership Statistical Database Act of 2000 FDI Foreign direct investment CCFICS Codex Committee on Food Import FOB Freign on board and Export Inspection and FCFS First-come, first-served Certification Systems FSC Foreign Sales Corporation CCGD Canadian Council of Grocery FTA Free trade area Distributors FTAA Free Trade Areas of the Americas CER Closer Economic Relations GATS General Agreement on Trade in (Agreement) Services CES Constant elasticity of substitution GATT General Agreement on Tariffs and CFF Compensatory Financing Facility Trade (IMF) GDP Gross Domestic Product CGE General equilibrium model GEF Global Environment Facility CGIAR Consultative Group on GI Geographical indication International Agricultural Research GM Genetically modified xv xvi Agriculture and the WTO GMO Genetically modified organism NAFTA North American Free Trade GSP Generalized System of Preferences Agreement GTAP Global Trade Analysis Project NFIDC Net-food-importing developing IARC International Agricultural Research country Centre NGO Nongovernmental organization IATRC International Agricultural Trade NTM Nontariff measure Research Consortium NTB Nontrade barrier ICPM Interim Commission on OECD Organisation for Economic Phytosanitary Measures Co-operation and Development IFPRI International Food Policy Research OIE International Office of Epizooties Institute OPS Output price support IGC Intergovernmental Committee on PBRs Plant Breeder's Rights Intellectual Property and Genetic PEM Policy Evaluation Matrix Resources, Traditional Knowledge PIC Prior informed consent and Folklore PNT Plants with novel trait ILA Agreement on Import Licensing PPM Processing and production method Procedures PROCAMPO Farmers Direct Support Program IMF International Monetary Fund (Mexico) IPGRI International Plant Genetic PRSP Poverty Reduction Strategy Papers Resources Institute PSE Producer Support Estimate IPPC International Plant Protection PVP Plant variety protection Convention QR Quantitative restriction IPS Input price subsidies R&D Research and development ISAAA International Service for the ROO Rules of origin Acquisition of Agri-biotech RTA Regional trade agreement Applications RUNS Rural-Urban North-South ISPM International standard for S&D Special and Differential (treatment) phytosanitary measures SP Sugar Protocol (EU) ITF International Task Force on SPS Sanitary and Phytosanitary Commodity Price Risk Agreement Management SSA Sub-Saharan Africa ITO International Trade Organization SSG Special safeguard ITPGRFA International Treaty on Plant STE State trading enterprise Genetic Resources for Food and SW Swiss formula Agriculture TAED Transatlantic Environmental IU International Undertaking on Plant Dialogue Genetic Resources for Food and TBT Technical Barriers to Trade Agriculture TNC Trade Negotiations Committee LDC Least-developed country TRIPS Trade-Related Intellectual Property LDBDC Least Developed Beneficiary TRQ Tariff Rate Quota Developing Country TSE Total support estimate LES Linear Expenditure System UNCTAD United Nations Conference on LMO Living modified organism Trade and Development MAT Mutually agreed terms UNEP United Nations Environment MERCOSUR Southern Common Market (Mercado Programme Comune del Sur) UPOV International Union for the Protec- MFN Most favored nation tion of New Varieties of Plants MPS Market price support (Union pour la Protection des MTA Material transfer agreement Obtentions Végétales) Acronyms and Abbreviations xvii URAA Uruguary Round Agreement on WHO World Health Organization Agriculture WIPO World Intellectual Property USITC U.S. International Trade Organization Commission USTR Office of the United States Trade Representative 1 What's at Stake? Developing-Country InterestS in the Doha Development Round Merlinda D. Ingco and John D. Nash Introduction resistance of domestic lobbies for protection of Developing countries have a huge stake in the suc- agriculture without securing support from other cess of the Doha Development Round. First, there sectors of the economy that would gain from trade are the potential gains from the strengthening of a reforms in other countries. In such cases, reciprocal rules-based global trading system for agriculture trade agreements under the auspices of the World and reducing distortions in global agricultural Trade Organization (WTO) may be easier to markets. Developing countries are the weaker play- undertake and are thus an important part of the ers in the trading system, and thus will benefit the political economy of policy reform in developing most when the dominant trading countries play by countries. In addition, supply response to reform common rules that discipline government activities depends upon the credibility of the reform process. supporting agriculture in the three key areas of Experience in many countries shows that the pri- domestic support, market access, and export com- vate sector does not invest if the persistence of petition. Developing countries will benefit from reforms is in doubt. Policy measures can be made further reforms in these three areas and will also more credible through a framework of a multilat- benefit from reforms in antidumping rules, which eral agreement that requires adherence to rules and are used in an increasing number of countries is equipped with built-in instruments that prevent (including developing countries). Small economies policy reversals and backsliding. and the least-developed countries will especially Third, the new round of trade negotiations benefit from the expansion of beneficial provisions should help advance global agricultural trade liber- for developing-country exporters. alization and thereby expand trade. In particular, Second, multilateral agreements and trade nego- the negotiations should achieve further cuts in tiations should help developing countries under- domestic and export subsidies in the Organisation take and lock-in their own trade and domestic pol- for Economic Co-operation and Development icy reforms needed to advance their development (OECD) countries and in reductions in high tariffs objectives. Ideally, countries will implement trade in OECD and developing countries. These reduc- and other policy reforms unilaterally because it is tions will be difficult to achieve outside the context in their interest to do so. However, in practice, most of global trade talks. Most developing countries will governments have difficulty in overcoming the benefit from increased market access to industrial 1 2 Agriculture and the WTO country markets for products such as sugar, beef, complementary measures and preconditions for and fruits and vegetables. Increasing trade will be trade policy reforms to have their greatest positive useful in its own right to expand markets and will impacts. also reduce the volatility of world prices (which is high partially because of the thinness of markets A Vibrant Agricultural Sector when so many countries try to insulate their Is Crucial domestic production). Since both producers and consumers in poor countries are especially vulnera- Agricultural sector growth is crucial for achieve- ble to large unpredictable price fluctuations, these ment of a number of development goals for groups have a special interest in reducing world developing countries. Among these goals are price volatility. Even though agricultural trade lib- enhancing overall economic growth and poverty eralization may raise world food prices slightly, reduction, improving food security, and conserving the net benefit should be positive for most develop- natural resources. ing countries, not just for current agricultural exporters. Those countries close to food self- Reducing Poverty through Economic Growth sufficiency may become net exporters following protection cuts in other countries. Other countries In low-income countries, owing to its relative size may remain net food importers only because they and its important growth linkages to the rest of the retain strong anti-agricultural domestic policies. economy, the agricultural sector is the primary Those countries, too, may benefit from global trade engine of overall economic growth (see box 1.1). liberalization because that would discourage some Agriculture is by far the largest employer in these of their resources from being employed in less countries, employing 68 percent of the labor force socially productive activities outside agriculture. and producing 24 percent of gross domestic prod- Even net importing countries, where many of the uct (GDP). In middle-income countries the share poor are in farm households, may benefit directly of GDP falls below 10 percent, but agriculture still from a global increase in prices and increased pros- accounts for one-quarter of total employment. perity in the rural sector. Many of the world's poor depend directly on agri- culture for their livelihoods (Fan, Hazell, and Tho- rat 1999). Increased agricultural productivity also The Plan of This Chapter provides cheaper food, which makes up a large This chapter elaborates on the basic theme that it is share of expenditures of poor households (Fan of critical interest for developing countries that the 2000). In addition, a modernizing agriculture cre- current Doha Round succeed in meeting its objec- ates jobs in agricultural processing and marketing, tives, particularly in reducing barriers to trade in input supply, and consumer products and services, agricultural and agroindustrial products. It begins and indirectly generates jobs for those leaving the by demonstrating the importance of growth in the farm. The agricultural sector thus contributes to agricultural sector, and discussing the connection growth both directly, through greater production between growth on the one hand, and on the other, and exports, and indirectly, by raising demand in the increasing trade and integration of the sector farm and rural communities for industrial goods into the world economy. It then shows how growth and services.1 in agricultural trade has lagged behind that in Growth in agriculture has a disproportionately manufactured products and examines why this is positive effect on poverty reduction, since poverty the case. It finds that trade reform in agriculture is predominantly a rural phenomenon.2 Approxi- has made much less progress than that in manufac- mately 75 percent of the poor reside in rural areas, tured goods, and that many barriers remain to both and the rural poor worldwide will outnumber their North­South and South­South trade in these urban counterparts for at least another generation products. Next, it quantifies the potential gains of (Alderman 2001; Ravallion 2000). For example, the a large-scale liberalization of trade that would incidence of rural poverty reported in seven of the presumably be an outcome of a successful round first Poverty Reduction Strategy Papers (PRSPs) of negotiations. Finally, it discusses some of the was between 10 and 40 percentage points greater What's At Stake? 3 BOX 1.1 The Importance of Agriculture to Developing Countries Historically, during the course of development, of the challenges currently being posed by the share of agriculture in both output and labor industrialized country agricultural trade policies, falls. This has led some development experts to the severity of rural poverty, and the central role view agriculture as only ancillary to develop- of agriculture in developing countries' economic ment. However, the fall in agricultural output growth, there is an urgent need to refocus on and labor can be a result of biased domestic this sector to take advantage of the comparative policies and international trade policies. The advantages most developing countries have in agricultural sector must be an engine of eco- agricultural production. nomic growth, especially in the very poorest Given the chronic incidence of rural poverty developing countries where agriculture still rep- and the concomitant harmful environmental resents a significant percentage of GDP and practices, a development framework for agricul- where the rural population accounts for a large ture must now focus on, among other things, percentage of the poor. equity and sustainability. Such a framework A well-integrated agricultural sector should should therefore include, at least: enhance food security, reduce real food prices (especially beneficial to the poor, who spend a · Economic policies in developing countries disproportionate share of their income on food), that are not biased against primary produc- increase employment and income, create impor- tion and export. tant economic linkages in production chains, and · Trade policies in rich economies that are not have a positive impact on the environment. biased against developing countries. During the 1990s, international support for · Public and private investments in infrastruc- agricultural development was weakened, as evi- ture, technical development, and credit, which denced by the declines in funds allocated to agri- are necessary for modernizing production cultural projects and research by governments and improving competitiveness. and international development agencies. In light The Role of Agriculture in Developing Countries by Region Agricultural Raw Agricultural Value Materials Exports Added (percent (percent of total Rural Population of total GDP) merchandise export) (percent of total) Region 1980 2000 1980 2000 1980 2000 East Asia and Pacific 24.4 12.6 12 2 78 65 Europe and Central Asia -- 10.5 -- 9 41 35 Latin America and the Caribbean 10.3 7.1 4 3 35 25 Middle East and North Africa 10.3 14.3 1 0 52 41 South Asia 38.0 25.1 10 1 78 72 Sub-Saharan Africa 17.6 17.0 6 4 77 66 (Continued) 4 Agriculture and the WTO Box 1.1 (Continued) The Role of Agriculture in Developing Countries by Income Group Agricultural Raw Agricultural Value Materials Exports Added (percent (percent of total of total GDP) merchandise exports) Income Group 1980 2000 1980 2000 Low income 33.6 23.9 12 4 Middle income 15.8 9.3 7 2 Lower middle income 24.7 13.3 -- 2 Upper middle income 11.2 6.6 7 2 Low and middle income 18.5 11.6 8 2 High income nd nd 4 2 -- Not available. Source: World Bank 2002b. TABLE 1.1 Rural­Urban Poverty Gap responded more to rural economic growth than to urban economic growth (Datt and Ravallion 1996). Percentage Points Difference Agricultural growth indirectly benefits urban and between the Incidence of Rural and Urban Poverty rural households by promoting higher wages, low- Country (Rural Percent­Urban Percent) ering food prices, increasing the demand for con- Bolivia 42 sumer and intermediate goods and services, Burkina- Faso 35 encouraging the development of agribusiness, rais- Honduras 23 ing the returns to labor and capital, and improving Mauritania 41 the overall allocative efficiency of factor markets. Mozambique 9 A 1997 study of 35 representative countries showed Nicaragua 38 that a 1 percent increase in agricultural GDP per Tanzania 16 capita created a 1.61 percent gain in the per capita incomes of the poorest 20 percent of the popula- Source: PRSP documents. tion (Timmer 1997). than in urban areas (table 1.1). Moreover, in most Improved Food Security developing countries the severity of poverty is greater in rural areas than in urban areas. People On a global scale, future food and feed needs are living in rural areas score lower on average than large and expanding, driven by population and urban residents by every quality of life indicator, income growth and by rapid demand for grain for and suffer more. A rural child has a much greater livestock feed. Projections by the International Food chance of dying before age five than an urban child; Policy Research Institute (IFPRI) indicate that public services in the rural areas, as measured by unless there is a renewed commitment to agricul- per capita public expenditure, are approximately ture through increased public and private invest- one-half that of urban areas; and more children ment and conducive policies, the long-term trend and adults in rural areas die or get sick from lack of toward lower food prices will not be maintained to access to clean water than in urban areas. 2020, and the international millennium targets for Perhaps not as obvious is the fact that urban reducing poverty and malnutrition will not be met poverty reduction can be accelerated by the growth (IFPRI 2001). At the national level, agricultural of the rural sector, especially in agriculture. For growth makes important contributions to access example, in India general poverty measures have to food--another dimension of food security--by What's At Stake? 5 increasing the incomes of the poor who depend on age rate of 5.7 percent in the four years following agricultural production for their livelihoods. the reform, in comparison with 2.8 percent in the year previous to reforms. When the reform pro- gram was partially or fully reversed, the analogous Conservation figures were 2.3 percent and 2.8 percent. Schiff and Agriculture is the greatest user of natural resources Valdés (1992a, 1992b) in a detailed study of the and has an important role in natural resource con- effects on agriculture of trade, pricing, and macro- servation. The deteriorating land and water base in economic policies of 18 developing countries many regions is a concern for many producers, and found that the protectionist, anti-export policies wider public awareness of environmental issues pursued by many of these countries had the effect is bringing urgency to conservation issues, many of reducing both agricultural and general eco- global in nature. In some countries the overuse of nomic growth. Similarly, when looking at a sample chemicals and pesticides--often encouraged by of 11 countries that had initiated trade liberaliza- input and output subsidies--contributes to envi- tion, Valdés (1998) found that sustained reforms ronmental problems. Removing these subsidies were associated with much higher growth rates of while raising incomes of farmers through improved both overall and agricultural GDP (5.2 percent and productivity are important components of a strat- 5.7 percent per annum) in the years following egy in reducing resource degradation. reforms than were reform episodes that collapsed (-1.5 percent and 1.1 percent). A World Bank (1994) study of 29 Sub-Saharan African countries Trade Liberalization Fuels found that when overall macroeconomic policies Prosperity (including trade policy and related exchange rate Openness and integration with the world econ- policy) are considered, the group of countries that omy promotes growth and reduces poverty. There showed a "large improvement" had a weighted is a preponderance of evidence in many countries agricultural growth rate of 3.5 percent per year that economy-wide trade liberalization and open- between 1986 and 1993 in contrast to 2.5 percent ness to trade increases the growth rate of income for those with a "small improvement" and 0.3 per- and output.3 In addition, numerous individual cent for those with a "deterioration" in macroeco- country studies performed from the 1970s nomic policies. through the 1990s have concluded that "trade Reducing barriers such as high tariffs to does seem to create, even sustain higher growth" imports of agricultural and food products has (Srinivasan and Bhagwati 1999). A country's trade another important poverty-reduction dimension. policy is the key link in the transmission of price Reducing barriers helps keep food prices low, and signals from the world market to the national low food prices benefit consumers. Since the poor economy. The undistorted price signals from spend a disproportionate part of their income on world markets, in combination with the exchange food, they will benefit disproportionately from low rate, allow resource allocation consistent with trade barriers. comparative advantage, thereby increasing pro- To support agricultural growth and poverty ductivity. An open trade and investment regime reduction goals, developing economies need to encourages integration into the global trading be better integrated into the world economy by environment and also encourages the import of reducing trade barriers--their own and, through diverse and modern technologies important for multilateral agricultural trade negotiations, those productivity improvements.4 of developed countries. However, to compete glob- More specifically, trade liberalization has been ally, developing countries will have to increase shown in a number of studies to be associated with their agricultural competitiveness and productivity enhanced growth in the agricultural sector. In through appropriate changes in cropping patterns, comparing episodes of trade policy reform, improvements in production techniques, the Michaely, Choksi, and Papageorgiou (1991) found upgrading of marketing channels, and other com- that if the liberalization efforts were sustained, plementary measures to realize any true benefits of then the agricultural sector would grow at an aver- trade liberalization. This means, of course, that in 6 Agriculture and the WTO FIGURE 1.1 Trade Has Propelled Growth, 1965­2000 Trade has outpaced output to propel growth... Ratio of trade growth to GDP growth 4 3 2 1 0 1965 1970 1975 1980 1985 1990 1995 2000 Source: World Bank 2001. ...and is now more important in all regions Trade (exports plus imports) as share of GDP (real) 70 1998-2000 60 50 1988-90 40 30 20 10 Developing EAP ECA LAC SSA MENA SA Source: World Bank 2001. Note: EAP = East Asia and Pacific. ECA = Europe and Central Asia. LAC = Latin America and the Caribbean. SSA = Sub-Saharan Africa. MENA = Middle East and North Africa. SA = South Asia. Agricultural Trade Lags Industrial trade negotiations there may have to be special con- Goods Trade sideration for those countries (see chapter 14). It also emphasizes the importance of providing ade- For more than a generation trade has been a driver of quate safety nets to cushion adverse effects of global growth across all global regions (figure 1.1). adjustment on vulnerable population groups, such In every year from the 1970s through the 1990s, as the rural laborer (see chapter 10). But the long- export growth worldwide has outpaced the growth term goal of lowering trade barriers should be fun- of total output by an average ratio of 1.5 to 1. More- damentally the same for developing and for more over, this trend has increased over time and, after a developed countries. pause in the mid-1980s, jumped to nearly 2.5 to 1 in What's At Stake? 7 FIGURE 1.2 Tariffs and Trade Growth Number Percent, 1950 = 100 70 2000 1800 1800 Manufacturing Tariffs Agricultural Tariffs Trade Growth 60 1600 60 50 1400 1200 40 38 1000 1000 30 800 20 600 17 600 400 9 10 250 6 4 200 100 0 0 47 GATT 62 Pre-Kennedy 72 Post-Kennedy 87 Post-Tokyo 94 Post-Uruguay the 1990s. Consequently, in the space of just the of total world trade has declined steadily through- 1990s, the average ratio of trade-to-GDP in devel- out the last century. This trend shows no sign of oping countries has risen from 29 percent to abating: The share of food and agriculture in total 43 percent. merchandise trade fell from 17 percent to 10 per- cent between 1980 and 1997. World Merchandise Trade High Tariffs and Other Barriers to Trade The volume of total merchandise trade has increased 18-fold since the founding of the General Overall, there has been some progress in global Agreement on Tariffs and Trade (GATT) in 1947, agriculture and trade policy reform, but this with much of the credit due to trade liberalization. remains fragile and fails to provide the liberaliza- Figure 1.2 presents an overview of world merchan- tion and technical support that developing coun- dise trade against the backdrop of continually tries had expected from the Uruguay Round falling tariffs for manufactured products that have (UR) accords: the spirit is missing. A review of the declined after each of the multilateral trade rounds, experience to date on the implementation of the and the continuing high tariffs for agricultural UR Agreement on Agriculture (URAA) indicates products. In the 1990s, the value of world exports that domestic policy reform and trade liberalization of agriculture products increased from US$274 bil- have been difficult to achieve in both the OECD and lion in 1990­2001 to US$394 billion in 2000­01, developing countries for four main reasons. more than 1.4 times its previous level. However, First, domestic policy distortions and support the increase in the manufacturing sector was remain high in many OECD countries. Rich country much larger: worldwide exports of manufactures border measures and subsidies that artificially increased to more than 1.8 times its 1990 level. increase production and distort trade have been By the end of the 1990s, global trade in food and reduced during the implementation period of the agricultural products was about 10 percent of the URAA, but still remain extremely high. Total sup- value of total merchandise trade. Although agricul- port amounted to US$311 billion in 2001 in the tural trade is increasing in absolute terms, its share OECD countries. While this amount was 1.3 percent 8 Agriculture and the WTO of their own GDP, it was roughly equal to the GDP passed legislation (a farm bill) that would sharply of all countries in the Sub-Saharan Africa region. increase subsidies when compared with those in the Of this support, 69 percent is administered via 1996 farm bill. However, the annual volume of sub- price support and output payments, mechanisms sidies forecast in the 2002 farm bill, while substan- that are the most distortional for production and tially larger than the subsidies planned in the 1996 trade (down though from 82 percent in 1986­88), bill, is less than the average volume of all subsidies and it accounted for 32 percent of total farm including both payments forecast in the 1996 bill as receipts (compared to 38 percent in 1986­88). On well as the government-provided ad hoc emergency average, prices received by OECD farmers were 31 relief payments. There is, of course, still some ques- percent above world prices (compared to 58 per- tion about whether the practice of giving such ad cent in 1986­88). While total support in OECD in hoc payments will continue. 2001 was slightly less than in 2000 by US$10 The 2002 farm bill is generally perceived as a billion, recent actions in the United States (U.S.) reversal of the U.S.'s commitment, embodied in the and the European Union (EU) make it unlikely that 1996 farm bill, to increase the reliance of farmers overall support will be further reduced, though on market-determined prices rather than govern- there is still scope to make it less trade distorting. ment support. A recent Japanese government pro- Producer support estimates (PSEs) varied gram sets an increased self-sufficiency goal in agri- widely across countries within the OECD. PSEs culture products of 45 percent, implying the ranged from 1 percent in New Zealand and 4 per- potential for even higher production subsidies and cent in Australia to 60 percent or more for some trade barriers. The proposal in the EU's latest countries such as Japan, Korea, Norway, and review of the Common Agricultural Policy (CAP) Switzerland. For the U.S., the average PSE is about would reduce intervention (support) prices and 20 percent and for the EU it is about 35 percent. increase reliance on decoupled payments, which are Producer support levels also vary greatly among less trade distorting than other forms of support. commodities: rice, 81 percent; sugar, 45 percent; However, adoption of this proposal is far from milk, 45 percent; sheep meat, 55 percent; wheat, assured. A recent decision fixed the total budget of 36 percent; beef and veal, 36 percent; oilseeds, the CAP through 2013 at a level close to its current 28 percent; pig meat, 16 percent; poultry, 16 per- level in real terms.5 cent; wool, 6 percent; and eggs, 10 percent. The Second, high tariff barriers and tariff escalation highest levels of support are directed at temperate on agricultural and agroindustrial products remain products, and it is sometimes argued that these are in many countries. Reflecting current levels of sup- not the products of greatest interest to developing port, agricultural tariffs remain extremely high, in countries in the trade negotiations. However, these spite of the reductions introduced with the URAA. products are significant in some developing coun- Recent estimates indicate average agricultural tariffs tries at present, and the heavy subsidies prevent are about six times as high as industrial tariffs (Gib- growers of tropical products from diversifying into son and others 2001; OECD 2001b). While new tar- these temperate crops in the current depressed iff rate quotas as a result of the URAA have replaced market conditions (see box 1.2). nontariff barriers, out-of-quota tariff rates remain Thus, the current levels of agricultural support high and sometimes prohibitive. Even in-quota in OECD countries are still a major factor influenc- tariff rates are often high, leading to the under- ing world production, distorting trade and depress- utilization of quotas. Tariffs of more than 50 per- ing world prices of agricultural products. More cent exist for 60 tariff lines in Canada, 71 in the EU, importantly, the structure of protection is such that 14 in Japan, and 8 in the U.S. (McCulloch, Winters, support (protection) mainly benefits large farms. and Cirera 2001). These harm nearly US$5 billion Meanwhile, the support is a significant burden to of developing countries' exports (despite the high both developing country farmers and low-income tariff rates) and are almost exclusively focused on consumers whose food consumption still consti- agriculture. Tariffs range from 129 percent for tutes a large share of total household expenditure. sugar in the U.S. to 162 percent for grains in the EU The future plans of the major OECD countries (Elbehri and others 1999). Protection also escalates are unclear to developing nations. The U.S. in 2002 with the level of processing, particularly in markets What's At Stake? 9 BOX 1.2 The Human Face of Policy Incoherence Policy incoherence refers to the interference of posed to be soothed by this antipoverty offen- government policy in one sphere of policy with sive are becoming more alienated and angry the achievement of strategic objectives in with the U.S. Mali, a predominantly Muslim another. Agricultural subsidies in OECD coun- country, has been largely peaceful since Septem- tries have often been cited as inconsistent with ber 11, 2001 (the terrorist attacks in the U.S.), the goals of those countries to support eco- but frustrations are increasing nonetheless. "This nomic development in poorer countries, and is where America is heading: It wants to domi- this point is made strongly in two investigative nate the world, economically and militarily," news reports in the U.S. newspaper, The Wall says a leader of the farmer's union in the Malian Street Journal (see Thurow and Kilman 2002 and regional center of Bougouni. According to the Fritsch 2002). article, citizens of the cotton countries of West The story by Thurow and Kilman contrasts and Central Africa, where Islam is the major reli- the situation of two families farming cotton. The gion, are crowding into the cities of Europe, and U.S. family lives comfortably, since the average those who remain are seeing more clerics from full-time American cotton-farming household Pakistan and the Middle East visit their mosques has assets worth about US$800,000, according and Quranic schools. In Mali, western diplomats to the U.S. Department of Agriculture. Owing to hear reports of some Malians crossing the Alger- the approximately US$3.4 billion in subsidy ian border for religious training abroad. checks, U.S. farmers last year harvested a record Finally, the article notes that these subsidies crop of 9.74 billion pounds of cotton, glutting undermine the program of the World Bank and the market and pushing prices far below the IMF to support a basic reform of Mali's state cot- break-even price of most growers around the ton marketing monopoly, which is being restruc- world. Under the 2002 U.S. farm bill, cotton tured to increase reliance on the private sector farmers can expect to receive about US$0.70 per to provide more money for the farmers. But it is pound (in comparison with a world price of not clear that these reforms will show significant around US$0.40). Many U.S. cotton growers are results without changes in cotton subsidy poli- expected to receive half of their income from cies elsewhere in the world. the government in 2002. The article cites the The story by Fritsch deals with the worldwide recent study which estimates that the increased depression in the coffee market, examining in U.S. cotton production due to subsidies--and detail the situation of farmers and laborers the consequent lower world prices--are respon- in Nicaragua, and finding that many of those sible for a reduction in revenue to West and Cen- farmers and laborers are destitute and on tral African countries of about US$250 million the verge of malnutrition. While coffee is not a annually. crop that is grown or subsidized by OECD In contrast, in Mali the price offered to Mali's countries on a large scale, the article makes cotton farmers this year is 10 percent lower than the point that there are important interconnec- in 2001, which was at the lowest level in three tions among markets. For many years, foreign decades. After the last harvest, once the farming aid donors and governments have advocated costs were paid, the Malian family was left with and financially supported the objective of diversi- less than US$2,000 for the year to support two fication out of traditional crops as a way of dozen family members and relatives. This year's increasing value and reducing poor farmers' lower prices, along with higher fertilizer and pes- vulnerability to the dramatic price fluctuations ticide costs, probably mean that the family will in these markets. This goal has become even be unable to replenish their cattle stock and may more pressing in the current severe depression in also have to stop supporting their youngest the coffee market. Yet, the article notes, farmers brother's high-school education. Given that the who consider switching out of coffee "are dis- U.S. spends US$40 million a year on education, couraged . . . by the experience of farmers who health, and other development programs in have grown peanuts and sesame. Those growers Mali, the article notes the obvious inconsistency now find themselves on the verge of bankruptcy and further notes that the subsidies are also after trying to compete against U.S. farmers inconsistent with the U.S. foreign policy goal of receiving generous subsidies from Washington." fighting terrorism. The people who are sup- 10 Agriculture and the WTO for processed tropical products, preventing the agricultural inputs high. Since those findings were development of value-added activities in develop- published in the 1980s, developing countries as a ing countries. Specific tariffs are used widely in group have improved their macroeconomic and agriculture and are generally regarded as less trans- trade policies, raising the question of whether these parent and more distortional than ad valorem tar- findings are still valid. While the magnitude of the iffs. Since the weight of specific tariffs (as a per- bias has clearly been reduced, it still remains signif- centage of the total price of the product) is lower icant in many countries owing to the continuing with higher-price products, there is therefore an protection of manufactured goods, the operation of incentive to supply higher-value products within inefficient state-owned marketing enterprises for each dutiable category. Given these levels of indus- key agricultural exports in some countries, and in trialized countries' trade barriers, improved mar- some cases direct taxation of exports. In addition to ket access offers the potential for huge increases in creating an overall bias against agriculture, the income in developing countries that can supply trade policies of developing countries create barri- these products.6 ers to increased South­South trade in agricultural Third, high levels of export subsidies continue products. Tariffs for both processed foods and pri- to distort world markets for key commodities. mary agricultural products are higher in develop- These high levels in OECD countries remain a ing countries than in industrialized countries, major factor in world food markets and have a wide although tariffs on some temperate crops and effect on world prices and market conditions. products and nontariff barriers (NTBs) in general Between 1995 and 1998, global export subsidies are higher in the industrialized countries. Both amounted to more than US$27 billion cumula- average tariffs and incidence of NTBs in developing tively, of which more than 90 percent is from the countries are also slightly higher for agricultural EU (Elbehri and Leetma 2002). Export subsidies imports than for manufactured goods (see tables reduce prices and make it difficult for potential 1.2, 1.3, 1.4, and 1.5). While both tariffs and NTBs agricultural exporting countries to compete. For were substantially reduced in the late 1980s and importers, export subsidies bring short-term bene- 1990s, the high levels continue to create very signif- fits in terms of lower import prices. But for both icant barriers to greater trade among the develop- groups of countries they can be detrimental to agri- ing countries, though there is some reason to cultural development in the longer run. The URAA believe that barriers in OECD countries are even placed limits on export subsidies for individual more effective in stifling trade. Further real reduc- commodities, but allowed for some flexibility in tions in trade barriers have the potential to improve their use as low usage levels early in the URAA productive efficiency based on true comparative implementation period (1995 was a period of high advantage. world prices) enabled some countries to bring for- ward unused subsidies and apply these when prices World Export Growth were low and the subsidy ceilings had been reached.7 Table 1.6 shows export growth over each of the last Fourth, policies of many developing countries two decades through 2001 for agricultural and continue to create a bias against their own agricul- manufacturing exports from developing and tural sectors and those of other developing coun- industrialized countries. Worldwide, both agricul- tries. Schiff and Valdés (1992a, 1992b) found that tural and manufacturing export growth in the in the past, developing countries have typically 1990s continued at about the same pace as in the taxed their agricultural sectors to some extent 1980s. Growth of developing countries' agricultural directly (for example, by taxes on exports or by exports rose, as did manufacturing export growth, controlled food prices) but also indirectly (for but this similarity masks an important difference. example, through trade barriers and macroeco- Within developing countries' agricultural exports, nomic policies that overvalued the exchange rate). the growth rate of those to other developing coun- In so doing, these policies turned the internal terms tries more than doubled, while those to developed of trade (manufactures prices vis-ŕ-vis agricultural countries stagnated. As a result of these trends, the prices) against agriculture and kept the prices of share of developing countries' agricultural exports What's At Stake? 11 b b 1.8 3.0 5.5 8.4 0.0 2.8 0.5 2.6 8.7 32.0 13.9 24.0 24.6 18.9 11.0 13.9 26.3 21.8 14.9 13.7 20.3 12.9 20.1 14.0 finance 1994­98 1994­98 equency equency Fr a Fr trade a NTB -- -- -- -- -- -- -- -- -- -- -- -- NTB -- 7.2 7.7 Materials 35.2 22.8 48.8 20.3 36.5 43.4 39.6 39.2 45.6 ictive Foods 1984­87 1984­87 restr periods. y Raw and time Primar 8.1 1.8 9.6 8.2 8.0 4.3 23.9 22.5 14.4 14.8 30.4 36.2 24.9 24.7 21.1 16.7 18.4 Foods 1994­98 ariffT 14.3 14.0 30.9 15.0 16.1 10.5 10.0 1994­98 prices, between Agricultural categories. Applied -- -- -- -- -- -- -- -- -- -- -- -- Applied import -- 5.9 9.7 barriers 18.8 11.9 23.9 10.6 39.4 21.2 25.5 17.0 16.9 1984­87 1984­87 control product 1994­98 b b nontariff and and 2.1 3.0 5.3 9.8 2.3 4.5 8.2 directly and 28.9 12.1 19.4 22.1 14.1 12.0 12.6 21.8 17.2 24.8 13.5 18.7 10.4 12.2 10.6 17.4 to 1994­98 1994­98 equency equency countries Fr Fr a a countries for 1984-87 NTB -- NTB -- -- -- -- -- -- -- -- -- -- -- -- measures 47.8 33.6 42.9 40.2 60.0 54.4 55.4 53.6 71.3 13.3 23.0 1984­87 1984­87 measures Foods developing Countries, restrictions, of NTB Food 6.2 6.2 25.0 25.2 13.9 15.2 26.8 41.8 27.0 27.2 20.3 17.9 17.1 26.4 27.1 14.4 15.6 25.5 47.1 28.8 29.6 20.0 18.8 19.4 ariffT 1994­98 ariffT and 1994­98 Processed import coverage tariff of Developing Applied -- Applied 8.3 data in 30.7 22.2 29.5 27.7 65.5 31.2 40.0 33.9 20.1 12.7 -- -- -- -- -- -- -- -- -- -- -- -- in 1984­87 1984­87 quantitative averages differences simple Protection (MENA) (MENA) including (LAC) (LAC) by are (ECA) (ECA) Africa Africa (EAP) (SSA) (LMIC) (UMIC) (LMIC) (UMIC) Asia (LIC) (HIC). (EAP) (SSA) Asia (LIC) (HIC). barriers, biased be statistics Caribbean North Caribbean North Region Africa Ctys Dev Region Africa Ctys Dev Bank. Agricultural Partner countries in Pacific & Income countries & (SA) Income Partner Income Income countries in Pacific (SA) may by & Central countries Central nontariff & East by & orld Asia Mid. Mid. & East Mid. Mid. available. W 1.2 Asia America, Income Income Asia America, Asia Income Income Protection ce: Group Group Not "Core" Change Exports Countries or Industrial Developing East Europe Latin Middle South Sub-Saharan Low Lower Upper High Exports Countries or Industrial Developing East Europe Latin Middle South Sub-Saharan Low Lower Upper High -- a. measures. b. Note: Sour ABLET 12 Agriculture and the WTO 7.9 2.7 3.5 2.5 8.0 0.7 1.2 4.4 8.5 5.8 8.7 3.2 7.3 0.1 0.2 0.3 2.0 0.2 0.9 2.0 OECD 47.4 45.5 13.0 14.7 HIC 4.8 2.7 7.3 0.6 0.4 1.1 6.3 0.7 2.2 2.1 13.4 22.2 54.1 29.4 20.0 13.6 23.2 49.3 16.4 23.6 14.3 16.0 12.2 16.0 7.4 8.2 7.0 6.1 5.7 8.3 6.1 6.3 8.6 UMIC 12.5 22.7 21.8 10.9 15.3 17.0 23.2 13.5 16.9 16.8 18.7 15.6 37.5 19.9 46.3 9.8 9.8 7.8 6.9 8.2 6.8 7.2 9.4 LMIC 13.5 31.2 23.1 11.7 20.4 21.8 20.4 15.4 20.2 20.5 23.1 17.3 27.2 20.2 24.0 10.5 LIC 16.5 30.3 25.8 14.0 25.5 30.7 25.9 14.3 18.8 22.8 25.8 29.3 32.8 29.6 46.4 30.6 42.5 17.3 10.9 13.2 15.0 11.9 12.8 14.9 SSA 9.9 16.8 30.9 25.8 13.7 26.4 30.5 25.9 14.4 18.7 20.4 26.5 30.4 33.3 30.5 34.1 30.3 36.9 16.4 11.7 14.4 11.4 11.4 14.5 SA 27.3 45.2 33.4 14.7 34.3 44.8 42.0 22.7 32.2 41.2 33.7 41.2 47.1 42.9 46.9 84.5 30.6 21.8 29.5 27.7 25.0 29.3 31.6 115.2 cent) 8.6 9.4 (Per MENA 27.2 54.9 32.4 12.4 17.9 30.6 23.0 11.6 13.7 20.8 24.5 24.2 31.0 21.7 45.1 24.7 33.9 10.1 10.7 12.2 10.6 12.8 LAC 8.3 9.3 7.4 8.0 7.3 6.5 7.1 6.2 7.0 8.2 1994­98 18.4 18.5 11.1 14.7 17.2 16.7 14.5 14.8 15.1 17.0 15.3 18.0 16.3 18.7 ECA 8.9 5.6 9.5 9.1 5.2 1.4 5.1 8.4 3.4 3.0 5.1 10.6 21.3 22.5 15.6 11.1 19.7 30.4 16.9 15.6 15.9 24.2 17.6 32.3 Countries, 8.1 5.8 3.9 5.4 4.1 3.4 4.2 5.3 EAP 22.1 21.1 37.0 31.9 27.3 21.9 20.5 13.1 36.6 22.8 28.5 21.1 46.7 20.2 42.9 10.1 Industrial and foods spices mats. stuffs & metals manufs. nuts honey vegetables fruits & raw. grains feed ores fuels oils preparations tea, skins rubber fibers products products & & processed & agr animals y seeds & Developing in Live Meat Dair Cereal egetablesV Fruits Sugar Animal Oil Fats Cereal Prepared Prepared Coffee, Beverages Other obaccoT Hides Natural Natural Other Mineral Mineral Non-ferrous Rates foods y ariffT materials Primar Processed foods Agricultural raw Miscellaneous y Applied Categor products y 1.3 Product Primar ABLET What's At Stake? 13 = 0.1 3.7 0.7 2.9 2.1 2.9 3.1 5.4 3.9 3.4 3.3 2.7 3.2 2.3 OECD 10.0 11.9 Europe LMIC values = HIC 2.8 6.6 5.4 3.4 3.5 4.4 5.1 3.0 4.9 5.1 4.5 8.0 4.5 7.6 8.3 3.4 ECA shaded countries, Pacific, (lightly) 5.7 5.0 9.5 UMIC 16.0 11.2 14.7 13.3 14.0 15.5 14.4 13.5 17.1 13.2 20.6 21.6 11.8 income and low Darkly Asia = 5.2 5.8 8.7 6.7 LMIC 16.7 11.7 12.8 16.4 13.5 17.6 14.3 20.8 14.7 22.3 20.1 10.8 East LIC = APE rica, ountries.c Af LIC 9.0 4.6 26.8 14.7 12.2 20.7 17.3 26.0 21.3 24.2 22.0 30.2 24.0 32.4 29.8 19.8 group. industrial y SSA 7.3 3.5 26.3 13.2 10.7 20.9 14.7 26.8 20.8 24.6 20.6 29.2 22.9 34.0 31.3 19.7 Sub-Saharan major countr = = SSA each SA 22.3 48.2 15.5 37.1 30.1 41.7 41.3 39.5 45.7 43.3 47.1 50.7 45.9 52.1 47.3 35.3 in OECD Asia, and South 8.2 MENA 22.7 11.8 11.1 10.7 15.7 14.3 21.6 19.4 22.7 19.8 28.9 19.1 27.8 27.3 13.0 protection = of SA countries, LAC 5.5 3.4 8.8 7.1 rates 14.4 11.5 11.0 14.3 11.6 12.8 11.7 17.1 11.8 17.8 17.6 10.8 Africa, of income North high ECA 3.0 9.6 4.3 5.9 4.4 6.6 7.9 9.0 6.2 8.5 7.1 9.6 7.5 = 11.9 13.7 11.5 averages and HIC East simple 3.5 3.2 6.2 5.1 9.4 7.6 EAP 14.7 14.4 11.8 13.0 16.5 12.4 17.9 12.1 20.2 19.1 group. are y Middle countries, = aluesV countr prods. y goods MENA income each fumes mach. min. equipment in per equipment travel clothing categories. middle & fertilizers y steel machiner & products & products products & Caribbean, upper protection Pharmaceuticals oiletrT Manuf. Iron Non-electric Electric ransportT ood Leather Rubber W Paper extilesT Non-metallic Furniture Footwear Professional product and = of of UMIC rates calculations.s y America steel & description Latin author' = countries, for Chemical Iron Machiner Other manufactures y 1.4 LAC 1999; (intermediate) income AD Asia, (Continued) table Categor highest UNCT See middle 1.3 Central ces: Product Manufactures Notes: and lower indicate Sour ABLET 14 Agriculture and the WTO TABLE 1.4 World Trade Simulation Model Product Categories Standard International Standard International Trade Classification Trade Classification Product Category (SITC), Rev. 2 Product Category (SITC), Rev. 2 Primary products (0 to 4) 68 Manufactured products (5 to 8) 68 Primary foods AGGR Chemicals 5 Live animals 0 Pharmaceuticals 54 Meat products 1 Toiletry & perfumes 55 Dairy products 02 025 Manufactured fertilizers 56 Cereal grains 041 to 045 Vegetables (0541 to 0545) 05481 Iron & Steel 67 Fruits & nuts 57 Sugar & honey 6 Machinery & equipment 7 Animal feed stuffs 08 08142 Non-electric machinery 71 to 75 Oil seeds 22 Electric machinery 76 77 Transport equipment 78 79 Processed foods AGGR Animal & vegetable (091 4) 4111 Other manufactured (6 8) ­ (67 68) oils & fats products Cereal meals, flours, 046 to 048 Leather & travel goods 61 83 & preparations Prepared vegetables (0546 to 056) 05481 Rubber products 62 Prepared fruits 58 Wood products 63 Coffee, tea, & spices 7 Paper products 64 Beverages 11 Textiles & clothing 65 84 Other processed 025 098 Non-metallic mineral 66 agricultural products products Furniture 82 Agricultural Raw AGGR Footwear 85 Materials Tobacco & tobacco 12 Professional equipment 87 88 manufactures Hides & skins 21 Natural rubber 232 All product categories 0 to 9 Natural fibers (261 to 265) 268 Other agricultural 29 raw materials Crude Fertilizer & 27 28 Mineral Ores Mineral Fuels 3 Non-ferrous Metals 68 Note: AGGR = sum of SITC numbers in the given product category. Negative sign (-) = "less." Sources: OECD 1998; UNCTAD 1994. What's At Stake? 15 0.0 0.6 7.1 0.3 OECD 28.4 32.1 62.1 39.3 30.8 27.9 34.3 21.6 23.0 27.6 27.6 20.8 22.8 21.5 33.3 27.4 16.9 25.4 12.1 14.4 indicate America income HIC 3.6 1.6 4.8 8.0 7.8 2.8 0.2 1.2 0.0 0.0 1.0 0.4 0.0 Latin values 36.6 33.4 28.0 23.4 22.4 18.2 18.8 10.2 10.0 17.8 16.2 = middle LAC shaded lower 9.7 0.0 4.6 6.6 2.4 UMIC 30.2 24.0 23.7 28.9 32.6 27.3 16.4 25.3 27.5 20.7 17.1 17.6 19.9 27.4 16.6 23.6 21.9 14.5 24.1 Asia, = ghtly) (li LMIC Central 9.0 9.2 8.0 9.7 8.7 7.8 3.9 0.7 6.5 5.1 5.2 3.6 LMIC 18.3 12.8 19.9 20.9 14.5 12.8 16.0 10.3 10.9 11.5 17.7 12.2 Darkly and countries, LIC 2.3 6.7 2.8 7.1 8.0 6.6 3.5 2.1 4.8 3.3 4.9 6.2 5.2 3.2 7.3 2.7 5.2 1.1 4.7 1.2 3.5 0.6 0.7 0.8 Europe = countries. income ECA low SSA 1.2 5.7 0.0 1.9 4.3 3.6 0.1 0.9 0.0 1.3 0.6 3.1 2.9 0.1 2.9 0.9 1.4 0.0 0.0 0.0 0.9 0.1 0.0 0.3 = industrial cent) Pacific, LIC 0 (Per and major SA 0.0 0.0 9.2 4.2 6.4 7.4 0.0 2.0 2.2 0.6 1.4 0.4 16.2 16. 12.6 11.0 11.6 16.2 14.2 14.2 26.4 12.6 13.2 14.0 Africa, = Asia OECD East 1994­98 8.5 9.8 7.0 1.6 8.0 6.8 0.0 3.6 0.6 4.8 0.6 = MENA 25.3 16.4 30.3 17.1 21.6 21.6 14.1 15.0 11.5 12.1 31.6 29.1 19.4 and EAP Sub-Saharan = LAC 1.0 2.0 8.0 8.3 5.4 SSA 27.4 21.5 23.0 33.8 29.7 25.7 15.5 20.8 30.8 12.0 21.9 20.0 20.7 21.6 18.3 17.5 14.8 18.5 15.7 group. countries, Countries, y Asia, countr income ECA 3.6 1.9 3.2 7.7 0.8 0.0 6.5 0.2 0.4 1.5 1.8 0.0 4.3 2.2 9.2 2.2 0.0 0.0 0.0 0.0 0.0 0.6 0.4 0.9 South = Industrial each high group. y in SA = and EAP 9.9 9.9 4.7 9.5 7.8 7.3 5.0 4.5 2.2 HIC 35.3 26.5 28.7 35.8 28.0 21.3 10.3 11.7 20.5 21.5 14.9 11.8 18.6 15.0 16.0 countr Africa, protection each of North countries, in foods spices Developing mats. and stuffs & metals rates in nuts honey egetablesV manufs. fruits & of skins East income grains feed ores fuels calculations.s oils Preparations tea, rubber fibers raw. protection products products & & & NTBs animals y seeds & processed agr of of Live Meat Dair Cereal egetablesV middle Fruits Sugar Animal Oil Fats Cereal Prepared Prepared Coffee, Beverages Oth. obaccoT Middle Hides Natural Natural Oth. Mineral Mineral Non-ferrous averages author' = rates upper y y 1999; simple MENA = . AD are Frequency Primar foods Processed foods Agr raw mat. Misc. UMIC Categor UNCT y aluesV (intermediate) 1.5 Caribbean, ces: Product Primar products Note: and countries, highest Sour ableT 16 Agriculture and the WTO TABLE 1.6. Export Growth Rates (Constant 1995 US$) Developing Countries' Export Growth Rates, percent World Export Growth Rates, Developing to Developing to percent Total Developing Industrial 1980/81­ 1990/91­ 1980/81­ 1990/91­ 1980/81­ 1990/91­ 1980/81­ 1990/91­ 90/91 00/01 90/91 00/01 90/91 2000/01 90/91 2000/01 Agriculture 4.5 3.6 3.5 4.8 3.8 7.9 3.4 3.4 Manufacturing 5.9 4.8 7.6 8.9 7.3 10.0 7.8 8.3 Note: Manufacturing exports are deflated by U.S. producer price increases for finished goods less food and energy. Agriculture exports are deflated by U.S. producer price indexes for farm products. Source: COMTRADE, calculated for World Bank 2003. TABLE 1.7 Shares of Developing and Industrialized Countries in World Exports (Percent) Developing Countries Industrial Countries 1980­81 1990­91 2000­01 1980­81 1990­91 2000­01 Agriculture Total 35.4 32.2 36.3 64.6 67.8 63.7 To Developing 9.5 8.9 13.4 16.0 12.2 14.5 To Industrialized 25.8 23.3 22.9 48.6 55.6 49.3 Manufacturing Total 19.3 22.7 33.4 80.7 72.3 66.6 To Developing 6.6 7.5 12.3 21.7 15.2 19.0 To Industrialized 12.7 15.2 21.1 59.0 62.1 49.6 Source: COMTRADE, calculated for World Bank 2003. to other developing countries increased between in industrialized countries have been more effective 1980­81 and 2000­01 from 9.5 percent to 13.4 per- than those in developing countries in reducing cent, while their share of agricultural exports to growth in exports. industrialized countries fell from 25.8 percent to Potential Gains of a Successful 22.9 percent (see table 1.7). Meanwhile, developing Doha Round Are Substantial countries' share in manufacturing exports to indus- trialized countries continued to boom, increasing According to the findings of a recent World Bank from 12.7 percent in 1980­81 to 15.2 percent in study (World Bank 2002a) on the economic wel- 1990­91 and 21.1 percent in 2000­01. fare benefits from global agricultural trade reform, Simple statistics such as these cannot by them- the potential to increase the aggregate welfare of selves prove or disprove any hypothesis. Nonethe- the developing world would be some US$142 bil- less, combined with what is known about world- lion annually (table 1.8). This does not include any wide barriers to trade, the data are certainly economic welfare gains from liberalization of consistent with the hypothesis that barriers have trade in services, or investments, or reductions in been more effective in stifling trade in agricultural imperfect competition. Most of these gains could products than in manufacturing. With respect to be realized from trade policy reforms within the agricultural products specifically, barriers to trade developing countries themselves (about US$114 What's At Stake? 17 TABLE 1.8 Gains from Removing All Trade Barriers in Agriculture and Food Globally, Post-Uruguay Round, 2005 (in 1997 US$ Billions) Liberalizing Benefiting With Fixed With Endogenous Region Region Productivity Productivity High income High income 73 144 Low income 31 99 Total 104 243 Low income and middle income High income 23 53 Low income 114 294 Total 137 347 All countries High income 106 196 Low income 142 390 Total 248 586 Source: World Bank 2002a. billion), while the impact of liberalization only in ing region. For example, 70 percent of the gains high-income countries upon developing countries from high-income countries' removal of agri- would be about US$31 billion,8 amounting to cultural trade distortions accrues to those same around 60 percent of the official development high-income countries. Even so, agricultural trade assistance for developing countries. When more reform contributes more than 40 percent (US$31 dynamic effects of liberalization, including invest- billion) of the total welfare gains to developing ment, are considered, the gains are potentially countries from industrialized countries' liberaliza- much larger. tion of all their merchandise trade (US$75 billion, not reported in table 1.7). As for developing countries liberalizing their All Gains from Liberalization own agricultural trade policies, 83 percent of the Overall, the potential gains from liberalization of benefits from their own reforms would stay with agricultural markets (US$248 billion) are estimated the developing countries themselves. And the gains to be substantially higher than the potential benefits from developing countries' liberalization con- from manufacturing (US$111 billion, not reported tribute more than half of the gains from those in table 1.8), even though agriculture accounts for a countries' overall merchandise trade reform. These much smaller share of total world merchandise large gains reflect not only the significant distor- trade. For the most part, this is due to the high level tions in those countries but also the fact that the of agricultural protection in high-income countries. food and agricultural sector is such a large part of Considering that agricultural and processed food the economy of developing countries. products in high-income countries contribute only There is of course a large literature on the gains around 4 percent of global GDP and only from agricultural trade liberalization, of which 6 percent of world trade, a disproportionately large the study discussed above is only one contribu- share of the global welfare gains would come from tion.9 Some other recent contributions to this lit- abolishing protection and trade barriers in these erature have estimated relatively small static wel- countries. fare gains to developing countries.10 While these The geographical distribution of the gains from gains are not trivial when compared to official full trade liberalization is also clear. As should be development assistance of around US$50 billion expected, most of the gains accrue to the liberaliz- per year, the gains seem small when compared 18 Agriculture and the WTO with trade volumes. There are, however, good rea- sector, since these still dominate food distribu- sons to believe that the gains underestimate the true tion, agroprocessing, and input supply in a num- value of liberalization in the context of a multilat- ber of countries. These markets need to be priva- eral agreement.All of these simulation studies (with tized and de-monopolized by allowing the the exception of the one discussed above) only esti- emergence of new and restructured private firms mate static welfare gains and so miss the dynamic in processing, input supply, and services. Finally, gains that increased trade flows can bring. But these financial institutions that serve the agricultural studies also understate on another level the gains to sector must be developed. a successful conclusion of the Doha negotiations in Investment in infrastructure (both physical and agriculture. It has become increasingly clear since human) is important in reducing barriers to and the beginning of these negotiations that an agree- the high costs of marketing, especially for small ment in agriculture is certainly a necessary--and producers. Some of the most critical investments some argue virtually a sufficient--condition for an for agricultural trade policy reform include those overall agreement. In that sense, the potential gains in trade-related infrastructure and institutions. attributable to an agreement in agriculture will Development of the entire agricultural production have to include at least partial credit for the gains in and marketing infrastructure is essential in captur- other areas, which are considerable for developing ing all the economic gains that trade liberalization countries. All of these simulations of course show offers developing countries (see box 1.4). large gains to the industrialized countries from lib- Agricultural imports into OECD and other eralization, particularly from removing their own major markets face an array of sanitary and phy- trade barriers in agriculture. tosanitary measures and other technical require- ments. These measures, though ostensibly designed to protect human, animal, and/or plant Reaping the Benefits of Liberalization life, can potentially restrict trade and pose major For developing countries especially, capturing the market access problems for developing country full gains from trade liberalization will require suppliers. This is an issue to be taken up in the complementary reforms and investments, includ- negotiations as a priority for developing countries ing the establishment of safety net mechanisms to (see chapter 11). Even when the standards are legit- cushion any negative impacts (see chapter 10). imate, developing countries may have serious diffi- "Decoupling" payments for agricultural support is culties proving that their exports actually meet one option for putting a safety net in place. Box 1.3 these standards, owing to the high cost of some discusses the experiences of two developing coun- testing and certification procedures. Technical tries that have carried out such reforms. capacity constraints--both in the public and pri- Changes are often needed in the rules governing vate sectors--may also seriously hinder developing factor markets to complement trade policy reforms, country compliance with the emerging array of either those undertaken unilaterally or through international standards and technical require- international negotiations. To either create factor ments. Investments are needed to build testing markets or make them more efficient, certain facilities and laboratories as well as to build up the reforms are particularly important. Privatization of necessary human capital. land and security of property rights are central to Most developing countries also lack a strong ensuring that private farming is the main compo- trade support service infrastructure. Investments nent of the farming system. Land markets should are needed to build networks of trade support facilitate entry to and exit from farming. Land mar- institutions capable of providing key services to kets have been especially rigid in some countries, exporters: trade policy information and commer- which is a serious problem when changing incen- cial intelligence; export promotion and marketing; tives require fluid movement of this and other fac- product development and financial services; and tors across different uses within agriculture, or training. The institutional building program also between this and other sectors. includes developing national institutional capacity Reform of inefficient state-owned enterprises for international negotiations, interministerial is another priority, particularly in the agricultural coordination, and public-private sector dialogue. What's At Stake? 19 BOX 1.3 Reforming Inefficient Support Systems: Recent Experience in Two Developing Countries A number of countries in both the industrialized the land must be in agriculture. An additional and developing world have begun to reform problem was that the payments were not ini- their agricultural support systems by removing tially assignable as collateral for loans, and this or reducing traditional price or income supports also was eventually remedied. and substituting payments that are decoupled These initial shortcomings notwithstanding, from (and do not, in principle, affect) farmers' the program has been quite successful. An ex current decisions on input usage or output. This post evaluation found that each peso in this form was the major principle motivating the changes of direct payment generated 1.5­2.6 pesos in in U.S. policy embodied in the 1996 farm bill income for the recipient (Sadoulet, de Janvry, (the "Freedom to Farm Act") and in the 1992 and Davis 2001). In addition, the program must MacSharry reforms in the EU's Common Agricul- be judged a success in the political sense, given tural Policy, which reduced support prices and that it has allowed the government to exceed its gave farmers area-based "compensatory pay- NAFTA commitments in opening the very sensi- ments." Neither of these reforms went far tive maize market to import competition. Before enough, and the U.S. measures have been par- NAFTA, there had been dire predictions that tially reversed, but they were clearly steps in the imports would impoverish smallholder maize direction of decoupling payments. growers, causing large-scale social unrest. The Among the experience of developing coun- share of output- and input-linked support in tries with decoupling, two countries stand out: gross farm receipts has been reduced to less Mexico and Turkey. In the mid-1980s, the gov- than half its previous level (see OECD 2002). ernment of Mexico embarked upon a reform Turkey's program is more recent, so its program that included extensive changes in its impact cannot yet be definitively judged, but agricultural support programs, which until then the experience is instructive from two perspec- had been almost totally dominated by direct tives. First, it demonstrates that with proper government interventions. It foresaw that as a preparation and consensus, this kind of program result of the North American Free Trade Agree- can be implemented quite rapidly. After a pilot ment (NAFTA) with the U.S. and Canada, its program of about one year, the large-scale roll- agricultural sector would have to become highly out of the program began in late 2001. By mid- competitive. As a result, Mexico began to 2002, more than half of the farmers in the coun- reduce support prices and make area-based pay- try (2.18 out of about 4 million total) had been ments to farmers instead under the "PRO- registered and given their first payments. This far CAMPO" program (starting in 1994). There exceeded expectations, given the incomplete were some startup problems. One major prob- state of the land records (cadastre) and registra- lem was caused by the advanced announcement tion system. The payments are area-based and that the payments would be based on the area completely decoupled from production or input planted in specific "program crops," using as a decisions, except that the land has to be used in baseline a three-year period including the year agriculture. The second lesson is one of political following the announcement, rather than using economy. The program has allowed the govern- the area from some past period as the baseline. ment to sustain its phase-out of the previous Farmers responded by greatly expanding their credit and fertilizer subsidies and support prices, plantings of these crops. Another shortcoming which had become so large as to be macroeco- was that in the early years of the program the nomically unsustainable. While an ex post evalu- payments were contingent upon the farmer's ation is yet to be carried out, it seems clear that planting crops on a rather limited eligible list. the program will distribute benefits more evenly That is, payments were not really very decou- than did the previous system, since all farmers pled (delinked from production), so these did are eligible--even the small noncommercial not leave farmers free to make their own plant- farmers who benefited little if any from the ing decisions based on market prices. The list of previous input- and output-based subsidies-- eligible land use was then greatly expanded, so and there is a cap on payments to the largest there are now few constraints, other than that farmers. 20 Agriculture and the WTO BOX 1.4 Cambodia Rice: Challenges to Integration Cambodia offers an interesting example of a government also distributes rice seed to farmers low-income developing country that is integrat- through the Cambodian Agricultural Research ing into the world marketplace for rice exports. Institute and the Agriculture Quality Improve- Still recovering from the devastation of the ment Project. Khmer Rouge years in the late 1970s, the coun- Improvement of post-harvest management is try just recently transformed itself into a net- a critical area for increasing Cambodian farm exporter of rice. Two of the world's largest rice income. Losses due to broken grains, moisture exporters, Vietnam and Thailand, are also damage, and pests during storage seriously located in the lower Mekong River basin, and erode crops suitable for consumption, sale, and because of Cambodia's proximity to China's export. Inconsistent grain quality and older industrializing coast, its potential to become a milling technology also cause losses. In Septem- successful agricultural exporter is evident. ber 2000, an International Finance Corporation Agriculture accounts for nearly half of Cambo- (IFC)-sponsored program estimated that rice mills dia's GDP and rice represents about 90 percent in Thailand produced 95 percent unbroken rice, of hectares under cultivation. At least 80 percent but for farmers in Cambodia grain successfully of the population spends 70 percent or more of milled for export was no more than 30 percent. household income on food, and rice is the largest The low literacy rate and weak communications single food item expenditure across all income infrastructure of Cambodia make educating levels. farmers about better management techniques a Inconsistency in the quality of the grain has challenge, but it is not insurmountable as demon- harmed Cambodia's ability to export product. strated by the corn contract pricing experience. Farmers source their seed from their own crops The 2002 Cambodian rice crop was seriously or from neighbors, causing seed varieties to mix imperiled owing to widespread drought fol- and often resulting in several different varieties lowed by flooding. The government initially of rice in the same paddy. Recent efforts at con- responded with charitable donations of rice and tract farming, where seed is provided to the additional rice seed so that farmers could replant farmer and financial incentives are given to lost seedlings. Further hardship caused by the ensure a quality crop, have proven effective. In extreme conditions was mitigated by the gov- one effort involving corn, the rejection rate of ernment and the Cambodian Red Cross. The product was reduced from 80 percent to 15 per- simultaneous goals of export expansion and cent in four years. A similar effort for rice was food security present great challenges to Cam- begun in 2001 by a private investor who also bodia, but ones that trends show can benefit the made a substantial investment in a new mill. The country in the long term. Sources: Cambodia PRSP; Ministry of Cambodia, IFC, Phnom Penh Post, www.foodmarketexchange.com There are some differences in priority between Notes the low-income and the middle-income countries 1. On the contribution of agriculture to growth and eco- in the area of related reforms and investments nomic development, see, among others, Adelman (1984), Adel- man and Robinson (1978), Johnston and Kilby (1975), Johnston (Valdés 1998). It is in the middle-income coun- and Mellor (1961), and Mellor (1966, 1995). On the issue of tries that the bottlenecks in infrastructure and "consumption linkage" effects critical to the influence of agricul- financial markets are felt most strongly. For the tural growth on the overall growth performance of rural and national economies of developing countries, see, for instance, low-income and transitional countries, the more Mellor and Lele (1973) and Ranis, Stewart, and Reyes (1989). important question is how to facilitate the creation 2. For a fuller discussion of the linkages between agricultural of factor markets. For the middle-income coun- growth and rural and urban poverty and growth see Johnson and Mellor (1961) and Timmer (1995, 1997). tries the issue is more about how to make these 3. See, among others, Ben-David (1993), Dollar (1992), markets more efficient. Furthermore, in middle- Edwards (1993, 1998), Frankel and Romer (1999), and Sachs income countries, where the agricultural sector and Warner (1995). World Bank (2002a) has a good survey. represents a smaller share of the economy, there 4. See Coe, Helpman, and Hoffmaister (1997) for evidence and (Romer 1994a, 1994b) for a further discussion. should be a more elastic factor supply than in 5. Chapter 6 discusses domestic support policies and options lower-income countries. for reform in more detail. What's At Stake? 21 6. Chapters 4 and 5 discuss how to improve market access. Fan, S. 2000. "Technological Change, Technical, and Allocative 7. Chapter 3 discusses how to discipline and eventually elim- Efficiency in Chinese Agriculture." Journal of International inate export subsidies, a topic of critical importance to develop- Development 12:1­7. ing countries. Fan, S., P. Hazell, and S. Thorat. 1999. "Linkages between Gov- 8. Note that the sum of gains from liberalization only by ernment Spending, Growth, and Poverty in Rural India." developed countries and only by developing countries (31 plus IFPRI Research Report 110, International Food Policy 114) are slightly greater than the gain from liberalization by both Research Institute, Washington, D.C. simultaneously (142). Frankel, Jeffrey, and David Romer. 1998. "Does Trade Cause 9. For a discussion, see, for example, Anderson and others Growth?" Unpublished, University of California at Berkeley. (2001) or Hertel and Martin (2000). Frankel, J., and D. Romer. 1999. "Does Trade Cause Growth?" 10. For example, ranging from US$14 billion per year American Economic Review 89(3): 379­99. (Rosegrant and Meijer 2002) to as low as US$7 billion per year Fritsch, Peter. 2002. "An Oversupply of Coffee Beans Deepens for only selected commodities (Vanzetti and Sharma 2002). Latin America's Woes." Wall Street Journal, July 8. Gibson, P., J. Wainio, D. Whitley, and M. Bohman. 2001. Profiles of Tariffs in Global Agricultural Markets. Agricultural Eco- Select Bibliography nomic Report 796. U.S.D.A. Economic Research Service, Washington, D.C. The word processed describes informally reproduced works that Hertel, Thomas, and Will Martin. 2000. "Liberalizing Agricul- may not be commonly available through libraries. ture and Manufactures in a Millennium Round: Implica- tions for Developing Countries." World Economy 23: 455­70. Adelman, I. 1984."Beyond Export-Led Growth." World Develop- IFPRI (International Food Policy Research Institute). 2001. 2020 ment 12: 937­49. Global Food Outlook: Trends, Alternatives, and Choices. Adelman, I., and S. Robinson. 1978. Income Distribution Policies IFPRI, Washington, D.C. in Developing Countries: A Case Study of Korea. Stanford, International Finance Corporation. 2000. "A Rice Industry Calif.: Stanford University Press. Revives, The Millers' Tale." IFC SME Facts 1:9. Alderman, Harold. 2001. "What Has Changed Regarding Rural Johnston, Bruce F., and John W. Mellor. 1961."The Role of Agri- Poverty Since Vision to Action?" Rural Strategy Background culture in Economic Development." American Economic Paper 5. World Bank, Washington, D.C. Review 51(4): 566­93. Anderson K., B. Hoekman, and A. Strutt. 1999. "Agriculture and Johnston, B. F., and P. Kilby. 1975. Agriculture and Structural the WTO: Next Steps." CIES Discussion Paper 99/14, Univer- Transformation. New York: Oxford University Press. sity of Adelaide, Australia. McCulloch, N., A. Winters, and X. Cirera. 2001. Trade Liberaliza- Anderson K., Betina Dimaranan, Joseph Francois, Thomas Her- tion and Poverty: A Handbook. London: Centre for Economic tel, Bernard Hoekman, and Will Martin. 2001. "The Cost of Policy Research. Rich (and Poor) Country Protection to Developing Coun- Mellor, J. W. 1966. The Economics of Agricultural Development. tries." Journal of African Economies 10(3): 227­57. Ithaca, N.Y.: Cornell University Press. Ben-David, Dan. 1993. "Equalizing Exchange: Trade Liberaliza- ------. 1995. Agriculture on the Road to Industrialization. Balti- tion and Income Convergence." Quarterly Journal of Eco- more: Johns Hopkins University Press. nomics 108(3). Mellor, J. W., and U. Lele. 1973. "Growth Linkages of the New Coe D., E. Helpman, and A. W. Hoffmaister. 1997. "North­South Foodgrain Technologies." Indian Journal of Agricultural Eco- R&D Spillovers." Economic Journal, 107: 134­9. nomics 28: 35­55. Datt, Gaurav, and Martin Ravallion. 1996. "Why Have Some Michaely, M., A. M. Choksi, and D. Papageorgiou. 1991. "The Indian States Done Better Than Others at Reducing Rural Design of Successful Trade Liberalization Policies." In A. Poverty?" Poverty, Income Distribution, Safety Nets, Micro- Koves and P. Marer, eds., Foreign Economic Liberalization: Credit Working Papers, 1594, World Bank, Washington, D.C. Transformation in Socialist and Market Economies. Boulder, Dollar, David. 1992. "Outward-Oriented Developing Economies Colo.: Westview. Really Do Grow More Rapidly: Evidence from 95 LDCs, Ministries of Agriculture, Trade, and Industry, Cambodia. 2001. 1976­85." Economic Development and Cultural Change, 40(3): Integration and Competitiveness Study Parts A, B, C, and D. 523­44. November. Edwards, Sebastian. 1993. "Openness, Trade Liberalization, and Nara, Lon. 2002. "Rains Arrive But Situation `Still Growth in Developing Countries." Journal of Economic Liter- Serious.'" Phnom Penh Post (Cambodia), Aug. 17. ature XXXI(3) (September): 1358­93. OECD (Organisation for Economic Co-operation and Develop- ------. Sebastian. 1998. "Openness, Productivity, and Growth: ment). 1998. "Background on Tables/Graphs." Workshop on What Do We Really Know?" Economic Journal 108(March): Emerging Trade Issues in Agriculture. October 26­27, Paris. 383­98. ------. 2001a. The Uruguay Round Agreement on Agriculture: Elbehri, A., and S. Leetma. 2002. "How Significant Are Export The Policy Concerns of Emerging and Transition Economies. Subsidies to Agricultural Trade? Trade and Welfare Implica- Paris: OECD. tions of Global Reforms." Paper presented at the 5th Annual ------. 2001b. OECD Observer. Policy Brief, November, Paris: Conference on Global Economic Analysis, June, Taipei. OECD. Elbehri, A., M. D. Ingco, T. W. Hertel, and K. Pearson. 1999. ------. 2002. Agricultural Policies in OECD Countries: Monitor- "Agriculture and WTO 2000: Quantitative Assessment of ing and Evaluation. Paris: OECD. Multilateral Liberalization of Agricultural Policy." Paper pre- Ranis, G., F. Stewart, and E. A. Reyes. 1989. "Linkages in Devel- sented at the WTO/World Bank conference on Agriculture opment: A Philippine Case Study." Working Paper Series and the New Trade Agenda in the WTO 2000 Negotiations. No. 89-02. Manila: Philippine Institute for Development October, Geneva. Studies. 22 Agriculture and the WTO Ravallion, Martin. 2000. Processed. "On the Urbanization of Timmer, C. Peter. 1995. "Getting Agriculture Moving: Do Mar- Poverty." World Bank, Washington, D.C. kets Provide the Right Signals?" Food Policy 20(5) (October): Romer, Paul M. 1994a. "New Goods, Old Theory, and the Wel- 455­72. fare Costs of Trade Distortions." Journal of Development Eco- ------. 1997. "How Well Do the Poor Connect to the Growth nomics. 43: 5­38. Process?" Consulting Assistance on Economic Reform Dis- ------. 1994b."The Origins of Endogenous Growth." Journal of cussion Paper 178. Harvard Institute for International Economic Perspectives 8(1): 3­22. Development, Cambridge, Mass. Rosegrant, M., and S. Meijer. 2002. "Agricultural Trade Liberal- U.S. Department of Labor. 2003. Global Economic Prospects. ization to 2020: Impacts on Trade, Prices, and Economic Washington, D.C. Benefits." Paper presented at the World Bank/International UNCTAD (United Nations Conference on Trade and Develop- Agricultural Trade Research Consortium Conference, June ment). 1994. Directory of Import Regimes. Geneva. 16­17, Whistler Valley, British Columbia, Canada. ------. 1999. Trade Analysis and Information System. Version Sachs, Jeffrey D., and Andrew M. Warner. 1995. "Economic 6.02, CD-ROM. Geneva. Reform and the Process of Global Integration." Brookings Valdés, A. 1998. "Trade Policy Reform and Agriculture." In J. Papers on Economic Activity (1): 1­118. Nash and W. Takacs, eds., Trade Policy Reform: Lessons and Sadoulet, Elisabeth, Alain de Janvry, and Benjamin Davis. 2001. Implications. Washington, D.C.: World Bank. "Cash Transfer Programs With Income Multipliers: PRO- Vanzetti, D., and R. Sharma. 2002. "Impact of Trade Liberaliza- CAMPO in Mexico." World Development 29(6): 1043­56. tion on Developing Countries: Results of the ATPSM Partial Schiff, M., and A. Valdés. 1992a. A Synthesis of the Economics in Equilibrium Model." Paper presented at the World Bank/ Developing Countries. Vol. 4. The Political Economy of Agri- International Agricultural Trade Research Consortium Con- cultural Pricing Policy. Baltimore: The Johns Hopkins Uni- ference, June 16­17,WhistlerValley, British Columbia, Canada. versity Press. World Bank. 1994. Adjustment in Africa: Reforms, Results, and the ------. 1992b. "Agriculture and the Macroeconomy." Policy Road Ahead. New York: Oxford University Press. Research Working Paper. 1967. World Bank, Washington, ------. 2001. Global Economic Prospects. 2001. Washington, D.C. D.C.: World Bank. Srinivasan, T. N., and Bhagwati, J. 1999."Outward-Orientation and ------. 2002a."Reshaping Global Trade Architecture for Devel- Development: Are Revisionists Right?" Economic Growth opment." In Global Economic Prospects. Washington, D.C.: Center Discussion Paper 806. Yale University, New Haven, World Bank. Conn. ______. 2002b. Rural Development Indicators Handbook. Wash- Thurow, Roger, and Scott Kilman. 2002. "U.S. Subsidies Create ington, D.C.: World Bank Cotton Glut That Hurts Foreign Cotton Farms." Wall Street ------. 2003. Global Economic Prospects 2003. Washington, D.C.: Journal, June 26. World Bank. 2 TRADE AGREEMENTS: ACHIEVEMENTS AND ISSUES AHEAD Merlinda D. Ingco and John Croome Introduction under the General Agreement on Tariffs and Trade (GATT) and continues to apply under its successor, Arguments in favor of trade liberalization as a the World Trade Organization (WTO), means that means of encouraging balanced economic develop- whatever liberalization is achieved is extended to all, ment and reducing poverty do not depend on and thus does not discriminate against economic whether the liberalization is undertaken unilater- efficiencies. These advantages explain why the GATT ally, bilaterally, on a regional basis, or as part of a engaged in successive liberalization "rounds"(formal multilateral effort such as the current Doha Round. trade negotiations) from 1948 to 1994, and why the Virtually all economists agree that the largest share WTO has now launched the Doha Round. of the welfare gains from liberalization accrues to the country that undertakes it, thereby reducing The Plan of This Chapter distortions to its economy. Each approach to liberalization has its advan- This chapter outlines the history of agricultural tages. A government that opts for unilateral liberal- negotiations in the GATT/WTO, briefly describes ization escapes the pressures, complications, and the provisions of the Uruguay Round Agreement delays involved in negotiating with other countries. on Agriculture (URAA), and discusses some issues Bilateral and regional liberalization agreements for developing countries in the WTO's current also have practical attractions, compared with round of trade negotiations. broad multilateral negotiations (see box 2.1). With fewer participants, there are usually fewer compet- How the Uruguay Round ing aims to be reconciled, which helps speed up and Agreement on Agriculture simplify the negotiating process.Very possibly there Came About may also be a closer identity of interests, and there- Legal Position in the GATT fore scope for greater ambition, than is feasible when more countries are involved. The importance of the URAA can be understood Multilateral negotiations, too, have real only by looking at the history of efforts to provide advantages--not least because the basic rule of rules for trade in agricultural commodities. Planners most-favored-nation (MFN) treatment that applied for the post-Second World War reconstruction and 23 24 Agriculture and the WTO BOX 2.1 The Price of Multilateral Negotiations Though multilateral negotiations have their levels, the percentage reduction in prospect, advantages, they also have their price. They are current trade flows, and the size of the potential long, slow, and complex, and make great market; and use this valuation in setting demands on governmental attention at both their own offers. All early GATT negotiations political and official levels. They are prone to were explicitly conducted, and their outcome recurrent delays and crises that may be trig- assessed, on this basis. More recent tariff negoti- gered by events such as foreign elections that ations have had as their starting point would normally be of no concern to most partic- the assumption that most reductions will be ipants. (Most of the major Uruguay Round crises, made according to an agreed formula. However, however, were directly linked to the agricultural the choice of formula is itself influenced by par- negotiations.) A further problem arises from the ticipants' calculations of its expected impact in linkages created among the disparate elements terms of "equality of sacrifice," and reciprocity of a multilateral round that is--like the Doha remains the basis for item-by-item bargaining Round--explicitly "a single undertaking." When on products excluded from the formula. Even "nothing is agreed until everything is agreed," with the formula approach, or in negotiations the most powerful or reluctant participants are on matters other than tariffs, there is a persistent given additional bargaining power with which tendency for any participant to welcome or reject they can make demands on other participants a negotiating proposal on the ground that it that may be economically unjustified but politi- appears to demand less or more movement on its cally unavoidable. This happened, for instance, part than is required of another participant. in the final days of the Uruguay Round. The very Perhaps fortunately, the valuation of tariff terms on which multilateral WTO negotiations offers is an inexact and subjective matter. Even in are conducted, with their focus on "reciprocity," mercantilist terms, and even if the negotiators "concessions," and so on, encourage a mercan- have access to the same detailed information tilist approach to trade policy in which attention about current trade flows and applied tariffs is focused on making other countries open their needed for product-by-product bargaining, all markets, while retention of one's own trade bar- parties to a tariff deal can usually convince them- riers and distortions becomes a mark of success. selves that they have "won." Each participant will In these circumstances, it is all too easy to lose have its own goals and will make its own judg- sight of the fundamental fact that trade liberal- ments, in light of advice from its own domestic ization is first and foremost to the economic producers, about how demand and supply will advantage of the country that undertakes it. respond to the tariff reductions they and others have conceded. These differences are usually suf- Reciprocity, Concessions, ficient for everyone to feel that, in their terms, and All That the outcome of a negotiation has been a success. Such a judgment may eventually be proved Reciprocity--the idea that each step a partici- wrong. Changes in consumer tastes and govern- pant in negotiations takes toward liberalization ment policies, technical developments, and is a concession, to be balanced or "paid for" by many other factors can result in unanticipated an equivalent concession from other participants and far-reaching shifts in the composition, vol- whose trade will benefit--is a concept that is ume, and value of trade flows. Nevertheless, deep-rooted in the work of the WTO and the what matters at the time, in political terms, is GATT. Its origins and clearest expression are in that every participating government can claim to bilateral tariff negotiations, in which countries have gained more than it gave away--and in the seek the reduction of import duties that longer run, provided the outcome of the negoti- hamper access of their export products to for- ations has been a shift in the overall direction eign markets; value the offers made by their of greater liberalization of trade, an economic negotiating partners in terms of present tariff judgment can probably be favorable, too. Trade Agreements 25 economic cooperation set up the International Trade management programs of the economically Organization (ITO) to oversee the operation of a advanced countries. Article XVI:3 determined that multilateral code of trade conduct (Josling, Tanger- agricultural export subsidies (illegal for industry) mann, and Warley 1996, p. 3). Trade issues then were legal, provided they were not used to gain included tariffs, quantitative restrictions, state-trading "more than an equitable share of world trade." On entities, export subsidies and preferential arrange- domestic subsidies, the GATT had no special rules ments, and the possibility of a special regime for but a number of provisions that covered agriculture trade in commodities (of concern to developing as well as manufactures: Article III:8(b) on like countries). Politics prevented the ITO's establish- treatment of domestic and imported products, ment, but its commercial policy provisions provided Article XVI:1 on notifications to the GATT, and the framework for discussing global trade. The 1947 Article XXIIII:1(b) on the nullification of benefits. Geneva trade conference resulted in a trade accord-- The Contracting Parties to the GATT engaged in the GATT 1947--signed by 23 Contracting Parties formal trade negotiations in 1947, 1949, 1951, and (countries). The features of the agreement--its lim- 1956. These negotiations did little to improve the ited mandate, qualified legal obligations, rudimen- conditions of agricultural trade as they focused on tary dispute settlement mechanism, improvised border measures, such as tariffs. In a significant institutional arrangements in Geneva, and unsatis- development in the thinking about international factory arrangements for agricultural trade--were to agricultural trade policy, the 1958 Haberler Com- stay with the GATT for the next 47 years. mittee Report and the work by subsequent com- Protectionism and discrimination are key fea- mittees linked national farm income support pro- tures of international trade in agricultural prod- grams and conditions of world agricultural trade, ucts. Governments have always used various meas- attempted to measure the extent of agricultural ures to assist or subsidize agriculture both in protection (increased transparency of agricultural domestic production and in the trade of agricul- policies domestically and internationally uncov- tural goods. Trade policy instruments such as ered the most trade-disruptive instruments), and domestic subsidies on products and inputs, export determined the impact of developed-country poli- subsidies, import barriers, quantitative restrictions, cies (surplus disposal issues) on developing coun- and nontariff barriers have dealt with the disposal tries. This work enabled the United States (U.S.) of surplus production associated with the cyclical and nonsubsidizing exporters to create the political nature of agricultural production, instability and momentum needed to make agricultural trade a inadequacy of agricultural prices and earnings, and major element of the Dillon and Kennedy Rounds. access to domestic markets. In addition, agricul- Both rounds aimed at removing distortions tural trade policies have been shaped by major caused by protectionist national agricultural political and economic developments: the 1930s' polices, providing improved and ensured access to Global Depression and the collapse in agricultural import markets for efficient exporters, and dealing prices; the Second World War and regulation of the with the particular trade problems of develop- agrifood sector with price controls, production ing countries (textiles and apparel). Little was planning, and food rationing; the 1956 process of achieved on agriculture in the Dillon Round. For forming the European Economic Community agriculture, the Kennedy Round was, in practice, a (EEC) with the 1958 adoption of a protectionist negotiation between the U.S. and the EEC. The and inward-looking Common Agricultural Policy; limited trade liberalization resulted in some and the increase in the number of countries tariff reductions on farm and food goods and the becoming Contracting Parties to the GATT. International Grains Arrangement. The Tokyo The 38 Articles of the GATT have always applied Round (1973­79) reached agreements affecting to agriculture as well as to manufactures. However, trade in temperate zone agricultural products for two "agricultural exceptions" to the general rules access, commodity arrangements (grains; dairy were built into GATT: Article XI:2(c) on imports products; and beef, veal, and cattle), and the code and Article XVI:3 on exports. Article XI:2(c) pro- on subsidies and standards, but agricultural subsi- vided for nontariff border measures (quantitative dies continued, with no specific definition and restrictions), used to enforce domestic market no obligation to reduce their use. Bilateral access 26 Agriculture and the WTO negotiations reduced tariffs, but the Swiss formula, mid-term review (1988, in Montreal) and the first applied to tariffs on manufactured products, was attempt at concluding the Round (1990, in Brus- not used for agricultural tariffs. sels) resulted in failure and in walkouts by delega- In the 1970s, world market developments in agri- tions of some agricultural exporting countries. It cultural goods suggested that the problem of agri- was only in December 1993 that an overall bargain culture was not oversubsidization and lack of mar- was reached among the Contracting Parties in ket access but global food scarcity and unreliable Geneva, allowing the round to be successfully con- supplies. In the 1980s, agricultural producers faced cluded in Marrakesh in 1994. The WTO was estab- weak world markets (owing among others to the lished and when compared with the GATT, was economic crises in Latin America), agricultural pro- much wider in scope, with a stronger institutional duction stockpiles in the U.S. and the European basis and with treaty status. The final outcome sub- Union (EU) and the increasing costs of supporting stantially changed the multilateral trading environ- their agriculture, a collapse in prices, and an increas- ment.1 The achievements of earlier rounds were ing number of disputes between the EU and the U.S. codified, new areas of trade were brought within With a farm policy crisis in most of the major the rules, a new system for settling disputes was industrialized countries, the 1982 Geneva Minister- established, and a separate agreement on agricul- ial outlined a work program for the 1980s and estab- ture was concluded. However, the URAA is a result lished a GATT Committee on Trade in Agriculture of numerous compromises and--as in any product (CTA) in preparations for a new GATT round. This of negotiations--its basic principles have a number examined all trade measures affecting market access of exceptions, and in some cases its wording is and supplies, as well as the operation of subsidies somewhat ambiguous.2 affecting agriculture--especially export subsidies and other forms of export assistance. This process The Agreement on Agriculture involved 41 countries and the European Community (EC) listing, by product, all measures affecting The URAA contains 21 articles and five annexes. It exports and imports, and the relevant GATT provi- applies to all agricultural products as defined by sions on which the measures were based. The CTA's Annex 1 of the agreement (essentially all agricul- discussions on a rules approach or tariff approach to ture and food products including raw fibers and reform were the start of the negotiating process. At hides, but excluding fish and forestry products). the same time in 1982, the Organisation for Eco- Other documents are important to understanding nomic Co-operation and Development (OECD) the process (see box 2.2). began an analytical study of agriculture policies and Other WTO agreements also apply to trade in trade that used the concepts of Producer Subsidy agricultural products. If a conflict arises between Equivalent (PSE) and Consumer Subsidy Equivalent the URAA and any other WTO agreements, the (CSE). The quantitative impact of policies was esti- URAA has priority. However, in many areas it is the mated using a multicommodity trade model devel- general GATT/WTO rules that apply, because the oped by the OECD Agriculture Directorate. The URAA does not include detailed implementation study provided transparency in agricultural policies provisions. For example, the administration of at the international level showing the value of agri- tariff quotas is governed by GATT Article XIII cultural production that resulted from government (Non-discriminatory Administration of Quantita- policies rather than from markets. tive Restrictions) and by the Agreement on Import Licensing Procedures (ILA). Many WTO disputes involving agricultural products have involved the The Uruguay Round GATT itself or other agreements. For example, cases Initial attempts to launch a round of trade negotia- on alcohol against Japan, the Republic of Korea, and tions at a GATT Ministerial meeting in Geneva in Chile involved the principle of national treatment 1982 failed. It took another four years of prepara- (GATT Article III); the EC­Bananas case (see below tions to succeed in starting the Uruguay Round at for details) involved GATT Article XIII, the Agree- Punta del Este, Uruguay, in 1986. For the next seven ment on Import Licensing Procedures, and the and a half years, negotiations continued. Both the General Agreement on Trade in Services; and the Trade Agreements 27 BOX 2.2 Uruguay Round Agreement on Agriculture: Important Documents Schedules of Commitments Some provisions that were negotiated after the AGST tables were submitted are not reflected in The 1994 WTO signatories' Schedules of Com- the tables. Countries acceding to the WTO after mitments, agreed to during the negotiations, the Uruguay Round have had to prepare their detail how the URAA commitments are to be own supporting tables, and for them the base implemented. For tariffs, for example, each period can be different from that used for the member's schedule sets out for each product the Uruguay Round negotiations. "Accession to the tariff rates (that is, the bound rate--the commit- WTO--Information to be Provided on Domestic ments not to increase the tariff above that stated Support and Export Subsidies in Agriculture," level or ceiling) at the start and end of the imple- WT/ACC/4; 18 March 1996. mentation period. New members acceding to the WTO after the Uruguay Round have to sub- mit their Schedule of Commitments following Modalities Document negotiations with WTO members. For many of the commitments made, such as the specific percentage reductions relating to Agriculture Supporting Tables domestic support and export subsidies, the (AGST) methods of calculation establishing these reduc- tions were set out in the important Uruguay For commitments on export subsidies and Round working document, "Modalities for the domestic supports, the starting point is the Establishment of Specific Binding Commitments "Supporting Tables Relating to Commitments under the Reform Programme" (MTN.GNG/ on Agriculture Products in Part IV of the Sched- MA/W/24). This has no legal force, but the ules" (GATT documents G/AG/AGST/Vols.1­3). reduction commitments, used to calculate mem- These tables show the amount each member bers' schedules, are legally binding. However, as country spent on different products and support in the case of the AGST tables, the commitments schemes during the base periods, and how the sometimes changed following negotiations starting point for reductions was calculated. between members. tariffs on milk powder case against the Republic of plemented by nontariff measures (NTMs), such as Korea involved the Agreement on Subsidies and quantitative restrictions (quotas, import bans, Countervailing Measures. To ensure compliance embargoes, the monopoly purchase power of state- with their obligations, members are required to owned entities). Some countries used the simple notify the WTO of their commitments (see this border protection method of tariffs while other chapter's Annex on Notification Requirements). methods used were more complicated, such as the Commencing in 1995, the implementation period EU variable levy system. This system maintained a for completing URAA commitments is defined in stable price within the EU by using the same post- Article 1 as six years for developed countries, but customs duty price for all imports. The lower the for the purpose of Article 13 (Peace Clause), a nine- border price of the product, the higher the tariff, year period applies. Article 15 provides an imple- resulting in a stable consumer price. With the stable mentation period of 10 years for developing coun- price objective, other countries restricted the import tries. The following sections discuss selected articles. volume, allowing imports when the domestic price increased and banning imports when it fell. These approaches stabilized prices within the importing Article 4: Market Access country, but usually at a high level and have been Before the Uruguay Round, trade liberalization blamed for amplifying international price fluctua- negotiations covered border protection for all goods, tions. If prices rose and imports were allowed, this except agricultural products for which there were meant that international demand increased, driving often no bound tariffs or for which tariffs were sup- up the international price. If prices fell, imports were 28 Agriculture and the WTO restricted, reducing demand on the international (Article 3 and legally enforceable through Article 4 of market and pushing international prices down. the URAA and Article II:1(b) of the GATT 1994). With the intention of aligning agricultural trade Developing countries were not required to rules with those applying to trade in other goods, undertake tariffication in the same way as devel- negotiators agreed that all barriers to imports, other oped countries. They were not required to convert than those in place for health and safety reasons, NTBs into tariffs giving the same protection, as should be subject to tariffs only--the only permitted estimated by taking the difference between domes- form of domestic protection (Article 4.2). All forms tic and border prices. Instead, they could opt to of import restrictions had to be converted into bind tariffs at arbitrary levels (often quite high) or tariffs--a process defined as "tariffication," with the through a combination of tariffication for some method differing according to country status (that products and bindings on others. is, developed or developing). The country commit- On its own, a tariff-only situation improves ted not to raise tariffs above a specific (bound) level transparency but does not necessarily result in bet- (also referred to as ceiling binding) for all products ter market access. For this reason, countries agreed covered by the agreement (see Appendix B). Coun- that to preserve existing market access and to create tries were free to use lower rates (applied)-- greater access, current access opportunities had to provided these were consistent with GATT Article I be maintained. With the removal of NTMs and and used on an MFN basis, or used in cases in which with some countries concerned about sudden the imports were subject to special arrangements surges in import volumes or a fall in prices of such as trade agreements or beneficiaries of prefer- imports, negotiators agreed that a special agricul- ential schemes. tural safeguard (Special Safeguard Provisions (SSG) In general, tariffication involved converting see Article 5) could be applied to certain products. NTMs into tariffs using the price-gap method, that The URAA's Annex 5 provided limited opportuni- is, the difference between domestic and world mar- ties for countries to undertake minimum import ket prices. Thus, if the world price for a product commitments for certain products, rather than was US$150 per ton, and the price inside the coun- adopting tariffs for them. This option was taken by try was US$200 per ton, then a tariff of US$50 per Japan, the Republic of Korea, and the Philippines ton could be the result from tariffication. for rice, and by Israel for certain sheep and dairy After establishing the tariff equivalent of an import products. Japan has since tariffied all rice imports. restriction, reductions were required: developed While ad valorem (that is, value-based) tariffs countries by an average of 36 percent and a minimum may be the simplest to establish and implement, of 15 percent over six years; and developing countries some countries have relatively complicated tariff by an average of 20 percent and a minimum of 10 per- structures, using fixed charges or a combination of cent over 10 years. These were simple averages and fixed and ad valorem tariffs. were not weighted for the volume of trade. However, an even more serious problem was that the reduction Tariff Peaks and Tariff Escalation commitments were based on "average cuts" rather Post-URAA agricultural tariffs remain high com- than cuts in the average tariff. That is, if a tariff of pared to those on industrial products, with tariff 1 percent was cut to one-half percent, this counted as peaks defined as rates exceeding 15 percent or three a 50 percent reduction. Thus, countries could achieve times the average nominal tariff, and tariff escala- the target cuts by reducing already-low tariffs a lot, tion defined as much higher tariffs on processed while making only the minimum reduction in sensi- products compared to unprocessed commodities. tive products with high tariffs. For example, the EU-applied tariff is 18 percent for Having made the necessary calculations for all the fresh grapes but 215 percent for grape juice. different tariff lines, countries drew up their sched- ules of commitments for agricultural products, Exceptions to Bound Tariffs showing bound rates and reduction commitments. The draft schedules, after being checked by other Article XX (General Exceptions) of GATT 1994, countries during a final verification phase, were then which lists the general GATT exceptions to the tariff- included in each Member's Schedule of Concessions only border measure rule, now applies to agricul- Trade Agreements 29 tural products in the same way as it does to indus- access commitments are listed in each member's trial products: Schedule of Concessions. In most cases, market access commitments were implemented by tariff · Article XX allows import restrictions for a quotas, which allow imports at low tariff rates up to number of reasons including the protection of certain volumes. This arrangement, known as a tar- human, animal, or plant life or health iff rate quota (TRQ), has not been without difficul- [Article XX(b), elaborated in the Uruguay ties, with members' notifications often questioned Round Agreement on the Application of Sani- in the Committee on Agriculture. Although the gen- tary and Phytosanitary Measures], reasons of eral principles are clear, many members believe that public morals [XX(a)], and protection of there is not enough clarity or detail in the rules on national treasures [XX(f)]. distribution of TRQs by countries and companies. · National security measures are exempt from the WTO Agreements (GATT 1994 Article XXI). GATT Article XIII · Balance-of-payments difficulties may justify restrictions (GATT 1994 Articles XII and Each member's market access concessions set out in XVIII). its Schedule of Concessions must comply with · General safeguard action is possible under GATT 1994 Articles, most notably Article XIII GATT 1994 Article XIX and the Agreement on (WTO 1997) and the Agreement on Import Licens- Safeguards. ing Procedures (ILA) (WTO 1997). As regards the · Antidumping action may be taken under GATT distribution of TRQs among supplying countries, 1994 Article VI and under the Agreement on the Article XIII requires that import restrictions be Implementation of Article VI, and countervail- applied in a nondiscriminatory manner: each sup- ing action may be taken under the Agreement plying country should have the same share of the on Subsidies and Countervailing Measures. import market as it would achieve in the absence of However, action under the subsidy provisions an import restriction. In other words, if country A may be restricted by the URAA "Peace Clause" supplies 10 percent of the import demand for a (Article 13, Due Restraint), which restricts retal- product in the absence of import quotas and then iation rights in cases where the provisions of the quotas are introduced, it should expect to supply URAA itself on domestic support and export 10 percent of the total import quota. subsidies are being complied with during the Article XIII:2 (d) gives some guidance--quotas period of implementation. allocated among new supplying countries can be allocated after consultations and agreement with countries that have a substantial interest. If this is Tariff Rate Quotas not practical then allocations should be based on As noted, tariffication did not necessarily result in some representative period. The WTO's Appellate liberalization or greater market access, although it Body clarified this requirement to some extent in did make the level of protection more transparent the dispute case, "European Communities-- by establishing tariff bindings (ceilings). However, Regime for the Importation, Distribution, Sale, and high tariffs can prevent realistic market access Distribution of Bananas." Its report stated that opportunities. Prior to the Uruguay Round, some in allocating quotas among supplying countries, member countries permitted limited imports at the importing country should allocate first to sub- relatively low tariff rates, but charged higher tariffs stantial suppliers and then to smaller suppliers (WTO on additional imports or did not allow imports 1997)--that is, allocation to the largest suppliers first over the quota limit. Members allowing import before it can allocate to the smaller suppliers. opportunities of more than 5 percent of the domes- For some TRQs, there may not be a representa- tic market agreed to maintain these opportunities, tive period, and it may not be possible to negotiate while other members undertook to create opportu- satisfactorily with all supplying countries. In the nities equivalent to 3 percent of domestic demand, EC­Bananas case, after the Appellate Body deci- rising to 5 percent by the end of the implementa- sion the EU tried to negotiate with supplying tion period. These current and minimum market countries. When this failed, it allocated quotas to 30 Agriculture and the WTO the principal suppliers on the basis of a historical Article 5.9 stipulates that the SSG is to remain in force period. However, under arbitration it was pointed for the duration of the reform period "as defined by out that to satisfy Article XIII the historical period Article 20." This sets no time limit for reform, which must be representative (WTO 1999). is defined only as "an ongoing process" of "substan- tial progressive reductions in support and protection resulting in fundamental reform." The SSG provi- Agreement on Import Licensing Procedures sions could thus remain in force for some time. How- Article 3 of the ILA on nonautomatic licensing ever, Article 5.9 can be interpreted as meaning that requires that in operating TRQs, the administration the right to use the SSG depends on the continuation system should not distort or restrict trade any fur- of the reform process, so that a failure of negotiations ther than the TRQ itself. It requires that information to continue this reform process would mean that the be provided to the WTO on the administration sys- right to use the SSG would lapse. tem. In the EC­Bananas case, the Appellate Body In the period 1995­2001, 10 of these 39 WTO found that although the ILA applies to import members notified the WTO of action under or, licenses made under TRQs, it does not apply to the in the case of the European Communities, made systems of licensing--only to the application of the operational by the SSG (WTO 2002b). system (WTO 1999). The systems themselves may not be fair and neutral, but they must be applied Price-Based SSG in a nondiscriminatory manner and should not increase any trade distortions arising from the The method of calculating the SSG is set out in quotas. Thus, a system that allocates licenses Article 5. For the price-based SSG, the trigger price among countries may be considered unfair by the is the average 1986­88 c.i.f. price with the addi- supplying countries, but it satisfies the ILA if no tional rate of duty depending on the difference individual country is discriminated against in the between the trigger price and the price of the administration of the license allocation system. imported product. As the price of the imported commodity falls, the SSG tariff rises. This compensates for the fixed Administration of TRQs tariff and reduces or eliminates the effect of falling There are different methods of administering prices on the domestic market. The degree of total TRQs, ranging from the relatively straightforward protection to the domestic market increases when method of first-come, first-served (first applica- international prices fall, and the SSG tariff blocks tions get allocations), to more complicated meth- the transmission of sudden price drops to the ods of auctioning quotas or allocating quotas on a home market. historical basis. Although no single method seems to operate perfectly and questions have been asked Volume-Based SSG about the different systems (especially the auction- ing system), there have been no formal challenges The volume-based SSG can be just as hard to calcu- made under WTO dispute settlement so far, apart late as the price-based SSG. Although the rate of from the EC­Bananas dispute. additional duty is up to one-third of the applied tariff, the calculation of the trigger volume is either 125 percent of the average imports over the past Article 5: Special Safeguard three years or calculated by a formula set out in Provisions Article 5 on the level of market access opportunities The SSG is available only to the 39 WTO members and the change in domestic consumption. that undertook tariffication, and reserved the right to have, subject to the relevant conditions being met, Article 6: Domestic Support recourse to the SSG in respect to certain designated Commitments products. These provisions may be used only on imports outside the tariff quota, with additional tar- Not all subsidies distort trade to the same extent. In iffs imposed if import prices fall or if volumes rise. some countries the widespread use of production- Trade Agreements 31 related subsidies, such as price supports, led the de minimis of total production is the Current to increasing agricultural production above the Total AMS established for a country. market equilibrium level. This excess production Reduction commitments apply only to the had to be stockpiled or exported. With world Current Total AMS and not to its components. market prices often much lower than domestic A country could redistribute supports among dif- prices, exports required export subsidies. Thus one ferent products, provided it has complied with its subsidy (domestic supports) led to another obligations to reduce the total figure. If no supports (export subsidies). Even without exports, domestic have been paid and no reduction commitments supports increase production and reduce demand apply, that country may use production-distorting for imports, thus affecting trade. supports--provided the total value of such sup- To apply reduction commitments to distorting ports is not greater than the de minimis level of agricultural subsidies, these were classified according 5 percent for developed countries and 10 percent to "boxes," using the traffic light approach, with red for developing countries. for prohibited subsidies, amber for subsidies that The reduction commitments, using the 1986­88 had to slow down, and green for nontrade-distorting base period, were 20 percent by 2000 for developed subsidies. The negotiators decided to treat export countries and 13.3 percent by 2004 for developing subsidies separately, so the red box disappeared, countries, with the following exemptions excluded while a new blue box covered direct payments to from the AMS calculation for developing countries producers under production-limiting programs, under Article 6.2: considered to be less trade-distorting than market price supports. In this way an exception was made · Investment subsidies generally available in from the general rule that production-linked subsi- developing countries dies should be reduced. · Input subsidies generally available to low- income or resource poor producers · Subsidies to encourage diversification away Amber Box from illicit narcotics production. Article 6 sets out the reduction commitments for domestic supports. These are expressed in terms of Green Box General Requirements the total Aggregate Measurement of Support (AMS), calculated according to provisions in Arti- Annex 2 lists the types of subsidies exempt from cle 6, Annex 3 and Annex 4. To establish a basis for reduction commitments. Exemption means that, reductions, each country set out its financial sup- subject to the provisions of Annex 2, countries may ports by product and program during the base increase spending or introduce or amend these period (1986­88) and data are in the Agricultural subsidies. To qualify as a green box measure, a sub- Supporting Tables (G/AG/AGST/) series of docu- sidy must satisfy the overall requirements set out in ments. Each country agreed to reduce its supports paragraph 1 of Annex 2: it must have no, or at on the basis of this data. most a minimal, trade-distorting effect; and must Countries with no amber box support agreed be provided through publicly funded government not to use supports over a de minimis level of 5 per- programs and not have the effect of providing price cent (10 percent for a developing country) of the support to producers. Despite these general total value of agricultural production. requirements, the green box covers a wide range of Calculations of compliance start at the product programs as listed below. level and if the value of support at the product level is less than the de minimis figure,it is treated as zero.For Government Programs nonproduct-specific supports the same rule applies, except that its value is compared with the total value Government programs set out in paragraph 2 of agricultural production. Only product-specific of Annex 2 are exempted from any reduction com- supports greater than de minimis and nonproduct- mitments. These cover many of the traditional gov- specific support greater than de minimis are then ernment duties associated with various economic added together, and only if that total is greater than sectors, such as expenditures for research, training, 32 Agriculture and the WTO and inspection services; and infrastructure services, transition from centrally planned economies including capital works programs for electricity, to market-oriented economies, this provision transport, irrigation, dams and drainage, and envi- has also been used by other countries with ronmental programs. Food security, addressed to structural problems in agriculture. Countries some extent in paragraphs 3 and 4, allows public trying to encourage younger farmers may have stockholding for food security and domestic food farmer retirement programs, while countries aid programs. trying to increase average farm size may have land purchase grants or subsidized loans. · Environmental programs support is provided to Direct Payments farmers to encourage compliance with environ- Direct payments to farmers are covered in Annex 2, mental rules or enrollment in environmental paragraphs 5 to 13. Paragraphs 5 and 6 require that, programs. The value of payments should be lim- for a direct payment to qualify under the green box, ited to the extra costs or reduced income associ- it must comply with the general requirement of ated with the environmental program. paragraph 1 and the specific criteria set out in para- · Regional assistance support is given to overcome graphs 7 to 13. Basically, direct payments must not the difficulties faced by producers in disadvan- be linked to current production and can be pro- taged areas. Although the payments should be vided even if there is no production. based on some historic reference period, they Payments can be made for income support, dis- cannot be related to current production and are aster relief, structural adjustment, and as part of limited to the extra costs or lower income asso- environmental and regional assistance programs. ciated with farming in such areas. Such programs are detailed below: Blue Box · Decoupled income support is provided as pay- ment based on production at some fixed time A special exemption from AMS commitments prior to the base period (1986­88) and not included in Article 6.5 covers payments made based on current prices, current production, or under production-limiting programs, provided other factors of production. No production is that such payments are based on fixed areas, crop necessary to receive this support. yields, or livestock numbers, or, if the payments · Income insurance support is allowed when are variable, on 85 percent of the base level of pro- income loss is greater than 30 percent, based on duction. These payments are used in some coun- the average for the previous three years. The tries (such as those in the EU) where traditional protection must not exceed 70 percent of the market support payments have caused problems income loss. of overproduction, have become too expensive to · Disaster relief support is possible when produc- maintain, or are too inefficient compared to the tion has declined by at least 30 percent com- objective of maintaining farm incomes or rural pared to the average for the previous three years, employment. and the payments must be related to loss of Direct payments to producers are much more income or of factors of production such as land efficient than market supports in maintaining or livestock. The value of the payments should incomes, planted areas, or livestock numbers. How- cover only the loss suffered and, when made ever, basing support on animal numbers or on along with income insurance payments, must areas under crops can distort trade. It has been not exceed 100 percent of the lost income. argued that this distortion is probably not as · Structural adjustment support is possible for great as would be the case under market support- retirement programs for producers (permanent type payments. retirement) and resource retirement programs (the land must not be used for agriculture for at Varying Domestic Support Levels among Members least three years). Payments can also be made through investment aids aimed at over- The level of domestic support provided by WTO coming structural disadvantages, or for land re- members varies widely. Members with high sup- privatization. Although targeted at countries in port levels during their base periods, and with Trade Agreements 33 reduction commitments, have tended to con- none of their commitments, and others are well tinue giving high levels of support but have shifted below their commitment levels for most products. support from amber box measures to blue and Export subsidies are concentrated on a few prod- green box measures. Examining notifications of ucts such as bovine meat and dairy. measures on domestic supports requires care, given different currencies (subject to varying fluctuations Article 9: Export Subsidy in exchange rates and inflation), the size of the agri- Commitments culture sectors, and the structure of support that is provided (often not evenly across all sectors). Export subsidies are subject to special treatment in Although the concept of the AMS calculation was the URAA under Articles 3, 8, 9, 10, and 11, with based on the OECD's PSE, there are several impor- Members' Schedules of Commitments detailing tant differences between the AMS and the PSE. The reduction commitments. Articles 3 and 8 require PSE includes supports that are excluded from the members to abide by their commitments and act in Current Total AMS because they fall within the blue conformity with the agreement. Article 9 lists the box, the exemptions for developing countries, or the types of subsidies subject to reduction commit- green box. For the past few years some OECD coun- ments, and Article 10 states that other measures tries have increased their levels of domestic support, cannot be used as a way to circumvent export sub- but as WTO members, these same OECD countries sidy commitments. have been within their commitments because they Only 25 countries listing reduction commit- reduced market and price supports in favor of direct ments are allowed to use such subsidies as payments decoupled from production. detailed in their Schedules. No new subsidies can be introduced, nor can existing commitments be transferred to other agricultural products. The Article 8: Export Competition reduction commitments in Article 9.2(b)(iv) Commitments require that members reduce export subsidies by Under Article XVI of GATT 1947, export subsidies 21 percent in volume and 36 percent in value over were prohibited. However, Article XVI:3 gave an the six-year period from 1995 to 2000. For develop- exemption for primary products provided they did ing countries, the reduction commitments are not give the concerned GATT Contracting Party 14 percent in volume and 24 percent in value over more than "an equitable share of world export the 10-year period from 1995 to 2004--a special trade." This provision's lack of clarity allowed some provision in Article 9.4 exempts developing-country Contracting Parties to use export subsidies to members from commitments on subsidies aimed at increase or maintain their shares of world agricul- reducing the cost of marketing, including internal tural trade. The negative effects of export subsidies and external transport, handling, and processing have been blamed for amplifying world market costs. This exemption is available only during the price variations because the level of subsidy tends to implementation period as defined by the URAA. increase in times of low world prices and decrease when prices rise. This exaggerates the swings in Rollover of Commitments world prices by reducing supply in times of high prices and increasing it in times of low prices. Article 9.2(b) allowed some flexibility in meeting Because of the extent to which export subsidies dis- export subsidy commitments. For the years tort competition, members agreed to introduce spe- 1996­99, countries could exceed their commit- cial rules for controlling their use. Export subsidy ments or carry over unused commitments, pro- reduction commitments are based on use during vided that by the end of 2000 the total amount for the years 1986 to 1990, but members were given the 1995­2000 did not exceed the total in their Sched- flexibility of using 1992 levels if these were higher ules. The annual commitment could not be exceeded than in the base period. Members using export sub- by more than 3 percent in value, or 1.75 percent in sidies agreed to reduce them, while those not using volume. Although an implementation period them agreed not to start doing so. Twenty-five described in Article 9.2(b) no longer applies, a members have export subsidy commitments cover- number of notifications are still being made show- ing 428 product groups. Some members have used ing members' usage of the rollover relief provision. 34 Agriculture and the WTO Article 10: Prevention food processing industry to develop, while others of Circumvention of Export consider export restraints a threat to their food Subsidy Commitments security. Article 10 reinforces the limitations on the Due Restraint use of export subsidies by trying to prevent circum- vention of commitments, by stating that export Article 13, the "Peace Clause"--a temporary subsidies not listed in Article 9.1 must not be used measure due to expire at the end of 2003--is to circumvent members' obligations. To reinforce intended to restrain members from taking action this, members undertake to work toward develop- against others with respect to their commitments ing disciplines on export credits. OECD negotia- on market access, domestic support, and export tions on disciplines on export credits covered by subsidies. The Article does not prevent all chal- Article 10.2, although not finalized, cover export lenges to subsidized agricultural production or credits with the aim of reducing their ability to cir- exports, but it does restrict the right to challenge. cumvent export subsidy commitments. Article 13 does not prevent a member from using Under Article 10.3, an exporting country must other provisions that allow protection such as show that any exports above its commitment levels safeguards (Agreement on Safeguards) and bal- receive no export subsidies.According to the dispute ance-of-payments difficulties (Articles XII and settlement panel in the EC request relating to the XVIII:B of GATT 1994). The WTO Agreement on U.S. Foreign Sales Corporations Tax (FSC) (WTO Subsidies and Countervailing Measures (SCM) is 2002a), a country with export subsidy commit- usually used to retaliate against subsidies, while ments must prove that any exports in excess of its the antidumping provisions of GATT 1994 (Arti- commitments do not benefit from the FSC subsi- cle VI) are used for goods sold at prices below the dies. If, however, a country has no export subsidy domestic market. Article XXIII of GATT 1994 is commitments, the country making the accusation used in cases in which another member takes needs to prove that subsidies were involved. action nullifying or impairing benefits from its Article 10 also requires that food aid comply tariff commitments. with the Food and Agriculture Organization's Green Box Article 13(a) prevents any action "Principles of Surplus Disposal and Consultative against green box subsidies, which have no, or at Obligations" and not be tied directly or indirectly most minimal, trade effect and therefore are to commercial exports of agricultural products to unlikely to qualify for countervailing duties. If they the recipient countries. The objective of this is to have more than a minimal effect they are, by defini- prevent members from dumping excess production tion, not green box subsidies. under the disguise of food aid, while continuing to Amber Box and Blue Box Article 13(b) pre- allow food aid donations. vents retaliation under the SCM Agreement, includ- ing under the serious prejudice provisions of Article 6. This means that if a member loses market share Article 12 and Article 13: Export because of subsidized production from another Restraints and Due Restraint country, it cannot retaliate. It is also not possible to Article 12 sets out the disciplines relating to use the nullification and impairment provisions of export prohibitions or restrictions. Countries are Article XXIII of GATT 1994 for retaliation. Action required to take into account the effect on an under Article VI of GATT 1994 is possible if injury, importing country's food security. Before imple- or threat thereof, can be determined, although due menting any restraint, a country must notify the restraint must be shown before initiating a counter- Committee on Agriculture and, if requested, enter vailing duty investigation. into consultations with any member having a sub- Export Subsidies A member's right to apply stantial interest in the products concerned. Article countervailing or antidumping duties on subsi- 12 does not apply to developing countries unless dized exports is restricted by Article 13(c), which they are net exporters of the foodstuff subject to requires a determination of injury or threat of the export restraint. Some countries feel that such injury and also requires that due restraint be restrictions may help encourage the domestic shown in initiating any countervailing duty inves- Trade Agreements 35 tigations. There is no prohibition on nullification reduction commitments on tariffs, domestic sup- and impairment action under Article XXIII of port, or export competition. GATT 1994, although this would require "reason- It has been argued that, apart from market able expectation" of a benefit and it could be access, many of the provisions on S&D treatment argued that a member would be expected to use are of little use to developing countries, given their any flexibility allowed by its export subsidy com- inability to provide subsidies (10 percent de min- mitments. It is not possible to take action under imis rule for market supports), and for input subsi- Article V or VI of the SCM Agreement for this pur- dies (exempted from the Amber Box calculations). pose, which means that actions in response to seri- In addition, for some developing countries that are ous prejudice (such as the loss of a third market) in transition from state control to more market- are not possible. orientated economies, or that are dealing with the Article 13 therefore does not prevent action to effects of natural or human-made disasters, their counter damage caused by export subsidies or by domestic reform, rather than WTO rules, will have domestic supports that have an effect on trade. a much greater impact on their agriculture. Some members have successfully introduced antidumping measures against other members based on these provisions and, although some have Article 16: The Marrakesh Decision been challenged and have withdrawn the meas- Trade liberalization is expected to improve global ures, others have been able to maintain them. prosperity and food security, but negotiators recog- Obviously, in applying countervailing duties a nized that there were some possible negative effects member must comply with the relevant provisions for least-developed and net-food-importing devel- of the Uruguay Round agreements concerned. The oping countries, arising from world market Article 13 prohibition from taking action in price increases. response to serious prejudice may restrict action The least-developed countries (LDCs) are from some countries. Concern about the exports defined as those recognized by the Economic and of another WTO member is not always about the Social Council of the United Nations. Net-food- exports into the domestic market--it can also arise importing developing countries (NFIDCs) are Bar- from competition for third-country markets. In bados, Botswana, Côte d'Ivoire, Cuba, Dominica, these cases, a countervailing duty to protect the Dominican Republic, Egypt, Honduras, Jamaica, home market is pointless as market share may only Jordan, Kenya, Mauritius, Morocco, Namibia, Pak- be regained if the subsidy is withdrawn. istan, Peru, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Senegal, Sri Lanka, Trinidad and Tobago, Tunisia, and Venezuela. Article 15: Special and Differential To address the concerns of LDCs and NFIDCs, Treatment the Marrakesh Ministerial Conference of 1994 Article 15 covers special and differential (S&D) approved a decision on the possible negative treatment for developing countries. There are many effects of the reform program providing for the references to such treatment throughout the agree- following: ment, including the Preamble, which notes that developed countries "would take fully into account" · A review of the level of food aid and initiation of the particular needs and conditions of developing negotiations on food aid commitments suffi- countries in the implementation of market access cient to meet the needs of developing countries. commitments and also recalls that S&D treatment · Adoption of guidelines to ensure that food aid is such as longer implementation periods had been an given in grant form and/or on concessional integral element of the Uruguay Round negotia- terms in line with Article IV of the Food Aid tions. Developing-country Schedules reflect the use Convention of 1986. of special treatment. On market access, many did · Technical assistance by members, in the context not tariffy all of their NTMs, instead opting for ceil- of their aid programs, to LDCs and NFIDCs ing bindings for a range of agricultural products. for improving agricultural productivity and Least-developed countries did not have to make any infrastructure. 36 Agriculture and the WTO Articles 17 and 18: Committee out in the URAA's Preamble; and (d) further com- on Agriculture mitments that are necessary to reach all of these long-term objectives. The Committee on Agriculture, established under Article 17, has as its main role to oversee the imple- Implementation of the Agreement mentation of the URAA and provide a forum for on Agriculture discussing matters related to agriculture trade. In practice, it currently functions almost as two sepa- A review of the experience of implementing the rate bodies. In its regular sessions, it reviews and new rules on market access, export subsidies, and discusses matters arising out of members' present domestic support reveals that the results have been obligations under the URAA, reporting regularly to modest. The reasons include weaknesses in many the WTO's Council for Trade in Goods. In separate specific aspects of the URAA, such as the modali- special sessions, largely carried out on an informal ties and the historically high support levels in basis (that is, without observers or formal records), developed countries in the base period (1986­88) it provides the forum in which the current round of from which reductions were made. In some coun- multilateral negotiations on agriculture takes place. tries such as the U.S., reforms undertaken prior to In regular sessions, held about four times a year, a the negotiations were adequate to fulfill the new principal task of the committee is to examine notifi- rules on reducing domestic support. cations made under the URAA provisions outlined A comprehensive analysis of protection indica- earlier in this chapter. The notifications are circu- tors (tariff protection in market access, nontariff lated to all members, who then put questions to the barriers, trade-distorting domestic policies such as notifying countries for answer at the committee market price support, and export subsidies) in the meetings. Under Article 18.6 other matters related to OECD and developing countries provides several the reform program can be raised, and under key conclusions, summarized below. Article 18.7 counter-notifications can be made when First, significant trade and domestic policy distor- one member has information that suggests another tions remain in national, regional, and global trade in should have notified the committee of a particular both OECD and developing countries. Tariffication-- measure or action. The committee is also responsible the conversion of nontariff barriers to tariffs--under for monitoring the implementation of the Decision the URAA was an important step forward, but in of the Possible Negative Effects of the Reform Pro- most OECD countries average agricultural tariffs are gramme on LDCs and NFIDCs, and developments higher (some exceed 500 percent) than applied tar- in trade in agricultural products. iffs for nonagricultural products and continue to restrict trade. The global, unweighted average bound tariff rate for agricultural products is over 60 percent Article 20: Continuation (Ingco 2000). For industrial countries, the un- of the Reform Process weighted average bound agricultural tariff rate is Article 20 required that the reform process on agri- 45 percent. These figures compare with tariffs for cultural trade continue and that negotiations start nonagricultural products, which are below 5 percent by 1 January 2000, one year before the end of the for most OECD countries and about 20 percent on main implementation period for Uruguay Round average for developing countries. Tariffs remain high commitments by developed countries. In its intro- in many countries with the level of tariff protection ductory paragraph, Article 20 states that the long- uneven, ranging from relatively open access for trop- term objective should be "substantial progressive ical raw materials to very high protection for prod- reductions" in agriculture protection and support. ucts such as milk, sugar, cotton, and rice. Some Factors to be taken into account in the negotiations countries have raised their applied tariffs for some are: (a) experience in putting the previous commit- commodities while others have made greater use ments into effect; (b) the effect of these commit- of export subsidies, export credits, or other export- ments on world trade in agriculture; (c) nontrade enhancing policies. concerns, S&D treatment for developing countries, Second, the objective of tariffs providing greater the objective of a fair and market-oriented agricul- transparency of protection levels through tariffica- tural trading system, and other objectives set tion has not been fully realized. Tariff Rate Quotas Trade Agreements 37 and their administration have resulted in more vention--possibly through the subsidy elements in complex tariff regimes, with different applied rates export credits, export restrictions, and revenue- for the same product, specific and mixed tariffs pooling arrangements in major products--is of with both ad valorem and specific elements, and concern. high tariffs often applying only to imports above the quotas. The Present Negotiations: Third, in many cases, tariffication has resulted in The Doha Round higher actual or potential protection than before the Uruguay Round (Ingco 1995, 1997). The tariffi- The current negotiations on agriculture were cation process allowed significant scope for flexibil- launched early in 2000, under the terms of ity, resulting in tariff bindings at rates much higher Article 20, to carry forward the reform process as than actual protection rates applied during the base defined by the URAA. The November 2001 launch period and prior to 1995. Tariff dispersion and tar- by the WTO Ministerial Conference in Doha of the iff peaks have increased in some countries. Tariffs broader Doha Round with its Doha Development escalated with the level of processing, particularly Agenda was seen as a positive development with the in markets for processed tropical products, thereby mandate for agricultural negotiations being further increasing the level of protection and obstructing the described in the Doha Declaration, particularly development of value-added activities in develop- paragraphs 13 and 14 (GATT Article XXXVI:8). ing countries. The texts closed off few options for the negotia- Fourth, domestic support remains highly con- tions, except to state that the overall aim remains centrated in a few OECD countries with levels "fundamental reform encompassing strengthened increasing in recent years. In 1994, the EU, U.S., rules and specific commitments on support and and Japan accounted for about 90 percent of total protection in order to correct and prevent restric- domestic support by OECD countries. Levels tions and distortions in world agricultural mar- declined during the first years of implementa- kets," so as to establish "a fair and market-oriented tion, coinciding with a marked increase in world trading system"(WTO 2000). A direct gain from prices of grains. This allowed countries to reduce the Doha launch of broader negotiations will be subsidies or in the case of the EU, to impose a tax tariff negotiations on industrial products (agricul- on cereal exports. Some countries shifted part of tural inputs) that are not within the scope of the their domestic support from amber box pro- URAA. Negotiations, in the context of the Agree- grams into less distorting green box programs ment on Trade-Related Aspects of Intellectual (which are exempt from global trade disciplines, Property Rights, could improve protection of geo- even though many measures are not totally pro- graphical indicators for food products. The Doha duction- or trade-neutral). In response to low Declaration also modified the terms of reference world prices since 1998, domestic support has for the agricultural negotiations themselves, increased in some OECD countries. although the practical effects of these modifications And fifth, although use of export subsidies (ille- remain to be seen. More substantially, the Doha gal since 1955 on nonagricultural products) has Round opens up possibilities for tradeoffs between been reduced, high levels of export subsidies trade sectors; WTO members that are unwilling to remain and continue to distort world markets. provide new market openings for agricultural During 1995­98, WTO members used 42 percent imports, or to limit subsidies to agriculture, may of the budgetary expenditure and 64 percent of the see an advantage in doing so now. The tradeoffs volume allowed for export subsidies, with the EU represent bargaining power to obtain commit- accounting for 90 percent of all OECD export sub- ments from agricultural exporters in other sectors, sidies. The URAA placed limits on export subsidies such as industrial products. Finally, agriculture can for individual commodities but allowed some flex- benefit from broader trade liberalization if this cuts ibility. With low usage levels early in the imple- farmers' costs for nonagricultural goods and serv- mentation period, when world prices were high, ices and, more generally, if it reduces an overall bias several countries carried forward unused export of national policy against agriculture. Additional subsidy credits to be used at a later date. Circum- benefits may be obtained if the liberalization is 38 Agriculture and the WTO expressed in terms of binding WTO commitments, was expected only after the 2003 Mexico meet- which can simultaneously provide bargaining ing, on the basis of each member's comprehen- power to obtain useful concessions from trading sive initial offer. Because final offers demand partners and add credibility, through the promise politically difficult decisions that, in a broad of permanence, to a country's own reforms. round of negotiations, are never made until gov- Prior to the Doha Declaration, the negotiating ernments have a clear view of potential overall process set out and discussed initial proposals gains and losses, this bargaining can be expected ("Phase 1" of the negotiations, March 2000­March to continue and intensify right up to the end of 2001) and examined these proposals in depth, issue the Doha Round. In any event, the March 31 by issue ("Phase 2," March 2001­February 2002). deadline was not met, thereby postponing deci- Altogether, 45 proposals were put forward in Phase 1, sions and underscoring the importance of the with 121 WTO members (some 85 percent of total Fifth Ministerial. WTO membership) sponsoring one or more pro- posals. During Phase 2, WTO members submitted WTO Committees and Pattern of Negotiations over 100 further papers, almost all informal in char- acter (and therefore not circulated publicly), most of The Doha Round as a whole is supervised by an which elaborated or commented on the proposals ad hoc Trade Negotiations Committee (TNC), made during Phase 1. As a result of this two-stage chaired by the Director-General of the WTO, under process, a fairly complete view of the ideas, ambi- the authority of the WTO's standing General tions, and concerns of participants was available. No Council, which itself is normally chaired by a decisions were taken on any of these proposals, national representative elected by the WTO mem- although some clearly enjoyed greater support than bership. Provision for collective political-level guid- others. Nor was the door closed to new ideas. ance, when necessary, is made primarily through The Doha Declaration (November 2001) came meetings of the WTO's Ministerial Conference. only toward the end of Phase 2 and it set three new Ministerial Conferences occurred in Singapore deadlines for the agriculture negotiations: (1996), Geneva (1998), Seattle (1999), Doha (2001), and Cancún (2003). · March 31, 2003--for the establishment of the Direct responsibility for the negotiations on modalities for future commitments (that is, the agriculture has been given to the Committee on ground rules that will shape members' offers, Agriculture meeting for this purpose in special ses- exceptions, and so on), including provisions for sion. This committee has been responsible for the special and differential treatment. agriculture negotiations since they were launched · September 10­14, 2003--the Fifth WTO Minis- in 2000, so the post-Doha decisions have not terial Conference (held in Cancun, Mexico) for changed their institutional arrangements signifi- members to submit their comprehensive offers cantly. WTO members automatically have full based on these modalities. membership in all of these bodies. If precedent is · January 1, 2005--for concluding the whole followed, the TNC will not, at least until the negotiating round. In the third phase of the agri- very end of the negotiations, intervene directly in culture negotiations, the main focus was on sectoral negotiations: its task will be to review developing the modalities for future commit- progress, ensure that the round as a whole contin- ments to meet the deadline of March 31, 2003. ues to move forward, and prepare decisions for This stage, the first to involve real negotiation, adoption by a future conference. required clarification of, and if possible deci- sions on, key points such as objectives in terms Issues for Developing Countries of overall percentage reductions in tariffs and in the Current Negotiations support, the starting points (reference levels) for these reductions, exceptions, and treatment The focus in WTO negotiations on considerations of past unilateral liberalization. Intensive of reciprocity--the idea that each step a partici- bargaining--the real core of the negotiations, pant in negotiations takes toward liberalization is and almost certainly their most difficult stage-- a concession, to be balanced or "paid for" by an Trade Agreements 39 equivalent concession from other participants bers, or on a basis that is nondiscriminatory among whose trade will benefit--is particularly unhelp- beneficiaries of preferential rates. ful to developing countries. It has been more than Negotiations on trade rules are also influenced 45 years since developed countries pledged in by the habit of expecting reciprocity, but less so GATT that they would "not expect reciprocity for than those on market access, since rule changes commitments made by them in trade negotiations lend themselves less readily to calculations in sup- to reduce or remove tariffs and other barriers to port of claims of victory or defeat than to the out- the trade of less-developed contracting parties" come of bargaining on specific trade barriers such (GATT Article XXXVI:8). However, as long as tar- as tariffs. Even so, such considerations will certainly iff commitments continue to be assessed by the arise also in the negotiations on domestic sup- traditional mercantilist measures, a lesser contri- port and export subsidies for agriculture, since a bution by developing countries, in terms of tariff key element in their outcome will be the choice cuts or bindings, is likely to reduce their bargain- of percentages and other figures that will de- ing power. fine permitted subsidy levels, and that, in terms of A related difficulty concerns negotiating credit consideration of reciprocity, will correspond to the for unilateral liberalization. "Credit" in tariff tariff reduction formulae applied in market access negotiations remains based on the extent to which negotiations. each country reduces its own import duties, or at The concept of reciprocity can help multilateral least binds them (in other words, commits itself negotiations by providing bargaining leverage to not to increase them above stated levels). If a persuade trading partners to reduce their trade bar- country has, since the previous round of multilat- riers. It is also vague enough that each participant eral tariff negotiations, reduced some import can claim, when the final bargain is struck, to have duties unilaterally, it is unlikely to be given credit given less than it has won, thereby enlisting domes- for having done so unless it can document the lib- tic support for the outcome. But it puts developing eralizing effect and WTO-conformity of these countries at some disadvantage, and it can be seri- changes, provide undertakings that they will be ously damaging--most of all to the countries maintained, and show that trading partners will in whose governments that invoke it--to the extent fact benefit from them. This issue has frequently that it encourages mercantilism and diverts atten- been raised by developing countries, but thus far tion away from real economic interests. developed countries have been unwilling to give A considerable problem for many developing negotiating credit for any liberalization that is not countries is their own limited resources available to expressed as a binding GATT obligation, included participate meaningfully in the negotiations. The in the scheduled commitments of the country giv- core institution for the agriculture negotiations will ing it. In addition, there is a perception that probably remain the special sessions of the Com- although negotiations are in principle concerned mittee on Agriculture, and its meetings--so far, held with bound (ceiling) tariff rates, less credit is about four times a year in Geneva--will set the pat- given in practice for binding a lower tariff rate tern of work in this sector. Typically, for agriculture that is already applied than for one that reduces as for other subjects dealt with in the WTO, the pat- the duty actually faced by imports. Unilateral lib- tern of multilateral work is a cyclical one, each cycle eralization is thus not always given the credit it being completed by a periodic meeting of the core deserves. institution. In parallel with all these multilateral For many developing countries, a further prob- activities directly tied to the periodic special lem in negotiations on tariffs and other obstacles to sessions of the Committee on Agriculture, there are market access is that for most products they are also many bilateral meetings, particularly in the generally among the smaller exporters, and as such later stages when bargaining is taking place on tar- may find it difficult to engage with participants iffs and other specific barriers to market access for with large markets. However, to the extent that particular products. The meetings make consider- larger suppliers can negotiate tariff cuts, smaller able demands on all participants but especially players also will benefit because the cuts must be those from smaller developing countries where the applied either on an MFN basis to all WTO mem- representation in Geneva and back-up in the 40 Agriculture and the WTO national capital is a significant drain on resources. Domestic Support (DS) And while many developing countries form coali- Notification requirements for domestic support tions of interests and can seek guidance from the take the following forms: WTO's Secretariat and other bodies, the final DS:1: Summary of Current Total AMS: All responsibility for the negotiations remains with members with base and annual commitment levels each country as its Schedule of Commitments shown in Section 1 of Part IV of their Schedule are becomes part of the WTO treaty obligations. required to make a notification 90 days after the end of their marketing or fiscal year. This table shows the result for the previous 12-month period. Chapter Annex: Notification Supporting tables DS:1 to DS:9 contain the details Requirements of the calculation. Market Access DS:1: Supporting Table DS:1: Lists the Green Box measures that are exempt from reduction com- Market Access commitments require notifications mitments. Under each of the headings contained in as follows: Annex 2 the reporting member is required to give a Market Access: 1 Administration of Tariff Quo- brief description of each measure, its value, and the tas: A single notification is required of method of data source. allocation, with ad hoc notifications made subse- DS:1: Supporting Table DS:2: Developing-country quently if there are changes. members using the exceptions listed in Article 6.2 Market Access: 2 Imports under Tariff Quotas: are required to list the measures under each of the This shows the volume of imports made under tar- three headings of Article 6.2 (investment subsidies, iff quotas, relative to the market access opportuni- input subsidies, and diversification from illicit nar- ties. Notification is required annually. cotic crops) along with a brief description of the measures, their value, and the data source. DS:1: Supporting Table DS:3: Those members Special Safeguards making use of the exception from AMS commit- For SSGs, notification requirements take the fol- ments under Article 6.5 for direct payments lowing forms: under production-limiting programs must list the Market Access: 3 Volume-Based SSG: The volume- programs, the monetary value of the measures, and based notification should as far as practicable be made the data source. before taking such action for the first time in any year DS:1: Supporting Table DS:4: This summarizes for each product, and in any event within 10 days of the product-specific AMS. Details of the individual the implementation of such action. elements are in other tables. Market Access: 4 Price-Based SSG: The price- DS:1: Supporting Table DS:5: This table includes based notification can be used either to provide an data on market price support schemes for each up-front notification of trigger prices or, on a case- product, and covers intervention schemes through by-case basis, for the first use of the price-based SSG which the state buys products at certain minimum for any particular product. The notification should prices. The data must include the description of the be to the extent possible in advance, but in any event basic product, the applicable year, the name of within 10 days of the taking of such action. the measure, the applied administrative price, the Market Access: 5 Annual Notification: An external reference price, the eligible production, annual notification should be made indicating the any associated levies or fees, and the total market use of the special safeguard provisions in any year. price support for each product. The notification should be submitted no later DS:1: Supporting Table DS:6: Includes data on than 30 days following the year in question. If nonexempt direct payments. This would normally the special safeguard provisions have not been cover deficiency price schemes whereby farmers are invoked in any year, a statement to this effect paid the difference between the price they get on should be made. the market and the target price set by the govern- Trade Agreements 41 ment. Similar to Supporting Table DS:5, the data those members that are significant exporters of the must include the basic product, the year, the name different product groups (that is, members of the measure, the applied administrative price, accounting for over 5 percent of world exports) the external reference price, the eligible production, make an annual notification showing the total the total price-related direct payments, other value of exports of the product group concerned. nonexempt direct payments, associated fees, and Statement of Non-Use of Export Subsidies: the total value of direct payments. Members with no export subsidy commitments DS:1: Supporting Table DS:7: If there are any must make an annual statement stating that no remaining product-specific supports, they are such subsidies have been used in the past year. included in this supporting table. This could Statement of Non-Use: Supporting Table ES:2: include subsidies per hectare planted or per animal Developing-country members using the exemption that are not related to price but are aimed at in Article 9:4 for marketing and internal transport increasing production. Also included in Supporting are required to list these subsidies, along with the Table DS:7 are the results from Supporting Tables volume of goods benefiting from them. DS:5 and DS:6. ES:3 Food Aid: Food aid donors are required to DS:1: Supporting Table DS:8: For market price make an annual notification showing the total vol- support schemes under which the price gap ume of food aid given in the previous year. Donors method of calculating the AMS is not practicable, with export subsidy commitments will have the Equivalent Measurement of Support is calcu- already provided this information in ES:1 notifica- lated as set out in Annex 4 of the Agreement on tion and therefore do not need to make an ES:3 Agriculture and notified in Supporting Table DS:8. notification. This could include an intervention scheme of lim- ited application or the cost of disposing of sur- Marrakesh Declaration pluses. The budgetary outlay associated with the scheme is usually used in the calculation. Notification takes the following form: DS:1: Supporting Table DS:9: This shows gen- NF:1: Notification under Article 16.2: Annual eral non-product-specific supports. notification sets out the quantity of food aid pro- vided to LDCs and NFIDCs, the proportion in fully grant form, and technical assistance provided for Green Box Measures developing productivity and infrastructure. Table DS:2: New Green Box Measures: New or modified domestic support measures exempt from Notes reduction are notified to the Committee on Agri- culture on an ad hoc basis. 1. All individual Uruguay Round agreements, decisions, understandings, and declarations are in "The Legal Texts: The Results of the Uruguay Round of Multilateral Trade Negotia- tions." Other documents pertaining to the WTO may be found Export Subsidies (ES) online through the organization's website. 2. Material for this section is largely drawn from Josling, Export subsidy notifications take the following Tangermann, and Warley's (1996) excellent description of the forms: history of agriculture in the GATT. ES:1 Budgetary Outlay and Quantity Reduction Commitments: Members with base and annual Select Bibliography commitments in Section II of Part IV of their Schedules must make an annual notification along Anderson, K., Erwidodo, and M. Ingco. 1999."Integrating Agri- culture into the WTO: The Next Phase." Paper prepared for with Supporting Table ES:1. the World Bank's Conference on Developing Countries and ES:1 Supporting Table ES:1: This shows the the Millennium Round, WTO, September 19­20, Geneva. details of the export subsidies provided. Croome, John. 1999. Reshaping the World Trading System: A His- tory of the Uruguay Round. New York: Kluwer. ES:2 Notification of Total Exports: Members Ingco, M. D. 1995. "Agricultural Trade Liberalization in with export subsidy commitments, along with the Uruguay Round: One Step Forward, One Step Back?" 42 Agriculture and the WTO Supplementary paper prepared for a World Bank Confer- Bananas." 25 September, WT/DS27/AB/R. Office of the Secre- ence on the Uruguay Round and the Developing Countries, tary General, Geneva. January 26­27, Washington, D.C. ------. 1999. "Decision by the Arbitrators. Regime for the ------. 1997. "Has Agricultural Trade Liberalization Improved Importation, Sale, and Distribution of Bananas: Recourse to Welfare in the Least-Developed Countries? Yes." Policy Arbitration by the European Communities under Article Research Working Paper No. 1748. International Trade Divi- 22.6 of the DSU." 9 April, WT/DS27/ARB. Office of the Secre- sion, International Economics Department, The World tary General, Geneva. Bank, Washington, D.C. ------, ed. 2000. The Legal Texts: The Results of the Uruguay Ingco, M. D., and F. Ng. 1998. "Distortionary Effects of State Round of Multilateral Trade Negotiations. Cambridge: Cam- Trading in Agriculture: Issues for the Next Round of Multi- bridge University Press. lateral Trade Negotiations." Policy Research Working Paper ------. 2001. "Doha Ministerial Declaration." 20 November, 1915, Development Research Group, World Bank, Washing- WT/MIN(01)/ DEC/1. Office of the Secretary General, Geneva. ton, D.C. ------. 2002a. "Report of the Appellate Body: U.S.--Tax Treat- Josling, Timothy E., Stefan Tangermann, and T. K. Warley. 1996. ment for `Foreign Sales Corporations' Recourse to Article Agriculture in the GATT. New York: Palgrave Macmillan. 21.5 of the DSU by The European Communities." 14 Janu- Mitchell, D. O., M. D. Ingco, and R. C. Duncan. 1997. The World ary, WT/DS108/AB/RW. Office of the Secretary General, Food Outlook. Trade and Development Series. Cambridge, Geneva. New York, and Melbourne: Cambridge University Press. ------. 2002b. "Special Agricultural Safeguard." Background WTO (World Trade Organization). 1997. "Appellate Body Report, Paper by the Secretariat, Revision. 19 February, G/AG/ EC­Regime for the Importation, Sale, and Distribution of NG/S/9/Rev.1. Office of the Secretary General, Geneva. 3 export competition policies Harry de Gorter, Lilian Ruiz, and Merlinda D. Ingco Introduction level (14 percent over a 10-year period for developing countries), and reduce the value of export subsidies The Uruguay Round Agreement on Agriculture by 36 percent (24 percent over 10 years for develop- (URAA) instituted important commitments to ing countries). The URAA provides flexibility by reduce agricultural protection in the area of export allowing countries to redistribute the value of subsi- competition. The purpose of this chapter is to dies or the volume of subsidized exports over years review and evaluate the export subsidy commit- but the cumulative totals through the year 2000­01 ments in the URAA in the implementation period. are not to exceed those that would have resulted from The effectiveness of the disciplines imposed on full compliance. Countries were also permitted to export subsidies are assessed. Options are presented aggregate products to a limited degree within a com- for new multilateral rules to improve disciplines on modity group in their commitments. The reductions export subsidies in the current World Trade Organi- apply to each of 23 product categories (see table 3.1). zation (WTO) trade negotiations on agriculture. The least-developed countries are not subject to A total of 25 WTO members can subsidize reduction commitments (Article 15.2). exports, but only for products on which they have World Trade Organization notifications on commitments in their schedules to reduce the sub- export subsidies required by each country annually sidies. Those without commitments cannot subsi- have the European Union (EU) spending the dize agricultural exports at all. Some among the 25 biggest share (over 80 percent) of total world have decided to greatly reduce their subsidies or export subsidies (table 3.2). The EU had a similar drop them completely since the URAA was signed. share of subsidies in the baseline period (fig- The agreement includes certain temporary exemp- ure 3.1). The United States (U.S.) export subsidy tions for developing countries, allowing them to expenditures were less than those of several coun- subsidize marketing and transport (Article 9.4).1 tries including South Africa (which eliminated Limits on and reductions in the volume and value export subsidies in 1997) and Switzerland. Canada of export subsidies are the key policy commitment eliminated transportation subsidies for exports and on export competition in the URAA. Each country thereby reduced its support in this category dra- agreed to reduce the volume of subsidized exports by matically (see table 3.3). 21 percent over six years from a 1986­90 base period 43 44 Agriculture and the WTO TABLE 3.1 Percentage of Total Export Volume Receiving Export Subsidies 1995­96 1996­97 1997­98 1998­99 Country All OECD All OECD All OECD All OECD Wheat and flour 7.1 6.0 20.4 21.0 22.0 23.0 24.1 24.0 Coarse grains 33.9 33.0 41.3 45.0 37.9 43.0 55.4 59.0 Rice 13.6 4.0 24.3 10.0 14.5 3.0 15.6 52.0 Oilseeds 0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.0 Vegetable oil 6.4 0.0 5.8 0.0 2.7 0.0 0.4 0.0 Oilcakes 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Sugar 15.2 19.0 20.1 29.0 26.7 30.0 21.6 28.0 Butter and butter oil 39.0 99.0 54.5 103.0 35.6 109.0 93.9 104.0 Milk powder 64.0 89.0 64.0 88.0 39.7 68.0 73.2 74.0 Cheese 11.9 78.0 43.6 76.0 9.9 76.0 7.6 61.0 Other milk products 92.9 81.0 57.2 79.0 54.6 78.0 53.7 74.0 Beef 44.8 58.0 43.8 70.0 31.5 61.0 30.3 40.0 Pig meat 15.2 38.0 9.0 31.0 7.3 20.0 22.8 52.0 Poultry meat 24.1 25.0 13.7 14.0 11.9 13.0 11.6 13.0 Sheep meat 5.6 49.0 4.8 17.0 6.3 6.0 1.9 2.0 Live animals 12.0 34.0 10.7 41.0 0.1 0.0 3.4 3.0 Eggs 9.4 52.0 NA 42.0 3.0 50.0 17.5 17.0 Wine 3.4 26.0 5.2 28.0 6.1 23.0 6.7 24.0 Fruits and vegetables 12.0 28.0 NA 33.0 65.8 29.0 NA NA Tobacco 5.4 4.0 3.6 1.0 1.1 0.0 1.9 0.0 Cotton 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Incorporated products 31.0 0.0 20.6 0.0 22.7 0.0 15.3 16.0 Other agricultural 8.7 23.0 1.9 72.0 2.4 56.0 2.1 0.0 products Source: Ruiz 2000. TABLE 3.2 Total Export Subsidy Commitments Base 1995­96 1996­97 1997­98 1998­99 1999­2000 2000 bound European Union (million US$) 14,800 6,292 6,684 4,915 5,835 5,588 9,400 United States (million US$) 930 26 122 113 150 80 600 World total (million US$) 20,000 6,991 7,489 5,729 6,477 5,972 13,400 Source: World Trade Organization. Most observers consider the rules on export Both "front-loading" and "rollover" occurred. In subsidies to be the most important element of the former, countries could designate a different the URAA (for example, IATRC 2001a, 2001b). base period while in the latter, subsidy commit- However, a closer examination of the evidence ments could be banked across years if unused, to reveals that the reduction commitments are less be used later in the implementation period. Sev- effective than originally thought. The base eral countries converted export subsidy policies period from which reductions were agreed was into domestic support, which was convenient unrepresentative, since it was a time of low because the restrictions on the aggregate meas- world prices and high levels of export subsidies. urement of support were weak (see chapter 6). Export Competition Policies 45 The Plan of This Chapter Export subsidies that are subject to reductions include direct export payments by governments to The following sections will address how effective the firms, industries, or producers of agricultural prod- disciplines on export competition are by evaluating ucts contingent on export performance; subsidized the ways in which countries have circumvented stock exports; producer-financed export subsidies; their commitments, the efficacy of value versus vol- export marketing cost subsidies; export-specific ume reduction commitments, consumer-financed transportation subsidies; and subsidies on goods export subsidies, and other issues such as how food incorporated into exports. Export subsidies may aid, export credits, public stocks, and state trading not be extended to commodities that are not subsi- enterprises (STEs) can subsidize exports. dized in the base period. Widely available export market promotion and advisory services, exempted from reduction commitments, were to be negoti- Circumvention of Export Subsidy ated later. Countries were also permitted to aggre- Reduction Commitments gate products within a commodity group in their The baseline for the export subsidy reductions was commitments. This flexibility allowed countries to calculated in both volume and value terms as an switch subsidies among commodities within the average level applied by the countries on export same group. subsidies from 1986­90. However, reductions could An important indicator of the use of export sub- be made from the 1991­92 levels if higher than the sidies is the proportion of total exports that are 1986­90 average. By taking advantage of this mech- subsidized. This is a measure of the overall reliance anism ("front-loading"), countries were able to on export subsidies for a given commodity. WTO increase their export subsidies expenditures from notifications reveal that the figures for subsidized the 1986­90 levels and start the reductions from a exports and total exports are different, with no higher level. explanation regarding the mechanism by which a The schedule of reductions imposed equal certain volume is subsidized and the remaining is annual reduction instalments on a commodity- not. Table 3.1 shows the share of total exports that specific basis. However, the agreement provided received export subsidies in each period. The two flexibility by allowing countries to redistribute columns for each period compare the share of total the value of subsidies or the volume of subsi- exports receiving export subsidies for all 25 coun- dized exports from the second to the fifth year of tries' exports with the share for Organisation for the implementation period. Allowing countries Economic Co-operation and Development (OECD) to "bank" unused subsidies from previous countries' exports only. periods was subject to conditions in Article 9.2(b): The disparity in the percentage use of export subsidies between the two columns for each period · The cumulative amounts of budgetary outlays and shows that, in general, OECD countries subsidize quantities of subsidized exports, from the begin- exports more. In addition, most countries did not ning of the implementation period through the make use of export subsidies for a significant share year in question, could not exceed the cumulative of their exports. However, most of the agricultural amounts that would have resulted from full com- exports for OECD (industrial) countries rely on pliance with the relevant annual outlay and quan- subsidization to some extent. For some commodi- tity commitment levels specified in the country's ties, the share of subsidized exports is even larger schedule by more than 3 percent and 1.75 percent, than 100 percent, because some countries report respectively, of the base period levels; and data from export certificates issued, which do not · The total cumulative amounts of budgetary out- always result in actual exports. The products that lays for such export subsidies and the quantities rely most heavily on export subsidies are dairy and benefiting from such export subsidies over the meat products with the least for oilseeds, sheep- entire implementation period could be no greater meat, and live animals. than the totals that would have resulted from full The use of the countries' respective export sub- compliance with the relevant annual commit- sidy value commitment for each period is shown in ment levels specified in the country's schedule.2 table 3.3, ranging from 0 percent use in Uruguay to 46 Agriculture and the WTO FIGURE 3.1 Baseline Allocation of the Value of Export Subsidies Poland 3.75% Mexico 3.58% Canada United States 2.71% 4.45% Switzerland-Liecht Australia 2.37% 0.44% Brazil Israel Norway 0.46% Slovak Republic 2.19% New Zealand 0.59% 0.42% 0.65% Colombia Venezuela 1.81% 0.17% South Africa 1.66% Indonesia Turkey 0.14% Other 0.67% Iceland 6.57% 0.12% Cyprus Hungary 0.08% 0.80% EC-15 Uruguay 70.92% 0.01% Bulgaria Romania 0.98% 0.01% Czech Republic 1.04% TABLE 3.3 Percentage Use of Value Commitments by Country Country 1995­96 1996­97 1997­98 1998­99 Australia 0 0 0 0 Brazil 0 0 0 0 Bulgaria 0 0 0 0 Canada 7 1 0 0 Colombia 4 5 6 6 Cyprus 18 17 12 23 Czech Republic 29 33 34 42 EC-15 42 51 44 58 Hungary 25 14 10 15 Iceland 21 3 1 0 Indonesia 0 0 0 0 Israel 34 24 11 2 Latvia 0 0 0 0 Mexico 0 0 0 0 New Zealand 0 0 0 0 Norway 57 59 94 86 Poland 0 36 21 30 Romania 0 0 0 0 Slovak Republic 15 22 34 31 South Africa 17 23 11 3 Switzerland 82 79 74 81 Turkey 53 91 76 66 Uruguay 0 0 0 0 United States 2 12 12 18 Venezuela 9 60 7 0 Source: Ruiz 2000. Export Competition Policies 47 TABLE 3.4 Percentage Allocation of Total Value Commitments by Commodity Period Base 1995­96 1996­97 1997­98 1998­99 Wheat and flour 17.86 21.54 21.40 20.86 19.93 Coarse grains 12.47 11.94 11.96 11.95 11.64 Rice 1.12 1.16 1.30 1.38 1.68 Oilseeds 0.58 0.52 0.54 0.53 0.30 Vegetable oil 0.82 0.91 0.91 0.90 0.92 Oilcakes 0.03 0.03 0.03 0.03 0.00 Sugar 5.62 5.52 5.57 5.62 5.80 Butter and butter oil 10.03 9.98 10.03 10.08 10.14 Milk powder 4.17 4.06 4.11 4.20 4.27 Cheese 4.07 4.66 4.60 4.45 4.32 Other milk products 9.39 9.59 9.54 9.54 9.91 Beef 13.21 13.60 13.58 13.65 13.69 Pig meat 2.31 2.30 2.29 2.26 2.34 Poultry meat 1.37 1.27 1.26 1.23 1.30 Sheep meat 0.31 0.17 0.17 0.17 0.06 Live animals 0.23 0.25 0.25 0.25 0.31 Eggs 0.49 0.48 0.49 0.49 0.51 Wine 0.51 0.47 0.47 0.47 0.48 Fruits and vegetables 2.44 1.99 2.09 2.41 2.73 Tobacco 0.67 0.77 0.74 0.71 0.61 Cotton 0.43 0.45 0.53 0.57 0.74 Incorporated products 6.32 6.45 6.17 6.23 6.28 Other agricultural products 0.67 0.43 0.39 0.47 0.47 Source: Ruiz 2000. 86 percent in Norway in the period 1998­99. Coun- rate effects. In terms of the actual expenditures, tries did not exceed their total expenditure com- meat, dairy, and related products used the largest mitments. The same aggregate analysis by country proportions of the total expenditures in export cannot be made with respect to volumes, owing to subsidies, averaging more than 70 percent for the the differences in units of the various commodity period 1995­99. (Meats accounted for 25 percent, groups. That is, it is problematic to aggregate met- dairy for 34 percent, and incorporated products for ric tons of commodities that differ greatly--such as 12 percent.) honey and whey powder. Table 3.3 shows that, on a Table 3.5 shows the percentage use of the value country-by-country basis, value commitments are allowances allocated to each commodity group.3 not binding. Coarse grains and wheat used a smaller proportion However, on a commodity basis, the analysis of of the allowances, owing to the high world prices in the use of value allowances generates different the beginning of the implementation period results. Table 3.4 shows the percentage allocation of (1995­98). Sugar exceeded the commitments, as total export subsidy expenditure commitments for well as wine during 1996­97. Pig meat exceeded the each period to each commodity group. The largest commitments during 1998­99. The largest percent- percentage allocation of value commitments is for age use of the commitments is that of dairy, meats, wheat, coarse grains, dairy products, and beef. The and incorporated products. differences in the percentage allowances during the Table 3.6 shows the percentage use of the volume implementation period are due to front-loading allowances allocated to each commodity group.4 (when reductions were made from a value larger Coarse grains and wheat used a small proportion of than the 1986­90 average) and also due to exchange the allowances during 1995­96 owing to the high 48 Agriculture and the WTO TABLE 3.5 Percentage Use of the Total Value Commitments Allocated to Each Commodity Group Period 1995­96 1996­97 1997­98 1998­99 Wheat and wheat flour 4 11 7 22 Coarse grains 18 24 18 57 Rice 18 41 19 13 Oilseeds 0 0 0 0 Vegetable oil 47 31 7 1 Oilcakes 0 0 0 0 Sugar 48 72 112 121 Butter and butter oil 19 42 25 25 Milk powder 31 46 38 70 Cheese 72 52 42 43 Other milk products 70 74 81 88 Beef 76 83 49 41 Pig meat 30 24 30 133 Poultry meat 68 43 52 64 Sheep meat 35 16 21 19 Live animals 58 32 0 1 Eggs 22 14 25 33 Wine 89 109 72 58 Fruits and vegetables 37 37 22 24 Tobacco 17 5 1 1 Cotton 0 0 0 0 Incorporated products 67 83 86 98 Other agricultural products 5 24 3 4 Source: Ruiz 2000. world prices. However, the use increased sharply limits have been more binding than the value limits. after 1996, and it became close to binding for This is a result of decrease in the domestic support coarse grains during 1998­99. Sugar, milk powder, price of several important commodities (the EU and other milk products have also increased the even taxed exports of cereals in the beginning of the use of volume allowances to levels very close to implementation period), and increases in world the limits. Pig meat exceeded the commitments market prices, which reduced the need for large per during 1998­99 because of a sharp increase of unit export subsidies. 350 percent in European subsidized exports. Dairy Reduction commitments have been more bind- products, meat (beef and pork) products, and ing for some specific commodities when analyzed wine make the largest percentage use of the vol- in a more detailed breakdown of commodity sec- ume commitments. tors and countries than that presented so far. But as The percentage use of value and volume indicated in tables 3.1­3.6, commitments have not allowances shows that the commitments are binding generally been binding in aggregate. The effective- for some commodity groups, while for others they ness of the limits imposed on value and volume are far above the actual volumes and values used. By on export subsidies depends on several factors, averaging the percentage use of commitments pre- such as world prices and shifts in excess supply sented in tables 3.5 and 3.6, the data show that, on and demand, "banking" of unused subsidies, and average, 36 percent of the value allowances have been "rollovers." Several other factors determine the used, while 45 percent of the volume allowances have effectiveness of the reduction commitments--such been used. These figures suggest that the volume as the particular baseline that was used. Export Competition Policies 49 TABLE 3.6 Percentage Use of the Total Volume Commitments Allocated to Each Commodity Group Commodity 1995­96 1996­97 1997­98 1998­99 Wheat and wheat flour 8.0 27.0 26.0 39.0 Coarse grains 33.0 54.0 42.0 96.0 Rice 17.0 44.0 30.0 29.0 Oilseeds 0.2 0.2 0.0 0.0 Vegetable oil 13.0 15.0 11.0 2.0 Oilcakes 0.0 0.0 0.0 0.0 Sugar 34.0 50.0 88.0 74.0 Butter and butter oil 28.0 55.0 37.0 35.0 Milk powder 66.0 63.0 54.0 80.0 Cheese 95.0 95.0 83.0 65.0 Other milk products 91.0 93.0 95.0 88.0 Beef 79.0 97.0 79.0 65.0 Pig meat 54.0 44.0 37.0 120.0 Poultry meat 74.0 71.0 77.0 74.0 Sheep meat 61.0 58.0 49.0 3.0 Live animals 56.0 41.0 0.4 2.0 Wine 66.0 99.0 98.0 83.0 Tobacco 11.0 7.0 2.0 4.0 Other agricultural products 13.0 2.0 4.0 7.0 Source: Ruiz 2000. Binding Commitments and the Rollover bound rates for several commodities during the of Unused Export Subsidies implementation period, and in some cases, includ- ing alcohol, pig meat, and sugar, exceeded the Several countries exceeded the export subsidy value annual commitments. and volume allowances throughout the implemen- The countries that reached or exceeded tation period. The excess is justified by the coun- the export subsidy volume limits are shown in tries as being "unused subsidies rolled over to table 3.8. The products that are close to or above subsequent years," in accordance to the flexibility the volume commitment levels most often are dairy, offered in Article 9.2(b) of the URAA. This flexibil- meats, sugar, and rice. The limits on coarse grains ity has also been called "banking." The "banking" became more binding during 1998­99, whereas in practice weakens the effect of the reduction com- the former periods, wheat and coarse grains had mitments, allowing strategic trade behavior in been well below the commitment levels. The vol- exploiting market share and allowing countries to ume commitments again became more binding as circumvent export subsidy commitments. Several the end of the implementation period approached. countries resorted to the "banking" practice, The same occurred with the value commitments. including the EU countries, U.S., Poland, and Norway. The Baseline and the Front-Loading Flexibility The countries that are close to or above the export subsidy value limits are shown in table 3.7. The reduction commitments are based on 1986­90, The value commitments have been binding on var- a period of unusually high export subsidies world- ious occasions, and it is clear that more countries wide. The base period, calculated as an average of exceeded the commitments as the end of the imple- the values and volumes applied during 1986­90, mentation period approached. Some commitments resulted in commitments that are, in general, were exceeded by more than 300 percent. The EU higher than the actual levels applied during the countries' expenditures were above the 2000­01 implementation period. Moreover, countries were 50 Agriculture and the WTO TABLE 3.7 Countries Using over 90 Percent of Value Commitments Member Commodity 1995­96 1996­97 1997­98 1998­99 Colombia Sugar -- 105 -- 95 Cyprus Cheese 406 100 100 100 EC-15 Alcohol -- 90 -- 106 EC-15 F & V 92 -- -- -- EC-15 Other milk products -- -- -- 92 EC-15 Pig meat -- -- -- 155 EC-15 Rice -- 141 -- -- EC-15 Sugar -- -- 122 134 EC-15 Wine -- 111 -- -- Hungary Corn 282 -- -- 413 Hungary Red pepper meal 147 141 -- -- Norway Bovine meat -- -- -- 106 Norway Cheese -- -- 113 117 Norway Poultry meat 243 -- -- -- Norway Processed products 97 -- 92 -- Norway Sheep meat -- -- 112 -- South Africa Cocoa and prep. 108 323 99 -- South Africa Tea 112 -- -- -- South Africa Waters -- 144 -- -- South Africa Wine products 94 179 107 -- Switzerland Cattle for breeding 94 -- -- -- Switzerland Fruits -- -- -- 91 Switzerland Processed prods. -- -- -- 100 Turkey Creams -- -- -- 98 Turkey Chocolate 97 -- -- -- Turkey F & V -- 99 99 100 Turkey Eggs/dozen -- -- 100 -- Turkey Poultry meat 99 -- -- -- United States Cheese -- -- -- 96 United States Other milk products -- -- 100 129 United States Skim milk powder -- -- -- 136 Venezuela F & V -- 96 -- -- Note: F & V = Fruits and vegetables. Source: Ruiz 2000. given the flexibility to start the reductions from the Tables 3.9 and 3.10 show selected cases in which 1991­92 levels, if higher than the baseline. This countries have a higher level of subsidies compared makes the commitments even less effective in to the baseline, in terms of both value and volume. reducing trade distortions. With this provision, the The column for 1986­90 shows the levels for subsidies can be higher (during all but the last value/volume in the baseline in table 3.9. The next year of the implementation period) than they column shows that the average levels for 1991­92 would have been if the 1986­90 baseline for reduc- are higher than the 1986­90 average baseline levels. tions had been maintained. An examination of the In these cases, the reductions started from the pattern of use over the implementation period 1991­92 average. The first year of the implementa- indicates that rather than a gradual reduction as tion period is shown in column 1995­96. The intended by the URAA, subsidies in some cases reduction commitment levels in 1995­96 are remained high or even increased in some years. higher than the 1986­90 baseline because they are Export Competition Policies 51 TABLE 3.8 Countries Using over 90 Percent of Volume Commitments Member Commodity 1995­96 1996­97 1997­98 1998­99 Colombia Fruits -- -- 115 138 Colombia Processed products 129 91 19 -- Colombia Sugar confection 316 473 332 1033 Cyprus Cheese 189 100 99 100 EC-15 Coarse grains -- 90 -- 123 EC-15 Sugar -- -- 118 112 EC-15 Rice -- 144 103 99 EC-15 Other milk products 98 100 102 91 EC-15 Cheese 99 99 -- -- EC-15 Olive oil 96 104 -- -- EC-15 Beef meat 90 110 94 -- EC-15 Poultry meat 96 99 105 99 EC-15 Wine 111 115 98 EC-15 F & V, fresh 99 99 98 93 EC-15 Eggs -- -- 90 104 Norway Sheep meat -- -- 142 106 Norway Pig meat -- -- 106 -- Norway Cheese -- -- 102 122 Norway Eggs and products -- 99 117 -- Norway Poultry meat 214 -- -- -- Poland Sugar -- 116 149 119 Slovak Republic Sugar -- -- 100 100 Slovak Republic Other dairy products -- 95 99 110 South Africa Beer 106 -- 105 -- South Africa Other milk products 139 -- -- -- South Africa Wine products 103 619 227 -- Switzerland Cattle for breeding 111 -- -- -- Turkey F&V, fresh or proc. 100 97 100 100 Turkey Poultry meat 100 -- -- -- United States Other milk products -- -- 100 107 United States Skim milk powder -- -- 104 154 United States Cheese -- -- 100 93 Note: F & V = Fruits and vegetables. Source: Ruiz 2000. based on the 1991­92 average. The last column milk products in the U.S. was increased by 426 per- shows the percentage increase in the base value and cent, as shown in table 3.9. volume when using the 1991­92 average instead of The 1980s were characterized by a period of inten- the 1986­90 average. sive use of export subsidies for cereals, coarse grains, Although Canada only subsidized its butter dairy, and other products. The EU used higher subsi- exports in the first year of the implementation dies in the baseline than at the beginning of the 1990s period (1995­96), the front-loading mechanism (except for wheat). Therefore, the export subsidy allowed the volume reductions to start at a volume commitment allowances, including the 2000­01 that was 141 percent higher than the 1986­90 base- bound levels,are higher than the actual levels in previ- line. The 1995­96 commitment level was above the ous years (from 1990­95) for some commodities, baseline level. In other cases, the increase in the which eliminated the need for dramatic changes in the baseline for reductions is still more dramatic. For domestic policies to accommodate the export subsidy example, the baseline value of export subsidies for reduction commitments. 52 Agriculture and the WTO TABLE 3.9 Value Front-Loading 1995­96 1986­90 1991­92 Outlay Increase in Member Commodity Outlays Outlays Commitment Base Outlays Canada Butter 17,227 44,444 38,874 158 EC-15 Beef meat 1,959 2,029 1, 923 4 EC-15 Cheese 534 550 594 3 EC-15 Incorporated products 648 702 717 8 EC-15 Poultry meat 142 147 136 4 EC-15 Raw tobacco 63 106 97 69 EC-15 Wheat 2,015 2,255 2,309 12 Norway Bovine meat 55 116 102 112 Norway Cheese 384 596 538 55 Norway F & V 1 1 1 56 Norway Whey powder 0.4 5 4 1, 250 USA Eggs (dozen) 2,505 8, 784 7,587 251 USA Other milk products 3,277 17,244 14,374 426 USA Rice 3,701 18,373 15,705 396 USA Vegetable oils 22,004 60,734 52,959 176 USA Wheat 568,460 845,836 765,499 49 Source: Ruiz 2000. TABLE 3.10 Volume Front-Loading Member Commodity 1995­96 1996­97 1997­98 1998­99 Canada Butter 4,431 10,657 9,464 141 EC-15 Beef meat 1,040,100 1,179,200 1,137,000 13 EC-15 Cheese 406,700 427,000 426,500 5 EC-15 Poultry meat 362,000 470,000 434,500 30 EC-15 Raw tobacco 140,300 206,000 190,000 47 EC-15 Wheat 18,276,000 20,255,000 20,408,100 11 Norway Bovine meat 1,895 3,610 3,258 90 Norway Cheese 20,516 24,333 22,979 19 Norway F & V 841 1,577 1,425 88 Norway Whey powder 30 143 123 377 U.S. Butter and butter oil 36,705 47,368 42,989 77 U.S. Cheese 3,836 3,989 3,829 4 U.S. Eggs (dozen) 8,758,991 34,930,255 30,261,813 299 U.S. Other milk products 4,300 14,940 12,456 247 U.S. Rice 48,802 318,281 271,660 552 U.S. Vegetable oils 178,860 676,786 587,538 278 U.S. Wheat 18,282,354 21,381,546 20,238,298 17 Source: Ruiz 2000. Export Competition Policies 53 The EU and the U.S. used only 42 percent and the amount of export subsidy expenditure. It is possi- 2 percent, respectively, of the value allowances in the ble that a larger amount of export subsidy used by a first year of the implementation period (table 3.3). given country will distort the world price minimally, This was due to the high international prices that while a smaller value applied by another country will caused the decrease in the per unit subsidy neces- have a larger impact in decreasing the world price. sary, rather than to the commitments themselves. In addition, the conditions of trade will influ- The EU increased the expenditures on sugar, other ence the way in which a country will allocate its milk products, pig meat, poultry meat, and eggs expenditure allowance. With no limit on the per during 1997­98 and the expenditures on almost all unit subsidy applied, the trade distortion might not products during 1998­99 over the previous peri- be decreased by the current rules, because value ods. It also increased the volumes subsidized of and volume constraints will rarely bind at the same almost all products during 1996­97 compared to time. For example, suppose that the volume limit the 1995­96 levels, of sugar and eggs during only is binding. If a shock in the world demand 1997­98 and of wheat, coarse grains, skimmed causes the world price to decrease, the exporting milk, pig meat, and eggs during 1998­99. The U.S. country will be able to increase the per unit subsidy, increased the expenditures on all subsidized and to continue to export the same quantity at a exports (except for poultry meat) during 1996­97 lower world price. In contrast, if limits were over 1995­96, and subsequent increases occurred imposed on the amount of distortion in the world in the dairy products, coarse grain, and poultry sec- price that a subsidy can cause, the country in ques- tors. The aggregate expenditures during 1996­97 tion would have to decrease the volume exported, and 1998­99 were higher than in the previous causing less distortion in the world price. period in each case. The volume subsidized Having no limits on per unit subsidies allows increased for all subsidized products during export subsidies to isolate the domestic market 1997­98 with respect to the previous period from changes in the world market. In the above (1996­97), for dairy products during 1996­97, and example, when the world price decreases because of for poultry and skimmed milk powder during an excess demand shock, if the country can increase 1998­99. the per unit subsidy and hence export the same vol- ume, the domestic price for this commodity will not change, despite the drop in price in world mar- The Per Unit Subsidy and Asymmetry of Protection kets (Ruiz 2000). In comparison with the discipline on tariffs, with The failure to discipline per unit subsidies cre- bound per unit or ad valorem duties, only the total ates another problem. As they are not directly amount of the expenditures (or total volume linked to the world price or the quantities exported, exported) is regulated for export subsidies. This the rate of subsidization across commodities differs results in less transparent constraints on the distor- significantly. Therefore, different incentives are cre- tion to trade caused by export subsidies, because ated for the commodities involved, because the the same value of export subsidy allocated to a effective per unit subsidy changes the relative commodity can have different effects on the vol- domestic prices with respect to the world markets ume exported and hence on the world price. (the nominal protection rates), adding to the Instead of a per unit export subsidy on all exports, distortions and efficiency losses caused by the governments may be targeting all export subsidies to export subsidies themselves. a few exports for the country in that year because of The resource misallocation caused by the use of uncertainty as to how the market will evolve. The per export subsidies can be represented through the unit subsidy would increase for those exports, measure of export subsidy equivalents (ESEs), enabling governments to maximize the export sub- which shows the relationship between the per unit sidy value (or to reach a spending target) by subsidiz- subsidy awarded and the world price of the com- ing fewer exports at higher subsidy rates, thereby modity. Table 3.11 shows ESEs for selected coun- circumventing quantity limits. The degree of trade tries and commodities, based on the per unit distortion created by an export subsidy depends on subsidy calculated in the countries' notifications to the conditions of excess supply and demand, and on the WTO, and free-on-board prices from the 54 Agriculture and the WTO TABLE 3.11 Export Subsidy Equivalents (ESEs) (Percent) Member Commodity Base 1995­96 1996­97 1997­98 1995­97 European Union Wheat and wheat flour 104 36 15 10 15 European Union Coarse grains (barley) 142 49 26 28 32 European Union Rice 226 191 130 148 145 European Union Sugar 195 146 151 164 154 European Union Butter and butter oil 248 102 138 112 118 European Union Skim milk powder 106 36 40 43 39 European Union Cheese 86 60 38 29 43 European Union Beef 102 63 57 49 57 European Union Pig meat 111 135 127 378 173 European Union Poultry meat 44 29 16 16 20 European Union Eggs 54 20 11 14 15 Hungary Pig meat 105 25 16 6 13 Hungary Sheep meat 30 11 7 5 8 Hungary Poultry 95 40 16 23 26 Norway Beef 201 143 141 207 166 Norway Pig meat 336 169 125 160 163 Norway Sheep meat 300 89 128 98 116 Norway Eggs 354 354 150 162 200 Norway Butter 127 81 98 90 89 Norway Cheese 184 149 151 155 152 USA Butter and butter oil 136 0 117 30 58 USA Skim milk powder 114 12 68 53 44 USA Cheese 98 28 37 53 39 Source: Ruiz 2000. OECD. The ESE is the ratio of per unit subsidy to countries. For all these reasons, there is a strong case world price, multiplied by 100. for regulating the ad valorem per unit export subsidy. Table 3.11 shows that in many cases the per unit Finally, there is the issue of export taxes that are export subsidies provided are very high with not adequately addressed in the URAA. During years respect to the world price of the commodity. It is in which world prices are high for traditionally subsi- not unusual that the amount the producers or dized export commodities, an export tax can be used exporters receive from the export subsidy is two or to prevent an increase in domestic prices relative to three times larger than the price paid in the world world levels, isolating the domestic market from the market. The table also shows the asymmetry of the world market. This protects the domestic processing ESE across commodities for the same country. In industry, and also can improve the terms of trade for the EU for example, the ESE for pig meat was 378 large exporting countries. This practice, although percent of the world price during 1997­98, while unregulated by the agreement, is also trade distorting only 16 percent for poultry meat. to the extent that it keeps world prices at levels higher Such asymmetry of incentives will not only distort than they would otherwise be. This has consequences the domestic productive sector and affect domestic on food security for developing countries in times of demand, but will also skew the relative incentives relative shortage. Making supply less responsive to (prices) in world markets. During 1995­96, the ESE prices only aggravates the problem, contributing to for skim milk powder in the EU was 36 percent, three more instability and uncertainty in the food supply. times that of the U.S. This difference reflects the dif- The EU used export taxes on wheat and coarse grains ferent levels of domestic price support, which absorbs (barley) after the rise in prices that occurred at the differences in the competitive advantages of these two beginning of the implementation period in 1995. Export Competition Policies 55 Expenditure versus Quantity Limits exports only the same quantity as it would export if it did not offer any export subsidies),whereas a 100 per- A key policy issue is whether the formula for fur- cent reduction in subsidy expenditure is needed to ther cuts in export subsidies should focus on vol- reach this free trade condition if there are no binding umes or expenditures or per unit subsidies. Ruiz volume reduction commitments. and de Gorter (2001) analyze the conditions under Within a dynamic framework, conditions of which volume or value limits will be more binding. trade change, and therefore the analysis needs to Factors affecting the effectiveness of volume ver- include cases where the free trade equilibrium sus value limits include the level of world prices and shifts during the export subsidy reduction imple- internal price supports, elasticities of excess supply mentation period, or thereafter. This makes the and demand, and the way the free trade equilib- results more complicated, but the conclusions of rium changes over time with shifts in excess supply the analysis in the annex are, in short, that volume and demand. (See the annex to this chapter.) constraints are binding in the case of a positive With regard to the current reduction commit- shift (that is, a shift that moves the free trade equi- ments, value reduction commitments will not gen- librium to the right, or alternatively stated, a shift erally be binding in a static framework (that is, in a that reduces the per unit subsidy necessary to model in which demand and supply schedules do achieve a given level of exports), and that expendi- not shift). For value commitments to be binding in ture constraints are always binding in the case of a such a framework, the percentage reductions in negative shift. volume and value would have to occur within spe- cific ranges. For example, for identical percentage Consumer-Only Financed Export reductions in volume and value, the volume reduc- Subsidies tion is always more effective. This result is inde- pendent of the elasticities of excess supply and The agreement has placed limits on taxpayer- and excess demand. producer-financed export subsidies, but has not rec- Ruiz and de Gorter (2001) determine that the ognized"consumer-only-financed"export subsidies.5 ratio of value over volume reduction commitments A consumer-only-financed export subsidy exists is always smaller than the ratio of the initial over when a government-sanctioned organization (for the reduced per unit export subsidy. This relation- example, a marketing board) price-discriminates ship allows us to predict the effects of bindings on (that is, charges a higher price in the domestic the reduction of per unit subsidies for a given set of market than in the export market for the same market parameters. Using this relationship, a coun- product) and pools revenues to producers (Sumner try can predict the final per unit export subsidy 1996; Schluep and de Gorter 2001). Such a scheme resulting from the proposed volume and value is similar to a taxpayer-financed export subsidy in reductions for each commodity sector. that it increases supply and decreases domestic Other relationships between value and volume demand alike, but the consumer-financed export reductions can also be established. For a larger ini- subsidy distorts trade more for a given domestic tial per unit export subsidy, other factors being price (Schluep and de Gorter 2001). Both a equal, volume reduction commitments are more taxpayer- and a consumer-only-financed export effective, and for larger initial quantities exported, subsidy scheme require import tariffs or nontariff again other things being equal, value reductions are barriers in order to be effective, since both raise the more likely to bind. This means that with more domestic price above the world price of the prod- elastic trade curves, when other factors are con- uct. To the extent that such import barriers are stant, value reduction commitments tend to be reduced under the market access part of the agree- more effective. ment, it may not be of paramount importance to Volume reduction commitments are always more recognize consumer-only-financed export subsi- effective than expenditure reduction commitments in dies as a separate issue. However, a number of the following sense: a less-than-100-percent reduc- taxpayer-financed export schemes are not covered tion in the volume of subsidized exports is needed to in the URAA at all. Examples include the reach a "free trade" condition (in which the country Canada­U.S. wheat agreement (according to which 56 Agriculture and the WTO BOX 3.1 Two Examples of Consumer-Only-Financed Export Subsidy Schemes The U.S. Federal Milk Marketing Order is an equalize and consumer fluid prices are higher in example of a consumer-financed export subsidy high-production-cost regions (with lower scheme because of higher domestic prices and weights for low class prices in the pool price), price pooling (Sumner 1996). In addition, an inefficiency in milk production does put some- implicit export subsidy prevails when one or what of a brake on total U.S. milk production more dairy products are sold in the domestic expansion arising from pooling. market at world prices. Supply increases to the By contrast, New Zealand dairy policy under export market for those products (sold in the control of the producer-controlled dairy domestic markets at world prices as well) for two board has price discrimination among multiple reasons: consumption of higher-priced products markets for many products in the world market. is reduced and pooling causes production to Pooling these revenues has the effect of a pro- expand. Exports are subsidized even if export duction subsidy (Schluep and de Gorter 2001). revenues for some products are not pooled to Price discrimination in world markets leads to farmers (but domestic sales at the world price higher prices for consumers worldwide and are). Schluep and de Gorter (2001) provide con- therefore it lowers consumption and trade. But ditions when such a scheme is a full-fledged prices in the more elastic demand markets-- export subsidy and when it approaches a pro- which are the markets in which New Zealand duction subsidy. competes with other exporters--are lower than The Federal Order system does not use they otherwise would be. The producer board explicit production controls, but regional price also pools downstream profits, domestic price differentials for fluid milk have a similar effect. premia, and import quota rents, resulting in an Because fluid milk prices are not allowed to expansion in farm milk supply. Canada "free-rides" on the U.S.-taxpayer-financed Producer-Financed Export export subsidy--see Peterson, Minten, and de Subsidies Gorter 1999) and fluid milk premia in several Mandatory or government-regulated producer- countries (see box 3.1). financed export subsidies are subject to reduction Domestic price discrimination for milk used in commitments. A producer-financed export subsidy different products in domestic markets alone acts as cannot exist on its own, without a taxpayer- and/or a consumer tax. Price discrimination between a consumer-only-financed export subsidy. Introduc- nontraded product such as fluid milk and traded ing a producer levy when a taxpayer-financed products such as butter, and the pooling of revenues export subsidy is already in place decreases the (so that the producer is paid a price that is an aver- price to both farmers and consumers. If the levy age of the two) also constitutes an export subsidy. maintains net price (and producer welfare), the Price discrimination in world markets combined price to consumers increases, as do the tax costs of with revenue pooling in the domestic market (such the program. A producer levy to finance part of the as the New Zealand Dairy Board's practices) acts as costs of a taxpayer-financed export subsidy pro- a production subsidy. In the presence of an export gram can reduce tax costs only if producer welfare subsidy, the degree of trade distortion from a pro- declines. duction tax differs, depending on whether the sub- The situation differs for a producer levy to par- sidy is taxpayer financed, or consumer-only (with tially or fully finance exports under a non-taxpayer- the latter always distorting trade). For this reason, or financed export subsidy scheme. Indeed, it turns out perhaps because of their effects, export subsidies that a producer levy with a consumer-financed have not been well understood, and consumer- export subsidy scheme results in an identical out- funded export subsidies are inadequately dealt with come as a consumer-financed export subsidy in in the URAA, the Agreement on Subsidies and terms of market effects, holding producer welfare Countervailing Measures, and Article XVI of the constant between the two scenarios (Schluep 1999; GATT 1994. Export Competition Policies 57 Schluep and de Gorter 2001). A levy imposed on a The subsidy rates for existing export credit consumer-only-financed export subsidy results in schemes are estimated to be small. An OECD study the same average revenue (pooled domestic and shows that the average subsidy for export credits export sales) curve because export revenues are not was 3.6 percent of export value in 1998, with the pooled at the world price anymore, but at some U.S. having the highest at 6.6 percent. These low higher price, which is used to justify the use of pro- subsidies notwithstanding, disciplines on the use of ducer levies in the first place. subsidized export credits and insurance need to be negotiated and enforced to ensure that they are not increased in the future. Other Issues Export Credit Programs Public Stock Disposal Export credit programs totaled US$7.9 billion in The Australian Bureau of Agricultural and Resource 1998 in the form of credit guarantees, interest rate Economics (ABARE) shows the direct correlation of subsidies, public assumption of risk, and insurance export subsidies and the rundown of public stocks. subsidies (OECD 2000). These programs are There is a large literature on government stock dis- designed to assist export sales by providing credit and posal procedures: some economists argue that such reducing risk associated with purchase and subse- schemes can decrease domestic prices and hence not quent default by the borrower. The export subsidy act as an export subsidy (Houck 1986; Chambers component is the taxpayer outlays for expected and Paarlberg 1991; Anania, Bohman, and Carter defaults, which can have major impacts on exports 1992). However, Peterson, Minten, and de Gorter because of the risk reduction effects. Export credits (1999) show that these traditional models of export can expand demand for exports, because they help to subsidy programs focus only on the effects of dis- alleviate financial constraints in the importing coun- posing public stocks on the world market. The con- try, and reduce the adverse effects of exchange rate clusions regarding economic effects of these export fluctuations during the length of the contracts. Use of subsidy programs are significantly different when export credits has increased for agricultural products, costs of acquiring these stocks are included. Stock but agricultural export credit programs are not cov- acquisition always causes the domestic price to rise. ered by the international agreement that regulates When stock acquisition costs are included in the export credits for manufactured goods. analysis, an export subsidy scheme can be shown to The export subsidy component of export credit be equivalent to cash export subsidies. guarantees is very complex to measure. Subsidized risk premiums need to be converted into a present Food Aid value of expected costs of defaults to be borne by taxpayers. It is possible that reducing the risk and Food aid can have market effects similar to those of enhancing purchasing power of importers may cor- an export subsidy. However, food aid is not included rect for a market failure that temporarily disrupts in reduction commitments on export subsidies. normal financial mechanisms. For example, export Food aid that displaces commercial exports (for credit guarantees are useful for food security pro- example, it is resold on the domestic market of the grams in importing countries suffering from finan- recipient country) depresses world prices while cial crises or food supply disruptions. emergency food aid that increases total world con- The same reasoning is valid for export insur- sumption (that is, given only to those who would ance. When the government or private agents offer not otherwise have purchased it) has the opposite export insurance at levels and rates that are not effect. Food aid and export promotion programs standard in the domestic markets of the countries have complex effects, which makes it difficult in a involved, this creates an incentive for exports and practical sense to determine whether they constitute may have the effect of decreasing the product's an export subsidy. Therefore, it is important to price to the buyer because it would reduce trans- determine when food aid is a result of commercial portation costs and risk. export diversion by the exporting country, as the 58 Agriculture and the WTO resulting effect may be trade distorting but not cate- government assistance during financial difficulties. gorized as an export subsidy. Specific disciplines should apply to all enterprises, On the other hand, when food aid is used to move public and private. surplus production (in other words, as a surplus dis- Other issues relating to exporting STEs are posal tool), it may be used to support prices in the discrimination among destinations ("pricing-to- exporting country, creating the prerequisite for an market"), and price pooling between domestic and export subsidy: a price gap with reduced domestic export sales (possibly leading to consumer-only- demand and increased domestic supply. Therefore, financed export subsidies dealt with above). Pricing- it has the same effect as a cash export subsidy in to-market is a very difficult issue both to analyze and depressing and destabilizing world prices, hurting to discipline because it often covers normal com- farmers and thereby reducing food production. If mercial practices related to transport costs, quality the donated commodity is resold in the domestic differences, and other similar factors. However, pric- market of the recipient country, its world price will ing-to-market can also function as predatory pric- only be affected if it displaces a significant volume of ing. In principle, this should be subject to discipline, potential imports (which depends, among other although it may be difficult in practice to agree on things, on whether the country would otherwise workable mechanisms for bringing that about. have had the possibility to pay for the commodity in a commercial transaction). If the commodity is re- exported from the recipient country, the trade dis- Options for Strengthening Disciplines on Export Subsidies tortion is more evident (Gardner 1996). in the Negotiations Although food aid can alleviate hunger in emer- gencies, a large proportion of it is currently being The successful conclusion of the Doha Develop- provided to support farm prices and dispose of sur- ment Round will depend on a complex mix of pluses (ABARE 2001). Alternative methods of assis- politics and economics, but the prospects for suc- tance can be adopted to help the poor countries, cess can certainly be enhanced by a close retrospec- including cash aid. tive look at what went right and wrong in the implementation of the URAA with respect to disci- plines on export subsidies. The analysis above finds Activities of State Trading Enterprises (STEs) the agreement's intent was undermined during the The activities of state trading enterprises/agencies implementation period (1995­2001 for developed were not disciplined in the URAA except for an countries), as countries took advantage of a number "understanding" on the interpretation of Article of omissions and loopholes. This has implications XVII of the GATT 1994. Progress on improving for the current negotiations, since an objective is the efficiency of international trade may result if "reductions of, with a view to phasing out, all forms a country with such entities agrees or is required of export subsidies." under proposed rules to open domestic markets for Some of the factors affecting the effectiveness of private trade, to create more competition for state the value and volume commitments in the URAA entities and thereby reduce their ability to control originated from the excessive flexibility allowed in exports. Otherwise the trading activity of state choosing some key parameters. Countries were trading entities needs to be subject to disciplines or allowed their choice of base period from which rules directly, because in many cases they have a reduction commitments would be measured (front- monopoly when buying commodities for export, loading), an option that allowed them to postpone benefit from government guarantees, and do not reductions. They were also given the option of have commercial objectives. However, Veemen, Ful- "banking" unused allowances in one period, to use ton, and Larue (1999) make a very strong argument them in a later period. This enabled an increase in that STEs can, in some situations, improve market the levels (value) of export subsidies in some peri- competition. They argue that STEs often operate ods during the overall implementation period. In on a commercial basis while private companies can addition, the per unit export subsidy was not regu- also have monopoly power, use the commercial lated so the reduction commitments did not limit practice of differential pricing, and may receive distortions in prices. This is because the same value Export Competition Policies 59 of subsidy can lead to different per unit subsidies, some kind of means testing), a formula should be and different impacts on export supply and there- developed for estimating the subsidy element of fore on world prices. The data on export subsidy food aid. This could be counted against the equivalents confirm that these differences are signif- export subsidy limits for purposes of meeting icant across countries, commodities, and over time. reduction commitments. Efforts could be made Export subsidies take a wide variety of direct and to encourage the use of cash aid in place of food indirect forms. Agreement on specific limitations aid to minimize adverse effects on markets. For will be difficult to reach because of the variety of these rules to produce their desired results, coun- institutions and programs involved. Notwithstand- tries would also have to untie their aid (that is, ing the difficulty, the analysis in this chapter indi- refrain from imposing requirements that the cates that meeting the following targets would help food be purchased from the country giving the the Doha Round Agreement to more fully meet the aid)--otherwise they would have no effect. goal of minimizing the unfavorable effects of export subsidies on the world trading system and on devel- Improving Rules on Circumvention oping countries in particular: The current rules on export subsidies can be · Establishing a firm deadline for eliminating strengthened to deal with the key problems and export subsidies. make the reduction commitments more effective by: · Establishing modalities to ensure significant reductions in MFN tariffs (since reductions in a. Banning the banking (or "rollover" to subse- import barriers will automatically strengthen dis- quent years) of export subsidies and agreeing on ciplines on export subsidies). annual targets for reductions. Without such lim- · Introducing ad valorem (percentage) limits on its, countries have excessive flexibility in meeting the per unit subsidy on a commodity-by- their commitments, which can lead to destabi- commodity basis (as a percentage of the world lization of world markets through higher subsi- price), combined with a ceiling on the total dization when world prices are low. value of exports that may be subsidized. The ad b. Making all commitments on a per product basis, valorem limit would place a constraint on the with a uniform agreed system of classifying difference between domestic and world prices, products, to ensure that countries cannot use and limit the distortions of relative prices across "tailor made" product aggregation in the base- different products, while the value limit would line as a means to circumvent commitments. constrain the impact of export subsidies on The current rules attempt to do this, in that world markets. The value limit became less products are aggregated into 23 commodity binding during the Uruguay Round implemen- groups (22 groups plus "other agricultural prod- tation period, so new commitments need to ucts"). However, the aggregation differs across "squeeze out some of the water" in the current countries, as well as the units reported. The level of commitments (in other words, reduce aggregation of different products into the same the commitment levels to make them binding) group (as "fruits and vegetables," for example) as a minimal starting point. poses problems when comparing figures across · Strengthening the monitoring of export subsi- countries that may refer to completely different dies by coordinating data collection with other products. The units used to report volumes for organizations (for example, the OECD), and meats can be carcass weight or product weight, possibly the WTO Trade Policy Review, provid- for instance, and for eggs they can be units, ing an annual evaluation of the effects of export dozens, tons, or tons of shell-equivalents. Such subsidies, focusing in particular on developing nonstandardization opens possibilities for countries. circumvention. · Improving the rules governing food aid. With the c. Banning front-loading (the practice of using dif- potential adverse impacts on domestic producers ferent bases from which export subsidy reduc- from food aid that is not targeted (WTO 2002) tions are measured), used in the Uruguay Round ("targeted" aid is only available to the poor via Agreements to postpone reductions. 60 Agriculture and the WTO d. Subjecting all expenditures involving direct in- cent reduction to reach this point, since any subsidy kind disposal of public stocks in export markets (without a volume constraint) would produce an to the same rules as export subsidies.6 equilibrium level of exports higher than Qft. Assuming that both volume and value reduc- Some holes left by the URAA need to be plugged tions are equally binding after the reductions have by: occurred, for exogenous shifts in excess supply and excess demand, volume constraints are always a. Establishing rules that directly constrain export binding in the case of a positive shift (that is, a shift credit and payment guarantees and direct financ- that moves the free trade equilibrium to the right, ing, as these have the effect of subsidizing exports. or alternatively stated, a shift that reduces the per Such rules are needed as part of the disciplines on unit subsidy necessary to achieve a given level of export subsidies. A methodology could be devel- exports), and expenditure constraints are always oped to include the export subsidy component of binding in the case of a negative shift. these measures, and this could be counted against Consider first a positive shift. In figure 3.A.1, reduction commitments. suppose that originally both volume constraints b. Establishing stronger disciplines on STEs by at are binding; that is, the country's commitments do a minimum subjecting them to the same rules not allow it to increase export volume above Q0, on export subsidies that apply to private sector nor its export subsidy expenditure above (Pd A B 0 enterprises, using notification and trans- Pw ). Now suppose that the excess supply curve 0 parency requirements to prevent disguised shifts outward from ES0 to ES1. If there was no export subsidies. An option would be to constraint on the volume of exports subsidized, strengthen and extend the disciplines of GATT but there was a constraint on expenditure, the gov- Article XVII and Article II on state trading ernment could increase exports, while lowering or imports to include limits on the ad valorem holding constant its expenditure. For example, it subsidies. could reduce the per unit subsidy such that the c. Establishing rules to constrain consumer- new expenditure level (Pd D E Pw ) would be the 2 2 financed export subsidies (for example, revenue- same as the previous level, while increasing the pooling arrangements). exports subsidized to Q1, causing the world price to decline to Pw . In contrast, if there was a con- 2 straint on volume subsidized, the government Chapter Annex: Volume versus could not affect the world price (since it could not Expenditure Limits on Export increase the supply of the product on world mar- Subsidies kets), regardless of whether there is an expenditure In a static analysis, excess supply and demand curves constraint or not. It would, in fact, have to lower do not shift over time. As figure 3.A.1 depicts, base- the subsidy to (Pd ­ Pw ) to keep exports from 1 0 line exports of Q0, which receive a per unit export exceeding their limit. In this case with a positive subsidy of Pd ­ Pw , represent the price gap 0 0 shift of the free-trade equilibrium, the volume between domestic and world prices. Assume that in constraint is necessary and sufficient to prevent an the baseline, both the volume of subsidized exports increase in subsidized exports and a consequent (Q0) and the value of export subsidy expenditures adverse impact on world prices, whereas an expen- (represented by the area Pd A B Pw ) are binding. 0 0 diture constraint alone would not be effective in Note that a reduction in the quantity of subsidized meeting this objective. The expenditure limitation exports to Qft (possibly a rather small reduction is not binding. That is, there can be water in this and certainly less than 100 percent) will result in a limit. free-trade equilibrium, in which the "price gap" is Next consider a negative shift, in which the excess zero and both the world price and the quantity supply curve shifts inward to ES2 in figure 3.A.2. exported are the same as if no export subsidies were With no constraint on expenditure, but with a con- offered. If the reduction commitment was made in straint on volume subsidized, the government terms of expenditure, it would require a 100 per- could increase the per unit subsidy to (Pd ­ Pw ), 3 0 Export Competition Policies 61 FIGURE 3.A.1 Excess Supply Curve Shifts Outward from ES0 to ES1 ES0 ES1 A Pd 0 Pd 2 D Pd 1 C Pw 0 B Pw 2 E ED 0 Qft Q0 Q1 Imports Note: (P0d -- P0w) < (P2d -- P2w) FIGURE 3.A.2 Excess Supply Curve Shifts Inward from ES0 to ES2 ES2 ES0 Pd 3 F Pd 4 G A Pd 0 Pw 4 H B Pw 0 ED 0 Q0 Imports maintain the same level of exports, and keep the allowing the world price to rise. In this case, it is the world price depressed. On the other hand, if there expenditure constraint that is necessary and suffi- was a constraint on expenditure, the government cient to enforce a reduction in subsidized exports would be forced to reduce the volume of exports and cause the world price to be higher than it other- that it subsidizes (such that the area Pd G H Pw 4 4 wise would be. Again, the volume limitation is not equals the original expenditure Pd A B Pw ), thereby 0 0 binding. There can be water in this limit. 62 Agriculture and the WTO Notes IATRC (International Agricultural Trade Research Consortium). 2001a. "Export Competition: Issues and Options in the Agri- 1. For more details, see the WTO Secretariat background cultural Negotiations." The International Agricultural Trade paper, "Export Subsidies," G/AG/NG/S/5, downloadable from Research Consortium Commissioned Paper 15, Washing- wto.org. ton, D.C. 2. The implication of this was that the country's budgetary ------. 2001b. "The Current WTO Agricultural Negotiations: outlays for export subsidies and the quantities benefiting from Options for Progress." The International Agricultural Trade such subsidies, at the end of the implementation period, should Research Consortium Commissioned Paper 18, Washing- be no greater than 64 percent and 79 percent of the 1986­90 ton, D.C. base period levels, respectively. For developing countries these Leetmaa, S. 1997. "EU Export Subsidy Commitments percentages were 76 and 86, respectively. Not Yet Binding, But Future Uncertain." Europe- 3. Not all the countries that have allowances use them. The International Agriculture and Trade Report, USDA, ERS, countries using subsidies and the commodity groups they subsi- WRS 97-5, Washington, D.C. dize are less than the total allowances of all countries for each Leetmaa, S., and K. Ackerman. 1998. "Export Subsidy Commit- commodity. Therefore, the percentage use of commitments ments: Few Are Binding, But Some Members Try to Evade would be different if calculated as a percentage of the commit- Them." Agriculture in the WTO, USDA, ERS, WRS-98-44. ments only for the countries and commodities that actually McCalla, A. 1999. Closing comments at the World Bank confer- make use of them. ence on "Agriculture and the New Trade Agenda from a 4. Eggs, and Fruits and Vegetables were excluded owing to Developmental Perspective: Interest and Options in the Next inconsistencies regarding the units reported in the notifications. WTO Negotiations." October 2, Geneva. Incorporated products are also excluded because no volumes are OECD (Organisation for Economic Co-operation and Develop- reported for this group. ment). 2000."An Analysis of Officially Supported Export Cred- 5. Taxpayer-financed export subsidies have consumer trans- its in Agriculture." (200)91/Final. OECD: Paris. fers as well because of the increase in domestic prices. Peterson, J. M., B. J. Minten, and H. de Gorter. 1999. "Economic 6. Economists have argued that such schemes can decrease Costs of the U.S. Wheat Export Enhancement Program: domestic prices and hence not act like an export subsidy, but Manna from Heaven or from Taxpayers?" International these types of analyses focus only on the effects of disposing pub- Agricultural Trade Research Consortium Working Paper 99- lic stocks on the world market and exclude acquisition costs. 2, Washington, D.C. Ruiz, L. 2000. "The Impacts of Export Subsidy Reduction Com- Select Bibliography mitments in the Agreement on Agriculture on International Trade: A General Assessment." M.S. thesis, August, Cornell ABARE (Australian Bureau of Agricultural and Resource Eco- University, Ithaca, New York. nomics). 2001. "Agricultural Export Measures in the WTO Ruiz, L., and H. de Gorter. 2001."The Impacts of Export Subsidy Negotiations." Research Report 01.12, Australian Bureau of Reduction Commitments in the Agreement on Agriculture Agricultural and Resource Economics, Canberra. on International Trade." In Lyn Kennedy, ed., Global Agricul- Abraham, F., I. Couwenberg, and G. Dewit. 1992. "Towards an ture in the New Millennium. Dordrecht, Netherlands: Kluwer EC Policy on Export Financing Subsidies: Lessons from the Academic Publishers. 1980s and Prospects for Future Reform." World Economy Schluep, I. 1999. "The Law and Economics of `Consumer Only' 15(3): 389­405. Financed Export Subsidies: A Context for the WTO Panel on Alston, J. M., and R. Gray. 1998."Export Subsidies and State Trad- Canadian Dairy Pricing Policy." M.S. thesis, Cornell Univer- ing: Theory and Application to Canadian Wheat." In T. sity, Ithaca, New York. Yildirim,A. Schmitz, and W. H. Furtan, eds., World Agricultural Schluep, I., and H. de Gorter. 2001. "The Definition of Export Trade. Westview Press: Boulder, Colo. Subsidies and the Agreement on Agriculture." In G. Peters, Anania, G., M. Bohman, and M. A. Carter. 1992. "U.S. Export ed., Tomorrow's Agriculture: Incentives, Institutions, Infra- Subsidies in Wheat: Strategic Trade Policy or Expensive Beg- structure and Innovations. Oxford and New York: Oxford gar-Thy-Neighbor Tactic." American Journal of Agricultural University Press. Economics 74(3): 534­45. Sumner, D. 1996. "The Role of Domestic Market Price Regula- Chambers, R. G., and R. L. Paarlberg. 1991. "Are More Exports tions in International Trade: The Case of Dairy Policy in the Always Better? Comparing Cash and In-Kind Export Subsi- U.S."Paper presented to the American Economic Association, dies." American Journal of Agricultural Economics 73: 142­154. January, San Francisco. de Gorter, H. 1999. "Market Access, Export Subsidies, and Veeman, M., M. Fulton, and B. Larue. 1999."International Trade Domestic Support Measures: Issues and Suggestions for New in Agricultural and Food Products: The Role of State Trad- Rules in the Agreement on Agriculture." In M. Ingco, ed., ing Enterprises." Agriculture and Agri-Food Canada Report Agriculture and the New Trade Agenda in the WTO 2000 #1998E, Ottawa. Negotiations. Forthcoming in 2003, Cambridge: Cambridge WTO (World Trade Organization). 1999­2000. "Country Noti- University Press. fications on Export Subsidies" G/AG/N. Office of the Secre- Gardner, B. L. 1996. "The Political Economy of U.S. Export Sub- tary General, Geneva. sidies for Wheat." In Anne Krueger, ed., The Political Econ- ------. 2002. "Submission of the World Bank, Response to the omy of American Trade Policy. Chicago: University of Proposal of the Least-Developed and Net Food-Importing Chicago Press. Countries." In Report of the Interagency Panel on Short- Houck, J. P. 1986. "The Basic Economics of an Export Bonus Term Difficulties in Financing Normal Levels of Commer- Scheme." North Central Journal of Agricultural Economics cial Imports of Basic Foodstuffs. WT/GC/62, G/AG/13. 8(2): 227­35. Geneva. 4 MARKET ACCESS: ECONOMICS AND THE EFFECTS OF POLICY INSTRUMENTS Harry de Gorter, Merlinda D. Ingco, and Laura Ignacio Introduction alization of agricultural trade. Bound tariffs are considered more transparent, allow trade to adjust This chapter aims to analyze and underline key points to changes in market conditions, limit the scope for for future action on the negotiating issues for market protection (including export subsidies), and can be access in agriculture in the context of the current disciplined through tariff reduction commitments. World Trade Organization's (WTO) Doha Develop- Hence, current negotiations can make decisive ment Round negotiations. Market access was at the progress for the future. core of past trade negotiating rounds, but agricul- The next tasks in the negotiations are to agree tural import barriers remain high with an average on formulas and other "modalities" for countries' global bound tariff of 62 percent in 2000 (the end of commitments. The purpose is to establish a fair and the implementation period of the Uruguay Round market-oriented trading system through a program Agreement on Agriculture [URAA]), and a large of fundamental reform that encompasses strength- variation of import protection rates among com- ened rules, and specific commitments on govern- modities and countries.1 The Agreement itself did ment support and protection for agriculture. In this not lead to much liberalization of agricultural way, restrictions and distortions in world agricul- trade,2 but it limits the type of border instruments tural markets can be disciplined and prevented. countries can use. Quantitative commitments have Developing countries as a whole have sizable also made market access issues more easily nego- potential gains from improved market access, but tiable in the current negotiations. The conversion of both the size and the distribution depend on the all nontariff measures into bound tariffs and the extent to which developing countries will be active introduction of access commitments in the form of in the negotiations. current and minimum access import quotas (as a Current negotiations should focus on creating share of domestic consumption) are two of the meaningful market access that would overcome the most important achievements of the URAA limitations of the URAA and this will require the (Tangermann 1996; IATRC 2001a, 2001b). Estab- identification of a set of feasible and effective lishing rules for acceptable agricultural policy meas- modalities. Market access issues that should be ures and requiring future mandatory reductions of addressed in the current negotiations and that are these measures was an important step for the liber- 63 64 Agriculture and the WTO particularly relevant for developing countries can iffs applied to imports above the quota). Another be categorized as follows: section provides empirical analysis of alternative modalities in liberalizing tariff quotas using the 1. Lowering tariffs, tariff escalation, effective pro- Policy Evaluation Matrix (PEM) model of the tection, tariff peaks, and tariff dispersion across Organisation for Economic Co-operation and commodities and countries. Development (OECD). The last section provides 2. Increasing tariff rate quotas. conclusions and policy recommendations. 3. Limiting the application of special safeguard tar- iffs, antidumping laws, countervailing duties, and other contingent measures so that they are Background: Evidence of Market Access Barriers in Agriculture not used to restrict market access unfairly. 4. Improving the rules on tariff rate quota admin- Barriers in market access were collapsed into two istration methods (including those concerning forms by the URAA: import tariffs and tariff rate state trading enterprises) to make them more quotas; and remaining contentious nontariff barri- transparent, to improve fill rates so that imports ers such as SPS standards. The tariffication process are not discouraged or blocked, and to ensure converted quantitative barriers into tariffs in the access by lowest-cost suppliers so that countries URAA and ensured access at bound tariffs for all are not discriminated against. commodities. Tariffs could be simple, specific, ad 5. Rationalizing product regulations (such as sani- valorem, or a combination. Each country was also tary and phytosanitary [SPS] rules, labeling able to designate TRQs for most commodities laws, and other nontrade concerns) so that they where quantitative barriers were tariffied. TRQs are cannot be formulated as nontariff barriers. commitments to allow access at least up to a quota 6. Developing rules for special and differential level at an in-quota tariff rate. The quota is either treatment measures for developing countries the level of imports in the 1986­88 base period or that provide greater benefits from a more liber- 3 percent of domestic consumption (increasing to alized trading environment and protection from 5 percent by the year 2000), whichever is larger. instability and food security concerns.3 Out-of-quota imports can occur but only at the (often much) higher out-of-quota tariffs that are The Plan of This Chapter bound in the agreement. The out-of-quota tariffs This chapter analyzes the first three issues identi- were reduced by an unweighted simple average of fied above. The fourth issue, on tariff rate quota 36 percent (minimum of 15 percent per tariff line) administration, is covered in the following chapter, by 2000­01 (and four years later for developing while the fifth and sixth issues are dealt with in countries). Requiring an average cut in tariffs other chapters on SPS (chapter 11), multifunction- rather than a cut in average tariffs meant that a ality (chapter 8). Concerns specific to developing country could make a large cut in a tariff that was countries are covered in later chapters too, such as already low (from 2 percent to 1 percent, represent- special and differential treatment (chapter 14), ing a 50 percent reduction) or in sectors with low managing low and volatile prices (chapter 10), and political sensitivity, while making only minimal food security concerns (chapter 9). The next sec- cuts in sensitive product categories. tion describes the issues and present empirical evi- Developing countries were not required to dence of tariff dispersion, tariff peaks, tariff water, undertake tariffication but instead could opt for preferential tariffs, and tariff rate quota fill rates. ceiling bindings. To address the concern of sudden This is followed by a section that provides an expla- import surges or depressed import prices, a special nation of the basic economics of tariff rate quotas safeguard for agriculture was introduced for those (TRQs), including the effects of water in the tariff who tariffied. However, the effectiveness of URAA (the difference between applied tariff rates and the market access commitments was compromised tariff equivalent of the binding quota4), binding because the out-of-quota tariffs were bound at such overhang (the difference between bound and high levels by both developed and developing applied tariffs), and over-quota versus out-of- countries as to effectively prevent all imports above quota import tariffs (the former has in-quota tar- the TRQ levels (ABARE 1999, Martin 2000). TRQs Market Access: Economics and the Effects of Policy Instruments 65 were sometimes understated to prevent a significant ward trend, however, implying a re-instrumentation increase in market access relative to the status quo toward domestic policy measures. During 1986­88, that prevailed before the 1994 agreement. On the the market price support from border protection other hand, countries with TRQs were allowed to accounted for 77 percent (61 percent) of the PSE follow significantly less restrictive import regimes (TSE), while by 2001 the shares had fallen to 63 per- than their commitment specifies by allowing over- cent (47 percent). quota imports. Applied in-quota tariffs (that is, the The shift away from border support in OECD actual tariff) may be less than the official in-quota countries is mainly concentrated in the grains and or out-of-quota tariff rates. oilseeds sectors (figure 4.2) with the overall rate of protection--support as a percent of farm rev- enues at world prices--dropping rapidly after Import Protection Levels 1986 and 1987 when world prices were low, and Agricultural trade regimes have changed in the has since stabilized. The overall percentage of pro- recent past. Developed countries have moved away tection started at close to 120 percent in 1986 and from domestic support and border protection to a fell to 50 percent in 1996; it stood at 77 percent as reliance on domestic support, while many develop- of 2001. The cyclical component of protection ing countries have reformed agricultural policies over time reflects changes in world prices more resulting in low rates of protection. Several middle- than changes in policy. Average protection for income countries, however, have provided more grains and oilseeds is 80 percent for the time protection in agriculture. period 1986­2001. The move away from MPS and The absolute level of support to agriculture in toward the various forms of direct payments, OECD countries has not changed very much since especially area payments and historical entitle- the conclusion of the Uruguay Round. Figure 4.1 ments, is also evident. Agricultural support, compares market price support (MPS)--that is from although still dominated by MPS, is declining, border protection--from 1986 to 2001 with both the with the growing components being area pay- total producer support estimate (PSE) and the total ments and historical entitlements. Purchased support estimate (TSE).5 The gap as a fraction of input subsidies and output price subsidies have MPS for either the TSE or PSE has shown a down- been fairly constant. FIGURE 4.1 Evolution of Border versus Total Support in OECD Agriculture (US$ Billions) 450 PSE MPS TSE 400 350 300 250 200 150 100 50 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 PSE 231 246 239 225 272 292 280 274 282 271 254 232 257 273 242 231 MPS 177 191 185 171 212 231 216 205 208 193 169 153 173 182 153 145 TSE 292 309 305 298 351 383 378 371 378 371 352 325 344 357 321 311 PSE: Producer support estimate, MPS: Market price support, TSE: Total support estimate. Source: OECD PSE/CSE Database 2002. 66 Agriculture and the WTO FIGURE 4.2 OECD: Grains and Oilseeds Input subsidies 120 Historical entitlements Area payments Output subsidies 100 MPS (border) trend-total average 80 (percent) trend-MPS 60 Protection 40 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Source: OECD database. "Protection" is the value of support over the value of production at world prices. FIGURE 4.3 OECD: Other Commodities Input subsidies Historical entitlements Area payments 120 Output subsidies MPS (border) average 100 trend-total 80 (percent) trend-MPS 60 Protection 40 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Source: OECD database. "Protection" is the value of support over the value of production at world prices. The rates of protection for all other commodities, tion period and therefore poor performance (low however, have a far less pronounced downward prices and export earnings) of global agricultural trend, although the average rate of protection is markets should not be blamed on trade liberaliza- lower at about 50 percent (figure 4.3). The value of tion because very little has transpired. Even though preferential tariffs to developing countries has not rates of protection are trending downward, mostly in eroded in this instance, but will if the trend away grains and oilseeds, in OECD countries the number from MPS, seen in the grains and oilseeds sectors, of farms has declined even faster and consequently spreads to other sectors. The significance of the fore- per farm subsidies have increased (Messerlin 2002). going discussion is that very little trade liberalization The distribution of bound "most favored nation" seems to have occurred in the URAA implementa- (MFN) tariffs across regions is summarized in fig- Market Access: Economics and the Effects of Policy Instruments 67 FIGURE 4.4 Average Agricultural Bound Tariffs by Region 2001 (in percent) 120 100 80 Average Agricultural Tariff (62%) 60 40 20 113 104 86 75 71 54 49 48 39 39 34 30 25 0 South Non-EU Caribbean Sub- North Central Eastern Middle Southern South Asia- European North Asia West Islands Saharan Africa America Europe East Africa America Pacific Union-15 America Europe Africa Source: Gibson and others 2001. FIGURE 4.5 Applied versus Bound Tariffs in Developing Countries (Tariff Binding Overhang) Argentina Brazil Colombia Ecuador El Salvador Bound Tariff Applied Tariff Tariff Overhang Guatemala Mexico Nicaragua Pakistan Panama Paraguay Philippines Tunisia Uruguay Venezuela, RB 0 20 40 60 80 100 120 Venezuela, Uruguay Tunisia Philippines Paraguay Panama Pakistan Nicaragua Mexico Guatemala El Salvador Ecuador Colombia Brazil Argentina RB Overhang 37 19 75 15 25 31 58 50 23 38 28 12 72 24 22 Applied 15 13 35 19 10 12 43 11 20 11 13 14 15 13 13 Bound 52 32 110 34 35 43 101 61 43 49 41 26 87 37 35 Source: Authors' calculations from ATPSM, OECD, TRAINS, and AMAD databases. ure 4.4.6 Regional average tariffs for WTO members protection because these "bound" tariffs are much range from an ad valorem equivalent of 25 percent to higher than "applied" tariffs ("tariff binding over- 113 percent. With the exception of Non­European hang") and fewer TRQs are used. Figure 4.5 gives the Union Western Europe, the regional groups with the level of tariff binding overhang for a selected group of highest average tariffs are developing countries.How- developing countries. The average tariff for agricul- ever, this comparison is not indicative of their actual tural products in developing countries declined from 68 Agriculture and the WTO FIGURE 4.6 Applied versus Bound Tariffs in Developed Countries (Tariff Binding Overhang) Turkey Norway Japan Mexico Canada U.S. European Union New Zealand Australia (50) 0 50 100 150 200 250 300 New European Australia U.S. Canada Mexico Japan Norway Turkey Zealand Union Applied Tariff 0.2 2.0 77.0 11.7 36.4 29.3 111.8 170.3 71.0 Bound Tariff 1.1 4.4 73.5 14.8 43.8 70.5 152.7 270.8 154.1 Tariff Overhang 0.9 2.3 (3.5) 3.1 7.4 41.2 40.9 100.5 83.1 Note: EU estimated overhang is ­3.5 because although a large majority of EU tariff lines have zero overhang, special circumstances cause some applied tariffs to exceed the bound rates, and there may also be some errors in the reported data. Source: Authors' calculations from ATPSM, OECD, and AMAD databases. 29.6 percent in 1990 to 22.9 percent in 1995 and was protection rates above the average of about 50 per- 18.4 percent in 2000--higher than 11.4 percent tar- cent. On the commodity side, only milk, sugar, and iff on manufactures in 2002 (GEP 2004, chapter 3). rice have protection rates above the average. Although far less common, developed countries Tables 4.1 and 4.2 provide data for 1999­2001 also exercise tariff overhang (figure 4.6). Significant on the level of producer support for 29 OECD gaps in bound versus applied tariffs must be taken countries and 10 commodity groups included in into consideration when assessing the effects of the OECD database. Table 4.1 column 1(a) shows trade liberalization. that total consumer transfers and taxpayer expen- Figure 4.7 compares average tariffs by commod- ditures on agriculture averaged US$329 billion ity. Of the 47 commodity aggregates shown, aver- from 1999 to 2001, mostly in rich countries.7 Col- age tariffs on 12 of the groups are above the global umn 1(c) shows that transfers to producers in agri- agricultural tariff average of 62 percent. These culture for a subset of countries, policies, and commodities include tobacco, dairy, meats, sugar, commodity sectors averaged US$126 billion.8 Col- and grains. Several sectors in OECD countries have umn 2 provides estimates of total transfers to farm- very high tariffs, raising tariff dispersion (discussed ers from trade barriers, with a breakdown of below). The incidence of non­ad valorem tariffs import quota rents and tariff revenues. Annual and the implications for measuring protection and average quota rents and tariff revenue column 2(b) transparency is discussed in box 4.1. are in the order of US$49 billion. Column 3 at Figures 4.8 and 4.9 present the nominal protection US$25 billion gives an estimate of transfers derived coefficient (by country and commodity, respectively, from domestic agricultural policy and indicates in the OECD countries and commodity groupings). the significance of domestic support. Nontrade A value of 1.0 means no protection. The Republic of transfers represent about 20 percent of total or Korea, Norway, Iceland, Switzerland, and Japan have consumer/taxpayers transfer to farmers. The final Market Access: Economics and the Effects of Policy Instruments 69 FIGURE 4.7 Average Bound Tariffs by Commodity, 2001 Meat: frozen beef, pork, or poultry 91 Dairy 86 Meat: fresh beef, pork, or poultry 80 Sugar beet 79 Meat: prepared 79 Meat: fresh, or frozen other meat 75 Sweeteners 69 Tobacco: unmanufactured 69 Live animals 67 Grain products 66 Tobacco: products 65 Eggs 63 Grains 62 Vegetable oils 62 All commodities 59 Food preparations 58 Average Agricultural Vegetables: fresh 58 Tariff is 62% Vegetables: frozen 57 Vegetable: preparations 56 Starches 55 Feed 55 Vegetable juice: tomato 54 Vegetables: frozen or prepared (other) 53 Horticulture: cut flowers and foliage 53 Fats and oils 52 Sugar cane 52 Vegetables: dried and fresh roots and tubers 51 Fruit juice 51 Cocoa beans and products 51 Fruit: preparations 50 Coffee: other 50 Coffee 50 Vegetables: dried 49 Nuts and fruit: dried, fresh, and prepared 49 Oilcake 49 Fruit: fresh 49 Oilseeds 48 Fruit: frozen 48 Essential oils 48 Tea and tea extracts 46 Nuts 45 Fruit: dried (raisins) 45 Skins and hides 43 Fruit: dried & fresh (coconuts, dates and figs) 43 Fiber 42 Horticulture: live 40 Spices 40 0 10 20 30 40 50 60 70 80 90 10 Source: Gibson and others 2001. column shows the proportion of transfers that are trade-distorting support as well as high levels of self-defeating because world prices decline due to the agricultural imports. country's own policies and the policies of other Table 4.2 presents the data by commodity group. countries. Much of these measured transfers simply Surprisingly, milk and sugar, among others, have offset the deleterious effects of the policies them- trade transfers that are less than nontrade transfers. selves, and the remaining transfers that appear to go Most of the wheat and oilseed transfers are offset by to farmers may be at least partially dissipated the reduction in world prices caused by wheat and because of economic inefficiencies and rent-seeking oilseed policies themselves, while sugar and rice trans- behavior by market participants. This occurs less in fers are mostly real increases in transfers to farmers Japan and the Republic of Korea because these two because the latter two commodities have high levels countries have extraordinarily high levels of total of market price support and are heavily traded. 70 Agriculture and the WTO BOX 4.1 Transparency and Tariffication The objective of providing greater transparency tries, all of them middle income, have a higher of protection levels through tariffication has not proportion of tariff lines with non­ad valorem been fully realized. Many of the tariffs within rates. On the other hand, Japan has 15 per- agriculture are still specific, compound, or cent, Canada 28 percent, U.S. 43 percent, and mixed. For these tariffs it is almost impossible to the EU 44 percent of all tariff lines that are spe- estimate the real protection levels because the cific, compound, or mixed. level of protection will change both over time In the sample reporting ad valorem equiva- and with the relative price of imports. The pro- lents, the duties are much higher than the ad tection will increase as the prices of products valorem rates (see table below). This suggests decline in the world markets and will be higher that tariffs are seriously underestimated in cases for lower-priced products originating from the with large numbers of non­ad valorem tariffs. developing countries. Tariffs are also nontransparent because of The transparency of tariffs in agriculture in many bilateral agreements to fix import developing countries is significantly higher than prices--such as those concerning tomatoes those in industrial countries. Of the 24 devel- from Mexico into the U.S. and apples from the oping countries included in this sample, 14 have U.S. into Mexico. These agreements are a result no non­ad valorem; two countries have less than of antidumping or countervailing actions and 1 percent of their tariff lines; four have less than end up not being reported as official tariffs in 5 percent of their tariff lines; and only six coun- the WTO. Ad Valorem and Specific Duty Rates Average Ad Average Ad Valorem Percentage of Valorem Tariff Tariff Equivalent Non­Ad Valorem Lines Australia 1.2 5.0 1.0 U.S. 8.8 12.3 43.8 European Union 10.6 35.2 43.6 Jordan 22.5 58.0 0.9 Source: WTO Integrated Database as presented in GEP 2004 (chapter 3). Although tariff quotas were to increase market these tariffs, resulting in water in the tariff. Rents are access for commodities that previously faced equal to the in-quota imports (including any over- quantitative barriers, the high in-quota and quota imports) times the difference between either out-of-quota tariffs still remain for these com- the out-of-quota applied tariff or the tariff equiva- modities with quotas. Across all WTO members, lent of the binding quota (depending on which the average out-of-quota tariff is 123 percent, while regime--quota, out-of-quota tariff, or in-quota tar- the simple in-quota average tariff is 63 percent iff regime--is the effective constraint) and the in- (approximately equal to the 62 percent average quota tariff. bound MFN tariffs for all countries and agricul- Figures 4.10 and 4.11 calculate the water in the tural commodities including non-quota commodi- tariff using values of the applied tariffs, and the MPS ties, Gibson and others 2001). The average tariff as tariff equivalent protection for the product from difference of out-of-quota tariffs over in-quota tar- binding TRQs. Although the data are rough esti- iffs is 336 percent. mates of world price gaps and do not include all Out-of-quota tariffs are overstated because dur- countries or commodities that have TRQs, there is ing the Uruguay Round negotiations, countries evidence that water in the tariff is an issue that can were given considerable discretion in calculating dilute the effects of any tariff liberalization. Market Access: Economics and the Effects of Policy Instruments 71 FIGURE 4.8 Nominal Protection Coefficient by OECD Country Switzerland Iceland Norway Korea, Rep. of Japan EU Czech Republic OECD All Canada 1986­88 2000 Slovak Republic U.S. Turkey Hungary Australia New Zealand Poland Mexico 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 a N New Slovak Czech Korea, Switzer Mexico Poland Australia Hungary Turkey U.S. Canada OECDAll EU Japan Norway Iceland Zealand Republic Republic Rep. of land 2000 1.18 1.11 1.01 1.00 2.00 1.31 1.14 1.11 1.13 1.34 1.06 1.33 2.45 2.90 2.31 2.27 2.71 86­88 0.91 1.00 1.02 1.05 1.14 1.15 1.19 1.29 1.40 1.58 1.74 1.87 2.51 3.36 3.38 3.78 3.85 Source: OECD PSE/CSE Database 2002. FIGURE 4.9 Nominal Protection Coefficient by Commodity (OECD) Rice Milk Sugar Other grains Sheep meat Wheat All commodities Other commodities 1986­88 2000 Beef and veal Maize Oilseeds Poultry Pig meat Eggs Wool 0.0 1.0 2.0 3.0 4.0 5.0 6.0 Beef Other All Sheep Other Wool Eggs Pig meat Poultry Oilseeds Maize and commod commod Wheat Sugar Milk Rice meat Grains veal ities ities 2000 1.01 1.20 1.23 1.27 1.27 1.31 1.44 1.47 1.58 1.68 1.87 1.97 2.33 2.69 4.91 86­88 1.02 1.07 1.18 1.20 1.22 1.18 1.29 1.27 1.34 1.15 1.18 1.14 2.11 1.77 5.28 Source: OECD PSE/CSE Database 2002. 72 Agriculture and the WTO b price policies to transfers in world OECD 4 by owing 42.4% 54.6% 48.3% 44.8% 54.1% 20.0% 16.0% 53.4% 44.7% 51.0% 60.5% 48.4% 48.7% 54.3% 51.4% 39.0% the of Reduction atederc decreases coverage y 2002. 3 71.6 11.4 190.7 769.6 306.0 253.5 439.0 602.1 276.7 138.9 690.7 504.7 June Non-trade transfers OPS+IPS 5,156.0 2,951.9 12,817.7 25,180.6 15-countr the a Canada,, to Millions radeT (b) 2.6 0.0 6.2 revenues import transfers 62.1 15.1 15.5 62.5 39.7 508.2 161.6 595.8 Whistler US$ omfr 2,373.8 9,701.2 1,023.4 ariffT and 34,362.7 48,930.4 correspond Quota Barriers Data ransfersT Conference, MPS 0.0 (a) 72.9 81.7 326.8 384.3 499.4 746.7 TRC Agriculture, otalT 1,462.8 2,401.2 1,991.1 2,222.0 database. 42,852.0 25,520.4 10,187.4 12,152.0 IA 100,900.7 at orld Outlook W a in w/ presented (c) imports 11.4 . 263.7 398.4 690.2 220.6 2,232.3 2,840.3 1,101.6 1,023.4 2,681.8 2,726.7 only Policies support 48,008.0 28,472.3 10,440.8 24,969.7 results rade-distortingT gross 126,081.3 Agriculturals to imports model therein. OECD' Due IFPRI and gross covered 12 ransfersT and (b) 37.1 470.5 402.0 703.8 238.7 using ransfersT support AD 2,232.3 3,550.3 1,629.0 1,144.6 2,681.8 3,960.9 database otalT 48,008.0 28,472.3 10,440.8 24,969.7 of rade-distortingT 128,942.0 UNCT calulated commodities from are PSE/CSEs the average) 4 Estimates (a) (total 759.9 162.2 332.3 to and 1,375.9 5,231.5 1,080.3 6,999.1 2,489.4 1,934.3 5,047.3 9,649.5 taken OECD' TSE support) 64,777.5 21,488.8 95,455.2 112,629.9 329,413.1 1(c) 2002 from changes Empirical (1999­2001 of Report Union columns y price Republic y Rep. Republic in Estimates 4.1 Zealand way orld ce: Countries Data W Countr Australia Canada Czech European Hungar Japan Korea, Mexico New Nor Poland Slovak Switzerland urkeyT U.S. All a. b. Sour Monitoring ABLET Market Access: Economics and the Effects of Policy Instruments 73 b price policies transfers to OECD 4 in world by due 8.0% 2.4% 40.3% 43.1% 50.0% 72.0% 38.8% 60.2% 52.4% 90.0% the of Reduction created decreases coverage y 2002. Millions b June transfers US$ 3 15-countr 974.9 400.3 390.8 the OPS+IPS 3,318.6 4,903.0 4,301.1 4,502.2 1,345.9 2,522.8 2,521.0 Canada,, to Nontrade Whistler Agriculture, a correspond orld rents W (b) revenues import Data 462.7 388.1 in trade 4,389.1 2,523.4 3,207.2 1,692.6 1,777.0 1,599.7 6,446.6 ariffT and quota 26,443.8 Conference, omfr TRC database. IA barriers at Commodity ransfersT Outlook MPS by (a) 390.7 otalT 3,735.1 8,384.8 4,257.7 4,108.8 1,296.0 3,063.0 presented 16,108.3 36,334.6 23,221.6 . only results Policies Agriculturals to a imports model therein. w/ OECD' Due (b) imports gross IFPRI and 8,638.1 4,892.9 9,730.7 5,232.6 4,509.1 1,686.8 5,584.0 covered support 19,426.9 40,635.7 25,744.4 and rade-distortingT gross using ransfersT 12 ransfersT AD database of otalT UNCT calculated commodities are from PSE/CSEs the average) (a) 4 Estimates support 8,665.4 4,897.4 9,769.9 5,543.5 6,217.0 2,026.0 5,691.3 to and 19,736.8 40,642.6 25,744.4 taken OECD' rade-distortingT 1(b) 2002 from changes Empirical (1999­2001 Report veal columns price grains sugar in Estimates 4.2 and y meat meat orld ce: Data W Commodity Beef Coarse Milk Oilseeds Pig Poultr Refined Rice Sheep Wheat a. b. Sour Monitoring ABLET 74 Agriculture and the WTO FIGURE 4.10 Water in the Tariff--Applied Tariffs versus Tariff Equivalent of Binding TRQs for Selected OECD Countries (%) U.S. Czech Rep. Korea Rep. Euro Union Hungary Canada Mexico Switzerland Poland Japan Norway 0 50 100 150 200 250 300 350 Norway Japan Poland Switzerland Mexico Canada Hungary Euro Union Korea Rep. Czech Rep. U.S. AppliedTariff 309 76 81 159 78 79 55 87 41 28 4 Tariff water 209 76 76 62 52 46 45 32 22 15 4 Tariff equiv. 100 0 5 97 26 33 10 55 19 13 0 Tariff equiv. Tariff water AppliedTariff Source: Applied tariffs data are from UNCTAD's ATPSM database. Tariff equivalent tariffs of binding TRQs are from the OECD's PSE/CSE database 2002. Because of different data sources and an arbitrary selection of countries and commodities, the estimates of water in the tariff are very imprecise and are meant only for illustrative purposes. FIGURE 4.11 Water in the Tariff--Applied Tariffs versus Tariff Equivalent of Binding TRQs for Selected OECD Commodities (%) Sugar Oilseeds Rice Maize Barley Sorghum Wheat Milk, conc. Pig meat Milk, fresh Poultry Bovine meat Sheep meat 0 20 40 60 80 100 120 140 160 180 200 Sheep meat Bovine meat Poultry Milk, fresh Pig meat Milk, conc. Wheat Sorghum Barley Maize Rice Oilseeds Sugar AppliedTariff 191 160 138 166 96 127 103 108 91 62 87 29 135 Tariff water 121 115 97 88 87 71 69 60 55 35 34 29 15 Tariff equiv. 70 45 41 78 9 56 34 48 36 27 53 0 120 Tariff equiv. Tariff water AppliedTariff Source: Applied tariffs data are from UNCTAD's ATPSM database. Tariff equivalent tariffs of binding TRQs are from the OECD's PSE/CSE database 2002. Because of different data sources and an arbitrary selection of countries and commodities, the estimates of water in the tariff are very imprecise and are meant only for illustrative purposes. Tariff Peaks and Dispersion median and mean tariffs in rich countries, indicating mega-tariffs (tariffs over 100 percent) or tariff peaks, Data in table 4.3 demonstrate the high dispersion in accompanied by a wide dispersion in tariff rates. The tariff rates, with a large number of tariff peaks for United States (U.S.) has 24 mega-tariffs with the products of interest to developing countries (Cernat, mean tariff four times the median tariff. The Euro- Laird, and Turrini 2002). Gibson and others (2001) pean Union (EU) has 141 mega-tariffs and the mean show that there is a large difference between the Market Access: Economics and the Effects of Policy Instruments 75 TABLE 4.3 Examples of Tariff Peaks and Dispersion in Agriculure Standard MFN-weighted Maximum Domestic peaks International Product Deviation average tariff MFN tariff (percent) peaks (percent) Beef 16.16 12.89 41.35 52.11 29.58 Sheep meat 9.02 0.84 21.25 3.45 3.45 Poultry 33.33 8.16 134.30 2.52 2.52 Milk 56.33 22.70 140.00 17.78 17.78 Milk concentrates 105.02 19.59 308.50 22.15 22.15 Butter 100.54 249.97 336.25 32.47 19.48 Barley 41.73 22.12 101.50 11.43 11.43 Maize 13.19 3.99 50.00 4.00 4.00 Wheat 28.93 39.51 81.50 13.11 9.84 Banana 9.07 4.27 27.95 22.73 13.64 Citrus fruits 7.10 4.62 25.65 6.10 8.54 Other tropical fruits 8.57 10.68 33.25 14.86 8.11 Nontropical fruits 5.60 0.77 17.75 1.45 2.90 Chocolate 40.55 22.72 276.50 34.21 14.33 Tobacco 97.97 44.86 350.00 6.25 6.25 Cigarettes 10.78 2.67 30.00 4.17 4.17 Cigars 6.95 10.14 17.00 0.00 10.00 Other tobacco 115.49 168.57 350.00 16.46 17.72 products Tea 5.96 3.82 17.75 11.11 11.11 Oilseeds 24.84 9.56 171.00 1.02 1.02 Vegetable oils 4.99 1.40 19.95 3.74 1.15 Source: UNCTAD TRAINS by Cernat, Laird, and Turrini 2002. tariff is over two and one-half times the median tariff also the ones where the highest maximum tariffs are while Japan has 142 mega-tariffs with the mean tariff found (between 300 and 350 percent). almost six times the median tariff. These mega-tariffs In terms of frequency of tariff peaks across agri- differ across countries for the same commodity. cultural products (expressed as the percentage of tar- A uniform tariff structure is preferred by econo- iff lines affected by tariff peaks in the total number of mists and policymakers if maximizing the benefits of lines), the most affected sectors by domestic tariff trade liberalization is the goal because misallocation peaks are beef (more than 52 percent) and chocolate of resources (land, capital, labor, consumption, (more than 34 percent). The highest frequency of investment, and so on) is exacerbated with an international tariff peaks is also found in beef, fol- increasing degree of tariff dispersion. Tariff peaks lowed by dairy products (milk and butter).Very high increase the economic inefficiency stemming from weighted out-of-quota tariffs are applied to butter protection as they reduce competition and special- and other tobacco products, which also have record ization according to comparative advantage, and dis- maximum out-of-quota tariffs. Other products with tort consumption. Uniform tariff rates are also more high tariffs have lower weighted average, probably as transparent and easier to administer than nonuni- a result of a large value of trade in items that have form tariffs, and less likely to be determined by the zero or very low ad valorem tariff rates. In summary, relative political power of domestic industries, the data show that although average tariff rates in bribery of customs officials, and administrative inef- developed countries have been reduced to low levels, ficiency. Data presented in table 4.3 reveal the degree tariff peaks on products of interest to developing to which applied tariff peaks affect various agricul- countries are of concern. One way developing coun- tural products. The highest tariff dispersion was tries may gauge success in the Doha Round negotia- found in tobacco products, milk concentrates, and tions is if tariff peaks on products of interest are butter. The highest standard deviation products are reduced, thereby improving market access. 76 Agriculture and the WTO Tariff Escalation and it is not necessary to demonstrate that serious injury is being caused to the domestic industry. Average tariffs on processed agricultural products The special agricultural safeguards (SSG) can only are found to be consistently higher than average tar- be used on products that were tariffied--which iffs on raw or low levels of processing, although bulk amount to less than 20 percent of all agricultural agricultural products have higher tariff rates than products (as defined by "tariff lines"). They cannot intermediate products in the U.S. and the EU be used on imports within the tariff quotas, and they (Wainio, Gibson, and Whitley 2001). An increase in can only be used if the government reserved the right tariffs with each stage of processing is termed tariff to do so in its schedule of commitments on agricul- escalation and affects trade flows in a number ture. Most developing countries cannot use these of products of interest to developing countries. Tar- safeguards because few TRQs were adopted by them iff escalation biases exports toward unprocessed and most opted for tariff ceiling bindings rather than resource-based commodities, characterized by for tariffication. In practice, the special agricultural low value-added. This may cause difficulties for safeguard has been used in relatively few cases. Ten commodity-dependent developing countries in their WTO members have notified the WTO as required attempt to diversify their agricultural export base. A that, from 1995 to 2001, they have taken such action number of important commodities have been cate- under the SSG. Thirty-eight WTO members that gorized according to primary, intermediate, and have reserved the right to use a combined total of processed stages in production chains, and table 4.4 5,043 special safeguards on agricultural products are provides the tariff levels by product and by processing shown in table 4.5. stage in the Quad markets (U.S., EU, Japan, and Taking U.S. dairy as an example, two types Canada). Tariffs can escalate between raw and semi- of special safeguards apply--price-based and finished products and also between semifinished and quantity-based. Price-based special safeguards finished products. On average, the escalation in allow additional duties (over and above the out- Canada, Japan, and the EU is higher between raw and of-quota tariff rate) to be imposed on out-of- finished goods, while in the U.S. the average highest quota imports when prices fall below a fixed escalation is found between semifinished and fin- trigger price (based on average prices during ished goods. 1986­88), and are invoked automatically on a A more detailed analysis of tariff escalation, dis- shipment-by-shipment basis. Price-based special tinguishing between markets of developing and safeguards duties operate like a variable levy developed countries, shows that tariff escalation is scheme whereby duties increase as the value of not just a feature of developed markets but in fact is imports declines. According to WTO notifica- present (sometimes even more prominently) in tions, price-based special safeguards were invoked developing countries as well. Tariff escalation also on a significant volume of out-of-quota imports impedes South­South trade. during 1995­98 on butter and cheese. Quantity- based special safeguards allow additional duties to be imposed on out-of-quota imports if actual Special Safeguard Tariffs imports exceed a certain trigger level. The trigger Safeguards are contingent restrictions on imports import level for each product is based on imports adopted temporarily to deal with special circum- over the previous three years, so these are stances such as a sudden surge in imports (WTO announced annually and decline 15 percent dur- 2002c). They normally come under the WTO Safe- ing the URAA implementation period. guards Agreement, but the URAA has additional special provisions (Article 5) on safeguards that The Basic Economics of Barriers apply only to tariffied agricultural products. In to Market Access agriculture, unlike safeguards provisions that apply to manufactured products, higher safeguards duties Import barriers such as tariffs and quotas reduce can be triggered automatically when the import the level of imports, thereby increasing the price volume of the agricultural products rises above a in the importing country and reducing world certain level, or if prices fall below a certain level; prices to exporters and other importers. Treasuries Market Access: Economics and the Effects of Policy Instruments 77 7.70 8.00 0.00 8.54 12.95 13.31 13.07 11.65 19.15 24.81 10.47 11.79 Finished Union Canada) 5.16 1.10 8.63 8.02 8.58 4.55 0.00 6.93 14.64 11.56 and opean Semifinished Eur Japan, Raw 1.53 6.27 9.34 1.35 0.00 0.11 8.12 0.71 0.00 4.47 17.30 EU, (US, 3.38 1.96 7.48 3.12 5.35 3.95 6.98 0.00 11.56 12.49 36.93 Finished 350.00 Markets U.S. 6.15 1.88 5.82 4.32 1.83 0.07 6.07 2.27 0.25 4.17 13.00 Quad Semifinished the in Raw 0.60 2.82 0.15 0.87 0.37 2.94 0.61 0.00 35.42 68.26 11.20 centage) 0.07 0.00 10.66 17.39 11.58 15.40 20.79 20.02 17.92 16.51 19.47 13.62 Finished Per in ariffsT Japan 5.10 1.00 4.20 8.44 0.20 0.64 6.94 12.92 12.86 10.60 13.43 Semifinished Applied Raw 0.88 3.91 6.37 0.14 1.63 7.07 0.00 -- 0.00 0.00 5.77 18.77 25.50 MFN 2002. 9.00 0.01 5.76 4.43 5.14 3.16 8.17 7.90 0.26 6.88 verage Finished 18.83 13.05 urriniT A and eighted Canada 1.53 6.25 3.85 3.00 0.00 4.56 5.70 3.17 0.00 3.83 Laird, 10.25 (W Semifinished Cernat, by Raw 0.11 1.94 0.01 0.00 2.75 0.00 0.08 0.89 7.79 0.01 0.00 1.23 Escalation TRAINS ariffT AD and oils skins Group tea, food UNCT 4.4 products and products products products and food and y products ce: egg Product Meat Dair Fish Sugar Cereal egetableV spices vegetables Coffee, Fruits obaccoT Other Animal Hides verageA Sour ABLET 78 Agriculture and the WTO TABLE 4.5 Special Safeguards Tabled in the URAA Country/Entity Number Country/Entity Number Country/Entity Number Australia 10 Iceland 462 Phillipines 118 Barbados 37 Indonesia 13 Poland 144 Botswana 161 Israel 41 Romania 175 Bulgaria 21 Japan 121 Slovak Republic 114 Canada 150 Korea, Rep. of 111 South Africa 166 Colombia 56 Malaysia 72 Swaziland 166 Costa Rica 87 Mexico 293 Switzerland-Lctn 961 Czech Republic 236 Morocco 374 Thailand 52 Ecuador 7 Namibia 166 Tunisia 32 El Salvador 84 New Zealand 4 U.S. 189 European Union 539 Nicaragua 21 Uruguay 2 Guatemala 107 Norway 581 Venezuela, RB 76 Hungary 117 Panama 6 Source: WTO 2002c. accumulate tariff revenues and product quota total imports and the import quota is depicted as Q. holders (such as companies, state trading enter- Pd represents the resulting domestic price in each prises, and others) obtain excess profits ("rents"). regime. The three regimes are as follows: Farmers in the protected import market increase supply and consumers reduce their demand. The · The "in-quota tariff regime" where the lower opposite occurs in exporting and other importing in-quota tariff t1 is effective, quota rents and countries. Hence, import barriers reduce economic out-of-quota revenues are zero, and in-quota efficiency by distorting relative prices in both the tariff revenues are equal to t1 times imports importing and exporting countries, and also (figure 4.12a). impose an externality in the form of lower world · The "quota-binding regime" where the import prices that adversely affects all farmers outside the quota determines price, quota rents equal the protectionist importing countries. Inputs such as difference between the domestic and world price land prices are also adversely affected in other plus t1 times the quota, in-quota tariff revenues countries, along with foreign exchange earnings. equal t1 times the quota, and out-of-quota tariff The analysis of alternative trade liberalization revenues are zero (figure 4.12b). scenarios is somewhat more complex than simple · The "out-of-quota tariff regime" where the tariffs or a simple import quota in the case of tariff higher out-of-quota tariff t2 is effective, quota quota schemes. It is important to understand that rents equal the difference in the domestic and only one of the import tariffs or the quota can be world price times the quota, out-of-quota tariff effective in the sense that it controls or limits imports revenues equal the difference between domestic at any one time, rendering the other two policy and world price times the difference in the total instruments redundant. Hence, there are three possi- imports and the quota, and in-quota tariff rev- ble regimes, depending on the level of the quota for a enues equal t1 times the quota (figure 4.12c). given market situation and relative tariff levels. Fig- ure 4.12 summarizes these three regimes, where PW While only one policy instrument can be effec- is the world price and the importer is assumed to be tive at a time, a change in world prices, a change in able to import as much as it needs without changing the policy instruments, or a change in domestic the level of the world price (Moschini 1991; de market conditions can result in a switch in the effec- Gorter and others 2001). The import demand curve tive tariff or quota regime. However, a change in the is given by ED (the horizontal difference between the redundant policy instrument may have no impact domestic demand and supply curves). M denotes on the volume of trade or prices unless that change Market Access: Economics and the Effects of Policy Instruments 79 FIGURE 4.12 The Three TRQ Regimes a. In-Quota Tariff Regime Price PW + t2 PW + t1 Pd PW t1 tariff revenues ED 0 M Q Imports b. Quota regime Price PW + t2 Pd quota rents PW + t1 PW t1 tariff revenues ED 0 M = Q Imports c. Out-of-quota tariff regime Price PW + t2 Pd quota rents t2 tariff PW + t1 revenues PW t1 tariff revenues ED 0 Q M Imports Source: Authors. is sufficiently large to cause a switch in the effective "applied" tariffs, often significantly lower than regime (Boughner, de Gorter, and Sheldon 2000). bound rates, are especially prevalent in developing countries. Most developed countries offer lower tar- iffs for developing countries under "preferential" The Implications for the Trade Negotiations tariff schemes with rates below either the in-quota Other types of tariffs may be used by countries so or out-of-quota tariffs, and can be equal to zero. that there are several variations of the in-quota and The basic implications of bound versus applied out-of-quota tariffs described above in terms tariffs are shown in figure 4.13 where there are of "applied" versus "bound" and "preferential" (see over-quota imports at the applied tariff t. In-quota box 4.2 for further explanation). As described below, tariffs are not necessarily bound, therefore they are 80 Agriculture and the WTO BOX 4.2 More on Tariff Quotas Tariff quotas are even more complicated than in which the WTO reports quota under-fill but described in figures 4.12­4.14. Nonquota imports does not report out-of-quota imports at the out- often occur in the same product category. For of-quota tariff levels, which may equal or even example, 50 percent of U.S. cheese imports are exceed the quota under-fill reported. In these nonquota products that do not compete with cases, rents are foregone by quota holders domestic production. Hence, comparing data because their costs are too high and out-of- on total imports with the quota level will be mis- quota tariff revenues are collected by the leading. Governments also import beyond the importing country in their stead. Some countries quota but at the in-quota tariff level. The quota do not even report any imports beyond quota is not a maximum but a minimum access level. levels. Finally, some imports come in under pref- The WTO data do not report any over-quota erential tariffs. Hence, it is very important to imports, so a fill rate of 100 percent may mean understand the details surrounding the defini- that more than the quota was imported. tions of imports, quota under-fill, and over- Another type of misleading statistics is the case quota versus out-of-quota imports. referred to here as the "official t1," and an applied tariff binding overhang. A small change in the tariff can be either the in-quota or out-of-quota quota, however, will cause an increase in imports tariff. If the applied tariff is below the tariff equiva- and a reduction in domestic prices. The situation in lent of the binding quota teQ, then the applied tar- which the out-of-quota regime is effective initially iff is simply designated as t. The intersection of is given in figure 4.14. A reduction in t2 will have an the implied quota with the excess demand curve impact but a negotiated increase in the quota will ED of the importer determines the domestic price need to be very large before having an impact on PW + teQ (where PW is the world price and teQ is trade. The in-quota tariff is effective only if the the tariff equivalent of the binding quota with over- import quota is to the right of point a in the figure. quota imports). The level of imports above the This will generate a situation of quota under-fill quota but at an applied rate below teQ implies over- because of the relative level of the three policy quota imports. This situation is very common in instruments (especially the effective tariff t) or both rich and developing countries. Quota rents market conditions. can be substantial. To maximize trade liberalization effects, nego- A small or moderate reduction in the bound t2 tiators would identify which policy instrument of can have no effect on trade. There are two potential the three is effective to begin with and change the sources for this: the gap between bound and corresponding policy instrument. They could applied tariffs (tariff binding overhang) and the focus on reducing out-of-quota tariffs, in those gap between the domestic price and world price cases with out-of-quota imports or if the out- plus applied t2 (water in the tariff). Many tariff of-quota tariff t2 is close to the tariff equivalent of quotas operate in a quota-binding regime pro- the binding quota. If the tariff equivalent of the tected by very high levels of out-of-quota t2 tariffs. binding quota is far below t2, increasing the quota Under these circumstances, only a very large reduc- will have a greater chance of liberalizing trade in tion in t2 will generate a switch to an out-of-quota the short run. A reduction in t1 will liberalize trade regime. A small reduction in the out-of-quota tariff only if t2 is close to and below t1, in which case rate will have no impact. Only if t2 falls below the both tariffs need to be reduced, or if under-fill is price gap (the tariff equivalent when the quota is significant because t1 is effective--otherwise, quo- binding) will it have an impact on liberalizing trade tas will also have to be increased to obtain trade- rather than simply reducing water in the tariff or liberalizing effects. Because governments may try Market Access: Economics and the Effects of Policy Instruments 81 FIGURE 4.13 Definitions of Alternative Tariffs Price PW + bound t2 tariff binding overhang PW + applied t2 water in the tariff PW + teQ PW + official t1 quota rents Pd PW + applied t1 t1 tariff revenues PW quota ED overhang 0 Quota Implied Imports Quota Notes: The applied t2 can be below teQ or anywhere below bound t2 (if applied t2 < teQ, no water in the tariff exists but tariff binding overhang remains). There can be quota underfill and out-of-quota imports. Source: Authors. FIGURE 4.14 Out-of-Quota Imports with Quota Underfill and Out-of-Quota Imports Price Quota underfill a PW + applied t2 Quota rents t2 revenue PW + official t1 t1 revenue PW In-quota Out-of-quota imports ED imports 0 Quota Implied Imports Quota Source: Authors. to protect farmers, they should not be given the of liberalizing trade. An increase in world prices choice of a trade-liberalizing instrument because a can cause a switch to the t1 regime. A downward redundant policy instrument might be chosen, trend can put trade into a t2 regime (although tar- thereby thwarting the objective of increasing mar- iff water and overhang are so large in some cases ket access. Such an approach relies on information that this switch would be extremely unlikely). In about each commodity sector and country, and the either case, the tariff quota is no longer the effec- resulting tariff policy conclusions may rapidly tive way of liberalizing trade. become obsolete. A switch in regimes can occur A broader approach to tariffs and quotas that with changes in market conditions--for example, focuses on more macro indicators such as minimum in a case in which the quota is initially effective and quotas relative to size of market (for instance, con- analysis would therefore have determined that an sumption) and maximum prices as a percentage of increase in the tariff quota was the effective means world prices (tariff) may be desired. The WTO gives 82 Agriculture and the WTO primacy to tariffs over quotas as the more effective 3. Should a reduction formula be negotiated on a instrument because tariffs are transparent and product-by-product basis? One major short- because they are more flexible to changes in market coming of the 36 percent reduction rate agreed conditions and allow comparative advantage to pre- to in the URAA was that it was not a reduction vail with changing volumes of trade. The URAA did in the average tariff rate for each country, but an provide for a minimum quantity of access but at a average cut in tariffs. Hence, countries could very low level, except for current access. It gave each allocate that average to individual tariff lines, as country flexibility in setting out-of-quota tariffs; they liked (as long as they reduced no single tar- most countries therefore set t2 at such a high level as iff by less than 15 percent). As a result, many tar- to effectively perpetuate the quantitative restriction iff peaks remain, and tariff dispersion is in many on imports that quotas replaced. cases even more pronounced than at the begin- A very large increase in the import quota would ning of the implementation period.9 result in in-quota tariff regimes that often provide 4. What formula should be adopted to reduce aver- much lower levels of domestic protection. A very age tariffs across products and reduce tariff peaks large decrease in out-of-quota tariffs (comparable and dispersion? The Swiss formula has been used to commodities not protected by tariff quotas) for industrial products in previous rounds and has could also result in out-of-quota tariff regimes with been proposed for agriculture by the U.S. and the significantly lower levels of protection. The "Swiss Cairns Group.10 Because higher tariffs inevitably formula" approach used for industrial products in contain more water and binding overhang than the Tokyo Round caps out-of-quota tariffs and this lower tariffs, a formula that results in more uni- would be effective in reducing peak tariffs and tariff form tariffs is likely to be the most effective. dispersion in agriculture. Such an approach offers 5. How much flexibility on reducing tariffs should the prospect of a substantial increase in market be given to individual countries? An interesting access for agricultural commodities. variation on the Swiss formula is a "cocktail" approach by which countries are allowed to choose among (a) a flat rate reduction of all tar- An Empirical Analysis of Options iffs and (b) a formula cut that yields an agreed for Liberalizing Tariffs and Tariff but somewhat smaller average reduction of tar- Quotas in OECD Countries iffs, but attenuates tariff dispersion. A modified Because of water in the tariff and tariff binding Swiss formula developed by Francois and Mar- overhang, another round of tariff reductions is tin (2002) would be a disciplined method of needed to improve market access for agricultural giving countries more political flexibility yet products. In principle, there should be no reason still addressing tariff peaks and dispersion, as why the same overall rate of reduction could not well as tariff escalation. be achieved that was agreed to in the URAA, There are a variety of possible liberalization sce- namely a reduction by 36 percent and a minimum narios for tariffs, applied or bound (including the of 15 percent per tariff line (for developed coun- tariff equivalents of quota binding situations): tries) over a six-year implementation period (2001 to 2006), but with more effective modalities 1. Proportional cuts; than were used in the URAA. This approach raises 2. The Swiss formula; the following questions: 3. Continuation of the URAA formula; 4. Across-the-board tariff cuts; and 1. What should the starting point be for calculating 5. Reduction of bound tariffs to applied tariffs the reductions? Zero tariffs are never reached if (including tariff equivalents of binding quotas). percentage reductions are always made from existing tariffs. We analyze two empirical simulations here: 2. What tariff rates should be disciplined? Bound or applied tariffs? And what should be done · OECD data on market price support (MPS) that about tariff equivalents of binding quotas-- reflect effective average tariffs or tariff equiva- simply expand quotas? lents for each commodity group (in ad valorem Market Access: Economics and the Effects of Policy Instruments 83 equivalents) for 29 countries and 11 commodity Note that some tariffs are reduced by more than groups (see "initial ad valorem tariffs" in the first 25 percent when a = 0.25 (and some less than part of table 4.6). 25 percent) but the maximum tariff is less than · OECD's PEM model with a subset of the com- 25 percent for any T0 . The average tariff must modities (4) and countries (20) but where be less than a. If the average initial tariff is more the effects of a binding quota are taken into than 100 percent, the reduction will be greater account explicitly as well as world price changes than 50 percent in this example of a 0.50. with liberalization. For each Ti , the maximum tariff will be less than a. For a T0, there is a 50 percent reduc- Analyzing the Swiss Formula Approach to Tariff tion in tariffs. The country with very high tariffs Reductions will have the highest average tariff ex post (but still less than a). Note that specific tariffs cannot be dis- The Swiss formula (SW) is defined as follows: ciplined using the SW because the parameter a will T1/T0 a/(a + T0) vary depending on the units and therefore will vary across countries and commodities with currencies, where T1 is the new tariff, T0 is the initial tariff and units used, and so on. a is the maximum tariff that is determined in The Agreement on Agriculture formula (AA) is advance by the WTO agreement. as follows: There are several properties of the SW if a com- mon parameter a is negotiated: T1,1 - T0,1 T1,2 - T0,2 + + T1,j /N 0.36 T0,1 T0,2 T0,j - T0,j 1. There exists a common maximum tariff for all countries and commodities determined by the where j = 1 to N countries. value of the parameter a chosen by the world In this case, there is an average reduction in tar- body politic (WTO) where 0 a . iffs (not a reduction in average tariffs as in the SW 2. High tariffs are brought down faster than low approach). tariffs (in terms of percentage reduction in ad valorem tariffs) across commodities and across Analyzing Liberalization Scenarios Using countries. Data in table 4.6 show: the PEM Model a. A high-tariff country's reduction in average There are tariff rate quota regimes for wheat and tariffs is higher than that of a low-tariff coarse grains in Mexico, Switzerland, and Japan in country; the OECD's Policy Evaluation Model (PEM--see b. A high-tariff country's final average tariff is OECD 2001a, appendix A, for detailed descriptions higher than a low-tariff country's final aver- of the PEM). There is also a tariff quota regime for age tariff; rice in the EU and Japan.11 Table 4.7 shows the level c. A high-tariff country's average tariff reduc- and value of production as well as tariff-quota- tion is higher than a low-tariff country's related protection for grains and oilseeds in the six average tariff reduction; and combined regions in the PEM. Total world produc- d. A high-tariff country's average of tariff tion ranges from 236 million tonnes (mmt) for reductions is higher than a low-tariff coun- oilseeds to 904 mmt for coarse grains. The total try's average of tariff reductions. amount of grains and oilseeds import quota in all six regions is only 43 mmt, ranging from 0 percent 3. There cannot be a common reduction in average of production for oilseeds to 3 percent for coarse tariffs across commodities and countries grains. The total value of production of tariff- because of different initial tariffs but the final quota-related market price support due to import tariffs will have greatly reduced dispersions and barriers (MPS) is US$15.9 billion, which was about similar maximum tariffs. 7.3 percent of the value of production (net of 4. There cannot be a common average tariff for all market price support). Other market nontariff- countries but the differences ex post will be quota-related protection amounted to US$0.7 bil- greatly reduced. lion, which indicates that tariff quotas are the 84 Agriculture and the WTO 22% verageA 7.0% 5.8% 6.9% 7.0% 5.6% 5.7% 9.5% 59.1% 15.3% 65.8% 90.9% 74.8% 55.3% 56.0% 29.4% 13.5% 75.0% 66.5% 71.6% 12.6% 16.3% 14.3% 10.8% 10.5% 10.0% 11.6% 0.12% 173.9% 155.3% U.S. 8% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 9.9% 1.8% 0.0% 0.2% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 7.1% 1.7% 0.0% 0.2% 0.0% 3.5% 56.2% 93.4% 12.4% 17.3% 19.7% 0.48% n.a. n.a. n.a. urkeyT 4.5% 5.4% 5.6% 19% n.a. n.a. n.a. 3.8% 9.6% 4.4% 4.5% 15.5% 57.1% 29.0% 71.1% 18.1% 50.0% 40.0% 17.4% 21.3% 13.4% 18.5% 10.5% 16.7% 12.0% 0.41% 143.3% n.a. n.a. n.a. n.a. 48.2% 59.6% 98.4% 34.9% 166% 16.5% 17.6% 19.9% 23.3% 22.6% 21.5% 22.3% 22.9% 14.6% 23.6% 23.2% 20.7% 0.10% 332.5% 237.5% 153.8% 202.8% 267.0% 415.8% 318.6% 197.2% witzerlandS Rep n.a. n.a. n.a. 2% n.a. n.a. n.a. Slovak 0.0% 0.0% 0.0% 0.0% 4.1% 5.8% 0.0% 0.0% 8.1% 0.0% 0.0% 0.0% 0.0% 3.5% 4.7% 0.0% 0.0% 3.7% 38.9% 31.9% 15.2% 14.0% 0.36% n.a. n.a. n.a. n.a. Poland 0.0% 0.0% 9.7% 0.0% 0.0% 0.0% 11% 0.0% 0.0% 7.0% 0.0% 0.0% 0.0% 7.6% 20.4% 34.7% 94.6% 23.1% 73.3% 23.3% 11.2% 14.5% 19.8% 12.0% 18.6% 0.64% way n.a. n.a. n.a. n.a. 0.0% 27% n.a. n.a. n.a. n.a. 0.0% 89.9% 99.8% 93.3% 18.3% 91.6% 90.4% 20.8% 19.6% 20.0% 21.1% 19.7% 10.6% 21.7% 19.6% 17.0% 0.52% Nor 122.5% 136.6% 162.0% 25%) of New n.a. n.a. n.a. n.a. n.a. n.a. n.a. 4% n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0.0% 0.0% 0.0% 0.0% 8.1% 9.8% 0.0% 0.0% 0.0% 0.0% 6.1% 3.8% Zealand 50.6% 16.7% 0.46% tariff n.a. n.a. 7% n.a. n.a. 5.3% 2.8% 6.3% 0.0% 0.0% 4.3% 9.2% 2.5% 5.0% 0.0% 0.0% 9.1% Mexico 33.3% 50.2% 14.5% 23.2% 58.5% 74.9% 24.5% 14.3% 16.7% 12.0% 17.5% 18.7% 0.50% of % n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 9.1% (maximum Korea, Rep. 18.8% 42.6% 14.3% 612% 23.2% 23.6% 24.2% 22.0% 21.5% 10.7% 15.8% 18.7% 0.37% 316.1% 409.4% 734.6% 182.1 152.7% 233.8% n.a. n.a. n.a. n.a. n.a. n.a. 0.0% 38.5% 80.9% 11.7% 17.0% 526% 0.0% 7.9% Formula Japan 23.9% 23.5% 23.9% 23.5% 22.9% 15.2% 19.1% 10.1% 17.0% 0.71% 538.5% 405.2% 564.4% 398.9% 274.9% 233.0% n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Swiss 8.0% Iceland 74.6% 98.4% 11.8% 17.2% 214% 18.7% 19.9% 20.6% 10.2% 23.6% 22.4% 17.6% 0.37% 115.7% 431.6% 212.6% 137.4% the y % n.a. n.a. 0.0% 0.0% 0.0% 0.0% 0.0% 22% n.a. n.a. 0.0% 0.0% 0.0% 0.0% 0.0% 9.9% 7.6% Hungar 19.9% 28.6% 22.2% 43.5% 16.5% 26.5% 160.7% 11.1 13.3% 11.8% 15.9% 21.6% 0.62% Using EU n.a. n.a. 0.8% 1.1% 0.0% 3.4% 0.0% 12.3% 53.3% 59.6% 22.9% 86.8% 336% 0.7% 8.2% 1.1% 0.0% 3.0% 0.0% 17.0% 24.1% 17.6% 20.7% 20.6% 12.0% 10.4% 0.86% 651.7% 120.0% 116.7% n.a. n.a. n.a. n.a. 5% n.a. n.a. n.a. n.a. Equivalents Czech Rep. 0.0% 0.0% 0.0% 8.8% 3.5% 0.0% 0.0% 0.0% 6.5% 9.8% 3.1% 16.1% 33.4% 46.7% 58.6% 18.6% 14.3% 16.3% 17.5% 10.7% 0.52% ariffT n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Canada 0.0% 0.0% 0.0% 0.0% 0.6% 0.0% 1.2% 10% 0.0% 0.0% 0.0% 0.0% 0.6% 0.0% 1.1% 3.5% 96.8% 17.3% 12.9% 19.9% 10.2% 0.49% tariffs n.a. tariffs n.a. em 0.0% 0.0% 2.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.2% 0.0% 0.0% 0.0% 1.9% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.2% Australia em 0.00% Analyzing valor valor ad veal veal 4.6 sugar ad sugar grains meat meat meat and y grains meat and y meat meat Initial ool Final 1. Wheat Maize Other Rice Oilseeds Refined Milk Beef Sheep W Pig Poultr Eggs verageA arianceV ool ABLET 2. Wheat Maize Other Rice Oilseeds Refined Milk Beef Sheep W Pig Poultr Eggs verageA arianceV Market Access: Economics and the Effects of Policy Instruments 85 % % 9.1% verageA 28.1 23.2% 27.4% 50.4% 28.1 65.0% 57.0% 43.3% 41.9% 22.3% 22.7% 40.0% 46.5% 38.1% 60.0% 70.0% 20.0% 88.9% 13.3% 30.8% 37.5% 66.7% 70.0% 13.3% 38.5% 37.0% 100.0% n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. U.S. 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 6.8% 0.0% 0.7% 0.0% 6.4% 69.2% 78.9% 28.3% 14.1% 38.3% 32.3% 69.5% 98.3% 31.1% % n.a. n.a. n.a. n.a. n.a. n.a. 5.8% 0.3% urkeyT 15.2% 38.3% 17.7% 69.5% 85.1 53.7% 74.0% 42.0% 18.2% 66.7% 48.1% 45.7% 65.6% 35.6% 30.9% 71.0% 54.5% 43.5% 21.7% 147.3% % n.a. n.a. n.a. n.a. 65.8% 70.5% 79.7% 93.0% 90.5% 86.0% 89.0% 91.4% 58.2% 94.3% 92.7% 82.8% 39.1% 50.8% 99.6% 134.7% 204.2% 190.5% 230.7% 105.8% 118.4% 156.3% 136.1 133.3% Switzerland % n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Rep 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 6.4% 60.9% 14.1% 18.8% Slovak 56.1 15.0% 75.3% 56.5% 40.3% 24.5% n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0.0% 0.0% 0.0% 0.0% 0.0% 44.9% 58.2% 79.1% 27.9% 48.0% 74.6% 30.2% 60.1% 21.6% 51.1% 20.2% 60.5% 36.3% Poland 106.8% % % % way n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0.0% 83.1 78.2% 80.0% 84.5% 78.9% 42.3% 86.6% 78.6% 68.0% 40.2% 95.4% 88.4% 86.2% 69.1% Nor 196.1 185.1 116.8% 109.7% n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. New New 0.0% 0.0% 0.0% 0.0% 2.7% 24.5% 66.9% 15.2% 38.7% 44.1% Zealand % % % n.a. n.a. n.a. n.a. n.a. n.a. 0.0% 0.0% 7.7% 57.1% 66.8% 17.4% 36.7% 48.2% 70.1 75.0% 10.2% 20.1 36.5% 36.6% 71.3% 31.5% 76.5% 11.5% 27.8% Mexico 103.6% 188.2% 27.1 of n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Korea, 92.7% 94.2% 96.7% 87.9% 85.9% 42.9% 63.0% 36.4% 75.0% 87.1% 54.2% 98.7% 88.6% 57.7% 21.6% Rep. 237.6% 243.9% 105.8% % % % n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0.0% 95.6% 94.2% 95.8% Japan 94.1 91.7% 60.6% 76.4% 31.8% 40.5% 68.1% 90.1% 44.7% 60.7% 40.2% 20.4% 12.9% 240.7% 243.1 236.1 102.5% % n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 74.9% 79.7% Iceland 32.1 82.2% 40.8% 94.5% 89.5% 70.5% 31.3% 84.3% 23.4% 79.3% 92.6% 95.8% 269.5% 136.6% y n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0.0% 0.0% 0.0% 0.0% 0.0% 6.4% 8.8% 0.6% 44.3% 53.4% 47.0% 63.5% 39.7% 86.5% 30.4% 31.9% 51.7% 86.3% 18.0% Hungar % % n.a. n.a. EU 3.0% 4.4% 0.0% 0.0% 0.0% 0.0% 32.9% 68.1 96.3% 70.4% 82.8% 82.4% 47.8% 12.1% 41.7% 89.5% 42.1% 84.1 35.2% 48.0% 23.5% 91.3% 96.8% 69.7% 17.0% 110.5% % % % n.a. n.a. n.a. n.a. eductionr n.a. n.a. 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 1.0% Czech Rep. 25.9% 39.1 57.2% 12.4% 65.1 70.1 30.0% 60.1% 31.4% 32.2% 45.4% 63.0% 50.9% tariff n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0.0% 0.0% 0.0% 0.0% 2.3% 0.0% 4.6% 0.0% 0.0% 0.0% 0.0% 0.0% 79.5% 40.9% 14.1% Canada average 39.4% 94.6% 88.5% 12.0% 17.3% of tariffs % in n.a. a n.a. 0.0% 0.0% 7.4% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.6% as 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 7.1% 85.3% Australia (continued) eductionr veal veal 4.6 sugar meat meat eductionr sugar meat cent grains and y grains meat meat Per ool riffaT and y meat ABLET 3. Wheat Maize Other Rice Oilseeds Refined Milk Beef Sheep W Pig Poultr Eggs verageA ool 4. Wheat Maize Other Rice Oilseeds Refined Milk Beef Sheep W Pig Poultr Eggs verageA 86 Agriculture and the WTO 0.0% verageA 69.4% 85.7% 20.0% 10.0% 14.3% 33.3% 42.9% 87.5% 14.3% 45.5% 64.5% 125.0% 100.0% 160.0% 100.0% n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 3.0% U.S. 71.5% 63.8% 24.8% 82.3% 93.8% 99.8% 82.7% 63.0% n.a. .2% n.a. n.a. 0.7% 1.7% 7.7% urkeyT 69.9% 92.4% 91.8% 37.3% 61 28.5% 67.7% 92.6% 24.8% 44.2% 44.5% n.a. n.a. 89.5% 28.9% 18.5% 49.5% 52.9% 289.8% 265.6% 105.5% 266.7% 376.5% 157.9% 454.3% 378.8% 175.3% 216.3% Switzerland % n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Slovak Rep. 53.6% 22.8% 74.9% 94.5% 89.5% 57.5% 88.7% 79.1 % % n.a. n.a. n.a. n.a. n.a. n.a. n.a. 2.8% 67.5% 65.5% 61.8% Poland 39.1 87.1 69.2% 10.1% 67.5% 52.1% n.a. n.a. n.a. n.a. n.a. way 81.2% 17.0% 36.6% 33.4% 66.5% 35.6% 37.7% 26.3% 72.5% 107.3% 146.9% 115.9% Nor % .1 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. New 12.0% 89.2% 23.9% 86.3% 56.5% Zealand 61 n.a. n.a. n.a. n.a. 9.7% 0.1% 62.7% 43.6% 92.0% 91.7% 74.4% 62.3% 94.9% 53.4% 65.8% 31.5% Mexico 228.4% of eductionsr n.a. n.a. n.a. n.a. n.a. Korea 92.0% Rep. tariff 32.5% 38.7% 43.2% 78.5% 380.3% 135.4% 707.9% 143.3% 176.2% 226.5% 182.5% n.a. n.a. n.a. Japan 92.7% countries' 33.5% 30.4% 84.5% 74.5% 811.4% 515.7% 224.5% 156.9% 267.4% 498.3% 225.4% 253.9% all of n.a. n.a. n.a. n.a. n.a. n.a. 0.3% 87.2% 25.6% 78.0% 78.9% 27.2% 91.9% Iceland 293.8% 475.3% 219.4% 144.9% average y to tariff n.a. n.a. n.a. n.a. n.a. n.a. n.a. 2.7% 71.3% 87.2% 61.7% 59.8% 22.3% 78.1% 63.0% 27.9% Hungar y elativer 141.5% y initial n.a. n.a. EU countr 88.0% countr 26.8% 98.7% 19.8% 98.3% 69.4% 20.4% 69.5% 95.4% 21.2% 21.2% average 319.6% 116.9% 108.3% each of each for n.a. n.a. n.a. n.a. n.a. n.a. n.a. for Rep. 42.6% 38.6% 94.4% 78.5% 39.7% 73.8% 37.7% 11.9% 74.1% 56.0% Czech centage tariffs tariff per a n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 4.5% 72.5% as 29.3% 98.9% 98.4% 74.0% 82.0% 60.5% Canada average average in in tariff n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 7.4% 89.3% (continued) Australia initials 98.9% 99.8% 98.9% eductionr eductionr y' 4.6 ariffT ariff veal countr sugar meat meat 7' centage centage grains and y meat e.vA Per Per Each ABLET 5. verageA 6. verageTA ool 7. Wheat Maize Other Rice Oilseeds Refined Milk Beef Sheep W Pig Poultr Eggs verageA Col. Market Access: Economics and the Effects of Policy Instruments 87 7.0% 5.0% 0.0% verageA 85.7% 20.0% 10.0% 14.3% 33.3% 42.9% 87.5% 14.3% 45.5% 85.7% 20.0% 10.0% 14.3% 33.3% 42.9% 87.5% 14.3% 45.5% 60.2% 125.0% 100.0% 160.0% 100.0% 125.0% 100.0% 160.0% 100.0% n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 6.4% 6.4% U.S. 38.3% 32.3% 69.5% 98.3% 62.9% 31.1% 38.3% 32.3% 69.5% 98.3% 62.9% 36.4% ­184.0% % n.a. n.a. n.a. n.a. n.a. n.a. 5.8% 0.3% 5.8% 0.3% urkeyT 45.7% 65.6% 35.6% 30.9% 71.0% 54.5% 43.5% 26.0% 21.7% 45.7% 65.6% 35.6% 30.9% 71.0% 54.5% 43.5% 26.0% 22.1% 147.3% 147.3% ­41.3 % % n.a. n.a. n.a. n.a. 39.1% 50.8% 99.6% 39.1% 50.8% 99.6% 134.7% 204.2% 190.5% 230.7% 105.8% 118.4% 156.3% 136.1 117.2% 133.3% 134.7% 204.2% 190.5% 230.7% 105.8% 118.4% 156.3% 136.1 117.2% 131.9% 260.5% Switzerland n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 6.4% 6.4% Slovak Rep. 75.3% 56.5% 40.3% 60.7% 24.5% 75.3% 56.5% 40.3% 60.7% 31.7% ­166.2% % % n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 60.1% 21.6% Poland 106.8% 51.1 20.2% 60.5% 20.7% 36.3% 60.1% 21.6% 106.8% 51.1 20.2% 60.5% 20.7% 28.2% ­1.7% 2002. % % % % way n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. OECD 40.2% 95.4% 88.4% 86.2% 69.1% 78.4% 40.2% 95.4% 88.4% 86.2% 69.1% 78.4% Nor 196.1 185.1 116.8% 109.7% 196.1 185.1 116.8% 106.2% 196.6% the by n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 2.7% New 38.7% 44.1% 60.0% 38.7% 44.1% 60.0% 18.2% Zealand ­92.7% published % % n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. as 7.7% 4.3% 7.7% 4.3% 36.6% Mexico 103.6% 188.2% 27.1 71.3% 31.5% 76.5% 11.5% 27.8% 36.6% 103.6% 188.2% 27.1 71.3% 31.5% 76.5% 11.5% 24.6% ­24.7% of n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. eductionsr Korea Rep. 87.1% 54.2% 98.7% 88.6% 57.7% 21.6% 96.6% 87.1% 54.2% 98.7% 88.6% 57.7% 21.6% 96.6% 237.6% 243.9% 105.8% countries 237.6% 243.9% 104.8% 186.1% tariff % % all % % % support"--MPS) n.a. n.a. n.a. n.a. of n.a. n.a. n.a. n.a. 90.1% 44.7% 60.7% 40.2% 20.4% 12.9% 78.5% 90.1% 44.7% 60.7% 40.2% 20.4% 12.9% 78.5% Japan 240.7% 243.1 236.1 102.5% 240.7% 243.1 236.1 100.1 177.2% price countries' eductionr all n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. of ("market Iceland 31.3% 84.3% 23.4% 79.3% 92.6% 85.0% 95.8% 31.3% 84.3% 23.4% 79.3% 92.6% 85.0% 94.4% 269.5% 136.6% tariff 269.5% 136.6% 159.0% y average countries n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 6.4% 8.8% 0.6% of all 31.9% 51.7% 86.3% 20.3% 18.0% average 6.4% 8.8% 0.6% 31.9% 51.7% 86.3% 20.3% 12.5% ­51.4% protections Hungar of of % % centage EU tariff n.a. n.a. n.a. n.a. n.a. n.a. 9.3% 9.3% per border 89.5% 42.1% 84.1 35.2% 48.0% 23.5% 91.3% 96.8% 69.7% 20.4% centage 89.5% 42.1% 110.5% 84.1 35.2% 48.0% 23.5% 91.3% 96.8% 69.7% 19.4% 110.5% a 54.0% of final per as % a % n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1.5% as n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1.8% Rep. Czech average 60.1 31.4% 32.2% 45.4% 63.0% 50.9% 11.7% 60.1 31.4% 32.2% 45.4% 63.0% 50.9% 21.4% ­97.3% of eductionsr equivalents % n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. a eductionr n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. tariff as 39.4% 94.6% 88.5% 12.0% 62.9% 38.9% 39.4% 94.6% 88.5% 12.0% 62.9% 43.7% tariff Canada tariff of 146.8% the tariff n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. using Australia finals 85.3% 98.4% 85.3% averages 85.3% 98.4% 91.8% averages 119.2% (continued) y' y' y' veal countr sugar 8' veal countr sugar 9' countr Calculated 4.6 grains meat meat meat and y meat ve.A grains meat and y meat ve.A Reducing ce: Each ool Each Each 8. Wheat Maize Other Rice Oilseeds Refined Milk Beef Sheep W Pig Poultr Eggs verageA ool Col. 9. Wheat Maize Other Rice Oilseeds Refined Milk Beef Sheep W Pig Poultr Eggs verageA Col. 10. ve.A Sour ABLET 88 Agriculture and the WTO TABLE 4.7 World Production and Value of Production and Tariff-Quota-Related Protection for Grains and Oilseeds in the Six Regions of the PEM Model Production Value of Production Total Quota Quota/Prod Total Quota MPS MPS/VOP Commodity (mmt) (mmt) (%) US$Bn US$Bn (%) Wheat 603 5 0.8 63.5 0.3 0.4 Coarse grains 904 27 3.0 74.6 1.3 1.7 Oilseeds 236 0 0.0 39 0.0 0.0 Rice 405 12 2.9 39.8 14.4 36.2 Total 2,148 43 2.0 216.9 15.9 7.3 Source: Authors' calculations using PEM (see Short 2002). TABLE 4.8 Import Quotas in PEM Imports in 2001 Tariff-quota Amount Share of Quota Over-quota Over-quota Crop/Region (mmt) Consumption (mmt) imports (mmt) imports Quota Mexico Wheat 3.003 51% 0.605 2.398 397% Maize 11.1 31% 2.501 8.6 344% EU Rice 1.848 75% 0.084 1.764 2100% Switzerland Wheat 0.300 33% 0.180 0.120 67% Coarse grains 0.151 19% 0.630 0.479 76% Japan Wheat 5.396 84% 5.740 0.344 6% Coarse grains 2.280 92% 1.369 0.911 67% Rice 0.763 8% 0.682 0.081 12% Source: Authors' calculations using PEM (see Short 2002). overwhelming form of border protection for grains the t1 regime is effective for coarse grains in and oilseeds in these regions. Switzerland, resulting in quota under-fill. Over- Table 4.8 shows the levels of quotas as well as the quota imports subject to the official in-quota rates level of gross imports in 2001. Imports are shown or less are allowed at the discretion of the import- in million metric tonnes as well as a share of con- ing country. This is a practice followed by many sumption. In many instances, imports are a signifi- countries. However, if current negotiations are cant share of consumption. The last two columns phrased in terms of each country's existing com- in table 4.8 show the degree of quota overhang mitment, it may have a significant impact on both reflecting the difference between total imports and the amount and the equity of additional market the tariff quota level at the applied tariff t (which is liberalization. Nevertheless, as shown in table 4.9, below teQ). Tariff quotas and total imports are the applied tariff is t1 in both cases in Switzerland almost identical in Japan, indicating that Japan is but the economics of trade liberalization will be fully applying its tariff quota entitlements. The very different. over-quota imports show that imports exceed tariff The significance of over-quota imports is that it quotas for Mexico and for wheat in Switzerland but affects how a country will or will not have to adjust Market Access: Economics and the Effects of Policy Instruments 89 TABLE 4.9 Tariffs Applicable to Tariff Quota Commodities World Pricec Bound t1 Bound t2 Crop/Region Bound t1 Bound t2 (US$/mt) (US$/mt) (US$/mt) Mexico Wheat 50%b 67%b 119.60 59.80 80.13 Maize 50%a 200%a 142.94 71.47 208.09 Other grains 50%a 119%a 99.42 49.71 118.31 EU Rice 21b,d 376b,d 169.40 18.81 336.74 Switzerland Wheat 222b,e 740b,e 224.12 131.60 438.67 Coarse grains 323b,e 500b,e 146.23 191.48 296.40 Japan Wheat 9.5%a 547%a 256.26 24.34 1,401.47 Coarse grains 0a 400%a 219.75 0.00 878.76 Rice 5%a 1291%a 277.38 13.87 3,580.99 Notes: a. Agricultural Market Access Database (AMAD) reported in the OECD Outlook database and described as follows: "The tariff and TRQ data are based on Most Favored Nation rates scheduled with the WTO and exclude those under preferential or regional agreements, which may be substantially different. Tariffs are averages of several product lines. Specific rates are converted to ad valorem rates using world prices in the outlook." b. The AMAD database directly, using weighted averages where this was appropriate. c. Reference prices are from the OECD PSE tables. d. In ECU/mt. e. In SF/mt. Source: Authors' calculations using PEM (see Short 2002). to increased trade pressure in the negotiations. For the OECD reference prices in 2001. The last two example, if quotas were increased by 20 percent as columns show tariffs in U.S. dollars per metric suggested by the U.S., there would be no impact on tonne that result from applying the tariff rates in either Mexico or Switzerland because their over- the first two data columns to the OECD reference quota imports are already much larger than 20 per- prices. cent of their current quota. The tariff equivalent rates (teQ) in the first data An increase in quotas of 20 percent in Japan column of table 4.10 are derived from the market would have an impact on wheat because there is no price support reported in the OECD PSE tables. It overhang but a priori the impact on rice would be is necessary to have an estimate of the average tar- small even though there are some over-quota iff actually charged for in-quota and over-quota imports, because the current quota is a small share of imports to divide the tariff equivalent rates into total consumption. (In this case a 20 percent increase benefits accruing to taxpayers in the form of duties in the quota would be less than 2 percent of current collected and those accruing to quota holders in the consumption, so the world market adjustments to form of quota rents. In some cases, the bound t1 tar- this change alone would be small.) iff rates are assumed to be the applied rates, while in Table 4.9 sets out tariffs applicable for the tariff other cases the applied rates are below the bound quota commodities in the PEM. (The bound t1 tar- rates. The applied tariffs for Switzerland were set at iff rates are applicable to tariff quota imports while the teQ rates since these were slightly less than the the bound t2 rates are applicable to out-of-quota bound t1 tariff rates estimated. The applied tariffs in imports.) The third data column of the table shows Mexico were 3 percent, 3 percent, and 2 percent on 90 Agriculture and the WTO TABLE 4.10 Tariff Rates (US$/mt) Tariff In-Quota Quota Rental Water and Water and Overhang Crop/Region Equivalents Applied Tariff Rate Overhang as a % of PW Mexico Wheat 39.85 3.59 36.26 97.69 82% Maize 52.25 4.29 16.51 149.31 104% Other grains 5.23 1.99 3.24 193.90 195% EU Rice 90.31 18.81 71.50 246.43 146% Switzerland Wheat 107.90 107.90 0.00 330.77 147% Coarse grains 113.52 113.52 0.00 182.88 129% Japan Wheat 109.73 24.34 85.39 1,291.73 504% Coarse grains 878.98 351.10 527.88 0.00 0% Rice 1,565.40 13.87 1,551.53 2,015.59 727% Note: The teQ rates are the rates of MPS derived from the PSE tables. Source: Authors' calculations using PEM (see Short 2002). an ad valorem basis for wheat, maize, and other · Reduction of all tariff-quota bound t2 tariffs by grains, respectively. application of the Swiss (25) formula; The case of coarse grains in Japan is somewhat · Reduction of tariff-quota bound t1 tariffs to unusual in that teQ in table 4.10 is almost exactly zero; and identical to the bound t2 tariff in table 4.9. There · Increase in tariff quotas by 20 percent. are significant over-quota tariffs for coarse grains in Japan. This suggests that the bound t2 tariff is deter- The Swiss formula achieves a reduction in the mining Japanese coarse grain prices rather than average tariffs (unlike the average 36 percent cut in applied tariff quotas. The applied tariff in this case the URAA, which allows members to meet the tar- is a combination of the in-quota tariff (table 4.9) get through large cuts in trivial tariffs, and small applied on the first 1.369 mmt of imports (the cuts in the highest and most important tariffs). It quota) and US$878.98/mt (the teQ) thereafter. also achieves a reduction in the highest rates by rel- The difference between the applied tariffs and atively more than others. the teQ rates are the quota rental rates listed in the Here the tariff-equivalent version of the model is third data column of table 4.10. used first, in which all commodities are modeled as Table 4.10 also contains an estimate of water in the though they are subject to a simple quota set at the tariff and binding overhang, found by subtracting the MPS rate in the baseline scenario but quotas are teQ rates from the t2-bound rate (from table 4.9). adjusted to ensure that the teQ of binding quotas decline with the MPS of nonbinding quota situa- tions. This scenario strips away any tariff quota pro- The Swiss Formula for Improving Market Access tection. The following two scenarios are run with in Tariff Rate Quota Regimes this "tariff" model: Swiss (25) and Swiss (50). The main features of the Swiss Formula for tariff The other scenarios were created using the PEM reductions are as follows:12 adapted to simulate the operation of the tariff rate quotas with imports restricted to the import quota · Reduction of all non-tariff-quota tariffs by and price determined nationally (the TRQ model). application of the Swiss (25) formula; Regimes switch automatically in the TRQ model Market Access: Economics and the Effects of Policy Instruments 91 TABLE 4.11 Impact of Tariff Quota Border Protection (US$/mt) Crop/Region Base Swiss (25) Swiss (50) Mexico Wheat 39.85 17.08 23.91 Maize 5.23 4.36 4.76 Other grains 52.25 17.49 26.21 EU Rice 90.31 29.80 45.04 Switzerland Wheat 107.90 36.89 54.96 Coarse grains 113.52 27.80 44.64 Japan Wheat 109.73 40.44 59.10 Coarse grains 878.98 51.89 97.91 Rice 1,565.40 67.77 129.78 Note: The teQ rates are the rates of MPS derived from the PSE tables. Source: Authors' calculations using PEM (see Short 2002). (Short 2002). In this way, the U.S. proposal can be world prices and the tariff bounds. This is true analyzed directly. even in Mexico where there are considerable However, if the national price becomes greater over-quota imports.13 than the world reference price for the country plus · The final level of border protection is very simi- the bound t2 (PW t2), imports increase above the lar to that resulting from the tariff-equivalent quota and price is PW t2. The second scenario in model. which the bound t2 rates are determined by the Because there is so little increase in world mar- Swiss formula with a 0.25 is meant to simulate the ket prices, producer prices decline almost entirely U.S. proposal for market access. The last two scenar- by the amount of the reduction in MPS. There is ios show more liberal versions obtained by increas- some offset with rice but baseline protection rates ing the a-parameter in the Swiss formula to 0.50. are so large for rice that this is hardly noticeable. Table 4.11 demonstrates the impacts on border protection by showing the specific tariff equivalents Policy Options for Market Access that would result in all three scenarios together with Commitments the levels of teQ rates in the base for comparison. All scenarios result in large changes in Japanese and This chapter provides a broad overview of the ways in Swiss agriculture because of the current extraordi- which market access improvements as a result of the narily high rates of effective protection. Mexico URAA have been limited across countries and com- would also be in a position of making extremely modities worldwide. Although policy data are only large changes except for the high degree of over- available on a comprehensive basis for developed quota imports in the baseline. countries and some middle-income developing There are perhaps two especially noteworthy countries, it can be concluded that most developing results from this simulation of the U.S. proposal countries have reduced protection in the past with the TRQ version of the PEM: decade by reducing tariffs. In the meantime, the · All commodities switch from a quota-binding absolute value of total transfers from consumers regime in which prices are determined nation- and taxpayers to agriculture has not changed very ally by the volume of imports to a t2-binding much for all developed countries and has even regime in which prices are determined by the increased for some middle-income developing 92 Agriculture and the WTO countries. Not only does import protection remain world trading system and on the developing coun- very high, it varies across countries and commodi- tries in particular. A basic set of principles that ties in terms of levels and the type of instrument should be incorporated in any deliberation of mar- used (tariffs versus tariff quotas), tariff escalation, ket access disciplines in the WTO negotiations is special safeguards, and tariff peaks and dispersion. presented. This means that the URAA in and of itself has done very little in liberalizing trade. The agreement Tariff Reductions is widely credited for bringing transparency through the process of tariffication, limiting the type of 1. Current tariff reduction commitments based on instruments that can be used, and making it easier unweighted average tariff reductions have for future negotiations to bring about meaningful allowed countries to have larger tariff reductions trade liberalization. Nevertheless, the analysis in this on commodities with low tariffs, which are of lit- chapter shows that transparency is not complete as tle importance to exporters, and politically less many non-ad valorem tariffs and complicated sensitive domestically. This shortcoming can be import quota schemes remain. mitigated if all tariff reduction commitments are Import protection rates have gone down for made on an individual product basis or using a developed countries, particularly in the grains and formula approach. oilseeds sectors as a re-instrumentation away 2. Tariff levels need to be reduced substantially from border protection to domestic support has because of gaps between bound tariffs and occurred. However, declines in protection rates applied tariffs and between applied tariffs and otherwise may be due more to world price changes tariff equivalents of binding import quotas. and exchange rate realignments than to policy 3. Non-ad valorem tariffs should be converted to changes. The share of total trade-distorting sup- ad valorem equivalents to increase transparency port from import protection versus production and then be reduced to levels consistent with and input subsidies (excluding area and historical reduction commitments for current ad valorem entitlement payments) has declined from 77 to 61 tariffs. percent in OECD countries. However, empirical 4. Tariff dispersion should be dealt with by lowering analysis in this paper indicates total trade rents and higher tariffs faster. For example, using a Swiss tariff revenue amount of approximately US$50 bil- formula­type approach, significant improve- lion, implying that revenues and rents for govern- ments were shown to occur. ments and quota holders may decline with trade 5. Tariff escalation should be dealt with by imme- liberalization. diately lowering all tariffs on processed foods to Empirical analysis of a variant of the Swiss for- current tariffs on imported intermediate or mula shows that tariff peaks and dispersion can be farm bulk products using a Swiss formula­type reduced significantly--therefore this is a useful approach. approach that should be taken seriously in the nego- 6. Discretionary duties in the EU and special safe- tiations. Using the PEM model, it was shown that guard duties for rich countries should be elimi- reducing tariffs is not enough because of gaps in the nated. applied versus bound tariffs (tariff binding over- 7. No tariffs should be employed to fulfill "non- hang) and applied versus tariff equivalent of binding trade" concerns. quota (water in the tariff). Hence, significant increases in quotas along with large tariff cuts are Expanding Quotas Versus Reducing Tariffs required for significant trade liberalization to occur. The Swiss formula with a maximum tariff of 25 per- Given that only minimal market access was ensured cent was used here as an illustration of such an effec- by the implementation of quotas, negotiators may tive mechanism. want to ensure a significant increase in access. The Some of the key options for the Doha Round analysis of this chapter suggests that the best strat- negotiations are summarized below that would egy to meet this goal depends on the conditions of enable it to more fully meet the goal of minimizing the market and the relative values of the quota and the unfavorable effects of import protection on the various tariff levels. Negotiators should especially Market Access: Economics and the Effects of Policy Instruments 93 focus on reducing out-of-quota tariffs, in those tariff. Still others include situations where the country imposes cases with out-of-quota imports, or if the out-of- an import tariff, but is an exporter and the domestic price is lower than the border price without tariff (in which case the quota tariff is close to the tariff equivalent of the entire tariff is water). This chapter discusses--and measures-- binding quota. If there are no out-of-quota those cases where the product is imported and is subject to a imports, increasing the quota will have a greater binding tariff rate quota. These are the cases that are relevant for policy discussions. chance of liberalizing trade in the short run. 5. "Total support" is a measure of all transfers from taxpay- A reduction in in-quota tariffs will liberalize trade ers and consumers, regardless of objectives and impact, while only if the out-of-quota tariff is close to and below the "producer support" is that intended to enhance farm incomes. However, most of these affect farm income indi- the in-quota tariff, in which case both tariffs need rectly--such as environmental payments, which can act as input to be reduced, or if under-fill is significant because subsidies, or expenditures on marketing and promotion (the lat- the in-quota tariff is binding. Otherwise, quotas ter alone amounting to US$20 billion). will also have to be increased to obtain trade-liber- 6. For those tariff quota commodities, these tariffs are the bound out-of-quota tariff levels. alizing effects. 7. The OECD country coverage includes only three devel- While reducing the extremely high out-of-quota oping countries (the Republic of Korea, Mexico, and Turkey) tariffs may seem an attractive political option to and four transition economies (Czech Republic, Hungary, Poland, and Slovak Republic). negotiators, the net result will be little or no 8. The difference between column 1(b) and 1(c) represents increase in the degree of trade liberalization if an estimate of non-import-barrier protection--for instance, either the quota or in-quota-tariff is actually the export subsidies. 9. As shown earlier, uneven rates of protection among prod- binding instrument--because there is a significant ucts closely related in production and/or use can distort the use amount of "binding overhang" and "water" in of resources in world agriculture even more than a slightly many of these tariffs. But as long as a two-tiered higher but more uniform level of protection. tariff rate quota regime remains, so does the possi- 10. A Swiss formula tariff cut will result in a common max- imum tariff across countries and high tariffs will be cut more bility of quotas and consequent rent seeking. than lower tariffs. Therefore, it may be better to reduce the out-of- 11. The EU tariff quota regime for wheat was in effect in quota tariff significantly and eliminate quotas all 2001 but the EU was a net exporter so it was not modeled in the PEM. together. For example, reducing the out-of-quota 12. Note that the recent Cairns proposal has similar charac- tariff will expand imports in those cases in which teristics while Japan and the EU have failed to provide numbers the tariff equivalent of the quota is very close to with respect to tariff-quota liberalization scenarios. 13. There is some uncertainty in designing scenarios for it--and even then, decreasing tariffs and increasing countries such as Mexico. Since these countries utilize only a quotas will maximize the gains from liberalization. fraction of the tariff quota protection available to them, it is dif- If, however, the tariff equivalent of the quota is sig- ficult to know how much protection they would choose to use nificantly below the out-of-quota tariff, increasing when the maximum protection is much reduced, as it would be with the U.S. proposal. The high degree of over-quota imports the size of the quota will have an immediate effect. gives them significant discretion in some instances. Their teQ rates are around US$40/mt for wheat and US$52/mt for maize at the current levels of imports. Applying the Swiss formula with Notes a 25 to their t2-bound rates and allowing world prices to adjust results in border protection of US$22/mt for wheat and 1. In contrast, most industrial tariffs are about 5 percent, US$23/mt for maize. In other words, they are forced out of the with many manufacturing tariffs now traded duty-free. Textiles quota-binding regime for these commodities. Their effective and a few other goods join agriculture as the last sectors with protection for other grains is much lower, however, so they may very high import protection levels. continue to allow over-quota imports at low tariffs and keep the 2. For such a critical assessment, see Ingco (1996), Ingco lower degree of protection they had in the base. Note that the and Hathaway (1996), and OECD (2001b), as well as chapters 3 tariff quotas for Mexico are to be nonbinding by 2008. and 6 of this volume. 3. As described in chapter 8 on food security, trade policy is usually not the best means to achieve food security goals. Select Bibliography 4. This is only one of several possible definitions of "water in the tariff." In the literature, water in the tariff is the amount ABARE (Australian Bureau of Agricultural and Resource Eco- by which a tariff charged exceeds the prohibitive tariff, with the nomics). 1999. "WTO Agricultural Negotiations: Important "prohibitive tariff" defined as the difference between the border Market Access Issues." Research Report 99.3, Canberra. price at which imports would be 0 and the domestic price; in AMAD (Agricultural Market Access Database--www.amad. org). other words, that portion of the tariff which is beyond what is Boughner, D., H. de Gorter, and I. Sheldon. 2000."The Econom- necessary to cut off all trade. Other definitions of water in the ics of Tariff-Rate Quotas in the Agricultural Agreement in tariff allow for the possibility that the domestic product is dif- the WTO." Agricultural and Resource Economics Review ferentiated from imports, and so is marked up by less than the 20(April): 58­69. 94 Agriculture and the WTO Cernat, Lucian, Sam Laird, and Alessandro Turrini. 2002. "Back Martin, W. 2000. "What Has the GATT/WTO Agricultural to Basics: Market Access Issues in the Doha Agenda." Trade Agreement Actually Done?" American Journal of Agricultural Analysis Branch, DITC, UNCTAD draft. Economics 82(August): 729­30. de Gorter, H., and D. Boughner. 1999."U.S. Dairy Policy and the Messerlin, Patrick A. 2002. "Agriculture in the Doha Round." Agreement on Agriculture in the WTO." Canadian Journal of Paper presented at the World Bank's Wilton Park Conference Agricultural Economics 47(5): 31­42. on Prospects for the New Trade Round, July. de Gorter, H., and I. Sheldon, eds. 2001. "Issues in Reforming Moschini, G. 1991. "Economic Issues in Tariffication: An Tariff-Rate Import Quotas in the Agreement on Agriculture in Overview." Agricultural Economics 5: 101­20. the WTO." International Agricultural Trade Research Consor- OECD. 2001a. "Market Effects of Crop Support Measures." tium (IATRC) Commissioned Paper 13, University of Min- OECD, Paris. nesota, St. Paul. iatrcweb.org/Publications/commiss.html. ------. 2001b. "The Uruguay Round Agreement on Agricul- Francois, Joseph, and Will Martin. 2002."Formula Approaches for ture: An Evaluation of Its Implementation in OECD Coun- Market Access Negotiations." World Economy 26(1): 1­28. tries." OECD, Paris. Gibson, Paul, John Wainio, Daniel Whitley, and Mary Bohman. ------. 2002. "Agricultural Policies in OECD Countries: Moni- 2001. "Profiles of Tariffs in Global Agricultural Markets." toring and Evaluation 2002." OECD, Paris. Agriculture Economic Report No. 796, USDA Economic Short, C. 2002. "Distributional Effects of Agricultural Policy Research Service, Washington, D.C. Reforms: Results from the Policy Evaluation Model." Report IATRC (International Agricultural Trade Research Consor- to ARD, World Bank, Washington, D.C. August 2. tium). 2001a. "Market Access: Issues and Options in the Tangermann, S. 1996. "Implementation of the Uruguay Round Agricultural Negotiations." Commissioned Paper 14, May. Agreement on Agriculture: Issues and Prospects." Journal of ------. 2001b. "The Current WTO Agricultural Negotiations: Agricultural Economics 47: 315­37. Options for Progress." Commissioned Paper 18, May. Wainio, J., P. Gibson, and D. Whitley. 2001. "Options for Reduc- IATRC, Washington, D.C. ing Agricultural Tariffs." In M. Burfisher, ed., "Agricultural Ingco, M. D. 1996. "Tariffication in the Uruguay Round: How Policy Reform in the WTO--The Road Ahead." USDA ERS Much Liberalization?" The World Economy 19(4) (July): Agricultural Economics Report No. 802., Washington, D.C. 425­47. World Bank. 2003. "Agricultural Policies and Trade." Chapter 3 Ingco, M. D., and D. E. Hathaway. 1996. "Implementation of the in Global Economic Prospects 2004. Washington, D.C.: World Uruguay Round Commitments in Agriculture: Issues and Bank. Practice." Paper for the Roundtable Discussion of the WTO (World Trade Organization). 2002a. "Tariff Quota and Uruguay Round Agreement on Agriculture, "Rural Well- Other Quotas." Background paper TN/AG/S/5, March 21. Being: From Vision to Action," sponsored by the Environ- ------. 2002b. "Tariff Quota Administration Methods and Tar- ment and Sustainable Development Department, World iff Quota Fill." Background paper TN/AG/S/6, March 22. Bank, Washington, D.C., September 25­27. ------. 2002c. "Special Agricultural Safeguard." Background Josling, T., S. Tangermann, and T. K. Warley. 1996. Agriculture in paper G/AG/NG/S/9, February 19. the GATT. Houndmills, U.K. and New York: Macmillan and St. Martin's Press. 5 Quota administration methods: Economics and effects with trade liberalization Harry de Gorter and Jana Hranaiova Introduction have TRQs in their Schedules annexed to the URAA. Since 1995, 1,425 tariff quotas have come into effect Quotas have resulted in the institutionalization of as a result of the World Trade Organization's preexisting rents for specific countries and firms or (WTO) Uruguay Round Agreement on Agriculture state trading enterprises (STEs), thereby potentially (URAA). Rents generated by import quotas provide maintaining resistance by these stakeholders to any an opportunity for firms to spend resources in further trade liberalization initiatives. competing for these rents, with the degree of dissi- The purpose of this chapter is to assess the pation depending critically on the method by problems and issues related to administering the which import quota licenses are allocated. Allocat- large number of TRQs. No specific provisions ing the importing rights directly to higher-cost were approved in the URAA regarding adminis- importing firms with nontransferable licenses can tration of the quotas, although relevant General also dissipate rents. Agreement on Tariffs and Trade (GATT) rules In 1995, the URAA put in place a set of rules that were to apply. Trade liberalization with TRQs is can have significant effects on the conditions for very complex, involving two tariffs, a quota, and market access for agricultural products. In most occasionally multiple situations like over-quota cases, bound tariffs replaced nontariff barriers such imports, quota under-fill, and preferential quotas as quotas, embargoes, and licenses. From that point and tariffs. In terms of administering TRQs, WTO on, rules facing exporters were to be more transpar- members use different methods, such as applied ent. In addition, minimum access commitments tariffs; auctioning; licenses on demand; and first- were made through the use of tariff rate quotas come, first-served (FCFS). Each method can lead (TRQs), with a lower tariff (in-quota tariff) for to differing inefficiencies and inequities. In addi- imports within the quota, and a higher tariff rate tion, other conditions placed on TRQ administra- (out-of-quota tariff) for imports exceeding the tion by WTO members such as domestic purchase quota. A total of 43 countries including the Organi- requirements (even though these are prohibited sation for Economic Co-operation and Develop- by existing GATT rules) or quota limits per firm ment (OECD) member countries (except Turkey) have the potential to generate inefficiencies. 95 96 Agriculture and the WTO Quota administration can have a direct influ- liberalization. In some cases, quotas are allocated to ence on both trade flows and the distribution of countries that are unlikely to be able to export the rents originating under the quotas, and is, there- relevant commodity. In other cases, tariffs under fore, a highly political issue. In the debate about the preferential agreements are lower than the in-quota URAA's implementation, members' dissatisfaction most-favored-nation (MFN) tariffs, so those mini- has been voiced regarding TRQ administration in mum access quotas are, de facto, filled with prefer- many specific cases, and, in some instances, dis- ential imports from particular countries. putes have been brought before the WTO. There is TRQs have various institutional designs with an urgent need to provide more information on respect to the distribution of quota shares among how TRQs are currently administered, what the countries and licenses (allocated by both importing economic implications are for TRQ reform and and exporting countries) among importing and trade liberalization, and what better rules for TRQ exporting firms or to STEs. A global quota has administration might look like. imports determined by market forces (provided TRQs at low or minimal tariffs provide market there are no biases in the licensing schemes) while access opportunities in agriculture, beyond the country-specific allocations allow the importing expected effects of the scheduled reduction in tar- country to assign shares to specific exporting coun- iffs. TRQs were put in place to mitigate the fact that tries. In the latter case, WTO rules state that all sub- tariffication of then-existing quantitative restric- stantial suppliers--defined as countries with a mar- tions would have shut off all trade in many cases. ket share of over 10 percent--have to receive a All countries with quotas were expected to allow share (see paragraph 7 of Article XXVIII:1, GATT access to their domestic markets for imports equiv- 1994). Licenses are often used as a means of admin- alent to at least 3 percent of domestic consumption istering TRQs and can be assigned to importing or in the 1986­88 base period. This proportion was to exporting firms (or to both such that an importing rise to 5 percent by 2000 (2004 for developing firm needs to present both an import and an export countries). These provisions refer to "minimum license to customs authorities). The share of rent access." When the normal level of imports did not going to importing or exporting countries will represent a sufficient percentage of domestic con- depend on the bargaining power resulting from any sumption, TRQs were applied to meet the URAA licensing requirements. minimum access commitments. An efficient TRQ administration method will be In addition to minimum access requirements, one that allows for full utilization of the import the URAA stated that pre-existing market access quota (where all quotas are allocated to importing also had to be preserved. That is, access conditions firms, which then fully use their allocation). Rules for historically established import quantities would such as the tradability of quotas and/or licenses (for be maintained by a provision referred to as "current example, selling or renting) will affect the incentives access." Hence, for a number of products, countries for utilizing TRQs. An understanding of the implica- established TRQs to meet the obligations of current tions of allocating nontradable, country-specific and minimum access. Quotas for these products export quotas and licenses to importing or exporting may be allocated on a yearly, half-yearly, or quar- firms is important. The method of allocating quotas terly basis. can have important implications for the impact of TRQs under minimum access are not always trade liberalization. For example, if export licenses allocated on a nondiscriminatory basis, as was are allocated to high-cost producers, a reduction of specified in the URAA's modalities. Countries have in-quota tariffs may result in increased quota fill, used the lack of rules to fill not only current access whereas an increase in the quota limit may result in TRQs but also, sometimes, minimum access TRQs quota under-fill. Other factors affecting the efficient with imports under preferential agreements. In administration of TRQs include simplicity, trans- such cases, one or a few countries are allowed access parency, and certainty. to the TRQ concerned and can take advantage of This chapter provides a conceptual analysis of the new trade opportunities. When this is the case, the negotiating issues for evaluating options for the it considerably limits the scope of the current func- reform of import quota administration methods in tioning of these URAA provisions in terms of trade agriculture and thereby increasing trade liberaliza- Quota Administration Methods: Economics and Effects with Trade Liberalization 97 tion. Improving the rules on quota administration United States [U.S.]). Accession negotiations by new methods (including STEs) is important as this will members since the Uruguay Round Agreements came make them more transparent, improve fill rates so into force determine their commitments. that imports are not discouraged or blocked, and The WTO has classified tariff quotas by means ensure access by lowest-cost suppliers so that these of broad product categories. Table 5.1 lists the countries are not discriminated against. product categories and the number of tariff quotas in each category (for all members combined). Fruits and vegetables have the highest number (370 The Plan of This Chapter tariff quotas), followed by meat products (258) and The first section below describes import quota cereals (226). Tobacco (13) ranks last among the administration methods. The next section provides 12 categories. Table 5.2 provides a breakdown of an overview of TRQs and WTO rules, using the the total number of tariff quotas by WTO member WTO bananas dispute as an example. This is fol- country, with the number of tariff quotas by mem- lowed by sections that discuss quota fill rates and ber ranging from 1 (Chile) to 232 (Norway). an overview of the basic economics of the alterna- During 1995­2001, the principal administration tive methods and the potential effects on trade. The method of about half of all applicable tariff quotas last section provides conclusions and policy recom- was Applied Tariffs, as shown in table 5.3. For a mendations for tariff quota administration reform. detailed description of each administration method, consult box 5.1. An applied-tariff regime means no quota shares are allocated and imports Alternative Quota Administration are allowed in unlimited quantities at the in-quota Methods tariff rate or lower. The second most frequently used As of 2002, 43 WTO members showed 1,425 individ- method (324 quotas) is License on Demand-- ual tariff quota commitments in their URAA Sched- where import licenses are allocated in relation ules.1 These tariff quotas originate from a number of to quantities demanded and requests are sources, including the Uruguay Round's "tariffica- typically reduced pro rata if they exceed the quota tion" methodology (current-access or minimum- volume--is used 25 percent of the time.2 First- access commitments), in place prior to the conclusion Come, First-Served (FCFS), the third most com- of the 1994 negotiations (such as sugar quotas in the monly used administration method (153 quotas), TABLE 5.1 Tariff Quotas by Product Category Product Category Number of Tariff Quotas Cereals 226 Oilseeds Products 129 Sugar and Sugar Products 59 Dairy Products 183 Meat Products 258 Eggs and Egg Products 21 Beverages 35 Fruits and Vegetables 370 Tobacco 13 Agricultural Fibers 20 Coffee, Tea, Spices, and Processed Agricultural Products 58 Other Agricultural Products 53 Total 1,425 Source: WTO 2002a, 2002b. 98 Agriculture and the WTO TABLE 5.2 Number of Tariff Quotas by Member Country Member Number of Tariff Quotes Members Number of Tariff Quotas Australia 2 Latvia 4 Barbados 36 Lithuania 4 Brazil 2 Malaysia 19 Bulgaria 73 Mexico 11 Canada 21 Morocco 16 Chile 1 New Zealand 3 China 10 Nicaragua 9 Colombia 67 Norway 232 Costa Rica 27 Panama 19 Croatia 9 Phillippines 14 Czech Republic 24 Poland 109 Dominican Republic 8 Romania 12 Ecuador 14 Slovak Republic 24 El Salvador 11 Slovenia 20 EC-15 87 South Africa 53 Guatemala 22 Switzerland 28 Hungary 70 Taiwan, China 22 Iceland 90 Thailand 23 Indonesia 2 Tunisia 13 Israel 12 U.S. 54 Japan 20 Venezuela, RB 61 Korea, Rep. of 67 All members (43) 1,425 Source: Data from WTO 2002a, 2002b. TABLE 5.3 Tariff Quotas by Principal Administration Method, 1995­2001 Number of Tariff Quotas Method 1995 1996 1997 1998 1999 2000 2001 Applied Tariffs 654 642 673 665 656 639 631 First-Come, First-Served (FCFS) 101 103 147 147 146 153 153 Licenses on Demand 290 299 302 298 310 297 324 Auctioning 41 39 59 59 59 52 52 Historical Allocation 68 82 88 97 97 119 100 Imports Undertaken By 22 22 20 19 20 20 21 State Trading Enterprises Producer Groups or Associations 8 8 7 7 7 9 8 Other 12 13 7 7 7 9 8 Mixed Allocation Methods 54 55 58 59 60 60 60 Nonspecified 9 10 6 6 14 7 20 Total Number of Applicable Tariff Quotas 1,259 1,273 1,367 1,364 1,376 1,365 1,377 Source: WTO 2002a, 2002b. Quota Administration Methods: Economics and Effects with Trade Liberalization 99 BOX 5.1 Tariff Quotas: Categories of Principal Administration Methods Applied Tariffs: No shares are allocated to importers. Imports of the products concerned are allowed into the territory of the member in unlimited quantities at the in-quota tariff rate or below. First-Come, First-Served (FCFS): No shares are allocated to importers. Imports are permitted entry at the in-quota tariff rates until such time as the tariff quota is filled; then the higher tariff automatically applies. The physical importation of the good determines the order and hence the applicable tariff. Licenses on Demand: Importers' shares are generally allocated, or licenses issued, in relation to quantities demanded and often before the physical importation is to take place. This includes methods involving licenses issued on a first-come, first-served basis and those systems under which license requests are reduced pro rata when they exceed available quantities. Auctioning: Importers' shares are allocated, or licenses issued, largely on the basis of an auctioning or competitive bid system. Historical Allocation: Importers' shares are allocated, or licenses issued, principally in relation to past imports of the product concerned. Imports Undertaken By State Trading Entities: Import shares are allocated entirely or mainly to a state trading entity that imports (or has direct control of imports undertaken by intermediaries) the product concerned. Producer Groups or Associations: Import shares are allocated entirely or mainly to a produce group or association that imports (or has direct control of imports undertaken by the relevant member) the product concerned. Other: Administration methods that do not clearly fall within any of the above categories. Mixed Allocation Methods: Administration methods involving a combination of the methods as set out above with no one method being dominant. Nonspecified: Tariff quotas for which no administration method has been notified. Source: WTO 2002a, 2002b. allows imports at the in-quota tariff rate until the istered by allocating a large share of the tariff quota is filled. Historical Allocation is a method quota concerned to historical importers, but sig- where the licenses are issued in relation to past nificant shares are also allocated by lottery or imports while Auctioning results in licenses allo- through an "exporter designated" method, both of cated on the basis of a competitive bid system. which fall into the Other category. These tariff Imports directly controlled by STEs (21) and Pro- quotas, and similar cases, were assigned to the ducer Groups (8) are the remaining types of admin- Mixed category since the administration methods istration methods of tariff quotas. Other adminis- are a mix of the Historical Importers and the tration methods (8) are those that do not fall into Other categories. the aforementioned categories, Mixed Allocation Additional conditions are also often specified (60) refers to procedures that are a combination of (box 5.2), such as a domestic purchase require- methods, and Nonspecified refers to tariff quota ment (a condition requiring the purchase of administration regimes that have not been notified. domestic production of the product in order to be Notwithstanding these administrative methods, eligible), limits on tariff quota shares (which limits products can still come into a quota-using country the maximum share or quantity of the quota but the tariff is paid at the out-of-quota rate. This allowed), export certificates (a condition that tariff may be prohibitive. As a result, effective entry requires an export certificate administered by the of the product is usually limited to the TRQ.3 exporting country), and past trading performance In a number of cases, it proved difficult to (which limits eligibility to established importers of identify the principal administration method, the product concerned). About 20 percent of because several methods were in place. For exam- the quotas were subject to additional conditions, ple, most dairy tariff quotas of the U.S. are admin- with 19 countries using conditions. The most 100 Agriculture and the WTO BOX 5.2 Tariff Quotas: Categories of Additional Conditions Domestic Purchase Requirement: An additional condition requiring the purchase or absorption of domestic production of the product concerned to be eligible to secure a share of the tariff quota. Limits on Tariff Quota Shares per Allocation: An additional condition involving the specification of a maximum share or quantity of the tariff quota for each importer or shipment. Export Certificates: An additional condition requiring the submission of an export certificate, certificate of authenticity, certificate of origin, or any kind of export license issued by the exporting country concerned to be eligible to secure a share of the tariff quota. Past Trading Performance: An additional condition limiting eligibility to secure a share of the tariff quota to established importers of the product concerned, although allocations are not made in proportion to past trade shares. No Other Conditions: None of the above was identified. Sources: WTO 2002a, 2002b. frequently used additional condition was "limits on administering agency to fully distribute import tariff quota shares per allocation." quotas to importers, and to do so relatively early in the quota period. Additional rules for operating the TRQ administrative system must be designed to Characteristics of More Efficient facilitate a full, rapid, and transparent distribution TRQ Administration Methods of the quota. Three broad objectives of TRQ administration are In terms of the second aspect, there are many important. The first one, from an international per- means of ensuring that importers holding the spective, is to allow market access opportunities up quota make full use of it. In a market economy, to the full amount of the TRQ level.4 This is how with the intention to preserve the profit motive for the URAA intended to deal with the remaining importers, the quota would be imported as long as high tariffs. The second objective, from a domestic domestic prices are higher than world prices by perspective of making the most efficient use of more than the cost of importation. This can be domestic resources, is to ensure that the lowest-cost accomplished by having private firms receive the firms do the importing. In other words, allocate import quotas or by allowing firms to compete to TRQs to those firms that can make the best use of obtain these import rights. Under these conditions, them by generating the highest profits from the the TRQ system should involve many importers importing activity. Finally, the third objective is to and not create a situation in which there is a operate efficiently--in other words, the TRQ monopoly importer. should not waste the country's resources. Similarly, allowing quotas to be rented or Given these objectives, the following describes bought and sold openly will create strong incen- what would constitute an efficient regime for tives for the firms that obtain the quotas to use administering TRQs by discussing the means for them fully. The application of carefully con- achieving each objective. structed additional regulations can also strengthen the incentives for firms to make full use of their quotas. One example is the widely observed rule, Full Utilization of the TRQ for all types of quota systems, that the quota To ensure full use of the TRQ, many methods and holder must "use it or lose it." The regulation procedures can be followed. The full use of TRQs removes the quota from holders if they do not use has two aspects: the aggregate TRQ for a country an agreed-upon percentage of it. Whether such should be fully allocated to importing entities added regulations are necessary is a separate ques- (firms), and the entities receiving TRQs should tion, but most countries seem to believe so because fully use their allocation. For the first aspect, the this type of regulation is almost universal across main means of ensuring full utilization is for the and within countries.5 Quota Administration Methods: Economics and Effects with Trade Liberalization 101 Giving TRQs to Firms That Make Best quota administration system can be kept as simple Use of Them as possible by minimizing the uncertainty and rule changes associated with the regime. Even if there The second objective is to ensure that those are several rules, if these rules are transparent, well importers receiving the quota are the most effi- publicized, and stable, the uncertainty factor facing cient ones in terms of net profit (lowest costs, quota users is substantially reduced. This is particu- highest revenues). One widely suggested method larly an issue in developing countries where quota of achieving this is to use quota auctions to allo- regimes are often characterized by little informa- cate the TRQ. Although this allocation mecha- tion and lack of transparency--and this can facili- nism is economically efficient, there may be legal tate corruption. WTO issues that inhibit its use. The fee that is An additional useful rule concerns a distribution paid in such an auction, although it is bid by the of quota rents and whether the recipient should pay would-be buyer, could be seen as a breach of the for the quota. (There are possible legal issues sur- tariff binding, the in-quota tariff in this case, and rounding any payment for the quota, beyond the that fee is not related to the cost of import service. in-quota tariff and a cost of service, as already men- Under allocation by auction, those firms that tioned.) The government may judge it desirable to make the highest net importing profit will acquire tax some of the profits (quota rents) accruing to the quota. quota holders. This can be done effectively by an However, other methods can achieve this same auction, or less thoroughly by imposing a tax to end. One effective but overlooked mechanism is to allow quota resale and transfer.6 However the quota acquire the quota. Imposing a tax on quotas has the advantage of generating some public revenue as is initially allocated, if there is a well-developed well as leaving some profits in the hands of the (and legal) market in quota resale and transfer, a quota recipient (although reducing those profits by firm that is unlikely to utilize its quota fully can sell the amount taxed). And the tax can be varied to it and realize the quota profits. In the process, the achieve any desired split in revenues (quota rents) quota passes along to a firm that will necessarily use between the quota recipients and the government it to recoup the costs of buying it. The point is often treasury. One disadvantage is the difficulty in lost that resale provisions will result in the quota knowing, at least at the outset, the amount of the ending up in the same hands (in other words, that tax that will fully extract all the rents. Observations it is as economically efficient) as with an auction. on the transfer price prevailing in private transac- This point has practical importance because many tions can be a guide to the total rents and to an jurisdictions find some reason for not allowing appropriate tax to levy on initial allocations. quota transfer and sale. One advantage of quota taxes is that they have the effect of reducing rent seeking or corruption by those wishing to obtain quotas. Rent seeking induced by a An Efficient TRQ Operating System quota allocation scheme, especially when quota val- The third objective is to have efficient quota admin- ues are high, can make the system extremely ineffi- istration and regulations. This can be accomplished cient in terms of the expense of time and money most effectively by following a basic rule in regulat- spent in lobbying. This can be reduced or prevented ing quota use: keep the regulatory system as simple as by making the receipt of quotas less lucrative by auc- possible. All firms (existing ones or newcomers) tioning them or charging a fee for them that is close should be allowed to acquire the quota; there is no to the auction price. In general, rent seeking can be reason for narrowly stipulating the quota validity reduced by keeping the quota allocation system rules- window (that is, firms should be able to use the full based, with clear reallocation criteria and a mechani- quota whenever they wish within the full quota cal reallocation process with no scope for case-by- period); and buying and renting of licenses should case adjustments or individual subjective judgments. be allowed for all firms regardless of the size of their (Keeping the system rules-based is still consistent facilities. Put differently, the rules state which com- with imposing penalties for quota-holder behavior modity can be imported and that imports must be that is considered undesirable by quota administra- completed by the end of the quota period. The tors. The key issue is that these penalties--which may 102 Agriculture and the WTO result in quota reallocations--be specified in advance terms only allow the import of processed eggs, not and not be discretionary.) table eggs, the TRQ will be valid only for low-valued Another way to make the quota system work egg products. This mix of commitment and adminis- more efficiently is to define two types of quotas-- tration detail effectively reduces the market access for permanent and annual. Tariff rate quotas are usu- the product under the TRQ. To maximize the value ally valid for only one year. In some cases, it may be of market access for a given TRQ, the commitment more efficient for a firm to own the quota outright, should be made across a broad commodity category, so that the amount of quota the firm will have in without further administrative constraint, and the future years is known with certainty. This can be private trade should determine which products to accomplished by defining a permanent quota, import within that broad commodity category. according to which the firm would receive the Country Reserves The country reserve or annual import rights every year in perpetuity (sub- preferential trade issue is one of equity in the distri- ject to the possible future demise of the regime, of bution of TRQs among different countries' exports. course, and subject to "use it or lose it" provisions). But limiting a TRQ to a specific country's exports Yet to have only such a "permanent" quota is less lowers the benefit in terms of trade liberalization of efficient than giving the permanent quota owner the TRQ compared with allowing any country to the flexibility of being able to rent out (or in) some export under that quota, as in the previous case dis- permanent quota from year to year. In other words, cussed. Certain country quota allocations existed an efficient quota system will involve both perma- prior to the URAA, sometimes under preferential- nent quotas (for acquisition for long-term reasons) type arrangements, and these were continued to and one-year quotas or the rental of permanent ensure that those countries would not lose access as quotas (for short-term reasons of fluctuating a result of an importer's URAA commitments markets and general flexibility). Several other issues under the guidelines for establishing current access regarding TRQ systems concern the efficiency of the commitments. quota administrative system, the profitability (size Restrictions Restricting import allocations to of quota rents) of the export opportunities opened industry segments, firms, and product end-uses up or restricted by the TRQ, and the equity of will also reduce the value of the market access rep- quota allocations. This allocation issue is as much resented by the TRQ. In effect, compared with about which entities within a country receive unrestricted, open market allocation of those import rights as it is about which countries gain the imports, such a restriction reduces the demand for right to export into the importing region through them. Although this restriction puts allocations the TRQs. into the hands of those who will use them, the Four issues must be addressed. First, how recipients are willing to pay less to get the alloca- aggregated are TRQ commitments and at what tions than others would be. If that were not the level of commodity aggregation are TRQs admin- case, the restriction would be unnecessary. istered? Second, should TRQs be targeted partly or Consequently this type of restriction has the completely to specific countries' exports ("coun- same effects as do country allocations and, try reserves")? Third, should import allocations arguably, allocating TRQs to state traders. Any be restricted to industry segments, firms, and restrictions on who can use or receive TRQs will product end-uses? And fourth, are there adminis- reduce the demand for and lower the implicit value trative matters concerning the handling of TRQs, of that TRQ, to the disadvantage of would-be such as validity periods and unfilled quota provi- exporters. There are several examples of this sions, that lead to fewer imports or lower-valued kind of TRQ allocation and it has usually arisen imports that lower the value of the TRQ to the for historical reasons, whereby pre-URAA end- exporting country? use allocations have been preserved in the current Aggregation On the subject of aggregation, to TRQ allocations. maximize the value of the TRQs, commitments Administrative With regard to administrative defined as broad aggregates and administered simi- restrictions in the handling of TRQs, such as limited larly are desirable. However, if the TRQ is defined validity periods for the quota and unfilled quota pro- broadly (for instance, as "eggs"), yet administrative visions,the tighter those restrictions,the more costly it Quota Administration Methods: Economics and Effects with Trade Liberalization 103 is to comply and the lower the demand for TRQ is inherently discriminatory. The WTO, established in imports. This could lead to fewer or lower-valued 1995, administers trade agreements negotiated by its imports,or to a reduction in import quota rents (or in members.In the enforcement of Article XIII,fair mar- the implicit value to the importing country of the ket access is the relevant provision and there is no pro- TRQ). This situation can harm the importing country vision relating to quota rents. However, it is the distri- as much as the exporting country, as discussed earlier. bution of rents that drives many trade conflicts and is the source of disputes over TRQs. Tariff quota administration concerns how the Tariff Rate Quotas: GATT, WTO, rights to import at the in-quota tariff are distrib- and URAA Rules uted. This determines the volume and distribution Article 4 of the URAA specifies disciplines for market of trade, as well as the distribution of quota rents. It access, one of the three key provisions or pillars of the is important to keep the distinction clear between agreement.7 Paragraph 2 of Article 4 states that the volume and distribution of trade and the vol- "members shall not maintain, resort to, or revert to ume and distribution of rents. The WTO is con- any measures of the kind which have been required to cerned only with how quota administration influ- be converted into ordinary customs duties." ences the volume and distribution of trade; it has While Article 4 established an obligation to con- no direct interest in the distribution of rents. How- vert nontariff barriers into customs duties, the URAA ever, it is the distribution of rents that motivates the left WTO members considerable discretion over how politics of TRQ administration. The choice of how to effect this conversion. Guidelines or "modalities" to administer a tariff quota becomes a political for establishing tariffs and tariff rate quotas were decision; many competing interests claim entitle- drafted but never formally adopted. Various eco- ments to quota rents, which are on the order of nomic studies show that several WTO members US$48 billion (as shown in chapter 4). engaged in "dirty tariffication," that is, at the time of Historically, four positions have been put the negotiations, they established higher tariffs on forward by various countries: sensitive agricultural products than the suggested method would have allowed. There are also guide- a. Quantitative restrictions are per se inconsistent lines for calculating minimum access volumes, that is, with MFN status (the principle of nondiscrimi- how to determine the 3 percent of base period nation). domestic consumption. These constructed tariffs and b. MFN status requires that each country be in-quota volumes were included in Members' Sched- assigned an equal share of the global quota. ules of concessions and commitments. A window for c. MFN status can be approximated by allotting challenging them existed between the time the coun- the global quota in proportion to the trade shares try Schedules were submitted and when the URAA of current suppliers. was signed. If a member submitted a tariff that was d. Quantitative restrictions should be filled on a "too high" or an in-quota volume that was "too low" First-Come, First-Served basis. and it was not successfully challenged, once it was accepted as part of the Member's URAA Schedule, it Because of conflicting interpretations of MFN was too late to be challenged. The tariffs and in-quota status, there was no consensus, but most agreed volumes became new WTO obligations. These obli- that "global, race-to-the-border quotas (now per- gations are the starting point for disputes over how mitted by Article XIII, GATT 1994) was inconsis- the obligations are implemented and administered. tent with MFN because it unduly favored countries Article XIII of the GATT 1994 "Non-discrimina- with geographical proximity and/or better trans- tory Administration of Quantitative Restrictions" port facilities" (Hudec 1988, p. 178, n. 14). The first governs the administration of quantitative restric- position claims there is no just way to solve the tions, including TRQs. Article XIII can be interpreted quota allocation problem. The second position as being inherently contradictory.It advocates nondis- argues for strict parity. The third position advocates crimination and the use of tariffs rather than quanti- proportionality, defined as the observed volume of tative restrictions, yet it also allows supplier tariff quo- trade in some recent representative period. The tas to be allocated on a historical basis, a method that fourth position asserts (literal) priority in the form 104 Agriculture and the WTO of first-come, first-served. The universal choice of counterfactual distribution. The economic princi- the best tariff quota administration method ple underlying the distribution of trade criterion is remains unresolved. the minimization of trade distortions given the Instead of advocating one principle of distributive TRQ constraint. The GATT principle of nondis- justice and proscribing all others,Article XIII allows a crimination asserts that trade shares should be conflicting set of distributive principles. Predictably, determined by the relative efficiency of suppliers this leads to trade conflicts over TRQ administration. and not by alternative, discriminatory criteria. The interpretation is as follows: quantitative restric- Subparagraphs in XIII:2 (c) and 2 (d), on sup- tions are inconsistent with MFN principles; however, plier quotas, are clearly contradictory in advocating if quota restrictions are administered as if they were both nondiscrimination and tolerance (if not advo- tariffs, they can be MFN-consistent. Two means of cacy) of discrimination. The subparagraphs allow administering TRQs as tariffs are auctioning TRQ for "supplier tariff quotas," TRQs that are allocated rights and allowing current TRQ holders to lease to supplying countries and require that "the TRQ rights to other suppliers. Actions and transfers imported product originate from a particular have radically different distributions of rent, but country or source." Thus, they allow importing identical expected distributions of trade. The countries a GATT-consistent means of discrimina- expected distributions of trade are also identical to tion. As for how the supplier tariff quota shares are that generated by a tariff, and thus consistent with apportioned, GATT Article XIII:2 (d) states that MFN principles. agreement should be sought among all interested The economic interpretation of GATT Article WTO members but that if this is "not reasonably XIII advanced here, and in Skully (1999, 2001a), practicable," concludes that the GATT advocates two criteria for judging whether the quotas under TRQs are being . . . the contracting party concerned shall allot properly administered: (1) quota fill and (2) distri- to contracting parties having a substantial inter- bution of trade. Quota fill requires that imports of est in supplying the product shares based upon the in-quota volume be allowed if market conditions the proportions, supplied by such contracting permit. That is, TRQ administrators should impose parties during a previous representative period, no impediments to imports beyond payment of the of the total quantity or value of imports of the in-quota tariff. If apparent profitable arbitrage product, due account being taken of any special opportunities are not realized, it may be because of factors which may have affected or may be the TRQ administration method. Of course, there affecting the trade in the product. may be other legitimate costs that have not been observed, thus zero-fill or under-fill does not neces- The passages quoted in this section have been sarily mean TRQ administration is the cause. the subject of further definition by the GATT in a As for the distribution of trade, GATT Article series of interpretative notes to Article XIII. The XIII, paragraph 2 states: convention has been to use an average of the three years prior to the imposition of a restriction as the In applying import restrictions to any product, representative period. Several disputes have arisen contracting parties shall aim at a distribution of over base periods during which there were other trade in such product approaching as closely as restrictions on trade. The GATT recommends that possible the shares which the various contract- shares be allotted according to the trade shares ing parties might be expected to obtain in the "which would correspond to what could reasonably absence of such restrictions. . . . have been expected in the absence of restrictions." Once again, this is the free trade counterfactual dis- In other words, one determines what the distri- tribution of trade, the operational equivalent of bution of trade (supplier market shares) would be nondiscrimination. if there were no trade restrictions. The allocation of With regard to the meaning of special factors, the the TRQ is then evaluated by how closely the GATT interpretation includes "changes in relative observed distribution of the restricted volume of productive efficiency" that may have occurred since trade (that is, under-tariff quota) approaches the the representative period "as between the various Quota Administration Methods: Economics and Effects with Trade Liberalization 105 foreign producers." Clearly, changes in competitive advantage of potential suppliers of the control advantage are viewed as an appropriate cause for product, the distribution of shares can become reapportioning supplier shares. increasingly malapportioned. The allocation of the Thus, XIII:2(c) and 2(d) instruct member gov- right to export via "country-specific" TRQs is also ernments that they are allowed to transfer TRQ contentious because an exporting firm has the rights to incumbent exporters, but that they should potential to obtain the rents available through bar- do so in such a way as to approximate the free trade gaining power, imperfectly competitive practices, counterfactual distribution of trade. This is not a and/or the issuance of an export license. simple task. The passage above elucidating the term An example of a dispute over the method by "special factors" gives the impression that exporter which country-specific export quotas are allocated shares can be (and, indeed, should be) reallocated is the "Banana Dispute" (WTO 1997a). Exporters in line with changing economic conditions. Logi- such as Ecuador argued that the methods used by cally this reapportionment should be without com- the European Union (EU) in allocating export quo- pensation. If quota rights are granted partially to tas were discriminatory and did not reflect recent compensate for lost market access due to the impo- trade patterns. Export quotas were allegedly allo- sition of a quota, quota rights should go to those cated to some countries but not others with com- suppliers actually harmed by the quota. If a parable or even greater historical trade levels. As supplier-granted quota suffers a loss of competitive upheld by the WTO Appellate Body in the Banana advantage and is incapable of exporting without Dispute, this is inconsistent with Article XIII, the quota rent, the quota clearly no longer denies although the latter rules that importers allocating market access and there is no basis for compensa- export quotas may "seek agreement with respect to tion. It is the lower-cost entrants that are impaired. the allocation of shares of the quota with all other However, once vested with quota rights, suppliers contracting parties having a substantial interest in aggressively defend what they view to be their prop- supplying the product concerned." erty rights to quota rents. If an agreement is not possible, export quotas are There is no case in which this kind of realloca- to be allocated to those countries having a "substan- tion has occurred in accordance with Article XIII. tial interest" based on shipments during a "previous The lack of such reallocations is hardly surprising. representative period." The EU was accused of not First, Article XIII:2(d) instructs the country impos- allocating quotas consistent with exports in the ing the quota to "seek agreement with . . . all other "representative period," with some countries receiv- contracting parties having a substantial interest in ing higher quotas than historical exports and other supplying the product concerned." As share reap- countries less. The EU also allocated export quotas portionment is a zero-sum game from the quota to non-WTO members; provided additional holders' point of view, unanimous agreement amounts to Lomé countries above and beyond that among them is unlikely. Second, the primary reason "required" by the preferential agreement between the government imposing the quota chooses to allo- the EU and African, Caribbean, and Pacific (ACP) cate "supplier quota" is to appease suppliers harmed countries (agreement known as the Lomé Conven- by the quota. In this regard, it is similar to a volun- tion and its successor, the Cotonou Agreement); and tary export restraint whereby the quota-constrained assigned shares to some, but not all, countries that exporter is partially compensated by the transfer of did not have a "substantial interest." rents from the importing country. For example, the The WTO Appellate Body also ruled on the U.S. tobacco, peanut, and sugar TRQs (and some in issue of the EU requiring only some countries to dairy) transfer quota rents from the U.S. to the issue export licenses to exporting firms for the holders in exporting countries of TRQ rights. The country-specific export quota. The EU was found quota rights are nontransferable, and the product to be in violation of Article I of GATT, which delivered in-quota must be the domestic product of requires that ". . . all rules and formalities in con- the exporter. Such compensation might have been nection with importation and exportation . . . be reasonably and nondiscriminatorily apportioned accorded immediately and unconditionally to the when the quota was imposed, but with the passage like product. . . ." Hence, not requiring export of time and changes in the relative comparative licenses for all countries with export quotas was 106 Agriculture and the WTO not in accordance with the MFN clause. Countries The licensing procedures were found to be with export licenses were given preferential bar- inconsistent not only with GATT Article I but also gaining power because the licenses allowed them with GATT Article III (national treatment clause) to extract a share of the quota rents. and Article X (applying different sets of rules) as The Banana Dispute also highlighted the prob- well as with General Agreement on Trade in Ser- lems of allocating import licenses. There are incon- vices (GATS) rules. Even though it is true that sistencies across countries in regard to the period of Article XIII of the GATT is not concerned with the validity for the import license, the size of the distribution of rents, rents cannot be arbitrarily licenses, eligibility requirements for an import distributed in a way that alters competitive condi- license, reallocation of unused licenses, and tions for firms in a discriminating manner. That is requirements for the use of the license. Overall, what the dispute around GATS was all about. Even firms importing from Latin America faced very though the EU had claimed that the distribution complicated licensing procedures in comparison to of quota rent was at its discretion and not within firms importing from the ACP countries. The first the scope of WTO rules, the panel stated precisely group of firms had unnecessary burdens imposed the fact that firms of complainants' origin, which on them; they were deemed to have been treated in were mostly category A operators (firms import- a discriminatory, trade-restrictive, and trade- ing from Latin America), had to purchase licenses distorting manner. These importing firms faced from EU/ACP firms, which were mostly category nonautomatic licensing and had to apply many B operators (firms importing from the EU/ACP times, which often delayed imports (sometimes for countries), to maintain their previous market the first three weeks of every quarter, according to share within the Latin American region. The price claims filed by Ecuador, a non-ACP country). The of these licenses sometimes usurped the entire panel ruled that licensing rules are generally covered quota rent. The European Commission had by Article 1 GATT as ". . . rules and formalities in reported to the Panel that the licensing regime was connection with importation and exportation . . . " designed to "cross-subsidize" bananas of EU and and therefore, the EU was again found in violation ACP origin, which means it intended this distri- of this GATT article. bution effect. To sum up, this panel decision The Agreement on Import Licensing in the makes clear that quota rent cannot be used by the GATT requires that the application process for importing country to manipulate competitive obtaining and renewing a license be as simple as conditions in a discriminating way in service sec- possible, and that all rules and information con- tors that are tied to the supply of the import- cerning the procedures should be published. The restricted good. Licensing Agreement provides for two types of These issues surrounding country-specific import licensing: automatic and nonautomatic. export quotas, exporting firm licenses, and import- Rules applied by importing countries for licensing licensing procedures highlight the problems of dis- procedures should ". . . be neutral in application crimination and exemplify the inefficiencies that and administered in a fair and equitable manner." can arise in the administration of tariff quotas. No licensing procedures should be trade distorting or restrictive and ". . . no more administratively Import Quota Fill Rates burdensome than absolutely necessary to adminis- ter the measure." The WTO publishes average quota fill rates across However, the importing country gets to decide commodity groupings, countries, and quota what is "fair and equitable" and which methods are administration methods (WTO 2002b). These least "administratively burdensome." The Licensing average fill rates can be misleading because of Agreement sets only vague guidelines, many of aggregation, whereby some products within a which are open to the interpretations of the group have high fill rates and others do not because importing countries. In the Banana Dispute, the of market conditions. The fill rates reported are EU's import licensing scheme was deemed to be also biased because trade volume or value is not "highly complex" for imports from Latin America used to weight them (that is, the fill rate of a com- (WTO 1997b). modity with a large volume of trade is averaged in Quota Administration Methods: Economics and Effects with Trade Liberalization 107 with those that are insignificant). Notification pro- they do not take into consideration any "over-fill" cedures also vary, with some countries reporting of a tariff quota. The fill rate represents imports as in-quota imports only, while others report imports a percentage of the volume of the tariff quota as under licenses only. notified in the WTO notifications rather than Data provided by the WTO (2002a, 2002b) indi- the scheduled quantity, in cases in which there is a cate that the fill rates are constant for the difference between the two.8 period 1995­2001 (table 5.4). However, these fill The spread of tariff quota fill in percentage rates are not directly comparable. First, the terms over five distinct ranges of fill rates is total number of applicable tariff quotas varies on a described in table 5.5 for the implementation yearly basis, as can be seen from the last column. period 1995­2001. For all implementation years, This is principally due to (a) the accession of new 52 percent of the tariff quotas for which notified members to the WTO over this period, (b) the data are available fall in the upper range of fill, that opening of additional tariff quotas as a result of is, 80­100 percent. Twenty-seven percent were less negotiations, and (c) the phasing out of certain tar- than 20 percent filled, and 21 percent were between iff quotas. Second, the number of tariff quotas 20 and 80 percent filled. included in, or excluded from, the calculation of fill There is no difference in the fill rate between rates also varies, as shown in the third and fourth global and country-specific export allocations, and columns, because of data availability problems. only a small difference across commodity groups, Nevertheless, with no changes in the fill rates over with fibers and beverages having the lowest fill rates time, it can be argued that imports have at least kept (approximately 40 percent) and oilseeds, sugar, and up with the mandated increase in the minimum tobacco having the highest fill rate of 67 percent. The access quota over the implementation time period. fill rates do vary significantly across countries, with For the purposes of calculating fill rates by prod- Australia, Brazil, and Indonesia showing 100 percent uct category and by member country, simple aver- fill rates and Costa Rica, Malaysia, and the Slovak ages of fill rates have been used. It should be noted Republic below 40 percent. Again, substantive con- that this measure is only a very broad indicator of clusions cannot be reached from these data because fill rates. The simple average does not differentiate of the many caveats regarding how the data are con- in any way between tariff quotas on the basis of size structed, plus the fact that almost 50 percent of the or economic importance. For example, a tariff administration methods are applied tariffs, in which quota of 200 kg. obtains the same weighting as a 100 percent fill rate would not be expected because another with a quantity of 6,000,000 tonnes. Like- the quota is greater than the imports under the in- wise, low-valued products are not differentiated quota tariff binding scenario. from high-valued products. The URAA did not mandate that each quota be As already noted, for reasons of consistency the filled but GATT XIII considers quota fill to be associ- fill rates are calculated up to 100 percent; that is, ated with proper TRQ administration. Nonetheless, TABLE 5.4 Tariff Quotas: Simple Average Fill Rates, 1995­2001 Implementation Simple Average Number of Tariff Quotas Year Fill Rate (%) Included Excluded Total 1995 66 1,028 231 1,259 1996 63 1,081 192 1,273 1997 62 1,166 201 1,367 1998 63 1,134 230 1,364 1999 62 849 527 1,376 2000 60 700 665 1,365 2001 54 71 1,306 1,377 Sources: WTO 2002a, 2002b. 108 Agriculture and the WTO TABLE 5.5 Distribution of Fill Rates, 1995­2000 Percentage of the Number of Tariff Quotas Included 0­20% 20­40% 40­60% 60­80% 80­100% All 1995 23 7 7 7 55 100 1996 27 7 6 7 52 100 1997 28 7 7 7 51 100 1998 27 7 6 7 53 100 1999 28 8 7 6 51 100 2000 30 8 6 6 50 100 Sources: WTO 2002a, 2002b. a low quota fill rate does not necessarily imply ineffi- respectively, protected by tariff quotas. These numbers ciency. There may be unavailable supply or insuffi- underestimate the true importance of OECD tariff cient demand such that the in-quota tariff is effec- quota commodities for at least three reasons: (a) the tive. A fill rate of 100 percent or more does not reported value of production for tariff quota com- necessarily imply efficiency. Filled quotas may occur modities in figure 5.1 is valued at world prices, while even if suppliers are high-cost importing firms or total value of agricultural production includes market export countries/firms, or STEs may have fulfilled price support; (b) many commodities not included as WTO commitments but have imported low-quality tariff quota regimes have policies that are very similar, product or destroyed imports (as in the case for rice if not identical (for example, rice in Korea, sugar in in the Republic of Korea and Japan, respectively). Japan, and fruits and vegetables in the EU); and Either way, inefficiencies in the administration of (c) total value of production in some instances quotas can be associated with fill rates greater than includes services and forestry. Of the total value 100 percent. Independent of export quotas or non- of production at world prices for all commodity tradability of licenses, the method of allocation of aggregates tracked by the OECD,85 percent of the sec- the import license itself can have a direct impact on tors have at least one tariff line with an import quota. the quota fill rate and hence on economic efficiency. An important indicator of administrative ineffi- Inefficiency of Quota ciency is when there is a fill rate of less than 100 per- Administration Methods cent and there are out-of-quota imports (de Gorter and Boughner 1999). Situations such as this raise the The chapter now outlines the incentives for ineffi- question of whether imports will increase with an ciencies with each tariff quota administration type. increase in the level of the quota. In other words, the Tariff quotas create economic rents, so the import- issue is whether the fill rate is proportionate to the ing country must develop procedures to administer quota, or whether in-quota imports are limited by the quota. There are various institutional designs other and non-economic factors, independent of the with respect to the distribution of quota shares. quota level. This becomes an important question Some countries require licenses for trading firms. A when determining the effectiveness of alternative global quota has imports determined by market trade liberalization scenarios. forces (provided there are no biases in the licensing While the numbers of tariff lines under tariff quo- schemes), while country-specific allocations involve tas are few, they cover some of the main commodities the importing country assigning shares to specific produced in the OECD countries.According to OECD exporting countries. In the latter case, WTO rules data,almost 28.5 percent of domestic agricultural pro- state that all substantial suppliers (defined as coun- duction is protected by tariff quotas (figure 5.1).These tries with a market share of over 10 percent) must rates range from a high of 50 percent in Eastern receive a share (see paragraph 7 of Article XXVIII:1 Europe to zero percent in Australia and New Zealand. of GATT 1994). Licenses are often used as a means The large economies,such as the EU and the U.S.,have of administering tariff quotas, and can be assigned 39.2 percent and 26.2 percent of their production, to importing firms or exporting firms, or to both Quota Administration Methods: Economics and Effects with Trade Liberalization 109 FIGURE 5.1 Share of Output under Tariff Quotas 100 80 60 50.0% 49.0% 39.2% 40 28.5% 26.2% 20 13.1% 13.6% 0.0% 0 OECD European U.S. Japan Eastern Australia, Other Other Average Community Europe New Zealand Industrial Developing Source: OECD, Agriculture Market Access Database (AMAD). (such that an importing firm needs to present both Almost 50 percent of the total 1,425 tariff quotas an import and an export license to import authori- scheduled in the WTO are administered by Applied ties). The share of rent going to importing or Tariffs. Licenses on Demand are used 25 percent of exporting countries will depend on the bargaining the time. Higher-cost firms and exporting coun- power between trading firms resulting from any tries are therefore guaranteed quotas, wasting eco- licensing requirements. nomic resources. FCFS is the third most commonly Controversy surrounds quota administration, as used administration method. In FCFS regimes, many quotas remain unfilled. As was shown earlier, firms waste resources racing to, or lining up at, the quota under-fill does not necessarily mean the quota border, so the uncertainty and increased transac- is the cause of economic inefficiency, nor does a tion costs generate economic waste. 100 percent quota fill rate necessarily imply that Historical Allocation is the next most common there is no inefficiency associated with the quota method and has licenses issued in relation to past itself. Rules such as tradability of quotas and/or imports. It is not responsive to changing cost condi- licenses (for instance, selling or renting) will improve tions, allowing inefficient traders to operate. Auc- the incentives for utilizing tariff quotas. The method tioning is one of the least common methods used, of allocating quotas can have important implications yet economists deem it to be the most efficient and for the impact of trade liberalization. High transac- fair method of disbursing licenses. tion costs exceeding the rent or allocating export Imports directly controlled by STEs and Pro- licenses to high-cost producers can result in quota ducer Groups are the remaining major types of under-fill. A reduction of in-quota tariffs may result administration methods of tariff quotas. The in increased quota fill, whereas an increase in the URAA has no precise rules on STEs, and control quota may result in quota under-fill. Other factors of imports by STEs and producer groups can affecting the efficient administration of tariff quotas discriminate across suppliers and reduce total include simplicity, transparency, and certainty, and imports. In a number of cases, combinations many exporters have demanded new rules or guide- or Other Methods are used (for example, U.S. lines for tariff quota administration. dairy quotas are Historical Importers, but lottery 110 Agriculture and the WTO and "exporter designated" fall into the "other" cat- the extra costs associated with importation. Thus, egory. Additional conditions are also often speci- relative to the auction, a license-on-demand fied (box 5.2). administration method generates waste through its The following sub-sections analyze the eco- inability to eliminate the inefficient firms. nomics of quota administration methods in The import quota will be filled if firms are greater depth to assess the effects on rent shares penalized for nonuse of awarded licenses. A good and economic efficiency. example of such a scheme is the EU regime for eggs. Any importer can apply for a license, but licensees are required to pay 20 percent of the value of the Licenses on Demand product in advance once a license is awarded. The Licenses are required to import at the in-quota tariff down payment acts as a deterrent for firms to let under a licenses-on-demand method of quota the licenses go unused--inducing the same result administration. If the demand for licenses is less than as under auctions, through which quota is com- the quota, each firm receives the amount requested pletely filled. But uncertainty and imperfect infor- and the system operates like it would according to the mation with risk-averse firms may result in license FCFS method (see below). If demand for licenses demand that is less than the quota, and conse- exceeds the quota, each firm's allocation of licenses is quently the remaining imports are subject to the reduced proportionately among all applicants, based inefficiencies of the FCFS method described later. on the ratio of each firm's request to the total Inefficiency increases under this scenario with requests. Firms behave strategically by adjusting larger variance of cost structures of importing requests for licenses to be as close as possible to their firms. This is because a higher cost variation in the individual desired quantities of imports. industry implies more high-cost firms. The differ- With no penalty for unused licenses, no bidding ence between the desired quantity and the allocated costs, and knowledge of cost structures among quantity is larger for high-cost firms, generating competing firms common, the lowest-cost firms more unused licenses than otherwise would have receive amounts less than their desired quantities been allocated to lower-cost firms had there been a while the highest-cost firms receive exactly the minimum penalty for unused licenses. amounts desired (Hranaiova, Falk, and de Gorter The above results are established in a one-period 2002). Each firm requests the maximum equal to setting when only bona fide importing firms par- the quota. If the firm receives more licenses than its ticipate in the bidding. As a consequence, no rent optimal level of imports, some licenses will be seeking occurs under licenses on demand. Being a unused. Each firm is allocated the same number of bona fide importing firm ensures a firm the right to licenses, 1/N, where N is the number of trading compete for licenses. Therefore no other firm has firms requesting licenses. This outcome is inde- an incentive to spend resources in the effort to pendent of the underlying cost structures of the obtain this right. However, this is not the case in participating firms. Inefficiency results because practice, where new higher-cost firms will enter the low-cost firms on average get fewer licenses than is industry to be able to bid on licenses. Also, no optimal, and quota under-fill also results. resources are assumed to be expended when com- Allocation of licenses to the high-cost firms peting for licenses, as bidding is assumed to be cost- increases the gap between the desired and actual less. This result is the same for a one-time auction, number of licenses for the low-cost participants. In since only the participants that can bid the highest addition, these allocated licenses are left unused by amount receive and pay for the licenses and there is high-cost firms, which use licenses only up to their no incentive to rent seek. desired levels of imports. This implies inefficiency The presence of fixed costs has an indeterminate in resource allocation. Allocating licenses by effect on economic inefficiency under a license- demand enables high-cost firms to remain in busi- on-demand system. If the fixed costs prevent the ness and achieve license allocation that under an highest-cost importers from bidding, inefficiency efficient outcome would be assigned to low-cost may decrease. Compared with zero fixed costs and firms. The inefficient distribution of licenses across identical variable cost functions among firms, fixed firms generates economic waste of quota rents with costs add to the deadweight cost because they con- Quota Administration Methods: Economics and Effects with Trade Liberalization 111 stitute an additional cost for the firms that would importers may exit the bidding process (the out- not be in business under an efficient auctioning come depends on the elasticity of excess system. The same reasoning applies when there is demand--and excess supply if the entity in ques- an exogenous decrease in the per unit quota rent tion is a large-country importer). If there is no caused by either an increase in the in-quota tariff exit, deadweight costs increase. A limit on the (for tariff quotas) or the world price, or a decrease import licenses received by a firm can increase (but in the domestic price owing to an economic shock. not decrease) deadweight costs but quota under- The absolute level of inefficiency may decrease fill never occurs under the assumptions of this because of an elimination of the highest-cost model unless the total quantity demanded is less importers, but relative distortion for the remaining than the quota. Note that the variance of costs is firms still exists. important in addition to the level of costs among An increase in the quota, on the other hand, importing firms in determining the degree of eco- decreases inefficiency by reducing the relative dis- nomic inefficiency. tortion. The high-cost firms that originally received To summarize, the following factors decrease the their optimal amount will maintain imports at inefficiency of the licenses-on-demand method: their optimal profit-maximizing levels and the additional quota will be transferred to lower-cost · Fixed costs firms. In addition, a decline in the per unit rent · Decrease in per unit quota rent because of: increases the probability that the highest-cost firm ­ an increase in the in-quota tariff will not participate. Hence, trade liberalization by ­ an increase in world price reducing in-quota tariffs or expanding quotas will ­ a reduction in excess demand reduce the inefficiency of the licenses-on-demand · Nonuse license penalties method. When a limit on the licenses received by · Increasing license fees each firm is imposed, inefficiency increases because · Reducing the limit on number of licenses the limit would be more binding on low-cost firms. per firm As will be noted later, this result is in sharp contrast · Allowing trade in licenses to that of nontransferable rents with the historical allocation method. Historical Allocation If marginal costs of importing firms are con- stant, each firm again bids the maximum amount Allocating import licenses to importing firms or equal to the quota and receives 1/N licenses. If fixed "country-specific" export quotas on the basis of costs are also present, deadweight costs increase by historical shares can lead to a waste of global the fixed costs of the high-cost firms remaining in resources if the lowest-cost exporting country or the bidding that otherwise would not be there, and importing firm does not receive the rights to decrease by the deadweight cost of the highest-cost import. Historical import allocation enables firms that exit. A decrease in the per unit quota rent high-cost importing firms and/or high-cost brought about by either an increase in the in-quota exporting countries to operate, leading to the par- tariff or change in world or domestic prices will tial dissipation of quota rents (provided licenses result in a decline in deadweight costs if the highest and quotas are nontradable). If licenses are cost firm is forced to exit the bidding process. The known to be allocated as a share of historical per firm deadweight costs of the remaining higher- imports, firms may act strategically to increase cost firms increase, but total deadweight costs market share. The U.S. company Chiquita is pur- decline. The identical effect on deadweight costs ported to have expanded banana imports in the occurs if a license fee is imposed that forces the EU in 1992 in anticipation of the new Common highest-cost firm to exit. Market Organization for Bananas (CMOB), An increase in the quota through trade negotia- which would have led to import licenses (and tions will have an indeterminate effect on dead- export quotas) being allocated as a proportion of weight costs because the increase causes domestic historical imports. Chiquita's resources were prices to fall (and world prices increase if the party therefore wasted in rent-seeking activities to is a large importer) and as a result high-cost obtain more licenses. 112 Agriculture and the WTO In summary, inefficiency results with high-cost the import market and with different seasons, gen- firms when historical shares is the method of allo- erating higher exporting costs at the beginning of cation because market conditions change over time the quota year. FCFS can also encourage low-value across trading firms or exporting countries, histor- bulk shipments as exporters cannot guarantee cus- ical importers may not be competitive in the first tomers regular shipments of finished products place, and strategic behavior in the form of "rent throughout the year. seeking" may occur. The effects of trade liberaliza- The waiting in line literature implies rent dissi- tion are as follows: a reduction of in-quota tariff pation through rent seeking if the value of the increases quota fill rate, and an increase in the product does not vary with the timing of the quota reduces quota fill rate. These results are the sales--imports in this case (Barzel 1974; Suen opposite of those yielded by Licenses on Demand 1989). Barzel shows that with perfect information because reducing a tariff increases high-cost firm rents are completely dissipated through waiting in participation while increasing the quota reduces line if agents are identical. Firms compete to be high-cost firm shares for a fixed rent (because the first, therefore, by arriving before the market opens quota is less of a constraint to low-cost firms under with the product and forming a queue. The waiting this scheme) and rent also declines. time rises, that is until in equilibrium, the money- plus-time price clears the market. Thus when firms are homogenous, waiting dissipates the rent com- First-Come, First-Served pletely. Every firm arrives at the same optimal A majority of tariff quotas in agriculture do not time--that is, before the supply is gone--and no require import or export licenses (WTO 2002a). unnecessary waiting occurs (each firm chooses an With no property rights assigned to these import optimal arrival time, given arrival times of all oth- quota rents, rent seeking can occur to the extent ers). With zero serving time, the queue forms that rents are dissipated. The WTO reports that immediately such that the marginal valuation almost 50 percent of the quotas are not binding in equals the price plus cost of waiting. that a tariff is operational (that is, applied tariff). Actions to mitigate costs of waiting--such as However, of the remaining administration methods paying storage costs or paying others to wait-- under which the quota is binding, the WTO lists may induce overdissipation of rents (Deacon and 147 official FCFS situations. FCFS, along with Sonstelie 1991). When firms are asymmetric (for Licenses on Demand and Historical Allocation example, with different opportunity costs of time), (accounting for the bulk of the other administra- every firm still waits for the same length of time but tion methods), are allocation methods that require only the marginal firm dissipates the rent com- some form of inefficient rationing (unlike well- pletely. The inframarginal participants are still able defined assignment of quota rights such as those to earn a part of the rent, resulting in underdissipa- determined through auctions). tion of total rents. Suen (1989) argues that the With no import rights allocated to either the extent to which rent is dissipated depends upon the importer or the exporter, an exporter may not risk degree of heterogeneity in individual valuations. the costs of shipping the product and finding that the Both the rent-seeking and waiting-in-line litera- quota has been filled, as this would result in having to tures assume that the marginal valuation (the price pay the costs of storage until the following quota or per unit rent) of the rationed good is fixed. This season, of paying the higher out-of-quota tariff, or will not be the case for FCFS import quotas in agri- of shipping the product elsewhere. The costs for culture where the rent is reduced as imports occur traders of establishing a business relationship over earlier in the year and domestic prices fall as a result. time with importers are also a factor contributing to Rent can still be dissipated in wasteful activities (for quota under-fill. Exporters do not have information example, seasonality in prices--and hence, rents-- on who holds the import license. The FCFS method results in rent dissipation) but more likely they will is prone to wasting resources by concentrating be transferred to (or "appropriated" by) domestic imports at the beginning of the season; increasing consumers because imports are "hurried up," caus- costs for importers that have to store the product; ing domestic prices to fall. Therefore, the economics and discriminating against exporters farther from of FCFS import quotas for agriculture differ from Quota Administration Methods: Economics and Effects with Trade Liberalization 113 these models in the literature because rents are mercial enterprises by market considerations, may endogenous. Skully (1999, 2001a) provides empiri- have no incentive to fill the quota. Another com- cal evidence of imports mostly in the first month plaint is that the STE, especially if it represents pro- and domestic prices falling with FCFS import quo- ducer interests, will choose to limit the quota to tas. In addition, agricultural products will have lower-valued imports within that category or to opportunity costs, including storage costs on ship pay the exporter lower prices for the good in ques- and cost of spoilage. If there is "serving time" (for tion than would be paid under a private market example, clearing customs, unloading, and so on), transaction. Therefore, allocating quotas to STEs importing firms still wait the same amount of time can reduce market access, even with 100 percent fill but come at times of the day (week or month) that rates, by discriminating across countries or choos- differ by the length of the serving time (Chau, de ing low-quality products. Some STEs deliberately Gorter, and Hranaiova 2002). In the meantime, allocate export quotas to higher-cost exporters for domestic producers receive only the world price and political reasons (ABARE 1999), resulting in ineffi- consumers enjoy lower prices for this part of ciencies and inequities. domestic production as well. This continues as long Many factors, as summarized in box 5.4, need to as imports occur. Once the import quota has been be taken into account when analyzing the effects of filled, the domestic price goes up and is above the an importing STE. Take for example an STE that level compared to when the quota was binding in favors domestic producers and controls imports. If traditional analysis. Rents to farmers are now higher the quota rents are given to producers, then it is a compared to the case of a standard import quota "producer group" STE. In theory, an exclusive right but only for the time period after the quota is filled. to purchase from producers and sell in the domestic There can be full rent appropriation by con- market should give monopoly power to such an sumers when the domestic price is being reduced to STE. In practice, political pressure and other factors world price levels as importers compete for the rents may curb this market power and the degree of the by racing to the border. Once the quota limit is power remains an empirical question. Also, there reached, the domestic price rises again but in the remains a problem of distributing the rents to farm- meantime domestic consumers have appropriated ers, which if blended with revenues from sales from the entire quota rent--rents are not dissipated in domestic production would cause an increase in wasteful rent-seeking activities. The same situation domestic production and consequently reduce arises when the product in question is nonperishable welfare to both producers and society. but storage costs are prohibitive. Box 5.3 provides a The control over imports leads the STE to trade summary of the economics of FCFS and the factors off the benefits from owning the quota rents and affecting the degree of rent appropriation versus dis- the loss in producer surplus through competition sipation. Factors affecting the outcome include the from imports. Failing to fill the quota is advanta- degree of perishability, storage costs, heterogeneity geous only if the quota rents are smaller than the of firms, and time sensitivity of domestic production loss in producer surplus due to increased imports. (some production is continuous and cannot be When relative levels of t1 (in-quota tariff), t2 (out- delayed--for example, milk consumed fresh). The of-quota tariff), and quota are such that quota amount of rent appropriated by consumers from would be binding under perfect competition, the domestic producers depends on the supply and STE will either choose to prevent imports com- demand elasticities, the level of the quota relative to pletely (autarky outcome) or fill the quota (quota- free trade levels, and the level of domestic produc- binding outcome). The relative surpluses of these tion (annual production at world prices) relative to two choices depend on elasticities of demand and the level of imports under free trade. supply and relative levels of TRQ instruments (that is, relative tariff levels, world prices, and quota lev- els), as well as the degree of market power of the State Trading Enterprises STE in the domestic market. The matter of STEs handling tariff quotas remains The out-of-quota inclusive price (Pw + t2) con- contentious. One argument is that the STE, often stitutes an upper boundary on the price the STE less influenced compared to firms or other com- can charge. The STE cannot charge a higher price 114 Agriculture and the WTO BOX 5.3 Tariff Quotas: Economics of First-Come, First-Served (FCFS) Situation 1: nonperishable product with zero storage costs · Race to border and rents dissipated by waiting in line · Money-plus-time price clears the market (all arrive at same time except if serving time, order of arrival is affected) · If heterogeneous importing firms, some inframarginal rents retained · Positive storage (fixed) costs can result in over-dissipation of quota rents Situation 2: perishable product or cost-prohibitive storage · No queuing and domestic price falls to world price Case A: production not time sensitive (instant at world price) · Consumers appropriate all quota rents and most of rents to farmers due to a standard import quota · Imports and domestic consumption at world price and then domestic prices spike normal quota levels Case B: both production and consumption constant flow (for example, milk) · Imports and domestic consumption and production at world price until quota filled; then domestic prices spike above normal quota level · Quota rents appropriated by consumers and producer rents for production at world prices; however, price higher than normal for end-of-year production, consequently producers get more rents than in Case A above Source: Chau, de Gorter, and Hranaiova 2002. because the demand would be filled by out-of- of inferior quality, thereby fulfilling the quota and quota imports. Once Pw + t2 becomes an effective at the same time maintaining income from domes- upper boundary price--that is, at lower levels of tic production, provided that the cross elasticity of t2--it reduces the surplus from the autarky out- demand is low. come and increases the probability of filling the The quota may be under-filled under the STE, quota. A switch to the quota-binding outcome then and the level of underfill depends on the degree of implies a discrete increase in social welfare. market power in the domestic market. The closer A decrease in t1 also increases the benefits to the STE is to a monopoly, the lower the proportion farmers of the quota-binding case as it lowers the of the quota that will be filled. Liberalizing through cost of purchasing the quota and thus increases the lowering t1 increases social welfare, while a decrease quota rents. Trade liberalization through quota in t2 again has no trade and welfare effects until expansion has ambiguous welfare effects. On the Pw + t2 becomes an effective upper boundary one hand, it increases the quota rents by increasing price--that is, Pw + t2 falls below the t1-binding the amount the STE can purchase at the low Pw + price. Further decreases of t2 result in a decrease in t1, but on the other hand the larger amount of producer welfare but no change in the social welfare. imports has a dampening effect on the domestic Positive out-of-quota imports occur in the price. The aggregate effect may be negative for the t2-binding case, which in the case of STE coincides quota-binding outcome and may lead the STE to with perfect competition. That is, when Pw + t2 is switch back to the socially inferior autarky solution. below the quota-binding competitive price, it is If the STE is obliged to fill the quota (market forced to behave competitively, acting as a price conditions permitting), the autarky solution taker at Pw + t2. It is always optimal for the produc- becomes less attractive as the STE is forced to incur ers (domestic) to fill the quota as producers the extra cost of purchasing in-quota imports. gain rents from the in-quota imports. Otherwise, However, demand and supply elasticities may be the demand would be filled by the out-of- such that it pays for the producers (if the STE acts quota imports that do not generate rents for pro- on behalf of producers' interests) to destroy the ducers. Trade liberalization has the usual positive imports rather than include them in the domestic trade and welfare effects as analyzed for perfect market. Or the STE could import a product that is competition. Quota Administration Methods: Economics and Effects with Trade Liberalization 115 BOX 5.4 Tariff Quotas: Factors Affecting the Impacts on Trade with an STE in Importing Country Objective Functiona · Maximize producer welfare · Maximize consumer welfare · Maximize regulatory intermediary welfare Degree of Control · Control of imports · Control of domestic market Obligation to Fill Quota · No obligation to fill quota · Obligation to fill quota if t1 not binding (option to destroy) Appropriation of Quota Rents · Quota rents redistributed to the favored group · Quota rents appropriated by taxpayers · Quota rents accrue to foreign country Relative Levels of t1, t2, and Quota · Effect of STE depends on what instrument is binding initially under perfect competition (quota, t1, t2, or autarky) · Effect of trade liberalization depends on what instrument is binding initially under the STE equilibrium (quota, t1, t2, or autarky) a.The list of objective functions is not meant to be exhaustive and may include job creation, revenue maximization, and the like. Notes: t1 = in-quota tariff. t2 = out-of-quota tariff. STE = State trading enterprise. Source: Hranaiova, Falk, and de Gorter 2002. Many STEs have both monopsony and monop- domestic price. If the STE is obliged to purchase the oly power (for instance, grain STEs in China, espe- quota imports and t2 is binding, the STE will be cially in the past, and BULOG in Indonesia). The indifferent between including the quota in domes- monopsony/monopoly power effectively prevents tic sales and letting the demand be filled by out-of- competitive imports from entering the country at quota imports, provided that the in-quota imports Pw + t2 and allows prices to rise above Pw + t2 as can be destroyed at no cost, or re-exported. much as the residual demand faced by the STE (net It should be noted that there is also some un- domestic demand) allows via "surcharges," as notified state trading cum TRQ activities in the is done in Japan, for example with rice. The Japanese sugar market as reported by Fukuda, domestic price ceiling, therefore, is above Pw + t2 Dyck, and Stout (2002). The Japanese government in these cases. uses sugar import targets and two-tiered "sur- What if the STE could not appropriate the in- charges" while keeping official tariffs low, but there quota rents and could not efficiently redistribute is no official TRQ. There may be other instances of them to producers? It would then never be optimal policies that act like TRQ schemes but are not to purchase the in-quota imports, or in the case of reported as such. the obligation to fill the quota, it would be optimal for producers (with the STE acting on behalf of Lottery producers or a producer organization and hence STE) to destroy the imports. This option is viable An option not discussed in the WTO's documents because no benefits accrue to domestic producers is issuing import licenses by lottery. This does from including the in-quota imports in the domes- occur (some U.S. butter and milk powder import tic sales, while it costs them in terms of a lower licenses, for example) and is efficient in that a firm 116 Agriculture and the WTO cannot affect the likelihood of obtaining the the government authority to importing or export- license a priori. Nevertheless, under this option ing firms (or to both, such that an importing firm each firm would have to comply with application needs to present both an import and an export procedures, and assuming each firm is allowed license to import authorities or to STEs). The share only one draw in the lottery, there are incentives to of rent going to importing or exporting countries break the firm down into many small firms to or firms will depend on the bargaining power increase the probability of receiving a license. Such resulting from any licensing requirements. rent-seeking activities involve economic inefficien- An efficient tariff quota administration cies. Furthermore, high-cost firms may win the method will be one that allows for full utilization lottery and this, if the licenses are nontradable, of the import quota (in terms of quotas allocated results in economic waste. to importing firms and of the latter fully using their allocation). Rules such as the tradability of quotas and/or licenses (for example, sold or Additional Conditions rented) will affect the incentives for utilizing tariff Additional conditions associated with the adminis- quotas. The lack of transparency in the adminis- tration methods discussed above can lead to tration of the TRQs has been magnified by their wasteful rent-seeking activities as well. A domestic variety, which could be reduced by broader prod- purchase requirement increases the cost for some uct definitions. A broader definition of products importing firms that otherwise would not be would reduce tariff dispersion--and yield one tar- involved in domestic production.9 Thus, part of the iff rate for several products under one category. quota rents is dissipated and fill rates would be lower The likelihood of using the URAA special safe- as domestic consumption declines and production guards provision may also be affected by the defi- increases. Limits on quota shares do not allow for nition of broader categories. economies of size and coordination, again resulting The method of allocating quotas can have in the dissipation of quota rents. Limits on quota important implications for the impact of trade shares discriminate against more distant suppliers liberalization. For example, if export licenses are for whom shipload amounts are the economic size of allocated to high-cost producers, a reduction of in- shipment, rather than truckload lots, for example. quota tariffs may result in increased quota fill, whereas an increase in the quota may result in quota under-fill. Other factors affecting the efficient Policy Options for Reforming Tariff administration of tariff quotas include simplicity, Quota Administration Methods transparency, and certainty. The objective of this chapter was to provide a con- To make quota administration rules more trans- ceptual analysis of the negotiating issues for market parent, to improve quota fill rates, and to ensure access through the use of tariff rate quotas, and access by low-cost suppliers, negotiators could con- evaluate options for their reform to improve mar- sider the following options: ket access in agriculture. Governments use different methods to administer tariff quotas, affecting the 1. All countries should adopt auctions of quotas to distribution of quota shares among countries and the highest bidder (lowest-cost trading firms licenses among importing and exporting firms. and exporting countries) and eliminate all other A global quota has imports determined by market methods. The license auction has a significant forces (provided there are no biases in the licensing advantage over other schemes because it ensures schemes), while country-specific allocations involve quotas are allocated to the lowest-cost producers the importing country assigning shares to specific and are fully utilized. exporting countries. In the latter case, WTO rules 2. All quotas and licenses should be tradable state that all substantial suppliers--defined as coun- between firms and countries. This has the same tries with a market share of over 10 percent--have advantages as explained in item 1 above. to receive a share (see paragraph 7 of Article 3. All preferential agreements should be collapsed XXVIII:1). Licenses are often used as a means of into the proposed "development box" put for- administering tariff quotas and can be assigned by ward by a number of developing countries as an Quota Administration Methods: Economics and Effects with Trade Liberalization 117 outcome of the current trade negotiations. This Bergsten, C. F., K. A. Elliott, J. J. Schott, and W. E. Takes. 1987. would ensure that only developing countries "Auction Quotas and U.S. Trade Policy." Policy Analyses in International Economics No. 19. Institute of International benefit--but in a fair way. Economics, Washington, D.C. 4. State trading enterprises (STEs) should be disci- Chau, N., H. de Gorter, and J. Hranaiova. 2002. "Rent Dissipa- plined to provide greater transparency in their tion versus Consumer Appropriation of Rents with First- Come, First-Served Import Quotas in Agriculture." Unpub- quota administration activities. Control over lished manuscript, Cornell University, Ithaca, N.Y. domestic prices and production can indirectly Deacon, R. T., and J. Sonstelie. 1991. "Price Controls and Rent affect imports. The degree to which STEs affect Dissipation With Endogenous Transaction Costs." American Economic Review 81(5): 1361­73. market access also depends on whether they con- de Gorter, H., and D. Boughner. 1999."U.S. Dairy Policy and the trol exports, whether they control imports and/or Agreement on Agriculture in the WTO." Canadian Journal of own import quotas, and whether they feel obli- Agricultural Economics 47(5): 31­42. gated to fill the quota. The objective of STEs is also de Gorter, H., and I. Sheldon. 2000. "Issues in the Administra- tion of Tariff-Rate Import Quotas in the Agreement on Agri- critical, whether they want to help domestic con- culture in the WTO: An Introduction." Agricultural and sumers or producers or maximize quota rents. Resource Economic Review 20 (April): 52­7. 5. All "additional conditions" should be eliminated de Gorter, H., and I. Sheldon, eds. 2001."Issues in Reforming Tar- iff-Rate Import Quotas in the Agreement on Agriculture in in the administration of tariff quotas. the WTO." International Agricultural Trade Research Consor- tium (IATRC) Commissioned Paper 13, University of Min- nesota, St. Paul.iatrcweb. org/Publications/commiss.html. Notes Fukuda, H., J. Dyck, and J. Stout. 2002. "Sweetener Policies in Japan." USDA-ERS Electronic Outlook Report SSS-234-01. 1. This paper draws on two good surveys of the issues USDA Economic Research Service, Washington, D.C. related to TRQs: IATRC Commissioned Paper 13 (de Gorter ers.usda.gov/publications/SSS/sep02/sss23401/ and Sheldon 2001) and the special issue of the Agricultural Hillman, A. L. 1989. The Political Economy of Protection. London: and Resource Economic Review (de Gorter and Sheldon Harwood. 2000). Hillman, A. L., and J. G. Riley, 1989. "Politically Contestable 2. Some license-on-demand regimes allocate licenses on a Rents and Transfers." Economics and Politics 1: 17­39. first-come, first-served basis. Holt, C. A. 1980. "Competitive Bidding for Contracts under 3. FCFS and Historical Allocation have shown the greatest Alternative Auction Procedures." Journal of Political Econom- percentage increases since 1995, suggesting that countries opt ics 88(3): 433­45. for administrative and political simplicity when selecting quota Holt, C. A., and R. Sherman. 1982 "Waiting-Line Auctions." Jour- administration methods. nal of Political Economics 90(2): 280­94. 4. This section relies heavily on chapter 2 in de Gorter and Hranaiova, J., and H. de Gorter. 2002. "The Economics of Tariff Sheldon (2001). Quotas with Imperfect Competition." Unpublished manu- 5. A "use it or lose it" stipulation could have firms importing script, Cornell University, Ithaca. when it is not economical to do so, because the import license Hranaiova, J., J. Falk, and H. de Gorter. 2002. "The Economics has capitalized value for future use. This can cause economic of Administering Import Quotas With Licenses-on- waste. Demand." Unpublished manuscript, University of New 6. There is a strong similarity between the TRQ and produc- Mexico, Albuquerque. tion quotas formerly used in U.S. agriculture (tobacco and Hudec, R. E. 1988. "Tiger, Tiger in the House: A Critical peanuts). For an interesting paper analyzing ways to improve Appraisal of the Case Against Discriminatory Trade Mea- the efficiency of the production quota allocation and the mar- sures." In M. Hilf and E.U. Petersmann, eds., The New GATT ket for production rights, see Rucker, Thurman, and Sumner Round of Multilateral Trade Negotiations: Legal and Economic (1995). Problems. Deventer, Netherlands: Kluwer. 7. This section relies heavily on Skully (2001b). Krishna, Kala. 1993. "Theoretical Implications of Imperfect 8. For a certain number of scheduled tariff quotas, the Competition on Quota License Prices and Auctions." World information notified by Poland covers specific tariff quota sub- Bank Economic Review (7): 113­36. items only, both in terms of notified quantities and notified Krishna, Kala, and Ling Hui Tan. 1998. Rags and Riches: Imple- imports. To ensure consistent treatment among members, these menting Apparel Quotas Under the Multi-Fibre Arrangement. partial data are not included in the analysis. Ann Arbor: The University of Michigan Press. 9. Domestic purchase requirements could have in the past Krueger, A. O. 1974."The Political Economy of the Rent-Seeking been considered a GATT Article XI quantitative restriction, Society." The American Economic Review 64(3): 291­303. which is now prohibited by the URAA. Lott, J. R. 1987. "Licensing and Nontransferable Rents." Ameri- can Economic Review 7(3): 453­55. Select Bibliography McCorriston, S., and I. M. Sheldon. 1994."Selling Import Quota Licenses: The U.S. Cheese Case." American Journal of Agricul- ABARE (Australian Bureau of Agricultural and Resource Eco- tural Economics 76(November): 818­28. nomics). 1999. "WTO Agricultural Negotiations: Impor- Rucker, R., W. Thurman, and D. Sumner. 1995. "Restricting the tant Market Access Issues." Research Report 99.3, Canberra. Market for Quota: An Analysis of Tobacco Production Barzel, Y. 1974. "A Theory of Rationing by Waiting." Journal of Rights With Corroboration from Congressional Testimony." Law and Economics 17(11): 73­95. Journal of Political Economy 103(1): 142­75. 118 Agriculture and the WTO Skully, D. 1999. "Economics of TRQ Administration." Interna- Negotiations. Cambridge and New York: Cambridge Univer- tional Agricultural Trade Research Consortium, Working sity Press. Paper 99­6. WTO (World Trade Organization). 1997a. "European Commu- ------. 2001a. "Economics of Tariff-Rate Quota Administra- nities-Regime for the Importation, Sale and Distribution of tion." U.S. Department of Agriculture, Economic Research Bananas." Report of the Appellate Body, 9 September, Service, Technical Bulletin Number 1893, Washington, D.C. WT/DS27/AB/R. Office of the Secretary General, Geneva. ------. 2001b. In de Gorter, H., and I. Sheldon, eds., "Issues in ------. 1997b. "European Communities--Regime for the Reforming Tariff-Rate Import Quotas in the Agreement on Importation, Sale and Distribution of Bananas, Complaint Agriculture in the WTO." Chapter 3. International Agricul- by Ecuador." Report by the Panel. 22 May, WT/DS27/R/ECU. tural Trade Research Consortium (IATRC): Washington, D.C. Office of the Secretary General, Geneva. Suen, W. 1989. "Rationing and Rent Dissipation in the Presence ------. 2002a. "Tariff Quota and Other Quotas." Background of Heterogeneous Individuals." Jounral of Political Economics paper. March 21, TN/AG/S/5. Office of the Secretary Gen- 97(6): 1384­94. eral, Geneva. Tullock, G. 1980. "Efficient Rent Seeking." In J. M. Buchanan, ______. 2002b. "Tariff Quota Administration Methods and Tar- R. D. Tollison, and G. Tullock, eds., Toward a Theory of the iff Quota Fill." Background paper. March 22, TN/AG/S/6. Rent-Seeking Society. College Station: Texas A&M Press. Office of the Secretary General, Geneva. WTO (World Trade Organization), ed. 1994. The Legal Texts: The Results of the Uruguay Round of Multilateral Trade 6 DOMESTIC SUPPORT: ECONOMICS AND POLICY INSTRUMENTS Harry de Gorter, Merlinda D. Ingco, and Laura Ignacio Introduction effects (classified as "green box" policies). For The objective of this chapter is to provide a concep- developing countries a wider list of policies was tual analysis of the negotiating issues related to gov- exempt and all least-developed countries were ernment-provided domestic support to agriculture, exempt from all subsidy reduction commitments. and to evaluate options in the current World Trade Finally, the URAA provision described as "the Peace Organization (WTO) trade negotiations to reform Clause," and due to expire in 2004, specified that or discipline such government action. In an unprece- green box policies were nonactionable (for exam- dented act, the 1994 Uruguay Round Agreement on ple, from countervailing duties) while amber and Agriculture (URAA) introduced disciplines on blue box subsidies were mostly protected, under domestic agricultural support programs that specific circumstances. encouraged production. Countries made commit- Several outstanding issues on domestic support ments to bind their level of support using relate to its measurement and the URAA disci- the Aggregate Measurement of Support (AMS). This plines. Reduction commitments on amber box provided a measure of domestic trade-distorting policies apply to total agricultural support and are policies (classified as "amber box" policies). not related to any specific agricultural sector. Developed countries agreed to reduce this support Hence, governments can reduce support in one by 20 percent by 2000, and developing countries sector and increase it in another without changing by 13 percent by 2004. Trade-distorting domestic total support for agriculture. The de minimis provi- policies that were below a de minimis standard (5 and sions for domestic support were large; they 10 percent of value of production in developed and have created the potential for the continuation of developing countries, respectively) were exempt in subsidies or an increase in support of commodity two separate categories: non-product-specific and production at high levels. product-specific support. In addition, other The AMS has two components: (1) price sup- domestic support policies exempted from disci- port, which is the difference between an "adminis- plines were certain direct payments with tered" price support and a fixed world "reference" production-limiting characteristics (classified as price; and (2) nonexempt domestic subsidies. In "blue box" polices) and policies that were consid- the first component, the domestic "administered" ered to have no or "minimally" trade-distorting support price does not represent the domestic 119 120 Agriculture and the WTO market price while the fixed-base world reference ative impacts on the environment and for provid- price does not represent the actual world market ing social amenities such as rural landscape. price. The price support element of the AMS also Government support for these "nontrade" concerns depends on the existence of the administered price can actually be disguised protection. Therefore, dis- as some countries simply did not report the refer- ciplines are also required for policies supporting ence price (or abandoned it), but kept the border these multifunctional aspects of agriculture. protection. Hence, there was no change in the level The URAA contains exemptions for developing of trade or prices. In addition, the quantity chosen countries on domestic support but does not provide varied among countries, some using production specific criteria for determining a country's devel- and others' exports. These facts mean that the opment status or details on what domestic support method of measuring the AMS specified in the policies should go in the "development box." URAA--which relies on these measures of domestic For most countries, the AMS has been the least and world prices--can be somewhat inaccurate. binding element of the URAA commitments As will be further elaborated below, the AMS because of an extremely high base period (1986­88) also includes support derived from border protec- upon which reduction commitments were made and tion (such as tariffs or export subsidies), which the ease with which commitments have been cir- results in double counting.1 In addition to includ- cumvented. (The base period support was high due ing a substantial proportion of support already to low world prices and the inclusion of blue box counted in the AMS market access and export sub- support in the base period only.) This, combined sidy support, the baseline AMS is overestimated with the deficiencies of the AMS mentioned previ- because it includes blue box support, but this sup- ously, implies that a major reform will be required in port is not measured in the AMS reduction require- the way the AMS is defined and measured. ments. The AMS is, therefore, a misleading Reform of domestic support is particularly indicator of domestic support and penalizes some important because a significant shift from border countries when policy reforms are undertaken and protection to domestic support has occurred, as favors others when they are not, depending entirely discussed in chapter 4. Furthermore, subsidies on on how programs are designed. inputs and outputs can almost be as trade-distort- The blue box is for support that is directed at ing as import barriers (OECD 2001). The purpose production-limiting programs for payments based of the URAA was to define, quantify, and reduce on no more than 85 percent of the base level of pro- trade-distorting policies, but we argue here that the duction. However, blue box support can be trade URAA has not properly defined and quantified distorting, depending on how the program is (and hence reduced) trade-distorting domestic sup- designed and implemented. port measures in many instances. Hence, a major Green box policies include expenditures on challenge is to address the imbalances in the defini- domestic food aid and food security, environmen- tion and quantification of agricultural support so tal programs, regional assistance, and infrastruc- that a level playing field is established and effective ture investments, and can also be trade-distorting. reductions in trade distortions are ensured. This For example, green box payments that are based will then make it possible to focus on trade-distort- on historical entitlements but require "land to be ing policies, negotiate reductions in their magni- kept in agricultural use" induce higher production tude, and provide an incentive for governments to because fixed costs are covered (which induces less change their domestic policies toward nondistort- exit by farmers); production risk is reduced; input ing measures. market constraints are removed; and expectations are formed for more support in the future (espe- The Plan of This Chapter cially when payment criteria are updated, as in the case of the 2002 United States (U.S.) Farm Bill). This chapter provides an overview of the amber, Rules on eligibility for decoupled payments have blue, and green boxes. The importance of domestic changed in some instances, making them coupled support versus border support is emphasized payments in practice. The green box is also being by presenting the results of the Organisation used for payments to farmers for reducing the neg- for Economic Co-operation and Development's Domestic Support: Economics and Policy Instruments 121 (OECD) Policy Evaluation Matrix (PEM) model around US$40 billion, but the share of trade- for the grains and oilseeds sectors. The theory of distorting domestic support as a percentage of total decoupling is then discussed and its weaknesses are support (including general services and the like) analyzed. The basic economics of the blue box as it has trended downward over time.2 This can be con- pertains to the European Union's (EU) use of it for strued as evidence of a move toward nondistorting area and headage payments is explained, followed support in agriculture among OECD countries. by a section on the specific issues and problems of On average from 1986 to 1991, approximately the AMS such as sector-wide aggregation, double 12 percent of total support is classified as trade- counting, insensitivity to changes in policy reform, distorting domestic support (figure 6.1). Tables 6.2 and the role of de minimis and non-product sup- and 6.3 provide data on domestic support as a port. Only limited attention is given to other issues share of trade-distorting support by country and such as how to reduce amber box support, the sta- commodity, respectively. The domestic support tus and future of the blue box, the definition of share varies widely across countries, with Australia, green box policies, how to accommodate non-trade New Zealand, the Slovak Republic, and the U.S. concerns and multifunctionality (see chapter 8 having trade-distorting domestic support above in this volume), the role and future of the Peace 50 percent. The EU, for example, has a low level of Clause, the role of Special and Differential treat- trade-distorting domestic support, but these pay- ment for developing countries, and the develop- ments exclude blue box expenditures (discussed ment box (see chapter 14 of this volume). later). Among the commodities, wheat, coarse grains, and oilseeds have the highest proportion of trade-distorting support in the form of domestic Brief Overview of Domestic subsidies (table 6.3). Support The composition of the URAA's measure of Carefully constructed estimates of domestic sup- domestic support by box and country is given in port to agriculture are provided by the OECD, and table 6.4. The green box is the largest (undisci- are presented in table 6.1. Much of the total sup- plined) box, followed by the amber box. port in agriculture is either border protection or so-called nondistorting domestic support. Data The Aggregate Measurement of Support (AMS) in table 6.1 show that border support averaged US$160 billion from 1999 to 2001, representing The AMS is calculated for specific commodities 65 percent of the producer support estimate (PSE) receiving "market support," nonexempt direct pay- and less than 50 percent of the total support esti- ments, or any other non-product-specific support to mate (TSE). Payments based on output and input producers. The last of these measures is aggregated use, while significant, represented only US$36.6 into a non-product-specific AMS. Amber box billion, followed by area/headage payments and policies include transfers from consumers such as historical entitlements. The category "other" government-administered price supports and includes payments based on input constraints and taxpayer-funded subsidies for both inputs and out- farm income, and general services such as infra- puts. The accounting method is either government structure and marketing costs. Table 6.1 also shows expenditures or price gaps calculated as the differ- that several countries or regions have large shares ence between a fixed external reference price and of protection derived from border support, espe- the applied administered price using the Equivalent cially Japan, the Republic of Korea, and Turkey, Measurement of Support (EMS). AMS includes while the U.S., the Slovak Republic, Norway, and market price support defined as a gap between Australia have above average shares of domestic the official administered price and the fixed world support. Only 10.2 percent of Australia's total sup- reference price (1986­88 average) times current port is from border protection--the rest is domes- production.3 The fixed world reference price means tic support. that market support as defined by the AMS does Figure 6.1 shows that trade-distorting domestic not change with world prices or with changes in the support (defined only as output and input subsi- domestic market price. Product-specific support dies) has trended up slightly since 1986­88 at that is less than 5 (10) percent of value of production 122 Agriculture and the WTO in developed (developing) countries is not included potential for governments to reach their ceilings may in the AMS--neither is non-product-specific have pressured them to re-instrument their policies domestic support that does not exceed 5 (10) per- towards green box policies. It is important to note cent of the total value of production. that the baseline AMS includes blue box payments. At least 30 WTO members made commitments to However, these payments are excluded from disci- reduce their trade-distorting domestic supports in pline and are therefore not included in the reduction the amber box as measured by their AMS. Members commitments for the amber box support during the without these commitments have to keep within implementation period. 5 percent of the value of production (that is, the The AMS provision was meant to focus on and de minimis level)--10 percent in the case of develop- distinguish distortions in border policies (which ing countries. Table 6.5 summarizes the values for were subject to other reduction commitments) from key countries and outcomes for 1995­2001. The EU, those of domestic support. In the case of taxpayer- Japan, and the U.S. account for over 85 percent of financed government payments, such as loan defi- total domestic support under the AMS. Table 6.6 ciency payments in the U.S. and area payments in the shows that all countries except Argentina and EU, measurement of the AMS is straightforward Iceland are below the baseline for 1995­2001. With a (but the AMS is still unable to distinguish among few exceptions, most countries have changed their policies with different degrees of trade-distorting domestic support policies to comply with the effects). On the other hand, consumer-financed URAA. The U.S. converted amber box subsidies into transfers to farmers in the form of price supports green box policies for crops by eliminating the target pose measurement problems for the AMS resulting price, while Canada made major changes in its in misleading comparisons and conclusions. domestic income support policies. Several countries For example, the EU's intervention price is con- are between 80 and 100 percent of commitment lev- sidered an administered price, so there is market els for 1995­2001 (see table 6.7) including Brazil, the price support for many products in the EU's calcu- Republic of Korea, Slovenia, Switzerland, and lation of the AMS. The reason for measuring mar- Tunisia (last three not shown) (Nelson and others ket price support using administered prices (rather 1998). Although the AMS level is under the limit, the than basing it on market prices) is that WTO FIGURE 6.1 Trends in Domestic Support DS (output + input subsidies) DS/TSE (DS/Total Support Estimate) Linear (DS/TSE) Linear DS 45 14% 43 13% 41 39 12% 37 35 11% Billions 33 10% 31 29 9% 27 25 8% 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Source: OECD Monitoring Report 2002. Domestic Support: Economics and Policy Instruments 123 4.9 22.5 -- 0.8 1.3 63.5 28.2 93.7 -- 147.0 589.7 364.4 120.0 Other 3,178.5 1,023.1 3,624.1 9,261.9 49.0 78.4 -- 20.1 -- -- -- -- -- -- 356.3 597.0 744.9 135.7 1,112.2 Historical 10,085.4 on 13,179.0 Entitlements Based Payments 13.7 92.4 55.0 -- -- 75.4 61.0 -- 17.2 96.6 -- Planted/ 303.5 377.5 530.3 Animal 2,722.1 Numbers 24,733.4 29,078.2 Area and Use 1.1 50.3 19.3 640.3 644.8 113.5 373.7 496.0 83 885.1 491.8 117.6 396.9 Output Input 3,982.4 1,293.7 10,184.3 16,162.0 verage) 36,682.8 A Border 10.2 51.3 55.9 61.1 48.7 48.2 90.4 94.8 63.7 69.9 39.2 69.5 16.6 59.4 78.1 36.4 64.5 (1999­2001 % Protection Millions 96.7 65.6 46.7 48.2 366.1 427.5 891.4 US$ Border 1,997.8 3,624.8 1,162.4 2,588.6 5,092.7 60,863.2 46,973.9 17,233.8 18,662.0 Protection 160,141.2 Countries, 66.7 otalT 946.7 655.3 878.6 135.9 290.7 3,892.1 5,692.6 2,274.1 1,672.7 4,354.4 6,522.0 OECD Producer Support Estimate 99,556.4 51,979.3 18,169.8 51,255.7 248,343.0 in 2002. Support taloT 760 156 162 332 1,376 5,231 1,080 6,999 2,489 1,934 5,047 9,649 Report 64,775 21,489 95,455 Support Estimate 112,628 329,564 Monitoring Agricultural of Republic y Rep. Republic OECD OECD 6.1 Zealand way ce: Australia Canada Czech EU Hungar Iceland Japan Korea, Mexico Nor New Poland Slovak Switzerland urkeyT U.S. otalT Sour ABLET 124 Agriculture and the WTO TABLE 6.2 Domestic Producer Support Relative to Total and Trade-Distorting Support in World Agriculture, US$ Millions (1999­2001 average) (1) (2) (3) Trade-Distorting Domestic Support-DS (input & output price subsidies) TSE TDS (Trade- (total distorting (a) (b) (c) Country support) Support) Total Value Percent of TSE Percent of TDS Australia 1,375.9 263.7 190.7 13.9 72.3 Canada 5,231.5 2,232.3 769.6 14.7 34.5 Czech Republic 759.9 398.4 71.6 9.4 18.0 EU 112,629.9 48,008.0 5,156.0 4.6 10.7 Hungary 1,080.3 690.2 306.0 28.3 44.3 Japan 64,777.5 28,472.3 2,951.9 4.6 10.4 Korea, Rep. of 21,488.8 10,440.8 253.5 1.2 2.4 Mexico 6,999.1 2,840.3 439.0 6.3 15.5 New Zealand 162.2 11.4 11.4 7.1 100.0 Norway 2,489.4 1,101.6 602.1 24.2 54.7 Poland 1,934.3 1,023.4 276.7 14.3 27.0 Slovak Republic 332.3 220.6 138.9 41.8 63.0 Switzerland 5,047.3 2,681.8 690.7 13.7 25.8 Turkey 9,649.5 2,726.7 504.7 5.2 18.5 U.S. 95,455.2 24,969.7 12,817.7 13.4 51.3 All Countries 329,413.1 126,081.3 25,180.6 7.6 20.0 Note: Data correspond to the 29-country coverage of the OECD Monitoring Report 2002 and the 11 commodity groups. Source: Estimates from OECD's PSE/CSE Database and OECD's Agricultural Outlook Database. TABLE 6.3 Empirical Measures of Domestic Support by Commodity in World Agriculture, US$ Millions (1999­2001 Average) (1) (2) Trade-Distorting Domestic Support Trade Distorting (a) (b) Commodity Support (TDS) Total Percent of TDS Beef and Veal 19,736.8 3,318.6 16.8 Coarse grains 8,665.4 4,903.0 56.6 Milk 40,642.6 4,301.1 10.6 Oilseeds 4,897.4 4,502.2 91.9 Pigmeat 9,769.9 1,345.9 13.8 Poultry meat 5,543.5 974.9 17.6 Refined sugar 6,217.0 400.3 6.4 Rice 25,744.4 2,522.8 9.8 Sheep meat 2,026.0 390.8 19.3 Wheat 5,691.3 2,521.0 44.3 Total 128,934.3 25,180.6 28.7 Note: Data correspond to the 15-country coverage of the OECD Monitoring Report 2002 and the commodities covered therein. Source: Estimates from OECD's PSE/CSE database and OECD's Agricultural Outlook database. Domestic Support: Economics and Policy Instruments 125 TABLE 6.4 Composition of Domestic Support by Country, 1995­1998 (percent) Country Total AMS Green Box Blue Box de minimis S&D Australia 11 89 0 0 n.a. Canada 21 51 0 28 n.a. Czech Republic 21 79 0 0 n.a. EU 55 22 23 1 n.a. Hungary 0 39 0 61 n.a. Iceland 82 16 2 0 n.a. Japan 48 51 0 1 n.a. Korea, Rep. of 26 67 0 6 0 Mexico 30 58 0 0 24 New Zealand 0 100 0 0 n.a. Norway 48 18 34 0 n.a. Poland 14 86 0 0 n.a. Slovak Republic 39 1 60 0 0 Switzerland 55 45 0 0 n.a. U.S. 12 83 3 2 n.a. OECD 38 46 12 1 0 Source: WTO 2002. members could not commit to reducing some- well above administered price support in the imple- thing over which they did not have direct control. mentation period (1995­2001), owing to export Setting an administered price at a particular level subsidies on dairy products and import barriers. requires a government decision of some kind.4 As a result, in the 2002 total AMS the U.S. has However, such reasoning leads to difficulties in US$4.3 billion in "unused" dairy support. comparing the AMS across commodities and Figure 6.2 shows how the trend in the AMS is countries because it is conflated with import barri- down even as the total producer support estimate-- ers and export subsidy measures, and conse- as well as trade-distorting support, as shown in fig- quently, the actual market price may not be equal ure 6.1--is going up. Table 6.7 shows the evolution to the support price. of the AMS versus the PSE over time and by coun- The URAA requires each country to identify try. In some instances, the AMS is greater than the market price supports in the form of administered PSE, as is the case for Mexico during 1986­88, prices, which must be included in the calculation of which in principle should not be possible. This the AMS. On the other hand, if there are import occurs even more often when comparing PSEs and barriers in place that keep domestic prices high but AMSs at the commodity level (Diakosavvas 2002). there is no administered price, no market price Looking at the world, the AMS was 71 percent of the support is estimated for the AMS. For example, PSE in the baseline and averaged only 38 percent of Canada has not been able to identify an adminis- the total domestic support reported in the amber, tered price for chicken, turkey, or eggs, so there is blue and green boxes (plus de minimis) for 1995­98 no market price support for Canada's AMS for (WTO 2000).5 It is therefore possible that the dou- these products. Japan simply eliminated an admin- ble counting with border support and the resulting istered price for one commodity, even though the overestimation of the AMS could be such that the commodity's level of protection did not change. sum of the three domestic support boxes plus de This is arbitrary action on whether an official price minimis support exceeds the total PSE. Indeed, for is reported or not. The U.S. reported an adminis- the EU alone in 1996, total domestic support (the tered price support for dairy products but these sum of the three boxes) was US$115 billion as meas- were mostly inoperative because market prices were ured by the WTO, while the total PSE as measured 126 Agriculture and the WTO TABLE 6.5 Use of Total Aggregate Measurement of Support (AMS) Commitments by Member, 1995­2001 (percent) Member 1995 1996 1997 1998 1999 2000 2001 Argentina 144 100 100 99 100 -- -- Australia 27 26 25 23 13 -- -- Brazil 28 35 30 8 -- -- -- Bulgaria n.r. n.r. 1 3 2 -- -- Canada 15 12 11 17 -- -- -- Chinese Taipei n.r. n.r. n.r. n.r. n.r. n.r. n.r. Colombia 15 1 4 3 2 -- -- Costa Rica 0 0 0 0 9 -- -- Croatia n.r. n.r. n.r. n.r. n.r. n.r. -- Cyprus 63 63 45 39 53 43 -- Czech Republic 7 11 7 7 31 35 -- EU 64 67 68 65 -- -- -- Hungary 51 -- -- -- -- -- -- Iceland 79 71 74 178 -- -- -- Israel 71 79 83 66 42 -- -- Japan 73 72 71 18 18 -- -- Jordan n.r. n.r. n.r. n.r. n.r. -- -- Korea, Rep. of 95 93 95 80 83 -- -- Lithuania n.r. n.r. n.r. n.r. n.r. n.r. -- Mexico 5 3 11 14 -- -- -- Moldova n.r. n.r. n.r. n.r. n.r. n.r. -- Morocco 12 33 12 17 24 -- -- New Zealand 0 0 0 0 0 0 -- Norway 71 79 82 88 90 90 -- Papua New Guinea n.r. -- -- -- -- -- -- Poland 6 6 8 8 7 -- -- Slovak Republic 58 59 73 70 66 78 -- Slovenia 93 91 87 97 85 25 -- South Africa 67 82 97 38 38 -- -- Switzerland-Liecht 83 74 72 71 -- -- -- Thailand 72 60 79 79 -- -- -- Tunisia 87 77 81 94 46 -- -- U.S. 27 26 29 50 -- -- -- Venezuela, RB 42 26 36 17 -- -- -- Note: n.r. = not reported. Source: WTO 2002. by the OECD was US$109 billion. If only OECD and therefore are not conflated with market access data are used for both domestic support and the or export subsidy policies. In this way, AMS reduc- PSE, total domestic support exceeds the total PSE tion commitments can have a maximum effect. To for the EU in 1996 by US$12 billion (OECD distinguish policies that have a market impact but 2000b). This erroneously implies that protection are not part of the import access or export subsidy afforded EU agriculture from import access and commitments, it would be fruitful to designate a export subsidies was negative and of the order of new amber box for such policies. At the present $12 billion.6 time, the only way a reduction in the AMS can have Trade negotiations need to single out amber box an impact is if it comes from a reduction in this policies that are truly domestic support policies kind of new amber-type policy intervention, or an Domestic Support: Economics and Policy Instruments 127 taloT 5 AMS n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Current 2001 1 taloT 80 17 53 14 67 34 62 15 AMS 472 969 520 n.a. 361 131 603 113 ment Level 4,301 3,972 1,798 Commit- 5 0 23 otalT n.a. AMS 214 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 134 325 n.a. n.a. n.a. n.a. Current 1,691 2000 taloT 80 71 54 41 2 67 34 27 AMS ment Level 472 969 520 n.a. 366 130 603 n.a. n.a. Commit- 4,301 3,973 1,798 7 2 4 80 62 10 29 48 otalT n.a. 939 n.a. n.a. 135 257 748 n.a. n.a. n.a. n.a. AMS Current 1,552 1999 taloT 81 17 55 14 69 35 27 492 983 520 n.a. 371 136 611 n.a. n.a. n.a. AMS ment Level 4,480 4,138 1,875 Commit- 2 1 4 taloT 81 83 14 10 22 47 98 120 790 n.a. 250 412 767 n.a. n.a. n.a. AMS Current 1,563 1998 taloT 82 71 55 51 72 37 72 AMS 511 997 520 n.a. 377 141 620 n.a. n.a. n.a. ment Level 4,659 4,304 1,952 Commit- 5 0 1 3 taloT 83 14 26 50 12 AMS 132 307 522 n.a. 109 524 n.a. n.a. n.a. Current 3,171 1,937 1997 otalT 83 18 56 15 74 38 28 AMS ment Level 531 635 n.a. 382 146 628 n.a. n.a. n.a. Commit- 1,011 4,838 4,470 2,029 4 0 2 1 taloT 84 36 51 11 144 363 n.a. 619 n.a. 108 501 n.a. n.a. n.a. AMS Current 3,330 1,967 1996 84 81 57 61 76 39 82 otalT AMS 551 650 n.a. 387 152 637 n.a. n.a. n.a. ment Level 1,025 5,017 4,635 2,106 Commit- 0 1 1 taloT 58 37 50 21 123 152 295 n.a. 777 n.a. 124 461 n.a. n.a. n.a. AMS Current 3,508 2,075 1995 Actual taloT 85 18 58 16 79 41 29 570 n.a. n.a. 392 157 645 n.a. n.a. n.a. AMS ment Level 1,039 5,197 4,801 2,183 Commit- ersusV Million Dinar Currency Million Million Million Million Million Million Million Billion Million Million Million Million Million Million Million Million Billion Billion US$ $A US$ ECU Can$ NT$ US$ US$ ŁC Kč ECU Ft SDR US$ Ą Million W US$ Mex$1991 SDR Commitments AMS of aipeiT Rica Republic y Rep. 6.6 Member Argentina Australia Brazil Bulgaria Canada Chinese Colombia Costa Cyprus Czech EU Hungar Iceland Israel Japan Jordan Korea, Lithuania Mexico Moldova ABLET 128 Agriculture and the WTO taloT 8 AMS n.a. 11 n.a. n.a. n.a. 14 n.a. n.a. n.a. n.a. n.a. n.a. Current see 2001 taloT 11 36 10 61 20 62 19 AMS 716 288 ment Level 1998 3,316 2,015 4,257 1,183 Commit- for 0 8 0 taloT 10 16 AMS 155 n.a. 336 439 n.a. n.a. n.a. n.a. figure Current 2000 AMS otalT 11 36 01 62 20 36 91 AMS ment Level 727 288 Commit- 3,329 2,015 4,257 1,200 otalT 0 7 taloT 11 55 92 71 180 n.a. 237 790 n.a. n.a. n.a. AMS Currents Current 1999 12 37 11 64 21 46 02 otalT 737 300 Iceland' AMS ment Level 3,457 2,099 4,434 1,218 to Commit- 0 8 11 65 16 16 01 otalT 126 n.a. 301 820 211 respect AMS Current 3,273 ith 1998 W taloT 12 37 11 67 21 56 12 AMS 748 312 ment Level 3,606 2,183 4,611 1,235 Commit- 0 8 taloT 91 11 60 17 35 6 n.a. WT/Let/292. AMS 296 457 Current 2,198 3,445 and 1997 taloT 13 38 11 70 21 66 21 758 324 AMS ment Level 3,745 2,267 4,789 1,252 Commit- 1998. 0 7 taloT 11 66 13 15 6 for 250 n.a. 227 331 AMS G/AG/N/ARG/4 Current 1,938 3,663 see AMS 1996 13 38 21 72 22 76 22 otalT otalT AMS 769 336 1995 ment Level 3,883 2,351 4,966 1,270 Commit- of 0 7 taloT 94 10 70 16 95 6 Current n.a. 255 542 figure AMS Current 1,640 4,287 AMS 1995 taloT 14 39 12 75 22 76 adjusted 23 779 348 AMS ment Level 4,022 2,435 ,1435 1,287 otalT and Commit- Currents unadjusted Currency Million Million the Million Million Million Million Million Million F Million Million Million Million Million Argentina' DH NZ$ NOK US$ US$ Sk ECU R Sw B D US$ US$ to regarding 2002. Guinea RB respect (Continued) ith New Republic Africa W WTO 6.6 Zealand way ce: Member Morocco New Nor Papua Poland Slovak Slovenia South Switzerland-Liecht. Thailand nisiauT U.S. enezuela,V Notes: G/AG/N/ISL/14 Sour ABLET Domestic Support: Economics and Policy Instruments 129 (3) 3.7 0.1 0.1 0.0 0.0 6.3 1.1 0.0 7.2 2.4 8.8 10.3 94.0 31.1 21.4 (2)/(1) 1998 (2) 2.0 5.2 6.2 0.1 7.3 PSE 31.9 11.4 17.6 37.8 20.0 12.4 48.6 102.3 227.6 256.7 (1) 0.1 0.5 0.0 0.1 0.0 0.0 5.9 1.1 0.4 0.0 1.4 0.3 2.3 AMS 10.4 22.5 (3) 4.3 8.6 0.4 0.1 0.0 0.0 1.3 0.0 8.1 4.5 (2)/(1) 10.6 33.2 20.3 17.1 447.2 Billions. US$ 1997 (2) PSE 2.3 4.4 7.9 9.1 5.9 0.2 6.6 7.2 92.7 67.2 19.2 31.6 18.4 30.7 231.8 (PSE), 0.1 0.4 0.0 0.1 0.0 0.0 2.0 0.4 0.0 1.5 0.3 2.4 6.2 (1) AMS 26.2 39.6 Estimate (3) 5.1 9.0 0.2 0.1 0.0 0.0 1.1 0.0 9.2 3.2 13.0 41.7 19.8 17.5 (2)/(1) 472.6 Support 1996 (2) PSE 2.2 5.0 8.2 6.5 0.2 7.1 7.1 26.2 91.7 18.8 11.0 17.7 29.9 104.5 254.1 (1) AMS 0.1 0.5 0.1 0.1 0.0 0.0 2.4 0.1 0.0 1.6 0.2 3.0 5.9 30.6 44.6 Producer and (3) 5.0 0.2 0.1 0.2 0.0 0.0 8.4 5.4 (2)/(1) 10.2 13.8 ­2.7 48.4 27.2 19.5 524.1 (AMS) 1995 (2) PSE 2.3 5.5 9.3 7.1 0.2 4.7 7.5 24.7 96.1 93.8 19.4 ­7.9 18.4 22.9 271.2 Support of (1) AMS 0.1 0.6 0.0 0.1 0.2 0.0 2.7 0.2 0.0 1.5 0.3 3.6 6.2 37.3 52.8 (3) 1.9 0.7 2.5 Measure (2)/(1) 22.1 54.8 94.9 21.7 23.4 11.7 42.8 57.1 69.8 467.7 5282.6 ­1648.3 (2) PSE 1.8 7.5 7.9 7.2 9.7 0.9 0.1 7.9 62.9 85.0 ­0.6 18.0 41.8 1986­88 123.9 238.9 Aggregate calculations. of (1) 0.4 4.1 1.2 0.9 0.2 2.1 9.6 0.2 2.1 4.2 3.4 AMS 80.7 33.8 23.9 166.8 applicable. Secretariat Evolution not Com. = y Republic y States OECD 6.7 Zealand n.a. way ce: OECD Countr Australia Canada Czech European Hungar Iceland Japan Korea Mexico New Nor Poland Switzerland United All Note: Sour ABLET 130 Agriculture and the WTO FIGURE 6.2 Trends in AMS versus PSE (1986­2001), US$ Billions 300 250 200 150 100 50 0 1986­88 1995 1996 1997 1998 1999 2000 2001 Source: Diakossavvas 2002. AMS PSE increase in market access or a reduction in export more meaningful reduction in trade distortion subsidies (because AMS measures are often con- arising from amber box policies and prevent coun- flated with the latter type of commitments). tries from shifting support from sector to sector as Another AMS issue requiring attention is that market conditions change. the world price is fixed at the 1986­88 base period and as a result, the AMS does not vary with world Blue Box prices but instead with changes in domestic pro- duction and support prices. Consequently, if the The blue box is an exemption from the general world commodity prices continue their decline, the rule that all subsidies linked to production must implied support from administered price supports be reduced or kept within defined minimal (de (with consumer transfers and border protection) minimis) levels. It covers payments directly linked will increase, but measured support in the AMS will to acreage or animal numbers, but under schemes not. This may have implications for the effective- that also limit production by imposing produc- ness of AMS reduction commitments as negotiated tion quotas or requiring farmers to set aside land. in the URAA. Only the EU, Iceland, Norway, Japan, the Slovak The aggregation of all policies and commodity Republic, Slovenia, and the U.S. have used these sectors (including sector-wide policies) into a sin- subsidies. The U.S. has since changed its blue box gle AMS has limited the effectiveness of amber box policies so that they fall in the green box cate- policy reductions. Countries have flexibility not to gory. Blue box payments were included in the reduce support in some sectors and even increase it baseline AMS only and were excluded from the in other sectors. The incentive to do this is con- AMS for the implementation period. The result is strained somewhat by the Peace Clause, which that the base AMS is overstated, which makes it states that policies are exempt from WTO policy easier for countries to reach their AMS reduction challenges such as countervailing duties, provided requirements. that support does not exceed the 1992 levels for that The distinguishing feature of blue box policies particular sector (see box 6.1). Support reduction is payments based on fixed acreage, yields, or live- commitments on a policy type and commodity sec- stock numbers on no more than 85 percent of tor basis rather than on a single AMS for all policies baseline production. Paying farmers not to pro- and sectors should be considered. This will secure a duce can be trade-distorting by either reducing Domestic Support: Economics and Policy Instruments 131 BOX 6.1 The Peace Clause and Domestic Subsidies The "Peace Clause" (Article 13 of the URAA) Agricultural subsidies hitherto have been shel- precludes most WTO dispute settlement chal- tered under various conditions from the applica- lenges against a country in compliance with the tion of several WTO provisions on subsidies. Agreement, but ends after 2003. Many agricul- However, amber and blue box policies have not tural subsidies will then be vulnerable to legal been entirely immune from countervailing duties challenge under Article XVI of GATT 1994, Arti- and GATT 1994 Article XVI:13 even with the cles 6.3(a)­(c) and 6.4 of the WTO Agreement Peace Clause in effect. The dispute between New on Subsidies and Countervailing Measures Zealand and the U.S. and Canada over the (SCM) or with claims of nullification or impair- impact of Canadian dairy policy, together with ment. Nonsubsidizing developing countries will the complaint by Brazil over U.S. cotton subsidies then be able to bargain using dispute settlement and the challenge by Australia, Brazil, and Thai- panels (or direct negotiations) to contest the land of the EU sugar regime reflect the reduced compatibility of domestic farm policies with protection from actionability relative to green these more strict disciplines generally available box and export subsidies. to nonagricultural sectors in the WTO. The rem- For a detailed analysis of the Peace Clause and edy would require elimination of the subsidy or its implications, read Richard H. Steinberg and reduce its adverse effects. The various subsidy Timothy E. Josling (2003), "When the Peace provisions of the WTO, therefore, represent Ends: The Vulnerability of EC and U.S. Agricul- potentially powerful disciplines on agricultural tural Subsidies to WTO Legal Challenge," Journal subsidies. of International Economic Law, Spring. production or even increasing production (the lat- continue to remain in agriculture. Hence, the EU ter because farmers have an incentive to stay in compensatory payment scheme appears to be at production and even expand in order to receive best only partially decoupled. Similar conclusions more payments in the future). The EU employs could be made for the original (that is, pre-1996 production controls (production quotas and Farm Bill) policies of the U.S. that were included in acreage constraints including acreage set-asides) the blue box but have since been transferred to the for major field crops, and deficiency payments are green box with the use of the production flexibility given for sectors such as oilseeds and sheepmeat. payment program. Acrerage premia are given for area set-asides and EU compensatory payments are blue box and, areas that are planted. There are slaughter premia although regarded as trade-distorting-type policies, for beef cattle and calves and "headage payments" are exempt from reductions. With the Peace Clause for beef and dairy (with the latter requiring land set to expire and blue box policies under considera- density targets to protect the environment). Even tion in the current trade negotiations, it is impor- though there are acreage constraints, production tant to understand the trade-distorting effects of may be distorted because farmers may increase blue box provisions for the EU relative to that of production by using inputs more intensively. decoupled taxpayer-financed programs such as his- Farmers may also be encouraged to plant more torical entitlements in the U.S. area--up to the ceiling--than they otherwise Area (formerly called compensatory) payments would. Acreage payments are based on a fixed for cereals totaled 14.6 billion Euro in 1999. regional base area but individual farmers still have However, acerage set-aside (both a mandatory an incentive to increase acreage to maximize their minimum and a voluntary set-aside with a maxi- share of the regional payments. It will be shown mum) along with set-aside payments (1.3 billion later that the EU program is fully coupled in Euro in 1999) restricted to a base level of acreage decisions, rendering the supply-reducing acres (with penalties if exceeded) are designed to impact of the base acreage constraint ineffective. reduce or even nullify the trade-distorting effects of A similar situation occurs for livestock headage such producer subsidies. Nevertheless, acreage set- payments and all payments that require farmers to aside has a cost, in that it can reduce production at 132 Agriculture and the WTO the same time by idling a valuable resource, as has expansion of animal numbers initially and then to been the experience with the U.S. Conservation lock in production near levels that are consistent Reserve Program (CRP). with the maximum number of animals eligible There is a need to identify the impacts of direct for payments. Those numbers reflect the very high income or "compensatory" payments for acres levels of support for several decades as well as the planted and diverted by cereal producers in the EU. incentives inherent in the headage payments Commentators argue that these payments fulfill (Roberts 2003).7 several criteria of Annex 2 of the URAA defining non-trade-distorting subsidies in that they are Green Box taxpayer-financed and do not provide price sup- port to producers (Cahill 1997; Guyomard, Baudry, Green box policies include decoupled payments and Carpentier 1996; Moro and Sckokai 1998). (which purportedly do not affect production However, payments are based on acres planted and decisions) and general policies such as environ- farmers are obligated to produce cereals on those mental programs, research, food aid, crop insur- acres. Because payments are made on a fixed base ance, and income safety net programs to correct acreage, economists thus far have assumed that the for market failures. These policies are generally effects are akin to those of inframarginal produc- taxpayer-funded and do not involve consumer tion subsidies. transfers. However, area payments in the EU are not Green box policies are supposed to meet the cri- decoupled because they are made on an aggregate teria of "no, or at most minimal, trade-distorting fixed area base that is set at the national or regional effects or effects on production." It will be shown level. Individual farmers do not have a base area-- later that the green box also encompasses some just eligible acres for which to receive payments-- policies that are not fully decoupled. and have area set-asides. If the regional base area is The URAA uses the term "decoupled" only in exceeded, the per unit subsidy is prorated down- connection with direct payments to producers ward proportionately for all farmers. This means called "decoupled income support." These provide that the area payments are fully coupled in plant- domestic support exempt from reduction commit- ings because there is no constraint for an individual ments. The issue is whether the rationale for the farmer not to plant up to the level of production at policy is to correct for market failures or to redis- which marginal costs equal market price plus the tribute income.8 If the rationale for the policy is to per unit subsidy (de Gorter, Hranaiova, and Tsur redistribute income, the policy should be designed to 2000). The penalty of the reduced subsidy due to fulfill the criteria for decoupling. Policies correcting the aggregate base area being exceeded is applied to for market failures, on the other hand, inevitably all farmers, even if another farmer or other farmers require a change in production (for example, pollu- are responsible for the excess acreage. Area pay- tion reduction policies), so these polices are not ments with a national base area are therefore not a required to meet the definition of decoupling as pre- limit on total acres planted. sented in the URAA. Of course, politicians like to For EU cattle, the headage payments that are mix the two motives in their call for multifunctional- under "production limiting" arrangements are ity. Hence, the URAA definition of decoupling may anything but production limiting because: (a) claims need to be expanded for other motives of subsidies for payments are the only limiting feature, not pro- to ensure minimally trade-distorting policies (see duction; and (b) the number of animals eligible is discussion below). not limited to numbers on farms prior to the pay- The green box also includes programs that are ments being instituted in 1992. Where numbers of not directed at particular commodities or prod- animals were below the maximum that could be ucts, such as income insurance programs, as well claimed per farm, farmers have had an incentive to as environmental protection and regional devel- expand their stock of animals held up to the opment programs. Table 6.4 provides the compo- limits on which payments are made. As such, the sition of domestic support with green box incentives in the program have been to encourage support as the largest category. Expenditures by Domestic Support: Economics and Policy Instruments 133 TABLE 6.8 Total Green Box Expenditures by Category, US$ Millions, 1995­98 Total Green Box Expenditures (US$ Million) Green Box Category 1995 1996 1997 1998 Research 3,251.5 3,428.9 2,660.3 713.1 Pest & Disease Control 1,339.7 2,705.7 620.1 368.3 Training Services 2,330.7 2,047.4 623.7 385.3 Extension & Advisory Services 2,620.4 3,264.5 2,923.8 597.8 Inspection Services 326.2 660.9 295.6 140.1 Marketing & Promotion Services 991.9 1,008.3 201.8 75.0 Infrastructural Services 28,507.3 23,664.4 18,834.0 3,179.5 Other General Services 6,020.1 4,173.6 3,514.2 364.5 Non-separated General Services 9,396.8 7,625.5 7,475.0 242.6 Public Stockholding for Food Security 2,927.3 1,375.4 1,202.8 264.5 Domestic Food Aid 40,771.1 40,131.2 37,718.2 351.5 Direct Payments to Producersa 339.8 349.4 0.0 0.0 Decoupled Income Support 2,586.5 7,316.4 7,254.3 793.4 Income Insurance/Safety Net Programs 39.8 53.9 8.3 0.0 Natural Disaster Relief 1,766.7 1,401.0 1,096.9 765.8 Producer Retirement Programs 1,505.5 2,035.2 757.3 7.4 Resource Retirement Programs 3,167.7 3,589.5 1,706.8 7.9 Investment Aids 12,065.7 9,847.2 3,751.3 2,019.5 Environmental Programs 5,237.9 7,459.4 2,292.0 1,056.5 Regional Assistance Programs 4,205.9 4,571.7 685.3 539.9 Other Measures 3,916.0 383.4 320.0 289.0 Total Green Box, of which (in %): 133,331.7 127,091.9 93,980.9 12,161.8 Shares by Category (in percent) Total General Services (para. 2 of Annex 2) 41.1 38.2 39.5 49.9 Total Public Stockholding & Domestic Food Aid (paras. 3­4 of Annex 2) 32.8 32.7 41.4 5.1 Total Direct Payments (paras. 6­13 of Annex 2) 23.2 28.8 18.7 42.7 Total Other Measures 2.9 0.3 0.3 2.4 Member Countries Covered 46 47 43 26 Note: "Direct Payments to Producers" relates to measures covered by the second sentence of paragraph 5 of Annex 2 of the Agreement. Source: WTO 2002. category of green box support are shown in of whether the programs listed in Annex 2 are truly table 6.8. non- or minimally trade-distorting, how green box Green box policies may require tighter defini- definitions can be tightened, what the criteria tions for the requirement that "no production shall should be to reduce the loopholes provided for be required in order to receive . . . payments," as continuing output-increasing subsidies, and stated in Annex 2 of the URAA. Do payments whether another category of green or nonaction- that require that eligible farmers keep their land in able subsidies should be established. Stricter rules, agricultural use meet that condition? The current definitions, and monitoring arrangements will be negotiations will have to address the questions required. 134 Agriculture and the WTO -- -- -- -- -- -- -- AMS 1997 0.1 otalT of % 1996 %1 0% 0% 0% 2% 0% 0% 0% minimus de %2 2% 0% 1% 1995 28% 0% 0% 0% -- -- -- -- -- -- -- 1997 2.45 minimus de w/o 1996 5.9 -- -- -- -- -- -- 2.81 AMS otalT 6.2 0.57 3.01 0.21 3.63 1.56 1995 62.13 37.29 -- -- -- -- -- -- -- 1997 0.28 minimus -- -- -- -- -- -- 1996 0.04 0.07 de taloT 0.11 1.06 0.13 0.22 0.04 -- -- -- 1995 -- -- -- -- -- -- 1997 2.73 2.38 AMS -- -- -- -- otalT 1996 5.94 2.87 2.96 1.64 Support 6.31 0.79 3.06 0.21 3.63 1.56 1995 61.07 37.42 minimis 2002. de of y Rep. WTO 6.9 way ce: Countr U.S. EU Japan Canada Korea, Mexico Switzerland Nor Sour ABLET Domestic Support: Economics and Policy Instruments 135 Finally, de minimis subsidies are defined as those According to OECD estimates of income transfer that fall within small limits. The levels of these sub- efficiency (that is, the actual amount that farmers sidies are summarized in table 6.9. Many countries receive from consumer and taxpayer transfers), no have expressed a general willingness to reconsider support policy linked to agricultural activity suc- de minimis levels for developing countries and pos- ceeds in delivering more than half the monetary sibly transition economies (most are bound by de transfers from consumers and taxpayers as addi- minimis levels rather than AMS reduction commit- tional income to farm households. In the case of ments). A possible loophole in the AMS are de mi- market price support due to border protection and nimis rules that allow for an exemption of 5 percent production subsidies, the share is one-fourth or (10 percent for developing countries) of total value less; for input subsidies it is less than one-fifth. On of production for each commodity plus another average, only 25 percent of producer support actu- 5 percent (10 percent for developing countries) for ally finds its way into the producer's pocket. Most of non-commodity-specific support. In the case of the market price support is paid back to input sup- Canada in 1995, de minimis support accounted for pliers, capitalized into land values, or compensates 28 percent of the total AMS (table 6.9). Unlike the for the market-depressing effects of government AMS itself, the de minimis clause operates on an policies (that is, offsetting the world price declines individual commodity basis only. It is not possible owing to all countries' policies). Hoekman, Ng, and to offset support above the de minimis level for Olarreaga (2002) find that high levels of protection one product by transferring it to another com- and domestic support for farmers in developed modity that is below its de minimis level. countries significantly affect many least-developed countries, both directly and indirectly through the price-depressing effect of agricultural support The Economics of Domestic policies. In assessing the impact of reducing Support and Trade Distortions domestic support for a sample of 119 countries, significant differences are found in the impact of a The Effect of Domestic Versus Border Support 50 percent cut in tariffs and a 50 percent cut in on Trade Distortion domestic support for developing countries com- Economists generally argue that domestic support pared to developed countries. However, for both is less trade-distorting than border protection groups tariff reductions have a much greater posi- because the latter reduces consumption as well. tive effect on exports and welfare. In the following But because the supply curve is very likely to be analysis, the relative trade-distorting effects of elastic and the demand for food inelastic in devel- domestic versus border support are analyzed using oped countries, the difference in trade-distorting the OECD's PEM model for grains and oilseeds.9 effects between domestic and border support may not be large. Analysis by the OECD (2001) using PEM Analysis Border support is implemented the PEM model for grains and oilseeds indicates through import tariffs/quotas and export subsidies that border protection does not distort trade much while domestic support refers to subsidies on out- more than does domestic support. Input subsidies puts and inputs that are classified as amber in the on purchased inputs, for example, can have a large WTO. Other payments such as historical entitle- distortion on production. This is more likely with ments or area payments depend on historical land a more elastic supply curve for purchased inputs uses and may be classified as green or blue. and a more inelastic supply curve of other inputs There is a great deal of variation in the relative (for instance, land). If purchased inputs are also a reliance on market price support derived from bor- large share of input costs, if the value of returns to der protection and from direct payments to provide scale in agricultural production is close to one, and support across countries and even across com- if the substitutability between inputs is high, then modities (wheat, coarse grains, oilseeds, and rice) an input subsidy will be more output distorting. within a country (table 6.10). All six countries/ These conditions do characterize much of agricul- entities in the model (Canada, Mexico, the U.S., the tural production. EU, Switzerland, and Japan) use domestic support. 136 Agriculture and the WTO TABLE 6.10 2001 Baseline Support by Commodity Region and Type (US$/tonne) Canada Mexico U.S. EU Switzerland Japan Border Wheat 0.00 39.85 0.00 0.83 107.90 109.73 Coarse grains 0.00 38.84 0.00 6.27 113.52 878.98 Oilseeds 0.00 42.37 0.00 0.00 653.01 0.00 Rice -- 20.80 0.00 90.31 -- 1,565.40 Domestic Wheat 15.96 54.97 68.77 83.53 165.75 980.28 Coarse grains 12.85 46.93 28.97 77.84 101.10 135.84 Oilseeds 35.81 173.82 53.70 131.60 812.02 840.56 Rice -- 47.04 81.38 39.53 -- 197.14 Source: See Appendix A. Support levels per tonne are lowest in Canada in all away from the commodity receiving extremely high cases. Domestic support for oilseeds tends to be levels of border support (rice and coarse grains) high in all countries except the U.S. and Japan. toward the commodity still receiving extremely Canada and the U.S. used no border protection high levels of direct payments (wheat and oilseeds) and the EU used very little border protection except and vice versa. The opposite happens for removing for rice. Mexico had moderate levels of border pro- direct payments only. Production of wheat there- tection for all four crops; Switzerland had high lev- fore increases by 0.9 mmt in this scenario because els of both border protection and direct payments input market adjustment from the even greater for wheat and coarse grains, and extremely high changes is affecting rice and coarse grains. border protection and direct payments for oilseeds. The world price of coarse grains increases by Japan employed extremely high border protection 3 percent while the price of oilseeds and rice combined with high direct payments for rice and increases by 1 and 7 percent, respectively. The coarse grains. It used the reverse combination for resulting patterns of adjustment are fairly easy to wheat and oilseeds--extremely high direct payments follow. The commodity most affected is rice because with high levels of border support. production in two regions, Japan and the EU, Table 6.11 shows the effect of removing border declines dramatically. Rice production increases in protection (setting the market price support the U.S., which relies on extensive direct payments [MPS] = 0). The world price of wheat increased by support for rice, and in the rest of the world. less than 0.5 percent because only a very small The most significant impact for Mexico is in amount of border protection was used in 2001. terms of wheat and coarse grains (maize) since OECD countries had net exports in 2001 without these are the staple crops. Reduction of MPS extensive use of export subsidies. The three countries reduces production by 20 and 18 percent, respec- that relied on MPS for wheat--Mexico, Switzerland, tively. Switzerland also has a large reduction in and Japan--accounted for 4.6 million tonnes (mmt) of production following the pattern of its MPS. production, which is less than 2 percent of world pro- The impact of removing domestic support on duction. Wheat production in Mexico and Switzerland increased world prices (from 5 percent for rice to 18 declined by 20 and 21 percent, respectively. percent for oilseeds) is greater than that of reducing Unbalanced baseline support patterns are very border support (0 percent for wheat to 7 percent for important in affecting the outcome of reducing just rice)--compare the last column of table 6.11 to that one type of support (border protection or direct of table 6.12. Two primary reasons account for this: payments). Removing border protection in Japan domestic support is higher (table 6.10); and direct results in movement of productive resources payments based on use of inputs with elastic supply Domestic Support: Economics and Policy Instruments 137 TABLE 6.11 Effect of Removing All Border Protection on Effective Producer Price and Production Canada Mexico U.S. EU Switzerland Japan ROW Effective Producer Price Wheat 0% 21% 0% 1% 32% 9% 0% Coarse grains 2% 25% 2% 4% 43% 78% 3% Oilseeds 1% 14% 1% 1% 77% 0% 1% Rice 0% 8% 4% 32% 0% 80% 7% Production Wheat 1% 20% 1% 1% 21% 128% 0% Coarse grains 3% 18% 3% 6% 48% 96% 0% Oilseeds 0% 4% 0% 3% 97% 66% 0% Rice 0% 2% 6% 46% 0% 90% 3% Note: ROW = Rest of World. Source: See Appendix A. TABLE 6.12 Effect on Effective Producer Price and Production of Removing All Direct Payments Support Canada Mexico U.S. EU Switzerland Japan ROW Effective Producer Pricea Wheat 13% 10% 8% 11% 4% 68% 12% Coarse grains 9% 4% 8% 11% 5% 1% 16% Oilseeds 15% 16% 7% 14% 3% 43% 18% Rice 0% 8% 35% 2% 0% 5% 5% Production Wheat 5% 37% 5% 9% 18% 100% 0% Coarse grains 1% 3% 5% 9% 22% 66% 1% Oilseeds 9% 15% 16% 4% 14% 43% 5% Rice 0% 15% 74% 10% 0% 6% 2% a. The effective producer price is the world price adjusted for transportation and quality differentials (OECD world price for each country) plus the MPS rate plus the rate of output price support in the form of direct payments. Note: ROW = Rest of World. Source: See Appendix A. curves generate larger increases in production small relative to the transfers that take place (OECD 2001). among the categories. Reduced deadweight loss Table 6.13 shows the impacts of agricultural from eliminating direct payments is US$4.2 billion, trade liberalization according to how each major US$11.9 billion from eliminating MPS, and US$14.9 category of people participates in these markets (as billion from eliminating both. taxpayers, consumers, farm operators, land owners, Deadweight loss is primarily a result of the very and other input suppliers) and by region. large distortion of consumer prices in Japan. The Economic efficiency is measured by deadweight consumer benefit of eliminating MPS is US$22.7 bil- loss and determined by the total welfare sum across lion. There is a benefit of US$1.7 billion in Mexico. the six countries. As expected, this number is As Japan, Mexico, and Switzerland rely on MPS, the 138 Agriculture and the WTO TABLE 6.13 Effect of Trade Liberalization (US$ Millions) Welfare Canada Mexico U.S. EU Switzerland Japan Sum No Border Support Taxpayers 4 450 133 202 194 3,039 3,618 Consumers 53 1,704 551 580 583 22,658 24,921 Other Farmer Owned 16 186 214 114 34 3,218 3,322 Land 22 494 320 218 28 2,718 3,115 Hired Labor 0 0 8 16 2 0 9 Purchased Inputs 13 272 189 248 36 2,579 2,932 Welfare Sum 25 302 47 187 290 11,105 11,925 No Domestic Payments Taxpayers 806 1,444 16,288 17,461 175 3,437 39,611 Consumers 504 135 4,598 3,116 8 158 8,519 Other Farmer Owned 111 4 243 43 13 547 738 Land 307 1,140 8,559 13,111 106 1,005 24,228 Hired Labor 0 0 27 10 1 0 37 Purchased Inputs 91 51 617 637 22 608 1,844 Welfare Sum 197 115 2,245 543 25 1,119 4,244 100% Liberalization Taxpayers 806 846 16,288 17,446 63 1,142 34,181 Consumers 555 1,028 5,139 2,526 502 22,325 15,636 Other Farmer Owned 127 170 20 132 36 3,669 3,900 Land 285 1,471 8,324 13,252 147 2,906 26,385 Hired Labor 0 0 18 26 2 0 46 Purchased Inputs 104 296 431 872 45 2,958 4,498 Welfare Sum 198 263 2,356 638 210 11,651 14,989 Notes: Mexico: 1991 U.S. dollars, n.a. not available. Source: See Appendix A. The Theory of Decoupling benefits of MPS elimination in Japan are more or less evenly split among farm operators, landowners, Re-instrumentation or change away from border pro- input suppliers, and taxpayers (through tariffs col- tection to domestic support and partially decoupled lected), but the benefits to all these are about half support accelerated in the mid-1980s, especially in the the cost to consumers. A similar pattern of benefici- major field crop sectors. For example, the U.S. intro- aries is seen in both Mexico and Switzerland. duced a "freeze" on payment yields in the 1985 Farm The main beneficiaries of eliminating direct Bill. The EU followed with similar direct payments in payments are taxpayers, particularly taxpayers in the major reform of the Common Agricultural Policy the U.S. and the EU because of the size of these (CAP) in 1992. Mexico replaced price supports with regions and because of their reliance on direct pay- decoupled payments in 1994, as did the U.S. in the ments in the form of historical entitlements and Federal Agricultural Improvement and Reform Act of acreage payments. In Mexico, Farmers Direct 1996. Aided by a World Bank loan, Turkey replaced Support Program (PROCAMPO) payments result price supports and input subsidies by phasing in a in a large transfer from taxpayers to landowners. decoupled payment scheme beginning in 2001. Consumers benefit somewhat from these policies Decoupling is of interest to developing countries and from other purchased input subsidies and defi- because a number of mostly middle-income countries ciency payments that reduce world prices. support agriculture. If effective decoupling is adopted Domestic Support: Economics and Policy Instruments 139 worldwide, world commodity prices will increase. For shape and the position of the supply and net commodity exporter developing countries, higher demand curves should not be changed. prices,due to decoupling would increase their share of The concept of full decoupling is more restrictive. world markets, increase rural income, and reduce Full decoupling requires that there be no impact on dependence on tropical commodity exports. the underlying supply response and adjustment A broadly accepted definition of decoupling in process--not just effects on equilibrium quantities, as production is when transfers to farmers are made in the first definition. It should be noted that measur- on inframarginal output only or are paid independ- ing equilibrium quantities is very difficult in practice. ent of current or future production. But the concept The URAA defines decoupling in terms of policy of decoupling has different meanings among econo- design, whereas the OECD defines decoupling in mists, policymakers, and trade negotiators. Some terms of policy effects. Policy design and policy define it in terms of trade distortions, others in effects are obviously interconnected but not terms of production distortions alone. Decoupling the same. Goodwin and Mishra (2002) offer an is sometimes defined only for those programs that alternative definition of a fully decoupled payment are designed to transfer income to producers; others whereby the payment level is fixed and guaranteed define it in terms of all programs including those, and thus is not influenced by ex post realizations of for example, to improve the environment. Still oth- market conditions (for example, low prices or area ers define decoupling to include the way in which yields). This is a definition in terms of rules, like the policy is implemented--that is, there is also no that of the URAA, but is more restrictive in that change in the underlying production structure. payments cannot change. Sometimes commentators evaluate the degree of Decoupling is important because most of the decoupling to include the dynamic, long-term existing income transfer mechanisms fall short of effects of a policy in increasing output owing to the meeting one of their main objectives: boosting policy's impact on such factors as uncertainty, farm household income. Three sources explain the investment, and expectations. gap between intended and actual transfers: (a) eco- The URAA uses the term "decoupled" in con- nomic inefficiencies; (b) distributive leakages, nection with taxpayer-financed payments to pro- whereby some of the benefits of support accrue to ducers that satisfy the following conditions (URAA groups (such as input suppliers) other than the Annex 2.6): intended beneficiaries; and (c) the self-defeating determined by clearly-defined criteria such effects of all countries' policies in lowering world as income, status as a producer or landowner, fac- prices. As a result, a large amount of support in tor use or production level in a defined and fixed most OECD countries has been transferred to pro- base period, and the amount . . . shall not be ducers but average farm household income has not related to, or based on, the type or volume of pro- increased as much. A major problem is that much duction . . . undertaken . . . in any year after the of the transfers go to farm households that do not base period . . . nor on the price, domestic or need them because the measures are still predomi- international, applying to any production under- nantly based on production or factors of produc- taken in any year after the base period. tion that fail to change the income distribution in any significant way. The OECD defines a policy to be effectively fully From an efficiency point of view, budgetary decoupled if: payments that are decoupled from agricultural activ- . . . it results in a level of production and trade ity altogether would transfer income to selected equal to what would have occurred if the policy farm households most efficiently. Economic distor- were not in place . . . The shape of the supply tions and distributive leakages would be minimized, or demand curves could be changed . . . even if providing a compelling case for further reform of the equilibrium production and consumption agricultural policies, and for increased market open- are not changed. ness. The benefits would be substantial, and include reduced costs to consumers and taxpayers, improved or to be fully decoupled if: trade opportunities for competitive suppliers, less . . . it does not influence production decisions stress on the environment, and more effective policies of farmers receiving payments . . . both the to achieve goals. 140 Agriculture and the WTO In order to design agricultural policies that are fixed per farm (based on historical production, decoupled from production, it is necessary to which means farmers do not have to produce to get understand the mechanisms through which decou- payments--for example, U.S. production flexibility pling may affect the farmer's production decision. contract payments).10, 11, 12The first three of these Decoupled payments may affect the farmer's pro- categories require production in order to receive duction decision in a dynamic dimension or in the decoupled payments. Chau and de Gorter (2001) presence of uncertainty. Producers make decisions model several consequences of domestic support about future production and what investments to payments such as (a) inducing exit or entry; undertake based on expectations of future prices, (b) biasing production incentives in domestic mar- policies, and output. Several mechanisms or cate- kets; and (c) cross-subsidizing production and gories of incentives can be identified whereby pro- exports. duction increases as a result of direct payments to Consumer-financed inframarginal production producers: fixed costs, risk reduction and wealth subsidies may also be referred to as decoupled pay- effects, government credibility problems with ments. Like milk and sugar quotas in California expectations about changes in future decoupled and the EU, respectively, and supply management payments or criteria to receive them, and imperfect programs in Canada, consumer-financed inframar- input markets. ginal production subsidies can act like a production subsidy because higher domestic prices imply an Fixed Costs Even if direct income payments are expansion in production owing to the role of pay- on inframarginal output only, fixed costs can be ments in improving farmers' ability to cover costs.13 offset. Often market revenues do not cover total Taxpayer-financed decoupled payments in the costs. But the addition of government payments U.S. are not linked to current market prices and are covers full economic costs in many cases. Hence, presumed to have little or no direct effect on pro- production beyond the amount for which the pay- duction decisions. Furthermore, program benefits ments are determined (for example, base or quota) are tied to the land,14 and farmers have flexibility in can occur because government payments deter what and how much to plant, except for some limi- exiting from farming or crop production. tations on planting fruits and vegetables. There are Taxpayer- or consumer-financed inframarginal implicit costs to remaining eligible to receive pay- (or decoupled) production subsidies to farmers ments, however, in that recipients of decoupled can effectively cross-subsidize production through payments may not use contract acreage for nonagri- their effects on farmers' ability to cover fixed and/or cultural commercial or industrial purposes. variable costs. When production is required to Decoupled payments can also distort production receive the payments, the farmers' ability to cover because, to receive these payments, farmers are these costs is improved, thereby allowing a farmer required to keep the land in"agricultural uses."This who would otherwise have to exit the industry to serves as an incentive to prevent some land from instead stay in business, and perhaps even expand being converted to nonagricultural use. In addi- output beyond the quota or base acreage. tion, as the schemes are currently designed in some Four broad categories of policies can be identified countries, farmers may not get direct payments if that encompass most "direct farm income payment they plant fruit and vegetable crops on their histor- schemes" in agriculture worldwide: (a) inframar- ical base area. ginal income payments financed by consumers (for Chau and de Gorter (2001) analyze the distribu- instance, the EU's sugar quotas, California's fluid tion of fixed costs across individual wheat farms in milk quotas in the U.S.); (b) inframarginal tax- the U.S. The output and export consequences of payer-financed income payments (for instance, for- three policy scenarios are studied: the removal of mer U.S. crop policy with fixed payment yield and coupled payments, removal of decoupled payments, base acreage); (c) inframarginal direct income pay- and removal of both. The removal of coupled pay- ments with fixed per unit production subsidies ments alone decreases exports by 56 percent financed by taxpayers (for example, EU oilseeds while the removal of decoupled payments alone and cereals); and (d) inframarginal income payments reduces exports by 41 percent. Hence, decoupled financed by taxpayers with the income payments payments can have a relatively large impact on Domestic Support: Economics and Policy Instruments 141 the exit decision on low-profit farm units. The and income, the degree of risk aversion on the part aggregate impact on output of the removal of both of the farmer, and the level of subsidy provided. can remain quite limited so long as the output level Because de Gorter and Tsur evaluated two different of the marginal farm is relatively small. If existing stabilization policies in Canada and Hennessy eval- income payments (both coupled and decoupled) uated an input subsidy for the U.S., changes would generate expectations of future payments that com- not necessarily be expected in the absolute level of pensate short-term losses, the reservation profit of output to be equal between the two studies. But it is the marginal farm may take on a negative value and very interesting that both studies find the relative the aggregate output and export distortion of magnitude of the stabilization effect on output decoupled payments can accordingly be consider- to be about three times as great as that of the sub- ably larger. sidy effect. The latter is usually debated in interna- tional trade negotiations, while the stabilization Risk and Wealth Agricultural production is char- effects on output because of farmers' risk aversion acterized by a high degree of uncertainty; therefore are ignored. the policies discussed here may affect a risk-averse The intuitive explanation for the empirical farmer's production decision by either mitigating result that the stabilization effect can have three the farmer's response to risk or reducing the times the subsidy effect revolves around the rela- amount of risk.15 tive magnitude of the mean versus variance effect. Two sources of output response can be identified A policy may not change the mean price very with a reduction in uncertainty as a result of direct much but stabilizes the price considerably. This government payments: depends on the design of the government pro- gram, the variability of market prices, and the level 1. The income effect, whereby higher incomes due of risk aversion on the part of farmers. It is possi- to direct government payments make farmers ble that the subsidy component is relatively large less risk-averse, thereby leading them to assume and the reduction in price variance very low more risk and increase production with uncer- (because of the design of the policy or because tainty in the marketplace. there is not very much variability in the price in 2. The risk effect, whereby direct payments reduce the first place). In this case, the subsidy effect the farmer's perception of the variability of farm would be expected to be higher than the stabiliza- revenue and hence degree of risk. The degree of tion effect. risk is only affected if total income varies with Young and Westcott (2000) evaluate the U.S. market prices and subsidies compensate at least crop and revenue insurance programs that provide partially for price declines. catastrophic coverage and subsidize premiums, reimbursing costs of selling and servicing policies, Hennessy (1998) describes how programs in the and underwriting risk protection. Government U.S. affect both the variance of total income and insurance reduces the variability of expected farm risk aversion of farmers, thus affecting their output revenues by reducing risk associated with crop decisions. Empirical studies by de Gorter and Tsur production variability. Subsidies encourage greater (1995), Hennessy (1998), and Mullen and others participation in insurance programs and conse- (2001), provide estimates of the absolute effects of quently increase hectares planted because of the the mean (subsidy) versus variance (stabilizing) greater level of revenue and lower variance (in effects of direct government payment programs in other words, again, the mean/variance compo- Canada and the U.S. In all of these studies, the vari- nents of risk-reducing policies). Because subsidies ance effect on output was approximately three are calculated as a percentage of the total pre- times that of the mean effect. The absolute change mium, which varies across crops and farms to in total output was 3 percent in the Hennessy study reflect different risks of loss associated with each and 10 to 20 percent in the de Gorter and Tsur crop and insurable hectare, premium subsidies are study. The impact of these direct payments on higher for production on riskier crops and absolute levels of output depends on the level of hectares. This subsidy structure favors production price or income stabilization, the variance of prices on hectarage with higher yield variability and thus 142 Agriculture and the WTO encourages production on land that might not support. But the credibility of the government is otherwise occur. now in question because of the additional emer- Young and Westcott report that crop insurance gency programs added since 1996 and the 2002 subsidies averaged US$1.4 billion during 1995­98, Farm Bill, which adds additional commodities to resulting in an increase of 600,000 hectares planted, the price support program and allows both pay- with wheat and cotton showing the largest percent- ment yields and base acreage to be updated. When age gains. Beginning in 2001, U.S. crop insurance future agricultural policies are considered as a subsidies almost doubled to about US$3.0 billion whole policy package evolving over time, new pro- owing to the reform passed by Congress to improve grams adding to old programs confirms an impres- the "safety net" for U.S. farmers. sion that one needs to be a farmer to receive these benefits. It appears politically impossible to design truly decoupled programs such as that purported Expectations About Future Policies and Dynamic in the last two U.S. Farm Bills without affecting Considerations Most disaster payments and "emer- farmers' expectations for future subsidies in the gency market loss" payments are made after the pro- event of adverse market conditions. Uncertainties duction decision has been made. Nevertheless, over future farm legislation in the U.S. about rein- producers will develop expectations of future assis- stating direct payments cause farmers to stay in tance based on past government actions, thereby business and produce cereals even if it is not eco- affecting current production decisions. As Barry nomical, because of the capitalized value of holding (1999) comments: "It is not hard to plan on govern- a base for future payments. ment assistance when it comes so easily."Furthermore, when market prices decline and "emergency market loss" subsidies increase, farmers build that into their Imperfect Input Markets Fully decoupled support expectations. Hence, production is inevitably in a static framework will not affect the farmer's increased not only through fewer exits and more entry investment decision if markets are perfect. of farms into crop farming such that land is not other- However, if input markets are imperfect, even wise left idle or used for nonagricultural purposes, but decoupled income support will affect the decision also through acreage expansion by existing farmers. through the investment mechanism. Farmers will A combination of a coupled payment and a therefore have an incentive to retain more resources quantity constraint that achieves an effectively in agricultural use and possibly produce more decoupled outcome will thus not qualify for the than justified by current market conditions. Hence, green box under the heading "decoupled income current policies can affect future production. support." However, this policy under current rules Direct payments can affect farmers' investment may well qualify for the blue box, which would and exit decisions if there are constraints facing them exempt it from the requirements to reduce domes- in capital and labor markets. For example, credit con- tic support. This would, for instance, be the case if straints because of imperfect information may pre- farmers expect the government to update the base vent farmers from investing under normal market year used for determining the amount of support conditions. However, direct payments allow banks to each farmer receives. This problem is explicitly make loans that they otherwise would not, thereby dealt with in the URAA, as green box decoupled allowing farmers to stay in agriculture and even income support payments may not be based on increase their investment in production capital and "output, prices or factors used in any year after a capacity. Another example of how imperfections in fixed base year" (OECD 2001). Taking dynamic input markets can result in direct payments affecting aspects and uncertainty into consideration suggests output is through the labor market. Farmers have additional channels through which agricultural specialized skills, and therefore have a difficult time policies may affect the farmer's production decision. finding off-farm employment that would remunerate For example, after the U.S. Federal Agricultural them enough to exit farming or to farm part-time. Improvement and Reform Act of 1996, American Government payments allow farmers to stay in agri- farmers were to receive generous transition pay- culture and thus avoid any costs associated with par- ments to compensate for the removal of the price ticipating in nonagricultural labor markets. Domestic Support: Economics and Policy Instruments 143 Toward an Ideal Decoupling Scheme Concluding Remarks and Options for the Negotiations The primary motivation for decoupling is to compensate farmers with transitional adjust- The URAA has not defined and quantified trade- ment assistance to free markets while at the same distorting measures in an optimal manner. Several time making the move politically palatable and categories of support are inappropriately catego- transparent. The ideal would be compensation rized as nondistorting. A major reform is needed in programs that are universal (open to all sectors the way in which the "aggregate measurement of in the economy, not just agriculture) or at least support" (AMS)--a measure for trade-distorting nonsector-specific within agriculture, such as domestic support policies--is defined and meas- the all-farm income or gross revenue stabiliza- ured. These trade-distorting policies were assigned tion programs used in Canada. Decoupling on a to the amber box category of support. URAA sector-specific basis is clearly a third-best option reduction commitments were supposed to measure because it is sector-specific within agriculture, is domestic support, independent of support due to not based on tax data (so targeting is not import barriers and export subsidies. In reality, achieved), and there are no effective limits per however, the AMS is double-counted with support farm. A non-product-specific payment scheme derived from trade policies. Total support as meas- covering all farmers dilutes the maximum pay- ured by the OECD's "producer support estimate" ment per farm and as a consequence minimizes (PSE) is often less than domestic support, as meas- inefficiencies and inequities. ured by the AMS. If each of these calculations Several features are necessary to make a decou- measured what it was supposed to measure, this pling policy work at the sector level (if universal or would not be possible. Hence, a new amber box whole-farm income insurance schemes for all of should be created that includes only domestic sup- agriculture are not available). The simplest way is a port that is trade-distorting (with perhaps adjust- one-time unconditional payment to all in farming ments downward for output-reducing measures), or considered in need of compensation (equal to and is not conflated with trade border measures. the difference between current income and income The method of measuring domestic support under free markets) as an annuity (bond), which is through the AMS is somewhat misleading and nontransferable to successors (Beard and Swinbank penalizes some countries. Furthermore, some green 2001). Any other scheme should meet the following box government payments induce higher produc- conditions: tion because fixed costs are covered (which induces less exit by farmers); production risk is reduced; · No production is necessary. input market constraints are removed; and expecta- · Land, labor, or other inputs do not have to be in tions are formed for more support in the future. "agricultural use." Hence, a major challenge in the current negotiations · Eligibility rules are clearly defined and not is to address the imbalances in the definition and allowed to change. quantification of agricultural support so that a level · A permanently fixed base period is established. playing field can be established and effective reduc- · Maximum number of years is required as a tions in trade distortions are consequently ensured. farmer to get payments (or if the farmer reaches The baseline AMS is overestimated not only because retirement age). it includes a substantial proportion of support · Payments are capped and vary inversely with already counted in market access and export subsidy farm size, and farm size or numbers cannot be measures, but also because it ignores the fact that manipulated. farm prices often diverge from the administered · Any other coupled support for agriculture is price support defined in the URAA. eliminated. In general, reduction commitments in the · Level of payments in aggregate and per farm URAA suffered from over-emphasis on border sup- and the terms described above must be bound port in relation to the amber box and under- in the WTO to prevent changes in policies by emphasis on green box support. With the end of governments. the URAA implementation (2001), the Peace 144 Agriculture and the WTO Clause set to expire on January 1, 2004, and new · "Administered" or "official" domestic prices negotiations, it is important to put the issues asso- rather than actual domestic prices used in the ciated with domestic support into perspective. implementation period result in over-counting With that outlook, the disparities in support levels or under-counting of true domestic support. and trade-distorting effects can be addressed in · Inclusion of "official" prices by some countries such a way as to strengthen the reform process results in countries being penalized if they do in general, and improve the fairness and effective- report domestic administered prices or have ness of future support reduction commitments. more conflation with border support than do Effective and balanced support reductions can be other countries. negotiated only if new categories of domestic sup- port and major changes in the current methods of 2. Abandon the "Peace Clause," which exempts measuring and classifying support are introduced. countries from countervailing duties under In the current WTO agricultural trade negotia- certain conditions (see box 6.1). tions, negotiators must be aware of the following: 3. Abandon the Blue Box. · The AMS has measurement problems. Collapse the blue box into the new amber box and · Two categories of de minimis reduces the effec- give no provision for supply controls--each coun- tiveness of support reduction commitments. try should be on its own on this and payments to · Blue box support is trade-distorting. have farmers idle resources should be counted · Some direct income payment policies in the as decoupled payments as described above.16 green box are trade-distorting. The EU's acreage decisions are fully coupled · The Peace Clause allows countries to circumvent while yields upon which payment amounts disciplines on domestic subsidies in the WTO in are calculated are decoupled, but the latter have general. output-distorting features such as decoupled infra- marginal production support and decoupled direct The following suggestions are designed to make payments. reduction commitments meaningful, measurable, and effective. 4. Introduce a New Amber Box. 1. Abandon or significantly modify the AMS. A new amber box covering domestic support with market effects (independent of market access and The AMS has the following problems: export subsidy support) will ensure that reduc- · The aggregation of the AMS commitment into a tions in domestic support will result in meaningful single measure covering all products and non- trade liberalization. Recommendations include the product-specific support maximizes the flexibility following: to avoid reducing support in some sectors and · Establish a new baseline of agreed-upon levels at even to increase it. the end of the URAA's implementation period. · Double counting with market access and export · Include so-called decoupled inframarginal pro- subsidy measures makes reduction commit- duction support financed either by consumers ments less effective. or taxpayers because such support induces exit · Fixed baseline world reference prices rather or entry, biases production incentives in domestic than actual world prices in the implementation markets, and cross-subsidizes exports in global period results in over-counting or under- markets. counting of true domestic support--with meas- · Include currently blue box support (the blue ured support very high when world prices are box should be abandoned with no credits given high (which means reductions in support are for inputs that are idled or restricted). not that pressing) and vice versa if world prices · Include so-called decoupled direct payments are low. (including crop insurance). These may reduce Domestic Support: Economics and Policy Instruments 145 both risk and risk aversion owing to the reduc- 3. The quantity eligible is also not well defined and therefore tion in uncertainty and an increase in wealth, open to manipulation (for example, one may use quantities pur- chased by the government instead of total production). respectively, resulting in higher levels of produc- 4. The specifics of how to measure "Market Price Support" tion than what otherwise would be. Direct pay- for AMS are in Annex 3 of the Agreement on Agriculture, in ments can also reduce imperfections in the paragraph 8. 5. The AMS is not the same as the "Producer Support input markets, thereby affecting exit/entry and Estimate" (PSE) as measured by the OECD. Only part of the investment decisions, and increasing production border support is included in the AMS for some countries and is in those countries that are only able to afford the excluded for others. In addition, fixed baseline reference prices are used to measure the current AMS (unlike for the PSE), and large direct payments to farmers. In addition, to the AMS excludes de minimis support as well as support con- make them truly minimally trade-distorting, the tained in the blue and green boxes. program design of decoupled payments should 6. The AMS can be greater than the PSE if world prices are be disciplined so that expectations for future above or below the AMS world reference price because the out- come also depends on the relative value of the domestic market payments are not based on the status of being price to the "Market Price Support". a farmer, of land being in agricultural use 7. These results need to be further refined to include mini- for particular crops (and perhaps even the level mum mandatory acreage set-aside requirements (farmers get paid the same as that for the minimum acres set aside) and max- of planting), or on previous levels of "emer- imum voluntary acreage set-aside, and "extra-ordinary" set- gency" payments, so production decisions are aside that has no financial compensation and is in proportion to not distorted. the excess acres. The historical and fixed character of yield pay- ments per hectare also affects the production decision. 8. Some economists argue that income inequality is a market 5. Revise other domestic support rules. failure. 9. The PEM is described in detail in Appendix A on the · Revise and expand the URAA definition of distributional effects of OECD policy reforms. 10. "Inframarginal" means the marginal cost for output- domestic support to include tax concessions receiving income payments is below the world price and farmers and other transfers of value to agricultural may or may not have to produce in order to receive payments production. (an example of not having to produce is the production flexibil- ity contract payment scheme of the 1996 U.S. FAIR Act). · Introduce a per unit support subsidy reduction 11. Standard analysis of consumer-financed inframarginal commitment along the lines of a tariff (not subsidies for the U.S. peanut sector (with and without quota only total value) to make reductions more transfers) is given by Borges and Thurman (1994); Rucker and Thurman (1990); and Rucker, Thurman, and Sumner (1995); effective. and for the U.S. dairy sector by Sumner and Wolf (1996). · Maintain the green box with defined limits 12. The approach also allows for the potential effect of for expenditures that provide for public goods decoupled payments on investment, given the specialized skills or prevent negative externalities (but limit of farmers and imperfect labor, information, and capital mar- kets (Roberts 1997; Skees 1999a, 1999b). both the level of product and nonproduct 13. The recent WTO Dispute Settlement Panel on Canadian expenditures)--and provide tighter defini- dairy policy illustrates the issues related to consumer-financed tions of programs listed in URAA Annex 2 as inframarginal subsidies. The panel ruled that milk sold at the world price below domestic prices was an export subsidy for rea- truly non- or minimally trade-distorting. Strict sons other than those discussed in this paper. For a critique on rules, definitions, and monitoring arrangements the WTO ruling, see Schluep and de Gorter (2001) and Annand, should be required. Cap green box expenditures Buckingham, and Kerr (2001). 14. Eligibility requires one of the following: (a) land is as a percentage of market value of output. enrolled in acreage reduction programs for any of the crop years 1991 through 1995; (b) land is planted to program crops under Notes program rules; or (c) land is enrolled in the Conservation Reserve Program (CRP) and has a crop acreage base associated 1. In some instances, total support as measured by the with it. OECD's "producer support estimate" (PSE) is less than total 15. Several variables can be stabilized such as price, output, domestic support. revenue, revenue net of costs (gross margins), and the like. 2. Data in figure 6.1 and tables 6.2 and 6.3 refer to 29 coun- Stabilization can also affect the choice of technique in production. tries covered by the OECD database and 11 commodity groups. 16. Paying farmers not to produce is like paying polluters not Trade-distorting domestic support refers to output and input to pollute--more firms enter the industry than would otherwise subsidies only. TSE refers to "Total Support Estimate" and and existing firms produce (pollute) more so as to get more includes expenditures on general services, etc. money later to produce (pollute) less. 146 Agriculture and the WTO Select Bibliography Guyomard, H., M. Baudry, and A. Carpentier. 1996. "Estimating Crop Supply in the Presence of Farm Programmes: The word processed describes informally reproduced works that Application to the Common Agricultural Policy." European may not be commonly available through libraries. Review of Agricultural Economics 23(4): 401­20. Hennessy, D. A. 1998. "The Production Effects of Agricultural ABARE (Australian Bureau of Agricultural and Resource Income Support Policies under Uncertainty." American Economics). 1998."Decoupling Farm Income Support."OECD Journal of Agricultural Economics 80(1): 46­57. Workshop on Emerging Trade Issues in Agriculture, Paris. Hoekman, B., F. Ng, and M. Olarreaga. 2002. "Reducing ------. 2001. "US and EU Agricultural Support: Who Does It Agricultural Tariffs versus Domestic Support: What's More Benefit?" Current Issues no. 2, Canberra. Important for Developing Countries?" Processed. World Annand, M., D. Buckingham, and W. Kerr. 2001. "Export Bank, Washington, D.C. Subsidies and the World Trade Organization." Estey Centre Innes, R. 1995. "An Essay on Takings: Concepts and Issues." Research Papers, Number 1, Estey Centre for Law and Choices 10(1): 4­7. Economics in International Trade, Saskatoon, Canada. Kuhn, B., and S. Offutt. 1999. "Farm Policy in an Era of Farm Barry, P. 1999. "Risk Management and Safety Nets for Farmers." Diversity." Choices 14(3): 37­38. Choices 14(3): 22­28. Moro, D., and P. Sckokai. 1998. "Modeling the 1992 CAP Beard, Nick, and A. Swinbank. 2001. "Decoupled Payments to Reform: Degree of Decoupling and Future Scenarios." Paper Facilitate CAP Reforms." Food Policy 26(2): 121­46. presented at AAEA Annual Meeting, 2-5 August, Salt Lake Blandford, D. 2000. "Are Disciplines Required on Domestic City, Utah. Support?" Paper Presented at the Canadian Agri-Food Trade Moschini, G., and P. Sckokai. 1994. "Efficiency of Decoupled Research Network Workshop on Agricultural Trade Farm Programs under Distortionary Taxation." American Liberalization, "Can We Make Progress?" 27­28 October, Journal of Agricultural Economics 76(3): 362­70. Quebec City, Canada. Mullen, K., N. Chau, H. de Gorter, and B. Gloy. 2001. "The Risk Borges, R., and W. Thurman. 1994. "Marketing Quotas and Reduction Effects of Direct Payments on U.S. Wheat Random Yields: Marginal Effects of Inframarginal Subsidies Production." Paper presented at the IATRC meeting, May 14, on Peanut Supply." American Journal of Agricultural Washington D.C. Economics 76(4): 809­17. Nelson, F., E. Young, P. Liapis, and R. Schnepf. 1998. "Domestic Brink, L. 2000. "Domestic Support Issues in the Uruguay Support Commitments: A Preliminary Evaluation." In Round and Beyond." Trade Research Series, Agriculture and "Agriculture in the WTO" WRS-98-44, U.S.D.A. Economic Agri-Food Canada, Ottawa. Research Service, Washington, D.C. Cahill, S. A. 1997."Calculating the Rate of Decoupling for Drops Orden, D. 2001."Should There Be a Federal Farm Income Safety under CAP/Oilseeds Reform." Journal of Agricultural Net?" Paper presented at the Agricultural Outlook Forum, Economics 48(3): 349­78. 22-23 February, Arlington, Virginia. Chau, N., and H. de Gorter. 2001. "Disentangling the OECD (Organisation for Economic Co-operation and Consequences of Direct Payment Schemes in Agriculture on Development). 1997. "Environmental Benefits from Fixed Costs, Exit Decisions and Output." Working Paper Agriculture: Issues and Policies." The Helsinki Seminar, 2001-16, Cornell University Department of Applied OECD, Paris. Economics and Management, Ithaca. ------. 1999. "A Matrix Approach to Evaluating Policy: Collins, K., and J. Vertrees. 1988. "Decoupling and U.S. Preliminary Findings from PEM Pilot Studies of Crop Policy Farm Policy Reform." Canadian Journal of Agricultural in the EU, the U.S., Canada and Mexico." COM/AGR/ Economics 36(4): 733­45. CA/RD/TC(99)117/Final. OECD, Paris. de Gorter, H., J. Hranaiova, and Y. Tsur. 2000. "Under- ------. 2000a. "Decoupling: A Conceptual Overview." standing the Production Effects of EU Direct Payments COM/AGR/APM/TD/WP(2000)14. OECD, Paris. for Acres Planted and Diverted." Processed. Department ------. 2000b. "A Preliminary Report of Domestic Support of Applied Economics and Management, Cornell Aspects of the Uruguay Round Implementation." COM/ University, Ithaca. AGR/APM/TD/WP(2000)9. OECD, Paris. de Gorter, H., and Y. Tsur. 1995. "Supply and Welfare Effects of ------. 2001. "Agricultural Policies in OECD Countries: Income Stabilization Programs: NISA versus NTSP." Monitoring and Evaluation 2001." OECD, Paris. Report to Policy Branch, Industry Performance and Roberts, I. 1997. "Australia and the Next Multilateral Trade Analysis Directorate, Agriculture and Agri-Food Canada, Negotiations for Agriculture." ABARE Research Report 97.6, Ottawa. Canberra. Diakosavvas, D. 2002. "How to Measure the Level of Agricultural ------. 2003. "Three Pillars of Agricultural Support and Their Support: Comparison of the Methodologies Applied by Impact on WTO Reforms." ABARE report 03.5, OECD and WTO." Workshop on Agricultural Policy Abareconomics, Canberra. Adjustments in China after WTO Accession, May 30­31, Rucker, R., and W. Thurman. 1990. "The Economic Effects of Beijing. Supply Controls: The Simple Analytics of the U.S. Peanut Gardner, B. L. 2001. "Benefit-Cost Economics of the FAIR Act." Program." Journal of Law and Economics 33(2): 483­515. Paper presented at the Fixing the Farm Bill Conference, Rucker, R., W. Thurman, and D. Sumner. 1995. "Restricting the 27 March, Washington, D.C. Market for Quota: An Analysis of Tobacco Production Gisser, M. 1993. "Price Support, Acreage Controls and Efficient Rights with Corroboration from Congressional Testimony." Redistribution" Journal of Political Economics 101(4): Journal of Political Economy 103(1): 142­75. 584­611. Rude, J. 2000a. "An Examination of Nearly Green Programs: Goodwin, B., and A. Mishra. 2002. "Are `Decoupled' Farm Case Studies for Canada, the U.S. and the EU." Agriculture Program Payments Really Decoupled? An Empirical and Agri-Food Canada, Economic and Policy Analysis Evaluation." Processed. Ohio State University, Columbus. Directorate Policy Branch, Publication 2010/E Ottawa. Domestic Support: Economics and Policy Instruments 147 ------. 2000b. "Green Box Criteria: A Theoretical Assessment." U.S. Department of Agriculture. 1997. "Agricultural Resources Agriculture and Agri-Food Canada, Economic and Policy and Environmental Indicators, 1996­97." USDA, Economic Analysis Directorate Policy Branch, Publication 2007/E Research Service, Natural Resources and Economics Ottawa. Division Agricultural Handbook 712. Washington, D.C. Schluep, I., and H. de Gorter. 2001. "The Definition of Export Vavra, P., and D. Coleman. 2000. "Land Allocation Under Arable Subsidies and the Agreement on Agriculture." In G. Peters, Area Payment Scheme." Paper presented at the International ed., Tomorrow's Agriculture: Incentives, Institutions, Association of Agricultural Economists Conference, Infrastructure and Innovations. Oxford: Oxford University August, Berlin. Press. Westcott, P. C. 1999. "Ag Policy: Marketing Loans Benefits Schmitz, A., and J. Vercammen. 1995. "Efficiency of Farm Supplement Market Revenues for Farmers." Agricultural Programs and Their Trade-Distorting Effect." In G. Rausser, Outlook, December, AGO-267. UDSA Economic Research ed., GATT Negotiations and the Political Economy of Policy Service, Washington, D.C. Reform, pp. 35­36. Berlin: Springer-Verlag. WTO (World Trade Organization). 2000. "Members' Usage of Skees, J. R. 1999a. "Agricultural Risk Management or Income Domestic Support Categories." 15 June, G/AG/NG/S/12. Enhancement?" Regulation 22(1): 35­43. Office of the Secretary General, Geneva. ------. 1999b. "Policy Implications of Income Insurance: ------. 2002. "Domestic Support Background Paper by the Lessons Learned from the U.S. and Canada." Paper presented Secretariat." March 20, TN/AG/S/4. Office of the Secretary to the European Association of Agricultural Economists General. meetings, Summer 1999, Warsaw, Poland. Young, E., and J. Westcott. 2000. "How Decoupled is U.S Sumner, D., and C. Wolf. 1996. "Quotas Without Supply Agricultural Support for Major Crops?" American Journal of Control: Effects of Dairy Quota Policy in California." Agricultural Economics 82(3): 762­67. American Journal of Agricultural Economics 78(2): 354­66. 7 THE DISTRIBUTIONAL EFFECTS OF AGRICULTURAL POLICY REFORMS Harry de Gorter, Merlinda D. Ingco, and Cameron Short Introduction the impact on poverty and income inequality on There has been dramatic growth in total world both urban consumers and rural farmers in a trade as successive rounds of multilateral trade developing country? These and other issues will negotiations have progressively reduced trade bar- be assessed in this chapter by using the Organisa- riers, providing an important engine of growth in tion for Economic Co-operation and Develop- the world economy. Nonetheless, very high barriers ment (OECD) "Policy Evaluation Matrix" model to trade remain in agriculture where markets are (PEM) for grains and oilseeds (see Appendix A for still heavily protected, especially in rich countries. complete details of this model).1 The grains and oilseeds sector is a good example of Such an analysis allows an evaluation of how extensive trade on the one hand, and high levels of poverty in both urban and rural areas should affect domestic subsidies and border protection (includ- attitudes toward trade liberalization in agriculture ing export subsidies) on the other. However, a and its implementation. There is overwhelming major transformation in terms of changes in policy evidence that trade generates higher economic intervention or re-instrumentation has also oc- growth and reduces poverty.2 However, income dis- curred in this sector over the past 15 years, tribution is inevitably affected, and although over- from border to domestic support and from cou- all poverty declines, some people become poor and pled subsidies to land area payments and historical some among the existing poor get poorer because entitlement payments. of trade liberalization (at least in the short run, Several key questions emerge: What are the before general equilibrium adjustments occur in impacts of alternative policy interventions on con- the labor markets). The other side to this is that sumers, taxpayers, farm operators, and land owners? additional aggregate income gives society more Does this impact vary across countries, commodity, resources to tackle poverty. or policy types? Is border protection or are domestic The purpose of this chapter is to illustrate how subsidy programs more important than other pol- agricultural policy reforms affect income distribu- icy interventions? What are the effects of unilateral tion both in high-income OECD countries and in a versus multilateral trade liberalization negotiated sample developing country, Mexico (also an OECD in the World Trade Organization (WTO)? What is member). This comparison can help policymakers 149 150 Agriculture and the WTO understand the effects on poverty and the need to The PEM framework is then used to analyze the consider complementary and sometimes compen- effects of these policies on the sample developing satory policies to alleviate some of these effects. The country, Mexico, for which the required data are broad policy implication of this analysis is that, in available. The various scenarios analyzed include most cases, trade liberalization would yield aggre- different types of policy reforms that might arise gate gains to a country and that analysis of its out of the Doha Round trade negotiations-- poverty consequences should be used to determine reforms of rich-country policies, as well as how these gains can be used to benefit the poor. reforms in Mexico's policies. The chapter reports Agriculture is the key sector for nearly all studies the distributional impacts by income level within on poverty analysis: the poor in developing coun- Mexico for both consumers and producers, as well tries are predominantly rural; food accounts for a as some selected results for other groups in other major share of their expenditure; agriculture is OECD countries. This analysis is intended to be their major source of income; and farm incomes illustrative of the methodology, rather than repre- have large spillovers to others in the rural sector sentative of developing countries as a group. Sev- and hence increases in farm incomes help to relieve eral further considerations of trade liberalization poverty throughout the rural economy. Many and poverty are discussed including imperfect developing countries have a comparative advantage markets and price transmission, and the implica- in agriculture because of their relatively large tions for subsistence farmers. The final section of endowments of land and unskilled labor. For these the chapter presents a summary and concluding countries, access to the large markets of the devel- remarks. oped countries and the prices received there are of central importance to their economic well-being. Agricultural trade liberalization affects poverty Policy Developments in the Grains in many ways.3 Changes in prices of agricultural and Oilseeds Sector outputs and inputs affect farm profits, land prices, The OECD identifies five major categories of sup- demand for rural laborers, off-farm labor opportu- port used in the PEM (see Appendix A): nities for farm households, and rural wages. The relative impact of these price changes on con- · Market price support (MPS) due to border pro- sumers, farmers, landowners, laborers, and input tection (including export subsidies). suppliers depends on the structure of the econ- · Output price support (OPS) such as deficiency omy--share of consumer budget spent on food, payments. income levels, demand elasticities, the production · Input price subsidies (IPS) such as fertilizer technology, input shares in production, and the subsidies. like. But the outcome also depends on farm size · Area payments such as those in the EU. distribution and whether farm operators own their · Historical entitlements such as those associated land or not. Trade liberalization effects can also be with grains and oilseeds in the U.S. affected by domestic policies. The PEM has six regions (the OECD-6): Canada, The Plan of This Chapter the European Union (EU), Japan, Mexico, Switzer- This chapter begins by describing some of the dif- land, and the United States (U.S.). Appendix A sum- ferent types of agriculture policies in the grains marizes the support for grains and oilseeds in the and oilseeds sectors in OECD countries. Trends in OECD-6 in the PEM and all OECD countries com- protection levels and changes in the policy mix are pared to all other commodities. Grains and oilseeds documented for the grains and oilseed sector in MPS represents 27 percent of total OECD MPS, but relation to all other commodities for the OECD as a grains and oilseeds receive a far higher share of total whole and for the six OECD regions covered in the support in each of the other four policy categories PEM. The chapter then outlines the agricultural of domestic support. Support across all policy types policy reforms in Mexico that are in many respects is far more uniform in the grains and oilseeds sector representative of such developing-country reforms. in all OECD countries as a group. The Distributional Effects of Agricultural Policy Reforms 151 The OECD-6 represents most of the support for landowners, input suppliers, taxpayers, con- grains in the OECD. In OPS, IPS, area payments, sumers, and world prices and trade. and historical entitlements, grains and oilseeds dominate in terms of amount of support relative to Agricultural Policy Reform in Mexico other commodities--and these policies dominate within the grains and oilseeds sector itself. This Prior to the 1980s, Mexico, like most developing means that the various types of payments are atyp- countries, followed the import substitution indus- ically far more important for wheat, coarse grains, trialization model of development with fixed and and oilseeds. The six regions analyzed comprise overvalued exchange rates and import tariffs on most of the support for each of the four commod- agricultural inputs. This imposed a tax on agricul- ity groups (wheat, coarse grains, oilseeds, and rice) ture and was typical of developing countries (the except for oilseeds (only 9.7 percent of total OECD so-called "urban bias"--see Krueger, Schiff, and support). A significant feature of area payments Valdés 1988). But governments also intervened and historical entitlements is its concentration in directly in agricultural markets using: (a) guaran- the grains and oilseeds sector in the countries and teed prices to farmers enforced by parastatal or regions in the PEM--especially the U.S., the EU, state trading purchases and food reserve purchases and Mexico. Both the U.S. and the EU will have a and fixed prices to consumers that were often lower major influence on the Doha Round because of the than producer prices (with either producer or con- size of their economies. sumer prices above or below world prices, depend- Appendix A shows that the rates of protection ing on the commodity, country, or year); (b) trade resulting from domestic policies are very low for controls (import licenses and import tariffs, or Canada in all policy categories, while the EU has direct controls by a state trading enterprise); and significant MPS for rice and high rates of support (c) subsidized inputs. Because of the focus on in area payments and IPS for all commodities. degree of self-sufficiency for the import-competing Japan has very high MPS and significant OPS and staple foods, policies promoted technical change IPS in oilseeds and rice only. Mexico uses and often subsidized purchased inputs such as fertil- all policies except area payments while the U.S. izers and improved seeds and other modern-sector relies heavily on OPS and historical entitlement purchased inputs.4 payments. Most developing countries including Mexico The trend downwards over time for MPS is sig- committed to major economy-wide and sectoral nificant and slightly more than that of total sup- reforms in the 1980s. This included trade liberaliza- port. Overall protection started at close to 120 per- tion, structural adjustment, deregulation, and cent in 1986 and fell to 50 percent in 1996--it privatization of state-run commodity bodies to stood at 77 percent in 2001. The move away from increase the role of market forces. The recent history MPS and toward the various forms of direct pay- of agricultural policy transformation in Mexico ments, especially area payments and historical for the import-competing staple field crops is entitlements, is also very evident. Purchased input a good example of developing-country policy subsidies and output price support have been fairly transformation. constant. In general, governments intervened through price One question this chapter seeks to address is controls, trade barriers, and marketing policies. In whether changing the composition of support Mexico, direct government intervention in agricul- away from border protection will make a major ture was heavily influenced by the National Company difference to developing countries. The alterna- of Popular Subsistence (CONASUPO) in its multiple tive policy framework proposed in the WTO by roles in shaping food production, consumption, and the 17-member Cairns Group of major nonsub- rural incomes. It was a typical developing-country sidizing agricultural exporting countries, and parastatal enterprise whose activities began to be more recently by other developing countries, reformed after the 1982 debt crisis and devaluation, calls for a substantial reduction in support. The and then further reformed and finally eliminated in PEM's strength is its ability to analyze the differ- the context of domestic economic reforms, trade lib- ential effects of alternative policies on farmers, eralization processes, and commitments under the 152 Agriculture and the WTO 1994 North American Free Trade Agreement and the There are two other important features about URAA. By 1999, the elimination of subsidies to con- direct payment policies in Mexico that have a signif- sumers for cornbread (or tortilla) consumption pre- icant impact on improving incomes of low-income ceded the liquidation of CONASUPO and com- and small farmers and hence on the distribution of pleted the liberalization of crop markets--which are income. Neither decoupled payments nor output now protected by modest import tariffs and import subsidies are distributed proportionately to produc- quotas that are binding for maize, barley/malt, and tion and therefore farm size. Decoupled payments dry beans only. There are no export subsidies. Farm- are given for a minimum of 1 hectare (regardless of ers currently receive three major categories of tax- the actual size of the farm) and a maximum of 100 payer-financed subsidies: hectares. Data furnished by the government of Mexico indicate that 1.2 million maize farms have · Area payments whereby farmers are required to less than one hectare of maize plantings, representing plant (but both the acreage receiving payments 36 percent of the total number of farmers but only and the payments themselves are not based on 9 percent of total area planted.5 current production decisions, and therefore Furthermore, output subsidies increase the far- payments are on inframarginal--that is, for the ther away a farm is from the major consumption or first units of output only, not for the last units of market center--and average farm size decreases output--acreage and hence deemed decoupled). with distance from major markets. These policy · Production subsidies (farmers are guaranteed details can have significant positive impacts on "concerted prices" through deficiency payments). income distribution compared to historical guar- · Input subsidies. anteed prices with a parastatal enterprise (where many small farmers did not benefit because they Consumers are now subsidized only for maize. A were often net buyers, sold goods at distressed summary of policy developments since 1980 is prices at harvest, or were not integrated with mar- given in Appendix A, table A.6. ket price centers owing to transactions costs and Parallel to the elimination of guaranteed crop the like).6 prices, a major reform in Mexican state interven- Consumer subsidy programs varied over the tion in staple food production was implemented in years but a subsidy to nixtamaleros (makers of corn 1991 through the creation of an agricultural dough to produce tortillas) and corn millers (pro- marketing agency, ASERCA (Support Services for ducers of flour to make tortillas), the two major Agricultural Marketing). Its functions are directed corn processing industries in Mexico, was signifi- toward marketing, but the agency does not buy or cant. These two industries represented 63 percent store agricultural commodities (now carried out by of the corn industry in 1996, and the subsidy the states, farmer organizations, or the private sec- allowed corn processors to sell flour and tortillas to tor). ASERCA is also in charge of a pilot hedging final consumers and to government retail shops at program for grains and oilseeds producers. low prices. In order to support the subsidy to tor- In 1992, extensive land reforms were imple- tilla consumers, CONASUPO sold maize to nixta- mented including the elimination of restrictions on maleros at a price that permitted a "reasonable" land transactions (previously, land in "Ejidos"--a profit from tortilla sales at subsidized prices. Corn form of communal ownership--could not be sold, millers received a cash subsidy for the corn that rented, mortgaged, or subdivided). Usufruct rights they bought directly in the domestic market. were contingent on occupation and cultivation of ASERCA, which administered the subsidy, has now the land. With these restrictions removed, land been replaced by a single targeted consumer sub- titling allowed decoupled payments to be imple- sidy program. mented. Some feared that poor, small farmers would be forced to sell their land in the face of cap- Some Empirical Results ital constraints, but small farmers in fact rent approximately 10 percent of their land to larger In this section, results are presented on the farmers (Olinto, Deininger, and Davis 2000). following two themes: The Distributional Effects of Agricultural Policy Reforms 153 1. Effects of an equal reduction in support rates in both in terms of level and type. This diversity makes all regions. it extremely difficult to design a "fair" result for sup- 2. Trade liberalization in Mexico versus trade liber- port reduction. How much of a reduction in U.S. alization in rich countries. historical entitlements is equivalent to a 10 percent reduction in EU area payments or a 10 percent It is clear, for example, that since Mexico is a net reduction in Japanese MPS? There are many other importer of grains and oilseeds, its citizens can in similarly possible deals that could be constructed-- the aggregate be made worse off by external meas- and interpreting fairness would still be subjective. ures that cause global prices of these commodities Instead of trying to construct equivalent sup- to increase. In this sense at least, Mexico is not rep- port reductions across regions, this chapter consid- resentative of developing countries as a whole, ers simply reducing support rates proportionately which are net exporters of agricultural products. in all regions by 10, 50, and 100 percent. Reducing But what is interesting is that this aggregate effect support rates rather than the amount of support masks important differences in how these policies per unit or the total amount of support implies affect different classes of citizens and Mexico larger reductions in regions with higher initial rates illustrates the benefits of a model that is capable of of support. It does not take into account whether disaggregating the impacts within groups. the type of support is normally associated with a high degree of trade distortion or not. Tables 7.1­7.4 summarize impacts by region for Trade Liberalization in All Regions wheat, coarse grains, oilseeds, and rice, respectively. Some of the material presented in the preceding sec- These tables also give a good idea of the overall tions illustrates the problem that government sup- responsiveness of the model by sector and region. port is highly variable from one region to another, The response in the wheat market is described in TABLE 7.1 Liberalization Effects on Wheat Price, Production, and Exports by Region, Base Levels, and Percentage Change Region Canada Mexico U.S. EU Switzerland Japan ROW Baseline Producer price (US$/mt) 92 190 108 110 332 1,192 105 Production (mmt) 20.7 3.4 53.3 97.6 0.5 0.7 426.7 Net exports (mmt) 12.6 2.6 19.3 13.6 0.4 5.7 36.8 10% liberalization Producer price 1% 3% 1% 1% 3% 8% 1% Production 0% 4% 0% 0% 4% 11% 0% Net exports 1% 7% 1% 1% 2% 1% 0% 50% liberalization Producer price 6% 15% 4% 5% 14% 39% 6% Production 2% 23% 1% 3% 19% 53% 0% Net exports 4% 37% 5% 3% 7% 5% 2% 100% liberalization Producer price 13% 30% 9% 11% 29% 77% 13% Production 5% 50% 2% 7% 30% 92% 0% Net exports 9% 81% 14% 4% 3% 4% 3% Notes: ROW = Rest of world, mt = metric tonne, mmt = million metric tonnes. Source: Calculated using the OECD's Policy Evaluation Model (PEM). 154 Agriculture and the WTO some detail. Only a brief description is provided of the most in Japan (from 11 percent to 92 percent) in the response for the other three commodities, since the 100 percent liberalization scenario; production to do more would be repetitive. decreases by 50 percent in Mexico and 30 percent in Of the six regions, Mexico, Switzerland, and Japan Switzerland even though producer prices decline by tend to have high rates of MPS and OPS for wheat, almost the same amount in both regions. The dif- which is reflected in the baseline producer prices (of ferences in the response among the three regions US$190/mt, US$332/mt, and US$1,192/mt, respec- that were not using high levels of MPS and OPS are tively) shown in the first row. Results for wheat in also instructive. EU wheat production declines by table 7.1 show that as a consequence, a proportional 7 percent despite the increase in commodity price reduction in all support rates will have the greatest because of the effect of the reduction in area pay- impact in these regions. The producer price of wheat ments. Production in both Canada and the U.S. with 100 percent liberalization falls by 30, 29, and increases slightly. Exports increase proportionately 77 percent, respectively in these three regions, but more from the U.S. because baseline exports are a increases in Canada, the U.S., and the EU. The last col- much larger share of production in Canada. umn, for the rest of the world ("ROW" in the table), Finally, it is perhaps worth noticing that there is a shows that the underlying world price increases by 1, fairly high degree of linearity in the results for prices 6, and 13 percent as the level of liberalization increases and production within each region. The change in to 10, 50, and 100 percent, respectively. prices in the 50 percent liberalization scenario, for Wheat production generally falls most in the example, is about five times the change in prices in regions that experience the greatest fall in producer the 10 percent liberalization scenario in each region. prices, but this too is moderated by the impact of This suggests that interpolating model results might other types of support. Wheat production decreases give a good approximation of these results. TABLE 7.2 Liberalization Effects on Coarse Grains Price, Production, and Exports by Region, Base Levels, and Percentage Change Region Canada Mexico U.S. EU Switzerland Japan ROW Baseline Producer price (US$/mt) 90 149 85 109 260 1,123 79 Production (mmt) 19.3 26.0 261.6 94.5 0.5 0.2 502.2 Net exports (mmt) 1.5 9.5 49.6 9.9 0.3 2.3 45.9 10% liberalization Producer price 1% 3% 1% 1% 4% 8% 2% Production 0% 3% 1% 1% 6% 7% 0% Net exports 6% 0% 1% 14% 13% 2% 2% 50% liberalization Producer price 6% 14% 5% 4% 19% 40% 9% Production 2% 13% 3% 6% 31% 39% 1% Net exports 31% 0% 8% 75% 69% 11% 9% 100% liberalization Producer price 12% 27% 11% 8% 38% 79% 20% Production 5% 25% 7% 15% 62% 81% 1% Net exports 68% 0% 21% 169% 140% 32% 17% Notes: ROW = Rest of world, mt = metric tonne, mmt = million metric tonnes. Source: Calculated using the OECD's Policy Evaluation Model (PEM). The Distributional Effects of Agricultural Policy Reforms 155 TABLE 7.3 Liberalization Effects on Oilseeds Price, Production, and Exports by Region, Base Levels, and Percentage Change Region Canada Mexico U.S. EU Switzerland Japan ROW Baseline Producer price (US$/mt) 187 302 204 201 849 1,205 162 Production (mmt) 6.4 0.1 79.6 13.4 0.0 0.2 136.5 Net exports (mmt) 1.5 4.4 29.5 20.7 0.2 4.8 0.9 10% liberalization Producer price 1% 3% 1% 1% 7% 4% 2% Production 1% 4% 2% 1% 8% 1% 0% Net exports 4% 0% 4% 1% 10% 0% 110% 50% liberalization Producer price 7% 15% 3% 7% 37% 22% 9% Production 5% 19% 8% 3% 44% 3% 2% Net exports 22% 2% 21% 5% 70% 2% 559% 100% liberalization Producer price 16% 29% 5% 15% 73% 43% 20% Production 10% 43% 17% 8% 96% 27% 5% Net exports 46% 4% 44% 11% 306% 9% 1153% Notes: ROW = Rest of world, mt = metric tonne, mmt = million metric tonnes. Source: Calculated using the OECD's Policy Evaluation Model (PEM). TABLE 7.4 Liberalization Effects on Rice Price, Production, and Exports by Region, Base Levels, and Percentage Change Region Canada Mexico U.S. EU Switzerland Japan ROW Baseline Producer price (US$/mt) 184 151 260 1,950 94 Production (mmt) 0.0 0.4 9.5 2.5 0.0 9.0 383.1 Net exports (mmt) 0.3 0.7 4.0 0.0 0.1 0.8 2.3 10% liberalization Producer price 2% 3% 3% 9% 1% Production 2% 6% 6% 7% 0% Net exports 0% 1% 15% 423% 0% 106% 68% 50% liberalization Producer price 9% 16% 16% 43% 5% Production 8% 33% 28% 39% 2% Net exports 0% 8% 77% 2122% 0% 621% 376% 100% liberalization Producer price 16% 30% 30% 85% 12% Production 16% 67% 55% 91% 5% Net exports 0% 16% 156% 4193% 0% 1773% 936% Notes: ROW = Rest of world, mt = metric tonne, mmt = million metric tonnes. Source: Calculated using the OECD's Policy Evaluation Model (PEM). 156 Agriculture and the WTO The high variability in results for net exports is Results for rice are the most dramatic and proba- also evident in the three tables for the other three bly the least reliable, especially for the 100 percent commodities (tables 7.2­7.4). It is interesting to liberalization scenario (table 7.4). Rice production in note that the EU switches from being a net exporter the four regions declines drastically, while produc- to a net importer of coarse grains in the 100 percent tion from the rest of the world increases by 5 percent liberalization scenario but not in the 50 percent lib- in response to 12 percent increase in average price. eralization scenario. The increase in oilseeds pro- This is possible because the rest of the world duction in Japan (by 27 percent with 100 percent accounts for 95 percent of production in the base, liberalization) should probably be regarded as a leaving only 5 percent from the endogenous regions. quirk resulting from the extremely high levels of Nearly all of the rest-of-the-world rice production support in the base and the model searching for comes from developing countries. The high levels of some way of using land in Japan. The U.S. does rel- OPS and/or MPS in the base for the endogenous atively less well in oilseeds (decrease by 17 percent regions is evident from a comparison of base prices. with 100 percent liberalization) because of the OPS Table 7.5 shows the welfare implications by level in its base. of liberalization and region. It shows the impact on TABLE 7.5 Welfare Effects of Liberalization by Region (US$ Millions) All six Region Canada Mexico U.S. EU Switzerland Japan regions 10% liberalization Taxpayer 81 129 1,749 1,853 15 1,279 5,106 Consumer 53 103 487 237 32 1,807 1,166 Farm operator 12 22 35 16 5 598 665 Landowner 33 163 853 1,439 15 645 3,148 Hired labor 0 0 3 2 0 0 5 Other purchased inputs 10 32 65 79 6 398 572 Sum 16 16 305 80 20 1,445 1,882 Percent PSE 16% 35% 27% 41% 60% 84% 39% 50% liberalization Taxpayer 403 554 8,487 9,086 41 3,885 22,455 Consumer 275 510 2,537 1,243 178 9,485 6,118 Farm operator 62 98 127 68 24 2,543 2,798 Landowner 151 783 4,102 7,028 69 2,503 14,636 Hired labor 0 0 13 8 1 0 22 Other purchased inputs 51 155 295 388 27 1,823 2,638 Sum 90 26 1,414 350 98 6,501 8,478 Percent PSE 10% 22% 16% 27% 47% 74% 24% 100% liberalization Taxpayer 806 846 16,288 17,446 63 1,142 34,181 Consumer 591 980 5,474 2,733 498 22,310 14,990 Farm operator 141 158 69 114 37 3,669 3,905 Landowner 262 1,488 7,669 13,166 133 2,906 25,624 Hired labor 0 0 20 21 2 0 43 Other purchased inputs 116 283 467 806 46 2,957 4,443 Sum 211 105 2,590 606 218 11,636 15,156 Percent PSE 3% 1% 1% 0% 10% 0% 1% Note: Percent PSE is the ratio of total transfers directly to producers and total farm revenues. Source: Calculated using the OECD's Policy Evaluation Model (PEM). The Distributional Effects of Agricultural Policy Reforms 157 economic welfare for each group of economic The concentration of program benefits on agents with a different role in the economy: tax- landowners raises some important questions about payers, consumers, farm operators, landowners, the distribution of agricultural land ownership.About hired labor, and purchased input suppliers. Farm- half the land in the U.S. and Canada is owned by farm ers are both farm operators and landowners, and operators with a significant portion of nonfarmer of course there are landowners who are not farm- landowners, probably retired farmers or their chil- ers. Similarly, farmers are also consumers. Though dren. If land ownership becomes skewed toward non- the consumer role can be ignored for farmers in farmers--and there are certainly tax and cultural the developed countries, consumption to fulfill incentives favoring this type of distribution of land own needs is a significant factor for poor farmers ownership--the distributional equity of agricultural in developing countries--the chapter will return support linked to land (area payments and historical to this issue below. Table 7.5 shows the impact on entitlements--that is, land payments) needs to be each role without regard to the interactions among seriously reconsidered. The situation in Mexico is the roles. probably somewhat unique because of the effect of Agricultural support in Japan chiefly consists the Ejido land tenure system. The interaction of the of a large transfer from consumers to farm opera- reform of Ejido tenure in 1992 and the Farmers Direct tors and landowners. In the U.S. and the EU, the Support Program (PROCAMPO) historical entitle- transfer is from taxpayers to landowners, with a ments introduced in 1994 on the distribution of land small share going to farm operators. Mexico and ownership and the issue of who ultimately benefits Switzerland are more even-handed on the financ- from Mexican agricultural policy is an intriguing ing side of this operation, taking from consumers question, but it will not be considered here. and taxpayers about equally but delivering most Tables 7.6 and 7.7 show that the deadweight loss of the benefits to landowners. of support for grains and oilseeds in these six TABLE 7.6 Distribution of the Effects of Global Liberalization on Mexico: Base Income Levels and Percentage Change Income Decile 1 2 3 4 5 6­10 Annual income per capita Farmer (US$/person) 365 466 687 954 1,526 Nonfarmer (US$/person) 362 656 859 1,089 1,217 3,799 10% liberalization Farmer Income effect 0% 1% 2% 2% 1% Consumer surplus 0% 0% 0% 0% 0% Total 0% 1% 2% 2% 1% Nonfarmer 0% 0% 0% 0% 0% 0% 50% liberalization Farmer Income effect 3% 7% 12% 8% 5% Consumer surplus 1% 1% 1% 1% 0% Total 1% 6% 11% 8% 5% Nonfarmer 1% 1% 1% 0% 0% 0% 100% liberalization Farmer Income effect 7% 15% 22% 16% 10% Consumer surplus 3% 2% 1% 1% 1% Total 4% 13% 20% 15% 9% Nonfarmer 2% 1% 1% 1% 1% 0% Source: Calculated using the OECD's Policy Evaluation Model (PEM). 158 Agriculture and the WTO TABLE 7.7 Welfare Effects of Liberalization by Region (US$ Millions) Liberalization Liberalization 100% liberalization in Mexico in Rich 5 (all regions) Region Rich 5 Mexico Rich 5 Mexico Rich 5 Mexico Taxpayer 83 846 33,336 120 33,336 846 Consumer 519 1,749 14,479 755 14,011 980 Farm operator 146 232 3,884 89 3,747 158 Landowner 286 1,652 24,368 260 24,135 1,488 Hired labor 9 0 51 0 43 0 Other purchased inputs 174 379 4,309 108 4,160 283 Sum 14 332 15,203 418 15,261 105 Source: Calculated using the OECD's Policy Evaluation Model (PEM). regions amounts to about US$15 billion per year. labor and a landless rural labor household class Landowners are invariably losers from liberaliza- would also provide better results. Currently all tion and this probably means that farmers also lose households involved in production on the com- from liberalization, with the one exception of farm- mercial farms are treated the same when in reality ers in Canada who own about 50 percent of the there is a division of benefits among farmers, land they operate. This means Canadian farmers on landowners, and hired operators that would be average probably do not lose from liberalization. available with better data. All regions gain from liberalization as shown in The columns in table 7.6 show the effects on table 7.5, but there is an important exception: Mex- both farmers and nonfarmers by income decile. ico gains at the 10 and 50 percent levels of liberal- Results for deciles 6­10 are combined. The first two ization but loses at the 100 percent level. Mexico's rows show baseline per capita income in U.S. dol- position here is probably atypical of developing lars per person per year. There is a small benefit to countries because its agricultural policy currently consumers from cereals liberalization in Mexico has a farm bias transferring large sums from tax- but it is at most equivalent to 2 percent of income payers and consumers to farmers. for the lowest-income urban consumer. (This result The welfare results in table 7.5 also provide an assumes that these consumers will continue to interesting insight into the interpretation of the receive the same level of targeted food subsidies.) commodity market results for rice in table 7.4. Farmers in income decile 1 are subsistence Even if there is a lack of confidence in the numeri- maize producers; in fact they are net maize buyers cal predictions in the latter, the quantitative result and maize is a relatively large part of their budget-- is clear. Rice policy in the developed countries so they too benefit as consumers. But they lose as results in a large shift of income among four farmers. The biggest losers are the small-scale com- groups of people: from consumers in the rich mercial farmers in income deciles 2­4. They are net countries and farmers in the poor countries, and to maize sellers, they receive a lot of their income from rich country farmers and developing-country con- maize production, and they find it more difficult to sumers. Since developing-country farmers are the adjust to lower subsidies. poorest of these groups and rich-country farmers the richest, this is a particularly regressive policy Trade Liberalization in Developed Versus framework. Developing Countries Finally, table 7.6 shows distributional effects in Mexico. This table is particularly affected by data Tables 7.7 and 7.8 contrast what happens when problems as the authors have not yet been able to trade liberalization occurs only in the rich countries identify the distribution of land ownership in terms in the model, with trade liberalization in Mexico of the households in the model. Including hired only. Again, these results are highly influenced by The Distributional Effects of Agricultural Policy Reforms 159 TABLE 7.8 Distribution of the Effects of Liberalization on Mexican Base Income Levels and Percentage Change Income Decile 1 2 3 4 5 6­10 Annual income per capita Farmer (US$/person) 365 466 687 954 1,526 Nonfarmer (US$/person) 362 656 859 1,089 1,217 3,799 Liberalization in Mexico Farmer Income effect 7% 17% 24% 17% 11% Consumer surplus 4% 3% 2% 1% 1% Total 4% 14% 22% 16% 10% Nonfarmer 3% 2% 2% 1% 1% 0% Liberalization in Rich 5 Farmer Income effect 0% 1% 4% 3% 2% Consumer surplus 1% 1% 1% 0% 0% Total 1% 0% 3% 2% 2% Nonfarmer 1% 0% 0% 0% 0% 0% 100% liberalization (All regions) Farmer Income effect 7% 15% 22% 16% 10% Consumer surplus 3% 2% 1% 1% 1% Total 4% 13% 20% 15% 9% Nonfarmer 2% 1% 1% 1% 1% 0% Source: Calculated using the OECD's Policy Evaluation Model (PEM). the baseline position with very high levels of 4 percent of income, with the greatest benefit for support in Mexico. the lower-income deciles. Elimination of support in Mexico has an enor- As might be anticipated, liberalization by only mous distributional effect within that country. the rich countries has the opposite effects. Results There is a net benefit overall of US$332 million and in table 7.7 show that this makes a small difference only a very small impact on the rich countries. only to the rich countries in the results of the Farm operators and landowners in the rich coun- 100 percent liberalization scenario already dis- tries benefit because of increased market access cussed. There is a huge difference in the distribu- while consumers lose from high prices and taxpay- tion of benefits in Mexico. Mexican farm operators ers from greater subsidies resulting from the small and landowners benefit modestly from the liberal- increase in production. ization in the rich countries because of higher Consumers and taxpayers in Mexico benefit by commodity prices. Table 7.8 shows that it is the about US$2.5 billion per year while the landowners large-scale commercial producers who benefit most are the main losers. Of course, many of the within the agricultural sector. Consumers and tax- landowners are farm operators at the same time. payers lose and the net effect is a modest negative. Table 7.8 shows that if all land benefits are attrib- uted to the farm operators, income for the small- scale commercial farmers drops by as much as Implications of Mexico's Change in the Policy Mix 22 percent. This group is hit the hardest by the pol- The effects of Mexico's shift in policy away from icy change. The consumer benefit ranges from 0 to input subsidies and border protection to historical 160 Agriculture and the WTO entitlements are illustrated by simulating the effect · Not surprisingly, therefore, a purchased input of a US$1 billion increase in each of the four major subsidy distorts trade more than any other policy. policy instruments: MPS, OPS, IPS, and historical · Small farmers benefit more from historical enti- entitlements in the form of area payments. The tlements in the form of area payments. It is likely results are summarized in table 7.9. Among the most that these results actually underestimate the interesting are the following: degree to which the system disproportionately benefits small farmers, since the model does not · A purchased input subsidy is the most ineffi- include three characteristics that increase pro- cient policy (even more so than MPS). This is gressivity: (1) the impact on the incomes of consistent with the argument earlier that elastic small farmers renting their land to large ones; purchased input supply curves can distort (2) the floor on payment so that those with land production more than border protection. under one hectare receive payment for a full TABLE 7.9 Impacts of Hypothetical Changes in Maize Policies in Mexico (US$ Millions) Change in Maize Farm Household Income by Decile Measure Subcategory 1 2 3 4 5 Total Purchase input subsidy (extra US$1 billion) 7.3 21.3 62.4 93.9 185.4 313.0 (all crops) Farm operators 3.1 8.8 16.5 24.8 49.0 78.4 Landowners 4.2 12.5 45.9 69.1 136.4 234.6 Output price subsidy (extra US$1 billion) 0.0 6.3 47.7 72.5 142.9 269.4 (maize only) Farm operators 0.2 0.6 6.9 10.4 20.6 38.3 Landowners 0.2 5.6 40.9 62.0 122.4 231.1 Market price support (extra US$1 billion) 0.6 13.2 90.9 136.5 269.5 510.7 (maize only) Farm operators 0.0 4.1 25.8 38.8 76.5 145.3 Landowners 0.6 9.1 65.0 97.7 193.0 365.4 Historical entitlements (extra US$1 billion) 10.1 50.2 103.7 155.5 307.1 626.7 (area payments) Farm operators 0.4 1.5 1.6 2.5 4.9 10.8 (all crops) Landowners 9.7 48.7 102.1 153.1 302.3 615.9 Grains and Oilseed Sectors1 Trade Purchased Distortion Input Inefficiency Measure Subcategory Maize Taxpayers Consumers Suppliers Costs Purchased input subsidy (extra US$1 billion) 58.2% 1,287.8 32.0 342.6 439.0 (all crops) Fram operators Landowners Output price subsidy (extra US$1 billion) 5.5% 1084.6 15.5 177.8 235.5 (maize only) Farm operators Landowners Market price support (extra US$1 billion) 60.4% 42.0 1040.4 194.4 328.6 (maize only) Farm operators Landowners Historical entitlements (extra US$1 billion) 6.9% 1033.0 3.7 18.0 81.7 (area payments) Farm operators (all crops) Landowners 1. The sum of transfers and inefficiency costs do not add up to US$1 billion because of indirect effects on other grains and oilseeds sectors (where policy intervention is held constant). Source: Calculated using the OECD's Policy Evaluation Model (PEM). The Distributional Effects of Agricultural Policy Reforms 161 hectare; and (3) output subsidies that increase Imperfect Price Transmission with distance from markets. Market failures (that is, transaction costs) create a · Farmers in the lower two deciles lose from an buffer for subsistence producers (about half of all input subsidy because land prices are bid up and maize producers in Mexico) from the change in the used by larger farm sizes. market price. Market imperfections, created by high transaction costs, mean many of Mexico's sta- Further Considerations ple farmers did not sell to the market. Transporta- This chapter has shown that heterogeneity across tion costs, marketing uncertainties, storage costs, households means that the effect of price declines and lack of credit often lead to harvest-time sales are felt differentially across household types-- at distressed prices. Being isolated from regional surplus growers are adversely affected. The imme- staple markets by poor infrastructure and high diate impact of eliminating the government- transaction costs makes people rely on local pro- guaranteed price of staples in Mexico is to duction to satisfy their consumption demands-- decrease the profitability of growing staples by and producers, finding it too costly to market their farm households that have access to the high output outside of the area, must seek markets for price. However, to see the full impact on these their surplus production within the local economy. households, food staples must be seen in the con- These conditions create a local market for the sta- text of the households' other income activities. In ple; the staple price is determined by local supply rural Mexico, households, including those pro- and demand, not by outside markets or govern- ducing staples, tend to be highly diversified. As a ment policy. result, decreases in the staple price would be The implication for trade liberalization and pol- expected to have small impacts on income. icy reform is that guaranteed market prices are not The analysis with the PEM, however, does not good instruments to improve efficiency or income incorporate the effects of several very important distribution among farmers and consumers. Direct details of Mexico's agricultural policy and eco- income payments such as those through PRO- nomic structure. For example, the characteris- CAMPO do get to the smaller farms farther away tics of the demand linkages among firms and from market centers. Furthermore, the new defi- households depend on the answers to several ciency payment program based on "concerted" questions: prices, that have payments increasing inversely with farm size (because farms on average become · Is price transmission imperfect because of mar- smaller and poorer with distance from market ket failures (transactions costs), owing to the centers), also improves the distribution of income. market power of intermediaries, government policy, or lack of rural infrastructure for trans- portation, information, credit, and the like? Imperfect Markets · Is the household a net consumer? Or do farmers facing high storage and transport costs need to Constraints exist owing to a variety of market fail- sell immediately because of a lack of credit, so ures, including missing or incomplete (that is, not that prices are depressed at harvest (and farmers functioning) capital, insurance, and labor markets. purchase for consumption later at higher However, market imperfections result in prices)? household-specific "shadow prices" that transmit · Is there a surplus pool of rural laborers or full income impacts to the production side of the employment? household-farm economy. If the household is · What are the opportunities of crop diversifi- liquidity-constrained (that is, lacks capital), the cation?7 PROCAMPO payments generate benefits relative · Are complementary policies relating to infra- to the constrained situation by increasing the structure and information available to establish amount of inputs used or investments in human markets, improve competition, and deepen or physical capital (and they also can be used market integration? as collateral). 162 Agriculture and the WTO To understand the impact of the transfer on quantities are responsive to changes in food prices. incomes and investments, it is necessary to follow In particular, it has been argued that by stimulating expenditure linkages to the labor-supplying house- food production and the demand for agricultural hold. Government payments (for decoupled labor, high food prices may benefit the rural poor income payments under PROCAMPO) stimulate through the induced wage response, even when the income growth in households that sell consump- poor are net demanders of food. tion goods or services to the household receiving But there has been little agreement on how the transfer. A PROCAMPO payment increases a responsive agricultural wages are to food prices. household's demand for staples (directly or indi- Contrast the results of Sah and Stiglitz (1984) with rectly, for example, through demand for livestock those of de Janvry and Subbarao (1986). Both stud- or other products). The local staple price increases. ies model the distributional effects of food pricing The higher staple price transmits benefits from the policies in economies when the rural poor depend transfer-receiving household to local staple surplus heavily on their earnings from supplying agricul- producers. Sadoulet, de Janvry, and Davis (2001) tural labor. Sah and Stiglitz contend that the food estimate the income multiplier for PROCAMPO price elasticity of the agricultural wage rate is close payments to range from 1.5 to 2.6. Hence, these to one, implying that a movement in the terms of market structures not taken into account by the trade against agriculture hurts everyone in the sec- PEM are critical in shaping local economy-wide tor, whether rich or poor. impacts of government policy. On the other hand, de Janvry and Subbarao assume that the nominal wage rate in agriculture is exogenously fixed, implying zero food price Net Suppliers or Demanders? elasticity, and, hence, quite adverse effects on the The empirical results show that all farm sizes for rural poor of higher food prices. Neither study maize in Mexico lose from price declines even presents sufficient empirical evidence to support though many of the decile 1 farmers are net buyers its assumptions. of food. Data limitations currently prevent the Ravallion (1990) examines the rural welfare dis- authors from having a net buyer category but it is tributional effects of changes in food prices under an important issue for those cases in which farmers induced wage responses for rural Bangladesh. He are net buyers. The traditional, neoclassical agricul- identifies the critical value of the elasticity of the tural household model used here assumes perfect agricultural wage rate with respect to the price of markets with decoupled payments and other exoge- food necessary for an increase in food price to be nous income shocks. Direct payments or exogenous welfare-improving. There can be little doubt that income changes do not affect production, because many poor households in rural areas are highly vul- an income transfer (from government) leaves nerable to at least the initial impact of an increase in the conditions for farm profit maximization the price of staple foods. But the welfare effects of unchanged. price changes have often focused on a single relative The welfare effects of low prices of food have price or income, whereas a better indicator would be been a contentious issue in numerous debates on the food purchasing power of the agricultural wage. development policy. A commonly held view is that A complete analysis of the welfare effects of a low food prices benefit urban groups but hurt the food price change, therefore, would require infor- rural population because they depend primarily on mation on how nonfood prices and incomes as well agriculture. However, many of the rural poor are as agricultural wages respond. It is not implausible, actually net demanders of food, even in Mexico. for example, that higher food prices would increase A great many among the poor do not produce village-level demand for the petty trading and serv- sufficient food for their own consumption, typically ice activities typically supplied by the poor. In a rel- supplementing their own farm incomes with agri- atively closed economy the prices of nonfood goods cultural labor earnings. Under regular partial equi- can also be expected to respond. The price elasticity librium conditions, such persons cannot benefit of the wage rate that is necessary for the household from high food prices. This conclusion is con- (net supplier or demander of food) and farm tentious when other welfare-relevant prices and laborer to benefit from a food price increase The Distributional Effects of Agricultural Policy Reforms 163 depends on several factors not analyzed here, the risk and uncertainty that poor households face, including the ratio of the worker's net food expen- although not always for the worse. It can also affect diture (after deducting the value of the worker's their ability to cope with risk and uncertainty. Poli- own production) to labor earnings, elasticities of cies such as improving access to credit markets can supply and demand for labor, and the share of help a great deal here, along with improvements in income from alternative income sources. land asset distribution and in the flexibility of local labor markets. Incorporating Off-Farm Labor Decisions Summary and Concluding Remarks The above discussion points to an important aspect of trade liberalization regarding its impact on the This chapter highlighted some important implica- allocation of labor to nonagricultural activities tions of alternative OECD policies on income distri- (including labor migration and income remit- bution, economic efficiency, and trade distortion tances). This would require the PEM to be extended across countries and commodity groups. Specific to a full household model that incorporates exoge- emphasis on Mexico was to illustrate the impacts on nous income sources and labor allocation to off- a developing country. The grains and oilseeds sector farm and noncrop agricultural activities. This would was chosen as an example using the OECD PEM incorporate multiple sources of income outside of framework. The countries and commodities agriculture, multiple crops within agriculture, off- explored are good examples because of the impor- farm versus farm labor decisions, and endogenous tance of grains and oilseeds for the six regions in rural wage. The results discussed earlier would be terms of policy intervention levels and value of pro- that much more complete with exogenous income duction and trade. There is also a wide range of pol- changes due to policy affecting labor decisions. icy instruments used and a major transformation of the sector in terms of policy re-instrumentation that has occurred over the past 15 years, from border to Complementary Policies domestic support and from coupled subsidies to For long-run benefits to accrue in terms of poverty land area payments and historical entitlement pay- alleviation, rural households will need to adjust ments. The policy mix also differs across both coun- their production patterns and farm/off-farm labor tries and commodities, allowing a wide range of activities.8 Trade liberalization should be accompa- policy effects to be examined. nied by sound domestic policies in areas such as Several key questions were addressed in this transport and communications infrastructure, chapter by analyzing the impacts of the alternative market facilitation, competition, education, and gov- policy interventions on consumers, taxpayers, farm ernance. Otherwise, it will fail to generate the invest- operators, and landowners and across countries, ment and productivity improvements needed for commodities, and policies. With high barriers to growth. More specifically, progress should be made trade in agriculture, the impact of trade liberaliza- in: investment in irrigation and rural roads to ensure tion and policy re-instrumentation has taxpayers that agricultural production can be connected to and consumers generally gaining from lower world markets; property rights to encourage invest- domestic prices and trade distortion declining as ment in the land; appropriate agricultural extension world prices rise. and mechanisms for the dissemination of market The effects on landowners, purchased input and technical information; and the development of suppliers, and farm operators were shown to markets for credit, agricultural inputs and services. depend critically on the policy instrument type Trade liberalization almost inevitably involves used. Purchased input subsidies and market costs of adjustment, notably job losses in formerly price supports were found to be inefficient, trade protected sectors. The best way to ease the pain of distorting, and ineffective in transferring income to transition is to protect social expenditure and farmers (the primary goal of agricultural policies) ensure appropriate targeting of the poor, providing while historical entitlements and area payments a cushion without undermining their incentives to were found to be the best (with output subsidies adjust. Trade liberalization can change the nature of falling in-between). 164 Agriculture and the WTO It is worth underscoring that OECD policy is As might be anticipated, liberalization by only particularly regressive in that it results in a large the rich countries has the opposite effects. The shift of income from farmers in the poor countries results show that this makes a small difference to (and rich-country consumers) to rich-country the rich countries compared to the full multilateral farmers (and developing-country consumers). The liberalization. There is a large difference in the dis- authors found substantial differences in the effects tribution of benefits in Mexico. Farm operators and of unilateral versus multilateral trade liberalization. landowners in Mexico benefit modestly from the Full trade liberalization in the PEM countries liberalization in the rich countries because of would result in a net savings of US$15 billion per higher commodity prices, and small-scale commer- year, with the largest part of this savings being cial producers benefit most within the agricultural made in Japan. Agricultural support in all regions sector. Consumers and taxpayers lose and the net involves much larger transfers from consumers effect is a modest negative. and taxpayers to farm operators and especially to The analysis demonstrates the importance of landowners. accurately simulating and fully evaluating the likely Mexico would not benefit from a full trade liber- consequences of partial liberalization of policies alization. Atypical for a developing country, Mexico such as tariffs, tariff rate quotas, input or output has farmers who now receive most support in the subsidies, area payments, and historical entitle- form of historical entitlements (that is, PRO- ments. The results here were highly sensitive to CAMPO), and less in terms of output price sup- details of market structure and the importance of port, input subsidies, and market price support. other support measures. Eliminating these latter policies would benefit poor Mexico's policy reform shifted away from fixed consumers but hurt small-scale commercial farm- and overvalued exchange rates, import tariffs on ers by much more. and subsidies to farmers for purchased agricultural All regions gain from liberalization, but there is inputs, and direct intervention in agricultural mar- a small, important caveat. Mexico gains at the kets with guaranteed prices to farmers and fixed 10 and 50 percent level of liberalization while lower prices to consumers, to direct area payments losing at the 100 percent level. Mexico's position (PROCAMPO) and production subsidies (ASERCA), here is probably atypical of developing countries which improved the efficiency and incomes of because its agricultural policy currently has a farm smaller farms. Consumer subsidies are also now bias transferring large sums from taxpayers and more targeted and therefore more effective. consumers to farmers. Furthermore, income distribution among farms Elimination of support in Mexico has a large has improved because land reforms allow small distributional effect within the country. There is a farms to rent out approximately 10 percent of their modest net benefit overall of US$332 million and land to larger farmers. Decoupled area payments only a very small impact on the rich countries. and output subsidies are distributed dispropor- Farm operators and landowners in the rich coun- tionately to smaller farm sizes. Decoupled pay- tries benefit because of increased market access ments are given for the minimum of 1 hectare while consumers lose from high prices and taxpay- (regardless of the actual size of the farm) and a ers lose because of greater subsidies resulting from maximum of 100 hectares. Output subsidies the small increase in production. increase the farther away a farm is from the major Mexican consumers and taxpayers benefit by consumption or market center (and average farm about US$2.5 billion per year while landowners are size decreases with distance from major markets). the main losers. Of course, many of the landowners These policy details can have significant are also farm operators. If all land benefits are positive impacts on income distribution compared attributed to the farm operators, income for the to historical guaranteed prices with a parastatal small-scale commercial farmers drops by as much enterprise--where many small farmers did not as 24 percent, making this group the hardest-hit by benefit because they were often net buyers, sold the policy change. The consumer benefit ranges prices at distressed prices at harvest, or were not between 0 and 4 percent of income with the great- integrated with market price centers owing to est benefit for the lower-income deciles. transactions costs and the like. The Distributional Effects of Agricultural Policy Reforms 165 Notes Krueger, A. O., M. Schiff, and A. Valdés. 1988. "Agricultural Incentives in Developing Countries: Measuring the Effect of 1. PEM is a multimarket policy model of the grains and Sectoral and Economywide Policies." World Bank Economic oilseeds sectors for OECD countries including Mexico. Review 1988, vol. 2(3): 255­71. 2. See Frankel and Romer (1999); Dollar and Kraay (2001); OECD (Organisation for Economic Co-operation). 2001. "Mar- and Ravallion and Chen (2001). For a critique of the empirical ket Effects of Crop Support Measures." OECD, Paris. results in the literature on the links between trade and ----. 2002a. "Agricultural Policies in OECD Countries: Moni- growth, see Rodrik (2000); and Nye, Reddy, and Watkins (2002). toring and Evaluation 2002." OECD, Paris. 3. Inequality can have an adverse impact on growth, espe- ----. 2002b."The Transfer Efficiency of Farm Support for Farm cially if there is inequality in asset distribution and education. It Household Income." OECD, Paris. is also well known that inequality can adversely affect political Nye, H., S. Reddy, and K. Watkins. 2002. "Dollar and Kraay on feasibility of policy reforms to liberalize trade. `Trade, Growth, and Poverty': A Critique." Paper, Inter- 4. As will be shown later, analysis of the OECD's PEMs sug- national Development Economics Associates, New Delhi. gests that this type of support is extremely trade-distorting, but networkideas.org/themes/human/ therefore also effective in promoting self-sufficiency. jun2002/hd19_Trade_Growth_Poverty.htm. 5. Data furnished by Patricia Aguilar through personal Olinto, P., K. Deininger, and B. Davis. 2000."Land Market Liber- communication. alization and the Access to Land by the Rural Poor: Panel 6. Payments vary by distance through the use of concerted Data Evidence of the Impact of the Mexican Ejido Reform." prices (prior to 2001, these payments were given to buyers). Pay- Processed. World Bank, Washington, D.C. ments are calculated through the "Local Prices of Indifference" Ravallion, M. 1989."Do Price Increases for Staple Foods Help or mechanism because in receiving the deficiency payment, the Hurt the Rural Poor?" Policy Research Working Paper buyer is indifferent between purchasing the domestic and WPS167. World Bank, Washington, D.C. imported crop. Prices of indifference vary according to the ----. 1990. "Rural Welfare Effects of Food Price Changes under region in which commodities are consumed or produced: there Induced Wage Responses: Theory and Evidence for is a different deficiency payment for each origin­destination Bangladesh." Oxford Economic Papers 42(3): 574­85. combination. The subsidy also serves as compensation to farm- Ravallion, M., and S. Chen. 2001. "Measuring Pro-Poor ers for selling at distressed prices at harvest owing to lack of Growth," Working Papers--Macroeconomics and Growth; credit, storage costs, and transportation costs. Stabilization; Monetary/Fiscal Policy, 2666, World Bank, 7. Coffee serves as an example here. Large, soft bean coffee in Washington, D.C. lowlands ("robusta") with lower costs in humid tropical regions-- Robilliard, A., F. Bourguignon, and S. Robinson. 2001. "Crisis for instance, Brazil and Vietnam--could easily switch to and Income Distribution: A Micro-Macro Model for cotton/sugar/beef/grains and oilseeds if desired, while hard bean, Indonesia." Paper prepared for the ESRC Development high altitude, high quality coffee such as grown in Central Amer- Economics/International Economics Conference, April 5­7, ica and Malawi/Kenya have economic conditions for coffee pro- Nottingham University. ducers who cannot switch to cotton/sugar/beef/grains because Rodrik, D. 2000. "Comments on `Trade, Growth, and Poverty,' by their land and climate make growing these crops unprofitable David Dollar and Aart Kraay." http://ksghome.harvard.edu/ (although some would migrate to lowlands if possible to plant ~.drodrik.academic.ksg/papers.html. these commodities subsidized by OECD countries). Sadoulet, E., and A. de Janvry. 1995. Quantitative Development 8. The ability of poor households to respond to the Policy Analysis. Baltimore: Johns Hopkins University Press. new opportunities presented by trade liberalization will be Sadoulet, E., A. de Janvry, and B. Davis. 2001. "Cash Transfer influenced by their location and demographic structure, and the Programs with Income Multiplier: PROCAMPO in Mexico." gender, health status, education, and assets of their members. In World Development 29(6): 1043­56. addition, benefits from trade liberalization will be minimal Sah, R. K., and J. E. Stiglitz. 1984. "The Economics of Price unless there is a sound investment climate, macroeconomic Scissors." American Economic Review 74(1): 125­38. stability and openness, governance and sound institutions and ------. 1987."Price Scissors and the Structure of the Economy." the rule of law, legal and judicial support for property rights, Quarterly Journal of Economics 102(1): 109­34. financial regulation, and competition policy. Singh, I., L. Squire, and J. Strauss. 1986. "A Survey of Agricul- tural Household Models: Recent Findings and Policy Implications." In Agricultural Household Models: Extensions, Select Bibliography Applications and Policy. Baltimore: Johns Hopkins Univer- sity Press. The word processed describes informally reproduced works that Winters, L. A. 2001. "Trade, Trade Policy and Poverty: What Are may not be commonly available through libraries. the Links?" Centre for Economic Policy Research Policy Paper No. 2382, London. de Janvry, A., and K. Subbarao. 1986. Agricultural Price Policy Winters, P., B. Davis, and L. Corral. 2001. "Assets, Activities and and Income Distribution in India. Delhi: Oxford University Income Generation in Rural Mexico: Factoring in Social and Press. Public Capital." Graduate School of Agricultural and Resource Dollar, D., and A. Kraay. 2001. "Trade, Growth, and Poverty." Economics Working Paper 2001-1, School of Economics, Uni- World Bank Policy Research Working Paper 2615. World versity of New England, Armidale, Australia. Bank, Washington, D.C. Frankel, J., and D. Romer. 1999. Does Trade Cause Growth? American Economic Review 89(3): 379­99. 8 THE "MULTIFUNCTIONALITY" OF AGRICULTURE AND ITS IMPLICATIONS FOR POLICY David Vanzetti and Els Wynen Introduction One can imagine negotiations stalled on valuing cultural heritage, biodiversity, or animal welfare. The Uruguay Round Agreement on Agriculture Continued support to agriculture in developed (URAA), in its built-in agenda for future negotia- countries distorts trade and may have detrimental tions, and the Doha Ministerial Declaration explic- effects on many developing countries. itly confirm that nontrade concerns will be taken into account in the current round of negotiations. Such concerns relate to nonfood outputs including The Plan of This Chapter the environment, food security, rural amenities, The aim of this chapter is to use some economic and viable rural communities. The major issue concepts to explore the issue of multifunctionality. concerning negotiators is the extent to which these In the next section, the positions of some of the benefits are jointly produced. The additional pro- major players on each side of the debate are briefly duction has provided an incentive for several highly described and analyzed. Legal dimensions are then protective developed-country members to raise described, followed by an economic perspective concerns about the multifunctional benefits of and some policy implications. Some implications agriculture that may be lost if further erosion of and conclusions draw the chapter to a close. coupled or trade-distorting agricultural support were to occur. Countries most active in espousing Positions of the Parties government support for the provision of these multifunctional benefits are Japan, the Republic of WTO members agreed in the Uruguay Round to Korea, Norway, and Switzerland, as well as the cut domestic support to agriculture. Support was European Union (EU). These countries have, coin- divided into three categories (so-called "boxes") cidentally or not, highly protected agriculture. depending on the degree of trade distortion. Multifunctionality conceivably presents itself as Support measures that are nondistorting, in the a stumbling block to the World Trade Organization sense that they have no or minimal effects on pro- (WTO) agricultural negotiations. The various ben- duction and trade, are grouped in the green box efits, such as landscape values, are notoriously diffi- (set out in Annex 2 of the URAA) and are not cult to value. It is thus tempting for proponents to subject to reduction disciplines. These include use inflated values to justify support to agriculture. measures such as general services (research, pest 167 168 Agriculture and the WTO and disease controls, infrastructure), food security and others 1999, p. 7). The crucial word here is stocks, domestic food aid, decoupled farm income "authenticity." Would agriculture become less support, income insurance, provision of safety nets, authentic if the intensity of production were to fall and environmental and regional assistance pro- following reductions in support? Assessing values is grams. This covers many of the nontrade concerns. also difficult. Many taxpayers may prefer wilderness The blue box covers domestic support measures to an agricultural landscape. of direct government payments to farmers under The Japanese government has offered flood mit- production-limiting programs. igation as a justification for subsidies to paddy rice More relevant, the amber box contains the most production. The water-buffering effect prevents production-distorting measures (such as govern- erosion and the potential destruction of valuable ment support for prices), which are subject to urban areas (ABARE 1999, p. 3). Opponents argue reductions. Some members argue that some that this hardly justifies raising rice prices to six production-stimulating measures should be per- times world levels, provided to all farmers regard- mitted precisely because of the need to stimulate less of their impact on flood mitigation. Further- agricultural production to provide the multifunc- more, alternative methods (dams, forestry) may be tional benefits and that some of the amber box equally as effective without stimulating agricultural measures need to be relocated to the green box, not production. subject to reduction commitments. Japan's government also stresses food security as As previously stated, the countries that most an important component of multifunctionality. It heavily promote the concept of multifunctionality maintains that short-term food insecurity may are Japan, the Republic of Korea, Norway, and increase in the future, owing to El Nińo and other Switzerland, as well as the EU. Perhaps coincidently, climate changes. Furthermore, increasing global as the U.S. Department of Agriculture (USDA) has demand for food, coupled with feed and supply observed, it is also these countries (along with constraints, may push up world prices (Japan Min- Slovenia) that have high levels of amber box sup- istry of Foreign Affairs 2000). The Japanese govern- port relative to what they are permitted under the ment emphasizes self-sufficiency as a means of URAA (Bohman and others 1999, p. 6). Although enhancing food security. the data used in the USDA study are dated, with the The EU proposal to the WTO negotiations gives EU switching support from the amber to the blue multifunctionality a central role in addressing soci- box in recent years, the point is well-made that fur- etal concerns related to agriculture (WTO 2002b). ther negotiated reductions in permitted amber box The proposal claims that this is necessary to gain support may act as a constraint on its support of public support to further trade liberalization. The agriculture. EU's claim is based on the notion that the European The Norwegian government has produced per- public fears the effects of trade liberalization on haps the best-articulated position in favor of multi- rural communities. Nonetheless, according to the functionality (see, for example, Norway Ministry of proposal, policies to achieve the multifunctional Agriculture 2001). Its argument progresses as fol- benefits of agriculture represent a much more tar- lows: the reform process agreed to in the Uruguay geted approach than the others discussed so far. For Round should not necessarily be finalized in the example, the EU proposal states that environmental current negotiations; countries are diverse, with policies should be used to tackle environmental differing concerns and production conditions; pro- problems. Moreover, these should be written into duction is required to supply the nontrade con- the URAA. The EU proposal also emphasizes ani- cerns; low-potential areas need government mal welfare and food safety issues, claiming that support to sustain production; and finally, green compensation for additional costs of meeting ani- box measures may be insufficient. The key to this mal welfare standards should be exempt from argument is the jointness of production. Elsewhere reduction commitments. Recent developments, the government maintains, "The value of the agri- including bovine spongiform encephalopathy and cultural landscape . . . [is] closely contingent foot-and-mouth disease in livestock, and geneti- upon the landscape's authenticity as a food pro- cally modified organisms in crops, have made food ducer" (Ministry of Agriculture, cited in Bohman safety issues a concern in the EU. The EU proposal The "Multifunctionality" of Agriculture and Its Implications for Policy 169 favors clarification of the precautionary principle as well as positive externalities. Finally, the dangers as a guide to food safety issues.1 of overvaluing hard-to-value multifunctional exter- The Republic of Korea and Switzerland are the nalities are stressed. The valuation problems allow other two countries best known for espousing the proponents to justify any expenditure and make concept of multifunctionality as a means of obtain- "wild and exaggerated claims as to the magnitude of ing flexibility in providing support to agriculture. the valuation of non-market rural goods,"according The Republic of Korea puts forward the additional to an Australian government representative (OECD view that agricultural productive capacity should 2000, p. 171). The Australian government is con- be maintained to allow for the possible reunifica- cerned that the concept of multifunctionality may tion of the two Koreas. (The Democratic People's be used to stifle reform of domestic support for Republic of Korea's agricultural capacity is cur- agriculture. In contrast to the U.S. and the EU, Aus- rently depleted.) tralia believes the production-stimulating effects of Although all countries have some multifunc- support, even blue and green box support, are sig- tional aspects associated with their agriculture, nificant (Roberts, Podbury, and Hinchy 2001, p. 5). most oppose the use of the concept to justify The Cairns Group proposes a review of green production-distorting government support. The box criteria to ensure that they are indeed mini- United States (U.S.) position on multifunctionality mally distorting. is articulated in a USDA publication, "The Use and The Cairns Group contains several agricultural- Abuse of Multifunctionality" (Bohman and others exporting developing countries that support the 1999). The title signals the flavor of the discussion, Group view with varying levels of enthusiasm. which emphasizes that the nonfood outputs from Some of these countries claim that domestic sup- agriculture can be provided without additional port outside the green box is a form of special and production-linked subsidies. Targeted measures to differential support for rich countries (WTO provide these multifunctional outputs are already 2002c). Outside the Cairns Group, developing included in the green box. This study makes the countries are a disparate grouping but in general point, mentioned earlier, that a country's support they oppose the use of the concept of multifunc- for the multifunctional agenda is related to its level tionality to justify domestic support. Relatively of protection for agriculture. Where the amber box few developing countries have submitted negoti- policies are a constraint, countries transfer their ating proposals to the WTO emphasizing non- production-linked support to the blue box or trade concerns. Mauritius has proposed that preferably to the unregulated green box. According developing countries, and in particular small to the authors, the concept has been misused by island states, be given greater scope to provide countries claiming that the nonfood outputs are production-distorting support to agriculture "jointly produced"--and therefore production- (WTO 2002a). Jordan has proposed domestic sup- linked, rather than targeted, policies are necessary port for its olive and sheep producers. The Democ- to increase production of these outputs. ratic Republic of Congo, Croatia, and Poland have As leader of the Cairns Group of agricultural also made proposals to the WTO concerning non- exporters, Australia holds a position that is rather trade concerns. The various African, Caribbean, similar to that of the United States (U.S.) (see and Pacific countries that currently receive prefer- Roberts, Podbury, and Hinchy 2001, p. 35 for a ential access to the EU markets tend to be sympa- detailed discussion). A note by the Australian thetic to the EU's negotiating position. Bureau of Agricultural and Resource Economics Many developing countries are concerned about (ABARE 1999) claims that multifunctionality is "a food security, one aspect of multifunctionality. The pretext for protection." These studies point to the problem they face is not so much inadequate flexi- need for targeted support, raise the question of bility to provide production-distorting domestic whether separate markets can be established to support to agriculture as it is the inability to finance determine the real values of environmental and it on a wide scale. For this reason, border measures cultural products, and emphasize the role of diver- rather than domestic support are commonly used sifying supply to ensure food security. They also to encourage agricultural production. Thus the EU, emphasize the importance of considering negative Japan, and other European countries appear to 170 Agriculture and the WTO have had little success as yet in building a substan- compensate farmers for the extra costs of meeting tial coalition of supporters of multifunctionality. more stringent animal welfare standards (WTO 2002c, p. 24). Food safety3 is another issue that sometimes finds a place on the multifunctionality Legal Dimensions bandwagon, the view being that domestic produc- Trade negotiations are concerned with changing tion ensures safe food (TAED 2000). The URAA the rules governing world trade. It is pertinent then does not cover this, but Article 5.7 of the WTO to outline what the current rules say on multifunc- Agreement on the Application of Sanitary and Phy- tionality and related issues. Article 20 of the URAA, tosanitary Measures refers to it, and sets out specific which contains an agenda for further negotiations conditions under which such measures may be and uses the term "non-trade concerns," an alterna- implemented. tive phrase for multifunctionality: "Commitments In essence, negotiations concerning multifunc- under the reform program should be made in an tionality hinge on whether production-distorting equitable way among all Members, having regard to support (including amber and blue box domestic non-trade concerns, including food security and support and border measures) should be main- the need to protect the environment. . . ." tained or perhaps increased to guarantee benefits The WTO Secretariat elaborates on this list, indirectly provided by agricultural production. specifying "non-trade concerns such as food secu- Opponents of that view claim that these external rity, the environment, structural adjustment, rural benefits from agriculture, if desired, can be pro- development, poverty alleviation and so on" (WTO vided without increasing production. The next sec- 2002c, p. 24). In 1947, GATT Article XX (General tion reviews some economic concepts to help the Exceptions) gave the then­GATT members the reader assess these conflicting claims. right to implement measures to protect the life or health of humans, animals, and plants and to con- serve natural resources. This applies particularly to Economic Perspectives multifunctional outputs of an environmental General Characteristics of Multifunctionality nature. However, recognizing the pitfalls, Article XX specifically mentions that these measures There may be an economic justification for govern- should not be applied in an arbitrary or discrimi- ments to provide key goods and services when mar- natory fashion, or be applied to restrict trade ket failure exists. Where the output of one good or (Raney and Tschirley 1999). service is directly related to the production of Some multifunctional outputs of agriculture another, in certain circumstances it may be desir- have specific rules. Food security is one.2 The WTO able to subsidize the second to obtain the benefits agreements include the "Decision on the Possible from the first. This is more likely to be the case if Negative Effects of the Reform Programme on the two goods are jointly produced, and the pro- Least-Developed and Net-Food Importing Devel- duction of the second involves externalities of such oping Countries," which is aimed at enhancing a nature that government intervention is desirable. agricultural productivity and infrastructure in For example, if a farmer's row of walnut trees pro- these developing countries (WTO 2002c, p. 23). vides some welcome shade along a dusty public However, most of the countries espousing food road, it may be desirable to subsidize production of security as a multifunctional benefit of agriculture walnuts from those particular trees to encourage are in fact developed countries, such as Japan, Nor- the farmer to maintain the shade-producing trees. way, and Switzerland. These essential concepts relating to multifunction- The issue of multifunctionality can be described ality are explained in turn below. as a "bandwagon" that has attracted numerous fel- low travelers hoping to use trade negotiations on Joint Production The main feature of multifunc- this issue to further their particular interests, legiti- tionality, in any industry, is that the secondary mate or otherwise. Animal welfare issues are not benefits are jointly produced and would not be covered under existing rules, but the WTO has available without the primary output. Wool is a received proposals allowing member countries to joint product of sheep meat. Although quality may The "Multifunctionality" of Agriculture and Its Implications for Policy 171 vary, it is not feasible to have one product without The author notes that these benefits are more the other. If strong jointness exists, the multifunc- closely related to input use, that is, if the land tional benefits cannot be produced separately from remains in farming most of the secondary benefits the agricultural output. If strong jointness does not are retained. exist, there may be a lower-cost means of producing the multifunctional benefit without subsidizing Government Intervention Government inter- agricultural output. vention depends on the nature of the externalities A recent Organisation for Economic Co-operation and the possibility of market failure. This inter- and Development (OECD) study synthesized infor- vention is more likely to be appropriate in cases in mation from 17 countries on the question of degree which the jointly produced multifunctional bene- of jointness between commodity and noncommod- fits are a public good, having the characteristics of ity outputs (Abler 2001, esp. p. 31). The report nonrivalry in consumption and nonexcludability. focused on physical rather than economic links and The earlier example used shade from walnut trees attempted to assess the scope for de-linking agricul- on a public road to illustrate this. The consump- tural production and its multifunctional benefits, tion by one person does not affect availability to and concluded that among the positive multifunc- others (at least until congestion occurs), nor can tional benefits only food security is strongly linked pedestrians be easily excluded and hence charged to commodity production. for the service. In the absence of a well-function- ing market, public goods (for example, shade) Externalities Externalities refer to beneficial or tend to be undersupplied. A government may harmful effects occurring in production, distribu- respond in several ways: (a) alter property rights tion, or consumption of a good or service that are to create a market (for instance, privatizing the not captured by the buyer or seller. This implies that road and allowing the farmer to charge a toll); (b) no market exists or that markets function poorly regulate (for instance, requiring farmers to grow and hence externalities are difficult to value. Jointly trees along roads); (c) impose taxes or subsidies to produced goods are not necessarily externalities. alter the behavior of buyers or sellers (for Wool is jointly produced but is not an externality instance, subsidizing tree planting along roads); because both buyers and sellers of sheep know the or (d) provide the service itself (for instance, value of wool, and it can be capitalized into the planting trees along roads). The appropriate value of a sheep. There is thus a market for the by- response depends on the precise nature of product. The market for external benefits may not the public good, the degree of jointness, the value exist because of an absence of property rights or of the benefits, transaction costs, the availability high transaction costs. Agricultural pollution is a of alternatives, and perhaps various location- common externality that has come about because specific issues. farmers, historically, have had the right to pollute. This has evolved from the days when farm pollution An Analytical Framework (for instance, water runoff containing pesticides) was minimal. A further example is landscape values. The OECD has attempted to provide a comprehen- Farmers are aware that passersby may enjoy the sive analytical framework to assist policymakers in scenery provided by their farm, but they find it diffi- deciding whether government intervention for the cult to charge them for this benefit. provision of positive multifunctional benefits is It is not clear how much agricultural output is justified (OECD 2001).4 This framework sets out a necessary to provide the desired secondary benefits. series of questions for decisionmakers: Does a decrease in agricultural output at the mar- gin lead to a fall in these positive externalities? A 1. Is there a strong degree of jointness between review (Burrell 2001, esp. p. 16) for the OECD of commodity and noncommodity outputs? reports from 17 countries has examined the issue of 2. Is there some market failure associated with the market failure, and concludes that almost no cases noncommodity outputs? are reported in which the level of multifunctional 3. Is government action required or are there bet- benefits is directly related to agricultural output. ter alternatives? 172 Agriculture and the WTO BOX 8.1 Aspects of Valuation Valuing multifunctional benefits is evidently entirely satisfactory. Travel costs methods likely to be difficult, yet policymakers must make attempt to value an attraction by measuring how decisions using explicit or implicit valuations. much people will pay to travel to it. Hedonic Apart from the absence of markets, there are pricing involves measuring a particular character- several additional difficulties relating to nonuse istic of a good. For example, housing situated valuations. near scenic agricultural land may have a different Existence values refer to the satisfaction value from otherwise identical housing situated gained from knowing that a good or resource elsewhere. Finally, contingent valuation involves exists. Many people who have never seen asking people hypothetical questions concerning whales or tigers nonetheless benefit from their their willingness to pay. continued existence. An obvious example is agri- In the absence of reliable economic valua- culture's role in preserving biodiversity. tions, policymakers have other indicators to Bequest values are the benefits to one gen- assess the desirability of a good or service. This eration of knowing that a resource will be passed may range from polling individuals to media down to the next generation. Forests may have coverage to visible protests (both in support or this characteristic. Rather than converting a for- opposition). The danger here is that particular est for agricultural use, there is some value in interest groups may claim to represent a larger preserving it to future generations. section of society than is in fact the case. Farm- Option values are the benefits of preserving ers may claim that the bulk of society is in favor a resource so that it can be used for an alterna- of preserving farmland, on the basis that taxpay- tive purpose later. ers have supported farmers for a generation. In Various methods exist to obtain nonmarket reality, many nonfarmers may prefer that the valuations. Unfortunately, in most cases none are land be returned to its natural state. All three questions must be answered in the affir- Valuation of Multifunctional Benefits mative to justify government support. The OECD A final conceptual point in setting up an economic emphasizes the scope to de-link commodity and framework helpful in analyzing multifunctionality noncommodity outputs by changing farming prac- as it relates to valuation, the estimation of demand. tices and technologies, or by pursuing low-cost Multifunctional benefits are commonly character- nonagricultural provision of noncommodity out- ized by externalities. By their nature, externalities puts. Even in the presence of strong jointness and are difficult to value (see box 8.1). How much of a market failure, the OECD questions whether options service should a government provide, given there is not linked to government (such as market creation no market to indicate its value to taxpayers who pay or voluntary provision) may not be a more efficient for the service? strategy than government subsidies, given the trans- This is an important point. When a valuation of action and administrative costs and risk of policy a service is not available, groups in society that failure associated with government intervention. want the service are likely to claim that it is much The OECD also notes the importance of nega- more valuable to society as a whole than others tive externalities associated with agriculture. who might favor alternative services. Governments Although it is obvious that they are significant, have to assess competing claims before providing these negative effects are rarely introduced into dis- the service, and given the conflicting advice from cussions of multifunctionality. The point that fur- stakeholders, bureaucrats, scientists, opinion polls, ther subsidization of agriculture may generate net and various interest groups, it is not clear how such negative rather than positive externalities is some- assessments can best be made. times overlooked. The "Multifunctionality" of Agriculture and Its Implications for Policy 173 The OECD (2000) recently examined the ques- Several policy implications can be drawn from tion of valuation and concluded that valuation stud- economic analysis of the multifunctional nature of ies can help policymakers. However, the OECD agriculture: warns that in many cases the results from hypotheti- cal contingent valuation studies should not be · There are both positive and negative externali- taken literally, and values should not be summed ties associated with agriculture production. It is across locations or different types of amenities. Nor desirable to internalize their effects, so that the should comparisons across countries be attempted. negative effects are minimized and the positive While of use, such studies should be used with care effects maximized. This does not, of course, and in conjunction with other approaches (OECD mean that the negative effects should be removed 2000, p. 176). With particular reference to the altogether--for example, agricultural produc- impact of further reductions in domestic support tion will inevitably involve some pollution. Nor for agricultural production, the OECD claims that does it mean that there should be no upper limit it is not clear whether these valuation techniques to the external benefits, as these will become shed any light on the impact of changes at the mar- increasing costly to provide, and at some point gin, and whether changes affect several multifunc- additional benefits will be unappreciated. tional outputs simultaneously. Finally, the · In circumstances involving more than one techniques have yet to gain the credibility required objective, multiple instruments are required. For for their use in international trade negotiations example, rather than using domestic support to (OECD 2000, p. 172). So far, the major elements of influence the level of output, additional instru- the debate on multifunctionality would seem to ments are required to control pollution and to center on the need for government intervention, encourage the provision of rural amenities. given the absence of a market of multifunctional · Sound policies imply targeting the problem as goods and the valuation of these goods. If govern- accurately as possible. Controlling agricultural ments are to intervene, it is desirable that their poli- output may be one way of controlling nitrate cies meet certain criteria. emissions, for instance, but a better way is to regulate or tax emissions directly. In that way, farmers change farming practices to sustain pro- Policy Implications duction and still meet the pollution objective. Policies applying to agriculture should be subject to the same criteria as other industries. Sound policies In some countries, input constraints have been are effective, efficient, equitable, transparent, and used to control output: farmers may be limited to a administratively simple. This means that such poli- given area. Area controls have been used in the U.S. cies actually work (effective), do so at least cost for this purpose. This policy is effective in limiting (efficient), are fair to providers and beneficiaries output only so long as there is a fixed relationship (equitable), easily understood (transparent), and between output and the limiting input (land, in this can be implemented at a reasonable cost (adminis- case). The typical response is for farmers to tratively simple). There are obviously tradeoffs increase the use of inputs that are not limited and among these criteria, but policies meeting the first thus raise the yields substantially, undermining the three criteria are usually well-targeted. Targeting policy objective. implies limiting the number of people affected by In general, negative externalities are related to the policy but may impose additional costs of input use rather than output, in particular the use transparency and administrative simplicity. If these of fertilizer and chemicals. These pollutants can be criteria are not taken into consideration, it is likely controlled most effectively by focusing directly on that policy failure will occur, or at least that alterna- the emissions. This would not be the case where tive policies would work better. inputs and outputs are in direct proportions, but 174 Agriculture and the WTO usually there is scope to vary the mix of inputs to ing landowners to charge fees to those wishing to meet environmental objectives at minimum cost. walk over their land. In this way land prices would Measures that influence output, such as tariffs more closely reflect their scenic value. Market-based and other trade measures and output-related solutions may not be the entire answer because domestic support measures, are not the most effec- nonuse values are hard to quantify (see box 8.1). tive instrument for tackling output-related positive A second approach taken by governments may or negative externalities. Environmental policies involve the payment of subsidies to encourage pro- are needed to achieve environmental outcomes. duction of the externality. The problem here is that This implies that to generate the optimal level of the government performs the function of the environmental benefits, a specific instrument market, and there is some question as to how is likely to be required, unless the externality is well it can provide the service where, when, and in generated in fixed proportion to output. If apple the appropriate quantity that consumers desire and blossoms are considered scenic, a subsidy should are willing to pay for. Switzerland provides scenic apply to the generation of blossoms and be related alpine landscapes, of value to locals and tourists to their duration and beauty, and the location of alike, and these are enhanced by the presence of live- admirers. There is little value in having scenery stock. There is little point in providing a scenic agri- with nobody to behold it (although some gain sat- cultural landscape in a valley where there are few isfaction for just knowing that it is there; see "exis- people to see it, or at a time when most consumers tence values" in box 8.1). (tourists) are elsewhere, or in quantities that are too few or too great. If this were the case then govern- ments could do better by subsidizing extensive live- Three Multifunctional Outputs from Agriculture stock operations alongside railways and motorways. All industries have multifunctional benefits to some degree, but this does not imply they should Agriculture and Food Security Food security is all be subsidized. Some claim that agriculture is an issue normally associated with developing coun- special or different in ways that justify government tries where food supplies are inadequate because of support. Three aspects of particular interest are low production or inability to import and distribute environmental benefits, food security, and rural adequate amounts.5 Perhaps somewhat surprisingly, amenities. These are examined in more detail. food security is also an issue in developed countries that can readily afford to import much of the food Agriculture and Environmental Benefits Agri- they consume. Japan, the Republic of Korea, and cultural production is inevitably associated with Norway, for example, claim to be concerned that positive and negative environmental externalities their supplies of imports may be disrupted because or spillovers. The negative spillovers are familiar; of wars, embargoes, price shocks, and perhaps, they include nutrient and pesticide runoff, soil ero- natural disasters. sion, methane emissions, loss of biodiversity, noise, Food security can be defined as a situation in odor, and increased likelihood of flooding. Positive which "all households have both physical and eco- environmental externalities are less obvious but nomic access to adequate food for all members, and real nonetheless. These include scenic vistas, where households are not at risk of losing such wildlife habitat, open spaces, recreational ameni- access" (FAO 1996). At the global level, food secu- ties, and opportunities for peace and solitude. One rity is obviously a distributional issue. In spite of important attribute in countries with steep terrain some 800 million malnourished individuals, there is flood control (Japan Ministry of Foreign Affairs is enough grain produced to feed all the people in 2000). Some WTO members claim, for example, the world, as well as livestock. Furthermore, poli- that rice paddies have an important function in cies such as production quotas, schemes to set aside preventing floods (Roberts, Podbury, and Hinchy productive land, and taxes are in place in many 2001, p. 39). countries to limit production. A feature of these externalities is the absence of a Food security fits rather uncomfortably with the market. One response of government may be to set other multifunctional benefits of agriculture, such up a market. This might involve, for example, licens- as environmental goods, for several reasons: The "Multifunctionality" of Agriculture and Its Implications for Policy 175 · Food security is not a nonfood item, which is sector, a secular decline that has been going on what multifunctional outputs of agriculture are since the 1800s, when the industrial revolution generally considered to be. made wages more attractive in industries located in · It is questionable whether food security is a pub- urban areas. As farm consolidation occurs, rural lic good, with the properties of nonrivalry and populations fall and the demand for services is nonexcludability, given that a functioning mar- reduced. As services such as hospitals, schools, bus ket exists. routes, and banks are reduced or removed, the local · Food security is not a joint product with pro- populations must find substitutes. At the extreme, duction, as it can be attained through trade and rural areas may depopulate altogether and coun- storage (Bohman and others 1999, p. 18). tries may be concerned about invasion from for- eign powers. Populating the land is seen as a means Arguments in favor of domestic support to of discouraging this (Norway Ministry of Agricul- increase food production, and hence food security, ture 2001). emphasize the insurance aspect of maintaining the This natural decline in farm populations is capacity to produce food domestically (Norway viewed by some policymakers as detrimental, and Ministry of Agriculture 2001). Taken to the extreme, they wish to subsidize agricultural production to evidence of self-sufficiency is obtained only when reverse this development. It is likely that subsidiz- all consumption is domestically produced. ing agricultural output is a very indirect means of A second argument in favor of food security supporting rural communities, as most of the sup- relates to the instability of international prices. For port goes to the large farmers. Even domestic sup- example, the Japanese view is that international port directed to inputs such as livestock and land, prices of staple commodities are likely to become such as that provided in the EU, is relatively inef- increasingly unstable with the El Nińo effect on cli- fective because 50 percent of it finds its way to only mate and increasing pressure on grain prices from 17 percent of the EU's farmers--those with the population and income growth in developing largest capacity (ABARE 2000). And in the U.S., countries (Japan Ministry of Foreign Affairs 2001). 9 percent of farms receive 41 percent of the gov- The arguments against using domestic support ernment's support, according to ABARE. to provide food security are along the lines that If it is thought desirable to support rural commu- domestic production does not necessarily provide nities, the most appropriate means is to directly pro- security, nor is it the best way to do so. Domestic vide missing services. This may mean governments production may be subject to disruption in the subsidizing medical practitioners in remote rural supply of imported inputs such as fuel, fertilizer, areas, providing infrastructure for wireless telephony and chemicals that would limit the ability of coun- (mobile phones), facilitating Internet banking, and tries to produce their own food in times of an perhaps subsidizing or cross-subsidizing transport embargo. Other means of achieving continuity in services. Such policies can be targeted to the specific supplies include holding stocks and long-term con- areas that need assistance. tracts, and maintaining a diversity of suppliers. In Another approach is to encourage rural the case of Japanese rice, in which the market for employment. Agricultural support encourages the favored Japonica rice grown in temperate cli- agricultural output, but its impact on employment mates is relatively thin, opening up the market is lessened by capital-labor substitution and would encourage suppliers to diversify into other increasing productivity. In many heavily populated production, thereby increasing market liquidity. developed countries--for example, Japan and the A further strategy could be for wholesalers to have EU countries--rural communities are not highly contracts directly with foreign producers, perhaps dependent on agricultural incomes, as off-farm even jointly owning the production facilities. incomes are a high proportion of total incomes. The share of agricultural employment in predomi- Agriculture and Rural Amenities Some, perhaps nantly rural regions ranges from 2 percent in Ger- most, countries place value on maintaining viable many to 8 percent in Norway, 11 percent in France, rural communities. This task is made more difficult and 14 percent in Japan (Bohman and others 1999, by the exodus of labor from the agricultural p. 19). 176 Agriculture and the WTO The relatively small share of agricultural tries of the EU--are those with high levels of govern- employment may be due to the proximity of ment support for agriculture. urban employment opportunities. Governments Opponents of multifunctionality maintain that can encourage the location of nonagricultural claims for the existence of the difficult-to-measure industries in particular rural areas through regu- external benefits of agriculture are merely a pretext lation or the provision of infrastructure. As tech- for additional support. Furthermore, they maintain nology changes work habits in countries, the need that these claims represent a substantial threat to for physical proximity in many occupations is further trade reform, that negative externalities lessening. As branch offices are replaced by tech- should also be considered, and that there are better nology call centers, the need to place employees in ways of providing the public goods that people a specific location is reduced. In rural areas there want. Indeed, the WTO has provision for such is a natural subsidy available to industry through policies (that is, the green box) and many such low-cost rent on housing and office accommoda- polices are already being used (Freeman and tions, but the need for high quality transport and Roberts 1999, p. 5). communications is paramount. Governments In analyzing the competing claims, perhaps the have a role in facilitating the provision of this major issue is the degree of jointness: the extent to infrastructure. Social infrastructure, such as edu- which output of agriculture and provision of the cation, training, and health, will help agricultural externality is jointly produced. If the requirement workers to find employment in other industries in was solely to generate the benefit, would it be nec- rural areas. essary to produce the same level of agricultural out- Thus far the main characteristics of multifunc- put as at present? If the degree of jointness is tionality--environmental benefits, food security, somewhat weak, a policy targeting the external and rural viability--have been discussed. Some benefit specifically will very likely be superior. claim that there are additional benefits, including A second issue is the valuation of the benefits, as cultural and heritage values (TAED 2000). These this determines the quantity to be provided. will not be discussed here explicitly; suffice to say Although by their nature externalities are difficult that the principles relating to their provision by to measure in the absence of market valuations, governments are similar to those covered above in measures to encourage the provision of the benefits the context of the main benefits. should be in proportion to the likely gains. For example, it may not be sound policy to subsidize all of agriculture forever to help prevent a flood once Implications and Conclusions in a lifetime or a price spike in a commodity once National sovereignty is a fundamental right of WTO every 50 years. members. This provides members with the right to Developing countries have long been concerned choose the nature of their agricultural policy objec- about limited access to developed-country markets tives, with the proviso, agreed among members, that because subsidized agriculture reduces market the policies to achieve these objectives are non- opportunities. Support to agriculture in OECD trade-distorting. In the run-up to the current round countries amounted to US$327 billion in 2000, more of multilateral trade negotiations, several developed than half the farm-gate value of production (OECD countries began emphasizing the need for output- 2002). In many cases, the subsidies exceed the value supporting policies to ensure that the external bene- added. Estimates by ABARE suggest that developing fits of agriculture were provided in sufficient countries would enjoy static welfare gains of quantities. These external benefits, by their nature US$14 billion from a 50 percent reduction in agri- difficult to measure, include environmental goods, cultural support levels and are likely to enjoy more food security, rural development, and other less con- stable world prices (Freeman and Roberts 1999, crete externalities such as cultural and heritage val- p. 38).While all modeling results should be used with ues. And as noted above, coincidentally or otherwise, caution, it can be said that developing countries as a members in favor of multifunctionality--Japan, the group are likely to be disadvantaged if the concept of Republic of Korea, Norway, Switzerland, and coun- multifunctionality leads to much agricultural sup- The "Multifunctionality" of Agriculture and Its Implications for Policy 177 port in some developed countries being reassigned FAO (Food and Agriculture Organization of the United Nations). into the green box in a way that locks in existing dis- 1996."World Food Summit: Technical Background Document No. 12."World Food Summit, 13­17 November. FAO, Rome. tortions in their potential markets. Freeman, F., and I. Roberts. 1999."Multifunctionality: A Pretext for Multifunctional benefits apply to all industries Protection?" Australian Bureau of Agricultural and Resource and have done so for many years. Sound policy Economics (ABARE), Current Issues 1999.3, Canberra. Japan Ministry of Foreign Affairs. 2000. "Negotiating Proposal involves identifying those benefits and seeing that by Japan on WTO Agricultural Negotiations." Committee on they are supplied in the right quantities, in the right Agriculture Special Session. www.mofa.go.jp. places, and at the right time. Norway Ministry of Agriculture. 2001. "Coexistence in a World of Agricultural Diversity--The Right of Every Country to Safeguard Non-Trade Concerns." Paper presented at the International Conference on Non-Trade Concerns in Agri- Notes culture, May 28­31, Mauritius. http://odin.dep.no. OECD (Organisation for Economic Co-operation and Develop- 1. Food safety covered under the Sanitary and Phytosanitary ment). 2000. "Valuing Rural Amenities--Territorial Econ- Measures (known as the "SPS agreement") is not strictly a mul- omy." Workshop Proceedings, June 5­6, Paris. tifunctionality issue, although some commentators regard it as ------. 2001. "Multifunctionality: Towards an Analytical such. This issue is addressed later. Framework." OECD Directorate for Food, Agriculture and 2. See also chapter 9. Fisheries, Workshop on Multifunctionality, July 2­3, Paris. 3. See also chapter 11. www.oecd.org. 4. The Abler (2001) and Burrell (2001) studies quoted ------. 2002. Agricultural Policies in OECD Countries: Monitor- above use this analytical framework to provide an empirical ing and Evaluation 2001. Paris: OECD. assessment. Podbury, Troy. 2000."US and EU Agricultural Support: Who Does 5. See also chapter 9. It Benefit?" Australian Bureau of Agricultural and Resource Economics (ABARE) Current Issues 2000.2, Canberra. Raney, T., and J. Tschirley. 1999. "Environment, Trade and Select Bibliography SARD: Concepts, Issues and Tools, Cultivating Our Futures Background Paper 4." Paper prepared for The FAO/Nether- ABARE (Australian Bureau of Agricultural and Resource Eco- lands Conference on "The Multifunctional Character of nomics). 2000."US and EU Agricultural Support: Who Does Agriculture and Land (MFCAL)." September 12­18, Maas- It Benefit?" Current Issues 2000.2, Canberra. tricht, The Netherlands. www.fao.org. Abler, D. 2001. "A Synthesis of Country Reports on Jointness Roberts, I., T. Podbury, and M. Hinchy. 2001. "Reforming between Commodity and Non-Commodity Outputs in Domestic Agricultural Support Policies through the WTO." OECD Agriculture." Paper presented at the Workshop on ABARE Research Report 2001.2, RIRDC Publication no. Multifunctionality, OECD Directorate for Food, Agriculture 01/07, Canberra. and Fisheries, July 2­3, Paris. TAED (Transatlantic Environmental Dialogue). 2000. "World Bohman, M., J. Cooper, D. Mullarkey, M. A. Normile, D. Skully, Trade, Production and Multifunctionality." Prepared by the S. Vogel, and E. Young. 1999. "The Use and Abuse of Multi- Food and Agriculture Working Group. www.tiesweb.org functionality." Economic Research Service, USDA, Washing- WTO (World Trade Organization). 2002a. "Negotiating Proposal ton, D.C. by Mauritius." G/AG/NG/W/90. Office of the Secretary Gen- Burrell, A. 2001. "Synthesis of the Evidence on the Possible eral, Geneva. Impact of Commodity Price Decreases on Land Use and ------. 2002b. "EC Comprehensive Negotiating Proposal." Commodity Production, and the Incidence on the Provision G/AG/NG/W/90. Office of the Secretary General, Geneva. of Non-Commodity Outputs." Paper presented at the Work- ------. 2002c. "WTO Agriculture Negotiations: The Issues, and shop on Multifunctionality, OECD Directorate for Food, Where We Are Now." Office of the Secretary General, Geneva. Agriculture and Fisheries, Workshop on Multifunctionality, July 2­3, Paris. 9 FOOD SECURITY AND AGRICULTURAL TRADE POLICY REFORM Merlinda D. Ingco, Donald Mitchell, and John D. Nash Introduction food-importing countries by structurally raising the cost of food imports. The chapter finally considers Article 20 of the Uruguay Round Agreement on Agri- what alternatives might be useful to reduce vulnera- culture (URAA), and the paragraphs of the Doha bility to shocks that threaten food security, whether Declaration dealing with the present negotiations on the shocks originate from trade policy reforms or agriculture focus on the continuation of reforms other sources. In particular, it considers whether there begun in the Uruguay Round, with "the long-term are ways in which trade rules can be developed to off- objective of substantial progressive reductions in sup- set the possible harmful effects, especially in develop- port and protection." Of the objectives and concerns ing countries; or whether there are additional mentioned in the URAA's Preamble, Article 20 refers international actions that can be taken (either in the specifically to "non-trade concerns, special and dif- context of the World Trade Organization [WTO] ferential treatment to developing country Members, negotiations or elsewhere) that would contribute to and the objective to establish a fair and market-ori- greater food security for these countries. ented agricultural trading system." Paragraph 13 of the Doha Declaration elaborates on these points and Whose Food Security? confirms that all three considerations will figure in the new negotiations. Among all the "nontrade concerns," the issue of food security is by far the most frequently men- The Plan of This Chapter tioned. This is especially true of developing coun- tries, but a significant number of developed This chapter considers the objective of food security, countries also mention it.1 one of the cluster of objectives collectively referred to Food security can be addressed at several levels: as"nontrade concerns."It first looks at several concep- food security of the individual or family, national tual levels of "food security" and examines how they food security, and international food security. have been or might be affected by multilateral trade reforms. It next examines more concretely and empir- Individual and Family Food Security ically the question of whether and to what extent the URAA has had the effect--as some allege--of reduc- Although the issue is not often framed in this ing the food security of some least-developed and net way, the most important level at which long-term 179 180 Agriculture and the WTO structural food insecurity occurs is the household.2 poverty, that is, policies aimed at economic devel- Hundreds of millions of individuals, largely in opment--especially rural development--and tar- developing countries, lack food security. The fun- geted programs of poverty alleviation to increase damental cause of this is poverty, that is, inade- the buying power of those most at risk. The ques- quacy of buying power. There are ample supplies of tion that must therefore be answered is whether the food in global markets to feed the poor, and a well- URAA or further reforms that are likely in the new established trading system to ensure that it gets trade negotiations tend to inhibit or encourage to areas where there is demand. If the poor had these kinds of policies. sufficient incomes, this would translate into effec- A strong case can be made that further reforms tive demand, and food insecurity would not be a that remove support and protection that distorts problem except in certain transitory situations, international agricultural markets will provide discussed later in the chapter. strong positive support for the necessary national Poverty in urban areas is due to scarcity of policies needed to increase rural development and employment opportunities and lack of human capi- economic growth in developing countries. This is tal to earn an income sufficient to purchase an ade- true for two reasons. quate diet. In rural areas, which still account for the First, most of the major distortions in world bulk of employment in most developing countries, agricultural markets arising from protection and poverty is due to the lack of productive resources, supports occur in developed countries. This has poor technology, and inadequate infrastructure, cou- two adverse impacts on developing countries. For pled with the need to use marginal land that often food-exporting developing countries, access to suffers from inadequate rainfall and other natural markets is limited, and the prices of their products disasters. Civil strife, ethnic violence, and corrupt or in international markets are depressed by unfair ineffective public institutions sometimes compound competition from developed-country export prac- these problems. Moreover, in many cases government tices. At the same time, developing countries that policies exacerbate poverty in rural areas by holding are not food exporters face international price sig- down internal prices of food and agricultural prod- nals that undervalue their internal production and ucts via the use of state trading enterprises, export may provide serious disincentives for necessary taxes, and overvalued exchange rates. investment in their agricultural development. Seen in this light, long-term structural food Moreover, these international policies that insecurity exists primarily because of a demand- adversely affect markets unwittingly support the side problem--lack of purchasing power by the flawed policies of developing countries that tax, poor--and must be addressed by raising incomes. rather than support, their agricultural sectors. Global trade negotiations can affect food security While there is good evidence that the URAA did in two ways. First, if the agreements result in a not have as large an impact in disciplining the use structural supply shift that increases the average of these policies as intended, one result of success- price of food, this could increase the long-term ful current negotiations will be to reform these incidence of food insecurity by causing more of the policies of developed countries, which will certainly poor to be unable to purchase an adequate supply improve incomes in rural areas in the developing of food, or to do so only at the cost of diverting so countries. much of their income into food that they have Second, careful examination of the policies that inadequate income remaining to meet their other were disciplined by the URAA--namely export basic needs. Such a structural supply shift could subsidies, certain kinds of domestic support, and also increase food insecurity by increasing the market access barriers--and that will be further budgetary requirement of any government or disciplined in the new round, suggests that the new privately-funded safety net programs that provide agreement is not likely to discourage developing food to the needy. This possibility is discussed later countries from adopting policies that would foster in the chapter. rural development and poverty reduction. The second channel through which the negotia- The URAA prevents all countries that did not tions could affect food security is through their declare the use of export subsidies during the base effect on the policies and conditions that alleviate period (1986­90) from introducing subsidies in the Food Security and Agricultural Trade Policy Reform 181 future. However, these subsidies represent a policy Apart from the fact that such domestic support that makes food available to foreign buyers at a policies are not good policies for rural development lower price than to a country's own population. It is and poverty alleviation, there are good reasons to hard, therefore, to see how a policy of using export believe that developing countries have not been subsidies would be an optimal development policy constrained in their use of amber box measures. in poor developing countries. In addition, the fiscal Countries were required to declare the level of cost of significant subsidies is beyond the fiscal amber box support they were using during the base capacity of most developing-country governments. period and were required to reduce their aggregate The case against export subsidies is strengthened by expenditures on subsidies in this category by the the fact that expenditures on export infrastructure end of the URAA implementation period. Develop- and internal transport costs--which have many ing countries had few amber box subsidies to benefits in addition to lowering export costs--are declare. The small number of developing countries exempt from export subsidy limits for developing providing some subsidies classified as amber box countries. did not have to make reductions as large as those Likewise, URAA constraints placed on domestic required of developed countries, and were allowed support to agriculture do not inhibit sound rural to make their cuts over a longer period--of up to development policies in developing countries. Of the 10 years. In addition, a de minimis exemption was three categories of domestic support defined in the included that exempted product-specific support URAA, green box support measures are those con- by developing countries if it did not exceed 10 per- sidered least-trade-distorting. They include spend- cent of the value of total agricultural production ing on general agricultural services including (compared with 5 percent of the value of produc- research, education, and payments to producers not tion for developing countries). Least-developed tied to current production. At present, there are no countries (LDCs) were not required to make limits to the expenditures that can be made in the reduction commitments. green box category, although certain countries have It is sometimes asserted that tariffs are prefer- called for some limits to the level of decoupled direct able to domestic support as a way of encouraging payments that can be made to farmers. However, agricultural sector growth in developing coun- even if such limits were adopted, they should not tries, since these countries cannot afford the kinds reduce the policy options of developing countries, of subsidies given to farmers in developed coun- which generally do not use such direct payments to tries. The argument usually proceeds that the producers because they are both expensive and diffi- URAA did not provide developing countries suffi- cult to administer. Possible changes to domestic sup- cient flexibility in this regard. However, it is not port measures in the blue box are even less relevant clear that poor consumers in these countries can to developing countries. It is difficult to imagine that "afford" to pay higher prices any more than the developing countries facing food supply constraints governments can "afford" to pay farmers directly. would adopt such production-control policies that Like subsidies, and for the same reasons, tariffs are reward producers for limiting production. a poor and untargeted policy for rural develop- The remaining category of trade-distorting ment and poverty alleviation. In any case, tariffs domestic support is the one classified in the URAA also were not well-disciplined by the URAA as the amber box. These are commodity-specific because of the large margin between the bound subsidies that support internal prices, are paid on and applied rates in developing-country tariffs (see the basis of prices, subsidize commodity-specific chapter 4). inputs, and otherwise distort the incentives to In summary, it is difficult to make a case that the produce and consume specific products. In general, URAA rules on domestic support impose con- they are not appropriate policies for rural develop- straints on the ability of developing countries to ment. They are also extremely inadequate tools for push forward with policies and programs that will poverty alleviation, since payments are propor- promote agricultural growth, improve agricultural tional to output or input use and therefore are incomes, and thus contribute to a significant greatest for large farmers and leave out the poorest increase in food security at the personal and family (landless laborers) entirely. levels, especially in rural areas. 182 Agriculture and the WTO In the new round, a number of developing coun- National Food Security tries have suggested that a development box be established. While this might be necessary or desir- Most discussions of food security are focused on able (the question is discussed further in chapter 14), food security at the national level, not at the indi- clear analysis is needed regarding what a develop- vidual level. Food security at the national level con- ment box could include that is not permitted under cerns having an adequate total food supply to the present green box rules, in combination with enable the population of the country on average to the higher de minimis rules and other special and maintain a reasonable level of consumption. differential treatment given to developing countries (See box 9.1 on Food Security Indicators that are by the URAA. considered proxies for measures of food security at Stronger arguments may exist for revisiting the the national level.) This approach to food security WTO rules on notification of existing subsidies by is not focused on internal income distribution or developing countries, and for promulgating rules individual food security. What is often obscured by that would prevent the unintended imposition of looking at food security in this way is that under real limits on policies by domestic inflation and cur- normal circumstances--barring war, natural disas- rency fluctuations, outside the sphere of agricultural ters, artificial barriers to trade, and macroeconomic policy. But the major implication for negotiations of disruptions sufficiently severe to impede imports-- this view of structural food security--that it is if individual households have adequate purchasing mainly a household level issue--is that food secu- power, national food security follows automatically. rity concerns are best addressed in the negotiations Households of course can purchase food of through a focus on ensuring that green box agricul- either domestic or foreign origin. However, there is tural support measures and targeted poverty pro- a tendency for governments, especially in countries grams are not constrained. with large populations, to believe that they can only BOX 9.1 Food Security Indicators Food security can be analyzed at the global, food imports from its total export revenues. regional, national, household, or individual level. The ratio of the cost of food import to the The focus of this chapter is at the national level, value of total exports is a more accurate food which is the level at which World Trade Organi- security indicator than the net food trade zation (WTO) negotiating categories are position, which is determined by subtracting defined. In a study done by International Food a country's total food imports from its food Policy Research Institute (IFPRI), "Food Security exports. and Trade Negotiations in the World Trade c. Calories per capita and protein per capita are Organization: A Cluster Analysis of Country used to estimate consumption levels and Groups," the authors describe key indicators of therefore food availability at the national food security that are considered proxies for level. It is understood that national averages three measures of food security at the national are limited in reflecting household and indi- level: food availability, access, and utilization. vidual food and nutrition security levels. How- The indicators are summarized below ever, national calorie-aggregates are important in explaining changes in malnutri- a. Food production per capita is a measure of a tion as defined by anthropometrical measures country's ability to feed its population. The of children. Food and Agriculture Organization (FAO) d. Nonagricultural population is used to analyze defines this variable as the total food produc- the nonagricultural sector in assessing how a tion in every year multiplied by the 1989­91 country may be affected by changes in world price in U.S. dollars, divided by the total trade and agricultural policies and the poten- population of the corresponding year. The def- tial distributive impact on the urban/rural inition of food is the one provided by the FAO population. This indicator, however, is not as Statistical Database. straightforward in its implications as the oth- b. The ratio of total exports to food imports is a ers. measure of a country's ability to finance its Food Security and Agricultural Trade Policy Reform 183 achieve food security in the national sense by main- therefore do not benefit from higher prices. Farm- taining a high degree of self-sufficiency in all of the ers in the commercial sector who are in a position basic foodstuffs. The argument is that the demands to grow the staples may benefit in the short run. of their populations are so great that food imports But even they may lose out from the protection of are in some sense more costly than domestic pro- these crops in the long run because of the disincen- duction. While the rationale is seldom made tives this creates for diversifying into nontradi- explicit, this line of thinking could have two possi- tional exports, which may be a better road to escape ble bases. One is that the shadow value (social from poverty (Christiaensen, Demery, and Pater- value) of foreign exchange is higher than its market nostro 2002; Kherallah and others 2000). Diversion value. This was undoubtedly true in many coun- of resources from export production also reduces tries in the past, when currency controls caused the long-term foreign exchange earning capacity of major distortions in foreign exchange markets, but the country and thereby undermines its structural is not an issue in the overwhelming majority of capacity to import food and other products. And countries today. A second possible rationale is that finally, supporting the agricultural sector through the elasticity of supply of basic foodstuffs is higher trade protection diverts attention from methods of in these countries with large populations than it is support that in the long run could be more produc- in the major food-exporting countries. Many tive. Simulations in a computable general equilib- developing countries with large populations have, rium (CGE) world model by the IFPRI (Diaz- however, already extended agricultural production Bonilla, Diao, and Robinson 2003) of the alternative to less favorable areas, and often have less devel- of raising cereal tariffs by half in different groups of oped infrastructure and support to expand agricul- developing countries (mostly net food-importers), tural output. It is clear that for these countries, the versus transforming the mainly implicit and pri- pursuit of food security via self-sufficiency will be vately collected tax on consumption in the first sce- an expensive route to food security. nario into an equivalent explicit tax, and investing Governments hoping to achieve food security that money in agricultural research, found the latter via self-sufficiency generally want to maintain high to be enormously more beneficial in increasing border protection and high internal prices to employment, income, and consumption in gen- encourage increased domestic output. This, how- eral--including, particularly, food. ever, has some adverse impacts on the food security While it is clear that policies aiming to increase of at least some groups. self-sufficiency are in many ways counterproductive First, the consumers of the country pay a consis- to the goal of structural food security, other justifi- tently higher price for the products, undermining cations for policies to increase self-sufficiency are the food security of the poor by reducing their pur- based on more short-term concerns--disruptions chasing power. This acts as a regressive tax, since to trade from hostile actions of other countries or the proportion of income spent on food is inversely from barriers to exports imposed by the exporting related to income. Consequently, poorer consumers countries themselves in times of international tend to be hit hardest by the higher prices. This is shortages or high prices. Such concerns tend to be especially troubling in light of evidence that there exaggerated; serious trade disruption would require may be a shift in poverty to urban areas (Haddad, concerted action by a large number of competitive Ruel, and Garrett 1999; Garrett and Ruel 2000), and exporters, which is most unlikely. The effects of that the poorest of the poor in rural areas are land- temporary disruptions could also be mitigated less laborers. Neither the urban poor nor landless through other measures--stockpiling of reserves, laborers derive any benefit from the higher food for example--that would be much less costly than prices, so clearly their structural food security is efforts to achieve self-sufficiency. But in any case, adversely affected by such policies. the preoccupation with self-sufficiency has diverted But second, even many members of the farming attention from the question of how the trading sys- community may be hurt or at best not be helped by tem could be used to prevent these short-term con- protectionist measures. The poorest farmers in the cerns. Too little attention has been given to the developing world are subsistence farmers, and these obligations of food-exporting countries to make by definition have no marketable surplus and supplies available to all food-importing countries 184 Agriculture and the WTO without export restrictions, taxes, or other discrim- world population at levels consistent with their inatory policies. This issue has been raised in the demands. Periodically agricultural commodity context of the current negotiations, and its resolu- prices rise substantially, notwithstanding a secular tion could give some comfort to countries con- downward trend. Whenever this occurs, there is cerned about disruptions and therefore make them concern about the ability of the world's agricultural more amenable to shifting their focus away from producers and the resources they command to con- self-sufficiency. tinue to meet world demand at constant or declin- More attention should also be given to the spe- ing real prices. Land and water resources available cial problems of developing countries that may for agricultural production are finite. However, the have difficulties in meeting their food import bills output from those resources has steadily increased because of unusual increases in import require- with the development and adoption of new tech- ments or because of increases in agricultural com- nology, greater investment, and improved human modity prices, either cyclical or longer-term resource skills engaged in agricultural production. structural increases. This latter issue underlies the Global food security depends on the continued Marrakesh Decision on "Measures Concerning the increase in the efficiency with which global agricul- Possible Negative Effects of the Reform Programme tural resources are used, on continued market on Least-Developed and Net Food-Importing incentives to develop and adopt improved technol- Developing Countries," discussed later in this chap- ogy, and on continued public investments in the ter. The implementation--or lack thereof--of this necessary institutions to support sustained agricul- Decision has increased the concern of many devel- tural growth. Further reduction of the major dis- oping countries and may drive their policies even tortions in agricultural markets can undoubtedly further in the direction of expensive attempts at make a significant contribution to the continued national food self-sufficiency. However, it is not achievement of global food security in two ways. clear that WTO agricultural negotiations are the First, it can encourage distribution of production forum in which to address this issue in a definitive among countries based on true comparative advan- way. The WTO is neither a multilateral financial tage. Second, trade barriers that inhibit the spread institution nor an international development assis- of new technologies that are embedded in agricul- tance agency. Attempts to address the problem tural inputs can significantly reduce global food within the context of the WTO are likely to lead security in the long term, and these barriers to to frustration and/or further problems, with the technology diffusion should be dismantled through implementation of agreements falling outside the successful negotiations. WTO's purview. Finally, while not a food security issue per se, the related argument that cyclically low commodity Aftermath of the URAA: The Marrakesh Decision prices or import surges can damage the relatively fragile agricultural structure in developing coun- The Decision on "Measures Concerning the Possi- tries may have some weight and is examined ble Negative Effects of the Reform Programme on separately (chapter 10). A relevant issue would be Least-Developed and Net Food-Importing Devel- whether such a mechanism should be available oping Countries" was adopted by the Ministerial only to developing countries in the name of food conference in Marrakesh in 1994 that concluded security, or on a broader basis. the Uruguay Round. It was intended to respond to the fears of the LDCs, as well as of a number of other developing countries that are net importers International Food Security of food (NFIDCs), that subsidy cuts imposed on The concept of international food security is rarely their food suppliers by the URAA reforms would if ever discussed, yet it may be the most important have the effect of raising the price of their food element of national food security. imports. The requirements of the Marrakesh Deci- International food security means that the sion focus on regular review of food aid commit- aggregate supply of food produced in the world is ments and flows, the adoption of guidelines to adequate to provide sufficient calories to feed the ensure that food aid is given in grant form or Food Security and Agricultural Trade Policy Reform 185 on appropriate concessional terms, and technical much higher cereal import prices following the assistance. Motivated in particular by a spike in URAA, and that these higher prices are due to the grain prices in the mid-1990s, there was a proposal agreement. The basis of this is a comparison of by 16 LDCs and NFIDCs for the establishment of a cereal import prices during 1993­94 with revolving fund to help meet short-term difficulties 1995­96, and the attribution of the change to the in financing normal levels of commercial imports. URAA. Cereal prices were low during 1993­94 This proposed fund was intended to fulfill the and high during 1995­96, as table 9.1 shows. terms of the Marrakesh Decision. The proposal However, prices have declined since 1996 and would establish a countercyclical subsidy mecha- average import prices are now below the levels of nism whereby food and financial aid are automati- 1993­94 for the LDCs and slightly above for the cally triggered by commodity price increases owing low-income food-deficit countries. to agricultural trade liberalization agreed to under the Uruguay Round. In this context, it is useful to The Evidence Does Not Support the Argument examine retrospectively the evidence on the effects That Cereal Prices Rose during 1995­96 of the URAA on world food prices. Because of the URAA Research into the real and potential effects of Cereal prices rose in 1995 and 1996 primarily trade liberalization on world grain prices leads to because of low world grain stocks, which followed five important conclusions, summarized below. from a 3.2 percent decline in world production in 1995­96 (see figure 9.1). Because of a poor United States (U.S.) harvest, cereal stock relative to use fell Cereal Import Prices for NFIDCs and LDCs Have to an unusually low level at the end of the 1995­96 Not Risen Significantly during the 1990s marketing year (13.4 percent compared to an aver- The proposal for a revolving fund was based on age of 18.8 percent in the previous five years). High the contention that NFIDCs and LDCs are facing fertilizer prices and strong economic growth TABLE 9.1 Average Cereal Prices (US$/ton) Price Difference Price Difference 1993­94 1995­96 95­96 vs 93­94 1997­1999 97­99 vs 93­94 Low-income food deficit 175.3 229.5 30.9 193.0 10.1 Least-developed countries 224.9 252.1 12.1 203.0 9.7 Developing countries 180.6 229.7 27.1 181.3 0.3 Trade weighted cereal price (free on board) 137.1 184.4 34.5 125.7 8.3 Source: World Bank Databases. FIGURE 9.1 Cereal Import Costs per Ton 300 250 Ton 200 per 150 US$ 100 50 1990 1992 1994 1996 1998 2000 Low-Income Food Def Least Developed World FOB Prices 186 Agriculture and the WTO during that period contributed to the sharp price target cuts in protection included in the agreement, increases. Stock levels recovered in 1996 and con- for reasons discussed elsewhere in this volume. tinued to increase through 1998, driving world cereal prices lower. The URAA was not directly Real Cereal Prices Have Shown a Strong Secular related to the price increases during 1995 and 1996, Declining Trend nor the price declines since. The fact that grain import bills in some countries continue to be Primarily because of the tremendous increase in pro- higher despite the price declines is another indica- ductivity and yields in the past decades, the long-run tor that trade liberalization is not the cause. prices of cereals (at least wheat and rice) have shown It is, of course, possible that the URAA increased a downward trend. Real prices of cereals are near food prices relative to what they would have been in extremely low levels as figure 9.2 shows. its absence, even though they fell relative to past levels. That is, it may be true that without the Prices Are Unlikely to Increase in the Near Term URAA, prices would have fallen further than they in fact did. This is plausible, but even by this stan- The failure of the Uruguay Round to achieve sub- dard of comparison, price "increases" attributable stantial agricultural liberalization implies that to the URAA are estimated to be quite small. One much is left to be done, so a successful current estimate suggests that enough liberalization was round of negotiations could potentially achieve a achieved in wheat, for example, to raise estimated great deal of liberalization, which in turn could world wheat prices by only 3.8 percent (Goldin and potentially raise world agricultural prices. Yet the van der Mensbrugghe 1996, p. 169). By contrast, impact is likely to be felt only gradually and in the the same study estimated that the prices of coarse medium term. The round itself is scheduled to be grains and sugar would rise by only around 2 per- concluded only in 2005, and it appears likely that cent, and dairy products by a little over 1 percent. any significant steps in sensitive areas such as agri- For a number of other agricultural commodities, culture would be phased in over an even longer including rice and cotton, the study estimates period. Other structural supply and demand that there was so little overall liberalization that shifters are also at work, the net impact of which is prices would actually fall relative to the prices of difficult to predict. But judging from the experience manufactures exports. of the last decades, it would appear likely that these One reason for this limited impact on food prices will operate to depress prices. In addition, to the is that the URAA did not achieve a great deal of agri- extent that trade barriers have been reduced (or cultural liberalization. While it achieved an enor- will be reduced in the future), this implies reduced mous amount in terms of establishing rules, the world price variability and therefore better ability liberalization that it achieved was much less than the to plan and to hedge against price hikes. FIGURE 9.2 World Grain Stocks and Price, 1990­2000 Million Tons World Grain Prices, US$ 500 250 400 World grain stocks 200 300 150 200 Grain prices 100 100 50 0 0 1990 1992 1994 1996 1998 2000 Source: World Bank Databases. Food Security and Agricultural Trade Policy Reform 187 Research Suggests That Multilateral Trade past, can be extremely important in least-developed Liberalization Produces Benefits to Net and net food-importing countries. One of the stud- Food-Importing Countries That in Some Cases ies, which covers a wide range of low-income and Offset the Costs from Higher Food Prices net food-importing countries, shows that in a Several World Bank studies that examined the number of cases, including Bangladesh, the Arab impacts of Uruguay Round liberalization on least- Republic of Egypt, and Tanzania, the distortion developed and net food-importing countries effects operated in the opposite direction to the focused on the impacts of round-induced changes terms-of-trade impacts, and converted an apparent in external prices on these countries.3 These studies negative impact of the URAA into a substantially divided the impacts of changes in world prices into larger gain.5 The distortion impacts are extremely two components--terms-of-trade effects and dis- complex however, and sometimes worked against tortion impacts. The terms-of-trade effects of liber- the country. In India, for example, the distortion alization are well known and have frequently impacts were negative, primarily because the world featured in policy discussions, but the distortion prices of some its commodities on which the coun- impacts have only received attention in a small try's own policies imposed a net tax, such as rice number of studies,4 none of which focused on the and cotton, were estimated to fall. These price least-developed and net food-importing countries. declines reduced India's output of these commodi- The terms-of-trade impacts resulting from ties, which was already too low because of the use of changes in world prices are relatively straightfor- export restrictions. On the other hand, in Kenya ward. They are the impacts of changes in world and Zimbabwe the distortion effects greatly rein- prices on the costs of net food imports, or the ben- forced an initial gain from the changed terms of efits obtained from net exports. For a net importer, trade. an increase in prices will raise costs, with the size of Overall, the results of these studies suggest that the additional cost depending on the level of the policy-related impacts increased the gains from imports. For a net exporter, an increase in prices the URAA to most low-income and net food- will increase returns, with the magnitude of the importing countries. However, the impacts in any benefits depending on the level of net exports. The future liberalization will clearly depend on the spe- distortion impacts are more complex, because they cific nature of the agricultural policies used in the depend on the nature of an entire array of policy country and on the impacts on world prices of the distortions in the economy. They arise from the resulting liberalization. impact of changes in world prices on the extent to The overall conclusion from the above points, which countries undertake activities that are worth taken together, is that any substantial increase in less (or more) than their true costs. If, for example, international agricultural prices that could poten- a country has an export tax on a particular com- tially result from the international agreements modity, the resulting depressed domestic prices will (a) has not occurred yet, (b) is probably a number of lead producers to produce less of the good than is years away, and (c) could well be offset by other eco- socially optimal, and consumers to consume more nomic factors at work in the meantime. Further- than is socially optimal. Both increases in produc- more, any losses to net food-importing countries tion and reductions in consumption of such a good would not be uniform across countries, and may be will make the country better off. A rise in the world partially or completely offset in individual countries price of this export good will, assuming it is passed by other gains from the overall agreement. on to the domestic market, lead to such a net gain. An increase in the import bill (caused, for example, by an increase in import prices) may contribute to Measures to Ameliorate the Impact of Shocks That Threaten an exchange rate depreciation, which will have the Food Security same kind of effect in the market for this export, tending to offset the adverse terms-of-trade effect Apart from the long-term structural food security for the country from the higher import prices. issues on which previous sections of this chapter A key finding of the studies is that the distortion focused, transitory food insecurity at the household impacts, which have typically been ignored in the level can arise because of shocks on either the 188 Agriculture and the WTO demand or supply side, or both. Wars, political dis- countries cope with price shocks, as is the CFF, but turbances, economic crises, or natural disasters it does have several instruments to provide urgent may temporarily disrupt the income-earning assistance when a country is struck by an emer- capacity of the population (demand side), as well as gency that seriously dislocates its economy and productive capacity and commercial channels for calls for a quick response. Several types of emer- delivering food (supply side). Short-term but dra- gency loans, as well as adjustment loans, work by matic spikes in international food prices may also quickly disbursing funds to a government's budget, disrupt import channels by exhausting the lines of thereby making foreign exchange available for credit of importers. In the section below, the exist- additional food imports. This mechanism shares ing and potential mechanisms for dealing with this most of the shortcomings and the advantages of the kind of shock are examined. CFF, though the trigger is not intended to be price movements. IMF and World Bank Facilities Commodity Price Risk Management Measures The IMF's Compensatory Financing Facility (CFF) is the only international mechanism that is specifi- A wide range of instruments is now available com- cally designed to help alleviate adverse effects of cer- mercially to protect against price movements in the tain commodities' price movements that result in short term. Options can be used by economic balance-of-payments difficulties. This facility was agents with large exposure to risk from price move- used by only four countries during 1993­99. ments either to lock in a specific price or to ensure A number of developing countries have noted that that price will not be greater (or lower) than a spec- the CFF has several shortcomings,6 including lim- ified level. These could be readily used by large pur- ited commodity coverage (the only import com- chasers of foodstuffs--either private sector modities that are eligible are grains), perceived commercial importers or government agents (food requirements of "policy conditionality," and the lack corporations or state-owned enterprises). They of concessionality in terms. With respect to the last could also be used by governments to insulate point, it is worth noting that the terms are almost safety nets (that is, targeted poverty programs such certainly better than commercial terms, though not as food stamps) from budgetary crises arising from as good as those of the IMF's concessional facilities. food price increases. In other words, a government Perhaps a more relevant criticism in the context of a agency that knows that its expenditures will discussion of food security is that the loan is taken increase when food prices rise (either because it by the government, and without some additional provides food in-kind and must procure the internal distribution mechanism, the benefits may higher-priced food itself, or because it must give not be shared with the individuals within the coun- higher income support to the poor) can purchase try whose food security is put at risk by the high options that would pay out when import prices food prices. This may then require that the govern- increase beyond a certain level. This would then ment rely on direct distribution to needy individu- ensure that a country would have the budgetary als, which undermines the development of efficient resources required without creating an unantici- private marketing channels. Nonetheless, use of the pated fiscal burden on the government. The price facility could at a minimum reduce the magnitude insurance would protect against short-term fluctu- of the disruption of imports and increase the aggre- ations in price, but not against long-term down- gate supply of food in a country. ward trends. Most of the World Bank's lending and nonlend- The International Task Force on Commodity ing services are aimed at the root cause of long- Price Risk Management, with its secretariat in the term food insecurity: poverty. Some investment World Bank, is currently implementing pilot proj- projects are also aimed at improving trade and dis- ects to provide commodity price insurance to tribution facilities, or improving a country's trade smallholder farms in developing countries, and this finance infrastructure, thereby reducing vulnerabil- could possibly be extended to consumers in food ity to disruptions of imports. The World Bank has importing countries as well. This insurance no facilities that are specifically designed to help scheme, while varied in its specific application by Food Security and Agricultural Trade Policy Reform 189 country, is focused on adapting the commodity because distribution is free or actual sales prices are price risk management tools available on interna- far below local prices. This tends to depress prices tional markets to provide a form of price insurance for local producers, lowering their incomes, reduc- to producers (and consumers in the case of import ing incentives for production, and thereby increas- prices). The World Bank also has developed instru- ing the country's future reliance on food imports. ments to support the development of crop yield In addition, the price reductions created by food insurance and has one operational pilot project in aid distort seasonal (interharvest) price movements Morocco, with others under preparation. and discourage the development of private storage facilities. Finally, food aid is in many cases distrib- uted through governmental channels, thereby Commodity Swaps undermining private sector development in mar- The World Bank began offering commodity swaps keting infrastructure. While such adverse effects in 1999, to help borrowing countries address may on specific occasions be acceptable costs to pay commodity price risk. Bank members whose debt- for mitigating or avoiding famine, they should be servicing capacity is heavily exposed to specific evaluated on a case-by-case basis. commodities may benefit from financial terms that Many of these same shortcomings are true of link the debt-servicing costs of their loans from the subsidized credit for food imports (including World Bank to movements in the price of a specific export credit from the exporting countries): the commodity or basket of commodities. Managing subsidized imports depress farmers' prices and dis- the risks associated with commodity prices is a courage production; if not done on a purely com- major challenge for many developing countries, and mercial basis, they can be poorly timed (for most commodity-dependent countries have inade- instance, in conformity with the requirements of quate access to risk-hedging instruments. It was to the export credit agency, not with the demands in address this need that the World Bank began offer- the importing country) and thereby distort sea- ing commodity swaps on a pilot basis to its borrow- sonal price movements; and they are often chan- ers with a demonstrated need to reduce exposure to neled through government agents, undermining commodity price changes. Such swaps are individu- private sector market development. ally negotiated transactions to exchange two sets of Both food donations and concessional credit cash flows at certain dates in the future, whereby programs also obscure the real value of the grant one set of cash flows is linked to the market price of element involved and run counter to the principle a commodity or index and the other is a pre-agreed of transparency. If a grant is justified, it should be fixed cash flow or a cash flow based on a floating or transparent and not disguised through an fixed rate of interest. Hedges could be structured to unknown grant element embedded in in-kind food reduce commodity price­related risks for both pro- aid or subsidized interest rates. ducers and consumers of energy, minerals, and agri- cultural products. New Approaches to Handling Short-Term Threats to Food Security Food Aid and Subsidized Credit Two other mechanisms have been proposed to Food aid is often the response of the international meet the concerns addressed in the Marrakesh community to disruptions of domestic supply in Decision, but not yet implemented, to deal with developing countries. While food aid has served a short-term food supply disruptions. These, sum- valuable function in cases of disasters when normal marized below, could be useful in meeting these supply channels are disrupted, it has in many cases concerns in the new round. had undesirable and unintended consequences. In general, food aid provides significant quantities Food Vouchers One proposal that would not that have a potentially large impact on the local mar- entail several of the shortcomings of direct food aid ket price. While an increase in supply will lower the is to establish a system to distribute food vouchers market price, even if it is sold at "market prices," this to the needy when food prices increase.7 These effect of distorting the market is often magnified vouchers could be redeemed through normal 190 Agriculture and the WTO commercial import and distribution channels and Notes therefore would not undermine private sector 1. Diaz-Bonilla, Diao, and Robinson (2003) have a good dis- development. They would also operate to increase cussion of the various submissions to the WTO on this subject demand, not supply, and as a result they would not by both developed and developing countries. 2. Some analysts distinguish between household and indi- distort domestic prices for either producers or con- vidual food security, but this distinction has little relevance for a sumers. The difficulty in this is that it requires a discussion of trade policy. system for identifying target recipients and distrib- 3. Anderson (1998a, b, c); Ingco (1997). These studies were uting the vouchers to them. (This problem is not undertaken using a relatively simple numerical model designed so that it could be applied to countries where data availability is unique to this scheme; any compensatory mecha- limited. These models allow the evaluation of the impacts on nism that does not identify those most in need will least-developed countries of different world price changes have a very high cost relative to the benefits it pro- resulting from future liberalization proposals. 4. These include Anderson and Tyers (1993); Tyers and vides to the poor.) Many countries already have at Falvey (1989); and Alston and Martin (1995). least some existing mechanisms that could be 5. See Ingco (1997, p. 16, Scenario II). adapted for this. Experience with similar programs 6. For a more detailed discussion, see WTO (2002). 7. See WTO (2002, annex 8, Submission by the World Bank). such as food stamps or other income support 8. See WTO (2002, annex 7, Submission by UNCTAD). mechanisms is accumulating as more countries begin to use them as part of their safety net pro- grams. Since there is significant lead time before Select Bibliography any structural price increases would be felt from the new round of trade negotiations, there is some Alston, J. M., and W. J. Martin. 1995. "Reversal of Fortune: Immiserizing Technological Change in Agriculture." Ameri- time to install such programs in countries that may can Journal of Agricultural Economics 77: 251­259. be vulnerable. Anderson, K. 1998a. "Economic Reform in Nepal and WTO Accession." Adelaide: Centre for International Economic Studies, Adelaide, Australia. An Ex Ante Fund for Trade Finance Sudden Anderson, K. 1998b. "Vietnam's Transforming Economy and increases in the international food prices may disrupt WTO Accession." Adelaide: Centre for International Eco- nomic Studies and Singapore: Institute of Southeast Asian imports if this causes the demand for trade finance to Studies, Adelaide, Australia. exceed the credit available to the importers in the Anderson, K. 1998c. "Lao Economic Reform and WTO Acces- country.8 One way of dealing with this would be to sion." Adelaide: Centre for International Economic Studies and Singapore: Institute of Southeast Asian Studies, Ade- establish a fund to make available to importers, guar- laide, Australia. antees that would enable them to access the necessary Anderson, K., and R. Tyers. 1993. "More on Welfare Gains to financing. The availability of these guarantees would Developing Countries from Liberalising World Food Trade", Journal of Agricultural Economics 44(2): 189­204. be triggered by objective criteria--agreed to in Christiaensen, L., L. Demery, and S. Paternostro. 2002. "Growth, advance--that are outside the control of the govern- Distribution, and Poverty in Africa: Messages from the ments. (Granting access based on an increase in the 1990s." Processed. The World Bank, Washington, D.C. import bill for food would reward poor economic de Janvry, A., Graff, G., Sadoulet, E. and D. Zilberman. (1999). "Agricultural Biotechnology and Poverty: How to Make the policies--for example, food subsidies or macroeco- Promise a Reality?" Preliminary draft, University of Califor- nomic mismanagement--which could trigger a nia at Berkeley. sharp increase in food imports.) An example of such Diaz-Bonilla, E., X. Diao, and S. Robinson. 2003. "Thinking Inside the Boxes: WTO Agricultural Negotiations and the an objective trigger would be some indicator of world Development and Food Security Boxes." Paper presented to wheat prices rising above a specified level or objective joint IATRC/USDA-ERS/UC Davis/University of Calabria warning indicators of famine. In such cases, "International Conference: Agricultural policy reform and the WTO: Where are we heading?" June 23­26, Capri. importers could access the guarantees up to an estab- Ferreira, F. H.G., and P. Lanjouw. 2000."Rural Non-Farm Activities lished limit. The feasibility of such a facility is being and Poverty in the Brazilian Northeast."Draft Mimeo. Develop- explored by a WTO panel. ment Economics Group,World Bank,Washington, D.C. Food Security and Agricultural Trade Policy Reform 191 Garrett, J. L., and M. Ruel. 2000. "Achieving Urban Food and Longhurst, R. and M. Lipton. 1989. "The Role of Agricultural Nutrition Security in The Developing World." IFPRI 2020 Research and Secondary Food Crops in Reducing Seasonal Focus 3. IFPRI: Washington, D.C. Food Insecurity." In D. Sahn (ed.), Seasonal Variability in Goldin, I., and D. van der Mensbrugghe. 1996. "Assessing Agri- Third World Agriculture: The Consequences of Food Security. cultural Tariffication Under the Uruguay Round." In W.J. Baltimore and London: John Hopkins University Press. Martin, L.A. Winters, eds., The Uruguay Round and the Sutherland, P. D. 1998. "Answering Globalization's Challenges," Developing Countries. Cambridge: Cambridge University ODC Viewpoint, October. Washington, D.C.: Development Press. Council. Haddad, L., M. Ruel, and J. L. Garrett. 1999. "Are Urban Poverty Tyers, R., and R. E. Falvey. 1989. "Border Price Changes and and Undernutrition Growing? Some Newly Assembled Evi- Domestic Welfare in the Presence of Subsidised Exports," dence." Food Consumption and Nutrition Discussion Paper Oxford Economic Papers, 41(2): 434­451. 63. IFRPI: Washington, D.C. World Bank. 2001. "Rural Poverty Reduction in Brazil: Towards Ingco, M. D. 1997. "Has Agricultural Trade Liberalization an Integrated Strategy." Processed. Washington, D.C. Improved Welfare in the Least-developed Countries? Yes." WTO (World Trade Organization). 1999. Annual Report, 1999. Policy Research Working Paper 1748, International Trade International Trade Statistics. Geneva: WTO. Division, International Economics Department, World ------. 2002. "Report of the Interagency Panel on Short-Term Bank. Washington, D.C. Difficulties in Financing Normal Levels of Commercial Kherallah, M., C. Delgado, E. Gabre-Madhin, N. Minot, and M. Imports of Foodstuffs." 28 June, WT/GC/62/G/AG/13. Johnson. 2000. "Agricultural Market Reforms in Sub-Saha- Office of the Secretary General, Geneva. ran Africa: A Synthesis of Research Findings." Markets and Structural Studies Division, IFPRI. Washington, D.C. 10 MANAGING POTENTIAL ADVERSE IMPACTS OF AGRICULTURAL TRADE LIBERALIZATION William Foster and Alberto Valdés Introduction countries. The paper considers national border When a country's agricultural sector is integrated measures particularly related to international price more closely into the world economy by reducing fluctuations. Other chapters address questions its trade barriers, its producers are exposed to relating to safety-net policies (such as revenue greater downside price risk. Policymakers in devel- insurance), international assistance measures (such oping countries have few instruments available to as food aid and compensatory financing), and the cushion farmers from episodes of very low world concerns regarding the effects of food price move- prices, which in many countries has made them ments on real wages and their potential macro- reluctant to make a serious commitment to signifi- economic repercussions. We take the perspective cantly reduce tariffs and maintain them at low lev- that the price risk management problem is not so els. This may help explain why so many developing much the instability of international prices per se, countries bound their agricultural product tariffs at but the difficulties associated with periods of low extremely high levels in the Uruguay Round. It and politically unsustainable price levels. The gov- would clearly be in the interests of both developed ernments of many developing countries under- and developing countries for all participants to standably distrust the present state of their domes- make commitments to significantly reduce their tic markets and institutions to manage the risks to import barriers in the Doha Round, but to induce farmers of extended periods of low prices. Such dis- developing countries to participate fully in the trust and the associated political pressures from Doha process may require some creative thinking farm groups--especially those competing with on how to meet this political concern over imports--would have the effect of reducing any increased price risk when tariff walls are lowered. enthusiasm for further trade liberalization. It is in In this paper several questions surrounding con- this context that the distortions in world prices and tinuing trade reform will be addressed as they relate the issue of the transmission of price risk become to the management of the price risk originating in critical in the political commitment to deepen world markets that affects farmers in developing trade reforms in developing countries. 193 194 Agriculture and the WTO The Plan of This Chapter larger numbers of developing countries--eventually including the transition economies of Eastern and The structure of this paper follows what are, in Central Europe--adopted trade liberalization, priva- the authors' opinion, the central, politically sensi- tive issues1 from the perspective of a developing tization, and deregulation. The degree and depth of these reforms, of course, varied significantly. Chile country in the process of expanding or deepening was an early and deep reformer, beginning with trade liberalization: a radical change in its trade regime around 1976. In · The enhanced transmission to domestic markets contrast, Venezuela and Caribbean countries, for of international price movements due to trade example, began reforms late, in the mid-1990s, coin- reforms, especially the elimination of quantita- ciding with the emergence else-where of the transi- tive restrictions (QRs), and the limited policy tion from command socialism policies. tools available in the present environment of The move toward the adoption of a world price trade commitments. regime--trade liberalization--meant (particu- · The definition of price risk in terms of periods larly in Latin America) the reduction in direct of very low prices facing producers in import- price interventions, the lowering of tariff levels, competing sectors. and the convergence of previously disparate tariffs · The resistance in developing countries to across commodities. And--perhaps most criti- lowering trade barriers on imports caused by cally in terms of enhancing price transmission-- developed-country subsidies that artificially trade liberalization meant the elimination of all lower world prices, which are in turn transmit- but a few quantitative restrictions on imports and ted to domestic developing-country markets. the removal of state trading in most countries. In One argument made in developing countries is the past these policies had the effect of insulating that activities that would otherwise be competi- domestic consumers and producers from the tive at world prices without distortions are not vicissitudes of world markets and international sustainable in the current environment of dis- trade. The decline of trade-restricting policies torted international prices. magnified the effects on domestic markets of · The stochastic nature of world prices is charac- changes in border prices--that is, it magnified the terized by occasional periods of persistently low transmission of price signals across borders as prices below trend. This persistence characteris- the determinants of a country's allocation of tic limits the type of policies that might be used resources. (This is what is meant when reference is to manage price risk. made to developing countries coping with a · The practical relevance of enhanced price trans- "world price regime.") Consequently, the argu- mission. The price risk facing farmers is not ments surrounding both price risk and the selec- merely caused by world price movements, but by tion of policies to counteract possible distortions other factors, such as exchange rate fluctuations. in international prices have grown in importance That price transmission is not the only question in domestic policy debates. of significance implies the reduced effectiveness In the more protectionist economic environ- of border instruments to manage price risk. ment of the past--especially with quantitative · Policy implications of price risk management, restrictions related to import licenses and quotas, in the present environment of World Trade variable levies, and state trading monopolies-- Organization (WTO) commitments and the logically there should have been less transmission functioning of world markets. of international price variability than under the present and evolving system, which emphasizes The Move to World Price Regimes: the use of tariffs only. And even where no quanti- Greater Price Transmission and the tative restrictions existed, high tariffs would have Limits to Managing International tended both to reduce the practical importance of Price Variability international price fluctuations, and to cushion In the mid-1980s, several developing countries, par- the effect of those shocks to whatever extent they ticularly in Latin America, began a series of eco- were transmitted from international to domestic nomic reforms. Continuing through the 1990s, markets. Managing Potential Adverse Impacts of Agricultural Trade Liberalization 195 Today the set of policy instruments available to export taxes and restrictions of many developing governments is largely restricted to tariffs and sur- countries on exportables. This had the combined charges (including safeguards), and the levels of effect of reducing the bias against export agricul- those tariff instruments are limited.2 For many ture. So while the same characteristics of world developing and transition economies, the present prices of importables also apply to exportables, the trade and policy environment has amplified politi- beneficiaries of trade liberalization tended to be cal pressures--in a situation of falling world producers in export-oriented sectors (and of course prices--to counteract the transmission to internal consumers). Furthermore, it appears to be a widely markets of the debilitating effects generated by per- noted fact that, with respect to policy issues, pro- ceived distortions in world prices caused by external ducers of importables usually are better-organized, interventions. Particularly worthy of attention are more vocal, and stronger lobbyists than producers the pressures arising in import-competing sectors. of exportables. The implications of low-price periods to pro- This political reality reinforces the prominence ducers in import-competing sectors, however, are of importables in policy debates over that what is really important about price risk in the con- of exportables. Simply put, the political economies text of trade reform. of the two types of products are different. On top of The adverse effect of high levels of protection in all of this, the present policy tools (for example, developed, industrial countries (addressed below) increasing tariffs) for the protection of importables is a widely recognized problem for developing are straightforward and bureaucratically remuner- countries as exporters of agricultural products. But ative. For exportables there is not much that fiscally what is likely an undervalued question (especially limited developing countries can realistically do, by economists in the "North"), and what generates unlike the case of richer economies, such as the the most complex domestic policy debates in many United States (U.S.) and the European Union (EU), developing countries, surrounds the pricing of where export subsidies represent a relatively minor importables. In practical political terms, for these and often inconspicuous burden on taxpayers. countries, the problem of access to foreign markets is not a contentious issue in internal debates over Greater Price Transmission Is domestic policies, although it is clearly an important Perceived as Increasing Producer issue in terms of foreign relations. Vulnerability The question of access is certainly central to the international diplomatic debate over trade with What evidence is there that greater price transmis- Organisation for Economic Co-operation and sion is perceived as increasing producer vulnerabil- Development (OECD) members. Nevertheless, as ity? The 1992 study of the political economy of we observe in the analysis of countries that have agricultural price policies in 18 developing coun- opened their markets significantly to trade, the tries by Schiff and Valdés shows that governments dilemma today is dealing with the episodes of tend to insulate domestic sectors to a greater degree "excessively low" border prices affecting some during periods of volatile world prices. If attaining import-competing activities. Moreover, the con- price stability was the purpose for such interven- cerns over low price episodes are reinforced by the tions, then in overall terms governments succeeded undeniable long-term declining historical trend in in achieving the objective, primarily through direct world agricultural prices. Often in economic litera- interventions. Relative to world prices, protective ture, price risk is understood, in the simplest terms, policies managed to reduce domestic price variabil- as the variance of prices. But in the context of trade ity by an average of 25 percent, and even more so in liberalization and efforts to encourage govern- products for which world prices were highly ments to move toward a world price regime, that volatile. Quiroz and Soto's 1995 cross-country definition is too narrow. Price risk should be con- analysis of the transmission of world agricultural sidered in terms of price levels, and not simply by price shocks reinforce these conclusions. In 67 of some measure of the variance of prices. 76 countries examined, it took at least two years Trade liberalization did two things: it reduced to transmit 50 percent of international price shocks border protection on importables and it removed to domestic markets. 196 Agriculture and the WTO The historical policy decisions of governments be competitive in world markets without distortions, could be interpreted as evidence, in the sense of but that they are unprofitable in the current environ- revealed preference, that they think reductions in ment of distorted international prices. In terms of price transmission effectively reduce the vulnera- political decisions in these countries, the implication bility of producers. This is what should be expected of this argument to the selection of price risk man- from a welfare and public choice perspective, where agement tools is to mix policy objectives. The politi- commodity prices are politically sensitive. In terms cal imperative is that such instruments should not of farmers in developing countries, governments merely guard against low prices, but simultaneously tend to be sensitive to price instability because compensate for an artificially caused downward shift farmers tend to be risk averse, there are also fewer in the long-term pattern of prices owing to devel- opportunities to hedge risk, credit markets are less oped-country subsidies. developed, and governments do not have the fiscal Is this a reasonable argument? What then is resources to provide nonborder support to farmers known about the manner in which developed- in years of low prices. In terms of consumers and country policies distort the levels of prices around labor markets, without the fiscal wherewithal to their long-term trends? In the context of the policy provide broad coverage of safety net programs, debate in many developing countries, reference is variation in food prices can have significant effects often found to the billions of dollars given to agri- on real wages and real household income.3 culture in developed nations. But it is an open As emphasized earlier, from the perspective of question as to the degree to which current com- politicians and farmers in import-competing sec- modity price levels and the declining trend in prices tors, the policy problem is not so much the volatility are due to protection and government support in of domestic prices per se, as those are influenced by developed countries--and to what degree they are world price fluctuations. It is rather the difficulties the result of productivity levels, factor use, income associated with periods of exceptionally low growth, and population.4 price levels falling below the long-term trends. Often explicitly, there is the presumption that Aggravating the political pressures to "do some- international prices are so distorted by external thing," producers and politicians in developing subsidies that they do not represent a sound basis countries tend to attribute these episodes to the dis- for the determination of the true competitiveness tortions in world prices brought about by the con- of domestically produced importables. Moreover, tinued subsidies of developed countries. there is often the confusion in the policy debate that the producers in developing countries should be compensated for the absolute value of the The Role of Developed-Country export subsidies, without a clear distinction Subsidies in Lowering World Prices between measurable subsidy levels and their over- From an economic perspective, the world price of a all final effect on world prices. From an econo- product, however determined, represents the true mist's perspective, of course, what should matter is cost of that product to an economy, and--abstracting the final effect of subsidies on world prices, not from dynamic considerations--the welfare of the absolute subsidy levels. The focus on absolute lev- economy as a whole is unambiguously maximized if els of subsidies leaves the question of the appropri- the world price is also the domestic market price for ate compensating tariff or surcharges with an both consumers and producers. In the political unrestricted upper limit.5 realm, however, the underlying determinants of the What should be considered most relevant for the international market conditions may matter a lot. discussion, in terms of the familiar steady-state The resistance in developing countries to lowering comparisons derived from modeling efforts, is that trade barriers on imports arises owing in part to the practically all quantitative analyses on agricultural widely held assumption that developed-country protection conclude that such protectionist subsidy subsidies artificially and significantly lower world policies depress world prices relative to what would prices. Lower prices are in turn transmitted to otherwise be the case with lower protection. This domestic developing-country markets. Many indus- point is concisely made by McCulloch, Winters, tries in developing countries argue that they would and Cirera (2002, 175­6): "Agricultural markets are Managing Potential Adverse Impacts of Agricultural Trade Liberalization 197 among the most distorted in the world, with both published studies on predicting the world price developed and developing countries maintaining effects of partial agricultural trade reform. In Ingco, high levels of intervention." for rice and coarse grains, world price effects on the But first one should keep in mind the impor- order of 5 percent or less are a good approximation. tance of distinguishing between the impacts of dis- For wheat, the estimates range from 1.2 percent to tortions from developing countries and those from 10 percent, where three of the six estimates are--for developed countries. The evidence indicates that all intents and purposes--approximately 6 percent. the direction of the effects of intervention differs For tropical products the effects are lower. between developed and developing countries. Goldin, Knudsen, and van der Mensbrugghe Developed countries usually protect their agricul- (1992) use the RUNS model8 to simulate both "par- tural sectors, and developing countries usually tax tial" and "full" agricultural reforms and multisec- their agricultural sectors. Certainly the historical toral reforms. For partial agricultural reforms, tariff evidence suggests that developed countries' policies equivalents are reduced by 30 percent; for full have depressed world prices, while developing agricultural reforms all tariffs are eliminated, countries' policies have increased those prices.6 In including those for inputs. Partial reforms are addition, complicating the issue in the case of similar to those mentioned above, but for full agri- developing countries, the effects of exchange rate cultural reforms the world price effects are consid- misalignment and industrial protection have been erably and surprisingly higher, especially when shown to be influential in depressing the internal compared to studies that address similar simula- prices of agricultural tradables, unlike the situation tions. For example, sugar prices rise from 10.2 per- in developed countries. cent in the case of partial agricultural reform to Regardless of past opinions, it is well recognized 59.3 percent in the case of full agricultural reform. today that domestic price and trade interventions Wheat prices rise 5.9 percent for partial reform to among the major producing and trading countries 39.2 percent for full reform. For multisectoral continue to have a significant effect on world prices. reform, the price effects in general are lower, owing The higher the level of protection on importables and to both demand and supply effects in a hypothetical the higher the export subsidies, the lower will be the world of more efficient use of resources. Notably, world price. From a welfare-analysis point of view, practically all studies conclude that the largest it is recognized that consumers of the country effects appear to be for sugar and dairy, on the order imposing these policies suffer the main costs of inter- of at least 10 percent in the case of partial reform, vention. For agriculture sectors, however, developed- and higher for full agricultural reform.9 country policies can also impose high costs, particu- larly in developing economies where agriculture is The Stochastic Nature relatively more important and where rural poverty is of World Prices of considerably greater social importance. There have been several efforts to quantify the What is known about price variability and the per- present effects of these interventions on world sistence of low prices? The measurement of price prices and trade flows. This literature is discussed in instability in world agricultural markets is a rela- publications including Goldin and Knudsen (1989) tively well-researched area. In recent years Alexander for the OECD, and various other sources. The mod- Sarris' (1997b) work on measurement, for example, eling efforts currently undertaken by the Global has been comprehensive in the case of cereals. He Trade Analysis Project and the OECD are worth arrives at several conclusions. With respect to the mentioning as the most significant in terms of question of whether or not there has been an recent work. While there are some methodological increase in interyear and intrayear price variability caveats to keep in mind,7 these studies can be used for cereals, the evidence shows no trend toward as points of reference that offer a sense of the orders greater world price instability. Harwood comes to of magnitude involved in world price distortions. the same conclusion for corn, using data from 1920 What can be concluded from the literature? Two to 1996. Apparently, no other comparable studies comparable papers, Valdés and Zietz (1995) and have been applied to other commodities of interest Ingco (1996) summarize the findings of various to developing countries. 198 Agriculture and the WTO TABLE 10.1 Coefficients of Variation of World Prices of Selected Commodities Period Corn Rice Sugar Wheat 1960­72 0.11 0.09 0.60 0.25 1973­85 0.31 0.44 0.81 0.42 1986­97 0.18 0.20 0.22 0.17 1960­97 0.37 0.46 0.90 0.44 Source: Valdés 2000, p. 21. Table 10.1 presents the coefficients of variation context of an open economy, a central problem of of world prices for four selected commodities for the design for policy instruments to deal with 1960 through 1997 (reproduced from Quiroz, instability is understanding the nature and dura- Foster, and Valdés 1999). The results show that tion of price cycles in world markets. Do shocks to sugar is considerably more unstable than the three international prices dissipate rapidly, or are they cereals throughout the period. Also notable is phenomena that persist for several years? There is that the sub-period 1973­85 shows a much higher now a rich literature on the times-series proper- coefficient of variation than the preceding and ties of commodity prices. Early research, often following sub-periods. Evidently the coefficient of framed within the context of unit-root analysis, variation of these commodities changes with time, led to the conclusion that prices exhibited a signif- and moreover, for wheat and sugar, there has been a icant degree of shock persistence. The results pre- decline in price instability for the period 1986­97 sented by Gersovitz and Paxson (1990) show that (which is consistent with the conclusions of Sarris). the hypothesis of a unit root could not be rejected A related question is whether the instability is for a number of commodity prices of African caused by the "system" of protection (see, for exam- exports. Later studies have been more cautious in ple, Josling 1980). Certainly establishing variable coming to the strict conclusion of unit roots levies and quotas with a targeted domestic price on deduced from simple measures of persistence importables results in a shift of domestic instability (Caner and Hansen 1997; Perron 1989; and Zivot to world markets. With the move toward removal and Andrews 1992). León and Soto (1995) of quantitative restrictions and variable levies, it conclude that real world prices do not exhibit unit would be expected that the contribution of devel- roots, but rather they are trend-stationary oped countries' distortions on world price instabil- processes. Nevertheless, there remains the stylized ity would be less. As a result of the Uruguay Round fact that commodity prices exhibit considerable tariffication by developed and developing coun- shock persistence (see also Deaton 1992). tries, one expects to see reduced price instability Especially pertinent--and in the authors' opin- but increased transmission of price instability to ion most convincing--is the 1999 study by Cashin, domestic producers and consumers. Tyers and Liang, and McDermott on the half-life of shocks to Anderson (1992) conclude from policy simulations world commodity prices. In the case of wheat, for of tariffication in industrial countries that example, international price shocks have a median "[t]he effect of tariffication is to reduce [world] half-life of 44 months, with a 90 percent confidence price volatility substantially" during the 1990s interval that implies a range from an extreme low (p. 264). Tariffication in developed countries helps half-life of 14 months to an extreme high of "infin- reduce instability in wheat, dairy products, and ity." It is significant that there is a 50 percent proba- beef, among others. For some commodities, how- bility of prices prevailing below the expected value ever, reduced instability would derive primarily (declining over time) for more than 44 months. from the tariffication policies of developing coun- The empirical evidence from Cashin, Liang, and tries, such as in the case of rice and sugar. McDermott (see also table 10.2) is that the distri- Instability per se is one issue, but more impor- bution of prices is not symmetric--low prices tant to understanding the price risk facing agri- endure for more months than high prices. cultural producers in developing economies is the As figure 10.1 amply demonstrates, the nature question of the persistence of low prices. In the of price movements is such that low prices have Managing Potential Adverse Impacts of Agricultural Trade Liberalization 199 TABLE 10.2 Descriptive Statistics for Selected Commodities Corn Rice Period Skewness Kurtosis Jarque-Bera Skewness Kurtosis Jarque-Bera 1960­72 ­0.34 2.34 5.87 ­0.51 1.87 15.14 1973­85 1.01 3.58 28.5 1.59 5.53 107.3 1986­97 1.14 4.73 47.76 1.25 4.73 51.64 1960­97 0.65 3.77 42.6 1.21 6.3 310.64 Sugar Wheat Period Skewness Kurtosis Jarque-Bera Skewness Kurtosis Jarque-Bera 1960­72 1.88 6.47 170.7 0.35 1.65 14.39 1973­85 2.05 8.51 306.98 1.69 6.06 134.79 1986­97 0.62 3.12 8.85 0.43 2.22 7.86 1960­97 3.25 17.66 48.35 1.41 6.1 325.08 Source: Quiroz, Foster, and Valdés 1999. FIGURE 10.1 Real Price of Selected Commodities (1960­97, US$ of July 1997) Corn Rice US˘/bu US˘/cwt 1400 120 1200 100 1000 80 800 60 600 40 400 20 200 0 60 65 70 75 80 85 90 95 60 65 70 75 80 85 90 95 Sugar US˘/lb Wheat US˘/bu 200 2500 2000 150 1500 100 1000 50 500 0 0 60 65 70 75 80 85 90 95 60 65 70 75 80 85 90 95 Notes: bu bushels, cwt hundred weight, lb pound. Years are represented on the X axis. Source: Quiroz, Foster, and Valdés 1999. 200 Agriculture and the WTO the tendency to persist for many months, with domestic prices. Covering 78 countries, Quiroz and occasional spikes of shorter duration. These char- Soto conclude, "In an overwhelming majority of acteristics of world price movements lead to cases transmission of international price signals in notable difficulties in the design of policies. The agriculture is either non-existent or low, by any use of futures markets would reduce the effect of reasonable standard." But the degree of price trans- short-term uncertainty but could not guard mission, of course, ranges across countries. For against the effects of consecutive years of low Australia, Canada, New Zealand, and Uruguay, prices. In the past, minimum import price 50 percent or more of any given shock from world schemes were popular, and several developing prices is transmitted within a year. But these few countries still have in place systems of price countries are in contrast to 30 countries that bands. Safeguards (increased tariffs) are always an exhibit price transmission characteristics such that applicable contingency measure, but under WTO it takes five years or longer to transmit half of any rules they are limited in duration, involve com- world price shocks. Surprisingly, another group of pensation, and are restricted in their frequency of 30 countries appears "virtually isolated from inter- application. national price signals." One major caveat of these mentioned studies is that the data correspond to trade regimes that The Practical Relevance of Price applied before the Uruguay Round. So with tariffi- Transmission in the Determination cation and the removal of QRs, transmission could of Price Risks Facing Farmers be higher today, and this offers a rich area of study. The degree of world price distortion due to developed-country subsidies and the stochastic The Relevance of Price Transmission nature of world prices set the stage in which price transmission--endogenously determined by poli- What is the practical relevance of enhanced price cies such as trade barriers--becomes a relevant transmission to the determination of real domestic question in the context of policy controversies in prices? The answer lies in price decomposition into developing countries. There has been a lively debate several elements. It is useful to return to the in the literature regarding the degree of price trans- Mundlak and Larson study (or that of Quiroz and mission. For European producers, Tyers and Soto 1995) to aid in understanding the alternatives Anderson (1992) demonstrate a transmission rate for specifying the question of the relevance of price of world price variation that is less than 10 percent transmission. The regression-oriented decomposi- for the short run, and 20 percent for the long run. tion of Mundlak and Larson results in an isolated More generally, Sarris (1997b) mentions transmis- elasticity of world price transmission, controlling sion coefficients of 0.24 and 0.58 in the short and for the variability of other factors. That is to say, it is long run, respectively. In a controversial paper, an elasticity derived from partial correlations Mundlak and Larson conclude that "the deviation underlying regression analysis, but in and of itself it from unitary elasticity is, on the whole, surprisingly says little about the magnitudes of the absolute small" (1992, p. 419). In their judgment, while poli- effects of world price changes on domestic producer cies affect price levels, they appear ineffective at prices. A different approach consists of decompos- insulating to a great degree domestic prices from ing the total variation of domestic prices as a sum of world price changes. The authors also note that changes in world prices, exchange rates, and border variations in world prices are the major contributor protection. In a practical sense, the isolated effect of to variations in domestic prices, a conclusion dis- price transmission per se might not be the most cordant with the evidence from the price decompo- important issue facing farmers in developing coun- sitions presented below. tries. The practical impacts of world prices on In contrast to Mundlak and Larson, Quiroz and domestic prices are, at a minimum, filtered through Soto (1995) made use both of spot prices (in place fluctuations in exchange rates, which are well of the Mundlak and Larson approach based on the known to be much more volatile in developing Food and Agriculture Organization [FAO] average economies. In addition, historically, changes in unit export values) and of an error-correction spec- border protection have been of significant impor- ification of the relationship between world and tance during the 1990s in many countries, influenc- Managing Potential Adverse Impacts of Agricultural Trade Liberalization 201 TABLE 10.3 Decomposition of Real Producer Price for Wheat in Transition Economies Percent Change in Domestic Border Real Residual Country Period Price Price Exchange Rate Policies Bulgaria 1994­95 2 6 ­8 4 1995-96 25 6 20 ­1 1996­97 ­2 ­18 22 ­6 1994­97 25 ­6 34 ­3 Poland 1994­95 ­8 ­14 ­8 14 1995­96 13 28 ­2 ­13 1996­97 ­9 ­24 3 12 1994­97 ­4 ­10 ­7 13 Romania 1994­95 ­8 ­6 ­1 ­1 1995­96 13 3 4 5 1996­97 ­13 ­13 ­3 3 1994­97 ­8 ­16 0 8 Russia 1995­95 17 14 ­4 7 1995­96 16 7 ­2 11 1996­97 ­6 ­12 2 4 1994-97 27 9 ­4 22 Ukraine 1994­95 ­30 ­5 ­35 10 1995­96 ­9 7 ­20 4 1996­97 ­4 ­12 ­5 12 1994­97 ­43 ­10 ­60 27 Germany 1994­95 ­4 14 ­5 ­12 1995­96 1 11 3 ­12 1996­97 ­4 ­10 6 ­1 1994­97 ­7 14 4 ­26 Source: Valdés 2000, p. 21. ing the ultimate effects of changes in world prices economies related to the "neighborhood effect" of on domestic producers. the various economic crises--the Asian Crisis, It is relevant to consider the extreme case of the Tequila (or Mexico) Effect, the Brazilian devalu- countries that have fixed exchange rates and dollar ation, and most recently the Argentinean crisis. convertibility, such as Argentina, until recently. In Even if the price transmission elasticity were one for such a case, and where the rate of inflation would be Chilean exported fruits and wine (which is likely the close to zero, a price transmission elasticity of one is real case), the final effect of world price changes on expected. Moreover, any decomposition of domestic real domestic prices might be overwhelmed by the price changes would show a near one-to-one total accumulated influence of exchange variation. correlation between domestic prices and world In the example of real price trends of wheat in prices (abstracting from marketing margin the transition economies of Bulgaria, Romania, the changes). In contrast, in Chile, the real exchange Russian Federation, and Ukraine (Valdés 2000), rate fluctuated significantly in spite of a stable these countries have experienced large yearly real macroeconomic environment (with an inflation producer price variations. The decomposition of rate of less than 3 percent, a fiscal deficit that is less these fluctuations is presented in table 10.3 for the than 1 percent of GDP, and with ample foreign period 1994­97, permitting an examination of the exchange reserves). This has been due mainly to contribution of price changes from changes in movements in capital accounts and also to the border prices, real exchange rates, and policy inherent instability associated with emerging interventions. What is notable from the data is 202 Agriculture and the WTO TABLE 10.4 Decomposition of Producer Price Changes: Argentina Cumulative Percentage Change in Real Real Real Domestic Border Exchange (1 Ti) Product Period Prices Prices Rate (1+Ti) Others Others Exportables Beef 1960­69 43.55 30.97 244.19 231.61 1970­79 22.29 43.58 67.67 46.38 1980­84 33.97 104.29 87.20 16.88 1985­89 32.68 13.16 39.36 19.62 0.22 19.84 1990­93 91.35 13.28 132.01 27.75 0.38 27.38 1994­95 2.45 3.31 2.02 7.78 Maize 1960­69 2.57 21.54 14.42 38.52 1970­79 102.78 4.04 83.34 23.49 1980­84 54.93 33.69 89.39 0.77 1985­89 38.40 38.74 14.36 4.65 3.52 9.23 1990­93 68.28 38.88 61.98 28.61 3.97 32.58 1994­95 25.97 13.08 2.02 14.92 Sorghum 1960­69 4.51 24.07 14.42 33.98 1970­79 38.49 4.86 58.54 15.19 1980­84 4.69 23.41 64.60 45.88 1985­89 15.72 1.86 35.62 2.65 15.39 18.04 1990­92 89.00 47.73 54.70 25.28 11.85 13.43 Soybeans 1960­69 -- -- -- -- -- -- 1977­79 112.07 51.20 55.40 5.47 1980­84 7.60 60.75 80.42 12.07 1985­89 41.02 10.87 48.30 ­3.52 0.07 3.59 1990­93 58.64 39.38 61.98 38.67 4.04 42.72 1994­95 41.36 52.61 2.33 13.58 Sunflowers 1960­69 -- -- -- -- -- -- 1977­79 89.08 24.34 55.40 9.33 1980­84 22.02 59.88 80.42 1.48 1985­89 56.64 17.50 35.62 2.01 1.51 3.52 1990­92 67.65 46.14 54.70 32.93 0.27 33.19 Wheat 1960­69 3.77 18.09 14.42 19.51 1970­79 87.73 6.85 92.60 33.80 1980­84 64.03 62.15 9.18 11.06 1985­88 3.72 19.11 12.32 20.18 7.53 27.71 1991­93 24.76 35.94 26.64 31.64 1.42 37.82 1994 2.61 5.93 1.44 4.77 Source: Valdés 1996. the imperfect correlation (sometimes negative) The evidence for various agricultural commodi- between border price changes and domestic pro- ties in Argentina, Chile, and Colombia (tables 10.4, ducer prices. For example, in the case of wheat in 10.5, and 10.6, respectively) shows a similar pattern Bulgaria during the period 1994­97, domestic with respect to the alternating relative weight of prices increased 25 percent while border prices fell border prices movements to the final determina- 6 percent, the increase owing primarily to a 34 tion of domestic real producer prices. In short, percent increase in the real exchange rate. price transmission per se is not the whole story. Managing Potential Adverse Impacts of Agricultural Trade Liberalization 203 TABLE 10.5 Decomposition of Producer Price Changes: Chile Cumulative Percentage Change in Real Real Real Domestic Border Exchange Product Period Prices Prices Rate (1+Ti) Others Exportables Apples, red 1960­70 133.77 41.02 30.94 17.14 44.67 1971­74 2.40 21.22 44.77 10.38 15.58 1975­83 6.40 8.40 46.47 6.77 24.90 1984­89 30.48 4.10 36.11 0.00 1.53 1990­93 429.10 64.88 14.79 0.00 349.43 Grapes, 1960­70 99.29 21.53 30.94 17.14 29.67 Thompson 1971­74 54.35 4.37 44.77 0.59 5.80 1975­83 41.51 20.91 46.47 16.55 9.31 1984­89 4.70 30.47 33.67 0.00 7.91 1990­94 22.24 21.37 16.80 0.00 17.67 Importables Beef 1960­70 41.20 9.57 30.94 0.00 0.69 1971­74 87.27 42.14 44.77 0.00 0.36 1975­83 51.72 76.06 46.47 0.00 22.12 1984­89 136.69 64.93 33.67 13.98 24.12 1990­95 30.05 16.88 26.84 3.54 16.54 Maize 1960­70 -- -- -- -- -- 1971­74 -- -- -- -- -- 1975­83 -- -- -- -- -- 1984­89 4.24 45.12 25.27 7.84 23.46 1990­93 22.18 23.84 16.25 3.54 21.44 Milk 1960­70 3.29 5.37 30.94 0.00 22.28 1971­74 77.44 14.27 44.77 0.00 18.40 1975­83 0.18 2.39 46.47 0.00 44.26 1984­89 6.25 9.46 33.67 13.98 63.35 1990­93 16.51 19.66 16.25 3.54 22.94 Sugarbeets 1960­70 -- -- -- -- -- 1971­74 -- -- -- -- -- 1975­83 -- -- -- -- -- 1984­89 1.35 39.16 25.27 7.84 57.93 1990­93 7.57 18.72 16.25 3.54 6.50 Wheat 1960­70 6.33 13.15 26.76 0.00 7.28 1971­74 53.81 80.13 9.89 0.00 36.21 1975­83 35.51 88.39 81.35 0.00 42.55 1984­89 0.42 32.83 37.84 7.84 2.41 1990­95 27.80 12.84 26.84 7.84 4.04 Source: Valdés 1996. 204 Agriculture and the WTO TABLE 10.6 Decomposition of Producer Price Changes: Colombia Cumulative Percentage Change in Real Real Real Domestic Border Exchange Product Period Prices Prices Rate (1+Ti) Others Exportables Beef 1960­69 -- -- -- -- -- 1970­79 -- -- -- -- -- 1980­89 6.45 40.26 63.16 29.22 0.13 1990­92 27.68 35.73 4.94 75.43 16.96 1993­95 24.77 15.87 32.89 -- 7.74 Coffee 1960­69 6.12 27.89 20.90 6.76 7.63 1970­79 47.24 78.81 21.43 34.88 24.75 1980­89 33.07 108.98 65.31 61.82 51.22 1990­92 47.12 61.91 4.94 28.45 18.60 1993­95 28.07 74.80 32.89 -- 13.83 Cotton 1960­69 7.30 20.82 20.90 0.13 7.52 1970­79 13.46 40.04 21.43 0.09 5.06 1980­89 4.43 52.62 65.31 0.11 8.37 1990­92 25.87 38.25 4.94 13.02 5.58 Source: Valdés 1996. Policy Implications for Managing · While there might be a concern about world Price Risk in the Context of WTO price instability per se and while it is likely more Commitments of a concern in developing countries, the real possibility of extended periods of "low prices" is In summarizing the concerns in developing coun- the biggest political difficulty in these countries. tries with respect to managing price risk, there are And the persistence problem makes more diffi- five points to keep in mind: cult any attempt to resolve that low price prob- · Perhaps the central political concern in develop- lem without resorting to border measures. ing countries is the question of whether or not · What can be done to deal with the low price prob- prices levels are very distorted--and excessively lem for developing countries? Special agricultural low--owing to OECD subsidies. This preoccu- safeguards (SSGs) were conceived as an instru- pation is intensified in the context of falling ment, but most developing countries in the prices in real terms. present WTO environment cannot use them, · From a perspective of developing and transition because they undertook tariffication before the countries, as the result of their trade liberalization Uruguay Round (see Konandreas 2000). (the reduction in price distortions and the · What is emphasized is direct price transmission, removal of QRs) and thus enhanced price trans- but there are other effects of trade liberalization mission,world price distortions and the possibility on the welfare of the poor, which go uncaptured of facing periods of low world prices are now more by simply looking at price effects. It should be rec- important in determining domestic producer ognized that there are also questions surrounding price risk. The reduction in protection and the what constitutes the important definition of change in the system of protection brought about "transmission." Normally price transmission by decisions in developing countries have aug- refers to the transmission from border prices to mented the importance of the possibly distorted domestic prices of tradables. But what is also levels and the stochastic nature of world prices. important from the perspective of poorer farm- Managing Potential Adverse Impacts of Agricultural Trade Liberalization 205 ers is the second-order transmission of world tional reforms. Obviously, those who believe in market changes through labor markets, capital the benefits of freer world trade have an interest in markets, and the influence of missing or poorly facilitating policy adjustments in developing functioning markets for staples (Taylor 2002). economies. But these adjustments must be designed Imperfectly functioning capital markets and high in such a way so as to overcome the potential politi- transactions costs--including those associated cal resistance likely to result from the exposure of with access to markets for new products--tend large agricultural sectors, in which a large share of to isolate poorer farmers and rural areas. Rural the poor are concentrated, to the risks of sustained households that would lose under trade liberal- price decreases. ization could conceivably move their resources-- Perhaps the first question an economist would primarily labor--to other economic sectors in ask is, "What about making use of market-oriented which transactions costs are lower and access to policies?" If problems associated with price risk capital is less expensive. Many do, but for some were merely related to year-to-year resource alloca- the net benefits of migration are low, especially tions, futures and other derivative markets would be for older persons and those with human capital the easiest and likely the most efficient solution to specific to farming and lower levels of education overcoming the risk-related political costs of further that might otherwise provide skills transferable trade reform that arises in countries that are risk to other sectors. averse and have underdeveloped capital markets, and other possible institutional problems. If private That world prices have been affected by govern- hedging was not feasible, then a government policy ment policies in both developed and developing based on a futures market would serve as a solu- countries appears to be a historical fact, although the tion. An expected low price would simply signal a degree to which such policies have distorted prices is decrease in resource use devoted to the commodity. subject to debate. Perhaps these distortions have And if a realized low price in the previous year led changed with the recent overall movements toward to an expectation that low prices would eventually liberalization, but certainly not the direction of their increase, resources would merely wait to reenter negative effect on world prices. Liberalization in production of the commodity until such a developing and transition economies has certainly time that price realizations signal their expected enhanced the transmission of world prices to their profitable use. economies, and thus the importance of the distor- Resource decisions, however, are usually matters tions that continue to exist. of multiyear commitments, and it can be concluded Particularly in the context of ongoing WTO from the above discussion of the stochastic nature negotiations, what can be said briefly about the pol- of world prices that there is a high probability that icy implications for managing price risk that one low-price events come grouped together in distinct can draw from this environment of price distortion episodes of series, or clusters, of months, if and price transmission? In this period of major pol- not years. This implies that futures and options icy reforms toward more open economies, there are markets would be inadequate to insure completely valid political and economic arguments for govern- against the unfavorable effects of exposing import- ments and the farming community in developing competing farm sectors to world price declines. countries to find effective interventions to deal with Certainly politicians have insisted on providing enhanced price transmission, especially in the con- some stabilization during the course of transition text of the persistence of low world prices. The toward integration into the world price regime. movement toward trade liberalization, toward Although in the case of New Zealand some 15 years allowing market price signals to determine the use ago, there was little government support to soften of resources, could be frustrated by the reluctance the transition from government support of farming of governments to expose further their farming con- to life without such support (see box 10.1). Other stituents, especially those in the import-competing strategies are called for to smooth income fluctua- sectors, to the risks of price instability and periods tions across years, such as the use of credit or equity of persistent low prices that would result from the markets, long-term contracts, vertical integration, enhanced price transmission associated with addi- and other means. Nevertheless, in developing 206 Agriculture and the WTO BOX 10.1 Farming Without Subsidies: The Experience of New Zealand New Zealand has a vibrant, diversified, and sus- caused long-term damage to agriculture by cor- tainable rural economy. Government handouts rupting market signals, reducing innovation, were discontinued some 15 years ago. Left to and misusing resources with their damaging face the market, family farmers have succeeded effects on the environment. by their own efforts. Their experience provides a Since 1986­87 the value of economic activity message of reassurance to farmers, policymak- in New Zealand's farm sector has grown by over ers, and others facing change. The New Zealand 40 percent in constant dollar terms, and its con- experience has debunked the myth that the tribution to the economy has risen from 14.2 farming sector cannot prosper without govern- percent of GDP in 1986­87 to 16 percent in ment subsidies. 1990­2000. The removal of farm subsidies has With agriculture based largely on pastoral proven to be a catalyst for productivity gains farming, New Zealand had just over 80,000 from 1 percent in 1986 to the current average farm holdings. Sheep and/or beef farms and annual 5.9 percent. dairy farms account for 58 percent and 29 per- The change was not achieved without signifi- cent of the total number of farms, respectively. cant effort. Early predictions of large numbers of Horticulture, forestry, cropping, and rural farmers leaving the land did not occur. Only tourism contributed to its agricultural sector, about 1 percent of the total number of farms which employed 11.4 percent of the work force. faced bankruptcy and were forced to sell. Land New Zealand exported about 80 percent of its prices, kept artificially high by subsidies, plum- farming outputs, and agricultural exports-- meted with their removal. Marginal land especially sheep meat and dairy products-- reverted to bush, and subsidy-driven land man- accounted for over 50 percent of total merchan- agement problems ended. Today, farmland val- dise exports. ues have more than recovered and farm prof- New Zealand has the lowest level of agricul- itability has been restored. Farmers reduced tural support for industrialized countries, which costs and focused on producing higher-value was around 2 percent of the value of output in products. Many restructured debts and contin- 1999 compared with an OECD average of ued farming, adjusting farm practices to reduce 40 percent. Most of this support relates to agri- input costs. With investment decisions now sub- cultural research funding. However, this has not ject to commercial and good farming disci- always been the case. In the mid-1980s the situ- plines, agricultural input suppliers were forced to ation was quite different. In 1984 nearly 40 per- become more competitive, thereby contributing cent of the average New Zealand sheep and to the competitiveness of the agricultural sector. beef farmer's gross income came from govern- The government's financial assistance to ment subsidies. With the New Zealand economy farmers to make the transition to an unsup- almost on the brink of bankruptcy and faced ported market was minimal: a one-off "exit" with deteriorating external markets, inflation, grant--the equivalent to about two-thirds of historically high interest rates, and agricultural their previous annual income, some access to support at 33 percent of output value in 1983, social welfare income, and limited financial the government suddenly removed all subsidies advice. There was no substantive effort to soften and took other market reform measures. In the the effect of change. New Zealand farmers have November 1984 budget, the government abol- proved far more resilient and adaptive than ished almost 30 different production subsidies, expected when subsidies were first removed. including fertilizers, taxation schemes, produc- They have proved to the rest of the world that, tion, and livestock support. These subsidies had for family farming, there is a life after subsidies. Source: Adapted from Life After Subsidies: The New Zealand Farming Experience--15 Years Later, Federated Farmers of New Zealand (Inc), (revised March 2001). countries the resources for implementing these effects of what seems to be the most disturbing market-based strategies are likely missing, or only characteristic of world commodity prices: their slowly becoming available. Therefore, to reduce the periodic tendency to persist at values below trend. resistance to reform, it would be worthwhile to Although there is a variety of possible instruments consider the development of other government currently in use, there is the question of both their price stabilization plans that would reduce the effectiveness and legality under the current WTO legal Managing Potential Adverse Impacts of Agricultural Trade Liberalization 207 framework. In a relevant technical note, Konandreas action under the de minimis provisions and the def- (2000) of the FAO discusses the current framework, initions of government support contained in the classifying permissible policies into two broad cate- green box, and are thus restricted in their use of gories: border measures through tariffs (within the nonborder support policies. In contrast, developed tariff ceiling bounds) and domestic support measures countries (and a few developing countries) (price and nonprice programs within the limits of reported high AMS. WTO commitments). Therefore, here is the situation: Developing With respect to the current state of border meas- countries have fewer fiscal resources to manage ures, most developing countries have bound their price risk and aid their farmers through domestic agricultural tariffs at a relatively high rate--100 per- supports, and have fewer alternative market instru- cent and sometimes more. Several countries, such as ments to compensate for the higher probability of the Philippines, Chile, Thailand, Argentina, and periods of low domestic prices that might result Malaysia, have bound their tariffs around 30 per- from further moves toward trade liberalization. cent. In many countries, the actual rates or applied This leaves many governments in developing coun- tariffs are lower than the bound tariff (so-called tries with the temptation to seek protection of their dirty tariffication). In Konandreas's assessment, a import-competing sectors through border meas- high bound tariff might be insufficient to avoid the ures. From an economist's perspective of the welfare politically difficult situation of very low prices. But gains from trade, and from a practitioner's perspec- more significantly, it is likely that the lack of other tive of facilitating the liberalization process, current instruments under the WTO has encouraged gov- WTO negotiations might well consider providing ernments to cynically set overly high bound tariffs. greater flexibility to developing economies with This permits considerable discretion in the selec- respect to well-defined and disciplined tariffs and tion of tariff levels, and thus a greater degree of surcharges. Such flexibility might come through uncertainty with regard to the effective trade permitting the application of price bands or price regime at any point in time. floors, the modification of the general WTO safe- In practical political terms, currently very few guards, and--perhaps the most promising--the countries have access to the SSG provision, an alteration or adaptation of the special safeguard exemption to tariff ceiling levels. Eligibility to this clause for countries with low bound tariffs. safeguard process is limited because the majority of developing countries did not "tariffy." Instead, Price Bands they offered ceiling bindings, or made pre-WTO commitments. An instrument that has been used in several With respect to domestic support policies, there countries in Latin America is a system of price are both bound supports under nonexempt policies bands. Price bands function in a similar manner to subject to the commitments of aggregate expen- the use of a simple price floor, although important diture ceilings (the aggregate measurement of points of difference exist. Price bands stabilize support--AMS--established during the Uruguay prices between a floor and a ceiling--both of which Round), and nonbound supports, exempt from typically are based on moving averages of past limits, operating within the green box. In addition, world prices, not a domestic target price. For exam- developing countries have access to a special cate- ple, in the case of Chile, the moving average gory of exempted supports under the "Special and for wheat is based on 60 months, and for sugar Differential Treatment" provision of the Uruguay 120 months.10 Such a scheme implies a moving Round Agreement on Agriculture--investment floor price level that would trigger government subsidies, input subsidies to low-income farmers, intervention. In the past the analytical justification and so on. Most developing countries (61 of 71) of price bands has been on stabilization per se rather reported zero AMS levels--and in the remaining than on the avoidance of extremely low price events. 10, these levels are very small. In part this was due In fact, the automatic, reflexive nature of determin- to fiscal limitations (Konandreas). The implication ing the price bands can lead to levels of floor prices is that most developing countries are limited in that would be difficult to justify in terms of protect- their support options to deal with price risk to ing against the risk of unsustainable low prices. For 208 Agriculture and the WTO FIGURE 10.2 Price Band of Sugar in Chile FIGURE 10.3 Price Band of Edible Oil in Chile 900 500 800 450 700 400 600 350 FOB/ton. 500 FOB/ton. 300 US$ 400 US$250 200 300 150 200 1986.01 1986.12 1987.11 1988.10 1989.09 1990.08 1991.07 1992.06 1993.05 1994.04 1995.03 1996.02 1997.01 1997.12 1986.01 1986.12 1987.11 1988.10 1989.09 1990.08 1991.07 1992.06 1993.05 1994.04 1995.03 1996.02 1997.01 1997.12 Ceiling Floor Reference Price Ceiling Floor Reference Price example, in the case of Chilean sugar, there have been greater than what would otherwise have been the periods in which the floor price in dollars has case. Productivity certainly has grown in wheat and exceeded the price ceilings in other periods (see fig- sugar, but not notably at a faster rate than in non- ure 10.2). Figures 10.2 and 10.3 show the evolution of protected commodities such as maize. the price floor and ceiling for sugar and edible oil in Chile. It is difficult to rationalize the narrowing over Price Floors time of the price bands in these cases, except by refer- ence to the manner by which the dynamics of In place of price bands as traditionally applied, one international price formation affect the mechanical policy option that might be worthy of considera- determination of the floor and ceiling. tion is a simplified fixed price floor scheme for the The price floors associated with price bands as elimination of the worst-case scenarios associated they have been typically applied are simply moving with several concurrent years of especially low averages, reducing near-term volatility but not world prices. For specific "sensitive" commodities, offering an assured minimum level as a fixed point a minimum world price or threshold level might of reference. Indeed, the adjusting price floor in a be defined, below which a government would price band scheme is in many cases merely a means commit itself to intervention in order to maintain of deferring, or delaying price changes that the domestic price received by producers.11 In inevitably must occur. For example, the 1998 floor terms of international acceptability, this threshold price for Chilean wheat was US$150 per ton, imply- price should be based on the minimum-average ing a minimum cost, insurance, and freight price of cost of the least-cost international exporter, and US$175 per ton, well above the long-term mini- therefore be unmistakably transitory, in the sense mum average production cost of the commodity that prices in the future would almost certainly in neighboring Argentina. Certainly there is noth- rise. For example, when sugar prices reached ing in price band schemes that involves moving- US$250 per ton in 1985, sugar production was averages price that guarantees some long-term unprofitable even in Australia, one of the most worst-case scenario for producers' incomes and cost-efficient producers in the world; one could ability to service debt. From recent experience with have concluded that world prices were unmistak- a relatively high price, supported in the short term ably low and resources would certainly move from by the price floor, it is difficult to assess whether or sugar production, and price would rise in the not the policy has caused investment in the wheat future. In fact, two years later, prices increased to sector (or generated the long-term financing) approximately US$400 per ton. Managing Potential Adverse Impacts of Agricultural Trade Liberalization 209 It is worth emphasizing that the objective of WTO.12 Unlike a variable levy, there would be no the type of price-floor policy discussed here is domestic target price. It appears clear that a basic more narrowly defined than that of price stabi- principle or criterion for establishing a feasible lization as pursued by most current stabilization floor price policy should be the limited nature of schemes. A price floor that is based on some fixed, the interventions. The record of most ad hoc price internationally recognized, most-efficient average stabilization programs is one of a proliferation of cost of production would be meant to provide a distortions and indiscriminate protections that safety net against the lowest prices, as a form of are difficult to justify.13 A second criterion would price insurance. While this would likely reduce the also be important: a product selected for special variance of domestic price, the objective would treatment with a price floor should be a standard- not be to insulate domestic prices in any general ized commodity of the type that has a near-perfect sense. Indeed, a simple floor policy would allow substitute on international markets.14 The appli- the flow of information between international and cation of these two criteria would guard against domestic markets through price signals in all peri- both the protection of goods that are not econom- ods except those with extremely low and certainly ically profitable in the long run and the tendency transitory low prices. A policy based on the free of governments to complicate and manipulate for transfer of price signals except at the lowest of short-term objectives programs that ought to be prices would leave unhindered the development of transparent and based on long-term economic risk markets, the use of futures, and other activi- feasibility. ties that "price stabilization" programs have It must be recognized, however, that this kind tended to impede. of scheme would not be free of practical difficul- While the principal advantage of a price floor ties. One difficult issue would be the determina- would be the protection of producers in long-run tion of the internationally lowest production cost. viable sectors that might not otherwise survive This was tried before in Australia, with the result extremely low prices in the short run, there are that the process became tainted by political influ- some other advantages worth mentioning. From ence. Some mechanism would need to be used to an economic perspective, in the present environ- guard against a similar outcome, and it is not clear ment of immature markets in which agriculture how this could feasibly be done. This concern finds itself in many developing countries, the would need to be considered in the calculus of the elimination of "worst-price scenarios" could pro- decision on what option is most likely to lower vide incentives to investment in promising sec- political resistance to dismantling high tariff bar- tors. While eliminating the worst-price risk, which riers without putting in their place something a developing-country economy is not yet able to equally as protectionist, or worse. diversify away by other means, a price floor based solely on the most efficient world producer would not distort long-run returns to investment and General WTO Safeguards would not protect long-run uncompetitive sec- and Contingency Measures tors. Moreover, in terms of credit and the ability of When there is either an abrupt and large surge of lenders to securitize producers' cash flows, vari- imports, or a sudden decline in import prices that ance-stabilizing policies, such as moving-average threaten a country's import-competing sector, price bands, do not have the advantage of an easily WTO safeguards or contingency measures are referenced and guaranteed forecast of a minimum applicable. These relatively well-known measures worst-case scenario that would result from a fixed permit the suspension of WTO obligations, at least price floor. temporarily and in some cases permanently. In terms of practical implementation, for General safeguard measures can be classified in the import-competing commodities, a price floor following manner (see Konandreas; Hoekman and policy could be implemented through a system of Kostecki 1996): variable surcharges that are only triggered by a predetermined, international reference price · Antidumping measures that counteract the along the lines of the safeguard mechanisms of the effects of private firms using price discrimination 210 Agriculture and the WTO that lowers export prices below home market from the Cairns Group. Nevertheless, it is possible prices. The legal test is that of material injury to to adapt or modify the special safeguard clause to a domestic industry. The most notorious use of allow effective border measures in the manage- antidumping measures has been by the U.S.; ment of the low-price risk relevant to developing- however, such measures have been used by some country import-competing sectors. developing countries, although with only partial The clearly stated impetus for modifying the success. Paz Caferata and Valdés (2000) observe special safeguard scheme would have to be to per- that in Latin American and Caribbean coun- mit countries to apply tariffs beyond their bound tries, of the 469 antidumping investigations ini- ceilings in the event that otherwise-competitive tiated during the period 1987­97, 60 were domestic producers face severe injury during peri- related to agriculture--but only 199 (of 469) ods of extremely low prices. This justification for investigations were adopted, and among those, any modification would be moot for commodities only three (of 60) applied to agricultural prod- for which bound tariffs are already high. But as ucts. While overall, 42 percent of the investiga- Konandreas notes, "There is no apparent relation- tions led to the adoption of antidumping meas- ship between access to the existing SSG and low ures, only 5 percent of the requests in bound tariffs. Eligibility to the existing SSG clause agriculture were adopted at the country level.15 has very little relationship to the problem that it presumably was meant to address" (p. 12). · Countervailing duties, which are also based on In practical terms, how might the special safe- material injury to a domestic industry, but apply guard clause be adjusted so that it would be ori- to the subsidization by the exporting country's ented toward protecting farmers from the risk of government. low world prices in the absence of a sufficiently · Emergency safeguards, which are explicitly tem- flexible (and WTO-legitimate) standard tariff sys- porary and limited in frequency of use. These tem, or where domestic supports are inadequate? are immediately applicable without formalities The first modification is to allow special safeguards in the event of imports threatening serious only for commodities and countries for which injury to domestic industries. The use of this bound tariffs are below some threshold (Konan- measure does not presume any "unfair" behav- dreas suggests 70 percent, although the authors ior on the part of exporting firms or govern- would suggest a limit of 35 percent). As a variation ments, but is restrictive both in time and repeat- on this theme, the upper limit to variable safe- able application.16 guards could be inversely related to the bound tar- · Measures related to balance of payments to safe- iff. This would induce the governments to guard a government's external financial position. set lower bound tariffs, because they would have · "General waivers," which enable countries to ask access to additional border protections triggered in for these. This requires formal authorization by emergencies. the WTO Council. But how would one define such emergencies? · Modifications of schedules and renegotiations The triggering of variable safeguards would have to of tariffs. be specified in terms of well-defined low price events, universally applicable to all countries, The Special Safeguards Provision (SSG) developed and developing. This would necessarily Special safeguards are applicable to agricultural involve a negotiation process within a WTO- producers under the Agreement on Agriculture, established framework, which would have to deter- restricted to products that were included in the mine coverage of importables that qualify for vari- Uruguay Round tariffication process. Many devel- able special safeguards and would have to establish oping countries simply are not eligible to use the a process of registration. SSG provision because they set bound tariffs out- A third possible modification of the special safe- side of the tariffication mechanism. Of course, the guards clause would be to make access to it contin- future status of special safeguards is questionable, gent on domestic support transfers. Countries with as they are especially vulnerable to the criticisms the ability to use other safety net mechanisms, Managing Potential Adverse Impacts of Agricultural Trade Liberalization 211 such as farm revenue insurance along the lines as practical wisdom derived from long experience. This is a of Canada's Net Income Stabilization Account, to political-economy argument for uniformity, not a neoclassical one" (Harberger 2001, p. 549). protect producers during periods of low prices 3. Another aspect, which we do not consider here, is the would be excluded. Konandreas suggests, for exam- effect of trade liberalization and price transmission on agricul- ple, that special safeguards be limited to countries tural value added, and not simply on prices. The case of Brazil after the 1991 reforms is very interesting: border protection on with domestic support below 15­20 percent of the tradable inputs fell rapidly and significantly, benefiting the pro- value of domestic production. ducers of exportables (whose export taxes were eliminated) and There is a technical discussion ongoing in Geneva partially compensating producers of importables who con- fronted tariff reductions on their products. The data suggest that and in the United Nations system on these issues. the reduction in price of purchased inputs and machinery The specificity of these measures, their legality, their tended to favor mid-size and larger farmers. operational implementation, and their relative effec- 4. Of course, productivity and factor use are endogenous tiveness are questions that would be major research and not independent of policy, but this issue is beyond the scope of the present discussion. efforts in and of themselves, and which are certainly 5. Recently in Chile farm lobbies proposed, with consider- worthy of the professional attention of economists able political support from legislators, a surcharge on imported and others committed to improving the perform- milk product prices that would compensate one-for-one the EU export subsidy on milk. If adopted, this proposal would lead to ance of the international trade system. internal Chilean prices approximately equal to the internal European price. This would grossly overestimate the true distor- tion in world prices. 6. See, for example, Zietz and Valdés (1989) for the OECD Notes (in Goldin and Knudsen, eds.), and Tyers and Anderson (1992) for work on developed and developing countries. 1. In this paper, we put emphasis on price risk and develop- 7. A number of methodological questions can be raised ing-country farmers in terms of further policy reforms. regarding the particular estimates of the effects of trade liberal- Nevertheless, we recognize that the topic of price risk is relevant ization derived from these models. There are three notable to consumers as well, particularly low-income consumers. With themes: trade reforms and world prices determining to a greater extent domestic prices, consumers could benefit from lower food prices, although these might be more volatile. On the negative · There is a complex interaction between domestic distor- side, however, there is a perceived problem of increased import tions and the transmission of these distortions into dependence. This is in essence a food supply management issue, domestic markets. What is assumed in estimation models and often examined in the context of food security. about price transmission is an important issue in and of 2. With policies restricted to tariffs, an important choice is itself (Coleman 1995; Taylor 2002). For example, prices in between differentiated and uniform tariffs. While most trade remote areas tend to exhibit independence from prices in economists favor low, uniform tariffs, one can always find tech- major trading centers. Notwithstanding, one can draw on nical economic arguments for applying higher or lower tariffs in various parameters to estimate export demand and sup- some products. But the process is almost always corrupted, a ply in order to predict the overall sector's response--even captive of special interests. Harberger concludes his 1984 edited if there are various subsets of producers that are not well volume, World Economic Growth, with several policy lessons, one integrated in the markets for tradables. This question is of which is entitled, "Some types and patterns of trade restric- one of distributional welfare implications, and we return tions are far worse than others" (p. 431). The author notes that briefly to this issue below. "[t]he only sure way to guarantee against catastrophic variations · How realistic are the assumptions about transmission? in rates of effective protection--even with moderate-looking Some earlier work (for example, Valdés and Zietz 1980) rates of nominal protection on final products--is to make the assumed perfect transmission, but more recent studies rate of nominal protection uniform across all products. . . . (such as Tyers and Anderson 1992) have allowed for the For only when all nominal rates of protection are equal are all possibility of lower transmission elasticities. This is an effective rates equal to this same nominal rate. Only a given uni- empirical question grounded in the nature of the data form rate of tariff can automatically avoid capricious and dis- available and the econometric subtleties required to give torting variations in the effective rates of protection actually flexibility to the analysis. achieved. Modification of tariff schedules in the direction of · What about the interdependence across products, both greater equality is thus one of the most important reforms advo- horizontally and vertically, extending upward through cated by professionals." And again in Frontiers of Development processed products? Processed agricultural products (for Economics, edited by Meier and Stiglitz (2001), Harberger states: instance, orange juice concentrate, tomato paste, "Yet, I have not the slightest doubt that, asked to choose between processed frozen and canned products, and wine) are Ramsey tariffs and uniform tariffs, or between a Ramsey-style rarely considered. It is a complex analytical problem, differentiated value-added tax and a broadly based, uniform involving a proliferation of parameters that might ade- one, my practicing professionals and Williamson's consensus quately account for cross-product effects. Some authors members would vote overwhelmingly in favor of the uniform- consider a fixed-proportion world, others are inclined to rate alternatives. In doing so they would be expressing not the an Armington-style approach. In our opinion this has implications of neoclassical theory but rather what they think of not been solved in a satisfactorily rigorous way. 212 Agriculture and the WTO 8. RUNS is an acronym for Rural-Urban North-South. This Cafferata, P., and A. Valdés. 2000. "Interests and Options in the model was developed as a joint OECD and World Bank project, WTO 2000 Negotiations: Latin America and the Caribbean." and includes 22 regions and 20 sectors, 15 of which are agricul- In M. D. Ingco and L. A. Winters, eds., Agricultural Trade tural, covering the years 1985­2002. Liberalization in a New Trade Round: Perspectives of 9. For further analysis of a more recent nature, we refer the Developing Countries and Transition Economies. Washington, reader to the OECD (2001) report on export subsidies, "A D.C.: World Bank. Forward-Looking Analysis of Export Subsidies in Agriculture," Cafferata Paz, J., and A. Valdés. 2001. "Agricultural Trade and the ERS/USDA study edited by Burfisher (2001). Liberalization in a New Trade Round." In M. D. Ingco 10. Also for the application of the bands, 25 percent of the and L. A. Winters, eds., World Bank Discussion Papers highest and lowest prices are eliminated for calculation of the No 418, World Bank, Washington, D.C. moving average. Cashin, P., H. Liang, and J. McDermott. 1999. "How Persistent 11. A government's commitment to intervene at very low Are Shocks to World Commodity Prices?" IMF Working prices should likely be reserved for one or only a few commodi- Papers, IMF, Washington, D.C. ties, the long-term international competitiveness of which is Coleman, D. 1995."Problems of Measuring Price Distortion and highly probable. This basic criterion of long-term comparative Price Transmission: A Framework for Analysis." Oxford advantage of selected commodities would preclude encouraging Agrarian Studies 23(1): 3­13. commodity production for the purposes of self-sufficiency or Deaton, A. 1992. "Commodity Prices, Stabilization, and Growth other goals inconsistent with developing patterns of free trade in Africa." Research Program in Development Studies, based on underlying differences in costs of production. For Discussion Paper 166. Princeton University. example, supporting rice production in Chile would not be con- Gersovitz, M., and C. Paxson. 1990."The Economics of Africa and sistent with this criterion, nor would promoting the production the Prices of Their Export." Studies in International Finance, in tropical climates of commodities associated with temperate 68, Princeton University Department of Economics, zones, as has been the case with crops such as wheat in Colombia Princeton, New Jersey. or sorghum in Venezuela. Goldin, I., and O. Knudsen. 1989. Agricultural Trade 12. For exported commodities, a fund built of producer con- Liberalisation: Implications for Developing Countries. Paris tributions during periods in which the floor policy is dormant and Washington, D.C.: Organisation for Economic could finance producers in periods of extremely low prices. In Development and Co-operation and the World Bank. the case of exportables, the use of put options could comple- Goldin, I., O. Knudsen, and D. van der Mensbrugghe. 1993. Trade ment a price floor policy and be useful in the initial years of Liberalisation: Global Economic Implications. Paris and operation before fund reserves could cover a string of low-price Washington, D.C.: Organisation for Economic Development events. and Co-operation and the World Bank. 13. Not only does the floating nature of price bands under- Harwood J., R. Heifner, K. Cable, J. Perry, and A. Somwaru. 1999. mine the use of any floor for guaranteeing medium-term agri- "Managing Risks in Farming: Concepts, Research, and cultural loans (and tend to preclude alternative market develop- Analysis." Agricultural Economics Report 774, U.S.D.A. ments that could reduce the consequences of price risk), as a Economic Research Service, Washington, D.C. broad rule bands have been applied indiscriminately to various Harberger, A. C., ed. 1984. "World Economic Growth: Case commodities and products. This has been notable in the cases of Studies of Developed and Developing Nations." Institute of Colombia, Republica Bolivariana de Venezuela, and Ecuador, Contemporary Studies, San Francisco. where more than 120 importable goods are covered by the pol- ------. 2001. "The View from the Trenches: Development icy. The administration of such a system of price bands has been Processes and Policies as Seen by a Working Professional." In cumbersome at best and has tended to make the structure of G. M. Meier and J. E. Stiglitz, eds., Frontiers of Development incentives and effective protection almost unpredictable Economics: The Future in Perspective. New York: Oxford (Quiroz 1996, 1997). University Press. 14. This criterion precludes stabilization programs for Hoekman, B., and M. Kostecki. 1996. "The Political Economy of agroindustrial products derived from basic commodities, which the World Trading System: From GATT to WTO." Oxford in effect become indistinguishable from crude protectionism, as and New York: Oxford University Press. has been the experience during the 1990s with price bands for Ingco, M. D. 1996."Progress in Agricultural Trade Liberalization "derivatives" in Ecuador, Colombia, and Venezuela. and Welfare of Least-Developed Countries." International 15. Argentina, Brazil, and Mexico initiated 87 percent of Trade Division, World Bank, Washington, D.C. Latin America and the Caribbean region's antidumping investi- Josling, T. 1980. "Developed Country Agricultural Policies and gations during 1987­97, and 88 percent of the investigations Developing Country Food Supplies: The Case of Wheat." related specifically to agriculture (Paz Caferata and Valdés 2000). (Research Report). International Food Policy Research 16. Notification to the WTO is required. Institute (IFPRI), Washington, D.C. Konandreas, P. 2000. "The Need for Effective Safeguards against Select Bibliography Very Low World Market Prices in the Context of the WTO Negotiations on Agriculture." United Nations Food and Burfisher, M. E., ed. 2001. "Agricultural Policy Reform in the Agriculture Organization (FAO) Informal Expert WTO--The Road Ahead." Market and Trade Economics Consultation, December 7­8, FAO, Rome. Division, ERS, USDA, Agricultural Economic Report no. León, J., and R. Soto. 1995. "Structural Breaks and Long-Run 802, Washington, D.C. Trends in Commodity Prices." Policy Research Working Caner, M., and B. Hansen. 1997."Threshold Autoregression with Paper 1406, World Bank, Washington, D.C. a Unit Root." Boston College Working Papers in Economics, McCulloch, N., L. A. Winters, and X. Cirera. 2002. Trade no. 381, Boston, Mass. Liberalization and Poverty: A Handbook. London: Centre for Economic Policy Research. Managing Potential Adverse Impacts of Agricultural Trade Liberalization 213 Mundlak, Y., and D. Larson. 1992. "On the Transmission of Tsetsekos, G., and P. Varangis. 1999. "The Structure of World Agricultural Prices." World Bank Economic Review, Derivatives Exchanges: Lessons from Developed and 6(3): 399­422. Emerging Markets." Paper prepared for the Commodity Risk OECD (Organisation for Economic Co-operation and Management Group, World Bank, Washington, D.C. Development). 2001. "A Forward-Looking Analysis of Tyers, R., and K. Anderson. 1992. Disarray in World Food Export Subsidies in Agriculture." Paris, OECD. oecd.org. Markets, A Quantitative Assessment. Cambridge: Cambridge Perron, P. 1989. "The Great Crash, the Oil Price Shock and the University Press. Unit Root Hypothesis." Econometrica 57(6): 1361­1401. Valdés, A. 1996. "Surveillance of Agricultural Price and Trade Quiroz, J., and R. Soto. 1995. "International Price Signals in Policy in Latin America during Major Policy Reforms." Agricultural Prices: Do Governments Care?" Documento de Discussion Paper 349, World Bank, Washington, D.C. Investigación no. 88, Programa de Postgrado en Economía, ------. 2000. "Measures of Agricultural Support in Transition ILADES/Georgetown University, Santiago, Chile. Economies: 1994­1997." In A. Valdés, ed., Agricultural Quiroz J., W. Foster, and A. Valdés. 1999. "Agricultural Price Support Policies in Transition Economies, World Bank Instability and Price Floors: A Proposal." World Bank, Technical Paper 470, Europe and Central Asia Washington, D.C. Environmentally and Socially Sustainable Development Sarris, A. 1997a. "żHa aumentado la inestabilidad de los merca- Series, Washington, D.C. dos internacionales de cereales?" Revista Espańola de Valdés, A., and W. Foster. 2002. "Reflections on the Policy Economía Agraria, 181 (September­ December): 157­82. Implications of Agricultural Price Distortions and Price ------. 1997b. "Risk Management in Cereal and Oilseed Transmission for Producers in Developing and Transition Markets." Paper prepared for the Commodity Risk Economies." Paper presented at the OECD Global Forum on Management Group, World Bank, Washington, D.C. Agriculture, Agricultural Trade Reform, Adjustment and ------. 1999a. "Commodity Risk Management for Developing Poverty, May 23­24, Paris. Countries: Towards Implementing a New Approach." Paper Valdés, A., and J. Zietz. 1980. "Agricultural Protection in OECD prepared for the Commodity Risk Management Group, Countries: Its Cost to Less-Developed Countries." World Bank, Washington, D.C. International Food Policy Research Institute, Washington, ------. 1999b. "Price and Income Variability." Paper prepared D.C. for the Commodity Risk Management Group, World Bank, Valdés, A., and J. Zietz. 1995. "Distortions in World Food Washington, D.C. Markets in the Wake of GATT: Evidence and Policy Schiff, M., and A. Valdés. 1992. A Synthesis of the Economics in Implications." World Development 23(6): 913­26. Developing Countries. Vol. 4 of The Political Economy of Zietz, J., and A. Valdés. 1989."International Interactions in Food Agricultural Pricing Policy. Baltimore and London: Johns and Agricultural Policies: The Effect of Alternative Policies." Hopkins University Press. In I. Goldin and O. Knudsen, eds., Agricultural Trade Taylor, J. E. 2002. "Trade Integration and Rural Economies in Liberalization: Implications for Developing Countries. Paris: Less Developed Countries: Lessons from Micro Economy- OECD. Wide Models with Particular Attention to Mexico and Zivot, E., and D. Andrews. 2002. "Further Evidence on the Great Central America." Working Paper, Latin America and Crash, the Oil Price Shock and the Unit Root Hypothesis." Caribbean Office of the World Bank, Washington, D.C. Journal of Business and Economic Statistics 20(1): 25­44. 11 THE SANITARY AND PHYTOSANITARY AGREEMENT, FOOD SAFETY POLICIES, AND PRODUCT ATTRIBUTES Simonetta Zarrilli1 with Irene Musselli2 Introduction difficulties when requested to meet requirements in The issue of human and animal health and plant the importing markets, based on international stan- protection is high on the agenda of several devel- dards. Transparency-related requirements usually oped countries, fueled by recent cases of food poi- represent a burden for developing countries, and soning, the spread of pests among animals, and they are often unable to benefit from them, owing to environmental contamination. International trade lack of appropriate infrastructure. With the relative is perceived as a magnifier of such problems. Devel- lack of scientific and technical expertise, they experi- oping countries appreciate that, in several cases, ence serious problems with testing and conformity these concerns are genuine, but they fear that devel- assessment. The regional conditions provision of the oped countries may use sanitary and phytosanitary SPS Agreement, which greatly benefits developing (SPS) measures for protectionist purposes. Their countries, has been little used because of difficulties concern is well founded, since the major difficulty related to its scientific aspects. The provisions relat- in dealing with SPS measures is likely to lie in dis- ing to special and differential treatment in favor of tinguishing those measures that are justified by a developing countries remain rather theoretical, and legitimate goal, and have a scientific justification, apparently have not led to any concrete steps. from those that are applied to shield domestic The need for specialist scientific or technical producers from other-country agricultural exports. knowledge makes restrictions imposed for health Developing countries are not well positioned to and safety reasons much more difficult to challenge address this issue. They lack complete information than some other barriers to trade. While the on the number of SPS measures that affect their requirement that SPS measures be based on scien- exports, they are uncertain whether these measures tific evidence helps secure trade policy objectives, it are consistent or inconsistent with the World Trade is perceived by some environmental and consumer Organization's (WTO's) SPS Agreement, and they protection groups in some developed and develop- have no reliable estimate of the impact of such ing countries as a dangerous limitation on the right measures on their exports. of governments to take precautionary measures to Most developing countries are unable to partici- protect their citizens and the environment against pate effectively in the international standard-setting risks that can have irreversible effects. Differences process relating to SPS measures and therefore face between "sound science" and the "precautionary 215 216 Agriculture and the WTO approach" to health and safety are causing acute For developing countries, the most promising tensions among countries. option to maintain and expand their agricultural The situation is further complicated by the and food exports is to respond to opportunities in emergence of biotechnology, and international their target markets by providing high-quality, safe trade in biotechnology products. Because of the products. This implies building up knowledge, scientific uncertainty related to the impact of skills, and capabilities. Strengthening domestic biotechnology products on health and on the capacities in the SPS domain would also help devel- environment and strong consumer resistance in oping countries to identify products that they may some countries to these products, restrictive trade wish to keep out of their markets because of the measures are increasingly being implemented by a actual or potential negative impact on human number of governments. While the same rules health and safety, animal health, or the environ- apply to all countries, developing countries are ment. However, for this goal to materialize, devel- facing several challenges in this field. As exporters, oping countries need the support of their devel- they may have to prove that their products do oped partners, international trade organizations, not contain any bioengineered inputs. This may and technical organizations. They also need a mul- imply a system of certification and segregation tilateral legal framework that will facilitate achieve- that can be costly and burdensome. As importers, ment of such a result. they have to justify the scientific basis for their trade measures. The SPS Agreement Equivalence of SPS measures is of special rele- Negotiating History vance to developing countries when one takes into account the share and destination of their agricul- When the Uruguay Round started, there was a con- tural exports, and considers that they face cli- sensus that the time had come for reform of inter- matic, developmental, and technological condi- national agricultural trade (Stewart 1993). The tions that often differ from those prevailing in Punta del Este Declaration, which launched the developed countries. However, equivalence has Round in September 1986, called for increased not yet been recognized in a significant number of disciplines in three areas in the agricultural trade transactions. sector: market access, direct and indirect subsidies, The Doha Ministerial Decision on Implemen- and sanitary and phytosanitary measures. On the tation-Related Issues and Concerns provides some lattermost, the negotiators sought to develop a flexibility to developing countries in relation to the multilateral system that would allow simplification timeframe for compliance with the SPS measures,3 and harmonization of SPS measures, as well as to the interval between the publication of an SPS elimination of all restrictions that lacked any valid measure and its entry into force, to the issue of scientific basis. equivalence, to developing-country participation in At the outset of the Round the negotiating the international standard-setting activities, and to positions were as follows: the United States (U.S.) technical assistance. and the European Communities (EUs) proposed broad harmonization efforts, based on the expert- ise of international organizations; and the Euro- The Plan of This Chapter pean Communities called for all standards to be This chapter aims to recall the main features of based on scientific evidence. The Cairns Group the SPS Agreement and its purposes and negotiat- endorsed the broad recommendations toward har- ing history, stress some of the main difficulties monization and suggested that the burden of justi- encountered by developing countries in this area, fication of SPS measures should be placed on the and formulate a number of suggestions on how to importing country. Japan supported harmoniza- improve developing countries' ability to use and tion efforts based on the work of international benefit from the agreement. The chapter also pro- organizations, giving preference to the develop- poses some actions that developing countries ment of guidelines rather than standards; the may wish to consider in the course of the current improvement of notification and consultation pro- trade negotiations. cedures and of the dispute settlement mechanism; The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 217 and special allowances for developing countries. fulfills the general objectives of the Punta del Developing countries strongly advocated the Este Declaration. removal of sanitary and phytosanitary measures that acted as nontariff barriers to trade. They sup- Salient Features of the SPS Agreement ported the international harmonization of SPS and the Legal Framework measures to prevent developed countries from imposing arbitrarily strict standards. The main goal of the SPS Agreement is to prevent At the Mid-Term Review of the Uruguay SPS measures from having unnecessary negative Round (December 1988), it was agreed that the effects on international trade and from being mis- priorities in the area of SPS were international used for protectionist purposes. However, the harmonization on the basis of the standards devel- agreement fully recognizes the legitimate interest of oped by international organizations, develop- countries in setting up rules to protect food safety ment of an effective notification process for and animal and plant health, and in fact allows national regulations, establishment of a system countries to give these objectives priority over for the bilateral resolution of disputes, improve- trade, provided there is a demonstrable scientific ment of the dispute settlement process, and basis for their food safety and health requirements. provision for the necessary input of scientific More specifically, the agreement covers meas- expertise and judgment, relying on relevant ures adopted by countries to protect human or ani- international organizations. mal life from food-borne risks; human health from The Working Group on Sanitary and Phytosani- animal- or plant-carried diseases; animals and tary Regulations, formed in 1988, produced a draft plants from pests and diseases; and the territory of text in November 1990. Its central feature was the a country from the entry, establishment, or spread proposal to establish disciplines for SPS measures of pests. In sum, SPS measures are meant to ensure in an agreement separate from the draft Agreement food safety and to prevent the spread of diseases on Agriculture. The negotiators reached consensus among animals and plants. on a number of points: SPS measures should not Sanitary and phytosanitary measures are typi- represent disguised trade barriers; they should be cally applied to both domestically produced and harmonized on the basis of international stan- imported goods, may address the characteristics of dards, guidelines, and recommendations and of final products, and may take the form of residue generally accepted scientific principles; special con- limits, conformity assessment certificates, inspec- sideration should be taken of developing countries tions, quarantine requirements, designation of and their difficulties in meeting standards; trans- disease-free areas, and import bans. parency should be ensured in setting regulations The agreement states that countries should base and in solving disputes; and an international com- SPS measures on science, and establish them on the mittee should be established to provide for consul- basis of an assessment of the risk involved. Sanitary tations regarding standards. Some areas, however, and phytosanitary measures should ensure that the remained unsettled. appropriate level of protection (as determined by As a result of deadlock in the agricultural nego- the country in question) is achieved. If interna- tiations, the Uruguay Round was not concluded in tional standards, guidelines, and recommendations December 1990. A year later, however, the Director- exist, the agreement urges countries to base their General of the General Agreement on Tariffs and SPS measures on them. It encourages countries to Trade (GATT) put forward a package of proposals play a full part in the activities of international (the so-called Dunkel draft) that incorporated both organizations to promote the harmonization of newly negotiated text and personal proposals by SPS regulations on an international basis; to accept chairpersons of the negotiating groups. For SPS the SPS measures of exporting countries as equiva- measures, the Dunkel text closely followed the lent to their own if they achieve the same level draft text produced by the working group in of SPS protection; and, where possible, to November 1990. The final text of the SPS Agree- conclude bilateral and multilateral agreements on ment, as approved at the end of the Uruguay recognition of the equivalence of specific SPS Round, was largely based on the Dunkel text and measures. 218 Agriculture and the WTO BOX 11.1 The Joint FAO/WHO Codex Alimentarius Commission The Joint Codex Alimentarius Commission (CAC) Committee that oversees the commission's work. is an intergovernmental body that reports to the The Codex Alimentarius is the complete collec- Directors General of FAO and WHO. It is open to tion of standards, codes of practice, guidelines, all members of these two organizations, and cur- and recommendations adopted by the commis- rently has 166 members. It meets every two sion to achieve its objectives. The standards, years. Its primary function is protecting the health guidelines, and recommendations established by of consumers while ensuring fair practices in the the CAC on food additives, veterinary drug and food trade. The commission's organizational pesticide residues, contaminants, methods of structure comprises 23 Codex Committees and analysis and sampling, and codes and guidelines Task Forces, responsible for drafting food stan- of hygienic practice are recognized by the SPS dards; six Regional Coordinating Committees to Agreement as the international reference for food ensure regional coordination; and an Executive safety requirements." Source: Information provided by the relevant organization at the Meeting of the SPS Committee held on March 14­15, 2001. The agreement requires countries to choose (WHO) Joint Codex Alimentarius Commission those measures that are no more trade restrictive (CAC) (see box 11.1), the Secretariat of the Inter- than required to achieve domestic SPS objectives, national Plant Protection Convention (IPPC) provided these measures are technically and eco- (see box 11.2), and the International Office of Epi- nomically feasible (for example, applying a quaran- zootics (OIE) (see box 11.3).4 The SPS Agreement tine requirement instead of an import ban). The permits governments to decline international stan- SPS Agreement recognizes that, owing to differences dards and adopt lower standards. The agreement in geographical, climatic, and epidemiological con- also permits the adoption of SPS measures on a ditions prevailing in different countries or regions, provisional basis as a precautionary step, in cases in it would often be inappropriate to apply the same which there is an immediate risk related to the rules to products coming from different regions or spread of diseases, food contamination, biodiver- countries. This flexibility should not, however, lead sity damage, and so on, but where the scientific evi- to any unjustified discrimination among foreign dence is insufficient. suppliers or in favor of domestic producers. On the All countries must maintain an "enquiry same lines, governments should recognize disease- point"--an office to receive and respond to free countries, or disease-free areas within coun- requests for information regarding domestic SPS tries, and adapt their requirements to products orig- measures, including new or existing regulations inating in such countries or areas. and decisions based on risk assessment. Countries The agreement allows countries to introduce are required to notify the WTO Secretariat of any SPS measures that result in a higher level of pro- new SPS requirement, or modification of existing tection than would be achieved by measures based requirements, that they are proposing to introduce on international standards, if there is a scientific domestically, if the requirements differ from inter- justification or if a country determines on the basis national standards and may affect international of an assessment of risks that a higher level of san- trade. The secretariat circulates the notifications to itary and phytosanitary protection would be all WTO members. Notifications are to be submit- appropriate. In carrying out risk assessment, coun- ted in advance of the implementation of the meas- tries are urged to use techniques developed by the ure, so as to provide other countries with the relevant international organizations. Since the opportunity to comment on them. In an emer- entry into force of the SPS Agreement, a substan- gency, governments may implement a measure tial amount of work has been undertaken in the prior to notification. Countries are also required-- area of risk analysis by the Food and Agriculture the original text of the paper uses the word Organization (FAO)/World Health Organization "requested," but the agreement states that members The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 219 BOX 11.2 The Office International des Epizooties (OIE) The Office International des Epizooties (OIE), an renowned scientists. Standards are adopted by intergovernmental organization with 158 mem- the OIE's highest authority, the International ber countries, is concerned with the occurrence Committee, meeting in General Session in May and course of epizootics that could endanger of each year. Standards, guidelines, and recom- animal or human health. Its objectives and func- mendations developed under the auspices of the tions include developing regulations designed to OIE principally refer to standards for interna- prevent the spread of transmissible diseases to tional trade in animals and animal products humans and animals through trade in animals (International Animal Health Code, 10th Edition, and animal products, and the harmonization of 2001); standardized diagnostic techniques and requirements for such trade, in order to avoid vaccine control methods for use in international unjustified trade barriers. OIE standards are rec- trade (Manual of Standards for Diagnostic Tests ognized by the SPS Agreement as international and Vaccines, 4th Edition, 2000); (International sanitary rules of reference. They are prepared by Aquatic Animal Health Code, 4th Edition, 2001; elected Specialist Commissions and by Working Diagnostic Manual for Aquatic Animal Diseases, Groups bringing together internationally 3rd Edition, 2000)." Source: Information provided by the relevant organization at the Meeting of the SPS Committee held on March 14­15, 2001. "shall ensure" that all SPS measures adopted are Article XX(b), especially relating to procedural obli- published promptly--to publish the sanitary and gations concerning appropriate risk assessment and phytosanitary measures they have adopted and, proper scientific experimentation. except in urgent circumstances, to allow a reason- It is important to be able to distinguish meas- able interval between the publication of a new SPS ures that fall under the SPS Agreement from those measure and its entry into force. falling under the Uruguay Round Agreement on The SPS Agreement provides for special and dif- Technical Barriers to Trade (TBT), which replaced ferential treatment in favor of developing countries the Standards Code. This distinction is relevant, as and least-developed countries (LDCs). It includes, there are some significant differences in the provi- under certain circumstances, longer timeframes for sions of the two agreements. Whether a specific compliance, time-limited exceptions from the measure is a technical regulation (and therefore obligations of the agreement, and facilitation of within the scope of the TBT Agreement) or a sani- developing-country participation in the work of tary/phytosanitary measure (under the SPS Agree- relevant international organizations. ment) depends on the objectives for which it has The agreement includes provisions for a two- been adopted. As a general rule, if a measure is year grace period for all developing countries, adopted to ensure the protection of human, ani- which expired at the end of 1997. For LDCs, a five- mal, and plant life, and the health and protection of year grace period expired at the end of 1999. the territory of a country from damage caused by Prior to the entry into force of the SPS Agree- the entry, establishment, or spread of pests, it is an ment, health and safety regulations affecting SPS measure. Measures adopted for purposes other imports were subject to the GATT and to the "pluri- than these are subject to the TBT Agreement. For lateral" 1979 Agreement on Technical Barriers to instance, a pharmaceutical restriction would be a Trade ("Standards Code").5 measure covered by the TBT Agreement (see WTO The GATT recognizes, in Article XX(b) ("Gen- 1998). Labeling requirements related to food safety eral Exceptions"), that protecting human, animal, are usually SPS measures, while labels related to the and plant life and health is a legitimate objective of nutrition characteristics or quality of a product fall governments, and therefore exempts from GATT under TBT disciplines. obligations, under specific and strict circumstances, As far as the legal relationship between the TBT measures designed to meet these objectives. The SPS and SPS Agreements and GATT 1994 is concerned, Agreement provides detailed rules that supplement once SPS applies, TBT cannot apply. This is because 220 Agriculture and the WTO BOX 11.3 The International Plant Protection Convention (IPPC) The IPPC is a multilateral treaty deposited with the comes into force, establishes priorities for standard Director General of the FAO. One hundred and setting and harmonization of phytosanitary meas- seventeen governments are currently contracting ures. IPPC is named by the SPS Agreement as the parties to the IPPC. Amendments to the conven- international organization responsible for phy- tion were unanimously adopted by the FAO Con- tosanitary standard setting and the harmonization ference in 1997 (referred to as the `New Revised of phytosanitary measures affecting trade. To Text of the IPPC') to update the convention and date, 17 international standards for phytosanitary reflect the role of the IPPC in relation to the WTO- measures (ISPMs) have been adopted. ISPMs 13 SPS Agreement. The convention is administered through 17 were adopted at the Fourth Session of through the IPPC Secretariat located in FAO's the ICPM (March 11­15, 2002). Additionally, a Plant Protection Service. The Interim Commission number of standards are under development and on Phytosanitary Measures (ICPM), established as revision. an interim measure by FAO until the IPPC (1997) Source: Information provided by the relevant organization at the Meeting of the SPS Committee held on March 14­15, 2001. the SPS Agreement has a very well-defined but lim- instructs countries to enter into consultations, ited scope of application and is more rigorous than upon request, with the aim of achieving bilateral the TBT Agreement in its requirements. Either the and multilateral agreements on recognition of SPS or the TBT Agreements and the GATT can the equivalence of specified sanitary or phytosani- apply concurrently. In the event of conflict between tary measures. TBT/SPS and GATT, the specific agreement prevails Equivalence is the best option when harmo- over GATT, according to General Interpretative nization of standards is not desirable or when Note to Annex 1A of the the Marrakesh Agreement international standards are lacking or are inap- Establishing the World Trade Organization. propriate. For developing countries--which face difficulties in trying to harmonize their standards Main Issues for Developing with those of importing countries--the recogni- Countries in the SPS Agreement tion of the equivalence of their SPS measures would represent a key instrument to enhance Equivalence market access for their products. These countries The November 2001 Doha Decision on Implementa- face climatic, developmental, and technological tion-Related Issues and Concerns instructs the WTO conditions rather different from those prevailing SPS Committee"to develop expeditiously the specific in developed countries; recognition of the equiv- program to further the implementation of Article 4 alence of their SPS measures to those applied by of the Agreement on the Application of Sanitary and the importing countries would represent a key Phytosanitary Measures" (WTO 2001a, para. 3.3). instrument to enhance market access for their Discussion on equivalency has been going on in products. the SPS Committee since 2000 and a decision on In this regard, the following issues might be of the implementation of equivalence was taken in special concern to developing countries. October 2001 (WTO 2001b). Article 4 of the SPS Agreement encourages coun- The Concept of Equivalence and Its Function tries to give positive consideration to accepting as equivalent the SPS measures of other WTO mem- The function of equivalence is to facilitate interna- bers, even if these measures differ from their own tional trade by recognizing that different measures or from those used by other countries, if the can achieve the same level of sanitary and phy- exporting country demonstrates that its measures tosanitary protection. Therefore countries have achieve the importing member's appropriate level flexibility about the kind of measures to adopt to of sanitary and phytosanitary protection. It also ensure adequate SPS protection. Equivalence, then, The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 221 is not about "duplication" or "sameness" of SPS agreements can incorporate a general part estab- measures. What is relevant is the achievement of lishing overall principles and long-term objectives, the appropriate level of protection sought by the which might accommodate developing countries' importing country. How this protection is achieved attitudes and concerns on the issue of equivalence, is not an autonomous issue. Methods might be rel- and which might be reflected in the way specific evant to the extent that the inquiry into methods is equivalence arrangements are forged. Second, even instrumental to assessing the achievement of the if equivalence agreements are time consuming appropriate level of protection, but do not have any and resource intensive, developing countries discrete relevance per se. The option of introducing might find it even more costly to seek equivalence additional requirements relating to how the level of on an ad hoc basis at the technical level. Argentina protection is achieved would add an unnecessary has proposed that equivalence agreements should burden to the recognition of equivalence. contain a general part establishing overall princi- ples, criteria, objectives, and long-term targets, and specific annexes for the products coming under Implementation of Equivalence the agreement (WTO 2001e). The concept of equivalence lends itself to various applications (WTO 2000b and WTO 2001c). Guidelines on the Recognition of Equivalence Variables on which the practical implementation of equivalence depends are, among others, the A related issue concerns the provision of guidance scope of the equivalence arrangement (specific on the recognition of equivalence. International products or product sectors, specific technical guidelines may be needed for the systematic appli- aspects of certain SPS measures, specific SPS cation of equivalence. In particular, in the absence measures, SPS systems, inspection and control of guidelines on methodology for judging equiva- systems, processing techniques); its level of for- lence, specific bilateral issues are more likely to mality (ad hoc recognition at the technical level,6 arise, and methodological concerns of developing unilateral determination of equivalence at the countries are more likely to be neglected. Interna- administrative level,7 formal agreements); and tional standard-setting bodies have been formally the parties involved (bilateral or multilateral) encouraged to elaborate guidelines, as appropriate, (WTO 2000a). Equivalence arrangements on spe- on the equivalence of SPS measures and equiva- cific technical matters play an important role in lence agreements.8 According to the March 2002 building confidence between laboratories and work program, draft guidance for the recognition certifying authorities in different countries and of equivalence of products historically or previ- usually represent a necessary step toward the con- ously traded, on the basis of categorization of trade clusion of broader arrangements. They may also patterns and risks, will be considered within the represent crucial learning experiences, since they SPS Committee (WTO 2002a). imply an intensive exchange of information and close contacts between relevant authorities. The North-South Dimension of Equivalence According to some WTO members, it might be Recognition of Equivalence versus Equivalence desirable not to emphasize the North-South Agreements dimension of equivalence as this is seen as oversim- In the ongoing debate on equivalence, emphasis is plistic and not exhaustive, disregarding the recipro- increasingly placed on the "recognition of equiva- cal nature of equivalence arrangements and the fact lence" rather than on formal "equivalence agree- that an increasing number of decisions are con- ments." Ad hoc acceptance of the equivalence of cerned with South-South trade. Even if equivalence particular SPS measures is largely recognized as the cannot be reduced to a North-South issue, this most effective way to apply the equivalence provi- dimension should not be underestimated. Equiva- sions. However, it might be worth considering lence is a particularly prominent issue for develop- some advantages associated with the negotiation of ing countries, reflecting their share and destination formal equivalence agreements. First, equivalence of agricultural and food exports,9 their technical 222 Agriculture and the WTO capability to comply with SPS requirements, and Burden of Proof and Costs Sharing the diversity of conditions prevailing in developed A related issue is the division of responsibilities and developing countries. Developing countries between the importer and the exporter in the deter- face climatic, developmental, and technological mination of equivalence. According to the October conditions that often differ from those prevailing 2001 Decision on the Implementation of Article 4 of in developed countries, and which might not the Agreement on the Application of Sanitary and be duly reflected in SPS systems designed and oper- Phytosanitary Measures (the "Decision on Equiva- ated in developed countries. In these cases, equiva- lence") (WTO 2001g), the importing country lence could contribute considerably to sound should explain the objective and rationale of the SPS results from a safety point of view, without unneces- measure at stake, identify the risks that the measure sary costs. is intended to address, indicate the appropriate level of protection that the measure is designed to achieve, Equivalence and the Importing Member's and give full consideration to requests by developing Appropriate Level of Protection countries for technical assistance to facilitate the The concept of equivalence is linked to the deter- implementation of Article 4. It is up to the importing mination of the appropriate level of protection. country to analyze all relevant evidence provided by A crucial issue for the exporting country is there- the exporting country on its SPS measures with a fore to identify the importing country's acceptable view to determining whether these measures achieve level of protection and the way the importing the appropriate level of protection. However, the member sets it. It might be worth emphasizing that importing country is not required to justify the the determination of the level of protection is not refusal for equivalence, though it should respond, an unquestionable sovereignty issue, being quali- normally within six months, to any request of equiv- fied by the wording of Article 5.4 (obligation to alence. A crucial issue in this respect is the impact on minimize negative trade effects) and Article 5.5 the competitiveness of the exported products of (obligation to avoid arbitrary or unjustifiable dis- costs related to assessing equivalence of SPS mea- criminations or disguised restrictions to interna- sures and the sharing of such costs between the tional trade). The appropriate level of protection importer and exporter. set by the importing country needs to be consis- Equivalence at the regional level, in the frame- tently met by domestically produced goods. SPS work of regional or subregional agreements, is eas- measures that are not enforced on domestic pro- ier to achieve. Developing countries may therefore ducers, but which are imposed on foreign ones, have an interest in analyzing the possibility represent disguised trade restrictions. Another rele- of including reference to the equivalency of SPS vant issue concerns transparency and cooperation. measures in the framework of regional and subre- In particular, the importing country should supply gional groupings.10 correct information on its acceptable level of risk so A necessary precondition for implementing that the exporting country can meet the require- equivalence is the capacity of the exporting country ment of objectively demonstrating that its SPS to provide scientific and technical information to measures are equivalent. In the salmon dispute support the claim that its measures achieve the (WTO 1998b), the WTO Appellate Body noted that appropriate level of protection identified by although the determination of the appropriate level the importing country. The latter may ask to check of protection is a prerogative of the member con- the laboratories and the testing facilities of the cerned, and there is no obligation to determine the exporting country to be assured about the reliability appropriate level of protection in quantitative of the information provided and the technical com- terms, this does not mean"that an importing mem- petence of the exporting country. Obviously a well- ber is free to determine its level of protection with prepared case, adequately justified from a scientific such vagueness or equivocation that the application point of view and supported by trustworthy certifi- of the relevant provisions of the SPS Agreement, cates, will have more chance of being considered pos- such as Article 5.6, becomes impossible" (WTO itively than an ill-prepared case. In this field, develop- 1998b, paras. 199 and 206). ing countries face enormous difficulties which could The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 223 jeopardize their capacity to benefit from equivalence sions and Articles 10.2 and 10.4 as nonmandatory and from the specific work program. Operationaliz- provisions (WTO 2002b). Presumably, Article 10.3 ing equivalence implies, therefore, the strengthening is also a nonmandatory provision. of developing-country scientific capacities, laborato- During the 1980s and in the early 1990s, the ries, and certification and accreditation authorities. S&D principle came under criticism: instead of The establishment of internationally financed encouraging good polices and practices in the national, regional, or subregional laboratories, certifi- developing countries, it was seen as contributing to cation bodies, and accreditation institutions could be development-unfriendly policies. This perception explicitly included as one of the suitable outcomes of influenced how S&D provisions were included in the work program on equivalence. the Uruguay Round agreements: they reflect the trend toward reducing the scope for S&D, in partic- ular for non-LDCs. The present rules and the Special and Differential Treatment related debate reflect a difficult equilibrium and Technical Assistance between the recognition of international asymme- The SPS Agreement includes a specific article (Article tries (implying that unequal countries cannot be 10) on special and differential treatment (S&D) for treated as equals) and the notion of "leveling the developing countries and LDCs, and another (Article playing field" (meaning that developing countries 9) on technical assistance. The Doha Decision on should not be shielded through discriminatory Implementation-Related Issues and Concerns pro- instruments in their favor, but helped to become vided some instructions and clarifications aimed at more efficient and able to compete fairly in interna- operationalizing Article 10, and indicated July 2002 tional markets). Special and differential provisions as the deadline for the identification of the S&D pro- are included in two different areas of international visions that should be made mandatory and for economic cooperation: financial and monetary examining additional ways in which S&D provisions instruments as provided by the international and can be made more effective (WTO 2001a, note 7, regional financial institutions; and trade disciplines para. 3.1 and para. 12.1 (i) and (ii)). at the multilateral and subregional levels. The lat- The WTO Secretariat has identified six types of ter are more prone to criticisms and challenges S&D provisions: (i) provisions aimed at increasing than the former (see UNCTAD Secretariat 2002; the trade opportunities of developing countries; Breckenridge 2002). (ii) provisions under which WTO members should Article 10.1 of the SPS Agreement states that safeguard the interests of developing countries; the special needs of the developing and least- (iii) provisions concerning the flexibility of com- developed countries shall be taken into account in mitments, of actions, and use of policy instru- the preparation and application of SPS measures. ments; (iv) provisions specifying transitional time Article 10.2 states: "Where the appropriate level periods; (v) provisions on technical assistance; and of sanitary or phytosanitary protection allows (vi) provisions related to LDCs. Provisions relating scope for the phased introduction of new sanitary to flexibility and transition times tend to specify and phytosanitary measures, longer time-frames exceptions to rules to which developing countries for compliance should be accorded on products of may have recourse if they choose. Provisions relat- export interest to developing country Members so ing to technical assistance, the safeguarding of the as to maintain opportunities for their exports." interests of developing countries, and measures to The Decision on Implementation clarifies that increase participation in world trade tend to specify "longer time-frames for compliance" shall be under- positive actions to be undertaken by developed stood to mean normally a period of not fewer than countries in favor of developing countries. Accord- six months. Moreover, "Where the appropriate level ing to this classification, Articles 10.1 and 10.4 of of sanitary and phytosanitary protection does not the SPS Agreement fall under the second category, allow scope for the phased introduction of a new Articles 10.2 and 10.3 fall under the fourth category, measure, but specific problems are identified by a while Article 9 falls under the fifth category (WTO Member, the Member applying the measure shall 2001g, pp. 5, 56­62). The WTO Secretariat has iden- upon request enter into consultations with the coun- tified Article 10.1 and Article 9 as mandatory provi- try with a view to finding a mutually satisfactory 224 Agriculture and the WTO solution to the problem while continuing to the agreement, of appropriate administrative achieve the importing member's appropriate level framework, and limited participation in the activi- of protection" (WTO 2001a, note 7, para. 3.1). ties of the international standard-setting organiza- Article 10.3 provides that specified, time-limited tions and of the SPS Committee make it difficult exceptions in whole or in part from the obligations for them to comply with the commitments under under the SPS Agreement could be granted to devel- the agreement.11 oping countries by the SPS Committee, upon As far as Article 10.3 is concerned, a delay in the request, to ensure that developing countries are able implementation of the agreement would be worth to comply with the provisions of the agreement. To pursuing under the condition that the transitional date, no request has been made under Article 10.3. period is used to strengthen capacities in develop- Article 10.4 refers to developing-country partic- ing countries to satisfy their trade partners' SPS ipation in the relevant international organizations. requirements, bring their domestic measures in The issue is addressed in the following section. conformity with international standards, and enter It is worth stressing that developing countries' into equivalence agreements. For this to happen, agricultural exports are concentrated in a few prod- international technical and financial support ucts and in a few markets and that the number of should be provided. Hence, developing countries exporting enterprises is also limited. This situation will need longer time frames for compliance and should make it easier to implement Articles 10.1, technical assistance to make it possible for their 10.2, and 10.3 of the SPS Agreement and paragraph products to meet the market requirements and for 3.1 of the Decision on Implementation. To further creating or strengthening the domestic institutional facilitate the functioning of such provisions, each framework necessary to comply with the provisions developing and least-developed country could pre- of the agreement. In fact, granting a longer time pare a list of the main agricultural products it frame for compliance without, at the same time, exports (perhaps a list of five to seven products), providing technical assistance, would have the identify the principal countries of destination effect only of postponing the problem. Transitional (again a list of five to seven markets), and circulate periods, longer time frames, and the like acquire it among WTO members. Whenever new SPS a real meaning for development when they are measures affecting the listed products are intro- accompanied by a policy package that facilitates the duced by a developed member, it should contact needed adjustments. the developing countries concerned and ensure The provisions on S&D treatment are therefore that the newly introduced measure is not going to very much linked to those on technical assistance. disrupt traditional trade flows. For this to happen, The SPS Agreement was apparently negotiated and the list should be "dynamic"--with products and concluded with little regard for the conditions neces- markets added or deleted as necessary and detailed sary for its effective implementation, particularly in (for example, including pesticides used on developing countries. Article 9.1 states that the assis- exported products) to facilitate the task of the tance provided to developing countries, either bilat- importing country in alerting the exporting coun- erally or through the appropriate international try about new SPS measures that may have an organizations,"may be, inter alia, in the areas of pro- impact on its exports. cessing technologies, research, and infrastructure, If an importing country implements an SPS including in the establishment of national regulatory measure that is scientifically justified but disrup- bodies, and may take the form of credits, donations, tive, either the implementing country should and grants." This is a mandatory provision that, if reconsider the measure, or, if this proves impossi- carried out well, would create greater coherence in ble, provide assistance to the affected developing SPS policy because it would enable developing countries to meet the new requirements to preserve countries to establish the necessary infrastructural existing trade flows. The option of equivalence and other conditions needed for the effective should also be considered. In this respect, develop- implementation of the agreement. Technical coop- ing countries have stressed that lack of adequate eration and financial support, however, are not a infrastructure, technology, finance, and skilled panacea and should not be used to replace the labor force, as well as lack of full understanding of removal of unnecessary SPS obstacles to trade. The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 225 Technical cooperation should be extended to forces on proposed changes to a specific SPS meas- cover capacity-building of the officials in charge of ure. For developing countries, it may be useful to the enquiry points, as transparency is a key issue for develop flexible alliances both among themselves the correct functioning of the agreement. It should and with developed countries, considering that the be extended to upgrade the technical skill of per- latter are often more experienced in bringing spe- sonnel working in laboratories, certification bodies, cific cases to the attention of other countries or to and accreditation institutions, since their having a the attention of the SPS Committee. Since technical certain level of qualifications and training is a pre- cooperation in the field of SPS measures is being condition for the international acceptance of cer- provided by several international organizations and tificates issued by them and represents the basis for by a number of developed countries, better coordi- the negotiation of equivalence agreements. Since nation among the different institutions would developing countries experience difficulties in deal- ensure that beneficiary countries fully benefit from ing with the scientific side of the agreement, in these efforts. particular risk assessment, technical cooperation The WTO Secretariat has identified four cate- should be extended on this matter. gories of technical assistance: information, training, The Decision on Implementation addresses the "soft" infrastructure development, and "hard" infra- issue of technical assistance, but only with reference structure development, and has developed a ques- to LDCs: it "(i) urges Members to provide, to the tionnaire for submitting requests for technical coop- extent possible, the financial and technical assis- eration (WTO 2001j). An informal meeting of the tance necessary to enable least-developed countries SPS Committee on technical assistance was held in to respond adequately to the introduction of any July 2001 and some specific suggestions were put new SPS measures that may have significant nega- forward to improve the effectiveness of technical tive effects on their trade; and (ii) urges members cooperation activities, namely: linking the private to ensure that technical assistance is provided sector to all initiatives; taking a regional approach; to least-developed countries with a view to integrating all forms of assistance into a coherent responding to the special problems faced by them program regardless of the source of assistance; focus- in implementing the Agreement on the Application ing on the development of human resources; and tar- of Sanitary and Phytosanitary measures" (WTO geting assistance to help developing countries to com- 2001a, note 7, para. 3.6). ply with measures in their export markets as well as to According to Article 9.2, "where substantial develop their own regulatory control systems to investments are required in order for an exporting ensure the safety of their domestic food supplies developing country Member to fulfill the sanitary (WTO 2001i). As a general rule, the more specific a or phytosanitary requirements of an importing request is, the more likely it is to be satisfied promptly Member, the latter shall consider providing such and adequately. On the other hand, vague and "shop- technical assistance as will permit the developing ping list" requests risk triggering actions that are country Member to maintain and expand its mar- rather unfocused and not particularly effective. ket access opportunities for the product involved." In conclusion, developing and least-developed Article 9.2 can be addressed along the same lines of countries may wish to request the establishment Article 10. First, this provision should be strength- of a clear and mandatory link between the S&D ened by requesting the country implementing an provisions included in the SPS Agreement and SPS measure that creates particular difficulties for those on technical assistance as one of the ways to developing countries, to reconsider it. Second, if make the S&D approach more effective, within the after reviewing its implications, the importing mandate included in the Decision on Implementa- country reconfirms the measure, the provision of tion-Related Issues and Concerns, paragraph technical cooperation, including the transfer of the 12.1(ii). This goal could be achieved through a necessary technology, should follow, considering change in the language of the agreement, or that Article 9.2 is a mandatory provision. Countries through an authoritative interpretation of it, pur- that experience the same trade problems in connec- suant to Article IX:2 of the Marrakesh Agreement tion with a specific SPS measure may wish to join Establishing the World Trade Organization. 226 Agriculture and the WTO The efficiency and fairness of the international International Standards and International standard-setting process is crucial. It has occurred Standardizing Organizations that standards developed by a limited number of The SPS Agreement aims to further the use of har- countries or approved by a narrow majority of par- monized measures based on internationally agreed ticipants have achieved the status of international standards. Hence, SPS measures that conform to standards. Because of the simple majority rules, for international standards "shall be deemed to be nec- instance, some Codex standards have been adopted essary to protect human, animal or plant life or or rejected by a relatively small majority of coun- health, and presumed to be consistent with the tries. This situation is well illustrated by the stan- relevant provisions of this Agreement and of GATT dard on maximum residue limits for growth hor- 1994" (WTO 1999b SPS Agr., Art. 3.2). The stan- mones (beef) that was approved by 33 votes in dards, guidelines, and recommendations developed favor, 29 against, and seven abstentions. The revised under the auspices of the CAC, OIE, and the Secre- standard for natural mineral waters was approved tariat of the IPPC are explicitly referred to in the by 33 votes in favor, 31 against, and 10 abstaining. SPS Agreement as the international bodies respec- As a result of increased criticism, international tively for food safety, animal health and zoonoses, standard-setting bodies have adopted and are and plant health (WTO 1999b SPS Agr., Annex A, developing new procedures to ensure that stan- para. 3(a)). For matters not covered by these organ- dards truly reflect the view of all member countries. izations, standards established by other relevant As regards the CAC, the Codex Committee on international organizations, as identified by the SPS General Principles, at its 14th Session, (April 19­23, Committee, may be recognized (WTO 1999b SPS 1999), discussed various options to ensure greater Agr., Annex A, para. 3(d)). fairness and efficiency in standard setting. Since Harmonization of international standards then, the commission has committed itself to pur- entails major benefits. It facilitates trade: the same sue consensus in the approval of its standards, as requirements applying in all countries that base opposed to a simple majority of votes cast. The their national measures on international standards. CAC, and specifically its 24th Session (July 27, It reduces disputes: since measures based on stan- 2001), continues to address the matter of develop- dards are presumed to be WTO-consistent they are, ing-country participation. The issue of participa- in principle, not challenged by trading partners. tion (and transparency) is being discussed in sev- And it fosters global dialogue on technical issues, eral areas, including as part of the commission's with no single country bearing the burden of risk Strategic Framework, which sets out the strategic assessment. Conversely, the divergence of standards priorities for the CAC and provides the basis for the and regulations creates costs for international elaboration of the Medium-Term Plan for the trade. Nevertheless, in some cases, these costs are period 2003­7.12 The participation of developing justified, since they arise from legitimate differ- countries in the Codex process is also likely to be ences in societal preferences, technological devel- considered within the current Review of the Joint opment, and environmental and health conditions. FAO/WHO Food Standards Program, which was In these cases, standards harmonization would not discussed at the extraordinary 49th Session of the be a desirable solution, while equivalence of SPS Executive Committee in September 2001 (CAC measures would provide a better option. 2001b, paras. 42­3). As to the IPPC, current proce- The benefits of harmonization may be impeded dures, entailing a nine-step elaboration and consul- if the process is captured by special interests in tation process including review after adoption, order to exclude market participants, or if it is not were adopted by the Interim Commission on Phy- adequately transparent. Against this, the adoption tosanitary Measures (ICPM) in 1999, based on of consultative and participatory procedures in interim procedures established by FAO (ICPM standard setting, in some cases including nontradi- 1999). Current procedures and policies emphasize tional stakeholders, makes the development and transparency, participation, and geographic repre- adoption of international standards more complex sentation in the IPPC's standard-setting processes. and time consuming, and implies that considera- All standards submitted to the ICPM have been tions of a nonscientific nature may play a role. adopted by consensus. Standards can be adopted by The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 227 a two-thirds majority vote if necessary; however, a At the 2001 WTO Ministerial Conference in vote cannot be requested for the adoption of a stan- Doha, the WTO, FAO, OIE, WHO, and the World dard on the first occasion it is submitted to the Bank issued a joint statement committing them- ICPM. The OIE has adopted a consensus approach selves to help developing countries participate to the development and adoption of standards. more fully in setting international norms for SPS However, it is reported that almost all written measures. The agencies are committed to coordinat- comments on draft Animal Health Code standards ing the technical assistance they give to developing come from fewer than 10 countries, with oral com- countries as part of this effort (WTO 2001h). ments from about the same number of countries. The costs of direct participation in standard set- Article 10.4 of the SPS Agreement states ting pose a constraint to participation by develop- that "Members should encourage and facilitate ing countries. For the CAC, its subsidiary bodies the active participation of developing country responsible for drafting proposed standards meet Members in the relevant international organizations." either annually or biennially, creating a burden on However, developing countries have repeatedly all member countries regarding participation costs, expressed their concern about the way in which inter- but affecting developing countries more. Hence, national standards are developed and approved, their participation directly in the standard-setting pointing out how their own participation is very lim- activities of the CAC is mainly in the plenary com- ited in both number and effectiveness. This makes mission sessions (where standards are formally international standards irrelevant to them, inappro- adopted). Conversely, the participation of develop- priate for use as a basis for their domestic SPS meas- ing countries in the committees responsible for ures, and extremely difficult to comply with when drafting proposed standards is still below the level incorporated by the importing countries in their that would be considered representative of the CAC national regulations.Developing countries'participa- as a whole.13 The OIE bears the cost of experts' par- tion in international standards setting is an "imple- ticipation in the working groups and specialist mentation" issue that has been discussed in the WTO commissions, as well as that of delegates attending General Council. The "three sisters" (CAC, OIE, and the annual General Session of the International IPPC) briefed members on participation in interna- Committee, where draft standards are discussed tional standard-setting bodies in a workshop before and adopted. Funds are available for developed- the meeting of the SPS Committee (March 13­15, and developing-country experts alike. 2001). The information showed that developing The most critical constraint, however, to effective countries are participating more than they were participation by developing countries in interna- before, but not necessarily in the most adequate man- tional standards setting is the lack of capabilities at ner (see WTO 2001k, 2001l, 2001m). the national level for the evaluation of draft stan- The Doha Decision on Implementation-Related dards and the formulation of positions in consulta- Issues and Concerns "(i) takes note of the actions tion with all interested parties. This means that taken to date by the Director-General [of the WTO] solutions such as sponsoring the participation of to facilitate the increased participation of Members developing-country delegates in plenary meetings at different levels of development in the work of the where international standards are formally relevant international standard setting organiza- approved are positive but far from sufficient. Ade- tions as well as his efforts to coordinate with these quate and effective participation of developing organizations and financial institutions in identify- countries in the international standard-setting ing SPS-related technical assistance needs and how process relies on their technical capacity to con- to best address them; and (ii) urges the Director- tribute to the process by proposing solutions and General to continue his cooperative efforts with criteria that are both scientifically sound and consis- these organizations and institutions in this regard, tent with their technological and developmental including with a view to according priority to the conditions. It is to this end that efforts are needed. effective participation of least-developed countries International cooperation aimed only at increasing and facilitating the provisions of technical and the number of developing-country delegates pres- financial assistance for this purpose" (WTO 2001a, ent at the official meetings of the international stan- note 7, para. 3.5). dardization bodies would be inadequate, and even 228 Agriculture and the WTO counterproductive, for developing countries, since unnecessary detrimental impact on international it would make it possible to define as "genuinely trade. However, variations in the quality and con- international" activities that in reality are not. tent of the information provided by countries in The work of standard-setting organizations may their notifications; short comment periods; delays also be of limited relevance to developing countries in responding to requests for documentation; and in the sense that standards are often developed for absence, at times, of due consideration for the com- products that are not of export interest to them. ments provided by other members are recurrent This situation has not helped to make developing problems limiting the effective implementation of countries particularly interested and, in conse- the transparency provisions. quence, active in the process of international stan- Paragraph 2 of Annex B of the SPS Agreement dardization. To a large extent, developing countries mandates members to allow "a reasonable interval see the international standards as developed by and between the publication of a sanitary or phytosani- for the developed countries. An effort should there- tary regulation and its entry into force in order to fore be made to develop standards on products that allow time for producers in exporting Members, are of immediate relevance to developing countries, and particularly in developing country Members, and that can facilitate their exports. Certain finan- to adapt their products and methods of production cial issues relating to directed funding for the devel- to the requirements of the importing Member." The opment of standards should be taken into account Decision on Implementation-Related Issues and when addressing this problem. Concerns clarifies that "the phrase `reasonable A recent development within the IPPC may well interval' shall be understood to mean normally a illustrate the case. Because of severe underfunding period of not less than 6 months." According to of the work program, financial assistance is pro- paragraph 5 of Annex B, an adequate time frame vided for standard-setting through direct financial has also to be provided between the notification of contributions from governments ("sponsorship of a proposed regulation and its adoption, because standards"). Concerns have been expressed by some this allows other members to provide comments on members that if funds are provided to assist with the draft. the development of a specific standard, the standard Language may be an obstacle to the effective might be afforded preferential treatment in the list capacity of countries to comment on draft regula- of priorities. Hence, rules have been developed to tions; therefore members have agreed that at least a ensure that no special treatment is given to the summary of the proposed regulation in one of the development of standards that are provided with official languages of the WTO (English, French, financial assistance. In view of the 4th Session of the and Spanish) should be made available by the noti- ICPM, held in Rome, March 11­15, 2002, the Infor- fying country, if the notifying country is a devel- mal Working Group recommended (ICPM 2002) oped country. In accordance with paragraph 9 of that the provision of external resources for standard Annex B of the SPS Agreement, the WTO Secre- setting should be applied only for standards that are tariat circulates copies of the notifications to all approved as priorities by the ICPM, and that it members and draws the attention of developing should follow the normal procedures, policies, and countries to any notification of special interest to practices of standard setting with no modification them. Notifications are circulated both by hard copy according to the preferences of the funding entities. and electronically. The ability of the enquiry points Accordingly, the ICPM has been invited to amend in developing countries to receive and provide the existing criteria for setting the topics and priori- information electronically is, therefore, crucial. ties in standard setting by removing the last crite- At times, even when countries are able to pro- rion listed--namely, the "availability of external vide comments on the draft regulation, those com- resources to support preparation of a standard" ments are not taken into account by the notifying (ICPM 1998, para. 13). country and the whole exercise becomes pointless. A possible solution to this problem could be that Transparency and Notification Provisions when comments and suggestions are not reflected Transparency is vital to make sure that SPS meas- in the final text of the measure, the notifying coun- ures are scientifically sound and do not have an try would have to explain the reason. The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 229 The SPS Committee is a forum in which coun- cedures, while keeping them scientifically sound, tries can discuss the implementation of the agree- and support for developing countries to prepare ment, bring their difficulties to the attention of their submissions for the recognition of pest- or other countries, and challenge specific SPS meas- disease-free areas or of areas of low pest or disease ures proposed or implemented by other members. prevalence. Developing countries have to determine Developing countries are still making limited use when it is feasible and cost-effective to make efforts of this forum (though their participation is grow- to eradicate a particular disease or pest from a zone ing), as well as of the other transparency provi- and whether they can get an appropriate return on sions included in the agreement. The limited use their investment. This is clearly an area where expert of the transparency provisions may be because assistance would facilitate the actual implementa- links between the public authorities and the pri- tion of the provision of the agreement by develop- vate sector are loose, with the result that public ing countries.14 Once a country or an area within a authorities are not fully aware of the difficulties country has been declared pest- or disease-free by that exporters face, while the private sector lacks the relevant international organizations, this status appropriate channels to bring the difficulties it should not be questioned again by individual trade experiences to the attention of the competent partners, which may, if necessary, request further authorities. Developing countries may, therefore, evidence in terms of Article 6.3, but should refrain consider making the necessary efforts to from requesting the concerned country to resubmit strengthen these links. all the evidence. The nonrecognition by a trade partner of the status of "pest- or disease free area" declared by the competent international organiza- Adaptation to Regional Conditions tion can be regarded as a trade-restrictive measure Within a given country, the situation regarding in terms of Article 2.3 of the agreement (see WTO plant or animal disease may not be uniform. The 1999a, Art. 6(2), Annex A(3)(B)). importing country should, therefore, consider whether there are zones within the exporting coun- try that represent a lesser danger, either as a result The Precautionary Approach to Health of the prevailing natural conditions or because the and Safety and Biotechnology exporting country has made efforts to eradicate the disease from such zones and has taken the In the present debate about health and safety, the necessary measures to prevent its reintroduction. "precautionary approach" is often opposed to a The adaptation to regional conditions, including "sound-science approach." Varying perceptions the recognition of pest- or disease-free areas or areas about public health risks, different levels of accept- of low pest or disease prevalence (SPS Agreement ance of them, divergent priorities, and dissimilar Article 6), is of key relevance to developing coun- economic interests have driven countries to take tries, especially large countries where geographical, quite opposite positions in this field and have cre- environmental, and epidemiological conditions ated acute tensions among them. International may vary considerably from one region to another. trade magnifies the problem. In some cases, adaptation to regional conditions has Article 5 of the SPS Agreement permits the facilitated trade in agriculture products. However, adoption of SPS measures on a provisional basis as efforts to eradicate a pest or disease from a specific a precautionary step in cases where there is an area may require a large investment and the proce- immediate risk of the spread of disease, but the sci- dures to prove that an area is pest-free or disease- entific evidence is insufficient. However, it also pro- free, or is an area of low pest or disease prevalence, vides that "Members shall seek to obtain the addi- are usually long and burdensome, and often require tional information necessary to a more objective the provision of complex scientific evidence. As a assessment of risk and review the sanitary or phy- result, developing countries have been unable to tosanitary measure accordingly within a reasonable fully benefit from Article 6, despite the support pro- period of time." The SPS Agreement thus includes a vided by the relevant international organizations. rather strict interpretation of the precautionary Possible solutions include simplification of the pro- approach. 230 Agriculture and the WTO In the mid-1980s the precautionary approach entific certainty about the effects of the genetically started to appear in multilateral environmental modified (GM) products on biodiversity and agreements. Its use increased further in the 1990s, human health. However, since the importing coun- especially when it was included as one of the 27 try is not obliged to seek the information necessary principles of the Rio Declaration on Environment for reaching scientific certainty, a trade-restrictive and Development (UNCED 1992). Principle 15 measure may be in force without time limits. The reads: "In order to protect the environment, the SPS Agreement, on the contrary, allows countries precautionary approach shall be widely applied by to provisionally adopt sanitary or phytosanitary States according to their capabilities. Where there measures when relevant scientific evidence is insuf- are threats of serious or irreversible damage, lack of ficient, but obliges them to seek the additional full scientific certainty shall not be used as a reason information necessary for a more objective assess- for postponing cost-effective measures to prevent ment of risk and to review the SPS measure within environmental degradation." By 1990, the precau- a reasonable period of time (see Zarrilli 2000). The tionary approach was also appearing in regional Preamble of the Biosafety Protocol states that it declarations and treaties. Since 1992, it has also shall not be interpreted as implying a change in the been reflected in the domestic legislation and case rights and obligations of the parties under existing law of an increasing number of countries.15 international agreements and that this recital is not The reference in multilateral, regional, and intended to subordinate the protocol to other inter- national legal texts to the precautionary approach national agreements. This provision is, however, does not make it less controversial in the context rather unclear and may prove not to be very helpful of international trade. Although it has been included if a trade conflict arises between countries with in a number of legal instruments dealing with the divergent interests or perceptions in the area of environment, its status within the framework of the biotechnology. international trading system is still unclear. While some developing countries have em- The SPS Committee discussed the precaution- braced biotechnology and are becoming producers ary approach in several of its meetings in 2000 and and exporters of bioengineered agricultural prod- 2001. While the EU has questioned the soundness ucts, others are highly skeptical and fear that their of Article 5.7 of the SPS Agreement and suggested a lack of scientific knowledge and familiarity with possible revision or broader interpretation of it in biotechnology issues may lead them to risk import- order to give countries more flexibility to protect ing products that, in the long run, may prove dan- their markets against products whose health, safety, gerous for health, safety, or the environment. or environmental impacts are uncertain, several Therefore they are putting in place restrictive trade developed and developing countries have stressed measures regarding the production and import of that Article 5.7 is adequate to deal with cases where GM products. However, if challenged, those coun- emergency measures are needed but related scien- tries may well face many problems in proving the tific evidence is not fully available. According to scientific basis of their trade measures. Until the these countries, a broader application of the pre- Cartagena Protocol enters into force, the SPS cautionary approach in international trade would Agreement will probably remain the most adequate lead to a situation of unpredictability in relation to multilateral legal instrument to deal with trade in market access, which would jeopardize the results genetically modified agricultural products--and as of trade liberalization. mentioned above, its provisions on precautionary The debate about the precautionary approach, measures are quite strict. and the way it is reflected in the multilateral legal Biotechnology is another field where developing instruments, acquires a specific significance when countries would greatly benefit from the strength- the issue is trade in biotechnology products.16 The ening of their scientific and technical capacities. precautionary approach is one of the main features Once a better knowledge of biotechnology is of the Cartagena Protocol on Biosafety.17 It allows acquired, developing countries would be able to importing countries to ban imports because of lack assess the related risks and benefits as applied to of scientific certainty. The ban may last until the them, and move from a situation of fear--and importing country decides that it has arrived at sci- related refusal of GM products--to a situation in The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 231 which they would decide which kinds of bioengi- and could likely result in a more meaningful partic- neered products they may have an interest in ipation of developing countries in such activities. producing and/or importing. Equivalence can facilitate international trade. Considering that it requires full confidence by the importing country in the laboratories and certifying Options authority of the exporting country, the establish- The benefits of trade liberalization in the agricul- ment of internationally financed regional, subre- ture sector achieved by the Uruguay Round negoti- gional, or national laboratories, certification bodies, ations, and those that may result from the Doha and accreditation institutions should be considered. work program, could be undermined by the protec- Special and differential provisions should help tionist use of sanitary and phytosanitary measures. developing countries to adjust their products and The SPS Agreement was negotiated to limit this process methods to the new requirements estab- danger, and represents a useful instrument for this lished by the importing countries, strengthen their purpose. However, this chapter has identified some institutional framework, upgrade their facilities, shortcomings of the agreement. It could thus be and become more capable of dealing with the sci- worth considering the introduction of certain entific and technical aspects of SPS measures as amendments to, or authoritative interpretations of, exporters as well as importers. S&D provisions that the agreement to ensure that the risk of SPS mea- only allow for longer time frames for compliance sures being used as border protection instruments are insufficient and may prove self-defeating in the is minimized, while all countries benefit equally long run. Technical assistance and S&D are, there- from the agreement. The Doha work program and fore, closely linked. the Decision on Implementation-Related Issues Adaptation to regional conditions is of key rele- and Concerns provide countries, especially devel- vance to developing countries. Clear reference oping countries, with a margin of flexibility in intro- should be made in Article 6 of the SPS Agreement duction and authoritative interpretation. to the scientific and administrative support needed Since it is quite obvious that governments are by developing countries to facilitate implementa- not willing to compromise public health, the most tion of the article. The disease-free status declared promising option for developing countries to by the competent international organizations maintain and expand their exports of agricultural should be recognized by all trade partners. and food products is to be able to meet the manda- The transparency provisions could be made more tory requirements and the expectations of con- development-oriented by adding new language in sumers in the target markets. At the same time, Annex B to stress the expectation that comments developing countries need to develop their own provided on the drafts are reflected in the final texts regulatory control systems to ensure the safety of and that, in the event they are not, explanations are their domestic production and of their imports, provided. Electronic access to information should be including those that result from the application of made easier for developing countries. biotechnology. Any possible modification and interpretation of the SPS Agreement should keep Notes these two goals in mind. 1. Simonetta Zarrilli is a legal officer of the Trade Negotiations International standards should be developed and Commercial Diplomacy Branch, Division on International through a fair process, based on consensus, where Trade and Commodities of the UNCTAD Secretariat. The views expressed in this paper are those of the author and do not neces- countries at different levels of development and sarily reflect those of the UNCTAD Secretariat or of its member from different geographical regions are effectively countries. E-mail: Simonetta.Zarrilli@UNCTAD.org. The author represented. To this end, support is needed to wishes to express her thanks to Gretchen Stanton, David Byron, improve developing-country capacity to evaluate and David Wilson for the information and comments provided. Any remaining errors are the author's responsibility. draft standards and formulate national, subre- 2. Irene Musselli holds a UN Fellowship by the United gional, and regional positions. The development of Nations Department of Economic and Social Affairs. standards for products that are of export interest to 3. Approved by Ministers at the 4th WTO Ministerial Con- ference in November 2001. It is separate from the Ministerial developing countries would increase the relevance Declaration that launched the current round of multilateral of international standard-setting activities for them trade negotiations. 232 Agriculture and the WTO 4. Annex A of the SPS Agreement defines risk assessment as and strengthened should further trade liberalization occur in the "the evaluation of the likelihood of entry, establishment, or agricultural and food products sectors. spread of a pest or disease within the territory of an importing 10. Equivalence of regulations is at present taking place Member according to the sanitary or phytosanitary measures among the member states of the European Communities, which might be applied, and of the associated potential biological among those of the North American Free Trade Agreement, and and economic consequences; or the evaluation of the potential between Australia and New Zealand. The states that are parties for adverse effects on human or animal health arising from the to the Southern Common Market (MERCOSUR) have adopted presence of additives, contaminants, toxins, or disease-causing three resolutions establishing the criteria, principles, and scopes organisms in food, beverages, or feedstuffs" (WTO 1999b). for determining the equivalence of control systems in MERCO- 5. The 1979 Standards Code is described as "plurilateral" SUR. The issue is currently being discussed in the context of the because, like other "codes" negotiated during the Tokyo Round Free Trade Areas of the Americas. negotiations of the 1970s, it was an agreement that applied only to 11. The WTO Secretariat has prepared a paper summarizing those countries that chose to sign it, rather than to the whole the positions taken by member countries, especially developing- GATT membership (GATT 1979). Its successor, the TBT Agree- country members, on S&D (see WTO 2001c). ment, applies to all WTO members (WTO 1999b). 12. See CAC 2001a Objective 5 Promoting participation fig- 6. New Zealand's experience on the implementation of ures among the objectives considered to be equally important to equivalence (WTO 2001d) provides various examples of ad hoc the overall achievement of the strategic vision. recognition of SPS measures as equivalent: the acceptance of 13. Codex Committees are organized by host governments, high-temperature forced air as effective fruit fly disinfestation which pay the operating costs of these meetings. The Rules of Pro- treatment for paw-paw, mangoes, and eggplants from the South cedure of the CAC specify that costs of delegates' participation are Pacific; acceptance of a "winter window" for cucurbit imported borne by the governments concerned. There have been several from Australia as equivalent to a postharvest chemical dip; and proposals to improve the participation of developing countries by acceptance of tamperproof official stickers for accompanied holding Codex meetings in developing countries, paid for by the consignments of fresh orchids from Singapore as equivalent to usual host country. The recent meetings of the Codex Committee certificates to verify official inspection of the consignment. on Food Additives serve to illustrate the potential of this option. 7. For example, a determination of equivalence by the Food Additionally, countries may still participate by correspondence. Safety and Inspection Service is a prerequisite for export of meat Although participation by accredited representatives to the parent and poultry to the U.S. At the successful completion of an equiv- organizations allows for greater nominal participation, the techni- alence determination process, basically consisting of document cal nature of the subject matter does not always ensure that the review and an on-site audit, a country becomes eligible to export participation is as effective as it could be. meat and poultry to the U.S. The determination of equivalence 14. For example, the European Communities have stated in a does not imply the stipulation of formal agreements between the communication on adaptation to regional conditions that, parties concerned (WTO 2000a). "When assessing the application of regionalization in an export- 8. The Codex Alimentarius Commission (CAC) has ing country, the primary element to be taken into consideration adopted Guidelines for the Development of Equivalence Agree- is the quality of the service in charge of implementation and ments Regarding Food Import and Export Inspection and Cer- control of the policy. Acceptance of trade when a decision on tification Systems (CAC 1999b). The Codex Committee on regionalization is taken requires full confidence in the certifying Food Import and Export Inspection and Certification Systems competent authorities" (WTO 1998c). (CCFICS) is continuing work on the guidelines in the area of 15. For an in-depth analysis of the precautionary principle in equivalence, mainly related to the judgment of equivalence of regional and national legislation and in case law, see Stilwell sanitary measures associated with food inspection and certifi- (forthcoming). For an analysis of the precautionary principle in cation systems (and also equivalence of technical regulations). international trade, see Ward (1999). The primary intention is to assist countries, and especially 16. The Convention on Biological Diversity defines biotech- developing countries, in the application of the SPS provisions. nology as "any technological application that uses biological The guidelines did not advance in the Codex Step process at the systems, living organisms, or derivatives thereof, to make or recent 10th CCFICS meeting mainly because of ongoing discus- modify products or processes for specific use." The biotechnol- sions in the WTO/SPS Committee on Article 4 of the agree- ogy industry provides products for human health care, indus- ment. In other words, several delegations at the CCFICS meet- trial processing, environmental bioremediation, and food and ing were trying to incorporate the rights and obligations of agriculture. WTO members into the Codex Guidelines. A lot of the debate 17. Negotiated under the auspices of the Convention on Bio- centered on the rights of exporters to obtain from the importers logical Diversity (Rio de Janeiro, 1992) and adopted on 29 Janu- the basis for their risk assessment when trying to establish ary 2000, the Cartagena Protocol provides rules for the safe equivalence. transfer, handling, use, and disposal of "living modified organ- 9. According to the World Bank, the value of agricultural isms" (LMOs), defined by the protocol as "any living organism exports as a proportion of total merchandise exports in Sub- that possesses a novel combination of genetic material obtained Saharan Africa averaged above 25 percent over the period through the use of modern biotechnology" [Article 3(g)]. Its 1980­97. Concerning the destination of such export flows, aim is to address the threats posed by LMOs to biological diver- developed countries still constitute the main recipients of agri- sity, also taking into account risks to human health. The proto- cultural and food export from developing countries (World col will enter into force 90 days after the 50th instrument of rat- Bank 1999). According to UNCTAD estimates, in 1997 devel- ification is received. As of April 2003, 103 countries, including oped-market-economy countries accounted for 54.8 percent of the countries of the European Community, had signed the pro- all food items exported from developing countries and territo- tocol and 48 countries, including those of the European Com- ries (UNCTAD 1997). These figures are likely to be confirmed munity, had ratified it. The Sanitary and Phytosanitary Agreement, Food Safety Policies, and Product Attributes 233 Select Bibliography ------. 1998c. "Review of the SPS Agreement, Submission by the European Communities." 23 November, G/SPS/GEN/ The word processed describes informally reproduced works that 101. Office of the Secretary General, Geneva. may not be commonly available through libraries. ------. 1998d. "Review of the SPS Agreement, Submission by Breckenridge, A. 2002. "Developing an Issue-Based Approach the European Communities." 23 November, G/SPS/GEN/ to Special and Differential Treatment." Paper presented at 101. Office of the Secretary General, Geneva. the Third Meeting of the Integration and Trade Network ------. 1999a. "Recognition of the Concept of Pest- or organized by the Inter-American Development Bank, Disease-Free Area as an International Standard, Guideline, March 19­20. Washington, D.C. or Recommendation, Submission by South Africa." CAC (Codex Alimentarius Commission). 1999a. "Guidelines 2 November 1999, G/SPS/GEN/139. Office of the Secre- for the Development of Equivalence Agreements Regard- tary General, Geneva. ing Food Import and Export Inspection and Certification ------. 1999b. The Legal Texts: The Results of the Uruguay Systems." CAC/GL 34-1999. Codex Alimentarius Commis- Round of Multilateral Trade Negotiations. Cambridge: sion, Rome. Cambridge University Press. ------. 1999b. "Improvement of Procedures for Adoption of ------. 2000a. "Equivalence, Submission from the United Codex Standards and Measures to Facilitate Consensus." States to the Committee on Sanitary and Phytosanitary March 1999, CX/GP 99/5. Codex Alimentarius Commis- Measures." 7 November 2000, G/SPS/GEN/212. Office of sion, Rome. the Secretary General, Geneva. ------. 2001a. "Codex Alimentarius Commission Strategic ------. 2000b. "Report by the Chairman, Summary of the Framework, Appendix II in Report of the Codex Alimen- Discussions of the SPS Committee. 29 November, G/L/ tarius Commission, 24th Session." July 2001, ALINORM 423. Office of the Secretary General, Geneva. 01/41. Codex Alimentarius Commission, Rome. ------. 2000c. "Special and Differential Treatment." ------. 2001b. "Report of The Forty-Ninth (Extraordinary) 9 May, G/SPS/W/105. Office of the Secretary General, Geneva. Session of The Executive Committee of the Codex Ali- ------. 2001a. "Decision on Implementation-Related Issues mentarius Commission." ALINORM 03/3. Codex Alimen- and Concerns." 20 November, WT/MIN(01)/17. Office of tarius Commission, Rome. the Secretary General, Geneva. GATT (General Agreement on Tariffs and Trade). 1979. ------. 2001b. "Decision on the Implementation of Article 4 "Tokyo Round Codes, Agreement on Technical Barriers to of the Agreement on the Application of Sanitary and Phy- Trade." LT/TR/A/5. Office of the WTO Secretary General, tosanitary Measures." 26 October, G/SPS/19. Office of the Geneva. Secretary General, Geneva. ICPM (Interim Commission on Phytosanitary Measures). ------. 2001c. "Second Report by the Chairman, Summary of 1998. "Report, Interim Commission on Phytosanitary Informal Discussion on Equivalence." 21 March, G/L/445. Measures." 3­6 November 1998, ICPM-98/REPORT. FAO Office of the Secretary General, Geneva. Headquarters, Rome. ------. 2001d. "Experience in Recognizing Equivalence of Phy- ------. 1999. "Report, Second Interim Commission on Phy- tosanitary Measures." Submission by New Zealand. 28 Febru- tosanitary Measures." 4­8 October 1999, ICPM-99 ary, G/SPS/GEN/232. Office of the Secretary General, Geneva. REPORT. FAO Headquarters, Rome. ------. 2001e. "Equivalence, Communication from Argen- ------. 2002."Directed Financial Assistance for Standard Set- tina." 15 August, G/SPS/GEN/268. Office of the Secretary ting (Sponsorship of Standards)." ICPM 02/15. Fourth General, Geneva. Session of the ICPM. 11­15 March 2002. FAO Headquar- ------. 2001f. "Decision on the Implementation of Article 4 ters, Rome. of the SPS Agreement (Equivalence)." 26 October, G/SPS/ Stewart, T. P., ed. 1993. The GATT Uruguay Round: A Negotiat- 19. Office of the Secretary General, Geneva. ing History. Deventer: Kluwer Law and Taxation Publishers. ------. 2001g. "Implementation of Special and Differential Stilwell, M. T. Forthcoming. "The Legal Relationship between Treatment Provisions in WTO Agreements and Decisions, the Precautionary Principle and Multilateral Trade Rules." Mandatory and Non-Mandatory Special and Differential United Nations Environment Programme (UNEP), Geneva. Treatment Provisions, Corrigendum." 4 February, WT/ UN (United Nations). 2000. Cartagena Protocol on Biosafety COMTD/W/77/Rev.1/Add.1/Corr.1. Office of the Secre- to the Convention on Biological Diversity. Secretariat of tary General, Geneva. the Convention on Biological Diversity, Montreal. ------. 2001h. "Agencies to Boost Developing Countries' Par- UNCTAD (United Nations Conference on Trade and Devel- ticipation in Setting Food Safety and Related Norms." WTO opment). 2002. "Note on the Work on Special and Differ- Press Release. 12 November, PRESS/ 254. Office of the Secre- ential Treatment." Processed, March. tary General, Geneva. Ward, H. 1999. "Science and Precaution in the Trading Sys- ------. 2001i. "Discussion on Technical Assistance and Coor- tem." Royal Institute of International Affairs and the Inter- dination-Informal meeting of the SPS Committee of national Institute for Sustainable Development, London. 9 July 2001-Report by the Chairman." 16 July, G/SPS/ World Bank. 1999. World Development Indicators 1998­99. GEN/267. Office of the Secretary General, Geneva. World Bank, Washington, D.C. ------. 2001j. "Questionaire on Technical Assistance, Com- WTO (World Trade Organization). 1998a. "WTO Agreements mittee on Sanitary and Phytosanitary Measures, Note by Series: Sanitary and Phytosanitary Measures." Office of the the Secretariat." 15 October, G/SPS/W/113. Office of the Secretary General, Geneva. Secretary General, Geneva. ------. 1998b. "Report of the Appellate Body: Australia-- ------. 2001k. "Note on Developing Country Participation in Measures Affecting importation of Salmon." 20 October, Codex Bodies, Codex Alimentarius Commission." 9 March, WT/DS18/AB/R. Office of the Secretary General, Geneva. G/SPS/GEN/236. Office of the Secretary General, Rome. 234 Agriculture and the WTO ------. 2001l. "Developing Country Participation in IPPC ------. 2002b. "Implementation of Special and Differential Standard-setting--Statistical Summary, Internal Plant Treatment Provisions in WTO Agreements and Decisions, Protection Convention." 8 February, G/SPS/GEN/227. Mandatory and Non-Mandatory Special And Differential Office of the Secretary General, Rome. Treatment Provisions." 4 February,WT/COMTD/W/77/ ------. 2001m. "Standard-setting Activities Related to the Rev.1/Add.1/Corr.1. Office of the Secretary General, SPS Agreement, World Health Organization." 28 February, Geneva. G/SPS/GEN/231. Zarrilli, S. 2000. "Trade in Genetically Modified Organisms ------. 2002a. "Equivalence-Program for Further Work, and Multilateral Negotiations: A New Dilemma for Devel- Decision by the Committee." 21 March, G/SPS/20. Office oping Countries." UNCTAD/DITC/TNCD/1, October 20. of the Secretary General, Geneva. UNCTAD, Geneva. 12 AGRICULTURAL BIOTECHNOLOGY: A PRIMER FOR POLICYMAKERS Donald J. MacKenzie and Morven A. McLean The Promise of Modern tion have historically been due to the expansion of Biotechnology for Developing land under cultivation. However, today most avail- Regions able arable land is already in use and any further expansion would mean incorporating fragile and Even though global food production has outpaced marginal areas that represent an increased risk of population growth over the last four decades and lost biodiversity and environmental degradation. there is a sufficient amount of food produced to adequately nourish the world's population (FAO The Plan of This Chapter 2000), hunger and malnutrition persist. More than one billion people, primarily in rural areas of This chapter discusses the meaning of modern developing countries, live in absolute poverty on biotechnology and the extent of global adoption of less than US$1 per day (World Bank 2000). Solu- biotech crops with their benefits and concerns. The tions to this dilemma cannot rely solely on food chapter outlines different government approaches for redistribution--they must address sustainable food managing the use of biotech crops,both legislative and production and economic growth in the regions regulatory, and the issues that these approaches raise, where these are needed. including regulatory triggers, transparency and public Constraints on existing food production include involvement, labeling, and safety assessment in the the availability of arable land and water for irriga- context of both food and the environment. The trade tion, crop losses due to pests and disease, and the implications of biotech crops are discussed by examin- inherent productivity potential of agricultural ing provisions in agreements on the environment (the crops. In addition, postharvest losses in tropical Cartagena Protocol on Biosafety) and trade agree- areas are higher than elsewhere, owing to fungal ments in the World Trade Organization (WTO). The and insect infestations, and compounded by a lack chapter also details the role of the Codex Alimentarius of appropriate storage facilities. Despite efforts to Commission (CAC). Given the important role of both prevent such crop losses, pests destroy more than the United States (U.S.) and the European Union (EU) half of the world food production. Except for in establishing rules on biosafety products, their trade recent decades, when rising crop yields have been dispute on these products has implications for future the major factor, increases in agricultural produc- trade and is covered in this chapter. 235 236 Agriculture and the WTO It has been argued that modern biotechnology enhanced resistance to diseases caused by fungal has the potential to contribute significantly to sus- and viral pathogens; improved tolerance to abiotic tainable gains in agricultural productivity and to stresses such as drought, high-salt soils, and heavy food security, especially in the developing world. In metal contamination; increased yield potential; the broadest sense, biotechnology has been defined improved nutritional quality or medicinal value; as the use of biological processes for the develop- vehicles for the delivery of oral vaccines and thera- ment of products such as foods, enzymes, drugs, and peutics; and factories for the production of phar- vaccines. In this view, biotechnology is simply a new maceutical proteins or industrial polymers. label for a process that humans have used for thou- As an example of a solution to high-salt soils, a sands of years to ferment foods such as beer, wine, new GM tomato has been developed that is not only bread, and cheese, for the purposes of food preserva- able to grow in 40 percent salt solution, but can also tion. Modern agricultural biotechnology, however, remove salt from the soil, thus combining the poten- generally refers to the application of molecular biol- tial of increased income with environmental biore- ogy, and more specifically, recombinant-DNA mediation.6 Examples of nutraceutical products (deoxyribonucleic acid) technology,1 to the genetic include the so-called golden rice, which has been improvement of crop plants and livestock animals. bioengineered to express pro-vitamin A (beta- Using modern biotechnology, plant breeders are carotene) in the rice endosperm (Ye and others able to introduce specific genetic material taken 2000), and tomatoes modified to contain four times from any species of plant, animal, or microorgan- the normal level of lycopene, a carotenoid pigment ism, or created synthetically within the laboratory, whose dietary intake has been correlated with a into many different species of plants or animals, to reduction in coronary heart disease and some types create so-called transgenic organisms.2 The "first of cancer (Ausich 1997). There are a number of generation" of genetically modified (GM) products examples of food products that are being developed has generally included crop plants with new to act as edible vaccines to combat agents such as the input traits, such as pest resistance or herbicide tol- hepatitis-B virus (Richter and others 2000); the Nor- erance. These have included insect-resistant vari- walk virus (Mason and others 1996) responsible for eties of maize, potato, and cotton, containing a viral gastroenteritis, which comprises about 25 per- toxin-encoding gene from strains of the bacterium cent of the cases of traveler's diarrhea; and the rabies Bacillus thuringiensis; plants with tolerance to virus (Modelska and others 1998). These products broad-spectrum herbicides, such as glyphosate have raised hopes of solving many of the problems (Roundup- tolerant canola, maize, and soybean), associated with the delivery of safe, effective vaccines and glufosinate ammonium-tolerant canola, maize, in developing countries (Mor and Arntzen 1999). rice, and soybean; and plants, such as melon, The application of transgenic animals for use in papaya, potato, and squash, with resistance to dis- human food production has lagged behind applica- ease caused by specific plant viruses. In addition, tions in crop plants, and there are currently no bio- there are a very few products with modified quality engineered animal species that have been approved traits, such as high­oleic acid canola and soybean, for food use. The first product likely to undergo reg- and delayed-softening melon and tomato.3 ulatory assessment in the U.S. is transgenic Atlantic To date there have been about 53 bioengineered salmon that have been genetically engineered to foods reviewed by the U.S. Food and Drug Admin- grow faster.7 Transgenic animals that have the istration (FDA), and about 51 novel foods author- potential to reduce the environmental impact of ized for marketing in Canada by Health Canada.4 A livestock food production are also being developed. smaller subset of products has also been authorized For instance, transgenic pigs express a phytase for commercial use in a number of other countries, enzyme that allows them to break down phytate,8 including Argentina, Australia, China, Japan, the which reduces phosphate contamination of agricul- Russian Federation, South Africa, and the countries tural soils. Bioengineered animals also have many of the EU.5 potential applications in medical research as models It is expected that over the coming years there of human disease, in the production of therapeutic will be a newer "second generation" of GM foods products such as pharmaceutical proteins in milk, proposed for market introduction. These next- and as sources of organ and tissue xenografts for generation products will include plants with interspecies transplants. Agricultural Biotechnology: A Primer for Policymakers 237 Current global trade in products of agricultural For example, between 2000 and 2002, China biotechnology is almost exclusively limited to plant increased its cultivation of insect-resistant Bt cotton biotechnology products. The remainder of this fourfold, from 1.25 million acres to 6.25 million acres chapter focuses on this class of GM products and (see box 12.1). does not address products such as recombinant Globally, the principal transgenic crops are herbi- vaccines or veterinary biologics. cide-tolerant soybean (62 percent of total acreage in 2002), insect- and/or herbicide-tolerant maize (21 percent), insect- and/or herbicide-tolerant cotton Global Adoption of Biotech Crops (12 percent), and herbicide-tolerant canola (5 per- For 2002, the International Service for the Acquisi- cent). Of the global aggregate area of these four crops tion of Agri-biotech Applications (ISAAA) has esti- (677.5 million acres), about 21 percent was used for mated that the global area of transgenic crops is planting transgenic varieties in 2002. Specifically, this 145 million acres--nearly 34-fold increase since represented 51 percent of total soybean acreage, 1996--grown by 5.5 million farmers in 16 countries 20 percent of total cotton acreage, 12 percent of total (James 2002; see figure 12.1). Nearly all of this area is canola acreage, and 9 percent of total maize acreage. accounted for by four countries: the U.S. (66 per- According to ISAAA, there is "cautious opti- cent), Argentina (23 percent), Canada (6 percent), mism that global area and the number of farmers and China (4 percent)--with 12 other countries planting GM crops will continue to grow in 2002." making up the remaining 1 percent. It is noteworthy Key events include the limited commercialization that the four principal countries growing transgenic of insect-resistant Bt cotton in India during 2003 crops are represented by two nations from the and the anticipated resolution of outstanding regu- "North"and two nations from the developing world. latory and legal issues surrounding the planting of Indeed, in recent years the rate of adoption of these herbicide-tolerant soybean in Brazil. In the fall of crops in developing nations (50 percent between 2002, the Philippines authorized the commercial 2000 and 2002) has outstripped that of industrial- planting of Bt maize (MON 810), which repre- ized countries (30 percent between 2000 and 2002). sented a milestone event for Southeast Asia. FIGURE 12.1 Global Area of Transgenic Crops Acres (millions) 150 145.00 130.00 125 40.00 109.20 33.75 98.60 100 26.75 17.75 75 105.00 69.50 11.00 96.25 50 80.85 82.45 28.50 25 5.00 58.50 4.30 1.00 23.50 3.30 0 1996 1997 1998 1999 2000 2001 2002 Source: Clive James 2002. Industrial Developing Source: Clive James 2002. 238 Agriculture and the WTO BOX 12.1 The Global Trade Effects of China Bt Cotton China attaches great importance to the applica- declined. Indiscriminant use of insecticides had tion of biotechnology to agriculture, although resulted in increased bollworm resistance to the recent developments relating to regulations, for- agents, produced significant yield losses, and eign investment in biotechnology, and criticism resulted in cotton becoming unprofitable. of genetically modified (GM) crops on environ- China is a significant importer of cotton and a mental, food safety, and ethical grounds have net exporter of textiles and apparel. Changes in created uncertainty. For the past 10 years in the China's cotton production have the ability to cotton industry, China has focused on reducing affect both global cotton production and trade its dependence on cotton imports and non- in textiles and apparel, which depends on con- Chinese GM cotton technology. The impact of tinued liberalization of the cotton trade. China's implementation of GM cotton goes well Anderson and Yao have quantified the poten- beyond its cotton crop sector to domestic textile tial global economic effects when China adopts production and foreign production of cotton, GM technology. The Global Trade Analysis Pro- apparel, and textiles. ject trade model is used with projections to Since the 1980s, China has invested signifi- 2005 and assumptions relating to China's acces- cantly in biotechnology research to minimize sion to the WTO and completion of the Uruguay dependence on foreign technology. Scientists Round undertakings in textiles and apparel have developed high-yielding, insect- and trade. According to the study, without GM cot- drought-resistant plant varieties of most major ton technology, China was projected to be pro- crops that have the potential to allow farmers to ducing and using around one-quarter of the produce more food on China's limited land area. world's plant fibers. Use of GM technology by The Chinese Cotton Research Institute, an insti- both China and outside of China can be tute of the Chinese Academy of Agriculture Sci- expected to have a large trade impact on China ences (CAAS) in Henan Province, has developed and other cotton-producing countries, including several varieties of Bt cotton and a large propor- other textile and apparel exporting countries. An tion of the planting seed used in Xinjiang. Bt increase in China's domestic GM cotton supply cotton produces its own pesticide through a lowers cotton imports and decreases global cot- gene borrowed from the bacterium Bacillus ton prices, as China's output is significant com- thuringiensis. Although earlier domestic varieties pared with total worldwide production. of Bt cotton were inferior, more recent types However, without China's voluntary export reportedly are quite effective. The industry restraints to the U.S. and EU, world cotton prices remains troubled by a shortage of Bt planting still fall, China's imports of cotton actually seed quality issues, and foreign content in increase, its exports of textiles increase, and this picked cotton. The latter two concerns are further expands the developing country's share affected by procurement price policy and grad- of the world textile market. ing standards, with no incentive to produce Bt cotton appears to have been largely unaf- higher strength cotton. fected by China's new biotech management CAAS estimates Bt cotton acreage during regulations (in 2002 the government published 2001­02 Marketing Year at 1.5 million hectares, regulations on labeling of GM foods which dis- over one third of the cotton planted area. Four rupted trade in soybeans from U.S. and South varieties of Bt cotton are approved for use in a America). The regulations were carefully phrased number of provinces, although anecdotal evi- to exclude biotech fibers, and to not affect cot- dence indicates that Bt cotton is being grown tonseed, which is processed into edible oil, and throughout the cotton-growing regions. Bt cot- into meal that is consumed as feed or added to ton continues to increase in area with the domi- foods. This could result in WTO violation of the nant varieties developed by Monsanto and by national treatment principle. There is also little CAAS. evidence that domestically-developed biotech Bt cotton played an important role on the cotton planting seeds are not being held to the return of production in the provinces of the Yel- same testing and approval standards set by the low River and Yangtze River, where acreage had government. Agricultural Biotechnology: A Primer for Policymakers 239 BOX 12.1 (Continued) The government now seems to be taking a development of the cotton industry by prohibit- more cautious approach to biotechnology. ing foreign investment in biotechnology. This Despite the importance of GM cotton in its cot- change of rules has placed existing biotech joint ton production, China recently limited the ventures in jeopardy. Sources: Anderson and Yao 2001. Benefits and Concerns pesticides for other agricultural applications. There are also fears that outcrossing between herbicide- As with any new technology, the potential benefits tolerant crop plants and closely related weeds could of applying modern biotechnology and genetic result in the creation of "superweeds," or in the engineering to food production are balanced by movement of engineered traits into neighboring concerns about potential adverse effects. All farm- crops. Finally, there are concerns that animals and ers, regardless of their economic status, are discern- insects consuming transgenic plants will be harmed ing judges of crop performance and the economic and that biodiversity will be diminished. advantage that may or may not be offered by new For developing countries, there are additional varieties. Viewed in this light, the rapid adoption of concerns that existing intellectual property rules will transgenic crops, both in developed and developing preclude access to the products of biotechnology regions, is testimony to the direct economic bene- research by resource-poor farmers, and that the new fits to farmers. Based on data from 1999, it has been technology will lead to exploitation and monopoly estimated that the global economic advantage to control of seeds with novel and desirable traits. farmers planting transgenic crops is about US$710 If biotechnology is to contribute to rural devel- million (James 2000). Transgenic soybean, cotton, opment and the achievement of food security and canola accounted for 61 percent, 35 percent, within developing countries, there must be and 9 percent of the net economic benefit, respec- increased emphasis on improving indigenous crops tively. The anomaly was the cultivation of Bt maize and agricultural practices, development of policies in the U.S., which represented a US$35 million loss and programs to foster biotechnology develop- to farmers in 1999.9 Countering these apparent ments as an adjunct to other agricultural practices, benefits to producers and technology providers are and establishment of adequate regulatory or over- criticisms that consumers have yet to realize any sight mechanisms to enable safe introduction of direct benefits. new technologies and products. There are striking differences of opinion regard- ing the environmental risks posed by GM crops. Some argue that insect- and disease-resistant GM Governance and Regulatory crops will reduce reliance on damaging pesticides, Oversight thus contributing to environmental sustainability, and could result in fewer cases of pesticide-related The potential for adverse environmental and farm worker illness or injury and less groundwater human health consequences arising from the intro- contamination. On the other hand, there are many duction of GM plants into agriculture and food concerns about the impacts of GM crops on both production has led to the development of regulatory agricultural and natural ecosystems, particularly in regimes that are specifically applied to assessing the the long term. These include concerns that plants safety of these products and controlling their intro- producing their own pesticides to combat insects duction into commerce. Based on the experiences of could accelerate the development of resistant insect those industrialized nations with established regula- populations, thus reducing the effectiveness of these tory systems, there is no consensus on a single best 240 Agriculture and the WTO approach. There are no examples of existing to review new dossiers and publish their opinions, biotechnology regulatory systems that were devel- which are generally consistent with scientific assess- oped anew from a comprehensive plan designed ments from other countries. from the outset to anticipate every contingency and to be integrated and coherent, both internally and Legislative Basis with other national policies. Overall, these systems were developed in piecemeal fashion, usually begin- Experiences from different countries have shown ning with voluntary guidelines and standards devel- that effective regulatory frameworks can be based oped cooperatively by academia, industry, and on nonstatutory guidelines,10 statutory regulations government, and ending up incorporated over time entrenched in existing or new legislation, or some into statutory instruments, either under existing combination of these approaches. Generally within legislation covering food and agricultural products, industrialized nations, the regulation of biotech- or under new legislation dealing specifically with nology was initiated as a system of voluntary infor- gene technology. mation guidelines, codes of practice, and risk Some countries have taken dramatically differ- assessment criteria. The competent authorities11 ent approaches to regulating GM products, which developed information guidelines, and technology can be particularly illustrated by comparing the developers abide by these. As examples, Argentina, U.S. and the EU. The U.S. has adopted an optimistic Australia, Canada, Japan, and South Africa have all mindset and approaches the evaluation of new used nonstatutory guidelines to manage the envi- products and technologies by asking "Why not?" ronmental safety of GM plants before promulgat- whereas the EU approach is more pessimistic and ing new acts or regulations. involves trying to predict the unknown and asking The FDA provides an example of a regulatory "Why?" Not surprisingly, these differences in philos- management system for bioengineered foods that ophy have been translated into unique regulatory sys- to date remains voluntary. The cornerstone of the tems with significant differences in intent (product FDA's 1992 policy for foods derived from new plant versus process), legislative basis (existing versus varieties is that foods produced through the appli- technology-specific laws), and location of decision- cation of genetic engineering techniques are not making authority (institutional versus political). inherently riskier than foods produced through Notwithstanding the divergence of philosophi- more conventional means (FDA 1992). Since pub- cal and/or political approach, at the scientific and lishing this policy, the FDA has conducted 53 vol- technical level each country asks very similar ques- untary (FDA 1997) consultations (FDA 2001) with tions when evaluating the potential environmental companies about the safety and composition of and human health risks of GM products. For exam- new bioengineered food products. ple, in evaluating novel foods each country takes a The benefits of implementing voluntary guide- comparative approach that focuses on the defined lines include the speed with which the guidelines can differences between the novel food and its tradi- be put in place and the flexibility they can afford. tional counterpart, and the effect these differences However, in the absence of a statutory instrument, have on composition, nutritional quality, toxicol- there is a limited capacity for independent, legally ogy, and potential allergenicity. While it may not enforceable auditing and monitoring of compliance. always be reflected in the regulatory decisionmak- Depending on the discretionary power of the com- ing process, the rate of approvals, or the ultimate petent authority, there may be no legal basis for the decision to allow marketing, there is a high degree imposition of penalties or other action in the event of consistency in the scientific opinion in different of noncompliance, or opportunities for the public to countries about the safety of novel foods that have seek redress through the courts should negligence be been put forward for commercialization to date. suspected. As important, the public may not have Even in the EU, where there has been a moratorium confidence that the government is adequately regu- on the approval of new GM products since 1998, the lating these products, or that developers are abiding European Commission's scientific committees (for by voluntary guidelines. In part because of political example, the Scientific Committee on Plants and and public pressures to do so, both Australia (OGTR the Scientific Committee on Foods) have continued 2001) and South Africa12 have implemented new Agricultural Biotechnology: A Primer for Policymakers 241 acts to specifically regulate gene technology and tory decisionmakers. Public involvement, on the genetically modified organisms (GMOs). In the U.S., other hand, refers to the extent to which the public the FDA has proposed a new rule requiring that all has input either into the formulation of regulatory new foods derived from biotechnology be subject to policy or into specific regulatory decisions. mandatory review prior to marketing. This pro- Although closely related, public information and posed new rule is subject to a public comment participation are not synonymous, as it is certainly period (that is, the public is given time to comment possible to have an open and transparent process on the proposed rule). that does not involve public input. The degrees of transparency and public involve- ment vary greatly among countries. As a mini- Regulatory Triggers mum, some countries publish information on the With the exception of Canada,13 some form of criteria for risk assessment and risk management "process" trigger is the rule in all countries that so that developers, stakeholders, and the public can have developed regulatory systems for GM foods. be confident that the biosafety system is both cred- This is also the case for the United Nations Carta- ible and predictable. Other jurisdictions have gena Protocol on Biosafety, which focuses specifi- improved on this and additionally notify the pub- cally on living modified organisms (LMOs), lic both when applications for the environmental defined as "any living organism that possesses a safety assessment of a GMO are received by the novel combination of genetic material obtained competent authorities and when a regulatory deci- through the use of modern biotechnology." Very sion is made. clearly, the protocol is limited to addressing Public participation may be sought at a number biosafety concerns that may be associated with the of levels throughout the development and imple- products of modern biotechnology, irrespective of mentation of a regulatory system. These include: the trait or traits that an LMO may express. representations to, or membership on, advisory While a "product-based" trigger for regulatory committees, particularly those tasked with evaluat- oversight is truest to the scientific principle that ing social, ethical, and economic dimensions of biotechnology is not inherently riskier than other biosafety; input at public hearings during the technologies that have a long and accepted history development of policy or regulation; and com- of application in agriculture and food production, ments during the risk assessment process. As exem- it is more difficult to implement than a "process- plified by recent proposed changes by the FDA and based" approach. In the absence of some form of the European Commission, the trend is toward ex ante risk assessment, it is extremely difficult to increasing openness and public involvement. design a regulatory system that captures the prod- ucts of concern without placing an undue burden Labeling Policies for GM Foods on less risky products. Considering that it is the rel- atively new process of genetic engineering that Notwithstanding the considerable scientific evi- causes the greatest public concern, regulatory over- dence supporting the safety of GM foods currently sight triggered by the application of this technology in the marketplace, there exists a level of consumer remains the most practical approach. concern about these products.14 For different peo- ple, these concerns may be precipitated by various factors, such as a lack of information about how Transparency and Public Involvement these products are evaluated for safety, a lack of Within the context of government regulatory sys- knowledge and/or trust in regulatory processes, or tems, transparency refers to the extent to which a philosophical opposition to the genetic engineer- governments provide information on why and how ing of food crops. certain products are regulated, and how risk assess- Provisions for the mandatory or voluntary label- ments are performed and decisions made as well as ing of GM foods have been perceived as the key the conclusions and decisions that have been means of providing consumers with the informa- reached. Transparency can also involve the per- tion they need to make choices in the marketplace. ceived independence and objectivity of the regula- Some countries, such as Canada and the U.S., have 242 Agriculture and the WTO taken a permissive approach and have not imposed developing a standard for the voluntary labeling of a regulatory distinction between GM and non-GM GM foods. foods when labeling for food safety. That is to say, Other countries, such as Australia and New there is a requirement for additional labeling only Zealand, the EU countries, Japan, the Republic in cases of significantly altered nutritional compo- of Korea, and Thailand have announced, or are sition, or potential for increased toxicity or aller- beginning to implement, mandatory labeling genicity. In January 2001, the FDA announced its requirements for GM foods that contain detectable draft guidance to industry on "Voluntary Labeling amounts of novel DNA or protein (that is, GM con- Indicating Whether Foods Have or Have Not Been tent). While these requirements or schemes are sim- Developed Using Bioengineering."15 The purpose ilar in intent, there are notable differences with of this guidance is to advise food manufacturers on respect to which foods must be labeled (for exam- the acceptability of various types of label claims ple, whole foods versus processed foods versus and to reinforce the fact that labeling must be restaurant foods), exemption thresholds for food truthful and not misleading. Of particular impor- additives that may have been derived from GM tance, the FDA reiterates the requirement that foods, and thresholds for the adventitious presence manufacturers must be able to substantiate label of GM foods in a processed product (see table 12.1). claims, and that in labeling a food as "not a product The effective implementation of labeling provi- of bioengineering" there should be no suggestion of sions, whether mandatory or voluntary, is severely superiority. In Canada, the Canadian Council of limited by the lack of objective, internationally Grocery Distributors (CCGD)16 and the Canadian accepted standards for verifying the presence or General Standards Board (CGSB)17 are currently absence of a GM food in a food product. Without TABLE 12.1 Labeling Requirements for Genetically Modified Foods Country/ Mandatory/ Jurisdiction Voluntary (M/V) Coverage Australia and M GM content in processed foods, fruits, vegetables; does New Zealand not cover whole foods that do not otherwise carry labels; 1 percent tolerance for adventitious presence. Canada V Mandatory labeling only for health and safety concerns, such as significantly altered nutritional composition or increased risk of toxicity or allergenicity; voluntary standard being developed. EU M GM content; 0.5 percent tolerance for adventitious presence and 0.9 percent threshold to trigger labeling; applicable to whole foods and to processed foods. Japan M GM content; labeling of food ingredients only if the ingredient is one of the top three by weight and comprises at least 5 percent of the total weight of the product; no labeling on packages smaller than 30 cm; labeling of bulk shipments with more than 5 percent GM content ("may contain"); no labeling of bulk shipments if less than 1 percent GM content. Korea, Rep. of M Processed foods with GM maize, soybean, or bean sprouts (and potatoes in 2002); if one of top five ingredients; 3 percent tolerance. Thailand M GM content in all foods and raw products; 3 percent or 5 percent tolerance. U.S. V Mandatory labeling only for health and safety concerns, such as significantly altered nutritional composition or increased risk of toxicity or allergenicity. Source: Adapted from Phillips and McNeill 2000. Agricultural Biotechnology: A Primer for Policymakers 243 adequate means of verification, the requirements that should be considered when evaluating a novel imposed by any labeling protocol are difficult to food (OECD 2000). They include the following: enforce. The lack of standardization and verification methods has been a chief criticism of the EU's · A description of the host organism that has been mandatory labeling of GM foods, which is viewed modified, including information on nutrient by some as arbitrary and trade-protectionist. composition, known antinutrients, toxicants and allergenic potential, and any significant changes in these that may result from normal Safety Assessment processing. Central to all biotechnology regulatory systems is a · A description of the donor organism, including framework for risk assessment that evaluates the any known associated toxicities and allergenici- characteristics of the organism, the introduced ties, and the introduced gene(s). trait, the environment into which the organism is · Molecular characterization of the genetic modifi- introduced, the interaction between these, and the cation, including a description of the modifica- intended application. tion process and the stability of the introduced trait. Food Safety Assessment and Substantial Equiva- · Identification of the primary and secondary gene lence To date, the safety assessment of genetically products, including a description of the charac- engineered and novel foods has been based on the teristics of the inserted gene. principle that these products can be compared with · Evaluation of the safety of expected novel sub- traditional foods that have an established history of stances in the food, including an evaluation of safe use (OECD 1993). Furthermore, this compari- any toxins produced directly by the modification. son can be based on an examination of the same · Assessment of the novel food's potential aller- risk factors (that is, hazards) that have been estab- genicity. lished for the traditional counterpart. Keeping in · Evaluation of unintended effects on food compo- mind that many conventional foods that are gener- sition, including (a) assessment of changes in ally considered safe may present specific risks the concentration of nutrients or naturally depending on conditions of processing, or to indi- occurring toxicants, (b) identification of antinu- viduals within a population, the intent is not to trient compounds that are significantly altered establish absolute safety but rather whether the new in novel foods, and (c) evaluation of the safety food can be said to be "as safe as" its conventional of compounds that show a significantly altered counterpart. concentration. This comparative approach has been embodied in the concept of substantial equivalence, which has In evaluating these safety issues, due consideration been both endorsed as a useful risk assessment tool is given to the processed version of the food (if the (FAO/WHO 1996) and criticized (Royal Society of food normally undergoes manufacturing or pro- Canada 2001), particularly for being subjective, cessing), as well as food consumption issues includ- inconsistent, and "pseudo-scientific" (Millstone, ing (a) identification of the potential human Brunner, and Mayer 1999). Even with its limita- population consuming the genetically engineered tions, the substantial equivalence approach remains foods and the amount they are expected to con- the most practical approach to addressing the sume, and (b) assessment of any effects that may safety of foods and food ingredients developed occur if intake of the modified food differs from through modern biotechnology. In fact, there are at intake of its conventional counterpart. present no alternative strategies providing a better assurance of safety. Environmental Safety Assessing the environ- International discussion and expert consultations mental safety of a transgenic plant requires famil- among Organisation for Economic Co-operation iarity with the biology of the plant itself and the and Development (OECD) countries, and within the agricultural practices employed in its cultivation. United Nations in the CAC of the FAO/WHO, have This concept of familiarity is a key approach used in resulted in a consensus on the specific safety issues identifying and evaluating environmental risks and 244 Agriculture and the WTO also in informing practices that may be needed to In addition to evaluating the risks associated manage recognized risks. For example, knowledge with the introduction of new transgenic varieties, about the biology of the plant can identify species- consideration must be given to the risks associated specific characteristics that may be affected by the with not using biotechnology to achieve desired introduced trait and permit the transgenic plant to goals. For example, the biodiversity of tropical rain- become "weedy," invasive of natural habitats, or be forests can be maintained only if these natural otherwise harmful to the environment. ecosystems are not destroyed by the expansion of Likewise, the introduction of a new trait may the agricultural land base. result in changed agricultural practices that affect the environment. The widespread cultivation of Trade Implications herbicide-tolerant rapeseed (Brassica napus, also known as canola) in the Canadian prairies has led The introduction of crop biotechnology products has farmers to switch to "no-till" cultivation. Farmers had a profound effect on international trade in agri- are able to seed herbicide-tolerant plants directly cultural commodities. A dominating force behind into the stubble of the previously harvested crop changing agricultural trade policies has been con- with no prior cultivation. This provides for both sumers' opinions, primarily in the EU, on the safety soil conservation (topsoil is held in place by the or necessity of GM food products. residue of the previous crop) and water conserva- European fears over the introduction of GM tion (the stubble cover allows for better water crops and foods have been widely ascribed to a retention and inhibits evaporation). generalized lack of confidence in science and regu- Transgenic plants expressing stress tolerance latory systems, largely as a result of food safety genes are much sought after, particularly for pro- crises such as bovine spongiform encephalopathy duction in soils damaged by salinization or alkalin- (BSE--better known as "mad cow disease") and ization, or in environments where water is the dioxin-tainted Belgian beef. In the BSE crisis, the limiting factor for food production. Stress toler- attempts by U.K. government officials in 1990 to ance may also be used to extend the typical zones of mollify initial public concern, which subsequently production of a crop plant. For example, cold toler- backfired, seriously eroded public confidence in ance is considered a desirable trait as it can be used the role of government in consumer protection. to limit frost damage to crops and consequently These issues, while not directly related to GM extend production seasons. Cold tolerance may foods, have contributed to the belief among con- also permit the introduction of novel plants into sumers that the whole story on GM foods is rarely areas where they have not previously been grown. told, and if it is, the risks are downplayed and the This approximates the introduction of an exotic benefits accentuated. species and therefore necessitates the same close These considerations, coupled with a highly examination for potential ecosystem disruption. organized campaign against GM products led by Other ecological risks that must be assessed European-based activist organizations, contributed include the potential spread of introduced traits to to the 1998 EU moratorium on new GM crop related plant species through outcrossing, the approvals, introduction of new regulations govern- potential build-up of resistance in insect popula- ing labeling and traceability of GM products, and tions to engineered insecticidal traits, unintended the ascendancy of the Precautionary Principle secondary effects on nontarget organisms, and within the risk analysis process for GM products.18 potential effects on biodiversity. In this regard, it is The Precautionary Principle has been expressed and important to distinguish between the biodiversity of implemented by different countries in forms rang- natural populations and that of crops and other ing from "weak" (that is, lack of full certainty is not organisms within the agro-ecosystem. It is the latter a justification for preventing an action that might be context that is most relevant within the environmen- harmful) to "strong" (that is, no action should be tal risk assessment process for transgenic plants, taken unless one is certain it will do no harm). It is which is designed to evaluate the incremental risks when implemented in the "strong" form, as in cases associated with replacing a conventional crop vari- where there is neither credible theoretical nor ety with a transgenic one. empirical evidence establishing the possibility of Agricultural Biotechnology: A Primer for Policymakers 245 harm, that the Precautionary Principle has received pean and Japanese consumer preferences for GM- its greatest criticism as a disguised trade barrier. free soybeans. The lack of European acceptance of GM prod- The following discussion presents a brief overview ucts affects not only the major growers of GM of the Cartagena Protocol on Biosafety, the only crops (U.S., Argentina, and Canada) but develop- international agreement to directly address interna- ing nations as well, which may be reticent to adopt tional trade of GM seeds and plants, and longer- the new technology for fear of losing export mar- established trade agreements under the WTO. This kets. Indeed, it could be argued that some coun- discussion attempts to highlight the relationship tries, such as Brazil, maintained an official between these agreements, particularly points of con- "GM-free" posture in order to capitalize on Euro- flict. (Additional resources are detailed in table 12.2). TABLE 12.2 Online Resources Resource Description Cartagena Protocol on Biosafety Home page for the Biosafety Protocol and gateway (Convention on Biological Diversity) to the Biosafety Clearing House mechanism. 393 Saint Jacques Street, Suite 300 Montreal, Québec, Canada H2Y 1N9 Tel: +1 (514) 288-2220 Fax: +1 (514) 288-6588 E-mail: secretariat@biodiv.org URL: www.biodiv.org Codex Alimentarius Commission International standards development with respect Secretariat of the Joint FAO/WHO Food to food safety, including the risk assessment of Standards Program foods derived from modern biotechnology. Food and Agriculture Organization of the United Nations Viale delle Terme di Caracalla 00100 Rome, Italy Email: codex@fao.org URLs: www.fao.org/ag/cgrfa/ www.codexalimentarius.net Essential Biosafety Information has been classified into three discrete Essential Biosafety Press LLC areas that include a database of safety information PO Box 475 on all GM plant products that have received Merrickville, Ontario regulatory approval; training tools in the form K0G 1N0 Canada of case studies for food and environmental risk Email: info@essentialbiosafety.info assessment; and a library of pertinent biosafety URL: http://essentialbiosafety.info references and online documents. Food and Agriculture Organization (FAO) Biotechnology in food and agriculture. Gateway to of the United Nations FAO documents and activities related to agricultural URL: http://fao.org biotechnology. Organisation for Economic Co-operation and The main focus of the work program is on Development (OECD) international harmonization of regulatory oversight OECD Environment Directorate in biotechnology that will ensure that environmental Environment, Health, and Safety Division health and safety aspects are properly evaluated, 2 André-Pascal 75775 Paris Cedex 16, France while avoiding nontariff trade barriers to products FAX: ** 33 1 45 24 16 75 of the technology. E-mail: ehscont@ OECD.org URL: http://www.oecd.org/EN/home/0,, EN-home-27-nodirectorate-no-no--27,00.html Note: This listing is not meant to be all-inclusive but rather to serve as a starting point. 246 Agriculture and the WTO The Cartagena Protocol on Biosafety With its emphasis on the transboundary move- ment of LMOs, the protocol intersects directly with Internationally, the need to ensure biosafety earlier (1995) international trade agreements made through national systems of risk assessment was under the WTO. The specific agreements with sig- recognized as a priority within the 1992 Conven- nificant overlap and potential for conflict with the tion on Biological Diversity, signed in Rio de protocol include the Sanitary and Phytosanitary Janeiro at the UN Conference on Environment and (SPS) Agreement, intended to protect animal, Development, and more explicitly within Chapter plant, and human health; the Technical Barriers to 16 of Agenda 21. Agenda 21 is a blueprint for sus- Trade (TBT) Agreement, which sets out technical tainable development in the 21st century, and regulations and conformity procedures such as Chapter 16 requires governments to consider inter- labeling; and the Trade Related Intellectual Prop- national cooperation on the "Environmentally erty (TRIPS) Agreement, which defines basic intel- Sound Management of Biotechnology" and to act lectual property standards internationally. appropriately to ensure that developing countries Guidance as to the nature of the relationship have effective participation in biotechnology between the protocol and trade agreements under research activities as well as priority access to the the WTO is included only in the preamble to the results and benefits of biotechnology on a fair and protocol. Among other things, the preamble: equitable basis (see table 12.3). One result of the Convention has been the Carta- · Recognizes that trade and environment agree- gena Protocol on Biosafety, which was adopted in ments should be mutually supportive. Montreal on January 29, 2000, and addresses the · Emphasizes that the protocol does not change safe transfer, handling, and use of LMOs.19 The pro- rights and obligations under existing agreements. tocol, which has been signed by 108 countries and · Understands that the protocol is not subordinate ratified by 49, requires the ratification of 50 govern- to other international agreements. ments before coming into force. On June 13, 2003, the small Pacific nation of Palau became the fiftieth The "soft" language of the preamble has left the country to ratify the Protocol, triggering its imple- relationship between the WTO and the Biosafety mentation on September 11, 2003. Protocol ambiguous and not likely to be resolved TABLE 12.3 The Cartagena Protocol on Biosafety What It Does: What It Does Not Do: Establishes an Internet-based "Biosafety Clearing Does not address food safety issues; these are House" for information exchange. addressed in other international fora. Advance Informed Agreement (AIA) procedure Does not subject shipments of bulk commodities requires exporters to seek consent of importers destined for direct processing or food prior to shipments of LMOs destined for use to the AIA procedure. environmental release. Does not require segregation of bulk Mandates bulk shipments of living modified commodities that may contain LMOs and does organisms (LMOs) destined for food processing not require detailed identification of bulk to be labeled as "may contain" LMOs and "not commodity shipments (these will be subject to intended for intentional introduction into the further negotiation within two years of the environment." protocol's coming into force). Includes a "savings clause" stipulating that Does not require consumer product labeling. existing rights and obligations under the WTO or other international agreements are not altered. Assists developing countries in building capacity for managing modern biotechnology. Source: http://usinfo.state.gov/topical/global/biotech/00021601.htm. Agricultural Biotechnology: A Primer for Policymakers 247 until a measure is taken under it. The preamble The protocol provides considerable flexibility in asserts that the protocol would not alter coun- terms of how importing countries may meet their tries' existing international rights and obliga- obligations with respect to risk management decision- tions, but the extent to which other agreements making and the implementation of these decisions. will preempt the Protocol has not been clarified For example, under the Biosafety Protocol AIA, an (Swenarchuk 2000). importing country can request that the exporter A number of the protocol provisions require an conduct the risk assessment of an LMO or that the action by the Parties that could have consequences exporter pay for the cost of the assessment. While for trade. These include the advanced informed this appears to be an equitable approach to address- agreement (AIA) procedure to export GM prod- ing the resource constraints (financial, scientific, ucts, risk management, confidential information, and infrastructural) of developing-country illegal transboundary movements, and socioeco- exporters, it may be that those asked to bear the nomic considerations in risk assessment (WTO costs associated with risk assessments will choose to 2000). Dispute about a national measure taken forgo the smaller markets, further constraining pursuant to the Biosafety Protocol could possibly access to the technology in countries that may ben- be resolved under the WTO, or under the dispute efit the most from it. This is particularly true in that procedure of the Convention on Biological Diver- the development of international standards for reg- sity (CBD) where parties must consent prior to ulatory harmonization around GM crops and going to arbitration, or by submission to an Inter- foods (for instance, by the OECD or the CAC) national Court of Justice (Cors 1999). Ongoing tends to be driven by industrialized countries, and discussions and analysis of the relationships is resulting in ever more complex and expensive between the WTO and multilateral environmental information and data requirements. These evolving agreements, such as the CBD, are assisted by the criteria, which do not necessarily ensure a higher United Nations Environment Programme (UNEP) safety standard, may be practically unattainable for to make trade and environment policies mutually resource-poor countries and will affect their ability supportive (WTO 2001). to import potentially useful products or technology The Biosafety Protocol entrenches a precaution- as well as their opportunities to access desirable ary approach in the risk assessment process as it export markets with their own, domestically pro- states that "lack of scientific certainty due to insuf- duced GM products. ficient relevant scientific information and knowl- edge" should not prevent countries from taking WTO: The SPS and TBT Agreements precautionary import actions regarding LMOs. The related provision in the SPS Agreement is less Because they were negotiated prior to the commer- restrictive as it emphasizes the provisional nature of cialization of any GM plants or food commodities, such measures. the SPS Agreement and the TBT Agreement contain Although most developing countries are very no provisions that are specific to these products. supportive of the Biosafety Protocol, its implemen- The SPS Agreement sets out the basic rules for tation will pose both scientific and economic chal- food safety and animal and plant health standards. lenges to poorer countries. Parties to the protocol Although it allows individual countries to set their must develop or have access to"the necessary capac- own standards, it requires that regulations be based ities to act on and respond to their rights and obli- on science, and that they be applied only to the gations." National, regional, and international extent necessary to protect human, animal, or plant agencies have recognized that successful implemen- life and health.21 Under the agreement, WTO Mem- tation of the protocol is contingent on the develop- bers are encouraged to adopt international stan- ment of national biosafety capacity in countries that dards (where they exist), but may define even have yet to establish, or are in the process of estab- higher standards provided these are based on a lishing, biosafety systems. Financial aid is being pro- sound scientific risk assessment and do not dis- vided to parties, most notably by the UNEP/Global criminate against imports. Recognizing that a com- Environment Facility (GEF) global project on the plete risk assessment may not be possible in the development of National Biosafety Frameworks.20 short term because of scientific uncertainty or the 248 Agriculture and the WTO lack of sufficient evidence, Article 5.7 of the SPS varieties.22 Countries have many options to fulfill Agreement allows member countries to adopt tem- this requirement. The U.S. and EU extend patent porary restrictive measures. In such cases, members protection to plants, while most other countries are expected to seek the additional information have implemented a sui generis system, such as plant required to complete a full risk assessment within a breeders' rights. In contrast, the CBD establishes the reasonable period. Maintaining restrictive measures rights of indigenous communities to genetic and indeterminately in the absence of scientific evidence biological resources that are held in common, a con- of risk solely for "precautionary" reasons is not cept that is largely ignored under intellectual prop- allowed. The SPS Agreement would apply to regula- erty systems. tions to protect the environment and biodiversity Market access and benefit-sharing are the two against introductions of harmful alien species and main crosscutting issues between the CBD and the LMOs via trade, pursuant to Articles 8(g) and 8(h) TRIPS Agreement. It is not clear if or how these of the CBD. issues will impact on the Biosafety Protocol. The TBT Agreement is intended to ensure that WTO Members do not use technical regulations FAO/WHO: Codex Alimentarius Commission and standards as disguised measures to protect domestic industries from foreign competition. In The CAC was created in 1962 by the Rome-based international trade law, health and environmental FAO and the Geneva-based WHO in order to standards and regulations, labeling, symbols, and establish internationally agreed-upon norms, packaging markings can be considered as technical directives, recommendations, or codes of practice barriers to trade. In the agrifood sector, the TBT designed to protect the health of consumers and Agreement applies to all rules other than those ensure that procedures followed in trade of food specifically covered by the SPS Agreement. The products are fair. The CAC comprises representa- TBT Agreement does not permit requirements for tives from 163 countries who meet every two years, the labeling of some products whereby "like prod- in Rome and Geneva alternately, in order to make ucts" remain unlabeled (Article 2.1). For example, final decisions on the adoption of texts. The Codex GM crop commodities that have been assessed and Alimentarius process is based on reaching consen- found to be "substantially equivalent" to their con- sus, and a new standard or norm can only be ventional counterparts would be considered "like adopted after eight stages of consultation. products," and thus would not require specific The importance of the Codex Alimentarius labeling. Critics of the WTO stance argue that "sub- process was heightened significantly when the new stantially equivalent" is an unacceptable outcome SPS Agreement came into force in 1995. Under the of the risk assessment of GM products and assert SPS Agreement, the CAC is responsible for main- that these products are not "like products" for pur- taining international standards relevant to food poses of labeling, and in fact, can be subjected to safety that should be recognized by the WTO, and labeling. Traditionally within the WTO, a consumer for determining whether national measures are suf- desire for a measure such as mandatory labeling ficiently "based on scientific principles" to comply would not be viewed as a legitimate objective with WTO rules. within the context of Article 2.2 of the TBT Agree- The CAC's Committees on Food Labeling, on ment. This article states that regulatory measures Food Import and Export Certification and Inspec- that effectively disrupt trade must achieve a legiti- tion Systems, on General Principles, and a special mate objective, and those regulations may restrict Task Force on Foods Derived from Biotechnology are trade no more than what is necessary to accomplish all currently considering issues relevant to GM foods. the legitimate objective (WTO 1994). Significant disagreements between the U.S. and the EU, particularly in the area of mandatory label- ing of GM foods, threaten to paralyze the CAC's WTO: Agreement on Trade Related Intellectual consensual approach. The U.S. supports mandatory Property Rights labeling of GM foods only in situations of signifi- One component of the TRIPS Agreement, Article cantly altered composition or increased risk of tox- 27:3(b), deals with plant variety protection, setting icity or allergenicity. The EU, on the other hand, a minimum standard for the protection of plant believes that labeling is required to inform con- Agricultural Biotechnology: A Primer for Policymakers 249 sumers regarding processing and production meth- Peru, and Uruguay, and unofficially by a number of ods (PPMs), such as the use of recombinant-DNA others. Should the U.S. and its partner countries win technology to produce new plants. This type of their action under the WTO, it is unclear whether PPM labeling has not been previously permitted by this will have the desired effect. Previous experience, the WTO, but in December 2001, the EU in a pro- such as the 1998 WTO ruling against the EU ban on posal to the WTO in the context of the current hormone-treated beef, has not been encouraging. In negotiations on agriculture proposed mandatory that case, the EU decided to meet its WTO obliga- labeling for PPMs to address "societal values or tions by allowing the U.S. (and other complainants) concerns."23 to retaliate against an equivalent value of EU Advocates of the Precautionary Principle, pri- exports. There has been no change in EU policy, marily from the EU, are pressing for incorporation making this a less-than-satisfying WTO victory. of the principle within a revised Codex Alimentar- Resolution of the growing U.S.­EU trade ius risk analysis framework, within the regulatory impasse has broader impacts, and it may ultimately harmonization work of the OECD, and within have its most profound effects on developing- anticipated revisions of the WTO SPS and TBT country farmers. Notwithstanding the obvious eco- Agreements. nomic benefits that GM crops may represent to farmers in industrialized countries, they can afford Future Developments to do without them. The same may not be true of resource-poor farmers in developing regions, Some of the most immediately visible impacts of whose economic prosperity could be dramatically changing European trade policies affecting GM increased by the adoption of new GM varieties that products are likely to be seen in U.S.­EU agricul- are resistant to pests, disease, and drought. tural trade. In 2001, the value of U.S. agricultural exports to the EU was US$6.3 billion, composed Notes mainly of soybeans, tobacco, and animal feed 1. There are other techniques used in biotechnology, such as (including corn gluten). The size of the reciprocal recombinant-RNA and cell fusion. The first recombinant deoxyri- U.S. market for EU agricultural products was even bonucleic acid (DNA) molecule was created in 1972 by larger, US$7.9 billion, and was accounted for pri- researchers at Stanford University. The group, led by Paul Berg, marily by exports of wine and beer. who received a Nobel Prize for the work, used enzymes found in bacteria called restriction endonucleases to cut DNA from two The EU moratorium on new GM product sources (a bacterium and a virus) and used a different enzymatic approvals and labeling requirements has had a lim- reaction to splice these two foreign pieces of DNA together into a ited negative impact on U.S. corn exports to the EU. functional, hybrid DNA molecule. In 1973, Stanley Cohen, also from Stanford, and Herbert Boyer, from the University of Califor- In 1999, the EU market represented about 4 percent nia at San Francisco, took this work to the next step by transfer- (US$180 million) of total U.S. corn exports. How- ring a recombinant-DNA molecule into a bacterium where it ever, the EU represents a much larger export mar- functioned alongside the bacterium's own genes. In doing so, they created the first "genetically engineered" organism. ket for corn byproducts, such as corn gluten used in 2. A transgenic organism is one into whose genome an iso- livestock animal feed, and accounts for more than lated gene sequence usually, but not always, from another 85 percent of total exports. In July 2001, the EU species has been integrated. In transgenic plants and animals, proposed new rules requiring the labeling of ani- the introduced gene sequence must be transmitted through meiosis to allow its inheritance by the offspring. mal feed products derived from GM plants or seeds 3. "Essential Biosafety Crop Database Synopsis." Available that do not currently require specific labeling. on the Internet at http://www.essentialbiosafety.info/dbase. These new labeling and traceability rules, which are php?action=Synopsis. 4. Canada's regulatory framework for biotechnology prod- viewed by many as clear violations of WTO trade ucts has a wider scope than just transgenic organisms produced rules, are certain to have a much greater impact on via recombinant-DNA technology. For environmental release, U.S.­EU trade. the focus is on plants with novel traits, which are defined as plants that exhibit neither familiarity nor substantial equiva- In May 2003, the U.S., together with Canada and lence that may have been produced by traditional plant breed- Argentina, announced its intent to challenge the ing or genetic engineering. With respect to food products, EU's de facto moratorium on GM foods in the WTO. Health Canada regulates novel foods, which could include any This action is being explicitly supported by a num- food without a history of safe use rather than just bioengi- neered foods. ber of countries, including Australia, Chile, Colom- 5. A comprehensive database of crop biotechnology bia, El Salvador, Honduras, Mexico, New Zealand, products that have received regulatory approval in various 250 Agriculture and the WTO jurisdictions can be accessed at http://www.essentialbiosafety. 18. In relation to biodiversity, the Precautionary Principle is info/dbase.php. listed as Principle 15 of the 1992 Rio Declaration on Environ- 6. High salinity affects about 20 percent of agricultural land ment and Development, which states that "In order to protect overall and about 40 percent of irrigated land in particular. Most the environment, the precautionary approach should be widely crop plants cannot tolerate high concentrations of salt, which applied by States according to their capabilities. Where there are leads to a decline in photosynthesis and an accumulation of threats of serious or irreversible damage, lack of full scientific deleterious metabolites (Zhang and Blumwald 2001). certainty shall not be used as a reason for postponing cost-effec- 7. These transgenic fish produce higher concentrations of tive measures to prevent environmental degradation." growth hormone, causing them to increase their size and weight 19. "Living modified organism" means any living organism up to six times as fast as the conventional salmon. (Their final that possesses a novel combination of genetic material obtained size is equivalent to that of normal Atlantic salmon, but they through the use of modern biotechnology. From the Cartagena achieve that size in a shorter period of time.) Protocol on Biosafety, Article 3 (g). 8. Seeds store phosphorus needed for germination in the 20. The three-year-old UNEP/GEF project is designed to form of phytate, a sugar molecule containing six phosphate assist up to 140 countries to develop their National Biosafety groups. Because phytate also strongly binds iron, calcium, zinc, Frameworks so that they can comply with the Cartagena Proto- and other mineral ions, it acts as an antinutrient in the human col on Biosafety. The project will also promote regional and sub- diet (as well as the diet of livestock animals), making these sub- regional cooperation on biosafety. See http://www.unep.ch/ stances unavailable for uptake. biosafety/. 9. This loss has been attributed to a record low level of infes- 21. For a fuller discussion, see chapter 9 on "Food Safety" of tation of European corn borer in 1999 that resulted in yields that this handbook. were below the economic threshold for cost-effective control. 22. See also chapter 13 of this handbook. Based on historical data, this can be expected to occur in one 23. "EU Paper Calls for WTO Approval for Process-Based year out of four. Labeling Schemes." Inside U.S. Trade, December 14, 2001, p. 13. 10. Voluntary guidelines may include standards for facilities and practices designed to prevent the unintended release of, or inadvertent exposure to, GMOs or recombinant DNA; condi- Select Bibliography tions for ensuring reproductive isolation and site monitoring during the conduct of confined field trials; and standards for risk Anderson, K., and S. Yao. 2001. "USDA Foreign Agriculture assessment that define criteria for conducting environmental or Attaché Reports: China, GMOS, and the World Trade in food safety assessments. Agricultural and Textile Products." Centre for International 11. The competent authority is the body responsible for Economics, Discussion Paper No. 0126, Adelaide, Australia. overseeing the development and implementation of, and compli- Ausich, R. L. 1997. "Commercial Opportunities for Carotenoid ance with, biosafety measures. It may be a government depart- Production by Biotechnology." Pure and Applied Chemistry ment or agency, or a statutory or nonstatutory committee. 69(10): 2169­73. 12. S. R. Moephuli, Registrar, GMO Act, SA. Personal Cors, T. A. 1999. "Biosafety and International Trade: Conflict or communication. Convergence?" Prepared for the Centre for International 13. Canada's regulatory system is unique in that it was not Development at Harvard University. http://www.biotech- designed to focus just on GM crops or GM foods, but rather on info.net/conflict_convergence.html. any new crop or food displaying novel, or unfamiliar, character- FAO (Food and Agriculture Organization). 2000. "Agriculture: istics. Because of this different conceptual approach, Canadian Towards 2015/2030." Technical interim report. FAO, Rome. regulators have adopted unique terminology and refer to plants fao.org/docrep/004/y3557e/y3557e00. htm. with novel traits and novel foods, respectively. FAO/WHO (Food and Agriculture Organization and the World 14. The true level of consumer concern is difficult to Health Organization). 1996. "Biotechnology and Food ascertain precisely. In unprompted surveys, most consumers Safety." Report of a joint Food and Agriculture Organiza- will identify issues such as microbial contamination or the tion/World Health Organization Consultation. FAO/WHO, presence of pesticide residues as their most immediate food Rome. safety concerns, with genetic modification of foods well down ------. 2000. "Safety Aspects of Genetically Modified Foods of the list. In a recent study prepared for the Australia New Plant Origin." Report of a Joint Food and Agriculture Orga- Zealand Food Authority in 1999, it was estimated that the nization/World Health Organization Consultation. underlying consumer concern over genetically modified foods FAO/WHO, Rome. ranged between 0.25 percent and 5 percent, and that 87 per- FDA (U.S. Food and Drug Administration). 1992. "Statement of cent of consumers made their buying decisions based on Policy: Foods Derived from New Plant Varieties." Federal "price, brand, and use-by date." Register 57: 22984­23001. 15. "Voluntary Labeling Indicating Whether Foods Have or ------. 1997. "Guidance on Consultation Procedures: Foods Have Not Been Developed Using Bioengineering." Draft guid- Derived from New Plant Varieties." FDA, Washington D.C. ance, U.S. Food and Drug Administration, Center for Food cfsan.fda.gov/~lrd/consulpr.html. Safety and Nutrition, Washington, D.C. January 17, 2001. ------. 2001. "List of Completed Consultations on Bioengi- Docket Number 00D-1598. Available on the Internet at http:// neered Foods."cfsan.fda.gov/~lrd/biocon.html. www.cfsan.fda.gov/~lrd/../ ~dms/biolabgu.html. James, C. 2000. "Global Status of Commercialized Transgenic 16. The CCGD is an organization representing about 80 Crops: 2000." ISAAA Briefs No. 23. ISAAA, Ithaca. percent of Canada's major food retailers and is involved in pub- ------. 2002. "Global Status of Commercialized Transgenic lic awareness and education activities for biotechnology. Crops: 2002." ISAAA Briefs No. 27: Preview. ISAAA, 17. The CGSB is a standards-development organization Ithaca. in the federal Department of Public Works and Government Mason, H. S., J. M. Ball, J.-J. Shi, X. Jiang, M. K. Estes, and C. J. Services. Arntzen. 1996."Expression of Norwalk Virus Capsid Protein Agricultural Biotechnology: A Primer for Policymakers 251 in Transgenic Tobacco and Potato and Its Oral Immuno- Swenarchuk, M. 2000. "The Cartagena Biosafety Protocol: genicity In Mice." Proceedings of the U.S. National Academy of Opportunities and Limitations. Canadian Centre for Policy Sciences 93(11): 5335­40. Alternatives." Briefing Paper Series: Trade and Investment Millstone, E. P., E. J. Brunner, and S. Mayer. 1999. "Beyond `Sub- 1(3). April 1. stantial Equivalence.'" Nature 401: 525­6. World Bank. 2000. World Development Report 2000/2001: Modelska A., B. Dietzschold, Z. F. Fu, K. Steplewski, D. C. Attacking Poverty. New York: Oxford University Press. Hooper, H. Koprowski, and V. Yusibov. 1998."Immunization WTO (World Trade Organization). 1994."The WTO Agreement against Rabies With Plant-Derived Antigen." Proceeding of on Technical Barriers to Trade (TBT Agreement)." Office of the U.S. National Academy of Sciences 95(5): 2481­85. the Secretary General, Geneva. Mor, T. S., and C. J. Arntzen. 1999. "Pharmaceutical Foodstuffs: ------. 2000. "Response of the Executive Secretary of the Con- Oral Immunization With Transgenic Plants." In A. Altman, vention on Biological Diversity (CBD) to the Chair of the M. Ziv, and S. Izhar, eds., Plant Biotechnology and in Vitro CTE." Communication from the CBD Secretariat. June 28, Biology in the 21st Century. Dordrecht, Netherlands: Kluwer. WT/CTE/W/149. Office of the Secretary General, Geneva. OECD (Organisation for Economic Co-operation and Develop- ------. 2001. "UNEP Meeting on Compliance, Enforcement, ment). 1993. "Safety Evaluation of Foods Derived by Mod- and Dispute Settlement in Multilateral Environmental ern Biotechnology, Concepts and Principles." OECD, Paris. Agreements and the WTO." Communication from UNEP. ------. 2000. "Report of the Task Force for the Safety of Novel 20 July, WT/CTE/W/199. Office of the Secretary General, Foods and Feeds." C(2000)86/ADD1. OECD, Paris. Geneva. OGTR (Office of the Gene Technology Regulator). 2001. "Gene Ye, X., S. Al-Babili, A. Kloti, J. Zhang, P. Lucca, P. Beyer, and I. Technology Act, 2000." http://scaletext.law. gov.au/html/ Potrykus. 2000. "Engineering the Provitamin A (Beta- pasteact/3/3428/top.htm Carotene) Biosynthetic Pathway into (Carotenoid-Free) Rice Phillips, P. W. B., and H. McNeill. 2000. "Labeling for GM Endosperm." Science 287(5451): 303­5. Foods: Theory and Practice." AgBioForum 3(4): 219­24. Zhang, H.-X., and E. Blumwald. 2001. "Transgenic Salt-Tolerant agbioforum.org. Tomato Plants Accumulate Salt in Foliage But Not in Fruit." Richter, L., Y. Thanavala, C. Arntzen, and H. Mason. 2000. "Pro- Nature Biotechnology 19: 765. duction of Hepatitis B Surface Antigen in Transgenic Plants for Oral Immunization. Nature Biotechnology 18: 1167­71. Royal Society of Canada. 2001."Elements of Precaution: Recom- mendations for the Regulation of Food Biotechnology in Canada." Royal Society of Canada, Ottawa. 13 GLOBAL INTELLECTUAL PROPERTY RIGHTS: A NEW FACTOR IN FARMING Geoff Tansey Why Intellectual Property one of the three pillars of the WTO--the others Rights Matter being trade in goods and trade in services. By plac- Since the early 1990s, the international rules affect- ing IPRs in the WTO and making them subject to ing agriculture have been changing as a result of its binding disputes procedure, proponents of a international negotiations ranging across issues of strong IPRs regime have made it possible for non- trade, the environment, food, and agriculture. The compliant WTO Members to face trade sanctions results of this process are only now emerging and are in any area if they fail to live up to its rules. This is likely to affect farmers everywhere. One of the most arguably the main reason why IPRs were put into hotly debated issues in these negotiations has been the WTO instead of the existing body concerned intellectual property rights (IPRs)--concerning with promoting IPRs, the World Intellectual Prop- among other things copyrights, patents, plant breed- erty Organization (WIPO). The TRIPS Agreement ers' rights, and trade secrets. The most controversial also includes rules on domestic enforcement proce- set of rules on IPRs is contained in the Agreement on dures and remedies. Trade-Related Aspects of Intellectual Property In the Uruguay Round negotiations (1986­94) Rights (TRIPS). This is one of the 1995 World Trade TRIPS rules were developed with very little public Organization (WTO) agreements, with which all involvement and introduced into the negotiations WTO Members must comply sooner or later. against strong, but in the end futile, opposition Before the TRIPS Agreement, countries could from developing countries. TRIPS originated from decide whether or not to provide any form of IPRs a small number of major business interests with a in agriculture. Most developing countries did not. handful of corporations and lobbyists responsible As a result of TRIPS, developed and developing for crafting its terms and pushing the agreement, members of WTO must adopt the same, relatively via various developed-country governments (see high, minimum levels of intellectual property pro- Drahos 1995; Drahos and Braithwaite 2002). tection. This removes from developing-country Today IPRs increasingly underpin the new members options that were used historically by knowledge economy: their ownership and control developed countries to adopt varying levels of will affect the distribution of wealth and power, IPRs, according to their needs. The agreement is and the direction of development. Intellectual 253 254 Agriculture and the WTO property rights allow creators and inventors to Other aspects of the TRIPS Agreement that may also exclude others from copying or using their work or affect agriculture--such as geographical indications invention without permission. The IPRs regime plays and trademarks, and other relevant international an important role in private sector-led innovation agreements--are also discussed briefly, as are the and establishing and maintaining market power.1 decisions taken at the fourth Ministerial Conference of the WTO in Doha in November 2001. (See table 13.1 for a list of online resources.) The Plan of This Chapter This chapter outlines the background to this change, Tradeoffs Involved in IPRs the questions for development raised by IPRs, key features of the TRIPS Agreement, and in particular, Overall, the economic effects from stronger IPRs the provisions on patents and plant variety protec- are far from simple, clear, or agreed. They are, as the tion most likely to affect agriculture. These provi- World Bank's World Development Report 1998/99 sions, initially reviewed in 1999, revealed wide argued, "a compromise between preserving the divisions between developed and developing coun- incentive to create knowledge and the desirability tries, and the review process became deadlocked. of disseminating knowledge at little or no cost." In TABLE 13.1 Online Resources Resource Description Convention on Biological Diversity The Secretariat of the Convention on Biological Diversity 393 Saint Jacques Street, Suite 300 works on developing access and benefit guidelines and the Montreal, Québec biosafety protocol, among other issues. Canada H2Y 1N9 Tel: +1 (514) 288-2220 Fax: +1 (514) 288-6588 Email: secretariat@biodiv.org URL: www.biodiv.org/ Commission on Plant Genetic The Commission deals with the International Treaty on Plant Resources for Food and Agriculture Genetic Resources for Food and Agriculture, and its Food and Agriculture Organization secretariat is housed in FAO. (FAO) of the United Nations Viale delle Terme di Caracalla 00100 Rome, Italy URL: www.fao.org/ag/cgrfa/ UNCTAD/ICTSD TRIPS and This partnership is producing a range of resources to help Development Capacity Building developing countries deal with IPR issues, including a policy URL: www.ictsd.org/iprsonline/ issues paper, series of case studies, annotated and updated bibliography on IPRs and sustainable development, and a detailed resource book on all aspects of the TRIPS Agreement. Drafts of these are put up on the Web site as the project progresses, with the final versions expected in mid-2003. It is also developing a portal to resources on IPRs available at www.ictsd.org/iprsonline/. Commission on Intellectual The Commission was set up by the British government Property Rights to look at how intellectual property rights can work better URL: www.iprcommission.org/ for poor people and developing countries. It reported its findings in September 2002. Its report and all background papers are available online. World Trade Organization (WTO) WTO is the only global international organization dealing URL: www.wto.org/ with the rules of trade between nations. At its heart are the WTO agreements, negotiated and signed by the bulk of the world's trading nations and ratified in their parliaments. The goal is to help producers of goods and services, exporters, and importers conduct their business. Global Intellectual Property Rights: A New Factor in Farming 255 TABLE 13.1 (Continued) Resource Description The International Union for the UPOV is an intergovernmental organization based on the Protection of New Varieties of International Convention for the Protection of New Varieties Plants (UPOV) of Plants. It provides guidance on how to implement Address: Boîte Postale (POB) 12, 34, the system of Plant Breeders' Rights. chemin des Colombettes CH-1211 Geneva 20, Switzerland Tel: +41-22 338 91 11 Fax: 41-22 733 03 36 Email: upov.mail@wipo.int URL : www.upov.org World Intellectual Property WIPO is an international organization dedicated to promoting Organization (WIPO) the use and protection of works of the human spirit. It has URL: www.wipo.org/ established an intergovernmental committee on genetic resources, traditional knowledge, and folklore. International Plant Genetic IPGRI is an international research institute with a mandate to Resources Institute (IPGRI) advance the conservation and use of genetic diversity for the Via dei Tre Denari 472/a well-being of present and future generations. It is a Centre of 00057 Maccarese (Fiumicino) the Consultative Group on International Agricultural Research Rome, Italy (CGIAR). It has produced a number of publications dealing Tel: (39) 06 6112.1 with IPRs issues and is continuing to work on these. Fax:(39) 06 61979661 Email: ipgri@cgiar.org URL: www.cgiar.org/ipgri International Seed Federation (ISF) The ISF was created in May 2002 by the merger of the FIS/ASSINSEL Secretariat International Seed Trade Federation (FIS) and the International Chemin du Reposoir 7 Association of Plant Breeders (ASSINSEL). ISF has a number 1260 Nyon, Switzerland of papers about its views on IPRs available online. Tel: +41 22 365 44 20 Fax: +41 22 365 44 21 Email: fis@worldseed.org or assinsel@worldseed.org URL: worldseed.org/ intellectual_property.htm Genetic Resources Action GRAIN is an international nongovernmental organization that International (GRAIN) promotes the sustainable management and use of agricultural Girona 25, pral. E-08010 biodiversity based on people's control over genetic resources Barcelona, Spain and local knowledge. It produces BIO-IPR, an irregular Tel: +34 933011381 list serve to circulate information about recent developments Fax:+34 933011627 in the field of intellectual property rights related to Email: grain@grain.org biodiversity and associated knowledge. BIO-IPR is a strictly URL: www.grain.org/ noncommercial and educational service for nonprofit organizations and individuals active in the struggle against IPRs on life. The views expressed in each post are those of the indicated author(s). To join the mailing list, send the word "subscribe" (no quotes) as the subject of an email message to bio-ipr-request@cuenet.com. The Action Group on Erosion, The ETC group, formerly Rural Advancement Foundation Technology and Concentration (ETC) International (RAFI), is an international civil society URL: www.etcgroup.org organization headquartered in Canada. It is dedicated to the advancement of cultural and ecological diversity and human rights, and monitors development in this area. Quaker UN Office (QUNO) QUNO represents worldwide Quaker concerns for peace 13 Avenue du Mervelet, Geneva, and justice at the international level. It works with Switzerland negotiators in Geneva to help develop their capacity URL: www.quno.org to deal with IPRs, food, biodiversity, and health issues and produces papers dealing with Article 27.3(b), traditional knowledge, plant variety protection, and food security and biotechnology. All are available on the organization's Web site--click on the link to Geneva pages. 256 Agriculture and the WTO theory, stronger IPRs should encourage more Development Impacts research and development (R&D) in countries The effects of IPRs protection become even more where they exist, but there is "limited empirical evi- complex when producers and users of knowledge are dence," even in industrial countries, that IPRs pro- in different countries with different economic levels tection leads to increased investment in R&D. This of development. Theoretically, "it is far from clear is partly because of the difficulty of separating that all countries should be required to maintain cause and effect--IPRs may stimulate more invest- the same level of intellectual property protection," ment, but countries and firms that invest more in argue Trebilcock and Howse. If a country has lim- R&D may demand more protection. ited innovative capabilities and primarily consumes Although IPRs restrict direct imitation, they can foreign innovations, stronger IPRs protection assist in the diffusion process of new knowledge may, they suggest, lead to "at least short-term con- within and among economies. Patents provide pub- sumer welfare losses and may discourage imi- lished information, which other researchers can also tation and adaptation by competitors, which use to develop innovations. The World Bank report themselves constitute valuable economic activities." found that the level of IPRs protection appears to For example, in some developing countries with influence the degree of foreign direct investment patent systems, the governments did not allow (FDI), the vertical integration of multinational patent protection on certain products, such as phar- firms, and direct technology transfers through tech- maceuticals. The absence of patents enabled these nology sales and licensing agreements, although the countries' infant industries to examine and copy relationship between protection and FDI is not well products and develop local production capacities-- established, according to other studies (UN 1993). as Swiss industry did in the 19th century. While There are costs related to the granting of IPRs as this may have inhibited inward investment, it may they increase the market power of the right's own- also have produced net economic benefits for the ers, which may lead to higher consumer prices. country. They also, according to the World Bank report, The World Bank report suggests that IPRs can "shift bargaining power toward the producers of disadvantage developing countries, however, "by knowledge, and away from its users." Stronger IPRs increasing the knowledge gap and by shifting bar- may lead to a higher cost of acquiring knowledge, gaining power toward the producers of knowledge, and therefore may adversely affect follow-on inno- most of whom reside in industrial countries." vations that draw on inventions whose patents have While accepting the point, some analysts see such a not yet expired. Tighter IPRs, notes the report, view of IPRs as equating knowledge producers with "may actually slow the overall pace of innovation. commercial and research-based producers, and However, there is no systematic empirical evidence they focus on the role played by developing- confirming this, just as there is none on the positive country farming communities in producing knowl- impact of IPRs on increased R&D." edge about plants and animals (Tilahun and Policymakers face the difficult task of defining the Edwards 1996). Intellectual property rights also scope of IPRs--the length and breadth of pose developing countries with another challenge protection--to maximize social welfare and to because "so many industrial-country firms are achieve certain distributional objectives. Protection acquiring strong intellectual property positions, that is too weak may lead firms to invest less than is often covering fundamental research tools as well as socially desirable in the creation of new knowledge. marketable products, that it may prove hard for Overly stringent protection may lead to wasteful new firms and researchers to elbow into this new R&D spending as firms compete to be first to inno- global industry." Firms and public research groups vate, which may make public R&D more socially in developing countries need to be able to negotiate desirable than private R&D. Only rarely, suggest agreements to use these technologies and "to par- Michael Trebilcock and Robert Howse (1995), will"a ticipate in the continuing debate about particular single level of protection for all technologies or sec- forms of intellectual property, to ensure that their tors maximize domestic welfare," as the tradeoff interests and those of their country are taken into between the economic benefits of innovation and account," states the report. imitation will depend on the sector involved. Global Intellectual Property Rights: A New Factor in Farming 257 Intellectual Property Rights encompassing patent requirements of Article 27. in International Agreements (This article states that "patents shall be available for any inventions, whether products or processes, TRIPS is not the only agreement concerned with in all fields of technology, provided that they are IPRs and food. The Convention on Biological new, involve an inventive step and are capable of Diversity (CBD), agreed at the 1992 United Nations industrial application.") Conference on Environment and Development There are concerns by a range of academics and (also known as the Rio Earth Summit), and the nongovernmental organizations (NGOs) about International Treaty on Plant Genetic Resources for the relatively high levels of IPRs protection and Food and Agriculture (ITPGRFA) both deal with judicial interpretations in the United States (U.S.) IPRs. The latter agreement was finally reached in and the European Union (EU), which provide November 2001 after seven years of difficult negoti- broad patent protection in software and biotech- ations. The International Union for the Protection nological inventions (Tansey 2002, p. 17). Recent of New Varieties of Plants (UPOV after its French moves in the EU to establish extensive rights in title, "Union Internationale pour la Protection des databases could also damage scientific research Obtentions Végétales") and the WIPO are also rele- (Commission on Intellectual Property Rights vant (see Tansey 2002). The most wide-ranging 2002, p. 108). Whatever the concerns about the agreement, however, is TRIPS. potentially damaging effects of excessive IPRs protection in developed countries, Keith Maskus in a major study of IPRs argues that it "is not too The WTO TRIPS Agreement early to claim that they are inappropriate for The 1995 WTO TRIPS Agreement covers eight developing economies and net technology areas: copyright and related rights, trademarks, importers" (Maskus 2000, p. 238). geographical indications, industrial designs, patents (and plant variety protection--PVP), lay- TRIPS Section 5: Patents out designs (topographies) of integrated circuits, protection of undisclosed information, and control One reason for greater interest in patents, especially of anticompetitive practices in contractual licenses. in the Organisation for Economic Co-operation Patents and PVP are the most important areas and Development (OECD) countries, is the rapid for agriculture. development of biotechnology and its application Unlike other WTO agreements, the only special in agriculture. A patent prevents someone from and differential treatment for developing countries making commercial use of what is claimed in the in TRIPS concerns the timetable for implementa- patent without the authorization of the patent tion. While developed countries had to implement holder. To be patentable, an invention must be: TRIPS within one year of the agreement's entry · Nonobvious for someone skilled in the art, that into force, developing countries had an extra four is, not simply be an extension of something that years, that is, until January 1, 2000. A similar delay already exists--it must require some inventive applied to economies in transition (from centrally step; planned to market economies). Least-developed · Novel or not previously known; and countries have a 10-year transition period, but may · Industrially applicable in some way. apply for extensions and thus could delay imple- menting TRIPS (Art. 66.1). New members acceding Patents can be given for products and processes and to the WTO generally do not benefit from the tran- are limited to a fixed period--at least 20 years sitional arrangements. under TRIPS--after which the invention moves Toward the end of the negotiations on TRIPS, into the public domain and can be used by anyone. as a result of strong resistance from a few develop- Patents are territorial and apply only in the coun- ing countries, various modifications were made tries in which they are granted. Each country to provide some degree of flexibility in its imple- decides its own patent rules and how to interpret mentation. These flexibilities include the lack of the meaning of different terms, but WTO Members definition of any terms and exclusions to the all- must meet the minimum standards in TRIPS. 258 Agriculture and the WTO In return for the temporary monopoly granted Obtaining a patent can be quite expensive. by the patent, the inventor must make a full disclo- Preparing a U.S. patent application in the early sure of the nature of his or her invention that is 1990s cost about US$20,000, while the cost in the understandable to anyone else skilled in the neces- EU was about twice that amount. Plant Breeder's sary arts or sciences. In this way, inventions do not Rights (PBRs), however, an alternative to patents "die with the inventor." Moreover, others can try to discussed later in this chapter, are cheaper--about invent something better, but sufficiently different a tenth the cost of a patent (Lesser 1997). A patent so as not to infringe the claim of the original applicant must apply for patents in every country patent. where they will be used, pay an annual fee to main- Clear evidence that the patent system has stimu- tain the patent, and pay patent agents' costs. The lated the development of new products and tech- costs of filing a patent also vary greatly, ranging nologies, which otherwise would not have been from US$355 to US$4,772 across 32 countries sur- developed, is available only for a few sectors (such veyed in the early 1990s (Helfgott 1993). For firms as pharmaceuticals). In other sectors of the econ- at the forefront of biotechnology research and use, omy, patents are sometimes considered to have establishing rights of ownership over new processes mainly anticompetitive effects: they serve to secure and plant varieties may also be a costly business as and strengthen the position of market leaders and firms engage in litigation to determine rights own- limit the entry of new competitors. In the extreme, ership and to secure their markets. they may actually slow the pace of innovation if a dominant firm possesses a powerful group of TRIPS Article 27.3(b): Patentable Subject Matter patents that limits the ability of other firms to and Exceptions improve existing products and technologies. Although policymakers have sought to limit Three subparagraphs in Article 27 of the TRIPS such adverse effects of patents through various Agreement permit exceptions to the basic rule on policies and regulations such as revised IPRs legis- patentability: lation, competition policy, and other business regu- lations, the anticompetitive implications remain a · Article 27.2 applies when members want to pre- cause of concern. During the past decade such con- vent the commercial exploitation of the invention cerns have regained momentum with the emer- to protect ordre public, or morality. This explicitly gence of patents on biotechnology products and includes inventions that are dangerous to human, processes that cover fundamental research tools, animal, or plant life or health, or are seriously prej- genetically engineered plants, human genes, and udicial to the environment. living organisms. · Article 27.3(a) provides exemptions for diagnos- Patenting life-forms, or parts of them, has raised tic, therapeutic, and surgical methods for the a range of concerns. Some are ethical, with various treatment of humans or animals. religious faiths and indigenous peoples rejecting · Article 27.3(b) permits WTO Members to the idea of intellectual ownership over life-forms. exclude from patentability "plants and animals Some are environmental, linked to the role the other than micro-organisms, and essentially bio- patents play in underpinning the private sector-led logical processes for the production of plants or development and in promoting the spread of animals other than non-biological and microbio- genetically engineered plants and animals with its logical processes. However, Members shall pro- possible consequences for the environment, biodi- vide for the protection of plant varieties either by versity, and human and animal health. Some argue patents or by an effective sui generis system or by that both patents and PVP systems favor commer- any combination thereof. The provisions of cial breeders at the expense of small, poor farmers this subparagraph shall be reviewed four years who are the basis of traditional, community-based after the date of entry into force of the WTO breeding--and fear that the adoption of these IPRs Agreement." will promote more intensive farming, which will marginalize small farmers, cause social dislocation, The language of this exception is deliberately and increase food insecurity. complex--described as "constructive ambiguity" Global Intellectual Property Rights: A New Factor in Farming 259 by some negotiators--and continues to be subject words, and legal cases (which may be lengthy) are to interpretation and legal argument among WTO likely to determine which competing interpretation members over its meaning (see Tansey 1999 for a will prevail. Currently, patenting principles and more detailed discussion of this issue). practices on biotechnological inventions remains in Members may also provide limited exceptions to a state of flux internationally, including in those the exclusive rights conferred by a patent, provided countries that have already gained experience in that such exceptions do not unreasonably conflict patenting genes. According to Carlos Correa (see with a normal exploitation of the patent and do not Correa 2000), where patenting is allowed, "the unreasonably prejudice the legitimate interests of patenting of genes at the cell level extends the scope the patent owner, taking account of the legitimate of protection to all plants which include a cell with interests of third parties (Art. 30). the claimed gene." In general, patents give patentees Patents must also be available, and patent rights the right to prevent any commercial use of the enjoyable, without discrimination as to the place of materials, including that for research and plant- invention and whether products are imported or breeding purposes. This could threaten commercial locally produced (Art. 27.1). According to Article plant breeding, especially with broadly drafted 28.1(a) of the TRIPS Agreement, patents relating to patents--for example, those that seek rights over products confer the right to prevent third parties processes used in any plant species. from "making, using, offering for sale or importing WTO Members may provide limited exceptions to for those purposes the product" without the paten- the exclusive rights conferred by a patent (Art. 30). tee's consent. This last point is important for agri- This provides some flexibility in drafting patent legis- cultural trade. For example, if something was lation and may allow members to include exemptions patented in the U.S. but it was not patented else- for research and breeding purposes. WTO Members where, then the holder of the U.S. patent could are also free to determine what "invention" means, block imports of that product unless royalty pay- and many developing countries, including Argentina, ments were made. This has happened in a number Brazil, and the Andean Pact countries, "exclude the of cases, for example, a contested yellow bean patentability of materials found in nature, even if iso- patent in the U.S., and has given rise to consider- lated therefrom," notes Correa. Other areas of flexi- able controversy.2 bility relate to the interpretation of the issues of In the case of process patents, the patentee may novelty and inventive step and the scope of claims prevent the use of the process as well as the commer- that will be admitted. cialization of a product "obtained directly by that process." Thus, if a process to produce a plant (for Plant Variety Protection--Sui Generis and UPOV instance, by genetic engineering) is patented, exclu- sive rights would also apply with respect to the plants A sui generis system of protection is a special system obtained with the process.Article 34.1 also places the adapted to a particular subject matter, as opposed burden of proof in process patents on the producer to protection provided by the existing patent or to show that it is not using the patented process. copyright systems or one of the other main systems of intellectual property protection. Thus, countries can make their own rules to protect new plant vari- TRIPS Article 27.3(b) eties with some form of IPRs provided that such The terms used in Article 27.3(b) are not defined in protection is effective. The agreement does not the TRIPS Agreement. The words that are open to define the elements of an effective system. In the interpretation are plants, animals, microorganisms, last resort, it will be a WTO dispute settlement essentially biological processes, nonbiological, panel that will interpret the provision relating to an microbiological, plant varieties, effective, and sui effective system. generis (of its own kind) system. These disputed One possible sui generis system likely to be rec- words are defined differently in different interna- ognized as effective is the UPOV system of PBRs. tional and national legislation. Various authors UPOV began in Europe in the 1960s and by early argue that there is considerable scope for individual 2002 had 50 members, of which 14 were developing national interpretations to be attached to these countries. PBRs were developed in Europe because 260 Agriculture and the WTO plant breeders found it difficult or impossible to appropriate for an agricultural sector characterized meet two of the fundamental requirements of primarily by subsistence farming (IPGRI 1999). Since patent law: inventiveness, and a written description both systems may exist in the same country, IPGRI of how to make and use the invention. PBRs were suggests that it may be worthwhile for countries to developed in response to the needs of commercial explore how options can be mixed and matched, breeders and allow for the registration of a plant including the prohibition of double protection and variety that has been discovered. UPOV defines a providing different levels of protection for varieties of breeder as the person who bred, or discovered and the same species, depending on their intended use. developed, a variety. Such a variety must be dis- IPGRI also stresses the need, whatever IPRs are tinct, stable, sufficiently uniform, and novel. Most used, for appropriate mechanisms to prevent any plant varieties developed and used by small farmers monopolistic effects of IPRs, particularly patents. in developing countries and by traditional and Such mechanisms include: antitrust laws; shifting indigenous communities do not normally meet the burden of proof in patent law so that patent these criteria. Some countries such as India are claimants have to prove that their wide-ranging developing their own systems of PVP, although claims will work, rather than challengers having to there is considerable pressure for most to adopt the prove that their claims will not; rigorously applying UPOV system. the inventive step and industrial application require- Two major differences in terms of the level of ments; mechanisms to balance the claims of initial protection offered made PBRs less restrictive com- and subsequent innovators; and limiting or pro- pared to patents. PBRs provided a breeder's or hibiting the use of functional claims. researcher's exemption, which allowed use of the protected varieties for further R&D, and a farmer's Seed Provision and Plant Variety Protection exemption or "privilege," which allowed farmers to save and replant seeds from the first harvest of the Formal seed production systems encompassing pub- protected crop. The scope of these exemptions has lic and private R&D with breeding companies domi- changed during various amendments to the UPOV nate seed provision in industrialized countries. More convention. For example, the 1961 version prohib- informal seed production systems, with production ited both patents and PVP on a variety, but the largely by a mixture of farmers and public institu- 1991 version does not. It also no longer requires a tions, exist in many developing countries. With the farmer's exemption, but rather leaves it is an option. exception of the U.S., and more recently the EU, in The UPOV system, however, produces quite a most countries plant varieties cannot be patented. strong IPRs regime for plant varieties geared to insti- There is now pressure from some developed coun- tutional breeding, which may not suit all countries. tries for patents to be extended to plant varieties as The alternative is for countries to develop their own well as PVP, and for PBRs to become more patent- solution with special legislation protecting plant like in their conditions. varieties appropriate to their situation. Both are pos- With the introduction of PVP, in particular sible, but developing an appropriate sui generis sys- PBRs, the plant and seed industry argues that tem is a challenging task that may take some time these will enable it to undertake breeding work (for a detailed discussion of this, see Dhar 2002). The and import seeds and breeding material into Organization of African Unity has developed a developing countries. The key questions, argues model law, and India has introduced legislation that Dwijen Rangnekar in a background study for the attempts to implement a new sui generis model and United Kingdom's Commission on Intellectual that encompasses farmers' rights as outlined in the Property Rights, are whether "the access to for- International Treaty on Plant Genetic Resources for eign-bred genetic material [has] enhanced Food and Agriculture (ITPGRFA). national capacity in plant breeding, and what is The International Plant Genetic Resources Insti- the impact on food security. Existing literature on tute (IPGRI), in producing a checklist for use in Kenya does not provide encouraging evidence on developing a sui generis system, argues that an IPR either of these two issues" (Rangnekar 2002). He suitable for an industrialized system of production finds that "(p)rivate sector breeding tends to limit geared toward export is unlikely to be suitable or itself to high value/low volume crops and hybrids. Global Intellectual Property Rights: A New Factor in Farming 261 Further, the agronomic qualities indicate that the for agriculture. Until relatively recently, agricultural target areas are characteristically the post­Green R&D was largely publicly funded. Research results Revolution areas. Accordingly, it appears unlikely were given to farmers through extension services. that the crop and agronomic needs of the wider The financial returns for publicly financed R&D farming populations, particularly low external- into improved farming productivity are high for input use communities, are consistent with this both developing and developed countries. The U.S. research priority." economy, for example, benefited from its long-term The Commission found that, based on evidence investment of US$134 million worth of support to from developed countries on the impact of patents international wheat and rice research aimed at or PVP on research, PVP could be used in these developing countries by up to US$14.7 billion, countries as a marketing strategy for product dif- according to research by the IPGRI. ferentiation as well as contribute to mergers in the In the OECD countries, private spending now seed industry (Commission on Intellectual Prop- accounts for about half of R&D. In many cases, erty Rights 2002, ch. 3). It is not only PVP legisla- governments have moved away from near-market tion that affects seed provision but also national research, which has immediate applicability on seed regulations and the effects of both on poor farms, to focus spending on basic research, which farmers that have to be considered. underpins future private R&D efforts. In some The key to determining what impact the intro- countries, resources have shifted into areas sup- duction of PVP required under TRIPS will have is porting agribusiness and food processing, which whether the system adopted takes account of local "may have reduced rather than increased the rate of conditions and provides a regulatory framework that return to public sector research," according to will support the various farming systems in the IPGRI researchers (Alston and others 1998). country--rather than advantage some and disad- One concern about patents is their effects on the vantage others. In a review of possible options, movement of breeding materials--animals and Biswajit Dhar concludes that "the sui generis legisla- plant germplasm. To date, the focus has been on tion that developing countries must introduce has to plants. There is evidence that the strengthening of take into consideration the interests of both the IPRs is leading to restrictions on the flow of farming communities and the plant breeders in the germplasm and therefore inhibits the development formal sector" (Dhar 2002, p. 27). He argues that of new plant varieties, particularly by publicly adopting the systems developed by the industrialized funded institutions such as those supported interna- countries and embodied in UPOV is not adequate, tionally by CGIAR. The seed industry itself is con- and suggests approaches that take into account cerned about this reduced flow and recognizes the farmers' rights, as outlined in the ITPGRFA, and that need to ensure that it is maintained. allow seeds bred by farmers to be covered. The private sector, naturally, invests in areas in Under patent law there is no farmers' exemption which it can hope for a return--with much work in to allow the use of farm-saved seed as allowed for agrochemicals. Today, former agrochemical compa- in UPOV. IPGRI notes, "Breeders and modern nies have expanded to become biotechnology/seed biotechnology companies often perceive the farm- companies (or life science companies, including ers' exemption as potentially reducing the profit, or pharmaceuticals), moving downstream to add value the expectation of profit. Consequently, there may to their products. Huge investments have gone into be strong opposition on the part of breeders and this area--over US$8 billion per year in the U.S. modern biotechnology companies to this exemp- alone, according to Ismail Serageldin, a vice presi- tion in countries where patent-like protection for dent of the World Bank and former chairman of the plant varieties is being considered" (see IPGRI CGIAR. Serageldin is concerned that this private 1999). proprietary science will focus on crops and innova- tions that will find rich markets, and ignore those of interest to poor, small farmers. Moves in developing Research and Development Priorities countries to a greater private sector role in breeding The importance of patents and PVP is increasing would need to be balanced by continued public partly because of changes in the funding of R&D service activity geared to their needs. 262 Agriculture and the WTO There is some concern among researchers over must not be obvious to one skilled in the art), the effects of the extension of copyright on access to restricting the scope of patent claims by making educational materials and databases. More generally, claimants prove the applicability over broad areas, since education is one of the prime areas for public and creating strong experimental use exemptions. investment that brings major returns in increasing productivity, if the extension of copyright affects Reviewing TRIPS Article 27.3(b) access to education, especially by the poor, it may and Geographical Indications have deleterious effects. Therefore, special considera- Reviewing 27.3(b) tion may be necessary for knowledge that affects food security--to ensure that the flow of informa- There has been little progress in the WTO on the tion is not inhibited to the detriment of farmers, mandated review of Article 27.3(b) that began in researchers, and others involved in food production. 1999. There is a wide range of views on what should happen and whether specific interpretations or amendments are needed, and many countries have Market Structure and Farmers' Options made proposals. Developing countries' proposals Market structure is also very important in assessing relate to the extension of exclusions from patentabil- the likely impact of changes in the IPRs regime and ity to all life-forms and extension of the timetable for the role different participants can play, according to implementation, prevention of biopiracy, respecting John Barton of Stanford Law School (Barton 1998). use of traditional knowledge and farmers' rights, There is an increasing trend toward economic con- amendment to take account of the Convention on centration of market power in larger and larger Biological Diversity (CBD) and International Under- enterprises throughout the developed countries, standing (now Treaty, see below), and what types of including in the seed industry (Tansey and Worsley sui generis systems of PVP are effective. Developed 1995). One argument for business mergers is that countries' proposals include deletion of the exclusion larger firms will be able to raise capital more easily and no lowering of standards of protection.3 than the small firms that make up the industry. Another is that there are economies of scale in R&D Geographical Indications and Other IPRs activities. The existence of an oligopoly also gives the firms greater pricing freedom and thus enables Although patents and PVP are likely to have the them to recover research costs. Such a structure most direct effects on agriculture, the major partici- provides an incentive for small firms to invest in pants in the food system use a mix of IPRs in their biotech innovation in the expectation that they will businesses. Many companies make considerable use recoup their costs and make money when they are of trademarks, investing substantial amounts of cap- sold to the large firms. ital in brand recognition to secure their markets. There are potential problems with this concen- Greater efforts to protect brands and increase market tration of market power, however, and not simply of share are increasingly likely by companies operating control over prices--it may produce declining globally. Mergers and acquisitions in the food indus- incentives for research. According to Barton, "the try that have become common over the past decade incentives for the industry leaders to conduct are still continuing, as firms prepare to serve global research are now limited . . . new smaller firms may markets and also to counter the growing power of now find it impossible to enter the business because multiple retailers. Achieving some kind of brand they face the assembled patent rights of the industry identity for products is a major challenge for small leaders and possibly also face contract restrictions farmers and traditional knowledge-based producers on access to marketed materials that would once as they try to differentiate and sell their products in have been available for further breeding." Barton markets dominated by brand advertising. sees the key challenge as finding a way to reverse the For some products, a combination of widely oligopoly, while maintaining the use of intellectual advertised branded (trademark) products and trade property incentives to encourage research. Such secrets can be used--Coca-Cola being the most incentives might include a tougher application of famous generally and hybrids perhaps the most the nonobviousness principle (that is, inventions notable use of trade secrets in agriculture. For some Global Intellectual Property Rights: A New Factor in Farming 263 producers, geographical indications defined as being knowledge as well as excluding plants and animals linked to a region and a method of production pro- from patentability as allowed in TRIPS. Moreover, vide a marketing tool that allows them to capitalize if indigenous crops such as quinoa, nuna, or yellow on their uniqueness--for example, Roquefort cheese beans are patented in developed countries, in what and Parma ham. Such designations normally come is now called biopiracy, this may foreclose export out of a well-established agricultural and/or arti- markets there--or if other crops have look- sanal activity that has national recognition and pro- alikes produced that are trademarked and widely duces items sought after by consumers. David marketed, this may also undermine the potential Downes and others in a study of these issues with markets for developing-country crops. five case studies of niche products (kava, Rooibos tea, quinoa, Basmati rice, and neem) concluded that The United Nations Convention on Biological "(b)oth geographical indications and trademarks Diversity show the greatest potential [to benefit local produc- ers] where traditional small-scale production is still The CBD is a framework agreement that leaves par- present, on the supply side, and where end-use prod- ties free to implement it through their own legisla- ucts are marketed directly to consumers. In other tion. Its three objectives are the conservation of words, they are less likely to be appropriate when the biological diversity, the sustainable use of its com- product is a commodity traded primarily in bulk" ponents, and the fair and equitable sharing of the (Downes and others 1999). benefits arising out of the utilization of genetic How far the use of geographical indications resources.4 (GIs) will affect agricultural producers is unclear The CBD recognizes, in Article 8(j), the need for because the economic costs and benefits of any in situ conservation of biodiversity and protection extension of stronger protection of GIs into food are of indigenous knowledge. Article 8(j) requires not clear. GIs are likely to be of use to traditional states, subject to their national legislation, to pre- communities producing products that could have a serve the knowledge, innovations, and practices of niche market domestically and abroad. They must indigenous and local communities insofar as that first be protected locally, however (Art. 22 [Protec- knowledge, innovation, and practice serve the goals tion of Geographical Indications] of TRIPS). There of conservation and sustainable use of biodiversity. is considerable disagreement among developing In agriculture, this comes about through its use countries about the economic benefits of extending and development in farming communities. Thus the stronger protection to geographical indications for impact of changes brought about by IPRs- foodstuffs, and this is under discussion at the WTO protected innovation in agriculture on those com- TRIPS Council. Some countries, such as India, favor munities becomes an issue. The CBD also requires this, believing they will gain from having protection states to diffuse that knowledge, innovation, and for a range of products such as Basmati rice. Others, practice with the cooperation of their holders, and such as Argentina, with a large segment of the pop- encourages the sharing of any benefits that arise ulation tracing their roots back to Europe and with from such diffusion. tastes for European-type foodstuffs, fear that pro- The CBD requires the equitable sharing of bene- duction of a local version of many products will fits arising from the commercial use of communi- become much more difficult if they are prevented ties' biological resources and local knowledge (Art. from using terms associated with the foodstuff that 15.7). It also requires that access to generic are likely to be reserved for products such as cheeses resources be subject to "prior informed consent of from Europe. the Contracting Party providing such resources, Notwithstanding the latter concern, there unless otherwise determined by that party" (Art. remain problems of misappropriation of tradi- 15.5). At the sixth meeting of the Conference of the tional knowledge of food crops, and the lack of a Parties (COP) in April 2002, parties agreed to a set system for ensuring benefit sharing with traditional of voluntary guidelines on access and benefit shar- and indigenous communities, despite the CBD (see ing (ABS) that aim to facilitate access to genetic Correa 2001). Other kinds of instruments besides resources on "mutually agreed terms" and on the IPRs may be needed to protect these communities' basis of the country of origin's "prior informed 264 Agriculture and the WTO consent" by providing guidance to parties in the riches to be found in compounds in plants with development of ABS regimes, while promoting minerals in the ground. Some developing countries capacity building, technology transfer, and the pro- felt that biodiversity-rich developing countries had vision of financial resources.5 undervalued wild biodiversity that was of use to Moreover, a section on the role of IPRs in imple- developed countries and industries, which had been menting ABS arrangements, contained in the deci- making use of them in patented products--such sion, invites parties to encourage the inclusion of as pharmaceuticals--generating enormous returns. disclosure requirements in IPRs applications and This winner-take-all, mining mentality toward the requests the CBD Executive Secretary and WIPO to exploitation of wild biodiversity pays little attention provide further information on this issue. to the differing nature of agricultural genetic In the CBD, which the U.S. has signed but not resources that have been developed, exchanged, and ratified, parties agree to undertake to provide bred around the globe for millennia. Indeed, some and/or facilitate access to and transfer of technolo- "wild" biodiversity-rich countries such as Brazil are gies to other parties under fair and most favorable agriculturally biodiversity-poor, depending for most terms (Articles 16.1 and 16.2). Such technologies of their agricultural production on nonindigenous include biotechnology and others "that are relevant crops such as soybeans. to the conservation and sustainable use of biologi- The COP supported the renegotiation of the cal diversity or make use of genetic resources and existing International Undertaking on Plant do not cause significant damage to the environ- Genetic Resources for Food and Agriculture (IU). ment" (Art. 16.1). Access to such technologies must Agreed on in 1983 at the UN Food and Agriculture be "on terms which recognize and are consistent Organization (FAO), the IU was premised on the with the adequate and effective protection of intel- concept of germplasm as a common heritage of lectual property rights" (16.2). This language mir- humankind, and needed to be restructured to be in rors that in the TRIPS Agreement. harmony with the CBD. The IU recognized that The CBD also aims to enable these developing today's crops had been developed by the activities countries where genetic resources originate to have of farmers over millennia. Renegotiating the IU access to the technology that makes use of those began in 1994 and concluded in November 2001, resources on mutually agreed terms, including tech- when an International Treaty (ITPGRFA) was nology protected by patents and other IPRs (Art. agreed to at the FAO conference in Rome. 16.3). The parties to the treaty also are required to The new treaty creates a mechanism that avoids cooperate to ensure that patents and other IPRs "are the high transaction costs involved in bilateral supportive of and do not run counter to" the CBD's exchanges of breeding material for food crops. It objectives (Art. 16.5). This reflects disagreement aims to ensure future food security by facilitating about whether or not IPRs support the CBD's objec- exchange of these materials through a multilateral tives and implicitly accepts that conflicts may well system, which will use material transfer agree- arise between IPRs and the CBD. A study for the ments. Such exchange is a necessity for future European Commission argued that legally TRIPS plant-breeding work. A rather limited range of 35 and the CBD are not in conflict but that conflicts crops and 29 grasses and forages are included, as may arise when the agreements are implemented, well as the ex situ collections of those crops held by depending on how that is done (CEAS Consultants the International Agricultural Research Centres in association with Geoff Tansey and Queen Mary belonging to the CGIAR. The treaty does not cover Intellectual Property Research Institute 2000). How- animals, although many breeds are threatened with ever, this congruence is not universally accepted. extinction. The treaty includes provisions on IPRs in Arti- cles 12 and 13, which recognize that any germplasm The International Treaty on Plant Genetic taken out of the general pool available for further Resources for Food and Agriculture breeding, by having patents taken out on it, would Only in decisions of the COP to the CBD did mem- create a loss to society as a whole--which should bers recognize the special needs of agriculture. The be compensated by some payment into a fund to CBD developed from an approach that equated promote the use of genetic resources for food and Global Intellectual Property Rights: A New Factor in Farming 265 agriculture. Considerable work remains to be done the CBD code and the ITPGRFA. The secretariat is to determine exactly how these and other articles of also working on documenting public domain tradi- the treaty will be implemented. This also applies to tional knowledge to ensure that patent examiners farmers' rights, which the treaty recognizes in Arti- can use such information to prevent misappropria- cle 9 but leaves to parties to realize as they wish. tion of this knowledge. This happened on a num- Governments should include at least three meas- ber of well-publicized occasions and gave rise to ures in their attempts to promote farmers' rights, concern in developing countries about biopiracy-- according to Article 9.2: the unauthorized commercial exploitation of the knowledge and resources of traditional and indige- · Protection of traditional knowledge relevant to nous communities in developing countries.7 plant genetic resources for food and agriculture. Although the IGC is most directly concerned · The right to equitably participate in sharing ben- with genetic resources and traditional farming efits arising from the utilization of plant genetic communities, other deliberations in WIPO could resources for food and agriculture. affect IPR use and remove the apparent flexibilities · The right to participate in making decisions, at negotiated into TRIPS, for example, through moves the national level, on matters related to the con- to harmonize requirements in national patent servation and sustainable use of plant genetic regimes. Harmonization would make the patent resources for food and agriculture. system of countries more like one another in terms of administrative procedures and rules, enforce- The negotiations associated with both the CBD and ment standards, and substantive law. the ITPGRFA have had a wider range of participa- A final area in which WIPO may affect the tion from civil society in their deliberation, both nature of IPRs in developing countries is through domestically and internationally, than has been the technical assistance provided to countries to the case with the WTO TRIPS Agreement, or the help them frame their laws and develop expertise in other IP-related bodies of WIPO and UPOV. In these areas. There have been concerns raised by a keeping with the farmers' rights provisions of the number of NGOs that this assistance is too nar- ITPGRFA, wider consultation with farmers is also rowly focused and has not supported countries needed in framing IPRs in agriculture. enough in using the flexibilities contained in TRIPS (MSF and others 2002). The World Intellectual Property Organization A Confusion of Fora WIPO is the specialized UN agency whose mandate Different government ministries and interests are is "to promote the protection of intellectual prop- involved in negotiating at these many different fora erty throughout the world through cooperation and there is considerable difficulty in achieving among States and, where appropriate, in collabora- coherence among them--or at least avoiding out- tion with any other international organization."6 right conflicts or contradiction. This was highlighted Developments on patents and other IPRs at WIPO in an aptly titled report, "Why Governments Can't could provide the basis for additional issues to be Make Policy--The Case of Plant Genetic Resources included in the TRIPS Agreement. An Intergovern- in the International Arena" (Petit and others 2001). mental Committee on Intellectual Property and The report reviewed decisionmaking in Brazil, Genetic Resources, Traditional Knowledge and France, Germany, India, Kenya, the Philippines, Swe- Folklore (IGC) was set up in 2001 to consider diffi- den, and the U.S. and found policy coordination cult issues such as IPRs relating to access to genetic problems in all of these countries. According to the resources and benefit sharing, as well as the protec- report, "(t)he combination of a complex interna- tion of traditional knowledge and expressions of tional negotiation process and a complex set of folklore. Following two meetings of the committee issues with tremendous long term social, economic in 2001, the WIPO secretariat is preparing model and political impact is the perfect setting for a IPR clauses for contractual agreements on ABS-- breakdown of international consensus on the issues although these will need to take into account both of genetic resources." The multiplicity of fora also 266 Agriculture and the WTO poses a challenge in many countries to developing inter alia, the relationship between the TRIPS greater coherence among different ministries in for- Agreement and the Convention on Biological mulating their polices and in the negotiations. Diversity, the protection of traditional knowledge and folklore, and other relevant new developments raised by Members pursuant to Article 71.1. In The Development Dimension undertaking this work, the TRIPS Council shall and TRIPS be guided by the objectives and principles set A central concern about the current rules concern- out in Articles 7 and 8 of the TRIPS Agreement ing IPRs is whether the one-size-fits-all approach of and shall take fully into account the develop- TRIPS rules, despite the apparent flexibilities, is ment dimension. adequate to deal with the diverse needs of commu- nities worldwide. In 2001, speaking to the first There is a concern about how extension of the cov- meeting of the IGC, the U.S. delegate questioned erage of patents for inventions in all fields of tech- whether "a comprehensive, uniform set of rules at nology, whether products or processes, will have an the international level to govern the use of genetic impact on agricultural R&D. While some of the resources, traditional knowledge and folklore" was concerns (for example, those over issuing over- either possible or desirable. Since many traditional broad patents) can be addressed within the terms knowledge farming communities are responsible of TRIPS through strict definitional and examina- for the development of agrobiodiversity, it is per- tion criteria, others, such as those over the length of haps worth asking the same question about TRIPS protection, cannot. While Article 27.2 could be when it comes to its impact on agriculture. interpreted to exclude certain inventions from The fourth WTO Ministerial meeting in Doha in patentability, these inventions must be prevented 2001 put a great deal of emphasis on there being a from commercial exploitation. This might address development dimension to the future work of the concerns of those who do not approve of the WTO. It made a number of specific comments patenting life-forms on a moral, ethical, religious, relevant to the future work on TRIPS including the or customary-law basis. However, it would not following: address the concerns of those who would prohibit In the Declaration on Implementation: basic processes from being patented, but allow them to be used for commercial purposes-- 11.2. . . . developed-country Members shall although the meaning of commercial exploitation submit prior to the end of 2002 detailed reports is not given in TRIPS. on the functioning in practice of the incentives Article 27.3(b) provides the greatest scope for provided to their enterprises for the transfer of revisiting the provisions, as a review was mandated technology in pursuance of their commitments in 1999 and highlighted in the Doha Ministerial under Article 66.2. Declaration. However, this review has been dead- locked and would require significant movement In the Ministerial Declaration: among the members to achieve a consensus. If, however, any clarifications or interpretations are 12. . . . issues related to the extension of the made acknowledging the special importance of protection of geographical indications provided food security and need for differentiation in that for in Article 23 to products other than area--linked as it is to nutrition, which is specifi- wines and spirits will be addressed in the Coun- cally mentioned in Article 8 on Principles--this cil for TRIPS pursuant to paragraph 12 of this might be the place to make such clarifications. Declaration. . . . Moreover, since food security is a crucial issue for 19. We instruct the Council for TRIPS, in pur- development, and the Doha Ministerial Declara- suing its work program including under the tion clearly required members to "take fully into review of Article 27.3(b), the review of the account the development dimension" in their implementation of the TRIPS Agreement under deliberations, there may be opportunity to do so in Article 71.1 and the work foreseen pursuant to future. The thrust of many of the concerns about paragraph 12 of this Declaration, to examine, the impact of IPRs on food security, and the appli- Global Intellectual Property Rights: A New Factor in Farming 267 cation of biotechnology, is that they will impede the to different needs. As the United Kingdom's developmental needs of the poorest people in many IPRs commission concludes, "the interests of countries. developing countries are best served by tailoring Given the many questions and uncertainties their intellectual property regimes to their par- surrounding IPRs' impact on agriculture and dif- ticular economic and social circumstances" ferent types of farmers, much more attention (Commission on Intellectual Property Rights must be paid to these issues and a wider range of 2002, p. 155). In other words, one size does not people must be involved in addressing the rules fit all, especially in agriculture (see box 13.1). BOX 13.1 The U.K. Commission on Intellectual Property Rights' Recommendations on Agriculture and Genetic Resources The U.K. government set up an independent eties that suit their agricultural systems. Such IPRs commission in May 2001 to see how IPR regimes should permit access to the protected regimes could best be designed to benefit devel- varieties for further research and breeding, oping countries. Its six members came from and provide at least for the right of farmers to Argentina, India, the U.S., and the U.K., and save and plant-back seed, including the possi- their wide-ranging and challenging report was bility of informal sale and exchange. published in September 2002. The key recom- · Because of the growing concentration in the mendations for agriculture and genetic seed industry, public sector research on agri- resources are here summarized. culture and its international component should be strengthened and better funded. The objec- · Developing countries should generally not pro- tive should be to ensure: that research is ori- vide patent protection for plants and animals, ented to the needs of poor farmers; that public as is allowed under Article 27.3(b) of TRIPS, sector varieties are available to provide compe- because of the restrictions patents may place tition for private sector varieties; and that the on the use of seed by farmers and researchers. world's plant genetic resource heritage is main- Rather they should consider different forms of tained. In addition, nations should consider the sui generis systems for plant varieties. use of competition law to respond to the high · Those developing countries with limited level of concentration in the private sector. technological capacity should restrict the · Developed and developing countries should application of patenting in agricultural accelerate the process of ratification of the biotechnology consistent with TRIPS, and FAO International Treaty on Plant Genetic they should adopt a restrictive definition of Resources for Food and Agriculture and the term "microorganism." should, in particular, implement the treaty's · Countries that have, or wish to develop, provisions relating to the following: biotechnology-related industries may wish to ­ Not granting IPRs protection of any mate- provide certain types of patent protection in rial transferred in the framework of the mul- this area. If they do so, specific exceptions to tilateral system in the form received. the exclusive rights, for plant breeding and ­ Implementation of farmers' rights at the research, should be established. The extent to national level, including: the protection of which patent rights extend to the progeny or traditional knowledge relevant to plant multiplied product of the patented invention genetic resources for food and agriculture; should also be examined and a clear exception the right to equitably participate in sharing should be provided for farmers to reuse seeds. benefits arising from the utilization of plant · The continuing review of Article 27.3(b) of genetic resources for food and agriculture; TRIPS should also preserve the right of coun- and the right to participate in making deci- tries not to grant patents for plants and ani- sions, at the national level, on matters mals, including genes and genetically modified related to the conservation and sustainable plants and animals, as well as to develop sui use of plant genetic resources for food generis regimes for the protection of plant vari- and agriculture. 268 Agriculture and the WTO Notes Sharing Benefits of Biodiversity and Related Knowledge-- Case Studies on Geographical Indications and Trademarks." 1. For a more detailed discussion of the wide-ranging policy Paper prepared for the UNCTAD Biotrade Initiative, issues involved in IPRs, see the bibliography on IPRs and Sus- Geneva. tainable Development as well as the draft policy issues paper on Drahos,P.1995."Global Property Rights in Information: The Story the UNCTAD­ICTSD Web site, listed in the online resources. of TRIPS at the GATT." Prometheus 13: 6­19. 2. See the ETC Web site, http://etcgroup.org for details. Drahos, P., with J. Braithwaite. 2002. Information Feudalism--Who 3. Seehttp://www.grain.org/publications/trips-countrypos-en.cfm. Owns the Knowledge Economy. London: Earthscan. 4. The CBD uses the following definitions: "Biological Helfgott, S. 1993. "Patent Filing Costs Around the World." Jour- resources" includes genetic resources, organisms or parts nal of the Patent and Trademark Office 75(7): 567­80. thereof, populations, or any other biotic component of ecosys- IPGRI (International Plant Genetic Resources Institute). 1999. tems with actual or potential use or value for humanity."Genetic "Key Questions for Decision-Makers--Protection of Plant resources" means genetic material of actual or potential value. Varieties under the WTO Agreement on Trade-Related 5. Reported in Bridges, Weekly Trade News Digest, 6(15), Aspects of Intellectual Property Rights." IPGRI, Rome. April 23, 2002. Available on the Internet at ictsd.org. ipgri.cgiar.org/system/page.asp?frame=publications/ 6. Article 3, Convention Establishing WIPO. July 14, 1967. indexpub.htm 7. See the Web site of the Action Group for Erosion, Tech- Lesser, W. 1997. "The Role of Intellectual Property Rights in nology and Concentration (ETC), formerly RAFI, for NGO cov- Biotechnology Transfer under the Convention on Biological erage of biopiracy cases, at etcgroup.org. Diversity." ISAAA Briefs 3, Cornell University, Ithaca. isaaa.org. Maskus, K. E. 2000. Intellectual Property Rights in the Global Select Bibliography Economy. Washington, D.C.: Institute for International Economics. Alston, J. M., P. G. Pardey, and V. H. Smith. 1998. "Financing Médecins sans Frontičres, Consumer Project on Technology, Agricultural R&D in Rich Countries: What's Happening and Oxfam, and Health Action International. 2002. "Conference Why." Australian Journal of Agricultural and Resource Eco- Report Implementation of the Doha Declaration on the nomics 42(1): 51­82. TRIPS Agreement and Public Health: Technical Assistance-- Barton, J. H. 1998. "The Impact of Patent Law on Plant Biotech- How To Get It Right." Conference Report on the March 28, nology Research." Intellectual Property Rights III--Global 2002 meeting, Geneva. Genetic Resources: Access and Property Rights. Crop Science Petit, M., C. Fowler, W. Collins, C. Correa, and C. G. Thorn- Society of America, Madison, Wisconsin. strom. 2001. "Why Governments Can't Make Policy--The CEAS Consultants in association with Geoff Tansey and Case of Plant Genetic Resources in the International Arena." Queen Mary Intellectual Property Research Institute. International Potato Centre (CIP) October, Lima. 2000. "Study on the Relationship between the Agreement Rangnekar, D. 2002. "Access to Genetic Resources, Gene-Based on Trade-Related Aspects of Intellectual Property Rights Inventions and Agriculture." Study Paper 3a, Commission and Biodiversity Related Issues." Prepared for the Direc- on Intellectual Property Rights. iprcommission.org/ torate-General for Trade of the European Commission. papers/pdfs/study_papers/sp3a_rangnekar_study.pdf. europa.eu.int/comm/trade/csc/dcs_trips.htm. Tansey, G. 1999. Trade, Intellectual Property, Food and Biodiver- Commission on Intellectual Property Rights. 2002. "Integrating sity: Key Issues and Options for the 1999 Review of Article Intellectual Property Rights and Development Policy." 27.3(b) of the TRIPS Agreement. Quaker Peace and Service, Report of the UK Commission on Intellectual Property London. geneva.quno.info/ main/publication.php?pid=113. Rights, London. www.iprcommission.org. ------. 2002. Food Security, Biotechnology and IPRs--Unpack- Correa, C. M. 2000. "Access to Plant Genetic Resources and Intel- ing Some Issues around TRIPS. Quaker UN Office, Geneva. lectual Property Rights." Originally published by FAO-CGRFA geneva.quno.info/main/publication. php?pid=113. in 1998. www.fao.org/ag/cgrfa and the author's book, Intellec- Tansey, G., and T. Worsley. 1995. The Food System--A Guide. tual Property Rights, the WTO and Developing Countries--The London: Earthscan. TRIPS Agreement and Policy Options. London and Penang: Zed Tilahun, S., and S. Edwards, eds. 1996. The Movement for Col- Books and Third World Network. lective Intellectual Rights. Addis Ababa and London: Insti- ------. 2001."Traditional Knowledge and Intellectual Property-- tute for Sustainable Development and The Gaia Issues and Options Surrounding the Protection of Tradi- Foundation. tional Knowledge." Quaker UN Office, Geneva, November. Trebilcock, M. J., and R. Howse. 1995. "Trade-Related Intellec- geneva.quno.info/main/publication.php?pid=113 and GRAIN tual Property (TRIPS)." In The Regulation of International at www.grain.org/publications/trips-countrypos-en.cfm. Trade. London: Routledge. Dhar, B. 2002. "Sui Generis Systems for Plant Variety Protection: UN (United Nations). 1993. "Intellectual Property Rights and Options under TRIPS." Quaker UN Office, Geneva, April. Foreign Direct Investment." Transnational Corporations and geneva.quno.info/main/publication. php?pid=113. Management Division, Economic and Social Development Downes, D. R., and S. A. Laird, with contributions by Graham Department. un.org/esa/ Dutfield and Wynberg. 1999. "Innovative Mechanisms for 14 RULES AND OPTIONS FOR SPECIAL AND DIFFERENTIAL TREATMENT Constantine Michalopoulos Introduction (GATS) resulting from the Uruguay Round and The Uruguay Round Agreement on Agriculture administered by the newly created World Trade (URAA) has been hailed as a milestone because it Organization (WTO). But at the same time, all the started the process of subjecting agricultural trade agreements, including the URAA, contained a large to the same rules as trade in other products. Many number of provisions calling for special and more have noted, however, that while the form of protec- favorable treatment (commonly called special and tion changed, little progress was made in actually differential treatment, or S&D) for developing reducing the level of protection used by some countries--clearly not a "one size fits all" situation developed countries--and their export subsidies, (WTO 2000c). while reduced, still remained at levels that under- Soon after the URAA was signed, it came under mined developing-country production and exports.1 considerable attack from developing-country crit- At the same time, the URAA's focus was on reform ics who felt that it maintained an imbalance in trad- in developed countries' agriculture, not on how ing rules which not only did not favor developing to increase agricultural production and trade in countries, but was actually tilted against them. developing countries. They argued that the S&D provisions were not More broadly, the 1994 Uruguay Round (UR) being implemented, or were not well conceived, or agreements were supposed to usher in an interna- even that the URAA had continued, albeit on a tional trade architecture in which all countries, reduced scale, the special and more favorable treat- developed and developing alike, would abide by the ment of developed countries in the international same rules--the notion that "one size fits all." This trading system. It is fair to say that the URAA was perception of the agreements has considerable always considered an interim agreement--a step- merit, as all countries were expected to have partic- ping stone--not a final one. It contained provisions ipated in previous multilateral agreements under for the initiation of new agricultural trade negotia- the General Agreement on Tariffs and Trade tions, which were duly started in 2000. Now, in the (GATT) such as that on Subsidies and Counter- aftermath of the 2001 Doha Ministerial Conference vailing Measures--and in new agreements, such as which is supposed to result in a "Development the General Agreement on Trade in Services Round," it is time to look carefully at the S&D 269 270 Agriculture and the WTO provisions of the URAA to ensure that indeed they Provisions Regarding Special further the process of development. The Doha and Differential Treatment Ministerial Declaration decided that the WTO The provisions introduced into the WTO agreements Committee on Trade and Development (CTD) to provide S&D fall into two broad categories:2 posi- should review the implementation of all Uruguay tive actions by developed-country members or inter- Round S&D provisions and make recommenda- national institutions, and exceptions to the overall tions to the WTO General Council by July 31, rules contained in the agreements--applicable to 2002 for changes that would make the provisions developing countries, with additional exceptions for contribute more effectively to the achievement of least-developed countries (LDCs) in some cases. development objectives. This and a subsequent Concerning agriculture, these provisions are deadline were missed, and in early 2003, the S&D included in three WTO texts: the URAA; the"Decision discussions in the CTD were deadlocked, having on Measures Concerning the Possible Negative Effects achieved little beyond an agreement that future of the Reform Programme on Least-Developed and S&D implementation should be subject to regular Net Food-Importing Developing Countries" (here- review. At the same time, negotiations on agricul- after, referred to as the Decision); and the Agreement ture continued, yielding in early 2003 a "Draft on the Application of Sanitary and Phytosanitary Modalities Text" containing many S&D provisions Measures (the SPS Agreement). (WTO 2003). The fundamental principle underlying these provisions should be that developing countries are The Plan of This Chapter subject to the same rules and disciplines as all other countries, with certain well-defined exceptions This chapter reviews S&D provisions in the URAA, designed to encourage development while having along with other aspects of the UR agreements the least possible distorting effects on trade. The that have a bearing on agriculture, with the aim main topic of this chapter is how this overall principle of making recommendations that are meaningful can be made operational in agriculture. and make developmental sense. It does not review all aspects of the agreement that have a bearing Positive Steps to Be Taken by Developed Countries on developing countries, even though some issues--for example, market access--are inevitably Developed countries have agreed to take three closely related to S&D. One of the major kinds of steps to support developing countries' problems addressed is that developing countries are participation in international trade: not homogeneous and have widely differing inter- ests in agriculture. Accommodating these varying a. Provide preferential access to developed- interests when designing rules for S&D is a country markets. major challenge. b. Provide technical and other assistance to permit The chapter is structured in four parts, includ- developing countries to meet their WTO obliga- ing this introduction. Part two discusses the con- tions and otherwise enhance the benefits they ceptual basis of S&D, with special attention to the derive from international trade. provisions concerning agriculture. The third part c. Implement the overall agreements in ways that assesses implementation of S&D in the after- are beneficial or least damaging to the interests math of the Uruguay Round and reviews new of developing and least-developed countries. proposals for S&D presented by developing coun- tries in the agriculture negotiations, in the run-up Preferential Market Access Developed countries to the Doha meeting, and in the most recent nego- have provided tariff preferences to exports from tiations on agriculture. The final part offers developing countries under the Generalized System some preliminary conclusions on approaches that of Preferences (GSP) schemes and, within that con- may be useful in structuring S&D provisions in the text, for special treatment of the LDCs. The GSP agriculture negotiations. An appendix lists all the schemes were given a GATT waiver in 1971 and S&D provisions in the UR agreements related made a permanent feature of GATT provisions to agriculture. through the 1979 Enabling Clause. By and large, the Rules and Options for Special and Differential Treatment 271 GSP programs of developed countries have focused a general nature and are expressed in broad "best on manufactures and have relatively limited cover- efforts" terminology.3 For example, the URAA's age of agricultural products. However, substantial Preamble says that "in implementing their commit- preferences on agricultural products have been ments on market access, developed-country mem- included in autonomous trade preference schemes bers would take fully into account the particular extended to smaller groups of developing coun- needs and conditions of developing-country mem- tries. Examples include those provided by the Euro- bers by providing for a greater improvement of pean Union (EU) to the developing countries in opportunities and terms of access for agricultural Africa, the Caribbean, and the Pacific (ACP) on products of particular interest to these members" such products as beef, rum, sugar, and bananas (for (see the chapter annex). There is a similar state- which a WTO waiver was extended through 2007); ment in Article 10.1 of the SPS Agreement. as well as those provided by the United States (U.S.) to countries in the Caribbean and Central America. A recent EU decision will provide tariff- and quota- Differential Commitments and Obligations free access to all LDC products except armaments by Developing Countries under a new agreement entitled "Everything But There are two fundamental ways in which develop- Arms" (EBA). ing and least-developed countries have accepted dif- ferential obligations under the WTO agreements. Technical and Other Assistance The WTO First, they enjoy freedom to undertake policies that agreements contain numerous references to the limit access to their markets or that provide support desirability of developed-country members and to to domestic producers or exporters in ways not international institutions providing technical assis- allowed to other members--all of which can be tance to developing and least-developed countries. viewed as exemptions from WTO disciplines. And The main objective of such assistance is to second, they are provided with more time in meet- strengthen the institutional capacity of developing ing obligations or commitments under the agree- countries and LDCs to enable them to meet the ments. In some cases, more favorable treatment WTO obligations. Interestingly, there are no refer- involves a combination of the two. ences to technical assistance in the URAA as such. Both the Decision and the SPS Agreement call for Exemptions from Disciplines The most general technical assistance to improve agricultural pro- and fundamental way in which developing countries ductivity and to meet the SPS standards. In addi- continue to be exempted from WTO disciplines tion, Article 10.4 of the URAA and the Decision regarding market access policies is the recognition of make recommendations on food aid, and the latter the principle of nonreciprocity in trade negotiations addresses the possibility that developing countries with developed countries to reduce or remove tariffs may face financing difficulties as a consequence of and other barriers to trade. This principle is recog- the implementation of the URAA (because reduced nized in GATT (1994) Article XXXVI and in the export subsidies of developed countries might lead Enabling Clause. Consistent with these provisions, in to higher prices and therefore to increased financ- the Uruguay Round in their schedules of commit- ing requirements for food imports). ments developing countries were able to bind tariffs at levels much higher than applied rates--although in agriculture, unlike industry, they had to bind Implementation of WTO Provisions in a Manner all tariffs. Favorable to Developing-Country Members The URAA also contains measures that exempt In their preambles, as well as in the substantive developing countries and, to an even greater extent, provisions, the WTO agreements contain many LDCs, from disciplines and obligations that apply references committing developed-country mem- generally. These measures may also provide for bers to implement the agreements in ways that take longer timetables or more modest reductions in into account the interests of developing and least- government support and subsidies than those that developed countries. These references are usually of apply to developed-country members. For example, 272 Agriculture and the WTO investment subsidies or input subsidies to low- this aspect of special and differential treatment income producers are exempted from the calcula- have to do with the realism of the time extensions, tion of aggregate measurement of support (AMS); relative to the actual time and cost required to build reductions in export subsidies are targeted both to the institutional capacity needed. be smaller and to occur over a longer period of time; and there are specific provisions regarding the Special Measures Concerning operation of government stockholding programs Least-Developed Countries aimed at enhancing food security as well as less demanding minimum access provisions regarding The Enabling Clause of 1979 provided the basis for primary agricultural products that are the predom- special treatment of LDCs "in the context of any inant staple in the traditional diet of a developing- general or specific measures in favor of developing country (see the chapter annex). countries." The UR agreements contain 17 provi- A second way in which developing countries sions applicable specifically to LDC members, in have been granted exemptions from general disci- addition to those applicable to all developing plines on the degree of protection is through the members. The URAA calls explicitly for exempting provisions of GATT Article XVIII, which give LDCs from all AMS and subsidy reduction com- developing countries the freedoms to grant the tariff mitments (see the chapter annex). A number of protection required for the establishment of a par- separate initiatives have also been adopted with ticular industry and to apply quantitative restric- regard to LDC members. Thus the 1994 WTO tions for balance-of-payments purposes. These have "Decision on Measures in Favour of Least Devel- been little-used for agricultural products. oped Countries" (WTO 1994) requires LDC mem- bers to undertake commitments and concessions only to the extent consistent with their individual Time Extensions The second fundamental way in development, financial, and trade needs, or which special and differential treatment is provided administrative and institutional capabilities. in the WTO is through the extension of the time- frame over which certain obligations under the agreements are to be implemented by developing Conceptual Issues Regarding Special and Differential Treatment countries and LDCs. Flexibility in transition times in Agriculture is provided in practically all WTO agreements: the URAA, for example, permits developing countries There are several conceptual problems related to to continue subsidizing exports for a period of time S&D provisions that characterize all UR agree- and in a variety of ways prohibited for other mem- ments, which are manifested in agriculture as well. bers, with flexibility taking the form of a longer The first problem is that developed-country implementation period. S&D commitments in the WTO agreements are The justification for the extension of additional either too broad and general in nature--such as time to implement agreed measures relates to those included in the many preambular statements, weaknesses in the institutional capacity of develop- including those in the URAA and in SPS--to be of ing countries and LDCs. It was assumed that, given any practical significance, or are of the best-efforts additional time (as well as technical assistance, variety, such as the provision of technical assistance, often expected to be provided), developing and which means that they are not legally enforceable least-developed countries would be able to (Kessie 2000) and, therefore, that developed coun- strengthen their institutions in ways that would tries cannot be held accountable for not imple- enable them to implement the agreements. In the menting them. case of subsidies, the presumption was that addi- Underlying this problem is a second issue, tional time would permit countries to develop the which, unless addressed, is likely to undermine any institutions and policies to implement alternative future efforts to provide developing countries with means of support (for agriculture or for exports in meaningful differential treatment by developed general) that were acceptable under the agreement. countries. This is the failure of all agreements to The main issues that arise in the implementation of recognize that the capacities of individual developing Rules and Options for Special and Differential Treatment 273 countries to export and compete in international interests in developed countries frequently succeed markets, including for agricultural exports, are in discriminating against these developing coun- vastly different--as are their needs for support and tries on products in which they enjoy comparative assistance. Despite this difference, most WTO rules advantage. If S&D provisions are not differentiated on the treatment of developing countries by devel- to exclude developing countries that do not need oped countries--except those rules concerning the them, there is very little likelihood that meaningful LDCs--are identical.4 Singapore and the Republic S&D commitments will be made by developed of Korea are to be treated the same way as Ghana countries. They will continue to resort to general, and Saint Lucia, Argentina and Brazil the same as best-efforts kinds of commitments whose imple- the Maldives and Kenya, and so on. mentation is hard to evaluate. On agricultural trade issues, at least four groups Developing countries have been treated differ- of developing countries can be identified, and this ently by developed countries in the latter's unilateral is often reflected in the support provided for differ- implementation of their various GSP schemes. But ent proposals in the agriculture negotiations. First, leaving to developed countries the decisions on there is the group of major agricultural-commodity definition of which countries are to get what pref- exporters, members of the Cairns Group. Second, erences and how much, invites the nonuniform there is a large group of the net food-importing treatment of countries in similar circumstances developing countries (NFIDCs) along with others and the introduction of extraneous political con- such as India, which have a significant agricultural siderations into the determination of benefits. sector and which produce but also import food, The final conceptual problem in the framing of and export various agricultural products. A third S&D provisions is that the commitments aimed group are countries with small nondiversified agri- at addressing developing countries' institutional cultural sectors that--either because of climatic constraints have been made without serious con- conditions or land constraints (for example, the sideration of how these will be implemented. small island economies)--face significant difficul- This was seen in the setting of the transition peri- ties in competing in agricultural trade. A fourth ods for implementing the various agreements. small group consists of generally higher-income These periods were part of the bargaining process developing countries that, like some developed at the late stages of the UR negotiations and with- countries, attach high priority to the many functions out much involvement of developing-country offi- that agriculture plays in their society, irrespective of cials familiar with how long it would take to build its efficiency or productivity. institutional capacity where it was inadequate or Food security is one area in which some effort totally lacking. In some cases, the time limits for the was made to apply the UR provisions in a some- extensions have already passed, and there is little what differentiated fashion regarding the benefici- evidence that countries have made sufficient aries: there was recognition, for example, of the progress in institution-building to permit them to different needs of the NFIDCs. But the category implement their obligations fully. On the contrary, itself does not adequately capture the groups of evidence gathered from the needs assessments countries facing the largest food insecurity situa- undertaken in the context of implementing the tions (Diaz-Bonilla and others 2000). In the context Integrated Framework for Trade Related Technical of export restraints, there was also recognition of Assistance for the Least Developed Countries sug- the fact that some developing countries are major gests that many institutional weaknesses remain in agricultural exporters. Most of the provisions, how- these countries and in others where even shorter ever, consider developing countries (except the timeframes for implementation had been envis- LDCs) as a single group, which clearly they are not. aged. While not all developing countries have given There is very little economic reason to suggest these issues the priority they deserve, it is clear that that some of the more developed of the developing the transition time provided was far too short to do countries cannot compete in the products in which any meaningful institution building. they have comparative advantage with developed A different aspect of the same problem is evident countries. And there is very little political support in the commitments made by developing countries for extending S&D to them. Indeed, protectionist to implement various new standards called for by 274 Agriculture and the WTO the UR agreements, including the SPS Agreement. offers a large opportunity for preferences,presumably These commitments were made without detailed under GSP schemes. At the same time, a number of consideration of the cost of their implementation developing countries have argued for improve- or whether building the institutional capacity ments in the existing GSP schemes, such as reduc- needed to permit developing countries to discharge ing the variability of product coverage from their responsibilities was a high development prior- country to country and from year to year, simplify- ity for them. ing the rules of origin to make these schemes more accessible, and binding tariff preference margins. Proposals have also been made to compensate Issues in the Implementation of Special and developing countries for the losses they sustain Differential Provisions in Agriculture when tariff or access preferences are eroded by Market Access The main market access problem reductions in MFN tariffs. These proposals have to in agriculture is that the present most-favored nation be evaluated in light of past experience with GSP (MFN) tariffs in some developed countries, as well schemes, the potential benefits that they would yield as the continuation of export subsidies, creates for- to developing-country beneficiaries, and in the con- midable barriers to developing-country exports text of other developments that will affect develop- and undermines the competitiveness of developing ing countries' trade in agriculture. countries in third markets. In developed-country The long experience with GSP schemes suggests markets for some agricultural products tariffs are that, while preferences can be and have been help- lower than overall agricultural tariffs, as this ful to some countries for some products and results from earlier liberalization in tropical prod- through some periods, preferential approaches ucts. Tariffication for some agricultural products suffer from economic drawbacks when used as under the UR provisions was often "dirty," resulting a means of improving generalized developing- in higher levels of protection than existed previ- country access to developed-country markets. First ously. The remaining protection on developing- of all, the bulk of the benefits tends to accrue to a country agriculture exports such as rice (Japan), few countries--frequently those that need them tobacco and groundnuts (U.S.), or grape juice (EU) the least. Second, preferential approaches tend to is staggeringly high (UNCTAD 1997; UNCTAD/ promote an attitude of dependency on particular WTO 1997). products and markets that is inimical to change in There is little evidence that the general hortatory response to changing market conditions or to pro- language of the URAA regarding liberalizing prod- mote competitiveness. And third, when aimed at a ucts of particular interest to developing countries special group of countries, for example the LDCs, resulted in greater cuts in protection on such prod- these preferences tend to impose certain costs on ucts in developed-country markets. It is unlikely other developing countries--in some cases low- that this central problem can be addressed by more income and possibly equally deserving of similar verbiage and nonbinding differential treatment treatment.5 commitments in the future. The political economy of trade preferences also As noted earlier, there is relatively limited provi- militates against generalized and unconditional sion of S&D in favor of developing countries in approaches because the preference-giving countries agricultural products under GSP-type schemes. At wish to preserve the freedom to choose the recipi- the same time, a number of preferential schemes-- ents, often using conditionality that includes non- most notably by the EU toward the ACP but also by economic criteria. The preferences offered to the the U.S. in favor of Caribbean and Central Ameri- LDCs by the EU's EBA initiative, and possibly by can countries and through a separate scheme for others in the future, are an exception to this rule African countries--are providing much deeper that can be explained in part by the fact that the preferences (that is, reduced tariffs) for the benefi- LDCs account for such a small portion of total ciary countries. trade as to have a very small possible impact on the The fact that in agriculture, unlike in industry, economies of developed countries. tariffs are high for some key agricultural products In judging the usefulness of preferences in the has led some developing countries to argue that this context of the negotiations on agriculture, other Rules and Options for Special and Differential Treatment 275 developments affecting agricultural trade in the countries in the agriculture negotiations, such near term should be taken into account. Most issues as including more products in GSP schemes important is the provision of preferential treatment and possibly also "binding" existing preference by the EU to the 80-odd ACP countries through margins by expressing them as a fixed percentage of Economic Partnership Agreements that essentially ad valorem tariffs.6 Given, however, that GSP will involve the establishment of regional trade schemes would really be of potential benefit only to arrangements (that provide preferences for all a relatively few countries, the question needs to be members) to be put in place after 2007 and the asked as to how much "negotiating capital" should expiry of the WTO waiver. Similarly, the U.S., in be expended in this direction. addition to the preferential scheme regarding An alternative approach that would make much African countries, is negotiating the establishment more economic sense is to expand the beneficiaries of of a free trade area that would include all the devel- the "deep" preferences now being extended to the oping countries in the Americas. The EU will also LDCs to cover a larger group of low-income, vulnera- have the EBA in place favoring the 49 LDCs, and ble, or "food-insecure" developing countries that other developed countries may follow with similar exhibit very similar problems to those of the LDCs programs for these countries. but which are not included in the current list (Tanger- Thus, a few years from now, the landscape in agri- man 2001). The problem with this approach is that it cultural trade for LDCs may include the deep and runs against the political economy of granting prefer- mostly unconditional preferences in most developed- ences to a list of countries unconditionally--and that country markets, while regional trade agreements the larger the group of countries included, the more providing deep preferential margins for many devel- difficult it is to cope with the developed-country oping countries may be in place in the two major domestic interests, affected by the expanded com- developed-country markets, the EU and the U.S. This modity and volume coverage. Finally, such an expan- situation would leave a number of countries in South sion of the LDC list would have to be accompanied Asia without additional preferences except for GSP, with clear graduation criteria to ensure that countries while some countries in the ACP and some in Latin do not continue to receive preferences when they do America would not enjoy preferences in the U.S. and not need them. EU, respectively. Included in this group would be a These considerations suggest that by far the best number of major developing-country agricultural way of addressing the issue of market access to exporters, such as Argentina, Brazil, Uruguay, the developed-country markets in the current agricul- Philippines, and Thailand, which arguably do not ture negotiations is not through efforts to expand need preferences in order to export. generalized S&D treatment, but rather through a In this setting, a "strengthened" GSP for agricul- general effort by developing countries to push for ture is really not likely to be especially "general- substantial reduction of developed-country protec- ized."At best, it could provide less deep preferences, tion on an MFN basis and for total elimination of and hence fewer benefits, for a relatively small export subsidies at the earliest possible date. A group of developing countries that should in any variety of approaches may be used for cutting case exclude those that do not need the preferences developed-country tariff protection, for instance by to export. Some of the proposals to strengthen GSP adopting a formula approach that would result in schemes, such as simplifying rules of origin, while more than proportional cuts in the tariff peaks that useful in the case of agroindustries, are far less are present in some developed-country agriculture relevant for agriculture than for manufactures. (see FAO 2002a). At the same time, relatively slow Other proposals, such as assistance programs for liberalization in developing countries would have countries--especially small economies dependent an impact on other developing countries that on only a few crops, which could be hurt sub- are major exporters, as trade among developing stantially as a consequence of the elimination of countries in agricultural products is important, preferences--may be implemented more easily in especially in Asia (Ruffer 2002). the context of regional arrangements that cover various aspects of economic integration. This Sanitary and Phytosanitary Issues The imple- leaves, as possibilities to be pursued by developing mentation of the new SPS Agreement has raised 276 Agriculture and the WTO significant problems for developing countries. ened. Article 9.2 of the SPS Agreement states that Some result from the lack of capacity to develop the where "substantial investments" are needed for a institutional arrangements that permit them to developing country to meet the SPS "requirements meet their SPS-related WTO commitments. Even if of an importing Member, the latter shall consider the capacity exists, these commitments are costly to providing such technical assistance as will permit implement: standards, along with testing and certi- the developing country Member to maintain and fication, represent between 2 and 10 percent of expand its market access opportunities for the overall product costs. Thus they impose a burden product involved." Developed-country members on developing countries even when the standards should have the obligation to provide the develop- are used for legitimate reasons and the countries ing country with both the financial and technical are able to meet them. assistance needed to permit the improvement in The greatest potential costs of the SPS Agree- product quality until the product meets the stan- ment, however, may lie in the market access issues dard. Ideally the obligation should be legally bound they raise. The agreement may serve to legitimize under the WTO, but even a broader "commitment" developed-country actions that result in substantial that makes the link may be a useful step. More problems for developing-country trade, even if the broadly, there should be increased coherence actions appear to be justified, for example, on health between trade commitments and assistance to meet grounds. Article 10.1 of the SPS Agreement says that those commitments, a more general point that "[i]n the preparation and application of sanitary requires additional coordination of actions by the and physosanitary measures, Members shall take international community. account of the special needs of developing-country Members and in particular of the least-developed country Members" (see the chapter annex). A Technical and Other Assistance As noted earlier, recent paper found that a 1998 European Commis- the URAA itself is silent on issues of assistance, sion regulation that raised the standards for the except in the case of food aid. The references to minimum level for certain types of aflatoxin (a assistance related to agriculture are included in the toxic substance found in foodstuffs and animal Decision and in the SPS Agreement (discussed feed), to levels higher than those required by the above). Regarding the provision of technical and Codex Alimentarius (the food standards code) is financial assistance to improve "agricultural estimated to cost close to US$700 million in lost productivity and infrastructure," the WTO Secre- revenue to African groundnut exporters, many of tariat prepared a compilation of such assistance pro- them LDCs (Otsuki, Wilson, and Sewadeh 2000). vided since 1995 as notified by members. It is It is difficult to argue against developed-country unclear whether this compilation includes all the actions that aim to protect consumer health. But assistance provided, including that by the inter- when such actions result in significant costs to national financial institutions. While the WTO is developing-country exports, it would seem fair that not the main international body where coordina- the developed countries be legally committed to tion of assistance in general, and on agriculture in take steps that will help the developing country particular, can be achieved, it is important that meet the problem that the developed-country regu- information regarding assistance flows available to lation has created. There are examples of such assis- the WTO be as complete as possible, especially on tance in the past: the EU has provided assistance to trade-related issues. Senegal in order to help exporters meet EU fish- The Decision also called for a new food aid con- eries standards. Yet a recent review by the WTO vention which was negotiated in 1999 and which Secretariat showed no information regarding the stipulates that, when allocating food aid, priority implementation of technical assistance provision should be given to LDCs and low-income countries. (WTO 2002a). Perhaps more assistance has been Other NFIDCs can also be provided with food aid provided in this area bilaterally but has not been "when experiencing food emergencies or inter- notified. nationally recognized financial crises leading to The existing provisions in the SPS Agreement on food shortage emergencies or when food aid opera- these issues are inadequate and need to be strength- tions are targeted on vulnerable groups." Donor Rules and Options for Special and Differential Treatment 277 performance regarding both food aid levels and from the CFF were included in the overall IMF allocations to LDCs and NFIDCs under the Food program to individual countries. But there is Aid Convention (FAC) 1999 has been monitored nothing--and there should be nothing--automatic closely by the WTO Committee on Agriculture. about the assistance provided. Indeed, if a need can These reviews have shown that donors collectively be shown to exist, the international response supplied food aid in excess of the commitments should not be limited to food aid but should extend they had made under the FAC, and that they fol- to all kinds of general-purpose financing on appro- lowed the guidelines regarding its concessionality. priate terms. The latter would be preferable to food At the same time there continues to be a problem in aid, which is frequently tied to procurement from a that food aid is most available when international particular donor and is determined by food stock prices are lowest and, hence, financing needs of availability in the donor rather than the needs of developing countries are less. the recipient. The main focus of developing-country discus- The new FAC could help, in a small way, in deal- sions on topics covered by the Decision has been on ing with some of the food security problems many the question of whether the URAA has resulted in face, but it is not a substitute for further liberaliza- additional financing needs for the NFIDCs and tion of agricultural trade. It should rather be LDCs on account of higher import prices for food. considered as a supporting element for such liberal- Perhaps because the URAA did not change agricul- ization. Reduced protection in developed-country tural trade that much, some of the fears of NFIDCs markets will improve market access prospects for about the possible adverse effect of the agreement both existing and potential exporters, while did not materialize. The Decision called for a study reduced export subsidies by developed countries will by the World Bank and the International Monetary reduce international market distortions that impede Fund (IMF) of the implications of the URAA on the expansion of developing-country agricultural food import prices for these countries and the pro- production. Where appropriate, reductions in devel- vision of appropriate compensatory finance to deal oping countries' own protection of agriculture, as with the problem. The two institutions found little part of the broader reduction in agricultural protec- evidence that the export subsidy reductions of the tion, could stimulate efficiency through improved URAA had led to an increase in import expendi- allocation of their own resources. tures of poor NFIDCs and concluded that there was Notwithstanding this experience, several devel- no need to provide financing over and above what oping countries have proposed the establishment of could be obtained through their regular facilities. a revolving fund facility whose objective would be Both the IMF and the World Bank have reiterated to stabilize international food prices to developing this position in subsequent reviews of the issue in countries. Following Doha, a new group was set up the WTO. to consider proposals in this area; its recommenda- Even so, it is legitimate to ask what the proper tions were submitted to the WTO General Council international response would be if developed- in June 2002 (WTO 2002c). This is discussed along country export subsidies were to decline. The prob- with other food security issues in chapter 9. The most lem relates to the possible adverse short-term recent draft modalities text developed in the context effects of eliminating trade-distorting measures on of the WTO negotiations on agriculture (WTO 2003) poor NFIDCs. These short-term effects are likely to contains a number of useful suggestions to improve be outweighed by the longer-term worldwide effi- the effectiveness of food aid--for example, that it ciency gains in agricultural markets, and are in fact should be fully in grant form and untied. likely to spread over time, as the distortions are des- tined for phaseout rather than elimination at once. Flexibility in Rules and Disciplines In some ways, It would also be very difficult to isolate the impact flexibility has emerged as the most widespread of the resulting price increases from other factors, instrument of special and differential treatment; including the developing countries' own policies. It is but the fundamental question of implementation is for the price effect reason that the IMF did not pro- whether the latitude permitted, in relation to WTO vide automatic financing from its Compensatory commitments, including tariff bindings or input Financing Facility (CFF) for cereals, but drawings subsidies, results in policies that are more suitable 278 Agriculture and the WTO to development. A more narrow issue is that devel- mostly middle- or higher-income that provided oping countries' rights to differential treatment in significant positive supports to agriculture, and for certain instances, including in the URAA, are con- an even fewer number--10--that had export sub- ditioned on their notification of the existence of sidy commitments. According to the most complete certain policies, namely subsidies as of a particular and recent information available, 40 developing time. While a number of developing countries have countries had notified the WTO on domestic sup- availed themselves of these provisions and pro- port to agriculture in the period 1995­2000 vided such notifications, there is a general impres- (ABARE 2002).7 Probably more countries have sion that many countries have not fully notified used the provisions than have notified the WTO, measures that they have been implementing, and and notifications for most countries are quite therefore, may at some point face challenges to these incomplete (FAO 2000). Only eight countries, policies because of their failure to notify them. mostly high-income, notified support that aver- In agriculture, there is a serious problem in aged in excess of 10 percent of value added in developing rules that are appropriate to promoting the agriculture sector. As value added is but a the interests of developing countries because the small fraction of gross value, none of these coun- URAA's overall focus is on improving the policy tries came even close to meeting the de minimis environment in developed countries, rooted in the provisions for exemptions. Indeed the bulk (over legitimate concern that agriculture in some devel- 50 percent) of the total amounts notified by such oped countries was protected and supported too countries as Brazil, the Republic of Korea, Mexico, much. The rules were, therefore, designed to pro- and Thailand were claimed as exempt and listed vide fewer government supports in a more trans- under the green box provisions. Few countries parent and less trade-distorting way that, if claimed support under the de minimis provisions, implemented, would have a beneficial impact on and two of these countries, India and Egypt, developing countries. accounted for over 85 percent of the total sup- In many developing countries, however, agricul- port claimed to be exempt under these provisions. ture was being penalized, not supported, by gov- Finally, provisions relating to investment and input ernment policy. This was certainly even more the subsidies for small and resource-poor farmers have case a decade ago, when the Uruguay Round was been utilized by a large number (27) of countries being negotiated and perhaps most so in the 1980s, but amounted, on average, to a little over 5 percent the reference years under the URAA for the meas- of total notified support. urement of supports. Even where agriculture was The list of 10 countries that notified export sub- supported, developing-country capacity to provide sidies is instructive, in that Indonesia is the only such support through budgetary transfers was (and one that is classified as low-income by the World still is) quite limited. In a way, the whole philoso- Bank. The others are classified either as middle- or phy that drove the URAA was upside down. How- upper-income.8 Four countries used marketing and ever, as the framework for the URAA was that of transport subsidies for exports (covered by an developed-country agriculture, the agreed S&D exemption in URAA Article 9.4); a few countries provisions and even subsequent developing-country notified outlays for public stockholding for food proposals to "improve" them have been couched in security purposes (under footnote 5 to URAA terms of the same upside-down framework. Thus Annex 2, paragraph 3) and domestic food aid to the URAA provisions in this area call for flexibility the poor (under footnotes 5 and 6 to Annex 2, by enabling developing countries to reduce their paragraph 4); and two countries (the Republic of AMS and export subsidies by less, and over a longer Korea and the Philippines) notified special treat- period of time, as well as to exempt a larger propor- ment provisions in the protection of rice, as a "pre- tion of the supports they provide from reduction dominant staple in the traditional diet"(Annex 5, commitments. Since most developing countries do section B) (see WTO 2002a; Youssef 1999). not have positive AMS, these S&D provisions are of Over the last several years, developing countries limited value. and others have made a number of proposals in the Flexibility, so far, has been provided for the rela- context of the ongoing negotiations on agriculture tively few--perhaps 15 or so--developing countries, for rebalancing the URAA to reflect a developmental Rules and Options for Special and Differential Treatment 279 perspective. Some pertain to changes in the URAA been unable to spend a lot of resources to pro- that would improve market access conditions, mote agriculture or enhance food security, they including, for example, eliminating tariff peaks, not should not be constrained from doing so in renewing the "Peace Clause" for developed coun- the future. tries, and the like. Others have dealt with issues of · Giving credit for negative product-specific sup- assistance to NFIDCs (already discussed), or port in the form of non-product-specific sup- involve the establishment of new special safeguard port in the calculation of permissible AMS, in measures available only to developing countries (to order to take account of situations in which be discussed below). Still others address the issue of countries keep food prices artificially low increasing the flexibility in the existing S&D provi- in order to achieve food security objectives for sions. Some countries have proposed that some or the poor and, therefore, have to compensate most of these provisions be included in a develop- farmers by providing them with inputs and ment box, which would parallel the green box other subsidies. exemptions--which, developing countries believe, · Exempting from the calculation of the AMS tend to focus on measures of primary significance not only capital costs for transport and other to developed-country agriculture. One submission marketing infrastructure but also the current (WTO 2002b) states: transport costs of shipping food from food sur- plus to food deficit regions within a country. Given the fundamental differences in the agri- · Continuation of export subsidies--irrespective culture production systems and the different of previous commitments. role that agriculture plays in developing · Lower rates of reduction of tariffs on agricul- and developed countries, there is a clear case ture than those agreed for developed countries for devising a development box whose provi- and possible renegotiation of existing tariff sions would apply only to developing countries. commitments. Agriculture is no doubt multifaceted, but argu- ments in this regard should not detract focus In addressing future changes in the S&D provi- from the concerns and specific problems of the sions of the URAA, consideration needs to be given rural poor of developing countries, which the to the weaknesses in developing-country institu- development box seeks to address. tions and in the operations of their markets--key characteristics of developing-country agriculture Proposals for Special and Differential Treatment that distinguish it from agriculture in developed in the Doha Round: What Makes Sense countries. Countries also need to consider, in shaping from a Development Perspective? future public sector involvement in agriculture, what has worked and what has not--that is, experience The main developing-country proposals for with "best practice" in government interventions. increasing flexibility in the existing S&D provisions A recent World Bank paper (WTO 2002) says in call for the following (see WTO 2000a­d; WTO this connection: 2001a­b; Green and Priyadarshi 2002.):9 Markets and provision of technology are often · Permanent exemptions in the calculation of the imperfect because of market failure. . . . The AMS for capital and input subsidies to resource- poor are in many cases more affected by market poor farmers. failure than the rich, rural areas more than · Exemptions from the calculation of the AMS for urban areas, and women more than men. Gov- measures that are targeted on all or some of ernments have major roles in enabling the trans- the following: poverty alleviation, rural develop- formation of subsistence oriented rural areas ment and employment, diversification, food with low incomes and little diversification into security. areas that are well integrated into markets and · Raising the de minimis level of exempt AMS, with access to modern technology. This either in general or for expenditures related to requires considerable public investments in enhancing food security, the justification being infrastructure, agricultural research and human that while in the past developing countries have capital. It also requires continuous efforts to 280 Agriculture and the WTO create the appropriate environment for the Article 6.2 subsidies may indeed be actionable development of markets and private sector under Article 13(b), if they exceed the budgetary investment. limit on subsidies fixed in 1992. This means that if a developing country provided subsidized credit to The recent assessment of country experience its poor farmers in excess of that provided in 1992, undertaken by the World Bank noted above, as well and was successful in stimulating increases in their as other literature on rural development, suggests production so that it displaced some imports, pre- that for developing countries to exploit the advan- vious suppliers to its markets could claim "serious tages of better access to developed-country markets prejudice" under the Agreement on Subsidies and and freer trade in general, they need investments in Countervailing Duties or "non-violation nullifica- capacity-building targeting the following: tion"under GATT Article XXIII (see Michalopoulos · The development of necessary infrastructure such 2001; Diaz-Bonilla and others 2002). It may be as irrigation and improved seeds, as well as better unlikely that a developed country that saw its distribution and communication systems. exports decline in these circumstances would com- · Institution building, in the form of research cen- plain in the WTO. But this is not the point: the ters that would improve farmers'access to modern point is that the philosophy and current legal basis production technologies as well as strengthening of the URAA is not development oriented. processing capacity technologies. Another technical improvement relates to the · The establishment of effective regulatory and way the AMS is calculated, on the basis of the gap food safety assurance bodies. between the current administered price and the · Education and training, particularly in technical fixed external reference price from the 1986­88 areas such as sustainable agriculture, quality con- base period. This base period was inserted in the trol, and packaging. URAA to help those developed countries that · Support for the development of local farm feared that declining world prices would force them organizations, especially for marketing and dis- to reduce their domestic supports. It did not address tribution. the concerns of developing countries, where infla- · Programs in support of product diversification, tion has typically been high, and where domestic including programs involving government assis- supports that are expressed in nominal prices may tance for risk management. prove meaningless in real terms (Tangermann and · Support of land reform. Josling 1999). Again, the URAA talks in vague terms about "due consideration" to be given to the influ- In light of these considerations, what should ence of "excessive rates of inflation" on the ability of be the thrust of the revisions in the URAA to make countries to honor their AMS commitments. It is it more development-friendly? Technical, proce- clear that the commitments in the AMS, whatever dural, and substantive changes are needed. they are, need to be made in real terms.10 One of the fundamental changes that develop- ing countries should seek to establish in the present Special and Differential Treatment negotiations is that certain policies that may make First, on the technical/procedural side, the S&D sense in a development context should always be provisions that exempt the support provided available to developing countries--or at least to through investment and input subsidies to poor some groups of them. To the extent that the green farmers under Article 6.2 are not in the green box box does not cover these policies, it should be list of subsidies that are permanently permitted amended, or a development box should be created, because they are nondistorting, but are included or the URAA should be amended in appropriate among the distorting measures of support that are ways to provide for such policies (WTO 2000a, permitted only temporarily. As such, they would 2002b; Green and Priyadarshi 2002). have to be reviewed before they can be continued-- The selection of policies to go into a development as if the problems afflicting poor farmers that box or the amendments needed would require dis- justify government intervention could be solved in cussion and negotiation. A starting point would be a few years' time. the existing URAA provisions exempting developing- Rules and Options for Special and Differential Treatment 281 country subsidies to low-income and resource-poor As regards the provision of subsidies to inputs, farmers from reduction commitments, as well as the such as credit, fertilizers, seeds, and so on, past devel- food security provisions on government stocks and oping-country experience has not been very good. prices. Many such programs have failed in numerous coun- First, the concept of "low-income or resource- tries. More recent efforts, aimed for example at devel- poor producers" to which the measures should apply oping market-based microfinance mechanisms, is difficult to define in practice. A more operationally appear to offer a more viable long-term approach to practical approach might be to exempt from the AMS sustainable rural credit institutions. Similarly, large those programs targeted to all households below a farmers have to a great extent appropriated the bene- certain poverty line, thus including households that fits of many traditional fertilizer and seed subsidy may not "produce" something but which work in programs. New approaches, such as the provision of agriculture and may be equally or more needy than vouchers to poor farmers, may address the problem "resource-poor producers." This approach would better. In general, market-based approaches involving include, for example, programs that have an impact the provision of direct transfers to farmers tend to be on rural labor or other poor consumers. superior and better-targeted measures of support On the question of targeting, there have been than traditional input subsidy interventions as a some suggestions for targeting "food security means of addressing market failure. crops" (Ruffer 2002). There are several problems Developing countries, however, are often in a with such targeting. First, the crops that enhance quandary. Serious budget constraints do not permit food security may differ from one country to them to provide a safety net for the urban poor another, which would lead to a complex system of through direct income transfers. At the same time, notifications of different crops for different coun- innovative approaches involving microfinance and tries. Second, it may well be that there are many direct approaches to input delivery to subsistence poor farmers who do not produce such crops. farmers have not been broadly implemented. In Should they be excluded from support? It would these circumstances, the countries sometimes opt seem both simpler and more equitable to exempt for a "low food price" policy as a second best, which from restraints all agriculture programs that can only be made viable through input subsidies to affect the rural poor, irrespective of the crops they farmers. Clearly, the best long-term option (which produce. Food security issues are discussed more for some developing countries, may be a long time extensively in chapter 9. indeed) is a combination of direct transfers with Second, the basic thrust of program support market-based pricing. In the interim, however, it should be in institution-building and improving may be unwise to limit the existing flexibility pro- the way agricultural markets work. On the question vided by the exclusion of input subsidies to poor of investment subsidies, there is little to add: given farmers from the calculation of the AMS. Similarly, the problems affecting agriculture and the poor in subsidies to internal transport costs to move food to most low-income developing countries, govern- deficit areas may be the most effective way, in the ment intervention would be necessary to strengthen short term, of promoting food security in such institutional capacity, improve market operations, regions--but these subsidies should also apply to and provide support to the urban and rural poor in moving imported food to these areas. various ways, including support to expanding the Operations of emergency food stocks should production of individual commodities. Additional clearly be included in some fashion in the develop- support is needed to help build small-farmer organ- ment box. Some clarification of the existing provi- izations, to promote land reform, to strengthen mar- sions on this score may well need to be considered keting, and to improve financing mechanisms, in (for details see Diaz-Bonilla and others 2002). But the addition to the more traditional government serv- stocks may need to be limited in total value, and also ices listed in the green box. The programs would not involve discrimination against foreign suppliers. need to differ according to whether they focused on Finally, programs in support of product diversifi- subsistence agriculture, smallholders, or commer- cation for small economies dependent on one or two cial farms. None of the subsidies involved should be export crops should be specifically exempted from included in the AMS. reductions in the AMS. At present, the exemption to 282 Agriculture and the WTO reductions in the AMS on account of support for perhaps even more so than many forms of domes- diversification in Article 6.2 is limited to diversifi- tic support, and most developing countries cannot cation from "growing illicit narcotic crops"--again, afford them; so it is in their interest to negotiate primarily a developed-country priority. On the tough rules on export subsidies to avoid getting other hand, small, low-income developing coun- into subsidy contests with developed countries or tries--including a number of small island the more advanced developing-country exporters. economies which need to diversify their production The problem here is more political than economic. and export structure in order to reduce their vul- Economically it may not make much sense for any nerability to external shocks--do not enjoy an country to subsidize exports. But because most exemption from the AMS for such programs. How- developed countries have been subsidizing exports ever, since these programs have to be limited to in the past and most developing countries have not, "minor" crops, they may be difficult to define reductions in developed-country export subsidies in practice. from their present high levels still leaves them with The developing countries should be able to pur- substantially more opportunities for subsidizing sue all of these programs as a matter of right. The exports and thus gaining a commercial advantage programs should be included in the URAA, whether against developing countries. An attempt to correct in a development box or elsewhere, and should not this asymmetry was made in the past by providing be considered aberrations or exemptions. developing countries with longer time periods during The question of raising the de minimis exclusion which to implement reductions in export subsidies. may no longer be relevant, if a well-defined devel- Greater differentiation in future commitments is opment box is agreed upon. There is no evidence probably needed in this area, not because export that the 10 percent exclusion has limited develop- subsidies are a good thing that developing coun- ing countries from providing the support to agri- tries should use, but in order to create a more level culture they need to provide, and hence it is unclear playing field.11 whether it is necessary to increase the de minimis provision from 10 percent to, say, 15 percent. Simi- Country Classification Issues Targeted programs larly it is unclear whether the proposal to permit to help the poor may be of importance to developing developing countries that have negative support for countries irrespective of their levels of income or certain items, as a consequence of consumer food size, and therefore should be available, in principle, subsidies, to offset these against positive support in for all developing countries to pursue. Perhaps the the calculation of the AMS would be necessary, if only exception may be major food-exporting coun- agreement can be reached on truly adequate devel- tries. A more restrictive definition could focus opment-oriented provisions. In this regard, it S&D provisions to address food security in "food- should also be noted that "low food price" policies insecure" countries, which would result in a some- tend to have a net negative effect on poor farmers what narrower list of eligible countries. However, (Ruffer 2002). there are several conceptual difficulties in defining The development box should exclude border "food insecurity" at the country level (see Stevens measures. Tariffs and similar measures raise prices 2002a, 2002b), and the existing listing of NFIDCs is to consumers and have a larger adverse impact on unsatisfactory, as it includes countries unlikely to be poor consumers, who spend a greater percentage of considered "food-insecure." The "product diversifi- their incomes on food, while these border measures cation" exception should be limited to small, low- benefit mostly the bigger agricultural (and food) income developing economies that are dependent on producers with larger quantities of products to one or two crops for the bulk of their export earnings. sell. A case could be made for setting a lower pace of Determining any "list" of beneficiary countries tariff reductions by developing countries in the involves problems. One is that the longer the list, negotiations (see Diaz-Bonilla and others 2002), the more likely the bulk of the benefits would go to but this is a matter of the overall negotiating princi- the more developed of the countries involved. At ples to be adopted. the same time, for the list to gain credibility and Finally, the issue of export subsidies should be political acceptance among developing countries, it considered separately. These are trade-distorting, would have to include the large economies in Asia Rules and Options for Special and Differential Treatment 283 such as China, India, and Pakistan where the bulk of in these countries, but this will take time to imple- the world's rural poor are to be found. The inclusion ment. In the meantime, something needs to be put of these countries is likely to increase opposition in place to deal with this issue. The answer could be from developed countries' protectionist interests. special safeguard provisions for agriculture, avail- The list of countries eligible for S&D benefits able only to developing countries. These provisions should also be consistent with the list of countries could be included in the development box, or could that need access to generalized preferences to be freestanding. Their key characteristics should be improve market share. In the end, of course, any list transparency and ease of use, for example, through would be the result of political compromise and the determination of price or quantity thresholds; tend to contain some degree of arbitrariness. But limited duration; and that the developing countries any listing that effectively takes into account the invoking them have not been able to seek compen- development considerations raised above would sation under the regular WTO safeguard provi- probably be an improvement over the situation that sions.12 The flexibility to use such provisions might prevails today. be made available only to developing countries that agree to reduce bound tariff rates significantly. Safeguards Another important concern of devel- oping countries is the limited capacity of their poor Transition Periods Many developing countries, farmers to adjust to a sudden upsurge in agricul- not only the LDCs, have had difficulty in imple- tural or food imports. This potential danger exists menting various WTO agreements, including the in general, and is exacerbated by the continued use SPS Agreement. Transition periods under a num- of export subsidies in agricultural trade. This issue ber of agreements have already expired and many is addressed more extensively in chapter 3, but a countries say they are experiencing difficulties in few observations are in order here. establishing the institutions needed for their imple- The problem of import surges is neither new mentation. Countries facing fiscal constraints often nor limited to agriculture. Most developing coun- have few resources to direct toward the areas of tries have a degree of leeway because their bound public administration responsible for overseeing tariff levels are much higher than their applied and coordinating the implementation of WTO rates. For 31 developing countries (excluding agreements, which is quite costly. These difficulties Cairns Group members), simple (unweighted) were supposed to be overcome through technical applied tariffs in agriculture averaged 25 percent, assistance and longer transition periods. compared to 66 percent for bound tariff rates. This The longer transition period of 10 years set by suggests that there is considerable scope for them to Article 15:2 for the implementation by developing increase their protection of agriculture, should they countries of reduction commitments under the wish to do so (Michalopoulos 2001). There are also URAA has not yet expired. In the case of SPS, it is general provisions for dealing with import surges notable that apparently no country has sought a under the WTO safeguards provisions. time-limited exception in whole or in part from The problem with the existing situation, how- obligations under the agreement, although longer ever, is twofold. First, the margins between applied timeframes (unspecified) have been explicitly pro- and bound tariffs are likely to shrink as part of the vided for in SPS Articles 10:2 and 10:3. It is actually overall liberalization resulting from the present hard to believe, in light of the frequently voiced con- negotiations. Second, the existing safeguards provi- cerns of developing countries in this area, that their sions may be difficult and time-consuming to implementation of the provisions of the SPS Agree- implement. At the same time, poor farmers in ment is proceeding entirely on time. It would appear developing countries do not have access to the kind far more likely that countries have just not requested of safety nets available to producers of farm or extensions from the WTO, possibly because of their other products in developed countries. This means fear of adverse reactions toward their exports. that if there is an upsurge of imports, it could have A very careful look at these transition periods is catastrophic effects on poor producers (see FAO needed in all areas in which these have been 2002a). Of course, the long-term solution to this extended on the grounds of institutional weakness, problem is the establishment of adequate safety nets including in SPS. In parallel, efforts are needed by 284 Agriculture and the WTO the developing countries themselves, with the sup- tional trade as well as to implement effective port of the international community, to establish trade policies. the institutions needed for discharging their e. Deficient regulatory regimes that are unable responsibilities under the SPS Agreement. to cope with weakness in the operations of markets.13 Least-Developed Countries The international community has made a special effort to address the The key issues for S&D provisions for the LDCs problems faced by the LDCs. These 49 countries are thus concern how concrete and effective support for exempted completely from the AMS reduction trade-related capacity-building measures can be commitments and the export subsidy disciplines. ensured in practically all areas of international trade. Developed countries and some developing coun- In this respect, the LDCs' needs are not qualitatively tries have also made additional preferential market- different from those faced by other low-income access commitments for LDCs. countries, and no additional provisions are needed It was estimated following the UR, that between in the URAA or related agreements to deal with their 80 to 90 percent of the value of LDC merchandise problems. At the same time, it is important that exports overall have duty-free access in their main LDCs recognize some of the implications and pitfalls developed-country markets. Subsequently, the EU of past developing-country experience with flexibil- announced the EBA initiative, aiming at both duty- ity in the application of WTO rules and disciplines. and quota-free access to its markets for all LDC It could be argued, for example, that existing S&D exports. At present there is pressure on other devel- provisions permit LDCs the most freedom of possi- oped countries to follow the EU example in the ble policy choice in areas, such as subsidies, that they context of the current WTO negotiations. As many can least afford. Tighter WTO disciplines in some pol- of the remaining constraints to LDC exports are in icy areas might actually be helpful to LDC govern- "sensitive" agricultural sectors (and textiles), ments that wish to introduce and gain domestic implementation of the proposal made in 1999 by consensus for trade policy reform. the Director General of the WTO and supported by the IMF and World Bank to establish duty- and quota-free access for all LDC exports could result The Ongoing Negotiations in significant further gains in agricultural exports on Agriculture for these countries (in part at the expense of other Many of the above suggestions for S&D treatment low-income developing countries), and the LDCs have been included in the most recent "modalities" would not have to "offer" any new liberalizing com- draft prepared by the chairman of the negotiating mitments in order to obtain improved market group which suggests approaches for future liberal- access. ization commitments in the current WTO negotia- On the other hand, an evaluation of the LDC tions on agriculture (WTO 2003). In particular, the needs assessments prepared under the Inte- draft includes most of the suggestions for strength- grated Framework exercise suggests that the main ening paragraph 6.2, which concerns the establish- constraints to the expansion of LDC exports derive ment of a special safeguard mechanism available primarily from weaknesses in institutional capacities only to developing countries, calculation of AMS in as well as other supply side factors, rather than real terms, and other provisions on food aid as from market access problems (WTO 1998). In par- noted earlier. For this reason, the draft should be ticular, these include the following: considered a good basis for further negotiations on a. Infrastructure deficiencies such as erratic power S&D in agriculture.14 supply, underdeveloped telecommunication At the same time, the draft also contains a networks, and the poor condition of terrestrial, large number of provisions permitting developing sea, and air transport links. countries far greater leeway in protecting agricul- b. Weaknesses in technological capacity. ture through border measures, such as tariffs and c. Underdeveloped financial and banking systems. tariff quotas, than would be the case for developed d. Shortfalls in a broad range skills and institu- countries. Such special and differential treatment tional capacity needed to participate in interna- could be justified because low-income developing Rules and Options for Special and Differential Treatment 285 countries do not have the fiscal capacity to support countries in integrating effectively into the interna- agriculture through less trade-distorting direct tional trading system. This has been done in the income supports. On the other hand, this addi- past, in both the GATT and the WTO, and needs to tional flexibility, over time, could also lead to the continue in the future. Some of this reorientation same kind of inefficiencies in agriculture that involves the general provision of S&D and some of have undermined the competitiveness of many it is specific to agriculture. developing-country industries nurtured behind The main priorities are as follows: high protective barriers. 1. Greater emphasis must be placed on strengthen- ing developing-country institutional capacity. Conclusions and Priorities The main differences between developed and for the Future developing countries are not in the trade poli- This analysis/assessment of S&D provisions in the cies they should pursue, but in the capacities of WTO agreements that are relevant to agriculture their institutions to pursue them. This requires suggests the need for a fundamental reorientation emphasis on S&D provisions related to technical of priorities both by developed- and developing- and financial assistance as well as longer transi- country members of the WTO. tion periods (which are linked to institutional Two major conclusions arise from the analysis. reform and capacity building). In this context: First, there is increasing recognition that institu- tions in many developing countries, including · Transition periods in the WTO agreements those required to implement agricultural trade linked to developing-country institutional policies and those that bear the costs of adjustment capacity weaknesses need to be reexamined. to globalization, are weak and inadequate to cope A number of these transition periods have with WTO obligations, and that markets in these already expired and need to be urgently countries do not work well, requiring government addressed. The review of the transition peri- intervention to make them work more efficiently. ods should also draw on panels of experts Second, supply-side constraints create important from governments and appropriate interna- impediments to the effective integration of many tional institutions that are knowledgeable developing countries into international trade, and about capacity-building efforts and require- require assistance, both technical and financial, to ments in the areas concerned. overcome them. The most extensive evidence · Developed countries must increase the coher- regarding trade-related institutional inadequacies ence between agricultural trade policy and aid and constraints has been developed in the context of to food and agriculture in developing coun- LDCs, but other low-income developing countries tries. This means, first, avoiding real incoher- suffer from similar inadequacies. This conclusion ence, when donors on the one hand provide has major implications about the need to continue assistance to increase agriculture or food pro- and emphasize certain dimensions of S&D. duction in developing countries and on the The overall objective of the international com- other undermine their own and the develop- munity should be to provide more meaningful ing countries' efforts by supplying surplus S&D treatment, through appropriate instruments, agriculture or food products at subsidized to countries that truly need it. A genuine Develop- prices into developing-country markets. Sec- ment Round of negotiations on agricultural trade ond, it means proactive coherence between must address the legitimate concerns of developing trade and aid policy, such as would result countries--especially those that are low-income from a developed-country commitment to and least-developed--about how to develop their support developing-countries in meeting SPS food and agriculture sectors. The conceptual standards in developed countries. underpinning for the reorientation of S&D should be provided by the evolving understanding of the 2. Sharper differentiation of developing countries is links between agricultural trade and development, needed in the provision of S&D. Many and of the constraints faced by different developing problems of institutional capacity that are 286 Agriculture and the WTO common to LDCs and other low-income devel- countries. Tariff preference margins, including oping countries with limited participation in in agriculture where protection is currently international trade are not faced by more very high, can be expected to decline over time advanced developing countries."Objective"indi- as tariffs are liberalized on an MFN basis, and cators of institutional capacity such as per capita no S&D provisions should inhibit this liberal- income and share of world trade need to be ization. One way to help developing countries introduced, which would focus assistance on that truly need assistance in breaking into for- needy countries and exclude more advanced eign markets is to exclude, through graduation, developing countries. Without a graduation pol- developing countries that do not need it. icy that would permit a narrower definition of 5. The agricultural negotiations should result in which countries should be eligible for S&D, agreement on measures to enhance food secu- developed countries will continue to make com- rity and stimulate agricultural production of the mitments to developing countries in general that rural poor in developing countries on a perma- are not concrete, make concrete commitments nent basis. These measures should be articulated only to LDCs that have a very small share of as developing-country rights, not as "excep- world trade, and rely on their own criteria-- tions" to reductions in the AMS. They should which is frequently nontransparent and politi- include elements of S&D now included in Arti- cally motivated--in determining which cle 6:2 of the URAA, expanded as appropriate to countries they will provide with more favorable provide the following: treatment or market access. · Direct and indirect investment and input Differentiation and graduation will be subsidies (including credit) or other sup- difficult for developing countries to accept. ports to households below a poverty line, in Substantial differentiation, however, exists order to encourage agricultural and rural regarding financial flows from all international development. Such supports could be gen- financial institutions and even from the United eral, or product-specific, provided they are Nations Development Programme. In the case effectively targeted to the rural poor. of the World Bank, some developing countries · Programs that support product diversifica- get no assistance at all, others are only eligible tion in the agriculture of small, low-income for loans on hard terms, others for soft loans, developing countries dependent on a very and still others for a mix.15 Why can the princi- small number of commodities for their ple that has been accepted without serious dif- exports, including programs involving gov- ficulty on issues of finance not be acceptable ernment assistance for risk management. regarding trade? It may be difficult to do, but · Foodstuffs at subsidized prices in targeted an effort needs to be made. programs, which do not discriminate against 3. Developing countries should seek improved imports, aimed at meeting food require- access to developed-country markets for ments of the poor (whether urban or rural) agricultural products on an MFN basis through as part of an overall effort to enhance food reductions of tariff peaks, elimination of security. remaining quantitative restrictions, and elimina- tion of developed-country export subsidies as 6. Finally, a new special safeguard provision should soon as practicable, but in any case at a faster rate be introduced in the URAA which would be than that applied to developing countries. available only to developing countries, perhaps 4. Efforts at improving GSP schemes affecting only to low-income ones and for others that agricultural products should focus on providing meet minimal requirements for reducing bound deep preferences, such as those currently avail- tariff rates. Its purpose should be to provide able or to be made available to LDCs, to other quick but time-limited protection against low-income, "food-insecure" or vulnerable import surges that could hurt poor producers. Rules and Options for Special and Differential Treatment 287 CHAPTER ANNEX: SPECIAL subsidies, at the conclusion of the implementation AND DIFFERENTIAL PROVISIONS period, are no greater than 64 per cent and 79 per IN AGRICULTURE cent of the 1986­1990 base period levels, respec- Agreement on Agriculture tively. For developing country Members these per- centages shall be 76 and 86 percent, respectively. Provisions Aimed at Increasing Article 9:4 Trade Opportunities During the implementation period, developing Preamble country Members shall not be required to under- Having agreed that in implementing their commit- take commitments in respect of the export subsi- ments on market access, developed country Members dies listed [below], provided that these are would take fully into account the particular needs and not applied in a manner that would circumvent conditions of developing country Members by pro- reduction commitments (Art. 9:1[c] and [d]): viding for a greater improvement of opportunities and terms of access for agricultural products of partic- · subsidies to reduce the costs of marketing exports ular interest to these Members, including the fullest of agricultural products,[...] including handling, liberalization of trade in tropical agricultural products upgrading and other processing costs, and the as agreed at the Mid-Term Review,and for products of costs of international transport and freight; and particular importance to the diversification of pro- · [providing] internal transport charges on duction from the growing of illicit narcotic crops. export shipments on terms more favourable than those for domestic shipment. Flexibility Article 6:2 Article 12:2 (Domestic Support Commitments). In accordance (Disciplines on export prohibitions and restrictions) with the Mid-term Review Agreement that govern- The provisions of [Article 12:1] shall not apply to any ment measures of assistance, whether direct or developing country Member, unless the measure is indirect, to encourage agricultural and rural devel- taken by a developing country Member which is a net opment are an integral part of the development food-exporter of the specific foodstuff concerned. programmes of developing countries, investment Article 15.1 subsidies which are generally available to agriculture In keeping with the recognition that differential and in developing country Members and agricultural more favourable treatment for developing country input subsidies generally available to low-income or Members is an integral part of the negotiation, spe- resource-poor producers in developing country cial and differential treatment in respect of commit- Members shall be exempt from domestic support ments shall be provided as set out in the relevant reduction commitments that would otherwise be provisions of this Agreement and embodied in the applicable to such measures, as shall domestic sup- Schedules of concessions and commitments. port to producers in developing country Members to encourage diversification from growing illicit nar- Annex 2, para. 3, footnote 5 cotic crops. Domestic support meeting the criteria of (Public stockholding for food security purposes) this paragraph shall not be required to be included in For the purposes of paragraph 3 of Annex 2, gov- a Member's calculation of its Current Total AMS. ernmental stockholding programmes for food security purposes in developing countries whose Article 6:4 (b) operation is transparent and conducted in accor- (Domestic Support Commitments--calculation of dance with officially published objective criteria or current total AMS). For developing country Mem- guidelines shall be considered to be in conformity bers, the de minimis percentage under this para- with the provisions of this paragraph, including graph shall be 10 per cent. programmes under which stocks of foodstuffs for Article 9:2(b)(iv) food security purposes are acquired and released (Budgetary outlays for export subsidies) at administered prices, provided that the difference The Member's budgetary outlays for export between the acquisition price and the external ref- subsidies and the quantities benefiting from such erence price is accounted for in the AMS. 288 Agriculture and the WTO Annex 2, para. 4, footnotes 5 and 6 differential treatment in favor of least-developed (Domestic food aid) and net food-importing developing countries. For the purposes of paragraphs 3 and 4 of Annex 2, Paragraph 5 the provision of foodstuffs at subsidized prices with As a result of the Uruguay Round certain developing the objective of meeting food requirements of urban countries may experience short-term difficulties in and rural poor in developing countries on a regular financing normal levels of commercial imports and basis at reasonable prices shall be considered to be in that these countries may be eligible to draw on the conformity with the provisions of this paragraph. resources of international financial institutions Transition time periods under existing facilities, or such facilities as may be Article 15.2 established, in the context of adjustment pro- Developing country Members shall have the flexi- grammes, in order to address such financing difficul- bility to implement reduction commitments over a ties. In this regard, Ministers take note of paragraph period of up to 10 years. Least-developed country 37 of the report of the Director-General to the CON- Members shall not be required to undertake reduc- TRACTING PARTIES to GATT 1947 on his consul- tion commitments. tations with the Managing Director of the International Monetary Fund and the President of Least-Developed Countries the World Bank (MTN.GNG/NG14/W/35). Article 16.1 Developed country Members shall take such action Paragraph 3(iii) as is provided for within the framework of the To give full consideration in the context of their aid Decision on Measures Concerning the Possible programmes to requests for the provision of techni- Negative Effects of the Reform Programme on cal and financial assistance to least-developed and Least-Developed and Net Food-Importing Devel- net food-importing developing countries to improve oping Countries. their agricultural productivity and infrastructure. Decision on Measures Concerning the Possible SPS Agreement Negative Effects of the Reform Programme on Article 10:1 Least-Developed and Net Food-Importing Developing Countries In the preparation and application of sanitary or phytosanitary measures, Members shall take Provisions under Which WTO Members Should Safe- account of the special needs of developing country guard the Interests of Developing Country Members Members, and in particular of the least-developed Paragraph 3(i) country Members. To review the level of food aid established periodi- Article 10:4 cally by the Committee on Food Aid under the Food Members should encourage and facilitate the active Aid Convention [1986] and to initiate negotiations in participation of developing country Members in the the appropriate forum to establish a level of food aid relevant international organizations. commitments sufficient to meet the legitimate needs of developing countries during the reform program. Transitional time periods Paragraph 3(ii) Article 10:2 To adopt guidelines to ensure that an increas- Where the appropriate level of sanitary or phy- ing proportion of basic foodstuffs is provided to tosanitary protection allows scope for the phased least-developed and net food-importing develop- introduction of new sanitary or phytosanitary ing countries in fully grant form and/or on appro- measures, longer time-frames for compliance priate concessional terms in line with Article IV of should be accorded on products of interest to the Food Aid Convention 1986. developing country Members so as to maintain opportunities for their exports. Paragraph 4 Ensure that any agreement relating to agricultural Article 10:3 export credits makes appropriate provision for With a view to ensuring that developing country Members are able to comply with the provisions of Rules and Options for Special and Differential Treatment 289 this Agreement, the Committee is enabled to grant duty-free treatment in various markets, the "binding" of prefer- to such countries, upon request, specified, time- ences would obviously apply to a much smaller set of countries. 7. Actually ABARE (2002) lists 41 countries, but Mongolia limited exception in whole or in part from obliga- had not been classified as a "developing country" when it acceded tions under this Agreement, taking into account to the WTO. their financial, trade and development needs. 8. The 10 countries are: Brazil, Colombia, Cyprus, Indonesia, Israel, Mexico, Romania, Turkey, Uruguay, and Republica Boli- variana de Venezuela (WTO 2002a). Technical Assistance 9. A number of these proposals (by India, the Small Island Article 9:1 Economies, the Dominican Republic, Pakistan, and Sri Lanka) were reiterated at the Special Session of the WTO Committee on Members agree to facilitate the provision of techni- Agriculture, February 4­8, 2002 (Ruffer 2002). cal assistance to other Members, especially develop- 10. Another issue that developing countries have raised is that ing country Members, either bilaterally or through because they claimed zero or de minimis overall support, they the appropriate international organizations. Such have to keep supports below the 10 percent threshold for each individual product. This again constrains unnecessarily the sup- assistance may be inter alia in the areas of process- ports they could offer by comparison with developed countries. ing technologies, research and infrastructure, 11. There are of course a number of other important issues in including in the establishment of national regula- the negotiations pertaining to the future rate of AMS reductions by developed countries, limitations on their use of Green Box subsi- tory bodies, and may take the form of advice, cred- dies, limitations on their use of exports credits, and so on. These are its, donations and grants, including for the purpose not discussed here as they are not S&D measures in favor of devel- of seeking technical expertise, training and equip- oping countries, but rather involve developing-country proposals to level the playing field by having developed countries accelerate ment to allow such countries to adjust to, and comply the reductions in their trade-distorting policies. with, sanitary or phytosanitary measures necessary 12.Variants for a safeguard mechanism have been proposed. In to achieve the appropriate level of sanitary or phy- one case, some developing countries proposed a safeguard mecha- tosanitary protection in their export markets. nism which would act more like a countervailing action against developed countries that subsidize their exports. There is little rea- Article 9:2 son, however, to differentiate the plight of poor farmers that have been affected by such actions from that of others who have been Where substantial investments are required in equally hurt by import surges caused by other factors. The pro- order for an exporting developing country Member posal has also been made that the special safeguard be limited to to fulfill the sanitary or phytosanitary requirements raising tariffs, or to raising them by a certain proportion (Ruffer 2002). It would appear that while these proposals have merit, of an importing Member, the latter shall consider they would also tend to complicate the use of the mechanism. providing such technical assistance as will permit 13. Based on responses to needs assessment questionnaire, as the developing country Member to maintain and recorded in WTO (1998). expand its market access opportunities for the 14. Curiously, an important S&D provision excluded from the draft is the one that would give the right to developing coun- product involved. tries to support export diversification. The inclusion of such a provision is especially important for small developing countries Notes dependent on one or two crops for the bulk of their export receipts. 1. There is extensive literature on these issues. See among 15. The principle has also been accepted in the establishment others, Das (1998), Diaz-Bonilla and Robinson (2001), and the in 2001of the Advisory Centre on WTO Law, an international various submissions and proposals by developing countries in institution established in Geneva, Switzerland to assist develop- the context of the ongoing negotiations on agriculture (WTO ing countries pursue cases in the context of the WTO dispute 2000a­d, 2001a­b). settlement mechanism. 2. For a discussion of the history and conceptual basis for special and differential treatment, see Oyejide (2002). 3. In a few cases, such as antidumping, there are more explicit provisions on how developing countries are supposed to Select Bibliography be treated more favorably or in ways that are least damaging to their interests. The word processed describes informally reproduced works that 4. There is another exception as well: in the Agreement on may not be commonly available through libraries. Subsidies and Countervailing, a special per capita income cutoff has been established. ABARE (Australian Bureau of Agricultural and Resource Eco- 5. There is extensive literature on this topic. Examples nomics). 2002. "Domestic Support to Agriculture." Current include Karsteny and Laird (1987), Onguglo (1999), UNCTAD Issues, May. (1999), and Topp (2001). Das, B. L. 1998. The WTO Agreements: Deficiencies, Imbalances 6. As the LDCs would be getting duty-free access, hopefully and Required Changes. Penang: Third World Network. in all major markets, and the ACP countries, many African and Diaz-Bonilla, E., and S. Robinson, eds. 2001. "Shaping Globaliza- several Latin American countries would be obtaining similar tion for Poverty Alleviation and Food Security." 2020 Focus 8, August 2001. IFPRI, Washington, D.C. 290 Agriculture and the WTO Diaz-Bonilla, E., S. Robinson, M. Thomas, and Y. Yanoma. 2002. Tangermann, S., and T. Josling. 1999. "The Interests of the "WTO, Agriculture and Developing Countries: A Survey of Developing Countries in the Next Round of WTO Agricul- Issues." Trade and Macroeconomics Paper 81, IFPRI, Wash- tural Negotiations." UNCTAD, November, Geneva. ington, D.C. Topp, V. 2001. "Trade Preferences: Are They Helpful Diaz-Bonilla, E., M. Thomas, S. Robinson, and A. Cattaneo. 2000. in Advancing Economic Development in Poor Countries?" "Food Security and Trade Negotiations in the WTO: A Cluster ABARE Report, Canberra. Analysis of Country Groups." Trade and Macroeconomics UNCTAD (United Nations Conference on Trade and Develop- Paper 59, IFPRI, Washington, D.C. ment). 1997. "Post-Uruguay Round Tariff Environment for FAO (Food and Agriculture Organization). 2000. "Experience Developing Country Exports." UNCTAD/WTO Joint Study, with the Implementation of the Uruguay Agreement on TD/B/COM.1/14, October. UNCTAD, Geneva. Agriculture." CCP 01/11, December. FAO, Rome. ------. 1999. "Quantifying the Benefits Obtained by Developing ------. 2002a. "A Special Agricultural Safeguard (SAS): But- Countries from the Generalized System of Preferences."UNC- tressing the Market Access Reforms of Developing Coun- TAD/ITCD/TSB/Misc. 52, October 7. UNCTAD, Geneva. tries." In Selected Issues Relating to the WTO Negotiations. UNCTAD/WTO. 1997. "Market Access Developments since the Rome: FAO. Uruguay Round: Implications, Opportunities and Chal- ------. 2002b. "Towards Improving the Operational Effective- lenges, in Particular for Developing Countries, and Least ness of the Marakesh Decision." In Selected Issues Relating to Developed Countries, in the Context of Globalization and the WTO Negotiations. Rome: FAO. Liberalization." UNCTAD, Geneva. Green, D., and S. Priyadarshi. 2002. "Proposal for a `Develop- WTO (World Trade Organization) 1994."Decision on Measures ment Box' in the WTO Agreement on Agriculture." Paper in Favor of Least-Developed Countries." LT/UR/D-1/3, presented at FAO Round Table on Special and Differential 15/04/1994. Office of the Secretary General, Geneva. Treatment in the Context of the WTO Agreement on Agri- ------. 1998."Market Access for Exports of Goods and Services culture. Geneva, February 1. from the LDCs: Barriers and Constraints." December, Karsteny, G., and S. Laird. 1987. "The Generalized System of WT/COMT/ LDC/W/11/REV.1. Office of the Secretary Gen- Preferences: A Quantitative Assessment of the Direct Trade eral, Geneva. Effects and of Policy Options." Discussion Paper No.18, ------. 2000a­b. "Agreement on Agriculture: Special and Dif- UNCTAD, Geneva. ferential Treatment and a Development Box." Proposal by Kessie, E. 2000. "Enforceability of the Legal Provisions Relating Cuba and others. June 23, G/AG/NG/ W/13 and G/AG/NG/ to Special and Differential Treatment under the WTO Agree- W/14. Office of the Secretary General, Geneva. ments." Paper presented at WTO Seminar on Special and ------. 2000c."Special and Differential Treatment of Developing Differential Treatment of Developing Countries. 7 March, Countries in World Agricultural Trade." Submission by Geneva. ASEAN. November 10, G/AG/NG/W/55. Office of the Secre- Michalopoulos, C. 2000. "Special and Differential Treatment for tary General, Geneva. Developing Countries in the GATT and the WTO." Policy ------. 2000d. "WTO Negotiations on Agriculture: Proposals Research Working Paper 2388, World Bank, Washington, by Small Island Developing States." December 29, G/AG/ D.C. NG/W/97. Office of the Secretary General, Geneva. ------. 2001. Developing Countries in the WTO. New York and ------. 2000e. "Implementation of Special and Differential Hampshire: Palgrave. Treatment Provisions in WTO Agreements and Decisions." Onguglo, B. P. 1999. "Developing Countries and Trade Prefer- WT/COMTD/W/77. Office of the Secretary General, ences." In M. R Mendoza, P. Low, and B. Kotschwar, eds., Geneva. Trade Rules in The Making. Washington, D.C.: Brookings. ------. 2001a. "WTO Negotiations on Agriculture: Proposals Otsuki, T., J. S. Wilson, and M. Sewadeh. 2000. "What Price Pre- by India." January 15, G/AG/NG/W/102. Office of the Secre- caution? European Harmonization of Aflatoxin Regulations tary General, Geneva. and African Food Exports." Processed. World Bank, Wash- ------. 2001b. "WTO Negotiations on Agriculture: Compre- ington, D.C. hensive Proposal by the Arab Republic of Egypt." March 21, Oyejide, A. 2002. "Special and Differential Treatment." In B. G/AG/NG/W/107. Office of the Secretary General, Geneva. Hoekman, A. Mattoo, and P. English, eds., The Trading Sys- ------. 2002a. "Information on the Utilisation of Special and tem and Developing Countries. Oxford: Oxford University Differential Treatment Provisions." February 7, WT/ Press for the World Bank. COMTD/W/77/Rev.1/Add.4. Office of the Secretary Gen- Ruffer, T., S. Jones, and S. Akroyd. 2002."Development Box Pro- eral, Geneva. posals and Potential Effect on Developing Countries." ------. 2002b. "The Development Box." Non-paper presented Oxford Policy Management, April. by the Dominican Republic, Kenya, Pakistan, and Sri Lanka Stevens, C. 2002a."Extending Special and Differential Treatment at the Special Session of the Committee on Agriculture, Feb- in Agriculture for Developing Countries." Paper presented at ruary 4­8. Office of the Secretary General, Geneva. FAO Round Table on Special and Differential Treatment in ------. 2002c. "Report of the Inter-Agency Panel on Short- the Context of the WTO Negotiations on Agriculture." Term Difficulties on Financing Normal Levels of Commer- Geneva, February 1. cial Imports of Basic Foodstuffs." June 28, WT/GC/62 and ------. 2002b. "The Future of Special and Differential Treat- G/AG/13. Office of the Secretary General, Geneva. ment for Developing Countries in the WTO." Background ------. 2003. "Negotiations on Agriculture, `First Draft of paper prepared for IDS seminar, May 27­28. Modalities for the Further Commitments' (revised)." March Tangermann, S. 2001. "The Future of Preferential Trade 18, TN/AG/W/1.Rev. 1. Office of the Secretary General, Arrangements for Developing Countries and the Current Geneva. Round of WTO Negotiations in Agriculture." Processed. Youssef, H. 1999. "Special and Differential Treatment for Devel- April, FAO/ESCP, Rome. oping Countries in the WTO." Working Paper 2, South Cen- tre, Geneva. 15 special trade arrangements to improve market access Helen Freeman Introduction The Plan of This Chapter The General Agreement on Tariffs and Trade (GATT) This chapter outlines the GATT/WTO rules relating and the World Trade Organization (WTO) agree- to trade preferences, provides a brief background of ments have provisions that acknowledge a need to trade preferences, examines the trade preference improve developing countries' access to world mar- schemes in the United States (U.S.) and the Euro- kets. In doing so, GATT and WTO provisions provide pean Union (EU), examines preferential aspects of for the formation of customs unions or free trade several trade agreements, examines the benefits and agreements and preferential treatment for developing problems associated with trade preferences, and countries. These are exceptions to the most-favored addresses some options for developing countries on nation (MFN) treatment principle and most action is preferential access in the trade negotiations. taken outside of the GATT/WTO. A major concern about current efforts to negotiate trade agreements is Background that these are a distraction from multilateral efforts to further liberalize trade. Most developing countries Multilateral, regional, and bilateral trade agree- have preferential access to industrial-country markets ments by their nature provide for preferential access through tariff preference schemes, but benefits are to members' markets--hence the importance of tar- limited or have been difficult to obtain for their key iffs in any agreement. Both the GATT (negotiated in products (agriculture, textiles, and apparel). The new 1947) and the WTO (its successor since 1995) are trade negotiations--the Doha Round--will result in based on the MFN obligation of "nondiscriminatory" further trade liberalization, including changes to treatment of all members. This requires that the market access and tariffs. These changes will affect same trading conditions apply to all members (that the current benefits that developing countries receive is, reciprocity). The MFN obligation precludes spe- under preferential schemes. Developing countries cial trading arrangements. will want to consider the effects of such reform on From its inception, the GATT allowed member preferential arrangements but will need to recognize countries to conclude customs unions and free that these arrangements are autonomous trade policy trade agreements under Article XXIV. During the instruments that are not subject to mandates result- 1947 discussion on the Havana Charter a statement ing from negotiations in the WTO. by Lebanon mentioned free trade areas. In 1958 291 292 Agriculture and the WTO European countries, in the Treaty of Rome, entered and the Caribbean Community (CARICOM 1973). into a customs union under GATT Article XXIV. Other examples of bilateral agreements are those WTO members may establish free trade areas between Australia and New Zealand (Australia- (FTAs) within which the duties and other restric- New Zealand Closer Economic Relations Trade tive regulations on commerce (unless not permit- Agreement--ANZCERTA 1987) and between ted within the WTO rules) are eliminated on Canada and Chile (1997)1 (see box 15.1). In the substantially all trade among member countries. period 1948­94, the GATT received 124 notifica- FTAs grant more favorable conditions (that is, pref- tions on RTAs/FTAs and since 1995 over an addi- erences) on trade with the other partners and are tional 100 arrangements have been notified to the inherently discriminatory in nature, in contrast to WTO. The majority fall under GATT Article XXIV, WTO agreements. FTAs are permitted under GATT 19 are under the 1979 Enabling Clause (see below), 1994, Article XXIV and WTO GATS Article V (the and fewer than 20 are under GATS Article V (WTO General Agreement on Trade in Services provision Website on Regionalism, 2003). relating to preferential agreements in services). The GATT/WTO rules do provide for exceptions Understanding on the Interpretation of Article from MFN obligations and since 1979, the XXIV of GATT 1994, an outcome of 1986­94 Enabling Clause has provided a permanent excep- Uruguay Round discussions on problems associated tion so that developed countries "may accord dif- the Article, applies to FTAs. ferential and more favorable treatment to Most WTO Members are now parties to regional developing countries" through "a system of gener- trade agreements (RTAs) and FTAs, which differ in alized, nonreciprocal preferences," usually referred membership, scope, and coverage. In 1996, the to as a Generalized System of Preferences (GSP). WTO Committee on Regional Trade Agreements These schemes are provided outside of the was set up to examine regional groups to assess their GATT/WTO structure and rules. Preferential treat- consistency with the WTO rules. Countries may join ment (such as lower tariffs) was the trade policy a RTA or FTA for a number of reasons: economic, instrument used by developed countries mainly political, or a combination of both. On the eco- for manufactures and semimanufactures, to help nomic side, the expansion of domestic markets after developing countries overcome their dependence barriers have been removed can create economies of on exports of raw materials with unfavorable long- scale. Regional and bilateral FTAs can result in both term price trends and pronounced short-term trade creation and trade diversion, and can raise quantity and price fluctuations. Temperate agri- world welfare if trade creation outweighs trade cultural products have not been part of preferen- diversion. Politically, by joining a group, smaller tial treatment under most GSP schemes as they countries may increase their bargaining power have been excluded or provided only limited access either regionally or multilaterally. An RTA/FTA is (quotas). Tropical products have MFN tariffs at one whereby tariff structures are lowered among relatively low levels--or at zero--and such action members but members maintain their own separate is a result of the 1979 Tokyo Declaration relating to and varying tariffs toward third countries, such as trade in tropical products. in the North American Free Trade Agreement Twenty-seven industrialized countries provide (NAFTA). A preferential trade agreement differs GSP programs, conceived as autonomous trade from an RTA/FTA in that one country or entity con- policy instruments, providing development- fers nonreciprocal preferences on another group of oriented preferences--usually lower tariffs--to countries, such as the EU and the Lomé Convention/ developing countries. WTO Members granting Cotonou Agreement. GSP preferences under the Enabling Clause include Regional and free trade agreements can take dif- Australia (the first country to introduce a GSP ferent forms and be regional in scope such as NAFTA scheme in 1966), Belarus, Canada, the Czech (1994) or nonregional such as between the U.S. and Republic, the EU, Hungary, Japan, New Zealand, Israel (also a bilateral agreement), and the EU agree- Norway, Poland, the Slovak Republic, Switzerland, ment with Mexico--or they can be customs unions and the U.S. Under GSP schemes, the beneficiaries, such as the European Communities (1958), the products, and type of preferences granted vary for Southern Common Market (MERCOSUR 1991), each donor country.2 Special Trade Arrangements to Improve Market Access 293 BOX 15.1 Trade Agreements: AFTA to SADC AFTA (ASEAN Free Trade Area): Brunei Europe Agreements: The EU has con- Darussalam, Indonesia, Lao PDR, Malaysia, cluded these with Bulgaria, Czech Republic, Myanmar, Philippines, Singapore, Thailand, Estonia, Hungary, Latvia, Lithuania, Poland, Vietnam. Romania, Slovak Republic, and Slovenia. ANZCERTA (Australia­New Zealand Euro-Med (Euro-Mediterranean Associa- Closer Economic Relations Trade tion Agreements) (First-generation): The Agreement): Australia and New Zealand. European Union has concluded these with APEC (Asia Pacific Economic Co- Cyprus, Malta, and Turkey. operation Forum): Australia, Brunei Darussalam, Euro-Med (Euro-Mediterranean Associa- Canada, Chile, China, Hong Kong (China), tion Agreements): The European Union has Indonesia, Japan, Republic of Korea, Malaysia, concluded these with Tunisia, Israel, Morocco, Mexico, New Zealand, Papua New Guinea, Peru, and the Palestinian Authority. Philippines, Russian Federation, Singapore, Taiwan Euro-Med (Euro-Mediterranean Co- (China), Thailand, U.S., Vietnam. operation Agreements): The European CARICOM (Caribbean Community): Union has concluded these with Algeria, Egypt, Antigua and Barbuda, Bahamas, Barbados, Belize, Jordan, Lebanon, and Syria. Dominica, Grenada, Guyana, Haiti, Jamaica, St. FTAA (Free Trade Area of the Americas): Kitts and Nevis, St. Lucia, St. Vincent and the Antigua and Barbuda, Argentina, Bahamas, Grenadines, Suriname, Trinidad and Tobago. The Barbados, Belize, Bolivia, Brazil, Canada, Chile, Bahamas does not participate in the common Colombia, Costa Rica, Dominica, Dominican market and Haiti is not yet a full member. Republic, Ecuador, El Salvador, Grenada, CEFTA (Central European Free Trade Guatemala, Guyana, Haiti, Honduras, Jamaica, Agreement): Bulgaria, Czech Republic, Hungary, Mexico, Nicaragua, Panama, Paraguay, Peru, Poland, Romania, Slovak Republic, Slovenia. St. Lucia, St. Kitts and Nevis, St. Vincent and COMESA (Common Market for Eastern Grenadines, Suriname, Trinidad and Tobago, and Southern Africa): Angola, Burundi, Uruguay, U.S., República Bolivariana de Comoros, Democratic Republic of Congo, Djibouti, Venezuela. Egypt, Eritrea, Ethiopia, Kenya, Madagascar, Group of Three: Colombia, Mexico, Malawi, Mauritius, Namibia, Rwanda, Seychelles, República Bolivariana de Venezuela. Sudan, Swaziland, Zambia, Zimbabwe. MERCOSUR (Mercado Común del EEA (Agreement on the European Sur/Southern Common Market Agreement): Economic Area): Austria, Belgium, Denmark, Argentina, Brazil, Paraguay, Uruguay. Finland, France, Germany, Greece, Iceland, NAFTA (North American Free Trade Ireland, Italy, Liechtenstein, Luxembourg, the Agreement): Canada, Mexico, U.S. Netherlands, Norway, Portugal, United SAARC (South Asian Association for Kingdom, Spain, Sweden. Regional Co-operation): Bangladesh, Bhutan, EFTA (European Free Trade Association): India, Maldives, Nepal, Pakistan, Sri Lanka. Iceland, Liechtenstein, Norway, Switzerland. SADC (Southern African Development EU (European Union): Austria, Belgium, Community): Angola, Botswana, Democratic Denmark, Finland, France, Germany, Greece, Republic of Congo, Lesotho, Malawi, Mauritius, Ireland, Italy, Luxembourg, the Netherlands, Mozambique, Namibia, Seychelles, South Africa, Portugal, United Kingdom, Spain, Sweden. Swaziland, Tanzania, Zambia, Zimbabwe. Source: OECD 2002. In May 2000, the U.S. enacted the African (LDC) products, excluding arms, and in April 2001, Growth and Opportunity Act (AGOA), and added Japan provided an additional list of LDC industrial more products benefiting from lower tariffs to its products for duty- and quota-free treatment. GSP list for designated sub-Saharan African coun- The Enabling Clause has also been used by tries. In September 2000, Canada enlarged its GSP developing countries to enter interregional or product coverage. In March 2001, the EU granted global agreements that include the reduction or duty-free access for all least-developed-country elimination of tariffs and nontariff barriers (NTBs) 294 Agriculture and the WTO on trade among themselves. The 1977 formation the Lomé IV Convention) in February 2002, the of ASEAN and the 1979 Global System of Trade parties to the agreement obtained a waiver from Preferences occurred under the Enabling Clause WTO members from obligations under Article 1:1 (see box 15.2). The Enabling Clause also provided of GATT 1994, once opposition from a number of that countries identified as LDCs by the United WTO members was withdrawn in the context of Nations be granted more favorable treatment. the Doha Round (see WTO Document These additional concessions for the LDCs allow WT/MIN/DI/IS, November 14, 2001). for discrimination among developing countries of Agriculture is of significant importance to many trading preferences based on economic capabilities developing countries. It is also a sensitive issue for and needs.3 developed countries, and the strength of the oppo- In 1996, WTO members in the Singapore sition to reform of trade in agriculture resulted Ministerial Declaration agreed to take measures in in few disciplines in the multilateral context. favor of LDCs "including provision for taking posi- This situation prevailed until the 1994 Uruguay tive measures, for example, duty-free access on an Round Agreement on Agriculture (URAA), which autonomous basis, aiming at improving their overall established more comprehensive rules for agricul- capacity to respond to the opportunities offered by tural trade. Developed countries' support for their the trading system."The 2001 Doha WTO Ministerial farmers has resulted in major distortions to world Declaration committed members to working toward agricultural trade. Many tariffs in developed coun- the objective of quota-free and duty-free access as tries for agricultural products are still extremely well as considering additional measures for progres- high, and even higher on "sensitive" products (usu- sive improvement in market access for LDCs (WTO ally sugar and dairy). Given this situation, preferen- document WT/MIN(01)/DE/W/1, paragraph 42). tial arrangements can be potentially valuable. In addition to the GSP and FTA/RTA exceptions, However, the literature on trade preferences and special waivers of MFN or WTO obligations can be in particular on trade preferences in agricul- granted to members with the approval of three- ture is surprisingly limited (Tangermann 2001; fourths of WTO members (GATT 1947 Article Messerlin 2001). XXV:5). The U.S., Canada, and the EU have sought waivers--U.S. and Canada in favor of Caribbean Preferential Trade: GSP Schemes countries and the EU in favor of the Africa, and Trade Agreements Caribbean and Pacific (ACP) countries. Following the signing of the ACP­EU Partnership Agreement A 2001 WTO Secretariat Note on "The Generalized (known as the Cotonou Agreement, which replaced System of Preferences. A Preliminary Analysis of BOX 15.2 A Brief History of the GSP Schemes Action on special treatment for developing coun- of former colonial powers about how to provide tries has been undertaken by both the trade preferences to newly independent devel- GATT/WTO and UNCTAD. In the 1954­55 oping countries. These trade preferences were Review Session of the GATT Contracting Parties seen as instruments of commercial and foreign the adoption of Article XXVII bis, dealing with policy (especially development assistance). tariff negotiations, contains one of the first indica- The Generalized System of Preferences (GSP) tions of differential and nonreciprocal treatment was proposed in 1964 by the UNCTAD Secretary for developing countries. With the Kennedy General Dr. Prebisch and subsequently adopted Round in the 1960s and the increasing number at New Delhi at UNCTAD 11(1968)--although of developing countries becoming GATT developing countries had already raised the issue Contracting Parties, work on the issue of nonrec- in the GATT. In 1971 in order to allow the GSP iprocity led to the 1964 addition to GATT, 1947 system to become legally operational, the GATT of Part IV on Trade and Development. This period Contracting Parties decided to waive Article 1 for and action coincided with the decolonization a period of 10 years and then permanently in period especially in Africa and Asia and concerns 1979 through the Enabling Clause. Special Trade Arrangements to Improve Market Access 295 the GSP Schemes in the Quad" (in other words, ing countries and U.S. businesses, and may be a fac- Canada, the EU, Japan, and the U.S.) comments on tor in the program's utilization. The program pro- the lack of data available on the application of GSP vides better access to the U.S. market for a range of schemes. The note provides a brief description of products imported from over 150 GSP-eligible the GSP schemes for the Quad countries, which developing economies (countries, territories, and imported about three-quarters of all LDC exports, associations). Least-developed countries are specif- with the EU and the U.S. accounting for 63 percent. ically designated in U.S. legislation (see table 15.1). Western Europe and North America account for 88 In 2001 the U.S. imported products were worth percent of LDC exports of manufactured goods but approximately US$16 billion under the program. only 45 percent of agricultural exports. Statistics The leading beneficiaries include Angola, Thailand, demonstrate that only a few countries have bene- Brazil, Indonesia, and India, with a concentration of fited from the GSP programs as a strong concentra- benefits to imports of manufactures and semiman- tion of benefits has gone to developing countries ufactured goods (in other words, in accordance with with relatively large and diversified economies, the intent of the program and its emphasis on man- including substantial manufacturing sectors. ufactures). Some agricultural food products and China is the leading beneficiary of schemes in sugar--with an imported value of US$718 million Canada, the EU, and Japan: it is excluded from the and US$276 million, respectively--benefit from U.S. scheme. There are no LDC countries among GSP status (see tables 15.2 and 15.3). the top 20 GSP suppliers to Canada. Bangladesh Eligibility for GSP status of both countries and is the only LDC in this category to the EU, and goods changes from time to time and this is seen in Mauritania the only LDC in this category to Japan the U.S. reaction through its GSP program to the (WTO 2001). Given the importance of the U.S. and economic crisis in Argentina during 2001­02 (see the EU economies, their preferential schemes are box 15.3). During 2000­01 Belarus and Ukraine provided in some detail below. The United Nations were suspended; in 2002 Malta and Slovenia were Conference on Trade and Development (UNCTAD) graduated; and in 2000 Nigeria and Georgia were Web site of the Generalized System of Preferences designated as Beneficiary Developing Countries. contains the Handbooks of the countries providing The 2001 graduation threshold was US$9,266 GSP preferences. (1999 GDP per capita) (see box 15.4). The original GSP list contained 4,650 articles that are duty-free for all eligible GSP beneficiaries; The United States' Generalized more than 10,000 products are imported into System of Preferences the U.S. In 1996, a second list added approximately The Office of the U.S. Trade Representative 1,770 duty-free articles for the Least-Developed (USTR), the key U.S. government body responsible Beneficiary Developing Countries (LDBDC), for trade negotiations, described the GSP as "an defined by the World Bank as those with annual important element of U.S. efforts to increase eco- incomes below US$786 (1996 GNP per capita). nomic development through the expansion of After the signing of the AGOA in 2000, the law trade opportunities" as all imports of GSP eligible expanded GSP benefits to eligible countries in sub- articles are duty-free (see USTR 2002c, 2002e). The Saharan Africa by providing GSP status on an addi- U.S. Congress provides the legislative authority for tional 1,800 products. the government to carry out the GSP program. Not all articles are eligible for duty-free access. Enacted in 1975, it has been renewed eight times Prohibited articles include most textiles, watches, over the past 25 years and it is now extended footwear, and others. However, GSP eligibility for six through December 31, 2008 by the Trade Act of categories of textile products applies when the GSP 2002. However, six of those eight renewals occurred beneficiary signs an agreement with the U.S. certi- during the past eight years of the program's exis- fying that products are handmade. Such agreements tence. From September 30, 2001 until its August have been signed with at least 15 countries including 2002 renewal there was no legislative authority for Botswana, Egypt, Malta, Romania, Tunisia, and the program. The stop-and-start nature of the pro- Thailand. Additional product exclusions include any gram has produced uncertainty for both develop- article that is determined to be import-sensitive (such 296 Agriculture and the WTO TABLE 15.1 U.S. GSP Beneficiaries, 1999 (Independent Countriesa) The following independent countries are GSP-eligible beneficiariesb: Albania Democratic Republic Lesotho* Senegal Angola* of Congo* Lithuania Seychelles Antigua and Barbuda Djibouti* Macedonia, Former Sierra Leone* Argentina Dominica Yugoslav Republic of Slovak Republic Armenia Dominican Republic Madagascar* Slovenia Bahrain Ecuador Malawi* Solomon Islands Bangladesh* Egypt Mali* Somalia* Barbados El Salvador Malta South Africa Belarus Equatorial Guinea* Mauritius Sri Lanka Belize Estonia Moldova Suriname Benin* Ethiopia* Morocco Swaziland Bhutan* Fiji Mozambique* Tanzania* Bolivia Gambia, The* Namibia Thailand Bosnia and Ghana Nepal* Togo* Herzegovina Grenada Niger* Tonga Brazil Guatemala Oman Trinidad and Tobago Bulgaria Guinea* Pakistan Tunisia Burkina Faso* Guinea-Bissau* Panama Turkey Burundi* Guyana Papua New Guinea Tuvalu* Cambodia* Haiti* Paraguay Uganda* Cameroon Honduras Peru Ukraine Cape Verde* Hungary Philippines Uruguay Central African Republic* India Poland Uzbekistan Chad* Indonesia Romania Vanuatu* Chile Jamaica Russia Venezuela, R.B. de Colombia Jordan Rwanda* Republic of Yemen* Comoros* Kazakhstan St. Kitts and Nevis Zaire Congo Kenya St. Lucia Zambia* Costa Rica Kiribati* St. Vincent and Zimbabwe Cote d'Ivoire Kyrgyzstan the Grenadines Croatia Latvia Sao Tome and Czech Republic Lebanon Principe* * Least-Developed Country (LDC). a. U.S. GSP Beneficiaries also include nonindependent countries and territories and Associations of Countries. b. Beneficiaries are added or deleted from the list and as of June 2000, Botswana, Gabon, Mongolia, and Western Samoa are now beneficiaries and in January 2002, Malta and Slovenia were graduated. Source: Adapted from USTR, U.S. Generalized System of Preferences Guidebook, Office of the United States Trade Representative, Executive Office of the President, Washington, D.C., March 1999 and UNCTAD Web site's GSP Handbook on the Scheme of the United States of America. as steel, glass, and electronics). When imports exceed The GSP program's annual review by the U.S. the competitive need-limitation (CNL) level they government involves the GSP Subcommittee of the no longer remain eligible for duty-free treatment. Trade Policy Staff Committee at the Office of This provision enables GSP benefits to apply to less USTR. Working to an annual timetable, this sub- competitive foreign producers by limiting access by committee considers petitions to modify the eligi- the more competitive Beneficiary Developing bility of articles and countries for duty-free Countries. treatment (see box 15.5). Special Trade Arrangements to Improve Market Access 297 TABLE 15.2 Leading Sources of U.S. GSP Imports, 2000 Beneficiary Total U.S. Imports Share of U.S. Imports Using GSP Developing (US$ Millions) (percent) Country (BDC) Duty-Free U.S. GSP-Eligible All Products GSP-Eligible All Products Angola 2,844 3,010 3,343 94.5 85.1 Thailand 2,205 4,130 16,301 53.4 13.5 Brazil 2,086 4,990 13,732 41.8 15.2 Indonesia 1,369 2,723 10,322 50.3 13.3 India 1,138 2,745 10,680 41.5 10.7 Philippines 745 1,545 13,943 48.2 5.3 Venezuela 745 11,353 17,429 6.6 4.3 South Africa 583 1,105 4,204 52.8 13.9 Russia 515 1,797 7,761 28.7 6.6 Turkey 435 853 3,027 51.0 14.4 Chile 419 1,524 3,258 27.5 12.9 Kazakhstan 326 393 431 83.0 75.6 Hungary 318 1,023 2,711 31.1 11.7 Poland 317 578 1,040 54.8 30.5 Czech Republic 280 470 1,069 59.6 26.2 Argentina 218 2,089 3,095 10.4 7.0 Congo (Kinshasa) 174 178 212 97.8 82.1 Equatorial Guinea 136 144 155 94.4 87.7 Sri Lanka 122 149 2,002 81.9 6.1 Slovenia 120 160 314 75.0 38.2 Total, Top 20 BDCs 15,095 40,959 115,029 36.9 13.1 Total, All BDCs 16,439 60,730 172,366 27.1 9.5 Source: The Trade Partnership 2001, Derived from U.S. Census data. TABLE 15.3 Leading Product Groups Imported by the U.S. Duty-Free under GSP, 2000 Share of Total Value GSP Imports Value or Products (US$ Millions) (percent) Duties Saved Oils and petroleum products 3,151 19.2 8.9 Electrical equipment and parts 1,712 10.4 56.3 Transportation equipment parts 1,149 7.0 29.0 Jewellery and parts 983 6.0 56.6 Machinery (including computers), parts 849 5.2 28.5 Organic chemicals 747 4.5 41.6 Agricultural food products (excl. sugar) 718 4.4 41.8 Plastics and plastic products 602 3.7 31.1 Wood and wood products 533 3.2 23.0 Iron and steel raw materials 467 2.8 13.7 Aluminum mill products 424 2.6 14.5 Rubber products 418 2.5 14.6 Iron and steel products 405 2.5 13.2 Sugar 276 1.7 8.2 Leather products 179 1.1 5.1 Furniture and parts 6 0.0 0.4 Total, Leading Products 12,617 76.8 386.5 Total, All GSP Products 16,439 100.0 555.3 Source: The Trade Partnership 2001, Derived from U.S. Census data. 298 Agriculture and the WTO The U.S. and the Caribbean Basin Initiative (CBI) entered the U.S. was greatest for CBERA and CBTPA U.S. trade preferences concerning the Caribbean, in 2002, with over one-third entering under these authorized by Congress and collectively known as programs (see table 15.4). the Caribbean Basin Initiative (CBI), have been used Benefits are conditioned on compliance with to facilitate the economic development and export legislated eligibility criteria such as labor rights. diversification of 24 Caribbean Basin and Central Conditionality provisions are used by the U.S. to American economies. From its 1994 inception, the bring about change on issues with CBTPA benefici- Caribbean Basin Economic Recovery Act (CBERA) aries. Guatemala provides an example of both the provided certain CBI goods--such as coffee, use of criteria and compliance. In mid-2000, the bananas, and minerals--open access to the U.S. U.S. government conducted an extensive review of market. These benefits were substantially expanded each beneficiary country. It suspended its review of by the Caribbean Basin Trade Partnership Act Guatemala's labor practices when, in May 2001, the of 2000 (CBTPA) with new preference provisions, Guatemalan government passed important labor including expanded benefits for apparel. By 2000, law reforms. manufactured products (such as apparel and electri- CBTPA (like its predecessor CBERA) applies to cal and nonelectrical machinery) amounted to over many of the same tariff categories covered under the 50 percent of CBI exports to the U.S. The value of U.S. GSP but it is broader in terms of additional the different schemes under which CBI products products and more liberal qualifying rules. Its duty- BOX 15.3 The U.S. Generalized System of Preferences (GSP) and Argentina's Economic Crisis, 2001­02 The U.S. government uses its GSP scheme to upon reaching a statutorily defined threshold assist countries undergoing difficult economic level of the U.S. market share. circumstances and in 2002 used it to help The petitioning process for the restoration of Argentina. Following an August 2002 presiden- product eligibility requires public notification in tial proclamation, the eligibility for 57 products the Federal Register by the Office of United (leather goods, nonsensitive agricultural prod- States Trade Representative (USTR), providing ucts, industrial chemicals, and others), which for a comment period and a date for the results accounted for over US$126 million in exports in of the petitioning. In 2002, Argentina, the 2001, was restored as products eligible for pref- Philippines, and Turkey petitioned the USTR to erential treatment. Previously they had been review 17 products and add these to the list of covered by the GSP program but were removed products eligible for GSP treatment. TABLE 15.4 U.S. Imports from CBERA Countries, Total and Under-Selected Import Programs, MFN-Free, GSP, and CBERA 1998 2002 Import 1998 (percentage 2002 YTD (percentage program (US$1,000) of total) (US$1,000) of total) CBERA 3,224,564 18.8 1,818,366 12.8 CBTPA 0 0.0 3,469,607 24.4 GSP 195,407 1.1 150,031 1.1 MFN-free 3,742,325 21.9 3,753,970 26.4 Subtotal 7,162,296 41.8 7,37,3608 64.7 Other 9,961,985 58.2 6,870,759 35.3 Total 17,124,281 100.0 14,244,367 100.0 Source: USTR Fourth Report to Congress on the Operation of the Caribbean Basin and Economic Recovery Act, December 31, 2001. Special Trade Arrangements to Improve Market Access 299 BOX 15.4 U.S. Generalized System of Preferences (GSP): Criteria and Conditions In deciding the eligibility of an import for GSP 2000, 128 concerned workers' rights. Failure treatment, the U.S. considers three main criteria. to meet this criteria led to temporary or per- The article must be: manent suspension of GSP privileges for Chile, Maldives, Mauritania, Paraguay, Sudan, · from a designated beneficiary country; and Syria and termination for Liberia, · an eligible article as defined for GSP treat- Nicaragua, and Romania (Laird and ment; and Safadi 2000). · meet the rules of origin. Not all beneficiary countries receive duty-free In addition, the following factors and condi- treatment on the entire list of articles. Exclusion tions are taken into account: may occur because of the following: · statement of the requesting country; · GSP imports of the article exceed the com- · benefits to furthering economic development petitive need-limitation (CNL) (based on dol- of a beneficiary country; lar value) or exceed more than 50 percent of · the anticipated impact on comparable U.S. total U.S. imports of that product; producers; · the country has been graduated from the · the extent of competitiveness of the benefici- program (per capita GNP exceeds the ary country's eligible products; threshold income level set for high-income · comparable GSP action by other major devel- countries by the World Bank) or has been oped countries; removed because it is deemed no longer to · market access for U.S. goods and services in be developing; the beneficiary country; · the value-added component in the benefici- · protection of U.S. intellectual property rights; ary country is insufficient to meet the GSP · reduction of trade-distorting investment rule-of-origin requirement (costs of material practices/policies; produced plus processing must equal at least · elimination of trade-distorting export prac- 35 percent of the appraised value of the tices; and product); and · provision for internationally recognized · the country fails to supply complete docu- worker rights (of association, of collective mentation. bargaining); forced or compulsory labor; minimum age for child employment and The President may waive CNL authority but acceptable work conditions regarding, for there are limitations on this waiver. GSP law instance, minimum wages, work hours, and automatically waives all CNLs for GSP beneficiar- occupational and health standards. (Workers' ies designated as LDBDCs. GSP law requires that rights criteria has been the single most com- all GSP articles be imported directly but GSP sta- mon issue cited in practices filed with the tus still applies if the indirect shipment of a prod- GSP Subcommittee. Of the 224 country prac- uct takes place through free trade zones or tices petitions filed with USTR from 1985 to entrepôts. free provisions have declined since 1998 as a propor- The U.S. and Andean Trade Preferences tion of total goods from the region.This corresponded U.S. trade preferences concerning this region have with an increase in the proportion (over 25 percent) been provided for strategic reasons as evident in the of CBI imports entering as MFN duty-free goods, 1991 Andean Trade Preference Act (ATPA) for reflecting the liberalization of U.S. tariffs under its Colombia, Peru, Ecuador, and Bolivia, reauthorized Uruguay Round commitments. Apparel, previously in 2002 as the Andean Trade Promotion and Drug excluded from preferential treatment in 2000, Eradication Act (ATPDEA). Under the new act, accounts for 43 percent of total exports. Other Ecuador was not then eligible for preferences. significant exports are cigars, mechanical and electri- In return for cooperating with the U.S. in controlling cal equipment, fruits, and coffee. 300 Agriculture and the WTO BOX 15.5 U.S. Generalized System of Preferences: Annual Timetable Action Target Date Deadline for the acceptance of review petitions June 1 Announcement for the petitions accepted for review July 12 Public hearings and submissions of written briefs September/October U.S. International Trade Commission (USITC) report on the economic December impact of petitioned actions Public comment on USITC report December/January Publication of the early warning list showing statistics over a 10-month February/March period of GSP imports that were close to exceeding the CNL Announcement of results of the reviews April 1 Effective date of changes July 1 Source: Office of the United States Trade Representative (USTR). narcotics production and trade, the U.S. offers sub-Saharan Africa with the African Growth and duty-free and duty-reduced access to the U.S. mar- Opportunity Act (AGOA) for the (then) 34 benefici- ket for eligible products (numbering 5,600). These aries, by expanding GSP benefits for eligible coun- countries benefit from both ATPDEA and GSP tries. AGOA increased preferential access of preferences. In Congress's reauthorizing legislation, beneficiaries' exports in key areas such as clothing. existing benefits were extended and expanded and Beneficiary countries receive duty-free and quota- duty-free access was provided for products not cov- free access for almost all products for eight years, ered initially such as apparel, footwear, and tuna. without the GSP competitive needs limit applying to Renewal also resulted in more conditions being their exports. An additional 1,800 products, includ- imposed, ranging from labor rights and intellectual ing selected apparel articles subject to special provi- property rights to resolution of commercial dis- sions (rules of origin and customs requirements), not putes. The U.S. government, following a public available to other GSP countries, have been added to comment period, determined that all countries GSP benefits. Substantially all products from sub- (including Ecuador) were eligible for the expanded Saharan Africa are now eligible to enter the U.S. duty- benefits. free. The U.S. imported US$8.2 billion of duty-free In 1999, some two-thirds of U.S. imports from goods in 2001 under AGOA, representing about the then­ATPA countries were already duty-free, 40 percent of U.S. imports from sub-Saharan Africa. with most entering under provisions other than Approximately 17.7 percent of all imports in 2000 APTA. APTA beneficiary products include fresh-cut were covered under the GSP program. Trans- flowers (Colombia accounts for close to half of all portation equipment (over 40 percent), miscella- U.S. imports), asparagus (Peru is the second-ranking neous manufactures (nearly 40 percent), and supplier), and processed tuna (not in cans, with electronic products (37 percent) are the sectors bene- Ecuador as a leading provider). In 1999 the portion fiting the most from GSP benefits. of exports that qualified exclusively under APTA AGOA benefits are conditioned on a number of was less than 10 percent. Of the region's products, legislative requirements. In addition to market- coffee (Colombia, Ecuador), shrimps and bananas based economy provisions such as eliminating (Ecuador) are unconditionally free of duty under trade barriers and efforts to combat corruption, the general GSP tariff rates (see Kornis 2000). there are requirements relating to U.S. national security, human rights, and international terror- ism. The U.S. government conducts reviews of the The U.S. and the African Growth GSP program and in 2000, Nigeria and Eritrea were and Opportunity Act (AGOA) added to the list of eligible countries while in 2001, In 2000 Congress established a new framework for a review of worker rights in Swaziland was termi- U.S. trade, investment, and development policy in nated because the U.S. had found that Swaziland Special Trade Arrangements to Improve Market Access 301 conformed to these GSP program requirements on highest in the fish and fruits categories with tariffs worker rights (see table 15.5). for the former ranging between 9 and 23 percent and for the latter from 9.0 to 118.1 percent (see table 15.6). European Union and Its The potential application of very high MFN tariffs Preferential Agreements enforces minimum import requirements and ensures The EU's GSP and preferential arrangements "form that imports do not exceed tariff rate quota (TRQ) a mosaic of tariffs, quotas and other restrictions on amounts. Most of the agricultural TRQs are part of EU agricultural imports" (see Hasha 2002). The EU preferential trading arrangements. EU concessions is the direct source of 40 percent of all preferential on agriculture remain limited (sugar, bananas, beef, trade agreements notified to the WTO and because and rum) as imports remain restricted to amounts there is convincing evidence of trade diversion on consistent with the Common Agricultural Policy's both import and export sides, future EU policy on (CAP) internal price objectives. The unconditional these agreements is relevant to the work of the opening of markets to LDCs under the EBA policy WTO (Messerlin 2001) (see box 15.6). was possible because their limited export agricultural In 1963 the European Communities and 18 newly potential in temperate products represented little independent countries (former colonies of three EC threat to EU interests and domestic pressure groups. member states) signed the First Yaounde Convention EU preferential trading arrangements do not create resulting in 18 de facto bilateral, nonreciprocal free trade but determine the source of imports. trade agreements. Since then, the EU has provided The EU's current GSP expires in 2004. Product special nonreciprocal tariff reductions for 77 ACP exclusion applies to certain beneficiaries and sec- countries under the Lomé Convention. This was tors. The 65 countries benefiting only from GSP renegotiated in 2000 as the Cotonou Agreement. preferences are the transition economies and some Further unrestricted duty-free access for LDC coun- Asian and Latin American countries, including tries for all products, excluding arms, has been pro- nine LDCs. Nine countries including Brazil and vided under the 2001 "Everything But Arms" China have graduated from nine sectors of the EU Agreement (EBA). (There are 48 LDC countries on economy; Brazil, Argentina, Malaysia, and Thailand the UN list and 39 are ACP countries. The non-ACP have lost preferences on specific agricultural com- LDCs are Yemen, Afghanistan, Bangladesh, Maldives, modities; and the Republic of Korea and Taiwan Nepal, Bhutan, Myanmar, Lao People's Democratic (China) have lost all preferences. Thailand and Republic, and Cambodia. For Myanmar all GSP and India lodged complaints in the WTO on the condi- EBA preferences are suspended). The EBA was ini- tions under which the EU grants tariff preferences tially proposed by the European Communities in to developing countries under the GSP scheme 1997 in the preparations for the 1999 WTO Seattle (that is, the preferential discrimination). The Ministerial Conference. Owing to domestic pressure European Communities settled a related WTO from agricultural groups, the proposal was amended complaint by Brazil (WTO document WI/DS/209) to provide for longer transition periods with the on the form of a tariff-free quota on soluble coffee. phasing-out of duty on bananas by 2006 and sugar In 2001 the EU revised its GSP scheme to sim- and rice by 2009. (See Page and Hewitt [2002], which plify the rules, to harmonize procedures on differ- argues that the EBA is trade diverting,discriminatory, ent arrangements, and to enhance the predictability and is provided for essentially political--and not relating to the eligibility of countries and sectors. development--motives.) Revisions addressed the issue of erosion of tariff In 1971, the EU introduced its Generalized preferences by applying "preference modulation" to Scheme of Tariff Preferences. It now applies to most two categories rather than four under the previous of the 143 independent countries and 36 depend- GSP scheme (modulation means that tariff reduc- ent countries and territories. tions vary by product according to the "sensitivity" Most-favored-nation tariffs on EU agricultural of the product for EU producers). Some countries products are high--for example, dairy (ranging receive incentives (that is, larger tariff reductions) from 38.4 to 209.9 percent), cereals (39.2 to 101.1 per- on the condition that they take action on labor, cent),and sugar (21.4 to 114.4 percent). Imports are environmental standards, and drugs. 302 Agriculture and the WTO TABLE 15.5 U.S.­Sub-Saharan Africa Trade: Major U.S. Import Suppliers under the Generalized System of Preferences, and the African Growth and Opportunity Act (YTD Jan.­Jun., AGOA-Eligible Countries Only in US$ Millions) AGOA AGOA Additional Additional including GSP including GSP items for items for provisions of provisions of GSP GSP AGOA AGOA the AGOA Act the AGOA Act 2001 2002 countries countries Country 2001 YTD 2002 YTD YTD YTD 2001 YTD 2002 YTD Nigeria 2,335,079 2,340,602 172 75 2,334,907 2,340,527 Gabon 448,501 608,356 18 121 448,483 608,235 South Africa 395,690 556,955 260,193 233,082 135,497 323,873 Lesotho 4,905 142,208 0 169 4,905 142,039 Madagascar 13,749 61,070 3,987 3,200 9,762 57,871 Mauritius 12,785 55,236 4,203 2,452 8,582 52,784 Kenya 17,887 48,875 1,738 2,243 16,149 46,632 Congo (ROC) 13,385 45,430 65 2,817 13,320 42,614 Swaziland 23 28,986 17 48 7 28,939 Ghana 24,784 22,904 5,839 5,856 18,945 17,048 Cameroon 16,541 22,799 293 142 16,248 22,657 Malawi 16,744 21,757 16,744 1,425 0 20,332 Cote d'lvoire 6,525 15,729 6,525 8,896 0 6,833 Botswana 558 1,881 558 440 0 1,441 Ethiopia 270 1,349 270 705 0 643 Tanzania 206 572 195 286 10 286 Senegal 312 222 312 221 0 0 Namibia 2 136 2 133 0 3 Mali 142 90 142 89 0 1 Sierra Leone 191 43 191 43 0 0 Zambia 724 42 714 18 10 24 Guinea 86 28 86 28 0 0 Mozambique 21 17 21 4 0 13 Mauritania 0 15 0 15 0 0 Uganda 72 15 72 6 0 9 Rwanda 33 10 33 10 0 0 Niger 42 6 42 6 0 0 Benin 0 0 0 0 0 0 Eritrea 0 0 0 0 0 0 Seychelles 2,233 0 2,233 0 0 0 Sao Tome & Prin 0 0 0 0 0 0 Djibouti 0 0 0 0 0 0 Cape Verde 75 0 75 0 0 0 Guinea-Bissau 0 0 0 0 0 0 Cen African Rep 0 0 0 0 0 0 Chad 0 0 0 0 0 0 Total 3,311,565 3,975,333 304,742 262,530 3,006,823 3,712,804 Note: Figures do not add to the totals shown due to rounding. Source: Compiled by the U.S. International Trade Commission 2002. Special Trade Arrangements to Improve Market Access 303 BOX 15.6 Parties to Regional Trade or Preferential EU Trade Agreements in Force as of April 2002 Europe Agreements: Bulgaria, the Czech Republic, Equatorial Guinea, Eritrea, Ethiopia, Republic, Estonia, Hungary, Latvia, Lithuania, Federated States of Micronesia, Fiji, Gabon, Poland, Romania, Slovak Republic, Slovenia. Gambia, Ghana, Grenada, Guinea, Guinea- Association Agreements: Cyprus, Malta, Bissau, Guyana, Haiti, Jamaica, Kenya, Kiribati, Turkey. Lesotho, Liberia, Madagascar, Malawi, Mali, Stabilization and Association Agree- Marshall Islands, Mauritania, Mauritius, ments: Croatia, Former Yugoslav Republic of Mozambique, Namibia, Nauru, Niger, Nigeria, Macedonia (FYROM). Niue Islands, Palau, Papua New Guinea, Euro-Mediterranean Association Agree- Rwanda, St. Christopher and Nevis, St. Lucia, ments: Israel, Morocco, the Palestinian St. Vincent and the Grenadines, Samoa, Sao Authority, Tunisia. Tome and Principe, Senegal, Seychelles, Sierra Cooperation Agreements (Euro-Med Leone, Solomon Islands, Somalia, South Africa, Association Agreements concluded, but not in Sudan, Suriname, Swaziland, Tanzania, Togo, effect, or under negotiation): Algeria, Egypt, Tonga, Trinidad and Tobago, Tuvalu, Uganda, Jordan, Lebanon, Syria. Vanuatu, Zambia, Zimbabwe. Other Free-Trade Agreements: Denmark Autonomous Trade Measures for the (Faroe Islands), Iceland, Liechtenstein, Mexico, Western Balkans: Albania, Bosnia-Herzegovina, Norway, South Africa, Switzerland. the Federal Republic of Yugoslavia, Kosovo. Other Customs Unions: Andorra, San Generalized System of Preferences Marino. (GSP) only: Afghanistan, Argentina, Armenia, Association of Overseas Countries and Azerbaijan, Bahrain, Bangladesh, Belarus, Bhutan, Territories (OCT): Anguilla, Antarctica, Aruba, Bolivia, Brazil, Brunei Darussalam, Cambodia, British Antarctic Territory, British Indian Ocean Chile, China, Colombia, Costa Rica, Cuba, East Territory, British Virgin Islands, Cayman Islands, Timor, Ecuador, El Salvador, Georgia, Guatemala, Falkland Islands, French Polynesia, French Honduras, India, Indonesia, Iran, Iraq, Kazakhstan, Southern and Antarctic Territories, Greenland, Kyrgyzstan, Kuwait, Lao PDR, Libya, Malaysia, Mayotte, Montserrat, Netherlands Antilles, New Maldives, Moldova, Mongolia, Myanmar, Nepal, Caledonia, Pitcairn, Saint Helena, Ascension Island, Nicaragua, Oman, Pakistan, Panama, Paraguay, Tristan da Cunha, South Georgia and the South Peru, Philippines, Qatar, Russian Federation, Sandwich Islands, St. Pierre and Miquelon, Turks Saudi Arabia, Sri Lanka, Tajikistan, Thailand, and Caicos Islands, Wallis and Fortuna Islands. Turkmenistan, Ukraine, United Arab Emirates, EU-African, Caribbean and Pacific Uruguay, Uzbekistan, Republica Bolivariana de (ACP) Partnership: Angola, Antigua and Venezuela, Vietnam, Republic of Yemen, American Barbuda, Bahamas, Barbados, Belize, Benin, Samoa, Bermuda, Bouvet Island, Cocos Islands, Botswana, Burkina Faso, Burundi, Cameroon, Cap Cook Islands, Gibraltar, Guam, Heard and Verde, Central African Republic, Chad, Comoros, McDonald Islands, Macao, Norfolk Island, Northern Congo, Cook Islands, Dem. Rep. of Congo, Cote Mariana Islands, United States Minor Outlying d'Ivoire, Djibouti, Dominica, Dominican Islands, Tokelau Islands, Virgin Islands (USA). Note: Least-developed countries (LDCs) (or economies) in italics. Source: Adapted from WTO Secretariat, based on Developing Country Trade (2001). The EU and Africa, Caribbean, and Pacific beef from six ACP countries and 1.2 million tons of Countries: Cotonou Agreement sugar from 13 other ACP countries and India. Access was further expanded under the EBA initia- Under the Cotonou Agreement, duty- and quota- tive for LDCs. free access is provided for all industrialized prod- ucts and 80 percent of agricultural products. The EU's Euro-Mediterranean Partnership Concessions on agriculture have been limited by quotas or only small tariff reductions on most The EU has provided Mediterranean countries with products supported by the EU's CAP. Of particular preferential market access reflecting long-standing importance are the ACP protocols for beef and trade relationships and important political associa- sugar--specifically, EU imports of 52,100 tons of tions. These countries receive reduced tariffs and 304 Agriculture and the WTO TABLE 15.6 EU: Applied MFN tariffs by HS Chapters 01­22, 2002 (percent and US$ Billions) MFN 2002 Imports Number Simple Standard 2000 HS of 8-digit average Min. Max. deviation (US$ code HS lines Description (%) (%) (%) (%) Billion) Total 10,400 6.4 0 209.9 11.3 807.4 2,354 HS Chapters 01­24 15.9 0 209.9 20.5 56.9 8,046 HS Chapters 25­97 3.8 0 39.8 3.7 750.5 1 55 Live animals 20.6 0 107.8 30.2 0.8 2 236 Meat and edible meat offal 28.3 0 192.2 31.1 2.7 3 324 Fish & crustacean, mollusk & 9.8 0 23 5.7 8.6 other aquatic invertebrate 4 175 Dairy prod.,birds' eggs, 38.4 0 209.9 36.8 1.0 natural honey, edible prod. 5 21 Products of animal origin, n.e.s 0.2 0 5.1 1.1 0.8 or included 6 42 Live trees & other plant, bulb, 6 0 10.9 3.4 1.0 root, cut flowers, etc. 7 108 Edible vegetables and certain 12.7 0 150.1 21.3 2.4 roots and tubers 8 128 Edible fruit and nuts, peel of 9 0 118.1 11.5 7.3 citrus fruit or melons 9 42 Coffee, tea, mate and spices 3.1 0 12.5 4.3 4.7 10 55 Cereals 39.2 0 101.1 27.7 1.4 11 83 Prod. Mill. Indust., malt, 22.2 1.2 84.4 16.8 0.1 starches, inulin, wheat gluten 12 79 Oilseeds, oleaginous, fruit, misc, 1.8 0 52.3 6.2 4.9 grain, seed, fruit, etc. 13 18 Lac, gums, resins & other 2.2 0 19.2 5.1 0.4 vegetable saps & extracts 14 8 Vegetable plaiting materials, 0 0 0 0 0.1 vegetable products 15 128 Animal/vegetable fats & 8.9 0 75.8 13.2 1.7 oils & their cleavage products 16 92 Prep. of meat, fish or 18.5 0 97.2 11.5 2.5 crustaceans, mollusks, etc. 17 47 Sugars and sugar confectionery 21.4 2.1 114.4 24 1.3 18 27 Cocoa and cocoa preparations 11.8 0 68.9 13 1.6 19 51 Prep. of cereal, flour, starch/ 16.4 7.6 49.6 10.5 0.5 milk, pastrycooks' products 20 321 Prep. of vegetables, fruit, nuts 20.6 0 146.9 13.2 2.7 or other parts of plants 21 42 Miscellaneous edible 9.6 0 21.1 5.1 1.0 preparations 22 175 Beverages, spirits and vinegar 5.5 0 58.6 9 2.8 Source: WTO 2002d. valuable import quotas for fruits and vegetables. The Liberation Organization. These association agree- trade objective of the "Euro-Mediterranean Part- ments replace earlier nonreciprocal arrangements nership"launched in 1995 is a free trade area by 2010. and provide for a greater degree of liberalization by Euro-Mediterranean association agreements have the EU's partners than did earlier cooperation agree- been concluded between the EU and Tunisia, Israel, ments and cover concessions on certain agricultural Morocco, Egypt, Jordan, and the Palestinian and fishery products and industrial products. They Special Trade Arrangements to Improve Market Access 305 constitute a progressive liberalization of trade WTO provisions for FTAs provide for free trade through a widening of tariff rate quotas as well. The for substantially all products and no quota restric- liberalization timetable varies for the partners--with tions. The EU's FTA agreements include TRQs for immediate application for the EU and with a transi- sensitive agricultural products that compete with tion period for full implementation by the other part- EU products. Its approach has been that these quo- ners, except for Israel, whose inclusion is immediate. tas are based on a historical relationship and can be The transition period is 15 years for Egypt and maintained without waivers in FTA agreements. 12 years for Morocco and Tunisia (see Guth and The EU/Mexico agreement covers key trade areas, Chomo 2002, and the EC Web site at http://www. market access, rules of origin, sanitary and phy- europa.eu.int/comm/external_relations/euromed/ tosanitary issues, and technical regulations. For index.htm). agricultural and fisheries products covered by the agreement, the EU is to eliminate tariffs for most products by 2008 and Mexico by 2010. Tariff quotas EU Trade Agreements and the WTO apply to certain products. The agreement provides EU agreements such as the Cotonou Agreement for the total liberalization of 95 percent of histori- have been criticized on the grounds of discrimina- cal EU imports, but for agriculture, only 62 percent tion (that is, non-ACP/Mediterranean countries are of historical trade will be fully liberalized. excluded) and product coverage. Hence, the chal- lenges in the GATT and more recently in the WTO Other Trade Arrangements: have been made on the grounds of discrimination AFTA, NAFTA, and APEC and noncompliance with international trade rules. The challenges focused on EU import regimes (for There are other agreements/arrangements that are of bananas) that favored EU distributors over other interest because agriculture is a sensitive area and distributors and former colonies over other coun- treated accordingly, and because of their contribution tries--and the general issues of graduation and tar- to trade liberalization and growth within the region. iff concessions requiring adherence to conditions Countries with FTAs benefit when lower-cost inputs on environment, labor, and drugs. Challengers see from partner countries displace higher-cost goods these provisions as inconsistent with the Enabling manufactured domestically (the trade-creating Clause's provision of general and nondiscrimina- effect). A possible cost is that trade diversion may tory preferences for all developing countries. occur if the removal of tariff barriers causes high-cost Since 1996 the EU has acted to make its arrange- imports from partner countries to replace low-cost ments compatible with WTO rules, by renegotiating imports from the rest of the world. A growing per- and converting current agreements to reciprocal centage of world trade has been created by regional free trade areas. In September 2002, the EU and trade liberalization. After controlling for distance, ACP countries launched negotiations for Economic economic size, and other factors, Frankel finds strong Partnership Agreements (EPAs) to shift to new trade effects for the Association of Southeast Asian Nations arrangements (EU Press Release, Brussels July 12, (ASEAN) countries with trade increasing an esti- 2002). The basic principles and timeframe for EPA mated fivefold for its members. The Andean Pact and negotiations were set out in the Cotonou MERCOSUR are both estimated to increase trade by Agreement. Seventy-six countries are eligible for a factor of 2.5. Intra-EU trade is estimated to be 65 EPAs. South Africa and Cuba are also ACP members percent larger than prior to the formation of the EU but EPAs will not be negotiated with them because because of the common market among members Cuba as is not a signatory to the Cotonou (see Anderson 2001). Agreement and South Africa concluded a Trade, Development, and Cooperation Agreement with the AFTA: The ASEAN Free Trade Agreement EU in 2000. In addition, the EU has FTAs with Mexico (2001) and various non-EU European The ASEAN Free Trade Agreement (AFTA), signed countries (see box 15.6). In 2000, the EU began in 1992, initially defined "free trade" as the reduc- negotiations with MERCOSUR and Chile on sepa- tion of tariffs to within the range of 0­5 percent but rate free trade areas. for "highly sensitive" products such as rice and 306 Agriculture and the WTO sugar the tariff rate was permitted to remain at and pears, cattle and calves, dairy products, higher levels at the end of the implementation and processed potatoes have shown the greatest period. ASEAN members have promoted economic increase. The greatest increase in agricultural cooperation through trade with the objectives of exports from Canada to the U.S. has come from creating a single ASEAN market and with helping beef and veal and wheat products, and from Mexico their members, through greater interregional trade, to the U.S. the items showing the greatest increase strengthen their competitive advantage with the are sugar, peanuts, cattle and calves, tomatoes, and rest of the world. To meet the requirements of wheat products. accession to AFTA, Cambodia and the Lao PDR were required to modernize their trade procedures. APEC: Not a Preferential Trade Agreement AFTA membership and regional linkages have But Still Free Trade by 2020 been a critical factor in growth of manufactures in Asian LDCs. Thailand's economic growth has In the 1994 Bogor Declaration, the Asia Pacific resulted in it being graduated from both the U.S. Economic Co-operation Forum's (APEC's) members and EU GSP programs. The creation of an indus- committed themselves to the establishment of free trial zone along the border between Thailand and trade and investment in the Asia-Pacific region, Cambodia provided Thai manufacturers with through a process based on "open regionalism." The an opportunity to relocate their production and declaration called for a tariff-free region for devel- thereby benefit from low Cambodian labor costs. oped economies by 2010 and for developing Cambodia's preferential market access to the EU economies by 2020. The focus of APEC's nondis- and U.S. markets under the GSP program criminatory approach to regional liberalization is in demonstrates that such an arrangement repre- accordance with its "open regionalism" approach, sents an opportunity for greater economic designed to overcome political considerations among growth in both countries. members. "Open regionalism" came to be under- stood as the gradual reduction of trade barriers on a nondiscriminatory basis; it is explicitly designed to NAFTA: North American Free Trade Agreement not be a preferential trading arrangement--as APEC For the three countries of this agreement--the has provided, throughout its existence, strong sup- U.S., Canada, and Mexico--agriculture has been a port for the WTO multilateral approach to the liber- sensitive subject. NAFTA provides for the retention alization of global trade. A key component of APEC of TRQs with prohibitive out-of-quota tariffs on a is building capacity for liberalization through the number of products, notably bilateral Canada­U.S. pooling of expertise and the development of shared trade in dairy products, U.S. and Mexican exports understandings on the principles of nondiscrimina- of poultry and eggs to Canada, Mexican exports of tion and "standstill"--that is, a commitment not to sugar to both U.S. and Canada, and U.S. maize raise barriers in the future in 15 action areas, includ- exports to Mexico. Canada's free trade agreement ing tariffs (see Scollay 2001). with Chile contains similar exclusions for dairy Within APEC, some members are also parties to products, poultry, and eggs. In these agreements preferential trade agreements with different there are other examples, again almost exclusively approaches to agriculture. For example, the bilateral agricultural, in which the period for phasing out Closer Economic Relations agreement (CER) be- trade barriers substantially exceeds the 10-year tween Australia and New Zealand (ANZCERTA) is period set in the Understanding on the Inter- the only agreement that meets the GATT 1994 Article pretation of Article XXIV of GATT 1994. For most XXIV requirement to cover "substantially all" trade. commodities NAFTA's influence has been relatively small, but for a handful of commodities the agree- Exploring the Benefits ment has had a much larger impact with increased of Preferential Treatment volumes of trade. Among U.S. exports to Canada, wheat products, beef and veal, and cotton have Lower tariffs and quotas provided market access shown the greatest increase while among through the various GSP-type programs. This U.S. exports to Mexico, rice, cotton, fresh apples access was expected to enhance the competitive Special Trade Arrangements to Improve Market Access 307 advantage of developing-country exporters and to and 73 percent for Japan. Utilization rates vary produce faster growth in their economies. Expected among LDCs and can be concentrated; for exam- benefits of trade preferences have been described in ple, Bangladesh was a key LDC exporter to the EU, terms of "hard" economic advantages such as bet- Norway, Canada, Japan, and Switzerland (Laird, ter access to developed-country markets, larger vol- Safadi, and Turrini). The utilization rate can be umes of exports, higher prices earned, improved affected by a number of factors such as capacity to economic welfare, more jobs, and more rapid eco- supply (that is, productive capacity), rules of origin nomic growth; or "soft" benefits such as a familiar- requirements relating to imported material and ity with markets in developed countries, growing components, social (labor) and environmental awareness of the need to produce high quality arti- requirements, insecurity relating to market access cles and more export-oriented attitudes, and new caused by the unilateral nature of GSP programs, business alliances (Tangermann 2001). and lack of human resources and institutional The critical factor in preference programs is capability to take advantage of preferential arrange- the product coverage benefiting from preferences ments, such as knowledge of the importing coun- (lower tariffs and quotas), and the ability of the try's tariff structure. Overcoming these constraints recipient country to provide those products. For requires investment, which is not encouraged by many developing countries, products of greatest the unpredictability of programs. The high utiliza- interest are agriculture and textile and apparel but tion rate associated with Japan's GSP program has these have been excluded or severely restricted been attributed in part to the program's stability-- as "sensitive" articles. Initially preferential pro- as evident from its expansion in 2001 for 10 years grams were provided for manufactured and semi- until March 31, 2011. In comparison, the U.S. manufactured products, with very limited coverage GSP program, which as noted above has been of agricultural products. Successive GATT negotia- renewed 6 times in the past 8 years, makes the tions have resulted in low MFN tariffs on most scheme's continuation difficult to predict and industrial products. For agriculture and processed affects both exporters and importers as well as products, even with the URAA, market access is potential investors. constrained by high MFN tariffs, with much higher Nontariff barriers (NTBs) such as quotas, tech- tariffs and limited quota access for processed agri- nical standards, and sanitary and phytosanitary cultural products. A World Bank study (Hoekman, measures may be a serious impediment to high uti- Ng, and Olarreaga 2003) on the economic impacts lization rates and trading opportunities for LDCs. of trade preferences on extending the EU EBA ini- These apply particularly to agricultural goods and tiative (duty- and quota-free market access for tar- textiles. It has been estimated that 42 percent of iff peak products) into Quad countries found that LDC exports of agricultural and fishery products LDC exports would increase by 11 percent. A sec- face NTBs in Quad countries; for textiles and cloth- ond study by UNCTAD/Commonwealth Secretariat ing, the corresponding figure is 66­69 percent (2002) showed only a 3 percent increase in exports. (UNCTAD 2002, table 21). Both studies ignore (but recognize) the problem of Improved market access is meaningless, how- low utilization rates, weak export capacity and sup- ever, if developing countries are unable to use such ply constraints, and that the benefits are likely to be access owing to domestic reasons or the qualified lower than the simulations (see table 15.7). nature of the access. According to UNCTAD, avail- Economists have tried to quantify the benefits of able estimates suggest that in the late 1990s about preferences by examining the utilization rate of half of LDC exports to Quad markets, potentially those preferences or by examining the value of eligible for GSP preferential treatment, did not margin of preference. Economists have also quali- qualify--because of NTBs. This resulted in unnec- fied the results of these studies (Tangermann 2001). essary payment of MFN customs duties, rejected UNCTAD reported that the utilization rate (that is, imports, unnecessary testing, spoilage, legal fees, the ratio between imports actually receiving the and foregone opportunities in general. preferences and total imports eligible for prefer- Preference margins and their size is another ences) was 77 percent for the U.S., 34 percent for way of trying to measure the benefits of preferences. non-ACP LDCs of the EU, 59 percent for Canada, The preference margin per unit of product 308 Agriculture and the WTO TABLE 15.7 Effective Benefits of Quad Countries: Generalized System of Preferences for Least-Developed Countries, Late 1990s Imports from LDCs Receiving Total/ GSP-eligible/ Total Dutiable Covered by preferential dutiable dutiable Utilization Utility imports imports GSP scheme treatment imports imports rate rate 1 2 3 4 (2/1) (3/2) (4/3) (4/2) US$ millions percentage Canada 256 92 10 6 36 11 59 6 EU 3,562 3,101 3,075 1,035 87 99 34 33 Japan 1,248 765 314 229 61 41 73 30 U.S. 4,975 4,247 2,282 1,747 85 54 77 41 U.S., excl. minerals 2,613 2,078 113 89 80 5 79 4 Total 10,041 8,205 5,681 3,017 82 69 53 37 Total, excl. U.S. minerals 7,679 6,036 3,512 1,359 79 58 39 23 Note: EU excludes Africa, Caribbean, and Pacific LDCs and U.S. excludes Haiti, a beneficiary of the Caribbean Basin Initiative. Canada and U.S. are 1998 data. EU is 1999 data. Japan is 1997 data. Source: UNCTAD 2002, part 2, ch. 3. exported to a given country is the difference Convention was applied to their access to the EU between the MFN tariff and the preferential tariff (see table 15.8). for that product, both tariffs being expressed as A study by Stoeckel and Borrell (2001) examines percentages. The identification and measurement and discusses the economic effects of preferences of margins is difficult because it requires establish- on the receiving country, on the granting country, ing which benefit was actually caused by trade and on third countries. To access the issue of diver- preferences, that is, the with-and-without compari- sion and the importance of preferences, the study son. Other factors may affect market access. uses the GTAP global trade model with trade data Tangermann (2001) discusses the work by a num- and preferences for 1997 under three scenarios: ber of economists but notes that "their results removal of all trade barriers facing non-Cairns should be interpreted with much caution. In par- Group developing countries, unilateral reform, and ticular, the actual welfare effects of trade prefer- multilateral reform. Results are summarized as fol- ences for the recipient countries may deviate lows: preferential trade can result in trade diver- significantly from such mechanically calculated sion; unilateral trade reform by developing preference margins and usually they will be far countries can be more beneficial than trade prefer- smaller. For the same reason the actual economic ences; and multilateral nondiscriminatory trade effects of MFN tariff reductions for preference- liberalization is the best policy. receiving countries may be considerably smaller than the erosion of preference margins estimated in Costs of Trade Preferences this mechanical way." With these caveats in mind, 1997 preference-margin data for products into the Distortion in Developing-Country Economies? EU show that for LDCs among African ACP coun- Preferential access may result in hindering both the tries (AACP) the aggregate preference margin economy and the industry by diverting trade oppor- under GSP would be only 7 percent less than under tunities away from preferred, more efficient produc- the Lomé Convention, while the remaining AACP ers to less competitive suppliers. It also biases members would lose around 85 percent of the pref- resource allocation to industries that are noncom- erence margin if GSP rather than the Lomé petitive and in the long run takes resources away Special Trade Arrangements to Improve Market Access 309 TABLE 15.8 Preference Margins for Selected Groups of Agricultural Products Exported from the AACP to the EU, Lomé Preferences and GSP Preferences Compared, 1999 EU Tariffs (1997 Trade Data) Value of preference margin, Value of preference margin, Least-Developed AACP Countries Other AACP Countries GSP GSP Lomé preferences Lomé preferences preferences (percentage preferences (percentage Product group (million C) = of Lomé) (million C) = of Lomé) Fish 41.7 41.7 (100) 115.1 14.9 (12.9) Tobacco 29.7 29.7 (100) 38.4 10.3 (26.9) Fresh fruit and vegetables 3.8 3.9 (101) 18.8 4.0 (21.0) Processed fruit and vegetables 0.7 0.5 (68.4) 19.8 2.2 (10.9) Cereals 0.0 0.0 (0) 0.0 0 Dairy products 1.3 0.0 (0) 0.0 0 Beef--general Lomé preferences 6.0 2.0 (33.1) 11.1 0 Sugar products--general Lomé preferences 0.5 0.0 (0) 0.5 0 Total of above products 83.7 77.8 (93.0) 203.7 31.4 (15.4) Note: In this table, 0.0 denotes a positive value less than 0.1, while 0 denotes a zero value. Source: Tangermann 2001. from more efficient industries and diverts attention 4.8 percent while under the SP it is 56.7 percent. As a from multilateral reform. The sugar and banana result, this relationship has locked resources into an industries in some developing countries illustrate industry that, despite subsidies through preferential these situations. access to the EU, is struggling--and represented a According to Page and Hewitt (2002) for many of loss to the economy in 1999. Issues of land allot- the small Caribbean countries sugar is probably ment, labor laws, and zoning have precluded a barely viable in the long term without the EU Sugar rationalization within the sugar industry, and its Protocol (SP). Caribbean countries could live with a structure and organizational model have remained gradual decrease in European prices but would not unchanged because the industry has been insulated be able to compete with Asian LDCs. St. Kitts and from world markets. Nevis, Trinidad and Tobago, and Jamaica would be The EBA Agreement provides additional sugar forced out of the sugar business if the EU protocol access for LDCs into the EU and given the uncer- were to be revoked. Trinidad and Tobago and tainty regarding the future of the Sugar Protocol, Barbados would probably benefit from abandoning some sugar producers in SP quota countries such as sugar as this would require them to diversify their Mauritius are already investing in sugar industry economies. For Jamaica, given its costs, it would not capacity in LDCs such as Mozambique, so that links be able to compete on the world market and the situ- to markets can be quickly established (UNCTAD ation would be difficult. The EU is a very important 2002). EU preferences for bananas have induced market for Mauritius with sugar representing some African countries to invest in banana produc- 46.6 percent of the value of total agricultural exports tion and to try to compete with the low-cost Latin to the EU in 1999 (see table 15.9). The value of the American producers, despite the long-term nega- preference margin under the Lomé Convention was tive impact of this crop on African land and water 310 Agriculture and the WTO TABLE 15.9 Preference Margins for Protocol Beef and Sugar Exported from Individual AACP Countries to the European Union under Lomé Provisions, 1999 EU Tariffs (1997 Trade Data) Percentage of value of total Value of preference agricultural export to EU of Product and country margin (million C­ ) country concerned Protocol beef Botswana 38.4 88.5 Kenya 0 0 Madagascar 2.3 1.2 Namibia 21.9 13.0 Swaziland 1.1 0.8 Zimbabwe 23.4 5.5 Protocol sugar Congo 4.1 28.0 Kenya 0 0 Madagascar 5.4 2.7 Malawi 11.0 5.3 Mauritius 168.8 46.6 Swaziland 63.3 48.9 Tanzania 4.0 2.9 Uganda 0 0 Source: Tangermann 2001. resources (see Stoeckel and Borrell 2001). paper (2002), the administrative requirements Developing countries receiving preferential access involved in documenting eligibility may explain for sugar (into the EU and the U.S.) and bananas the limited use made of AGOA provisions.By October (into the EU) are likely to face difficulties if and 2002 only 15 countries had used the AGOA provi- when their preferential access erodes significantly sions, with most benefits accruing to four countries-- or is eliminated. Gabon, Lesotho, Nigeria, and South Africa--with fuel accounting for 85 percent of AGOA imports (see also Rules of Origin Mattoo, Roy, and Subramanian 2002). The value of preferences can be eroded by rules of ori- Under each GSP scheme ROO reduces the effi- gin (ROO), which are unilateral domestic content ciency of the trade involved and may absorb scarce requirements.These are seen as a tool of protectionism resources and political capital that might otherwise and are described by Hirsch (2002) as"constituting an be used in support of broader-based multilateral indispensable component of any discriminatory and regional initiatives. As tariffs are reduced, the arrangement."As there is no uniformity in the various widening use of ROO is increasingly seen as a GSP schemes, there is no uniformity of rules. If condi- measure to restrict trade (Hirsch 2002).5 tions relating to ROO are complex, preferences often The 1994 WTO Agreement on Rules of Origin remain unused because meeting these rules is simply deals with harmonization at the global level too burdensome for many smaller LDCs. In the and therefore is not concerned with preferential NAFTA there are over 200 pages devoted to ROO. In arrangements. A survey of domestic content rules developed countries, most rules are formulated in an included in the various trade agreements and non- extremely complex and technical form, requiring spe- preferential regulations shows that most local con- cial expertise to assess their impact. tent rates are within the range of 35­62.5 percent, Under AGOA margins of preference are substan- with 62.5 percent applying to certain automobiles tial for textile and apparel products, other light man- under NAFTA and 35 percent provided for in the ufactures, and food products. According to an 1985 Israel­U.S. Free Trade Agreement and in pro- International Monetary Fund (IMF)/World Bank grams such as GSP and CBI. Special Trade Arrangements to Improve Market Access 311 BOX 15.7 The Americas: Trade Diversion--Displacement from Markets Different preferential treatment relating to duty ranging from 6 to 35 percent, the tariff tuna by the U.S. in its NAFTA and ATPA agree- advantage provided an incentive for Mexico to ments had the potential to create a problem build up its tiny tuna industry. Such a build-up for AFTA countries. Until the 2002 extension of would have created a serious threat to ATPDEA ATPA (now the ATPDEA)--which, among other countries as this preferential access to the U.S. provisions, lowered tariffs on tuna in would have made a new competitor out of pouches--ATPA members were concerned that Mexico and Central America. Legislation creat- the NAFTA agreement would result in a new ing comparable market access conditions (and competitor in the tuna trade with the U.S. lower tariffs) for these beneficiary countries may because under NAFTA, U.S. duty on canned present a problem for the Philippines, which tuna from Mexico will be eliminated by 2008. exports tuna to the U.S. market under less favor- With tuna from other countries subject to a able access conditions. Multilateral Agreements Versus 1996 Singapore WTO Ministerial as a result of a Free Trade Agreements: Political well-structured and coordinated APEC initiative. Economy Costs? Another political economy cost can result when a trade arrangement creates a lack of interest in lib- During the 1990s, with the proliferation of trade eralizing trade in the product where the preferences agreements, there was concern among trade policy- currently apply. The most often cited examples of makers that such agreements would detract from this are the access provided on sugar by the U.S. and efforts to obtain multilateral trade liberalization on sugar and bananas by the EU. Given the value of through the WTO. This concern has continued to the preferences some developing countries have lit- the extent that WTO Director General Supachai tle interest in seeing the liberalization of trade for warned members in November 2002 that "by dis- these two commodities. Some reports indicate that criminating against third countries and creating the EU has been able to obtain support in the WTO a complex network of trade regimes, such agree- agriculture negotiations for its position on the mul- ments pose systemic risk to the global trading tifunctional approach to agriculture from some system" (quoted in the Financial Times--see countries that clearly benefit from access to EU de Jonquieres 2002). With their tariff preferences, markets. trade agreements meant that there were winners In terms of the budget/development debate in and losers: in other words, the article or commod- developed countries and the politics of budgetary ity that received the preference and the country support for developing countries, trade preferences that was a member of the agreement was a may be a factor. These are not as visible in budgets "winner," while nonmembers were seen as "losers" as direct aid and therefore may be more easily (see boxes 15.7 and 15.8). obtained, provided that domestic industry interest Given such a situation, trade agreements were groups do not take political action to counter considered a poor substitute for multilateral trade efforts to provide such preferential access. liberalization. However, fears that these trade agreements represented stumbling blocks and not Alternative Options in Trade building blocks for trade liberalization have not Negotiations: What Is the Future eventuated. APEC is not a preferential trade agree- of Trade Preferences? ment but its ultimate goal is trade liberalization. An examination of APEC's careful policy approach of As the current round of trade negotiations will open regionalism shows that its activities reinforce seek further reduction in tariffs (especially on the multilateral approach. If anything, APEC has processed agricultural products), greater access acted as a catalyst for negotiations on a range under TRQs, and simplification of quota adminis- of issues. This was evident in the case of the tration, there is likely to be some erosion in the Information Technology Agreement adopted at the value of preferences (in other words, reduced tariffs) 312 Agriculture and the WTO BOX 15.8 Chile: Strategy for Trade Growth--Access or Diversion? Chile currently follows a strategy of negotiating sion costs. Owing to preferential market access, "additive regionalism," that is, bilateral free trade however, Chile's strategy is likely to yield gains agreements with all of its significant trading part- that are greater than static welfare gains from ners. In addition to its free trade area with unilateral free trade. Economic modeling of the MERCOSUR, it has agreements with Canada, results of NAFTA, MERCOSUR, and the additive Mexico, and the EU and in June 2003 signed one regionalism approach all point to the usual with the U.S. One study examining Chile's impetus of improved market access in preferen- regional arrangements found that only Chile's tial trading areas. Even for the Chile/ agreements with Northern partners (Canada MERCOSUR agreement, if Chile were to lower its and the EU and eventually with the U.S.) provide external tariff, reducing trade diversion would be sufficient market access to overcome trade diver- beneficial. provided to developing countries under current pref- The European Communities' proposal to the erential arrangements (see table 15.10; see also WTO's Special Session of the Committee on Tangermann [2001] for a discussion on the issues of Agriculture, December 3­5, 2001, called for an both erosion and compensation for such erosion). examination of the issues raised by preferential These arrangements, whether resulting from GSP- agreements. The EC proposal noted that the maxi- type schemes or from trade agreements, have been mum value of preferences is limited by the size of established outside of the GATT/WTO system. the corresponding MFN tariff and that the impor- Developing countries may want to consider whether tance of preferences will decline as MFN rates are attention to the issue of preferences detracts from reduced; that where preferences exist a sudden their own domestic liberalization efforts. Ozden and removal could cause significant problems; and that Reinhardt's analysis (2003) of the U.S. GSP program the possibility that removal of the preference is using a dataset of 154 developing countries from in itself a limit on the preference's effectiveness, 1976 through 2000 finds that countries removed because this uncertainty limits investment. from GSP benefits adopted more liberal trade poli- Given the discrimination in current preferential cies than those remaining eligible. This suggests that trade arrangements under GSP schemes affecting developing countries may be better served by fully some key developing countries, and the marginal integrating into a reciprocity-based world trading benefit of preferential arrangements to their devel- system rather than continuing with GSP-style special opment, some countries may question whether the preferences. Of the 154 countries eligible for GSP preferential approach is of use--or whether negoti- benefits, 36 have graduated from the program since ating capital should be directed toward greater 1976 (including Singapore, Hong Kong, Taiwan global liberalization for all, discarding the use of [China], the Republic of Korea, Malaysia, Mexico, preferences and instead insisting on their WTO and Botswana).The list of major countries remaining rights to achieve greater global market access. eligible include Brazil, India, Indonesia, Turkey, Developing countries will need to decide on South Africa, and Thailand. In addition to preferen- whether to make an issue of preferential arrange- tial tariff schemes' possible negative effect on the ments and to improve their use through "best incentive to liberalize, critics have pointed to their endeavors" language, knowing that mandatory lan- underachieving nature, seen in the low utilization guage cannot apply to GSP programs. Their deci- rates and especially in the increasingly stringent sion will depend on the chances of suggesting ROOs. A further shortcoming of these schemes is changes that will provide real benefits compared that developing countries have no way of protecting to the existing preferential treatment. Given their "rights" in the various GSP programs, as these the complexity of preferential arrangements arrangements are outside the WTO jurisdiction and (involving beneficiaries, product coverage, and subject to the political pressures of the GSP host rules administration), however, even "best endeav- country. ors" language may be difficult to achieve. Special Trade Arrangements to Improve Market Access 313 TABLE 15.10 Preference Margins for Selected Groups of Agricultural Products Exported from the AACP to the European Union Under Lomé Provisions, Hypothetical EU Tariffs after the Next WTO Round (1997 Trade Data) Value of preference margin (percentage of value (percentage decrease in of AACP export to EU of preference margin Product group (million C­ ) product concerned) relative to 1999 levels) Fish 75.0 6.3 52.1 Tobacco 24.5 5.1 64.0 Fresh fruit and vegetables 8.9 2.9 60.6 Processed fruit and vegetables 6.2 6.2 61.9 Cereals 0.0 0.5 10.9 Dairy products 0.4 8.0 71.7 Total of above products 115.1 5.4 56.6 Source: Tangermann 2001. Nevertheless, there are some aspects of prefer- Notes ences for which all developing countries may want 1. See various documents from the WTO, OECD, and UNC- to take a concerted stand because of their effects on TAD on the issue of trade preferences. Examples include "The LDCs. These include the following: Generalized System of Preferences: A Preliminary Analysis of the GSP Schemes of the Quad" (WTO); "Mapping of Regional Trade Agreements" (WTO); "Market Access Issues Related to · Providing technical assistance for information Products of Export Interest Originating from Least Developed services and training courses for local producers Countries" (WTO); "The Relationship between Regional Trade and exporters using GSP programs. Agreements and Multilateral Trading System" (OECD). See also the UNCTAD Web site of the Generalized System of Preferences · Providing predictability and harmonization of (GSP); U.S. ITC International Economic Reviews including rules in preference programs for all developing Andriamananjara (2001) and Anderson (2001) and Laird, countries or specifically for LDCs through "best Safodi and Turrini. 2. The variability of the schemes means that beneficiaries endeavors" language on the issue. It should be can change from year to year; as a result, the lists here may noted that increasing the stability and pre- already be out of date. dictability of trade preferences through a set of 3. For an interesting examination of GATT/WTO provi- multilaterally agreed criteria has already been sions on the current trend of "soft law" status for developing countries resulting in their disadvantage inside the WTO legal under consideration. The issue of language system, in terms of access to the disputes settlement process, see "binding" members was raised in preparation Olivares (2001). for the 1999 Seattle WTO Ministerial but the 4. The European Communities (EC) is the signatory to the WTO. Doha WTO Ministerial only committed mem- 5. For an interesting discussion of Rules of Origin see bers to working toward the objective of quota- Hirsch (2002). Hirsch describes the factor of information and duty-free market access as well as to asymmetry (in other words, the gap between the knowledge of the developed- and developing-country exporters), the considering additional measures for the pro- widening use of ROOs, and their challenge to the current trad- gressive improvement in market access for ing system. LDCs (WTO document WT/MIN(01)/DE/W/1, paragraph 42). Select Bibliography · Proposing "best endeavors" language that would result in the adoption of the same approach to Agama L., and S. Zeleke. 2000. "New Trade and Investment Framework Agreement between the United States and the tariff cuts and product coverage in preferential Common Market for Eastern and Southern Africa." U.S. schemes. ITC International Economic Review January/February. 314 Agriculture and the WTO Agama, L., 2001. "Assessing the Desirability of a Free-Trade Area Jennings, T. F. 2002. "USITC Reports That CBERA Imports Will in Southern Africa." U.S. ITC International Economic Review, Likely Increase." U.S. ITC International Economic Review, July/August. January/February. Anderson, M. M. 2001. "Preferential Trade Agreements: Trade Kornis, M. 2000. "U.S. Trade With the Beneficiaries of the Diversion and Other Worries." U.S. ITC International Andean Trade Preference Act." U.S. ITC International Economic Review, May/June. Economic Review, September/October. Andriamananjara, S. 2001. "Preferential Trade Agreements and the Kose, M. A., and R. Riezman. 2002. "Small Countries and Multilateral Trading System." U.S. ITC International Economic Preferential Trade Agreements: How Severe Is the Innocent Review, January/February. Bystander Problem?" Pacific Economic Review (U.S.) 7(2): Barfield, C. 2002. "Preferential Pacts Hurt Trade." Far Eastern 279­304. Economic Review, July 4. Laird, S., R. Safadi, and A. Turrini. 2002. "The WTO and Borrell, B., and D. Pearce. 1999. "Sugar: The Taste Test of Trade Development." Update of Paper for conference, Tulane Liberalization." Paper prepared for Centre for International University, New Orleans. Economics, Canberra. Ludema, R. 2002. "Increasing Returns, Multinationals and Borrell, B. 1999. "Bananas: Straightening Out Bent Ideas on Geography of Preferential Trade Agreements." Journal of Trade as Aid." Paper prepared for Centre for International International Economics 56(9): 329­58. Economics, Canberra. Mattoo, A., D. Roy, and A. Subramanian. 2002. "The Africa Bradsher, K. 2002. "Quandary on Trade" (News Analysis), New Growth and Opportunity Act and Its Rules of Origin: York Times, May 21. Generosity Undermined?" World Bank Policy Research Burki, S. J., E. P. Guillermo, and S. Calvo. 1998. Trade: Toward Paper, Washington, D.C. Open Regionalism. Washington, D.C.: World Bank. Messerlin, P. A. 2001. "Measuring the Costs of Protection in Chang, W., and L. A. Winters. 2001. "Preferential Trading Europe: European Commercial Policy in the 2000s." Institute Arrangements and Excluded Countries: Ex-Post Estimates of for International Economics, Washington, D.C. the Effects on Prices." The World Economy 24(6): 797­807. Organisation for Economic Co-operation and Development Dean, Judith M. 2002. "Do Preferential Trade Agreements (OECD). 2002. "The Relationship between Regional Trade Promote Growth? An Evaluation of the Caribbean Basin Agreements and Multilateral Trading System, Trade Economic Recovery Act." Working Paper, U.S. International Facilitation." TD/TC/WP(2002)17/Final, April 11. Working Trade Commission, Office of Economics, Washington D.C. Party of the Trade Committee, Paris. de Jonquieres, G. 2002 "The Challenge for the Multilateral Trading Olivares, C. 2001. "The Case for Giving Effectiveness to System."FinancialTimes.com. GATT/WTO Rules on Developing Countries and LDCs." European Communities. 2001. "Tariff Preferences for Journal of World Trade 34(3): 545­52. Developing Countries." (WTO) Special Session of the Ozden, C., and E. Reinhardt. 2003. "The Perversity of Committee on Agriculture, Note. Office of the Secretary Preferences: GSP and Developing Country Trade Policies, General, Geneva. 1976­2000." Working Paper 2955, World Bank, Washington, Guth, J., and V. Chomo. 2002. "The Euro-Mediterranean D.C. Partnership." U.S. ITC International Economic Review, Page, S., and A. Hewitt. 2002. "The New European Trade March/April. Preferences: Does `Everything But Arms' (EBA) Help the Guth, J. 2000. "European Union and Mexico Conclude Free Poor?" Development Policy Review 20(1): 91­102. Trade Agreement." U.S. ITC International Economic Review, Peters, G. 2001. "Mexico Leverages Free Trade." Global Finance, April/May. June, 49­50. Guth, J. 2000. "New EU Agreements With African Countries Pollard, W. 2000. "Impact of the Caribbean Basin Economic Liberalize Two-way Trade." U.S. ITC International Economic Recovery Act Declines." U.S. ITC International Economic Review, November/December. Review, April/May. Harrison, G. W., T. Rutherford, and D. Tarr. 2002. "Trade Policy Pollard, W. 2001. "Renewal and Expansion of ATPA Could Options for Chile: The Importance of Market Access." The Enhance Effectiveness of the Program." US ITC International World Bank Economic Review 16(1): 49­79. Economic Review, July/August. Hasha, G. 2002. "Trade Among Unequal Partners Changing EU Scollay, R. 2001. "The Changing Outlook for Asia-Pacific Trade Arrangements With Developing Countries." Regionalism." The World Economy 24(9): 1135­1160. Agricultural Outlook, USDA Economic Research Service, Stoeckel, A., and B. Borrell. 2001. "Preferential Trade and September. Washington, D.C. Developing Countries--Bad Aid, Bad Trade." Prepared for Hensen, F. 2001. "Trade at the Crossroads." Business Credit, Cairns Group Farm Leaders meeting, Uruguay, Rural June. New York. Industries Research and Development Cooperation Hillberry, R., and C. McDaniel. 2002. "A Decomposition of (RIRDC), Canberra. North American Trade Growth Since NAFTA." U.S. ITC Tangermann, Stefan. 2001. "The Future of Preferential Trade International Economic Review, May/June. Arrangements for Developing Countries and the Current Hirsch, M. 2002. "International Trade Law, Political Economy Round of Agricultural Trade Negotiations on Agriculture." and Rules of Origin." Journal of World Trade 36(4): 171­89. Paper prepared for FAO, Rome. Hoekman, B., F. Ng, and M. Olarreaga. 2003. "Reducing Topp, V. 2001. "Trade Preferences: Are They Helpful in Agricultural Tariffs Versus Domestic Support. What is More Advancing Economic Development in Poor Countries?" Important for Developing Countries." Working Paper 2918, Australian Bureau of Agricultural and Resource Economics World Bank, Washington, D.C. (ABARE), Canberra. International Monetary Fund and the World Bank. 2002. Trade Partnership. 2001. "The U.S. Generalized System of "Market Access for Developing Country Exports." Staff Preferences Program: An Update." Washington, D.C. Paper. Washington, D.C. Special Trade Arrangements to Improve Market Access 315 UNCTAD (United Nations Conference on Trade and WTO (World Trade Organization). 2001. "The Generalized Development). 2002. "Escaping the Poverty Trap." The Least System of Preferences: A Preliminary Analysis of the GSP Developed Countries Report 2002, June 18. Schemes in the Quad." Committee on Trade and U.S. ITC. 2002. "United States-Sub-Saharan Africa Trade Data, Development, WT/COMTD/ W93. Office of the Secretary Updated Quarterly." U.S. Trade and Investment with Sub- General, Geneva. Saharan Africa, July. ------. 2002a. "Mapping of Regional Trade Agreements." Note USTR United States Trade Representative. 1995. "Comparison by the Secretariat, Committee on Regional Trade of NAFTA, AFTA, CER Trade Agreements." Draft, Agreements, WT/REG/W/41, October 11. Office of the Washington, D.C. Secretary General, Geneva. ------. 2002. "New Andean Trade Benefits," Fact Sheet. Office ------. 2002b. "Market Access Issues Related to Products of of the USTR, Washington, D.C. Export Interest Originating from Least Developed ------. 2002a. "Fourth Report to Congress on the Operation of Countries." Note by Secretariat, Subcommittee on LDC the Caribbean Basin Economic Recovery Act." USTR Web Negotiating Group on Market Access, WT/COMTD/LDC/ site, Washington, D.C. W/28, TN/MA/S/7. Office of the Secretary General, Geneva. ------. 2002b. "Comprehensive Report on U.S. Trade ------. 2002c."Material on Regionalism, Scope of RTAs." Office and Investment Policy toward Sub-Saharan Africa and of the Secretary General, Geneva. Implementation of the African Growth and Opportunity ------. 2002d. "Trade Policy Review: European Union, Report Act." Office of the USTR, Washington, D.C. of the Secretariat." WT/TPR/S/102. Office of the Secretary ------. 2002c. "Generalized System of Preferences (GSP) General, Geneva. Guidebook, Part I: Most Frequently Asked Questions." May, Zahniser S., and J. Link. 2002. "Effects of North American Free Washington, D.C. http://www.ustr.gov/reports/ gsp/faq. Trade Agreement on Agriculture and the Rural Economy." html. Economic Research Service (ERS) and Trade Report No. ------. 2002d. "2002 Trade Policy Agenda and 2001 Annual WRS0201, July. U.S. Department of Agriculture, Report of the President of the United States on the Trade Washington, D.C. Agreements Program." USTR Web site. ------. 2002e. "USTR Team to Provide Trade Seminars in Argentina and Uruguay." Press release, September 9, Office of the USTR, Washington, D.C. Appendix A oecd policy evaluation matrix and trends in policy for various commodities Cameron Short and Harry de Gorter Introduction through A.5 summarize the support for grains and oilseeds in 2001 for both the OECD-6 in the PEM This appendix begins by describing the commodity and all OECD countries compared to all other and policy coverage of the Organisation for commodities. Tables A.1 and A.2 show that MPS Economic Co-operation and Development's for all commodities represents US$145,269.3 mil- (OECD) Policy Evaluation Matrix (PEM) model lion (63.0 percent) of total support in the OECD, (see OECD 2001). Trends in protection levels and which was US$230,746.0. Area payments are changes in the policy mix are documented for the the next largest policy category at 12.6 percent of grains and oilseeds sector in relation to all other total OECD support, and these payments equaled commodities for the OECD as a whole and for the US$29,056.7 million. six OECD regions covered in the PEM. The appen- Table A.3 shows that grains and oilseeds MPS dix then outlines the basic economic structure of represents 17.0 percent of total OECD MPS, but the model. receive a far higher share of total support in each of the other policy categories of domestic support. The Policy Developments in the Grains column titled G&O Total in table A.2 shows that and Oilseeds Sector grains and oilseeds MPS represents only 38.1 percent The OECD identifies five categories of support used of the total grains and oilseed support in the OECD in the PEM: market price support (MPS) due to bor- and 30.8 percent for the OECD-6. Support across all der protection (including export subsidies); output policy types is far more uniform in the grains and price support (OPS) such as deficiency payments; oilseeds sector for both the OECD and the OECD-6 input price subsidies (IPS) such as fertilizer subsi- countries. dies; area payments such as those in the European Table A.1 shows that the OECD-6 represents Union (EU); and historical entitlements such as most of the support for grains and oilseeds overall those in the United States (U.S.). There are also pay- in the OECD with the exception of support for ments based on overall farm income and miscella- oilseeds alone (compare the bottom rows of the neous payments but these are relatively small. OECD and OECD-6 sections). For output support, The PEM has six regions: Canada, the EU, Japan, input support, area payments, and historical entitle- Mexico, Switzerland, and the U.S. Tables A.1 ments, grains and oilseeds dominate in terms of size 317 318 Agriculture and the WTO TABLE A.1 OECD Support for Grains and Oilseeds 2001 (in US$ Millions) Coarse Other Wheat Grains Rice Oilseeds Commodities Total G&O Total OECD Market price support 1,197.2 1,889.0 21,266.9 321.0 120,595.1 145,269.3 24,674.1 Based on output 471.3 1,935.7 1,520.8 3,514.4 6,761.4 14,203.5 7,442.2 Based on area planted/ 7,331.3 6,695.3 164.2 1,743.2 13,122.7 29,056.7 15,934.0 animal numbers Based on historical 3,270.2 5,661.4 175.5 117.1 3,223.9 12,448.0 9,224.2 entitlements Based on inputs* 1,919.9 2,564.9 1,054.8 1,227.7 19,324.6 26,091.9 6,767.3 Total** 14,332.8 18,970.1 24,339.6 7,114.3 165,989.2 230,746.0 64,756.8 OECD-6 Market price support 969.5 1,851.2 14,397.2 31.2 107,517.2 124,766.2 17,249.1 Based on output 354.5 1,899.3 1,520.8 3,514.4 5,847.6 13,136.7 7,289.0 Based on area planted/ 7,232.9 6,551.2 90.2 1,711.0 12,505.2 28,090.4 15,585.3 animal numbers Based on historical 3,264.7 5,658.7 175.5 114.5 2,674.0 11,887.5 9,213.4 entitlements Based on inputs* 1,665.6 2,383.1 888.8 1,177.5 17,081.0 23,195.9 6,115.0 Total** 13,607.2 18,547.5 17,082.2 6,734.1 148,034.8 204,005.8 55,971.0 Canada Market price support n.a. n.a. n.a. n.a. 1,856.5 1,856.5 0.0 Based on output 17.7 93.3 n.a. 9.2 125.7 245.9 120.2 Based on area planted/ 129.3 71.8 n.a. 102.0 102.6 405.7 303.1 animal numbers Based on historical 136.8 67.5 n.a. 86.5 232.2 522.9 290.8 entitlements Based on inputs* 46.4 15.2 n.a. 30.4 243.2 335.2 92.0 Total** 414.6 274.5 n.a. 289.1 2,949.4 3,927.7 978.2 European Union Market price support 81.1 592.2 224.3 0.0 53,392.1 54,289.6 897.6 Based on output 39.9 30.6 1.3 10.3 3,592.8 3,674.8 82.1 Based on area planted/ 6,938.5 6,071.3 77.8 1,358.8 10,588.2 25,034.6 14,446.4 animal numbers Based on historical 12.9 50.7 0.0 12.8 480.2 556.6 76.4 entitlements Based on inputs* 1,162.5 1,201.4 19.1 381.9 6,455.0 9,220.0 2,764.9 Total** 8,206.0 7,923.0 322.4 1,758.0 74,873.6 93,083.0 18,209.4 Japan Market price support 704.6 181.1 14,164.5 0.0 27,437.4 42,487.7 15,050.2 Based on output 0.0 4.9 968.9 129.2 384.4 1,487.5 1,103.0 Based on area planted/ n.a. n.a. n.a. n.a. n.a. n.a. 0.0 animal numbers Based on historical n.a. n.a. n.a. n.a. n.a. n.a. 0.0 entitlements Based on inputs* 107.8 23.0 814.9 68.3 2,253.9 3,268.0 1,014.0 Total** 812.5 209.1 15,948.3 197.6 30,075.7 47,243.2 17,167.5 Appendix A 319 Table A.1 (Continued) Coarse Other Wheat Grains Rice Oilseeds Commodities Total G&O Total Mexico Market price support 134.5 1,023.2 8.5 6.1 2,871.6 4,043.9 1,172.3 Based on output 103.4 192.2 8.2 11.2 3.7 318.6 315.0 Based on area planted/ 0.2 1.7 0.0 0.0 70.1 72.1 1.9 animal numbers Based on historical 69.5 928.0 9.2 8.5 242.5 1,257.7 1,015.2 entitlements Based on inputs* 12.4 114.5 1.8 5.3 646.1 780.0 134.0 Total** 321.0 2,267.2 27.8 31.1 3,889.6 6,536.8 2,647.1 Switzerland Market price support 49.2 54.7 n.a. 25.1 2,152.6 2,281.6 129.0 Based on output 0.0 0.0 n.a. 0.0 225.6 225.6 0.0 Based on area planted/ 5.0 3.7 n.a. 19.9 506.1 534.6 28.6 animal numbers Based on historical 42.3 27.0 n.a. 6.8 681.6 757.6 76.1 entitlements Based on inputs* 28.3 18.1 n.a. 4.6 245.2 296.2 51.0 Total** 131.4 107.7 n.a. 57.3 3,917.8 4,214.2 296.4 United States Market price support 0.0 0.0 0.0 0.0 19,807.0 19,807.0 0.0 Based on output 193.6 1,578.2 542.5 3,354.5 1,515.5 7,184.3 5,668.8 Based on area planted/ 159.9 402.8 12.3 230.3 1,238.1 2,043.4 805.3 animal numbers Based on historical 3,003.3 4,585.6 166.3 0.0 1,037.5 8,792.7 7,755.2 entitlements Based on inputs* 308.0 1,010.9 53.0 687.0 7,237.6 9,296.5 2,058.9 Total** 3,721.7 7,766.0 783.7 4,400.9 32,328.6 49,000.9 16,672.3 * This combines Based upon input use and Based upon input constraints. ** The total includes Based on overall farm income and miscellaneous payments. relative to other commodities and within the grains gories. By contrast, in coarse grains, the OECD-6 sup- and oilseeds sector itself. Table A.3 shows that grains port expenditures comprised 97.8 percent of the total and oilseeds overall received 52.4 percent of total OECD support. OECD output support, 25.9 percent of input sup- A significant feature of the experiments with area port, 54.8 percent of area payments, and 74.1 per- payments and historical entitlements is their con- cent of historical entitlements in 2001. This means centration in the grains and oilseeds sector that the various types of direct payments are atypi- in the countries and regions in the PEM as found in cally far more important for wheat, coarse grains, the OECD data--especially the U.S., the EU, and and oilseeds. Mexico, the first two of which will have a major The six regions analyzed comprise most of the sup- influence on the Doha Round because of the size of port for each of the four commodity groups, as shown their economies. The grains and oilseeds sector's in table A.5. However, for oilseeds individually, the share of producer support estimate (PSE) is 28.1 OECD-6 accounted for only 9.7 percent of total percent, almost twice its share of value of produc- OECD MPS support. OECD-6 total support for rice at tion of 17 percent, indicating that it is disproportion- 70.2 percent of the OECD total support was also less ately protected. Although MPS is 17.0 percent, most comparable to these countries' efforts in other cate- of this (14.6 percent) is for rice. The grains and 320 Agriculture and the WTO TABLE A.2 Type of Support as a Percentage (%) of Total Support for a Commodity Coarse Other G&O Wheat Grains Rice Oilseeds Commodities Total Total (%) OECD Market price support 8.4 10.0 87.4 4.5 72.7 63.0 38.1 Based on output 3.3 10.2 6.2 49.4 4.1 6.2 11.5 Based on area planted/ 51.2 35.3 0.7 24.5 7.9 12.6 24.6 animal numbers Based on historical 22.8 29.8 0.7 1.6 1.9 5.4 14.2 entitlements Based on inputs* 13.4 13.5 4.3 17.3 11.6 11.3 10.5 Total** 100.0 100.0 100.0 100.0 100.0 100.0 100.0 OECD-6 Market price support 7.1 10.0 84.3 0.5 72.6 61.2 30.8 Based on output 2.6 10.2 8.9 52.2 4.0 6.4 13.0 Based on area planted/ 53.2 35.3 0.5 25.4 8.4 13.8 27.8 animal numbers Based on historical 24.0 30.5 1.0 1.7 1.8 5.8 16.5 entitlements Based on inputs* 12.2 12.8 5.2 17.5 11.5 11.4 10.9 Total** 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Canada Market price support 0.0 0.0 n.a. 0.0 62.9 47.3 0.0 Based on output 4.3 34.0 n.a. 3.2 4.3 6.3 12.3 Based on area planted/ 31.2 26.1 n.a. 35.3 3.5 10.3 31.0 animal numbers Based on historical 33.0 24.6 n.a. 29.9 7.9 13.3 29.7 entitlements Based on inputs* 11.2 5.5 n.a. 10.5 8.2 8.5 9.4 Total** 100.0 100.0 n.a. 100.0 100.0 100.0 100.0 EU Market price support 1.0 7.5 69.6 0.0 71.3 58.3 4.9 Based on output 0.5 0.4 0.4 0.6 4.8 3.9 0.5 Based on area planted/ 84.6 76.6 24.1 77.3 14.1 26.9 79.3 animal numbers Based on historical 0.2 0.6 0.0 0.7 0.6 0.6 0.4 entitlements Based on inputs* 14.2 15.2 5.9 21.7 8.6 9.9 15.2 Total** 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Japan Market price support 86.7 86.6 88.8 0.0 91.2 89.9 87.7 Based on output 0.0 2.4 6.1 65.4 1.3 3.1 6.4 Based on area planted/ n.a. n.a. n.a. n.a. n.a. n.a. n.a. animal numbers Based on historical n.a. n.a. n.a. n.a. n.a. n.a. n.a. entitlements Based on inputs* 13.3 11.0 5.1 34.6 7.5 6.9 5.9 Total** 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Appendix A 321 TABLE A.2 (Continued) Coarse Other G&O Wheat Grains Rice Oilseeds Commodities Total Total (%) Mexico Market price support 41.9 45.1 30.5 19.6 73.8 61.9 44.3 Based on output 32.2 8.5 29.4 35.8 0.1 4.9 11.9 Based on area planted/ 0.1 0.1 0.1 0.0 1.8 1.1 0.1 animal numbers Based on historical 21.6 40.9 33.2 27.3 6.2 19.2 38.3 entitlements Based on inputs* 3.9 5.0 6.3 17.1 16.6 11.9 5.1 Total** 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Switzerland Market price support 37.4 50.8 n.a. 43.7 54.9 54.1 43.5 Based on output 0.0 0.0 n.a. 0.0 5.8 5.4 0.0 Based on area planted/ 3.8 3.4 n.a. 34.6 12.9 12.7 9.6 animal numbers Based on historical 32.2 25.1 n.a. 11.8 17.4 18.0 25.6 entitlements Based on inputs* 21.6 16.8 n.a. 8.0 6.3 7.0 17.2 Total** 100.0 100.0 n.a. 100.0 100.0 100.0 100.0 United States Market price support 0.0 0.0 0.0 0.0 61.3 40.4 0.0 Based on output 5.2 20.3 69.2 76.2 4.7 14.7 34.0 Based on area planted/ 4.3 5.2 1.6 5.2 3.8 4.2 4.8 animal numbers Based on historical 80.7 59.0 21.2 0.0 3.2 17.9 46.5 entitlements Based on inputs* 8.3 13.0 6.8 15.6 22.4 19.0 12.3 Total** 100.0 100.0 100.0 100.0 100.0 100.0 100.0 * This combines Based upon input use and Based upon input constraints. ** The total includes Based on overall farm income and miscellaneous payments. oilseeds sector accounts for 54.8 percent of the OECD production of oilseeds, which was 42 percent entire area payments and 74.1 percent of historical of world oilseed production. Wheat production in entitlements in the OECD countries.1 Deficiency the six regions was 29 percent of world production. payments are disproportionately used to support The six regions account for 86 percent of both the oilseed producers. volume of production and of the PSE for OECD The significance of grains and oilseeds in terms grains and oilseeds. They account for 99 percent of of agriculture in the OECD countries can be derived the area payments and historical entitlements and from additional computations to the OECD data- 93 percent of other direct payments (again primarily base (grains and oilseeds accounted for 17 percent deficiency payments). of the total agricultural production by value). The Table A.2 shows that the amounts of protection six regions accounted for 33 percent of world grains are very low for Canada in all categories while the and oilseeds production by weight in 2001 but EU has significant MPS for rice and high rates of 86 percent of production in OECD countries. The support in area payments and IPS. Japan has very six-region share of world production by weight is high MPS and significant output support and input higher for coarse grains at 44 percent and signifi- support in oilseeds and rice. Mexico uses all poli- cantly lower for rice at only 5 percent. The cies except area payments while the U.S. relies heav- six regions also accounted for 85 percent of the ily on output support and historical entitlement 322 Agriculture and the WTO TABLE A.3 Support to a Specific Commodity as a Percentage (%) of Total Type of Support Coarse Other G&O Wheat Grains Rice Oilseeds Commodities Total Total (%) OECD Market price support 0.8 1.3 14.6 0.2 83.0 100.0 17.0 Based on output 3.3 13.6 10.7 24.7 47.6 100.0 52.4 Based on area planted/ 25.2 23.0 0.6 6.0 45.2 100.0 54.8 animal numbers Based on historical 26.3 45.5 1.4 0.9 25.9 100.0 74.1 entitlements Based on inputs* 7.4 9.8 4.0 4.7 74.1 100.0 25.9 Total** 6.2 8.2 10.5 3.1 71.9 100.0 28.1 OECD-6 Market price support 0.8 1.5 11.5 0.0 86.2 100.0 13.8 Based on output 2.7 14.5 11.6 26.8 44.5 100.0 55. Based on area planted/ 25.7 23.3 0.3 6.1 44.5 100.0 55.5 animal numbers Based on historical 27.5 47.6 1.5 1.0 22.5 100.0 77.5 entitlements Based on inputs* 7.2 10.3 3.8 5.1 73.6 100.0 26.4 Total** 6.7 9.1 8.4 3.3 72.6 100.0 27.4 Canada Market price support 0.0 0.0 n.a. 0.0 100.0 100.0 0.0 Based on output 7.2 38.0 n.a. 3.7 51.1 100.0 48.9 Based on area planted/ 31.9 17.7 n.a. 25.2 25.3 100.0 74.7 animal numbers Based on historical 26.2 12.9 n.a. 16.5 44.4 100.0 55.6 entitlements Based on inputs* 13.9 4.5 n.a. 9.1 72.5 100.0 27.5 Total** 10.6 7.0 n.a. 7.4 75.1 100.0 24.9 EU Market price support 0.1 1.1 0.4 0.0 98.3 100.0 1.7 Based on output 1.1 0.8 0.0 0.3 97.8 100.0 2.2 Based on area planted/ 27.7 24.3 0.3 5.4 42.3 100.0 57.7 animal numbers Based on historical 2.3 9.1 0.0 2.3 86.3 100.0 13.7 entitlements Based on inputs* 12.6 13.0 0.2 4.1 70.0 100.0 30.0 Total** 8.8 8.5 0.3 1.9 80.4 100.0 19.6 Japan Market price support 1.7 0.4 33.3 0.0 64.6 100.0 35.4 Based on output 0.0 0.3 65.1 8.7 25.8 100.0 74.2 Based on area planted/ n.a. n.a. n.a. n.a. n.a. n.a. n.a. animal numbers Based on historical n.a. n.a. n.a. n.a. n.a. n.a. n.a. entitlements Based on inputs* 3.3 0.7 24.9 2.1 69.0 100.0 31.0 Total** 1.7 0.4 33.8 0.4 63.7 100.0 36.3 Appendix A 323 TABLE A.3 (Continued) Coarse Other G&O Wheat Grains Rice Oilseeds Commodities Total Total (%) Mexico Market price support 3.3 25.3 0.2 0.2 71.0 100.0 29.0 Based on output 32.4 60.3 2.6 3.5 1.2 100.0 98.8 Based on area planted/ 0.3 2.4 0.0 0.0 97.2 100.0 2.8 animal numbers Based on historical 5.5 73.8 0.7 0.7 19.3 100.0 80.7 entitlements Based on inputs* 1.6 14.7 0.2 0.7 82.8 100.0 17.2 Total** 4.9 34.7 0.4 0.5 59.5 100.0 40.5 Switzerland Market price support 2.2 2.4 n.a. 1.1 94.3 100.0 5.7 Based on output 0.0 0.0 n.a. 0.0 100.0 100.0 0.0 Based on area planted/ 0.9 0.7 n.a. 3.7 94.7 100.0 5.3 animal numbers Based on historical 5.6 3.6 n.a. 0.9 90.0 100.0 10.0 entitlements Based on inputs* 9.6 6.1 n.a. 1.5 82.8 100.0 17.2 Total** 3.1 2.6 n.a. 1.4 93.0 100.0 7.0 United States Market price support 0.0 0.0 0.0 0.0 100.0 100.0 0.0 Based on output 2.7 22.0 7.6 46.7 21.1 100.0 78.9 Based on area planted/ 7.8 19.7 0.6 11.3 60.6 100.0 39.4 animal numbers Based on historical 34.2 52.2 1.9 0.0 11.8 100.0 88.2 entitlements Based on inputs* 3.3 10.9 0.6 7.4 77.9 100.0 22.1 Total** 7.6 15.8 1.6 9.0 66.0 100.0 34.0 * This combines Based upon entitlements input use and Based upon input constraints. ** The total includes Based on overall farm income and miscellaneous payments. payments. Figure A.1 displays the trends in support All other commodities, however, have a far less for grains and oilseeds over time. The trend down- pronounced downward trend in protection ward for MPS is significant and slightly more than although the average rate of protection is lower, at that of total support. The overall percentage of pro- about 50 percent (figure A.2). Figure A.3 shows the tection started at close to 120 percent in 1986 and sharp decline in Canada's protection for grains and fell to 50 percent in 1996 but stands at 77 percent in oilseeds, unlike for other commodities in this 2001. Average protection is 80 percent for the time country (figure A.4). There has been no downward period 1986­2001. The move away from MPS and trend in U.S. protection for grains and oilseeds, toward the various forms of direct payments, espe- although there is a big shift in policies away from cially area payments and historical entitlements, is area payments to historical entitlements and OPS also evident. Agricultural support is still dominated (figure A.5). Interestingly, the trend in protection by MPS but this is declining and the growing com- for all other commodities in the U.S. is declining ponent is made up of area payments and historical somewhat (figure A.6). entitlements. Purchased input subsidies and output Figure A.7 shows the dramatic drop in MPS for the price support have been fairly constant. EU while total support has only a slight downward 324 Agriculture and the WTO TABLE A.4 Support as a Percentage (%) for Grains and Oilseeds Distributed Among Specific Commodities Coarse Other G&O Wheat Grains Rice Oilseeds Commodities Total Total (%) OECD Market price support 4.9 7.7 86.2 1.3 n.a. n.a. 100.0 Based on output 6.3 26.0 20.4 47.2 n.a. n.a. 100.0 Based on area planted/ 46.0 42.0 1.0 10.9 n.a. n.a. 100.0 animal numbers Based on historical 35.5 61.4 1.9 1.3 n.a. n.a. 100.0 entitlements Based on inputs* 28.4 37.9 15.6 18.1 n.a. n.a. 100.0 Total** 22.1 29.3 37.6 11.0 n.a. n.a. 100.0 OECD-6 Market price support 5.6 10.7 83.5 0.2 n.a. n.a. 100.0 Based on output 4.9 26.1 20.9 48.2 n.a. n.a. 100.0 Based on area planted/ 46.4 42.0 0.6 11.0 n.a. n.a. 100.0 animal numbers Based on historical 35.4 61.4 1.9 1.2 n.a. n.a. 100.0 entitlements Based on inputs* 27.2 39.0 14.5 19.3 n.a. n.a. 100.0 Total** 24.3 33.1 30.5 12.0 n.a. n.a. 100.0 Canada Market price support n.a. n.a. n.a. n.a. n.a. n.a. n.a. Based on output 14.7 77.6 0.0 7.6 n.a. n.a. 100.0 Based on area planted/ 42.7 23.7 0.0 33.7 n.a. n.a. 100.0 animal numbers Based on historical 47.0 23.2 0.0 29.7 n.a. n.a. 100.0 entitlements Based on inputs* 50.5 16.5 0.0 33.1 n.a. n.a. 100.0 Total** 42.4 28.1 0.0 29.6 n.a. n.a. 100.0 EU Market price support 9.0 66.0 25.0 0.0 n.a. n.a. 100.0 Based on output 48.6 37.3 1.5 12.6 n.a. n.a. 100.0 Based on area planted/ 48.0 42.0 0.5 9.4 n.a. n.a. 100.0 animal numbers Based on historical 16.9 66.4 0.0 16.8 n.a. n.a. 100.0 entitlements Based on inputs* 42.0 43.5 0.7 13.8 n.a. n.a. 100.0 Total** 45.1 43.5 1.8 9.7 n.a. n.a. 100.0 Japan Market price support 4.7 1.2 94.1 0.0 n.a. n.a. 100.0 Based on output 0.0 0.4 87.8 11.7 n.a. n.a. 100.0 Based on area planted/ n.a. n.a. n.a. n.a. n.a. n.a. n.a. animal numbers Based on historical n.a. n.a. n.a. n.a. n.a. n.a. n.a. entitlements Based on inputs* 10.6 2.3 80.4 6.7 n.a. n.a. 100.0 Total** 4.7 1.2 92.9 1.2 n.a. n.a. 100.0 Appendix A 325 TABLE A.4 (Continued) Coarse Other G&O Wheat Grains Rice Oilseeds Commodities Total Total (%) Mexico Market price support 11.5 87.3 0.7 0.5 n.a. n.a. 100.0 Based on output 32.8 61.0 2.6 3.5 n.a. n.a. 100.0 Based on area planted/ 11.3 86.6 1.6 0.5 n.a. n.a. 100.0 animal numbers Based on historical 6.8 91.4 0.9 0.8 n.a. n.a. 100.0 entitlements Based on inputs* 9.3 85.4 1.3 4.0 n.a. n.a. 100.0 Total** 12.1 85.6 1.1 1.2 n.a. n.a. 100.0 Switzerland Market price support 38.1 42.4 0.0 19.4 n.a. n.a. 100.0 Based on output n.a. n.a. n.a. n.a. n.a. n.a. n.a. Based on area planted/ 17.5 12.9 0.0 69.6 n.a. n.a. 100.0 animal numbers Based on historical 55.6 35.5 0.0 8.9 n.a. n.a. 100.0 entitlements Based on inputs* 55.6 35.5 0.0 9.0 n.a. n.a. 100.0 Total** 44.3 36.3 0.0 19.3 n.a. n.a. 100.0 U.S. Market price support n.a. n.a. n.a. n.a. n.a. n.a. n.a. Based on output 3.4 27.8 9.6 59.2 n.a. n.a. 100.0 Based on area planted/ 19.9 50.0 1.5 28.6 n.a. n.a. 100.0 animal numbers Based on historical 38.7 59.1 2.1 0.0 n.a. n.a. 100.0 entitlements Based on inputs* 15.0 49.1 2.6 33.4 n.a. n.a. 100.0 Total** 22.3 46.6 4.7 26.4 n.a. n.a. 100.0 * This combines Based upon input use and Based upon input constraints. ** The total includes Based on overall farm income and miscellaneous payments. TABLE A.5 OECD-6 as a Percentage (%) of OECD Total for Selected Commodities and Support Types Coarse Other G&O Wheat Grains Rice Oilseeds Commodities Total Total (%) OECD-6 Market price support 81.0 98.0 67.7 9.7 89.2 85.9 69.9 Payments based on output 75.2 98.1 100.0 100.0 86.5 92.5 97.9 Payments based on area 98.7 97.8 54.9 98.2 95.3 96.7 97.8 planted/animal numbers Payments based on historical 99.8 100.0 100.0 97.8 82.9 95.5 99.9 entitlements Payments based on inputs* 86.8 92.9 84.3 95.9 88.4 88.9 90.4 Total** 94.9 97.8 70.2 94.7 89.2 88.4 86.4 * This combines Based upon input use and Based upon input constraints. ** The total includes Based on overall farm income and miscellaneous payments. 326 Agriculture and the WTO FIGURE A.1 OECD--Grains and Oilseeds 140 Input support 120 Hist. entitlements Area payments OPS 100 trend-total MPS average 80 (percent) trend-MPS 60 Protection 40 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. FIGURE A.2 OECD--Other Commodities Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. trend. A large transformation of policy re-instru- protection with a period average protection compa- mentation has therefore occurred away from MPS rable to that of grains and oilseeds (figure A.10). and IPS to area payments. EU oilseeds have seen a Figure A.11 shows that support for grains and significant reduction in total support (figure A.8) oilseeds has increased in Japan since 1986, with along with rice after 1995 (figure A.9). All other very high protection rates of almost 500 percent commodities in the EU have an upward trend in on average. Support for all other commodities, Appendix A 327 FIGURE A.3 Canada--Grains and Oilseeds 120 Input support 100 Hist. entitlements Area payments OPS 80 MPS (percent) trend-total 60 40 Protection average 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. FIGURE A.4 Canada--Other Commodities 50 Input support 45 Hist. entitlements trend-total Area payments 40 OPS MPS 35 average 30 (percent) 25 20 Protection 15 10 5 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. however, have remained flat and averaged just over entitlements (figure A.13). However, support was 80 percent (figure A.12). MPS in all cases is the bulk very high in the mid-1980s, reaching a high of of support given to agriculture in Japan. almost 450 percent. The trends in support for grains and oilseeds For Mexico, the trends in support for grains and in Switzerland mirrors that of most other countries oilseeds has varied greatly since 1986 although the (other than Japan) with sharp downward trends trend in total support is positive (figure A.14). The in total support and a movement toward historical reader should notice, however, the significant 328 Agriculture and the WTO FIGURE A.5 U.S.--Grains and Oilseeds 70 60 Input support Hist. entitlements Area payments 50 OPS MPS 40 trend-total (percent) 30 average ction 20 Prote 10 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 ­10 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. FIGURE A.6 U.S.--Other Commodities 40 35 Input support Hist. entitlements Area payments 30 OPS trend-total MPS 25 average (percent) 20 trend-MPS 15 Protection 10 5 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. Appendix A 329 FIGURE A.7 EU--Grains and Oilseeds 180 Input support 160 Hist. entitlements Area payments 140 OPS MPS 120 trend-total 100 (percent) average 80 trend-MPS 60 Protection 40 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. FIGURE A.8 EU--Oilseeds 200 180 Input support Hist. entitlements 160 trend-total Area payments OPS 140 MPS 120 (percent) average 100 80 Protection 60 40 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. 330 Agriculture and the WTO FIGURE A.9 EU--Rice 160 Input support trend-total 140 Hist. entitlements Area payments OPS 120 MPS 100 average 80 (percent) 60 40 Protection 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 ­20 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. FIGURE A.10 EU--Other Commodities 140 Input support Hist. entitlements 120 Area payments OPS MPS trend-total 100 average 80 (percent) trend-MPS 60 Protection 40 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. Appendix A 331 FIGURE A.11 Japan--Grains and Oilseeds Input support 700 Hist. entitlements Area payments OPS 600 MPS 500 average trend-total trend-MPS 400 (percent) 300 Protection 200 100 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. FIGURE A.12 Japan--Other Commodities 120 Input support Hist. entitlements Area payments OPS 100 MPS trend-total 80 average (percent) 60 40 Protection 20 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. 332 Agriculture and the WTO FIGURE A.13 Switzerland--Grains and Oilseeds 500 450 Input support Hist. entitlements 400 Area payments OPS 350 trend-total MPS 300 average (percent) 250 200 trend-MPS Protection 150 100 50 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. FIGURE A.14 Mexico--Grains and Oilseeds 100 Input support Hist. entitlements 80 Area payments OPS MPS 60 average (percent) trend-total 40 trend-MPS 20 Protection 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 ­20 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. change in the mix of policies, from MPS and IPS negative protection in the 1980s as did wheat to historical entitlements. (OPS policies were intro- (not shown). Figure A.16 shows that for other com- duced in 2001.) Among the grains and oilseed modities, protection follows a similar pattern except crops, figure A.15 shows that rice had significant more negative support in the 1980s and lower sup- Appendix A 333 FIGURE A.15 Mexico--Rice 60 Input support Hist. entitlements Area payments OPS 40 MPS trend-total 20 average (percent) 0 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 Protection­20 ­40 ­60 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. FIGURE A.16 Mexico--Other Commodities 60 Input support 50 Hist. entitlements Area payments OPS 40 MPS 30 trend-total 20 (percent) average 10 0 Protection 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 ­10 ­20 ­30 Note: "Protection" is the value of support over the value of production at world prices. Source: OECD 2002a. port in the past decade. A summary of policy devel- policy framework called for by the Cairns Group opments in Mexico since 1980 is given in table A.6. and more recently by developing countries is the One question this appendix seeks to address is elimination of the special treatment that devel- whether maintaining the current level of support oped-country agriculture has received in Uruguay but changing its composition will make a major Round Agreement on Agriculture (URAA), which difference to developing countries. The alternative is to say, the elimination of support. A strength of 334 Agriculture and the WTO TABLE A.6 Policy Instruments for Grains and Oilseeds in Mexico (1980­Present) 1980­ 1982­ 1989­ 1991­ 1994­ 1996­ Instrument 1981 1988 1990 1993 1995 2000 2001­ MARKET ACCESS 1 EXPORT 1.1 Licenses 1991­1992 1.2 Tariffs 1.3 NTB's (VER, Phytosanitary (fruits, vegetables, meat, pork) barriers, etc.) 2 IMPORT 2.1 Licenses 1991­1992 2.2 Fixed-rate tariff 2.3 Ad-Valorem tariff 2.4 Seasonal 2.5 Tariff escalation 2.6 TRQ­GATT 2.7 TRQ­NAFTA DOMESTIC SUPPORT 3 STATE TRADING ENTERPRISE 3.1 Guarantee prices Maize and Dry B., till 1994 3.2 Reference prices Maize from 95­98 3.3 Strategic Reserve of Prop. 2001 Grains (SRG) 4 DEFICIENCY PAYMENTS (per ton) 4.1 Local prices of indifference (P. Discrimination) 1994 4.2 Averaged prices of indifference 1996­1998 4.3 Auctions (domestic or foreign market) 1998­2000 4.4 Supporting profitability (direct payments) 5 DIRECT PAYMENTS 5.1 Compensatory payments (per ton) 5.2 Phitosanitary payments (per ha) 1993 5.3 Switched crops (per ha) 1996, 2000 5.4 Transportation subsidies (per ton) 5.5 PACE 6 DECOUPLED PAYMENTS (per ha) 6.1 PROCAMPO w/o cap 6.2 PROCAMPO w/cap (assumed) Assumption 6.3 PROCAMPO w/o cap + yield Assumption weighted (assumed) 7 HEDGING PROGRAM 7.1 Hedging federal budget 1992­1993 1994 7.2 Hedging budget & prices of 1994 indifference Appendix A 335 TABLE A.6 (Continued) 1980­ 1982­ 1989­ 1991­ 1994­ 1996­ Instrument 1981 1988 1990 1993 1995 2000 2001­ 7.3 Service at hedging 7.4 Farmers price support 7.5 Hedging farmers & consumers 1998­2000 8 CREDIT 8.1 Market failures at interest rates 1991­1992 8.2 CCC recycling funds 1993 8.3 Loans in dollars 8.4 Marketing loans 9 MARKETING INFORMATION 1992 10 PRODUCTION SUBSIDIES 10.1 Input subsidies Limited, electricity fees for irrigation 10.2 Alianza para el campo EXPORT SUBSIDIES 11 Deficiency payments 1997 12 Cost of transportation 1997, 1998 13 Exchange--rate loss CONSUMER SUBSIDIES 14 General 1996­1997 15 Selective 16 Added-value taxes 17 Import subsidies 1992­93 1994 1996­1997 18 Official prices Source: Authors. the PEM is that it has been designed to analyze the and usually through a case-study approach, rather differential effects of alternative policies on farm- than through an integrative multicountry or global ers, landowners, input suppliers, taxpayers, con- model.2 One reason for the paucity of research on sumers, and the world price and trade. this important topic is that tracing links between trade and poverty requires detailed analysis of poli- cies and institutions that are country-specific Conceptual Framework (Winters 2001). Chapter 1 of this volume cites several quantitative Exceptions are the research reported in studies estimates of the impact that agricultural trade liber- by Levinsohn, Berry, and Friedman (1999); Hertel alization in rich countries would have on develop- and others (2002); and Robilliard, Bourguignon, ing countries as a whole. There are other estimates and Robinson (2001). Hertel and others' study, for of this from other sources, with broadly consistent example, explicitly models the trade policy­poverty conclusions. These tell us little, however, about connection through factor markets, using the the impact on income distribution and poverty Global Trade Analysis Project general equilibrium within individual countries. This topic has been framework to model the global linkages, mainly addressed, but for only a few developing countries through the effects of liberalization on global price 336 Agriculture and the WTO levels. It draws on national household survey data price provides closure in a national commodity for individual developing countries to estimate market when the quota is binding. expenditures and reliance on different sources When the TRQ is not binding, the volume of of factor income by income strata. This is quite a imports is influenced by the tariff rate and domestic useful framework, since it allows for modeling in demand and production. When the TRQ is binding, broad terms the effects of all trade policy changes, however, then it alone determines the volume of agricultural and nonagricultural, on all house- imports. Imports affect domestic prices and the holds, agricultural and nonagricultural. world market in both situations. These differences As with all modeling, a computable general equi- in market structure can have important conse- librium model (CGE) has some disadvantages--it quences for policy. For example, raising OPS may does not model, for example, the effects of agricul- have very different effects on trade distortion and tural support policies administered through non- transfer efficiency, depending on whether there is a border measures, or the nontrade barriers to binding quota or not. agricultural trade.As indicated below, given the idio- Although the PEM analyzes only four commod- syncratic nature of agricultural policies and the dif- ity groups, it includes linkages with the factor ferences in the effects of different support policies, markets, thereby showing the impact on factor careful modeling of these can be quite important in earnings from various policy changes. There are accurately reflecting the impacts of modifying them. two specialized input markets in the PEM, one for This point is underscored by the fact that many of land and one for other farmer-owned inputs. the recent changes in OECD agricultural policies A specialized inputs market is one that links one have been in the type of domestic support; this and source of demand to each source of supply. The non-tariff barriers will be major issues for the devel- farmer-owned inputs market is modeled like that oping countries in the Doha Round. for purchased inputs. This appendix presents an alternative approach This approach, combined with household sur- to examining the trade­poverty linkages and the vey data on sources of income and expenditure, effects on income distribution within developing allows a detailed mapping from policy instrument countries of trade reform by rich countries and by to effect on each type of household in its role of the developing countries themselves. The approach producer, consumer, and taxpayer. This informa- uses the OECD's Policy Evaluation Matrix (PEM) tion should be useful in examining not only the model. The strengths of this approach are that impacts on the poor in developing countries, but (a) this model was specifically developed to capture also the question of who gains and who loses in the the differential effects of the many types of agricul- rich countries. In the context of the Doha Round tural support policies, and (b) it specifies produc- negotiations, both of these are important consider- tion functions for particular crops (wheat, coarse ations in the political economy of reform. grains, rice, and oilseeds) that are protected more In the PEM, production is represented with con- than other agricultural products on average and in stant elasticity of substitution (CES) production particular are given a disproportionately high level functions in producer prices for wheat, coarse of support through non-border measures such grains (including maize), oilseeds, and rice for six as area payments, historical entitlements, and other major producing regions. The input markets in direct payments. These crops are also of great inter- each region are simulated with demand (first order est to a large number of developing countries either conditions) derived from the production functions. as net exporters or importers. Food security con- Factor supply is for the most part simulated with cerns in many developing countries focus on the constant elasticity factor supply functions, as in most important staple crop, which is most fre- many CGE models, but elasticities are specified tak- quently rice, maize, or wheat. ing into consideration the scope of the model and In some scenarios used in this report, the market the constraints on resource mobility in agriculture. structure may be varied to represent the operation In this framework, commodity supply is repre- of tariff rate quotas (TRQs). In this situation, the sented in terms of an aggregate production func- link between domestic prices and world prices tion and the associated factor demand and factor depends on the regime. A separate domestic market supply functions (see table A.7 for a summary of Appendix A 337 TABLE A.7 Model Outline Symbol Definition 1: Symbols Defined qsir,qdir quantities supplied and demanded of commodity i in region r psir,pdir,pi0 the supply or demand price of commodity i in region r; the world price of commodity i xskir,xskr,xdkir the quantity of input k supplied or demanded for the production of output i in region r; the index i is dropped for aggregate supply wskir,wskr,wdkir the supply or demand price of input k for the production of output i in region r; the index i is dropped for aggregate supply price skir, kir the share of input k used in the production of output i in region r and the corre- sponding share parameter for the production functions rir,sir the elasticity parameter for commodity i in region r, and the corresponding elasticity of substitution for the production functions sijr, dijr the elasticity of supply or demand for commodity i, with respect to the price of commodity j in region r bkjir,bkir,bkr the input supply elasticity of input k for commodity i in region r, with respect to input price k, used for commodity j; in region r; the input supply elasticity with respect to own price for commodity i in region r; and the input supply elasticity with respect to own price in region r rir mps ,rops,rir,rr k k ir the support rate for commodity i, input k, in region r; the superscript mps and ops are used for market price support and output price support, respectively, while the superscript k is shown for policies that affect factor markets csir,cdir the constants that reflect quality differences (location, season, varieties, and so on) which cause differences between world and regional commodity prices i, j the indexes for the four commodities; i, j 1, 2, 3, 4 for wheat, coarse grains, oilseeds, and rice, respectively k the indexes for the four inputs; k 1, 2, 3, 4 for land, other farmer-owned inputs, hired labor, other purchased inputs, respectively r the index for region; r 1, 2, 3, 4, 5, 6, 7 for Canada, Mexico, the United States, the European Union, Switzerland, Japan, and the rest of the world, respectively; r 0 is used in the equations that feature the world price There are four parts to the PEM model: · commodity demand, commodity supply or production, and equilibrium conditions for the commodity markets; · the demand, supply, and equilibrium conditions for the factor markets; · the price relationships; and · the equilibrium conditions. 2: Commodity Markets Commodity demands are expressed as a function of demand prices and elasticities with a constant elasticity functional form: [A.1] qdir qdoirpd1 di1r di2r di3r di4r r pd2r pd3 r pd4 r i 1,2,3,4 r 1,2, . . . ,7 The quantity supply for the rest of the world region has the same form: [A.2] qsi7 qsoi ps17 ps27 ps37 ps47 si17 si27 si37 si47 7 i 1,2,3,4 Supply for the six endogenous regions are based on a constant elasticity of substitution (CES) production function. Production is generally a function of four inputs: land, other farmer-supplied inputs, hired labor, and other purchased inputs, except in the cases of Mexico and Japan where a hired labor input is not specified separately: [A.3a] qsir ( 1ir xd1­rir ir 2ir xd2 ­rir ir 3irxd3 ­rir ir 4irxd4 ) ­rir ­1/rir ir i 1,2,3,4 r 1,3,4,5 or [A.3b] qsir ( 1ir xd1­rir ir 2ir xd2 ­rir ir 4irxd4 ) ­rir ­1/rir ir i 1,2,3,4 r 2,6 No rice produced in Canada or Switzerland. 338 Agriculture and the WTO TABLE A.7 (Continued) Market equilibrium requires equating world supply and demand for each of the four commodities: 7 7 [A.4] qsir = qdir i = 1,2,3,4 r =1 r =1 3: Factor Market Relationships The market structure for factors reflects the fact that both land and farmer-owned inputs are relatively specialized and do not freely change use. Even in the case in which farms produce a mix of crops, these will be grown in rotations that restrict the ability to change the output mix. This is especially the case for paddy rice land, which is highly unlikely to be used in the production of other crops in the model, at least within the range of scenarios that the model is suitable for evaluating. A separate supply function is specified for land by commodity and region, although the supply function for land includes cross-price terms except in the case of rice: [A.5] xs1ir xso1 ws1 b11 ir ir i 1,2,3 r 1,2, . . . ,6 ir ws2 ws3 b12ir b13ir ir ir The supply function of land for rice production is: [A.6] xs14 r xso14rws14r b14 r r 2,3,4,5 A separate supply function is specified for other farmer-owned inputs by commodity and region: [A.7] xs2ir xso2 ws2 b2 ir ir i 1,2,3,4 r 1,2, . . . ,6 ir Neither purchased inputs nor hired labor is specialized. A single supply function by region is specified for hired labor and purchased inputs: [A.8] xs3r xso3rws3r b3r r 1,2, . . . ,6 and [A.9] xs4r xso4 ws4 b4 r r r 1,2, . . . ,6 r Factor demand in all cases is represented by the first order condition: îqsir [A.10a] wdkir psir îxskir k 1,2,3,4 i 1,2,3,4 r 1,2, . . . ,6 For the CES production function, the first order condition can be written as: ir kirpsir [A.10b] xdkir qsir wdkir Market equilibrium equates supply and demand in each input market in each region. For land and other farmer-owned inputs these are: 4 [A.11] xskir xdir k 1,2 i 1,2,3,4 r 1,2, . . . ,6 r =1 For hired labor and other purchased inputs, the factor market equilibrium conditions are: [A.13] xskr xdkir k 3,4 r 1,2, . . . ,6 4: Commodity and Factor Price Relationships Agricultural policy and other institutional factors create a set of wedges between world prices and demand and supply prices. [A.14] pdir pi mps 0 rir cdir i 1,2,3,4 r 1,2, . . . ,6 [A.15] psir pi mps ops 0 rir rir csir i 1,2,3,4 r 1,2, . . . ,6 [A.16] wdkir wskir rirk k 1,2 i 1,2,3,4 r 1,2, . . . ,6 [A.17] wdkir k wskr rr k 3,4 r 1,2, . . . ,6 the model structure). This allows one to analyze Cost-Benefit Analysis a change in the value of some exogenous policy In calculating the costs and benefits of support parameter, for example an administered price, an measures, inputs were classified according to area payment, or an input subsidy. whether they were supplied by or purchased by Appendix A 339 FIGURE A.17 Production and Input Supply Responses to Increased Commodity Price d a p Qw Q* Q Da Da Qa Qb Db Db Qa Qb farm households. Inputs supplied by farm house- initial incidence, though, the price and quantity holds--for instance, land and farm household links in the model tie the factor and output markets labor--are called "farm owned." At least three together so that a change in a policy measure that kinds of inputs purchased by farm households were affects any one of the crop or factor markets will end distinguished: fertilizer; hired labor; and a residual up having some effect on all of them. Figure A.17 aggregate, "other purchased." contains supply and demand diagrams illustrating An important indicator of the effect of a policy the basic components of this representative model. is the "initial incidence" or "which price is affected The upper panel shows the commodity supply first." Market price support has its initial incidence curve (the demand curve is ignored) and the lower in commodity markets while area payments and two panels show supply and demand curves for two subsidies to purchased inputs have their initial inci- of the four factors of production (purchased inputs dence in the associated factor markets. Regardless of and land, respectively). 340 Agriculture and the WTO Figure A.17 shows how price wedges correspond- its due to the increase in price from Pw to Pd. Area ing to unit market price support or output price sup- c d + e represents the increase in costs due to the port are represented in the PEM crop models. The higher input prices. The increase in the farm opera- wedge separates prices paid to domestic producers, tor's welfare is now smaller than would be the case if Pd, from the corresponding price on world markets, input prices were constant, but still greater than zero. Pw. The accompanying level of price support is The areas marked i and j in the middle panel of indicated by the area a d e f g h and figure A.17 represent the impact of the change in represents the sum of taxpayer and consumer market price support on net incomes of firms sup- costs (or taxpayer costs only if output price support plying purchased inputs. Correspondingly, areas only). marked l and m in the lower panel show increased Consider first the case of no change in input profits for landowners. Which of these is the greater factor prices as production of the crop expands depends on the elasticities of factor supply and sub- with increased producer prices. The farmer would stitution as well as on the relative importance of the move along the original supply curve S and pro- factor bundles in crop production. duce at Q. Total revenue would increase by The price increase will always be the greater for (Pd­Pw)Qw (Q­Qw)Pd. However, farmers' costs the factor (or factor bundle) exhibiting the lowest increase by area g h o p (albeit with no supply elasticity, which is land in many cases. changes in input prices by assumption) so the net However, this does not guarantee that the largest gain in producer profits is area a d e f. share of total benefits of support go to this factor However, if the supply curve for purchased inputs since that depends on factor shares as well as elas- and land were upward sloping (as depicted in the ticities. The essential point is that there will be bottom two panels in figure A.17), the increase in some sharing of the economic benefits of increased demand for inputs as the production of the com- support among these two groups of economic modity increases (owing to the policy of increasing agents. But if land supply is very inelastic and sup- producer prices initially) will generate an increase in ply of other inputs very elastic, then land captures input prices. a disproportionate amount of the rents. The bottom panels of figure A.17 show the Area payments is a category of support that has associated factor market effects. The hypothesized gained importance in recent years. Such payments increase in the producer price due to increased mar- increase the rental rate received by a landowner for ket price support translates into outward shifts in land used in crop production as compared to the demand for both purchased factors of production rental rate that land would earn in other uses and land as shown by the new input demand curves (unless mitigating effects such as payment limita- labeled Da and Db in the figure. This causes the tions, set-aside requirements, voluntary set-aside quantities and prices of both factors to rise, the and base area limitations override it). degree to which clearly depends on relative elasticities Area payments create a wedge between the price (slopes) of the factor supply schedules and the mag- a farmer earns from using his land and other owned nitude in the shift in input demand curves (which, in factors in crop production. Similarly, subsidies to turn depends on the share of input in production and purchased inputs create a difference between the the relative production elasticities for the inputs). price suppliers receive and the price farmers pay The increase in input prices will cause the supply for them. curve for output to shift upward to S in the top A payment to land that comes with a requirement panel of figure A.17. Output will now increase to Q* to plant induces an increase in the area of land used only and the supply curve with input prices chang- in crop production. This payment drives a wedge ing is given by S*. S* is the equilibrium adjustment between the effective rental rate received for land supply curve that plots the production response to used in crop production and the associated rental the increase in prices received by farmers in the pres- rate charged (perhaps implicitly) for that land. The ence of upward-sloping input supply curves (shown equilibrium supply price for land rises, while the for the purchased input and land markets). Farm equilibrium demand price falls, leading to an increase operators and input suppliers gain area a minus area in the quantity of land and other farm-owned factors c d + e.Area a is the increase in farm operator prof- demanded and used in crop production, which will Appendix A 341 be greater or less depending on the elasticity of sup- Graphical analysis is helpful in illustrating dif- ply of crop land reflected in the slope of the supply ferences in the channels through which support curve for farm-owned factors. In turn, this increased measures produce their effects, and differences factor use would show up as a rightward shift of the in the direction of impacts on selected outcomes. commodity supply schedule that is shown in the top However, the PEM simulations to follow will indi- panel of figure A.17 and would cause the associated cate differences in both the direction and the rela- increases in supply and net trade. tive magnitude of effects of small changes in The increased supply results in some drop in market price support, payments based on output, world market prices. To the extent that the world payments based on input use, and payments based market price falls, and that this price fall is trans- on area planted. Taken together, these findings con- mitted to the domestic market, consumption of the firm that the estimated trade and welfare effects of commodity will increase. This is one reason to sus- a given amount of support differ depending on the pect that increased support made in the form of support measure used to provide that support. a payment tied to land would have less of an effect The outcome depends on economic efficiency on trade than the same amount of support provided losses and distributive leakages (reallocations) to as market price support. input suppliers, consumers, taxpayers, and the rest of Another reason has to do with the fact that the the world through market price changes. The factors payment is a subsidy targeted to one factor of pro- governing the impacts of agricultural policies include duction--land--while market price support may be the structural characteristics of the agrifood sector viewed as a subsidy spread more or less evenly across (notably the share of inputs supplied from off-farm all inputs. The former has less effect on output, sources, the share of income coming from off-farm based on the assumption that the elasticity of supply activities, and the extent of competition upstream of land is less than the elasticity of the supply of and downstream from the farm gate); the response of nonland factors of production. There is an income these characteristics to changes in the level of support gain for farm households and input suppliers as (production technology, supply elasticities of inputs, well. Input suppliers gain, in this example, because demand elasticity of the output, world market condi- the increase in commodity production creates addi- tions, and so on); and the form and distribution tional demand for purchased inputs. This is repre- through which support is provided (input subsidy sented as the rightward shift in the demand curve versus a tariff, and so forth). for purchased factors. The reader should notice that the extra produc- Structure of PEM tion constitutes only one of two channels through which suppliers of purchased inputs may be The technique used in the PEM shares many features impacted by the area payments. It should be recalled with general equilibrium modeling but also draws that the model assumes farm-owned factors and on some of the techniques used in the equilibrium purchased inputs are substitutes in production. displacement models found in the agricultural eco- Insofar as production decisions are concerned, an nomics literature. The latter use multimarket equi- area payment reduces the cost of land (rental rate) librium elasticities to determine the incidence on all relative to nonland factors of production. If this were stakeholders of interventions in agricultural input the only effect, the demand for purchased inputs and output markets. These are based on differentiat- would decline. The increased demand shown reflects ing a system of equations representing production a "net" substitution effect that is more than offset by functions, first-order conditions, factor supply, and the scale effect associated with increased production. commodity demand. The results are then calibrated There is again a sharing of the benefits of sup- with cost share and elasticity parameters and solved port among farm households and input suppliers, for the multimarket equilibrium parameters. but this time farm households might be expected to Multimarket equilibrium models share four get the lion's share if they are landowners as well. In characteristics of general equilibrium models: fact, this result, like the result for commodity trade, also depends heavily on the relative elasticity of · The model is calibrated from base period data supply of land. (consisting of demand, production, and factors 342 Agriculture and the WTO of production or production shares) and a set of portation, wholesale, and retail services. Goods and elasticities. services that are genuine inputs are transformed into · Each industry is treated as a representative other products that do generate derived demands for firm. Consumption is based on a representative intermediate inputs. When the PEM is changed by consumer. including demands for different households these · The payments to capital are determined residu- intermediate demands need to be considered. ally; in the case of unincorporated businesses Finally, many of the expenditure items associated these may not be dissociated into the compo- with cereals and oilseeds in household budget sur- nent parts of value added. veys reflect a great deal of other value added. · The models simulate normative adjustments to Consumers seldom buy wheat but rather flour, a change in policy or exogenous variables by the bread, wheat tortilla, and a great variety of other representative firms and consumers based on similar products. (Flour may be classed as a primary competitive market behavior (usually) rather product in CGEs because input-output tables some- than the positive behavior built into econo- times treat milling as a service industry rather than a metric models. manufacturing industry.) Much of the maize and rice purchased is in the form of a grain or flour so The behavior assumption used in both types of there is far less value added. A method is needed, models means that they are inherently medium- or therefore, to reconcile values at the farm gate with long-term mapping outcomes from one equilib- intermediate use and values at the consumer level. rium position to another. One possibility is to go to an input-output table--or shipments data that lie behind input- A Disaggregated System output tables--to determine quantities used in feed, of Demand and Production manufacturing, and final demand. The resulting for Mexico data can be used to construct simple two-input This section of the appendix describes how a disaggre- gross output production functions (in primary gated approach can be used to model both production input and value added) following Gardner (1975).3 and demand for Mexico.It can show the impact on the The easier--if less satisfactory--approach used here distribution of impact across consumers of different is to disaggregate feed use based on the attribution income levels and it does not differentiate impact and of MPS found in the PSE tables. supply response between subsistence and commercial Table A.8 shows expenditure by income decile farmers.Disaggregated supply and demand sectors are from a 2000 household expenditure survey of needed to simulate the effects of changes in policy on about 10,000 Mexican households. Expenditures different types of farm households in Mexico. for wheat, maize, and rice are aggregations by class The production function for coarse grains is of the categories listed in table A.9. replaced with a household production system for To estimate expenditures in 2001, the expenditure five types of maize-producing farms varying princi- pattern found in the household survey (showing pally by farm size. A single production function con- greater expenditures than the value of farm produc- tinues to be used for wheat, other coarse grains, tion) is "rased" with the farm-level value of con- oilseeds, and rice. The constant elasticity demand sumption from table A.10 and production and total function is replaced with a linear expenditure system expenditures for each decile in 2001 as control totals. for each of the five maize-producing farm house- Total expenditures in 2001 is total expenditures in holds and 10 others varying by household income. 2000 grossed up by 10.2 percent, which is the growth in private final consumption expenditures (OECD 2002c, p. 3, converted to current U.S. dollars). This Expenditure and Consumption Distributions procedure essentially assigns the value added in mar- The PEM bases demand upon commodities valued keting and food manufacturing industries into the at producer prices except for differences between other demand. Table A.11 shows the resulting alloca- consumer and producer prices caused by subsidies. tion of consumption in 2001 expressed in U.S. dol- It follows CGE practice, whereby goods are valued in lars per capita. Dividing these by the effective producer prices with the value added in the market- consumer prices calculated from the PSE dataset ing chain resulting in a separate demand for trans- (using a unit price for "other") gives quantities. Appendix A 343 TABLE A.8 Expenditure by Decile, 2000 in Million Pesos (Per Capita in Parenthesis, US$) Decile Wheat Maize Rice Other Total 1 988 (2.7) 2,173 (24.4) 340 (1.6) 28,380 (351) 31,881 (379) 2 1,456 (3.9) 3,559 (39.7) 409 (1.9) 50,828 (624) 56,253 (669) 3 1,823 (4.8) 4,035 (44.6) 353 (1.6) 68,720 (841) 74,931 (892) 4 2,048 (5.4) 4,250 (46.8) 341 (1.6) 86,132 (1,050) 92,771 (1,104) 5 2,362 (6.2) 4,070 (44.7) 372 (1.7) 108,119 (1,315) 114,924 (1,368) 6 2,636 (6.9) 4,051 (44.3) 364 (1.7) 130,671 (1,586) 137,721 (1,639) 7 3,209 (8.4) 4,074 (44.5) 404 (1.8) 165,809 (2,010) 173,497 (2,065) 8 3,478 (9.0) 3,988 (43.3) 331 (1.5) 212,290 (2,565) 220,087 (2,619) 9 3,575 (9.2) 3,788 (41.0) 299 (1.4) 288,830 (3,477) 296,492 (3,528) 10 4,370 (11.2) 3,093 (33.2) 478 (2.1) 783,026 (9,367) 790,967 (9,413) Total 25,946 (6.8) 37,083 (40.6) 3,690 (1.7) 1,922,805 (2,319) 1,989,524 (2,368) Source: Private communication, Allesandro Necita (July 2002). TABLE A.9 Allocation of Household Survey Expenditure Categories Code Description Class A001 Maíz en grano, pozolero, palomero Coarse grains A002 Harina de maíz Coarse grains A003 Masa de maíz Coarse grains A004 Tortilla de maíz Coarse grains A005 Fécula de maíz (maicena, polvo para atole) Coarse grains A006 Otros productos de maíz: tostadas, hojuelas, pinole, etc. Coarse grains A007 Harina de trigo (refinada o integral) Wheat A008 Tortilla de harina Wheat A009 Galletas saladas Wheat A010 Galletas dulces Wheat A011 Pan blanco: bolillo, telera, baguette Wheat A012 Pan de dulce: en pieza o empaquetado Wheat A013 Pan para emparedado, hamburgesa, hot-dog Wheat A014 Pasteles y panecillos en pieza o empaquetado Wheat A015 Pasta para sopa Wheat A016 Otros productos de trigo: pasta para fritura, Wheat hojuelas, harina preparada, etc. A017 Arroz en grano Rice A018 Otros productos de arroz: harina, tostado, etc. Rice A019 Avena Coarse grains A020 Otros cereales: centeno, granola, cebada, etc. Coarse grains A021 Frituras procesadas de trigo o maíz Coarse grains Source: Authors' calculations. Calibration of the Linear Expenditure tive to the fixed proportions would be the constant System (LES) elasticity of transformation among the three com- modities, but this option is not pursued here. In this appendix the three cereals are treated as a sin- Arc income elasticities were calculated for the gle commodity within the linear expenditure system change in expenditures between each income decile and expenditures are allocated in fixed proportions in table A.8 and smoothed using a three-category by income class among the three cereals. An alterna- 344 Agriculture and the WTO TABLE A.10 Maize Production Systems Subsistence Very Sum or Item Small Small Medium Large Average Farm type (size range, Ha) 2 2­5 5­10 10­18 18 Average size (Ha) 1.3 3.4 7.3 14 37.8 11.3 Total land area (Ha) 108,625 549,250 1,161,199 1,741,706 3,439,220 7,000,000 Share of households (%) 13.5 26.1 25.7 20.1 14.7 100 Number of households 83,557 161,544 159,068 124,408 90,985 618,943 Yield (t/Ha) 1.46 1.88 2.64 2.64 2.64 Adjusted production (t) 164,618 1,071,829 3,182,058 4,772,832 9,424,563 18,615,900 Labor (no/Ha) 1.63 0.68 0.41 0.41 0.41 Total labor 177,142 371,552 477,205 714,100 1,410,080 3,150,078 Population 531,425 1,114,655 1,431,616 2,142,299 4,230,240 9,450,234 Expenditure per capita 3,415 4,353 6,418 8,913 14,261 (pesos) Sources: Winters, Davis, and Corral (2001); Authors' calculations; and SARH-DGE 1991 as reported in Heinegg (2001). TABLE A.11 Expenditures Per Capita by Decile, 2001 (U.S. dollars) Decile Wheat Maize Rice Other Total 1 2.7 24.4 1.6 351 379 2 3.9 39.7 1.9 624 669 3 4.8 44.6 1.6 841 892 4 5.4 46.8 1.6 1,050 1,104 5 6.2 44.7 1.7 1,315 1,368 6 6.9 44.3 1.7 1,586 1,639 7 8.4 44.5 1.8 2,010 2,065 8 9.0 43.3 1.5 2,565 2,619 9 9.2 41.0 1.4 3,477 3,528 10 11.2 33.2 2.1 9,367 9,413 Average 6.8 40.6 1.7 2,319 2,368 Source: Authors' calculations. moving average. The resulting income elasticities share income such that it generates satisfactory values are shown in table A.12. overall in terms of price elasticities and the minimum One approach to calibrating an LES therefore consumption parameters. The individual minimum might be to solve equations for the parameters from consumption parameters can then be found simply base period shares and a set of own price elasticities by inverting the demand functions. Using this proce- and income elasticities found in the literature. This dure will generally result in reasonable values for the approach is sensitive to the values for the income and minimum consumption parameters consistent with own price elasticity selected. If those are not consis- their interpretation. The value for the minimum con- tent with the overall LES assumption and the shares, sumption bundle can be set to give the most reason- it is easy to generate contradictory signs on the cali- able values for all own price elasticities. brated parameters and unreasonable elasticities. To calibrate the minimum consumption param- The calibration approach used here is to set the eters, the minimum consumption bundle was set at value of the minimum consumption bundles as a 75 percent of observed expenditure of the lowest Appendix A 345 TABLE A.12 Expenditure Parameters and Elasticities for the Base Period, 2001 Decile Quantitya Alpha Gammaa Income Elasticity Own Price Elasticity Cereals: 1 0.219 0.069 0.169 0.632 0.283 2 0.348 0.051 0.199 0.526 0.458 3 0.392 0.025 0.275 0.303 0.316 4 0.412 0.015 0.318 0.210 0.240 5 0.402 0.014 0.289 0.230 0.290 6 0.403 0.011 0.292 0.211 0.285 7 0.416 0.006 0.340 0.126 0.187 8 0.410 0.000 0.401 0.013 0.021 9 0.392 0.000 0.383 0.013 0.022 10 0.350 0.000 0.341 0.013 0.026 Other goods: 1 351 0.931 262 1.045 0.948 2 624 0.949 259 1.051 0.979 3 841 0.975 249 1.063 0.993 4 1,050 0.985 243 1.061 0.997 5 1,315 0.986 247 1.048 0.997 6 1,586 0.989 246 1.043 0.998 7 2,010 0.994 240 1.041 0.999 8 2,565 1.000 232 1.036 1.000 9 3,477 1.000 234 1.026 1.000 10 9,367 1.000 239 1.010 1.000 a. The units are tonnes for cereals. The quantity of other goods are measured by an index. Source: Authors' calculations. TABLE A.13 Maize Production Coefficients Input: Subsistence Very small Small Medium Large Farmer-owned 0.53 0.35 0.15 0.15 0.16 Purchased inputs 0.00 0.28 0.59 0.59 0.57 Source: Authors' calculations. income class or US$285 per year. This implies ­1.0, especially for the high-income deciles. The a value of 4 for the flexibility of income for this income elasticity of cereals is about 0.6 for low- income decile, which is high in absolute terms but income households, declining with increasing not unreasonably so for the lowest income decile in income to nearly zero for high-income house- a developing country. holds. Own price elasticity is 0.283 for lowest- The minimum consumption levels for cereals income households but it rises to 0.458 for range from 169 to 400 kilograms per person per households in decile 2. This is because of the year or 0.5 to 1.1 kilograms per person per day. increase in the size of the gamma parameter. Cereals are such a small part of the budget that Thereafter the absolute value of own price elastic- income elasticities for other goods and services ity declines to nearly zero. The own price demand are close to 1.0 and own price elasticities are near elasticity for wheat, maize, and rice in the original 346 Agriculture and the WTO PEM are 0.76, 0.15 and 0.24, respectively, so coefficients are based on Heinegg, with some judg- the values calibrated from the LES should give ment by the authors. similar overall response to a change in cereals Following WDC, maize production is repre- prices--although they would vary considerably in sented by five different types of farms varying by their details. size (table A.10). The average farm size of 11.3 Demand by farm households is disaggregated hectares can be calculated directly from the distri- from the total demand in table A.12 by solving the bution of farm types and the average size by farm appropriate demand functions by decile. Both sub- type. Heinegg reports 6.9 million hectares of land sistence and very small farms are in decile 1 while for maize in 1991. Dividing 7 million hectares by small, medium-size, and large farms are in deciles the average farm size gives the estimate of 620,000 2, 3, and 5 respectively. (The farm households are households and the number of households by each described in the next section.) Both the nonfarm farm type. population and this population's income can be Combining Heinegg's yields with the new distri- found by subtracting out the farm portions. Since bution of land by farm types gives a total estimate the demand systems are linear, total demand is not of maize production of 17.9 million tons. This was changed. The resulting demand system then has grossed up 4 percent to match with the value for 15 household types. production in the OECD data. Combining Heinegg's labor per hectare with the new distribution of land by farm types gives a total Disaggregated Maize Production estimate of labor use of 3.1 million people (for at The production coefficients in the PEM are based least part of the year). Assuming two additional on national costs of production surveys by Mexico's family members per worker gives a total population Ministry of Agriculture, SARH-DGE. Unfortunately partially dependent on maize production of 9.4 mil- this report is not yet published so an estimate of lion, which is about 37 percent of the rural popula- costs of production has been made here based on tion in 2001. It should be possible with better other available sources.4 The result allows the model information on the actual share of labor on maize to be fully calibrated to illustrate its application. production to refine these estimates and include There has been a major change in the structure a labor supply function as part of the LES. of farming in Mexico in the 1990s, perhaps because Heinegg also reports household expenditure for of the land reform in 1992 and the major changes maize and other crops. Total income is estimated by in the structure of support after Mexico joined the applying this ratio to maize expenditure. Results for North American Free Trade Agreement (NAFTA) 1991 are then grossed up to 2001 according to in 1994. Heinegg (2001) shows 2.8 million house- the ratio of rural value added in 2001 to value holds producing maize in 1991, 63 percent of which added in 1999. had less than two hectares. Winters, Davis, and Production shares reported in table A.13 are as Corral 2001 (WDC) have 86.5 percent of their reliable as the data in table A.10. The land income by households with more than two hectares. This is a farm type was calculated by taking the total returns rather critical level, because, according to Heinegg, to labor for maize production in PEM and allocat- at about two hectares households switch from ing to each farm type according to its share of land being net buyers of maize to net sellers. in table A.13. Dividing land income by the value of The Heinegg distribution of households is production gives the land share. The share for other inconsistent with expenditure levels in the 1996 farm-owned input is found analogously using the and 2000 household expenditure surveys: it gener- distribution of labor in table A.10. Purchased inputs ates too many people in the lowest expenditure were calculated as a residual except for subsistence decile. The structure used in the appendix therefore farms in which the farmer-owned input was the is based on WDC for the present, but production residual and purchased inputs set to zero. Appendix A 347 Notes Levinsohn, J., S. Berry, and J. Friedman. 1999. "Impacts of the Indonesian Economic Crisis: Price Changes and the Poor." 1. A large component of the other 39 percent in this category Paper prepared for the Conference on Poverty and the are EU headage payments. International Economy, organized by the World Bank and 2. Reimer (2002) has a useful survey. Swedish Parliamentary Commission on Global Development, 3. See Short (2002) for an application of this method in a Stockholm, October 20­21, 2000. PEM context. Both two- and three-input gross output process- Löfgren, H. 1999. "Trade Reform and the Poor in Morocco: A ing sector production functions were calibrated with value Rural-Urban General Equilibrium Analysis of Reduced added disaggregated in the three-input case. Protection." IFPRI TMD Discussion Paper 38. 4. Principally Heinegg (2001), which makes use of 1991 OECD (Organisation for Economic Co-operation and Develop- SAH-DGE and Winters, Davis, and Corral (2001), which is ment). 2001."Market Effects of Crop Support Measures."Paris. based on the de Janvry survey of 951 Ejido households in 1997 ------. 2002a. "Agricultural Policies in OECD Countries: for the World Bank. Monitoring and Evaluation 2002." Paris. ------. 2002b. "The Transfer Efficiency of Farm Support for Farm Household Income." Paris. Select Bibliography ------. 2002c. "Main Economic Indicators." June. Reimer, J. 2002. "Estimating the Poverty Impacts of de Janvry, A., and K. Subbarao. 1986. Agricultural Price Policy Trade Liberalization." GTAP Working Paper 20. and Income Distribution in India. Delhi, India: Oxford Purdue University Center for Global Trade Analysis University Press. and Department of Agricultural Economics. West Lafayette, Dewbre, J. H., and C. Short. 2002. "Alternative Policy Instruments Indiana. for Agriculture Support: Consequences for Trade, Farm Robilliard, A.-S., F. Bourguignon, and S. Robinson. 2001. "Crisis Income and Competitiveness." Presented at the Canadian and Income Distribution: A Micro-Macro Model for Agricultural Economics Association annual meeting in Indonesia." Paper prepared for the ESRC Development Calgary, May 30­June 1. Economics/International Economics Conference, Nottingham Dewbre, J. H., J. Anton, and W. Thompson. 2001. "The Transfer University, U.K., April 5­7, 2001. Efficiency and Trade Effects of Direct Payments." American Rodrik, D. 2000."Comments on `Trade, Growth, and Poverty,' by Journal of Agricultural Economics 83:1204­14. David Dollar and Aart Kraay." Available at European Commission. 1997. "A Matrix Approach to Evaluating http://ksghome.harvard.edu/~.drodrik.academic.ksg/papers Changes in the EU's Crop Regime." Brussels, Belgium, .html. September. Sadoulet, E., and A. de Janvry. 1995. Quantitative Development Gardner, B. L. 1975. "The Farm Retail Price Spread in a Policy Analysis. Baltimore: the Johns Hopkins University Competitive Food Industry." American Journal of Agricul- Press. tural Economics 57:399­409. Short, C. 2002. "A Methodology to Evaluate the Impact of Cost Gardner, B. L. 1987. The Economics of Agricultural Policies. New Recovery in Canadian Agriculture and Agri-Food Chains." York: Macmillan. Research and Analysis Directorate, Agriculture and Agri- Hertel, T., P. Preckel, J. A. L. Cranfield, and M. Ivanic. 2002. Food Canada. "Poverty Impacts of Multilateral Trade Liberalization." GTAP Singh, I., L. Squire, and J. Strauss, eds. 1986. Agricultural Working Paper 16. Purdue University Center for Global Household Models: Extensions, Applications and Policy. Trade Analysis and Department of Agricultural Economics. Baltimore: The Johns Hopkins University Press. West Lafayette, Indiana. Tyers, R., and K. Anderson. 1992. Disarray in World Food Heinegg, A. 2001. "Effects of NAFTA and PROCAMPO on Markets: A Quantitative Assessment. Cambridge, U.K.: Mexican Maize Producers: A Farm-Household Micro- Cambridge University Press. simulation Model." M.Sc. Thesis, Department of Applied Winters, A. 2001. "Trade, Trade Policy and Poverty: What Are Economics and Management, Cornell University, Ithaca, the Links?" Centre for Economic Policy Research Policy New York. Paper No. 2382, London. Ianchovichina, E., A. Necita, and I. Soloanga. 2001. Winters, P., B. Davis, and L. Corral. 2001. "Assets, Activities and "Implications of Trade Reform on the Distribution of Income Generation in Rural Mexico: Factoring in Social and Income in Mexico." Policy Research Working Paper 2667. Public Capital." Working Paper, Agricultural and Resource The World Bank Development Research Group, August. Economics, University of New England, Australia, June. Appendix B The Agreement on agriculture AGREEMENT ON AGRICULTURE as it appears on the WTO Website. Members, Having decided to establish a basis for initiating a process of reform of trade in agriculture in line with the objectives of the negotiations as set out in the Punta del Este Declaration; Recalling that their long-term objective as agreed at the Mid-Term Review of the Uruguay Round "is to establish a fair and market-oriented agricultural trading system and that a reform process should be initiated through the negotiation of commitments on support and protection and through the establishment of strengthened and more operationally effective GATT rules and disciplines"; Recalling further that "the above-mentioned long-term objective is to provide for substantial progressive reductions in agricultural support and protection sustained over an agreed period of time, resulting in cor- recting and preventing restrictions and distortions in world agricultural markets"; Committed to achieving specific binding commitments in each of the following areas: market access; domestic support; export competition; and to reaching an agreement on sanitary and phytosanitary issues; Having agreed that in implementing their commitments on market access, developed country Members would take fully into account the particular needs and conditions of developing country Members by pro- viding for a greater improvement of opportunities and terms of access for agricultural products of particu- lar interest to these Members, including the fullest liberalization of trade in tropical agricultural products as agreed at the Mid-Term Review, and for products of particular importance to the diversification of pro- duction from the growing of illicit narcotic crops; Noting that commitments under the reform programme should be made in an equitable way among all Members, having regard to non-trade concerns, including food security and the need to protect the envi- ronment; having regard to the agreement that special and differential treatment for developing countries is an integral element of the negotiations, and taking into account the possible negative effects of the imple- mentation of the reform programme on least-developed and net food-importing developing countries; Hereby agree as follows: 349 350 Agriculture and the WTO Part I rated by reference in Part IV of a Mem- ber's Schedule; and Article 1 Definition of Terms (ii) with respect to support provided during In this Agreement, unless the context otherwise any year of the implementation period requires: and thereafter, calculated in accordance with the provisions of Annex 4 of this (a) "Aggregate Measurement of Support" and Agreement and taking into account the "AMS" mean the annual level of support, constituent data and methodology used in expressed in monetary terms, provided for an the tables of supporting material incorpo- agricultural product in favour of the producers rated by reference in Part IV of the Mem- of the basic agricultural product or non-prod- ber's Schedule; uct-specific support provided in favour of (e) "export subsidies" refers to subsidies contin- agricultural producers in general, other than gent upon export performance, including the support provided under programmes that export subsidies listed in Article 9 of this qualify as exempt from reduction under Annex Agreement; 2 to this Agreement, which is: (f) "implementation period" means the six-year (i) with respect to support provided during period commencing in the year 1995, except the base period, specified in the relevant that, for the purposes of Article 13, it means tables of supporting material incorpo- the nine-year period commencing in 1995; rated by reference in Part IV of a Mem- (g) "market access concessions" includes all mar- ber's Schedule; and ket access commitments undertaken pursuant (ii) with respect to support provided during to this Agreement; any year of the implementation period (h) "Total Aggregate Measurement of Support" and thereafter, calculated in accordance and "Total AMS" mean the sum of all domestic with the provisions of Annex 3 of this support provided in favour of agricultural pro- Agreement and taking into account the ducers, calculated as the sum of all aggregate constituent data and methodology used in measurements of support for basic agricultural the tables of supporting material incorpo- products, all non-product-specific aggregate rated by reference in Part IV of the Mem- measurements of support and all equivalent ber's Schedule; measurements of support for agricultural (b) "basic agricultural product" in relation to products, and which is: domestic support commitments is defined as (i) with respect to support provided during the product as close as practicable to the point the base period (i.e. the "Base Total AMS") of first sale as specified in a Member's Schedule and the maximum support permitted to and in the related supporting material; be provided during any year of the imple- (c) "budgetary outlays" or "outlays" includes rev- mentation period or thereafter (i.e. the enue foregone; "Annual and Final Bound Commitment (d) "Equivalent Measurement of Support" means Levels"), as specified in Part IV of a Mem- the annual level of support, expressed in mon- ber's Schedule; and etary terms, provided to producers of a basic (ii) with respect to the level of support actu- agricultural product through the application ally provided during any year of the of one or more measures, the calculation of implementation period and thereafter (i.e. which in accordance with the AMS methodol- the "Current Total AMS"), calculated in ogy is impracticable, other than support pro- accordance with the provisions of this vided under programmes that qualify as Agreement, including Article 6, and with exempt from reduction under Annex 2 to this the constituent data and methodology Agreement, and which is: used in the tables of supporting material (i) with respect to support provided during incorporated by reference in Part IV of the the base period, specified in the relevant Member's Schedule; tables of supporting material incorpo- Appendix B 351 (i) "year" in paragraph (f) above and in relation Article 5 Special Safeguard Provisions to the specific commitments of a Member 1. Notwithstanding the provisions of paragraph refers to the calendar, financial or marketing 1(b) of Article II of GATT 1994, any Member year specified in the Schedule relating to that may take recourse to the provisions of para- Member. graphs 4 and 5 below in connection with the importation of an agricultural product, in Article 2 Product Coverage respect of which measures referred to in para- graph 2 of Article 4 of this Agreement have This Agreement applies to the products listed in been converted into an ordinary customs duty Annex 1 to this Agreement, hereinafter referred to and which is designated in its Schedule with as agricultural products. the symbol "SSG" as being the subject of a con- cession in respect of which the provisions of Part II this Article may be invoked, if: (a) the volume of imports of that product Article 3 Incorporation of Concessions and Commitments entering the customs territory of the Member granting the concession during 1. The domestic support and export subsidy any year exceeds a trigger level which commitments in Part IV of each Member's relates to the existing market access Schedule constitute commitments limiting opportunity as set out in paragraph 4; or, subsidization and are hereby made an integral but not concurrently: part of GATT 1994. (b) the price at which imports of that product 2. Subject to the provisions of Article 6, a Mem- may enter the customs territory of the Mem- ber shall not provide support in favour of ber granting the concession, as determined domestic producers in excess of the commit- on the basis of the c.i.f. import price of the ment levels specified in Section I of Part IV of shipment concerned expressed in terms of its Schedule. its domestic currency, falls below a trigger 3. Subject to the provisions of paragraphs 2(b) and price equal to the average 1986 to 1988 refer- 4 of Article 9, a Member shall not provide export ence price2 for the product concerned. subsidies listed in paragraph 1 of Article 9 in 2. Imports under current and minimum access respect of the agricultural products or groups of commitments established as part of a conces- products specified in Section II of Part IV of its sion referred to in paragraph 1 above shall be Schedule in excess of the budgetary outlay and counted for the purpose of determining the quantity commitment levels specified therein volume of imports required for invoking the and shall not provide such subsidies in respect of provisions of subparagraph 1(a) and para- any agricultural product not specified in that graph 4, but imports under such commitments Section of its Schedule. shall not be affected by any additional duty imposed under either subparagraph 1(a) and Part III paragraph 4 or subparagraph 1(b) and para- graph 5 below. Article 4 Market Access 3. Any supplies of the product in question which 1. Market access concessions contained in Sched- were en route on the basis of a contract settled ules relate to bindings and reductions of tariffs, before the additional duty is imposed under and to other market access commitments as subparagraph 1(a) and paragraph 4 shall be specified therein. exempted from any such additional duty, pro- 2. Members shall not maintain, resort to, or vided that they may be counted in the volume revert to any measures of the kind which have of imports of the product in question during been required to be converted into ordinary the following year for the purposes of trigger- customs duties1, except as otherwise provided ing the provisions of subparagraph 1(a) in for in Article 5 and Annex 5. that year. 352 Agriculture and the WTO 4. Any additional duty imposed under subpara- (b) if the difference between the import price graph 1(a) shall only be maintained until the and the trigger price (hereinafter referred end of the year in which it has been imposed, to as the"difference") is greater than 10 per and may only be levied at a level which shall cent but less than or equal to 40 per cent of not exceed one third of the level of the ordi- the trigger price, the additional duty shall nary customs duty in effect in the year in equal 30 per cent of the amount by which which the action is taken. The trigger level shall the difference exceeds 10 per cent; be set according to the following schedule (c) if the difference is greater than 40 per cent based on market access opportunities defined but less than or equal to 60 per cent of the as imports as a percentage of the correspon- trigger price, the additional duty shall ding domestic consumption3 during the three equal 50 per cent of the amount by which preceding years for which data are available: the difference exceeds 40 per cent, plus the (a) where such market access opportunities additional duty allowed under (b); for a product are less than or equal to 10 (d) if the difference is greater than 60 per cent per cent, the base trigger level shall equal but less than or equal to 75 per cent, the 125 per cent; additional duty shall equal 70 per cent of (b) where such market access opportunities the amount by which the difference for a product are greater than 10 per cent exceeds 60 per cent of the trigger price, but less than or equal to 30 per cent, the plus the additional duties allowed under base trigger level shall equal 110 per cent; (b) and (c); (c) where such market access opportunities for (e) if the difference is greater than 75 per cent a product are greater than 30 per cent, the of the trigger price, the additional duty base trigger level shall equal 105 per cent. shall equal 90 per cent of the amount by In all cases the additional duty may be which the difference exceeds 75 per cent, imposed in any year where the absolute vol- plus the additional duties allowed under ume of imports of the product concerned (b), (c) and (d). entering the customs territory of the Member 6. For perishable and seasonal products, the con- granting the concession exceeds the sum of (x) ditions set out above shall be applied in such a the base trigger level set out above multiplied manner as to take account of the specific char- by the average quantity of imports during the acteristics of such products. In particular, three preceding years for which data are avail- shorter time periods under subparagraph able and (y) the absolute volume change in 1(a) and paragraph 4 may be used in reference domestic consumption of the product con- to the corresponding periods in the base period cerned in the most recent year for which data and different reference prices for different peri- are available compared to the preceding year, ods may be used under subparagraph 1(b). provided that the trigger level shall not be less 7. The operation of the special safeguard shall be than 105 per cent of the average quantity of carried out in a transparent manner. Any Mem- imports in (x) above. ber taking action under subparagraph 5. The additional duty imposed under subpara- 1(a) above shall give notice in writing, includ- graph 1(b) shall be set according to the follow- ing relevant data, to the Committee on Agricul- ing schedule: ture as far in advance as may be practicable and (a) if the difference between the c.i.f. import in any event within 10 days of the implementa- price of the shipment expressed in terms tion of such action. In cases where changes in of the domestic currency (hereinafter consumption volumes must be allocated to referred to as the "import price") and the individual tariff lines subject to action under trigger price as defined under that sub- paragraph 4, relevant data shall include the paragraph is less than or equal to 10 per information and methods used to allocate these cent of the trigger price, no additional changes. A Member taking action under para- duty shall be imposed; graph 4 shall afford any interested Members the Appendix B 353 opportunity to consult with it in respect of the poor producers in developing country Members conditions of application of such action. Any shall be exempt from domestic support reduc- Member taking action under subparagraph tion commitments that would otherwise be 1(b) above shall give notice in writing, includ- applicable to such measures, as shall domestic ing relevant data, to the Committee on Agricul- support to producers in developing country ture within 10 days of the implementation of Members to encourage diversification from the first such action or, for perishable and sea- growing illicit narcotic crops. Domestic support sonal products, the first action in any period. meeting the criteria of this paragraph shall not Members undertake, as far as practicable, not be required to be included in a Member's calcu- to take recourse to the provisions of subpara- lation of its Current Total AMS. graph 1(b) where the volume of imports of the 3. A Member shall be considered to be in compli- products concerned are declining. In either case ance with its domestic support reduction com- a Member taking such action shall afford any mitments in any year in which its domestic interested Members the opportunity to consult support in favour of agricultural producers with it in respect of the conditions of applica- expressed in terms of Current Total AMS does tion of such action. not exceed the corresponding annual or final 8. Where measures are taken in conformity with bound commitment level specified in Part IV paragraphs 1 through 7 above, Members under- of the Member's Schedule. take not to have recourse,in respect of such meas- 4. (a) A Member shall not be required to include ures,to the provisions of paragraphs 1(a) and 3 of in the calculation of its Current Total AMS and Article XIX of GATT 1994 or paragraph 2 of shall not be required to reduce: Article 8 of the Agreement on Safeguards. (i) product-specific domestic support 9. The provisions of this Article shall remain in which would otherwise be required to force for the duration of the reform process as be included in a Member's calculation determined under Article 20. of its Current AMS where such sup- port does not exceed 5 per cent of Part IV that Member's total value of produc- tion of a basic agricultural product Article 6 Domestic Support Commitments during the relevant year; and 1. The domestic support reduction commit- (ii) non-product-specific domestic sup- ments of each Member contained in Part IV of port which would otherwise be its Schedule shall apply to all of its domestic required to be included in a Mem- support measures in favour of agricultural ber's calculation of its Current AMS producers with the exception of domestic where such support does not exceed 5 measures which are not subject to reduction in per cent of the value of that Member's terms of the criteria set out in this Article and total agricultural production. in Annex 2 to this Agreement. The commit- (b) For developing country Members, the de ments are expressed in terms of Total Aggre- minimis percentage under this paragraph gate Measurement of Support and "Annual shall be 10 per cent. and Final Bound Commitment Levels". 5. (a) Direct payments under production-limit- 2. In accordance with the Mid-Term Review Agree- ing programmes shall not be subject to the ment that government measures of assistance, commitment to reduce domestic support if: whether direct or indirect, to encourage agricul- (i) such payments are based on fixed area tural and rural development are an integral part and yields; or of the development programmes of developing (ii) such payments are made on 85 per countries, investment subsidies which are gener- cent or less of the base level of pro- ally available to agriculture in developing coun- duction; or try Members and agricultural input subsidies (iii) livestock payments are made on a generally available to low-income or resource- fixed number of head. 354 Agriculture and the WTO (b) The exemption from the reduction com- producers, or to a marketing board, con- mitment for direct payments meeting the tingent on export performance; above criteria shall be reflected by the (b) the sale or disposal for export by govern- exclusion of the value of those direct pay- ments or their agencies of non-commer- ments in a Member's calculation of its cial stocks of agricultural products at a Current Total AMS. price lower than the comparable price charged for the like product to buyers in the domestic market; Article 7 General Disciplines on Domestic Support (c) payments on the export of an agricultural 1. Each Member shall ensure that any domestic product that are financed by virtue of gov- support measures in favour of agricultural ernmental action, whether or not a charge producers which are not subject to reduction on the public account is involved, includ- commitments because they qualify under the ing payments that are financed from the criteria set out in Annex 2 to this Agreement proceeds of a levy imposed on the agricul- are maintained in conformity therewith. tural product concerned or on an agricul- 2. (a) Any domestic support measure in tural product from which the exported favour of agricultural producers, including product is derived; any modification to such measure, and any (d) the provision of subsidies to reduce the measure that is subsequently introduced that costs of marketing exports of agricultural cannot be shown to satisfy the criteria in products (other than widely available Annex 2 to this Agreement or to be exempt export promotion and advisory services) from reduction by reason of any other provi- including handling, upgrading and other sion of this Agreement shall be included in processing costs, and the costs of interna- the Member's calculation of its Current tional transport and freight; Total AMS. (e) internal transport and freight charges on (b) Where no Total AMS commitment exists export shipments, provided or mandated in Part IV of a Member's Schedule, the by governments, on terms more favourable Member shall not provide support to agri- than for domestic shipments; cultural producers in excess of the rele- (f) subsidies on agricultural products contingent vant de minimis level set out in paragraph on their incorporation in exported products. 4 of Article 6. 2. (a) Except as provided in subparagraph (b), the export subsidy commitment levels for each year of the implementation period, as specified Part V in a Member's Schedule, represent with respect Article 8 Export Competition Commitments to the export subsidies listed in paragraph 1 of this Article: Each Member undertakes not to provide export (i) in the case of budgetary outlay reduc- subsidies otherwise than in conformity with this tion commitments, the maximum level Agreement and with the commitments as specified of expenditure for such subsidies that in that Member's Schedule. may be allocated or incurred in that year in respect of the agricultural product, or group of products, concerned; and Article 9 Export Subsidy Commitments (ii) in the case of export quantity reduc- 1. The following export subsidies are subject to tion commitments, the maximum reduction commitments under this Agreement: quantity of an agricultural product, (a) the provision by governments or their or group of products, in respect of agencies of direct subsidies, including pay- which such export subsidies may be ments-in-kind, to a firm, to an industry, to granted in that year. producers of an agricultural product, to a (b) In any of the second through fifth years of cooperative or other association of such the implementation period, a Member Appendix B 355 may provide export subsidies listed in 4. During the implementation period, develop- paragraph 1 above in a given year in excess ing country Members shall not be required to of the corresponding annual commitment undertake commitments in respect of the levels in respect of the products or groups export subsidies listed in subparagraphs of products specified in Part IV of the (d) and (e) of paragraph 1 above, provided Member's Schedule, provided that: that these are not applied in a manner that (i) the cumulative amounts of budgetary would circumvent reduction commitments. outlays for such subsidies, from the beginning of the implementation Article 10 Prevention of Circumvention of Export period through the year in question, Subsidy Commitments does not exceed the cumulative amounts that would have resulted 1. Export subsidies not listed in paragraph 1 of from full compliance with the relevant Article 9 shall not be applied in a manner annual outlay commitment levels which results in, or which threatens to lead to, specified in the Member's Schedule by circumvention of export subsidy commit- more than 3 per cent of the base ments; nor shall non-commercial transactions period level of such budgetary outlays; be used to circumvent such commitments. (ii) the cumulative quantities exported 2. Members undertake to work toward the devel- with the benefit of such export subsi- opment of internationally agreed disciplines to dies, from the beginning of the imple- govern the provision of export credits, export mentation period through the year in credit guarantees or insurance programmes question, does not exceed the cumu- and, after agreement on such disciplines, to lative quantities that would have provide export credits, export credit guaran- resulted from full compliance with tees or insurance programmes only in con- the relevant annual quantity commit- formity therewith. ment levels specified in the Member's 3. Any Member which claims that any quantity Schedule by more than 1.75 per cent exported in excess of a reduction commitment of the base period quantities; level is not subsidized must establish that no (iii) the total cumulative amounts of budg- export subsidy, whether listed in Article 9 or etary outlays for such export subsidies not, has been granted in respect of the quantity and the quantities benefiting from such of exports in question. export subsidies over the entire imple- 4. Members donors of international food aid mentation period are no greater than shall ensure: the totals that would have resulted (a) that the provision of international food from full compliance with the relevant aid is not tied directly or indirectly to annual commitment levels specified in commercial exports of agricultural prod- the Member's Schedule; and ucts to recipient countries; (iv) the Member's budgetary outlays for (b) that international food aid transactions, export subsidies and the quantities including bilateral food aid which is mon- benefiting from such subsidies, at the etized, shall be carried out in accordance conclusion of the implementation with the FAO "Principles of Surplus Dis- period, are no greater than 64 per posal and Consultative Obligations", cent and 79 per cent of the 1986-1990 including, where appropriate, the system base period levels, respectively. For of Usual Marketing Requirements developing country Members these (UMRs); and percentages shall be 76 and 86 per (c) that such aid shall be provided to the cent, respectively. extent possible in fully grant form or on 3. Commitments relating to limitations on the terms no less concessional than those pro- extension of the scope of export subsidization vided for in Article IV of the Food Aid are as specified in Schedules. Convention 1986. 356 Agriculture and the WTO Article 11 Incorporated Products (a) domestic support measures that conform fully to the provisions of Annex 2 to this In no case may the per-unit subsidy paid on an Agreement shall be: incorporated agricultural primary product exceed (i) non-actionable subsidies for pur- the per-unit export subsidy that would be payable poses of countervailing duties4; on exports of the primary product as such. (ii) exempt from actions based on Article XVI of GATT 1994 and Part III of the Part VI Subsidies Agreement; and Article 12 Disciplines on Export Prohibitions (iii) exempt from actions based on non- and Restrictions violation nullification or impairment of the benefits of tariff concessions 1. Where any Member institutes any new export accruing to another Member under prohibition or restriction on foodstuffs in Article II of GATT 1994, in the sense accordance with paragraph 2(a) of Article XI of paragraph 1(b) of Article XXIII of of GATT 1994, the Member shall observe the GATT 1994; following provisions: (b) domestic support measures that conform (a) the Member instituting the export prohi- fully to the provisions of Article 6 of this bition or restriction shall give due consid- Agreement including direct payments that eration to the effects of such prohibition conform to the requirements of paragraph or restriction on importing Members' 5 thereof, as reflected in each Member's food security; Schedule, as well as domestic support (b) before any Member institutes an export within de minimis levels and in conform- prohibition or restriction, it shall give ity with paragraph 2 of Article 6, shall be: notice in writing, as far in advance as (i) exempt from the imposition of coun- practicable, to the Committee on Agricul- tervailing duties unless a determina- ture comprising such information as the tion of injury or threat thereof is nature and the duration of such measure, made in accordance with Article VI of and shall consult, upon request, with any GATT 1994 and Part V of the Subsi- other Member having a substantial inter- dies Agreement, and due restraint est as an importer with respect to any mat- shall be shown in initiating any coun- ter related to the measure in question. The tervailing duty investigations; Member instituting such export prohibi- (ii) exempt from actions based on para- tion or restriction shall provide, upon graph 1 of Article XVI of GATT 1994 request, such a Member with necessary or Articles 5 and 6 of the Subsidies information. Agreement, provided that such meas- 2. The provisions of this Article shall not apply to ures do not grant support to a specific any developing country Member, unless the commodity in excess of that decided measure is taken by a developing country during the 1992 marketing year; and Member which is a net-food exporter of the (iii) exempt from actions based on non- specific foodstuff concerned. violation nullification or impairment of the benefits of tariff concessions Part VII accruing to another Member under Article II of GATT 1994, in the sense Article 13 Due Restraint of paragraph 1(b) of Article XXIII of During the implementation period, notwithstand- GATT 1994, provided that such meas- ing the provisions of GATT 1994 and the Agreement ures do not grant support to a specific on Subsidies and Countervailing Measures (referred commodity in excess of that decided to in this Article as the "Subsidies Agreement"): during the 1992 marketing year; Appendix B 357 (c) export subsidies that conform fully to the gramme on Least-Developed and Net Food- provisions of Part V of this Agreement, as Importing Developing Countries. reflected in each Member's Schedule, shall be: 2. The Committee on Agriculture shall monitor, (i) subject to countervailing duties only as appropriate, the follow-up to this Decision. upon a determination of injury or threat thereof based on volume, effect Part XI on prices, or consequent impact in accordance with Article VI of GATT Article 17 Committee on Agriculture 1994 and Part V of the Subsidies A Committee on Agriculture is hereby established. Agreement, and due restraint shall be shown in initiating any countervail- Article 18 Review of the Implementation ing duty investigations; and of Commitments (ii) exempt from actions based on Article XVI of GATT 1994 or Articles 3, 5 1. Progress in the implementation of commit- and 6 of the Subsidies Agreement. ments negotiated under the Uruguay Round reform programme shall be reviewed by the Committee on Agriculture. Part VIII 2. The review process shall be undertaken on the Article 14 Sanitary and Phytosanitary Measures basis of notifications submitted by Members in relation to such matters and at such intervals Members agree to give effect to the Agreement on the as shall be determined, as well as on the basis of Application of Sanitary and Phytosanitary Measures. such documentation as the Secretariat may be requested to prepare in order to facilitate the Part IX review process. Article 15 Special and Differential Treatment 3. In addition to the notifications to be submitted under paragraph 2, any new domestic support 1. In keeping with the recognition that differen- measure, or modification of an existing meas- tial and more favourable treatment for devel- ure, for which exemption from reduction is oping country Members is an integral part of claimed shall be notified promptly. This notifi- the negotiation, special and differential treat- cation shall contain details of the new or mod- ment in respect of commitments shall be pro- ified measure and its conformity with the vided as set out in the relevant provisions of agreed criteria as set out either in Article 6 or this Agreement and embodied in the Schedules in Annex 2. of concessions and commitments. 4. In the review process Members shall give due 2. Developing country Members shall have the consideration to the influence of excessive flexibility to implement reduction commit- rates of inflation on the ability of any Member ments over a period of up to 10 years. Least- to abide by its domestic support commit- developed country Members shall not be ments. required to undertake reduction commitments. 5. Members agree to consult annually in the Committee on Agriculture with respect to Part X their participation in the normal growth of world trade in agricultural products within the Article 16 Least-Developed and Net Food- framework of the commitments on export Importing Developing Countries subsidies under this Agreement. 1. Developed country Members shall take such 6. The review process shall provide an opportunity action as is provided for within the framework for Members to raise any matter relevant to the of the Decision on Measures Concerning the implementation of commitments under the Possible Negative Effects of the Reform Pro- reform programme as set out in this Agreement. 358 Agriculture and the WTO 7. Any Member may bring to the attention of the (a) the experience to that date from imple- Committee on Agriculture any measure which menting the reduction commitments; it considers ought to have been notified by (b) the effects of the reduction commitments another Member. on world trade in agriculture; (c) non-trade concerns, special and differen- tial treatment to developing country Article 19 Consultation and Dispute Settlement Members, and the objective to establish a The provisions of Articles XXII and XXIII of GATT fair and market-oriented agricultural 1994, as elaborated and applied by the Dispute Set- trading system, and the other objectives tlement Understanding, shall apply to consulta- and concerns mentioned in the preamble tions and the settlement of disputes under this to this Agreement; and Agreement. (d) what further commitments are necessary to achieve the above mentioned long-term objectives. Part XII Article 20 Continuation of the Reform Process Part XIII Recognizing that the long-term objective of sub- Article 21 Final Provisions stantial progressive reductions in support and pro- tection resulting in fundamental reform is an 1. The provisions of GATT 1994 and of other ongoing process, Members agree that negotiations Multilateral Trade Agreements in Annex 1A to for continuing the process will be initiated one year the WTO Agreement shall apply subject to the before the end of the implementation period, tak- provisions of this Agreement. ing into account: 2. The Annexes to this Agreement are hereby made an integral part of this Agreement. ANNEX 1 PRODUCT COVERAGE 1. This Agreement shall cover the following products: (i) HS Chapters 1 to 24 less fish and fish products, plus* (ii) HS Code 2905.43 (mannitol) HS Code 2905.44 (sorbitol) HS Heading 33.01 (essential oils) HS Headings 35.01 to 35.05 (albuminoidal substances, modified starches, glues) HS Code 3809.10 (finishing agents) HS Code 3823.60 (sorbitol n.e.p.) HS Headings 41.01 to 41.03 (hides and skins) HS Heading 43.01 (raw furskins) HS Headings 50.01 to 50.03 (raw silk and silk waste) HS Headings 51.01 to 51.03 (wool and animal hair) HS Headings 52.01 to 52.03 (raw cotton, waste and cotton carded or combed) HS Heading 53.01 (raw flax) HS Heading 53.02 (raw hemp) *The product descriptions in round brackets are not necessarily exhaustive. 2. The foregoing shall not limit the product coverage of the Agreement on the Application of Sanitary and Phytosanitary Measures. Appendix B 359 ANNEX 2 DOMESTIC SUPPORT: (d) extension and advisory services, including THE BASIS FOR EXEMPTION FROM the provision of means to facilitate the THE REDUCTION COMMITMENTS transfer of information and the results of research to producers and consumers; 1. Domestic support measures for which exemp- (e) inspection services, including general tion from the reduction commitments is inspection services and the inspection of claimed shall meet the fundamental require- particular products for health, safety, ment that they have no, or at most minimal, grading or standardization purposes; trade-distorting effects or effects on produc- (f) marketing and promotion services, tion. Accordingly, all measures for which including market information, advice and exemption is claimed shall conform to the fol- promotion relating to particular products lowing basic criteria: but excluding expenditure for unspecified (a) the support in question shall be provided purposes that could be used by sellers to through a publicly-funded government reduce their selling price or confer a direct programme (including government rev- economic benefit to purchasers; and enue foregone) not involving transfers (g) infrastructural services, including: electricity from consumers; and, reticulation,roads and other means of trans- (b) the support in question shall not have the port, market and port facilities, water supply effect of providing price support to pro- facilities, dams and drainage schemes, and ducers; plus policy-specific criteria and infrastructural works associated with envi- conditions as set out below. ronmental programmes. In all cases the expenditure shall be directed to the provi- sion or construction of capital works only, and shall exclude the subsidized provision of Government Service Programmes on-farm facilities other than for the reticula- 2. General services tion of generally available public utilities. It Policies in this category involve expenditures shall not include subsidies to inputs or oper- (or revenue foregone) in relation to pro- ating costs, or preferential user charges. grammes which provide services or benefits to 3. Public stockholding for food security purposes5 agriculture or the rural community. They shall Expenditures (or revenue foregone) in relation not involve direct payments to producers or to the accumulation and holding of stocks of processors. Such programmes, which include products which form an integral part of a food but are not restricted to the following list, shall security programme identified in national leg- meet the general criteria in paragraph 1 above islation. This may include government aid to and policy-specific conditions where set out private storage of products as part of such a below: programme. (a) research, including general research, The volume and accumulation of such research in connection with environmen- stocks shall correspond to predetermined tal programmes, and research pro- targets related solely to food security. The grammes relating to particular products; process of stock accumulation and disposal (b) pest and disease control, including general shall be financially transparent. Food pur- and product-specific pest and disease con- chases by the government shall be made at trol measures, such as early-warning sys- current market prices and sales from food tems, quarantine and eradication; security stocks shall be made at no less than (c) training services, including both general the current domestic market price for the and specialist training facilities; product and quality in question. 360 Agriculture and the WTO 4. Domestic food aid6 (e) No production shall be required in order Expenditures (or revenue foregone) in relation to receive such payments. to the provision of domestic food aid to sec- 7. Government financial participation in income tions of the population in need. insurance and income safety-net programmes Eligibility to receive the food aid shall be (a) Eligibility for such payments shall be subject to clearly-defined criteria related determined by an income loss, taking into to nutritional objectives. Such aid shall be account only income derived from agri- in the form of direct provision of food to culture, which exceeds 30 per cent of aver- those concerned or the provision of age gross income or the equivalent in net means to allow eligible recipients to buy income terms (excluding any payments food either at market or at subsidized from the same or similar schemes) in the prices. Food purchases by the government preceding three-year period or a three- shall be made at current market prices and year average based on the preceding five- the financing and administration of the year period, excluding the highest and the aid shall be transparent. lowest entry. Any producer meeting this 5. Direct payments to producers condition shall be eligible to receive the Support provided through direct payments (or payments. revenue foregone, including payments in (b) The amount of such payments shall com- kind) to producers for which exemption from pensate for less than 70 per cent of the reduction commitments is claimed shall meet producer's income loss in the year the pro- the basic criteria set out in paragraph 1 above, ducer becomes eligible to receive this plus specific criteria applying to individual types assistance. of direct payment as set out in paragraphs 6 (c) The amount of any such payments shall through 13 below. Where exemption from relate solely to income; it shall not relate to reduction is claimed for any existing or new type the type or volume of production (includ- of direct payment other than those specified in ing livestock units) undertaken by the paragraphs 6 through 13, it shall conform to cri- producer; or to the prices, domestic or teria (b) through (e) in paragraph 6, in addition international, applying to such produc- to the general criteria set out in paragraph 1. tion; or to the factors of production 6. Decoupled income support employed. (a) Eligibility for such payments shall be (d) Where a producer receives in the same determined by clearly-defined criteria year payments under this paragraph and such as income, status as a producer or under paragraph 8 (relief from natural landowner, factor use or production level disasters), the total of such payments shall in a defined and fixed base period. be less than 100 per cent of the producer's (b) The amount of such payments in any total loss. given year shall not be related to, or based 8. Payments (made either directly or by way of gov- on, the type or volume of production ernment financial participation in crop insur- (including livestock units) undertaken by ance schemes) for relief from natural disasters the producer in any year after the base (a) Eligibility for such payments shall arise period. only following a formal recognition by (c) The amount of such payments in any government authorities that a natural or given year shall not be related to, or based like disaster (including disease outbreaks, on, the prices, domestic or international, pest infestations, nuclear accidents, and applying to any production undertaken in war on the territory of the Member con- any year after the base period. cerned) has occurred or is occurring; and (d) The amount of such payments in any shall be determined by a production loss given year shall not be related to, or based which exceeds 30 per cent of the average on, the factors of production employed in of production in the preceding three-year any year after the base period. period or a three-year average based on Appendix B 361 the preceding five-year period, excluding resources which involves the production the highest and the lowest entry. of marketable agricultural products. (b) Payments made following a disaster shall (d) Payments shall not be related to either the be applied only in respect of losses of type or quantity of production or to the income, livestock (including payments in prices, domestic or international, applying connection with the veterinary treatment to production undertaken using the land or of animals), land or other production fac- other resources remaining in production. tors due to the natural disaster in question. 11. Structural adjustment assistance provided (c) Payments shall compensate for not more through investment aids than the total cost of replacing such losses (a) Eligibility for such payments shall be and shall not require or specify the type or determined by reference to clearly-defined quantity of future production. criteria in government programmes (d) Payments made during a disaster shall not designed to assist the financial or physical exceed the level required to prevent or restructuring of a producer's operations in alleviate further loss as defined in crite- response to objectively demonstrated rion (b) above. structural disadvantages. Eligibility for (e) Where a producer receives in the same such programmes may also be based on a year payments under this paragraph and clearly-defined government programme under paragraph 7 (income insurance and for the reprivatization of agricultural land. income safety-net programmes), the total (b) The amount of such payments in any of such payments shall be less than 100 given year shall not be related to, or based per cent of the producer's total loss. on, the type or volume of production 9. Structural adjustment assistance provided (including livestock units) undertaken by through producer retirement programmes the producer in any year after the base (a) Eligibility for such payments shall be deter- period other than as provided for under mined by reference to clearly defined crite- criterion (e) below. ria in programmes designed to facilitate the (c) The amount of such payments in any retirement of persons engaged in mar- given year shall not be related to, or based ketable agricultural production, or their on, the prices, domestic or international, movement to non-agricultural activities. applying to any production undertaken in (b) Payments shall be conditional upon the any year after the base period. total and permanent retirement of the (d) The payments shall be given only for the recipients from marketable agricultural period of time necessary for the realiza- production. tion of the investment in respect of which 10. Structural adjustment assistance provided they are provided. through resource retirement programmes (e) The payments shall not mandate or in any (a) Eligibility for such payments shall be way designate the agricultural products to determined by reference to clearly defined be produced by the recipients except to criteria in programmes designed to require them not to produce a particular remove land or other resources, including product. livestock, from marketable agricultural (f) The payments shall be limited to the production. amount required to compensate for the (b) Payments shall be conditional upon the structural disadvantage. retirement of land from marketable agri- 12. Payments under environmental programmes cultural production for a minimum of (a) Eligibility for such payments shall be three years, and in the case of livestock determined as part of a clearly-defined on its slaughter or definitive permanent government environmental or conserva- disposal. tion programme and be dependent on the (c) Payments shall not require or specify any fulfilment of specific conditions under alternative use for such land or other the government programme, including 362 Agriculture and the WTO conditions related to production methods price support, non-exempt direct payments, or or inputs. any other subsidy not exempted from the (b) The amount of payment shall be limited reduction commitment ("other non-exempt to the extra costs or loss of income policies"). Support which is non-product spe- involved in complying with the govern- cific shall be totalled into one non-product- ment programme. specific AMS in total monetary terms. 13. Payments under regional assistance pro- 2. Subsidies under paragraph 1 shall include both grammes budgetary outlays and revenue foregone by (a) Eligibility for such payments shall be lim- governments or their agents. ited to producers in disadvantaged 3. Support at both the national and sub-national regions. Each such region must be a level shall be included. clearly designated contiguous geographi- 4. Specific agricultural levies or fees paid by pro- cal area with a definable economic and ducers shall be deducted from the AMS. administrative identity, considered as dis- 5. The AMS calculated as outlined below for the advantaged on the basis of neutral and base period shall constitute the base level for objective criteria clearly spelt out in law or the implementation of the reduction commit- regulation and indicating that the region's ment on domestic support. difficulties arise out of more than tempo- 6. For each basic agricultural product, a specific rary circumstances. AMS shall be established, expressed in total (b) The amount of such payments in any given monetary value terms. year shall not be related to, or based on, the 7. The AMS shall be calculated as close as practi- type or volume of production (including cable to the point of first sale of the basic agri- livestock units) undertaken by the pro- cultural product concerned. Measures directed ducer in any year after the base period at agricultural processors shall be included to other than to reduce that production. the extent that such measures benefit the pro- (c) The amount of such payments in any ducers of the basic agricultural products. given year shall not be related to, or based 8. Market price support: market price support on, the prices, domestic or international, shall be calculated using the gap between a applying to any production undertaken in fixed external reference price and the applied any year after the base period. administered price multiplied by the quantity (d) Payments shall be available only to produc- of production eligible to receive the applied ers in eligible regions, but generally avail- administered price. Budgetary payments made able to all producers within such regions. to maintain this gap, such as buying-in or stor- (e) Where related to production factors, age costs, shall not be included in the AMS. payments shall be made at a degressive 9. The fixed external reference price shall be rate above a threshold level of the factor based on the years 1986 to 1988 and shall gen- concerned. erally be the average f.o.b. unit value for the (f) The payments shall be limited to the extra basic agricultural product concerned in a net costs or loss of income involved in under- exporting country and the average c.i.f. unit taking agricultural production in the pre- value for the basic agricultural product con- scribed area. cerned in a net importing country in the base period. The fixed reference price may be adjusted for quality differences as necessary. ANNEX 3 DOMESTIC SUPPORT: 10. Non-exempt direct payments: non-exempt CALCULATION OF AGGREGATE direct payments which are dependent on a MEASUREMENT OF SUPPORT price gap shall be calculated either using the 1. Subject to the provisions of Article 6, an Aggre- gap between the fixed reference price and the gate Measurement of Support (AMS) shall be applied administered price multiplied by the calculated on a product-specific basis for each quantity of production eligible to receive the basic agricultural product receiving market administered price, or using budgetary outlays. Appendix B 363 11. The fixed reference price shall be based on the price support shall be made using the applied years 1986 to 1988 and shall generally be the administered price and the quantity of pro- actual price used for determining payment duction eligible to receive that price or, where rates. this is not practicable, on budgetary outlays 12. Non-exempt direct payments which are based used to maintain the producer price. on factors other than price shall be measured 3. Where basic agricultural products falling using budgetary outlays. under paragraph 1 are the subject of non- 13. Other non-exempt measures, including input exempt direct payments or any other product- subsidies and other measures such as market- specific subsidy not exempted from the ing-cost reduction measures: the value of such reduction commitment, the basis for equiva- measures shall be measured using government lent measurements of support concerning budgetary outlays or, where the use of budget- these measures shall be calculations as for the ary outlays does not reflect the full extent of corresponding AMS components (specified in the subsidy concerned, the basis for calculating paragraphs 10 through 13 of Annex 3). the subsidy shall be the gap between the price 4. Equivalent measurements of support shall be of the subsidized good or service and a repre- calculated on the amount of subsidy as close as sentative market price for a similar good or practicable to the point of first sale of the basic service multiplied by the quantity of the good agricultural product concerned. Measures or service. directed at agricultural processors shall be included to the extent that such measures ben- efit the producers of the basic agricultural ANNEX 4 DOMESTIC SUPPORT: products. Specific agricultural levies or fees CALCULATION OF EQUIVALENT paid by producers shall reduce the equivalent MEASUREMENT OF SUPPORT measurements of support by a corresponding 1. Subject to the provisions of Article 6, equiva- amount. lent measurements of support shall be calcu- lated in respect of all basic agricultural products where market price support as ANNEX 5 SPECIAL TREATMENT defined in Annex 3 exists but for which calcu- WITH RESPECT TO PARAGRAPH 2 lation of this component of the AMS is not OF ARTICLE 4 practicable. For such products the base level Section A for implementation of the domestic support reduction commitments shall consist of a mar- 1. The provisions of paragraph 2 of Article 4 shall ket price support component expressed in not apply with effect from the entry into force terms of equivalent measurements of support of the WTO Agreement to any primary agri- under paragraph 2 below, as well as any non- cultural product and its worked and/or pre- exempt direct payments and other non- pared products ("designated products") in exempt support, which shall be evaluated as respect of which the following conditions are provided for under paragraph 3 below. Sup- complied with (hereinafter referred to as "spe- port at both national and sub-national level cial treatment"): shall be included. (a) imports of the designated products com- 2. The equivalent measurements of support pro- prised less than 3 per cent of correspon- vided for in paragraph 1 shall be calculated on ding domestic consumption in the base a product-specific basis for all basic agricul- period 1986-1988 ("the base period"); tural products as close as practicable to the (b) no export subsidies have been provided point of first sale receiving market price sup- since the beginning of the base period for port and for which the calculation of the mar- the designated products; ket price support component of the AMS is (c) effective production-restricting measures not practicable. For those basic agricultural are applied to the primary agricultural products, equivalent measurements of market product; 364 Agriculture and the WTO (d) such products are designated with the 5. Where the special treatment is not to be con- symbol "ST-Annex 5" in Section I-B of tinued at the end of the implementation Part I of a Member's Schedule annexed to period, the Member concerned shall imple- the Marrakesh Protocol, as being subject ment the provisions of paragraph 6. In such a to special treatment reflecting factors of case, after the end of the implementation non-trade concerns, such as food security period the minimum access opportunities for and environmental protection; and the designated products shall be maintained at (e) minimum access opportunities in respect the level of 8 per cent of corresponding domes- of the designated products correspond, as tic consumption in the base period in the specified in Section I-B of Part I of the Schedule of the Member concerned. Schedule of the Member concerned, to 4 6. Border measures other than ordinary customs per cent of base period domestic con- duties maintained in respect of the designated sumption of the designated products from products shall become subject to the provi- the beginning of the first year of the sions of paragraph 2 of Article 4 with effect implementation period and, thereafter, from the beginning of the year in which the are increased by 0.8 per cent of correspon- special treatment ceases to apply. Such prod- ding domestic consumption in the base ucts shall be subject to ordinary customs period per year for the remainder of the duties, which shall be bound in the Schedule of implementation period. the Member concerned and applied, from the 2. At the beginning of any year of the implemen- beginning of the year in which special treat- tation period a Member may cease to apply ment ceases and thereafter, at such rates as special treatment in respect of the designated would have been applicable had a reduction of products by complying with the provisions of at least 15 per cent been implemented over the paragraph 6. In such a case, the Member con- implementation period in equal annual instal- cerned shall maintain the minimum access ments. These duties shall be established on the opportunities already in effect at such time and basis of tariff equivalents to be calculated in increase the minimum access opportunities by accordance with the guidelines prescribed in 0.4 per cent of corresponding domestic con- the attachment hereto. sumption in the base period per year for the remainder of the implementation period. Section B Thereafter, the level of minimum access opportunities resulting from this formula in 7. The provisions of paragraph 2 of Article 4 shall the final year of the implementation period also not apply with effect from the entry into shall be maintained in the Schedule of the force of the WTO Agreement to a primary Member concerned. agricultural product that is the predominant 3. Any negotiation on the question of whether staple in the traditional diet of a developing there can be a continuation of the special treat- country Member and in respect of which the ment as set out in paragraph 1 after the end of following conditions, in addition to those the implementation period shall be completed specified in paragraph 1(a) through 1(d), as within the time-frame of the implementation they apply to the products concerned, are com- period itself as a part of the negotiations set plied with: out in Article 20 of this Agreement, taking into (a) minimum access opportunities in respect account the factors of non-trade concerns. of the products concerned, as specified in 4. If it is agreed as a result of the negotiation Section I-B of Part I of the Schedule of referred to in paragraph 3 that a Member may the developing country Member con- continue to apply the special treatment, such cerned, correspond to 1 per cent of base Member shall confer additional and acceptable period domestic consumption of the concessions as determined in that negotiation. products concerned from the beginning Appendix B 365 of the first year of the implementation accorded to developing country Members period and are increased in equal annual under this Agreement. instalments to 2 per cent of correspon- ding domestic consumption in the base period at the beginning of the fifth year of Attachment to Annex 5 the implementation period. From the beginning of the sixth year of the imple- Guidelines for the Calculation of Tariff Equivalents mentation period, minimum access for the Specific Purpose Specified in Paragraphs 6 opportunities in respect of the products and 10 of this Annex concerned correspond to 2 per cent of corresponding domestic consumption in 1. The calculation of the tariff equivalents, the base period and are increased in equal whether expressed as ad valorem or specific annual instalments to 4 per cent of corre- rates, shall be made using the actual difference sponding domestic consumption in the between internal and external prices in a trans- base period until the beginning of the parent manner. Data used shall be for the years 10th year. Thereafter, the level of mini- 1986 to 1988. Tariff equivalents: mum access opportunities resulting from (a) shall primarily be established at the four- this formula in the 10th year shall be digit level of the HS; maintained in the Schedule of the devel- (b) shall be established at the six-digit or a oping country Member concerned; more detailed level of the HS wherever (b) appropriate market access opportunities appropriate; have been provided for in other products (c) shall generally be established for worked under this Agreement. and/or prepared products by multiplying 8. Any negotiation on the question of whether the specific tariff equivalent(s) for the pri- there can be a continuation of the special treat- mary agricultural product(s) by the pro- ment as set out in paragraph 7 after the end of portion(s) in value terms or in physical the 10th year following the beginning of the terms as appropriate of the primary agri- implementation period shall be initiated and cultural product(s) in the worked and/or completed within the time-frame of the 10th prepared products, and take account, year itself following the beginning of the where necessary, of any additional ele- implementation period. ments currently providing protection to 9. If it is agreed as a result of the negotiation industry. referred to in paragraph 8 that a Member may 2. External prices shall be, in general, actual aver- continue to apply the special treatment, such age c.i.f. unit values for the importing country. Member shall confer additional and acceptable Where average c.i.f. unit values are not avail- concessions as determined in that negotiation. able or appropriate, external prices shall be 10. In the event that special treatment under para- either: graph 7 is not to be continued beyond the 10th (a) appropriate average c.i.f. unit values of a year following the beginning of the implemen- near country; or tation period, the products concerned shall be (b) estimated from average f.o.b. unit values subject to ordinary customs duties, established of (an) appropriate major exporter(s) on the basis of a tariff equivalent to be calcu- adjusted by adding an estimate of insur- lated in accordance with the guidelines pre- ance, freight and other relevant costs to scribed in the attachment hereto, which shall the importing country. be bound in the Schedule of the Member con- 3. The external prices shall generally be converted cerned. In other respects, the provisions of to domestic currencies using the annual aver- paragraph 6 shall apply as modified by the rel- age market exchange rate for the same period evant special and differential treatment as the price data. 366 Agriculture and the WTO 4. The internal price shall generally be a repre- not the measures are maintained under country-specific deroga- sentative wholesale price ruling in the domes- tions from the provisions of GATT 1947, but not measures maintained under balance-of-payments provisions or under tic market or an estimate of that price where other general, non-agriculture-specific provisions of GATT adequate data is not available. 1994 or of the other Multilateral Trade Agreements in Annex 1A 5. The initial tariff equivalents may be adjusted, to the WTO Agreement. 2. The reference price used to invoke the provisions of this where necessary, to take account of differences subparagraph shall, in general, be the average c.i.f. unit value of in quality or variety using an appropriate the product concerned, or otherwise shall be an appropriate coefficient. price in terms of the quality of the product and its stage of pro- cessing. It shall, following its initial use, be publicly specified and 6. Where a tariff equivalent resulting from these available to the extent necessary to allow other Members to guidelines is negative or lower than the current assess the additional duty that may be levied. bound rate, the initial tariff equivalent may be 3. Where domestic consumption is not taken into account, established at the current bound rate or on the the base trigger level under subparagraph 4(a) shall apply. 4. "Countervailing duties" where referred to in this Article basis of national offers for that product. are those covered by Article VI of GATT 1994 and Part V of the 7. Where an adjustment is made to the level of a Agreement on Subsidies and Countervailing Measures. tariff equivalent which would have resulted 5. For the purposes of paragraph 3 of this Annex, govern- mental stockholding programmes for food security purposes in from the above guidelines, the Member con- developing countries whose operation is transparent and con- cerned shall afford, on request, full opportuni- ducted in accordance with officially published objective criteria ties for consultation with a view to negotiating or guidelines shall be considered to be in conformity with the provisions of this paragraph, including programmes under appropriate solutions. which stocks of foodstuffs for food security purposes are acquired and released at administered prices, provided that the Notes difference between the acquisition price and the external refer- ence price is accounted for in the AMS. 1. These measures include quantitative import restrictions, 5. & 6. For the purposes of paragraphs 3 and 4 of this Annex, variable import levies, minimum import prices, discretionary the provision of foodstuffs at subsidized prices with the objective import licensing, non-tariff measures maintained through state- of meeting food requirements of urban and rural poor in develop- trading enterprises, voluntary export restraints, and similar bor- ing countries on a regular basis at reasonable prices shall be consid- der measures other than ordinary customs duties, whether or ered to be in conformity with the provisions of this paragraph. INDEX Tables, figures, boxes, and notes are indicated by t, f, agricultural trade vs. industrial goods, 6­16 b, and n respectively. amber box, 119 commitments to reduce domestic supports and, access. See market access 122 acreage set-asides, 130­32, 145n7 countries in support of, 168 administered prices, 121, 125 distortion caused by, 181 Africa due restraint and, 34 See also specific countries recommendation for new amber box, 126, U.S. preferential treatment of, 274, 275 144­45 African Growth and Opportunity Act (AGOA), scope of policies of, 121, 130, 168 293, 300­301, 302t15.5, 310 URAA provisions, 31 AFTA. See ASEAN Free Trade Area U.S. conversion of subsidies into green box, 122 Aggregate Measurement of Support (AMS), 31­33, AMS. See Aggregate Measurement of Support 121­30 Andean Trade Preference Act (ATPA), 299­300, administered prices in calculation of, 121, 125 311b15.7 background of, 119­20 Andean Trade Promotion and Drug Eradication commitment vs. actual, 122, 127t6.6 Act of 2002 (ATPDEA), 299­300, 311b15.7 components of, 119­20 animals de minimis support. See de minimis support bioengineered, 236 evolution of AMS and PSE, 125, 129t6.7 transgenic, 236 reforms needed, 143, 144 welfare of, 168, 170 reprint of Annex 3 on, 362­63 antidumping measures, 209­10 special and differential treatment and, 280­81 developing countries and, 1, 289n3 trends in AMS vs. PSE, 125, 130f6.2, 145nn5­6 duties on subsidized exports, 34­35 use by member, 120, 122, 126t6.5 Latin American and Caribbean investigations, aggregation to maximize value of TRQs, 102 212n15 AGOA. See African Growth and Opportunity Act recommendations for Doha Round, 64 Agreement on Agriculture. See Uruguay Round APEC. See Asia Pacific Economic Co-operation Agreement on Agriculture (URAA) Forum Agreement on Trade-Related Aspects of Intellec- Applied Tariffs, 97, 98t5.3, 99b5.1, 109 tual Property Rights. See TRIPS agreement Argentina agricultural laborers bioengineered foods, approval of, 236, 240 Nicaragua coffee laborers, 9b1.2 decomposition of producer price changes, 202, wages of, 162 202t10.4 agricultural sector growth financial crisis of, 201 in developing countries, 3­4b1.1 transgenic crops in, 236 importance of, 2­5 U.S. GSP and, 295, 298b15.3 improved food security, 4­5 ASEAN Free Trade Area (AFTA), 293b15.1, 305­6, poverty reduction and, 2­4 311b15.7 agricultural subsidies. See domestic support com- Asia Pacific Economic Co-operation Forum mitments; export subsidies (ES) (APEC), 293b15.1, 306 367 368 Agriculture and the WTO Association of Southeast Asian Nations (ASEAN), future developments for, 249 294, 305 intellectual property rights and, 248 ATPA (Andean Trade Preference Act), 299­300 See also intellectual property rights ATPDEA (Andean Trade Promotion and Drug labeling policies for GM foods, 241­43, Eradication Act of 2002), 299­300 242t12.1 auctioning, 98t5.3, 99, 99b5.1, 101, 116 legislative evolution to deal with, 240­41 licenses on demand and, 110 online resources, 245t12.2 Australia pros and cons of, 239 agreement with New Zealand, 292 public participation in regulatory system on, bioengineered foods, approval of, 236, 240 241 determination of lowest production cost in, 209 regulation of, 239­44 domestic support shares of, 121 safety assessment of, 243­44 fill rates, 107 SPS and TBT agreements, relationship to, import quota, 108 246­48 labeling of GM foods, 242 techniques of, 236, 249n1 on multifunctionality, 169 trade implications of, 244­49 producer support estimates (PSEs) in, 8 transgenic crops, 236, 237, 237f12.1 Australian Bureau of Agricultural and Resource transparency and, 241 Economics, 57 TRIPS Agreement, relationship to, 248 authenticity and agricultural landscape, 168 blue box, 32, 120, 130­32, 144, 168 baseline AMS including, 122 bacillus thringiensis (Bt) due restraint and, 34 cotton, 237, 238f12.1 expectation of government subsidy and, 142 insect resistant crops and, 236 reforms needed, 144 Banana Dispute, 105­6, 305 border measures EC-Bananas case, 29, 30 in developing countries, 207 bananas development box to exclude, 282 See also Banana Dispute effect of domestic support vs. border support import licenses for, 111 on trade distortion, 135­38 preferential access for, 310 effect of removing all border protection on Bangladesh effective producer price and production, food price changes under induced wage 136, 137t6.11 responses and, 162 evolution of border vs. support in OECD agri- as LDC qualifying for special trade arrange- culture, 65­70, 65f4.1 ments, 295 PEM analysis, 135­38 banking of unused export subsidies, 49 tariff quota border protection and, 91 recommendations for Doha Round, 59 bound tariffs. See tariffs barriers to trade, 7­10 bovine spongiform encephalopathy, 244 See also market access brand identity of products, 262­63 baseline and front-loading flexibility, 49­53 Brazil beef beneficiary of U.S. GSP program, 295 mad cow disease, 244 bioengineered soybean in, 237 tariff peaks (mega-tariffs), 75 domestic support levels of, 122 biotechnology, 235­51 effect of trade liberalization on agricultural See also Cartagena Protocol on Biosafety value in, 211n3 Codex Alimentarius Commission. See Joint fill rates, 107 Codex Alimentarius Commission (CAC) Brussels Round (1990), 26 defined, 236 Bt. See bacillus thringiensis developing countries and, 235 environmental safety and, 243­44 CAC. See Joint Codex Alimentarius Commission Index 369 Cairns Group, 82, 93n12, 151, 169, 216, 333 cattle special and differential (S&D) treatment and, headage payments under production limiting 273 arrangements, 132 URAA negotiating position of, 216 mad cow disease, 244 Cambodia rice exports, 20b1.4 CBD. See Convention on Biological Diversity Canada CBI (Caribbean Basin Initiative), 298­99 administered prices in, 125 cereals agreement with Chile, 292 EU tax on exports of, 37 baseline support by commodity region and import costs per ton, 185, 185f9.1 type, 136 liberalization in Mexico and, 158 bioengineered foods, approval of, 236, 240, prices, 185­86, 185t9.1, 186f9.2 249n4, 250n13 CFF (IMF's Compensatory Financing Facility), butter exports and baseline level, 51 188 dairy pricing as source of dispute, 131b6.1, Chile 145n13 agreement with Canada, 292 de minimis support by, 135 bilateral free trade agreements with, 311, domestic support policies of, 122, 320tA.2, 312b15.8 321 decomposition of producer price changes, 202, export subsidies spending by, 43 203t10.5 farmers as landowners in, 158 EU trade with, 305 grains and oilseeds vs. other commodities, pro- financial crisis of, 201 tection of, 323, 326­333fA.1­A.16 milk prices, surcharge on imported, 211n5 gross revenue stabilization programs of, 143 price bands, 207 labeling of GM foods, 242 of edible oil, 208, 208f10.3 supply management programs in, 140 of sugar, 207­8, 208f10.2 tariff escalation in, 76 of wheat, 207, 208 tariff rates of, 8 trade liberalization in, 194 transgenic crops in, 236 China transparency of tariffs in, 70b4.1 Bt cotton, 237, 238f12.1 wheat production, 154 GSP schemes that benefit, 295 Canada-U.S. wheat agreement, 55­56 transgenic crops in, 236 Canadian Council of Grocery Distributors Chiquita and banana import licenses, 111 (CCGD), 242, 250n16 circumvention of export subsidy reduction com- Canadian General Standards Board (CGSB), 242, mitments, 45­54 250n17 baseline and front-loading flexibility, 49­53 canola, bioengineered, 237, 239, 244 recommendations for Doha Round, 59­60 Caribbean Basin Economic Recovery Act reprint of Agreement on Agriculture provisions, (CBERA), 298, 298t15.4 355 Caribbean Basin Initiative (CBI), 298­99 URAA Article 10 and, 34 Caribbean Basin Trade Partnership Act of 2000 coarse grains. See wheat and coarse grains (CBTPA), 298­99 Codex. See Joint Codex Alimentarius Commission Caribbean Community (CARICOM), 292, (CAC) 293b15.1 coffee market Caribbean countries duty free, 300 antidumping investigations initiated by, growers' ability to switch to another crop, 165n7 212n15 Nicaragua farmers and laborers and, 9b1.2 sugar and EU Sugar Protocol, 309 Colombia trade liberalization in, 194 Andean Trade Preference Act (ATPA), 299­300 Cartagena Protocol on Biosafety, 230, 245, 246­47, decomposition of producer price changes, 202, 247t12.3 204t10.6 370 Agriculture and the WTO Committee on Agriculture expectations about future policies and dynamic responsibility for Doha negotiations on agricul- considerations, 142 ture, 38, 39 fixed costs and, 140­41 URAA Articles 17 and 18 and, 36 full decoupling, 139 reprint of, 357­58 green box policies and, 132 commodities. See specific commodity ideal approach to, 142 Common Agricultural Policy (CAP) imperfect input markets, 142 adoption of, 25 risk and wealth, 141­42 direct payments in 1992 reform, 138 de minimis support, 31, 119, 134t6.9, 135, 181, 282, MacSharry reforms (1992), 19b1.3 289n10 producer support and, 8 See also blue box concessions, 24b2.1 developed countries Conference of the Parties (COP April 2002), 263, See also OECD countries 264 domestic support commitments of, 119 conservation, 5 special and differential (S&D) treatment and, consumer-only financed export subsidies, 55­56, 270­71 56b3.1 tariff binding overhang in, 68 Doha Round recommendations on, 60 trade liberalization in, 158­61, 158t7.7, 164, Convention on Biological Diversity (CBD), 196­97 232nn16­17, 246, 247, 257, 262, 263­64, developing countries 268n4 See also least-developed countries (LDCs) COP. See Conference of the Parties agricultural sector growth in, 3b1.1 copyright issues, 262 applied vs. bound tariffs and, 79, 181 See also intellectual property rights average tariffs of, 67, 67f4.4 corruption. See rent seeking biotechnology and, 235, 239 Costa Rica fill rates, 107 border measures in, 207 cost-benefit analysis, 338­41 credit for unilateral liberalization by, 39 Cotonou Agreement, 105, 294, 301, 303, 305 decoupled payments, effect on, 138­39 cotton de minimis support for, 135, 289n10 Bt cotton, 237, 238f12.1, 239 development box, suggestion of, 182 dispute over U.S. subsidies, 131b6.1 Doha Development Round issues of, 1­22, U.S. vs. Mali farmers of, 9b1.2 38­40, 193 country reserves and TRQs, 102 domestic support commitments of, 119, 120, credit for unilateral liberalization, 39 181 credit guarantees as export credit programs, 57 farmers as consumers in, 157 crop insurance subsidies in U.S., 141­42 International Task Force on Commodity Price CTA (GATT Committee on Trade in Agriculture), Risk Management pilot projects for, 26 188­89 Cuba and EU trade, 305 multifunctionality and, 169, 176 price distortion in, 204 dairy products. See milk and dairy products reciprocity and, 39 deadweight costs and licenses on demand, 111 special and differential (S&D) treatment and, deadweight loss 271­72, 273, 280­82, 288 economic efficiency and, 137 differentiation of countries needed, 285­86 of support for grains and oilseeds, 157, 157t7.6, SPS Agreement and, 215­16, 220­29 158t7.7 tariff binding overhang in, 67­68, 67f4.5 decoupling of agricultural support, 18, 19b1.3, 120, taxes on agricultural sector, 10 138­43 trade liberalization, effect on, 204­9 defined, 139 transparency of tariffs in, 70b4.1 EU policy, 131 development box, suggestion of, 182, 280­82 Index 371 Dillon Round (GATT), 25 Committee on Agriculture role. See Committee direct payments on Agriculture exemption from AMS commitments (blue box), deadlines for agriculture negotiations, 38 32 developing country interests in, 1­22, 38­40, as green box measures, 32 193 "dirty tariffication," 103, 274 export subsidies negotiations, 58­60 disaster relief support, 142 food security as issue for, 179 food aid. See food aid options in negotiations of, 37­41, 63­64, as green box measures, 32 143­45 World Bank and IMF facilities to assist with, special and differential (S&D) treatment pro- 188 posals in, 279­80, 285­86 discrimination trade liberalization, 5­6 EU agreements alleged to discriminate, 305 TRIPS issues for, 266­68 international trade and, 25 domestic purchase requirement, 100b5.2 milk and domestic price discrimination, 56 domestic support commitments, 37, 119­47 TRQs and, 96 amber box. See amber box disease control. See Sanitary and Phytosanitary AMS. See Aggregate Measurement of Support Agreement (SPS) blue box. See blue box dispersion in tariff rates. See market access decoupling, theory of, 138­43 distortions. See trade distortions See also decoupling of agricultural support distributional effects of agricultural policy de minimis support, 31, 125t6.4, 134t6.9, 135 reforms, 149­65 direct payments, 32 complimentary policies, 163 domestic producer support relative to total and imperfect markets, 161­63 trade-distorting support, 121, 124t6.2 imperfect price transmission, 161 economics of domestic support and trade dis- implications of Mexico's shift in policy mix, tortions, 135­42 159­61 baseline support by commodity region and Mexico, agricultural reform in, 151­52, 157t7.6, type, 135, 136t6.10 158, 163­64 effect of removing all border protection on net suppliers or demanders, 162­63 effective producer price and produc- off-farm labor income and, 163 tion, 136, 137t6.11 policy developments in grains and oilseed sec- effect of trade liberalization, 137, 138t6.13 tors, 150­52 effect on effective producer price and pro- trade liberalization duction of removing all direct pay- See also trade liberalization ments support, 136, 137t6.12 in all regions, 153­58, 153t7.1, 154t7.2, PEM analysis, 135­38 155t7.3, 155t7.4 empirical measures of, by commodity, 121, complementary polices to, 163 124t6.3 in developed vs. developing countries, government programs, 31­32 158­61 green box. See green box welfare effects of, 156­57, 156t7.5, 158, measure by box and country, 121, 125t6.4, 132 158t7.7 New Zealand transition to no support, 205, wages of agricultural laborers, 162 206b10.1 diversification in crops, programs in support of, notification requirements, 40­41 281­82, 289n14 in OECD countries, 121, 123t6.1 Doha Development Round overview of, 121­35 agricultural sector growth, importance of, 2­5 "Peace Clause" and, 131b6.1 agricultural trade vs. industrial goods trade, reforms needed, 120, 143­45 6­16 reprint of Agreement on Agriculture provisions, benefits from liberalization, 1­2, 16­20, 18t1.8 353­54 372 Agriculture and the WTO trends in, 122f6.1 cattle headage payments under production lim- URAA Article 6 and, 30­33 iting arrangements, 132 varying domestic support among members, coarse grain trade and, 156 32­33 consumer-financed subsidies in, 140 due restraint Cotonou Agreement. See Cotonou Agreement amber box and, 34 domestic support levels of, 37, 122, 126 blue box and, 34 Economic Partnership Agreements with, 275, domestic subsidies and, 131b6.1 305 export subsidies and, 34­35 equivalence, adoption of, 232n10 green box and, 34 Euro-Mediterranean Partnership, 303­5 URAA Article 13, 34­35 export subsidies spending by, 43 reprint of provisions, 356­57 export taxes on wheat and coarse grains, 54 duty-free access food safety issues in, 168­69 industrial goods, 93n1 grains and oilseed sector compared to other U.S. GSP and, 295 commodities, 323, 324, 329fA.7­A.8 import quota fill rates, 108 Eastern Europe intervention price and AMS, 122 import quota fill rates, 108 labeling of GM foods, 242 real price trends of wheat in, 201 most-favored-nation (MFN) tariffs of, 301, EBA. See "Everything but Arms" Agreement 304t15.6 EC-Bananas case. See bananas multifunctional benefits, support for, 167, 168, Economic Partnership Agreements (EU), 275, 305 169 Ecuador Precautionary Principle, 244, 249, 250n18 Andean Trade Preference Act (ATPA), 299­300 preferential agreements and GSP arrangements, Banana Dispute and, 105 301­5, 303b15.6 emergencies regional conditions, adaptation of, 232n14 ex ante fund for trade finance in, 190 rice compared to other commodities, food stocks for, 281 330fA.9­A.10 environmental issues, 5, 168, 170 sugar regime dispute, 131b6.1 agricultural pollution, 171, 173 tariff escalation and, 76 positive vs. negative externalities of, 174 tariff peaks (mega-tariffs) and, 74­75 programs as green box measures, 32, 132 tariff rates of, 8 transgenic plants and, 243­44 trade-distorting domestic support of, 121 equivalence and SPS Agreement, 220­23, 231, transparency of tariffs in, 70b4.1 232nn6­8 unused importer licenses for eggs, 110 Equivalent Measurement of Support (EMS), 121 URAA negotiating position of, 216 reprint of Annex 4 on, 363 value commitments and, 44t3.3, 53 ESEs. See export subsidy equivalents wheat production, 154 Euro-Mediterranean Partnership, 303­5 "Everything but Arms" Agreement (EBA), 271, European Union (EU), 293b15.1 301, 309 See also Common Agricultural Policy (CAP) export certificates, 100b5.2 acreage set-asides, 131­32 export competition policies, 43­62 agricultural support from taxpayers to farm See also export subsidies (ES) operators and land owners, 157 activities of state trading enterprises (STEs), in Banana Dispute, 105­6, 305 58 banking of unused export subsidies, 49 Doha Round recommendations on, 60 baseline for export subsidies, 51 circumvention of reduction commitments, bioengineered foods, approval of and morato- 45­54 rium on, 236, 240, 244­45, 249 See also circumvention of export subsidy blue box payments used by, 130, 131 reduction commitments Index 373 consumer-only financed export subsidies, value commitments 55­56, 56b3.1 by commodity, percentage use of, 47, 47t3.4 Doha Round recommendations on, 60 countries using over 90 percent of, 49, 50t3.7 expenditure vs. quantity limits, 55, 60­61, by country, percentage use of, 44t3.3, 47 61f3.A.1, 61f3.A.2 total value commitments, percentage use of, export credit programs, 57 47, 48, 48t3.5 Doha Round recommendations on, 60 volume commitments food aid, 57­58 countries using over 90 percent of, 49, 51t3.8 producer- financed export subsidies, 56­57 export volume receiving, percentage use of, public stock disposal, 57, 62n6 44t3.1, 45 URAA Article 8 and, 33 total volume commitments, percentage use export credit programs, 57 of, 47, 48, 49t3.6 Doha Round recommendations on, 60 export subsidy equivalents (ESEs), 53­54, 54t3.11 food aid, 57­58 export subsidies (ES) FAC (Food Aid Convention), 277 banking of unused, 49 FAIR Act. See Federal Agricultural Improvement recommendations for Doha Round, 59 and Reform Act of 1996 (U.S.) baseline FCFS. See first-come, first-served (FCFS) for quota allocation of value of, 46f3.1 administration front-loading flexibility and, 49­53 FDA. See Food and Drug Administration, U.S. binding commitments, 49 FDI (foreign direct investment) and intellectual circumvention of. See circumvention of export property rights, 256 subsidy reduction commitments Federal Agricultural Improvement and Reform Act consumer-only financed, 55­56, 56b3.1 of 1996 (U.S.), 138, 140, 145n10 Doha Round recommendations on, 60 Federal Milk Marketing Order (U.S.), 56b3.1 deadline for eliminating, 59 fertilizers. See environmental issues distortion effect of, 37, 53, 282 financial crises and world prices, 201 Doha negotiations on, 58­60 first-come, first-served (FCFS) for quota adminis- due restraint and, 34 tration, 97, 98t5.3, 99b5.1, 109, 112­13 food security and, 181 costs of waiting, 112 front-loading, 44 economics of, 114b5.3 flexibility, 49­53 serving time and, 113 recommendation to ban, 59 storage costs and, 113, 114b5.3 value front-loading, 51, 52t3.9 fixed costs volume front-loading, 51, 52t3.10 direct income payments and, 140­41 negative effects of, 10 licenses on demand and, 110­11 notification requirements, 41, 43, 182 fixed world reference price, 121 per unit subsidy flood mitigation and Japan's rice production, 168 ad valorem limits on, 59 food aid, 34, 57­58, 276­77 asymmetry of protection and, 53 recommendations for Doha Round, 59 policy incoherence and, 9b1.2 subsidized credit and, 189 producer-financed, 56­57 Food Aid Convention (FAC), 277 reprint of Agreement on Agriculture provisions, Food and Drug Administration, U.S. (FDA) 354­55 labeling guidance for GM foods, 242 rollover of unused, 33, 49 review of bioengineered foods, 236, 240­41 recommendations for Doha Round, 59 food safety issues, 168, 170 taxpayer-financed, 55, 56, 62n5 See also biotechnology; Sanitary and Phytosani- total commitments, 44t3.2 tary Agreement (SPS) URAA Article 9 and, 33 food security, 179­91 URAA Article 12 and, 34­35 agricultural sector growth and, 4­5 374 Agriculture and the WTO cereal import prices and, 185­86 tariff rate quotas, 103­6 commodity price risk management and, 188­89 General Agreement on Trade in Services (GATS) description of, 179­84 rules on discrimination, 106 ex ante fund for trade finance in emergencies, Generalized System of Preferences (GSP), 292­93 190 benefits of, 306­8 food aid. See food aid costs of, 308­10 food vouchers, 189­90 effect on developing countries, 274­75 IMF and World Bank, 188 effect on least-developed countries, 274, 295, indicators, 182, 182b9.1 307, 308t15.7 for individual and family, 179­82 of European Union, 301­5 at international level, 184 agreements in force as of April 2002, Marrakesh Decision. See Marrakesh Decision 303b15.6 measures to ameliorate threats, 187­90 beneficiaries of, 301 multifunctionality and, 168, 169, 170, 174­75 preference margins for AACP exports to EU, multilateral trade liberalization, 187 308, 309t15.8, 309t15.10, 312 at national level, 182­84 future of, 311­13 short-term threats to, 189­90 history of, 270­71, 294b15.2 special and differential treatment and, 273, 281, improvements recommended for, 286 286 trade agreements and, 294­95 URAA Articles 12 and 13 on export restraints of United States, 295­301 and due restraint, 34­35 annual timetable, 296, 300b15.5 food vouchers, 189­90 Argentina's economic crisis and, 295, foreign direct investment (FDI) and intellectual 298b15.3 property rights, 256 beneficiaries of, 295, 296t15.1 Freedom to Farm Act (U.S. 1996), 19b1.3 criteria and conditions, 295, 299b15.4 free trade areas (FTAs), 292, 305 imports, 297t15.2, 297t15.3, 298t15.4 See also specific agreements (e.g., Euro-Mediter- utilization rate of, 307 ranean Partnership) genetically modified foods (GM foods), 230, 236 front-loading, 44 labeling requirements. See labeling require- flexibility, 49­53 ments recommendation to ban, 59 second generation of, 236 value, 51, 52t3.9 U.S., Canada, and Argentina to challenge EU volume, 51, 52t3.10 moratorium on, 249 geographical indications (GIs) and trademarks, GATT Committee on Trade in Agriculture (CTA), 262­63 26 Global Trade Analysis Project general equilibrium General Agreement on Tariffs and Trade (GATT) framework, 335 Agreement on Import Licensing (Licensing GM foods. See genetically modified foods Agreement), 106 government programs agricultural exceptions of, 25 See also domestic support commitments Article XIII, "Non-discriminatory Administra- as green box measures, 31­32 tion of Quantitative Restrictions," 103­6 intervention. See intervention benefits of, 7 grains and oilseed sectors history of agricultural negotiations in, 23­26, border protection for, 136 24b2.1 domestic support levels and, 37 history of efforts to provide rules for trade in Mexican pilot hedging program for, 152 agricultural commodities, 25­26 OECD support for, 317­21, 318tA.1, 326fA.1 market access by developing countries as goal policy developments in, 150­52, 317­41 of, 291 in Policy Evaluation Model (PEM), 83, 88t4.7, MFN treatment under, 23 135, 149 Index 375 shift away from border support, 65­66, 66f4.2 notification of subsidies by, 278 See also market access industrial goods trade liberalization and, 153­54, 155t7.3 agricultural trade vs., 6­16 green box, 31, 119, 132­35 tariff rates, 12t1.3 due restraint and, 34 duty-free, 93n1 environmental use of, 120 Information Technology Agreement, 311 expectation of government subsidy and, 142 infrastructure, reforms needed for trade-related, expenditures by category, 133, 133t6.8 18 multifunctionality and, 169 in-quota tariffs notification, 41 average vs. out-of-quota average, 70 reforms needed, 145 background of, 95 social amenities and, 120 over-quota imports, when allowed, 88 special and differential treatment and, 280 reduction in, 109 GSP. See Generalized System of Preferences tariff regime, 78, 78f4.12a insurance programs Haberler Committee Report (1958), 25 as export credit programs, 57 Health Canada, 236 U.S. government, 141­42 historical allocation of tariff quotas, 98t5.3, 99, Integrated Framework for Trade Related Technical 99b5.1, 109, 111­12 Assistance for the Least Developed Coun- tries, 273 Iceland integration into world markets, 20b1.4 blue box payments used by, 130 intellectual property rights, 253­68 nominal protection coefficient, 68 See also TRIPS agreement ICPM (Interim Commission on Phytosanitary developing countries and, 239 Measures), 226­28 development impacts of, 256 IFPRI (International Policy Research Institute), 4 geographical indications and, 262­63 IMF's Compensatory Financing Facility (CFF), 188 importance of, 253­54 imperfect markets, 142, 161­63 international agreements on, 257­62 imperfect price transmission, 161 lack of international consensus over, 265­66 import quotas market structure and, 262 See also tariff rate quotas (TRQs) online resources for, 254t13.1 fill rates, 106­8, 108, 109f5.1 plant genetic resources for food, 264­65 distribution of, 108t5.5 plant variety protection (PVP) and, 259­61 simple average fill rates, 107t5.4 research and development and, 256, 261­62 under-fill, 109, 114, 116 seed provision and, 260­61 in PEM, 88, 88t4.8 tradeoffs involved in, 254­56 imports WIPO and, 257, 265 See also import quotas; market access interest rate subsidies as export credit programs, licensing procedures, 30 57 protection levels, 65­70 Interim Commission on Phytosanitary Measures undertaken by state trading entities, 99b5.1 (ICPM), 226­28 income insurance as green box measures, 32, 132 International Finance Corporation program, income support as green box measures, 32 20b1.4 India International Office of Epizootics. See Office Inter- basmati rice, geographic indication of, 263 national des Epizooties (OIE) beneficiary of U.S. GSP program, 295 International Plant Genetic Resources Institute indigenous knowledge, protection of, 263 (IPGRI), 260 Indonesia International Plant Protection Convention (IPPC), beneficiary of U.S. GSP program, 295 218, 220b11.3, 226­28 fill rates, 107 International Policy Research Institute (IFPRI), 4 376 Agriculture and the WTO International Service for the Acquisition of Agri- nominal protection coefficient and, 68 biotech Applications (ISAAA), 237 producer support and, 8 International Task Force on Commodity Price Risk state trading of sugar, 115 Management pilot projects, 188 tariff escalation and, 76 International Trade Organization (ITO), 25 tariff peaks (mega-tariffs) and, 75 International Treaty on Plant Genetic Resources tariff quotas and total imports, 88 for Food and Agriculture (ITPGRFA), 257, tariff rates of, 8 260, 261, 264­65 trade liberalization's effect on, 164 International Undertaking on Plant Genetic transparency of tariffs in, 70b4.1 Resources for Food and Agriculture (IU), URAA negotiating position of, 216 264 wheat and MPS, 136, 154 International Union for the Protection of New Joint Codex Alimentarius Commission (CAC), Varieties of Plants (UPOV), 257, 259, 261 218, 218b11.1, 226, 227, 232nn12­13 intervention GM foods, consensus on safety issues related to, domestic price and trade intervention, 197, 243, 248­49 211n7 government intervention, 171 Kennedy Round (GATT), 25 price and AMS, 122 Korea, Republic of IPGRI (International Plant Genetic Resources domestic support levels of, 121, 122 Institute), 260 fill rates and, 108 IPPC. See International Plant Protection Conven- labeling of GM foods, 242 tion multifunctional benefits, support for, 167, 168, IPRs. See intellectual property rights 169 ISAAA (International Service for the Acquisition nominal protection coefficient, 68 of Agri-biotech Applications), 237 ITO (International Trade Organization), 25 labeling requirements, 219 ITPGRFA. See International Treaty on Plant for animal feed products derived from GM Genetic Resources for Food and Agriculture plants or seeds, 249 IU (International Undertaking on Plant Genetic CAC process and, 148­149 Resources for Food and Agriculture), 264 for GM foods, 241­43, 242t12.1 land markets Japan in Mexico, 152 administered prices in, 125 reforms needed, 18 agricultural support from consumers to farm landowners operators and land owners, 157 liberalization's effect on Mexican landowners, bioengineered foods, approval of, 236, 240 158, 164 blue box payments used by, 130 as recipients of agricultural support programs, border protection, use of, 136 157 deadweight loss and consumer prices in, 137 Latin America domestic support levels of, 37, 121 See also specific countries domestic support policies of, 320tA.2, 321 antidumping investigations initiated by, 212n15 eliminating MPS, benefits for, 137­38 price bands, use of, 207­8 fill rates and, 108 least-developed countries (LDCs), 35 flood mitigation and rice production, 168 agreements with, 293 grains and oilseeds policies compared with cereal prices and, 185, 185t9.1 other commodities, 156, 326­27, Committee on Agriculture role, 36 331fA.11­A.12 domestic support in developed countries, effect labeling of GM foods, 242 on, 135 multifunctionality, position on, 167, 168, 169, exemption from subsidy reduction commit- 175 ments, 43, 119 Index 377 Integrated Framework for Trade Related Tech- evolution of border vs. support in OECD agri- nical Assistance for the Least Developed culture, 65, 65f4.1 Countries, 273 import protection levels and border support, Marrakesh Decision and, 184 65­70, 66f4.2, 66f4.3 preferences and, 274, 295, 307, 308t15.7 border vs. total support in OECD countries, reprint of Agreement on Agriculture provisions, 65, 65f4.1 357 liberalizing tariffs and tariff quotas in OECD rules of origin and, 310 countries, 82, 88­91 special treatment of, 272, 284 PEM model, 83, 88­90, 88t4.7, 88t4.8, 89t4.9, liberalization. See trade liberalization 89t4.10 licenses Swiss formula (SW), 83, 84­87t4.6, 90­91, Agreement on Import Licensing (GATT), 91t4.11 106 nominal protection coefficient on demand, 97, 98t5.3, 99b5.1, 109, 110­11, by commodity, 68, 71f4.9 117n2 by OECD country, 68, 71f4.8 deadweight costs and, 111 policy options for market access commitments, fixed costs and, 110­11 91­93 inefficient methodology of, 108­9, 116 expanding quotas vs. reducing tariffs, 92­93 inequity in allocation of import licenses, 106 tariff reductions, 92 nonuse penalties, 110 reprint of Agreement on Agriculture provisions, tariff rate quotas (TRQs), 96 351 linear expenditure system (LES), calibration of, special and differential provisions and, 343­46 274­75 Lomé Convention, 105, 294, 301, 309, 310t15.9 special safeguard tariffs, 76, 78t4.5 lottery, 109, 115­16 tariff binding overhang, 82, 93 low price episodes. See prices in developed countries, 68, 68f4.6 in developing countries, 67­68, 67f4.5 macroeconomic policies and agricultural growth tariff escalation, 76, 77t4.4 rate, 5 tariff peaks and dispersion, 74­75, 75t4.3, 92 MacSharry reforms (EU 1992), 19b1.3 trade arrangements to improve, 291­315 mad cow disease, 244 See also Generalized System of Preferences maize (GSP) See also Mexico URAA Article 4, 27­30, 63, 103 bioengineered (Bt), 237, 239 water in the tariff, 82, 93 Malaysia fill rates, 107 by commodities, 70, 74f4.11 Mali cotton farmers, 9b1.2 by countries, 70, 74f4.10 market access, 63­94 defined, 93n4 average bound tariffs market price support (MPS), 82­83, 145n4 by commodities, 68, 69f4.7 border vs. total support in OECD countries, 65, by region, 66­67, 67f4.4 65f4.1 background, 63, 64­76, 91­94 wheat and, 136 basic economics of barriers to, 76, 78­82, 78t4.5 Marrakesh Decision, 26, 35, 184­87 complexity of quotas, 80b4.2 notification requirements, 41 implications for negotiations, 79­82 Mauritania as LDC qualifying for special trade three TRQ regimes, 78 arrangements, 295 empirical estimates of transfers MERCOSUR, 293b15.1 due to country policies, 68­69, 72t4.1 as customs union, 292 due to policies by commodities, 69, 73t4.2 equivalence, adoption of, 232n10 evidence of market barriers in agriculture, EU trade with, 305 64­76 mergers and acquisitions, 262 378 Agriculture and the WTO Mexico milk and dairy products agricultural reform in, 149, 151­52, 157t7.6, Canadian pricing as source of dispute, 131b6.1, 158, 163­64 145n13 applied tariffs in, 89­90 Chile, surcharge on imported milk prices, ASERCA (Support Services for Agricultural 211n5 Marketing), 152, 164 domestic price discrimination, 56 base income levels and percentage change, empirical estimates of transfers, 69 157t7.6, 158 New Zealand dairy policy, 56, 56b3.1 bilateral agreements and nontransparency, nominal protection coefficient, 68 70b4.1 special safeguard (SSG) provisions, 76 decoupled payments in, 138, 152 tariff peaks (mega-tariffs) and, 75 disaggregated system of demand and produc- tariff quotas and, 99 tion for, 342­47 U.S. administered price support for, 125 distribution of effects of liberalization on Mexi- U.S. Federal Milk Marketing Order, 56b3.1 can base levels and percentage change, mixed allocation methods for tariff quotas, 98t5.3, 159t7.8 99, 99b5.1, 109 eliminating MPS, benefits for, 137­38 Montreal Round (1988), 26 EU trade with, 305 most-favored-nation (MFN) tariffs, 23 Farmers Direct Support Program (PRO- bound tariffs by region, 66­67, 67f4.4 CAMPO), 19b1.3, 138, 157, 161­62, developing countries and, 274, 286 164 European Union's, 301, 304t15.6 grains and oilseeds policies, 327, 332, 332fA.14, exceptions from, 292 334tA.6 reductions in, 59 imperfect markets in, 161­63 TRQ administration and, 103­4 implications of liberalization for, 158­59, multifunctionality, 167­77 159t7.8, 164 analytical framework for, 171­72 imperfect price transmission, 161 benefits, valuation, 172­73, 172b8.1 implications of shift in policy mix of, 159­61, characteristics of, 170­71 160t7.9 economic perspectives on, 170­76 imports in excess of tariff quotas, 88, 93n13 environmental benefits and, 174 landowners in, 158 externalities and, 171 agricultural support from consumers and food security. See food security taxpayers to, 157 government intervention and, 171 liberalization's effect on, 159, 164 implications of, 176­77 maize policy of, 152, 158, 162 joint production of secondary benefits in, disaggregated production and, 346 170­71 impacts of hypothetical change in, 160t7.9 legal dimensions of, 170 National Company of Popular Subsistence outputs from agriculture, 174­76 (CONASUPO), 151, 152 policy implications of, 173­74 rice policies compared with other commodities, as pretext for protection, 169 332, 333fA.15­A.16 proponents of vs. opponents of, 167­70, 176 small farmers in rural amenities and, 175­76 historical entitlements and, 160­61 valuation of benefits, 172­73, 172b8.1 liberalization's effect on, 159 multinational negotiations support programs, 19b1.3, 136 See also trade agreements elimination of, 152 advantages of, 23 wheat and MPS, 136, 154 effect of, 24b2.1 middle-income countries import protection levels, 65 NAFTA. See North American Free Trade Agree- reform issues for, 20 ment Index 379 net-food-importing countries (NFIDCs), 35 fill rates and, 108 Committee on Agriculture role, 36 GM foods, consensus on safety issues related to, reprint of Agreement on Agriculture provisions, 243 357 grains and oilseed sector policies of, 150­52, special and differential (S&D) treatment and, 317­21, 318tA.1, 322tA.3 273 compared to other commodities, 323, neutraceutical products, 236 326fA.1­A.2 New Zealand import protection levels of, 65 agreement with Australia, 292 import quota fill rates of, 108, 109f5.1 dairy policy, 56, 56b3.1, 131 liberalizing tariffs and tariff quotas in, 82, 88­91 equivalence implementation by, 232n6 nominal protection coefficient by country, 68, import quota fill rates, 108 71f4.8 labeling of GM foods, 242 policy incoherence in, 9b1.2 producer support estimates (PSEs) in, 8 tariff rate quotas for, 95 transition from government support to no sup- tariff reductions in, 1 port, 205, 206b10.1 URAA and policy distortions in, 7­10, 36 NFIDCs. See net-food-importing countries Office International des Epizooties (OIE), 218, Nicaragua coffee farmers and laborers, 9b1.2 219b11.2, 226, 227 nondiscriminatory application of tariff cuts, 39 online resources nonoboviousness principle, 262 biotechnology, 245t12.2 nontariff barriers (NTBs) intellectual property rights, 254t13.1 frequency of, 10, 15t1.5 Organisation for Economic Co-operation and preferences and, 307 Development (OECD) North American Free Trade Agreement (NAFTA), countries of. See OECD countries 293b15.1, 306 on decoupling policy, 139 equivalence, adoption of, 232n10 recommendations on valuation studies, 173 Mexican commitments to, 19b1.3, 151­52, 346 study of agricultural policies and trade, 26 preferential treatment and tuna industry, 311, study of export credit programs, 57 311b15.7 study on jointness between commodity and North-South trade and equivalence dimension, noncommodity outputs, 171 221­22 out-of-quota tariffs Norway average vs. in-quota average, 70 banking of unused export subsidies, 49 background of, 64, 95 blue box payments used by, 130 with quota underfill and out-of-quota imports, domestic support shares of, 121 80, 81f4.14 multifunctional benefits, support for, 167, 168 tariff regime, 78, 78f4.12c, 80 nominal protection coefficient, 68 notification requirements, 40­41 partial reforms, effect on tariff reduction, 197 of export subsidies (ES), 41, 43, 182, 278 past trading performance, 100b5.2 of green box measures, 41 patents market access and, 40 See also intellectual property rights Marrakesh Declaration and, 41 nonoboviousness principle, 262 of special safeguards, 40 TRIPS Agreement and of SPS Agreement, 228­29 Article 27.3(b) and geographical indications, 262­63, 266 OECD countries Article 27 and patentability, 258­59 domestic support estimates of, 121, 123t6.1, 126 definition of invention as excluding materi- empirical estimates of transfers, 68­69, 72t4.1 als found in nature, 259 evolution of border vs. support in, 65f4.1 section 5, 257­58 export subsidies (ES) in, 10, 44t3.1, 45 Peace Clause (URAA provision), 29, 34, 119, 279 380 Agriculture and the WTO See also due restraint Chile, 202, 203t10.5 domestic subsidies and, 130, 131, 131b6.1 Colombia, 202, 204t10.6 reforms needed, 143­44 financial crises and world prices, 201 PEM. See Policy Evaluation Model intervention per unit subsidy and AMS, 122 ad valorem limits on, 53, 59 domestic price and trade intervention, 197, asymmetry of protection and, 53­54 211n7 recommendations for subsidy reduction, 145 low price problem pesticides. See environmental issues for developing countries, 193, 204, 205 Philippines and Bt maize, 237 government intervention, 212n11 Plant Breeder's Rights (PBRs), 258, 259­60 for producers, 194, 195, 212n12 plant variety protection (PVP), 259­61 for rural poor, 162 Poland market price support. See market price support banking of unused export subsidies, 49 (MPS) scheduled tariff quotas, 117n8 movement, 198, 199f10.1, 199t10.2, 200 Policy Evaluation Model (PEM), 83, 88­90, reduction in volatility, 2 165n1 risk. See risk conceptual framework using, 335­38 trade liberalization and. See trade liberalization grains and oilseeds protection, 83, 88t4.7, 135, transmission. See trade liberalization 149, 150­51, 317­41 world price fluctuations, 197­200, 198t10.1 import quotas, 88, 88t4.8 domestic volatility and, 196 structure of, 341­42 pricing-to-market, 58 tariff rates, 89t4.10 principal administration methods, categories of tariffs applicable to tariff quota commodities, tariff quotas, 99b5.1 89t4.9 processed foods policy incoherence, 9b1.2 protectionism and, 8, 10 population shift from rural to urban areas, 175 tariff escalation and, 76, 77t4.4 poverty tariffs on, 10, 28 agricultural sector growth reducing, 2­4, 3b1.1, producer-financed export subsidies, 56­57 150 producer groups or associations, 98t5.3, 99, 99b5.1, food security issues. See food security 109, 113 rural-urban poverty gap, 4, 4t1.1 producer support estimates (PSEs) trade liberalization's effect on, 165n8 evolution of AMS and PSE, 125, 129t6.7 Poverty Reduction Strategy Papers (PRSPs), 2, 4, evolution of MPS, TSE, and PSE, 65, 65f4.1 4t1.1 measure of producer support, 93n5 Precautionary Principle and GM products, 244, in OECD countries, 8 249, 250n18 trends vs. AMS, 125, 130f6.2 preferential agreements, 96, 116 prosperity and trade liberalization, 5­6, 6f1.1 See also Generalized System of Preferences protectionism (GSP); specific agreements in developing countries, 11t1.2 Andean Trade Preferences, 299­300 international trade and, 25 European Union, 301­5 for processed products, 8, 10, 11t1.2 prevention of circumvention of export subsidy PRSP. See Poverty Reduction Strategy Papers commitments. See circumvention of export PSEs. See producer support estimates subsidy reduction commitments public assumption of risk as export credit pro- prices grams, 57 administered prices, 125 public stock disposal procedures, 57, 62n6 cereal prices after URAA, 185­86, 185t9.1 Punta del Este Declaration, 216, 217 determination of real prices, 198 purchased input subsidies, 160 Argentina, 202, 202t10.4 PVP (plant variety protection), 259­61 Index 381 quota administration methods, 95­118 rules of origin, 310 See also tariff rate quotas (TRQs) RUNS model, 197, 212n8 quota-binding regime, 78, 78f4.12b rural areas communities, support for viability of, 175­76 reciprocity, 24b2.1, 39 poverty, extent of, 2, 4, 4t1.1 recombinant-DNA technology, 236 Russian Federation See also biotechnology bioengineered foods, approval of, 236 regional assistance support as green box measure, real price trends of wheat in, 201 32, 132 regional conditions, consideration in SPS Agree- salmon, genetically engineered, 236, 250n7 ment, 229, 231 Sanitary and Phytosanitary Agreement (SPS), 18, regional trade areas (RTAs), 292 215­34 rent seeking burden of proof and costs sharing, 222­23 FCFS and, 112 conflict with Cartagena Protocol, 246, 247­48 licenses on demand and, 110 developing countries and, 215­16, 220­29 by those wishing to obtain quotas, 101­2 equivalence issue, 220­23, 231, 232nn6­8 Republic of Korea. See Korea, Republic of features and legal framework of, 217­20 research and development, 256, 261­62 governments' reaction to, 215­16 See also intellectual property rights international standards and organizations, retirement programs as green box measures, 32 226­28 rice multifunctionality and, 170 border protection for, 136 negotiating history of, 216­17 Cambodia exports, 20b1.4 options, 231 EU rice policies compared to other commodity precautionary approach of, 229­31 policies, 326, 330fA.9­A.10 recommendations for Doha Round, 64, 231 fill rates and, 108 regional conditions, consideration of, 229, golden rice, bioengineering of, 236 231 India and basmati rice, 263 reprint of Agreement on Agriculture provisions Japan's position on multifunctionality and, 168, on, 357 175 special and differential treatment, 223­25, 231, nominal protection coefficient, 68 275­76 special treatment as staple in traditional diet, transparency and notification procedures, 278 228­29, 231 tariff quota regime for, 83 scenic landscapes as positive externalities, 174 trade liberalization and, 153, 155t7.4, 156, 158, seeds and patenting issues, 260­61 197 serving time and first-come, first-served, 113 Rio Declaration on Environment and Develop- Slovak Republic ment, 230 blue box payments used by, 130 risk domestic support shares of, 121 commodity price risk management, 188­89, fill rates, 107 205 Slovenia in context of WTO commitments, 204­9 blue box payments used by, 130 consumers and price risk, 211n1 domestic support levels of, 122 decoupling of agricultural support and, 141­42 small farmers price transmission and price risks facing farm- International Task Force on Commodity Price ers, 195, 200­202 Risk Management pilot projects for, public assumption of risk as export credit pro- 188­89 grams, 57 Mexico rollover of unused export subsidies, 33, 49 historical entitlements and, 160­61 recommendations for Doha Round, 59 liberalization's effect on, 159 382 Agriculture and the WTO South Africa SPS. See Sanitary and Phytosanitary Agreement bioengineered foods, approval of, 236, 240 state-owned enterprises, reforms needed for, 18 EU trade with, 305 state trading enterprises (STEs), 58, 109, 113­15 export subsidies spending, 43 Doha Round recommendations on, 60 South-South trade factors affecting trade with, 113, 115b5.4 barriers to, 10 fill rates and, 108, 117 equivalence dimension and, 221 imports undertaken by, 98t5.3, 99, 99b5.1 tariff escalation and, 76 monopoly and monopsony power of, 115 soybeans, bioengineered, 237, 239, 245 recommendations to improve transparency of, special and differential (S&D) treatment, 223­25, 117 231, 269­90 STEs. See state trading enterprises agreements, 287­89 storage costs and first-come, first-served, 113, conceptual issues regarding, 272­80 114b5.3 country classification for, 282­83 Sub-Saharan African countries developed countries' role, 270­71 See also African Growth and Opportunity Act developing countries and, 271­72, 280­82, 288 (AGOA) Doha Round proposals on, 279­80, 285­86 agricultural growth rate in, 5 exemptions from disciplines, 271­72 trade with U.S., 300­301, 302t15.5 flexibility in rules and disciplines, 277­79, 287 subsidies. See domestic support commitments; future priorities, 285­86 export subsidies (ES) implementation of provisions, 271, 274­79 substantial suppliers, 96, 108, 116 least-developed countries and, 272, 284, 288 sugar low-income or resource-poor producers and, dispute over EU regime, 131b6.1 281 empirical estimates of transfers, 69 market access and, 274­75 EU Sugar Protocol, 309, 310t15.9 negotiations on, 284­85 nominal protection coefficient, 68 preferential market access, 270­71 price band in Chile, 207­8, 208f10.2 provisions, 270­72, 280­84 price floors of, 208 recommendations for Doha Round, 64 state trading in Japan and, 115 reprint of Agreement on Agriculture provisions, trade liberalization and, 197 287­89, 357 U.S. GSP program and, 295 Annex 5, 363­66 sui generis system of plant variety protection, safeguards for, 283, 286 259­60 sanitary and phytosanitary issues. See Sanitary supply management programs, 140 and Phytosanitary Agreement (SPS) support systems technical and other assistance, 223, 271, 276­77 GATT and, 25 time extensions as, 272 reforming insufficient, 19 transition periods for, 283­84, 285, 288 Swiss formula (SW) URAA Article 15 and, 35 for improving market access in TRQ regimes, WTO provisions, 271 90­91 special safeguard (SSG) provisions, 76, 210­11 for tariff reductions, 83, 84­87t4.6, 92, 93n10 by country, 78t4.5 Switzerland developing countries not able to use, 204 agricultural support from consumers and tax- notification requirements, 40 payers to land owners, 157 price-based, 30 applied tariffs in, 89 recommendations for Doha Round, 64 domestic support levels of, 122 reprint of Agreement on Agriculture provisions, eliminating MPS, benefits for, 137­38 351­53 export subsidies spending by, 43 URAA Article 5, 30, 76 grains and oilseeds policies, 327, 332fA.13 volume-based, 30 imports and tariff quotas for, 88 Index 383 multifunctional benefits, support for, 167, 168, giving TRQs to firms that make best use of 169 them, 101 nominal protection coefficient and, 68 restrictions on import allocations, 102 wheat and MPS, 136, 154 permanent vs. annual quotas, 102 profitability tied to size of quota allocations, tariffication 102 "dirty tariffication," 103, 274 recommendations for Doha Round, 64 effect in developed countries, 198 recommendations to improve transparency, transparency and, 70b4.1 116­17 URAA and, 28, 36, 37, 64 rent seeking by those wishing to obtain quotas, tariff rate quotas (TRQs), 80b4.2, 95­118 101­2 See also out-of-quota tariffs STEs and, 115b5.4 administration of, 30, 36­37 substantial suppliers' share, 96, 108, 116 alternative methods, 97­100, 97t5.1, 98t5.2, supplier tariff quotas and, 104 98t5.3 tradability of quotas and/or licenses, 109, 116 inefficiency of methods, 108­16, 115b5.4 for wheat and coarse grains, 83 policy options for reforming, 116­17 tariffs restrictions on, 102­3 applied tariffs, 97, 98t5.3, 109 alternative regimes for, 78, 79f4.12 vs. bound tariffs, 79­82, 81f4.13, 181 annual vs. permanent quotas, 102 average bound tariffs background of, 64, 65, 95 by commodities, 68, 69f4.7 categories of principal administration methods, by region, 66­67, 67f4.4 99b5.1 binding overhang, 82, 90t4.10, 93 country-specific allocations, 96, 105, 108, 116 in developed countries, 68, 68f4.6 distribution of quota rents and, 101, 106 in developing countries, 67­68, 67f4.5 equity of quota allocations, 102 bound tariffs European Union's, 301 applied tariffs vs., 79­82, 81f4.13, 181 first-come, first-served, 97, 98t5.3, 99b5.1, average bound tariffs by commodity, 68, 112­13 69f4.7 economics of, 114b5.3 in developing countries, 207 GATT Article XIII, 103­6 exceptions to, 28­29 GATT, WTO, and URAA rules on, 29, 95, 103­6 definitions of alternative tariffs, 79, 81f4.13 global quota, 96, 108, 116 administration of under URAA, 30 inefficiency of methods, 108­16, 115b5.4 three TRQ regimes, 78, 79f4.12 additional conditions, 100b5.2, 110, 116, 117 escalation, 76, 77t4.4 historical allocation, 98t5.3, 99, 99b5.1, 109, recommendations for Doha Round, 64 111­12 high tariffs and trade barriers, 5, 7­10, 11t1.2, import quota fill rates, 106­8, 109f5.1 12t1.3, 14t1.4, 15t1.5 licenses on demand, 97, 98t5.3, 99b5.1, 109, liberalizing tariffs and tariff quotas in OECD 110­11 countries, 82, 88­91 state trading enterprises, 113­15, 115b5.4 See also market access licenses to administer, 96 peaks and dispersion, 74­75, 75t4.3, 92 lottery, 115­16 recommendations for Doha Round, 64 more efficient administration methods for, 96, Swiss formula (SW) for reductions, 83, 100­103, 116­17 84­87t4.6 administration restrictions, 102­3 trade growth and, 7­10, 7f1.2 aggregation, 102 transparency of, 70b7.1, 92 country reserves, 102 uniform vs. differentiated, 211n2 efficient TRQ operating system, 101­3 unilateral reduction of, 39 full utilization of TRQs, 100 URAA Article 4: Market Access, 27­30 384 Agriculture and the WTO taxes on agricultural sector, 10 preferences and, 308­10 export taxes, use of, 54 URAA and, 7­10, 36, 120 taxpayer-financed export subsidies, 55, 56, 62n5 trade liberalization, 16­20, 153­58 technical assistance to developing countries, 41, adverse impacts of, 193­213 222, 271, 273, 276­77, 285, 289 agricultural value and, 211n3 Technical Barriers to Trade Agreement (TBT), approaches to, 23 219­20 background of, 194­95 conflict with Cartagena Protocol, 246, 247­48 benefits of, 1­2, 17­18, 164 Thailand See also Doha Development Round beneficiary of U.S. GSP program, 295 coefficients of variation of world prices, 198, labeling of GM foods, 242 198t10.1 tobacco products complementary polices to, 163 out-of-quota tariffs and, 75 descriptive statistics of selected commodities, TRQs, 97 199t10.2 Tokyo Declaration (1979), 292 developed-country subsidies and, 196­97 Tokyo Round, 25 distributional effects of. See distributional tomatoes effects of agricultural policy reforms added lycopene, 236 low price episodes and, 193, 194, 195, 204, 205 genetically modified for high-salt soils, 236 price bands, 207­8, 212nn10 & 13 total support estimate (TSE) of edible oil Chile, 208, 208f10.3 evolution of MPS, TSE, and PSE, 65, 65f4.1 of sugar in Chile, 207­8, 208f10.2 measure of total support, 93n5, 145nn1­2 price floors and, 208­9 tradability of quotas and/or licenses, 109, 116 price instability and, 197­200, 198t10.1 trade price risk and WTO commitments, 204­9 agricultural vs. merchandise, 7, 7f1.2 price transmission problems, 200­202, 204­5 as driver of economic growth, 6, 6f1.1 imperfect price transmission, 161 trade arrangements to improve market access, as increasing producer vulnerability, 195­96 291­315 price risks facing farmers and, 200­202 See also Generalized System of Preferences prosperity and, 5­6, 6f1.1 (GSP) real price of selected commodities, 198, Trade Act of 2002, 295 199f10.1 trade agreements, 23­42, 293b15.1 reaping benefits of, 18, 20 See also specific agreement (e.g., North American special safeguards provision, 210­11 Free Trade Agreement) See also special safeguard (SSG) provisions discrimination and, 25 welfare effects of, 156­57, 156t7.5, 158, 158t7.7 history of efforts to provide rules for trade in world price regimes and, 194­95 agricultural commodities, 23­26 WTO safeguards and contingency measures, multilateral agreements, 311 209­11 multinational negotiations, effect of, 24b2.1 trademarks, 262­63 protectionism and, 25 Trade Related Intellectual Property Agreement. See URAA, 26­37 TRIPS agreement legal position in GATT, 23, 25­26 trade secrets, 262­63 trade distortions transgenic plants and animals, 236, 237, 237f12.1, developing vs. developed countries and, 197 249n2 domestic support as cause of, 119, 120, 121, transparency 122f6.1, 180 biotechnology and, 241 effect of domestic support vs. border support quota administration and, 116­17 on, 135­38 SPS Agreement and, 228­29, 231 export subsidies and, 37, 53, 282 tariffication and, 70b4.1 PEM analysis, 135­38 TRIPS agreement, 253, 257­59 Index 385 Article 27 and patentable subject matter, Generalized System of Preferences program of, 258­59 295­301 Article 27.3(b) and geographical indications, See also Generalized System of Preferences 262­63, 266 (GSP) development dimension to, 266­68 grains and oilseed sector compared to other DOHA negotiations in context of, 37 commodities, 323, 326fA.5­A.6 relationship to Cartagena Protocol, 246, 248 import quota fill rates, 108 relationship to CBD, 264 insurance programs, 141 Section 5 on patents, 257­58 on multifunctionality, 169 TRQs. See tariff rate quotas oilseeds production, 156 tuna industry and preferential treatment, 311, patent costs, 258 311b15.7 possibility of reinstating direct payments, effect Tunisia and domestic support levels, 122 of, 142 Turkey producer support estimates (PSEs) in, 8 decoupled payments in, 138 tariff escalation and, 76 domestic support programs of, 19b1.3, 121 tariff peaks (mega-tariffs) and, 74 tariff rates of, 8 unilateral reduction of import duties, 39 transgenic crops in, 236 United Kingdom transparency of tariffs in, 70b4.1 Commission on Intellectual Property Rights' URAA negotiating position of, 216 recommendations on agriculture and value commitments and, 44t3.3, 53 genetic resources, 267, 267b13.1 wheat production, 154 patent costs, 258 UPOV. See International Union for the Protection United Nations of New Varieties of Plants See also Convention on Biological Diversity URAA. See Uruguay Round Agreement on Agricul- (CBD) ture Development Programme, 286 urban-rural poverty gap, 2, 4, 4t1.1 Environment Programme (UNEP), 247, Uruguay Round Agreement on Agriculture 250n20 (URAA), 26­27, 36­37, 119, 287­89 on special safeguard measures, 211 Agricultural Supporting Tables (AGST), 27b2.2, UNCTAD study on economic impact of trade 31 preferences, 307 Article 4: market access, 27­30, 103 United States See also market access agricultural support from taxpayers to farm Article 5: special safeguard (SSG), 30 operators and land owners, 157 See also Special Safeguard (SSG) provisions banking of unused export subsidies, 49 Article 6: domestic support commitments, bioengineered foods, review of, 236, 240 30­33 blue box payments used by, 130 See also domestic support commitments border protection, use of, 136 Article 8: export competition commitments, 33 conversion of subsidies into green box See also export competition policies amber box subsidies, 122 Article 9: export subsidy commitments, 33 blue box subsidies, 130 Article 10: prevention of circumvention of decoupled payments in, 138, 140 export subsidy commitments, 34 domestic support levels of, 37, 121, 122 Article 13: due restraint, 34­35 export subsidies spending, 43 Article 15: special and differential treatment, 35 Farm Bill (1995), 138 See also special and differential (S&D) treat- Farm Bill (1996), 19b1.3 ment Farm Bill (2002), 8, 120, 142 Article 16: Marrakesh Decision, 35 Federal Agricultural Improvement and Reform Article 17 and 18: Committee on Agriculture, Act of 1996, 138, 140, 145n10 36 386 Agriculture and the WTO Article 20: nontrade concerns, 179 welfare effects of liberalization, 156­57, 156t7.5, conflicts with, 26 158, 158t7.7 on decoupling policy, 139 wheat and coarse grains disappointing results of, 7­10, 120, 143, 180, border protection for, 136 186, 269 decomposition of real producer price, 201t10.3 documents, 27b2.2 export taxes on, 54 Dunkel draft, 217 fixed costs of U.S. farms, 140 history leading to, 23, 25­26, 216­17 price bands in Chile, 207, 208 implementation of Agreement on Agriculture, real price trends of, in Eastern Europe, 201 36­37 tariff rate quotas for, 83, 88, 93n11 legal position in GATT, 23, 25­26 trade liberalization and, 153­54, 153t7.1, Mid-Term Review of, 217 154t7.2, 156, 197 Modalities Document, 27b2.2, 103 WIPO. See World Intellectual Property Organiza- negotiations during, 24b2.1, 216­17 tion Peace Clause. See Peace Clause Working Group on Sanitary and Phytosanitary priority of, 26 Regulations, 217 reprint of, 349­66 World Bank Schedule of Commitments, 27b2.2, 40, 103 differentiation in treatment of developing Technical Barriers to Trade Agreement (TBT), countries, 286 219­20 food security, role of, 188 U.S. Trade Representative (USTR) and GSP pro- International Task Force on Commodity Price gram, 295, 296 Risk Management, 188 studies and reports vaccines from bioengineered foods, 236 on AGOA documenting of eligibility, 310 value commitments on intellectual property rights, 254, 256 countries using over 90 percent of value com- on Sub-Saharan African countries, 5 mitments, 50t3.7 on trade preferences, 307 percentage use of on Uruguay Round liberalization on external allocated to each commodity group, 47, 48, prices, 187 48t3.5 world export growth, 10­16, 16t1.6, 16t1.7 by commodity, 47, 47t3.4 World Intellectual Property Organization (WIPO), by country, 46t3.3, 47 257, 265 value front-loading, 51, 52t3.9 world prices Venezuela and trade liberalization, 194 AMS issues and, 130 volume commitments financial crises and, 201 countries using over 90 percent of volume com- removal of border protection and, 136, mitments, 51t3.8 137t6.11, 137t6.12 percentage use of total volume commitments removal of domestic support and, 136 allocated to each commodity group, 47, trade liberalization and, 197­200, 198t10.1 48, 49t3.6 domestic volatility and, 196 volume front-loading, 51, 52t3.10 price regimes, 194­95 World Trade Organization (WTO) wages biotechnology dispute over GM foods, 249 of agricultural laborers, 162 Committee on Regional Trade Agreements, 292 of off-farm labor, 163 Committee on Trade and Development (CTD), waiting costs and first-come, first-served, 112 270 water in the tariff, 82, 90t4.10, 93 dispute settlement relating to U.S. Foreign Sales by commodities, 70, 74f4.11 Corporations Tax, 34 by countries, 70, 74f4.10 import quota fill rates, 106­8, 109f5.1 defined, 93n4 distribution of, 108t5.5 Index 387 simple average fill rates, 107t5.4 intellectual property rights subject to, 253 See also TRIPS agreement market access by developing countries as goal of, 291 special and differential treatment provisions, 271 tariff rate quotas, 103­6 trade liberalization, 204­11 Trade Negotiations Committee (TNC) to supervise Doha Round, 38 TRIPS agreement and, 257 T R A D E A N D D E V E L O P M E N T S E R I E S D eveloping countries have a major stake in the outcome of trade negotiations conducted under the auspices of the World Trade Organization (WTO). Agriculture and the WTO: Creating aTrading System for Development explores the key issues and options in agricultural trade liberalization from the perspective of these developing countries. Leading experts in trade and agriculture from both developed and developing countries provide key research findings and policy analyses on a range of issues that includes market access, domestic support, export competition, quota administration methods, food security, biotechnology, intellectual property rights, and agricultural trade under the Uruguay Round Agreement on Agriculture. Material is covered in summary and in comprehensive detail with supporting data, a substantial bibliography, and listings of online resources. This book will be of interest to policymakers and analysts in the fields of development economics and commodities pricing and trade. THE WORLD BANK ISBN 0-8213-5485-X