e ' f X MSS;; * % s ts S f ;; g0W g g: ;SS f ff f t~~~ ,,~~ S ,_S. , % @,.>:.- .?- N l~ ~~~~~~~~~~~~~~~~~~~~~~~~~ 4 Global Jo a ~~. Economic Prospects and the Developing Countries 2001 Copyright C) 2001 by the International Bank for Reconstruction and Development/The World Bank 1818 H Street NW, Washington, DC 20433, USA All rights reserved Manufactured in the United States of America First printing December 2000 This publication has been compiled by the staff of the Development Prospects Group of the World Bank's Development Economics Vice Presidency. The World Bank does not accept responsibility for the accuracy or completeness of this publication. Any judgments expressed are those of World Bank staff or consultants and do not necessarily reflect the views of the Board of Executive Directors or the governments they represent. The material in this publication is copyrighted. The World Bank encourages dissemination of its work and will normally grant permission promptly. Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use is granted by the World Bank, provided that the appropriate fee is paid directly to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470. Please contact the Copyright Clearance Center before photocopying items. For permission to reprint individual articles or chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978-750-4470. All other queries on rights and licenses should be addressed to the Office of the Publisher, World Bank at the address above or faxed to 202-522-2422. ISBN 0-8213-4675-X ISSN 1014-8906 Library of Congress catalog card number: 91-6-440001 (serial) Contents Acknowledgments vii Summary ix Abbreviations, Acronyms, and Data Notes xiii Chapter 1 Prospects for Developing Countries and World Trade 1 Long-term growth in industrial countries is projected to be higher 4 World trade remains on a long-term high-growth path 11 Private capital flows remain volatile 18 Commodity prices exhibit divergent recoveries 21 Developing countries' recovery is unexpectedly rapid, and prospects for long-term growth have improved 25 Vulnerabilities are significant 32 Recent trends and prospects for poverty reduction 34 Notes 44 References 45 Annex 1 Membership of selected major regional integration agreements (RIAs) and dates of formation 47 Chapter 2 Trade Policies in the 1990s and the Poorest Countries 49 Reductions in barriers to trade 51 Trends in trade and economic growth 53 Weaknesses in domestic trade-related policies 59 Protection in industrial countries 65 Notes 71 References 73 Annex 2 Sample countries in various charts and tables 76 Chapter 3 Standards, Developing Countries, and the Global Trade System 81 The regulation of standards: setting the stage 82 Product standards and regulatory barriers to trade 83 Labor standards and trade sanctions 88 Environmental standards and trade 93 Notes 101 References 103 iii G LO B AL EC ONO MI C PRO SP EC TS Chapter 4 Electronic Commerce and Developing Countries 111 Emergence of electronic commerce 112 The digital divide 113 Effects on productivity in industrial and developing countries 116 Effects on international trade in developing countries 120 Effects on income distribution 122 Impediments to Internet use in developing countries and the role of policies 122 Challenges to regulatory regimes in developing countries 128 Notes 130 References 131 Annex 4 Firm interviews and website survey 134 Companies Participating in Survey of Alibaba B2B Website Users 135 Appendix 1 Regional Economic Prospects 137 Appendix 2 Global Commodity Price Prospects 159 Appendix 3 Global Economic Indicators 175 Technical Notes 190 Classification of Economies 191 Figures 1.1 Industrial production in developing regions 2 1.2 GDP growth for major industrial countries, 1998-2002 5 1.3 U.S. retail sales and the NASDAQ index 7 1.4 Total retail sales of durable goods and autos 8 1.5 IT (information technology) investment growth per employed person and productivity growth, 1980-99 9 1.6 Japanese corporate profits and private capital spending 9 1.7 German exports, foreign orders, and manufacturing output 11 1.8 Growth of GDP per employed person: United States and European Union 12 1.9 Trade versus GDP growth 12 1.10 East Asia-5 industrial production and import volume 13 1.11 Export volume and market growth, 1997-2000 14 1.12 GDP and export volume growth 15 1.13 Number of WTO notifications of regional integration agreements (RIAs) 16 1.14 Intra-RIA exports as a share of RIAs' total exports 17 1.15 Net capital flows to developing countries, 1985-2000 18 1.16 Sectoral breakout of bond financing by developing countries, January 1997-June 2000 19 1.17 FDI flows to developing countries, 1990s 20 1.18 Crude oil prices, January 1990-October 2000 21 1.19 Crude oil prices, 1960-2010 22 1.20 Divergent recoveries of commodity prices 23 1.21 Real commodity prices, 1900-2000 25 1.22 Developing regions' real GDP growth, 1999-2002 26 1.23 GDP per capita growth, 1990-2010 27 iv C O N T E N T S 1.24 Growth of real per capita GDP, developing countries as a group, 1963-2008 33 1.25 Growth of real per capita GDP, Latin America and the Caribbean, Sub-Saharan Africa, Middle East and North Africa, 1965-2000 34 1.26 Nonperforming loans of commercial banks in the East Asia-5 35 2.1 Average unweighted tariff rates by region 53 2.2 Merchandise export growth and GDP per capita growth in developing countries in the 1990s 55 2.3 Merchandise export and GDP per capita growth in poor developing countries in the 1990s 58 2.4 Real effective exchange rate volatility and growth in the 1990s 62 2.5 Real effective exchange rate behavior in selected poor countries, 1993-99 63 2.6 Imports of manufactures from developing countries as a percentage of apparent consumption 69 2.7 Share of developing countries in world trade 70 3.1 WTO enquiry point notification, by country group, 1995 and 1999 89 3.2 Number of notifications under the TBT agreement, 1995-2000 90 3.3 Number of notifications under the SPS agreement, 1995-2000 90 4.1 Estimates of electronic commerce in industrial countries, 1999-2000 113 4.2 Estimates of Internet access, 1990-2000 114 4.3 Regional Internet access 114 4.4 Access to telecommunications 115 4.5 Cost savings from electronic commerce 120 4.6 New customers gained from alibaba website 121 4.7 Increased sales reported because of alibaba website 122 4.8 Internet monthly access charge as a percentage of GDP per capita, 1998 123 4.9 Telephone mainlines per 100 inhabitants, developing countries, 1998 124 4.10 Developing country privatization in telecommunications, 1990-98 125 4.11 Information technology jobs unfilled because of skill shortages, 1998 126 4.12 Internet use in industrial countries by knowledge of English, 1998-99 127 4.13 Bond versus bank financing 129 Tables 1.1 Global conditions affecting growth in developing countries and world GDP growth 3 1.2 Intra- and extraregional trade 17 1.3 Current account effects for a sample of developing countries from a $10 increase in oil prices 23 1.4 Annual percentage change in nominal oil and non-oil commodity prices, 1981-2010 25 1.5 Growth of world GDP, 1998-2002 27 1.6 Growth of world GDP per capita, 1980s through 2010 28 1.7 Forecast assumptions: developing countries 29 1.8 Population living on less than $1 per day and head count index in developing countries, 1987, 1990, and 1998 36 1.9 Population living on less than $2 per day and head count index in developing countries, 1987, 1990, and 1998 37 1.10 Population estimates and projections, developing countries, 1998-2015 41 v G LO BAL E CON OM IC PROS PE CTS 1.11 Poverty in developing countries under scenarios of base case growth (scenario A); low case growth (scenario B); and 1990s average growth, 1990, 1998, 2015 42 1.12 Regional breakdown of number of people living on less than $1 per day and head count index in developing countries, under scenarios of base case growth (scenario A) and low case growth (scenario B), 1990, 1998, and 2015 42 1.13 Regional breakdown of number of people living on less than $2 per day and head count index in developing countries, under scenarios of base case growth (scenario A) and low case growth (scenario B), 1990, 1998, and 2015 43 2.1 Standard deviation of tariff rates 53 2.2 Frequency of total core nontariff measures for developing countries, 1989-98 53 2.3 Countries imposing restrictions on payments for current account transactions 54 2.4 Average black market premium 54 2.5 The international environment 55 2.6 GDP and merchandise export growth rates 55 2.7 GDP, services, and merchandise export growth rates 57 2.8 Decomposition of merchandise export growth for the sample countries 57 2.9 Growth rates by income level for the sample countries 59 2.10 Tariffs in selected African countries 64 2.11 Developing-country exports to Quad countries facing tariffs of more than 50 percent 67 2.12 Average tariff rates by importing and exporting region 67 2.13 Producer support estimates for OECD countries 68 3.1 Summary of economywide studies assessing the impacts of trade liberalization on pollution 97 3.2 Evidence on international competitiveness and environmental regulation 98 4.1 Future Internet access speeds 115 Boxes 1.1 U.S. labor productivity and information technology 6 1.2 North-South regional arrangements 16 1.3 Trends in inequality 38 2.1 Openness and growth-evidence, old and new 52 2.2 Trends in volatility 56 2.3 Economic factors contributing to conflict 58 2.4 Exchange rate overvaluation in the CFA countries 61 2.5 The integrated framework for least-developed countries 66 2.6 Food processing 70 3.1 Mutual recognition agreements 88 3.2 The Trade-Related Intellectual Property Agreement (TRIPs) and developing countries 94 3.3 Evidence on the "race to the bottom" 99 4.1 Electronic data interchange (EDI) systems 117 4.2 The Internet and primary commodity exporters 119 ,V Acknowledgments T his report was prepared by the Development Prospects Group, and drew from resources throughout the Development Economics Vice Presidency, the Poverty Reduction Board, and World Bank operational regions. The principal author of the report was William Shaw, with direction by Uri Dadush. The chapter authors were Hans Timmer (chapter 1), Ataman Aksoy (chapter 2), Dominique van der Mensbrugghe (chapter 3), and William Shaw (chapter 4). The report was prepared under the general direction of Jo Ritzen and Nicholas Stern. The report drew on inputs by other staff of the Development Economics Vice Presidency and from throughout the Bank. Ibrahim Al-Ghelaiqah, Caroline Farah, Himmat Kalsi, Robert Key- fitz, Annette I. De Kleine, Robert Lynn, Fernando Martel Garcia, Dominique van der Mens- brugghe, Shoko Negishi, and Mick Riordan contributed to the analysis of global economic trends and prospects in chapter 1. Tamar Manuelyan Atinc, Shaohua Chen, Valerie Kozel, Giovanna Prennushi, Martin Ravallion, and Aristomene Varoudakis contributed to the discussion of pov- erty. Betty Dow, Faezeh Fouraton, Carol Gabyzon, Theresa Goldberg, Dorsati Madani, Donald Mitchell, Ashish Narain, Francis Ng, and Konstantin Senyut contributed to chapter 2. Constan- tine Michalopoulos and John S. Wilson contributed to chapter 3. Carol Gabyzon, Somik Lall, Ashish Narain, Andrew Sunil Rajkumar, and David Wheeler contributed to chapter 4. And John Baffes, Betty Dow, Donald Mitchell, and Shane Streifel contributed to the analysis of commodity prices in chapter 1 and the annex. Many others from inside and outside the Bank provided inputs, comments, guidance, and support at various stages of the report's publication. John Beghin, David Rohland-Holst, and Matthew Slaughter wrote background papers on trade issues. Henry Ergas and lain Little wrote a background paper on electronic commerce. Gary Hufbauer, Arvind Panagariya, Francisco Rodriguez, and Alan Winters served as outside reviewers. Carlos Braga, Shanta Devarajan, Richard Newfarmer, and Gene Tidrik were discussants at the Bankwide review. We would par- ticularly like to thank Gordon Betcherman, Milan Brahmbhatt, Sara Calvo, Richard Eglin, David Ellerman, Michael Finger, Carsten Fink, Andrea Goldstein, Bernard Hoekman, Albert Keidel, loannis Kessides, Michael Klein, Amy Luinstra, Will Martin, Aaditya Mattoo, Marcelo Olarreaga, Gary Pursell, David Tarr, and Edith Wilson for their helpful comments. The Development Data Group contributed to the appendix. Betty Sun served as the External Affairs task manager, Robert King managed dissemination from the Development Prospects Group, and Phil Hay managed media arrangements. Sarah Crowe served as the principal assistant to the team and Katherine Rollins assisted with chapter 1. Book design, editing, and production were directed and managed by the Production Services Unit of the World Bank's Office of the Publisher. vii Summary T ECHNOLOGICAL INNOVATIONS AND THE growth. The same applies in developing coun- dismantling of trade barriers over the tries, where liberalization of markets, more past decade have contributed to an stable macroeconomic policies, and techno- acceleration of growth in global trade. This logical change have promoted integration. In- acceleration has been associated with faster dicators of human capital, including school growth in developing countries as a group. enrollment and literacy rates, show broad im- However, many of the poorest countries have provement across most developing regions. not kept pace. This year's Global Economic However, cyclical and structural aspects Prospects focuses on international trade and of the current boom have increased imbal- discusses policies that are required if devel- ances and tensions in the global economy. Eas- oping countries are to benefit from global ier monetary policy in the United States and integration. increased fiscal stimulus in Japan boosted growth from the depths of the financial crisis, but these policies also increased the already Prospects for developing countries large U.S. current account deficit (4.5 percent and world trade of GDP) and Japanese government debt (115 T he global economy is likely approaching a percent of GDP). The strong global recovery T cyclical high in 2000, boosted by a further of 1999-2000, coupled with the sharp reduc- acceleration of growth in the United States, the tion in OPEC (Organization of Petroleum Ex- recovery in Europe and Japan, and the sharp porting Countries) supply, caused a surge in rebound in countries affected by the global fi- oil prices. Structural reforms and rapid tech- nancial crisis. World trade volumes are likely nological change have also generated political to increase by 12.5 percent, the highest rate of tensions. The fast pace of global economic growth since before the first oil shock of the integration has accentuated competition and 1970s. A moderation of growth in the crisis increased uncertainty, particularly for firms countries and slower consumption growth in in declining industries and their workers. In- the United States are likely to lead to a decel- equality both among and within countries ap- eration of output growth over the next year. pears to have risen, in part the result of tech- The apparent shift upward in trend pro- nological progress. ductivity growth in the United States, increased A low-case scenario assumes a less favor- labor market flexibility and product market able resolution of these imbalances and ten- competition in Europe, and steps toward fi- sions, marked by continued high oil prices and nancial and corporate restructuring in Japan a reversal of international investment flows have improved the prospects for long-term from the United States. The resulting reces- ix G LO B AL EC ONO MI C PRO SP EC TS sion, coming on the heels of the global finan- ularly affected the poorest countries, because cial crisis, may feed "reform fatigue" and thus a host of other domestic policy and institu- lower developing countries' long-term growth tional weaknesses inhibit their diversification potential. into less restricted sectors. Trade policies in the 1990s Standards, developing countries, and the poorest countries and the global trade system O ver the past decade, developing countries f roduct standards (rules governing the char- 0 reduced the level and dispersion of tariffs, 1 acteristics of goods that are generally im- dismantled nontariff trade barriers, and in- posed to protect health and safety) are critical creased reliance on market forces to allocate to the effective functioning of markets and foreign exchange. These policies, coupled with provide important support to the trade system. other market reforms, were associated with an However, many developing countries (particu- acceleration of output and export growth, ex- larly the poorest ones) lack the technological cept for countries that were affected by conflict and financial resources to develop product or the breakup of the Soviet Union. The per standards effectively, meet industrial countries' capita income of small, low-income countries import requirements, and bring disputes when (thus excluding China and India) declined standards are used to discriminate against their during the 1990s, but growth averaged 1 per- exports. cent a year if countries involved in conflict Adherence to labor and environmental stan- and countries in transition are excluded. This dards (for example, the right to form unions and represents a significant acceleration compared limits on pollution) is critical to economic ef- with the 1980s but is still well below the aver- ficiency and welfare. However, pressures to use age of middle-income countries. trade sanctions to support labor and environ- Weaknesses in trade-related policies con- mental standards threaten to restrict develop- tinued to impede growth in many of the poor- ing countries' access to international markets est countries. Appreciated real exchange rates while doing little to improve welfare. Labor and high real exchange rate volatility have and environmental standards generally improve often been associated with a muted export re- as countries develop, but low labor and envi- sponse to trade liberalization; per capita in- ronmental standards are not usually a signifi- come growth was significantly faster in poor cant source of competitive advantage. The im- countries with relatively stable real exchange position of trade sanctions is vulnerable to rates. The absence of effective duty exemp- capture by protectionist interests and hurts tion/drawback programs, coupled with fiscal workers by reducing demand for the goods reliance on tariffs on intermediate and capital they produce. Even if the threat of sanctions goods, has increased costs for exporters. Fi- improves conditions for some workers, aver- nally, weak export infrastructure, inadequate age working conditions in the economy are ancillary export services, and high transport unlikely to improve. Similarly, empirical stud- costs-often in part the result of policy short- ies show that imposing trade sanctions on ex- comings-have left many countries (particu- porters can cause considerable output losses larly the landlocked ones) at a competitive dis- while doing little to reduce pollution. advantage on international markets. High trade barriers imposed by industrial countries on agriculture and processed food Electronic commerce and imports, along with agricultural subsidies, developing countries have contributed to the decline in developing mhe Internet will boost efficiency and en- countries' share of world trade in these com- T hance market integration, particularly in modities. These trade distortions have partic- developing countries that are most disadvan- .N SUMMARY taged by poor access to information. The Inter- that lack the reputation to bid on the new on- net will raise productivity through increased line exchanges or the technology to interact ef- procurement system efficiency, strengthened ficiently with more sophisticated firms could inventory control, lowered retail transaction see reduced demand. While the growing use of costs, and elimination or transformation of in- cell phones and other technologies should in- termediaries. The cost of reaching industrial crease Internet access rapidly over the next 10 country markets will fall, generating large gains years, access is likely to remain limited in per from trade. Developing-country firms that sell capita terms, especially in the poorest countries. labor-intensive, differentiated products (for ex- Taking advantage of electronic commerce ample, crafts, software, and business services- requires an open economy to promote compe- particularly services involving the remote pro- tition and diffusion of Internet technologies; cessing of routine information) will experience improved international coordination (for ex- increased demand. Developing-country firms ample, in confronting challenges to domestic also will benefit from the opportunity to leap- tax and financial systems); and efficient so- frog to the most advanced technologies. cial and infrastructure services, in particular a Nevertheless, Internet access is grossly un- competitive telecommunications sector and a equal across countries, and the Internet also well-educated labor force. The importance of brings increased danger of economic marginal- network effects and first-mover advantages ization to countries that cannot access it effec- emphasizes the importance of government sup- tively. For example, developing-country firms port for achieving these goals. xi Abbreviations, Acronyms, and Data Notes APEC Asia Pacific Economic Cooperation CAP Common Agricultural Policy CEE Central and Eastern Europe (Central and Eastern European countries are CEECs) CFA Communaute Financiere Africaine CIS Commonwealth of Independent States EBRD European Bank for Reconstruction and Development ECA Europe and Central Asia EMU European Monetary Union EU European Union FAO Food and Agriculture Organization of the United Nations FDI Foreign direct investment GATT General Agreement on Tariffs and Trade GDP Gross domestic product GSP Generalized System of Preferences HIPC Heavily indebted poor countries HIV/AIDS Human immunodeficiency virus/acquired immune deficiency syndrome ILO International Labour Organisation IT Information technology LAC Latin America and the Caribbean LIBOR London interbank offered rate LDC Least-developed countries M&A Mergers and acquisitions MNA Middle East and North Africa Mercosur Latin America Southern Cone trade bloc (Argentina, Brazil, Paraguay, and Uruguay) MFN Most favored nation xiii G LO B AL EC ONO MI C PRO SP EC TS MFP Multifactor productivity MRA Mutual recognition agreement MUV Manufactures unit value index NAFTA North American Free Trade Agreement NASDAQ National Association of Securities Dealers Automated Quotation NIE Newly industrializing economy NTBs Nontariff barriers OECD Organisation for Economic Co-operation and Development OPEC Organization of Petroleum Exporting Countries PSE Producer support estimate RIA Regional integration agreement saar Seasonally adjusted annualized rate TBT Technical barriers to trade TRIPs Trade-related intellectual property rights SPS Sanitary and Phytosanitary Standards UNAIDS Joint United Nations Programme on HIV/AIDS WTO World Trade Organization Data notes The "classification of economies" tables at The following norms are used throughout: the end of this volume classify economies by income, region, export category, and indebt- * Billion is 1,000 million. edness. Unless otherwise indicated, the term * All dollar figures are U.S. dollars. "developing countries" as used in this vol- * In general, data for periods through 1998 ume covers all low- and middle-income are actual, data for 1999 are estimated, countries, including the transition economies. and data for 2000 onward are projected. xiv Prospects for Developing Countries and World Trade W rTORLD ECONOMIC ACTIVITY DURING on high oil revenues and more fragile than in 2000 is proceeding at the fastest pace East Asia. With oil prices expected to ease in in over a decade, with developing- the medium term and the effect of the 1998 country output growth expected to exceed 5 ruble devaluation wearing off, the Russian percent. World trade volumes are expected to Federation's current GDP growth of about 7.2 rise by a record 12.5 percent in the year. Al- percent is expected to slow significantly over though oil prices have surged by more than the medium term. Sub-Saharan Africa has ex- 50 percent, inflation in both industrial and de- perienced a less uniform recovery, with oil veloping countries continues, thus far, to be rel- exporters gaining and commodity-dependent atively subdued. But developments in oil mar- oil-importing nations suffering large terms- kets remain a major uncertainty in the outlook, of-trade losses. These synchronous recoveries as do the sustainability of the remarkable non- have carried developing-country growth to a inflationary U.S. expansion and the general peak of 5.3 percent in 2000-0.7 percent- fragility of financial systems in East Asia. This age points faster than projected nine months chapter reviews the cyclical and structural fac- ago in the World Bank's Global Development tors responsible for the robust economic ex- Finance 2000-with a slight slowing to 5.0 pansion and discusses the major challenges and percent expected next year (table 1.1). Growth risks ahead, in both the short and the medium in the industrial countries may also be near- terms. The main conclusions are: ing a turning point; it is expected to slow from this year's rapid 3.7 percent pace to 2.9 The world economy recovered remarkably percent in 2001. Moderation of consumer well and is likely approaching a cyclical demand in the United States, following in- high in 2000 terest rate increases and stock market de- Many of the developing countries that experi- clines, is the principal factor behind this mod- enced a sharp rebound after the 1997-98 re- est deceleration. cession appear to have reached cyclical peaks, The current double-digit growth of world with the five East Asian countries hit hardest trade, the strongest since before the first oil by the financial crisis the clearest example of shock of the early 1970s, is clearly a cyclical this development (figure 1.1). The strength of phenomenon tied to robust world activity lev- the recovery in Latin America has been im- els. During the upswing, as inventories were re- pressive, but momentum appeared to be wan- plenished and investments accelerated, trade ing in the second half of the year. And the re- expanded much faster than the economy as a bound in the Russian Federation has also been whole. Once stocks of durable goods and capi- unexpectedly strong, though largely dependent tal goods have adjusted, growth rates of trade 1 G LO B AL EC ON 0 M I C P R O S P E C T S Figure 1.1 Industrial production in developing regions Three-month moving average, percent year over year 1 8 East Asia All developing regions 12 C GEE 6 t ~~~LAC , I/ 0 -6 Jan. 1997 July 1997 Jan. 1998 July 1998 Jan. 1999 July 1999 Jan. 2000 July 2000 Note: I atest data for East Asia are from Jan, for (FE (Central and Eastern Europe) arc fromn june, for decloping regionls are from June, and for LAC (Iatiri Anierica and the Caribbean) are from Jlnu. Source: Darastre.im and World Bank staff estimiiates. should moderate to around 8 percent, which is improved business confidence. And Japan ap- still a high level by historical standards. pears to be emerging from a long period of sluggish growth. This follows the initiation of Foundations for longer-term growth have serious efforts toward financial and corporate improved in many industrial and restructuring, although a lack of self-sustaining developing regions . . . effective demand, especially from private con- Industrial countries have been undergoing a sumers, is still a danger. period of accelerated transformation, restruc- Liberalization, accompanying policy mea- turing, and adjustment that is now starting to sures, and technological change in many devel- pay off. The United States appears to have cre- oping countries have led to a spectacular in- ated an institutional and policy environment crease in openness during the 1990s. Foreign that supports the adoption of new informa- direct investment (FDI) flows into developing tion and communications technologies at a countries rose from 0.5 percent of developing rapid pace, contributing to a substantial accel- countries' GDP in 1990 to 2.7 percent at the eration in productivity growth. Most Euro- end of the decade. Despite the financial crisis, pean countries have made some progress in exports of goods and services from developing rendering labor markets more flexible and ex- countries increased by 10 percent a year during posing product and service markets to greater the 1990s, contrasted with less than 4 percent competition; these processes have been facili- during the 1980s. Competition from both do- tated by regional integration, including, most mestic and foreign sources has increased in this recently, the introduction of a single currency. more open environment, and macroeconomic The recent decline in Euro Area unemploy- policies have become more prudent, keeping ment rates, and the more than doubled value inflation low and reducing some of the larger of merger and acquisitions (M&A) activities fiscal deficits. And indicators of human capital, and corporate bond issues in 1999, offers some including school enrollment and illiteracy indication of accelerated restructuring and rates, have shown broad improvement across PROS PE C TS F OR DEVE LOP ING CO UN T RI ES AND W O RL D TRA DE Table 1.1 Global conditions affecting growth in developing countries and world GDP growth (percentage change from previous year, except interest rates and oil price) Current Cufrent March 2000 Estmate Forecasts Forecasts 1999 2000 2001 2002 2W0 2001 2002 Global Conditions World trade (volume) 5.8 12.5 8.0 6.8 8.3 6.9 6.5 Inflation (consumer prices) G-7 OECD countriesab 1.2 2.0 1.9 1.9 I.S 13 2.0 United States 2.2 3.4 3.0 2.8 2.7 2.5 2.6 Commodity prices (nominal $) Commodity prices, except oil ($) -11.2 -0.8 3.4 4.9 5.6 3.9 3.3 Oil price ($, weighted average), $/bbl 18.1 28.0 25.0 21.0 23.0 19.0 18.0 Oil price, Percent Change 38.3 55.0 -10.7 -16.0 . 27.3 -17.4 -5.3 Manufactures export unit value ($)c -2.7 -2.3 3.6 3.7 2.5 2.5 2.6 Interest rates LIBOR, 6 months (US$, percent per year) 5.5 6.7 6.8 6.2 6.5 6.5 5.5 EURIBOR, 6 months (Euro, percent per year) 3.0 4.5 5.0 4.6 World GDP growth 2.8 4.1 3.4 3.2 3.5 3.1 3.1 High-income countries 2.7 3.8 3.0 2.8 3.2 2.7 2.6 OECD countries 2.7 3.7 2.9 2.7 3.0 2.6 2. United States 4.2 5.1 3.2 2.9 3.8 2.7 2.8 Japan 0.3 2.0 2.1 2.2 i.2 1.4 1.6 Euro Area 2.4 3.4 3.2 2.8 3.4 3.1 2.8 Non-OECD countries 4.2 6.3 5.1 5.1 4.6 4.8 5.1 Developing countries 3.2 5.3 5.0 4.8 4.6 4.8 4.8 East Asia and Pacific 6.9 7.2 6.4 6.0 6.6 6.3 6,1 Europe and Central Asia (ECA) 1.0 5.2 4.3 3.9 I2S 3.4 3.6 Latin America and the Caribbean 0.1 4.0 4.1 4.3 3.6 3.8 4.4 Middle East and North Africa 2.2 3.1 3.8 3.6 3.5 3.6 3.6 South Asia 5.7 6.0 5.5 5.5 5.9 5.8 5.5 Sub-Saharan Africa 2.1 2.7 3.4 3.7 3,2 3.7 3.8 Memorandum items East Asia-5 countriesd 6.7 6.9 5.5 5.1 5.7 5A 5.1 Transition countries of ECA 2.5 5.0 4.2 3.7 2.1 3.0 3.3 Developing countries Excluding the transition countries 3.3 5.3 5.1 5.0 5.0 5.0 5.1 Excluding China and India 2.2 4.7 4.4 4.3 3.8 4.0 4.2 ... Not available. a. Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States. b. In local currency, aggregated using 1995 GDP weights. c. Unit value index of manufactures exports from G-5 to developing countries, expressed in U.S. dollars. d. Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand. Source: Development Prospects Group, baseline, October 2000; and GDF projections of March 2000. most developing regions. With these structural . . . but these favorable cyclical and changes, many countries in Latin America, structural conditions contain built-in Central Europe, and Sub-Saharan Africa ap- tensions pear to have considerably improved their Developments during the global financial crisis growth potential. Assuming continued corpo- sowed the seeds for some severe imbalances rate and financial restructuring to deal with that have remained or become evident during the debt overhang left by the crisis, countries in the current boom. The adoption of an easier East Asia should achieve high rates of growth monetary policy in the United States to avert over the next decade. a global recession in late 1998 contributed to 3 G LO B AL EC ONO MI C PRO SP EC TS an acceleration of U.S. demand growth and a Asia and Latin America more severely. The widening of the current account deficit, which sharp growth slowdown that would result, is likely to breach 4.5 percent of GDP in coming on the heels of the global financial 2000. Fiscal stimulus in Japan, while helping crisis, may feed "reform fatigue" in developing to sustain demand during the worst part of the countries, resulting in low growth. The low- crisis, has further increased the huge burden case scenario below illustrates the importance of government debt to some 115 percent of of reducing short-run imbalances to safeguard GDP. Nonperforming loans in the Asian crisis the long-term prospects for growth. countries reached 30 to 50 percent of GDP This chapter is organized as follows. First and have been declining only gradually. The the cyclical environment and the long-term financial vulnerabilities translated into an av- growth potential in the industrial countries are erage decline of more than 30 percent in the discussed, and a review of recent develop- equity markets in these countries between Jan- ments and prospects for world trade and fi- uary and November. The strong global recov- nancial flows to developing countries follows. ery of 1999-2000, coupled with the sharp re- The section on commodity prices focuses on duction in OPEC supply (following the plunge the sharp hike in oil prices, one of the major in oil prices to $10 per barrel in 1998), caused threats to the current outlook. And the fol- a surge in oil prices. lowing two sections summarize the conse- Structural reforms and rapid technological quences of these trends for developing regions change have also generated political tensions. in the short and longer terms, including elabo- The fast pace of global economic integration ration of a low-case scenario. Finally, the con- has accentuated competition and increased un- sequences for poverty alleviation are explored. certainty, particularly for firms in declining industries and their workers. Inequality, both among and within countries, and in part tied Long-term growth in industrial to technological change, appears to have in- countries is projected to be higher creased. A backlash against globalization could C- rowth in the high-income Organisation result in a slower pace of reforms, especially if Vfor Economic Co-operation and Devel- the current expansionary phase is broken. opment (OECD) countries may average 3.7 percent in 2000 (the fastest growth recorded These tensions could reduce growth in in over a decade), driven by a sharp accelera- both the short and longer terms tion of exports, strong carryover effects of The baseline scenario assumes a soft landing the U.S. consumer boom of late 1999 to mid- for the U.S. economy, smooth private sector 2000, broadening and strengthening of eco- adjustment, and prudent policy reactions to nomic activity across the Euro Area, and a the current oil price shock. However, a less pickup in Japanese private and public invest- favorable resolution of the tensions now af- ment spending. Growth rates in the three fecting the global economy is possible. Supply major blocs are expected to move toward con- interruptions or unexpectedly high demand vergence, yielding OECD growth of 2.9 per- could lead to a sharper and more protracted cent in 2001 and 2.7 percent in 2002 (figure spike in oil prices, while uncertainty about fu- 1.2). But this outlook is subject to important ture oil prices could severely affect business risks, including the potential for a hard land- and consumer confidence. These adverse reac- ing in the United States because of investor tions could be reinforced by a tightening of concern over the burgeoning current account monetary policies. A reversal of international deficit, higher inflation and the likelihood of investment flows to the United States, triggered monetary tightening if the present spike in oil by increasing current account deficits and a prices is sustained, and a disruption of the change in sentiment in the stock market, could Japanese recovery because of fragile financial accentuate the global downturn affecting East conditions. PROS PE CT S FOR D EVEIO PING CO UN TRI ES AND WORLD TRADE Figure 1.2 GDP growth for major industrial countries, 1998-2002 Percent 6 * United States O Euro Area 0 Japan 3 0 -3 1998 1999 2000 2001 2002 Note: Growth for 2001-02 is estimated. Source: Darastream and World Bank staff estimates. Structural transformation may lead to in information and communications technol- stronger long-term growth ogy (box 1.1). Nevertheless, cyclical factors Technology-driven productivity growth in the have played an important role in the boom. United States, market reforms and adjustment Increasing job opportunities, rising incomes to a common currency in the European Union and wealth, and strong corporate profits have (EU), and corporate and financial restructur- boosted consumer and business optimism to ing and deregulation in Japan offer the poten- record levels and encouraged rapid growth tial for rapid growth in the long run. However, in expenditure. Equity price movements have important challenges remain in reaping the exerted a large impact on consumer behavior benefits of these new technologies, expanding (figure 1.3). Over 1995-98, household net the EU to the east, and adjusting to slower wealth grew each year by some 30 percentage population growth. Moreover, the huge U.S. points more than disposable incomes.1 Partly external deficit and Japan's rising government as a result, the personal saving rate dropped debt will continue to pose major risks. As- from 7.6 percent in the first half of the 1990s suming effective policies to confront these to negative territory (-0.2 percent) in the third challenges, growth for the industrial countries quarter of 2000. over 2003-10 has been upgraded from earlier Consumer price inflation has risen by 1.5 forecasts to 2.8 percent. percentage points over the last year, partly in response to the 50 percent rise in oil price. Cyclical and structural forces are shaping Compensation pressures are rising, as the Em- the path of U.S. expansion ployment Cost Index increased by 4.4 percent United States. The remarkable performance of during the first three quarters of the year. the U.S. economy since the mid-1990s has its However, the pass-through of rising input roots in prudent monetary, fiscal, and regula- costs to core inflation has been limited, in tory policies that encouraged private sector large measure because of strong productivity activity. It also stems from the availability of growth (4.7 percent through the third quarter venture capital and a flexible labor force that from a year ago)-suppressing any increase facilitated productivity-enhancing innovations in unit labor costs-and the appreciation of 5 G LO B AL EC ONO MI C PRO SP EC TS o 1 U.a pd iiinformtation ioXispr has i- the underlying structureo- n dc-.ty While the,tastiso ftpourvygis ~rtl nteown ut od to the service sector has n i- i e data5 it is clear that th de-af mtained strong, maki it e t t _ t t gains in, IT prductvt ileaiueo;nfht ai - .- .. .. ..S , . S .- - - ,. . 4 percent. b. Ihdeed, Gardon (_ -vuo i e -= fts- he argues tha tharevy ialo Pe i u- 4 e rir productivity wil not he sustakidet c. For a. number Of seri idsre ffor exapleeduction), theaaarewthipg that- real out and price changes mov to -thtsoay ta rs noloro oth ol . Figure 1.3 U.S. retail sales and the NASDAQ index Sales: percentage change over three months ago, seasonally adjusted annualized rate: NASDAQ: percentage change over three months ago Sales NASDAQ 18- 60 Retail sales 50 12 40 30 6 20 10 0 0 -10 NASDAQ -6 -20 Jan. 1998 June 1998 Nov. 1998 April 1999 Sept. 1999 Feb.2000 July 2000 Dec.2000 Souryce: U.S. Department of Commerce and I)atastream. the dollar on the heels of massive capital momentum of consumer demand growth over inflows.2 the course of the first half of the year, with The Federal Reserve's increase in the Fed- interest-sensitive sectors such as automobiles eral Funds rate (by 175 basis points in six steps and housing being particularly affected (figure from June 1999 to May 2000) reduced the 1.4). The slowing of consumption growth was 7 G LO B AL EC ONO MI C PRO SP EC TS Figure 1.4 Total retail sales, durable goods and autos Retail sales and autos: three-month/three-month, percentage change, seasonally adjusted annualized rate 20 15 \ ~~~~~~~~~~~~~~~~~~~~Autos Total retail sales d yr -d t Durable goods sale asinly d Jan. 1999 Apr. 1999 July 1999 Oct.1999 Jan.2000 Apr. 2000 July 2000 Oct. 2000 Source: U.S. Census Bureau; Datastream; and WVorld Bank staff calculations. short-lived, however, and third-quarter data plus, which would tend to increase the current revealed a rebound in spending to 4.5 percent account deficit yet further. Current financial growth. Nonetheless, GDP advanced at a 2.7 tensions in the high-yield sectors may be a first percent pace in the third quarter representing a sign that financing of large U.S. private debt is dramatic slowing to about one-half the rate of becoming increasingly difficult. the previous year. A sharp decline in business Strong productivity growth is likely to con- avxed investment was a major factor in the tinue over the medium term (box 1.1), as the slowdown, as an unwinding of the high-tech rapid growth in IT investment (which has spending boom appears to have begun. risen over the 1990s at four times the rate of Still, prospects remain favorable for a soft other private capital-spending components) landing and we expect that GDP growth will despite cyclical up- and downturns is likely to average 5.1 percent in 20003 and about 3 per- continue at high rates on a secular basis (figure cent on average in 2001-02. The consensus 1.5). With demographic factors likely to slow view of financial analysts is that the Federal growth of the labor force to rates below 1 per- Reserve is likely to raise interest rates further cent per year over the coming decade,4 long- in 2001 against the background of still rapid term potential growth could be as high as 3 or domestic demand growth, high oil prices, and 3.5 percent, without risk of significant infla- continued wage pressures. With a slackening tionary pressure. But achieving this potential in the pace of economic activity over the course growth will present policy challenges, as cor- of 2001, policy as well as long-term interest rection of the persistent external deficit will rates should ease moderately in 2002. The un- require extended periods of low import de- derlying risk of a harder landing remains, how- mand, a fall in the value of the dollar, or both. ever, since domestic savings are not expected to recover and the current account deficit is Japan emerges from recession, but its likely to register $450 billion to $475 billion in financial underpinnings are fragile 2000-02 (4.5 percent of GDP). The possibility Japan. GDP rose by 10.3 percent (seasonally of tax cuts following the November elections adjusted annualized rate, or saar) in the first suggests a reduction of the public sector sur- quarter of 2000 and 4.2 percent in the second, 8 PROS PE CTS FOR DEvE LOPING C OU N TRI ES AND WORLD TRADE Figure 1.5 IT (information technology) investment growth per employed person and productivity growth, 1980-99 Percentage change Percent (/T/E) Percent (VA/hr) 25 -5 IT investment 20 4 3 15 2 10~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 5 - Productivity growth 0 (right side) 0 I I I I I I I I --1 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 Source: U.S. Bureau of Labor Statistics; U.S. Department of Commerce; Bureau of Economic Analysis. as public investment increased and a sharp re- sumer confidence the boost necessary for the covery in profits supported private capital recovery to maintain momentum. The Bank of spending (figure 1.6). There are now signs that Japan has abandoned its "zero" policy interest household demand is rising (after a decade of stance, suggesting that the pickup in activity is stagnation or decline), grounded in improved sufficiently grounded to withstand the 25-basis labor market conditions. This could give con- point rise. With evidence that industrial corpo- Figure 1.6 Japanese corporate profits and private capital spending Percentage change year over year 40 Operating proi 20 ~~~~~~Private capital expenditures -20 -40 Q1 1997 Q3 1997 01 1998 03 1998 01 1999 Q3 1999 Q1 2000 Q3 2000 Note: All measures are in nominal yen. Source: Datastream. 9 G LO B AL EC ONO MI C PRO SP EC TS rate recovery is more advanced than antici- Growth solidifies in the Euro Area, but pated, that public works-related investment is weak currency is underpinning now filtering through the economy, that a inflationary pressures nascent upturn in consumer demand could Euro Area. During the second half of 1999, consolidate with rising incomes, and that pros- improvements in world activity, a competitive pects for Japanese exports remain favorable, exchange rate, and buoyant domestic demand we have upgraded projections for GDP growth delivered a rebound for the Euro Area from in 2000 to 2 percent, and to a range of 2-2.2 the crises of 1998, with GDP growth averag- percent over 2001-02. ing 3.8 percent on an annualized basis. This Recent efforts in the corporate and financial pace of growth continued unabated in the first restructuring required for long-term recovery quarter of 2000, slowing to an annualized 3.5 from a decade of slow growth show progress. percent in the second. A key to the recovery Announcements of corporate restructuring was the momentum underlying export growth, plans (mostly by larger firms) surged during which continued to build during the first half 1999 and 2000, and many of these plans con- of 2000 toward rates of 10 to 15 percent, with tained commitments to refocus on core activi- thickening export order books and rising man- ties, improve long-term profitability, strengthen ufacturing production (figure 1.7 highlights financial control, and forge links with foreign the case of Germany). partners. The government is drafting more The European Commission's surveys of workable insolvency laws to help facilitate consumer and business confidence reached labor mobility and the scrapping of excess ca- record highs during the first half of 2000, with pacity, is providing loans and credit insurance retail sales rising 3.5 percent in the year to to startups and venture firms, and is easing the June. Notable after several years of stagnant process for mergers and acquisitions and em- employment growth has been the creation of ployee buyouts. Successful restructuring over over one million jobs in 1999, bringing down the next decade could generate significant gains Euro Area unemployment to 9 percent from in productivity, which together with the ex- 11 percent in 1998. The economic expansion pected decline in the labor force would imply has also become more broadly based across output growth modestly above 2 percent per the region, although Italy remains weak in year. part because of tightened fiscal policies in Nevertheless, critical challenges remain. the run-up to the European Monetary Union Uneven corporate restructuring continues to (EMU). Preliminary figures for the third quar- pose a threat to the near-term recovery. The ter point to a slight slowing and stabilization number of business failures soared to a record of activity, partly as a consequence of the oil in the first seven months of 2000, and debt as- related terms-of-trade shock and rising inter- sociated with the failed firms has skyrocketed. est rates. Higher oil prices and the weak euro Events triggered by the still fragile state of have boosted the harmonized index of con- several financial institutions and nonmanufac- sumer prices by 2.8 percent in the year to turing firms could impair consumer and busi- September, well above the European Central ness confidence, as evidenced by the bank- Bank's (ECB) target of 2 percent year-on-year ruptcy of the Sogo department stores (carrying growth. In response, the ECB has tightened $17 billion in debt) after the withdrawal of a monetary policy since November 1999, grad- proposed government bailout. And Japan's ually raising the repurchase rate by 225 basis general government gross liabilities will reach points to 4.75 percent in October. Further 115 percent of GDP in 2000; massive expen- hikes in policy rates appear likely in order to diture compression and an overhaul of the tax prevent a translation of high current inflation system will be required to address the debt into higher price and wage expectations-or overhang in the medium term. so-called second-round effects. 10 PROS P ECT S F OR D EVELO PING CO UN T RIES AND WO RL D TRADE Figure 1.7 German exports, foreign orders, and manufacturing output Percentage change, three-month moving average, year overyear 25 Export orders 20 1 5 Exports (volume) 10 5 < ~~~~~Manufacturing \ ~~~~~~~~~~~~~~~~output 0 -5 Jan. 1998 May 1998 Sept. 1998 Jan. 1999 May 1999 Sept. 1999 Jan. 2000 May 2000 Sept. 2000 Source: Datastrearn. Recovery in 2000 will likely result in Euro the possibility of "New Economy" contagion. Area growth of 3.4 percent, up from 2.4 per- EMU comes on the heels of increased compe- cent in 1999. Looking forward, growth should tition in the financial field stemming from the be supported by continued firm consumer de- internal market, deregulation, and rapid tech- mand-bolstered by tax reductions in France, nological process, thereby accelerating the move Germany, Italy, and Spain-with stronger spill- toward integrated and more efficient capital overs to fixed investment, and the expected markets. The more than doubling of the value unwinding of the terms-of-trade shock as oil of M&A activities and of corporate bond is- prices fall. Yet growth will be restrained by sues in 1999 is some evidence of the early im- the higher interest rate environment and slow- pact of the EMU. The eastward expansion of ing from exceptionally rapid growth in a num- the EU could enhance the positive growth sce- ber of smaller countries (such as Belgium, the nario outlined above. Alternatively, difficulties Netherlands, and Spain). These factors sug- in absorbing substantial new population blocs gest a slight moderation in growth toward 3.2 into the union could present risks to future percent in 2001 and further to 2.8 percent in growth.5 Questions regarding intra-EU labor 2002. mobility and especially the Common Agricul- Economic performance in the major Euro- tural Policy (CAP) will become more pressing pean countries is expected to improve sub- as expansion moves forward. stantially over the next decade compared with the 1990s, when low productivity growth (1.3 percent during the second half of the decade- World trade remains on a figure 1.8), persistent unemployment, and slug- long-term high-growth path gish capital spending limited GDP growth to T he 1990s witnessed a dramatic acceleration less than 2 percent per year, compared with T of world trade, both in comparison with the more than 3 percent in the United States. Po- 1980s and in relation to growth in GDP, driven tential growth rates may be as high as 2.8 or 3 by technological change and the removal of percent underpinned among other things, by trade barriers (figure 1.9). World trade is likely the introduction of the euro; the growing par- to continue to grow strongly, although some- ticipation of women in the labor force; and what below the current record pace. 11 G LO BA L EC ONO MI C PROS PE CT S Figure 1.8 Growth of GDP per employed person: United States and European Union Percentage change 3 European Union 2 0 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Source: U.S. Bureau of Labor Statistics; U.S. Department of Commerce; OECD. Figure 1.9 Trade versus GDP growth Percent 12 --4 12 A Trade ~~~~~~~~~~~~~~Ratio of trade/GDP growth rates /T 9 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~3 6 -2 3 1 0 0 -3 - 1965 1970 1975 1980 1985 1990 1995 2000 Note: Trade is defined as the average of real exports and imports of goods and nonfactor services. Trade-to-GDP growth-rate ratio is based on five-year moving average. Source: World Bank staff estimates. Global trade is now at a cyclical high supported by strong demand growth in indus- World trade accelerated in the second half of trial countries and the recovering economies of 1999, peaked at 14 percent (year on year) in the East Asia (which contributed 25 percent of the first quarter of 2000, and is expected to average growth in world demand in 1999). After the fi- a remarkable 12.5 percent for the year as a nancial crisis, industrial production in the crisis whole, the highest annual rate of growth since countries surged to refill inventories and stocks before the first oil crisis. This robust growth was of capital goods and consumer durable goods 12 PROS PE CTS F OR DEVELOP ING C OU N T RI ES AND WORLD TRADE Figure 1.10 East Asia-5 industrial production and import volume Percentage change year over year 25 15 Industrial 5 ~~~~~~ct prduton GDP (bar) -15 Iport \ / ~~volume -25 ; I 01 1996 01 1997 01 1998 Q1 1999 Q1 2000 Note: The East Asia-S countries are Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand. Source: Datastream and World Bank staff estimates. (figure 1.10). Demand for foreign durable with China's record of no growth in this area goods and intermediate inputs increased at the between October 1998 and April 1999. In same rate. As industrial production will rise contrast, Latin American countries (excluding faster than GDP only temporarily, the extraor- Mexico) experienced significant losses in mar- dinarily strong import demand is only transi- ket share in 1999 (figure 1.11, second panel), tory. Other regions recovering from the crisis and the rebound witnessed in the first half showed similar, although weaker, patterns. of 2000 was weak in comparison to that of In addition, real exchange rate depreciation East Asia. Export volumes continued to grow fueled developing countries' export volumes. strongly in Mexico throughout the crisis pe- East Asian countries' real exchange rates de- riod of 1997-99 and averaged about 15 per- preciated by an average of 23 percent in 1999 cent in the first half of 2000, despite an appre- compared with June 1997 levels, resulting in ciating real exchange rate, owing to strong strong gains in market share-though there links to U.S. manufacturing developed through were short-lived losses in U.S. dollar terms- the globalization of production and cemented (figure 1.11, first panel). Brazil, Colombia, by the North American Free Trade Agreement Ecuador, and Peru also undertook large ex- (NAFTA). Similarly, exports from Central Eu- change rate adjustments in early 1999 (al- ropean economies benefited from their in- though the average real exchange rate in Latin creasingly close ties to Western Europe (partic- America in 1999 was only 7 percent below ularly Germany) as they progress toward full precrisis levels). accession to the EU (figure 1.11, third panel). Even China, which initially gained export market share in U.S. dollar terms because of Structural factors boosted trade during the its policy decision to hold the renminbi fixed 1990s during the crisis period, benefited handsomely Developing countries' exports increased by 10 from the cyclical upturn with export volumes percent per year during the 1990s, triple the growing in excess of 35 percent year on year in growth rate during the 1980s (figure 1.12). the first half of 2000. This can be compared Privatization and more intense competition in 13 G LO BAL ECO N OM IC PROS P ECT S domestic markets increased the incentive to Figure 1.11 Export volume and market find lower-cost intermediate inputs and to growth, 1997-2000 search for new export markets. Technological Three-month moving average, year over year advances reduced communications and trans- portation costs, greatly facilitating marketing Percent EastAsIa-5 and outsourcing of production (World Bank 40 1992, 1997). And regional and multilateral export growth agreements have reduced barriers and greatly 30- contributed to the acceleration in trade. 20 V \ __Multilateral agreements. Negotiations under the General Agreement on Tariffs and Trade 10 (GATT) and the World Trade Organization o , , , , (WTO) have provided an enormous impetus to marketgrowth trade. Multilateral agreements were primarily -1 m responsible for the reduction in average tariff Oct. Apr. Oct. Apr. Oct. Apr. July rates in industrial countries and the removal of 1997 1998 1998 1999 1999 2000 2000 Note: The East Asia-5 countries are Indonesia, the Republic of a wide range of nontariff barriers through the Korea, Malaysia, the Philippines, and Thailand. id1990 when the Round was Source; Dataseream; and World Bank staff estimates, mid -1Vsw.'...s.en, til okyo fon a ully implemented. Further, the GATT negotiations Latin America have exerted important influences on other ne- 40 gotiations and trade policy in general. Prece- dents established under the GATT have guided 30 regional arrangements.6 The GATT has pro- 20 vided an important venue for many countries to participate in trade negotiations, sometimes 10 for the first time; has established a wide variety ot >9ro tZZI Sof standards (such as tariffication, import val- o, , , uation, standards for trade in food and ani- -10 export growth mals [SPS agreement], protection of intellectual -10 Oct. Apr. Oct. Apr. Oct. Apr. July property [TRIPs agreement], and so forth); has 1997 1998 1998 1999 1999 2000 2000 contributed immeasurably to maintaining sta- Note: The Latin American countries in the graph above are Argentina, Brazil, Chile, and Colombia. ble rules of the game in international trade re- Source: Datastream; and World Bank staff estimates. lations, by facilitating dispute settlement and Central Europe constraining unfair trade practices;7 and has Percent heightened awareness of the importance of in- 40 ternational trade and encouraged significant 30 improvements in countries' capacity for trade export growth administration and negotiation.8 20 - \/\ r Regional agreements. Regional agreements 10 ~ \ ,. / u\ played an increasingly important role in the marketggrowth global trading system during the 1990s (box 0 , gt 1.2). They have often provided opportunities for more comprehensive dismantling of trade 1Oct. Apr. Oct. Apr. Oct. Apr. July barriers and greater harmonization of rules 1997 1998 1998 1999 1999 2000 2000 governing trade than can be accomplished Note: The Central European countries in the graph above are under multilateral negotiations. This is par- the Czech Republic, Hungary, and Poland. Source: Datastream; and World Bank staff estimates. ticularly true of the EU and NAFTA, both of which developed important precedents for 14 PRO SP EC TS F OR DE V EL O PING CO UN TR I ES AND W OR L D T RA DE Figure 1.12 GDP and export volume growth Percentage change per year 12 * 1970s U 1980s 7 199 10 8 6 4 2 0 GDP Exports GDP Exports GDP Exports High-income Developinga World a. This excludes countries in Europe and Central Asia. Sozurce: World Bank staff estimates. multilateral negotiations and other regional lateral agreements that lead to increased arrangements. There are many reasons for en- growth may spur intraregional exports be- tering regional trade agreements-many of a cause of lower transport costs (than outside political economy nature. However, there are the region) and other agglomeration effects significant concerns over their economic bene- (for example, greater knowledge of closer fits. Regional trade agreements shift import markets than of extraregional ones). Con- supply from external countries to countries versely, regional arrangements can stimulate within the free trade area. This may lead to re- global trade through improving the efficiency duced efficiency for the countries within the and hence competitiveness of regional produc- free trade area if external suppliers are lower- ers and expanding demand for inputs from cost suppliers. Also, those outside the agree- nonregional sources. Nevertheless, the existing ment suffer from lost market share or lower data do indicate that some regional arrange- supply prices. ments have been associated with expanded A myriad of other regional integration trade. The growth of intraregional trade was agreements have evolved (figure 1.13 and the significantly greater than the growth of ex- annex).9 Some of these agreements are de- ports outside the region in NAFTA and the EU signed to address similar leverage and harmo- during the 1980s, and in NAFTA and Merco- nization issues that faced the EU and NAFTA. sur (the Latin America Southern Cone trade Some countries have undertaken more am- bloc) during the first half of the 1990s (table bitious efforts at regionalism-for example, 1.2). The EU during 1990-95 is an exception, the members of the Association of Southeast owing to the relatively slow growth in Europe Asian Nations and the Asia Pacific Economic following German reunification.10 Cooperation. Many of the other regional arrangements It is extremely difficult to measure the rela- lack the economic diversity required to meet tive importance of regional and multilateral the bulk of their trade needs. Only three of the agreements to the expansion of trade. Multi- non-NAFTA and EU agreements have more 15 G LO B AL EC ONO MI C PRO SP EC TS $ex 1.2 rob . g ien, flFigure 1.13 Number of WTO notifications of regional integration agreements (RlAs) t8 15 e*tet~sive their repciergos(iue11) oe ariiaigi h T a ral opi gains ~ cve frroigohreo tradeEa intefuue tisto~ fou moeo egoa rrneet 16~~~~~~~ enhance pro ource;WorldTr4de rganiation ternltaifshani 20 pecnd fteei irmaverag trd ihn Tevs ices ntenme of4 coutres persectve,thesrrse i clude renrgioens (fiuepr 1.4.Nn- priiaigi h T a ral opi market,. acelerste fregigna cinttgraion aranemet a ctdngoitos,afc ta a laon Figurea 1roi3 Number of Wtrad inothefutionseo triegtona fntegrmtron oagreements arranemest 16~~~~~~~ PROS PE CT S F OR D EVELO PING CO UN TRIES AND WORLD TRADE Table 1.2 Intra- and extraregional trade growth to 12-13 percent in 2000. However, (annual percentage change in exports) world trade growth is likely to slow over the 1980-90 1990-95 course of the year, in line with the expected slowing of world industrial production. In- thiin Oustide Within Outside dustrial production in key developing regions regiws rcgiwr region region (such as East Asia and Latin America) had al- NAFrA 15.6 9.8 9.9 5.3 ready slowed by the second quarter. While European Union 16.1 10.9 3.2 6.6 some upturn is likely for these countries in the Mercosur 4.3 9.3 27.5 4.0 second half, overall momentum is unlikely to Source: World Bank staff data. return to the rates experienced in the latter half of 1999 and the first quarter of 2000. Growth in world trade volumes is projected with smaller memberships, where reciprocal to slow to 8 percent in 2001 and 6.8 percent in concessions can be more transparent and 2002, for a number of reasons. First, the cycli- immediate (thus facilitating the negotiating cal pattern of world GDP growth is expected process). Smaller memberships may also make to move toward more sustainable long-run it easier to negotiate the increasingly impor- rates, thereby reducing import demand. For tant issues inherent in product standards (see example, U.S. import growth, which reached chapter 3). 13 percent (year on year) in the first half of 2000, is likely to slow toward 7 or 8 percent Prospects for trade growth in 2001-02, helping to stabilize the widening Strong growth momentum in industrial coun- trend in the current account deficit. This is un- try import demand in the first half of the year likely to be offset completely by increases in will bolster developing-country export volume import demand in other major trading coun- Figure 1.14 Intra-RIA exports as a share of RIAs' total exports Percent 80 70 U 1990 * 1996 60 50 40 30 20 1 0 0 Note: The names and abbreviations on the horizontal axis represent the names of organizations involved in regional integration agreements. See the annex for membership and dates of formation. Source: World Bank 2000e. 17 G LOB AL ECON OM IC PROSP ECT S tries. Second, gross private capital flows to de- While participating countries will continue to veloping countries are expected to rise by only benefit from increased integration, it is un- 15 to 20 percent over the next two years, well likely that further reductions in trade barriers below the rate of increase in 1996-97, when will be of the same magnitude. large capital flows permitted some developing Other forces may boost world trade growth regions (such as Latin America) to boost im- in comparison with the 1990s. For example, ports. Third, the terms of trade for oil-import- there may well be improvements in informa- ing countries are likely to remain soft in the tion technology (see the section on industrial near term, as oil prices stay relatively high and countries and chapter 4), and another round non-oil commodity prices rebound weakly. of trade negotiations may be successfully con- This, in combination with fairly sluggish pri- cluded (despite the derailing of the launch of vate capital flows, would tend to limit the abil- a new round in Seattle in December 1999). ity of oil-importing countries to sustain rapid While any quantitative comparison of these import growth for an extended period. How- influences is extremely speculative, on balance ever, none of the above factors are expected to we anticipate some decline in the ratio of world cause a massive deterioration in world trade trade growth to output growth. growth in the near term. In the longer term (2003-10), world trade is projected to grow by 6.8 percent a year. Private capital flows remain The long-term forecast for trade growth is 2.1 volatile times the projected rate of world GDP growth, he surge in globalization during the lower than what was observed in the 1990s l 1990s was even more spectacular in cap- but still much higher than in the 1980s. The ital flows than in trade flows. Net long-term very high ratio of the 1990s was in part due to capital flows to the developing countries the one-time increases in integration repre- surged from $80 billion in 1989 to $344 bil- sented by the EU single-market initiative and lion just before the financial crisis, before NAFTA as well as large-scale trade liberaliza- falling to $280 billion in 1999 (figure 1.15). tion in a number of developing countries. FDI flows grew steadily to $180 billion in Figure 1.15 Net capital flows to developing countries, 1985-2000 Billions of U.S. dollars 400 350 - J Official *Other privaten 300 - *FDI 250 200 150 100 50 0 1985 1990 1995 2000 Note: Amounts for 2000 are estimated. Source: World Bank data and staff estimates. 18 PRO SP EC TS F OR D EVE LOP ING CO U NTRI ES AND WORLD T RAD E 1999, almost eight times their level at the be- conditions, combined with large current ac- ginning of the decade. Other private flows count surpluses, reduced the need for interna- have been extremely volatile-increasing ten- tional financing. At the same time, the current fold between 1989 and 1996, but declining account deficits in the high-income countries 70 percent during the last three years of the increased from $9 billion in 1998 to $175 bil- decade. Total official flows fluctuated around lion in 1999, and they are expected to reach $50 billion, with significant dips in 1996 and $250 billion in 2000. With an increased do- 1997. Preliminary data for 2000 covering gross mestic savings shortfall from $218 billion to flows suggest that total inflows stabilized, with $435 billion during the last two years, the the share of FDI declining somewhat from its United States (which saw an investment boom) high level of 1999. was the main source of the deterioration of the current account in the industrial world. Stabilization of capital flows Continued uncertainty and risk aversion fol- in the short run lowing the financial crisis constrained market- The stabilization of international capital flows based flows (bonds, bank loans, and equity) to into developing countries was initially driven several of the emerging market economies in by a reduced supply of funds by international 2000. The average risk premium on developing- investors, but now it increasingly reflects im- country secondary market debt remained high. proved domestic credit conditions and a sharp New financing primarily targeted less risky rise in capital demand in the industrial world. borrowers: 60 percent of total developing- Most countries affected by the financial crisis country bond issuance came from sovereign brought inflation rapidly under control while borrowers (compared with 55 percent in 1999), achieving currency stability after large devalu- and the share of private borrowers remained ations and current account adjustments, and low (figure 1.16). Moreover, a substantial pro- this opened the way for more accommodating portion of bank lending (55 percent) went to monetary policies. Improved domestic credit finance the rollover of upcoming liabilities or Figure 1.16 Sectoral breakout of bond financing by developing countries, January 1997-June 2000 Percent | Private M Sovereign F1 Public 100 75 50 25 0 HI 1997 H2 1997 Hi 1998 H2 1998 Hi 1999 H2 1999 Hi 2000 Source: World Bank data and staff estimates. 19 G LO BAL ECO N OM IC PROS P ECT S took the form of less risky lending, such as past years, from $111 billion in 1993 to $52 trade finance or securitized lending. billion in 1998 and $41 billion in 1999.13 The volatility of capital flows in the second quarter underlined the continued vulnerabil- In the long term, capital flows should ity of developing countries to shifts in inves- regain momentum tor sentiment. A sharp correction in the U.S. FDI flows to developing countries are likely to NASDAQ market was associated with a jump rise over the long term, as rapid international in volatility in developing-country stock mar- integration continues (witness the recent wave kets,11 and the risk premium on developing- of cross-border mergers and acquisitions country external debt rose to 850 basis points among corporations in the industrial coun- (compared with 760 basis points at the start of tries),14 and developing countries' growth rates the year). In April, the volume of capital flows continue to exceed growth rates in the indus- to developing countries dropped by 75 percent trial world. Renewed cross-border M&A ac- over March, and it declined marginally further tivity in Korea and in other East Asian coun- in May before recovering in June to almost the tries could raise FDI inflows to the region. March level. And political commitment to removing obsta- For the first time in over a decade, prelimi- cles to privatization may accelerate postponed nary data suggest a contraction in FDI flows to projects in a number of Central and Eastern developing countries in 2000 from the $180 European economies. However, the growth of billion recorded in 199912 (figure 1.17). The FDI is unlikely to be as spectacular as it was in downturn in FDI was brought about by re- the 1990s. duced commitments for new projects in major Other private capital flows are expected to recipient countries, combined with a slowdown regain some momentum from their current de- in M&A activity, and completion of large-scale pressed levels. A narrowing of current account privatization projects. China, the largest recip- imbalances may increase demand in some de- ient of FDI, experienced a substantial reduction veloping countries, and further progress in fi- in the value of new commitments during the nancial reforms should go some way toward Figure 1.17 FDI flows to developing countries, 1990s Billions of U.S dollars Percent 200 40 180 35 160 30 140 A s _ wol FDl 120 25 100 20 80 15 50 1 0 40 20 5 0 0 1991 1992 1993 1994 1995 1996 1997 1998 l999a a. Preliminary. Source: World Bank Debt Reporting System and UNCTAD Investment Yearbook. 20 PRO SP EC TS F OR D EVE LO P I N G C OU N TRI E S AND WO RLD T RAD E restoring the confidence of international in- inventories fell dramatically, and prices sky- vestors. However, capital market flows will re- rocketed (figure 1.18). OPEC has responded main volatile, in turn contributing to the un- to the near-term shortage in the market by certainty in the real economy. For that reason, raising its production ceiling back to the levels FDI flows are likely to continue as the primary of early 1998. A combination of supply in- source of international funding for developing creases and some decline in demand (from countries, in the process helping to reduce vul- higher prices) should reduce oil prices from an nerability to financial shocks. average of $28 per barrel in 2000 to $25 per barrel in 2001 and $21 per barrel in 2002. Plausible worst-case scenarios (for exam- Commodity prices exhibit ple, an unusually cold winter or unanticipated divergent recoveries supply disruptions) could see prices averaging Oil prices. The present oil price shock is $30 a barrel in 2000 and 2001, with tempo- O expected to be temporary, since it was rary spikes running to $50 or more. Depend- generated by the confluence of a number of ing on policy and private sector reactions, unexpected short-term factors. The spike in such higher prices could pose a substantial oil prices has its roots in the reaction to the threat to global expansion, particularly if the 1998 price decline in the wake of the financial shock contributes to steep declines in the sev- crisis-a decline that in real terms placed the eral highly valued industrial country equity oil price at one-quarter of its peak level of markets (the implications are explored in the 1980. OPEC members, along with some non- low-case scenario-see below). However, it is members, agreed on production cuts in 1999 difficult to see significantly higher prices being to boost prices, while low prices also led to a sustained for more than a year or two, given slowdown in the growth of non-OPEC pro- that non-OPEC production would increase in duction and in investment in the oil sector. response. Prices are expected to average about The drop in production coincided with the un- $18 to $19 per barrel for the rest of the de- expectedly strong rebound in world economic cade, as technological improvements (for in- activity in 1999, and hence in oil demand. Oil stance, better methods of locating and recov- Figure 1.18 Crude oil prices, January 1990-October 2000 US$ per barrel 40 Dubai Brent - WTI 35A 25 20 A f 15 10 I 5 I l I Jan. 1990 Jan. 1992 Jan. 1994 Jan. 1996 Jan. 1998 Jan. 2000 Oct. 2000 Source: World Bank data and staff estimates. 21 G LO B AL EC ONO MI C PROS PE CT S Figure 1.19 Crude oil prices, 1960-2010 USsIbbl 60 l Current US$ 1990 US$ 50 40 30 20 10 0O I lI I I lI I I 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Source: World Bank data and staff estimates. ering crude oil) boost energy production and tries, table 1.3 presents the impact of a $10 per conservation efforts continue. barrel increase in price (the average increase The impact of the current oil price rise on anticipated in the baseline for 2000) on current industrial countries has been less than the im- account positions for a sample of 92 countries. pact of price rises during the oil price shocks While the current account balance of oil- of 1973-74 and 1979-80, because the current exporting developing countries is expected to increase is smaller and output is much less improve by about $135 billion (at unchanged dependent on oil than before. Oil prices in oil trade volumes) as a result of the oil price in- 2000 should average about half the level of crease, that of oil-importing developing coun- the 1979-80 oil shock in real terms (figure tries is expected to deteriorate by about $40 1.19).15 Nevertheless, the oil price rise has billion, or a little over 1 percent of GDP. increased inflationary pressures and trade Because the oil shock is expected to be tem- deficits in some of the industrial countries, as porary, there is a good economic case for oil- well as exacerbating tensions over the level of importing countries to meet higher bills for oil gasoline taxes. and gas imports through temporary balance of Oil-importing developing countries have payments deficits and external financing been more severely affected than industrial rather than through adjustment. However, there countries, because they consume more energy is a good deal of uncertainty about how high per unit of output and have less access to the prices will go and for how long, and even a external financing required to sustain expen- temporary shock could make international diture levels until oil prices decline. Moreover, lenders jittery about the sustainability of prices for their primary commodity exports countries' external debt. This uncertainty in- (especially tropical beverages and other agri- creases the risk of a sudden withdrawal of ex- cultural goods) have continued to drop over ternal finance. It is thus likely that risk-averse the course of 1999 and 2000, so their terms of policymakers in oil-importing countries will trade have fallen precipitously. undertake some degree of prudent adjustment. To illustrate the effects of higher crude oil The oil-importing emerging market econ- (and natural gas) prices on developing coun- omies should be able to smooth the impact of 22 PROS PE CT S F OR D EVE LO PING C OU N TRIES AND WORLD TRADE Table 1.3 Current account effects for a sample of developing countries from a $10 increase in oil prices Ol lashs 'u Oil exporters ' i dUI ' Nuise in II $US - Number in $US as Nme iSU as smap bYn %GDP sample bln. % GDP sit e bI % GDP East Asia and Pacific 7 -16 -L.0 3 7 2.0 10 -9 4.7 South Asia S - -0.9. 0 0 0.0 5 -5 -0.9 Latin America is -4 -0.7 7 22 2.0 22 to 0.8: Sub-Saharan Africa -13 2 -0.7 5 13 19.5 -,18 11 3.2 Europe and Central Asia .18 14 -1.7 3 27 10.3 21 13 15 Middle East and North Africa 6 -. -LI 10 66 11.4 16 64 8.6 Total developing countries 64 -43 -1.1 28 135 5.7 92 92 1.5 Memo item: HIPC 13 -2 -1,4 6 5 19.0 19 3 1.7 Note: The table shows the direct current account impact (keeping volumes constant) of a $10/bbl increase in crude oil and refined products and a (similar) 54 percent increase in the gas price. Source: World Bank staff estimates. the shock with private finance, though some a time, the net additional call on international with already large current account deficits will donors does not appear insurmountable. have to proceed with caution. Oil-importing Non-oil commodity prices. Non-oil com- developing countries without access to private modity prices began to decline in early 1997 capital markets will face an additional official and then plummeted with the East Asian crisis financing need of about $18 billion (without (figure 1.20). While the global economic re- adjustment). Since countries will be undertak- covery has led to some recovery of metals and ing some degree of adjustment (leading to a minerals prices, agricultural prices continue lower financing need), and the need for official to languish near their cyclical troughs. This aid flows to oil exporters may be much less for divergent recovery is not surprising, since met- Figure 1.20 Divergent recoveries of commodity prices Index, January 1997=100 Oct. 2000 150 Crude oil 125 100\ Metals and minerals\'> 100 75 50 ~~t \ w_/ Agriculture 25 Jan. May Sept. Jan. May Sept. Jan. May Sept. Jan. May Sept. 1997 1997 1997 1998 1998 1998 1999 1999 1999 2000 2000 2000 Source: World Bank staff estimates. 23 G LO B AL EC ONO MI C PRO SP EC TS als, which are used as inputs to industrial pro- countries in Latin America and Sub-Saharan duction, have higher short-run income elastic- Africa, have seen substantial deterioration in ities than food and beverages. their commodity terms of trade. C6te d'Ivoire, After the price declines in 1998, metals and Ghana, Kenya, and Uganda all receive 40 to minerals producers cut production at high- 60 percent of export earnings from agriculture cost mines and smelters, leading to some slow- (mainly coffee and cocoa), and fuel imports down in production growth. For example, constitute 20 to 30 percent of import costs. copper production slowed to 3 percent growth Most Asian countries have been less affected, in 1999, from 4 percent in 1998. At the same since they are less dependent on agricultural time, the strong global economic recovery exports and fuels are a smaller share of total boosted demand for metals. Consumption of imports. copper rose 4 percent in 1999 and will rise an Non-oil commodity prices are expected to in- expected 6 percent in 2000, while aluminum crease in the near term, gradually aligning with consumption rose 6 percent in 1999 and is up the continued expansion of the global economy 5 percent in 2000. Slower production growth (table 1.4). Metals and minerals prices, which and accelerating demand have reduced stocks, rose about 14 percent in 2000, are expected to and metals and minerals prices are estimated increase about 2 percent per year in nominal to have risen above 14 percent in 2000, to a terms over the next several years, but more rapid level about 20 percent above the cyclical trough. increases are possible if global economic growth In contrast, agricultural prices remained is higher than anticipated. stagnant for most of this year. Despite this, the The recovery in agricultural prices is ex- United Nations' Food and Agriculture Or- pected to remain slow, as supplies continue to ganization's index of global agricultural pro- increase at nearly the same pace as consump- duction rose by 1.6 percent in 1999 (slightly tion. But experience shows that current low below the 30-year trend growth rate of 2.2 prices in agriculture could give way to a surge percent), which contributed to further stock in the near to medium term. While it is difficult buildups. Consequently, world stocks of most to predict when such an event might occur, his- agricultural commodities remain high-and in torical evidence indicates that it could begin some cases stocks have continued to increase. about two to three years after the cyclical low. Sugar stocks, for example, rose for the fifth Over the longer term, non-oil commodity consecutive year in 1999, while cocoa stocks prices are likely to decline in real terms, con- reached the same levels as in 1990-91, when tinuing the trend over the past 100 years (real the International Cocoa Organization was op- non-oil commodity prices fell by nearly two- erating a buffer stock mechanism. An excep- thirds during the twentieth century, and by tion to this trend is cotton, for which pro- half over the last two decades-[figure 1.21]). duction is expected to decline by 2 percent, There appears to be no letup in the improve- contributing to a 15 percent reduction in stocks. ments in technology that boost commodity Moreover, recovery in demand has been weaker supplies at lower cost. Crop yields continue than in metals. Grain consumption is expected to increase along historical trends, and new to be roughly unchanged in 2000; but con- plant-breeding techniques offer the prospect sumption of raw materials is recovering, led of further increases. Improved mining and re- by cotton, which is expected to increase 2 per- fining techniques reduce the cost of recovering cent next year. ore and producing metals. On the demand Recent trends in commodity prices have side, population growth is projected to slow obviously favored food importers (particularly from 1.4 percent during the 1990s to 1.1 per- the oil-exporting countries, which simultane- cent during the first decade of the 21st century ously have benefited from higher oil revenues), and 0.9 percent during the second decade. In while net agricultural exporters, such as many Asia, where the demand for commodities has 24 PROS P ECT S FOR D EVELO PING CO UN TRIES AND WORLD TRA DE Table 1.4 Annual percentage change in nominal oil and non-oil commodity prices, 1981 -2010 Forecasts Commodity 1981-90 199147 1991 1999 2000 2001 2002-10 Oil -4.7 -2.5 -31.8 38.3 55.0 -10.7 -3.0 Non-oil -2.2 2.3 -15.7 -11.2 -0.8 3.4 2.8 Agriculture -3.2 3.7 -163 -13.9 -5.2 3.9 3.3 Food -3.3 2.2 9.8 -16.5 -3.9 5.1 2.4 Grains -,9 1.6 -9.7 -14.7 -9.4 7.5 3.8 Beverages -5.8 7.9 -17.7 -23.4 -16.9 1.5 4.3 Raw materials -0.4 1.9 -23.2 1.4 3.7 4.2 3.5 Metals and minerals 0.6 -E2 -16.2 -2.3 13.6 2.2 1.6 Fertilizers -2.5 Z6 2.0 -6.6 -6.3 4.7 1.1 Memorandum item G-5 manufactures unit value 3.3 1.1 -1.9 -2.7 -2.3 3.6 2.2 Note: The G-5 countries are France, Germany, Japan, the United Kingdom, and the United States. Source: World Bank data and projections update, November 2000. grown most rapidly, population growth will be Developing countries' recovery is even slower. This may be partially offset by unexpectedly rapid, and prospects faster growth of world real incomes (projected for long-term growth have at 3.4 percent over 2000-10 compared to 2.7 improved percent during the 1990s). However, since in- eveloping countries' recovery from the come elasticities of demand for commodities D/ 1997-98 financial crisis at 5.3 percent are low, the overall impact of more rapid in- growth has been faster and much stronger than come growth on commodities will be small. anticipated.16 All regions have experienced Figure 1.21 Real commodity prices, 1900-2000 Index, 1900=100 250 200 Crude oil 150l 100 t ~~~~~~ ~ ~ ~~Non-oil t 50 1900 1920 1940 1960 1980 2000 Source: World Bank data and staff estimates. 25 G LO B AL EC ONO MI C PRO SP EC TS Figure 1.22 Developing regions' real GDP growth, 1999-2002 Percent 1 1999 * 2000 E 2001 * 20021 4 2 0 East Asia South Asia Latin America ECA MNA Sub-Saharan Attica Source: World Bank staff estimates. stronger growth in 2000, although there has by 5.3 percent in 2000, matching peak years been diversity across regions. Contributing fac- 1983 and 1997. Inflation came down quickly tors include easier monetary policies in the following the crisis (when sharp exchange rate industrial countries and in East Asia, which depreciations led to rapid price rises in several lowered interest rates and stimulated domestic countries) and remains moderate despite the demand; the depreciation of many developing spike in oil prices. Despite this favorable pic- countries' currencies, which boosted exports; ture, financial tensions are building up once and more recently the rise in oil prices, which again in East Asia and Latin America. The de- has supported economic activity in some of the cline in stock markets and the recent increase economies hit by the crisis or in those near in spreads make several countries vulnerable in crisis (such as Indonesia, Nigeria, and the Rus- the short run. The risks associated with these sian Federation). Industrial production in most vulnerabilities are explored later in this chapter of the crisis-affected countries of East Asia re- where the possibilities of a strong global down- bounded at double-digit growth rates in late turn are discussed. The baseline forecast, how- 1999 and into 2000. Latin America also is re- evet, features a moderate slowdown from the covering sharply, albeit at a slower rate than in cyclical peak in early 2000. With this moderate the wake of the Mexican peso crisis. And Rus- deceleration, all developing regions are expected sian growth (a large segment of the growth in to enjoy near-term increases in per capita in- the Europe and Central Asia region) was un- come, ranging from nearly 6 percent in East expectedly strong, boosted by oil revenues (fig- Asia to about 1.5 percent in the Middle East ure 1.22 and table 1.5). China and India con- and North Africa and Sub-Saharan Africa. tinue to exhibit sustained rapid growth, and Middle Eastern countries are benefiting from Payoffs to domestic reforms and improved high oil prices and recovery in the Euro Area. external conditions favor long-term growth Even the non-oil exporters in Sub-Saharan The cyclical recovery is expected to be followed Africa increased GDP by 3.2 percent, despite by an acceleration of long-term growth, al- low non-oil commodity prices. Altogether, de- though the outlook varies considerably across veloping countries' GDP is expected to increase regions (figure 1.23). Population growth in the 26 PRO SP EC TS F OR DE V EL O PING CO UN TR I ES AND W OR L D T RA DE Table 1.5 Growth of world GDP, 1998-2002 (percentage change in real GDP) Forecast 1998 1999 2000 201 2002 World total 1.9 2.8 4.A 3.4 3.2 High-income countries 2.1 2.7 3.8 3.0 2.8 OECD 2.1 2.7 3.7 2.9 2.7 United States 4.4 4.2 5.1 3.2 2.9 Japan -2.5 0.3 2.0 2.1 2.2 Euro Area 2.7 2.4 3.4 3.2 2.8 Non-OECD countries 0.7 4.2 6.3 5.1 5.1 Developing countries 1.0 3.2 5.3 5.0 4.8 East Asia and Pacific -1.4 6.9 7.2 6.4 6.0 Europe and Central Asia (ECA) 0.0 1.0 5.2 4.3 3.9 Latin America and the Caribbean 2.0 0.1 4.0 4.1 4.3 Middle East and North Africa 3.3 2.2 3.1 3.8 3.6 South Asia 5.6 5.7 6.0 5.5 5.5 Sub-Saharan Africa 2.0 2.1 2.7 3.4 3.7 Memorandum items East Asia-5 countries' -8.2 6.7 6.9 5.5 5.1 Transition countries of ECA -0.7 2.5 5.0 4.2 3.7 Developing countries Excluding the transition countries 1.2 3.3 5.3 5.1 5.0 Excluding China and India -0.6 2.2 4.7 4.4 4.5 Note: All countries listed in the "classification of economies" section at the end of this volume are included as components of the regions presented in tables 1.5 and 1.6 (as well as the world summary table [table 1.1]). Exceptions for which sufficient historical data or projections are unavailable include: 11 low-income countries (among which are Armenia, Honduras and Nicaragua); seven middle-income countries (among which are Iraq, Georgia, and Guyana); and two high-income countries (Cyprus and Iceland). a. Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand. Source: Development Prospects Group, baseline, October 2000. Figure 1.23 GDP per capita growth, 1990-2010 Percent change per year 6 i | *~~~~~~~~~I 1 990s E 2000s E] low cas 4 2 0 _ -2 East Asia South Asia ECA LAC MNA SSA Sozurce: World Bank staff estimates. 27 G LO B AL EC ONO MI C PRO SP EC TS developing world is slated to slow from 1.6 reap the benefits of reforms carried out over percent annually in the 1990s to 1.3 percent the past decade (selected indicators are high- during 2000-10. And output per capita in de- lighted in table 1.7). In effect, these factors veloping countries is projected to rise by 3.7 constitute the initial conditions from which percent per year over the next decade, more longer-term prospects may be drawn. A num- than double the 1990s rate, in large part re- ber of clear improvements can be discerned. flecting the turnaround from output declines Median inflation rates have been halved, and in the transition economies (table 1.6). Other central government budget deficits are lower developing regions are expected to achieve now than in the late 1980s, contributing to more modest increases in growth rates. Ex- improved investor confidence. And developing ternal conditions are assumed to be more countries are much more open now than they favorable than during the 1990s, as higher were 10 years ago, as trade liberalization and productivity-led per capita growth in indus- stronger trade growth have helped raise trade to trial countries (2.6 percent versus 1.9 percent, GDP ratios by 50 percent on average. In addi- respectively) and further progress in trade lib- tion, better policies have attracted FDI (which eralization should support the growth of de- increased from 0.5 percent of developing coun- mand for developing-country exports at high tries' GDP in 1988-90 to 2.7 percent in 1998- levels. And capital flows to developing coun- 2000). Moreover, rapid growth in exports fa- tries should resume within an environment of cilitated a significant decline in debt-to-export low inflation and low interest rates. ratios compared with the late 1980s. It is important to note that developing Many developing countries have made sub- countries all over the world are expected to stantial investments in human capital. For ex- Table 1.6 Growth of world GDP per capita, 1980s through 2010 (annual average percentage change) Forecast Baseline Low cs I9aOe 199~~s 2000-10 20-1 World total 1. 1 2.3 1.3 C L.0 High-income countries a No t 2.7 1.7 4. OECD . f i 2.6 1.6 -0 O United States 42.5 1.2.9.3 Japan 3. I2.3 1.0-1 Euro Area .1.93.0 2.4 -, Non-OECD countries of 3 CA 4.1 2.3 Developing countries 0 S 3.7 2.3 -1.4 East Asia and Pacific 5. ~5.4 3.945 Europe and Central Asia (ECA) 0. 204.1 3.0-1 Latin America and the Caribbean .9.0 1.4 -1.6 Middle East and North Africa -0 ~ 041.7 0.7 40 South Asia the 3i K 3.9 2.5 Sub-Saharan Africa .12 -.61.3 -0.1 -. Memorandum itemis East Asia-5 countries' . . 4.2 2.9 -. Transition countries of ECA 0. 264.1 3.1 -. Developing counitries Excluding the transition countries 13 303.7 2.2 -. Excluding China and India .0.52.9 1.6 -. a. Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand. Source: Development Prospects Group, baseline and low-case, October 2000. 28 PRO SP EC TS F OR D EVE LO PING CO UN TRI ES AND WO RL D TRA DE Table 1.7 Forecast assumptions: 1999 at a rate of 6.7 percent in contrast with developing countries their 1998 crisis decline of 8.2 percent. They Initial conditions 1998--0 1t&84000 consolidated further with growth near 7 percent in 2000. A low-inflation, low-interest-rate envi- 1. Ratio of real GDP per capita: ronment has been particularly beneficial to the industrial / developing countries 19.S 20.5 2. Trade (X+M) / GDP ratio (real) 29.0 43.5 process of unwinding the domestic debt prob- 3. Median inflation rate 12.6 6.1 lems faced by firms and consumers in these cri- 4 Median fiscal balance / GDP -2.7 -l,8 sis countries. Corporate and financial restruc- 5. Investment I GDP (real) 23.1 24,3 6. Investment / GDP (nominal) 25. 24.3 turing and rehabilitation of the financial sectors 7. Gross national savings / GDP 25.2 23.3 are being pursued, though perhaps at a slower 7a. Gross domestic savings / GDP 27.1 25.1 8. Current account balance / GDP -0.7 - 8 pace than warranted. The slow pace could be a 9. FDI / GDP 0.5 2.7 detrimental factor to near-term growth, if inter- 10. External DOD / exports* 172.6 142.2 est rates rise rapidly or demand falters leading to 11. School enrollment rates Primary 78.0 82.0 diminished cash flow. Robust export growth Secondary 56.0 63.0 and firming export prices have helped maintain 12. Illiteracy rate 31.0 26.0 a positive current account balance. Though the 13. Under-5 mortality rate 91.0 79.0 14. Life expectancy 63.0 65.0 recovery of imports and higher oil prices have narrowed the balance in many countries, rising Exogenous assumptions 1990s 2001-10 reserves and the improved term structure of for- 1. Population growth 1.6 1.3 eign debt have strengthened external positions 2. OECD GDP growth 2.4 2.9 3. Oil price $ per barrel (avg.) 182 20. v2s-a-vis precrisis levels. 4. World trade growth 6.5 6.8 Growth in China during the postcrisis pe- riod has ranged between 7 and 8 percent. A *Exports of goods and services plus workers remittances. fioff ingetwth, c nd with t Note: Real indicators use 1995 as base year. falloff in export growth, combined with the Source: World Bank database, World Bank staff estimates. short-term impact of reform programs for the state enterprises and the financial system, led ample, school enrollment rates are substantially initially to a drop in domestic demand and a higher than in the late 1980s, and illiteracy period of deflation. The real depreciation of rates fell from 31 percent in 1990 to 26 percent the yuan, coupled with the global recovery, in 1998. And health indicators show improve- eventually led to a resurgence of exports. ment: under-five mortality rates dropped from Combined with fiscal pump priming, and an 91 per 1,000 live births to 79, and life ex- incipient increase in FDI, export growth has pectancy has increased from 63 years to 65 produced improving conditions in China, with years. These developments suggest that new- GDP growth accelerating in the first half of comers to the labor force should be better edu- 2000 and the deflationary cycle ending. In cated and more capable of working than those 2001-02, output for the East Asia region is who retire-a positive development for absorp- likely to begin a general process of moderating tion of new technologies and for innovation. and converging toward longer-term growth With real per capita incomes today still only paths. The two most vulnerable countries are one-twentieth that of the industrial countries, Indonesia and the Philippines. These countries developing countries that remain open to trade also suffer from political weaknesses, civil dis- and FDI can achieve higher rates of growth turbances, and a perception (from the point of through maximizing the new technology and view of investors) that business operating prac- skills embodied in these flows. tices have not changed substantially from less East Asia. On average, output in the five than transparent modes. countries most affected by the financial crisis East Asia should continue to achieve the (Indonesia, the Republic of Korea, Malaysia, the most rapid rates of growth over the longer Philippines, and Thailand) recovered smartly in term, although some deceleration from the 29 G LO B AL EC ONO MI C PRO SP EC TS last decade's pace is likely. Growth in the re- the surge of world trade growth have sup- gion's higher-income economies is expected to ported a broad resumption of economic activ- converge toward more moderate OECD aver- ity across the region. At the same time, infla- age rates. Lower-income countries that have tion eased or held steady in most countries, achieved high growth rates through strong re- allowing interest rates to continue on a general form programs may find the future reform declining trend. Exchange rates stabilized in agenda (particularly strengthening the finan- several countries that experienced periods of cial sector) more difficult to implement. free fall during 1999 (for example, Brazil and South Asia. GDP growth in South Asia has Ecuador), improving the outlook for domestic risen to 5.7 percent in 1999 and is likely to demand growth, especially in Brazil. register 6 percent in 2000, owing to better Global conditions are expected to be more than expected agricultural sector performance supportive of growth in the region over the in Bangladesh, India, and Pakistan, as well as next two years. However, recent experience an acceleration of India's industrial produc- suggests that volatility in financial markets tion to double-digit rates and strong advances and primary commodity prices remains a sub- in services output. Burgeoning foreign de- stantial threat to near-term recovery. Private mand for IT-related services from Bangalore capital inflows fell dramatically in the second and a pickup of FDI inflows ($2.2 billion in quarter of 2000, tied to the worldwide decline 1999) are major factors underlying India's im- in equity markets, and the recovery in indus- proved export performance. To facilitate the trial production among the large countries of growth of Indian services exports, legislation the region appeared to have faltered. The has been introduced to support the IT sector surge in the price of oil, concomitant with and develop electronic business infrastructure. weakness in commodity prices of critical im- Average growth for the region is expected to portance to the region (particularly the prices slow to 5.5 percent in 2001-02. Financial dif- of coffee, grains, and soybeans) produced ficulties are likely to restrain growth in Pak- terms-of-trade losses for a large number of istan. In addition, the region is heavily depen- countries. Nonetheless, consolidation of the dent on energy imports and (especially in the region's recovery in 2001-02 is likely, as ad- case of the smaller countries) on agricultural justment in Brazil has been impressive so far, exports such as cotton, tea, and rubber. The and new governments in Argentina and Mex- necessity of adjusting to terms-of-trade losses ico appear set to embark on a path of deep- from the recent, adverse movements in pri- ened reforms. Regional output growth is ex- mary commodity prices may dampen growth pected to reach 4.1 percent in 2001 and to rise in the near term. By contrast, South Asian further to 4.3 percent in 2002. economies may raise per capita growth rates Latin America is poised to enter a phase of in the long term if they can manage to reduce sustained moderate growth over the next fiscal deficits (while still maintaining growth- decade that is due to the past trend toward enhancing expenditures) and make necessary market-friendly policies in the larger coun- progress in trade liberalization. For example, tries; relatively strong banking and financial India's average tariff for all goods, while con- sectors; potential for technology spillovers siderably reduced from that of 10 years ago, from the United States; the largest rise in FDI remains at 40 percent. among developing regions (much of which Latin America's GDP is expected to rise by went into infrastructure such as telecommuni- 4 percent in 2000, although the dispersion of cations, utilities, ports, and so forth); and the growth across the region is wide, ranging from potential strengthening of Mercosur through over 6 percent in Mexico and Chile to nearly trade links with Europe and NAFTA. But low 2 percent in Colombia and Uruguay, and to lit- national savings and large debt overhangs that tle growth in Argentina, Ecuador, and Jamaica. will need to be rolled over on a continuing Stabilization of global financial markets and basis make the region vulnerable to swings in 30 PRO SP EC TS F OR D EVELO PING CO UN T RI ES AND WORLD TRADE external financing and are likely to constrain sian Federation, while the trajectory of growth growth below the rates expected in Asia. in world trade and output should support Europe and Central Asia. Average GDP steady gains in other CIS states. growth is expected to rise to 5.2 percent in Sub-Saharan Africa. Fallout from the 1997- 2000, significantly above the 1 percent ad- 99 crisis continued to exert a depressing effect vance of 1999. The 50 percent rise in oil and on the region in 2000, as non-oil commodity gas prices has transformed the Russian pri- prices remained near cyclical lows. But higher mary fiscal position from deep deficit to sur- oil revenues boosted growth for the region's oil plus, allowing reductions in government wage exporters, and output in South Africa strength- arrears and contributing to higher disposable ened moderately to 2.2 percent growth follow- incomes.17 Moreover, Russian industry con- ing several years of subdued performance. On tinues to benefit from the sharp devaluation of average, the region experienced an acceleration August 1998, although import-substitution ef- of growth to 2.7 percent from 2.1 percent in fects are diminishing with the recent real ap- 1999, and per capita income gained an average preciation of the ruble. Higher energy prices of 0.2 percent. Countries with better policy en- and economic spillovers from the Russian vironments-Botswana, Uganda, and several Federation are contributing to stronger output countries of the Communaute Financiere growth among hydrocarbon-rich members Africaine (CFA) zone-tended to perform bet- of the Commonwealth of Independent States ter than average, with GDP gains of 4.4 per- (CIS). The Central and Eastern European coun- cent. Countries experiencing civil strife or tries (CEECs) and the Baltic countries are bene- major political disruption-Angola, the Demo- fiting from growing demand from Western Eu- cratic Republic of Congo, Ethiopia, Sierra rope and to a lesser degree from the Russian Leone, and Zimbabwe-registered the weakest Federation.18 Growth in Turkey is approaching performances, averaging a decline of 1.5 percent 6 percent in 2000, up from the sharp 5.1 per- during the year. cent contraction in 1999, principally because of Growth is projected to accelerate to 3.4 per- a rebound domestic demand linked to declines cent in 2001 and 3.7 percent in 2002. Oil pro- in real interest rates. ducers, including Angola, Nigeria, and Sudan, Growth performance for the region through are scheduled to bring further supply on- 2002 is expected to remain relatively strong in stream, while continued high prices through aggregate, stabilizing at around 4 percent. De- 2001 should abet revenue growth. The terms of velopments in the EU export market, policy im- trade for commodity exporters should stabilize plementation related to EU accession for the or improve moderately from their current low CEECs, and the path of the oil price will be crit- levels as non-oil commodity prices firm. The ical factors in shaping the outlook. The Russian HIPC (Heavily Indebted Poor Countries) Ini- Federation and other hydrocarbon exporters of tiative is gaining momentum, with nine African the CIS may experience a slowing of growth countries-Benin, Burkina Faso, Cameroon, beginning in 2001, as oil prices retreat from Mali, Mauritania, Mozambique, Senegal, Tan- current high levels. The region's longer-term zania, and Uganda-now having qualified for prospects have improved considerably after the a total of close to $9 billion (net present value) difficulties experienced during the initial period of relief. And several more countries are ex- of transition to market economies in the 1990s. pected to reach completion points in the near Countries anchored by the EU accession process term. The enhanced HIPC Initiative is worth have strong incentives to implement reforms nearly $30 billion in net present value terms, and are positioned for stronger growth than with some 80 percent of the program ear- other countries in the region. The baseline as- marked for Sub-Saharan Africa. sumes improved economic management and Progress in reform programs and in debt- some progress in implementing recently pro- relief has improved the prospects for growth. posed social and economic reforms in the Rus- Per capita income is projected to rise by 1.3 31 G LO B AL EC ONO MI C PRO SP EC TS percent per year over the next decade. This growth in the oil exporters. For the diversified prospect is far better than the decline that con- exporters, the positive effects of higher exter- tinued over the 1990s, but the increase is only nal demand are being counterbalanced by rel- one-third the average rate of Asian economies. atively strong currencies, high fiscal deficits in Economies in Sub-Saharan Africa will continue Egypt and Lebanon, as well as recent declines to confront the severe problems of poor trans- in stock markets. Moreover, the ongoing na- port and communications infrastructures, a lack ture of recent conflict in the Levant may also of investor confidence that encourages capital have dampening effects on confidence in the flight and constrains private investment rates, rest of the MNA region. and continued low levels of official assistance. Progress in structural reforms and im- It is important to realize that HIV/AIDS will proved fiscal behavior with respect to com- have a substantial negative impact on a num- modity price booms and busts should support ber of countries. According to estimates by some acceleration of per capita growth over UNAIDS (2000), Sub-Saharan Africa contains the next decade. However, large and ineffi- 24.5 million (or 70 percent) of the 34.3 million cient public sectors, a shortage of social safety existing cases worldwide and 12.1 million of a nets, and low savings and private investment total of 13.2 million AIDS orphans. In the rates should limit growth rates to well below longer term, lower human capital accumula- those of most other regions. Moreover, with- tion may well emerge as the biggest cost, and out more substantial diversification of pro- in the worst-affected countries, labor force duction, these economies will remain exposed growth could slow by 1 or 2 percentage points, to unfavorable terms-of-trade shocks. with a depressing effect on growth. Middle East and North Africa. Develop- ments for both oil exporters and diversified ex- Vulnerabilities are significant porters in the region have been quite favorable, W hile the baseline scenario of solid growth with GDP growth of 2.2 percent reported in W in all regions is realistic and achievable, 1999 and growth of 3.1 percent anticipated for history cautions that cyclical downturns or 2000. Many of the major oil producers had crises induced by commodity or financial shocks formulated budgets around an assumed oil are difficult to anticipate. To explore the im- price of $22 per barrel, and higher revenues plications of less favorable outcomes, a low- have contributed to lower borrowing require- case scenario has been developed that com- ments, lower deficits, and a decline in domestic bines a downturn of the global economy in the arrears. Strong growth in Western Europe has short run with lower potential growth rates in fueled a boom in tourism, with record numbers the long run. In the short run, continued high of tourist arrivals in many North African and oil prices especially characterized by short- Mediterranean countries. The economic revival lived "spikes," contribute to inflationary pres- in Europe has also led to stronger gains in sures and increased uncertainty, triggering se- non-oil exports and workers' remittances. For rious cuts in demand and restrictive monetary example, remittance flows to Tunisia rose by policies. Additionally, investor concern over 75 percent during 1999. And the ending of the high U.S. current account deficit leads to drought conditions in many countries boosted a rapid reversal of foreign funds and a large agricultural incomes and exports and led to de- stock market correction, while the associated clines in required food imports. fall in demand, depreciation of the dollar, and Activity is expected to pick up moderately rise in interest rates have significant spillovers to 3.8 percent in 2001 and 3.6 percent in to other regions through trade, capital flows, 2002. With an average oil price of $25 per and debt service. The East Asian countries, in barrel for 2001 and $21 in 2002, export process of financial restructuring, would be revenues should continue to support income particularly affected. The ensuing global re- 32 PRO SP EC TS F OR D EVE LOP ING COUNT RIES AND WORLD T RAD E cession exacerbates the strains inherent in velopments, while for Pakistan the worsening rapid globalization and structural adjustment, of international financial conditions has very leading to a hold on reform programs that severe consequences. slows expected gains in productivity leading Since the structural risks are mainly of do- to long-term growth. On average, these ele- mestic origin, they are by nature quite differen- ments bring down the potential growth rate of tiated across countries. Nevertheless, there are the developing countries as a group by almost some common elements that follow from past 1.5 percentage points over the period to 2010 trends in the regions. Sub-Saharan Africa and (table 1.6 and figure 1.24). the Middle East and North Africa are at the The implications of the short-term global end of two decades of stagnation or decline, recession differ greatly across developing re- while growth in Latin America has picked up gions. Latin America, with high levels of debt from the "lost decade" of the 1980s (figure and relatively high dependence on exports to 1.25). Reform programs in many of these coun- the United States, is hit hardest by the global tries have greatly improved the conditions for downturn and the higher interest rates. East growth. However, high indebtedness and the Asia, which has a similar export orientation fragility of the reforms make these regions very toward the United States, is directly hurt by vulnerable to adverse global conditions, espe- the fall in U.S. demand, the depreciation of the cially a rapid rise in interest rates. The main dollar, and mounting domestic financial diffi- risk for the oil-importing countries in these re- culties. Central and Eastern Europe, the Mid- gions is that a global downturn, combined with dle East and North Africa, and Sub-Saharan a deterioration of the terms of trade and a lack Africa, all with a stronger focus on Europe, ex- of immediate improvements, could bring about perience a more moderate downturn in the social unrest and "reform fatigue." For oil ex- short run, as the growth slowdown in Europe porters, the danger is that the temporary surge is not as severe as in the United States. In South in oil revenues might suggest that reform is not Asia, the impact of the global downturn is di- urgent anymore. Such a reversal of the reform verse. As during earlier crises, India exhibits momentum in both oil-exporting and oil- some resilience to less favorable external de- importing countries could reduce the growth Figure 1.24 Growth of real per capita GDP, developing countries as a group, 1963-2008 Compound growth rate, centered five-year moving average, percent 5 4 ~~~~~~~~~~~~~~~~~~~~~~~~Base case 2 Source: World Bank staff estimates. 33 G LO B AL EC ONO MI C PRO SP EC TS Figure 1.25 Growth of real per capita GDP, Latin America and the Caribbean, Sub-Saharan Africa, Middle East and North Africa, 1965-2000 Compound growth rate, five-year moving average, percent 6 4 Latin America and the Caribbean 2 0 Sub-Saharan Africa -2 \\-X tJ\/ W ~~~~Midde East and North Africa -4 N__ 1965 1970 1975 1980 1985 1990 1995 2000 Source: World Bank staff estimates. potential for the coming decade. And in Sub- Europe, this increases domestic tensions and Saharan Africa, diminished government rev- reduces FDI flows, bringing down trend growth. enues tied to terms-of-trade losses could make The oil-exporting countries in the CIS experi- HTV/AIDS prevention and alleviation campaigns ence a delay of necessary reforms, similar to more difficult, further increasing economic the delay in the Middle Eastern and North losses associated with the epidemic. African oil exporters. When, ultimately, the oil In Asia, by contrast, many countries achieved price declines quickly as a result of the eco- rapid rates of growth through strong reform nomic downturn, the lack of reforms trans- programs during the 1980s and 1990s. Never- lates into lower potential growth. theless, continued rapid growth in the larger countries requires further reforms, including trade liberalization (in China measures related to Recent trends and prospects for WTO accession, in India reduction of high ex- poverty reduction isting tariffs), strengthening of the financial sec- Doverty trends during the 1 990s. Our esti- tor (through much of East Asia)-(figure 1.26) l mates for poverty in developing countries and strengthening of the fiscal position in India. have changed slightly since last year's Global In the alternative scenario, a backlash to reform Economic Prospects because of the availability programs reduces the long-term growth poten- of new information from household surveys. tial in Asia by about 1.5 percentage points a year. These revisions do not affect the major conclu- The transition economies experience some sions about poverty trends. Extreme poverty weakening of reform momentum that lowers declined only slowly in developing countries long-term productivity, without repeating the during the 1990s: the share of the population disastrous experience of the 1990s. Central living on less than $1 a day fell from 28 per- European countries' accession to the European cent in 1987 to 23 percent in 1998, and the Union is postponed because the global down- number of poor people remained roughly con- turn reduces growth in Europe and increases stant as the population increased.19 The share the perceived costs of accession. For Central and number of people living on less than 34 PROS PE CT S F OR DE V EL O PING CO UN TR I ES AND W OR L D T RA DE Figure 1.26 Nonperforming loans of commercial banks in the East Asia-5 Percent of total commercial bank loans Indonesia _MMarch 1999 _______________ June 2000 Rep. of Korea June 2( 00 Nov. 1998 Malaysia uly 2 )00 Nov. 1999 Philippines July 2000 Thailand -My _ q It ~~~~~~August 2000 0 15 30 45 60 75 ElI Latest estimate U Dec. 1999 * Peak Note: See source for details. Because of the redefinition of nonperforming loans in Korea, there is no estimate for the peak. This excludes loans sold to asset management corporations. Source: ADB, Asia Recovery Report, October 2000, page 11. $2 per day-a more relevant threshold for countries that experienced stagnation or con- middle-income economies such as those of East traction. Indeed, the overall decline in extreme Asia and Latin America-showed roughly sim- poverty during the 1990s was driven by high ilar trends (tables 1.8 and 1.9). rates of growth in countries with large num- It should be emphasized that these historical bers of poor people. For example, China ac- estimates are subject to some uncertainty. Up- counted for a fourth of the total number of to-date survey and price data are not available poor at the start of the decade, and per capita for all countries, and the quality of household GDP during the 1990s rose by 9 percent per surveys can vary considerably among countries year, so by 1998 China's share of the world's and over time. Some country surveys yield in- poor was less than one-fifth. Nevertheless, the come measures of living standards, while others decline in poverty in rapidly growing coun- yield consumption measures, and these two tries was slowed by increases in inequality in a sources are likely to give different poverty esti- number of countries with large numbers of mates for the same underlying population.20 poor, in particular in China, Bangladesh, India, Further, the international measure of poverty and Nigeria.21 Income inequality is an impor- used here is subject to error because of the diffi- tant factor in determining poverty outcomes culties involved in estimating purchasing power (box 1.3). parity exchange rates. Despite these weaknesses, In East Asia, poverty declined most rapidly the estimates provide a fairly reliable view of during the 1990s, falling sharply in China. poverty trends at the aggregate level, because However, growth in China's poorer and more of the substantial increases in the coverage of rural western provinces was much slower than household surveys and in data accuracy over the in the more industrialized east. This diver- past few years. gence reflects slow growth in rural incomes In general, poverty declined in countries related to declining prices for agricultural that achieved rapid growth, and increased in products and reduced opportunities for off- 35 GLOBAL EC ON OMI C PROS PE CTS Table 1.8 Population living on less than $1 per day and head count index in developing countries, 1987,1990, and 1998 Popula"on Number of people living on less than $1 a day covered by at (millions) leat one survey Region (percent) 1987 1990 1998 new 1998 (GEP 2000) East Asia and Pacific 90.8 417.5 452.4 267.1 278.3 Excluding China 71.1 114.1 92.0 53.7 65.1 Europe and Central Asia 81.7 1.1 7.1 17.6 24.0 Latin America and the Caribbean 88.0 63.7 73.8 60.7 78.2 Middle East and North Africa 52.5 9.3 5.7 6.0 5.5 South Asia 97.9 474.4 495.1 521.8 522.0 Sub-Saharan Africa 72.9 217.2 242.3 301.6 290.9 Total 88.1 1,183.2 1,276.4 1,174.9 1,198.9 Excluding China 84.2 879.8 915.9 961.4 985.7 Population Head count index covered by at (percent) kestt one survey Region (perent) 1987 1990 1998 new 1998 (GEP 2000) East Asia and Pacific 90.8 26.6 27.6 14.7 15.3 Excluding China 71.1 23.9 18.5 9.4 11.3 Europe and Central Asia 81.7 0.2 1.6 3.7 5.1 Latin America and the Caribbean 88.0 15.3 16.8 12.1 15.6 Middle East and North Africa 52.5 4.3 2.4 2.1 1.9 South Asia 97.9 44.9 44.0 40.0 40.0 Sub-Saharan Africa 72.9 46.6 47.7 48.1 46.3 Total 88.1 28.3 29.0 23.4 24.0 Excluding China 84.2 28.5 28.1 25.6 26.2 Note: The $1 a day is in 1993 purchasing power parity terms. The numbers are estimated from those countries in each region for which at least one survey was available during the period 1985-98. The proportion of the population covered by such sur- veys is given in column 1. Survey dates often do not coincide with the dates in the above table. To line up with the above dates, the survey estimates were adjusted using the closest available survey for each country and applying the consumption growth rate from national accounts. Using the assumption that the sample of countries covered by surveys is representative of the re- gion as a whole, the numbers of poor are then estimated by region. This assumption is obviously less robust in the regions with the lowest survey coverage. The head count index is the percentage of the population below the poverty line. Further details on data and methodology can be found in Chen and Ravallion 2000. Source: World Bank staff estimates. farm employment. This widening of income In South Asia, the share of the population inequality slowed the rate of poverty reduc- living in poverty declined moderately through tion for the country as a whole.22 Elsewhere in the 1990s, but not sufficiently to reduce the the region, poverty increased in the aftermath absolute number of poor. Household survey of the 1997-98 financial crisis. In Indonesia, data indicate limited growth in average con- the government responded to the crisis by sumption in rural areas, reflecting slow growth strengthening safety nets, which helped cush- in agriculture.23 Urban poverty appears to have ion the impact of the crisis. However, the inci- declined at twice the rate of poverty in rural dence of poverty still increased substantially, areas. However, the Indian poverty data are doubling from its precrisis level. Since early subject to considerable uncertainty. In particu- 1999, there have been indications that poverty lar, private consumption as measured in the has declined significantly as rice prices have national accounts has grown about three fallen, and real wages are starting to recover times faster over the 1990s than household (Suryahadi and others 2000). consumption as measured by the National 36 PROS PE CT S F OR D EVELO PING C OU N TRI ES AND WORLD T RAD E Table 1.9 Population living on less than $2 per day and head count index in developing countries, 1987,1990, and 1998 popuWtiol Number of people living on less than $2 a day covered by at (miillons) ast one suvey Region {percenei 1987 1990 1998 new 1998 (GEP 2000) East Asia and Pacific 90.8 1,052.3 1,084.4 894.9 892.2 Excluding China 71.1 299.9 284.9 252.1 260.1 Europe and Central Asia 81.7 16.3 43.8 98.2 92.9 Latin America and the Caribbean 88.0 147.6 167.2 159.0 182.9 Middle East and North Africa S52 65.1 58.7 85.4 62.4 South Asia 97.9 911.0 976.0 1,094.6 1,095.9 Sub-Saharan Africa 72.9 356.6 388.2 489.3 474.8 Total 88.1 2,549.0 2,718.4 2,811.5 2,801.0 (excluding China) 842 1,796.6 1,918.8 2,178.7 2,168.9 PopuDation Head count index eoverd by at (percent) las one survey Region (pereent) 1987 1990 1998 new 1998 (GEP 2000) East Asia and Pacific 90.8 67.0 66.1 48.7 49.1 Excluding China 71.1 62.9 57.3 44.3 45.0 Europe and Central Asia 81.7 3.6 9.6 20.7 19.9 Latin America and the Caribbean 88.0 35.5 38.1 31.7 36.4 Middle East and North Africa 52.5 30.0 24.8 29.9 21.9 South Asia 97.9 86.3 86.8 83.9 84.0 Sub-Saharan Africa 72.9 76.5 76.4 78.0 75.6 Total 88.1 61.0 61.7 56.1 56.0 Excluding China 84.2 58.2 58.8 57.9 57.6 Note: The $2 a day is in 1993 purchasing power parity terms. See the note to table 1.8. Source: World Bank staff estimates. Sample Survey. Discrepancies are to be ex- been key in holding back the reduction of pected, as the two sources track different ag- poverty in rural areas.25 gregates.24 Moreover, the survey data tend to In Latin America, both the share and the understate the consumption of high-income number of poor declined between 1990 and households. Nevertheless, the size of this dif- 1998. In Brazil, successful stabilization has ference and the slowness of poverty reduction stepped up the reduction of poverty, with the revealed in the survey data are difficult to ac- poor gaining from stronger growth and the count for, particularly given the improvement decrease in inflation. Nonetheless, their liveli- in human development indicators. Thus more hoods remain vulnerable. Evidence from em- accurate data could indicate more rapid pov- ployment surveys in metropolitan areas shows erty reduction than our current estimates. In large swings in poverty, with an upturn in the Bangladesh, steady growth reduced the inci- poverty rate in the wake of the 1997-99 crisis dence of poverty during the 1990s, in contrast and a decrease since late 1999, thanks to the re- to the relative stagnation experienced in the bound in growth. Low educational attainment 1980s. Poverty in urban areas fell at a consid- has helped to perpetuate income inequality and erably faster rate than rural poverty, partly re- poverty by preventing the poor from taking ad- flecting slower growth in rural wages and vantage of opportunities created by growth higher rural unemployment. Landlessness has (World Bank 2000a). 37 GLOBAL ECONOMIC PROSPECTS Box 1.3 Trends in inequality C ountries with high levels of initial inequalitysznallholder agriculture), they benefit mote; if have reduced poverty less for 'given rates of gr&wth takes place in areas or sectors that are growth than countries with low initial inequality not accessible to the Poor. inequality can increase. (World sank 2000d}, and if growth is accompanied Domestic policy distortiojis that hinder agriculture, by increasing inequal4 its inipacron po'rty vi1l I (along with international frade ba'rrier) have re- be reduced. However, our understanding of long- stained growth in rural incomes in many coun- term trends in inequality is linilted, partly because of tries. This has also been reflected in rising' regional weaknesses in the data.5 Trends in iueqtsaliry have inequality, as in poor regions farming is often the been extremely diverse. For exampl, Maja ia saw dominant sector of activity declines in inequality (as measured by the Gini coef- * Changes in incotue ineqiality teflect changes in flie ficient) during the 1980s, bu this treiid>was reversed "distribution of gssers'(for example, education) and in in the 1990s. Korea and hidunesia experienced rapid the return to these asiets. hi some couiitries such as growth during the 1R0s vith little betige t ii- Mexico,n e4ucated workers saw larger increases equality, while China and R a eperieftc'&d large n earnings than did cthers workers, and thesegans increases  lvei the sante peri9d. contributed to ittcreasi income inequality> The available data show no stable relaionsh ' Gender bias and ithe fdrrns of dIscrimination between growth and inequalityJ' On average, income have led to increasing inequality where the groups inequality within countries has neithet decreased nor that are discriminated against are poorer than oth- increased over the last 30 years. Howeve sinot ers to rarr withl For exainple discrminazion led' within-country inequality has increased in sonte. to lower returns to education and lower overall in populous countries, overall more pe4ple have bedn" comes for ethnic minorities in Vietnam and indige affected by increases in inequality than vby 4ecteases " nous groups in brim America. What drives inequality? 1-lere, too, 'l-; * The impact of liberalIzation programs on inequal edge is limited.  both crdescounry. ity has diffeted among countries. If preform con analyses and case studies have generated ighs iit inols'benef,i higher-income 'groups disproportion the link between inequality and se4eral pniicy and ately, fdrms can narrow ineqaality If, or the instituriongl factors " other hand, prereform controls favor the poor lib eralizatin caw have the opposite effct (Ravallion # Policies fostering stable macroeconomic oi,di- 2QO0. For example, in the transition to an open tions, openness to trade, and moderate  of..trade regi''e 'the 'poor may suffer if sector where government rend to stimulate growth hut have they have a'stal& are subjected to competition. been found in one study nor to y reinatically.This ajay haPpen eecialiY in muddle-income d& affect the distribution of income (Dollar nd veloping economies with 'intermediate skill endow Kraay 200(1), However, policies tharredue infla- merits. Thee econqmies ma' have a coinparative non from very high levels appear to benefit the 'advantage regrdihg g&eds that re4uire'mediuni- poor more' than the average intensity skili& These couinries are likely no pro- * If growth is strong in areas where the poor live '' sect secors intensive in unskilled labor where low- and sectors where they ate employed (for example, paidworkers can be £ound.d a. Inequality is estimated with a certain degree t,f uncerrainry as it is based on samLe surveys. Thus changes over time need to be considered carefully to assess whether they are sjgtificanr to a certain dteree or whether diy fail within the margn of error. The estimation of standard errors is complex, and wgl on this' just beginning. 1,. See for example Deininger and Squire I96; ltaeallioii and Chen 1997; runo, Ravllion and 'Squire t99 Dollar 'and Kraay 2000. c. For example, in the Indian state fUttar Pradesh-whith has a population of 160 million and a poverty rate of about 48 per- cent-agriculture accounts for 40 perceistnfGDP and provides 75. percent of eruploynient. d. For example, in Mexico, a country tha tiiplemented one of the most ambitious trade policy reform programs from 1985 'to 1958, the nominal tariff and import licenat coveralle in appel and footwear were among 'the highest' in manufacturing (Revenga' 1995). A similar prereform pattern ofprorectionwa alsofotnid in Morocco Currie and Harrison 1997). 38 PRO SP EC TS F OR D EVE LOP ING C OU N T RI ES AND WORLD T RAD E In Africa, slow growth increased both the crisis. Inequality widened dramatically during share and the number of the poor over the the transition, with the Gini coefficient of con- 1990s; Africa is now the region with the largest sumption expenditure rising from an estimated share of people living on less than $1 per day. 0.24 in 1988 to about 0.49 in 1998. Increas- In Nigeria, the number of people living in ing disparities in poverty across regions have extreme poverty rose steeply following the re- also surfaced, exacerbated by a inefficient sys- versal of the 1985-92 reforms, reaching an es- tem of fiscal decentralization that left the timated 70 million (66 percent of the popula- more backward regions short of resources to tion) based on the national definition (rather assist the poor. than the international $1-a-day definition used Prospects for poverty. As noted above, here). Nigeria now accounts for nearly one- progress in reducing extreme poverty during fourth of Sub-Saharan Africa's poor. Urban the 1990s was constrained by increasing in- poverty has grown faster than rural poverty, equality in a few countries that accounted for owing to massive migration from rural areas to a large share of the world's poor. As in last the cities, with the incidence of urban poverty year's Global Economic Prospects, this year's now matching that of rural poverty. By con- poverty scenarios show that continued in- trast, the rural poverty rate fell in Ethiopia, creases in inequality, coupled with less than Sub-Saharan Africa's second most populous robust growth, would imply failure to reach country and one of the poorest. The reforms the poverty target for developing countries as implemented after the end of the civil war in a group; in particular, the scenarios indicate the early 1990s spurred a strong recovery, end- substantial increases in the number of poor in ing a two-decade slump. The benefits of agri- Sub-Saharan Africa. Given the uncertainty sur- cultural price liberalization have spread quickly, rounding the historical estimates for poverty boosting growth of rural incomes. Urban pov- and the risks associated with long-term growth erty, on the other hand, has been stagnant. projections, these scenarios should not be viewed Urban inequality has risen, in part because of as presenting the full range of poverty rates large population movements resulting from the that are likely to occur. civil war, and in part as a result of economic re- The three poverty scenarios outlined below form, as agricultural price liberalization raised require a projection of growth of the economy consumer prices in urban areas and civil service as a whole (and of population growth), a pro- rationalization reduced urban employment. jection of the average growth rate in per capita Unfortunately, progress is likely to have been consumption for the household sector (mea- slowed by the border conflict. sured by household surveys)26; and a projec- In the Middle East and North Africa, the tion of changes in the distribution of per capita percentage of people living on less than $1 per consumption. day declined slightly, but the proportion living Income growth. The three scenarios differ below $2 per day increased, from 25 to 30 per- only in terms of the assumed growth rate for cent of the population, because of increases in the economy as a whole. Scenario A reflects Egypt, Morocco, and Yemen. the base case growth rates, and scenario B re- Poverty also rose markedly in the transition flects the low case growth rates described above. economies during the 1990s. In the Russian A third scenario assumes that the growth rate Federation, the breakup of the central plan- of each developing-country region is reduced ning system was accompanied by a steep fall proportionately from the low-case forecast, so in output and a sharp increase in inflation. that the average growth for developing coun- Poverty as measured by the national definition tries as a group is equal to that experienced in had jumped from an estimated 11 percent dur- the 1990s (1.7 percent in per capita terms). ing the Soviet period to 43 percent by 1996, Consumption trends. In previous poverty and probably increased further with the 1998 forecasts, the projected growth rate of per 39 G LO B AL EC ONO MI C PRO SP EC TS capita consumption for households was taken the consumption levels of the rich. Thus, the from forecasts of private consumption from projections for India and the ECA region as- the national income accounts. By contrast, the sume that the share of national accounts scenarios outlined below take account of re- growth reflected in the survey mean will equal cent research that shows that the growth in 51 percent over the forecast period, the lower household consumption from survey data has bound of the 95 percent confidence interval been lower on average than private consump- for the estimate for the developing world as tion growth as measured by the national in- a whole (excluding China, India, and Europe come accounts. Data for 142 time periods and Central Asia).29 (during the 1980s and 1990s) for 60 countries Distribution. The other determinant of the suggest that the growth of per capita con- incidence of poverty is in the distribution of sumption from household surveys was an esti- household consumption. Long-term cross- mated 87 percent of the growth rate in private country evidence suggests that most countries consumption from the national accounts.27 have not experienced a systematic trend in The most likely explanation for this discrep- household consumption inequality as mea- ancy is that the surveys do not pick up fully sured using household survey data. Thus, the the growth in living standards of the rich.28 As assumption for the bulk of the developing the poverty estimates are based on consump- countries is that inequality will not change tion from household surveys, we assume in over the forecast period. poverty forecasts for most developing coun- However, there are exceptions. The 1990s tries that the growth rate of this variable will did witness a dramatic rise in inequality in the equal 87 percent of the growth rate of private Europe and Central Asia region. We assume consumption from the national income ac- that this was a transitional phenomenon and counts. The failure to adjust the forecast of will not continue. Further, the available data household consumption growth to reflect the do indicate a rise in inequality in China and historical divergence from the national income India over the past decade,30 in part because of accounts has resulted in substantial overesti- slower growth in rural areas, where the major- mation of the rate of poverty reduction in past ity of the poor live, than in urban areas. We as- forecasting exercises. sume that inequality will continue to rise in The discrepancy between consumption both countries over the forecast period. In growth from the household surveys and the China, the liberalization of trade in agricultural national accounts is larger in China and India commodities and land markets is likely to allow (which together account for more than half of a shift to more remunerative crops and larger the world's poor) and in the Europe and Cen- landholdings. Since good quality land is scarce, tral Asia region. For China, the time series ev- the consolidation of landholdings and higher idence indicates that 72 percent of a gain in returns to good quality land are likely to lead to private consumption is reflected in household higher levels of inequality in rural areas. More- consumption, and this adjustment is used in over, continued integration with the world the projections. For India, only 28 percent of economy will increase the demand for skilled an increase in private consumption is reflected labor. Inequality within urban areas may rise, in the household consumption, and in Europe as wages increase rapidly for skilled workers in and Central Asia the time series evidence for manufacturing and some services while low- the 1990s suggests virtually no correlation be- skill service workers experience lagging wages tween the two consumption aggregates. It is under the twin pressures of migrant laborers difficult to understand these unusually large and laid-off workers from the state enterprises. discrepancies, which probably reflect serious Rising demand for skilled labor may also in- data problems, as well as the failure to capture crease inequality between urban and rural 40 PRO SP EC TS F OR D EVE LO PING C OUI N T RIES AND WO RLD TRADE areas, as the gap in educational attainment be- Table 1.10 Population estimates and tween the two is high. Thus, both scenarios as- projections, developing countries, 1998-2015 sume that urban incomes will increase more (millions of people) rapidly than rural incomes, and that inequality Region 1998 2015 within both the rural and the urban sectors will increase slightly, in the form of a 10 percent East Asia and Pacific 1,817 2,099 Excluding China 569 708 higher Gini coefficient in each sector by 2015. Eastern Europe In India, rising inequality during the 1990s and Central Asia 475 483 appears to have slowed the rate of poverty re- Latin America and the Caribbean 502 623 duction relative to that of the previous decade. Middle East So far, reforms have largely bypassed the econ- and North Africa 286 390 omy in rural areas, where the majority of South Asia 1,305 1,676 Sub-Saharan Africa 627 914 the poor live, leading to a wide divergence of growth between urban and rural areas. Weak Excluding China 3,763 4,794 infrastructure services, limited education, and inadequate health care have made it difficult Source: World Bank staff estimates. for the poor to share equally in the country's rapid growth. For example, the liberalization process is increasing returns to education, while of about 1 billion.31 Finally, if aggregate GDP education is inequitably distributed (one-third growth in developing countries over the next of men, and 60 percent of women, over the 15 years were to equal the average of the age of 15 are illiterate). The forecasts assume 1990s, then progress in poverty reduction that the divergence in consumption growth would be even slower than in scenario B, and between rural and urban areas will continue the number of people living on less than $1 a along past trends. day at the end of the forecast period would be Scenarios. In scenario A, with base case only marginally lower than in 1998. The num- growth (adjusted for historical differences be- ber of poor based on the $2 per day level tween household survey and national income would actually increase. Table 1.11 provides a accounts consumption) and rising household summary of the poverty forecasts, and tables consumption inequality in China and India, 1.12 and 1.13 give regional details for the two the world as a whole would be on track to scenarios that use the base case and low case reach the International Development Goal of growth rates. reducing the share of people living on less than The preceding scenarios highlight the im- $1 per day by 2015 to half of what it was in portance of achieving fast growth and dis- 1990. The total number of poor people would tributing the benefits of growth equitably. decline to about 800 million (see table 1.10 for Without macroeconomic stability, sustained the forecasts of total population in developing structural reforms, prudent and transparent countries). But not all regions would be on use of public resources, improvements in the track: Africa would be far from reaching the provision of public services and infrastructure goal even under this favorable growth sce- to the poor, and actions to reduce vulnerabil- nario. With low case growth rates (scenario B), ity and give the poor more voice in develop- the world as a whole would not reach the tar- ment choices, the pattern of sustained, inclu- get. Only the countries of East Asia would be sive growth that underlies the best scenario able to reduce poverty beyond the target of will not be realized and millions more people half the 1990 incidence. The total number of will remain enslaved in poverty. Achieving the poor people in the world (excluding China) poverty reduction targets also will require an would remain unchanged from the 1990 level increase in aid flows to the poorest countries. 41 G LO B AL EC ONO MI C PROS PE CT S Table 1.11 Poverty in developing countries under scenarios of base case growth (scenario A); low case growth (scenario B); and 1990s average growth, 1990,1998, 2015 $lada? S~~~~~~~~~2 a day Head count rasia Nuasbee of poor Head count ratio Number of poor ~~percept~~ (mil11ou~~~ (percent) (mijUions) 1990 2 _,276 61.7 2,718 1998 3 , 56.1 2,812 2015: scenario A (base case growth) 26 7 36.7 2,272 2015: scenario B (low case growth) 6 1,1143.2 2,672 2015: growth as in 1990s 1. 1,157 47.5 2,938 Source: World Bank staff estimates. Table 1.12 Regional breakdown of number of people living on less than $1 per day and head count index in developing countries, under scenarios of base case growth (scenario A) and low case growth (scenario B), 1990,1998, and 2015 Region X 9 l a East Asia and Pacific 10 0.7 6 Excluding China Europe and Central Asia 7.1 17.6 9. 0 6.3 Latin America and the Caribbean 73. 0 7 8, 42.8 Middle East and North Africa South Asia Bank staff estimat96. Sub-Saharan Africa24.30646.306 Total127.117.1,12765 Excluding China a al, 9ion 93re 720.9 42015 2015 Region 9019 o cs aecs East Asia and Pacific 2.147.83.1 Excluding China 18594281.3 Europe and Central Asia1.3.1913 Latin America and the Caribbean168219.69 Middle East and North Africa2.2.1613 South Asia440402.517 Sub-Saharan Africa 4. 4, 673. Total2.02416426 Excluding China 2. .. 941. Source; World Bank staff estimates. With slow growth and increases in inequality, lion more people worldwide would remain progress would be much slower everywhere, mired in poverty. If policies are inadequate to the target would be out of reach for all regions achieve more than the slow growth of the apart from East Asia, and more than 200 mil- 1990s, then the number of people living in ex- 42 PRO SP EC TS F OR DE V ELO PIN G CO UN T RI ES AND W OR L D T RA DE Table 1.13 Regional breakdown of number of people living on less than $2 per day and head count index in developing countries, under scenarios of base case growth (scenario A) and low case growth (scenario B), 1990, 1998, and 2015 Number of peopl living on less than $2 pe day 201t 2015 Region 1990 199 low case base case East Asia and Pacific 1,084A 884.9 472.2 323.2 Excluding China 284.9 252.1 187.2 114,6 Europe and Central Asia 43.8 98.2 57.6 46.9 Latin America and the Caribbean 167.2 159.0 161.6 132.9 Middle East and North Africa S8.7 85.4 79.7 57.5 South Asia 976.0 1,094.6 1,213.6 1,077.8 Sub-Saharan Africa 388.2 489.3 690.3 636.7 Total 2,718.4 2,811.5 2,675.0 2,275.1 Excluding China 1,918.8 2,178.7 2,390.0 2,066.5 Head wunt sdex (percent) 201$ 2015 Region 199 1998r low case base case East Asia and Pacific 66.1 48.7 22.5 15.4 Excluding China 57.3 44.3 26.4 16.2 Europe and Central Asia 9.6 20.7 11.9 9.7 Latin America and the Caribbean 38.1 31.7 25.9 21.3 Middle East and North Africa 24.8 29.9 20.4 14.7 South Asia 86.8 839 72.4 64.3 Sub-Saharan Africa 76.4 78.0 75.6 69.7 Total 61.7 56.1 43.3 36.8 Excluding China S8.8 57.9 49.9 43.1 Source: World Bank staff estimates. treme poverty would remain near current lev- ing on less than $2 per day in 2015. Thus, the els for the next 15 years. global war on poverty is likely to be with us In Africa, the number of people living in well into the twenty-first century. poverty would increase under all scenarios. If In closing, it is important to note that these the lack of progress observed over the last projections have some serious limitations. First, decade with respect to other dimensions of despite enormous progress in measuring pov- poverty-life expectancy, school enrollment, erty over the past 10 years, the database has and child mortality-continues, as may well be significant weaknesses: recent data are missing the case if the AIDS epidemic is not stemmed, for a number of countries, especially in Africa, then the gap between the region and the rest of where renewed efforts are needed to institu- the world could widen significantly. This would tionalize survey work that began in the 1990s. be a grim outlook, not just for Africa but for Major questions remain as to the trends for the whole world, and efforts are needed in the India. In addition, our understanding of trends region and elsewhere to break with the recent in inequality and the divergence between na- pattern of conflict and crisis, and to deal with tional accounts and household-based measures the AIDS epidemic. of private consumption is limited. Research to Even if the most optimistic scenario is address some of these limitations, including fur- achieved, 2.3 billion people would still be liv- ther analysis of the data for India, is underway. 43 G LOB AL EC ONO MI C PROS PE CT S Notes 15. Oil prices deflated by the U.S. Dollar Manufac- 1. See Gale and Sabelhous 1999. tured Export Unit Value (MUV) index for France, Ger- 2. GSee caale indfelhows (largely portfolioflows many, Japan, the United Kingdom, and the United 2. Gross capital inflows (largely portfolio flows) ex- States. The latter index has been essentially flat over ceeded $750 billion in 1999. See U.S. Department of the 1990s. Commerce, Survey of Current Business, various issues. 16. See appendix 1, Regional Economic Prospects, 3. This figure is strongly influenced by large carry- for further details. over in GDP levels from late 1999 and first-half 2000; 17. The percentage of the population on wages growth on an annualized basis is anticipated to fall 'below subsistence" remains high at 27.6 percent, ac- within a range of 3 percent during the second half of cording to official estimates as of June 2000. However, the year. it has declined significantly from the 34 percent aver- 4. This assumes little or no increase in participation age of 1999. rates from the average of the 1990s. 18. The EU market now accounts for 60 to 80 per- 5. The total population of the EU, assuming all cent of Central and Eastern European countries' exports. countries now under consideration (excluding Turkey) 19. Figures for 1998 were updated in September join, will be close to 500 million in 2010, with the new 2000 using data from surveys that have become avail- members representing close to 25 percent of the total. able only recently, and they differ slightly from the pre- 6. There is an entire literature on strategic linkages liminary estimates included in last year's Global Eco- between multilateral and regional integration agree- nomic Prospects. ments. See World Bank 1999, Section 5, for more de- 20. The estimates of global poverty given here are tailed discussion. based on consumption, and income data are adjusted 7. For more discussion, see Hoekman and Kostecki accordingly. 1995. 21. A common way to measure inequality is to cal- 8. Nonetheless, many developing-country members culate the Gini coefficient. The Gini coefficient would of the WTO-particularly the least developed-still be equal to 0 if all had the same income and to 1 if one face significant impediments in being able to par- person had all the income and everybody else had ticipate fully in the workings of the WTO and other in- none. We observe Gini coefficients for income in the ternational bodies related to international trade (see range of 0.2 to 0.6 (the Slovak Republic has the lowest chapter 3). Capacity building, technical assistance, and Gini, 0.195, while Swaziland and Brazil have the high- financial resources to help developing countries im- est (0.6); among OECD countries, Austria has the low- prove their presence in Geneva are major items on their est Gini at 0.23 and New Zealand the highest at 0.44 agenda for the new post-Uruguay Round negotiations. (World Bank 2000c). 9. Forty-two of the 108 notifications listed in figure 22. The data for China pose several problems. First, 1.13 represent extensions of the EU or NAFTA. consumption per capita, as estimated by surveys, has 10. inceintrregonaltrad isusualy aong been growing less rapidly than estimates of private 10. Since intraregional trade is usually among closer substitutes than extraregional trade, the former consumption from the national accounts would sug- can be more vulnerable to the business cycle. gest. Second, urban household surveys do not include can Theoe vulerfrable tof sthek businetss i cyce, eop rural migrants. Third, savings rates are very high in 11. The performance of stock markets in develop- China, even among the poor, so poverty estimates ing countries was heavily influenced by the technology based on consumption measures yield a higher poverty and telecommunications sectors, which accounted for incidence than those based on income. Moreover, it ap- some 65 percent of total equity placements in the first pence than te b ased on ongcome. pooreover pears that savings rates increased among the poor over half of 2000. this period. The estimates above differ from official es- 12. The 1999 figure of $180 billion reflects a revi- timates, and new survey work will be needed to recon- sion from the estimate of $192 billion presented in cile the differences (work on urban surveys is under Global Development Finance 2000 because of lower way). These discrepancies cast doubt on the estimates levels of inflows to China and Saudi Arabia during the for China and therefore on the global estimates, given year. the size of the country. 13. However, the recent downtrend in FDI to China 23. Unfavorable trends in agriculture partly reflect may be reversed, as the value of approved projects rose inefficiencies in public support to farming, as well as 25 percent year on year during the first five months of limited reform, in contrast to the deregulation of the 2000. urban sector. 14. Global crossborder acquisitions of a more than 24. The major differences are that consumption 10 percent-stake reached $720 billion in 1999, up 35 measures from household surveys sometimes do not in- percent from 1998. clude imputed housing, and private consumption in the 44 PROS PE CT S F OR DEVELOP ING C OU N T RI ES AND WORLD T RAD E national accounts typically includes spending by non- References profit enterprises (nongovernmental organizations, po- Bosworth, Barry P., and Jack E. Triplett. 2000. "Num- litical parties, churches, charities, and so on) as well as bers Matter." Brookings Institution Policy Brief households. 63 (July) Washington D.C. 25. For example, a household with at least 2.5 63(uy. Wahntn D.C. 25 For example,householwithateast25 Bruno, Michael, Martin Ravallion, and Lyn Squire. acres enjoyed 43 percent higher per capita consump- 1998. "Equity and Growth in Developing Coun- tion in 1995-96 than did a landless rural household tries: Old and New Perspectives on the Policy Is- (World Bank 1999). sues." Policy Research Working Paper 1563. 26. This excludes consumption by other private en- World Bank, Washington, D.C. tities such as nonprofit organizations, political parties, Chen, Shaohua, and Martin Ravallion. 2000. "How unincorporated enterprises and so forth that are often Did the World's Poorest Fare in the 1990s?" Pol- included in the national accounts estimate of private icy Research Working Paper 2409. World Bank, consumption. Washington, D.C. 27. See Ravallion 2000. India, China, and Europe Council of Economic Advisors. 2000. "Economic Re- and Central Asia are excluded from this estimation. port of the President." Washington, D.C., GOP. 28. There is a presumption that higher-income groups Available at http://w3.access.gpo.gov/usbudget/ tend to underreport consumption. Moreover, consump- fy2001/pdf/2000 erp.pdf. tion measures from household surveys sometimes do Currie, Janet, and Ann Harrison. 1997. "Sharing the not include imputed housing expenditures, which, in Costs: the Impact of Trade Reform on Capital fast-growing economies, are likely to grow rapidly in and Labor in Morocco." Journal of Labor Eco- the higher-income groups. There are other explanations nomics (U.S.) 15:S44-S71. for the divergence between survey mean household con- Deininger, Klaus, and Lynn Squire. 1996. "A New Data sumption growth rates and those from the national Set Measuring Income Inequality." The World accounts, including the fact that (for most countries) Bank Economic Review 10(3):565-91 (Septem- private consumption in the national accounts includes ber). Data available at hrtp://www.worldbank.org/ spending by nonprofit enterprises (such as churches and research/growth/dddeisgu.htm charities and so on) as well as households, and the share Dollar, David, and Aart C. Kraay. 2000. "Growth Is of the nonprofit sector is probably rising. Good for the Poor." Policy Research Working 29. This assumption has an important impact on Paper (forthcoming). World Bank, Washington, the forecasts. For example, if in India consumption D.C. Available at http://www.worldbank.org/ were assumed to rise at 87 percent (rather than 51 per- research/growth/pdfiles/growthgoodforpoorpdf. cent) of the national income accounts growth rate, then Gale, William G., and John Sabelhaus. 1999. "Per- by 2015 the forecast (using base case growth rates) for spectives on the Household Saving Rate." Brook- extreme poverty in South Asia would be only about ings Papers on Economic Activity (1):181-224. half the 22 percent rate shown for scenario A. Gordon, Robert J. Forthcoming. "Does the 'New' Econ- 30. For example, the per capita consumption (as omy Measure up to the Great Inventions of the measured in the household survey) of the bottom 10 Past?" Draft for Journal of Economic Perspectives. percent of China's population increased by 2.5 percent IMF (International Monetary Fund). 2000. World Eco- per year during 1990-98, while per capita consump- nomic Outlook: 2000. Preliminary edition, Sep- tion of the top 10 percent increased by 11 percent per tember. Washington, D.C. year. In India, per capita consumption of the bottom 10 Jorgenson, Dale W., and Kevin J. Stiroh. 2000. "Indus- percent did not increase at all during 1985-97, while try-level Productivity and Competitiveness be- the top 10 percent saw a rise of 4.7 percent per year. tween Canada and the United States: U.S. Eco- 31. These results are roughly similar to the poverty nomic Growth at the Industry Level." American forecasts in Global Economic Prospects 2000. In sce- Economic Review 90 (2): 161-7. nario A, the head count poverty index is 15.9 percent Kasman, Bruce. 2000. "U.S. productivity growth: It's in 2008 (the last year of the GEP 2000 forecasts), com- breadth that matters." J. P. Morgan Global Data pared with 12.3 percent in GEP 2000. This year's fore- Watch. 18 August, 7-10. Available online to sub- cast is more pessimistic because we assume that growth scribers at http://www/morganmarkets.com. in household consumption will be slower than in pri- Kotlikoff, Laurence J., and others. 2000. vate consumption in the national income accounts. OECD (Organisation for Economic Co-operation and Conversely, the GEP 2000 forecast for scenario B was Development). 1999. Economic Survey of Japan. more pessimistic (head count index of 21.9 percent November. Paris. compared with 19.5 percent now), because last year we Oliner, Stephen D., and Daniel E. Sichel. 2000. "The assumed a rise in inequality in all regions. Resurgence of Growth in the Late 1990s: Is In- 45 G LOBA L EC ONO MI C PROS PE CT S formation Technology the Story?" Federal Re- ness Economics, April 1999. 34(2):13-17. Also serve Board. May. Washington, D.C. available at http://www.brook.edu/es/research/ Ravallion, Martin. 2000. "A Note on Forecasting areas/productivity/papers/trip4 99.pdf. Poverty Using National Accounts Growth Rates." U.S. Department of Commerce. 1996. "Prospects for A background paper for Global Economics Pros- Growth in Japan in the 21st Century." Office of pects and the Developing Countries 2000. the Chief Economist Economics and Statistics Ad- Ravallion, Martin, and Shaohua Chen. 1997. "What ministration Research Series: OMA 2-96, No- Can New Survey Data Tell Us about Recent vember. Washington DC. Changes in Distribution and Poverty?" World . Various years. Survey of Current Business. Bank Economic Review 11(2):357-82. Washington, D.C. Revenga Ana. 1995. "Employment and Wage Effects UNAIDS (United Nations 2000. Report on the Global HI VIAIDS Epidemic. Geneva: joint United Na- of Trade Liberalization: The Case of Mexican tions Programme on HV/AIDS), http://www.unaids. Manufacturing." Policy Research Working Paper org/epidemic_update/report/Epi report.pdf. 1524 World Bank, Washington, D.C. World Bank. 1999. Shah, Shekhar, Martin Ravallion, and Quentin Wodon. World Bank. 2000a. Attacking Brazil's Poverty: A 1999. Bangladesh: from Counting the Poor to Framework for Sustainable Poverty Reduction- Making the Poor Count. A World Bank country with a Focus on Urban Areas. Forthcoming. Wash- study. Washington, D.C. ington, D.C. Suryahadi, Asep, and others. 2000. "The Evolution of . 2000b. Global Economic Prospects and the Poverty during the Crisis in Indonesia, 1996 to Developing Countries 2000. Washington, D.C. 1999 (Using Full Susenas Sample)." Social Moni- . 2000c. World Development Indicators 2000. toring and Early Response Unit Working Paper. Washington, D.C. Jakarta. Available at http://www.smeru.or.id/data/ . 2000d. World Development Report 2000/01. report/evolpov2.pdf. Washington, D.C. Triplett, Jack E. 1999. "Economic Statistics, the New . 2000e. Trade Blocs. New York: Oxford Uni- Economy, and the Productivity Slowdown." Busi- versity Press. 46 PRO SP EC TS F OR DE V EL O PING CO UN TR I ES AND W OR L D T RA DE Annex 1 Membership of selected major regional integration agreements (RIAs) and dates of formation INDUSTRIAL AND DEVELOPING ECONOMIES European Union (EU): formerly European Economic Community (EEC) and European Community (EC), 1957: Belgium, France, Germany, Italy, Luxembourg, Netherlands; 1973: Denmark, Ireland, United Kingdom; 1981: Greece; 1986: Portugal, Spain; 1995: Austria, Finland, Sweden. European Economic Area (EEA): 1994: EU, Iceland, Liechtenstein, Norway. Euro-Mediterranean Economic Area (Euro-Maghreb): Bilateral agreements, 1995: EU and Tunisia; 1996: EU and Morocco. EU bilateral agreements with Eastern Europe: 1994: EC and Hungary, Poland, 1995: EC and Bulgaria, Romania, Estonia, Latvia, Lithuania, Czech Republic, Slovak Republic, Slovenia. Canada-US Free Trade Area (CUSFTA): 1988: Canada, United States. North American Free Trade Area (NAFTA): 1994: Canada, Mexico, United States. Asia Pacific Economic Cooperation (APEC): 1989: Australia, Brunei Darussalam, Canada, Indonesia, Japan, the Republic of Korea, Malaysia, New Zealand, Philippines, Singapore, Thailand, United States; 1991: China, Taiwan (China), Hong Kong (China); 1993: Mexico, Papua New Guinea; 1994: Chile; 1998: Peru, the Russian Federation, Vietnam. LATTN AMERICA AND THE CARIBBEAN Andean Pact: 1969: revived in 1991, Bolivia, Colombia, Ecuador, Peru, Republica Bolivariana de Venezuela. Central American Common Market (CACM): 1960: revived in 1993, El Salvador, Guatemala, Honduras, Nicaragua; 1962: Costa Rica. Southern Cone Common Market, Mercado Comun del Sur (Mercosur): 1991: Argentina, Brazil, Paraguay, Uruguay. Group of Three (G-3): 1995: Colombia, Mexico, Republica Bolivariana de Venezuela. Latin American Integration Association (LAIA): formerly Latin American Free Trade Area (LAFTA), 1960: revived 1980, Ar- gentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Mexico, Paraguay, Peru, Uruguay, Venezuela. Caribbean Community and Common Market (CARICOMk: 1973: Antigua and Barbuda, Barbados, Jamaica, St. Kitts and Nevis, Trinidad and Tobago; 1974: Belize, Dominica, Grenada, Montserrat, St. Lucia, St. Vincent and the Grenadines; 1983: The Bahamas (part of the Caribbean Community but not of the Common Market). MIDDLE EAST AND ASIA Association of Southeast Asian Nations (ASEAN): 1967: ASEAN Free Trade Area or AFTA was created in 1992, Indonesia, Malaysia, Philippines, Singapore, Thailand; 1984: Brunei Darussalam; 1995: Vietnam; 1997: Myanmar, Lao People's Democratic Republic; 1999: Cambodia. Gulf Cooperation Council (GCC): 1981: Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, the United Arab Emirates. South Asian Association for Regional Cooperation (SAARC): 1985: Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, Sri Lanka. AFRICA Cross-Border Initiative (CBI): 1992: Burundi, Comoros, Kenya, Madagascar, Malawi, Mauritius, Namibia, Rwanda, Seychelles, Swaziland, Tanzania, Uganda, Zambia, Zimbabwe. East African Cooperation (EAC): 1967: formerly East African Community (EAC), broke up in 1977 and recently revived, Kenya, Tanzania, Uganda. Economic and Monetary Community of Central Africa (CEMAC): 1994: formerly Union Douaniere et Economique de l'Afrique Centrale (UDEAC), 1966: Cameroon, Central African Republic, Chad, Congo, Gabon; 1989: Equatorial Guinea. Economic Community of West African States (ECOWAS): 1975: Benin, Burkina Faso, Cape Verde, COte d'lvoire, Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone, Togo. Common Market for Eastern and Southern Africa (COMESA): 1993: Angola, Burundi, Comoros, Djibouti, Egypt, Ethiopia, Kenya, Lesotho, Malawi, Mauritius, Mozambique, Rwanda, Somalia, Sudan, Swaziland, Tanzania, Uganda, Zambia, Zimbabwe. Indian Ocean Commission (IOC): 1984: Comoros, Madagascar, Mauritius, Seychelles. Southern African Development Community (SADC): 1980: formerly known as the Southern African Development Coordination Conference (SADCC), Angola, Botswana, Lesotho, Malawi, Mozambique, Swaziland, Tanzania, Zambia, Zimbabwe; 1990: Namibia; 1994: South Africa; 1995: Mauritius; 1998: Democratic Republic of the Congo, Seychelles. Economic Community of West Africa (CEAO): 1973: revived in 1994 as UEMOA, Benin, Burkina Faso, Cote d'lvoire, Mali, Mauritania, Niger, Senegal. West African Economic and Monetary Union (UEMOA or WAEMU): 1994: Benin, Burkina Faso, C6te d'lvoire, Mali, Niger, Senegal, Togo, 1997: Guinea-Bissau. Southern African Customs Union (SACU): 1910: Botswana, Lesotho, Namibia, South Africa, Swaziland. Economic Community of the Countries of the Great Lakes (CEPGL): 1976: Burundi, Rwanda, Democratic Republic of the Congo. Source: World Bank 1999. 47 Trade Policies in the 1990s and the Poorest Countries URING THE 1960s AND 1970s, DEVELOP- This chapter reviews the export and ing economies exhibited severe trade- growth performance of developing countries related distortions, including quantita- in the 1990s, giving special attention to the tive restrictions on imports and exports, very poorest economies. It reviews the decline in high tariffs, overvalued exchange rates, and trade barriers during the 1990s, examines out- administrative controls on foreign exchange put and export trends, and analyzes domestic allocation. Although growth remained rapid trade-policy constraints. Finally, it considers against a background of a favorable external external barriers to developing countries' ef- environment up to the first oil shock in 1973, forts to accelerate their export growth. policies of import control and substitution in- The chapter reaches the following con- duced inefficiencies as well as rigidities in eco- clusions: nomic structure. Often, they resulted in peri- odic balance of payments crises. Subsequently, * Trade regimes were significantly liberal- the failure of many countries to adjust ade- ized during the late 1980s and 1990s. De- quately to the external shocks of the 1970s veloping countries cut the average tariff and early 1980s underlined the importance of rate by half, narrowed tariff dispersion in reforms designed to encourage responsiveness many instances, and greatly reduced the to market signals, improve the investment cli- incidence of nontariff barriers to trade. mate, and to enhance export diversification. Most countries now rely on market forces One result was that most developing coun- rather than administrative fiat to allocate tries implemented significant liberalization of foreign exchange, and black market pre- their trade regimes during the late 1980s and miums have declined significantly. Al- the 1990s. Their export growth accelerated though the degree of trade protection is during the 1990s, and kept pace with the 6 still high in many developing countries, percent per year expansion of world trade gross distortions in trade regimes have in volume terms. However, average per capita been greatly reduced. growth rates in developing countries as a group * Despite the reforms and improved global remained well below those of the rich countries economic conditions, developing coun- in the 1990s. Though the giant low-income tries' average real per capita incomes in- countries China and India embarked on market creased by less than 1 percent per year reforms and grew rapidly, growth in a large during the 1990s, compared with more number of small, poor countries was disap- than 2 percent in industrial countries. pointing. This led many observers to question This outcome is partially affected by the the success of liberalization programs. political shocks and various foreign and 49 G LO B AL EC ONO MI C PRO SP EC TS civil conflicts. Eighteen developing coun- couraging rapid diversification. Appreci- tries were severely affected by conflict, ated real exchange rates and high real ex- and as a group they suffered declines in change rate volatility have often been per capita income of more than 1 percent associated with a muted export response per year. Incomes also fell in most transi- to trade liberalization and other reform tion economies following the breakup of measures. Per capita income growth was the Soviet Union. Excluding countries hit significantly faster in countries with rela- by these severe political shocks, develop- tively stable real exchange rates. Addition- ing countries saw per capita incomes rise ally, institutional weaknesses, such as by 1.5 percent a year, about 1 per cent the absence of effective duty exemption/ faster than in the 1980s. drawback programs, coupled with the * These countries also saw merchandise ex- need to use revenues from tariffs on inter- port growth of 6.4 percent per year dur- mediate and capital goods as a revenue ing the 1990s, about 2 percent faster than source, have acted as an effective tax on during the 1980s. Export outcomes were exports in many of the poorest developing very uneven, however. Regions that saw countries. Finally, weak export infrastruc- the largest declines in trade barriers, in- ture, inadequate ancillary export services, cluding East Asia, South Asia, and Latin and high transport costs-often in part the America also saw the largest acceleration result of policy shortcomings-have left in exports. By contrast, growth in export many of the poorest countries (particu- volumes in Sub-Saharan Africa averaged larly the landlocked ones) at a competitive only 2 percent per year, in part because disadvantage on international markets. world trade of the products Africa exports * External barriers to exports from devel- grew at half the rate of growth of world oping countries, especially agricultural and trade. Countries in Sub-Saharan Africa and labor-intensive products, continue to im- in the Middle East and North Africa also pede the integration of the poorest eco- saw market share decline in their tradi- nomies into the world market. While the tional exports. share of developing countries in world * The poorest countries were those most af- trade of manufactures rose sharply in the fected by conflict and political shocks. Ten 1990s, their share of world trade in agri- of the 32 low-income countries were af- cultural products and processed foods has fected by conflict. Average per capita in- declined. In part this development is the come of the low-income, small countries result of domestic policies that restrain declined during the 1990s, but averaged agricultural exports. But high trade bar- 1 percent a year in real terms if countries riers imposed by industrial countries on involved in conflict are excluded. This rep- agriculture and processed food imports, resents a significant acceleration compared and agricultural subsidies in industrial with the 1980s, but is still well below the countries, are also important and have be- average of middle-income countries. Ex- come even more important with the do- ports also accelerated in the small low- mestic policy reforms in developing coun- income countries not involved in conflict, tries. These barriers particularly penalize but grew about 3.5 percent slower than in rural areas where the majority of the the middle-income countries in the 1990s. poor in developing countries reside, and * Despite significant progress, many of the impede growth in the poorest countries poorest countries have not put in place the in areas of their comparative advantage. policies necessary to raise living standards Various restrictions and subsidies in in- by improving (or even maintaining) ex- dustrial countries also hamper the poorest port shares in traditional markets and en- countries' efforts to diversify into down- 50 T RA DE PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST C O U N T R I E S stream processing, higher value added, reduced its average tariff from 100 percent and faster-growing products. The poorest in 1986 to around 33 percent in 1998, and countries are the least equipped to deal Bangladesh's average tariff fell from 82 per- with these external barriers because a cent to 24 percent during the same period. host of other domestic policy and institu- Tariffs in Africa have also been reduced, tional weaknesses inhibit their diversifica- though more moderately: the average un- tion into less restricted sectors. weighted tariff remains at almost 20 percent. Some countries have embarked on rapid re- forms, however. Tariffs fell in Kenya from Reductions in barriers to trade 41 percent in 1980-85 to 13.5 percent in D eveloping countries made substantial 1996-99 and in Guinea from 76.4 percent in D progress in reforming their trade and ex- 1978-80 to 10.8 percent in 1990-95. A few change rate policies during the 1990s. Tariffs countries, such as Zimbabwe, increased their were cut, their dispersion declined in many average tariffs (from 10 percent in 1980-85 to countries, fewer products were covered by 22.7 percent in 1996-99), reflecting in part the quantitative restrictions (QRs), the number of conversion of an import surcharge into tariffs.3 countries allocating foreign exchange through The Middle East and North Africa region has administrative means (as measured by the In- seen little reduction in average tariffs, although ternational Monetary Fund) dwindled, and the signing of trade agreements with the Euro- the black market premium narrowed. Although pean Union in recent years by several countries any one of these measures is an imperfect guide in the region should eventually pave the way to the restrictiveness of the trade regime,1 for significant reductions. taken together they show enormous progress Nontariff barriers (NTBs). Developing in opening the developing economies to in- countries in all regions have substantially re- ternational trade. Several studies have found duced the coverage of NTBs (such as licensing, that increased openness is associated with eco- prohibitions, quotas, and administered pricing) nomic growth (box 2.1). during the 1990s (table 2.2).4 In many coun- Tariffs. The average tariff rate in develop- tries where NTBs remain more or less preva- ing countries has been cut by at least half in lent (for example, India and Korea), commit- the last 20 years, from 32 percent in the first ments have been made to liberalize them in the half of the 1980s to 15.6 percent in the second future (Michalopoulos 1999). NTB coverage half of the 1990s. Tariff reductions have been has been reduced significantly in a number of significant in most regions (figure 2.1), but Latin American countries, with the remaining have been largest in South Asia (where, how- NTBs mainly on agricultural products (Dorn- ever, they remain the highest of any region), busch and Edwards 1995). Most countries in Latin America, and East Asia. In many devel- South Asia have reduced NTBs significantly, oping countries, the degree of tariff dispersion though India still has restrictions on a relatively has also declined (table 2.1). Despite this prog- large number of imports. ress, tariffs remain high in many countries, av- Excbange rate regimes. The 1990s also eraging more than 15 percent in three of the saw a general move toward market-based for- six geographical regions in 1996-98. eign exchange regimes. In 1991, 66 countries Latin American countries reduced tariffs had restrictions on payments for current ac- substantially during the 1990s. Colombia, for count transactions, but by 1995 only 28 did example, slashed import tariffs by 65 percent (table 2.3). Many countries also undertook sig- in just one year in 1991, while Argentina and nificant exchange rate reforms. The average Nicaragua reduced them from an average of black market premium for developing coun- 110 percent to 15 percent in one bold move in tries, one indicator of the restrictiveness of for- 1992 (Dornbusch and Edwards 1995). India eign exchange allocations as well as of macro- 51 G LO BAL EC ON OM IC PROS PE CT S economic imbalances, fell almost 70 percent and North Africa, the considerable reductions between the 1980s and the 1990s.5 in black market premiums in many countries The average black market premium hss aeas toeA masked by two outliers (table 2.4). remained low or fallen further in all regions. Reforms by income groi p. The trend to- East Asian countries t htad low black mar- ward lowering tariffs and eliminating foreign ket premiums in recent years because they exchange restrictions is evident in both low- were early adopters of an outward-oriented income and middle-income countries. The development strategy. Latin America also average tariff rate for low-income countries achieved impressive gains in the late 1980s, as fell from almost 45 percent in the early 1980s did South Asia in the first half of the 1990s. In to 20 percent in the late 1990s, only slightly Sub-Saharan Africa and in the Middle East above the average rate for middle-income 52 T RA DE PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST CO UN TR I ES Figure 2.1 Average unweighted tariff rates by region Percent 70 60 _ 1980-85 * 1986-90 D 1991-95 Cl 199648 | 60_ 50 40 20 10 0 South Asia Latin America East Asia Sub-Saharan Middle East Europe and Industrial and the and Pacific Africa and Central Asia economies Caribbean North Africa Source: World Bank data, and World Trade Organization data. Table 2.1 Standard deviation of tariff rates countries (figure 2.2). By the late 1990s, most 1.991>-94 1995 98 of the low-income countries had eliminated current account restrictions and reduced the South Asia black market premium to negligible levels. Bangladesh 114.0 14.6 India 39.4 12.7 SriLanka 18.1 15.4 Trends in trade and economic Sub-Saharan Africa growth South Africa 11.3 7.2 Malawi 15.5 11.6 ; he lowering of trade barriers and wide- Zimbabwe 6.4 17.8 T spread adoption of market-based foreign East Asia and Pacific exchange regimes during the 1990s was ac- Philippines 28.2 10.2 Thailand 25.0 8.9 Indonesia 16.1 16.6 China 29.9 13.0 Table 2.2 Frequency of total core nontariff Latin America measures for developing countries, and the Caribbean 1989-98 Argentina 5.0 6.9 Brazil 17.3 7.3 Region 1989-94 1995-98 Colombia 8.3 6.2 Mexico 4.4 13.5 East Asia and Pacific (7) 30.1 16.3 Middle East and Nortb Africa Latin America and the Caribbean (13) 18.3 8.0 Egypt, Arab Rep. of 425.8 28.9 South Asia No.a (4) 43836 Tunisia 37.4 11.7 Sub-Saharan Africa (12) 26.0 10.4 Turkey 35.7 5.7 Notes: Average number of commodities subject to nontariff Notes: Country observations are for one year in the time measures as a percentage of total. Figures in parentheses are period noted above. the number of countries in each region for which data are Source: World Bank, World Development Indicators 1998; available. 2000; WTO, Trade Policy Reviews. Source: Michalopoulos 1999. 53 G LO B AL EC ONO MI C PRO SP EC TS Table 2.3 Countries imposing restrictions on payments for current account transactions (percent) Region F 1991 East Asia and Pacific (9) 33e a, 33 South Asia (5) ~Ua aa 100a Middle East and North America m6) pm67u Sub-Saharan Africa (23) 83 Latin America and the Caribbean (30) Ia60n Eucope and Central Asia (17) an 94 Industrialized economies (12) h de8 Total (102) 65 2 Notes: Figures in parentheses are the number of countries in each regional grouping. Source: IMF, Exchange Arrangements and Exchange Restrictions 1981 1992d 1996i Table 2.4 Average black market premium (percent) Region 1990a9 Total' ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~a78.2 East Asia and Pacific a m a a 3.6 Middle East and North Africa atility 351.6 a thr Excluding outliersb a a a i a 8.8 Latin America and the Caribbean 13.1 ya South Asia a 45.1 Sub-Saharan Africa 28.6 Excluding Nigeria 2 5c.8 a. Sample of 41 developing countries. b. Algeria and the Islamic Republic of Iran. Source: World Bank data. companied by an acceleration of trade in de- export expansion-has somnetimes been cited veloping countries. The improvement in the in- as evidence that increased integration with the ternational environment in the 1990s, marked global economy has not helped developing by lower intereSt rates and inflation but higher countries, particularly the poorest. In fact, out- world trade growth and non-oil commodity put declines in countries that were hit by se- prices than in the 1980s, also contributed to vere political shocks, including the breakup the acceleration of exports (table 2.5). More- of the Soviet Union and various external or over, developing countries faced a more stable internal conflicts, had a large effect on the global economy in the 1990s, as the volatility developing-country aggregates, and the rise of of these key indicators declined. The growth of incomes in countries that did not suffer these developing-country exports almost doubled in calamities was considerable (table 2.6). Exclud- real terms, to 6.2 percent a year in the 1990s, ing the transition economies and countries compared to 3.7 percent a year in the 1980s.6 involved in conflict, per capita GDP growth Per capita income also accelerated, though mnore rates in developing countries averaged 1.5 per- modestly, from no growth in real terms in the cent per year during the 1990s, versus 0.4 per- 1980s to 0.7 percent in the 1990s. cent in the 1980s, and export growth rates The observation that developing countries averaged 6.4 percent in the 1990s, compared as a group continued to exhibit relatively slow with 4.3 percent in the 1980s. Furthermore, growth in per capita income in the 1990s- while developing countries became more open about one-third the rate of advance in the rich during the 1990s, volatility declined in most countries, despite large-scale liberalization and countries (see box 2.2). 54 T RA DE PO LI C I ES IN T HE I 9 9 Os AND T HE PO ORE ST C O U N T R I E S Table 2.5 The international environment Figure 2.2 Merchandise export and 1980 1990-98 GDP per capita growth in developing countries in the 1990s u.s. real LIBOR 4.S 2.4 Percent peryear G-7 CPI (percent per year) 5,3 2.7 G-7 real GDP growth (percent per year) 2.8 2.1 7 World export growth (percent per year) 4.4 6.4 Oil price (index, 1987 = 100) 13SA 113.1 6 Non-oil commodities price (index, 1987=100) 115.6 123.0 5 Note: The G-7 countries are Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States. 4 Source: World Bank data. 3 in tariffs during the 1990s also saw the most 2 rapid increases in GDP and exports.7 GDP and export growth rates more than doubled in 1 Latin America in the late 1980s and 1990s 0 (table 2.7), a period that followed large ad- Developing countries Developing countries vances in trade liberalization as well as the 184 countres] excluding ECA and countries in conflict adoption of important macroeconomic reforms [59 countnes] and widespread privatization. South Asia and * Merchandise export growth, average 1990-98 East Asia also saw very sharp increases in ex- | GDP per capita growth, average 1990-98 port growth rates and continued high GDP growth. By contrast, countries in Sub-Saharan Source: World Bank data. Africa and in the Middle East and North Af- rica made relatively less progress in reducing tariffs, and both GDP and export growth was With the exception of the countries in considerably slower than in the other regions. transition and conflict, there is little evidence Factors affecting developing country ex- that developing regions that engaged in rapid ports. The acceleration of exports of develop- trade liberalization and companion reforms ing countries in the 1 990s reflected higher saw a deterioration in performance. In fact, de- growth of world trade (itself a reflection in veloping regions that saw the largest reductions part of the increased integration of developing Table 2.6 GDP and merchandise export growth rates (percent per year, in constant prices; group simple average)a 'GDP PF cpita Merchandise exportsb Number of 1t989 199098 1980-89 1990-98 00sw4d World 0.6 1.2 4.1 6.4 133 Industrial countries 2.1 2.2 5.6 6.8 30 Countries that lack consistent data -0.4 1.8 3.6 6.4 20 Developing countries (excluding those that lack consistent data) 0.3 0.7 3.7 6.2 83 Europe and Central Asiac 2.6 -1.1 0.4 7.6 6 Countries in conflict -1.1 -13 3.0 5.0 18 Developing countries excluding ECA and countries in conflict 04 1.5 4.3 6.4 59 a. All growth rates are estimated using least squares for the sample periods. b. Merchandise exports are deflated by constant 1987 U.S. dollar export prices. c. To give comparable data over 1980s and 1990s, the republics of the former Soviet Union are treated as one country. Source: World Bank data. 55 G LO B AL EC ONO MI C PRO SP EC TS r he incrsnopennress, andel accgelter dvriiation of dh oa iveaniialcrtisis of nre in8-§. u the MdecdleEs grow theduring thexports were assoiatrd with d oftheGD9cpita as rchan c Aerit ed bMeuchadSarae oxpir a smal decliegiadotnan inroed asketi s hae oailit shaditi tha t soedvrsftont, butdt tw g loalrgedessions andmr of output nd exportgrowth indevelopin cou 198e dramatica 9ly afSecte (~impvolat piity deflaoil xb Centraaresltbl .8. Wie ot Asia, (EA u elfrohrdseo gEast kae oe to * sh are b in traiton ld ofrkets countrie as a goup andin threeof the ive 98eo 1e990e debts t h 1990S 1o98i0te to9a90aSi Asamand stigi and Latin low erolatility inw bthieasthes red Sub-Saha anAfi6 7sthnd out b s t nmmcsoc tanetshe i9n0g The traitia mret re of n g icount~ exen smunder sl eregrowth of world (198756S5)a 1287 USS0 15pr peet All countries 4.5 4~~ ~ ~~~~~~~.2 15. .2 14.107. 15.2 Developing 5.0 4~~ ~ ~~~~~~~.4 17.6 16.0 20.2 17.1 tercountries) 6.4elasgraer4.1siiato 24.8vesi 19.6on 28.0rie 18.1 idlEs D heveoinr excludin base and, founriesoedvlpn anNotAfiandSbahrnficsw Sapegin,4.3vd akt hr 3.5 15.1iion11.2m 18.0iictin bu33.2eelieinmr mres(al28)8WieSouth Asia2.1t2.6 11.8 e 8.2 14.0tina 10ret.0 Ai,adLtnA eiaswbtinraeinSub-Saharan Africa 4.6a3.7 18.4 2.3a20.2 14. rgosimrvdmarket share in traditionalmaktan some dierifcaion,eprine buch aslawrgerdelin in mar-d 56 TR AD E PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST CO UN TR I ES Table 2.7 GDP, services, and merchandise export growth rates (percent per year, in constant prices; group simple average) Exports GEIPW Merchandise Goods and servicesa Number of t9809 19S 1950-89 1990-98 1980-89 1990-98 - : 1- couatries Totalb 3.1 3S 4.3 6.4 4.5 6.3 .. 41 59 East Asia and Pacific 5.4 6.3 5.7 14.5 5.9 14.3 4 7 Middle East and North Africa 3.6 3.4 5.3 6.0 5.5 5.0 9 Latin America and the Caribbean 1.8 3.6 3.9 9.1 4.8 8.1 t. -.. 14 South Asia 4.9 5.2 5.8 9.3 5.0 10.3 . a Z ; 5 Sub-Saharan Africa 2.5 2,9. 3.5 2.0 3.4 2.6 $ 24 a. All growth rates are simple arithmetic averages of individual rates computed using least squares. Nonfactor services are de- flated by constant 1987 GDP deflator; other series are expressed in constant U.S. dollars. b. Refers to a sample of 59 developing countries. c. Denotes average annual growth rate of GNFS export deflated by merchandise import prices. Source: World Bank data. Table 2.8 Decomposition of merchandise export growth for the sample countries (percent per year, current U.S. dollars) 1980/81-1989/90 World TOW Trade Market Total ; Expo8ts Growth Share 1 sif.aeilm Exports Diversification Total 3.9 4.1 -1.0 0. . .' - - 0.6 East Asia and Pacific 8.5 5.2 2.1 11.9 6.S . 0.1 Middle East and North Africa 2.4 3.1 -1.3 4 .6 3.0 5" 4 , 1 0 5 Latin America and the Caribbean 3.9 5.4 -2.2 0.8 .3 61 0s - 0.9 South Asia 8.9 5.9 1.5 I 014- 10.0 - .2.4 0.6 Sub-Saharan Africa 1.0 2.9 -1.6 0.8 3.0 . `- 0.6 Source: COMTRADE database, and authors' calculations. trade in its export basket in both decades than significant increases in per capita incomes in other developing regions. Reflecting the contin- the 1990s, per capita income declined in the ued decline in primary commodity prices and 1990s in the poorest developing countries as low-income elasticities of demand, world trade a group (figure 2.3). Since 1980, 10 of the 32 growth rates during the 1990s were less than low-income countries with consistent data 2 percent annually for exports of some of have been involved in foreign or civil wars (see the poorest African countries, including Benin, box 2.3). In 1999, one African in five lived in Chad, Mali, and Mauritania, among others. a country severely disrupted by conflict (World Trends in the poorest developing coun- Bank 2000a). Excluding countries affected by tries. Excluding China and India, which saw conflict, per capita GDP growth rates in small 57 G LO B AL EC ONO MI C PRO SP EC TS X, gae 3 2 M erchandise export growth and (DP per capita growth min poor developing 11`ok,i.i-,,ls!, the 390s g Percent per year 5 * Merchandise export growth, average 1990-98 4 ;; 4 _ A ~~~~~~~GDP per capita growth, average 1990 98 |~ 3 2 0 -1 Low-income small Low-income small Least-developed Least-developed countries [32 countries] countries excluding countries countries excluding countries in conilict [26 countries] countries in conflict [22 countries] [16 countries] Source: World Bank data. low-income countries averaged 1 percent per ing countries affected by conflict) averaged year during the 1990s, well below the 1.5 per- only 4.1 percent per year during the 1990s, up cent rate of growth achieved by middle-income from 2.9 percent in the 1980s but still well countries (table 2.9). below performance in the middle-income coun- Merchandise export volume growth rates tries. More than two-thirds of these small low- in the 22 small, low-income countries (exclud- income countries (16 of 22) are in Sub-Saharan 58 T RA DE PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST CO UN TR I ES Table 2.9 Growth rates by income level for the sample countries (percent per year) GDP : Merchandise exports I GPper capita Number of isO. 19 1980s 1990s 19f 1990s countries Total 3.1 3.8 4.3 6.4 0.4 1.5 59 Low income (large countries) 7.9 8.4 6.7 14.1 6. 6.9 2 Low income (small countries) 2.7 3.6 2.9 4.1 -0.1 l.0 22 Least developed 2.2 3.5 2.9 3.2 -0.6 0.8 16 Middle income 3.0 3.6 5.1 7.4 0.5 1.5 35 Note: The sample (59 developing countries) excludes countries in conflict, transition economies, and those with limited data. Source: World Bank. Africa. An important factor accounting for the the export sector include overvalued real ex- slow growth of the poor countries' exports was change rates, inconsistent or erratic macroeco- slow growth of world trade in their traditional nomic policies, a high share of government export baskets. Several of the poorest develop- spending in aggregate expenditure, excessive ing countries lost market share in traditional reliance on tariff revenues and on taxes on exports and were unable to diversify rapidly agriculture, and a variety of direct public inter- enough into new products. ventions in product and factor markets, espe- Summarizing, developing countries not in cially price controls and requirements that bank conflict and not in transition saw significant credit be allocated to the public sector (World acceleration in incomes and exports in the Bank 1989, 1994, 2000a). Weak infrastructure 1990s, following adoption of trade and com- and inadequate provision of ancillary services panion reforms. The poorest small countries to exports also severely constrain export per- were among those most affected by conflicts formance in many countries. As discussed in and political shocks. The poorest small coun- previous sections, some of the most severe pol- tries that avoided these calamities performed icy distortions present in developing countries better, but still not as well as the middle- in the 1980s are being alleviated: average in- income countries, on average. Thus, despite sig- flation and fiscal deficits have declined signifi- nificant progress, it appears that policy regimes cantly (Easterly 2000); real exchange rates in many of the poorest countries are still in- have been adjusted in many cases; tariffs and adequate to improve or even maintain export quotas have been cut; and export taxes and re- competitiveness in traditional markets or to strictions on agriculture have been reduced or encourage rapid diversification. The next two eliminated in many countries. sections will discuss the key trade-related pol- This section examines three aspects of icy impediments to rapid integration of the trade-related domestic policies that remain im- poorest countries in the world economy, in- portant impediments to integration of many of cluding trade barriers to their exports in indus- the poorest developing countries into the global trial countries. economy: overvalued and unstable real ex- change rates, high cost of imported inputs for exporters, and the higher costs of transporta- Weaknesses in domestic trade- tion and other trade-supporting infrastructure. related policies Exchange rate management. A competi- Domestic policies that directly restrict ex- tive and stable real exchange rate ensures that D ports or deter investment in the export producing tradable goods is profitable, and to sector remain important in most of the poorest underpin investor confidence in the export sec- developing countries (Yeats and others 1997). tor. In many of the poorest countries, pro- Policies that tend to discourage investment in tracted episodes of real exchange rate overval- 59 G LO B AL EC ONO MI C PROS PE CT S uation have impaired the competitive position tries depreciated during the 1990s (Easterly of local firms. At different times, they have 2000), and many countries adopted more flex- contributed to unsustainable trade deficits, ible nominal exchange rate arrangements.10 In and prompted increased tariffs, quantitative some cases, for example the CFA countries restrictions, and foreign exchange controls in (see box 2.4), these efforts resulted in a sharp futile attempts to contain these deficits. Epi- turnaround in economic performance. sodes of overvaluation have also resulted in Several of the poorest countries failed to foreign exchange crises, economic instability, achieve stable and competitive real exchange and the deterioration of growth over the long rates during the 1990s. Overall, the average run. At the same time, unsustainable macro- volatility of real exchange rates in the small, economic policies have also contributed to the low-income countries increased. High exchange overvaluation of real exchange rates (Bhagwati rate volatility was associated with weak eco- 1978; Krueger 1978; Papageorgiou, Choksi, and nomic performance. Among the countries in Michaely 1991; and Rodrik 1996). our sample (excluding countries affected by The effects of real exchange rate overval- conflict), the poor countries that exhibited low uation have been studied extensively. Dollar exchange rate volatility saw exports rise at over (1992) shows that Africa and Latin America 6 percent a year and per capita incomes rise had relatively overvalued exchange rates and at nearly 2 percent a year, much faster than large real exchange volatility, while East Asia countries with high exchange rate volatility had relatively undervalued exchange rates (figure 2.4). lower real exchange rate volatility, and more As countries became more open to inter- rapid growth. Using a wide variety of mea- national trade, the adverse effects of real ex- surements and techniques, econometric stud- change rate volatility on stability and growth ies have fairly uniformly found that overval- may have become even more severe. In some uation and associated real exchange rate cases, trade liberalization combined with sharp volatility are negatively correlated with both real exchange appreciation severely impaired export and GDP growth.9 the profitability of domestic firms.11 Ex- The factors that lie behind protracted change rate overvaluation and high volatility episodes of real exchange rate overvaluation, also has a negative effect on investor con- despite awareness of their adverse implica- fidence and can delay the supply response to tions for growth, are varied. They include use liberalization.12 Figure 2.5 shows six coun- of the exchange rate as a nominal anchor for tries in Africa that moved to flexible exchange the stabilization program and difficulties in rates and liberalized their trade regimes during exiting from the peg; rigid exchange rate the early 1990s. These countries subsequently regimes that fail to accommodate secular dete- experienced appreciation and high volatility rioration in the terms of trade and adjust to of the real exchange rates and low rates of widening productivity differentials; dispropor- growth. Per capita income growth in these six tionate influence of urban elites who are con- countries averaged only 0.5 percent a year in sumers of imported consumer goods; concerns the 1990s, and their export effort (changes in about the effect on the urban poor dependent market share and diversification) fell by 2.3 on imported food staples or energy; and aver- percent a year. sion to incur the immediate fiscal costs reval- The causes of high real exchange rate uating foreign currency liabilities. volatility have been widely explored. They in- Perhaps because of increased openness, in clude unstable or inconsistent macroeconomic recent years policymakers appear to have policies, which result in inflationary pressures become more aware of the need to maintain and high fiscal and current account deficits. competitive and stable real exchange rates. The Many countries have tried to maintain stable average exchange rate of developing coun- nominal exchange rates, despite the formal 60 TR AD E P O L I C I ES IN T HE 1 9 9 0s A N D T H E PO ORE ST CO UN TR I ES 4A thngrtenr1aoithe CFA , 'at -'C,PA- - ~~ X sW4 *:AAs p_S nt s-r-k- i annua ratef inceae in exportsdue to rin -e--- s __=X X = ;< - wse-iom~~~~~~~e-4.8- Th. 1W JUS*~~po'' ktZ ifhares anid'diversification, w from pnt~tferyea in 98693 to7.8 ecn per ya -a-o- Exohange rates and CFA countries - - - - of $e 9t i . - .---1986-93 1994-98 Average GDP growth 0.7 4.7 r;i4~at)ia t _taWp a o-r Average REER index 99.4 63.0 t*_ dfl4&eehot F The CPA Snc Was de- Export effort5 -4.8 7.8 . - Note: The countries incduded are Burlina Paso, Camneroon, Central African Republic, Chad, C6te d'lvoire, Gabon, rtratkct ti~~ Mali, Niger, Senegal, antd Togo. a. The merchandise exrt growth rate achieved through r as changes in market share and export diversification. will e:e tina of world rra* incte CA xi Source: IMF, International Financial Statistics. adoption of more flexible exchange rate re- Thus, the record of the 1990s underlines gimes during the 1990s (Calvo and Reinhart the difficulties that poor countries face in main- 2000).13 However, attempts to maintain nom- taining a stable and competitive real exchange inal exchange rate stability in the face of high rate, reflecting their thin foreign exchange and variable inflation rates or various domes- markets, their vulnerability to weather-related tic or foreign shocks imply that the real ex- shocks, and their dependence on primary change rate will be volatile. For example, 17 commodity exports that are subject to sharp of the 26 countries that exhibited high real ex- changes in price.15 change rate volatility were also rated by the Exchange rate policies are hampered by World Bank as exhibiting relatively unstable the lack of clear signposts as to the exchange macroeconomic policies.14 Among the 22 low- rate regime that minimizes real exchange rate income countries in our sample, only two volatility. The choice of exchange rate regime managed to achieve both macroeconomic sta- will depend on many factors, including the bility and low real exchange rate volatility. currency composition of public and private In addition, terms-of-trade shocks are an external debt and domestic financial assets, important factor contributing to high real ex- the track record and credibility of the mone- change rate volatility in commodity-dependent tary authorities, and the nature of external countries. Hausman and others (1999) cite the and internal shocks.16 volatility of capital flows as a driving force be- Free trade status for exporters. High tar- hind the volatility of exchange rates in many iffs coupled with inefficiencies in customs and middle-income countries, particularly in con- tax administration have increased the costs of junction with large debts denominated in for- exporting from many developing countries. eign currency. Access to inputs at world prices is critical to 61 G LO B AL EC ONO MI C PRO SP EC TS Figure 2.4 Real effective exchange rate volatility and growth in the 1990s Percent per year 9 8 _ | * ~~~~~~~~~~~Merchandise export growth, average 199G-98L L | * ~~~GDP per capita growth, average 1990-98 l 7 6 5 4 3 2 0 Countries with high Countries with low Low-income small Low-income small volatility in 1990s volatility in 1990s countries with high countries with low [26 countries] [33 countries] volatility in 1 990s volatility in 1 990s [ 13 countries] ]9 countries] Source: World Bank data; IMF, International Financial Statistics. export competitiveness, so that high tariffs on ensure that reimbursement or exemption is intermediate and capital goods can reduce ex- granted quickly and fairly. ports, particularly of processed goods that re- By contrast, the poorest developing coun- quire substantial intermediate inputs. At the tries have faced severe difficulties in ensuring same time, many developing countries depend duty-free access to inputs by exporters, for two on tariffs for an important share of govern- reasons. First, a substantial share of govern- ment revenues. Thus governments often face a ment revenues in many of these countries is gen- difficult tradeoff between the desire to encour- erated from taxes on international trade. To re- age export production through low tariffs on duce distortions affecting domestic production imported inputs and the need to generate suf- and to simplify administration, tariff reform in ficient revenues. many poor countries involves increasing low Successful exporters such as the Asian tariffs while reducing peak tariffs (Falvey and newly industrializing countries, in the early Kim 2000; Harberger 1988). As a result, many stages of their export drive, have enabled ex- of the poor countries have significant tariffs on port firms to obtain their inputs at world mar- intermediate and capital goods (table 2.10). Al- ket prices through reimbursing tariff duties on though the average tariff rates for this sample of inputs (duty drawback systems), providing ex- 15 African countries are not high compared emptions on duties for exporters (duty exemp- with those of many other developing countries tion systems), and setting up export process- (figure 2.1), the duties on capital and interme- ing zones where intermediate inputs can enter diate goods do represent a considerable tax on the country free of duty.17 These systems have export production.18 These duties are essen- been effective in reducing or eliminating tariffs tially designed to collect revenues, rather than on exporters' inputs, while maintaining a tar- to protect domestic production.19 iff structure that meets the revenue needs of Second, most of the poorer countries lack the government. These systems, however, re- the administrative resources required to estab- quire considerable administrative resources to lish effective drawback/exemption schemes. 62 TR AD E PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST CO UN TR I ES Figure 2.5 Real effective exchange rate behavior in selected poor countries, 1993-99 Ghana Kenya 100 140 130A AA 90 120 120 ~~~~~~~~~~~~~120 80 ~~~~~~~~~~~~~~110 100 70 ~~~~~~~~~~~~~~90 80 - v 60 t70 1, II I 1993 1994 1995 1996 1997 1998 1999 1993 1994 1995 1996 1997 1998 1999 Malawi Tanzania 120 160 100 A140 80 1200 60 100 40 80 1993 1994 1995 1996 1997 1998 1999 1993 1994 1995 1996 1997 1998 1999 Zambia Nigeria 180 200 140 A 8A 160 12014 120 10012 100 80 v ~~~~~~~~~~~~~~80 60 ~~~~~~~~~~~~~~60, 1993 1994 1995 1996 1997 1998 1999 1993 1994 1995 1996 1997 1998 1999 Note: 1990= 100, upward movement indicates appreciation. Source: IMF, International Financial Statistics; World Bank data. 63 G LO B AL EC ONO MI C PRO SP EC TS Table 2.10 Tariffs in selected African countries Tanzana UTrade weighted andShare of Attempts toesalihsuhytemaverage tariff revenues All goods 16.2 14100.0 Capital goods 13.2 g r no 12.9 Intermediate goods 14e2 42.4 Consumer goods imposed5.3 a 22.4 Note: Textiles, energy and passenger cars are not incloded in capital, intermediate, and consumer goods. Data are for 1996 for Benin, Borkina Faso, Cameroon Cape Verde, Coted dvoire, Ghana, Malawi, Maoritius, Senegal, South African Customs Union, Tanzania, Uganda, Zambia, and Zimbabwe; data for Mali are for 1997. Source: World Bank data. Attempts to establish such systems during the move to a tariff structure that has zero or very 1990s were generally not successful. The re- low rates for capital and intermediate goods quirement of examining individual import along with much lower rates for consumer transactions imposed a substantial administra- goods (coupled with reduced exemptions). tive burden on exporters and the government. Infrastructure for trade expansion. Weak In many countries, the government failed to infrastructure constrains economic growth in provide (or significantly delayed) promised re- many of the poorest countries and has a severe imbursements of duties paid on imported in- impact on trade performance. A firm's ability puts.20 Even exporters' value added taxes were to compete in the world economy depends in not reimbursed consistently. Ongoing work on part on the cost and availability of supporting 15 reforming Sub-Saharan African countries services, such as transport, communications, shows that none had a working duty drawback and finance. While some of the problems these system. Mauritius operated a successful export countries face in providing adequate infra- processing zone program, while Cape Verde structure and other services are to some extent and South Africa had low duties on intermedi- the result of location, technological deficien- ate goods. The 12 other countries imposed cies, and a lack of capital, many of these prob- significant duties on inputs to export produc- lems result from restrictive regulatory policies tion without effective means of compensating that limit competition. 22 exporters. Many of the least-developed countries have There is no easy solution for poor coun- weak and expensive service suppliers. Amjadi, tries attempting to ensure duty-free access to Reincke, and Yeats (1996) conclude that high inputs by exporters.21 Further efforts are re- transport costs in low-income African countries quired to strengthen the administration of cus- are a more important trade barrier than tar- toms and to increase reliance on other sources iffS.23 These high costs add to the cost of ex- of revenues such as income taxes. However, porting, and (other things being equal) lower its such efforts take time. In the short term, de- profitability in many of these countries. Coun- pending on the structure of export activity (for tries in Africa normally absorb all, or most, of example, degree of processing) and the level the transport charges for penetrating external of institutional development, governmnents markets. Africa's net freight and insurance pay- need to decide whether drawback/exemption ments in 1990/91 were about $3.9 billion, ap- systems or free trade zones are feasible, or al- proximately 15 percent of the total value of ternatively whether reductions in intermediate the region's exports. For developing countries and capital goods tariffs are necessary to en- as a whole similar payments averaged 5.8 per- courage export production. Since all but five of cent, about one-third of Africa's ratio. Thus, in the countries in table 2.10 have total collection order to be competitive, an average producer in rates of less than 10 percent it is possible to Africa has to be 10 percent more efficient than 64 T RA DE PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST C O U N T R I E S firms in other developing countries. For the 10 ple, allowing exporters to charter their own landlocked countries in Africa, average net vessels in Cote d'Lvoire halved the costs of freight and insurance payments in 1990 were shipping bananas to the United States, and re- 42 percent of their total exports, eight times the duced cocoa freight costs one-quarter (World average for other developing countries. Bank 2000a). Improving the operating effi- Limao and Venables (1999) also find that ciency of the Nacala rail line through Mozam- higher transport costs and weak infrastructure bique, with little new investment, would in- explain a significant portion of Africa's poor crease the GDP of Malawi by 3 percent. These trade performance. This is especially true for efforts may have a significant role in improv- landlocked countries. Exporters from land- ing export performance over the next decade locked economies face final city destinations (AfDB 1999). Donors also are engaged in the that are on average four times further from the efforts to help the low-income countries to sea than that of exporters from coastal eco- overcome the major institutional impediments nomies. The median transport cost for land- that inhibit their integration into the world locked countries is 58 percent higher than the economy (see box 2.5). median for coastal countries. Delays and coor- dination problems at the border, higher insur- ance costs due to uncertainty and delays, and Protection in industrial countries direct charges made by transit countries also Jmport restrictions and subsidies in industrial add to the transport costs of landlocked coun- Icountries limit the growth of developing tries. Improving transport infrastructure in countries' exports by supporting less efficient landlocked countries and their transit coun- production in industrial countries. Export sub- tries can dramatically reduce costs and increase sidies to industrial country producers further trade flows. Improving a country's worldwide limit the expansion of developing-country ex- rank from the 75th percentile to the 50th per- ports to third markets. Moreover, these poli- centile in the distribution of infrastructure cies make it difficult for developing countries quality would double the volume of trade. to diversify into products for which world de- Poor availability of communications and mand is high or increasing, and in line with energy can also constrain exports. Unreliable their evolving competitive advantage. service can be even more damaging to com- Industrial-country import restrictions. Al- petitiveness than high costs. Production stop- though the average tariffs in the Quad (Can- pages, missed delivery dates, and lack of re- ada, European Union, Japan, and the United liable communications make it difficult to States) countries range from only 4.3 percent compete and become part of global produc- in Japan to 8.3 percent in Canada, their tariffs tion networks. Despite liberalization of the fi- and trade barriers remain much higher on nancial systems in many countries, access to many products exported by developing coun- credit is limited, interest rate spreads are high, tries (Finger and Laird 1987). The Uruguay and in many countries weakness in regulatory Round contributed to a sharp decline in the institutions have kept financial sector reforms use of nontariff barriers (NTBs) in the OECD. from having their anticipated benefits. In the Quad, only 1.2 percent of tariff lines Reforms are being undertaken in many are subject to NTBs. However, most of the developing countries to liberalize the policy NTBs are found in agriculture (tariff quotas, regimes for these sectors. Foreign participa- for example)25 and textiles and clothing (Multi- tion is being allowed to improve the level of fiber Arrangement), where developing coun- technology and efficiency, and also to generate tries have a comparative advantage (Finger the financing for the required investments. and Schuknecht 1999). The potential gains from these reforms and re- Products with high tariffs in Quad coun- sulting investments are very large. For exam- tries include (1) major agricultural staple food 65 G LO B AL EC ONO MI C PRO SP EC TS products, such as meat, sugar, milk, dairy prod- percent range for a large number of products. ucts, and chocolate, where tariff rates fre- These are sectors in which developing countries quently exceed 100 percent; (2) tobacco and have a comparative advantage. some alcoholic beverages; (3) fruits and vegeta- For example, in the United States only 31 1 bles-including 180 percent for above-quota of 5,000 tariff lines are above 15 percent. 26 bananas in the European Union and 550 per- Yet 15 percent of exports from least-developed cent and 132 percent for shelled groundnuts countries to the United States face these tariffs in Japan and the United States, respectively; (Hoekman, Ng, and Olarreaga 2000). Thus, (4) food industry products, including fruit j'uices, there might be considerable potential for the canned meat, peanut butter, and sugar confec- least-developed countries to increase their ex- tionery, with rates exceeding 30 percent in ports if U.S. tariffs were reduced. several markets; and (5) textiles, clothing, and Some of the highest tariff rates in indus- footwear, where tariff rates are in the 15 to 30 trial countries are applied to products that are 66 T RA DE PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST C OU N T R I E S Table 2.11 Developing-country exports to erential margins and imposes country or sector Quad countries facing tariffs of more than quotas, or both (Michalopoulos 1999). 50 percent The potential for growing exports in the Average most favored nation tariff M% 113 restricted categories is illustrated by those de- Range of tariffs (%) 50-343 veloping countries that have managed to accel- Exports to Quad (billion US$) 5.0 eaeterarclua xot lruhfe Share of developing countries in erate their agricultural exports through free total imports of Quad (%) 10 trade arrangements with industrial countries Exports to the world (billion US$) 26.6 (see chapter 1 for a discussion of the benefits of Source: DECD, 2000b; UN COMTRADE; and staff North-South regional trade agreements). For calculations. example, since Mexico joined NAFTA in 1994, the value of its agricultural exports has in- creased by 15 percent per year-twice the rate typically exported by developing countries. of the previous five years. After joining the EU For example, almost $26 billion of exports in 1986, Portugal's agricultural exports grew from developing countries in 1999 to the by 9.2 percent per year in nominal U.S. dollars, world were products that would have faced and Spain's exports grew by 10.9 percent. In tariffs above 50 percent in the Quad countries. the prior five years, both Portugal and Spain Only about $5 billion of that sum was actually had seen declining export earnings from agri- exported to the Quad countries. On the other culture. These experiences suggest that further hand, Quad countries imported about $50 bil- trade liberalization in agriculture is likely to lion of the same goods, most of it from other have a major effect on developing countries' industrial countries.27 This suggests some po- abilities to increase agricultural exports. tential for developing countries to expand ex- Developing-country trade barriers. Devel- ports of these products to the Quad if the tar- oping countries' exports are subject to much iffs were lowered. Although $26 billion is only higher trade barriers in other developing coun- about 2 percent of total developing-country tries than in industrial countries. The average exports, individual developing countries (in- tariff developing countries face in their exports cluding among the least-developed countries of manufactures to other developing countries in the above example) may be more affected. is 12.8 percent, more than three times the av- The tariff rates in table 2.11 do not reflect erage tariff on their manufactured exports to the Generalized System of Preferences (GSP) industrial countries (table 2.12) (Hertel and and other preferential schemes operated by Martin 1999). Tariffs in developing countries Quad countries. These schemes tend to reduce the tariff rates applicable to imports of these products from some developing countries. Table 2.12 Average tariff rates by North-South preferential trade agreements importing and exporting region such as NAFTA or the Euro-Mediterranean (percent) Agreements also provide duty-free entry for some developing countries.28 However, in most cases where tariff peaks are present, the sensi- Exporting region o1 tes tivity of domestic industry to imports exclude various products from preferential schemes High-income 10.9 limit the amount that can be imported under Developing 3.4 12- 8 the preferential rates, or restricts the number of World 1.5 11.5 Agriculture countries that are eligible. For example, the High-income .9 21.5 United States has no GSP or least-developed Developing 15.1 18.3 country preference for products facing tariffs World 15.6 20.1 of more than 50 percent.29 The EU limits pref- Source: Hertel and Martin 2000. 67 G LO B AL EC ONO MI C PRO SP EC TS Table 2.13 Producer support estimates for OECD countries Percent 0 0 0 US$ billion :1:E986-88i;;'9900 1997-99 00 1986-88 1997-99 Lower-income OECD countries 37.5 28.7 28.8 39.4 Australia, New Zealand 9.5 4.s 1.8 1.4 Canada, United States 2. 18.5 47.5 47.8 European Union 44. 0 44.0 95.2 116.6 Other non-EU OECD countries X 66.7 7.8 7.8 Japan 67. 0 5610 53.6 53.1 OECD 4. 34.9 234.8 266.2 Source: OECD 2000a. are somewhat higher in agricultural markets as countries to spur rapid export growth. The well. These high tariffs are becoming all the more successful middle-income countries have more important to developing countries, as the achieved rapid progress by increasing exports share of South-South trade in their exports has of labor-intensive manufactures, in part replac- risen from about 26 percent in 1980 to 40 per- ing production in industrial countries. This de- cent in 1999. celeration or decline in industrial countries' Industrial-country agricultural subsidies. share in labor-intensive manufactures has al- Agricultural subsidies in industrial countries lowed developing countries to increase their limit the growth of developing countries' agricul- exports at a much higher rate than the growth tural exports by supporting inefficient producers. of world income.32 By allowing the least-cost Furthermore, surplus agricultural products have producers to capture a greater share of de- been exported at a loss, further reducing the op- mand, the efficiency of global production has portunities for many developing-country pro- increased. Despite remaining quotas and other ducers in third markets. Estimates of agricultural restrictions on many labor-intensive products producer support declined as a share of gross (such as textiles), the shares of developing farm receipts between the mid-1980s and the countries' manufacturing exports in world mid-1990s. Over the past few years, however, trade (figure 2.6) and in the consumption of the when agricultural prices have declined, subsidies Quad countries have increased. But even in have actually increased.30 By 1999 the average 1995 the shares were not high-ranging from producer support estimate reached 40 percent, 6.8 percent of consumption in Canada and the almost equal to the average of 1986-88 (ta- United States to only 3.4 percent in Japan. ble 2.13). Virtually all OECD countries, except Tariff escalation in industrial-country New Zealand and Australia, have increased their markets has restricted the market access of support levels. The biggest increases in the rate of developing-country producers of finished goods, protection took place in lower-income OECD thus hampering industrialization.33 The Uru- countries, but the largest subsidy in absolute guay Round has made some progress in re- terms was given by the EU. During 1997-99, the ducing the degree of overall tariff escalation. average annual value of subsidies was about 60 Yet, in a number of sectors (such as food pro- percent of total world trade in agriculture, and cessing) that are of particular interest to de- almost twice the value of agricultural exports veloping countries, high levels of tariff escala- from developing countries.31 tion are still present (box 2.6). Many products The impact of trade restrictions on di- are also protected by some form of quota. As versification and trade growth. The lowering some of these quotas are allocated on the basis of industrial country trade barriers has com- of historical trade shares, new and more effi- bined with improved efficiency in developing cient countries cannot enter these markets. 68 T RA DE PO LI c I ES IN T HE 1 9 9 Os AND T HE PO OR EST CO UN TR I ES Figure 2.6 Imports of manufactures from developing countries as a percentage of apparent consumption Percent 8 | 1980 * 1985 El 1990 E 1995| 7 6 5 4 3 2 0 European Union United States and Canada Japan Source. UNCTAD 1996. Developing-country exporters of agricul- growth and poverty reduction. Many of the tural products have not achieved even the lim- poorest countries are still primarily agricul- ited penetration of industrial-country markets tural exporters, 40 to 60 percent of their pop- that occurred in manufactures. The share of de- ulation (and the majority of the poor-World veloping countries' agricultural exports in world Bank 2000c) lives in rural areas, and expan- trade have actually decreased (figure 2.7). The sion of agricultural exports is one of their few relatively low level of developing-country agri- avenues to accelerating growth (at least in the cultural exports can be attributed in part to medium term), given their level of technology greater protection and subsidies in agriculture and human capital base. Many of the poorest than in manufacturing. Higher protection al- countries made progress during the 1990s in lows only the most efficient agricultural pro- removing domestic policy constraints on agri- ducers in developing countries to enter in- cultural exports; however, policy reforms and dustrial country markets and relatively more investments in rural areas, which are neces- inefficient producers in industrial countries to sary for poverty alleviation, are unlikely to maintain their market share. The success of yield significant improvements unless the de- many developing countries in products that face mand for many of these products can be ex- lower protection and subsidies, such as cut flow- panded through exports to world markets. ers from Africa, and more stable trade shares in Numerical estimates of gains from re- fruits and vegetables also suggest that if protec- duced agricultural protection vary consider- tion in agriculture is lowered, many of the poor- ably, but reducing protection could have a est countries could expand their exports.34 particularly important impact on the poorest Implications for the poorest countries, countries. lanchovichina, Mattoo, and Olar- While external constraints are not the primary reaga (2000) show that if all the Quad coun- reason for slow export growth and declining tries gave free trade access to the low- terms of trade of the poorest countries,35 nev- income African countries, their net exports ertheless, industrial-country trade restrictions would increase about 6 percent. The negative and subsidies are having an adverse impact on impact of this expansion on other developing 69 G LO B AL EC ONO MI C PRO SP EC TS )H> ~~~~~~ ci~>k~1s ~u4 ~< Imports of processed food from developing - Figure 2.7 Share of developing countries as a percentage of apparent in world traeceasisversmllncmpaistconsumption / < Percent *M 1980 ~E1985 *1990 ~19 253 2 T~~~~~~~~~ arf lessrdcalassmtion thnin food mauacterin 20 nada 7 U dapan European Union United States Japan and Canada First ~ ~ r stageer 3n poin 35d obudrie Semiprocessed 8 18 36 ~~~~~Source: UNCTAD 1996. FtOly processed 4n 2g te atrial bariers woud be Figure 2r7Share Aof developo countie.i Mosi of these lo Manufactureslcountries andvonwindustrial-countriesgwouldab n world trade crease is very small in comparison to the Percent world trade. 36 Almost all the gains come from 40 the expansion of agricultural trade. Hertel, Hoekman, and Martin (2000) report that a 40 percent reduction in industrial countries' agri- 3970 19751980 1985199 5cultural tariffs and export subsidies by 2005 (a less radical assumption than in the paper 20 ~~~~~~~~~cited above) would increase income in most developing regions by less than 1 percent. However, one point needs to be underlined 10 when discussing the trade barriers facing the Agriculture exports of the poorest counties. Most of these - -Manufacture countries have weak export-supporting infra- 0 structure and skills that make it difficult to 1970 1975 1980 1985 1990 1995 switch from domestic markets to exporting, and Source: FAOSTAT, World Bank data, vice versa, in response to changes in relative prices, or to diversify out of traditional com- 70 T RA DE PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST C O U N T R I E S modities facing artificially constrained mar- mium may also reflect the degree of overvaluation as kets. These weaknesses are reflected in little well as capital account restrictions. export-market diversification, imperfect capital 6. This chapter uses simple averages, which give equal weight to individual countries regardless of size. These figures differ from the weighted average growth timely credit, low capitalization of many enter- rates reported in chapter 1 and annex 1, which give prises, small numbers of firms, and a very diffi- greater weight to bigger countries. China and India in cult environment for transactions. The fact that Asia, and South Africa and Nigeria in Africa, dominate many of these countries have a history of policy the regional averages. Weighted averages are appropri- reversals and mismanagement implies that the ate for measuring the total regional increase in GDP or credibility that must underpin investment is exports, while simple averages are appropriate in ana- lackin hsthyaenamchwaelyzing the impact of policy on growth (since the rela- lacking. Thus they are in a much weaker posi- tionship between policy and performance is just as rel- tion than some of the more successful countries evant for a small country as for a large one). to overcome industrial countries' barriers to 7. Tariff reductions are clearly not the only deter- their traditional exports through diversification. minant of export growth rates, and some regions with relatively high tariff rates (such as South Asia) had rapid growth in exports. Nevertheless, the reductions Notes in tariff rates during the 1990s were, on average, asso- 1. For example, import-weighted calculations of ciated with rapid export growth. average tariffs may understate the importance of very S. To measure the role of world trade conditions as high rates that block all imports, while the number of opposed to other factors that determine export perfor- QRs may not accurately reflect their impact on trade. mance, export growth (in nominal dollars) is decom- Moreover, data availability for these series is uneven posed into three components using COMTRADE data: both over time and across countries. Nevertheless, they (a) growth caused by the expansion of world trade for are preferable indicators of trade policy than many products defined as traditional exports in the initial pe- used in the literature (see Edwards 1997). Some of riod, (b) gains in the share of individual countries in them (such as share of trade to GDP, or growth of ex- world trade in these commodities, and (c) growth attrib- ports) are endogenous variables that depend on loca- utable to diversification. COMTRADE data are not tion, country size, and other policies; some rank trade identical to the trade data used in the rest of the report. regimes as perceived by users (the Heritage index), They are based on the reported imports from OECD and while others measure the degree of trade distortions other reporting countries and cover about 90 percent of only at one point in time (the Sachs-Warner index). world trade. Because most countries in Sub-Saharan Most of these measures have even more limited coun- Africa and the Commonwealth of Independent States try and time period coverage. (CIS) do not report their imports fully, they underesti- 2. See Barro and Sala-i Martin 1995 for a review of mate intra-African and intra-CIS trade. The decomposi- the econometric literature. Sensitivity of results to both tion into growth in traditional markets, increased mar- variables and specifications are highlighted by Levine and ket share, and diversification depend on the definition of Renelt 1992 and Sala-i Martin 1997a, 1997b. the traditional or initial export basket. In table 2.8 the 3. One concern is that in many developing coun- export baskets of 1979-80 and 1989-90 were taken and tries tariff bindings have been applied to only a small all exports above US$5 million were classified as major. percentage of tariff lines, and wherever applied, these Growth of new exports and exports that were less than ceilings in general have been higher than the actual ap- $5 million during the initial period are classified as di- plied tariffs. In a sample of 42 countries, the bound versified. (See Yeats 1998, and Ng and Yeats 2000 for a ceilings were 49 percent, while the average applied rate similar exercise.) was only 18 percent. The large gap between tariff bind- 9. See Dollar 1992; Elbadawi 1998; Ghura and ings and actual levels may make the government sus- Grenness 1993; Razin and Collins 1997; and Sekkat ceptible to protectionist pressures and creates uncer- and Varoudakis 2000. Edwards and Savastano (1999) tainty about the future direction of reforms. have a critical review of the measures of exchange rate 4. These averages need to be used with caution, as misalignment. See also Hinkle and Montiel 1999 and data are available for only a few countries in each re- Williamson 1994 for a detailed analysis of the mea- gion and may not exist for both the time periods under surement issues and economic implications of ex- consideration, change rate misalignment. 5. The black market premium is used here to mea- 10. According to IMF, 97 percent of its members in sure the administrative allocation of foreign exchange 1970 were classified as having a pegged exchange rate; for trade purposes. However, a large black market pre- by 1999, this share had come down to very low levels 71 G LO B AL EC ONO MI C PRO SP EC TS (Calvo and Reinhart 2000). As of 1999, of the 185 18. There has been significant tariff reduction since countries reporting their exchange rate arrangements, 1996 in most of these countries. Also, there are large dif- 84 had pegged regimes, 75 had floating rates, and 26 ferences among the countries included (the trade- managed a system of limited flexibility. Of the countries weighted tariff ranges from 25 percent in Zimbabwe to with floating rates, 27 maintained a managed float, and about 7 percent in the South African Customs Union). 48 had an independent float (Shatz and Tarr 2000). 19. Many of these products do not have local sub- 11. The fears of deindustrialization by many stitutes, and it is unlikely that they will be produced in African countries are usually attributed to trade re- these countries over the medium term. In that sense, form, while the exchange rate behavior probably they are pure revenue tariffs. played an even more important role. 20. See Nash and Fourotan 1997 for the experi- 12. Some studies have found a negative relation- ences in Africa, and Rajapatirana 1997 for Latin Amer- ship between indicators of real exchange rate volatility ica and the Caribbean. and private investment (Aizenman and Marion 1995, 21. Just lowering tariffs on intermediate and cap- 1996). Similar results were obtained by Darby and oth- ital goods without corresponding reductions in tariffs ers 1998 for a group of five OECD countries and by on consumer goods will increase the effective rate of Serven 1998 for developing countries. Other studies protection for domestic producers, and thus may im- have found no relationship between real exchange rate pair the efficiency of resource allocation. volatility and aggregate investment (Bleaney 1996; 22. Most infrastructure services in developing Goldberg 1993; Ramey and Ramey 1995). countries are supplied by governments or public de- 13. They cite various evidence, including the partments with administrative barriers to entry. This lower volatility of nominal exchange rates in develop- lack of competition leads to higher costs, poor mainte- ing countries than in several industrial countries, nance, and a lack of investment for technological up- higher volatility of foreign exchange reserves than grading. Opening up many of these services to compe- would be expected under floating rates, that countries tition under a transparent regulatory environment will use nominal interest rates to smooth exchange rate lower costs, attract foreign and private capital, and im- fluctuations, and that exchange rate rigidity occurs in prove delivery of services. For examples of recent pri- the face of commodity price shocks. vatization and demonopolization efforts and their pos- 14. The World Bank prepares ratings of develop- itive impact on performance, see AfDB 1999. ing countries' macroeconomic policy stances using a 23. In a sense these high costs act as an export consistent framework, for the purpose of cross-country tax, increasing the costs of production over those of comparisons. Although exchange rate volatility macro- suppliers that do not pay these duties. economic stability ratings are highly correlated, the re- 24. See various needs assessment papers issued lationship is not perfect. There are 21 countries which through the Integrated Framework for Trade-Related either had high exchange rate volatility and were rated Technical Assistance to Least-Developed Countries as having stable macroeconomic policies or vice versa. process. They include references to the state of standards Only 16 countries managed to achieve both macroeco- and conformity assessment systems and infrastructure nomic stability and low real exchange volatility. These needs in these countries (www.ldcs.org/index.htm). countries also achieved much higher per capita GDP 25. Under tariff quotas, a fixed quantity of prod- growth rates than other countries. ucts from specified countries can be imported at a 15. Thin markets lead to large changes in ex- lower tariff and anything above that level is subject to change rates due to small changes in demand and sup- the normal tariff. ply of foreign exchange. 26. The average most favored nation (MFN) tariff 16. A recent IMF study on exchange rate regimes for these 311 lines is 21 percent; for the least-developed comes to the same conclusions (Mussa and others 2000). countries it is 18 percent. 17. In many countries, initial policies did not in- 27. It is not clear why so few of these products are clude broad trade liberalization, but what can be called imported into Quad countries. But the significant dif- "compensatory" policies for exports. These range from ference between the United States' (86 percent) and the elaborate input coefficients for duty exemption systems European Union's (9 percent) share of developing coun- in Korea and duty drawback systems in Taiwan to ex- tries' exports suggest that intra-EU trade makes up the port processing zones, and so on, which gradually gave difference. way to elimination of NTBs, lower tariffs, and opening 28. Note that GSP and North-South free trade up of the domestic economy. Mauritius relied on an ef- arrangements imply some trade diversion that will ben- fective free trade zone, while the success of Bangladesh efit some developing countries, but hurt others. Forth- garments is due to a well-working bonded warehouse coming work by Hoekman, Ng, and Olareaga (2000) scheme, which is a form of duty exemption system. argues that these preferences are not very extensive. 72 T RA DE PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST C O U N T R I E S 29. The NAFTA preference is 50 percent for these Bhagwati, Jagdish. 1978. Foreign Trade Regimes and tariffs. Economic Development: Anatomy and Conse- 30. The producer support estimate (percent PSE) quences of Exchange Control Regimes. Cam- is calculated as the annual monetary value of produc- bridge, Mass.: Ballinger. tion-related support to agricultural producers as a Bigsten, Arne, and others. 1998. "Exporting and Effi- share of gross farms receipts (OECD 2000a). ciency in African Manufacturing." (Paper pre- 31. The recent proposal by the EU to grant duty- sented for the Conference on Policies for Compet- free access to all exports from least-developed countries itiveness in Manufacturing in Sub-Saharan Africa, and the move to income support rather than direct sup- Johannesburg, South Africa. 6-7 November.) port to production should improve market access for de- Bleaney, Michael. 1996. "Macroeconomic Stability, In- veloping countries. vestment and Growth in Developing Countries." 32. UNCTAD (TDR 1999) has shown that in Journal of Development Economics 48: 461-77. low- and medium-skill manufactures, small changes in Calvo, Guillermo A., and Carmen Reinhart. 2000. import penetration ratios significantly increase the ex- "Fear of Floating." University of Maryland, Col- port growth rates for these products from developing lege Park, Md., and National Bureau of Economic countries. Research, Cambridge, Mass. Draft. May. 33. Tariff escalation means that the tariff rate Clerides, Sofronis, Saul Lach, and James Tybout. rises with the level of processing. Forthcoming. "Is Learning by Exporting Impor- 34. For example, developing countries' share in tant? Micro-Dynamic Evidence from Colombia, fruits and vegetables trade, which is less protected, has Mexico, and Morocco." Quarterly Journal of not declined over this period. Economics. 35. Previous sections described the impact of con- Collier, Paul, and Anke Hoeffler. 1999. "Greed and flicts and trade-related policies on developing coun- Grievance in Civil War." Working Paper 2355. tries' performance. In addition, expansion of agricul- World Bank, Washington, D.C. tural exports (such as tropical commodities) that are Darby, Julia, and others. 1998. "The Impact of Ex- subject to few trade restrictions in industrial countries change Rate Uncertainty on the Level of invest- tends to reduce prices and hence the terms of trade. ment." Discussion Paper 1896. Centre for Eco- 36. The effect on agricultural exports is much nomic Policy Research, London. greater. The model reduces the exports of other prod- Dollar, David. 1992. "Outward-Oriented Developing ucts that faced lower tariffs because of factors moving Economies Really Do Grow More Rapidly: Evi- to the expanding agricultural sector. dence from 95 LDCs, 1976-85." Economic Devel- opment and Cultural Change 40: 523-44. April. Dornbusch, Rudiger, and Sebastian Edwards, eds. 1995. References Reform, Recovery and Growth. Latin America and AfDB (African Development Bank). 1999. African De- the Middle East. Cambridge, Mass.: National Bu- velopment Report 1999: Infrastructure Develop- reau of Economic Research. ment in Africa. New York: Oxford University Press. Easterly, William. 2000. "The Lost Decades ... and Aizenman, Joshua, and Nancy Marion. 1995. 'Volatil- the Coming Boom? Policies, Shocks, and Devel- ity, Investment and Disappointment Aversion." oping Countries" Stagnation 1980-1998." World Working Paper 5386. National Bureau of Eco- Bank, Washington, D.C. May. nomic Research, Cambridge, Mass. Easterly, William, Roumeen Islam, and Joseph E. . 1996. "Volatility and the Investment Re- Stiglitz. 2000. "Explaining Growth Volatility." sponse." Working Paper 5841. National Bureau World Bank, Washington, D.C. January. Processed. of Economic Research, Cambridge, Mass. Edwards, Sebastian. 1997. "Openness, Productivity Amjadi, Azita, Ulrich Reincke, and Alexander Yeats. and Growth: What Do We Really Know?" Work- 1996. "Did External Barriers Cause the Margin- ing Paper 5978. National Bureau of Economic alization of Sub-Saharan Africa in World Trade?" Research, Cambridge, Mass. March. World Bank Discussion Paper 348. World Bank, Edwards, Sebastian, and Miguel A. Savastano. 1999. Washington, D.C. March. "Exchange Rates in Emerging Economies: What Barro, Robert J., and Xavier Sala-i Martin. 1995. Eco- Do We Know? What Do We Need to Know." nomic Growth. Cambridge, Mass.: MIT Press, Working Paper 7228. National Bureau of Eco- McGraw Hill. nomic Research, Cambridge, Mass. July. Ben David, Dan. 1993. "Equalizing Exchange: Trade Elbadawi, Ibrahim. 1998. "Real Exchange Rate Policy Liberalization and Income Convergence." Quar- and Non-Traditional Exports in Developing Coun- terly Journal of Economics 108(3). tries." Research for Action 46. UNU World Insti- 73 G LO B AL EC ONO MI C PRO SP EC TS tute for Development Economics Research (UNU/ IMF (International Monetary Fund). 1981. Exchange WIDER), The United Nations University, Helsinki. Arrangements and Exchange Restrictions. Wash- Falvey, Rod, and Cha Dong Kim. 2000."Timing and ington, D.C. Sequencing Issues for Trade Liberalization in . 1992. Exchange Arrangements and Exchange Africa." In Trade and Fiscal Adjustment in Africa, Restrictions. Washington, D.C. edited by David Bevan, Paul Collier, Norman . 1996. Exchange Arrangements and Exchange Gemmell, and David Greenaway. Great Britain: Restrictions. Washington, D.C. Macmillan Press Ltd. . 1998. Trade Liberalization in IMF-Supported Finger, J. Michael, and Sam Laird. 1987. "Protection in Programs. February 1998. IMF, Washington, Developed and Developing Countries: An Over- D.C. view." Journal of World Trade Law 21 (6). . International Financial Statistics. Washington, Finger, J. Michael, and Ludger Schuknecht. 1999. D.C. "Market Access Advances and Retreats: the Kraay, Aart C. 1997. "Exports and Economic Perfor- Uruguay Round and Beyond." Policy Research mance: Evidence from a Panel of Chinese Enter- Working Paper. World Bank, Washington, D.C. prises." Development Research Group, World Frankel, Jeffrey A., and David Romer. 1999. "Does Bank, Washington, D.C. Processed. Trade Cause Growth?" The American Economic Krueger, Anne 0. 1978. Liberalization Attempts and Review 89 (3): 379-99. June. Consequences. A Special Conference Series on Ghura, Dhaneshwar, and Thomas J. Grenness. 1993. Foreign Trade Regimes and Economic Develop- "The Real Exchange Rate and Macroeconomic ment. Volume X. Cambridge, Mass.: National Bu- Performance in Sub-Saharan Africa." Journal of reau of Economic Research; New York: Ballinger Development Economics 42: 155-74. Pub. Co. Goldberg, Linda S. 1993. "Exchange Rates and Invest- Levine, Ross, and David Renelt. 1992. "A Sensitivity ment in United States Industry." Review of Eco- Analysis of Cross-Country Growth Regressions." nomics and Statistics 75: 575-88. American Economic Review 82: 942-63. Harberger, Arnold C. 1988. Reflections on Uniform Limao, Nuno, and Anthony Venables. 1999. "Infra- Taxation. Processed. structure, geographical disadvantage and trans- Hausman and others. 1999. "Financial Turmoil and port costs." Policy Research Working Paper 2257. the Choice of Exchange Rate Regime." Working Development Research Group-Trade, World Bank, Paper, Inter-American Development Bank, Wash- Washington, D.C. December. ington, D.C. Little, Iain, Malcolm David, Tibor Scitovsky, and Mau- Hertel, Thomas, Bernard Hoekman, and Will Martin. rice Fitzgerald Scott. 1970. Industry and Trade in 2000. "Towards a new round of WTO negotia- Some Developing Countries. Oxford, U.K.: Ox- tions: issues and implementation for developing ford University Press. countries." (Paper presented at the ABCDE meet- Michalopoulos, Constantine. 1999. "Trade Policy and ing. World Bank, Washington D.C. April.) Market Access Issues for Developing Countries: Hertel, Thomas, and Will Martin. 1999. "Would De- Implications for the New Millennium Round." veloping Countries Gain From Inclusion of Man- Policy Research Working Paper 2214. World Bank, ufactures Trade in the WTO 2000 Negotiations?" Washington, D.C. (World Bank Paper presented at the Conference Mussa, Michael, and others. 2000. "Exchange rate on the Millennium Round, WTO, Geneva. Sep- regimes in an increasingly integrated World Econ- tember 20-21.) omy." Occasional Paper 193. International Mon- - 2000. "Liberalising Agriculture and Manufac- etary Fund, Washington, D.C. Also available on turers." The World Economy 23 (4): 455-69. April. (http://www.imf.org/EXTERNAL/PUBS/CAT/lon Hinkle, Lawrence E., and Peter J. Montiel. 1999. Ex- gres.cfm?sk&sk=3518.0). change rate misalignment: Concepts and Mea- Nash, John, and Faezeh Foroutan, eds. 1997. Trade surement for Developing Countries. Washington, Policy and Exchange Rate Reform in Sub-Saharan D.C.: World Bank. Africa. National Centre for Development Studies, Hoekman, Bernard, Francis Ng, and Marcelo Olar- The Australian National University. reaga. 2000. "Problems Created by Protection in Ng, Francis. 1997. "A Profile of Tariffs, Para-tariffs, the Markets for Developing Country Exports." Non-Tariff Measures, and Economic Growth in World Bank, Washington, D.C. July. Developing Countries." World Bank, Washing- lanchovichina, Elena, Aaditya Mattoo, and Marcelo ton, D.C. Olarreaga. 2000. "Duty-Free Access for LDCs' Ng, Francis, and Alexander Yeats. Forthcoming. "On Exports: How Much Is It Worth and Who Pays?" the Recent Trade Performance of Sub-Saharan World Bank, Washington, D.C. June. African Countries: Cause for Hope or More of 74 TR AD E PO LI C I ES IN T HE 1 9 9 Os AND T HE PO OR EST C O U N T R I E S the Same?" Africa Region Technical Paper, World Shatz, Howard J., and David G. Tarr. 2000. "Exchange Bank, Washington, D.C. Rate Overvaluation and Trade Protection: Lessons Organisation for Economic Co-Operation and Devel- from Experience." Policy Research Working Paper opment (OECD). 1996. "Indicators of Tariff and 2289, World Bank, Washington, D.C. February. Non-Tariff Trade Barriers." . 1996. "Handbook of International Trade and . 2000a. "Agricultural Policies in OECD Coun- Development Statistics." Geneva. tries: Monitoring and evaluation 2000." Agricul- UNCTAD, 1999. Trade and Development Report, ture and Food, Paris. Geneva. . 2000b. "Tariffs and Trade: OECD Query and Wang, Zhen Kun, and L. Alan Winters. 2000. "Putting Reporting System." CD Rom. Paris. 'Humpty' Together Again: Including Developing Papageorgiou, Demetrios, Armeane M. Choksi, and Countries in a Consensus for the WTO." Policy Michael Michaely, eds. 1991. Liberalizing Foreign Paper 4. Centre for Economic Policy Research, Trade in Developing Countries-The Lessons of London. Experience. Washington, D.C.: World Bank; Ox- Williamson, John (ed). 1994. Estimating Equilibrium ford: Blackwell. Exchange Rates. Washington, D.C: Institute for Rajapatirana, Sarath. 1997. Trade Policies in Latin International Economics. America and the Caribbean: Priorities, Progress World Bank. 1989. Sub-Saharan Africa: From Crisis to and Prospects. San Francisco, Calif.: Interna- Sustainable Growth. A Long-Term Perspective tional Center for Economic Growth. Study. Washington, D.C.: World Bank. Ramey, George J., and Valerie A. Ramey. 1995. . 1994. Adjustment in Africa: Reforms, Results, "Cross-Country Evidence on the Link between and Road Ahead. World Bank Policy Research Volatility and Growth." American Economic Re- Report. New York: Oxford University Press. view 1138-51. . 1995. Global Economic Prospects and the De- Razin, Offair, and Susan M. Collins. 1997. "Real Ex- veloping Countries. Washington, D.C.: World Bank. change Rate Misalignments and Growth." Work- . 1996. "Trade Policy Reform in Developing ing Paper 6174, National Bureau of Economic Countries since 1985." Discussion Paper 267. Research, Cambridge, Mass. September. Washington, D.C. Rodriguez, E, and Dani Rodrik. 1999. "Trade Policy . 1997a. Global Economic Prospects and the and Economic Growth: A Skeptic's Guide to the Developing Countries. Washington, D.C.: World Cross-national Evidence." May, Discussion Paper. Bank. Centre for Economic Policy Research, London. . 1997b. World Development Indicators. Wash- Rodrik, Dani. 1996. "Understanding Economic Policy ington, D.C. Reform." Journal of Economic Literature 24 (1). . 1998. World Development Indicators. Wash- March. ington, D.C. . 1998. "Where did all the Growth Go? External - . 2000a. Can Africa Claim the 21st Century? Shocks, Social Conflict and Growth Collapses." Washington, D.C.: World Bank. Working Paper 6350, National Bureau of Eco- . 2000b. Trade Blocs: A World Bank Policy Re- nomic Research, Cambridge, Mass. search Report. New York: Oxford University Press. Sachs, Jeffrey D., and A.Warner. 1995. "Economic Re- . 2000c. World Development Report: Poverty forms and the Process of Global Integration." and Development. Washington, D.C. Brookings Papers on Economic Activity 1-118. . 2000d. World Development Indicators. Wash- Sala-i-Martin, Xavier. 1997a. "I Just Ran Four Million ington, D.C.: World Bank. Regressions." Working Paper 6252. National Bu- WTO (World Trade Organization). Trade Policy Re- reau of Economic Research, Cambridge, Mass. views-Quad Countries. Geneva. . 1997b. "I Just Ran Two Million Regressions." . 1998. Trade Policy Reviews for 31 Developing American Economic Review, Papers and Proceed- Countries 1995-98. Geneva. ings 87 (2): 178-83. Yeats, Alexander. 1998. "Have Policy Reforms Influ- Sekkat, Khalid, and Aristomene Varoudakis. 2000. enced the Trade Performance of Sub-Saharan "Exchange Rate Management and Manufactured African Countries." Trade Research Team. World Exports in Sub-Saharan Africa." Journal of De- Bank. Washington, D.C. velopment Economics 61 (1). February. Yeats, Alexander J., with Azita Amjadi, Ulrich Reincke, Serven, Luis. 1998. "Macroeconomic Uncertainty and and Francis Ng. 1997. Did Domestic Policies Private Investment in LDCs: An Empirical Investi- Marginalize Africa in International Trade? Direc- gation." Washington, D.C. World Bank. Processed. tions in Development, World Bank, Washington, September. D.C. 75 G LOBA L E C ON O M I C P R O S P E C T S - p. X, Aestr - ', 5 U > ; ie X , various charts and table Within core sample ECA Countries Countries All countriea Least- Low-income with high with low devloping Cosustrits (excluding developed small REER REER countries in conDfict Yugoalavia) Core sample countries countries volatility in volatility in No. Region [1o5] [191 [61 [59] [16] [22] 1990s [26] 1990s [33] 1 Africa Angola Angola 2 Africa Burundi Burundi 3 Africa Benin Benin Benin Benin Benin 4 Africa Burkina Faso Burkina Faso Burkina Faso Burkina Faso Burkina Faso 5 Africa Botswana Botswana Botswana 6 Africa Central Central Central Central Central African African African African African Republic Republic Republic Republic Republic 7 Africa C6te d'lvoire C6te d'Ivoire C6te d'Ivoire 8 Africa Cameroon Cameroon Cameroon 9 Africa Congo Congo 10 Africa Comoros 11 Africa Cape Verde 12 Africa Djibouti 13 Africa Ethiopia Ethiopia 14 Africa Gabon Gabon Gabon 15 Africa Ghana Ghana Ghana Ghana 16 Africa Guinea 17 Africa Gambia Gambia Gambia Gambia Gambia 18 Africa Guinea-Bissau 19 Africa Equatorial Guinea 20 Africa Kenya Kenya Kenya Kenya 21 Africa Liberia Liberia 22 Africa Lesotho 23 Africa Madagascar Madagascar Madagascar Madagascar Madagascar 24 Africa Mali Mali Mali Mali Mali 25 Africa Mauritania Mauritania Mauritania Mauritania Mauritania 26 Africa Mauritius Mauritius Mauritius 27 Africa Malawi Malawi Malawi Malawi Malawi 28 Africa Namibia 29 Africa Niger Niger Niger Niger Niger 30 Africa Nigeria Nigeria Nigeria Nigeria 31 Africa Rwanda Rwanda 32 Africa Sudan 33 Africa Senegal Senegal Senegal 34 Africa Sierra Leone Sierra Leone 35 Africa Sao Tome and Principe 36 Africa Swaziland 37 Africa Seychelles Islands 38 Africa Chad Chad Chad Chad Chad 39 Africa Togo Togo Togo Togo Togo 40 Africa Tanzania Tanzania Tanzania Tanzania Tanzania 41 Africa Uganda Uganda 42 Africa South Africa South Africa South Africa 43 Africa Democratic Democratic Republic of Republic of the Congo the Congo 44 Africa Zambia Zambia Zambia Zambia Zambia 45 Africa Zimbabwe Zimbabwe Zimbabwe 46 East Asia China China China 47 East Asia Fiji 48 East Asia Indonesia Indonesia Indonesia Indonesia 76 T RA DE PO LI C I ES IN T HE 1 9 9 0s AND T HE PO OR EST CO UN T RI ES Annex 2 Sample countries in various charts a .r- _Wthin core sample ECA Countries Countries All counries Least- Low-income with higb with low developing Countries )exduding developed small REER REER countries in conflict Yugosavia) Core sample countries countries volatility in volatility in No. Region [105] 1191 161 1591 116] [22] 1990s [261 1990s [33] 49 East Asia Republic of Republic of Republic of Korea Korea Korea 5M East Asia Malaysia Malaysia Malaysia 51 East Asia Philippines Philippines Philippines 52 East Asia Papua New Guinea 53 East Asia Thailand Thailand Thailand 54 East Asia Myanmar Myanmar Myanmar Myanmar Myanmar 55 ECA Former Former Soviet Union Soviet Union 56 ECA Bulgaria Bulgaria 57 ECA Former Former Czechoslovakia Czechoslovakia 58 ECA Hungary Hungary 59 ECA Poland Poland 60 ECA Romania Romania 61 ECA Yugoslavia, Yugoslavia Federal Federal Republic Republic of (Serbial of (Serbia/ Montenegro) Montenegro) 62 LAC Argentina Argentina Argentina 63 LAC Bolivia Bolivia Bolivia 64 LAC Brazil Brazil Brazil 65 LAC Barbados 66 LAC Chile Chile Chile 67 LAC Colombia Colombia Colombia 68 LAC Costa Rica Costa Rica Costa Rica 69 LAC Dominican Dominican Dominican Republic Republic Republic 70 LAC Ecuador Ecuador Ecuador 71 LAC Guatemala Guatemala 72 LAC Guyana 73 LAC Honduras Honduras Honduras 74 LAC Haiti Haiti 75 LAC Jamaica Jamaica Jamaica 76 LAC Mexico Mexico Mexico 77 LAC Nicaragua Nicaragua 78 LAC Panama 79 LAC Peru Peru 80 LAC Paraguay Paraguay Paraguay 81 LAC El Salvador El Salvador 82 LAC Suriname 83 LAC Trinidad and Tobago 84 LAC Uruguay Uruguay Uruguay 85 LAC Venezuela, Venezuela, Venezuela, Rep. Bol. de Rep. Bol. de Rep. Bol. de 86 MNA Turkey Turkey Turkey 87 MNA United Arab Emirates 88 MNA Kuwait Kuwait 89 MNA Morocco Morocco Morocco 90 MNA Bahrain 91 MNA Algeria Algeria Algeria 77 G LO B AL EC ONO MI C PRO SP EC TS an~i tab'tes 'continued) Within core sample ECA Countries Countries All countries Least- Low-income with high with low developing Countries (excluding developed small REER REER countries in conflict Yugoslavia) Core sample countries countries volatility in volatility in No. Region 1105] [19] 16] 159] 116] 122] 1990s [26] 1990s [33] 92 MNA Egypt, Egypt, Egypt, Arab Arab Arab Rep. of Rep. of Rep. of 93 MNA Iran, Islamic Iran, Islamic Iran, Islamic Rep. of Rep. of Rep. of 94 MNA Iraq Iraq 95 MNA Jordan Jordan Jordan 96 MNA Oman Oman Oman 97 MNA Saudi Arabia Saudi Arabia Saudi Arabia 98 MNA Syrian Arab Syrian Arab Republic Republic 99 MNA Tunisia Tunisia Tunisia 100 MNA Yemen, Yemen, Rep. of Rep. of 101 South Asia Bangladesh Bangladesh Bangladesh Bangladesh Bangladesh 102 South Asia India India India 103 South Asia Sri Lanka Sri Lanka Sri Lanka Sri Lanka 104 South Asia Nepal Nepal Nepal Nepal Nepal 105 South Asia Pakistan Pakistan Pakistan Pakistan 78 Standards, Developing Countries, and the Global Trade System P RODUCT STANDARDS, OR RULES GOVERN- to play an effective role in the design and ing the characteristics of goods, are crit- implementation of product standards and ical to the effective functioning of mar- thus constrain their access to some mar- kets and provide important support to the kets. Many developing countries, particu- trade system. For example, government test- larly the poorest ones, lack the technolog- ing and certification of the bacteria content of ical capabilities and financial resources to imported beef safeguards health and increases participate effectively in the development consumer acceptance of imported products. of product standards, to meet industrial Product standards in the international trade countries' import requirements, and to system do, however, raise difficult issues for bring disputes when standards are used developing countries. These countries' limited to discriminate against their exports. For technical capability and financial resources example, the European Union (EU) is har- make it hard for them to participate effectively monizing standards for levels of aflatoxin, in negotiations governing standards or to a substance that may cause liver cancer, in bring disputes. In addition, pressures some- food products. The new standard, which is times exerted to use trade sanctions in support more stringent than would be suggested by of labor and environmental standards-legit- internationally accepted standards, would imate and desirable as these standards may lower risks by approximately 1.4 cancer be intrinsically-threaten to restrict develop- deaths per billion per year.2 The new stan- ing countries' access to international markets dard has the potential for substantially re- without achieving their professed goals. ducing exports of cereals from developing The rapid growth of international trade has countries into Europe (Otsuki, Wilson, greatly increased the importance of effective and Sewadeh 2000). Few developing coun- regulation of standards at the international tries have the technology to evaluate the level. This chapter examines how standards dangers of aflatoxin, nor do they have imposed by governments in importing coun- the capabilities in scientific analysis to ad- tries affect developing-country exporters and dress the new EU standard. Furthermore, discusses the international regulation of some considerable legal and financial resources of the more prominent standards addressed in are needed to initiate a review under the global trade negotiations.1 Its main messages World Trade Organization's (WTO's) dis- are as follows: pute resolution mechanism. One achieve- * Insufficient technical and financial re- ment of the Uruguay Round agreement sources limit developing countries' abilities was to strengthen international rules gov- 81 G LO B AL EC ONO MI C PRO SP EC TS erning product standards in order to mini- dards and because better standards tend mize their use for protectionist purposes to encourage technological change to and to create a level playing field. None- economize on inputs. Studies have found theless, the lack of capacity in developing only limited evidence that low environ- countries, particularly the poorest, limits mental standards increase competitive- the ability of these countries to benefit ness or attract more foreign direct invest- from the new rules (Wilson 2000b). ment. Experience in both industrial and The adoption and respect of core labor developing countries shows that the cost standards-including freedom from dis- of appropriately designed environmental crimination, from exploitative child labor, protection is often low in terms of both forced labor, and the freedom to associ- forgone growth and the capital cost of ate and bargain collectively-are desirable abatement. Keeping labor standards low and essential. However, the threat of trade is not an effective way of gaining a com- sanctions or the imposition of trade bar- petitive advantage over trading partners. riers are likely to be excessively costly Indeed, low labor standards are likely to instruments for raising labor standards, erode competitiveness over time because and could even be counter-productive in they reduce incentives for workers to im- some cases. Barriers to a country's exports prove skills and for firms to introduce hurt workers by reducing demand for the labor-saving technology. country's products. Even if sanctions force * The international community has more improvements in some sectors, they are effective means than trade sanctions to unlikely to improve average working con- encourage improved environmental and ditions in the economy. For example, the labor standards in developing countries. result of foreign pressure to reduce the use Efforts to support development, such as of child labor in the production and export increasing assistance to countries with of garments in Bangladesh was that many good policies, will raise standards. Encour- of the laid-off chiLdren were employed in aging greater openness to trade and to for- more harmful occupations, such as prosti- eign direct investment (FDI) will facilitate tution or brick-breaking, and in factories the diffusion of cleaner technology that that did not produce for export (Financial can reduce environmental degradation Times, August 24, 1999). The imposition and improve worker productivity, thereby of trade barriers to improve labor stan- promoting better labor standards. Re- dards is vulnerable to capture by well- gional collaboration is appropriate for organized interests in domestic markets addressing environmental issues that have that would benefit from limiting imports. a clear regional component, such as trans- Similarly, trade sanctions are usually inef- boundary emissions and shared water fective in addressing environmental degra- resources. dation. Empirical studies show that impos- ing trade sanctions on exporters can cause considerable losses in output while doing The regulation of standards: little to reduce pollution. setting the stage * Although labor and environmental stan- Jn the broadest sense, regulations are estab- dards generally improve as countries de- i lished because of perceived market failures, velop, low labor and environmental stan- when reliance on voluntary market transactions dards are not usually a significant source is not efficient from the standpoint of society.3 of competitive advantage. Labor and en- For example, market prices may not reflect the vironmental standards are positively cor- full cost of production because firms use public related with income, both because higher waterways to dispose of waste; consumers may incomes stimulate demand for better stan- lack information about product defects that can 82 S TAN DAR DS, DE V EL O PING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T E M have serious consequences (unsafe automobiles, kinds of standards. Briefly put, product stan- for example); and collusion and monopoly may dards are necessary to support markets and mean increased costs for consumers. Establish- must be applied in a nondiscriminatory fash- ment of regulatory standards is appropriate ion. Environmental standards should be ad- when the benefits of correcting these market dressed by the community affected by the rel- failures exceed the costs. For example, current evant market failure. The impact of pollution auto safety standards have significantly reduced is normally limited to domestic or, sometimes, the chances of injury and death but have not regional markets, although some issues, such eliminated them, presumably because the cost as those related to global warming and deep- of doing so is too high. Costs include not only sea fishing, require global action. Differences the direct costs facing the regulated firm but in labor standards do not impose costs on for- also the costs of monitoring compliance and of eign markets and hence are not an appropriate any potential spillovers in other areas. (Taxing area for international trade negotiations. gasoline at the pump to limit pollution imposes a direct cost on consumers, but it also imposes a cost on filling stations and refineries as a re- Product standards and regulatory sult of lower demand.) The more detailed the barriers to trade rules are in defining what goods are produced TV nsuring that imported products meet ap- and consumed, and how, the greater the costs in Epropriate standards for protecting health terms of stifling innovation, reducing choice, and safety has become increasingly important and monitoring compliance. Thus, regulatory with the rapid expansion of trade over the instruments should use the market as much as past decade. Discriminatory regulations im- possible to encourage flexibility and choice of posed at the border can disadvantage foreign products and of production techniques. For ex- producers and distort commercial markets. ample, taxes and tradable permits have proved The reduction of tariffs and quotas through to be an effective and efficient means of con- multilateral trade negotiations has highlighted trolling air and water emissions in certain cir- the use of product standards as trade barriers. cumstances and to be less onerous than tradi- Tariffs, quotas, and subsidies continue to re- tional regulations that specify maximum levels strict trade in several sectors (see chapter 2), of pollution. but other barriers-technical requirements, Because preferences and policy options dif- testing, certification, and labeling that affect fer from country to country, regulatory re- imports-have emerged as important new is- gimes should be determined as much as possi- sues for liberalization efforts (World Bank ble by the communities to which they apply, 2000b). Two significant achievements of the unless there are spillovers to other communi- Uruguay Round, the agreement on Technical ties. Given different preferences and different Barriers to Trade (TBT) and the agreement on access to information, regulation that is ac- Sanitary and Phytosanitary Standards (SPS), countable to the community and meets locally were designed to address some of these issues. defined needs is likely to be more efficient and The TBT agreement essentially relates to man- legitimate than regulation imposed from afar. ufactured goods; the SPS agreement applies In an international context, it is important to to food (sanitary standards) and animals and ensure that regulation (a) does not discrimi- plants (phytosanitary standards). nate between domestic and foreign producers, (b) relates to products or activities that impose The role of product standards costs on domestic markets, (c) is restricted ge- Product standards are critical to the effective ographically to the markets affected, and (d) is functioning of markets and play an important implemented locally. role in supporting international trade. For These simple principles have powerful im- consumers, standards provide information plications for the appropriateness of different and help ensure quality. (For example, food la- 83 G LO B AL EC ONO MI C PRO SP EC TS beling requirements allow easier comparison Health and safety standards typically require across products, and regulations increase con- testing and conformity assessment for all pro- sumer confidence that electrical fixtures are ducers, but costs will be greater for exporters safe.)4 Standards are critical for "component" than for domestic producers if the exporters goods such as consumer electronics and com- must conform to standards different from puters, where the ability to mix and match those in their own market or if they are subject components is important. They also help to duplicative tests (Hoekman and Konan achieve public objectives such as cleaner air; 1998). For example, an EU regulation requires auto emissions standards and fuel economy that dairy products be manufactured from milk regulations are examples. Because the export produced by cows kept on farms and milked of goods that are physically dangerous or of mechanically. This rule precludes imports from agricultural products that are harmful to many developing countries, particularly those human health obviously damages the export- with many small producers for whom mech- ing country's (and the firm's) credibility and anization is not cost-effective (Henson and the acceptance of its products in the interna- others 2000). A country may have relatively tional trade system, there are important incen- stringent regulatory requirements owing to a tives for self-regulation. different view of the tradeoff between risks and For producers, standards can facilitate scale price. Such requirements may pose a significant economies and the efficient combination of compliance cost for exporters but would not be parts and components in production. Stan- viewed as discriminatory, since they apply to dards can also be used to gain access to intel- both domestic and foreign producers. lectual property and technology. For example, The need to comply with varying standards the European Union's (EU's) licensing of tech- can raise entry barriers in the form of in- nology based on European Telecommunication creased one-time costs of product redesign Standards Institute (ETSI) standards facilitated and creation of an administrative system. For the spread of wireless telephones in the Euro- example, manufacturers may need to keep re- pean market, highlighting the importance of designing automobile seat belts to meet chang- the relationships between standards and trade ing standards for multiple export markets. in goods and services (Wilson 1997). Stan- Standards may also diminish the ability to dards can facilitate coordination of produc- compete, owing to the recurrent costs of main- tion that might not be achieved through mar- taining quality control, testing, and certifica- ket forces. For example, countries can improve tion. Often, firms must decide whether to es- their integration into global information and tablish a costly platform design that can easily telecommunication networks by adhering to in- accommodate small modifications-for exam- ternational compatibility requirements for elec- ple, a car chassis that can serve multiple mar- trical products. Shared standards can reduce kets-or to design a product solely for the entry barriers by lowering inspection and test- home market, even though costly modifica- ing costs that typically arise from imperfect tions are required for export. A classic exam- information concerning the quality of traded ple of the latter is the right-hand or left-hand goods (Moenius 2000). placement of car steering wheels. Costs also may be incurred in meeting pre- Standards as barriers to trade cise technical regulations and carrying out con- Mandatory standards can also act as nontariff formity assessment-that is, in evaluating barriers to trade, whether or not the intent is whether a product "conforms" to a regulatory discriminatory; regulatory requirements may requirement. These requirements present the raise foreign firms' costs relative to those of largest potential technical barrier to future domestic firms even if both are subject to the trade. Governments in importing countries may same requirements in the domestic market.5 refuse to recognize tests performed in foreign 84 S TAN DAR DS, DE VE LO PING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T E M laboratories or by foreign public authorities mechanism should help establish precedents and may not accept declarations of conformity for determining what is acceptable under by a foreign manufacturer. For example, Mex- WTO disciplines. This should help resolve ico used to allow only Mexican organizations disputes earlier and restrain discriminatory and laboratories to test products subject to government initiatives that clearly conflict Mexican regulations. (Under the North Ameri- with principles of nondiscrimination. In addi- can Free Trade Agreement, or NAFTA, Mexico tion, greater reliance on private initiatives, as agreed to allow U.S. and Canadian firms to opposed to government fiat, in designing perform testing and certification.) Such re- product standards is desirable. For example, quirements may represent legitimate concerns whereas voluntary agreements account for a regarding the quality of administration in the large proportion of the standards (except for exporting countries, or they may result from those related to health or the environment) in administrative shortcomings in the importing industrial countries, in developing and tran- country (delays, arbitrary inspections, redun- sition countries such as China, Russia, and dant tests, and the like) that affect both foreign Ukraine standards in important areas of eco- and domestic firms. nomic activity continue to be developed and The use of product standards for protec- promulgated by governments. Reliance on pri- tionist purposes is a clear threat to an open vate norms in developing countries would trade regime. In principle, it is possible to dis- reduce the use of standards as trade barriers tinguish between the "normal" costs of trade (industry-based standards may have protec- (the kinds of frictional costs described above) tionist intent but can be difficult to enforce and barriers that are designed to limit compe- unless backed up by government regulations), tition from imports. The SPS agreement pro- and they can help ensure appropriate expertise vides that trade restrictions can be imposed in designing standards. only to the extent necessary to protect life or health, that they must be based on scientific Empirical evidence on standards as trade principles, and that they cannot be maintained barriers if scientific evidence is lacking. Where the A large proportion of internationally traded weight of scientific evidence is clear and well- goods is subject to standards, including about accepted, this approach has helped to resolve 60 percent of U.S. exports and 75 percent of disputes. For example, the United States suc- intra-EU trade (Wilson 1997). The coverage cessfully challenged Japanese technical regula- of standards has increased significantly in the tions on the ground that there was no evidence past few years (Hoekman and Konan 1998). that costly fumigation tests were necessary for Few attempts have been made to measure the each new variety of fruit imported into Japan. general impact of product standards on traded At times, the scientific community is unable to goods.6 The Organisation for Economic Co- assess risks because the damages are only evi- operation and Development found that dif- dent ex post (as was the case with asbestos), fering standards and technical regulations, or the relative newness of the technology may along with costs of testing and certification, call for caution in accepting existing evidence, can represent between 2 and 10 percent of as is happening with genetically modified or- overall product costs, and the European Com- ganisms (Messerlin and Zarrouk 2000). mission (1996) found that the average fric- Given differences in historical experiences, tional costs of differing standards among EU levels of development, and risk preferences, countries prior to the single-market initia- differences in product standards among coun- tive ranged between 2 and 3 percent of the tries will remain an important feature of the value of trade. The U.S.-EU mutual recogni- trade system. Over time, the accumulation of tion agreement on telecommunications and in- case law through the WTO dispute settlement formation technology products, if fully imple- 85 G LO B AL EC ONO MI C PRO SP EC TS mented, could reduce costs by 5 percent of the number of trade disputes over standards and value of goods traded (Wilson 1997). These technical barriers during the past five years. costs are greater than the average tariff on (The increase is evident in the U.S. annual re- intra-OECD manufacturing trade (less than 1 ports in the National Trade Estimates series percent in 1995) and on developing countries' and the EU's annual reports on trade barriers.) manufactured exports to industrial countries, In addition, most countries' submissions for which was 3.4 percent (Hertel, Hoekman, and the 1999 ministerial conference of the W'TO Martin 2000). in Seattle stressed the need to address techni- There is some evidence that the adoption cal barriers in the context of new trade talks of common standards tends to reduce imports (Wilson 1999). The most prominent standards from other sources. Sectors of EU economies cases in recent years have been in agriculture, for which common trade regulations were such as the dispute between the EU and United adopted as part of the move to the single mar- States over hormone-treated beef.8 The use of ket represent one-third of EU value added and genetically modified organisms (GMOs) in ag- one-third of intra-EU trade, but only one- riculture is also generating trade tensions. By fourth of EU imports from the rest of the the end of January 1999, the WTO Dispute world. Conversely, sectors in which the estab- Settlement Body had considered 25 disputes lishment of common trade regulations was less that referenced either the SPS or the TBT (Wil- successful represent one-third of both intra-EU son 1999). Nine of the disputes centered on trade and EU imports from the rest of the food safety regulations, five involved technical world (Messerlin 1998). Surveys and simula- regulations tied to customs requirements, and tion exercises confirm the role of standards in the remainder were in areas such as quotas, increasing costs. An OECD (1999) survey of import bans, and disputes over environmental 55 firms in Germany, Japan, the United King- laws. Most of the complaints brought to the dom, and the United States found that tech- WTO are from industrial countries; of the 25 nical standards and conformity assessment complaints considered by the WTO through procedures imposed significant costs on dairy January 1999, 16 were brought by industrial products, auto parts, and telecommunications. countries against other industrial countries, 3 Typical problems included requirements for were brought by industrial countries against testing of each product consignment both be- developing countries, and 6 were brought by fore shipping and at the port of entry and for developing countries against industrial coun- frequent tests following design changes. Simu- tries. No low-income country other than India lations with a computable general equilibrium has brought cases to the WTO under the TBT (CGE) model found that a 2.5 percentage point or the SPS or has been challenged under these decrease in border costs within the EU (the es- agreements.9 Pursuing a case through WTO timated result of adoption of uniform stan- procedures is expensive and resource-intensive, dards) would generate a short-term welfare which may explain in part why many develop- gain of up to 0.5 percent of the gross domestic ing countries have not done so. product (GDP) of EU countries (Harrison, One indication of the increased focus on Rutherford, and Tarr 1996), in part because of WTO dispute settlement-including cases re- scale economies and increasing competition.7 lated to standards-by members is the invest- The benefit could reach 2.4 percent of GDP ment by the United States in new staff in over the long term as investment increases as a the office of the U.S. Trade Representative result of a rise in the real return to capital. (USTR). The budget request for fiscal 2001 in- cludes an increase of 14 percent for additional Trade disputes on product standards staff, all of whom would focus on dispute set- One indication of the importance of standards tlement case work at the WTO (Hufbauer, in restricting trade is the marked increase in the Kotschwar, and Wilson 2000).1o The least- 86 S TAN DAR DS, DE V EL O PING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T E M developed countries (a UN-designated group ducers may thus confront higher costs of entry of 48 developing countries) are likely to find it in markets than do producers from countries difficult to match this type of investment in that can certify compliance through an MRA WTO dispute settlement processes. (see box 3.1). Furthermore, governments and Disputes over standards as barriers to trade firms in more advanced countries can establish will undoubtedly become more important as strategic standards that shut out developing- (a) the share of trade in world output increases country firms or that alter the terms of compe- and developing countries' weight in world tition or the terms of trade in favor of domes- trade rises, (b) exports of finished goods by tic firms (Fischer and Serra 2000; Gandal and developing countries grow, and (c) large de- Shy 1999; Matutes and Regibeau 1996). veloping and transition countries, such as Full implementation of the commitments China, Russia, and Ukraine, whose domestic made in the SPS and TBT agreements will ben- regulatory systems and import rules require efit both developing and industrial countries deep reform, join the WTO. A recent review and will strengthen the multilateral system. of Ukraine's standards and regulatory system There have, however, been reservations about commissioned by the World Bank, for exam- developing countries' abilities to meet specific ple, reveals serious economic distortions in the provisions of these agreements. The SPS agree- design of the government's standards, testing, ment, for example, encourages the use of rele- and certification systems (World Bank 2000a). vant international standards; although a coun- try may apply other standards at the border, Capacity in developing countries it has the burden of demonstrating their scien- Product standards may work to the disadvan- tific merit. Since most standards were designed tage of developing countries, where capacity by industrial countries, they may not be appro- to engage in standards development and to priate for the technology mix or preferences in comply with standards in export markets is developing countries. "Thus for a country to limited. Because of lack of resources, many effectively use the WTO agreement to defend developing countries find it difficult to diffuse its export rights or justify its import restric- best-practice information on quality standards tions, it will have to upgrade its SPS system to such as those in the International Organiza- international standards" (Finger and Schuler tion for Standardization (ISO) 9000 series and 2000). Upgrading standards and providing risk to adopt appropriate process and production assessments for proposed standards can be methods (World Bank 2000b). Certification costly. There are similar questions regarding re- costs can be particularly significant for small quirements in the TBT agreement which em- firms. ISO 9000 certification for a single plant body the concept that trade is best facilitated can cost up to $250,000, with additional au- by harmonizing international standards. diting costs after initial approval. Limits on Effective compliance with requirements for capacity are particularly important for the WTO enquiry points (offices that provide in- least-developed countries. formation regarding national technical regula- Developing countries lag behind industrial tions) can involve substantial costs, including countries in their capacity for effective certifi- the costs of establishing governmentwide in- cation and accreditation of testing facilities formation systems to report regulatory changes (Stephenson 1997), and authorities in indus- and respond to requests.11 Formal compliance trial countries may not trust developing coun- with enquiry point requirements has improved tries' inspection procedures (Baldwin 2000). in developing countries, but it remains less Developing countries thus find it difficult to than 60 percent for the SPS agreement and 75 develop standards based on international percent for the TBT agreement (figure 3.1). It norms and to reach mutual recognition agree- is not clear whether these enquiry points meet ments (MRAs) with other nations. Their pro- all the provisions of the agreements. The num- 87 G LO B AL EC ONO MI C PRO SP EC TS ber of notifications by developing countries of requirements in light of development needs, in- new technical regulations and certification cluding extension of the time frame for com- rules, as required by TBT and SPS agreements, plying with some provisions and modification has grown (figures 3.2 and 3.3), although the of the rules governing notification of new tech- increase may not reflect a greater ability to nical regulations. (Providing 60 days to com- meet product standards for export goods. ment on new regulations is of questionable All in all, the costs involved in comply- value to developing countries that lack the ing with TBT and SPS requirements are sub- capacity to analyze and formulate positions stantial and are likely to be equal to an entire on technical requirements quickly.b)14 A serious year's development budget in some least- and thorough use of the results of the Second developed countries. (This calculation includes Triennial Review of the TBT agreement, sched- the costs of meeting trade-related aspects of uled to conclude in November 2000, would intellectual property rights, or TRIPs, require- help address developing countries' concerns. ments; see (Finger and Schuler 2000 ).)12 Such expenditures can improve a country's capacity to participate in international trade, but they Labor standards and trade must be evaluated in light of other development sanctions priorities.13 fA doption and compliance with core labor Many developing countries have recom- f estandards is desirable on moral grounds, mended a targeted review of TBT and SPS and necessary for promoting broad-based and 88 S TAN DAR DS, DE VE LO PING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T EM Figure 3.1 WTO enquiry point notification, by country group, 1995 and 1999 Percent 100 90 19 80 70 60 50 40 30 20 110 TBT SPS TBT SPS Developing countries Industrial countries Note: Percentage of countries with WTO enquiry points under the rules of the Technical Barriers to Trade (TBT) and Sanitary and Phytosanitary Standards ISPS) agreements. Source: G/TBT/ENQ,WTO . inclusive economic development (World Bank lower labor standards generally provide a 1995 and Aidt and others 2000). However, competitive advantage. imposing trade sanctions to bring about im- proved labor standards is unlikely to enhance Core labor standards and their either global welfare or the welfare of devel- relationship to development oping countries. In terms of the criteria for Core labor standards are commonly defined regulatory decisionmaking outlined in the be- to include freedom of association and collec- ginning of the chapter, labor standards should tive bargaining; nondiscrimination in employ- not be the subject of trade negotiations be- ment; no exploitative child labor; and no forced cause the level of standards in one country labor (for example, slavery).15 Each of these does not affect the welfare of its trading part- core standards is covered by at least one In- ners, and the workers whom trade sanctions ternational Labour Organisation (ILO) con- are designed to protect have no role in decid- vention. By the mid-1990s, only 27 countries ing whether sanctions are imposed. Although worldwide and only 10 OECD countries had higher labor standards are associated with im- ratified all core ILO conventions, although proved living conditions and development, the ratifications increased in the second half of the imposition of trade sanctions is a remarkably 1990s (OECD 2000). Several countries that costly mechanism. Furthermore, trade sanc- have not ratified some of these standards are tions are vulnerable to capture by domestic in- regarded as being in compliance with them in terests, and are likely to hurt the workers the practice. Their reasons for nonratification ap- sanctions are designed to assist. Lower labor pear not to relate to objections on principle, standards abroad are not a serious threat to but rather to specific details of the conven- the livelihoods of workers in industrial coun- tions or their interpretations by ILO bodies tries; neither theory nor evidence suggests that (OECD 1996).16 Of course, ratification does 89 G LO B AL EC ONO MI C PROS PE CT S Figure 3.2 Number of notifications under the TBT agreement, 1995-2000 700 | Low- and middle-income countries U High-income countries 600 500 400 300 200 100 0 1995 1996 1997 1998 1999 2000 Note: TBT, Technical Barriers to Trade. Data are for 2000 through October 4, 2000. Source: G/TBT/N, aWTO . not necessarily imply that core labor standards of association is used because it is easier to will be observed, which requires legislative and measure than some of the other standards.) regulatory changes, as well as monitoring and On average, more-developed countries have enforcement to stop abuses. better-than-average compliance, while compli- Adherence to core labor standards, as mea- ance in many of the poorest countries is inad- sured by freedom of association, is weakly equate. Higher income levels stimulate demand correlated with both higher levels, and higher for better standards, and higher standards con- growth rates, of GDP per capita."7 (Freedom tribute to growth by increasing work effort 90 S TAN DAR DS, DE V EL O PING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T E M and stimulating innovation, in order to econo- knowledge of the competitive conditions in the mize on labor. affected labor markets. There is less evidence that adherence to In some cases, improvements in labor stan- labor standards is correlated with other mea- dards may have unintended consequences and sures of economic development, such as real not necessarily improve the welfare of work- wages. In the newly industrializing countries of ers. In several countries, labor standards are East Asia, rising real wages have been associ- lower in export-processing zones (EPZs) than ated with improved bargaining rights (Maskus in the rest of the country, mainly because of 1997). In a larger sample of countries, how- bans on unions or restrictions on strikes ever, there is no clear correlation between free- (OECD 1996).20 Workers in most EPZs, how- dom of association and changes in real wages ever, earn higher wages and enjoy better work- or changes in manufacturing output per worker ing conditions than their counterparts else- for the period 1973-92. In a sample of 17 where in the country (ILO 1993; Maskus countries that had recorded discrete improve- 1997).21 It is not clear what effect better labor ments in legislation and practice regarding standards would have on investors the EPZs freedom of association, there was no uniform are designed to attract. Advocates of improv- tendency for growth to accelerate after the ing labor standards in EPZs must understand changes (OECD 1996). not only the domestic labor market but also the negotiating position of the developing coun- Labor standards and economic welfare try relative to investors. Adherence to core labor standards can make Labor standards and competitiveness important contributions to improving welfare. It is often argued that low labor standards im- The welfare impact depends on the structure pose low wages and thus enhance domestic of domestic institutions and policies. For ex- competitiveness at the expense of trading part- ample, if monopsonist firms hire workers be- ners' workers. In some cases, employer collu- low their marginal revenue product, allowing sion, in the absence of collective bargaining worker association and collective bargaining rights for workers, may reduce wages below could raise both worker wages and efficiency what they would be with effective labor stan- by boosting employment.18 Moreover, orga- dards, thus potentially raising production.22 nized labor can contribute to raising efficiency Such an outcome would require nationwide and welfare in ways that go beyond the adju- collusion since if a firm pays below the pre- dication of wages. For example, unions can vailing wage, it will eventually lose its employ- contribute to firm-specific knowledge and or- ees to other sectors.23 The key point in this dis- ganizational capital, thus raising productivity, cussion is that standards need to be ratcheted and can help improve domestic labor stan- up in a coordinated and economywide fashion; dards by overcoming a "prisoner's dilemma" a sectoral or partial approach will have spill- low-standards equilibrium (Stiglitz 2000).19 over effects which could in many cases be But if the economy starts from a competitive detrimental to a broad group of workers. equilibrium, collective bargaining that raises Over time, artificially imposed low labor wages above their marginal revenue product standards are likely to erode competitiveness may lower efficiency. If collective bargaining is because they reduce incentives for workers to supported by measures to restrict entry (and in improve their skills; the earnings gain that can competitive conditions, collective bargaining is be achieved by upgrading skills is limited by not likely to have a long-term impact on wage labor market conditions. Similarly, low labor levels otherwise), the excluded workers are standards reduce incentives for firms to intro- clear losers. Thus, evaluating policies for in- duce labor-saving technology, because the sav- ducing higher labor standards requires detailed ings are worth less if wages are low. 91 G LO B AL EC ONO MI C PRO SP EC TS The data do not indicate that core labor ther employment and wages in the sector. If standards play a significant role in shaping the monopsonist were large, that would lead trade performance (OECD 1996). Countries to pressures to reduce wages in the economy with higher labor standards had higher growth as a whole (Maskus 1997). rates in their share of world manufacturing Trade sanctions on particular export goods exports from 1980 to 1990, but it is not pos- are unlikely to improve labor standards for sible to infer the direction of causality from the economy as a whole, even if the sanctions these results. Of six countries that achieved change the behavior of particular firms. For significant improvements in labor standards, example, barring child labor in one firm or half saw a decrease in the growth of their share sector without addressing the fundamental in world manufactures, while half saw an in- causes of child labor is likely to shift children crease. Differences in endowments and tech- to less-remunerative and perhaps more dan- nology are much more important than labor gerous occupations in other sectors. In Ban- standards in determining patterns of compara- gladesh in 1993, the threat of U.S. sanctions tive advantage. led owners of garment factories in Dhaka to The OECD study (1996) also examined the dismiss all children under age 16. Anecdotal relationship between labor standards and FDI. evidence suggests that many of these children Most world FDI flows from OECD countries found employment in workshops and facto- into other OECD countries, which generally ries not producing for export, or as prosti- have high labor standards. As for the inflows tutes, brick-breakers, or street vendors (Pana- of FDI to non-OECD countries, it is not clear gariya 1999a). There are effective measures that countries with low standards are the pri- for combating abusive child employment, in- mary destinations.24 cluding income-support programs and subsi- dies for education, but trade sanctions are not Effectiveness of trade sanctions in among them. improvisengss lboftra standards tn The political-economy arguments against imposing trade sanctions on countries with Even where low labor standards reduce eco- low labor standards are even more compel- nomic efficiency and welfare, sanctions are ling. As discussed above, determining whether unlikely to improve workers' welfare. It is a particular improvements in labor standards familiar principle of economics that the most would raise welfare requires considerable in- efficient way to remove a distortion is to ad- formation on labor and product market con- dress it directly. For example, setting a tariff is ditions. Determining whether labor arrange- an inefficient way to encourage domestic pro- ments in exporting countries will affect wage duction of a good. Imposing trade sanctions is rates in importing countries is even more com- a vastly inefficient way to encourage better plicated, requiring estimates of various param- labor standards. eters such as demand and supply elasticities Take a favorable (to those advocating trade in different markets and factor intensities of sanctions) case, in which government-supported goods (Maskus 1997). The complexity of these barriers to entry enable monopsonist employ- issues, and the decentralized nature of the ers to pay workers below their marginal rev- costs of protection to consumers, increase the enue product. Both employment and wage potential for decisions to be captured by well- rates are lower than if the market were com- organized domestic interests that would ben- petitive. Assume that the rest of the world im- efit from trade barriers. The fact that labor poses a tariff on exports of the product, thus unions and producers in some protected indus- reducing the export price. The monopsonist's tries in industrial countries favor using the response to any reduction in demand under WTO system to improve labor standards un- this market structure would be to reduce fur- derlines this concern. 92 S TAN DAR DS, DE V EL O PING CO UN TR IE S, AND T HE G LO B AL TR AD E SYS T E M While adherence to core labor standards Environmental standards improves welfare, integrating labor standards and trade into the WTO is contentious. Under tradi- he past decade has seen increasing debate tional criteria, which focus on product stan- T over the contribution of trade to environ- dards, not process standards, labor standards mental degradation. In part, this debate has would not be considered for trade discussions. reflected concern about the role of growth in The TRIPs agreement has, however, widened depleting cross-border public goods; specific the scope for broadening the traditional crite- issues include the dangers of global warming ria (see box 3.2). and the unsustainable pace of fishing and water Inadequate labor standards and poor work- use in some regions (Nordstr6m and Vaughan ing conditions are, first and foremost, a devel- 1999). Workers and firms in industrial coun- opment challenge that affects sizable popula- tries fear that their competitive position is tions, whether or not they are involved in being undermined by environmental regula- trading activities.25 It may be easy to identify tions that force pollution-intensive industries some blatant abuses linked with goods that to move to developing economies. Greater trade enter industrial markets, but the large majority integration and access to information, while of workers in developing countries may suffer boosting global welfare, are increasing the in- from even worse conditions than workers em- tensity of disputes and the potential for do- ployed in export activities. The keys to improv- mestic interests to be injured by the actions of ing workers conditions-beyond development foreigners. itself-lie in assisting countries with the devel- Although environmental concerns are opment of domestic institutions to support clearly legitimate, the trade system is rarely the workers' rights and improve working condi- appropriate instrument for addressing them, tions, and coordinating policies across develop- given the principles outlined at the beginning ing countries to ratchet up standards and escape of this chapter. Only a limited set of environ- a low-standard equilibrium. mental issues affect more than one country. To the extent that environmental damage is lim- The ILO has been actively pursuing these ited to a single country, decisions on whether activities since its creation in 1919. The ILO to resmgle contr envionmether regularly ~~ . moitr wokn.odtosi t to restrict production for environmental rea- regularly monits w n csons should not be imposed through trade ne- member countries, and traditionally provides gotiations. Imposing trade sanctions to achieve incentives (such as technical assistance) to environmental goals is likely to be inefficient encourage improved compliance with ILO and perhaps counterproductive. Countries have conventions. However, it is able to invoke eco- different priorities, which are in large part a nomic sanctions (Article 33 of the ILO Con- reflection of different levels of development. stitution) and did so for the first time in 2000 Poorer countries are likely to make different (against Myanmar), although implementation choices in facing tradeoffs between growth and of the sanctions was postponed to allow the environmental goals than do industrial coun- country time to comply.26 Strengthening the tries-today's industrial countries did the same ILO and enhancing its cooperation with other when they were developing. It is important that international organizations would be an effec- developing countries retain access to the in- tive step toward ameliorating working condi- ternational trade system, even if their domestic tions around the world. The private sector, environmental policies are not those preferred particularly multinational firms, should also by richer countries. Several international in- play a more active role by promoting uniform stitutions-such as the Joint United Nations corporate codes of conduct and using best- Environment Programme (UNEP)-and the practice production methods in all countries international environmental summits, have an where they or their affiliates operate. environmental mandate and should be the 93 G LO BAL EC ONO MI C PRO SPEC TS AgevemetoTpP)an evloig utre stuandardsfrpoetnonelculprpry toefrtetn IS rmlra oeo h e drigt (icldnpaet,cprgt ntrdmks tieydane veP cnnis yaeo. an o etbiholgtosrgrigteefc- poueteedusdtusiaI.T &ot~ meto ihs iptsudrTIP r ujc o cudivk n xsrn ~teTW~seint thing nertddsut eteetsse. gatcipisr iese o h oetcrdc dusrilcutiswrstogavctsoTRs, ono rg.Tharemtalwsfrcpusy andeeoigcntismyavaceetoito lcsigndreriscd n,nebigba achievprgesisetrofipractote, feieoito ewenhelclgermtan suhaI giutueadPxieooh &eg nnuarr~rgr'n h esso TRP ae infcn dmn o hne n wic h nniatrrwul ewligi uii itellctulpoetheie,priual nmn h di~tcmrenete ae opnaini deelopnconreinwihpoeioofielc- deth oei atnhldrLsadacdctn 9 u4 poet osnrme iiu RP tn rismyhv ifciiti rdcnieori S TAN DAR DS, DE V EL O PING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T E M forums for discussing environmental goals. In outward-oriented economies have lower pollu- addition, donor countries and international tion intensity of aggregate output than inward- agencies can and do condition their assistance oriented ones. During the 1980s outward- on achievement of environmental goals, in- oriented growth was associated with declining cluding those that affect important aspects of pollution intensity because the industrial ac- the global commons. tivities of outward-oriented economies became more diversified, shifting away from heavy The impact of trade integration on the manufacturing (Lucas, Wheeler, and Hettige environment 1992).31 FDI and the use of technology-laden Trade integration influences growth, the tech- imported inputs have helped transmit cleaner nology mix, and the composition of output. In- technologies from the regulated industrial- creased openness will raise economic growth country market to developing countries-for and living standards, which, other things being example, in the paper and pulp industry equal, will increase environmental degrada- (Wheeler and Martin 1992) and the steel in- tion. This scale effect is empirically important, dustry (Reppelin-Hill 1999). especially for countries that are specialized in Conversely, in many countries import- environment-intensive activities, such as min- substitution strategies have been pollution- ing, fisheries, and forestry, as in Chile, and and resource-intensive because of price distor- wood and wood products, industrial chemi- tions and lack of competitive discipline. There cals, and petroleum, as in Indonesia (Lee and is strong evidence that under an import- Roland-Holst 1997). substitution strategy, countries have special- Although the scale effect is always positive ized in pollution-intensive manufacturing ac- (as long as trade integration induces growth), it tivities in which they are not truly competitive. can be counterbalanced by two other effects: The resource content of goods in such coun- the technique effect and the composition ef- tries is much higher than that of comparable fect.29 Trade integration changes access to tech- goods in open economies (Jha, Markandya, nology (through, for example, capital goods and Vossenaar 1999; Vukina, Beghin, and So- imports), and this technique effect may have a lakoglu 1999). Some distortions have stronger positive or negative impact on environmental environmental consequences than others. For degradation. New technology may result in example, subsidized energy usually implies a savings on energy and other inputs, reducing more resource-intensive economy and there- the pollution intensity of growth. The compo- fore more emissions. sition effect may also have a positive or nega- Trade liberalization and other reforms have tive impact on environmental degradation. helped correct policy distortions that subsidize Trade integration and growth affect the com- environmental degradation. For example, en- position of output, owing to changes in the ergy use per unit of aggregate product in 12 relative endowments of factors, the increas- former centrally planned economies declined ing consumption of (relatively cleaner) services drastically with market reform, in part be- that accompanies higher incomes, and the in- cause of the rise in domestic oil prices and the creased affordability and desirability of pollu- cleaner composition of manufacturing output tion reduction, which indirectly lead to better following trade and price liberalization. En- environmental protection.30 ergy intensity in China fell by 30 percent be- The impact of trade integration on the en- tween 1985 and 1997 as market-oriented re- vironment has varied considerably, depending forms were introduced (Vukina, Beghin, and on the nature and strength of these three ef- Solakoglu 1999; World Bank 1997). Similar fects, but outward orientation has reduced findings emerge for use of natural resources. the pollution intensity of output in several For example, in Sri Lanka, trade liberalization countries (Birdsall and Wheeler 1992), and increased the demand for land to be planted in 95 G LO B AL EC ONO MI C PRO SP EC TS tea, which is less erosive than other crops, thus tries, setting off a "race to the bottom." The generating both environmental and economic emergence of such a race is theoretically pos- benefits (Bandara and Coxhead 1999). sible (Klevorick 1997; Wilson 1997), particu- Some countries do show increased pollu- larly in political and regulatory environments tion following trade liberalization, owing to that are not transparent and are vulnerable both scale and composition effects. Beghin to capture by dirty-industry interests. (Cap- and Potier (1997) suggest that some countries ture by "green" interests is also possible-en- faced more domestic pollution following trade vironmental protection would exceed public liberalization because their aggregate activities preferences.) The several methodological ap- expanded, not necessarily because they spe- proaches used to study this question generally cialized in "dirty" activities. Several countries, find mixed evidence as to whether environ- however, did see increased specialization in mental regulation is eroding competitiveness dirty activities following trade liberalization in relatively "clean" countries (see table 3.2 because they happened to be competitive in and box 3.3). these activities. In this category are Indonesia (Lee and Roland-Holst 1997; Strutt and An- The cost of environmental protection derson 1999); China (Dean 1999; Dessus, One reason for the paucity of evidence that Roland-Holst, and van der Mensbrugghe environmental regulations impair competitive- 1999; Jha, Markandya, and Vossenaar 1999); ness is that the cost of environmental protection Costa Rica (Abler, Rodriguez, and Shortle is often low, as measured by forgone growth or 1999; Dessus and Bussolo 1998); and Turkey the capital cost of abatement. Despite the inef- (Jha, Markandya, and Vossenaar 1999). Fer- ficiency of the command-and-control approach rantino and Linkins (1999), using simulations that most OECD countries have used in ad- with a CGE model to estimate the effects of dressing pollution, the cost of compliance to in- trade liberalization on output of toxic emis- dustries has been surprisingly small, and abate- sions, suggest that specialization is more im- ment has been significant (Jaffe and others portant than scale in determining the impact 1995). Simulations using applied general equi- of trade liberalization on pollution. Table 3.1 librium models of developing economies have summarizes the evidence from economywide found that the cost of abatement for most types studies on the relationship between trade lib- of emissions is modest in terms of forgone GDP eralization and pollution. Panel studies found growth. This finding was robust, having been a mixed effect of outward orientation. Rock generated from models of seven developing eco- (1996) found that the composition effect of nomies with different assumptions on abate- outward orientation was positive or ambigu- ment possibilities and for 13 types of pollution. ous. Lucas, Wheeler, and Hettige (1992) found The only type of pollution that was found to be a negative composition effect. Negative results expensive to abate was bioaccumulative toxic in a study by Vukina, Beghin, and Solakoglu releases in water (Beghin, Roland-Holst, and (1999) were robust. van der Mensbrugghe forthcoming). Detailed One concern about trade and financial in- qualitative case studies of individual industries tegration is that countries with relatively weak undertaken by the United Nations Conference environmental regulations will attract dirty in- on Trade and Development (UNCTAD) con- dustries away from countries with stronger firm these findings (Jha, Markandya, and Vos- regulations, and that because of competitive- senaar 1999). ness concerns integration will inhibit the im- Malaysia provides an interesting case of position of strong environmental regulations specialization in resource-intensive activities ("regulatory chill"). A related conjecture is accompanied by environmental protection that states could strategically decrease envi- (Jha, Markandya, and Vossenaar 1999). The ronmental protection to attract new indus- palm oil industry adapted to a rapidly imple- 96 S TAN DAR DS, DE V EL O PING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T E M Table 3.1 Summary of economywide studies assessing the impacts of trade liberalization on pollution Pq ehasg Scale Uosnton Technique Total polution Mexicoa Trade liberaliation + - ... Small derease United Statesa with NAFTA + + ... Increase Canadaa + + increase Mexico, Trade piberaliaton + + ... Increast United Statesa with NAFrA plas + - ... Increase Canadaa invesimet + + ... hKreaSe lberalization Mexicob Trade liberalization, +2.8 to 3.7% -4.3 to 2.6% -0.7 to 3.5% -0.2 to 6.4% beter term of trade with United Ssates and Canad Costa Rica' Trade liberalitios 9.4% 5.6 to 10.6% + but small 15 to 20% Vietnamd Trade liberalation S to 8.8% -6.3 to 8% 1.1 to 7.5% 0.8 to 23.1% lndonesiae Trade liberalization 0.87% -.36 to 2.86% ... 0.51 to 3.73% with Japan Japane Trade liberalization 0% -0.9 to -0.02% ... -0.09 to -0.02% with Indonesia Globalf Multilateal ... ... -0.02 to 0% -4.32 to 0% libealization ... Not available. Note: NAFTA, North American Free Trade Agreement. The data cited in notes a-f are reproduced from Beghin and Potier 1997. a. Grossman and Krueger 1992; percentages not available. b. Beghin, Roland-Holst and van der Mensbrugghe 1995. The scale effect range refers to production and absorption. The ranges for composition and technique effects refer to 13 measures of pollution emissions. c. Dessus and Bussolo 1998. The scale effect is the increase in output. The composition effect is the difference between total and scale effects. d. Dessus and van der Mensbrugghe 1996. e. Lee and Roland-Holst 1997. The range of composition effects refers to 10 pollutant types. The authors also report a human toxicity index. f. Ferrantino and Linkins 1999, tables 7 and 9. Scale and composition figures are not disaggregated. mented set of environmental regulations and tion of the latest technology (Jha, Markandya, taxes. Compliance is high, and exports are sta- and Vossenaar 1999). ble, even though opportunities to pass the cost increase on to consumers were limited by the Trade policy and environmental highly competitive nature of the industry. State- protection funded research helped develop commercial Tariffs are usually ineffectual instruments for by-products from palm meal, reducing the tackling pollution and environmental degra- cost of compliance by generating revenues dation. Only when the externality originates from the by-products instead of treating them in trade are trade taxes effective in addressing or dumping them and paying fines and fees the problem (Subramanian 1992). A ranking (Jha, Markandya, and Vossenaar 1999; Khalid of instruments for addressing pollution emis- and Braden 1993). The Malaysian electronics sions follows the targeting principle (Bhagwati industry also continued to grow despite tighter and Srinivasan 1997), which, broadly, says environmental regulations, in part because the "the closer, the better." Hence, emissions taxes strong FDI presence facilitated the introduc- are the best instrument for dealing with pollu- 97 G LO B AL EC ONO MI C PRO SP EC TS Table 3.2 Evidence on international competitiveness and environmental regulation Approach Study Conclusion Cross-sectional Kailet0 19 88 U.S. manufacturing exports negatively affected by Heckscher-Ohlin (H-O) model environmental regulation Tobey 1990 World trade in dirty commodities nor affected by environmental regulation Har 1996 Small negative impact of regulation, decreasing over time :aliuru and Peterson 1997 Grain trade not affected by environmental regulation Diakosauvas 1994 Exports of the five most polluting crops negatively affected by regulation VXu 1999 00 ..00000: 0 ....Environmentally sensitive exports of 34 countries not influenced by regulation Investigations of FDI flows Albrech t 19 United States found to import pollution-intensive industries more than it exports them Eskeland and Harrison 1997 No pollution-intensive bias in French and U.S. FDI in developing economies XigandKotad 1995 0 00t;00f} 0U.S. FDI influenced by weak regulation only in chemical industries Plant location: firm surveys UNCTAD:193 Negative effects of environmental policy on location Levinson 9,stye 0D:0Mary Marginal impact of compliance cost except for self-declared U.S. dirty industries Plant location: econometric in I 7 No effect approach Bartk 198 Small and negative effect Maxsi, Pargaland Huq 1997: t Positive effect of one measure of environmental stringency on plant location Meral20:00 t000000:/b5200X090\00 Negative effect of regulatory stringency on small U.S. livestock operators tion emissions and minimizing distortionary There have been few trade disputes over effects elsewhere in the economy. If emissions technical requirements related to the environ- taxes are not feasible, input taxes are prefer- ment. Whalley and Hamilton (1996) report able to production taxes, which in turn are only a limited number of environment-related preferable to tariffs (Beghin, Roland-Holst, trade disputes for the period 1982-96, and very and van der Mensbrugghe 1997; Lloyd 1992; few such disputes have been brought to the Ulph 1999). This point has been documented WTO since 1995 (WTO website). Only two of empirically in the case of forestry products the 43 requests from developing countries- (Barbier and Rauscher 1994), as well as for concerning reformulated U.S. gasoline and the the Indonesian economy (Lee and Roland- U.S. ban on certain seafood products-involve Holst 1997). With increasing economic in- environmental objectives. Of 300 cases of trade tegration, Indonesia is tending to specialize impediments to U.S. agricultural exports, only in resource- and pollution-intensive activities. one was based on environmental goals; most in- Pollution emissions at the national level (as volved food safety and protection of crops and distinguished from the sector level), however, livestock from pests and disease. It is not clear cannot be decreased even modestly by using whether the paucity of environment-related dis- tariffs. By contrast, production taxes propor- putes reflects the limited impact of environmen- tional to the pollution content of output make tal regulations on traded goods, the high costs the targeted pollution abatement feasible at a of litigation, or the scope of disputes provided reasonable cost in forgone growth. for under WTO rules. 98 S TAN DAR DS, DE V EL O PING CO UN TR I ES, AND TH E G LO B AL TR AD E SYS T E M Box 3.3 Evidence on the "race to the bottom" T mpirical. studies of the pattern of trade, the allo- pollution4itensive bias in the allocation of French and E cation of FM, plant location, and profitability U.S. FDI flows going into manufacturing industries in have found limited or no evidence that environmental Cate d'lvoire, Mexico, Morocco, and the Republica regulations have reduced investment or lowered com- Bolivariana de Venezuela. Xing and Kolstad (1995) petitiveness. As night be expected, evidence for a find that U.S. FDI in chemical industries seems to be race to the bottom is somewhat stronger for the dirti- influenced by weak environmental regulation, as prox- est industries, although even here there are conflict- ied by sulfur dioxide emissions, but they also find that ing results. There is no evidence that intracountry FDI in cleaner industries was not influenced by envi- differences in environmental regulations affect ronmental stringency. investment. Large firms appear better-able to accom- Studies have found only limited evidence that modate environmertal reulations than snaller firms. environmental regulation has a significant effect on Stiudies of the patterns of trude have used the plant location. Surveys of the relocation of transna- cross-sectional Heckscher-Ohlin (M-0) model, which tional corporations provide some support for the no- explains specialization on the basis of environmental tion of a race to the bottom (Runge 1994; UNCTAD abu-ndance, to examine indirectly the effects of envi- 1993). Surveys, however, tend to be less reliable than rornmentai reglation on international competitive- actual data because they report what is said rather ness. The results are mixed. Using 1977 data, Kalt than what is done (Levinson 1997a). Levinson finds, (1988) found that U.S. environmental regulation for many industries and measures of stringency, that had a significantly negative effect on competitive- interstate differences in environmental regulations do ness, as measured by net exports of manufacturing not systematically affect the location choices of most goods. Tobey (1990), using 1975 data, found no evi- manufacturing plants in the United States. Mani, dence that increased regulation affected output in Pargal, and Huq (1997) find, surprisinWy, that a pollutionrintensive industries. Han (1996) tested the proxy for different levels of enforcement of federal environmental H-0 model using panel data (across environmental policy in Indian states is positively industties and over time) and actual expenditure related to decisions on the location of new manufac- data on pollution abatement as a imeasure of the turing plants for a wide range of manufacturing environmental input. He found that increased envi- industries and for the smaller subset of poUlution- ronmental regulation has had a significantly negative intensive industries. It is possible that the proxy for effect on competitiveness, but that this effect has de- stringency (the share of the state budget spent on creased over time as many countries tightened their environmental programs) measures the efficiency of regulations and as abatement costs fell with new cap- state administration, which induces firms to locate in ital vintages, learning by doing, and new technolo- states with higher environmental expenditures. gies. Valluru and Peterson (1997) and Diakosauvas Several other studies have looked at the impact (1994) found little evidence that environmental regu- of environmental regulation in agriculture, but mostly lations have had a signiicant negative econornic in OECD countries. Metcalfe (2000) finds that strin- effect on agricultural trade except for the most- gency had little impact on the location of U.S. hog polluting commodities such as cotton and tobacco. production across states and over time. Stringency Xu (1999) found that the export performance of en- did have a negative impact on small operators but vironmenrtaly sensitive industries in 34 countries was not on large, modem, confinement livestock produc- unchanged between the 1960s and the 1990s despite ers. Hettige and others (1996) found evidence of the erergence of environmental standards in most economies of scale in environmental compliance for industrial countries since 1970. many other industries in several countries. The evidence on the allocation of PFD provides Finally, studies have found a positive relation- little support for the existence of pollution havens. ship between environrnental performance and the The United States is importing more pollution- profitability of U.S. firms (Cohen and Fenn 1997; intensive industries than it is exporting, and dirty Repetto 1995). Although environmental compliance industries are no more likely to invest abroad than is not free, it creates new market opportunities and other industries (Albrecht 1998, cited in Nordstrom may induce further efficiency gains that may offset its and Vaughan 1999; Eskeland and Harrison 1997). (small) cost. Environmental performance appears to Eskeland and Harrison (1997) find no evidence of be systematically associated with higher profitability. 99 G LO B AL EC ONO MI C PRO SP EC TS Several global environmental treaties have cation under industrial-country labeling schemes been concluded over the last 25 years, notably may be difficult for developing countries to the Convention on International Trade in En- obtain (Jha, Markandya, and Vossenaar 1999; dangered Species of Wild Fauna and Flora Jha and Zarrilli 1994; OECD 1997a; Zarsky (CITES)32 protecting trade in endangered 1994). For example, none of the 48 licenses species, and the Montreal Protocol33 banning granted under the EU Commission's ecolabel the use of ozone-depleting chemicals (for exam- went to a developing-economy firm, although ple CFCs, widely used as a coolant in refrigera- it is not clear whether any of these firms ap- tors and air conditioners). These agreements plied (Nimon and Beghin 1999). Ecolabeling typically provide incentives for compliance schemes can be used in a discriminatory way, through both technical and financial assistance. especially in markets dominated by develop- In addition, many also provide for trade sanc- ing economies, such as textiles. Domestic tions to enforce compliance. The compatibility industries have more say in defining ecostan- with WTO rules-of-trade sanctions potentially dards than do foreign competitors. The stan- allowed by such treaties has not been tested. dards are likely to favor technologies that are feasible in industrial countries rather than the Alternative policies for environmental input mix and technology set of developing protection countries. Although trade sanctions are not effective Local ecolabels are emerging in developing means of inducing environmental protection, countries, especially in timber-based products, foreigners can affect environmental choices in but also in textiles, to promote better practice other ways. In some cases, foreign countries and preempt discriminatory labeling in indus- could provide subsidies to encourage better trial countries. For example, Malaysia sup- environmental practices. For example, in the ports ecolabels and standards that apply to all U.S.-Mexican dispute over protecting dolphins, types of timber and are based on internation- an alternative policy would have been for the ally agreed standards, not merely on standards United States to equip Mexican fishermen with developed by one or a few countries (Jha, improved nets.34 The cost of this option would Markandya, and Vossenaar 1999). have to be compared with the overall losses Another approach is to help trading part- resulting from trade restrictions. This is to ners implement market-based environmental some extent an empirical issue, but the option policies that have proved effective in tackling would at least reduce dolphin kill, which nei- environmental problems in developing coun- ther trade sanctions nor a consumer boycott is tries. Reducing subsidies on pollution-intensive likely to do. activities or raising taxes on polluting activi- Ecolabeling schemes enable foreign con- ties, through discharge, input, or output taxes, sumers to choose goods produced in an envi- has reduced pollution and increased tax rev- ronmentally benign way. These schemes can enues in Bangladesh, Brazil, Indonesia, and be a source of trade friction, even though the other countries (World Bank 1997). Market- markets they cover are still relatively small, based instruments also provide incentives to because of the increased production costs in- save on the taxed resource and become more volved in the certification process. For exam- resource-efficient. The more targeted the in- ple, ecolabeling schemes in textiles require strument, the better. Some countries, such as multiple production standards for dyes, fibers, China and Malaysia, have used emissions and bleaching chemicals (OECD 1997a). In charges with some success. When the cost of addition, most schemes impose fees. Canada's monitoring is not prohibitive, the market in- Environmental Choice Program imposes a 0.5 strument can be very targeted; for example, percent charge, based on the price of the good, many countries use stumpage fees to foster on sales up to Canadian $1,000,000. Certifi- sustainable forest management (World Bank 100 STANDAR DS, D EVE LOPING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T E M 1997). China has been successfully abating many developing countries, but it is not the main sub- pollution for the past 20 years by using levies ject of this chapter. The effects of voluntary product (Wang and Wheeler 2000). standards are touched on summarily. Privatzatio and competitin o2. Many international food standards are set by the Privatization and competition, or incre- Codex Alimentarius Commission, which is based in mental reform in this direction, can promote Rome and is a joint commission of the Food and Agri- better resource management. Several studies culture Organization of the United Nations (FAO) and identify state firms as worse polluters than the World Health Organization (WHO). firms in the private sector (Pargal and Wheeler 3. This framework is taken, in part, from Rollo and 1996) or centrally planned economies as worse Winters 2000. than market economies (Vukina, Beghin, and 4. For a primer on standards and trade, see Na- tional Research Council 1995. Solakoglu 1999). Incentives to economize, 5. This section draws on Maskus and Wilson combined with increased resources for better forthcoming. management, have improved the performance 6. The Development Economics Research Group of public entities in many countries. For ex- of the World Bank is carrying out a major project on ample, in several countries, water-user associ- trade and standards that includes construction of a ations have been substituting for the govern- new global database on standards barriers to support ment in allocating irrigation water. future empirical and policy research in this area. 7. The estimate of the decrease in costs attributable Engagement of the public iS essential to suc- to the adoption of uniform standards is taken from cessful environmental protection. This process Gasiorek, Smith, and Venables 1992. can foster partnership among the public, firms, 8. An overview of all WTO dispute settlement cases and authorities. The government can be a fa- may be accessed through the WTO website at cilitator for private industry by disseminating and the WTO information on new technology and environ- document distribution facility at . mental regulations. Alternatively, the process The U.S.-EU case on hormone-treated beef is cataloged can be coerclve, relying on disclosure of vio- under WT/DS26 and WTIDS48. 9. WTO cases involving only industrial countries lation of environmental regulations, such as may have implications for developing-country ex- illegal discharges. The coercive approach has porters' market access. For example, the EU's restric- been effective in developing economies such as tions on U.S. exports of genetically modified grains China (Dasgupta and Wheeler 1997), although could have major implications for the exports of simi- complaints tend to be positively associated with lar products from countries, such as Argentina and higher income and greater human capital. Brazil, where varieties of genetically modified organ- Regional apaisms have been widely planted. Regional approaches to environmental stan- 10. The budget of the office of the USTR in fiscal dards may prove more effective than global ap- 2000 was $25.5 million, and the office had 178 (full- proaches, particularly on issues with a clear time equivalent) staff members; see . emissions and shared water resources. A re- 11. For additional background, see Wilson 2000a gional approach does not imply uniform stan- and 2000b. dards for domestic environmental problems; 12. The (unweighted) average development assis- the case against harmonization of policies is tance budget as a share of GDP for low-income coun- the cse aginst armoizatin of olicis is tries was 11.2 percent in 1998. overwhelming in most settings because of dif- 13. The World Bank has an active program to assist ferent valuations of the marginal benefits of en- developing countries in improving their standards in- vironmental protection. frastructure. Further information is available at . 14. This discussion is taken from formal posi- Notes tions submitted to WTO General Council, January- 1. The regulation of standards in a local or national November 1999. economy, particularly with respect to appropriateness 15. These core labor standards were enunciated in and efficiency impacts, is another important topic for the June 1998 ILO Declaration on Fundamental Prin- ciples and Rights at Work. 101 G LO B AL EC ONO MI C PRO SP EC TS 16. The 1996 OECD study was updated in a more reasons, but the attraction of a large and rapidly grow- recent report (OECD 2000), reaching broadly the same ing market is the most significant motive. conclusions as the earlier study. 25. Labor markets in developing countries have 17. This analysis should be viewed with some cau- been a subject of increasing involvement by the World tion, for several reasons: simple correlations provide Bank-including the seminal 1995 World Develop- no information on the direction of causality; the lack ment Report on "Workers in an Integrating World"- of a theoretical model of the determinants of growth in large part because of the recognition that they play means that the measured correlations between stan- a key role in poverty reduction and economic develop- dards and growth may be misleading; and it is difficult ment. Bank projects with a labor market component to construct adequate quantitative measures of the ex- have increased dramatically since the early 1990s. Ef- tent of adherence to core labor standards. forts are also underway to enhance dialogue with 18. Alternatively, the worker association could re- NGOs and other external partners, including regular strict entry of workers and enable them to bargain for consultations with representatives from the Interna- a higher wage, improving the welfare of workers in the tional Confederation of Free Trade Unions (ICFTU) to association at the expense of excluded workers. The discuss areas of mutual concern. outcome would depend on the goals of the association 26. For more information, see http://www.ilo.org/ and the relative bargaining power of workers and cap- public/english/bureau/inf/pr/2000/27.htm. ital (Maskus 1997). 27. To date, six cases against developing countries 19. The "prisoner's dilemma" in this context refers have been initiated, all by the EU and the United States. to the fact that if a country attempts to improve stan- Four cases deal with pharmaceuticals and agricultural dards it will lose a competitive edge if it acts alone. As chemicals and two concern compatibility of domestic a result, in the absence of coordination, no country will regulations with TRIPs obligations. The countries in- attempt to improve standards. volved are Argentina, Brazil, India, and Pakistan; see 20. Maskus (1997) points out, however, that labor http://www.wto.orglenglish/tratop_e/dispu_e/stplay-e. turnover in EPZs is rapid, in part because assembly em- doc. ployment is dominated by women, who leave to marry. 28. The transfers refer to the net present value of In any event, high labor turnover results in low union- payments from 1988 on, based on the 1988 structure ization rates, even in EPZs in which union organization of patents. and the right to strike are protected. 29. The scale effect, almost by definition, has an 21. Firms in EPZs may pay higher wages than other elasticity of one with respect to growth. Thus, if an domestic firms for several reasons: they benefit from economy grows by x percent, all else being equal, emis- less burdensome regulations and can operate more flex- sions will also increase by x percent. ibly; they tend to be larger and thus enjoy scale 30. A significant body of literature on the 'envi- economies; they are governed by the policies of foreign- ronmental Kuznets curve" (EKC) posits that pollution owned firms that are bound by their headquarters' best intensity follows an inverse-U-shaped curve with re- practices in labor standards; they need to attract labor spect to income. At low levels of development, pollu- to move to the area; or pressures to maintain quality to tion tends to increase with economic growth; above a satisfy export requirements may encourage them to pay certain income level, it declines. There is evidence, at higher wages to induce greater effort (Maskus 1997). least for some types of pollutants, that the turning 22. Even here, the impact of higher exports on point in some developing countries is occurring at wages in importing countries is likely to be small, par- lower levels of income than was witnessed in industrial ticularly in the familiar case of highly labor intensive countries earlier. If this tentative evidence is borne out, goods such as apparel, footwear, and electronics it suggests that several factors are working in favor of (Maskus 1997). a more rapid transformation to a cleaner environment 23. This analysis depends on the structure of labor in developing countries. These factors include techno- market conditions. For example, employers who col- logical diffusion of both cleaner production processes lude to reduce labor standards can benefit if there are and abatement technologies and greater awareness of effective barriers to labor mobility. the costs of environmental damage on the part of both 24. In the 1990s China, whose labor standards officials and the general public. This literature is sum- have been criticized, was the largest beneficiary of FDI marized succinctly in Nordstrom and Vaughan 1999. flows among developing countries. Significant anecdo- 31. See Rock 1996, however, for a critique of the tal evidence indicates, however, that U.S. FDI in China measurement of openness and market integration. is establishing above-market-wage, high-standards op- 32. For more information see http://www.wcmc. erations. Clearly, firms are investing in China for many org.uk/CITES/index.shtml. 102 S TAN DAR DS, DE V EL O PING CO UJ N TR I ES, AND T HE G LO B AL T RAID E SYS T E M 33. For more information on the Montreal Proto- Basu, Kaushik. 1999. "International Labor Standards col, see http://www.unep.org/ozone/montreal.htm. and Child Labor." Challenge 42 (5, September- 34. The United States placed trade restrictions on October): 80-93. the import of Mexican tuna because Mexican tuna Beghin J., and M. Potier. 1997. "Effects of Trade Lib- fishing techniques led to the indiscriminate killing of eralisation on the Environment in the Manufac- dolphins. turing Sector." World Economy 20 (4): 435-56. Beghin J., B. Bowland, S. Dessus, D. Roland-Holst, and D. van der Mensbrugghe. 1999. 'Trade, Environ- ment, and Public Health in Chile. Evidence from References an Economywide Model." In Per G. Fredriksson, Abler, D. G., A. G. Rodriguez, and J. S. Shortle. 1999. ed., Trade, Global Policy, and the Environment, "Trade Liberalization and the Environment in 35-54. Discussion Paper 402. World Bank, Wash- Costa Rica." Environment and Development ington, D.C. Economics 4 (3): 357-73. Beghin, John, David Roland-Holst, and Dominique Albrecht, J. 1998. "Environmental Policy and Inward van der Mensbrugghe. 1995. "Trade Liberaliza- Investment Position of U.S. Dirty Industries." In- tion and the Environment in the Pacific Basin: tereconomics (July-August): 186-94. Coordinated Approaches to Mexican Trade and Anderson, K. 1992. "The Standard Welfare Economics Environment Policy." American Journal of Agri- of Policies Affecting Trade and the Environment." cultural Economics 10 Aug 1995. 77(3):778. In The Greening of World Trade Issues, edited by . 1997. "Trade and Environment Linkages: K. Anderson and R. Blackhurst. Chapter 2. Ann Piecemeal Reform and Optimal Intervention." Arbor: University of Michigan Press. Canadian Journal of Economics 30: 442-55. Anderson, K., and R. Blackhurst, eds. 1992. The , eds. Forthcoming. Trade and Environment in Greening of World Trade Issues. Ann Arbor: Uni- General Equilibrium: Evidence from Developing versity of Michigan Press. Economies. Dordrecht, the Netherlands: Kluwer Bailey, R., ed. 1995. The True State of the Planet. New Academic Publishers. York: Free Press. Benjamin, Dwayne, Loren Brandt, Paul Glewwe, and Baldwin, Richard E. 1992. "Are Economists' Tradi- Guo Li. 1999. "Markets, Human Capital, and In- tional Trade Policy Views Still Valid?" Journal of equality: Evidence from Rural China." Develop- Economic Literature 30: 804-29. ment Research Group, World Bank, Washington, . 2000. "Regulatory Protectionism, Developing D.C. Processed. Nations and a Two-Tier World Trade System." Bhagwati, J. N. 1997. "The Demands to Reduce Di- Discussion Paper 2574. Centre for Economic Pol- versity among Trading Nations." In Fair Trade icy Research, London. and Harmonization: Prerequisites for Free Trade? Baldwin, Robert E., and Anthony J. Venables. 1997. edited by J. N. Bhagwati and R. E. Hudec. Cam- "International Economic Integration." In Hand- bridge, Mass.: MIT Press. book of International Economics, edited by G. Bhagwati, J. N. and T. N. Srinivasan. 1997. "Trade Grossman and K. Rogoff. Vol. 3. Amsterdam: and the Environment: Does Environmental Diver- North-Holland. sity Detract from the Case for Free Trade?" In Bandara, J. S., and 1. Coxhead. 1999. "Can Trade Lib- Fair Trade and Harmonization: Prerequisites for eralization Have Environmental Benefits in De- Free Trade? edited by J. N. Bhagwati and R. E. veloping Country Agriculture? A Sri Lankan Case Hudec. Cambridge, Mass.: MIT Press. Study." Journal of Policy Modeling 21 (3): Bhagwati, J. N., and R. E. Hudec, eds. 1997. Fair 349-74. Trade and Harmonization: Prerequisites for Free Barbier, E. B., and M. Rauscher. 1994. "Trade, Tropi- Trade? Cambridge, Mass.: MIT Press. cal Deforestation and Policy Interventions." Envi- Birdsall, N., and D. Wheeler. 1992. "Trade Policy and ronment and Resource Economics 4: 75-90. Industrial Pollution in Latin America: Where Are Barrett, S. 1994. "Strategic Environmental Policy and the Pollution Havens?" In Patrick Low, ed., In- International Trade." Journal of Public Econom- ternational Trade and the Environment, 159-67. ics 54: 325-38. Discussion Paper 159. World Bank, Washington, Bartik, Timothy J. 1989. "Small Business Start-Ups in D.C. the United States: Estimates of the Effects of Brainerd, Elizabeth. 1998. "Winners and Losers in Characteristics of States." Southern Economic Russia's Economic Transition." American Eco- Journal 55 (4, April): 1004-18. nomic Review 88 (5): 1094-1116. 103 G LO BAL E CON OM I C PROS P ECT S Brander, J., and B. J. Spencer. 1985. "Export Subsidies Davis, Donald R. 1996. "Trade Liberalization and In- and International Market Share Rivalry." Journal come Distribution." Working Paper 5693. Na- of International Economics 18: 83-100. tional Bureau of Economic Research, Cambridge, Brander, J., and S. M. Taylor. 1997. "International Mass. Trade between Consumer and Conservationist Dean, J. M. 1999."Testing the Impact of Trade Liber- Countries." Working Paper 6006. National Bu- alization on the Environment. Theory and Evi- reau for Economic Research, Cambridge, Mass. dence." In Per G. Fredriksson, ed., Trade, Global Brown, Drusilla K., Alan V. Deardorff, and Robert M. Policy, and the Environment, 55-63. Discussion Stern. 2000. "U.S. Trade and Other Policy Op- Paper 402. World Bank, Washington, D.C. tions and Programs to Deter Foreign Exploitation Deininger, Klaus, and Lyn Squire. 1996. "A New Data of Child Labor." In Topics in Empirical Interna- Set Measuring Income Inequality." The World tional Economics: A Festschrift in Honor of Bank Economic Review 10 (3): 565-91. Robert E. Lipsey, edited by Magnus Blomstram Dessus, S. and D. van der Mensbrugghe. 1997. "Trade and Linda S. Goldberg. Chicago: University of Reform and the Environment: The Case of Viet- Chicago Press nam." OECD Development Centre, Paris. Carruth, Alan A., and Andrew J. Oswald. 1981. "The Processed. Determination of Union and Non-Union Wage Dessus, S., and M. Bussolo. 1998. "Is There a Trade- Rates." European Economic Review 16: 285-302. off between Trade Liberalization and Pollution Chang, Seung Wha. 1997. "GATTing a Green Trade Abatement? A Computable General Equilibrium Barrier: Eco-Labelling and the WTO Agreement Assessment Applied to Costa Rica." Journal of on Technical Barriers to Trade." Journal of World Policy Modeling 20 (February): 11-31. Trade 31 (1): 137-159. Dessus, S., and D. van der Mensbrugghe (1996). Chichilinsky, G. 1994. "North-South Trade and the "Trade Liberalization and the Environment in Global Environment." American Economic Re- Vietnam," OECD Development Centre, Paris, view 84: 851-74. mimeo. Choksi, A., M. Michaely, and D. Papageorgiu. 1991. Dessus, Sebastien, David Roland-Holst, and Do- "The esig of uccesful radeLibealiztionminique vani der Mensbrugghe. 1999 (inl prepara- "The Design of Successful Trade Liberalization tion or forthcoming) "Trade and Environment in Policies." In Foreign Economic Liberalization, China," in Beghin et a]. (eds.), op citum. edited by A. Koves and P. Marer. Boulder, Colo.: _- . Forthcoming. "Trade and Environment in Westview Press. China." In Trade and Environment in General Cohen, M., and S. Fenn. 1997. "Environmental and Fi- Equilibrium: Evidence from Developing Econo- nancial Performance: Are They Related?" Van- mies, edited by J. Beghin, D. Roland-Holst, and derbilt University, Department of Economics, D. van der Mensbrugghe. Dordrecht, the Nether- Nashville, Tenn .. lands: Kluwer Academic Publishers. Copeland, B. R. "International Trade and the Environ- Diakosauvas, D. 1994. "The Impact of Environmental ment: Policy Reform in a Polluted Small Open Policies on Agricultural Trade." Journal of Inter- Economy." journal of Environmental Economics national Development 6 (2): 207-18. and Management 26: 44-65. Eskeland, Gunnar S., and Ann E. Harrison. 1997. Copeland, B. R., and S. M. Taylor. 1995. "North- "Moving to Greener Pastures? Multinationals South Trade and the Environment." Quarterly and the Pollution Haven Hypothesis." Policy Re- Journal of Economics 109: 755-87. search Working Paper 1744. World Bank, Wash- Currie, Janet, and Ann Harrison. 1997. "Trade Reform ington, D.C. and Labor Market Adjustment in Morocco." European Commission. 1996. Journal of Labor Economics 15 (3): S44-S72. Feenstra, Robert C., and Gordon H. Hanson. 1997. Dasgupta, Susmita, and David Wheeler. 1997. "Citizen "Foreign Direct Investment and Relative Wages: Complaints as Environmental Indicators: Evi- Evidence from Mexico's Maquiladoras." Journal dence from China." Policy Research Working of International Economics 42 (3-4): 371-93. Paper 1704. World Bank, Washington, D.C. Ferrantino, M. J., and L. A. Linkins. 1999. "The Effect Dasgupta, Susmita, Hua Wang, and David Wheeler. of Global Trade Liberalization on Toxic Emis- 1997. "Surviving Success: Policy Reform and the sions in Industry." Weltwirtschaftliches Archivl Future of Industrial Pollution in China." Policy Review of World Economics 135 (1): 128-55. Research Working Paper 1856. World Bank, Finger, J. Michael and Philip Shuler. 2000. Developing Washington, D.C. Countries and the Millennium Round. Mimeo. 104 S TAN DAR DS, DE V EL O PING CO UN TR I ES, AND T HE G LO B AL T RA DE SYS T E M Finger, J. Michael, and Philip Schuler. 1999. "Imple- Countries." Policy Research Working Paper mentation of Uruguay Round Commitments: The 1710. World Bank, Washington D.C. Development Challenge." Policy Research Work- Haskel, Jonathan, and Matthew J. Slaughter, 2000. ing Paper 2215. Trade Development Research "Have Falling Tariffs and Transportation Costs Group, World Bank, Washington, D.C. Raised U.S. Wage Inequality?" Working Paper . 2000. "Developing Countries and the Millen- 7539. National Bureau of Economic Research, nium Round." Processed. Cambridge, Mass. Fink, Carsten. 2000. "How Stronger Patent Protection Henson, Spencer, Rupert Loader, Alan Swinbank, in India Might Affect the Behavior of Transna- Maury Bredahl, and Nicole Lux. 2000. "Impact of tional Pharmaceutical Industries." Policy Re- Sanitary and Phytosanitary Measures on Develop- search Working Paper 2352. Trade, Development ing Countries." Centre for Food Economics Re- Research Group, World Bank, Washington, D.C. search, Department of Agricultural and Food Eco- Fischer, Ronald and Serra, Pablo. 2000. "Standards nomics, University of Reading, Reading, U.K. and protection." Journal of International Eco- Hertel, Thomas W., and Will Martin. 1999. "Develop- nomics. 01 Dec 2000. 52(2):377. ing Country Interests in Liberalizing Manufac- Fischer, Ronald. The New Protectionism and Latin tures Trade." Background research paper for America. Introduction Chapter. Seattle Ministerial Meeting of the WTO. World Freeman, Richard B., and David L. Lindauer. 1999. Bank Institute, Washington, D.C. "Why Not Africa?" Working Paper 6942. Na- Hertel, Thomas W., Bernard M. Hoekman, and Will tional Bureau of Economic Research, Cambridge, Martin. 2000. "Developing Countries and a New Mass. Round of WTO Negotiations." World Bank Insti- Gandal, Neil, and Oz Shy. 1996. "Standardization Pol- tute, World Bank, Washington D.C. icy and International Trade." Working Paper Hettige, H., M. Huq, S. Pargal, and D. Wheeler. 1996. icy and International Trade. "Determinants of Pollution Abatement in Devel- 12-96. Sackler Institute for Economic Studies, Tel Aviv. oping Countries: Evidence from South and South- Gasiorek, Michael, Alasdair Smith, and Anthony J. east Asia." World Development 24: 1891-1904. Venables. 1992. "1992: Trade and Welfare-A Hoekman, Bernard, and Denise Eby Konan. 1998. G"In Trade Plows and Hoekman, Bernard M., and Denise Eby Konan. 1999. Generale Equliby afterium92, oditel byL.A.Winters. "Deep Integration, Nondiscrimination, and Euro- Trade Policy after 1992, edited by L. A. Winters. Mediterranean Trade." Policy Research Working Cambridge, U.K.: Cambridge University Press. Paperr2130. developmen Research Group, Gotcal,Pte,ad iohyM medn. 97 Paper 2130. Development Research Group, Gortschalk, Peter, and Timorhy M. Smeeding. 1997. World Bank, Washington, D.C. "Cross-National Comparisons of Earnings and Hufbauer, Gary, Barbara Kotschwar, and John S. Wil- Income Inequality." Journal of Economic Litera- son. 2000. "Trade Policy, Standards, and Devel- Grossman, G. M., and A. B. Krueger. 1992. "Environ- opment in Central America." Paper prepared for a World Bank Institute seminar, Panama City, mental Impacts of a North American Free Trade June 27-29. World Bank Institute, Washington, Agreement." Discussion Paper 644. Centre for D.C. Economic Policy Research, London. ILO (International Labour Organisation). 1993. Multi- Han, K. 1996. Dissertation on U.S. Manufacturing and nationals and Employment. Geneva. Environmental Regulation. University of Illinois, . 2000. Review of Annual Reports unlder the Fol- Urbana-Champaign, llt. low-Up to the ILO Declaration on Funzdamental Hanson, Gordon H., and Ann Harrison. 1999. "Trade Principles and Rights at Work. Geneva. Liberalization and Wage Inequality in Mexico." Jaffe, A., S. Peterson, P. Portney, and R. Stavins. 1995. Industrial and Labor Relations Review 52: "Environmental Regulation and the Competitive- 271-88. ness of U.S. Manufacturing: What Does the Evi- Harrison, Glenn W., Thomas F. Rutherford, and David dence Tell Us?" Journal of Economic Literature G. Tarr, 1996. "Increased Competition and Com- 33: 132-63. pletion of the Market in the European Union: Sta- Jha, V., A. Markandya, and R. Vossenaar. 1999. Rec- tic and Steady State Effects." Journal of Eco- onciling Trade and the Environment: Lessons nomic Integration 11 (3): 332-65. from Case Studies in Developing Countries. Chel- Hartman, Raymond S., Mainul Huq, and David tenham, U.K.: Edward Elgar. Wheeler. 1997. "Why Paper Mills Clean Up: De- Jha, V., and S. Zarrilli. 1994. "Eco-labeling Initiatives terminants of Pollution Abatements in Four Asian as Potential Barriers to Trade." In Life-Cycle 105 G LO B AL EC ONO MI C PRO SP EC TS Management and Trade, 64-73. OECD Docu- . 1997b. "NIMBY Taxes Matter: States Taxes ments. Paris. and Interstate Hazardous Waste Shipments." Uni- Jian, Tianlun, Jeffrey D. Sachs, and Andrew M. versity of Wisconsin, Madison, Wis. Warner. 1996. "Trends in Regional Inequality in Li, Hongyi, Lyn Squire, and Heng-fu Zou. 1998. "Ex- China." Working Paper 5412. National Bureau of plaining International and Intertemporal Varia- Economic Research, Cambridge, Mass. tions in Income Inequality." Economic Journal Johnson, George. 1997. "Changes in Earnings Inequal- 108 (January): 26-43. ity: The Role of Demand Shifts." Journal of Eco- Lister, Bruce A. 1987. "Comparison of U.S. Laws and nomic Perspectives 11 (2): 41-54. Regulations Concerning Labeling of Prepackaged Johnson, Harry G., and Peter Mieszkowski. 1970. Foods with the Codex Alimentarius Draft Gen- "The Effects of Unionization on the Distribution eral Standard for Labeling of Prepackaged of Income." Quarterly Journal of Economics 84 Foods." Food, Drug, and Cosmetic Law Journal (4): 539-61. 42: 175-83. Jones, Ronald W. 1971. "Distortions in Factor Lloyd, P. J. 1992. "The Problem of Optimal Environ- Markets and the General Equilibrium Model of mental Policy Choice." In The Greening of World Production." Journal of Political Economy 79: Trade Issues, edited by K. Anderson and R. 437-59. Blackhurst. Chap. 2. Ann Arbor: University of Kalt, J. R 1988. "The Impact of Domestic Environ- Michigan Press. mental Regulatory Policies on U.S. International Lucas, R. E .B., D. Wheeler, and H. Hettige. 1992. Competitiveness." In International Competitive- "Economic Development, Environment Regula- ness, edited by A. M. Spence and H. A. Hazard. tion and the International Migration of Toxic In- Cambridge, Mass.: Ballinger. dustrial Pollution: 1960-1988." In Patrick Low, Katz, Lawrence E, and Kevin M. Murphy. 1992. ed., International Trade and the Environment, "Changes in Relative Wages, 1963-1987: Supply 67-86. Discussion Paper 159. World Bank, Wash- and Demand Factors." Quarterly Journal of Eco- ington, D.C. nomics (February): 35-78. Mani, Muthukumara, and David Wheeler. 1998. "In Khalid, R., and J. B. Braden. 1993. "Welfare Effects of Search of Pollution Havens? Dirty Industry in the Environmental Regulation in an Open Economy: World Economy 1960-1995." Journal of Envi- The Case of Malaysian Palm Oil." Journal of ronment and Development7 (3). Agricultural Economics 44 (January): 25-37. Mani, Muthukumara, Sheoli Pargal, and Mainul Huq. Klevorick, Alvin. 1997. "Reflections on the Race to the 1997. "Does Environmental Regulation Matter? Bottom." In Fair Trade and Harmonization: Pre- Determinants of the Location of New Manufac- requisites for Free Trade? edited by J. N. Bhag- turing Plants in India 1994." Policy Research wati and R. E. Hudec. Cambridge, Mass.: MIT Working Paper 1718. World Bank, Washington, Press. D.C. Leamer, Edward E. 1984. Sources of International Martin, Will, and Keith E. Maskus. 1999. "Core Labor Comparative Advantage. Cambridge, Mass.: MIT Standards and Competitiveness: Implications for Press. Global Trade Policy." World Bank, Washington, - 1998. "In Search of Stolper-Samuelson Link- D.C. Processed. ages between International Trade and Lower Maskus, Keith E. 1997. "Should Core Labor Standards Wages." In Imports, Exports, and the American Be Imposed through International Trade Policy?" Worker, edited by Susan M. Collins. Pp. 141-202. Policy Research Working Paper 1817. Develop- Washington, D.C.: Brookings Institution Press. ment Research Group, World Bank, Washington, Lee, H., and D. Roland-Holst. 1997. "The Environ- D.C. ment and Welfare Implications of Trade and Tax . 2000a. "Regulatory Standards in the WTO: Policy." Journal of Development Economics 52: Comparing Intellectual Property Rights with 65-82. Competition Policy, Environmental Protection, Levinson, A. 1997a. "Environmental Regulations and and Core Labor Standards." World Bank, Wash- Industry Location: International and Domestic ington, D.C. Processed. Evidence." In Fair Trade and Harmonization: . 2000b. "Intellectual Property Rights in the Prerequisites for Free Trade? edited by J. N. Global Economy." Processed. Bhagwati and R. E. Hudec. Cambridge, Mass.: Maskus, Keith E., and John S. Wilson, eds. Forthcom- MIT Press. ing. Quantifying the Impact of Technical Barriers 106 S TAN DAR DS, DE V EL O PING CO UN T RI ES, AND T HE G LO B AL T RA DE SYS T E M to Trade: Can It Be Done? Ann Arbor: University Panagariya, Arvind. 1999a. "Labor Standards in the of Michigan Press. WTO and Developing Countries: Trading Rights Matutes, C., and P. Regibeau. 1996. "A Selective Re- at Risk." University of Maryland, College Park, view of the Economics of Standardization: Entry Md. Deterrence, Technological Progress and Interna- . 1999b. "TRIPs and the WTO: An Uneasy tional Competition." European Journal of Politi- Marriage." University of Maryland, College Park, cal Economy 12.(2): 183-209. Md. Messerlin, Patrick A., and Jamel Zarrouk. 2000. Pargal, Sheoli, and David Wheeler. 1996. "Informal "Trade Facilitation: Technical Regulations and Regulation of Industrial Pollution in Developing Customs Procedures." World Economy 23 (4): Countries: Evidence from Indonesia." Journal of 577-93. Political Economy 104 (December): 1314-27. Metcalfe, M. R. 2000. "Environmental Regulation and Repetto, Robert. 1995. Jobs, Competitiveness and Envi- Implications for the U.S. Hog and Pork Indus- ronmental Regulation: What Are the Real Issues? tries." Ph.D. dissertation. North Carolina State Washington, D.C.: World Resources Institute. University, Raleigh, N.C. Reppelin-Hill, V. 1999. "Trade and Environment: An Michaely, M., D. Papageorgiu, and A. Choksi. 1991. Empirical Analysis of the Technology Effect in the Liberalizing Foreign Trade: Lessons of Experi- Steel Industry." Journal of Environmental Eco- ence in the Developing World. Cambridge, Mass.: nomics and Management 38 (3): 283-301. Basil Blackwell. Richardson, J. David. 2000. "The WTO and Market- Moenius, Johannes. 2000. "Information versus Prod- Supportive Regulation: A Way Forward on New uct Adaptation: The Role of Standards in Trade." Competition, Technological, and Labor Issues." Northwestern University, Kellogg Graduate Review (Federal Reserve Bank of St. Louis) 82 School of Management, Evanston, Ill. Processed. (4): 115-26. National Research Council. 1995. Standards, Confor- Robbins, Donald J. 1996. "Evidence on Trade and mity Assessment, and Trade: Into the 21st Cen- Wages in the Developing World." Technical Paper tury. Washington, D.C.: National Academy Press. 119. OECD Development Center, Paris. Nimon W., and J. Beghin. 1999."Ecolabels and Inter- Roberts, Donna. 1998. "Implementation of the WTO national Trade in the Textile and Apparel Mar- Agreement on the Application of Sanitary and ket." American Journal of Agricultural Econom- Phytosanitary Measures: The First Two Years." ics 81: 1078-84. IATRC Working Paper 98-4. International Agri- Nordstrom, H., and S. Vaughan. 1999. Trade and En- cultural Trade Research Consortium. Available at vironment. Special Studies 4. Geneva: World . Trade Organization. Rock, M. T. 1996. "Pollution Intensity of GDP and O'Connor, D. 1994. Managing the Environment with Trade Policy: Can the World Bank Be Wrong?" Rapid Industrialisation: Lessons from the East World Development 24 (3): 471-79. Asian Experience. Paris: Development Centre Rollo, Jim, and L. Alan Winters. 2000. "Subsidiarity Studies. and Governance Challenges for the WTO: Envi- OECD (Organisation for Economic Co-operation and ronmental and Labor Standards." World Econ- Development) 1996. Trade, Employment and omy (4): 561-76. Labour Standards. Paris. Runge, C. E, with E Ortalo-Magne and P. Van de . 1997a. "Eco-Labeling: Actual Effects of Se- Kamp. 1994. Freer Trade, Protected Environ- lected Programmes." (97)105. Paris. ment: Balancing Trade Liberalization and Envi- . 1997b. Regionalism and Its Place in the Mul- ronmental Success. New York: Council on For- tilateral Trading System. Paris. eign Relations Press. . 1999. An Assessment of the Costs for Interna- Slaughter, Matthew J. 1999. "Globalization and Wages: tional Trade in Meeting Regulatory Require- A Tale of Two Perspectives." World Economy 22 ments. Paris. (5): 609-30. . 2000. International Trade and Core Labour - . Forthcoming. "Trade Liberalization and Per Standards. Paris. Capita Income Convergence: A Difference-in- Otsuki, Tsunehiro, John S. Wilson, and Mirvat Se- Differences Analysis." Journal of International wadeh. 2000. "Saving Two in a Billion: A Case Economics. Study to Quantify the Trade Effect of European Stephenson, Sherry M. 1997. "Standards and Confor- Food Safety Standards on African Exports." mity Assessment as Nontariff Barriers to Trade." World Bank, Washington, D.C. Processed. Policy Research Working Paper 1826. Develop- 107 G LO B AL EC ONO MI C PRO SP EC TS ment Research Group, World Bank, Washington, tion Levy System." Policy Research Paper 2336. D.C. Infrastructure and Environment, Development Re- Stiglitz, Joseph E. 2000. "Democratic Development as search Group, World Bank, Washington D.C. the Fruits of Labor." Keynote address to the In- Wang, Zhen Kun, and L. Alan Winters. 2000. "Putting dustrial Relations Research Association, Boston, 'Humpty' Together Again: Including Developing January. Processed. Countries in a Consensus for the WTO." Policy Strutt, A., and K. Anderson. 1999. "Will Trade Liber- Paper 4. Centre for Economic Policy Research, alization Harm the Environment? The Case of London. Indonesia to 2020." In Per G. Fredriksson, ed., Whalley, John, and Colleen Hamilton. 1996. The Trad- Trade, Global Policy, and the Environment, ing System after the Uruguay Round. Washington, 13-34. Discussion Paper 402. World Bank, Wash- D.C.: Institute for International Economics. ington, D.C. Wheeler, D., and P. Martin (1992). "Price, Policies, and Subramanian, Arvind. 1992. "Trade Measures for En- the International Diffusion of Clean Technology: vironment: A Nearly Empty Box?" World Econ- The Case of Wood Pulp Production," in P. Low omy 15: 135-52. Ed., International Trade and the Environment, - 1999. "TRIPs and Developing Countries: The World Bank Discussion Papers 159, Washington, Seattle Round and Beyond." Paper presented D.C., The World Bank: 197-224. to the Conference on Developing Countries and Wilson, J. D. 1997. "Capital Mobility and Environ- the New Multilateral Round of Trade Negotia- mental Standards: Is There a Theoretical Basis for tions, Harvard University, Cambridge, Mass. a Race to the Bottom?" In Fair Trade and Har- November. monization. Prerequisites for Free Trade? edited Sykes, Alan 0. 1995. Product Standards for Interna- byJ. N. Bhagwati and R. E. Hudec. Pp. 393-428. tionally Integrated Goods Markets. Washington, Cambridge, Mass.: MIT Press. D.C.: Brookings Institution. Wilson, John S. 1999. "The Post-Seattle Agenda of the Szekely, Miguel, and Marianne Hilgert. 1999. "In- WTO in Standards and Technical Barriers to equality in Latin America during the 1990s." Trade: Issues for the Developing Countries." De- Inter-American Development Bank, Washington, velopment Research Group, World Bank, Wash- D.C. Processed. ington D.C. Tiebout, C. M. 1956. "A Pure Theory of Local Expen- 2000a "The Development Challenge in Trade: ditures." Journal of Political Economy 64: S an "he vopmnt Change i ade: 416-24. Sanitary and Phytosanitary Standards." Paper Tobey, J. A. 1990. "The Effects of Domestic Environ- prepared for the WTO Sanitary and Phytosani- mental Policies on Patterns of World Trade: An tary Standards Committee. Development Re- Empirical Test." Kyklos 43: 191-209. search Group, World Bank, Washington, D.C. Ulph, A. M. 1999. Trade and the Environment: Se- . 2000b. "Technical Barriers to Trade and Stan- lected Essays of A. M. Ulph. Cheltenham, U.K.: dards: Challenges and Opportunities for Develop- Edward Elgar. ing Countries presented by the World Bank." UNCTAD (United Nations Conference on Trade and Paper submitted to the Technical Barriers to Trade Development). 1993. Program on Transnational Committee Meeting, World Trade Organization. Corporations. Environmental Management in Development Research Group, World Bank, Wash- Transnational Corporations: Report on the Bench- ington, D.C. mark Environmental Survey. Geneva. Wood, Adrian. 1997. "Openness and Wage Inequality Valluru, S. R. K., and E. W. F. Peterson. 1997. "The in Developing Countries: The Latin American Impact of Environmental Regulations on World Challenge to East Asian Conventional Wisdom." Grain Trade." Agribusiness 13 (3): 261-72. The World Bank Economic Review 11 (1): 33-58. Vukina, T., J. C. Beghin, and E. G. Solakoglu. 1999. World Bank. 1997. "Five Years after Rio. Innovations "Transition to Markets and the Environment: Ef- in Environmental Policy." Environment Depart- fects of the Change in the Composition of Manu- ment, Washington, D.C. facturing Output." Environment and Develop- - . 2000a. "Report on Standardization and Con- ment Economics 4 (4): 582-98. formity Assessment in Ukraine." Washington, D.C. Wang, Hua, and David Wheeler. 2000. "Endogenous . 2000b. Trade Blocs. Policy Research Report. Enforcement and Effectiveness of China's Pollu- New York: Oxford University Press. 108 STAN DA RD S, DEVELOP ING CO UN T RI ES, AND T HE G LO B AL T RA DE SYS T E M WTO (World Trade Organization) "Overview of the Xu, X. 1999. "Do Stringent Environmental Regula- State-of-Play of WTO Disputes." Geneva. Avail- tions Reduce the International Competitive- able at . Global Perspective." World Development 27 (7): Xing, Y., and C. Kolstad. 1995. "Do Lax Environmen- 1215-26. tal Regulations Attract Foreign Investments?" Zarsky, L. 1994. "Towards an International Ecola- Working Papers in Economics 06/95. University belling Framework." In Life-Cycle Management of California, Santa Barbara. and Trade, pp. 194-205. OECD: Paris. 109 Electronic Commerce and Developing Countries T HE INTERNET IS GLOBALIZATION ON STER- The Internet will boost productivity in de- oids. It will boost efficiency and en- veloping countries by increasing the efficiency hance market integration domestically of the procurement system, strengthening in- and internationally, particularly in developing ventory control, lowering retail transaction countries that are most disadvantaged by poor costs, and eliminating or transforming inter- access to information. Although the Internet mediaries. Virtual proximity to industrial should enhance global growth, it also brings in- country markets will increase as the Internet creased danger of economic marginalization to reduces the costs inherent in operating at a countries that cannot access it effectively. Tak- distance. Given the wide differences in returns ing advantage of electronic commerce requires to factors in developing versus industrial policies similar to those needed to capitalize on countries, this increased proximity will gener- the opportunities for trade: improved interna- ate large gains from trade in sectors that lend tional coordination, for example in ensuring themselves to electronic commerce. interoperability of communications technology Consumers will benefit from increased and confronting challenges to domestic tax and competition and market transparency, but the financial systems; an open economy promoting benefits to firms will vary greatly, depending competition and diffusion of Internet technolo- on the sector, degree of product differentia- gies; and efficient social and infrastructure ser- tion, and level of technological sophistication. vices, in particular a competitive telecommuni- Developing-country firms that sell labor- cations sector and a well-educated labor force. intensive, differentiated products (such as Despite the obvious benefits of the Internet, crafts, software, or business services-particu- uncertainty exists about the implications of larly services involving the remote processing this technology and its likely rate of diffusion. of routine information) will experience in- This chapter provides a tentative view of the creased demand. These firms also will benefit implications of electronic commerce for devel- from the opportunity to leapfrog to the most oping countries, based on the theoretical liter- advanced technologies and from easier access ature, inferences from experience in industrial to advertising on global markets. countries, and anecdotal evidence. The discus- The impact of electronic commerce on devel- sion is inevitably somewhat more speculative oping countries' sales to global supply chains is than in other chapters. The evidence, however, uncertain. Reduced transactions costs should warrants four broad conclusions: provide greater interaction among multi- Firms in developing countries should enjoy nationals and technologically sophisticated productivity gains and expanded demand with firms in developing countries. Many developing- the spread of electronic commerce. country firms may lack the reputation required 111 G LO B AL EC ONO MI C PRO SP EC TS to bid on the newly created online exchanges, developing countries. Access, supported by however. Industrial-country multinationals also the growing use of cell phones as a major link may prefer integrating their operations more to the Internet, is expected to rise at a faster closely with a reduced number of the most ad- rate in developing countries than in industrial vanced firms, given the opportunities for mana- countries during the next 10 years. Internet ging tightly linked production processes through access is likely, nonetheless, to remain limited the Internet. in per capita terms, especially in the poorest Government action is critical to removing countries, and to remain well below levels al- impediments to electronic commerce. ready achieved in industrial countries. Access Network externalities imply that market by firms in developing countries may increase prices may not fully reflect the gains to the so- significantly, but the poorest developing coun- ciety from increased levels of Internet access; tries may still see their competitiveness im- hence the government has a role in speeding paired because of a lack of human capital and Internet diffusion. Complementary inputs, in- complementary services required for effective cluding telecommunications, transport and participation in electronic commerce. power infrastructure, and a well-educated labor force are critical to exploiting the Internet's po- Emergence of electronic commerce tential. Governments have also encouraged the ransacting business using electronic aids expansion of the Internet by subsidizing Inter- T is as old as the telegraph, which was in- net connections and investing directly in infra- troduced in the mid-nineteenth century. More structure, although such investments can crowd recently, electronic data interchange (EDI) sys- out private initiatives and may quickly become tems not residing on the Internet have facili- obsolete due to rapid changes in technology. tated business transactions worth trillions of Other policies can also contribute to boost- dollars. This chapter, however, explores how ing Internet use. An open foreign direct invest- the relatively new phenomenon of the Internet ment regime is necessary to promote dissemi- is likely to affect commerce. Several defini- nation of information technology and training. tions of electronic commerce exist, including Governments can help facilitate services that transactions where the Internet is used to evaluate and attest to the quality of output gather information, to order goods or services, from domestic firms, which could support and to make payments. A reasonable definition their access to global Internet exchanges. of electronic commerce would include com- Governments must provide a supportive legal mercial operations in which two of these three framework for electronic transactions, such as steps are taken electronically.1 recognition of digital signatures and legal ad- Although the chapter focuses on electronic missibility of electronic documentation. Gov- commerce (in keeping with the overall theme ernments can also encourage Internet expansion on trade), this is by no means the only, or nec- by moving procurement and administrative re- essarily the most important, way the Internet quirements (tax forms and permits, for exam- will affect developing countries. Increased ac- ple) online. Finally, it is desirable to avoid high cess to information holds enormous promise levels of taxation on critical inputs to elec- for bettering noncommercial aspects of the tronic commerce such as personal computers lives of people in the developing world-pro- and telecommunications equipment. viding health and education services from a The gap in Internet access between indus- distance, and more efficient government ad- trial and developing countries will persist ministration are but two examples. At the through the next decade. same time, the growth of the Internet will in- Access to the Internet is grossly unequal, crease the exposure of developing countries to with 30 percent of the U.S. population online material, such as pornography, that may be compared with an average of 0.6 percent in viewed as undesirable. 112 E L E C T R O N I C CO MM ERC E AND D EVE LOP ING C OU N T R I E S Electronic commerce in industrial countries The digital divide has grown rapidly, from next to nothing in the he distribution of Internet access among mid-1990s to $100 to $200 billion in 1999- 1 countries is severely unequal. Despite 2000 (figure 4.1). Nevertheless, the dollar value rapid growth in Internet access in developing of electronic commerce transactions is less than countries, industrial countries still account for 1 percent of the total U.S. gross domestic prod- the majority of Internet subscribers (figure 4.2). uct (GDP) of $23 trillion (and the business-to- More than 30 percent of U.S. residents had ac- business data refer to total turnover, not just cess to the Internet in 1999, compared with value added). Consumer Internet purchases eualu about two-thi erd onercent of retail 0.5 percent in Sub-Saharan Africa (figure 4.3). sales of goods in the United States (U.S. Depart- Electronic commerce is also relatively small in salen of goommercthe 1999),excludin ses (U.S.Dep most developing countries. In Latin America, ment of Commerce 1999), excluding services for example, electronic commerce is estimated such as travel, tickets, and financial brokers, for mllonin commer is coma and about one-third of 1 percent in the United at $459 million i 1999 (Lapper 2000), com- Kingdom and Germany (OECD 2000a).2 pared with a GDP of about $2 trillion. The importance of electronic commerce Internet access in the developing world rests not in its current size but in the likely varies greatly. Some countries, particularly in speed of its establishment as a significant ve- East Asia, have achieved impressive penetra- hicle for commerce and the potential for fu- tion rates. For example, the share of Internet ture growth. Electronic commerce is projected subscribers in Korea has grown rapidly and is to reach $4 trillion to $6 trillion in the United estimated at 20 percent of the population in States alone within the next three to four 2000, above rates in most European countries years (Bermudez and others 2000; Economist (Grebb 2000). Although per capita subscriber 1999).3 Electronic commerce may account for rates in China and India remain low, these as much as 25 percent of world trade by 2005 countries are so large that they have a critical (UNCTAD 1999). mass of subscribers ready to benefit from the Figure 4.1 Estimates of electronic commerce in industrial countries, 1999-2000 Billions of US. dollars 160 140 * Business to business 120 * Business to consumer 120 * [: Business to business and consumer 100 80 60 20 Suttle 2000 Teo 1999 OECD 2000a Phillips and Meeker Gartner 2000 Source: See Suttle 2000 in the chapter references. 113 G LO B AL EC ONO MI C PRO SP EC TS Figure 4.2 Estimates of Internet access, 1990-2000 Millions of subscribers 250 | United States * Western Europe E Japan * Developingcountries 200 150 100 50 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Source: Pyramid. Figure 4.3 Regional Internet access Percentage of population with access to the Internet 35 30 25 20 15 10 5 0 Industrial United Other Middle East Sub- Eastern Latin Asia countries States and Saharan Europe America North Airica Africa and the Caribbean Source: Pyramid. Internet, a situation that increases the potential which is the availability of telecommunications for electronic commerce transactions. services. Canning (1999) finds strong evidence that the quantity and quality of telecommuni- Analysis of Internet diffusion cations services provided in a country is a sig- The nascent condition of Internet diffusion in nificant determinant of the existence of Inter- many developing countries reflects the con- net connections and the level of Internet use; to straints on Internet use, the most important of date, almost all Internet users have depended 114 E L E C T R O N IC CO MM ER C E AND DE V EL O PING CO UN TR I ES on telephone lines for connection. The trends in "Internet intensity"-the ratio of Internet Figure 4.4 Access to telecommunications subscribers to available telephone lines-are Per 1,000 persons remarkably similar across developing and in- 600 dustrial countries, however. Urban density and _ OECD countries the policy environment for private sector de- 500 U Latin America velopment are strongly related to growth in In- and the Caribbean ternet intensity.4 Many developing countries 400 exc Sub-Saharan Africa, (including, on average, those in Asia, Latin e America, and Sub-Saharan Africa) are experi- 300 encing much more rapid diffusion of the Inter- net for the given availability of telephone lines 200 than is the United States. The digital divide re- sults from differential access to telecommuni- 100 cations, not from the use of the Internet after telecommunications are available. 0 Unfortunately, the gap in telecommunica- Telephone mainlines Mobile phones tions services between industrial and develop- Source: World Bank 2000b. ing countries is large, so the digital divide is likely to remain wide for some time. The gap is also wide among developing countries, with rope. Approximately 10 million users in Japan, the poorest countries being particularly disad- or 40 percent of total users there, accessed the vantaged. For example, the average OECD Internet through mobile telecommunications country had 70 times, and the average Latin devices in May 2000 (Reuters 2000). Low- American country had 17 times, the number of Earth-orbit satellite systems also have potential telephone mainlines than did countries in Sub- for reaching areas where telephone service is Saharan Africa (excluding South Africa) (fig- poor (Wood 1999).6 Work is underway to in- ure 4.4). By some indicators, the digital divide vestigate the feasibility of Internet transmission is widening. Wilson and Rodriguez (2000) find through power lines. that an index of between-country inequality in All of these links to the Internet will bypass access to communications (the components are the often inefficient and difficult-to-build tele- personal computers, Internet hosts, fax ma- phone networks now used for Internet access chines, mobile phones, and televisions) deteri- and will have substantially higher access orated substantially during the 1990s.5 speeds, although the forecasts in table 4.1 are speculative. The rapid diffusion of cellular Prospects for Internet access phone and other systems with Internet access Given the enormous investments required for will mean that much of the Internet's capacity telephone lines (and in some countries the con- tinued dominance of the telephone system by Table 4.1 Future Internet access speeds inefficient monopolies), hopes for narrowing the digital divide rest largely on the spread of Platform Yer avaible Potential bandwidth alternative means of accessing the Internet. Cellular 1999 144 kilobits The availability of cable, cellular phone, and 200 1.6 megabits satellite systems is likely to reduce dependence 2003 5.2 megabits Cable Iwo200 1-10 megabits on telephone lines for access to the Internet Satellite 205 64 megabits during the next decade. Digital cellular tele- Power grid (?) 2.5 gigabits(?) phone systems with Internet access are already Source: Harrow 2000; www.teledesic.com; www. spreading rapidly in Japan and Western Eu- mediafusioncorp.net. 115 G LO B AL EC ONO MI C PROS PE CT S may be available at relatively low cost in many new technologies (although at the same time developing countries. Although the most re- developing-country firms will benefit from lower cent experience with wireless Internet access prices and increased access to services offered has been disappointing in some respects (wit- by industrial-country firms). Finally, an overly ness the slow diffusion of Internet-enabled restrictive interpretation of current rules on in- wireless phones in the United States), over the tellectual property rights could constrain devel- medium term these alternative Internet plat- oping countries' access to some of these new forms are likely to give a significant boost to technologies, which have largely been devel- the spread of the Internet. oped in the industrial world. One potential Some insight into the implications of new issue is the patenting of business processes and platforms for Internet access and electronic methods linked to the Internet; for the time commerce can be gained by looking at the being that practice is found only in the United prospects for diffusion of cellular telephones. States. International recognition of such patents During the 1990s cell phone diffusion within could constrain the ability of firms in other countries was strongly influenced by per countries to compete. capita income, the change in per capita in- come, the size of the urban population, and the strength of the policy environment facing Effects on productivity in the private sector.7 Assuming future per capita industrial and developing income growth and policy performance will be countries equal to the 1990s experience, this equation E lectronic commerce will generate produc- forecasts very rapid growth in cell phone, and Etivity gains by reducing transaction costs. hence Internet, penetration during the next The rapid dissemination of information, the decade in all developing regions. Cell phone substitution of digital for paper record keep- use in developing countries as a group would ing, and the networking capabilities of the quadruple by 2010 compared with 1998. Internet will improve flexibility and respon- Cell phone penetration would remain low siveness, encourage new and more efficient in- relative to population, a projected 6 percent termediaries, increase the use of outsourcing, in 2010, compared with 2 percent in 1998. reduce time to market by linking orders to However, this figure does not imply that only production, and improve internal coordina- 6 percent of developing country residents will tion. Although the effect of electronic com- have access to cell phones, given the potential merce on productivity has probably been small for multiple use (although privacy concerns to date (Oliner and Sichel 2000), simulations may constrain multiple use in some circum- have indicated that electronic commerce could stances). For example, hundreds of people have raise output levels by some 5 percent in the access to the single cell phone provided to major industrial countries (Mann, Eckert, and each village participating in the Bangladesh Knight 2000; OECD 2000a).8 Firms can ex- Village Pay Phone program. pect productivity gains through improved sys- Increased access to the Internet is only one tems for procurement and inventory control precondition for effective participation in elec- and reduced costs of intermediation and sales tronic commerce. Many developing countries, transactions, as well as through more rapid particularly the poorest ones, lack the human diffusion of technology. Consumers also will capital and complementary services required benefit through reduced search costs, thus in- to make effective use of the latest technolo- creasing competition and reducing prices. gies. There also is concern that developing country firms will face increasing challenges in Procurement and inventory control competing with the leading firms in industrial Firms in developing countries can use the In- countries, which have a headstart in using these ternet to achieve the kinds of procurement and 116 EL EC TRON IC CO MM E RC E AND D EVE LO PING C O U N T R I E S inventory savings now enjoyed only by the The use of online auctions also can reduce largest firms that have established EDI sys- the price of inputs by improving transparency tems, simply by purchasing "out of the box" and facilitating competitive bidding. General electronic commerce applications (box 4.1). Electric, for example, has cut the cost of pur- Goldman Sachs (1999) estimates that 30 per- chased inputs by 10 to 20 percent through cent or more of the total cost of intermediate online bidding. The potential savings from in- goods typically are "process costs," or the creased transparency varies with the infor- costs of administering transactions and main- mation content of the good. Goldman Sachs taining inventories. The potential for savings (1999) estimates that the savings from pur- can be divided into reduced processing costs chasing online may vary from 2 percent in of procurement transactions, reduced price of relatively undifferentiated products (such as inputs attributable to increased competition, coal) to 40 percent in highly differentiated ones and improved inventory control. (some electronic components, for example). Substituting the Internet for paper-based Keeping an electronic inventory and trans- systems can reduce the cost of processing or- ferring information on replenishment needs ders by saving staff time, speeding up the over the Internet enables producers and retail- process, and reducing processing errors.9 Esti- ers to reduce the time that components and mates of the savings in processing costs of raw materials spend at each processing stage. Web-based procurement are 90-95 percent Even relatively small reductions in inventory (Schwartz 2000; U.S. Department of Com- holding time in retail trade can mean substan- merce 1999). tial increases in profits because the average Dox 4.1 iElctronic data iterchange (EDI) systems rm-E' systems provideoe viewof the potential effi- Internt-based systems are less than 1 percent of EDI eicygains fro rely the hnternet for pro- systems (Xie 2000). cesing ptocurement and inetory control. These That large firms are willing to make huge fixed sysems se propretary Isoftware to connep- investments and pay high operating costs for EDI chaser and supp ls computersd to utomate systems indicates the substantial gains that firms can the tr4nsaction inrma ex-captu by transferring from paper-based to elec- c-nngt fl)~ is to ssippoct abut$3 trillon trozic syste-ms. The smaller initial investment and coormn, - I tesone loroperatg costs in Intermet-based systems ISlips and 2*O0).- Abo percent of the (along with greater flexibility and transparency of among opeations) mearns that small- and mnedium-size firms .-, tune 500 companies in rI U0d captur these gains. d-cred through *3)1 Systems {Dermudn and others The migration of EDI systems to the Internet 2-001. D-eipite tis- only about 100 coss and to improve flexibility) rest -tof-million U. oa e i likely to boo the dollar value of elctonic witPor nrt T arge is - co ionsover the next few years. g*ited todevelopproprietarysoftwe cischanep wilbe Slow because of the reluctance -nvolin integrati-g new supprs effectively bars to bandoai the huge fixed costs represented in s- all firms from nsis~g-EPl. By conrrasrt, large £01 sysrem and the 'cost of conversion to ternet- :e qE utW invs requiredor sium i s stger 1999). Concen over, over the I-ternet has -headygbeeh mae d interac- wr of the Internet compared with txvny with other in fly pro- tha of Erop LDI systems also may limit vided. Also, it.is esSa1toperating costs in conve-rsins.- 117 G LO B AL EC ONO MI C PRO SP EC TS cost to retailers of holding inventory for a year cess to information, the Internet has enabled is at least 25 percent of the price, and margins the elimination of retailers, wholesalers, and may average only 3-4 percent (OECD 1999). (in the case of intangible products) even dis- Improved inventory control will enable firms tributors in some sectors. More commonly, to become more integrated with suppliers, existing middlemen have been replaced by thereby saving time and allowing greater pro- new approaches to intermediation made pos- duction specialization. Increased production sible by the technology-for example, online integration has led to a boom in specialized auctions and aggregators (firms that represent manufacturing firms that produce compo- collections of buyers that can demand lower nents for more well-known companies. A fa- prices for bulk purchases). mous example is Cisco, whose components The Internet also can generate significant are made to Cisco's specifications by suppli- cost savings in transport. The advertisement ers, tested through a connection to the Inter- and trading of empty truck space over the net, and then shipped directly to the buyer. Web is reducing costs per ton in the U.S. Procurement in most developing countries is trucking sector (Economist 1999). According slower, less efficient, and more labor-intensive to one industry estimate, $15 billion to than in industrial countries, so the technical ef- $20 billion annually in cost savings (4 to ficiency gains from transferring procurement 5 percent of output in the U.S. trucking indus- systems to the Internet could be relatively large try) may be realized (Business 2.0 1999). (although the lower cost of labor in developing Eliminating or transforming intermediary countries means that the economic gains could functions will enable developing-country pro- be more limited than in industrial countries). ducers to access both domestic and foreign The savings in working capital from reduced markets at lower cost. By contrast, firms in de- holding of inventories also would be significant veloping countries whose main purpose is to in developing countries, where the cost of help domestic companies trade with interna- capital is high and credit is often rationed or tional markets will be at particular risk. Net- unavailable. The lack of reliable telecommu- work externalities, combined with a low mar- nications networks and complementary ser- ginal cost of adding new users, mean that vices-for example, transport facilities-may the market for providing intermediary services limit these gains, however. Some limited survey offers considerable advantage to the first com- evidence (see annex 4 for description of survey) pany on the scene. Thus the later-arriving indicates that North American firms that were and less technologically sophisticated firms in better at supply-chain management to begin many developing countries may have difficulty with are cutting these costs by an even larger competing with industrial-country firms as amount when using Internet-based inventory Internet-based intermediaries (UNCTAD 2000). systems. This may be because an adequate sup- Developing countries also may not be able to ply of high-skilled workers and a flexible orga- capture the cost savings from reduced inter- nization are required to reap the full benefits of mediation in some sectors, such as primary these systems. commodity exports, where purchasers are likely to be the major beneficiaries of any cost Reduced intermediary costs savings (box 4.2). Productivity gains can be derived from elimi- nating or improving the efficiency of interme- Retail transactions diaries involved in marketing and distribution. The Internet offers the potential for savings Middlemen often charge substantial markups in retail transactions compared with tradi- because of their knowledge about and con- tional systems. OECD (1998b) suggests that tacts with suppliers. By greatly expanding ac- the greater availability of information to the 118 ELE CT R ON IC CO MM ER C E AND DE V EL O PING CO UN TR I ES Box 4.2 The Internet and primary commodity exporters E lectronic commerce may affect relatively homo- (for example, Forrester Research 1999a), but the Egeneous primary commodities less than it does efficiency gains will vary depending on the corn- more differentiated products because most of the modity. Commodities already traded on established necessary information is contained in the product exchanges (such as wheath, with widely disseni- price. Although electronic commerce is likely to pro- nated information on prices and centralized trading} vide some benefits to producers by increasing the ef- may not be greatly affected. Online auctions will ficiency of commodity markets, the major benefits have a greater role in reducing margins for com- will accrue to purchasers. modities (such as fertilizer} that trade through More timely access to market information about brokers with limited price transparency. For exam- prices could generate benefits to producers. Small- ple, brokerage fees in the sugar trade, which range holder farmers in remote areas could check the prices from 0.5 to 1.0 percent of the value of the com- in the nearest market (which could be a considerable modity, may decline as more trade is done over journey if done in person) before deciding whether to the Internet. sell to local middlemen; such a capability could po- Producers in developing countries are unlikely tentially improve the farmers' bargaining position to see substantial increases in incomes from lower with local and foreign buyers. The Internet also trading and -marketing costs; rathet; these gains will could provide producers with better information accrue to the consumer through lower prices for final about input prices and product availability and eas- products. Lower consumer prices may increase de- ier access to training about best production practices. mand only slightly, because the demand for most Several firms have projected that a large share commodities is price inelastic and the reduction in of commodity trade will occur over the Internet marketirg costs will probably be smalL consumer and savings on providing services reflects the transfer of costs from producers to could increase the productivity of sales staff in consumers in the form of time spent searching OECD countries by a factor of 10. The evi- the Internet. dence on the sale of goods over the Internet so The lower cost of service transactions is far does not show large savings, however. Pre- likely to have a less significant effect in devel- liminary studies found that goods sold on the oping than in industrial countries because the Internet were priced the same or higher than lower wages paid in developing countries mean in stores (Goldman Sachs 1997; Krantz 1998; that firms have less incentive to undertake the Lee, Westland, and Hong 2000; OECD 1998a). fixed costs involved in setting up electronic sys- Other studies estimated that books and com- tems. Also, poor distribution systems, inade- pact discs (CDs) were 10 percent cheaper on quate protection against credit card fraud, and the Internet (Economist 2000; Oliner and limited consumer Internet access constrain the Sichel 2000).10 The potential savings in ser- potential for business-to-consumer commerce vice transactions are more impressive. For ex- in many developing countries. ample, the total cost (including investment) of bank transfers over the Internet is half that of Knowledge acquisition and technology existing automated systems and one-eighth diffusion that of transactions using tellers (WTO 1998; Easier access to knowledge through the Inter- figure 4.5). Note that a portion of this savings net will speed technology diffusion, which is reflects efficiency gains, while another portion of critical importance to developing countries 119 G LOB AL E CON OMIC PROSPECTS Figure 4.5 Cost savings from electronic commerce Ratio of cost of traditional process to Intemet 35 30 25 20 15 10 5 0 Bank transfer Airline tickets Sending documents Software Notes: Bank transfer: total cost of teller and ATM transaction versus online. distribution Airline tickets: travel agent booking via computer reservations system versus consumer booking online. Sending documents: send a 42-page document by fax, courier, and airmail relative to e-mail. Software distribution: download to CD and ship versus electronic transmission. Source: Kibati 1999; OECD 1999; VWTO 1998. because they tend to operate within the tech- working" is greatly facilitated by the Internet. nological frontier. Electronic commerce can Harris (1998) quotes a Neilson survey that reduce the costs of communication between found business's primary use of the Internet geographically distant partners and lower the was for gathering information. search-and-compare costs involved in finding potential business partners and technologies. Moreover, the Internet provides a radial struc- Effects on international trade in ture for interpersonal communication net- developing countries works. Bulletin boards and news servers allow B y opening markets to a wider range of po- individuals to exchange information faster B tential buyers and sellers, the Internet is and within a wider environment than with likely to foster a greater volume and variety of networks based on telephone and fax. Con- trade. The Internet could erode an important nolly (1998) found that differences in commu- advantage now enjoyed by firms in industrial nication and transportation infrastructure countries: proximity to wealthy customers. were significantly related to differences in the For example, the Internet reduces the cost of rate of product imitation encouraged by for- producing customized products designed for eign direct investment (although this does not distant markets; consumers in the United necessarily mean that electronic commerce has States can purchase a hand-sewn suit made by an independent positive effect). Grossman and a tailor in Shanghai without ever visiting Helpman (1991) argue that international con- China (Xie 2000). tacts enable a country to obtain foreign tech- nologies and adjust them to domestic use, an Service exports important channel through which the produc- The Internet will reduce barriers to the sale tivity levels of industrial and developing coun- of services embodying skilled labor (Harris tries are interrelated. Such international "net- 1998).11 In the Philippines, for example, com- 120 E L E C T R O N I C CO MME RC E AND D EVE LO PING C OU N T R I E S panies use the Internet to provide accounting the International Standards Organization and services, process insurance claims, and track the International Electrotechnical Commis- credit card defaulters for firms in industrial sion) to assess independently the quality of countries (Jordan and Hilsenrath 2000). In new firms' products and services. However, India workers have been transcribing U.S. relatively few small firms, even in industrial physicians' oral records into written files countries, use the certification services these since 1996, at one-tenth the cost of U.S. tran- bodies provide, because of the cost and fears scription services (Mills 1996).12 Schuknecht that certification may not fully address buyers' and Perez-Esteve (1999) suggest that financial, concerns in the markets where small firms insurance, and other business and communica- compete (Callaghan and Schnoll 1997). tion services are likely to see the greatest im- pact from electronic commerce. Online advertisement New websites are emerging that provide a Multinational supply chains venue for smaller firms to advertise their The Internet's impact on access to multina- goods, and buyers to advertise their product tional supply chains by firms in developing needs.13 A survey of one of the best-known ad- countries is uncertain. Increased information vertising websites (www.alibaba.com), which on these firms may improve their access to has grown to 200,000 subscribers since its multinationals, which tend to use suppliers April 1999 inception, indicates that the impact with whom they have experience. Goldman has been modest so far, although it is too early Sachs (1999) estimates that, because of poor to judge the site's full potential. Most par- research, firms' purchasing managers tend to ticipants say they have found only a few new award 90 percent of their procurement con- customers so far. Only 13 percent of firms re- tracts to about 20 percent of suppliers. At the ported that total sales had gone up consider- same time, suppliers with poor hardware, ably since posting on the website, whereas software, and Internet transmission capabil- 64 percent reported some increase and the rest ities may be unable to compete with better- none (no firm reported a decline in sales). connected companies. It is unclear whether However, 87 percent of the firms viewed the the new online auction systems have resulted in the expansion of supply networks. General Electric, for one example, may be increasing Figure 4.6 New customers gained from the number of its suppliers through its online alibaba website bidding site for procurement. Percentage of respondents A lack of credibility may make it difficult 70 for firms in developing countries to access on- 60 line auctions. Purchasers need to have confi- dence that suppliers will provide input on time 50 and in conformance with specifications, and product quality may not be known ex ante. 40 More than half of 35 large firms using online 30 auction or exchange sites said that they would not do business through online Web sites with 20 firms they did not know (Forrester Research 1999b). Interview results indicate buyers- 10 typically firms in industrial countries-see an o especially high risk in purchasing from firms None Few Many in developing countries. Over time, greater use Source: See annex 4. may be made of certification agencies (such as 121 G LO B AL EC ON O M I C P R O S P E C T S 85 percent of Internet users are in Nairobi Figure 4.7 Increased sales reported (Jensen 1999; Kibati 1999). because of alibaba website Some aspects of electronic commerce could Percentage of respondents mitigate its impact on inequality, particularly 70 on the poor, in developing countries. The fall in production costs is likely to increase the de- 60 mand for all workers, despite the fall in the 50 per unit labor input in production. Although inequality may increase, the income of the 40 poor may rise. Also, electronic commerce in- creases market transparency, thus reducing 30 search costs and reliance on intermediaries. 20 These effects reduce the price of skill-intensive goods, thus raising the real incomes of work- 10 ers generally, although the poor are unlikely to o l _ l _ - l benefit greatly as consumers because of their None Moderate Considerable limited access. Source: See annex 4. A direct impact of technological change on inequality in developing countries has been difficult to show empirically, perhaps because website as helpful or very helpful, with the the adoption of new technologies has coin- smaller firms showing the greatest enthusiasm. cided with economywide structural reform (Rodriguez 2000). Studies have shown that, for industrial countries, the recent rise in in- Effects on income distribution come inequality has occurred during a period T he Internet improves access to and use of of rapid technological change in the informa- T information, which increases the produc- tion technology sector.16 tivity of capital, thus raising its return relative to labor (Rodriguez 2000). The Internet also increases the demand for skilled labor, partic- Impediments to Internet use and ularly in the information technology sector. By the role of policies in developing contrast, better information will tend to re- countries duce the demand for, and hence the relative re- he presence of network externalities, turn to, unskilled labor involved in the routine T where all participants gain from each ad- processing of transactions and (perhaps) retail dition to the network, implies that market sales. 14 There is a risk that this divergence in prices may not fully reflect the total benefit demand for skilled versus unskilled labor to society from increased Internet access. Thus could increase inequality between industrial government has an important role in speeding and developing economies as well as within Internet diffusion. Inappropriate policies and developing economies.15 Any rise in inequality the lack of complementary services, particu- may be exacerbated by opportunities for mi- larly affecting the telecommunications sector, gration as skills shortages in the information other infrastructure, human capital, and the in- technology sectors intensify in industrial coun- vestment environment severely constrain In- tries. Electronic commerce also may increase ternet access in developing countries. regional inequality. In many developing na- tions, Internet services rarely spread beyond Telecommunications the country's capital and a few large urban Poor telecommunications will limit the growth centers. In Kenya, for example, more than of electronic commerce. Required telecommu- 122 ELECTRON IC CO MM E RC E AND D EVE LOP ING C O U N T R I E S nications facilities include transmission facili- even traffic that originates and terminates do- ties connecting a country's domestic network mestically can cost the same as international to the greater Internet, the domestic Internet transmission. backbone, and connections from homes and The high cost of Internet access, the lack of businesses to the backbone network. The de- local loop infrastructure necessary for basic fects of domestic telecommunications services dial-up modem access, and the poor quality of may be less important for the larger firms in the local loop infrastructure that does exist all developing countries; these firms may find it impede connections to the domestic backbone. profitable to invest in telecommunications fa- Country comparisons show a strong relation- cilities (such as wireless) that bypass the local ship between usage price and Internet penetra- network. tion; for industrial countries the correlation State or privatized monopolies that control between the Internet hosts per capita and the international connections impose inefficient average cost of Internet access from 1995 to pricing structures and conditions,17 which means 2000 is -.73 (EU Commission 2000; OECD that many Internet service providers cannot 2000c). Developing countries face much higher afford to buy enough transmission capacity costs relative to incomes than do industrial for electronic commerce applications to func- countries (figure 4.8). For example, surveys in- tion without congestion. This poor state of dicate that some users in Beijing may spend an domestic backbone networks results in a large average of 35 percent of take-home salaries on volume of domestic Internet traffic being sent Internet access charges (Rosen 1999). to the United States before being returned to Internet use appears to be higher in coun- its region of origin (Cukier 1999). A growing tries where local phone service is charged at a number of African Internet sites are hosted on fixed rate than in those where callers are billed servers in Europe or the United States because by the minute. For example, in most Latin of poor infrastructure (Jensen 1999). Hence, American countries (Mexico being the major Figure 4.8 Internet monthly access charge as a percentage of GDP per capita, 1998 United States Australia Finland Japan Turkey_ Mexico Senegal Guinea Mozambique Ethiopia Uganda Sierra Leone 0 20 40 60 80 100 120 140 Source: ITU 1999. 123 G L OBAL EC ON OMI C P ROSPE CT S exception), local calls are charged per unit of can overcome inadequate local loop infrastruc- time (Oxford Analytica 1999), and telephone ture have been either shared facilities or wire- charges account for 40 percent of monthly ac- less local loop. Shared facilities, which involve cess costs (E-Marketer 2000). In the United local entrepreneurs selling the use of a com- States the marginal cost of local calls is zero. puter with Internet access, are a fast and rela- Note that subsidizing Internet access through tively cheap way of increasing Internet use. For flat rate charges for local calls may discourage example, the number of Internet users on each investment in alternative forms of Internet ac- Internet account in Egypt is estimated to be be- cess that ultimately could be more efficient tween 2.5 and 4.5 people (El-Nawawy and Is- (EU Commission 2000). mail 1999). Public access Internet caf6s exist in For many developing countries, the most some 110 countries (Rao 1999). important issue is the lack of telephone service Wireless and satellite technologies also pro- to homes and businesses. Despite increases in vide an alternative to the high costs and ineffi- rates of telephone line penetration during the ciencies of many domestic telecommunications 1990s, more than one-third of the 130 devel- systems. Although currently used primarily for oping countries (excluding small islands) with voice, mobile phones "soon will be a much data for 1998 have fewer than 5 telephone better device for many of the usual Internet ap- lines per 100 inhabitants (figure 4.9). The com- plications," according to some technologists.18 parable level in the United States is 66. Cellular phones in some developing countries The quality of access also is important, as have experienced strong growth rates and rela- some electronic commerce applications that tively high penetration, similar to those in in- rely on sophisticated technology and high user dustrial countries. In Haiti, for example, poor interactivity require low congestion and high telephone service (0.9 phone lines per 100 bandwidth transmission between the user's ac- people, less than half Africa's average, and cess device and the host server. The most pop- huge waiting lists for new lines) has led to the ular alternatives by which developing countries growth of wireless service (Peha 1999). In 1998, Ecuador, the Slovak Republic, and West- ern Samoa had ratios of cellular phone sub- Figure 4.9 Telephone mainlines per 100 scriptions to regular phone service similar to inhabitants, developing countries, 1998 those of industrial countries (ITUJ 1998). On Number of countries average, however, cellular phone penetration 60 remains well below industrial-country levels. Sub-Saharan Africa averages 5 cellular phones 50 per 1,000 people, compared with 265 cellular phones for every 1,000 people in high-income 40 countries (World Bank 2000b). Impressive increases in penetration can be 30 achieved through increasing competition, al- though in some cases privatization has meant 20 reducing subsidies to local calls, with a nega- tive impact on Web access, at least in the short 10 term (Zambrano 1999). The relative level of capital spending on communication infra- Fewer than 5 5 to 20 More than 20 structure and development of Internet applica- tion software generally tends to be higher and Note: United States = 66 telephone mainlines per 100 inhabitants. more advanced in those industrial countries Source; World Bank data. that liberalized telecommunications markets earlier (OECD 2000a). Perhaps as a result of 124 E LE CT RON IC CO MM ERC E AND D EVE LOP ING C O U N T R I E S privatization and liberalization, Africa has re- American consumers are unwilling to pur- corded its highest annual growth rate in tele- chase goods over the Internet because credit phone mainlines for a decade (AED 1998a). card companies will not compensate holders Figure 4.10 reveals the growing trend of pri- for fraudulent use of cards (in many industrial vatization in the telecommunications sectors countries, cardholders have only a limited ex- of developing countries. posure to loss). This lack of security does not make consumer purchases on the Internet Other infrastructure impossible. In China, companies are depend- Poor infrastructure services (other than tele- ing on cash payments and local distribution communications) are an important constraint through taxis and bicycles to reach consumers on electronic commerce. Frequent and long (Fan 2000). power interruptions can seriously interfere with data transmission and systems performance, Human capital so many Bangalore software firms have their A critical mass of highly skilled labor is needed own generators (Panagariya 1999). Mail ser- in developing countries to supply the necessary vice can be unreliable, expensive, and time- applications, provide support, and disseminate consuming in many developing countries. For relevant technical knowledge for electronic example, the unreliability of postal services in commerce. The work force in many develop- Latin America has meant that more expensive ing countries lacks a sufficient supply of these courier services must be used to deliver goods skills, and the demand for this specialized ordered over the Internet, and in response, in- labor from industrial countries has further ternational courier services are setting up spe- strained the supply of this labor in developing cial distribution systems in Miami (Lapper countries. In the mid-1990s North America 2000; Oxford Analytica 1999). and Europe had unfilled demand for profes- The lack of safeguards against fraud can se- sionals trained in information technology (fig- verely restrict credit card purchases, the most ure 4.11). The wages of workers in informa- common means of conducting transactions tion technology industries continue to rise over the Internet. For example, many Latin more rapidly than those of workers in other Figure 4.10 Developing country privatization in telecommunications, 1990-98 Billions of US. dollars 30 25 20 15 10 5 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 Source: World Bank 2000b. 125 G LO B AL EC ON O M I C P R O S P E C T S be taxed at rates exceeding 50 percent. The Figure 4.11 Information technology jobs overall environment for private sector activities unfilled because of skill shortages, 1998 is a significant determinant of Internet service Thousands diffusion. An open foreign direct investment 400 regime helps promote technology diffusion, which is important to the growth of electronic 350 commerce. Foreign direct investment also is one 300 channel that could facilitate certification of do- mestic firms for access to online auctions. 250 Governments also can play an important 200 role in supporting the certification of firms by 150 providing information on certification proce- dures, promoting access by domestic firms to 00 U Uinternational organizations and firms that 50 provide certification, and perhaps subsidizing the costs of certification to demonstrate the 0 l ~ l ~ l ~ l _ l - kinds of resources available in the domestic United Other Germany Canada United States Kingdom market. This role will be particularly impor- Source: OECD 1998b. tant (at least in a transitional sense) as the intermediaries that formerly helped connect developing-country firms to international markets are replaced by Web-based intermedi- U.S. industries (U.S. Department of Commerce aries that may have less information on devel- 1999). U.S. firms will be able to fill only half of oping countries. the estimated 1.6 million positions open in Governments must provide a supportive 2000 (ITAA 2000), while the shortfall of in- legal framework for electronic transactions, in- formation technology workers is estimated at cluding recognition of digital signatures; legal almost 1 million in Japan (OECD 2000b). admissibility of electronic contracts; and estab- At the same time, Harris (1998) notes that lishment of data storage requirements in paper the Internet also facilitates the mobility of skilled form, intellectual property rights for digital labor services. Workers can choose to remain content, liability of Internet service providers, in their own country while exporting labor privacy of personal data, and mechanisms for services to higher-paying industrial countries.19 resolving disputes. The U.N. Commission on Developing countries may also reap some ben- International Trade Law has a "Model Law on efits from migration. For example, Indians in Electronic Commerce" that offers national leg- Silicon Valley have played a role in provid- islatures legal principles and guidelines for ing capital, expertise, and business contacts dealing with some of these issues (Price Water- to Indians in the software exporting firms of house Coopers 1999; UNCTAD 2000). Bangalore. Governments also have had considerable impact on Internet use through direct inter- Investment environment ventions. Singapore is providing grants to Several regulatory impediments to the wide- local companies to encourage participation in spread adoption of electronic commerce exist in electronic commerce (Price Waterhouse Coop- many developing countries. Duties and taxes on ers 1999). Malaysia is wiring a zone south of computer hardware and software and comrnmu- Kuala Lumpur with fast communications nication equipment increase the expense of con- (Bickers 1999). The "Wiring the Border" proj- necting to the Internet. For example, a com- ect is providing subsidies to small businesses puter imported into some African countries may along the Mexico-U.S. border to finance Inter- 126 E L E C T R O N I C CO MM ER C E AND DE V EL O PING CO UN TR I ES net access. The U.S. Department of Defense played a critical role in developing the initial Figure 4.12 Internet use in industrial networking technologies (Goodman and oth- countries by knowledge of English, ers 1994). The U.S. government also financed 1998-99 the original Internet backbone until increased Percentage of population demand for services led to the creation of 35 commercial backbones; a similar pattern was followed in several other industrial countries 30 (Braga and Fink 1997). Despite some success 25 stories, however, the rapidity of technological change greatly increases the riskiness of gov- 20 ernment interventions to support Internet ac- _ cess. The expenditure of billions of dollars to connect schools to the Internet through tele- 10 phone lines could be wasted if wireless or power line technology turns out to be less ex- 5 pensive (Davis and Seib 2000). Furthermore, 0 government investments may crowd out pri- English is Knowledge of Other vate sector initiatives that could provide ser- first language English is vices more efficiently. for many common v ices ly, m or e reff ently. cansupportthespread First groups Australia, Canada, Ireland, United Kingdom, Finally, government can support the spread United States. of the Internet by switching to online services Second group: Denmark, Finland, Norway, Sweden. Third group: Austria, Belgium, France, Germany, Greece, for its own transactions. Public sector procure- Italy, Japan, Portugal, Spain, Switzerland. ment, many aspects of tax and customs ad- Source: www.headcount.com. ministration, the processing of routine applica- tions (such as car permits and real estate licenses) and other governmental functions can often be carried out through the Internet. iarity with English was the principal determi- Decisions on the use of the Internet in public nant of Internet use in Slovenia, for all eco- administration should be based on the costs nomic groups (Vehovar, Batagelj, and Lozar of providing services relative to paper-based 1999). Conversely, Internet content is limited processes, the capability of government per- in the local language of most developing coun- sonnel, and the extent of demand. Neverthe- tries. From a commercial aspect, Schmitt less, greater government use of the Internet can (2000) found that just 37 percent of For- play a role in encouraging public participation. tune 100 websites support a language other than English.20 Language The amount of non-English material on the That most Internet business is conducted in Web is growing, however. Spanish websites in English is currently an important constraint particular are increasing, in part to serve the on using the Internet. Estimates of the share of large Spanish-speaking community in the English used on the Internet range from 70 to United States (Vogel and Druckerman 2000). 80 percent, but only 57 percent of Internet Improvements in translation services (by peo- users have English as their first language (ITU ple and machines), as well as Web browsers 1999; Vehovar, Batagelj, and Lozar 1999). Per that recognize characters of different lan- capita Internet use averages about 30 percent guages, should ease language constraints (U.S. in those industrial countries where English is Department of Commerce 1999). There is grow- common, compared with about 5 percent in ing recognition that English-only content is in- other industrial countries (figure 4.12). Famil- sufficient for an international economy.21 127 G LO B AL EC ONO MI C PRO SP EC TS Challenges to regulatory regimes transactions, which also are difficult to moni- in developing countries tor. Barter exchanges using digital money are E lectronic commerce will pose difficult small now, but the potential for growth could Echallenges for government regulation of be great when a critical mass of participants tax and financial systems. The growth of elec- is reached (Washington Post 2000). Reduced tronic commerce will encourage tax compe- transaction costs also might make it easier for tition and may facilitate some forms of tax taxpayers to hide potentially taxable activi- evasion. Competitive pressures in domestic ties. The impact of electronic commerce on banking systems will rise, generating substan- tax evasion should not be overestimated, how- tial benefits to consumers and firms but po- ever. The sale of domestic goods will still be tentially lowering the franchise value of exist- controlled by monitoring companies' transac- ing banks. tions, and imported goods will be controlled at the border. Tax policy Tax authorities will require greater exper- The growth of electronic commerce will pre- tise in information technology, both to im- sent some challenges to tax enforcement in prove the efficiency of tax administration and developing countries and place increased em- to enhance government ability to obtain and phasis on improving the technological sophisti- understand records of electronic transactions. cation of tax authorities and on international In this respect, the greater transparency and coordination of tax collection efforts. The In- ease of retrieval of electronic transactions (com- ternet reduces the cost to firms of being physi- pared with paper processes) could assist tax cally far away from customers and increases enforcement. the ability of companies to relocate production, because a substantial share of the work in- Financial sector and capital flows volved in organizing production is carried out Electronic commerce could pose a significant by computers that can be located anywhere. challenge to government regulation of the fi- Thus multinationals will find it easier to shift nancial sector by reducing the franchise value activities to low-tax regimes. Governments may of domestic banks, thus increasing the incen- find it more difficult to impose desired income tive for banks to undertake excessive risk. At tax levels on existing corporations, and com- the same time, electronic commerce will gen- petition among developing countries for in- erate substantial benefits to consumers and vestment by multinationals may rise. This situ- firms that rely on banking services. Competi- ation will place a greater premium on efforts to tion in domestic banking systems will increase reach agreements among developing countries because of reduced costs, greater access by to limit this kind of competition. foreign banks, and greater reliance on capital In addition, some of the new transactions markets by former bank borrowers. conducted through electronic commerce will Online banking may reduce banking trans- be difficult to monitor. Governments may not action costs by 15-25 percent compared with be able to detect the transfer of digitized ma- regular accounts (Morgan Stanley Dean Wit- terial and thus know that a taxable transac- ter 1999). Enhanced transparency and compe- tion has occurred.22 It may also be difficult to tition will mean that a large part of these cost control such transactions through the supplier reductions are transmitted to the consumers of because companies can easily provide such banking services. In fact, some degree of price services from other jurisdictions. Thus effec- competition among banks is already becoming tive international agreements to assist with tax apparent in the emerging economies (Gold- enforcement will be important. man Sachs 2000). The Internet increases the potential for con- The Internet will greatly increase the ability sumer-to-consumer purchases and for barter of foreign banks to compete in developing 128 ELECTRONIC CO MM ER C E AND DE V EL O PING CO UN T RI ES countries' domestic markets. The geographic location of the consumer and the service pro- Figure 4.13 Bond versus bank financing vider will become increasingly irrelevant, Millions of U.S. dollars eroding the local banks' advantage of having 1,200 large branch networks. First-mover advan- tages, economies of scale, and reputational 1,000 advantages backed by strong supervisory sys- tems will strengthen the competitive position 800 Comml banks of industrial-country banks. They will be in a 6 better position than domestic banks to offer 600 integrated trade payments systems, in which customers can obtain applications for a letter 400 of credit, care of credit approvals, telegraphic 200 transfers, invoices, and confirmations (Granit- sas 2000). One indication of the potential for 0 foreign firms' inroads into developing-country 1980 1990 1999 banking systems comes from a 1997 study, Source: World Bank 2000. which indicated that U.S. and U.K. firms dom- inated global Internet financial services mar- ket because of their reputations, head start, and the predominance of English on the Inter- of private source debt that was held in bonds net (OECD 1999). in developing countries rose from about one- Furthermore, the Internet will reduce the fourth in 1990 to more than one-half in 1999 advantage that domestic banks enjoy from (figure 4.13). Currently, leading rating com- having a monopoly on information about panies rate only several hundred corporate their clients. Local banks often are better and foreign bonds. A private initiative is un- placed than foreign lenders to monitor the fi- derway to rate millions of companies on the nancial position of domestic firms (Eichen- Internet by merging credit management tech- green and Mody 1999). This advantage has niques with collecting and organizing infor- enabled domestic banks to play an important mation on companies (UNCTAD 2000; www. role in onlending international capital flows cface.com). Furthermore, with a huge pool of to domestic borrowers; domestic banks ac- rated companies, it would become possible to counted for as much as 46 percent of net long- issue guarantees based on estimated default term capital flows to developing countries in probabilities and to securitize these guarantees 1997 (World Bank 1999). Domestic banks may through the capital markets. experience sharper competition for making loans in their own markets, as the Internet International coordination makes information about borrowers more ac- Electronic commerce raises several regulatory cessible to international lenders. issues that should be addressed through im- The easier dissemination of credit informa- proved international coordination. The im- tion made possible by the Internet is likely to portance of international cooperation to tax strengthen the trend toward greater reliance administration was noted earlier. Ensuring on bond financing in developing countries, open access for the international transmission potentially at the expense of domestic banks. of goods delivered electronically would facili- The number of firms in developing countries tate the continued growth of electronic com- with access to international bond markets in- merce. Members of the World Trade Organi- creased more than sixfold from 1991 to 1998 zation have decided provisionally to exempt (Eichengreen and Mody 1999), and the share such goods from customs duties.23 A more im- 129 G LO B AL EC ONO MI C PRO SP EC TS portant contribution to ensuring open access 6. Companies such as Teledesic and Skybridge would be to secure agreement on barrier-free plan to launch satellite systems capable of supporting treatment of electronically delivered services, high-speed Internet access for millions of users. So far by streatm het ing commitmen ss entervies, in the failure of the Iridium system has not derailed these by strengthening commitments entered Into plans. under the General Agreement on Trade in Ser- 7. The regression equation is log(cell 99) - log(cell vices (Mattoo and Schuknecht 2000). 90) = -9.16 - .841og(cell 90) + .781og(urban pop 90) + More generally, the rise in the volume of 2.12[log(income 99) - log(income 90)] + .78log(in- cross-border transactions promised through come 90) + 1.061og(policy). R-squared is .78. Obser- the spread of the Internet will raise a host Of vations = 99. the spread of the Internet will raise a host of 8. These studies are by their nature speculative, issues requiring international agreement. Ex- however, and the results require several restrictive as- amples include improved procedures for allo- sumptions that may or may not reflect actual events cating domain names used for Internet loca- (OECD 2000a). tions;24 agreement on privacy standards for 9. At Cisco the replacement of phone, fax, and e- data generated through commercial transac- mail ordering by electronic commerce systems cut the tions; closer coordination of antitrust reviews number of orders that had to be reworked from 25 per- to reduce the administrative burden imposed cent to 2 percent (OECD 1999). to . 10. The use of discounts to increase market share, on companies; and efforts to harmonize the coupled with the lack of profits, leaves some doubt as laws governing electronic commerce transac- to whether these lower prices reflect efficiency gains, tions. Electronic commerce will further speed however. the process of globalization and thus will un- 11. See World Bank 1995 for an earlier discussion derline the importance of an effective interna- of the potential for long-distance service exports from tional framework for cross-border transactions. developing countries. tional framework for cross-border transactions. 12. The potential for exports of Internet-based services by developing countries in the medical sector is staggering. About one-third of the cost of health care Notes in the United States, or some $300 billion a year, rep- 1. See OECD 1999 and UNCTAD 2000 for a resents the cost of capturing, storing, and processing more detailed discussion of the definition of electronic information such as records, physicians' notes, test re- commerce. sults, and insurance claims (Evans and Wurster 1997). 2. These estimates exclude sales where the Inter- 13. Examples include www.indiaonestop.com; net is used as an information resource only. In some www.in-business.com.ar/mall/;www.maquilamarket. service sectors, the penetration of electronic commerce com; and mm.malaysiadirectory.com/b2b/, which focus is relatively high. In the United States, for example, mainly on markets in India, Argentina, Mexico, and electronic commerce accounts for some 25 percent of Malaysia, respectively. The World Bank is providing stock trades. support to the Virtual Souk, an online marketplace for 3. These figures vary greatly depending on the de- artisans in the Middle East and North Africa, owned by finition of electronic commerce used. Projections of local nongovernmental organizations and cooperatives. trillions of dollars of electronic commerce in part re- 14. Note that the Internet also enables service flect the migration of EDI systems to the Internet. transactions that would not otherwise have occurred, Moreover, these forecasts are extremely speculative. As which increases the demand for literate but not highly indicated earlier, estimates of even the current size of skilled workers in developing countries. electronic commerce differ substantially. 15. Inequality is affected by many factors, and the 4. The regression is log(I/T 1997) - log(1/T 1990) strength of this effect is unknown. Thus, whether the = -14.76 - 1.041og(I/T 1999) + .861og(urban pop Internet will ultimately have a significant impact on in- 1999) - .lOlog(income 1990) + .61log(policy). All co- equality remains uncertain. efficients are significant except income. Regional 16. For example, see Autor, Katz, and Krueger dummy variables are also included. R-squared is .97. 1998; Bresnahan and Brynjolfsson 1998; Krueger 1993. ITr stands for Internet intensity. 17. In Egypt, for example, El-Nawawy and Ismail 5. Between-country inequality of access to com- (1999) report that the cost of an international half cir- ponents of this index (such as mobile phones) appears cuit can be 2.5 times the international tariff. to have declined during the 1990s. Thus trends in the 18. Helft 2000. digital divide may differ significantly among different 19. Harris (1998) goes further to say that "the kinds of communications media. Internet can eliminate the scale disadvantage of small 130 EL EC T RON IC COMM ERC E AND D EVELO PING C O U N T R I E S regions in producing services ... [and] then can po- vate Sector. Note No. 122. World Bank, Wash- tentially lead to in-migration of skilled labor to the ington, D.C. July. region." Bresnahan and Brynjolfsson. 1998. "Information Tech- 20. DePalma, McCarthy, and Armstrong (1998) nology, Workplace Organisation and the Demand found that although 49 of 50 companies surveyed had for Skilled Labour: Firm-level Evidence." Stan- operations outside the United States, and 80 percent ford University, Palo Alto, Calif. Processed. print marketing material in other languages, only five Business 2.0. 1999. "Wise Load." (www.business2. said that their multilingual sites were as functionally com[MaylD. rich as their English sites. Callaghan, Nancy, and Leo Schnoll. 1997. "ISO 9000 21. Advertisements have appeared in U.S. busi- for Small Companies." Quality Digest Online ness magazines highlighting the need for Internet (www.qualitydigest.com[August]). content to be written in other languages in addition Canning, David. 1999. "Infrastructure's Contribution to English. Schmitt (2000) warns businesses that to Aggregate Output." Policy Research Working English-only sites are no longer feasible for interna- Paper 2246. World Bank, Washington, D.C. tional companies. Connolly, Michelle. 1998. "The Dual Nature of Trade: 22. According to Bach and Erber (1999), it is vir- Measuring Its Impact on Imitation and Growth." tually impossible to enforce taxes on electronic trans- Duke University Working Paper 97-34 (http:/H actions. This conclusion is uncertain, however, and its www.econ.duke.edu/Papers/Abstracts97/abstract. accuracy will depend on whether the evolution of In- 97.34.html). ternet technology ultimately favors privacy over gov- Cukier, Kenneth. 1999. "Bandwidth Colonialism? The ernment monitoring. Implications of Internet Infrastructure on Interna- 23. The impact of this decision on government tional E-Comrnmerce." (Paper delivered at the an- revenues should be slight because the tariff lost if nual INET conference sponsored by the Internet no taxes are levied on digitizable goods is less than Society, San Jose, Calif., (http://www.isoc.org/ one-fifth of 1 percent of total revenues in the major inet99/proceedings/le/le_2.htm). developing-country importers (Matoo and Schuknecht Davis, Bob, and Gerald Seib. 2000. "Technology Will 2000). Test a Washington Culture Born in Industrial 24. The World Intellectual Property Organization Age." Wall Street Journal. May 1. issued a detailed report on the intellectual property DePalma, D., J. McCarthy, and A. Armstrong. 1998. issues associated with domain names and has devel- "Strategies for Global Sites." The Forrester Re- oped an online system to assist in resolving disputes port 3 (3) (May). (UNCTAD 2000). Economist. 1999. "A Survey of Business and the Inter- net." June 16-July 2, 351 (8125) (B1-39). Eichengreen, Barry, and Ashoka Mody. 1999. "Lend- References ing Booms, Reserves, and the Sustainability of AED (Africa Economic Digest). 1998a. "African Tele- Short-Term Debt: Inferences from the Pricing coms: A Renaissance." 19 (10). of Syndicated Banking Loans." Working Paper . 1998b. "Roadbuilding on the Information Su- 7113. National Bureau of Economic Research, perhighway." 19 (10). Cambridge, Mass. Autor, Katz, and Krueger. 1998. "Computing Inequal- El-Nawawy, Mohamed A., and Magda M. Ismail. ity: Have Computers Changed the Labour Mar- 1999. "Overcoming Deterrents and Impediments ket?" Quarterly Journal of Economics 113(4): to Electronic Commerce in Light of Globalisa- 1169-213 (November). tion: The Case of Egypt." (Paper presented at the Bach, Stefan, and Georg Erber. 1999. "Opportunities annual INET conference sponsored by the Inter- and Risks of Global Electronic Business Transac- net Society, San Jose, Calif. (http://www.isoc.org/ tions." Economic Bulletin. German Institute for inet99/proceedings/lg/lg_3.htm). Economic Research 36 (4) (April). E-Marketer. 2000. "Strong Net Growth South of the Bermudez, John, Bob Kraus, David O'Brien, Bob Border." March 22. (http:www.emarketer.com). Parker, and Larry Lapide. 2000. B2B Commerce EU Commission. 2000. Europe: An Information Soci- Forecast: $S.7T by 2004. AMR Research (http:// ety for All. Progress report for Special European www.amrresearch.com). Council on Economic Reforms and Social Cohe- Bickers, Charles. 1999. "Asia's Race to Go Digital." sion. Brussels. March 23-4. Far East Asian Economic Review July 1 (2). Evans, Philip B., and Thomas S. Wurster. 1997. "Strat- Braga, Carlos A., and Carten Fink. 1997. "The Private egy and the New Economics of Information." Sector and the Internet." Public Policy for the Pri- Harvard Business Review 75(5):70-82. 131 G LO B AL EC ONO MI C PRO SP EC TS Fan, Grace. 2000. "Pedicarts Link Shanghai's Streets to Krueger, Anne 0. 1993. "How Computers Have the Internet." New York Times. March 29. Changed the Wage Structure: Evidence from Mi- Forrester Research. 1999a. "Online Energy Industry to crodata, 1984-1989." Quarterly Journal of Eco- Reach $266 Billion by 2004." October. Cam- nomics 108 (1): 33-60 (February). bridge, Mass. Lapper, R. 2000. "Tropical Nets." Financial Times. _. 1999b. "Managing e-Marketplace Risks." De- February 3. cember. Boston, Mass. Lee, Ho Guen, J. Christopher Westland, and Sewon Goldman Sachs. 1997. "Cyber Commerce: Internet Hong. 1999-2000. "The Impact of Electronic Tsunami." New York. August 4. Marketplaces on Product Prices: An Empirical - . 1999. "B2B: 2B or Not 2B? Version 1.1" Study of AUCUNET." International Journal of - 2000. E-Finance in Asia, Part 2: Citibank-an Electronic Commerce. Winter 4(2): 45. Asian Internet Mandarin in the Making? Mann, Catherine L., Sue E. Eckert, and Sarah Cleeland Goodman, S. E., L. 1. Press, S. R. Ruth, and A. M. Knight. 2000. Global Electronic Commerce: A Rutkowski. 1994. "The Global Diffusion of the Policy Primer. Washington, D.C.: Institute for In- Internet: Patterns and Problems." ternational Economics. Granitsas, Alkman. 2000. "Tangled in the Web: The Matoo, Aaditya, and Ludger Schuknecht. 2000. Trade Rush into Internet Banking May Not Be Justified Policies for Electronic Commerce. World Bank, by the Number of On-Line Customers in Asia." Washington, D.C. Processed. Far Eastern Economic Review May 4. Mills, Mike. 1996. "In the Modem World, White Col- Grebb, Michael. 2000. "Korea's Digital Quest." Busi- lar jobs Go Overseas." Washington Post. Septem- ness Week. September 25: 68-72. ber 17, p. Al. Grossman, Gene M., and Elhanan Helpman. 1991. In- Morgan Stanley Dean Witter. 1999. "The Internet and novation and Growth in the Global Economy. Financial Services." New York. August. Equity Cambridge, Mass. MIT Press. Research, North America. Harris, R. G. 1998. "The Internet as a GPT: Factor OECD (Organisation for Economic Co-operation and Market Implications.," In General Purpose Tech- Development). 1998a. "Electronic Commerce: nologies and Economic Growth, edited by El- Prices and Consumer Issues for Three Products: hanan Helpman. Cambridge, Mass.: MIT Press. Books, Compact Discs, and Software." Paris. Harrow, Jeffrey. 2000. "The Rapidly Changing Face . 1998b. The Economic and Social Impact of of Computing" (http://www.compaq.com/rcfoc/ Electronic Commerce: Preliminary Findings and [April 24]). Research Agenda. Paris. October. Helft, D. 2000. "Latin America Looking to Wireless." - . 1999. The Economic and Social Irmpact of The Standard. 9 May 2000. Electronic Commerce: Preliminary Findings and ITAA (Information Technology Association of America). Research Agenda. Paris. 2000. Bridging the Gap: Information Technology . 2000a. "E-Commerce: Impacts and Policy Skills for a New Millennium. Arlington, Va. Challenges." Economics Department Working ITU (International Telecommunications Union). 1998. Paper 252. Paris. ITU Telecommunication Indicators Handbook. _ 2000b. Information Technology Outlook. Geneva. Paris. - 1999. Challenges to the Network-Internet _ 2000c. "Local Access Pricing and E-Com- for Development. Geneva. February. merce." Directorate for Science, Technology and Jensen, Mike. 1999. African Internet Status. Industry, Paris. Jordan, Miriam, and Jon E. Hilsenrath. 2000. "Amer- Oliner, Stephen D., and Daniel E. Sichel. 2000. "The ica Talks, India Types Up the Transcript." The Resurgence of Growth in the Late 1990s: Is In- W'all Street Journal. March 16. formation Technology the Story?" Finance and Kibati, Mugo. 1999. "What Is the Optimal Technologi- Economics Discussion Series 2000-20. Federal cal and Investment Path to 'Universal' Wireless Reserve Board, Washington, D.C. Local Loop Deployment in Developing Coun- Oxford Analytica. 1999. Latin America: Electronic tries." (Paper presented at the annual INET con- Commerce. September 16. http://wvw.oxan.com. ference sponsored by the Internet Society, San Jose, Panagariya, Arvind. 1999. "E-Commerce, WTO, and Calif. (http:H/www.isoc.org/inet99/proceedings/ Developing Countries." UNCTAD. Geneva. July 1c/Ic_2.htm). (http://www.unctad.org/en/docs/ecwto/pdf). Krantz, Michael. 1998. "The Internet Economy." Time Peha, Jon M. 1999. "Alternative Paths to Internet In- 152 (July 20): 34-41. frastructure: The Case of Haiti." Paper prepared 132 E L E C T R O N IC COMM ERC E AND D EVE LO PING C OU N T R I E S for the annual INET conference sponsored by the . 2000. Building Confidence: Electronic Com- Internet Society, San Jose, Calif. (http://wvw.isoc. merce and Development. UNCTAD/SDTE/ org/inet99/proceedings/3f/3f.2htm) MISC.1 1. Geneva. Phillips, Charles, and Mary Meeker. 2000. The B2B U.S. Department of Commerce. 1999. The Emerging Internet Report. Morgan Stanley Dean Witter Eq- Digital Economy 11. (http://www.ecommerce.gov/ uity Research. New York. April. ede.report.html). Price Waterhouse Coopers. 1999. SME Electronic Vehovar, Vasja, Zenel Batagelj, and Katja Lozar. 1999. Commerce Study. Final report, Asia Pacific Eco- "Language as a Barrier." (Paper presented at the nomic Cooperation Telecommunications Work- annual INET conference sponsored by the Inter- ing Group. net Society, San Jose, Calif.). Rao, MaDanmohan. 1999. "Bringing the Net to the Vogel, Thomas T., and Pamela Druckerman. 2000. Masses: Cybercafes in Latin America." On the "Latin Internet Craze Sets Off Alarm Bells." Wall Internet. 5 (1). (http://www.indialine.com/net. Street Journal. February 16. edirorial/editorial/70.html). . 1996. Washington Post. Reuters. 2000. "It's a Wireless World in Japan." June Washington Post. 2000. "The Wired Economy: A Spe- 20. cial Section." April 5 issue. Washington, DC. Rodriguez, E C. 2000. "Does Information Technology Wenninger, John. 1999. "Business-to-Business Elec- Raise Inequality?" University of Maryland, De- tronic Commerce. Current Issues in Economics partment of Economics. College Park, Md. and Finance 5 (10) (June). Rosen, Daniel H. 1999. "Hype versus Hope for E- Wilson, E., and Rodriguez, F 2000. Are Poor Coun- Commerce in China." The China Business Re- tries Losing the Internet Revolution? Report pre- view (July/August) 26(4): 3842. pared for World Bank. Schmitt, E. 2000. "The Multilingual Site Blueprint." Wood, L., 1999. "Lloyd's Satellite Constellations." J ' * - * ~~~~~~~~~~~~(w ww. ee. surrey. ac .u k /Personaif/L .Wood! lThe Forrester Report June 2000 (http://www. (w .esuryaukPsol/Wod The forrester ReportcJuneo2000m(bttp://www. constellations/overview.html). Report prepared forreste.ncom). for the World Bank (http://www.infodev.org/ Schuknecht, L., and Perez-Esteve, R. 1999. "A Quanti- library/working.htm.) tative Assessment of Electronic Commerce." World Bank. 1995. Glohal Economtc Prospects and the WTO Working Paper ERAD 99-01. World Trade Bank. 1995. Waspectond the Organization, Economic Research and Analysis Developing Countries 1995. Washington. D.C. -___ 1999. Global Development Finance 1999: Division. Geneva. September. Analysis and Summary Tables, Washington, D.C. Schwartz, Nelson D. 2000. "Playing the Internet's . 2000a. "The Networking Revolution. Oppor- Next Gold Rush." Fortune 141 (10) 178-82. tunities and Challenges for Developing Coun- (May 15). tries." InfoDev Working Paper. June. Global In- Suttle, Philip. 2000. "The Digital Economy and the formation and Communications Technology Global Economy." World Financial Markets. Department. The World Bank, Washington, D.C. Morgan Guaranty Trust Company Economic Re- . 2000b. World Development Indicators 2000. search. January 14. Washington, D.C. Teo, Pebble. 1999. "Business to Business Electronic WTO (World Trade Organization). 1998. Electronic Commerce: the Asian Experience." Presentation Commerce and the Role of the WTO. Special to OECD Workshop. Oslo, June 17. Study 2. Geneva. UNCTAD (United Nations Conference on Trade Xie, Andrew. 2000. Global Economic Forum. Morgan and Development). 1999. "Report of the Pre- Stanley Dean Witter, New York. UNCTAD-X Workshop on Exchange of Experi- Zambrano, Raul. 1999. "Internet Users in Latin Amer- ences among Enterprises in the Area of Electronic ica." Presentation at UNCTAD Workshop on Commerce. TD(X)/pc/3." Geneva. September 14. E-Commerce, Lima, August 4-5. 133 Annex 4: Firm interviews and website survey Firm interviews * participants in a conference on the May The interview information included in the 24, 2000, "Wiring the Border" program, chapter is based on conversations with: including representatives from small- business associations, technology special- s managers or executives from five multi- ists, government officials, and academic national firms: Ingram Micro, The Gap, experts. Ford, General Electric, and Infosys. These specialize in computer hardware and soft- ware distribution; apparel; automobiles; Website survey electrical and lighting equipment; and The conclusions on the alibaba website are customs software, respectively. The first based on a survey of firms via fax, conducted four appeared on this year's Fortune 200 in May 2000. To keep the survey manageable, list, while the last is a major Indian-based the firms to be contacted were chosen from se- multinational software company. lected economic subsectors. Within these sub- * representatives from: (i) the U.S.-Mexico sectors, we included all firms posting offers in and U.S.-Philippines Chambers of Com- the last two weeks of April 2000-whether as merce; as well as from (ii) the offices of sellers or buyers. We received 105 complete the Commercial or Trade Attaches repre- replies to our questionnaire from the 800 firms senting Taiwan (China) and Brazil in the surveyed. A list of all respondents' names and United States. websites (if any) is attached. 134 ELECTRON IC CO MM E RC E AND D EVE LO PING CO UN TRI ES Companies Participating in Survey of Alibaba B2B Website Users China Yangzhou Weiteli Motor- Green Source International Group Belgraver Asia Pte. Ltd. Manufacturing Co., Ltd. Wellmade Industry Corp., Ltd. W. & J. Co., Ltd. Tianjin Printronics Circuit Corp. Zhejiang Light Industrial Products ChangZhou Rui Da Trading Company Horman Company Import and Export Corp. Well Hung (Australia) Pte. Ltd. China National Electric Wire and Cable iSquare Design and Development Renaissance International I/E Corp. (Xiamen Branch) Shandong Xingfa Foodstuffs Wuhan Zhongbai Group Co., Ltd. Hebei Xin Hua Li Da Sale Department Mideast Mercantile Ltd. CNACC International Co., Ltd. K.O.G. International Philippines., Inc. Xiamen Zhongxin Metal Products Co., Fujian Coal Import and Export Corp. Jiangxi Wire and Cable General Factory Ltd. Tung Kong Handbag Mfy Young Eak Trading Co., Ltd. Cintel International Ambp Enterprises Co., Ltd. Shenzhen Tonghaisheng Investment Sandstone International Co., Ltd. Software International Development Co., Ltd. Shanghai Yang Ning International Xiamen Gemachieve Enterprise Jiangsu lI/E (Group) Corp. (Heiteng Co., Trading Co., Ltd. Shandong Metals and Minerals Import Ltd.) Pacific Silk Route Pte. and Export Corp. Sinochem Hebei (Shenzhen Toomly) China Shaanxi Techrun Technology D.P. International Import and Export Co., Ltd. Company Jitco Group Ltd. Zhejiang Yongkang Crown Power Tools Praphan Ceramica Co., Ltd. Beijing Orient Sotoma Garment Co., Ltd. Manufacturing Co., Ltd. China Tea Import and Export Corp. Ishida Co., Ltd. Truly Sales Co. Hebei Sanli Cashmere Products Co., Ltd. Ningbo Free Trade Zone Sino-Dubai Hongguang Electronics Import and Yixing Tanghan Ceramic Co., Ltd. International Trading Co., Ltd. Export Co. (Guangzhou Office) Xiamen Zhenhua Ind. Corp. Shriya Impex Chengdu Guoxin Maida Electronic Co., Ningbo Economic and Technical Regan (H. K.) Ltd. Ltd. Development Zone Import and Export Citic Shanghai Import and Export Co., Catic Electronics Corp. Ltd. H20 Electronics Co. Starscom Info-Tech Co., Ltd. Chengde Bearing Co., Ltd. Dong Young International Corp. Seanet RS Manray Concept Sinoleather.com Ltd. Cixi Kaida Bearing Co., Ltd. Atul Exports Chew The World Kedi Hi-Tech Industrial Co., Ltd., W. H. Enterprises Gusung Machinery Xiamen Office Guangdong Yangchun Bearing Co. Ltd. Eros Group Shijiazhuang Gulf Semiconductor Co., S.L.S. Partnership Xiangshui Bearing Accessory Co., Ltd. Ltd. Adore International Shandong Gifts and Decorations Import Tao's Inc. Dorly International Enterprises Inc. and Export Corp. Feidong Foreign Economics and Trade Shandong Commercial Group Dalian ET.Z. Zhengxing International Corp. Corporation Trading Co., Ltd. China Dalian Aidi International Trade Goldsense Technology Ltd. Yuyao Kingfan Industry and Trade Co., Company CV RJR International Ltd. Aurora Translation Services Cixi Fuda Bearing Co., Ltd. Hanbit Ebenezet Shen Zhen Xinhaowei Industrial Co., Ltd. G.K. Trading Corp. Giga Technology Co., Ltd. On Time Taiwan Ltd. Suzhou Arts and Crafts I/E Corp. 135 Appendix 1 Regional Economic Prospects East Asia and Pacific nance 2000 were generally conservative, and we have upgraded the 2000 forecast for most Recent developments countries. Growth for developing East Asia, in E 7 AST ASIA HAS CONTINUED TO CONSOLIDATE particular, is likely to be nearer to 7.2 percent its recovery from the deep crisis-induced than to earlier projections for 6.6 percent _E_recession of 1997-98, albeit with sub- growth. stantial variation across countries in the re- A common element across the region over gion. Developing countries in the region regis- the last 12 to 18 months has been low and sta- tered 6.9 percent growth in 1999, up from a ble inflation and interest rates, and these have decline of 1.4 percent in 1998. The Republic been strong positive factors in the recovery of Korea experienced the sharpest "V-shaped" process. For example, inflation in the East recovery, with GDP growth of 10.7 percent in Asia-5 has stabilized at a rate of 1.5 to 2 per- the year. Indonesia, at the other extreme, cent, after a rapid but brief acceleration caused barely reached positive territory, following its by the devaluation of 1997-98 (figure A1.1). difficult economic performance of 1998. On Surging oil prices have translated into mild in- average, output in the five most affected crisis flationary pressure in Korea (3.9 percent year countries (Indonesia, the Republic of Korea, on year in September) but have had consider- Malaysia, the Philippines, and Thailand- ably less impact to date in Malaysia or Thai- known as the East Asia-5) recovered smartly at a rate of 6.7 percent from their 1998 crisis lanued to be largely accommodative, though decline of 8.2 percent. China suffered a minor tinueat be largely acc atitoug dip in output growth in 1999-though still central bank officials are carefully monitoring growing at a rclip of 719pret-as reovr the situation. The low-inflation, low-interest growing at a clip Of 7.1 percent-as recovery rt niomn a enpriual eei in exports did not occur until the second half rate environment has been particularly benefi- of the year. The newly industrializing econ- cial to the process of unwinding the domestic omies (NIEs-Hong Kong [China], Taiwan debt problems faced by firms and consumers [China], and Singapore), which suffered sub- in the crisis countries. Similarly, governments stantial spillover effects of the crisis, saw a have been able to limit the growth of public rebound to growth of 4.8 percent in 1999 from debt (as a share of GDP) below the worst lev- 1.1 percent in 1998. And momentum con- els initially feared. Indonesia stands out from tinues to be fairly strong in the region. Data the other East Asian economies. It has been covering the first three quarters of 2000 sug- buffeted by continuing instability-the rupiah gest that the near-term projections published dropping 20 percent through October since the nine months ago in Global Development Fi- beginning of the year-and inflation started a 137 G LO B AL EC ONO MI C PROS PE CT S Figure A1.1 Inflation in the East Asia-5 countries, 1991-2000 Percent change, year over year 25 20 15 10 O f II Il I l l lI I 01 1991 01 1992 01 1993 Q1 1994 Q1 1995 Ql 1996 Q1 1997 Q1 1998 Q1 1999 Q1 2000 Note: GDP-weighted average. The East Asia-5 comprises Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand. Source: IMF, International Financial Statistics, and World Bank staff estimates. rapid rise in the third quarter (6.6 percent year combined with softening internal demand as on year in September). a result of reforms in the state enterprises In all the crisis countries, real effective ex- and the financial system, led to a deflationary change rates have stabilized at rates well above cycle. Deflation effectively yielded a real depre- crisis troughs, but some 15 to 30 percent be- ciation of the currency, and export growth has low precrisis levels. Thus real devaluation has boomed since the second half of 1999, with persisted and facilitated a double-digit boom in merchandise exports 35.4 percent above year- exports. Robust export growth and firming ago levels in 2000 (year-to-date through Au- export prices have abetted the maintenance gust in dollar terms). of a positive current account balance, though The main weakness in many countries has the recovery of imports and higher oil prices appeared in the equity markets. On average, has narrowed the balance in many countries. stock market indexes in the five East Asian cri- Nonetheless, rising reserves and the improved sis countries declined by over 30 percent (in term structure of foreign debt have led to a local currency terms) since the beginning of the substantial improvement in the external posi- year (through early November). Globally, there tion of the region compared to the precrisis has been a flight to "quality" instruments, and position. China's competitiveness-through this has depressed financial markets in almost maintaining stability in its own foreign ex- all emerging markets. Gross financial capital change market-suffered in the wake of the inflows into East Asia appeared to have picked crisis, but ultimately improved as a conse- up in the first half of 2000 compared to 1999. quence of price deflation and VAT (value However, the flows have been dominated by added tax) export rebates. The loss of compet- some large issues, particularly by China. For itiveness, combined with the sharp import re- example, China received $10.4 billion of the versals in the crisis countries, including Japan, regional total of $11.6 billion in equity in- led to a sharp falloff in export growth-1.7 flow, and less than $600 million flowed into percent in 1998 (in dollar terms), followed by the equity markets of the East Asia-5 countries. 5.8 percent in 1999. Weak external demand, Net flows remain negative, and in particular, 138 RE G I ONA L E C ON OM IC P R O S P E C T S commercial banks continue to unwind their practices have not changed substantially from local positions. earlier, less than transparent modes. Some of the smaller island nations have also suffered Near-term outlook from political turmoil (for example, Fiji and In 2001-02, output for the group is likely to the Solomon Islands), whereas newly formed begin a general process of moderating and con- East Timor is in a slow and lengthy process of verging toward longer-term growth paths, with nation building. growth easing to 6.4 percent in 2001 and 6 percent by 2002 (table A1.1). Export growth, Long-tenn prospects sizzling in 2000, should ease considerably in Long-term prospects are little changed from 2001 and 2002 in line with slower external earlier projections. Average per capita income growth. External risks for East Asia are similar grows in our long-term baseline (2003-10) to what we have assessed over the last 12 by 5.4 percent per year-somewhat below months: a hard landing in the United States, a 1990-2000 per capita growth of 5.9 percent. renewal of financial difficulties in Japan, and a The factors underlying slower growth vary weakening of the electronics cycle. But these from country to country. The upper-middle- risks have generally been pushed back in time. income countries and NIEs are converging And domestic risks in aggregate have dimin- with (or in some cases have exceeded) OECD ished from past high levels. Nonetheless, the income levels. They are maturing economies, process of working-out from under the post- with already highly educated work forces; and crisis financial difficulties is far from finished. it is likely that GDP growth will ease gradually Higher interest rates or slower growth could toward the OECD average over the next sev- further worsen financial conditions for many eral years. The lower-income countries, partic- firms and consumers still saddled with high ularly China, are unlikely to sustain the high debt. The two most vulnerable large countries growth rates of the past decade. Many of the are Indonesia and the Philippines. These coun- low-income countries-as well as the crisis- tries also suffer from political weaknesses, civil affected middle-income countries-will have disturbances, and a perception (from the point to devote resources in order to overcome the of view of investors), that business operating legacy of past institutional failures: addressing Table A1.1 East Asia and Pacific forecast summary (percent per year) Baseline forecast Growth rates/ratios * -$# . 2000 2001 2002 2000-10 Real GDP growth ;, IA> ; 7.2 6.4 6.0 6.3 Consumption per capita - .Z- 4. . 5.9 6.1 6.1 5.8 GDP per capita 14 43 5*2. 6.1 5.4 5.0 5.4 Population - 1Z - - ii El 1.0 0.9 0.9 I 0.8 Median inflation, 2 7 9.2 3.8 5.4 5.8 4.8 Gross domestic investment/GDP i- r*6. 30.3 31.1 31.9 1 32.6 Median central gvt. budget/GDP .I4.. .- 43 - -2.8 -2.1 -1.7 -0.8 Export volumeb .-. e2 - 19.4 9.3 8.6 8.5 Current account/GDP 4.1 33 2.5 2.3 2.1 Memorandum item GDP East Asia-5 countriesc -.. 6.7 6.9 S.5 5.1 5.5 a. GDP deflator. b. Goods and nonfactor services. c. Indonesia, Republic of Korea, Malaysia, Thailand, and the Philippines. Source: World Bank baseline forecast, October 2000. 139 G LO B AL EC ONO MI C PRO SP EC TS Table A1.1a Forecast assumptions- The countries of East Asia-with their ever East Asia and Pacific increasing involvement in the so-called new Initial conditions economy-are well placed to meet the chal- lenge, but they are lagging far behind the more 1. Ratio of real income per capita: advanced countries. In 1999, the East Asia-5 industrial / East Asia and Pacific .2 27 2. Trade (X1M) E GDP ratio (real) countries had only half the number of Internet 3. Median inflation rate (percent) hosts (per 10,000 persons) that Brazil or Mex- 4. Median fiscal balance / GDP ico had, and only 5 percent compared to the 6. Investment / GDP (nominal) NIEs. And though markets for the Internet 7. Gross national savings! GDP and mobile phones have been growing at some 7a. Gross domestic savings / GDP 40 percent per year in East Asia, they have 9. Furrent aount balance / GDP ~been growing at over 50 percent in Brazil and 10. External debt / exports* 0 9 Mexico, in part as a result of deeper reforms 1]. School enrollment rates and greater competition in the telecommuni- Primary (pct of eligible population) cations sectors of the latter countries. There is Secondary also the possibility of reform fatigue or even 12. Illiteracv rate (pct of people reversal. Malaysia's recent decision to renege 15+ years) 2< 1. 13. Under-5 mortality rate on removing import tariffs on automobiles (per 1,000 live births) could signal a weakening of a commitment to 14. Life expectancy at birth (years) 4 regional free trade. Exogenous assumptionsn 1. Population growrh . South Ai 2. Market's GDP growth M 3ot Asia 3. Oil price S/bbl (avg.) 5, 4. Market's import growth Recent developments 'Exports of goods and services plus workers' remittances. DP growth in South Asia averaged 5.1 Note: Market growth is trade-weighted partner GDP / import Gpercent in 1997-98, as the larger econ- growth. omies-relatively to international trade- Source: World Bank database, World Bank staff estimates. closed were successful in mitigating losses of agricul- tural income tied to commodity price declines nonperforming loans in the financial sector, in the wake of the East Asian crisis. Output disposing of distressed assets, and reducing growth accelerated to 5.7 percent in 1999 the state's active role in the economy while en- and is estimated to reach 6 percent in 2000. hancing its regulatory role and competition. Better-than-expected agricultural sector per- Initial conditions for sustained high growth formance in Bangladesh, India, and Pakistan in East Asia at the beginning of the millen- has accounted for a fairly large proportion of nium appear better than at the beginning of the recent improvement in growth outturns. the 1990s, the end-of-decade financial crisis In addition, the rate of growth in industrial notwithstanding (table A1.la). Openness in- production in Bangladesh and India climbed creased by more than 20 percentage points to more than 10 percent during the first half over the 1990s and was, if anything, enhanced of 2000 (figure A1.2). Output in Bangladesh during the crisis, presenting both an opportu- was boosted by the recovery from the mas- nity as well as a challenge. The opportunity sive flooding of 1998. The burgeoning Indian comes from the ability to import new tech- service sector also has maintained strong nologies, knowledge, and business practices. advances, at rates of more than 8 percent The challenge comes from increased competi- through 1999 and into 2000. Exports of goods tion and the need to develop institutions that and services continue to grow at rapid rates- enhance flexibility and speed of adaptation. by more than 10 percent in India, Pakistan, 140 REG IO NA L EC ONO MI C PRO SP EC TS Figure A1.2 Industrial production in South Asia three-month moving average, year-on-year, percentage change 20 15 10 5 -op\s Pakistan Bangladesh -10 Jan. 1998 May 1998 Sept. 1998 Jan. 1999 May 1999 Sept. 1999 Jan. 2000 May 2000 Source: IŽAF, Interrratr,nal Fi-anc,ial Statistics. and Sri Lanka. At the same time, manufactur- $210 billion. Recently, however, in tandem with ing production has fallen sharply in Pakistan, global financial volatility, there was a reversal given financial constraints and other difficul- of portfolio flows, which affected the stock ties. And the surge in the oil price and contin- market and exerted some pressure on the rupee. ued weakness in non-oil commodity prices (for Nonetheless, steps toward improving super- example, the prices of Sri Lanka's main export vision and restructuring of the banking systems commodities-tea and natural rubber-are in India, Pakistan, and Sri Lanka have yielded now some 20 and 30 percent below recent some positive results and have improved confi- highs) is exacting a moderate toll from the re- dence in the region to a degree. gion's growth momentum. Recent steps to make South Asian economies Near-term outlook more open to capital flows and strengthen the Average growth for 2001-02 is anticipated to financial system have also supported growth. be 5.5 percent for the region (table A1.2). India eased some restrictions on FDI to encour- Underlying this aggregate figure, however, are age foreign flows into the energy sector, where a number of driving and restraining forces it is most needed. FDI registered $2.2 billion in shaping the near-term view. Among positive 1999 and is expected to achieve similar levels in factors are improved prospects for capital in- 2000. But foreign investment is broadening in flows, as the Indian government in particular scope across the economy, supplementing do- undertakes efforts to boost foreign investment mestic investment in such sectors as the soft- and relax direct exchange controls. And to fa- ware industry, which has achieved remarkable cilitate the growth of services exports, legisla- growth of almost 50 percent over the last year. tion has been introduced to support the IT sec- Portfolio flows to India also increased, to a high tor and develop "e-business." External factors of $3 billion in 1999-2000, attracted by (and such as continued strong advances in world contributing to) the boom in India's stock mar- trade and prospects for an eventual moderate ket. Equity prices increased by more than 50 firming of non-oil commodity prices should percent from the first quarter of 1999 to the support growth across countries of the region, first quarter of 2000, and capitalization rose to especially in Bangladesh and Sri Lanka. 141 G LO B AL EC ONO MI C PRO SP EC TS Table A1.2 South Asia forecast summary (percent per year) Aoxrce: World Bank baseline forecast Growth rates/ratios 2000 2001 2002 2000-10 Real GDP growth 60 5.5 5.5e 5a Consumption per capita 3.7 3.3 3.5 3.6 GDP per capita 40t 3.7 3.8 3.9 Population A 7 1.8 1.7 1.7 ' 1.5 Median inflation, 5 nd cnr g e n fis Gross domestic investment/GDP 237 24.1 24.3 25.0 Median central govt. budget/GDP .4-4.7 -4.5 -.4.4 I -3.7 Export volumeb sm $4 bli 11a5 4P4 8b 7.9 Current account/GOP -w32.6 -2.9 -2.2 -3.0 a. GDP deflator. b. Goods and nonfaceor services. Source: World Bank baseline forecast, October 2000 n Recent developments in oil markets will re- achieved some progress in a number of areas strain growth in the near term, however. South supportive of longer-term growth (table A1.2a). Asia is one of the more energy import-intensive Although the region remains in large part developing regions, with crude oil and other closed to foreign trade (in part because of the energy commodities constituting 20 percent of large scale of the Indian domestic economy), total imports in India and 15 percent in Pak- median inflation and central government fis- istan (representing 2 percent of GDP in both cal deficits have declined modestly; external countries). The 50 percent rise in the oil price debt ratios have been brought down signifi- over the past year has increased India's im- cantly, and domestic investment and FDI as a port bill by some $4 billion and Pakistan's by share of GDP have increased from generally $650 million, increasing pressure on balance low levels. Indicators of human capital have of payments positions, especially for Pakistan, also improved, with school enrollment rates where external financing difficulties are ex- rising, illiteracy falling, and life expectancy in- pected to continue. Moreover, uncertainty gen- creasing by three years over the last decade. erated by the high debt levels and precarious Estimates for longer-term growth assume fiscal position of central and state governments that the region's high potential, as embodied is likely to constrain private sector activity, in the initial conditions above, will be fully used. Relative to the 1990s, total factor pro- Long-term prospects ductivity in India, for example, is expected to Average GDP growth for South Asia over the continue growing at a slightly higher base (by 2003-10 period is anticipated to register 5.4 0.2 percent) in the next decade. The abundant percent, about 0.3 percentage points higher supply of Indian workers with training in high than in projections prepared one year ago. technology sectors should continue to provide This pace of output growth, combined with strong momentum to the software industry. declining rates of population growth, should Total investment is expected to maintain support advances in per capita incomes of growth of 8 percent throughout the next close to 4 percent per year over the 2000-10 decade, with most growth emanating from the period, a marked improvement over the 1990s private sector. Demand for the region's ex- record of 3.5 percent growth (table A1.2). ports is expected to continue to grow rapidly, South Asia begins the new decade after having with import growth in South Asia's principal 142 REG IONA L EC ONO MI C P R O S P E C T S Table A1.2a Forecast assumptions- However, countries such as India and Pak- South Asia istan face major challenges in achieving the Initial conditions potential rate of output growth over the next ipl3}:v;' decade. High levels of domestic debt and large industrial inSouth AsiaptA' fiscal deficits present substantial difficulties in 2. Trade (X+M) / GDP ratio (rea achieving fiscal consolidation while maintain- 3. Median inflation rate (percent) ing expenditures that are necessary for growth. 4. Median fiscal balance! GDP(e A reduction in unproductive subsidies and 6. Investment / GDP (nominal) i stepped-up investment in human capital and 7. Gross national savings / GDP infrastructure are essential to this effort. Infra- 7a. Gross domestic savings / GDP structure bottlenecks and in 8. Current account balance / GDP delays privatiza- 9. FDI / GDP tion may limit the acceleration of growth in 10. External debt/ exports* the real and financial sectors. Also, much re- 11. School enrollment rates Primary (pct of eligible mains to be done to improve the competitive- population) ness of the region's export industries. The in- Secondary < c , , creased focus of the government of India on 12. Illiteracy rate (pct of people 4, 15+ years) trade liberalization has coincided with some 13. Under-5 mortality rate increases in tariffs and an intensified use of (per 1,000 live births) < g anti-dumping measures. High tariff rates, for 14. Life Expectancy at birth (years) <* . ; 14 LieEpcanyaitherexample, an average of 40 percent for all Exogenous assumptions , goods in India in 1999-2000, limit exporters' 1. Population growth access to cheaper, more efficient industrial in- t . Population growth ---1* <, 2. Market's GDP growth puts, and serve in the long term to limit pro- 3. Oil price $/bbl (avg.) < - ductivity gains. 4. Market's import growth *Exports of goods and services plus workers' remittances. Note: Market growth is trade-weighted partner GDP / import Latin America and the Caribbean growth. Source: World Bank database, World Bank staff estimates. Recent developments The economic recovery in Latin America export markets rising from 6 percent in the T has been broadly favorable, with the re- 1990s to 6.2 percent over 2000-10. Intra- gion's GDP expected to rise by 4 percent in regional trade and economic integration with 2000. Stabilization of global financial markets the world are assumed to accelerate as an eas- and the burgeoning of world trade growth ing of import substitution policies and trade have come to support a general resumption of and industrial restrictions takes place. Smaller economic activity across the region. As in East countries such as Bhutan, Maldives, Nepal, Asia, this has been complemented on the do- and Sri Lanka will benefit from the reduc- mestic front by a steady improvement in most tion in larger-country import barriers. But an macroeconomic indicators through the course important factor likely to restrain growth is of 2000. Inflation declined or held steady in the dependency of the region-and especially most countries (Ecuador was a notable excep- the smaller countries-on a limited number of tion), allowing interest rates to continue on a export crops, for example, cotton, tea, and falling trend. Unemployment dropped and real rubber. Volatility and secular decline in com- wages rose in Brazil, Chile, and Mexico com- modity prices are likely to continue to pres- pared with 1999 averages, but unemployment sure merchandise export receipts. remains high in Argentina and Colombia. Ex- 143 GLOB AL EC ONO MI C PRO SP EC TS Figure A1.3 World and selected Latin American export growth in U.S. dollars (Percent, three-month moving average year over year) 30 Mexico 25 15 10 < ~ \ / / World -10 0 -5 -10 ~~~~~~~~~~~~~~~~~~~~A,ecuig Mexico -15 June 1997 Dec. 1997 June 1998 Dec. 1998 June 1999 Dec. 1999 June2000 Note: LAC in this case refers to Argentina, Brazil, Chile, Colombia, and Mexico; world refers to 30 countries that are responsible for 82 percent of the world's exports. Source: World Bank data. change rates stabilized in several countries the "engine" for world activity through this that experienced periods of freefall during 1999 period. Mexican growth has continued to be (such as Brazil and Ecuador), restoring a de- buoyed by the U.S. import boom in 1999- gree of purchasing power and improving the 2000, with business cycles in the two countries outlook for private consumption and invest- becoming more closely aligned-and likely ment spending. reaching high points in 2000. In addition, the In real terms, most exchange rates have ap- usual exchange rate difficulties that Mexico preciated in 2000, but they are still low enough experienced with earlier electoral cycles was to facilitate rapid export growth in countries noticeably absent this time, in part because of such as Brazil, Chile, Colombia, and Peru, prudent macroeconomic policies that helped to leading to an improvement in trade balances restrain inflation under 10 percent for the first for most countries (including oil importers). time since the 1994 peso devaluation. As Mex- Merchandise exports in dollar terms from the ico approaches a peak in its growth cycle, while region's largest economies (excluding Mexico) others are escaping the trough, an implication exhibited a sharp recovery from the lows ex- is that near-term growth (2000-01) for Latin perienced in 1998, growing by over 17 per- America as a whole is unlikely to display the cent year on year during January-June 2000; distinct V-shaped pattern of recovery evident in Mexico's exports advanced by 25 percent (fig- East Asia. ure A1.3). Mexico is an exception within the region Near-term outlook with respect to the positioning of countries on Volatility in financial markets and primary the recovery and growth cycle. Whereas most commodity prices continues to pose a threat to Latin American countries experienced negative the recovery in Latin America. Sharply declin- or slow growth in 1999 because of fallout from ing equity prices during the first half of 2000 the Asian and Brazilian crises, Mexico bene- contributed to a period of uncertainty in global fited from its special trading relations through financial markets at a time when key commod- NAFTA with the United States, which remained ity price movements for the region also di- 144 REG IONA L EC ONO M IC PROS PE CTS Table A1.3 Latin America and the Caribbean forecast summary (percent per year) Baseline forecast Growth rates/ratios 1990-2000 19 1999 2000 2001 2002 2000-10 Real GDP growth 3.4 2.0 0.1 4.0 4.1 4.3 l 4.3 Consumption per capita 0.9 0.8 -2.9 1.8 2.3 2.6 2.3 GDP per capita 1.7 0.4 -1.5 2.4 2.5 2.8 3.0 Population 1.7 1.6 1.6 1.6 1.6 1.5 1.4 Median inflationa 16.6 7.6 9.3 8.2 9.0 7.0 7.2 Gross domestic investment/GDP 195A 20.9 19.5 19.8 20.3 20.8 21.7 Median central gvt. budget/GDP -2.8 -2.3 -2.9 -1.9 -2.0 -2.0 -1.3 Export volumeb 8.4 7.5 7. 8.9 7.8 7.3 7.0 Current account/GDP -2.3 -4.0 -2.9 -2.8 -2.6 -2.6 -2.0 a. GDP deflator. b. Goods and nonfactor services. Source: World Bank baseline forecast, October 2000. verged. The oil price rose sharply, while non- Table A1.3a Forecast assumptions- energy commodity prices of importance to the Latin America and the Caribbean region weakened-particularly coffee, grains, initial conditions 19890 19982000 and soybeans. Although most of the large countries experienced strong gains in industrial 1. Ratio of real income per capita: output in the first quarter of 2000, the recovery 2. Trade (X+M) / GOP ratierica 2.6 51.5 appeared to have faltered in the second quar- 3. Median inflation rate (percent) 24.4 6.1 ter, except for the oil exporters Ecuador and 4. Median fiscal balance / GDP -1.4 -23 Venezuela. And private capital inflows fell dra- 5. Investment / GDP (real) 18,4 202 6. Investment / GDP (nominal) 21.4 19.9 matically. Argentina was particularly hard hit 7. Gross national savings /GDP 203 17.4 by these developments as they coincided with 7a. Gross domestic savings / GDP 23.9 19.3 strong fiscal adjustment and a political crisis 8. Current account balance / GDP -0.7 -3.6 9. FDI / GDP 0.8 3.9 that weakened investor confidence and delayed 10. External debt / exports' 279.1 202. the economic recovery. Nonetheless, consoli- 11. School enrollment rates dation of the region's recovery in 2001-02 is Primary (pct of eligible likely, as adjustment in Brazil has been impres- population) 84.0 94.0 Secondary S8.0 66.0 sive so far, and new governments in Argentina 12. Illiteracy rate (pct of people and Mexico appear set to embark on a path 15+ years) 1S.0 12.0 of deepened reforms. Global conditions are ex- 13. Under-5 mortality rate (per 1,000 live births) 49.0 38.0 pected to remain supportive of growth in the 14. Life expectancy at birth (years) 68.0 70.0 region, with above-average world trade growth, gradual recovery in key non-oil commodity Exogenous assumptions 199t 20010 prices, declining but still moderately high oil 1. Population growth 1. Populatlon growth1.7 1lA prices, and a modest increase in private capital 2. Market's GDP growth 2.7 3.1 flows. The Caribbean islands, with their in- 3. Oil price $/bbl (avg.) 18.2 20.2 creasing reliance on tourism revenues, are also 4. Market's import growth 7.8 6.2 expected to benefit from moderately strong in- 'Exports of goods and services plus workers' remittances. come growth in North America and Europe Note: Market growth is trade-weighted partner GDP / import growth. over the next two years. Latin American region Source: World Bank database, World Bank staff estimates. 145 G LO B AL EC ONO MI C PRO SP EC TS output growth is expected to reach 4.1 percent macroeconomic policies to deal with volatility, in 2001 and to rise further to 4.3 percent by which may have contributed to the observed 2002 (table A1.3). cycle of booms and busts in many countries during the 1980s and 1990s. Moreover, vul- Long-term prospects nerability of the region to swings in external Per capita GDP growth over the long term financing is likely to remain a concern in the (2003-10) is likely to average around 3 per- long run. Low national savings and the persis- cent, about 0.2 percentage points higher than tence of large debt overhangs will require in projections prepared one year ago. Some of rollover on a continuing basis. And this fun- the elements supporting this cautious opti- damental exposure to international financial mism are highlighted in table A1.3a and in- conditions underlines our view that per capita clude: a definitive movement toward greater growth potential is unlikely to breach 3 to 3.3 domestic macro stability, as median inflation percent over the next decade, which would, rates dropped from 24 percent to 6 percent nonetheless, be twice as fast as occurred during over the decade; a two-point increase in real the 1990s. investment as a share of GDP, supported by strong FDI inflows, surging from less than 1 percent to almost 4 percent of regionwide out- Europe and Central Asia put. And openness to investment flows has been complemented by a remarkable increase Recent developments in integration with global trade flows-with r DP growth in the Europe and Central this measure doubling as a proportion to GDP Asia region (ECA) is expected to acceler- over the last 10 years. Finally, indicators of ate sharply through 2000, after hitting a human capital have improved, with primary trough in 1998 and early 1999 following the and secondary school enrollment rates rising August 1998 financial crisis in the Russian by some 10 points. Federation. Growth in 2000 is anticipated to The improved state of initial conditions for rise to 5.2 percent, significantly higher than the outlook joins with a more definitive trend the 1 percent realized in 1999. Notably, for toward market-friendly policies in the larger the first time since the onset of the transition countries, such as Argentina, Brazil, and Mex- and the breakup of the Soviet Union, almost ico, and potential for technology spillovers all of the countries in the region are expected from the United States (particularly for Mex- to record positive growth. Short-term projec- ico). Banking and financial sectors in the large tions stand well above those prepared for the economies weathered the global financial crisis Global Economic Prospects report one year of 1997-98, in part because of reforms enacted ago, primarily because of the unexpected in many countries following the Mexican cri- strength of the rebound in Russia. And a com- sis. Further strengthening of prudential regula- mon element supporting near-term growth tion and supervision should support financial across the region is a substantial pickup in ex- deepening and help to diminish the incentives ports, in large part because of rising import for capital flight. Finally, the strong inflows of demand from the Euro Area. FDI in recent years into areas of the economy In Russia, President Putin's apparent will- that could raise growth of productivity sub- ingness to introduce reforms has eased some stantially-telecommunications, utilities, ports, political uncertainties and improved business and so forth-should produce dividends in the confidence. Russian industry has continued to next decade compared with the relatively poor benefit from the impact of import substitution performance of the last 10 years. However, driven by the sharp devaluation of 1998; this there is need for more progress in financial and can be witnessed in an industrial production 146 REG ION AL EC ON OM IC P R O S P E C T S growth rate of close to 10 percent during the The Central and Eastern European coun- first half of 2000. This fillip to growth is di- tries (CEECs) are benefiting from growing de- minishing, however, with the recent real appre- mand from Western Europe and, to a lesser ciation of the ruble. An additional, more recent degree, from Russia. This is boosting growth driver to Russia's unanticipated recovery is the in Hungary (5.8 percent) and Poland (4.4 per- windfall increase in oil and natural gas export cent), in particular. In the Baltic countries, es- revenues. Higher oil prices improved the fiscal pecially Estonia and Latvia, GDP growth of balance from -4.2 percent of GDP in 1999 some 4.8 and 3.8 percent, respectively, in to +1.6 percent during the first half of 2000, 2000 is also largely export-driven. However, and has yielded a primary surplus of 4.8 per- high oil prices, combined with the deprecia- cent. This has supported continued reductions tion of the euro, are putting additional pres- in government wage arrears, contributing to sure on external balances and contributing to higher disposable incomes.1 The current ac- higher inflation. Turkey's economy has contin- count surplus is expected to register close ued to recover, reaching growth above 6 per- to $30 billion in 2000, or some 15 percent cent in 2000, up from the sharp 5.1 percent of GDP, boosting foreign reserve holdings contraction experienced in 1999, because of a considerably and easing the need for near- rebound in domestic demand tied to marked term external financing. Ukraine is now regis- declines in real interest rates and reconstruc- tering gains in output, although the political tion expenditures in the wake of the 1999 context remains difficult. For several of the earthquake. Business confidence has also im- hydrocarbon-rich Commonwealth of Indepen- proved in response to implementation of an dent States (CIS), higher oil and gas prices are IMF-sponsored stabilization program begun providing extraordinary export earnings and in January 2000. Recent data, however, show contributing to higher output. Oil and gas ex- that the current account deficit will be larger port volumes should increase as well, as export than targeted, reflecting strong domestic growth, markets (especially in Western Europe) are ex- real exchange rate appreciation and higher oil pected to achieve stronger growth. prices. Figure A1.4 Eurobond spreads for selected ECA countries Basis points 5,000 |- Czech Republic Romania - Russian Federation Turkey 4,500 4,000 3,500 3,000 2,500 2,000 1,500 500 _- . . i ~_ _ _ ___ 0 ~Cp 5e9aX- \ 1 i 0 gG 4 se 59 Source: Bloomberg and World Bank data. 147 G LO B AL EC ONO MI C PRO SP EC TS Table A1.4 Europe and Central Asia forecast summary (percent per year) Baseline forecast Growth rates/ratios 2000 2001 2002 2000-10 Real GDP growth -1e a ntGD 5.2 4.3 3.9 9 4.1 Consumption per capita 1 . . 4.9 4.1 3.6 4.2 GDP per capita Eu o p -0 j5.1 4.2 3.8 4.1 Population 0.1 0.1 0.1 0.1 Median inflationa 33 1Z 4 8.0 6.9 5.5 I 4.6 Gross domestic investment/GOP 5. 230 24 22.6 22.9 23.1 i 23.6 Median central gvt budget/GDP 3, -.0 45 -4.0 -3.7 -3.5 -3.6 Export volumeb 0. . O8.3 6.5 6.4 6.8 Current account/GDP -01 17 6 2.8 1.4 0.6 -1.3 Memorandum item GDP Central and Eastern Europe 1.' 25 25 3.9 3.9 4.2 , 4.4 GDP CIS states -. -33 26 5.9 4.5 3.3 3.5 a. GOP deflator. b. Goods and nonfactor services. Source: World Bank baseline forecast, October 2000. In most ECA countries, exchange rates have Table A1.4a Forecast assumptions- broadly stabilized since mid- to late 1999, and Europe and Central Asia inflationary pressures remain largely under con- initial conditions trol. Early in 2000, several currencies faced up- ward pressure against the euro, including the 1. Ratio of real income per capita: industrial I Europe and Central Asia 69 1. Polish zloty, the Czech and Slovak koruny, as 2. Trade (X+M) GDP ratio (real) Asi well as the Turkish lira following the introduc- 3. Median inflation rate (real) 4 5 tion of a crawling peg regime at the beginning 4. Median fiscal balance G GDP 5. Investment!/ GOP (real) .4 21 of the year. Interest rates have eased in Turkey 6. Investment / GDP (nominal) and in Russia, where higher fiscal revenues 7. Gross national savings I GDP 2 have reduced pressures on central bank financ- 7a. Gross domestic savings GDP 8. Current account balance / GDP 21 -. ing. Interest rates have been reduced in the 9. FDI/ GDP . Czech Republic and Hungary to stimulate do- 10. External debt / exports' mestic demand, in contrast with rate increases 11. School enrollment rates Primary (pct of eligible in Poland intended to slow domestic demand population) and import growth. Investor perceptions of po- Secondary tentially improved prospects for the region, in- 12. Illiteracy rate y pcr of people 15+ years)4. 4. cluding reduced political uncertainty in Russia, 13. Under-S mortalitv rate have led to a large decline in spreads on sec- (per 1,000 live births) . ondary market bonds (figure A1.4). 14. Life expectancy at birth (years) Exogenous assumptions 1 Near-term outlook I h growth Growth performance for the region through 2. Market's GDP growth 2002 is expected to remain relatively strong 3. Oil price $/bbl (avg.) 0X2 202 in the aggregate (table A1.4). Output growth 4. Market's import growth is expected to stabilize near 4 percent over 'Exports of goods and services plus workers' remittances. 2001-02. The moderate slowdown reflects Note: Market growth is trade-weighted partner GDP / import ina growth. short-term effects of structural reforms in a Source: World Bank database, World Bank staff estimates. 148 REG IONA L EC ONO MIC P R O S P E C T S number of countries; a tightening of policy to percent (though largely concentrated in the avoid overheating in a few Central European CEECs), and trade openness has doubled, re- countries; and anticipated easing of energy flecting earlier western reorientation of trade prices, affecting CIS performance. But devel- and a nascent recovery of intraregion flows. A opments in export markets, policy changes re- principal area of weakness, particularly in lated to European Union (EU) accession, and contrast with other emerging-market regions, the highly uncertain path of the oil price will remains the paucity of investment, and there be critical factors in shaping the outlook. has been a 10-percentage point decline in gross Output among the CEECs is expected to national saving over the period. Despite in- pick up in 2001-02 to 3.9 and 4.2 percent re- creasing inequality in Russia and other CIS spectively, as export markets remain firm and countries, the basic quality of the labor force deeper domestic reforms contribute to improved remains potentially strong, and it is an asset macro stability and continued dynamism of the that, combined with improved physical capi- private sector. These countries should continue tal, could sow the seeds for more rapid growth to attract high levels of FDI flows, linked tightly in productivity and living standards. to the EU accession process.2 Although fiscal Given these initial conditions, GDP for the and current account pressures will remain high region is projected to expand at a fairly robust in most of the CEECs, they are expected to ease 4.1 percent per year for the period 2000-10, gradually because of policies that converge more contrasted with its decline of -1.9 percent over rapidly with Western European norms. In con- 1990-2000. But the regionwide forecast again trast, the outlook for Russia and other hydro- masks expected divergences in growth outcomes carbon exporters of the CIS is particularly un- at the country level. Countries anchored by the certain, given the state of flux in current and EU accession process have achieved a greater prospective oil market developments. Growth in degree of stability and realignment of institu- these countries is expected to slow moderately tions and markets, positioning them for stronger in 2001, with sharper deceleration possible in growth compared to most of the states of the 2002, as oil prices retreat from current high lev- CIS, which have not wholly supported reform els (see the commodities section of this report). as extensively. The current long-term projection for the region is more optimistic than that pre- Long-term prospects sented in forecasts of one year ago (3.4 percent The coming decade is likely to be character- average GDP growth for the period 1999- ized by substantially higher average growth 2008)-a revision taking into account three rates than witnessed during the difficult initial main factors. transition period of the 1990s. As is appar- The new projections reflect a substantively ent, relative performance of countries within revised assumption that over the near to ECA has varied tremendously over the last medium term, Russia's new administration decade, with Poland having re-attained pre- will be partly successful in improving eco- transition GDP levels by 1996, while Russia nomic management and in implementing re- and Ukraine still languish some 40 percent cently proposed social and economic reforms. below that level. Despite less-than-satisfactory And growth prospects in the CEECs will begin outturns in a number of transition countries, to reflect benefits associated with the Decem- there has been a degree of underlying progress ber 1999 Helsinki Accord of the European in strengthening some of the fundamentals Commission, a decision to extend invitations for longer-term growth (table A1.4a). Median for EU membership to Bulgaria, Latvia, inflation is now one-quarter of its 1988-90 Lithuania, Romania, the Slovak Republic, value, while fiscal balances have been main- and Turkey. (This is in addition to invitations tained at moderate levels. FDI has risen as a previously extended to the "early accessors": share of regional GDP from 0.1 percent to 2.3 the Czech Republic, Estonia, Hungary, Poland, 149 GLOBAL ECONOMIC PROS P ECT S Figure A1.5 Nonenergy commodity prices-world and Sub-Saharan Africa Index, July 1996= 100 110 100 ~~~~~ r\ ~~~~~~~~~~Sub-Saharan Africa index* 90 80 World index 70 July Nov. March July Nov. March July Nov. March July Nov. March July 1996 1996 1997 1997 1997 1998 1998 1998 1999 1999 1999 2000 2000 Weighted index of non-oil commodities traded by Sub-Saharan African countries. Source: World Bank data. and Slovenia.) Finally, the somewhat higher Sub-Saharan Africa trajectory of growth in world trade and West- ern European output in the new baseline has Recent developments a positive impact on expansion in the ECA allout from the 1997-99 crisis continued region. Fto exert a dampening effect on much of the Despite the upward revision in aggregate Sub-Saharan African economy in 2000, as growth, there are a number of downside risks. non-oil commodity export prices continued to In Russia, risks center on the extent to which decline over the first half of the year (figure the new government will be able to transform A1.5). But higher oil revenues boosted growth the current windfall gains into long-term for the region's oil exporters, while South growth. Policy measures among the EU candi- Africa also strengthened modestly to 2.2 per- date countries have contributed to increased cent growth following several years of sub- economic stability and have helped attract sub- dued performance. These and other support- stantial foreign capital. But policy mismanage- ing factors helped to raise GDP growth to ment could expose the CEECs to a sharp re- an estimated 2.7 percent from 2.1 percent in versal of these inflows. There is also a risk that 1999, and per capita income rose by 0.2 disputes within the European Commission percent. over reforms to its institutions, decisionmaking Substantial variation in performance was procedures, and budget and agricultural subsi- apparent across the region. Countries with bet- dies-which must be undertaken to accommo- ter policy environments-Botswana, Uganda, date an expanded membership-could delay and the Communaute Financiere Africaine the accession process, thus deterring foreign in- countries (excluding C6te d'Ivoire)-tended vestors. Similarly, within the applicant coun- to perform better than average, with GDP up tries, as more contentious reforms are intro- 4.4 percent. Oil producers benefited from duced, political support for joining the EU may buoyant export receipts and strong investment diminish, potentially slowing the accession pro- and grew by 3.S percent. In East and southern cess and growth within these countries. Africa, many countries lagged behind. The 150 REG IO NA L EC ONO MI C PRO SP EC TS Table A1.5 Sub-Saharan Africa forecast (percent per year) Baseline forecast Growth rates/ratios 1990.2000 1998 1999 2000 2001 2002 2000-10 Real GDP growth 2.1 2.0 2.1 2.7 3.4 3.7 3.6 Consumption per capita -0.6 -0.6 -1.0 0.3 0.6 0.6 0.8 GDP per capita -0.6 -0.5 -0.3 0.2 0.9 1.3 1.3 Population 2.6 2.6 2.5 2.5 2.5 2.4 2.3 Median inflationa 10.4 5.3 5S1 4.3 3.6 3.5 4.1 Gross domestic mnvestment/GDP 19.5 ZO2. 19.9 20.1 20.7 21.0 21.9 Median central gvt. budget/GDP -3.7 -3.3 -3.0 -2.9 -2.4 -2.0 -1.8 Export volumeb 4.6 0.7 2.4 6.1 5.3 5.9 6.0 Current account/GDP -1.5 -3.5 -2.3 -2.7 -2.2 -3.1 -3.1 Memorandum item GDP major oil exportersc 2.5 2.3 3.2 3.5 4.0 3.8 3.8 GDP region x S. Africa/oil-X 2.6 3.6 2,9 3.0 3.6 4.1 4.1 a. GDP deflator. b. Goods and nonfactor services. c. Angola, Gabon and Nigeria. Source: World Bank baseline forecast, October 2000. weather was partly to blame, as drought in 2000. Country-specific factors, especially con- Kenya and Ethiopia, floods in Mozambique cern over the value of the rand and political and South Africa, and hurricanes in Mauritius developments in the region, were mainly re- contributed to a string of disappointing results. sponsible. A conservative fiscal stance helped Countries experiencing civil strife or major to reinforce foreign confidence, though a sec- political disruption-the Democratic Republic ond successive year of declining real public of Congo, Ethiopia, Sierra Leone, and Zim- consumption did nothing to counter the weak- babwe-registered the weakest performances, ness elsewhere in the economy. with GDP falling 1.5 percent during the year. A strong rise in oil-related export revenue Near-term outlook particularly afforded some breathing space to Sub-Saharan Africa should further consolidate Nigeria's new, democratically elected govern- its recovery, as growth accelerates to 3.4 per- ment. While the country faces enormous cent in 2001 and 3.7 percent in 2002 (table short-term problems, a much needed boost to A1.5). Oil exporters will benefit from prices government revenues and strong foreign in- that are expected to remain at high levels vestment inflows underwrote growth of some through 2002. Further, Angola, Equatorial 3.1 percent and yielded a sharp improvement Guinea, Nigeria, and Sudan are all scheduled in the balance of payments and fiscal ac- to bring additional supplies and exports on- counts. This is helping to tide the country over stream.3 However, these gains will in part be until a possible resumption of International offset by terms-of-trade losses to many other Monetary Fund lending and hoped-for debt countries in the region, especially beverage ex- relief. In South Africa, a weak performance by porters, who are facing the lowest prices in a agriculture was the main cause of growth that generation (see the commodities section of this was somewhat below expectations, though report). At least, the worst is now likely over, the country also experienced renewed turbu- and terms of trade are expected to stabilize or lence in financial markets, as investor senti- improve modestly as non-oil commodity prices ment turned negative in the fourth quarter of begin to firm. Also, despite the price weakness, 1999 and remained so through the first half of privatization and deregulation are promoting 151 G LO B AL EC ONO MI C PRO SP EC TS Table A1.5a Forecast assumptions- having received a total of close to $9 billion of Sub-Saharan Africa relief in net present value terms. Several more Initial conditions African countries are expected to reach deci- sion points in the near future. 1. Ratio of real GDP per capita: Industrial / Sub-Saharan Africa 3. 7. 2. Trade (X+M) / GDP ratio (real) 3 , Long-term prospects 3. Median inflation rate (real) 1 Despite the growth slowdown of the late 4. Median fiscal balance / GDP 1990s, recent performance continues to sup- 5. Investment!/ GDP (real) 13 1. 6. Investment / GDP (nominal) 17. port the view that fundamental structural 7. Gross national savings/ GDP change and institutional strengthening will 7a. Gross domestic savings / GDP , . 4 have a significant impact on Sub-Saharan 8. Current account balance / GDP - i 4F . 9. FDI / GDP - Africa's prospects. The forecast is for a halt to 10. External debt / exports2 the region's lengthy decline and marginaliza- 11. School enrollment rates tion and even for a moderate reversal. The Primary (pct of eligible population) 4 longer term (2003-10) outlook is for sus- Secondary . , tained GDP growth-around 3.7 percent- 12. Illiteracy rate (pct of people = with per capita incomes rising 1.5 percent per 15+ years) 5. '4. 13. Under-5 mortality rate year. The primary driving force behind the (per 1,000 live births) . outlook remains better governance and ongo- 14. Life expectancy at birth (years) ing reforms to the policy environment. Table Exogenous assumptions A1.Sa highlights improvements in a number 2. ge of economic outturns over the last decade that 1. Population growth 2 serve as the initial conditions for development 2. Marker's GDP growth Z 3 3. Oil price $/bbl (avg.) 1 4 into the next 10 years. Median inflation and 4. Market's Import growth . fiscal deficits have been reduced, while moder- , Not available. ate gains in real investment and in FDI have 'Exports of goods and services plus workers' remittances. also been achieved. To a degree, these have Note: Market growth is trade-weighted partner GDP / import been countered by a fall in savings rates and a growth. Source: World Bank database, World Bank staff estimates. rise in external deficits (characteristic of the secular decline in commodity prices). Increases in school enrollment and the decline in illiter- greater supply and export growth in key mar- acy rates are more encouraging indicators for kets such as cotton and cocoa in west Africa the stock of human capital. However, HIV/ and copper in Zambia. Nevertheless, the im- AIDS is expected to carry substantial negative pact on real incomes will result in weaker do- effects for the future labor force (see below). mestic demand and slower growth in the near Growth is likely to remain modest com- term. The baseline also assumes a return to pared to that of other emerging regions, re- more normal weather patterns, which will fur- flecting a range of negative factors still to be ther boost agricultural production and exports. overcome, including poor transportation and Trade liberalization, particularly in COMESA, communications infrastructure, low savings SADC, and the South African-EU FTA, should and private investment rates, and limited ac- spur greater trade and regional cooperation. cess to foreign capital (World Bank 2000). And finally, the Heavily Indebted Poor Coun- Moreover, without substantial diversification tries (HIPC) Initiative is gaining momentum, of production, economies in the region will with nine African countries-Benin, Burkina remain overexposed to irregular weather con- Faso, Cameroon, Mali, Mauritania, Mozam- ditions and unfavorable terms-of-trade shocks. bique, Senegal, Tanzania, and Uganda-now Unfortunately, on balance, there seems very 152 REG IO NA L EC ON OM IC P R O S P E C T S little prospect for achieving widespread per cision. Nevertheless, a growing body of survey capita growth rates on the order of 4 or 5 per- work has attempted to measure the effects on cent or more, which have characterized East households and businesses, communities, and Asia's best performers. governments. Various studies have identified Access to foreign savings may also prove a wide range of costs. These include reduced problematic. Apart from the HIPC Initiative, household savings and labor supply caused by foreign aid is likely to diminish further. The the expenditure of time and money in caring enhanced HIPC Initiative is worth nearly for sick family members or raising orphaned $30 billion in net present value terms, with children; lower productivity in the business some 80 percent of the program earmarked sector because of illness, absenteeism, skill for Sub-Saharan Africa. However, even this shortages, and higher training costs; and diver- level of resource transfer-roughly 9 percent sion of government budgets from expenditure of 1998 GDP-is small compared to the re- on education and infrastructure. The impacts gion's requirements, and it will be necessary to are more severe in sectors such as transpor- attract more private capital. The main poten- tation, construction, and power generation, tial benefit of the HIPC Initiative may be the where male workers live away from their fam- impetus it gives toward strengthening the pol- ilies (UNAIDS 2000, pp. 26-36; Bollinger and icy framework and poverty reduction objec- Stover 1999). Macroeconomic impacts have tives in the region. been modeled using various approaches at both AIDS and the economy. The forecast at- a regional level and in country-specific studies tempts to factor in more carefully the eco- of Cameroon, Kenya, Swaziland, Tanzania, and nomic effects of HIV/AIDS-an issue of huge Zambia. These studies yield similar estimates importance for Sub-Saharan Africa. Accord- of the prospective cost, generally a fall in per ing to U.N. estimates (UNAIDS 2000), Sub- capita growth of 1 percent or more annually Saharan Africa is home to 24.5 million (or 70 for countries with high, though not the most percent) of the 34.3 million existing cases extreme, incidence rates (Over 1992; Bollinger worldwide and 12.1 million of a total of 13.2 and Stover 1999). million AIDS orphans. Moreover, the epidemic Projecting the medium-term economic im- appears to be increasingly concentrated in pact is further complicated by variations in Africa, where 4 million of the 5.5 million new government policy, which can have a major in- infections occurred in 1999. Southern Africa fluence on the course of the disease. In Uganda is particularly affected, with Botswana, Swa- and Zambia, concerted government efforts ziland, Zimbabwe, Lesotho, Zambia, South have helped to lower significantly incidence Africa, and Namibia having incidence rates rates in high-risk populations (see, for exam- in the adult population between 36 percent ple, UNAIDS 2000, p. 10). Nevertheless, until (Botswana) and 19 percent (Namibia). By now such interventions have been all too rare. contrast, incidence rates in north and western Thus, fairly pessimistic estimates of the im- Africa are typically below 3 to 5 percent, pact on population growth and productivity though they range up to 8 percent in Cam- must be factored in, particularly for countries eroon and 11 percent in C6te d'Ivoire. The where the incidence is currently high, a cate- fact that victims tend to be working adults in gory that includes some of Sub-Saharan the prime of their lives amplifies not only the Africa's consistently strongest performers. In tragic human impacts but the social and eco- the worst-affected countries, population growth nomic disruption as well. is expected to slow by 1 to 2 percent annually Because the scale of the HIV/AIDS epidemic (possibly even to turn negative), while per is unprecedented in recent history, it is difficult capita income growth may slow by nearly that to gauge its macroeconomic impacts with pre- much again. 153 G LO B AL EC ONO MI C PRO SP EC TS Figure A1.6 Oil prices and current account balances, Middle East and North Africa, 1995-2000 Millions of U.S. dollars US$/bbl 15,000 - 30 15|OW - Current account balance 4 Average oil pnce | 10,000 - 25 5,000 oe- 20 0 Right axi 15 -5,000 -10,000 - 10 Left axis -15,000 - ,5 -20,000 - . -0 1995 1996 1997 1998 1999 2000 Source: World Bank staff estimates. Middle East and North Africa ing export revenues and incomes in the oil exporters. As OPEC increased production Recent developments quotas, export volumes have risen, and most espite some favorable developments, in- OPEC countries are now producing at peak D cluding higher oil prices, the Middle East capacity. The impact on fiscal positions has and orthAfria reions grwth rospcts been favorable; major oil producers had for- and North Africa region's growth prospects muaebdgtarndnasmd l rc remain modest in the short and medium mulated budgets around an assumed oil price terema modverstdevelopments in the shm edm of $22 per barrel, and higher revenues have terms. Adverse develomencontributed to lower deficits and borrowing East peace process are also casting a renewed requirements. For example, in 1999 Kuwaiti pall on prospects for tourism, investment, and oil revenues grew by 39 percent and public trade in the Levant. GDP growth of 2.2 per- revenues increased by 30 percent. External cent was reported in 1999, and growth of debt-financing constraints have eased and do- 3.1 percent is expected in 2000. Cyclical fac- mestic arrears have fallen. In Saudi Arabia, tors have played a role in the region's recovery, bank claims on the public sector had grown by with external factors such as the surge in oil 7.7 percent in 1998 and 3.5 percent in 1999, prices, recovery in traditional export mar- but fell by 8 percent in May 2000. As a result, kets-especially the Euro Area-and some im- pressure on exchange rates that were experi- provement in weather conditions beginning enced in 1998 and early 1999 in several Gulf to relieve drought; these factors are leading countries have eased significantly, and mone- to modest improvements in growth outturns. tary authorities in most oil-exporting countries But significant impediments to higher potential have built up sufficient foreign reserves to de- growth remain, as the countries in the region fend their exchange rate pegs. Current account face major challenges in the reform of domestic positions have shown large improvement, with policies affecting trade openness, exchange a turnaround of over $42 billion between 1998 rates, investment, and the labor force. and 2000, representing about 6 percent of oil Oil prices remained above $30 per barrel exporters' annual GDP (figure A1.6). Increased for much longer than anticipated in 2000, rais- imports by the Islamic Republic of Iran and Al- 154 RE GI ONA L E C ON OMI C P R O S P E C T S geria have diminished the magnitude of the re- in food imports. But drought conditions have gion's surplus to a degree. continued in Morocco for the second consecu- For the diversified exporters, favorable ex- tive year, with a significant impact on overall ternal conditions have been largely overshad- growth. Price stability in many diversified ex- owed by negative domestic effects. Growth porters has been maintained in a time of rose somewhat to 3.6 percent in 2000 from higher oil prices through energy subsidies to 3.3 percent during 1999. Strong growth in domestic consumers and maintenance of tight Europe has fueled a boom in tourism, with monetary policy in support of fixed exchange record numbers of tourist arrivals and receipts rate pegs to dollar-dominated baskets. How- being experienced in many North African and ever, there are some troubling elements in the Mediterranean countries. Tourist arrivals in- picture for diversified exporters that are limit- creased by 37 percent in the Arab Republic of ing GDP growth: Egypt, rose to a record 4.8 million in Tunisia, and are rising by 10 percent year on year in * Exchange rates in many countries are Jordan for the first four months of 2000. The pegged to an appreciating dollar, but these economic revival in Europe has also led to countries' major export markets are in the gains in some export categories (food and pri- Euro Area. Prices in the region are rising mary commodities, some mechanical goods, more rapidly than in the United States, im- and energy), but exports of labor-intensive plying that exchange rates are becoming goods such as textiles and clothing remain flat overvalued relative to other currencies. As a in value terms. Workers' remittances have also result, export growth is generally lower than been boosted (75 percent in 1999 in Tunisia) export market growth would suggest, and by the improvement in economic activity in several countries, particularly Egypt, are ex- the broader Euro-Mediterranean and Gulf re- periencing pressures on their exchange rates. gions. The gradual easing of drought condi- * Lower confidence in the group stemming tions in many countries improved agricultural from both economic and political factors is incomes and exports and led to some decline a second force restraining growth. Capital Table A1.6 Middle East and North Africa forecast summary (percent per year) Baseline forecast Growth rates/ratios 199-2O 1998 1999 2000 2001 2002 2000-10 Real GDP growth 3.l 3.3 2.2 3.1 3.8 3.6 3.6 Consumption per capita 04 -1.2 -0.3 1.2 1.5 1.3 1.3 GDP per capita 0.9 1.3 0.3 1.1 1.9 1.7 l 1.7 Population 2.2 2.0 2.0 2.0 1.9 1.9 l 1.9 Median inflationa ... 0.7 4.3 5.2 3.9 3.8 3.6 Gross domestic investment/GDP 21.6 22.0 22.3 22.6 22.8 23.1 23.7 Median central gvt. budget/GDP -1.6 .-3.3 -34 -2.0 -1.8 -1.7 -1.2 Export volumeb 4.6 -2.5 4.2 5.6 4.5 5.1 5.3 Current account/GDP 0A4 -1.0 1.7 1.3 1.0 0.5 E 0.7 Memorandum item GDP oil-dominant economies 2.5 0.9 1.C 3.2 3.3 2.9 3.1 GDP diversified exporters 3.9 5.5 3.3 3.6 4.7 4.9 4.4 ... Not available. a. GDP deflator. b. Goods and nonfactor services. Note: Excluding Iraq. Source: World Bank baseline forecast, October 2000. 155 GLOBAL EC ON OMI C PROS PE CT S markets have responded with disfavor to Table A1.6a Forecast assumptions- higher-than-anticipated fiscal deficits in Middle East and North Africa Egypt and Lebanon, with several down- Initial conditions j grades undertaken by ratings agencies for Lebanon, large declines in stock market in- I.Ratio of real GDP per cap Industrial / MNA region %,S dexes (by 50 percent in Egypt from January 2. Trade (X+M) / GDP ratio (real) to October), and rising spreads. Because of 3. Median inflation rate (percent) the effects on confidence of the recent con- 4. Median fiscal balance / GDP S. Investment /GDP (real) 3 flict in the West Bank and Gaza, and the 6. Investment /GDP (nominal) lower likelihood of a peace agreement in the 7. Gross national savings / GDP i near future in the Levant, the climate for in- 7a. Gross domestic savings / GDP , 8. Cuirrent account balanice I GDP -2 I vestment (and tourism) will be poorer and 9. FDI / GD?P0o6 12 GDP growth will show only modest, rather 10. External debt/exports2 than strong, recovery in 2000. 11. School enrollment rates Primary (pct of eligible population)8.0 0 Near-term outlook Seconidaryti The recent modest improvement in economic 12. Illiteracy rate (pet of people 15+ years) 44 $7 activity in 2000 and the effects of reform pro- 13. Uider-5 mortality rate grams under way in many countries will influ- (per 1,000 live births) ence the near-term outlook for the region fa- 14. Life Expectancy at birth years) vorably. Activity is expected to pick up to Exogenous assumptions 3.8 percent in 2001 and to slow slightly to o g 3.6 percent in 2002 (table A1.6). For oil- 2. Population growth , 7 exporting countries, the outlook is conditioned 3. Oil price $/bbl (avg.) by the expected path for oil prices in the period 4. Market's import growth and the lagged response to higher incomes. 'Exports of goods and services pis workers' remittances. With an average price of $28 a barrel for 2000 Note: Market growth is trade-weighted partner GDP / import and $25 in 2001, oil revenues should continue growthi. and $2Supp income 2 rowh renuhes shoild eporte, Source: World Bank database, World Bank staff estimates. to support income growth in the oil exporters, but large domestic and external arrears and business climate. And in a reversal of patterns structural budget deficits imply continued need of the 1990s, trade regimes are also expected for public sector expenditure restraint and re- to become more open, as several countries, forms. Improved agricultural conditions in Al- such as Saudi Arabia and Oman, seek and gain geria and the Islamic Republic of Iran, com- membership in the World Trade Organization. bined with continuing attempts to privatize The near-term outlook for the diversified state industries, and planned investments in the exporters indicates that growth will rise to hydrocarbons sector through both domestic 4.7 percent in 2001 and to 4.9 percent in and foreign investment, will make positive 2002. Favorable factors assisting growth in- contributions to growth. Moreover, the grad- clude the smooth changes in leadership in Jor- ual reform of the multiple exchange rate dan, Morocco, and the Syrian Arab Republic, regime in the Islamic Republic of Iran is ex- which contributed to lower political uncer- pected to continue, given the success of the tainty; the potential for reduction of political "export" rate in recent months. In the Gulf conflict in the Western Sahara; and efforts to Cooperation Council countries, continued in- jump-start the UMA (Arab Maghreb Union). vestments in hydrocarbons, reform of invest- The public sector in several countries will con- ment regimes (particularly the taxation of for- tribute significantly to growth, with bond is- eign firms), and a lowering of corporate taxes sues in Lebanon and Egypt to finance higher generally should provide a more favorable fiscal expenditures. The recent gradual down- 156 REG IO NA L EC ONO MI C P R O S P E C T S ward adjustment to the Egyptian pound is ex- others, with the exception of South Asia, with pected to continue into 2001, improving the no change in the ratio of trade to GDP over a prospects for net exports somewhat. Broaden- 10-year period. And investment has remained ing of privatization programs into areas such stagnant, with little improvement in private as telecommunications, airlines, and the fi- capital spending-a critical element in foster- nance sectors in Egypt, Jordan, Tunisia, and ing efficiencies and contributing to higher Morocco will attract additional foreign invest- long-term potential growth rates. ment, helping to deepen equity markets, and Other substantial obstacles to higher long- improve efficiency in operations of the sectors. term growth persist. Public sectors continue to However, unfavorable factors persist. The po- be large and inefficient in delivering services, litical uncertainty in the Levant appears to be and the institutional and regulatory capabili- continuing, with significant adverse impacts on ties in many sectors are not geared to fostering confidence and investment. The effects on in- private sector development. Moreover, the vestment may be felt outside the Levant, partic- region continues to rely heavily on narrow ularly because of "wrong neighborhood" ef- sources of external revenues, whether they are fects. Additionally, the pace of reforms in the product or export-market dominated, suggest- public sector, privatization programs, and mi- ing the potential for longer-term volatility in croeconomic reform in the private sector is very export earnings. Overvaluation of exchange slow, suggesting that the gains will not be felt rates in several countries will continue to have fully in the near term. And with the perfor- dampening effects on export growth. And the mance of the large agricultural sectors in many problem of noncompetitiveness of basic indus- countries heavily reliant on weather conditions, try will be further exacerbated by the proposed volatility in growth resulting from unpredict- abolition of the Multifibre Agreement (MFA), able weather conditions will continue. which will allow greater competition by low- cost producers in traditional export markets Long-term prospects for North African textiles and clothing. At this juncture, the moderately optimistic Recent political events have cast a pall near-term picture for the region does not trans- over confidence in the region, overshadowing late into significantly higher long-term growth the potential for improvements in regional in- potential. Despite reforms in many of the tegration resulting from the EU Association countries in the region, much of the improved Agreements and peace agreements in the Le- performance in the recent past owes much to vant. While tensions remain high, prospects cyclical and external factors such as weather, for intraregional trade, as well as foreign in- oil prices, and export market-growth. Signifi- vestment and tourism, will be poor. Even if cant structural impediments to higher long- some form of detente is reached in the Levant, term growth remain. Table A1.6a shows that it may result in a "cold peace" where coun- there have been fundamental (and likely secu- tries may not be in conflict, but the coopera- lar) improvements in some regional macro- tion required for trade relations may be lack- economic conditions and in the nurturing of ing. For oil exporters, the windfalls from the labor force over the last decade. Reduction higher oil prices are expected to be temporary, in inflation and fiscal imbalances, attraction and oil prices are expected to fall in the me- of additional FDI flows, and lowering of ex- dium term, suggesting that structural changes ternal debt ratios are significant. Moreover, need to be made to improve fiscal positions considerable improvements in school enroll- and increase the scope of private activity in ment, reduction of illiteracy, and better health domestic economies. This implies that while indicators are positive forward-looking indi- conditions in oil markets and major trading cators for the future labor force. But the re- partners are so favorable, the region should gion remains less open to foreign trade than take advantage of the current trends to cast a 157 G LO B AL EC ONO MI C PRO SP EC TS wider net and proceed more quickly with their lations with OPEC, a third liquid natural gas (LNG) domestic reform agendas. train is scheduled to come onstream at Bonny Island in 2002, raising production by 50 percent. The addi- tional output has already been presold. See, further, Notes (http://www.eia.doe.gov/emeu/cabs/nigeria.html). 1. The percentage of the population on wages "below subsistence" remains high at 27.6 percent, ac- cording to official estimates as of June 2000. However, References it has declined significantly from 33.5 percent as of Bollinger, Lori, and John Stover. 1999. "The Economic April. Impact of AIDS." The Futures Group Interna- 2. The EU market now accounts for 60 to 80 percent tional, Glastonbury, Conn. Available at (http:// of Central and Eastern European countries' exports. www.iaen.org/impact/stovboll.pdf). 3. Note that Angola, Equatorial Guinea, and Over, Mead. 1992. "The Macroeconomic Impact of Sudan are not OPEC members. For Nigeria, recent AIDS in Sub-Saharan Africa." World Bank. crude oil production has been marginally below the Processed. Available at (http://www.worldbank. OPEC quota of 2.157 million barrels per day. How- org/aids-econ/macro.pdf). ever, the Obasanjo administration has ambitions to in- UNAIDS. (Joint United Nations Programme on HIV/ crease production well beyond the current quota limit, AIDS). 2000, Report on the Global HIV/AIDS to 3 million barrels per day by 2003. A warming of Epidemic. Geneva: Joint United Nations Pro- relations with Nigeria's foreign joint venture partners gramme on HIV/AIDS. Available at (http://www. and a recent surge in exploration and development ac- unaids.org/epidemic update/report/EpiLreport.pdf). tivity indicate strongly that an increase of this magni- World Bank. 2000. Can Africa Claim the 21st Cen- tude will likely be feasible. Less problematic for re- tury? Washington, D.C. 158 Appendix 2 Global Commodity Price Prospects G LOBAL COMMODITY PRICES HAVE FOL- in prices in 1999 and 2000. Metals and min- lowed many different paths since the erals prices are projected to rise 2.2 and 2.4 lows after the Asian crisis, with crude percent, respectively, in 2001 and 2002 after oil prices rising sharply, agricultural prices re- rising 13.6 percent in 2000. Agricultural prices maining low, and metals and minerals prices continued to fall in 2000, with a decline of 5.2 staging a modest recovery. The recovery of percent, but are expected to increase 3.9 per- non-oil commodity prices lagged behind that of cent in 2001 and an additional 6.0 percent in oil prices because supplies of non-oil commodi- 2002 as global stocks begin to fall and de- ties were slow to adjust to low prices while oil mand increases in response to current low production was significantly reduced by OPEC prices and rapid economic growth. producers. Producers of non-oil commodities Over the balance of the decade, real com- have been left with large inventories that still modity prices1 are expected to reverse recent need to be absorbed before prices can rise moves as energy and metals prices fall and significantly. Metals and minerals prices have agricultural prices rise (see annex tables A2.2 begun to recover, rising 27 percent since their and A2.3 for real price forecasts for individual lows. However, agricultural prices remain near commodities and indexes). Real energy prices their cyclical lows (after a brief rally that was are projected to fall sharply from current lev- not sustained), because of continued produc- els, with real petroleum prices down 47 per- tion increases and large stocks. Rapid global cent by 2010 compared to 2000 levels as OPEC economic growth, which contributed to the and non-OPEC supplies increase. Agricultural sharp increase in crude oil prices in 1999 and prices are low by historical comparison, and 2000, is expected to fuel a recovery in non-oil real prices are expected to rise modestly over commodity prices during the next several years. the balance of the decade. By 2010, real agri- The near-term outlook is for the divergence cultural prices are projected to rise 9 percent in commodity prices to be reduced with de- relative to 2000. Metals and minerals prices clines in energy prices, further increases in have already made a significant recovery from metals and minerals prices, and a recovery in the lows of 1999, and by 2010 they are pro- agricultural prices (see annex tables A2.1 and jected to fall 7.6 percent from the 2000 levels. A2.3 for nominal price forecasts for individual This would still leave metals and minerals commodities and indexes). In nominal terms, prices above the 1999 levels. The long-term crude oil prices are expected to decline 11 per- decline in real commodity prices, which has cent in 2001, relative to 2000, and an addi- been observed for many decades, is expected to tional 16 percent in 2002 as OPEC and non- continue. However, these trends will largely be OPEC supplies increase in response to the surge dominated during the decade by the reaction 159 G LOB AL EC ONO MI C PROS PE CT S of prices to recent extremes, which have seen decades, from 2.7 percent per year during the energy prices rise and agricultural prices fall. 1970s to 1.7 percent growth during the 1980s and 0.8 percent growth during the 1990s3, and this has led to nearly stagnant world trade Agriculture since the late 1970s. While consumption and trade have seen slow growth, world grain Food yields have been increasing at 1.4 percent per T he World Bank's index of nominal food year over the last decade, and an even more T prices has declined by one-third since the rapid 1.7 percent when the countries of the recent high in 1997. In real terms, food prices former Soviet Union (FSU) and Eastern Europe are down by more than half since 1980 (see are excluded. The yield increases have been figure A2.1). The decline in real food prices re- rapid enough to meet global demand at declin- flects the combined impact of countries' agri- ing real prices and still allow total world crop- cultural policies, improved technology, and land devoted to grains to fall by 8 percent since changes in demand, which, on balance, have the peak in 1981. Among major grain-exporting caused food supplies to increase faster than countries4, cropland planted to grain has de- food demand and prices to decline relative to clined 21 percent since the peak. Much of this manufactures prices. Despite the price declines, idled cropland will not likely return to grain the FAO's index of world food production in- production, but it represents substantial capac- creased by 20 percent from 1990 to 1999, and ity that could return if prices rise enough to per capita production increased by about 5.5 justify its use. Grain prices are not expected to percent. Our forecast is for real food prices rise in real terms for any sustained period be- to stabilize over the decade following recent cause of continued yield increases, the surplus declines. production capacity in major exporting coun- Grain prices are severely depressed, with tries, and continued moderate demand growth. nominal prices near the lows of the past decade However, prices are projected to increase over and real prices at all-time lows.2 Several fac- the next several years, as prices recover from tors account for current low prices. Consump- current severely depressed levels. This will tion growth has slowed over the last few likely be followed by further price declines be- Figure A2.1 Food Price Index, 1960-2010 index, 1990=100 400 350 300 250 200 Real 150 100 a 50 Nomninal 01 °l III 1960 1970 1980 1990 2000 2010 160 G LO BAL CO MMO DI TY P RICE P R O S P E C T S ginning about mid-decade as production in- sugar prices averaged 17.6 cents per kilogram creases exceed demand growth. in the world market in 2000 compared to an Vegetable oil prices remain depressed fol- average of 24.5 cents per kilogram during the lowing the declines in 1999. Prices of major decade ending in 1998. World production and vegetable oils, such as soy and palm, have de- stocks are expected to fall in 2001, and prices clined by nearly one-half since their 1998 highs, should continue to recover. However, the price while prices of other oils, such as coconut and recovery is expected to take several years, with groundnut, have fallen by about one-quarter prices rising to about 20 cents per kilogram by since their 1999 highs. Unlike most other agri- 2005. Real prices are projected to remain al- cultural commodities, vegetable oil prices in- most unchanged from 2000 to 2010. creased during the Asian crisis, as Indonesia (a major exporter) imposed export taxes on palm Beverages oil in an effort to stabilize domestic prices. After falling sharply in 1998 and 1999, the These taxes were gradually removed starting in index of nominal beverage prices is expected 1999, as the crisis eased, and this caused ex- to increase modestly in 2001 and more rapidly ports to increase and all vegetable oil prices to in 2002 (figure A2.2). The decline in prices fall. Global supplies of vegetable oils are ex- began as the Asian crisis weakened demand pected to increase 5.0 percent in 2000, com- and followed several years of high prices in the pared to the long-term average of 3.5 percent, mid-1990s, which had stimulated global pro- and this could keep prices depressed for at duction. The sharp drop in prices has not yet least another year. Palm oil production has been reversed despite falling beverage stocks grown by 7.5 percent per year over the past and rising imports. Currency devaluations in decade, compared to 5.5 percent for soy oil, the major exporters: Brazil (for coffee), C6te and this growth is expected to continue as d'Ivoire (for cocoa), and Kenya (for tea) con- more Southeast Asian and Latin American pro- tributed to lower U.S. dollar export prices.5 ducers expand palm oil production. Palm oil Weak currencies in major importers, such as could displace soy oil as the dominant oil pro- the European Union and the Russian Federa- duced within five years, and this would con- tion, also weakened import demand. Beverage tribute to long-term weakness in the entire veg- prices have historically been among the most etable oils complex as palm oil use displaces volatile commodity prices, and a supply dis- soy and other oils. Palm oil is already the most ruption in a major producer could quickly re- heavily traded oil, with a 40 percent market verse the recent price declines. However, bar- share, while soy oil is second with a 20 percent ring such an event, prices are expected to be share. The index of nominal vegetable oil prices slow to recover because of new capacity added fell 8.6 percent in calendar year 2000 and is by major exporters. The index of nominal bev- projected to rise 6.0 percent in 2001. By 2005, erage prices is expected to rise 1.5 percent in nominal prices are projected to increase 30.9 2001 and 8.7 percent in 2002. Real prices are percent from 2000 levels. Real prices are pro- expected to increase about 20 percent from jected to rise less than 3 percent between 2000 2000 to 2005 and then decline as productivity and 2010. increases allow supplies to meet demand with Other food prices have been mixed, with falling real prices. beef and shrimp prices strong because of the Cocoa prices reached a three-decade low in rapid global economic growth, while banana February 2000, as production increased 6 per- and citrus prices have remained weak because cent in the 1999 season compared to a decade- of large supplies. Sugar prices have recovered long growth rate of 1.4 percent. Cocoa con- from 1999 lows despite large stocks resulting sumption rose in response to lower prices and from five consecutive seasons when global increased global economic growth, but not production has exceeded consumption. Raw enough to keep stocks from rising 12 percent. 161 G LOB AL EC ON OM IC PRO SP E CT S Figure A2.2 Beverage Price Index, 1960-2010 Index, 1990=100 600 500 400 300 Real 200 100 O I I ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~~~I I II 1960 1970 1980 1990 2000 2010 Prices are expected to begin to recover in 2001 all weakness in all coffee prices. In response as demand increases in major markets accom- to low prices, Brazil and Colombia, the two panying projected strong economic growth. largest arabica producers and dominant mem- By 2002, nominal cocoa prices are projected bers of the Association of Coffee Producing to rise 22 percent from 2000 levels. The longer- Countries (ACPC), agreed to an export reten- term outlook is for real prices to rebound from tion scheme to withhold 4.5 million bags of current severely depressed lows. By 2005, real production from export during 2000 and prices are projected to rise 45 percent from 2001. This could support arabica prices and the lows of 2000 and then remain about un- would be more effective if other ACPC coun- changed at that level, but this would still leave tries joined the scheme. However, this will not real prices at one-third of the 1980 level. One change the longer-term issues of weak demand of the factors that should keep prices from re- growth, excess production capacity, and large turning to previous highs is the 20 percent in- stocks, which have been with the industry for crease in world cocoa planted areas during the many years. Barring a weather-related supply 1990s as low-cost producers such as C6te disruption, prices are expected to be slow to d'Ivoire, Ghana, and Indonesia expanded pro- recover, with arabica prices increasing only duction capacity. 7.4 percent by 2002 and robusta prices in- Coffee prices declined sharply during 1999 creasing 16.2 percent. Real prices are pro- and 2000, with arabica prices down 37 per- jected to rise over the next 10 years (from cur- cent and robusta prices down 48 percent. Over- rent extremely depressed levels), with arabica production, the Brazilian currency devalua- prices up 7 percent by 2010 compared to tion in January 1999, and weak demand in 2000, and robusta prices up 55 percent. Europe and the United States all contributed Tea prices have remained the strongest of to the price declines. Vietnam emerged as the the three major beverages, with a 10 percent largest robusta producer and exporter, and decline in 1999 compared to 1998 and a 2.8 became the second-largest overall coffee ex- percent increase in 2000. The strength was porter, following Brazil. This contributed to largely due to poor weather-related growing the greater decline in robusta prices compared conditions in Kenya, which reduced quality to arabica prices but also contributed to over- exports, and the recovery of demand in coun- 162 G LO BA L C OM MO DI TY PRI C E P R O S P E C T S tries that benefited from increased crude oil response to the Asian crisis. Prices are now set prices. Many of the major oil exporters of the to recover from the lows of 1998 and have in- Middle East as well as the Russian Federation creased about 5 percent during 1999 and 2000 are major tea buyers. The return of Iraq as a (figure A2.3). We project a further increase of tea importer, following the lifting of U.N. 4.2 percent in 2001 and 5 percent in 2002. By sanctions on food imports, also contributed to 2010, real prices are projected to increase 23 the overall price strength. However, supplies percent relative to the 1998 lows, which would are now increasing in major exporters, and still leave the index well below the cyclical nominal prices are not projected to increase highs of the mid-1990s. However, raw materi- significantly over the next several years. Tea als prices are responsive to global economic yields in Sri Lanka, a major exporter, in- conditions and would likely rise further if the creased 48 percent from 1990-92 to 1996-98 global recovery exceeds current forecasts. in response to tea estate privatization in the Cotton prices have remained around 150 early 1990s, which led to increased investment cents per kilogram (nominal) for the past two and improved management of the tea estates. decades, and there is no reason to think this Nominal prices are expected to rise about 11 will change soon. Prices rose 66 percent from percent by 2010 relative to 2000, while real 1993 to 1995, from 128 to 213 cents per kilo- tea prices are expected to fall about 13 per- gram, and then fell back to 117 cents per kilo- cent. There is some prospect that rapid con- gram in 1999. Global consumption rose sharply sumption growth in major producing coun- during the 1980s as clothing fashions favored tries, such as India and China, could offset cotton. However, those trends have changed weak demand in industrial countries and pro- and global consumption stagnated during the vide a firmer price outlook. 1990s. Global production has been very erratic in response to wide swings in prices and policy Agricultural Raw Materials changes in major producers such as China and The index of nominal agricultural raw materi- the United States. Consequently, cotton prices als prices rose by 35 percent during the first have been volatile, but without a clear trend, half of the 1990s, as the global economy since about 1980. Prices have begun to recover boomed, and then fell sharply by 35 percent in from the recent lows, with nominal prices up Figure A2.3 Agricultural Raw Materials Price Index, 1960-2010 Index, 1990=100 250 200 150 100 50 Nominal °l I I I I 1960 1970 1980 1990 2000 2010 163 G LO B AL EC ONO MI C PROS PE CT S about 9.2 percent in 2000 and projected to rise rise over the next several years as tropical tim- about 6.9 percent in 2001. By 2005, nominal ber becomes scarcer, environmental regula- prices are projected to rise to 159 cents per tions become tighter, and demand continues to kilogram, and by 2010, prices are expected to increase. However, real price increases will also reach 181 cents per kilogram. In real terms, be moderated by improved production tech- prices are forecast to rise 24.1 percent relative niques that allow better use of timber. Real to the 1999 lows by 2010. prices of Cameroon log are projected to in- Rubber prices were severely depressed in crease 9.9 percent from 2000 to 2010. 1999 because Indonesia, Malaysia, and Thai- land (which account for 70 percent of rubber exports) devalued their currencies as a result Energy of the Asian crisis. The price of rubber in U.S. rude oil prices have tripled since the lows dollars tumbled to a 24-year low in 1999- of early 1999, to well over $35 per barrel, down 60 percent from the 1995 high. Prices as significant production cutbacks by OPEC have stopped falling, but the recovery has (as well as reductions by Mexico and Norway) been modest as record production, weak de- and strong demand growth, reduced stocks to mand, and high stocks have kept prices near historically low levels. Product stocks, partic- the low reached in 1999. The International ularly gasoline and middle distillate, have also Natural Rubber Organization, which was the been drawn to extremely low levels, and a last U.N.-backed commodity price stabiliza- tight gasoline market is expected to turn into tion body, was formally dissolved in October a tight heating oil market this winter. In addi- 1999 after the withdrawal of key members in tion, steep backwardation of futures prices the wake of the rubber price collapse and cur- (near-term futures prices lower than distant rency devaluations. Buffer stocks held by the futures prices) has discouraged stock building organization (amounting to 2.5 percent of an- above immediate requirements. The U.S. gaso- nual trade) are yet to be liquidated, but they line market has been additionally affected by will eventually find their way into the market. capacity outages, the introduction of Phase II Prices are expected to recover slowly and are reformulated gasoline (RFG), the phaseout of unlikely to reach the highs seen in the mid- methyl tertiary butyl ether (MTBE), and Uno- 1990s. Our near-term forecast is for nominal cal's RFG patent, which makes it more costly prices to rise about 6 percent per year in both to manufacture gasoline. 2001 and 2002, following the 12 percent in- OPEC responded to the dramatic price in- crease in 2000. Real prices are projected to creases by raising production quotas 7.5 per- rise 9.2 percent between 2000 and 2010. cent in April 2000, 2.9 percent in July 2000, Asian tropical timber has been one of the and 3.1 percent in October 2000. But these few commodities that have seen rising real increases were not enough to reduce prices. prices over the past two decades. However, OPEC also introduced a price band mecha- prices fell following the Asian crisis as demand nism for its basket of crudes in mid-2000. If weakened dramatically. Prices of Malaysian the average price of the OPEC reference basket logs have since risen 18 percent from the 1998 exceeds $28 per barrel each day for 20 consec- lows, and the recovery in Asian economies utive trading days, OPEC production, exclud- will likely support further price increases. ing Iraq's, will increase by 0.5 million barrels Malaysian log prices are expected to increase per day. If the average price falls below $22 per 18.6 percent, in real terms, from 2000 to 2010. barrel for 10 consecutive trading days, OPEC African tropical timber is mostly imported into production, excluding Iraq's, will decrease by Europe, and prices did not decline as sharply an additional 0.5 million barrels per day. This as those of Asian timber did following the mechanism was triggered in late October when Asian crisis. The improving growth prospects prices exceeded $28 per barrel for 20 consecu- in Europe suggest prices of African timber could tive trading days, and OPEC announced plans 164 G LOB AL C OM MO D ITY PRI C E P R O S P E C T S to increase production by 0.5 million barrels of Mexico and Brazil give promise of similar per day. Saudi Arabia, the largest OPEC pro- development in many other locales around the ducer, has stated that it would like to see prices world. The costs of non-conventional oil re- settle around $25 per barrel. Iraq remains out- sources, such as oil sands development in side the quota system because of U.N. sanc- Canada, have fallen significantly in recent tions, but its production has risen to nearly 3 years, and new projects have proceeded under million barrels per day. In response to persis- the assumption of low oil prices. Several coun- tent high prices, the United States released 30 tries have invited in, or back, foreign oil com- million barrels from its Strategic Petroleum panies, including some OPEC countries, and Reserve and set up emergency heating oil in- these actions will result in increased produc- ventories in the northeast region. tion capacity. Inventories are now rebuilding, although High oil prices will reduce demand and en- stocks will likely remain low in the near term, courage substitution of other fuels for oil, as depending on demand and the severity of win- occurred when prices spiked during the past ter. OPEC's new price band and accompanying three decades. For example, world oil demand production restraint are designed to stabilize (excluding demand in the FSU) grew by 2.3 oil prices. However, the impact on prices is percent in the 1990s compared with global expected to be short-lived because oil produc- growth of 7.5 percent prior to the first oil price tion costs are substantially below current prices, shock in 1973. Efficiency improvements, how- and advances in technology and improved ever, have slowed significantly in recent years managerial practices continue to result in ever as real prices have declined. This has been no- lower development costs. In addition, the costs table in the United States, with the surge in of competing fuels and non-conventional en- demand for fuel-thirsty sport-utility vehicles ergy sources continue to fall and are becoming (SUVs). In addition, U.S. corporate average fuel increasingly competitive when oil prices are economy (CAFE) standards for conventional high. Non-OPEC oil supplies are expected to automobiles have not been raised since 1990. continue to increase, despite the significantly Thus significant potential exists for improving slower growth in 1998-99, because of low oil efficiency in transport and other uses. Higher and gas prices. Capital expenditures by the pe- prices will also encourage the substitution of troleum industry have been relatively modest, other fuels, notably natural gas, and also of re- despite the rebound in prices, as companies newable energy sources. Environmental pres- grapple with large merger activities, debt pay- sures to reduce local pollution, reduce con- down, share buyback programs, and a cautious gestion, and curb greenhouse gas emissions attitude to the "new" price regime. However, will push policymakers to improve energy effi- major oil companies have had large earnings ciency and restrain consumption of oil and increases, and this could eventually lead to sig- other carbon-based fuels. More ominously for nificant spending programs, which would re- oil producers, the development of transport sult in higher oil production in future years. fuel-cell technology6 looms on the horizon, al- Significant advances in oil development though the costs remain high and a single pre- technology in recent years, such as 3-D com- ferred fuel has not yet been established. puter seismic, horizontal drilling, and floating We expect oil prices to average $28 per bar- production systems, have helped reduce devel- rel in 2000 because of tight underlying market opment and operating costs and shifted supply conditions and OPEC's resolve to keep prices curves outward. New frontiers still remain for within its new price band. Prices are then ex- substantial oil development, for example, off- pected to fall to $25 per barrel in 2001 as shore, deepwater, heavy oil, and the Caspian higher production from both OPEC and non- Sea. Large increases in production from off- OPEC sources allows stocks to rebuild and tilt shore West Africa are expected in the next few the market back into surplus. However, most years, and deepwater advances in the U.S. Gulf OPEC countries are at or near capacity, with 165 G LO B AL EC ONO MI C PROS PE CT S only Saudi Arabia having significant spare ca- crisis in the 1990s, domestic fertilizer con- pacity. This, along with a delayed non-OPEC sumption declined along with grain demand, supply response, could keep OPEC in firm and firms directed their fertilizer production control of the market for an extended period- to the export market. This led to aggressive perhaps several years. Over the longer term, price-cutting and competition for market real oil prices (figure 2.4) are expected to de- share in the nitrogen fertilizer market, espe- cline because of abundant low-cost global sup- cially by the Russian Federation and Ukraine. plies, increasing competition from non-OPEC The competition was less intense in the phos- producers and non-oil fuels, environmental con- phate and potash markets because the FSU cerns, and technological advances. countries had smaller market shares and be- cause other major phosphate and potash pro- ducers responded to increased exports from Fertilizer the FSU by cutting production rather than by T ertilizer prices, like the prices of many lowering prices and competing for market 1 other commodities, have followed very share. Other factors also contributed to the divergent paths over the past several years. different price behavior, including the decision Nitrogen fertilizer prices declined from more by China (the major nitrogen fertilizer im- than $200 a ton to near $60 a ton (for bulk porter) to ban nitrogen imports in 1997. urea), while phosphate fertilizer prices de- Nitrogen fertilizer prices have increased clined only 20 percent (for triple super phos- nearly 45 percent in 2000 compared to 1999 phate, or TSP), and potash fertilizer prices as major producers in Europe and the U.S. cut continued to rise. The differences in price be- production. However, the price recovery is ex- havior were due to the different impact that pected to slow as the industry faces large ex- the economic collapse of the FSU had on fer- cess capacity and continued aggressive export tilizer markets, the different industry market competition. Weak grain prices contribute to structures, and different export firm behavior. weak demand and further delay a significant The FSU was both a major producer and a price recovery, since more than 50 percent of major consumer of fertilizer prior to 1990. nitrogen fertilizer is used for grain production. When these countries faced severe economic Real urea prices are projected to rise 55 per- Figure A2.4 Petroleum Price Index, 1960-2010 Index, 1990=100 250 200 150 100 _ __ 50 Nonminal 01 I l lI 1960 1970 1980 1990 2000 2010 166 G LO B AL CO MM OD IT Y PR ICE PRO SP EC TS Figure A2.5 Metals Price Index, 1960-2010 Index, 1990=100 200 180 160 Real 140 120 100- 80 1. /- Q? - 60 Nominal _ 40 ,- .- -_ 20 01 l I . l I 1960 1970 1980 1990 2000 2010 cent by 2010 compared to 1999 lows, but still MOP prices are projected to increase about 1 remain 30 percent below the highs of 1996. percent per year until 2005 and then remain Phosphate prices fell less, and will likely about unchanged for the balance of the decade. reach new highs sooner, than nitrogen fertil- In real terms, prices will decline, as nominal izer prices. The industry is faced with surplus price increases will not be large enough to off- capacity, but demand has been strong, as set overall inflation. By 2010, real MOP prices many developing countries have increased im- are projected to fall about 19 percent from the ports of phosphate in order to improve the 2000 level. balance of fertilizer applications. After falling 19 percent from 1998 to 2000, TSP prices are projected to increase 7 percent in 2001. Nom- Metals inal prices are expected to increase an addi- he World Bank's nominal index of metals tional 7 percent by 2005 as improvements in T and minerals prices has increased nearly world grain prices boost fertilizer demand. By 27 percent from the lows in early 1999 be- 2010, real prices are expected to decrease as cause of production cutbacks and strong de- new capacity comes onstream, causing real mand growth (figure A2.5). However, abun- prices to fall 5 percent from 2000 levels. dant supplies and high stocks have prevented Potash prices have increased about 5 per- even larger price gains. Much of the increase cent since 1998, while most other commodity in the price index has been a result of the more prices fell. This was possible because of strong than doubling of nickel prices and the 30 to import demand from developing countries and 40 percent increases in aluminum and copper the willingness of major producers to close prices. Other metals prices have failed to move production capacity rather than see prices fall. higher because of abundant supplies and in These industry trends are expected to continue some cases relatively weak demand. Tin prices and should lead to gradually increasing muri- have risen slightly, while gold and silver prices ate of potash (MOP) prices. At some point, are relatively unchanged since early 1998. enough new capacity may be developed to Lead prices have fallen because of weak de- threaten this price stability, but this probably mand and rising stocks. Many metals prices will not occur for several more years. Nominal are poised to increase in the near term in re- 167 G LO B AL EC ONO MI C PRO SP EC TS sponse to demand growth, which accompa- likely that copper will retain its position in ex- nies expected strong economic growth, and isting applications. Real prices are expected to more favorable supply balances. rise about 3 percent between 2000 and 2010. Aluminum prices have recovered from the Gold has traded between $275 and $300 lows of early 1999, but high stocks, rising per troy ounce during most of the past three production, and forward selling by producers years because of central bank sales, declining have kept a lid on price gains. In mid-2000, production costs, and forward selling by pro- the market balance began to tighten because ducers. A number of central banks have been of producer cutbacks in the United States. selling gold reserves to exchange their low- Nearly 40 percent of the U.S. aluminum ca- interest assets for investments that yield higher pacity is located in the Pacific Northwest, interest. The Netherlands, Switzerland, and the where deregulation of the U.S. power industry, United Kingdom are in the midst of large gold along with strong summer demand, has driven sale programs, while other countries are con- up power prices and led to lower aluminum templating such actions. In September 1999, 15 production. Further production cuts or strong European central banks agreed to limit gold demand growth could lead to a period of higher sales to 2,000 tons over the next five years and prices. In the longer term, real prices are ex- restrict their lending activities, and several pected to weaken because of improved tech- producers announced that they would limit or nologies, new lower-cost capacity, and demand- suspend their gold hedging programs. How- side pressures from substitution of low-cost ever, this failed to lift prices. Prices are ex- materials such as plastics. By 2010, real alu- pected to remain under pressure as supplies minum prices are expected to fall by about 5 from all sources will be more than adequate to percent from 2000 levels. meet demand. Price movements above $300 Copper prices have risen about 40 percent per troy ounce will probably face reduced de- from the lows of early 1999 because of strong mand, provide greater incentives for produc- demand and significant reductions in high-cost ers to sell forward, and encourage central banks production-much of it in the United States. to increase sales. Real prices are expected to World copper consumption is expected to grow decline by about 1.7 percent per year between by more than 5 percent in 2000, and far out- 2000 and 2010. stripped global production growth, resulting Nickel prices rose from under $4,000 per in a 50 percent decline in London Metals Ex- ton in December 1998 to more than $10,000 change (LME) inventories. As a result, the per ton during 2000. Various supply problems market balance has moved into deficit, follow- contributed to the tight market, particularly ing large surpluses during the 1997-99. Further technical problems bringing on new capacity shortfalls are anticipated, causing nominal in Australia and labor strikes in Canada. prices to rise by 8.2 percent in 2001 and an ad- Nickel demand has also been very strong be- ditional 3.8 percent in 2002. However, higher cause of the strength of the global economic prices will bring forth investment in new capac- recovery and large growth in steel production. ity, along with reactivation of idle plants, which This has depleted stocks, causing LME inven- will prevent escalation of real prices over the tories to fall to the lowest level in nine years. forecast period. Prices will remain cyclical, with Prices are expected to fall as nickel production the cost structure of the industry essentially de- increases substantially in the coming years and termining the low point in the cycle. New tech- large amounts of scrap metal are brought to nologies will continue to reduce production market. The supply deficit is expected to di- costs, leading to declining real prices later in the minish in 2001, and the market is expected to forecast period. On the demand side, new cop- be in better balance going forward. Real prices per alloys could regain some market share pre- are expected to decline by nearly 40 percent viously lost to aluminum. Although the threat between 2000 and 2010, mainly reflecting the of substitution from new materials exists, it is lofty level of prices in 2000. 168 G LO B AL CO MM OD I TY PR ICE PRO SP EC TS Table A2.1 Commodity prices and price projections in current dollars Actual lrojections Commodity Unit 1970 1980 1990 1998 1999 2000 2001 2002 2005 2010 Energy Coal, U.S. $lMt - 43.10 41.67 34.38 33.17 33.00 33.00 33.50 35.00 37.30 Crude oil, avg, spot SVbbl 1.21 36.87 22.88 13.07 18.07 28.00 25.00 21.00 18.00 19.00 Natural gas, Europe $/mrmbtu - 3.40 2.55 2.42 2.13 3.80 3.75 3.20 2.75 2.75 Natural gas, U.S. mrmmbtu 0.17 1.55 1.70 2.09 2.27 4.00 4.00 3.50 2.75 3.00 Non-Energy Commodities Agriculture Beverages Cocoa c)kg 67.5 260.4 126.7 167.6 113.5 90.0 95.0 110.0 150.0 170.0 Coffee, other milds c/kg 114.7 346.6 197.2 298.1 229.1 195.0 195.0 209.4 253.5 265.0 Coffee, robusta c/kg 91.4 324.3 118.2 182.3 148.9 94.8 97.0 110.2 149.9 187.4 Tea, auctions (3) average cekg 83.5 165.9 205.8 204.6 183.9 189.0 192.0 192.0 195.0 210.0 Food Fats and oils Coconut oil Vimt 397.2 673.8 336.5 657.9 737.1 444.0 500.0 540.0 620.0 650.0 Copra S/mr 224.8 452.7 230.7 411.1 461.5 310.0 425.0 435.0 460.0 483.0 Groundnut oil lImt 378.6 858.8 963.7 909.4 787.7 700.0 740.0 775.0 820.0 850.0 Palm oil $/mnt 260.1 583.7 289.8 671.1 436.0 322.0 340.0 360.0 400.0 450.0 Soybean meal $/mt 102.6 262.4 200.2 170.3 152.2 185.0 195.0 200.0 215.0 226.0 Soybean oil $lmt 286.3 597.6 447.3 625.9 427.3 340.0 360.0 380.0 430.0 460.0 Soybeans $1mt 116.9 296.2 246.8 243.3 201.67 210.0 220.0 230.0 250.0 270.0 Grains Maize $/mt 58.4 125.3 109.3 102.0 90.2 86.0 95.0 110.0 125.0 130.0 Rice, Thai, 5% $1mt 126.3 410.7 270.9 304.2 248.4 202.0 215.0 235.0 275.0 300.0 Sorghum $/mt 51.8 128.9 103.9 98.0 84.4 85.0 88.0 100.0 120.0 125.0 Wheat, U.S., HRW $/mt 54.9 172.7 135.5 126.1 112.0 112.0 120.0 130.0 160.0 170.0 Other food Bananas, U.S., new series S/mt 166.1 377.3 540.9 489.5 373.8 430.5 465.2 490.5 529.1 567.7 Beef, U.S. clg 130.4 276.0 256.3 172.6 184.3 194.0 198.4 202.8 209.4 225.0 Oranges $/mt 168.0 400.2 531.1 442.4 438.2 365.0 400.0 500.0 565.0 600.0 Shrimp, Mexican c/kg - 1,152 1,069 1,579 1,461 1,503 1,515 1,530 1,550 1,590 Sugar, world dckg 8.2 63.16 27.67 19.67 13.81 17.84 18.10 18.10 20.00 24.00 Agricultural raw materials Timber Logs, Cameroon $/cum 43.0 251.7 343.5 286.4 269.3 275.0 285.0 300.0 330.0 385.0 Logs, Malaysia SlCum 43.1 195.5 177.2 162.4 187.1 192.0 198.0 210.0 245.0 290.0 Sawnwood, Malaysia $/Cum 175.0 396.0 533.0 484.2 600.8 600.0 620.0 655.0 750.0 900.0 Other raw materials Cotton c/kg 67.6 206.2 181.9 144.5 117.1 127.9 136.7 141.1 158.7 180.8 Rubber, RSS1, Malaysia c/kg 40.7 142.5 86.5 72.2 62.9 70.6 75.0 79.4 88.2 99.2 Tobacco $/mt 1,076 2,276 3,392 3,336 3,041 2,985 3,000 3,100 3,250 3,300 Fertilizers DAP $/Mt 54.0 222.2 171.4 203.4 177.8 155.0 165.0 175.0 195.0 205.0 Phosphate rock $/Mt 11.00 46.71 40.50 43.00 44.00 43.80 44.00 44.00 44.00 46.00 Potassium chloride $/mt 32.0 115.7 98.1 116.9 121.6 122.5 124.0 124.0 125.0 127.0 TSP $1/mt 43.0 180.3 131.8 173.1 154.5 140.0 150.0 155.0 160.0 170.0 Urea, E. Europe, bagged S/mt 48.0 222.1 130.7 103.1 77.8 112.0 120.0 130.0 140.0 150.0 Metals and minerals Aluminum $Snt 556 1,456 1,639 1,357 1,361 1,575 1,600 1,650 1,800 1,900 Copper s/mt 1,416 2,182 2,661 1,654 1,573 1,825 1,975 2,050 2,200 2,400 Gold $/toz 36.0 607.9 383.5 294.2 278.8 280.0 280.0 275.0 275.0 300.0 Iron ore, Carajas dd/mtu 9.84 28.09 32.50 31.00 27.59 29.00 29.50 30.25 32.00 33.00 Lead c/kg 30.3 90.6 81.1 52.9 50.3 46.0 50.0 55.0 60.0 64.0 Nickel $/mt 2,846 6,519 8,864 4,630 6,011 8,600 7,500 7,000 6,000 6,800 Silver cItoe 177.0 2,064 482.0 553.4 525.0 505.0 500.0 510.0 525.0 550.0 Tin c/kg 367.3 1,677 608.5 554.0 540.4 545.0 550.0 560.0 590.0 610.0 Zinc c/kg 29.6 76.1 151.4 102.5 107.6 114.0 116.0 117.0 120.0 125.0 - Not available. $/mt, dollars per metric ton; $/bbl, dollars per barrel; $/mmbtu, dollars per million British thermal units; c/kg, cents per kilogram; $/cum, dollars per cubic meter; $/toz, dollars per troy ounce; cidmtu, cents per dry metric ton unit of iron (fe). Note: Projections as of November 14, 2000. Source: World Bank, Development Economics, Development Prospects Group. 169 G LO B AL EC ONO MI C PRO SP EC TS Table A2.2 Commodity prices and price projections in constant 1990 dollars Actual Proleetions Commodity ni 1970 1980 1990 1998 1999 2000 2001 2002 2005 2010 Energy Coal, U.S. S/mt - 59.86 41.67 32.40 32.10 32.70 31.57 30.91 30.13 29.18 Crude oil, avg, spot $Jb 4.82 51.21 22.88 12.31 17.49 27.74 23.91 19.38 15.49 14.78 Natural gas, Europe S/mmbtu - 4.72 2.55 2.28 2.06 3.76 3.59 2.96 2.37 2.14 Natural gas, U.S. 5/;ubro 0.68 2.15 1.70 1.97 2.19 3.9 3.83 3.23 2.37 2.33 Non-energy commodities Agriculture Beverages Cocoa c/kg 268.9 361.6 126.7 157.9 109.9 89.2 90.9 101.5 129.1 132.3 Coffee, other milds /k 456.8 481.4 197.2 280.9 221.7 193.2 186.5 193.3 218.2 206.2 Coffee, robusta c/kg 364.0 450.5 118.2 171.7 144.1 939 92.8 101.7 129.0 14S.8 Tea, auctions (3) average c/k 332.7 230.5 205.8 192.8 178.0 11873 183.7 177.2 167.8 163.4 Food Fats and oils Coconut oil S/mt 1582.4 935.9 336.5 619.9 713.5 439.9 473 498.3 533.7 505.7 Copra m 895.8 628.8 230.7 387.3 446.7 307 406.5 404 395.9 375.8 Groundnut oil Sm 1508.2 1192.7 963.7 856.8 762.4 693.6 707.9 715.1 705.8 661.3 Palm oil Sm 1036.0 810.7 289.8 632.3 422.0 3190 325.2 332.2 3443 350.1 Soybean meal S/mt 408.7 364.5 200.2 160.5 147.3 183.3 186.5 184.5 18. 175.8 Soybean oil S/mr 1140.8 830.0 447.3 589.7 413.6 336.9 4 35.6 37.1 357.9 Soybeans S/mt 465.8 411.4 246.8 229.2 195.2 08.1 20.5 212.2 215.2 210.1 Grains Maize /m 232.7 174.0 109.3 96.1 87.3 82 9. 101.5 107.6 101.1 Rice, Thai, 5% Sm 503.2 570.5 270.9 286.6 240.5 00.1 205.7 216.8 236.7 233.4 Sorghum S/mr 206.4 179.0 103.9 92.4 81.7 84.2 84.2 92.3 103.3 97.3 Wheat, U.S., HRW S/nse 218.7 239.9 135.5 118.8 108.5 1110 114.8 120.0 137.7 132.3 Other food Bananas s/ 661.7 524.0 540.9 461.2 361.9 46 450 452.6 455.4 441.7 Beef, U.S. c/kg 519.6 383.3 256.3 162.6 178.4 192.2 1i8 9. 187.1 180.2 175.1 Oranges m 669.5 555.8 531.1 416.8 424.2 361.6 382.6 4613 486.3 466.8 Shrimp, Mexican c/kg .. 1,600 1,069 1,488 1,414 1,48 1 ,,449 1,412 1,334 1,237 Sugar, world / 32.8 87.7 27.7 18.5 13.4 1. 17.3 16.7 17.2 18.7 Agricultural raw materials Timber Logs, Cameroon 171.3 349.6 343.5 269.8 260.7 272.5 27.6 276.8 284.0 299.5 Logs, Malaysia 5/Cue 171.8 271.6 177.2 153.0 181.1 0. 189. 193.8 210.9 225.6 Sawnwood, Malaysia S/oi 697.2 550.0 533.0 456.1 581.6 594.5 593.1 604.4 645.6 700.2 Other raw materials Cotton c/kg 269.4 286.4 181.9 136.1 113.4 12.6.7 130.8 130.2 136.6 140.7 Rubber, RSS1, Malaysia c/k 162.2 197.9 86.5 68.0 60.8 69.9 71.7 73.2 75.9 77.2 Tobacco S/mt 4,287 3,161 3,392 3,143 2,944 2,5 2,80 A2,860 2,797 2,567 Fertilizers DAP m 215.1 308.6 171.4 191.7 172.1 1 1 61 1 159.5 Phosphate rock S/ 43.8 64.9 40.5 40.5 42.6 43.4 42. 0.6 379 35.8 Potassium chloride S/Is 127.5 160.7 98.1 110.1 117.8 98.8 TSP S/e171.3 250.4 131.8 163.0 149.5 187 435 4.0 3.7 132.3 Urea, E. Europe, bagged S 191.2 308.5 130.7 97.1 75.3 1. 148 21 105 116.7 Metals and minerals Aluminum S/mt 2,215 2,022 1,639 1,279 1,317 1,560 1, 1,478 Copper S/5 5,640 3,031 2,661 1,558 1,522 ,08 1,889 1,891 1 1,894 1,867 Gold S/a 143.2 844.3 383.5 277.1 269.8 2 2 253.7 2367 233.4 Iron ore c :mts 39.2 39.0 32.5 29.2 26.7 8 28.2 279 27.5 25.7 Lead ck 120.7 125.8 81.1 49.8 48.7 45.6 447 SO.8 51.6 49.8 Nickel S/m 11,339 9,054 8,864 4,362 5,819 8,521 7,174 6,459 5,164 5,291 Silver crt 705.2 2866.1 482.0 521.4 508.1 500.4478.3 0.6 45.9 427.9 Tin c/kg 1463.5 2329.8 608.5 522.0 523.1 540. 052611 516.7 507.8 474.6 Zinc c/leg 117.9 105.7 151.4 96.5 104.2 130 111.0 108.0 103.3 97.3 - Not available. $/mr, dollars per metric ton; $/bbl, dollars per barrel; S/mmbtu, dollars per million British thermal units; c/kg, cents per kilogram; $/cum, dollars per cubic meter; $/toz, dollars per troy ounce; c/dmtu, cents per dry metric ton unit of iron (fe). Note: Projections as of November 14, 2000. Source: World Bank, Development Economics, Development Prospects Group. 170 G LO B AL CO MM OD I TY PR ICE PROS PE CT S Table A2.3 Weighted indexes of commodity prices and inflation Acul Projections' Index (1990 = 100) 1970 1980 1990 1998 1999 2000 2001 2002 2005 2010 Current dollars Petroleum 5.3 161.2 100.0 57.1 79.0 122.4 109.3 91.8 78.7 83.0 Non-energy commoditiesb 43.8 125.5 100.0 A.1 88.0 87.3 90.3 94.8 105.7 115.6 Agriculture 45,8 138.1 100.0 107.8 92.8 88.0 91.4 96.9 110.6 122.3 Beverages 56.9 181.4 100.0 140.6 107.7 89.5 90.8 98.6 121.7 132.9 Food 46.7 139.3 100.0 104.9 87.6 84.2 88.5 93.2 102.3 110.0 Fats and oils 64.4 148.7 100.0 132.8 1050 96.0 101.8 106.4 116.8 125.7 Grains 46.7 134.3 100.0 101.3 #6.4 78.3 84.1 93.2 110.1 117.4 Other food 32.2 134.3 100.0 84.1 74.0 77.9 80.0 82.4 86.2 93.0 Raw materials 36.4 104.6 100.0 87.3 88.5 91.8 95.7 100.5 113.0 130.3 Timber 31.8 79.0 100.0 90.9 111.8 112.0 115.7 122.3 140.4 168.2 Other raw materials 39.6 122.0 100.0 84.8 72.7 78.1 82.1 85.6 94.3 104.4 Fertilizers 30.4 128.9 100.0 122. 114.1 106.9 111.9 114.3 116.7 123.3 Metals and minerals 40.4 94.2 100.0 75.5 73.7 83.8 85.6 87.6 92.7 98.6 Constant 1990 dollars' Petroleum 21.1 223.8 100.0 53.8 76.5 121.3 104.5 84.7 67.7 64.6 Non-energy commodities 174.7 174.3 100.0 93.4 85.2 86.5 86.4 87.4 91.0 90.0 Agriculture 182.4 191.8 100.0 101.6 89.8 87.2 87.5 89.4 95.2 95.1 Beverages 226.6 252.0 100.0 132.4 104.2 88.6 86.8 91.0 104.8 103.4 Food 186.0 193.4 100.0 98.9 84.8 83.4 84.6 86.0 88.1 85.6 Fats and oils 256.4 2065 100.0 125.2 101.7 95.1 97.4 98.2 100.5 97.8 Grains 186.1 186.5 100.0 95.4 83.6 77.5 80.4 86.0 94.8 91.4 Other food 128.4 186.6 100.0 79.3 71.6 77.1 76.5 76.0 74.2 72.3 Raw materials 145.1 145.2 100.0 82.3 85.7 91.0 91.6 92.7 97.3 101.4 Timber 126.6 109.7 100.0 85.7 108.2 111.0 110.7 112.8 120.8 130.8 Other raw materials 157.7 169A 100.0 79,9 70.3 77.3 78.5 78.9 81.2 81.3 Fertilizers 121.1 179.0 10.0 115.0 110.4 105.9 107.0 105.5 100.5 96.0 Metals and minerals 160.8 130.8 100.0 71.i 71.3 83.0 81.9 80.8 79.7 76.7 Inflation indexes, 1990 = lood MUV index' 25.10 72.00 100.00 106.14 103.31 100.93 104.54 108.38 116.18 128.53 Percentage of change per year 11.11 3.34 0.75 -2.67 -2.30 3.58 3.68 2.35 2.04 U.S. GDP deflator 33.59 65.93 100.00 119.32 121.11 123.89 126.87 129.91 138.54 152.96 Percentage of change per year 6.98 4.25 2.23 1.50 2.30 2.40 2.40 2.17 2.00 aCommodity price projections as of November 14, 2000. 'The World Bank primary commodity price indexes are computed based on 1987-89 export values in U.S. dollars for low- and middle-income economies, rebased to 1990. Weights for the subgroup indexes expressed as ratios to the non-energy index are as follows in percent: agriculture 69.1, fertilizers 2.7, metals and minerals 28.2, beverages 16.9, food 29.4, raw materials 22.8, fats and oils 10.1, grains 6.9, other food 12.4, timber 9.3, and other raw materials 13.6. cComputed from unrounded data and deflated by the MUV index. dlnflation indexes for 2000-10 are projections as of November 3, 2000. MWV for 1999 is an estimate. Growth rates for years 1980, 1990, 1998, 2005, and 2010 refer to compound annual rate of change between adjacent endpoint years; all others are annual growth rates from the previous year. 'Unit value index in U.S. dollar terms of manufactures exported from the G-5 countries (France, Germany, Japan, the United Kingdom, and the United States) weighted proportionally to the countries' exports to the developing countries. 171 G LO B AL EC ONO MI C PR OS PE CT S Description of Price Series in Commodity Price Tables Aluminum (LME) London Metal DAP (diammonium phosphate), bulk, Rubber (Malaysian), RSS 1, f.o.b. Kuala Exchange, unalloyed primary ingots, f.o.b. U.S. Gulf. Lumpur. high grade, cash price. Gold (U.K.), London afternoon fixing. Sawnwood (Malaysian), dark red Bananas (Central and South American), Groundnut oil (any origin), c.i.f. seraya/meranti, select and better import price, free on truck (f.o.t.) Rotterdam. quality, kiln dry, cost and freight U.K. U.S. Gulf. Iron ore (Brazilian), Companhia Vale do Silver (Handy and Harman), refined, Beef (Australian/New Zealand); frozen Rio Doce (CVRD) Carajas fines, New York. boneless; 85 percent chemical lean; contract price to Europe, f.o.b. Ponta Sorghum (U.S.), no. 2 milo yellow, f.o.b. cost, insurance, and freight (c.i.f.) da Madeira. Gulf. U.S. East Coast. Lead (LME), refined, settlement price. Soybean meal (any origin), c.i.f. Coal (U.S.) thermal, free on board Logs (West African), sapele, high quality Rorterdam. (f.o.b.) Hampton Roads/Norfolk. (Loyal and Marchand LM), f.o.b. Soybean oil (Dutch), crude, f.o.b. Cocoa (ICCO), International Cocoa Cameroon. ex-mill. Organization daily price. Logs (Malaysian), meranti, Sarawak, Soybeans (U.S.), c.i.f. Rotterdam. Coconut oil (Philippines/Indonesian), Tokyo import price. Sugar (world), International Sugar bulk, c.i.f. Rotterdam. Maize (U.S.), no. 2, yellow, f.o.b. U.S. Agreement daily price, raw, f.o.b. Coffee (ICO), International Coffee Gulf ports. Caribbean ports. Organization indicator price, other Natural Gas (Europe), import border Tea, average of quotations at Calcutta, mild Arabicas, average New York price. Colombo, and Mombasa/Nairobi. and Bremen/Hamburg markets. Natural Gas (U.S.), Henry Hub, Tin (LME), refined, settlement price. Coffee (ICO), International Coffee Louisiana. TSP (triple super-phosphate), bulk, f.o.b. Organization indicator price, Nickel (LME), cathodes. U.S. Gulf. Robustas, average New York and Le Oranges (Mediterranean exporters), EEC Urea, (varying origins), bagged, f.o.b. Havre/Marseilles markets. indicative import price, c.i.f. Paris. Eastern Europe. Copper (LME), grade A, cathodes and Palm oil (Malaysian), bulk, c.i.f. N. W Wheat (U.S.), no. 1, hard red winter, wire bar. Europe. export Gulf. Copra (Philippines/Indonesian), bulk, Phosphate rock (Moroccan), 70 percent Zinc (LME), special high grade, c.i.f. N.W. Europe. BPL, contract, free alongside ship settlement price Cotton ("Cotton Outlook A Index"), (f.a.s.) Casablanca. c.i.f. Northern Europe. Potassium chloride, f.o.b. Vancouver. Crude oil, average spot price of Brent, Rice (Thai), 5 percent broken, white rice, Dubai, and West Texas Intermediate, milled, indicative survey price, f.o.b. equally weighed. Bangkok. 172 G LO BAL COMM OD I TY P RICE P R O S P E C T S Notes 3. However, the growth during the 1990s was re- duced by a 40 percent decline in grain consumption in 1. Real prices are obtained by deflating nominal the FSU countries and smaller declines in Eastern Eu- prices by the unit value index in U.S. dollar terms of rope. When these countries are excluded, world grain manufactures (MUTV) exported from the G-5 countries consumption grew by 2.0 percent per year during the (France, Germany, Japan, the United KingdOm, and the 1990s. Growth rates in China and India, with 46 per- (Franited Statermany, Japan,hthed pr rteKiona tom the o cent of developing-country populations, has been 1.9 United States) weighted proportionally to the coun- and 1.5 percent, respectively, during the 1990s. tries' exports to the developing countries. 4. The five largest grain exporters are Argentina, 2. Grains account for 55 percent of the world's Australia, Canada, the European Union, and the United food supplies (calories) and occupy nearly one-half of States. Together, these entities account for about 85 the world's cultivated cropland (FAO). Grains prices percent of world exports. are important as an indicator of overall food prices be- 5. For example, the Brazilian real depreciated 68 cause of the close substitutability of grains with other percent from 1997 to 1999, the CFA franc depreciated food crops in production and consumption. Sugar and 9 percent, and the Kenyan shilling depreciated 16 per- vegetable oils account for about 10 percent each of the cent (IFS, August 2000) world's total calorie supplies while animal products 6. Fuel cells convert energy stored in a fuel directly and fish account for about 16 percent. The remaining into electricity and heat without combustion. Using hy- roughly 10 percent of world food supplies come from drogen as fuel, they emit only water and heat as waste fruits, nuts, pulses, roots, tubers, and vegetables. products. 173 Appendix 3 Global Economic Indicators G LO B AL EC ONO MI C PRO SP EC TS Table A3.1 Growth of real GDP, 1971-2010 (GDP in 1995 prices and exchange rates-average annual percentage growth) 1999GDP US. do)Eattr) 1971-S0 1981-90 1991-99 I 0 World 20t.,2 3.6 3.0 2A 2.8 4 3.4 High-income economies 23,665 3.3 3.1 2.2 2.7 3.8 3.0 Industrial 23,135 3.2 3.0 2.1 2.7 3.7 2.9 G-7 f9,895 3.3 3.1 2.1 2.5 3.6 2.8 United States 8,710 2.8 3.2 3.0 42 35.1 3.3 Japan 4,3S5 4.5 4.0 1.3 0.3 2.0 2.2 G-4 Europe 6,050 2.9 2.3 1.6 8.9 3.1 2.8 Germany' 2,080 2.7 2.3 1.7 1 3.1 2.8 Euro Area 6,375 3.2 2.4 1.8 2.4 3.4 3.0 Other industrial 3,240 3.1 2.6 2.4 :3. : 40 3.2 Other high-income 735 7.9 5.5 5.3 4.2 6.3 5.3 Asian NIEs 530 9.5 7.4 5.9 :4. 6.9 5.9 Low- and middle-income economies 6,555 5.3 2.7 3.2 3.2 5.3 5.0 Excluding Eastern Europe and CIS 5,75S 5.5 3.4 4.7 3.3 5.3 5.2 Asia 2,490 5.4 6.8 6.7 6.6 6.9 6.1 East Asia and Pacific ,8 6.6 7.2 7.1 6.9 7.2 6.3 China : 90 5.3 9.1 10.4 . 7.5 Korea, Rep. of 4 7.6 7.7 5.8 0 8.7 Indonesia 145 7.9 5.6 4.1 :03t 44 South Asia 595 3.1 5.8 5.4 5.7 6.0 4 India 445 3.0 5.9 5.7 : 6.4 Latin America and the Caribbean 2,OSS 5.9 1.1 3.2 0.1 4.0 4.3 Brazil 750 8.5 1.5 2.4 1.0 4.1 Mexico 485 6.7 1.9 3.1 3.7 6.8 Argentina 280 3.0 -1.5 5.1 3.1 0.8 Europe and Central Asia 1,095 4.9 1.2 -2.3 8.0 5.2 4.2 Russian Federationb 400 4.8 -0.3 -5.8 3.2 7.2 Turkey 5is 4.2 4.0 3.7 -.5.8 6.2 Poland 155 5.0 -0.3 3.9 4 :4.1 4.4 Middle East and North Africa 505 6.6 2.4 3.1 2.4 3.1 3.6 Saudi Arabia 140 10.3 0.4 2.1 0.4 2.9 Iran, Islamic Rep. 110 1.8 2.7 3.8 2.5 2 . Egypt, Arab Rep. 90 6.6 5.5 4.3 590 5,3 Sub-Saharan Africa 330 3.3 1.8 2.0 2.3 2.7 3.6 Republic of South Africa 130 3.5 1.3 1.4 1. 2 _2. Nigeria 35 4.7 1.1 2.5 4.1 l31 a. Data prior to 1991 covers West Germany. b. Data prior to 1992 covers the former Soviet Union. Source: World Bank data and staff estimates. Figure A3.1 Real GDP growth. 1991-2010 Pawn,en 0.0 -2._ -4.0 qiSh-i amc ECat Aaa Souat Asi Lat, Amenca E&cpe am MiddleEnat Sb-Sahinsn eoomise arts Pacitic and the Central Asia anm Narth Atrica Camrihsaan AtriGa 176 G LO B AL EC ONO MI C IND IC AT OR S Table A3.2 Growth of real per capita GDP, 1971-2010 (GDP in 1995 prices and exchange rates-average annual percentage growth) 1999C . per r.*a U5. d.la) 1971-80 1981-90 1991-99 19f l0o 2loo-lo World 5,oss 1.7 1.3 1.0 I'5 2.7 2.3 High-income economies 26,560 2.4 2.4 1.6 2.2 3.2 2.7 Industrial 27445 2.4 2.4 1.5 2.2 3.2: 2.6 G-7 22,920 2.5 2.5 1.5 2.0 3.1L 2.5 United States 31,9S 1.8 2.2 2.0 3.2 4, 2.5 Japan .34,7245 3.3 3.4 1.0 .0. 1.9: ..3 G-4 Europe 23,465 2.6 2.1 1.3 18 2.9 2.8 Germanva 25,370 2.6 2.1 1.4 - 1.4 3.1 3.0 Euro Area 21,.86 2.7 2.1 1.5 2.2 3.3 3.0 Other industrial 21,320 2.3 2.0 1.9 3.1 3.7 3.0 Other high-income 17,165 5.2 3.7 3.6 2.6 4.6 4.1 Asian NIEs 16,56 7.2 5.9 4.6 3.6 5.6 5.0 Low- and middle-income economies 1,290 3.2 0.8 1.6 L7 3.7 3.7 Excluding Eastern Europe and CIS 1,230 3.2 1.3 3.0 1.7 3.6 3.7 Asia 810 3.2 4.9 5.1 5:2 5.4 4.9 East Asia and Pacific 1,030 4.5 5.6 5.8 5 8 6. L 5.4 China 790 3.4 7.5 9.2 6.1 6.6 - Korea, Rep. of S,68s 5.7 6.4 4.7 9.7 7.8 Indonesia 690 5.4 3.7 2.4 .11 2.7 - South Asia 45S0 0.6 3.5 3.5 3.8 4.0 3.9 India 450 0.7 3.6 3.8 4.3 4. Lann America and the Caribbean 4,035 3.4 -0.9 1.5 1 5 2.4 3.0 Brazil 4,470 5.9 -0.4 1.0 .-03 2.8 MN Lexico .4,65 3.6 -0.2 1.3 2... 51 Argentina 7,745 1.3 -2.9 3.8 -4.3 -0.4 Europe and Central Asia 2,300 3.9 0.4 -2.5 0.9 5. 4.1 Russian Federationb 2),740 4.2 -0.9 -5.7 3.5 7.5 Turkev 2,85 1.8 1.6 2.2 :±6S 4.6 Poland 3,925, 4.1 -1.0 3.7 4.0 4.3 NMiddle East and North Africa 1,975 3.6 -0.6 0.9 0.3 1.1 1.7 Saudi Arabia 6,505 5.1 -4.8 -1.3 -2.8 -0.4 Iran, Islamic Rep. 1,760 -1.4 -0.7 2.2 05 0i.. Egypt, Arab Rep. 1,430 4.4 2.9 2.3 4.2 3.6 - Sub-Saharan Africa 515 0.5 -1.2 -0.6 . 0.2 1,3 Republic of South Africa 3,11S 1.2 -1.2 -0.6 415 0.5 Nigeria Zts 1.7 -1.9 -0.4 . I 0S 4.7 a. Data prior to 1991 covers West Germanv. b. Data prior to 1992 covers the former Soviet Union. Source: World Bank data and staff estimates. Figure A3.2 Real per capita GDP growth, 1991-2010 Percent S.-t LE 200-_10 40 2.0 55 -2 0 -4-0 High-income East Asia South Asia Lati Ameoca Eurape a-d Middle East Sub-Sahanan coanomies and Pacific and tha Central Asia and North Airica Car,,bean Atrica 177 G LO B AL EC ONO MI C PRO SP EC TS Table A3.3 Inflation: GDP deflators, 1971-2010 (percentage changea) Estimate Forecast 197140 19&I-9ft 1991-99 1999 2000 2000-10 World .2.9 2.8 2.5 High-income economies 7 52.0 2.1 2.0 Industrial 72 58 ' .62.0 1.9 2.0 G-7 7l E 5 2.0 3.8 2.0 United Statres 5.7 65 2.3 1.5 2.1 2.1 Japan i 34iS-0.9 -1.5 0.3 G-4 Europe 6. . 12.0 1.8 1.7 Germany 0.9 0.0 1.2 Euro Area 7 3 1.2 1.2 1.7 Other indusrrial F e d e 6a t ion3.1 2.0 2.1 Ocher high-income 7o7 6.2 4.9 4.2 5.0 3.8 Asian NIEs 67.14V2.0 3.0 2.9 Low- and middle-incomne economies 6.7 5.8 4.8 Excluding Eastern Europe and CIS 16.6.2 5.2 4.7 Asia prio 71 1996.5 4.9 5.2 East Asia and Pacific 1.a741.4 3.3 5.0 China Z2 5671-2.2 2.1.. Korea, Rep. of 23. 7l1 196-20.6 0.4. Indonesia 229 891317.2 3.3.. South Asia 122L 88.8 5.4 5.7 India 9. A885.5 5.4.. Larin America and the Caribbean 43 24010'6.5 8.2 7.0 Brazil 4$ 629 3729.0 8.1.. Mexico 15.9 12.0 ... Argentina 1.7 7536 -2.0 0.9.. Europe and Central Asia 7.5 11.4 5.3 Russian Federarion'c03 1, 4, 56.3 43.3.. Turkeyv23 472 7. 52.2 53.7.. Poland 6. 0. 586.9 7.8.. Middle East and North Africa 138)488.7 5.5 3.6 Saudi Arabia 26 271912.8 5.5.. Iran, Islamic Rep. 209 1. 5211.8 14.1.. Egypt, Arab Rep. 1. 33945.6 6.13. Sub-Saharan Africa 104 9.3 9.4 5.1 4.4 4.5 Republic of South Africa 6.9 6.4 Nigeria140 17 347' 1199. Note: Deflators are in local currency units: 1995=100. a. High-income group inflation rates are GDP-weighted averages of local currency inflation. Low- and mniddle-income groups are medians. World is GOP-weighted average of the two groups. b. Data prior to 1991 covers West Germanyv c. Data prior to 1992 covers the foirmer Soviet Union. Source: World Bank data and staff estimates. Figure A3.3 GOP inflation, 1991-2010 Percent 25,0 ~ ~ ~ U2thi 15.0 to.e H7y-droe- East Asia Sc-th Asia Lati A-merica . crpe and Mididle East Sub-Sahar- ece-ies and Pacific arc the Ce-tral Asia and North Africa Carrbbean Air no 178 G LO BAL EC ONO MI C IND I CATO RS Table A3.4 Current account balances, 1970-2010 (percentage of GDP) 1999 currmt (biio-s of Estimate FPecast U.S. dolarS) 1970-80 1981-90 1991-99 1999 2000 2000-10 World -130U 0.1 -0.4 -0.1 -0.5 -0.8 -0A High-income economies -152.4 0.1 -0.2 0.1 -0.8 -1.1 -0.S Industrial -194.1 0.0 -0.5 -0.1 -1.0 -1.3 4.6 G-7 -220.4 0.2 -0.4 -0.1 -LI -1.5 -1.0 United States -331.5 0.0 -2.0 -1.7 -3.7 -4.6 -3,6 Japan 106.9 0.7 2.4 2.5 2.5 2.5 2.2 G-4 Europe 64 0.2 0.3 0.1 0J1 0.0 0.4 Germanva -19.3 0.5 2.6 -0.7 -1.0 -1.0 -0A Euro Area 21.8 0.0 0.4 0.5 0.5 0.7 1.1 Other industrial 26.3 -0.9 -0.9 0.4 -0.2 0.4 1.4 Other high-income 41.7 7.2 9.8 3.9 5.9 4.3 2.6 Asian NIEs 37.6 0.1 6.8 5.0 7.2 4.8 3.3 Low- and middle-income economies 22.3 0.0 -0.8 -1.1 0.3 0.2 -0.2 Excluding Eastern Europe and CIS 22.2 -0.6 -2.1 -1.5 0.2 -0.3 40.3 Asia 73.3 -1.1 -1.4 -0.3 3.0 1.3 1.0 East Asia and Pacific 79.0 -1.4 -1.1 0.1 4.1 3.3 2.1 China 15.7 -0.4 0.1 1.5 1.6 04 . Korea, Rep. of 24.5 -6.1 0.7 0.7 6.1 2.6 Indonesia 5.8 -1.4 -3.1 -1.0 4.1 4.5 ... South Asia -5.7 -0.S -2.0 -1.7 -1.6 -2.6 -3.0 India -2.8 0.3 -1.7 -0.2 0.6 -2.0 Latin America and the Caribbean -55.6 -2.6 -1.8 -2.5 -2.9 -2.8 -2.0 Brazil -25.1 -4.1 -1.6 -1.8 -4.4 -3.8 Mexico -14.2 -3.1 -0.7 -3.8 -2.9 -3.1 Argentina -12.3 -0.3 -2.1 -3.1 -4.3 -3.8 Euriope and Central Asia 0.2 0.5 2.2 0.2 1.6 2.8 -1.3 Russian Federationb 25.0 2.0 3.6 2.1 6.7 8.4 Turkey -1.4 -2.0 -1.3 -0.6 -0.7 -3.0 - Poland -12.5 -0.9 -1.4 -2.5 -7.5 -7.0 - Middle East and North Africa 13.3 6.6 -3.5 -2.2 -1.7 1.3 0.7 Saudi Arabia -1.7 19.8 -7.2 -1.4 -1.3 2.3 Iran, Islamic Rep. 5.5 2.4 -0.4 0.6 2.5 2.7 ... Egypt, Arab Rep. -1.6 -4.9 -3.4 1.1 -3.1 -1.5 - Sub-Saharan Africa -8.9 -1.9 -2.8 -1.9 -2.3 -2.7 -3.1 Republic of South Africa -0.5 -1.7 0.6 -0.2 -0.4 -1.2 Nigeria 0,5 0.8 -0.7 -0.9 1.2 -0.2 - a. Data prior to 1991 covers West Germany. b. Data prior to 1992 covers the former Soviet Union. Soturce: World Bank data and staff estimates. Figure A3.4 Ratio of current account balance to GDP, 1991-2010 Percenit 20 1.0 E 2000 10 n.s -1 0 -30 -o - Rgh---ncsme East As a South Asia Latin America Europe ana M ddle East Sub-Saharan aconomies and Pasific and the Central Asia and North Atrifa Carribbean Afo- 179 G LOB A L EC ONO MI C PRO SPEC TS Table A3.5 Exports of goods, 1998 (percent) AveraCnragAea Aver.5agAeia 104 31 . East sia nd Ameni 329 .. 4anAnulal3767 n.2a 8 Pacific 524,083 ~12.9 86 zebijn978. Bosn 20 1 29899 China 18629 1.24 . BLarun 713mer4.iCMideroot nd 1' 0 1. . FAll 3d3v4.3ping Brulgaray49 -. . COTendIsia e4,7 5.4 cnountresia 4SVenezu9.9 7.4 Czeeh Rep,2,9 . 67 Ehoi 568 4.71 . Korea, Re. 1212 12.85 . Estopanda260 . 6 Gu-aboh2ara81n. Malaysia 72,517 10.2 8,4 G~~eotrgal Asi . Ghfi'an2,9 9.71 . MEantAsaard 1,34 135m.9 Hngai2 7 52 3 Kna 2,gol3 5.02 . Papifa New7 Azerkhijan 3Bo1 s 64 vadagsa53 5.59 . Guinea 1,7 15.2 7.7 lKrgys ReC35am ieriao,n7 3.7 . Phiippns494 .3. B7.7 lgatria 2, 11.4 . S6eegalv1o1r0 2.97 . Thaneiand3,4 13.19 Litchuai 3R92p.6 SothiAfrica 492 .57 . Virenam 9,363 1.8 EstMonia G4 3. uabn 396 14. . Malavsia 10.2 Ge~~olgan 3247G05h. amba 87 9-.7 7. SGuthAiane96 0. oaa 5.6,302g2.2R6.. Zimbrabw,2 3.37 . Phangpadesh ,4 13.7 LatRvsianSnga . India 34 6 11.1 7.' Federationi8 . . High.income Nepal 42 10.3 .2 Slovak Rep. 1,20 . economies 4,8, S .8 76 Pakistan 842 5.8 78 Tajikistan 8 Sri Lanka 473 11.5 66 TFYR Industrial 36542 5.6 7. Macedonia32 .. 40 Latin America ad .Turkmenistan >~ '6 .G-7 26,03 5.2 77 the Caribbean 2.545 8.2 8. Turkey 12.' 10.6 61 Canada 2. 774 6.4 6. Argentina 2444 7.8 '>,'Ukraine 11 9 .France >3065 5.6 68 Bolivia 144 64 9~5 Uzbekistan ~ ..Germanv 59,4 5.2 6. Brazil 5 6 4.4 8. talv.24A9 5.4 7. Chile 43I 6.9 , c Middle East and> Japan . 99 2.5 9. Colomnbia 1 7.2 >,4 North Africa '4s 3.5 7,2 United Costa Rica '16.0 '~7pAlgeria I,$ 2.5 66 Kingdom 21 ~~ 5.2 7. Dominican Rep. 418 10.0 b, Bahrain -1.0 5 United States 47 ~ 6.7 8. Ecuador 4t9 6.1 & Egypt, Arab El Saivador 242 11.9 4 Rep. 4,4 .0 Other Guatemnala 4,* 9.1 ' Iran, Islami industrial ~ 949 6.6 4i Jamaica ,i6 . 9.2 6. >Rep. ''9Z 3.1 7'Australia '~42 6.5 78 Mexico 11'$4 11.7 . Iraq >8 -3.2 ,4 Austria 6~9 8.4 6 Paianama>,Z 9.3 . Jordan 4 5.7 4 Belgium' .4~4 4.8 2 Paraguay -32" 16.3 '1,> Morocco /4.4 6 ` . Denmark ~ 98 4.4 Peru , 8.1 3. Oman 8 6.1 Finland 3,4" 7.7 2 Trinidad and , Saudi Arabi 2.5 ' AGreece 9,~ 4.8 >5 Tobago ;' ,i8 4.7 'Syrian Arab Rep& . 4.5 ,9 Iceland -1.6 ". 180 G LO BAL EC ONO MI C IN DI CATO RS Table A3.5 Exports of goods, 1998 (continued) (percent) MereIsviie Average Effectivc Merausl Average Efledve Merchandise Average Efftve eXport annual air exports annual mtodtt expotnt annual Was (81w growth grOWIh fUS$ growth growth (S$ growth growth ioeee) 1989-98 19m99 moen 1989-98 t9 ' M mons) 1989-98 19w9 - Other industrial (cotisaed) Other high- Other high-incowm Ireland 63,513 16.4 6.2 income 457,135 7.6 9.1 (continued) Netherlands 172,795 4.6 6.4 Brunei 1,894 0.2 8.7 Singapore 110,591 8.8 10.0 New Zealand 12,71 2.8 7.6 Hong Kong, Taiwan, Norway 40,637 7.9 6.4 China 175,784 9.3 9.2 China 110,046 6.3 8.9 Spain 109,690 10.4 6.7 Israel 22,972 6.6 7.1 United Arab Sweden 83,369 7.1 6.6 Kuwait 9,618 3.2 8.6 Emirates 27,059 0.7 6.8 Switzerland 92,845 5.0 6.9 Qatar 4,377 2.7 7. .. Not available a. Effective market growth is a weighted average of import volume growth in the country's export markets. b. Includes Luxembourg Source: See technical notes. Figure A3.5a Merchandise exports as a share of GDP. 1998 Percent 40.0 30.0 Wordd 20.0 M m _ 0.0 Industrial Sub-Saharan East Asia Souah Asia Latin America Europe and Middle tEast econmies Afri.a and Pacific and the Central Asia and North Canibbean Africa Figure A3.5b Average annual growth rate of export volumes 1989-98 12.0 - 100 o Word 8.0-* so 4.0 -_ 2.0 rncustriat Sub-Saharan Eas Asia South Asia Lati America Europe ancl Middle East economies Africa and Pacdic wnd the Central Asia and Norh Carribbean Africa 181 G LO BAL EC ONO MI C PRO SPEC TS Table A3.6 Imports of goods, 1998 (percent) Mges*ieAverage M"rdas- Mrssds Average Merdla.Merladise Average Merdian. ipo* annual -iciprs anual die spot annual di-e f* girowth Iusrl US growth Ipes US growth Imports/ ~e~it) 1989-98 G?mlon) 1989-98 GD ilo) 1989-98 GDP World SZ,55 6.4 184 Latin America an Middle East and the Caribbean (nind)North Africa (cntned) All developing Uruguay 361 11.4 17.3 Tunisia 785 7.9 39.5 countries ,5188 8.0 213 Venezuela 1486 1.2 15.6 Yemen, Rep. 2,201 3.6 36.7 Asia 4292 9.0 188 Europe and Sub-Saharan Central Asia 3,66 5.6 30,3 Africa 81,791 5.2 26.8 East Asia and Armenia 0 . 42.4 Angola 209 3.8 27.7 Pacific 3420 9.3 20. Azerbaijan 1,2 . 41.9 Botswana 1,983 6.7 40.7 China 36IS 11.3 145 Belarus 8,6 . 33.4 Cameroon 1,452 1.4 16.4 Fiji 62 4.3 388 Bulgaria 454 -7.5 37.3 C6te dIlvoire 2,705 4.3 24.8 Indonesia 3192 8.4 339 Czech Rep. 2,8 .. 51.4 Ethiopia 1,042 0.8 15.9 Korea, Rep. 9045 6.0 285 Estonia 3,0 . 73.1 Gabon 765 -0.9 16.1 Malaysia 5449 12.7 751 Georgia 1,6 . 20.4 Ghana 2,897 10.8 38.8 Myanmar ,40 15.8 101 Hungary 2311 9.0 48.3 Kenya 3,029 4.9 26.2 Papua New Kazakhstan 6,7 . 30.4 Madagascar 693 7.9 18.5 Guinea 108 -2.8 290 Kyrgyz Rep. 75Z.4. Ngra9,211 7.3 22.3 Philippines 2954 13.4 454 Latvia 3,4 . 49.1 Senegal 1,245 2.5 26.6 Thailand 3676 6.9 328 Lithuania 5,8 . 51.0 South Africa 27,216 4.2 20.3 Vietnam 1031 21.9 381 Moldova 1,4 . 64.5 Sudan 1,732 6.4 16.2 Poland 4533 12.9 28.8 Zambia 1,022 3.8 30.5 South Asia 6862 7.4 123 Romania 1097 3.4 28.6 Zimbabwe 2,019 4.9 31.9 Bangladesh ,76 9.6 157 Russian India 4488 8.0 104 Federation 5771 . 0.9 High-income Nepal ,3 6.4 259 Slovak Rep. 1301 . 4.2 economies 4,0967 5.9 17.7 Pakistan 1067 3.7 167 Tajikistan 73 . 41.1 Sri Lanka 532 10.2 338 TFYR Industrial 3,59,07 5.6 16.3 Macedonia 1,2 .. 51 Latin America adTurkmenistan 1,3 . 42.0 G-7 2,622,338 5.6 13.9 the Caribbean 3,08 13.0 164 Turkey 4552 12.8 22.9 Canada 204,554 5.7 34.4 Argentina 2958 19.1 99 Ukraine 1,8 .. 38.4 France 279,506 4.6 19.6 Bolivia 1,6 10.5 206 Uzbekistan 2,1 . 16.1 Germany 459,188 6.0 21.8 Brazil 57,73 14.1 .4Italy 202,782 4.7 17.2 Chile 1737 12.9 227 Middle East and Japan 251,254 4.1 6.6 Colombia 1407 11.0 136 North Africa 9130 3.1 190 United Kingdom 305,730 4.9 22.0 Costa Rica 5,71 1.0 553 Algeria 65-18.7 18 United States 919,324 6.8 11.0 Dominican Rep. 757 16.7 479 Bahrain 329 3.5 6. Ecuador 518 1 1.5 264 Egv,pt, Arab Other El Salvador 377 13.4 313 Rep. 1467 4.7 177 industrial 966,729 5.5 30.2 Guatemala 426 10.5 226 Iran, tslamic Australia 83,433 S.3 17.0 Jamaica 210 7.6 422 Rep. 1368 2.2 11,9 Austria 67,988 6.1 32.9 Mexico 1534 15.3 306 Iraq ,25 -18.7 0.5 Belgiuma 149,230 5.1 56.2 Panama 766 11.6 842 Jordan 344 3.2 46.0 Denmark 44,021 5.5 25.4 Paraguay 342 13.8 453 Morocco ,43 7.9 26.6 Finland 30,903 3.7 24.6 Peru ,20 10.5 131 Oman 527 7.0 34.9 Greece 19,17 4.8 15.9 Trinidad and Saudi Arabia 2755 2.7 21.4 Iceland 2,279 4.5 28.8 Tobago 299 10.4 470 Syrian Arab Rep 307 5.4 19. Ireland 41,896 12.9 48.6 182 G LO B AL EC ONO MI C IND IC AT OR S Table A3.6 Imports of goods, 1998 (continued) (percent) Merchandise Average Merdan- Mrcaoadise Average Mcilrehan- ferrhmse Average Mchaw1 imports annual dise imports annual dise imports annual disc fUs$ growth Inports (USS growth Impot SS growth hmports miuions) 1989-98 GD'P millo) 1989-98 GDP - m ) 1989-98 (L Other industrial (continued) Other high- Other high- Netherlands 152,247 4.8 40.5 income 440,629 8.8 61.5 income (continued) New Zealand 11,334 S.0 21.5 Brunei 1,718 12.0 35.4 Qatar 3,322 10.1 30.8 Norway 39,070 5.1 26.6 Hong Kong, Singapore 95,781 8.2 115.7 Spain 127,740 7.9 22.9 China 183,503 10.6 112,6 Taiwan, China 98,949 7.4 37.4 Sweden 66,237 3.9 29,1 Israel 26,197 6.6 21,0 United Arab Switzerland 92,882 2.5 35.4 Kuwait 7,714 1.6 30.5 Emirates 24,995 12.0 52.9 Not available a. Includes Luxembourg Source: See technical notes. Figure A3.6a Merchandise imports as a share of GDP, 1998 Pretent 40.0 30.0 _ World_ 20 0 \_ 10,0_ Industrial Sub-Saharan East As,a Soulh A-l Lawm Amenca Europe and M,ddle East econo.-e Afrnca and Paoert and th. C.rtal Aina nd No,th Ca,nbbean Africa Figure A3.6b Average annual growth rate of import volumes, 19B9-98 Pencent 14.0 12.0 10.0 Worid_ B.0 \ 4.0 25 0 0.0.I 0 Industrial Sub-Saharan East Asia South Asia Latin A-erica ECrope and Midde Eaw economies Afnca and Pacihc andthe Cennal Asia and North Carribbean Af rica 183 G LO B AL EC ONO MI C PRO SP EC TS Table A3.7 Direction of merchandise trade, 1998a (percentage of world trade) Hihicm im'ete Low- and middle-income importers Latin Middle America AUl Sub- East Europe East and low- Other All Other All Saha- Asia and and the and :n ted- iMus. indiss. =igh. hig- ran and South Central North Carib- middle- Source of exports t L n ta ta acn ,e, Africa Pacific Asia Asia Africa bean income W:-ld High-income econ. 1. 2.7 3 138 5. 5. 5.4 1.0 6.1 0.7 3.8 1.7 4.7 17.9 7. Industrial 0 2.0 4 4 8 5. 0.9 3.8 0.5 3.7 1.6 4.5 14.8 66. United States .. 3 2 1 1.3 76 . : 0.1 1.1 0.1 0.2 0.3 2.8 4.8 EU-1534 2. . 0. 87 13 3. 0.6 1.0 0.2 3.2 1.0 1.1 7.0 3. Japan 2.5 0.1 1.4 0.1 0.1 0.2 0.4 2.2 : .8 Other industrial 0si 1 9. 85 3. 12.2 00.2 0.6 01 1.0 0.3 0.4 2.5 Other highbincomeb 21 14 07 .8 .6 .2 .8 0.1 2.3 0.2 0.1 0.1 0.2 3.189 Low- and middle- income economies 61 .0 20 26 4. 2. 178 0.5 1.8 0.5 2.0 0.6 1.5 6.9 2. Sub-Saharan Africa 02 0.4 0 it 0.2 0.1 0.1 0.0 0.0 0.0 0.4 1.2 East Asia and Pacific 22 16 15 10 57 24 81 0.2 1.2 0.2 0.2 0.2 0.3 2.4 1. South Asia 03 03 01 01 06 01 08 0.0 0.1 0.1 0.0 0.1 0.0 0.31. Europe and Central Asia 03 24 01 08 29 01 30 0.0 0.1 0.0 1.6 0.1 0.1 2.0 4 Middle East and North Africa 02 06 02 02 10 02 12 0.0 0.3 0.1 0.1 0.1 0.0 0.61. Latin America and Caribbean 2.9 0.7 0.1 0.3 3 0.0 0.1 0.0 0.1 0.1 1.1 1.4 5.4 World 18.2 357 5. 0 1. 61.5 7.9 1.2 5.8 2.3 6.2 24.8 00.0 a. Expressed as a share (percent) of total world exports. World merchandise exports in 1998 amounted to some $5,360 billion. b. Other high-income group includes the Asian newly industrializing economies, several oil exporters of the Gulf region, and Israel. Source: IMF, Direction of Trade Statistics. 184 G LO BAL ECON OM IC IND I CATO RS Table A3.8 Growth of current dollar merchandise trade, by direction 1989-98 (average annual percentage growth) I4sseoue Isportrn Low- and middle-income importers Latin Middle America All Sub- East Europe East and low- Other All Othe MA Saha- Asia and and the and tjnjs.4 iss4sss. is,sJut, high- high ran and South Central North Carib- middle- Source of exports Stabes M.IS Japan til trial ioon inom Africa Pacific Asia Asia Africa bean income World High-income economies 6.1 4.0 3.8 3J3 4.6: 80 49 2.9 8.6 3.1 11.8 4.7 12.0 9.9 5.7 Industrial 5.9 3.8 3.5 3.0 4.4 7.1 4.6 2.6 7.0 1.0 11.7 4.7 11.9 9.6 5.3 United States ... SA 4.4 4.2 6A 7.5 6.5 6.1 8.8 1.8 6.7 6.9 12.9 10.8 7.7 EU-1S 6.2 3.7 4.5 2.9 3.9 8.1 4.1 2.4 8.2 0.1 13.5 4.4 11.8 11.0 4.8 Japan 2t9 2.5 ... 2.0 25 6.3 3A -0.5 5.6 0.2 -0.4 3.8 8.9 5.6 3.8 Other industrial 5.1 3.2 3.0 2.4 3.4 6.2 3.5 2.5 7.8 -0.1 10.8 2.5 10.8 9.6 4.1 Other high-income' 6.9 8.2 4.7 8.2 6.8 12.0 7J7 5.8 12.0 10.3 17.7 5.6 14.8 11.6 8.9 Low- and middle- M1 10.5 612 9.5 1.4: 11,0 105 12.3 17.6 12.1 13.0 7.8 15.9 16.7 11.4 income economies - Sub-Saharan Africa 5.9 4.1 8.5 2.5 5.2 28.4 5.9 14.1 21.3 26.1 2.9 12.0 11.6 15.8 7.8 East Asia and Pacific 11.0 12.1 6.9 14.3 10.3 11.5 10.6 10.3 20.1 12.8 12.2 8.1 23.7 17.3 11.6 South Asia 12.9 8.9 1.6 6.3 9.4 13.0 9.9 17.0 14.5 13.7 -3.2 8.7 32.4 14.3 9.7 Europe and Central Asia 13.7 10.1 -1.9 3.5 9.9 13.4 10.0 4.2 4.7 -2.6 6.1 3.2 8.8 12.6 8 .0 Middle East and North Africa 1.7 2.8 5.I 0.4 2.8 3.2 2.8 15.6 18.5 8.1 0.3 7.2 2.7 10.5 4.6 Latin America and Caribbean l4.5 3.4 1.4 2.0 10.4 6.4 10.3 5.1 6.0 9.4 -0.5 6.3 15.0 12.5 10.6 World 7.7 4.5 4.5 3.8 S.5 S.9 5.8 4.9 9.9 5.4 9.3 5.1 12.8 10.9 6.5 a. Other high-income group includes the Asian newly industrializing economies, several oil exporters of the Gulf region, and Israel. Note: Growth rates are compound averages. Source: IMF, Direction of Trade Statistics 185 G LO B AL EC ONO MI C PRO SP EC TS Table A3.9 Structure of long-term public and publicly guaranteed (PPG) debt, 1998 (percentage of long-term PPG debt) Non-concessional Non-concessional Coneessis,nal Variable Fixed Cioncessional Variable Fixed All developing economies 25.7 35.1 39.2 Europe and Central Asia (continued) Bulgaria 2.1 84.9 13.0 Asia 37. 31.4 31.5 Czech Republic 1.1 43.4 55.6 East Asia and Pacific 26.6 37.7 35.7 Estonia 12.7 62.8 24.5 China 27.0 35.1 37.9 Georgia 58.9 11.2 29.9 Indonesia 4. 38.6 19.9 Hungary 3.3 23.0 73.6 Korea, Rep. 0.1 54.2 45.7 Kazakhstan 8.7 62.0 29.3 Malaysia 12.9 24.8 62.4 Kyrgyz Republic 67.8 23.8 8.4 Myanmar 88,1 0.0 11.9 Latvia 18.6 78.5 2.9 Papua New Guinea 59.9 19.9 20.2 Lithuania 9.6 52.7 37.7 Philippines 40.5 32.6 26.9 Moldova 24.5 48.1 27.4 Thailand 26.9 31.5 41.6 Poland 22.5 58.7 18.8 Vietnam 22.2 16.5 61.3 Romania 8.3 34.8 56.9 Russian Federation 1.9 54.9 43.2 South Asia 60.3 17.6 22.1 Slovak Republic 7.2 36.5 56.2 Bangladesh 990- 0.1 0.8 Tajikistan 88.2 9.2 2.6 India 48.0 19.3 32.7 Turkmenistan 6.8 79.1 14.1 Nepal 98.5 0.0 1.5 Turkey 11.6 21.6 66.9 Pakistan 64.6 27.6 7.8 Ukraine 2.9 59.6 37.5 Sri Lanka 89.7 5.8 4.6 Uzbekistan 19.0 55.2 25.9 Latin America and Middle East and the Caribbean 7.7 42.4 49.9 North Africa 4S.6 29.8 21.6 Argentina 2.9 34.0 63.1 Algeria 11.5 53.8 34.7 Bolivia 748 10.2 15.0 Egypt, Arab Rep. 84.8 4.4 10.8 Brazil 1:6 53.5 44.9 Jordan 52.1 28.2 19.7 Chile 7.6 78.1 14.2 Morocco 33.6 38.3 28.2 Colombia 5.3 39.7 55.0 Oman 24.9 45.0 30.1 Costa Rica 23.0 25.0 52.1 Syrian Arab Republic 92.4 0.0 7.6 Dominican Republic 4316 36.5 19.9 Tunisia 28.3 25.1 46.6 Ecuador 15.4 54.2 30.4 Yemen, Rep. 90.3 2.2 7.5 El Salvador 48.5 26.8 24.7 Guatemala 44.3 23.2 32.6 Sub-Saharan Africa S1.9 13.1 35.0 Jamaica 34.1 23.4 42.5 Angola 27.3 12.8 59.9 Mexico 1.5 37.3 61.1 Botswana 58.2 11.4 30.4 Panama 7.2 49.1 43.7 C6te d'Ivoire 45.8 39.0 15.2 Paraguay 58.2 19.9 22.0 Cameroon 54.5 13.4 32.1 Ethiopia (excludes Peru 18.9 42.1 39.1 Eritrea) 91.6 0.2 8.2 Trinidad and Tobago 0.9 52.2 47.0 Gabon 24.7 14.2 61.1 Uruguay 4.7 43.8 51.5 Ghana 82.9 0.4 16.7 Venezuela 0.3 54.7 45.1 Kenya 73.5 4.7 21.8 Madagascar 70.9 5.7 23.3 Europe and Nigeria 6.8 19.7 73.5 Central Asia 8.1 46.9 45.0 Senegal 78.8 11.6 9.7 Armenia 56.3 22.9 20.8 Sudan 51.0 14.6 34.4 Azerbaijan 74.5 25.5 0.0 Zambia 68.6 13.4 18.1 Belarus 1.5 67.7 20.9 Zimbabwe 45.5 16.2 38.3 186 G LO B AL EC ONO MI C IND I CATO RS Figure A3.9a Structure of long-term PPG debt, by group, 1998 Pencent 100 80 60 East Asia Europe and Latin America Middle East South Asia Sub-Saharan and Pacific Central Asa aid the and North Aftrica Carribbean Africa Figure A3.9b Top ten ratios of non-concessional debt to GDP, 1998 Vietnam_ Turkmenistan Gabon Guyana Bulgaria Congo, Dern. Rep. Guinea-!issau Angola Congo, Rep. Of Nicaragua 0 20 40 60f 80 100 120 1t0 Percent 187 G LO BAL EC ONO MI C PRO SP ECT S Table A3.10 Long-term net resource flows to developing countries, 1998 (millions of U.S. dollars) Asia ~ ~ ~ 5,24 4.25 7480 -,4 785 938 20,594 11,543 9,051 East Asia and Pacific 8288 4.94 6,4 -519 412, 907 15,588 6,857 8,732 China 4520 4.78 4266 -,4 3'i 125 2,554 1,360 1,194 Indonesia -88 -0.86 -379 -,5 36 20 2,951 1,618 1,333 Korea, Rep. 13,214 4.16 7,4 487 545 406 5,557 -19 5,576 Malaysia 8,59 11.77 825 ,73 500 92 235 66 169 Myanmar 27 . 13 8 0 0 119 120 -1 Papua New Guinea 48 11.23 30 29 10 0 188 229 Al1 Philippines -274 4.25 256 49 173 44 178 481 -303 Thailand 897 8.02 725 148 6,123 1,162 355 807 Vietnam 240 7.91 83 38 120 0 1,318 1,304 14 South Asia 1256 2.23 7,8 41 369 31 5,005 4,686 319 Bangladesh 133 3.05 88 -3 08 3 1,015 1,026 -11 India 764 1.77 ,11 ,74 235 42 1,453 1,010 443 Nepal 23 5.28 - -112 0 253 253 0 Pakistan 181 2.94 06 36 00 0 1,066 1,179 -113 Sri Lanka 18 5.21 35 16 13 6 493 488 5 Latin America and the Caribbean 367Z 6.71 1,84 573 6923 148 10,118 3,522 6,595 Argentina 1953 6.56 1889 1,9 ;$ 0 654 -165 818 Bolivia 114 13.96 60 -2 7 334 445 -111 Brazil 5933 7.63 4,8 13 39l3 4Z 5,008 106 4,902 Chile ,19 12.04 9,5 ,2 ,3 7 -62 47 -110 Colomhia 3,797 3.69 3,3 ~ ,3 6 168 30 138 Costa Rica 796 7.59 80 21 59 0-5 -30 25 Dominican Rep. 807 5.09 77 9 436 42 -6 Ecuador 88 4.25 58 27 ~ 0 254 152 102 El Salvador 48 3.60 24 3 2 0 186 109 77 Guatemala 87 4.76 62 5 7$ 0 276 173 103 Jamaica 34 8.32 56 27 39 0 -52 29 -81 Mexico 2248 5.47 2318 1,2 028 70 -760 -31 -730 Panama 160 17.50 1,5 5 ,0 ) 141 -14 155 Paraguay 35 3.51 36 20 56 069 20 50 Peru 304 4.82 2,2 2b 190 14 299 204 95 Trinidad and Tobago 73 11.48 76 1 70 0 -28 7 -35 Uruguay 66 3.34 46 32 14 0 201 6 195 Venezuela 808 8.43 6,6 ,6 ,3 4 1,142 13 1,129 Europe and Central Asia S956 .94 3,4 68 2450 ,94 6,220 5,791 429 Armenia 31 16.91 22230. 89 99 -9 Azerbaijan 1,7 28.62 10 5 t,2 097 65 33 Belarus 26 0.84 12 -7 1994 3 91 Bulgaria 673 5.49 98 3 41 6 175 95 81 Czech Repuhlic 397 5.67 3,3 4. 254 19 -135 100 -235 Estonia 70 15.00 74 8 1 5366 56 11 Georgia 27 3.97 7 7 50 0 150 164 -14 Hungary 385 7.98 463 248 ,96 59 -869 97 -965 Kazakhstan 237 10.63 183 25 18 353 95 258 Kyrgyz Republic 23 17.95 10 0 ~185 178 7 Latvia 50 8.29 36 &164 70 94 Lithuania 183 11.02 98 7 ~ 6 0 200 86 114 188 G LOB AL EC ON OM IC IND I CAT ORS Table A3.10 Long-term net resource flows to developing countries, 1998 (continued) (millions of U. S. dollars) private Official Official Percentage Debt development Total of GDP Total flows net FMI Portfolio Total assistance Other Europe and Central Asia (condtiuet) Moldova 100 6.22 62 -23 85 0 39 12 27 Poland 9,716 6.18 9,653 2,319 6,365 969 62 319 -256 Romania 1,825 4.78 1,826 -247 2,031 42 -1 227 -228 Russian Federation 20,142 7.28 19,347 16,286 2,764 296 796 12 784 Slovak Rep. 1,691 8.30 1,480 918 562 0 211 70 141 Tajikistan 70 3.91 -3 -21 18 0 72 72 0 Turkmenistan 601 22.20 473 343 130 0 128 75 54 Turkey 1,584 0.80 1,641 -179 940 880 -57 -219 163 Ukraine 2,438 5.75 2,087 1,344 743 0 351 133 219 Uzbekistan 732 4.35 592 392 200 0 140 162 -22 Middle East and North Africa 11,472 1.94 9,222 3,290 5,OS4 878 2,249 4,067 -1,818 Algeria -1,427 -3.01 -1,321 -1,328 5 2 -106 62 -167 Egypt, Arab Rep. 2,458 2.97 1,385 -186 1,076 494 1,073 1,258 -184 Iran, Islamic Rep. -325 -0.28 588 564 24 0 -913 -5 -908 Jordan 632 8.56 207 -114 310 11 425 377 48 Morocco 936 2.63 965 470 322 174 -29 373 -401 Oman -248 -1.65 -214 -330 106 10 -34 -29 -5 Syrian Arab Rep. 143 2.12 76 -4 80 0 67 104 -37 Tunisia 619 3.10 694 4 650 40 -76 -17 -59 Yemen, Rep. 6 0.10 -210 0 -210 0 216 233 -17 Sub-Saharan Africa 14,895 4.46 3,452 -1,621 4,394 679 11,444 12,387 -943 Angola 249 3.31 40 -320 360 0 209 238 -30 Botswana 107 2.19 91 -5 95 0 16 40 -24 Cameroon 238 2.68 1 -49 50 0 237 368 -131 C6te d'lvoire 729 6.67 181 -260 435 6 548 719 -171 Ethiopia 500 7.64 6 2 4 0 494 495 -2 Gabon -64 -1.36 -57 -7 -50 0 -8 37 -45 Ghana 579 7.74 42 -29 56 15 537 571 -35 KIenya 149 1.29 -57 -72 11 4 206 301 -94 Madagascar 414 11.08 i5 -1 16 0 399 417 -18 Nigeria S98 1.45 1,028 -2S 1,051 2 -430 -143 -287 Senegal 341 7.28 24 -16 40 0 317 350 -33 Sudan 558 5.21 371 0 371 0 187 188 -1 Zambia 281 8.37 40 -32 72 0 241 289 -49 Zimbabwe 68 1.08 -217 -296 76 3 285 216 69 Source: World Bank data. 189 Technical Notes The principal sources for the data in this ap- have been supplemented by UNCTAD and UN pendix are the World Bank's central databases. Comtrade databases or by World Bank staff es- Regional aggregates are based on the clas- timates. Trade figures for countries of the for- sification of economies by income group and mer Soviet Union reflect the total of non-CIS region, following the Bank's standard defini- and intra-CIS exports and imports. tions (see country classification tables that fol- Tables A3.7 and A3.8. Growth rates are low). Debt and finance data refer to the 137 compound averages and are computed for cur- countries that report to the Bank's Debtor Re- rent dollar measures of trade. porting System (see the World Bank's Global Table A3.9. Long-term debt covers public Development Finance 2000). Small economies and publicly guaranteed external debt but ex- have generally been omitted from the tables cludes IMF credits. Concessional debt is debt but are included in the regional totals. with an original grant element of 25 percent Current price data are reported in U.S. or more. Nonconcessional variable interest dollars. rate debt includes all public and publicly guar- anteed long-term debt with an original grant element of less than 25 percent whose terms Notes on tables depend on movements of a key market rate. Tables A3.1 through A3.4. Projections are This item conveys information about the bor- consistent with those highlighted in Chapter 1 rower's exposure to changes in international and Appendix 1. interest rates. For complete definitions, see Tables A3.5 and A3.6. Merchandise ex- Global Development Finance 2000. ports and imports exclude trade in services. Im- Table A3. 10. Long-term net resource flows ports are reported on a c.i.f. basis. Growth are the sum of net resource flows on long-term rates are based on constant price data, which debt (excluding IMF) plus non-debt-creating are derived from current values deflated by rel- flows. Foreign direct investment refers to the net evant price indexes. Effective market growth is inflows of investment from abroad. Portfolio the export-weighted import growth rate of the equity flows are the sum of country funds, de- country's trading partners. The IMF's Balance pository receipts, and direct purchases of shares of Payments database is the principal source by foreign investors. For complete definition, for data through 1998; in some cases these data see Global Development Finance 2000. 190 Classification of Economies G LO B AL EC ONO MI C PRO SP EC TS Table 1 Classification of economies by income and region, July 2000 Europe and | Middle East Sub-Saharan Arica Asia Central Asia and North Africa East and Eastern Income southern West East Asia South Europe and Rest of Middle North group Subgroup Africa Africa and Pacific Asia Centrai Asia Europe East Africa Ameica Low- Angola Benin Cambodia Afghanistan Armenia Yemen, Rep. Haiti income Burundi Burkina Faso Indonesia Bangladesh Azerbaijan of Ni ua Comoros Cameroon Korea, Dem. Bhutan Georgia Congo, Dem. Central Afiman Rep. of India Kyrgyz Rep. of Republic of Lao PDR Nepal Republic Eritrea Chad Mongolia Pakistan Moldova Ethiopia Congo, Rep. of Myanmar Tajikistan Kenya Cte d'lvoire Solomon Turkmenistan Lesotho Gambia, The Islands Ukraine Madagascar Ghana Vietnam Uzbekistan Malawi Guinea Mozambique Guinea- Rwanda Bissau Somalia Liberia Sudan Mali Tanzania Mauritania Uganda Niger Zambia Nigeria Zimbabwe Sao Tomes and Principe Senegal Sierra Leone Togo Middle- Lower Namibia Cape Verde China Maldives Albania Turkey Iran, Islamic Algeria Beeize income Swaziland Equatorial Fiji Sri Lanka Belarus Rep. of Djibouti Bolivia Guinea Kiribati Bosnia and Iraq Egypt, Arab Clma Marshall He na Jordan Rep. of Rica Islands Bulgaria Syrian Arab Morocco Cuba Micronesia, Kazakhan Republic Tunisia Dominica Fed. Sts. of Latvia West Bank eul Papua New Lithuania and Gaza Ecudo Guinea Macedonia, El Sa Philippines FYRa enala Samoa Romnania G;uyan Thailand Russian Honduras Tonga Federation Jamoaca Vanuatu Yugoslavia, araguay Fed. Rep. Perui of St. Vincent an1d the Grenadines Upper Botswana Gabon American Croatia Isle of Man Bahrain Libya Aiaa Mauritius Samoa Czech Lebanon Malta Barsid Mayotte Korea, Rep. of Republic Oman Arenin Seychelles Malaysia Estonia Saudi Arabia South Africa Palau Hungary il |Poland0 Chile :; Slovak Dominica Republic Grenad Sublolal 0-157 25 23 23 8 0-80 2t0; 10 7 192 REG IO NA L EC ONO MI C PRO SP EC TS Table 1 Classification of economies by income and region, July 2000 (continued) Europe and Middle East Sub-Sahaa Africa Asia Central Asia and North Africa East and Eastern Income souhrn West East Asia South Europe atnd Rest of Middie North group Subgroup kfrica Arica and Pacific Asia Centtal Asia Europe East Africa Americas High- OECD Australia Austria Canada income Japan Belgium United New Zealand Denmark States Finland Francec Germany Greece Iceland Ireland Italy Luxembourg Netheriands Norway Portugal Spain Sweden Switzerland United Kingdom Non-OECD Brunei Slovenia Andorra Israel Aruba French Channel Kuwait Bahamas, Polynesia Islands Qatar The Guam Cyprus United Arab Bermuda Hong Kong, Faeroe Emirates Cayman Chinad Islands Islands Macao, Greenland Netherlands Chinar Liechtenstein Antilles New Monaco Virgin Caledonia Islands N. Mariana (U.S.) Islands Singapore Taiwan, China Total 25 23 35 8 27 27 14 7 41 a. Former Yugoslav Republic of Macedonia. b. Federal Republic of Yugoslavia (Serbia/Montenegro). c. The French overseas departments French Guiana, Guadeloupe, Martinique, and Reunion are included in France. d. On 1 July, 1997, China resumed its exercise of sovereignty over Hong Kong. e. On 20 December, 1999, China resumed its exercise of sovereignty over Macao. Source: World Bank data. Definitions of groups the group are experiencing similar development or that other For operational and analytical purposes, the World Bank's economies have reached a preferred or final stage of develop- main criterion for classifying economies is gross national ment. Classification by income does not necessarily reflect de- product (GNP) per capita. Every economy is classified as low- velopment status. income, middle-income (subdivided into lower-middle and This table classifies all World Bank member economies, upper-middle), or high-income. Other analytical groups, and all other economies with populations of more than based on geographic regions and levels of external debt, are 30,000. Economies are divided among income groups accord- also used. ing to 1999 GNP per capita, calculated using the World Bank Low-income and middle-income economies are sometimes Atlas method. The groups are: low-income, $755 or less; referred to as developing economies. The use of the term is lower-middle-income, $756-$2,995; upper-middle-income, convenient; it is not intended to imply that all economies in $2,996-$9,265; and high-income, $9,266 or more. 193 G LO B AL EC ONO MI C PRO SP EC TS Table 2 Classification of economies by income and indebtedness, July 2000 Income Sub- Not classified group group Severely indebted Moderately indebted Less indebted by indebtedness Low- Afghanistan Mali Bangladesh Armenia Liberia income Angola Mauritania Benin Azerbaijan Burkina Faso Mozambique Cambodia Bhutan Burundi Myanmar Chad Eritrea Cameroon Nicaragua Gambia, The Korea, Dem. Rep. of Central Niger Georgia Lesotho African Nigeria Ghana Nepal Republic Rwanda Haiti Solomon Islands [Comoros Sao Tome India Tajikistan Congo, Dem. and Principe Kenya Ukraine Rep. of Sierra Leone Kyrgyz Republic Uzbekistan Congo, Rep. Somalia Moldova C6te Sudan Mongolia d'lvoire Tanzania Pakistan Ethiopia Uganda Senegal Guinea Vietnam Togo Guinea- Zambia Turkmenistan Bissau Yemen, Rep. of Indonesia Zimbabwe Lao PDR Madagascar Malawi Middle- Lower Bolivia Algeria Albania Namibia Marshall Islands income Bosnia and Belize , Belarus Paraguay Micronesia, Fed. Sts. of Herzegovina Colombia Cape Verde Romania West Bank and Gaza Bulgaria Equatorial China Sri Lanka Cuba Guinea Costa Rica Suriname Ecuador Honduras Djibouti Swaziland Guyana Jamaica Dominican Tonga Iraq Macedonia, FYR5 Republic Vanuatu Jordan Morocco Egypt, Arab Yugoslavia, Peru Papua New Rep. Fed. Rep. of'b Syrian Arab Guinea El Salvador Republic Philippines Fiji Russian Guatemala Federation Iran, Islamic Samoa Rep. of St. Vincent and Kazakhstan the Grenadines Kiribati Thailand Latvia Tunisia Lithuania Turkey Maldives Upper Argentina Chile Antigua and Oman American Samoa Brazil Hungary Barbuda Poland Isle of Man Gabon Lebanon Bahrain Saudi Arabia Mayotte Malaysia Barbados Seychelles Palau Mauritius Botswana Slovak Puerto Rico Panama Croatia Republic Uruguay Czech South Africa Venezuela, Rep. Bol. de Republic St. Kitts and Dominica Nevis Estonia St. Lucia Grenada Trinidad and Korea, Rep. Tobago Libya Malta 194 REG IO NA L EC ONO MI C PRO SP EC TS Table 2 Classification of economies by income and indebtedness, July 2000 (continued) Income Sub- [ Not classified group group severely idebted Moderately indebted Less indebted by indebtedness High- OECD Australia Japan income Austria Luxembourg Belgium Netherlands Canada New Zealand Denmark Norway Finland Portugal Francec Spain Germany Sweden Greece Switzerland Iceland United Ireland Kingdom Italy United States Non- Andorra Kuwait OECD Aruba Liechtenstein Bahamas, Macao, The Chinad Bermuda Monaco Brunei Netherlands Cayman Antilles Islands New Caledonia Channel N. Mariana Islands Islands Cyprus Qatar Faeroe Singapore Islands Slovenia French Taiwan, China Polynesia United Arab Greenland Emirates Guam Virgin Hong Kong, Islands (U.S.) Chinac Israel --ta 46 _ 43 L59 | 59 a. Former Yugoslav Republic of Macedonia. b. Federal Republic of Yugoslavia (Serbia/Montenegro). c. The French overseas departments French Guiana, Guadeloupe, Martinique, and Reunion are included in France. d. On 20 December ,1999, China resumed its exercise of sovereignty over Macao. e. On 1 July, 1997, China resumed its exercise of sovereignty over Hong Kong. Source: World Bank data. Definitions of groups ther of the two key ratios exceeds 60 percent of, but does not This table classifies all World Bank member economies, and reach, the critical levels. For economies that do not report de- all other economies with populations of more than 30,000. tailed debt statistics to the World Bank Debtor Reporting Sys- Economies are divided among income groups according to tem (DRS), present-value calculation is not possible. Instead, 1999 GNP per capita, calculated using the World Bank Atlas the following methodology is used to classify the non-DRS method. The groups are: low-income, $755 or less; lower- economies. Severely indebted means three of four key ratios middle-income, $756-$2,995; upper-middle-income, (averaged over 1996-98) are above critical levels: debt to $2,996-$9,265; and high-income, $9,266 or more. GNP (50 percent); debt to exports (275 percent); debt service Standard World Bank definitions of severe and moderate to exports (30 percent); and interest to exports (20 percent). indebtedness are used to classify economies in this table. Se- Moderately indebted means three of the four key ratios ex- verely indebted means either: present value of debt service to ceed 60 percent of, but do not reach, the critical levels. All GNP exceeds 80 percent or present value of debt service to other classified low- and middle-income economies are listed exports exceeds 220 percent. Moderately indebted means ei- as less indebted. 195 -Nicholas Stern, SeniorVice President and Chief Economiiist The global economiiy is likelv approaching a cvclical high in 2000, while policy reforms and improvements in health and education in developing countries over the past decade have substantially increased their poteiitial for lonig-ternm growth. However, volatility in the global econoiic environmiient presents major risks for developing countries that could depress prospects. This eleventh annual edition of Global Economic Prospects 2001 • Analyzes the prospects for the global economy and the imiplications for poverty reduction E Explainis whly econiomic perfornmanice in the poorest developing countries has lagged behilnd • Shows that trade sanctions are ineffective in raising labor and environmental standards in developing countries • Explores the potential imipact of electronic commerce on productivity and trade in developing counitries Global Economlic Prospects 2001 provides essential informationi for those concerned with developmnents shaping todav's global economy. ISiS H Sticet, N.W. Washingtoi. DI. . 2t434, U.S.A. Telphllhne: 2)02 477 1234 F1iceimilc: 202 47 6n39 t Telex: NICI 64145 WORLDBANK NCKI 248423 WN:OLIDB1ANK illtfernct: www.woI (dba nk.org F-mail: feed h.ick(awor(l dl1ant k .or-g For more in foriratiot,) sce: wVwv.wV(oridbao lk.(org'ipr-osp'cts ISBN-0-821 3-4675-X