―園・……LOBAN& - ．叫 Global Economic Prospects and the Developing Countries 2002 o 2002 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 All rights reserved. 01 02 03 04 05-10 9 8 7 6 5 4 3 2 1 The findings, interpretations, and conclusions expressed here do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is copyrighted. No part of this work may be reproduced or trans- mitted in any form or by any means, electronic or mechanical, including photocopying, recording, or inclusion in any information storage and retrieval system, without the prior written permission of the World Bank. The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint, please send a request with complete information to the Copyright Clearance Center, Inc, 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, www.copyright.com All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org ISBN 0-8213-4996-1 ISSN 1014-8906 Library of Congress catalog card number: 91-6-440001 (serial) Contents Acknowledgments ix Summary xi Abbreviations and Data Notes xxi Chapter 1 Prospects for Developing Countries: Coping with a Global Slowdown A simultaneous downturn in the industrial countries 4 Global environment: trade 9 Global environment: financial markets 13 The outlook for developing countries 16 Risks to the outlook 24 Long-term prospects: growth and poverty reduction 27 Conclusions 33 Notes 33 References 34 Chapter 2 Market Access and the World's Poor 37 A changing landscape of merchandise trade 38 Labor-intensive exports can spur pro-poor growth 38 Market access barriers limit export opportunities of developing countries 44 Liberalizing trade to promote development 56 Notes 63 References 64 Chapter 3 Trade in Services: Using Openness to Grow 69 Surging trade and investment in services 70 Service reforms can promote efficiency and growth 76 Domestic policy: emphasizing competition and regulation 81 Multilateral engagement: buttressing domestic reforms 84 Notes 90 References 92 GLOBAL ECONOMIC PROSPECTS Chapter 4 Transport Services: Reducing Barriers to Trade 97 High transport costs penalize exports 98 Why some countries pay more for transport services: geography and income 103 Why some countries pay more: policy-driven factors 109 Unleashing competition in international transport: policy implications 120 Notes 123 References 125 Chapter 5 Intellectual Property: Balancing Incentives with Competitive Access 129 Intellectual property rights and development 130 Costs of enforcing IPRs 136 IPRs policies for promoting development 139 Other policies can support technological progress 144 Multilateral actions and IPRs in a development round 145 Notes 149 References 149 Chapter 6 Envisioning Alternative Futures: Reshaping Global Trade Architecture for Development 153 Reshaping global trade architecture for development 154 Envisioning alternative futures 166 Conclusions 176 Annex 1 177 Notes 178 References 182 Appendix 1 Regional Economic Prospects 187 Appendix 2 Global Commodity Price Prospects 211 Appendix 3 Global Economic Indicators 233 Figures 1.1 Industrial production in the G-3 countries falls in 2000-2001 5 1.2 European industrial production falls 8 1.3 U.S. NAPM and manufacturing industrial production excluding high tech, 1994-2001 9 1.4 GDP growth in OECD countries 9 1.5 Import growth across industrial centers 10 1.6 Export shares for developing countries excluding transition economies 11 1.7 Distribution of countries by share of primary commodities in total merchandise exports 11 1.8 Episodes of world growth slowdown and agricultural and mineral export prices 12 1.9 OPEC output and crude oil prices 13 1.10 Gross capital market flows to developing countries 14 iv CONTENTS 1.11 Emerging market spreads and share of global capital flows 14 1.12 Industrial- and developing-country GDP growth, 1981-2003 17 1.13 Composition of developing-country exports 18 1.14 Major destinations for developing-country exports 19 1.15 Total external debt in developing countries, 2000 19 1.16 Emerging market stripped spreads, 1999-2001 20 1.17 Global dynamics of recessions in industrial countries 26 1.18 Income and population shares 29 1.19 World poverty, 1820-1998 31 1.20 Under-5 mortality-hopes and aspirations 32 2.1 Changing global trade patterns 39 2.2 A rising share of exports in GDP is associated with faster growth 40 2.3 Increases in exports and agricultural production go hand in hand 41 2.4 In globalizing economies the poor participate in stronger growth 44 2.5 Tariffs still impede trade 45 2.6 Support to agriculture in the Quad is growing . .. partly due to the fall in commodity prices 51 2.7 Despite preferences, LDC exports to the Quad often face high tariffs 54 2.8 LDC exports can grow fast when tariff preferences are significant 56 2.9 Opposite patterns of tariff incidence in manufactures and agriculture 56 3.1 Trade in services has grown faster than trade in goods-and developing countries' share in world exports has increased, 1985-98 71 3.2 Transport has declined while "other" services have increased 72 3.3 FDI in services is concentrated in the OECD countries-but the growth rates are higher for many developing countries 73 3.4 Software is cheaper to develop in India 75 3.5 Services liberalization indices: telecoms & financial services 79 3.6 Greater liberalization in services is associated with more rapid growth 79 3.7 WTO members have been reluctant to make market access commitments on the movement of natural persons 86 4.1 Transport costs are often higher than tariffs 100 4.2 Tourism earnings in developing countries, 1998 101 4.3 Potential market access explains variations in income 102 4.4 Shipping a container from Baltimore, Maryland, around the world: Distance is only half the costs story 105 4.5 Ocean freight rates, 1970-99 106 4.6 Decomposing the costs of door-to-door shipments 108 4.7 Potential door-to-door cost savings on containerized imports in Brazil 109 6.1 Regional integration agreements are proliferating-and now span the globe 155 6.2 Developing countries could reap income gains of over $500 billion from full trade liberalization 168 6.3 Unskilled wages rise substantially relative to cost of living-implying a substantial reduction in poverty 175 6.4 Reform has costs, but they are largely outweighted by the gains 176 6.5 World trade booms, particularly in food and agriculture 177 V GLOBAL ECONOMIC PROSPECTS Tables 1.1 Global conditions affecting growth in developing countries and world GDP growth 3 1.2 Merchandise export volumes, annual average percentage change 10 1.3 All developing countries: key indicators 18 1.4 First year effects of a 2% of GDP decline in investment in the United States, Europe, and Japan 25 1.5 Short-term claims of international banks outstanding in selected developing countries 26 1.6 Withdrawal of short-term lending by industrial-country banks to selected developing regions: the first-year impact on GDP 27 1.7 Long-term prospects: forecast and scenario growth of world GDP per capita 28 1.8 Regional breakdown of poverty in developing countries 30 2.1 Major export booms in textiles and clothing and effects on economic performance and poverty 42 4.1 Ad valorem freight rates for U.S. imports: 1938, 1974, and 1998 105 5.1 TRIPS: who gains? 133 5.2 TRIPS-consistent IPRs standards: options for developing countries 141 6.1 Agriculture accounts for the bulk of the gains from merchandise trade liberalization 171 6.2 Global gains are sensitive to productivity-openness linkages 171 6.3 Services liberalization generates substantial windfall gains for developing countries 172 6.4 Labor's share of national income rises substantially 173 6.5 Developing countries increase their market share 176 Boxes 1.1 Japan and the developing countries 7 2.1 The aftermath of trade liberalization in agriculture: lessons from Haiti 43 2.2 U.S. sugar policy and its impact on imports 48 2.3 Wheat production with CAP support 49 2.4 Bringing support to agriculture and export subsidies under multilateral rules: a long-awaited endeavor 50 2.5 A primer on the agreement on textiles and clothing 52 2.6 Anti-dumping-and better alternatives 53 2.7 Mushroom wars 55 2.8 Calculating effective tariffs faced by the poor 57 2.9 Designing appropriate safety nets to ensure trade forms are pro-poor 59 2.10 The banana dispute: good intentions ... bad policies? 61 3.1 Why do services matter for development? 70 3.2 Whose regulations and for what purpose? Challenges in electronic commerce 74 3.3 Welfare gains from service liberalization: the case of Tunisia 78 3.4 Challenges in implementing procompetitive regulation 83 3.5 Financial sector liberalization: the need for policy coherence 85 3.6 Ensuring barrier-free trade in electronically delivered products 88 4.1 The Kenyan-European cut-flower supply chain 99 Vi CONTENTS 4.2 Inefficient internal transport systems contribute to the concentration of China's export industries in coastal regions 104 4.3 Lessons from customs reforms in Mexico 111 4.4 Maritime shipping in West Africa 113 4.5 How important are public and private barriers to trade in maritime services? 115 4.6 Lessons from reforming Argentina's ports 117 4.7 EU noise regulations and their potential effect on air service to Central Asian countries 119 5.1 An overview of intellectual property rights 131 5.2 Pharmaceutical policies and the limits of TRIPS 138 6.1 Reshaping global trade architecture for development: the four-part policy agenda 158 6.2 The recently renovated integrated framework 160 6.3 Environmental standards and trade 161 6.4 Improving labor standards in a way that works 163 6.5 Standards development facility: coordinated action to bridge the standards gap 165 6.6 The complexities of measuring openness and growth 169 6.7 World Bank programs: activities to support trade-led pro-poor growth 177 Vii Acknowledgments his report was prepared by the Economic Policy and Prospects Group and drew from re- sources throughout the Development Economics Vice Presidency and the World Bank operational regions. The principal author of the report was Richard Newfarmer, with direction from Uri Dadush. The chapter authors were Hans Timmer (chapter 1), Aristomene Varoudakis (chapter 2), Aaditya Mattoo (chapter 3), Carsten Fink (chapter 4), Keith Maskus (chapter 5), and Dominique van der Mensbrugghe and Richard Newfarmer (chapter 6). Bernard Hoekman provided ideas, suggestions, and comments on the trade chapters. The report was pre- pared under the general guidance of Nicholas Stern. The report benefited from contributions from many Bank staff. Annette 1. De Kleine, Caroline Farah, Himmat Kalsi, Robert Keyfitz, Robert Lynn, Fernando Martel Garcia, Shoko Negishi, Dilip Ratha, Mick Riordan, Virendra Singh, and Bert Wolfe contributed to the global trends; Shaohua Chen and Martin Ravallion contributed to the poverty analysis; and David Roland- Hoist assisted with the international development goals in chapter 1. Ataman Aksoy, Betty Dow, Dilek Aykut, Don Mitchell, John Baffes, Marcelo Olarreaga, Simon Evenett, and Francis Ng con- tributed to chapter 2. Yong Zhang, Stijn Claessens, Antonio Estache, Charles Kenny, Fernando Garcia Martel, Cristina Neagu, Randeep Rathindran and Taizo Takeno provided inputs into chapter 3. Shweta Bagai, Ileana Cristina Neagu and Ranga Rajan Krishnamani contributed to chapters 4 and 5. The following provided comments on chapter 4: Mark Juhel, Juan Gaviria, Ronald Kopicki, Aaditya Mattoo, Marcelo Olarreaga, Uma Subramanian, Simon J. Evenett, and Tony Venables. Ataman Aksoy, Shweta Bagai, Mirvat Sewadeh, and John S. Wilson contributed to Chapter 6. The regional and statistical annexes were prepared by Milan Brahmbhatt, Caroline Farah, Robert Keyfitz, Annette I. De Kleine, Robert Lynn, Mick Riordan, and benefited from the guidance of the Bank's regional chief economists, Sadiq Ahmed, Alan Gelb, Homi Kharas, Mustafa Nabli, Guillermo Perry, and Marcelo Selowsky. John Baffes, Betty Dow, Don Mitchell, and Shane Streifel contributed to the analysis of commodity prices in chapter 1 and the appen- dix. Mark Feige edited the report with a special competence. Shweta Bagai, Yeling Tan, and Yong Zhang provided research assistance. Awatif Abuzeid was the task assistant for the report, and Katherine Rollins assisted with chapter 1. Many others from inside and outside the Bank provided inputs, comments and suggestions that immeasurably improved its content. Milan Brahmbhatt and Roberto Zagha were overall peer reviewers, and other reviews of specific chapters were invaluable: James Hanson, Odin Knudsen, Kym Anderson, James Hodge, Elizabeth Twerk, Carlos Braga, Eric Swanson, Anne ix GLOBAL ECONOMIC PROSPECTS Kenny McGuirk, Todd Schneider, Ejaz Syed Ghani, Manjula M. Luthria, David Hummels, Mauricio Carrizosa, Will Martin and T. N. Srinivasan. The report also benefited from the com- ments of Ian Goldin, Shahrokh Fardoust, Larry Hinkle, Ernesto May, David Tarr, Dimitri Diakosavvas, and Edith Wilson. The Development Data Group was instrumental in the prepara- tion of the appendix. Robert King coordinated the production and managed the dissemination from the Development Prospects Group, working closely with Heather Worley and Susan Gra- ham. Book design, editing and production were managed by the Production Services Unit of the World Bank's Office of the Publisher. X Summary A 2001 DRAWS TO A CLOSE, THE GLOBAL have become important actors in the global economy is slipping precariously to- system. In contrast to the early rounds of ward recession. Developing countries global trade negotiations-the Dillon Round have seen their economic growth rates plunge. in 1960 had only 39 participants, mostly from Growth in trade has undergone one of the industrial countries-the next round will have most severe decelerations in modern times- more than 142 WTO members, 70 percent of from over 13 percent in 2000 to 1 percent in which are developing countries. This mirrors 2001. Developing countries are confronting a the increased weight of developing countries 10 percentage point drop in the growth of de- in the global economy. They have grown to ac- mand for their exports. Though the weight of count for more than one-third of merchandise evidence still points to a probable recovery in trade-and they have much to gain from a new mid-2002, the risks posed to recovery are the round. gravest in a decade. The terrorist attacks in the On the other hand, they worry that the United States, although it is still too early to multilateral system, in leaving intact barriers evaluate them fully, have unleashed new and to markets whose removal would otherwise unpredictable forces that have substantially stimulate pro-poor growth, has become less raised the risk of a global downturn. fair and less relevant to their development Against this uncertain backdrop, world concerns; that the trade agenda is being ex- leaders have launched an intense discussion panded to include only issues in which the de- about whether to begin a new round of global veloped countries have an interest; and that trade negotiations at the ministerial meeting of multilateral rules are increasingly becoming a the World Trade Organization (WTO) in No- mere codification of existing laws and rules vember 2001. A round would offer an oppor- prevalent in developed countries, but which tunity to renew progress on multilateral rules are inappropriate or unenforceable in devel- that open markets and expand trade. A reduc- oping countries (Ganesan 2000). tion in world barriers to trade could accelerate Nor is support for new trade initiatives growth, provide stimulus to new forms of pro- universal among industrial countries. New op- ductivity-enhancing specialization, and lead to position to "globalization" in general-and a more rapid pace of job creation and poverty expanded trade in particular-has emerged reduction around the world. forcefully, questioning the very premises that However, the fate of new trade talks is as more open markets can raise people's incomes, uncertain as the global outlook. Many devel- especially those of the poor. The downturn in oping countries have lingering doubts about the global economy may inflame protectionist new trade negotiations. On the one hand, they sentiment. Xi GLOBAL ECONOMIC PROSPECTS The international community thus faces a pie, full implementation of the Agreement on clear choice: whether now is the time to con- Trade-Related Intellectual Property Rights tinue down the path toward greater openness (TRIPS) may not be suitable for all countries. that has led to greater integration and pros- Transportation cartels enjoy official sanction perity for more than five decades, or whether but are costly to developing countries, and to allow the hiatus in the wake of the WTO some standards may be set with little regard meetings in Seattle (1999) to endure. If trade for their effects on developing countries. talks are to succeed in underpinning a new Protection is not solely an issue for high- wave of global prosperity, and at the same income countries. Developing countries have time contribute to raising the incomes of the also placed high barriers on agriculture, labor- poorest in the global community, they will intensive manufactures, and other products and have to ensure that the world's poorest coun- services. Developing-country tariffs in manu- tries and poorest people will benefit. facturing average four times higher for imports from developing countries than are tariffs in in- The world's poor could benefit from dustrial countries on imports from developing reshaping the global architecture countries (12.8 percent as opposed to 3.4 per- of trade- cent). Restrictions on services trade are usually Poor people-those living below the interna- more common than in industrial countries. tional poverty line of $2 per day-work pri- This report argues for reshaping the global marily in agriculture and labor-intensive manu- architecture of world trade to promote devel- factures. These sectors confront the greatest opment and poverty reduction. The report fo- trade barriers, putting the world's poor at a cuses on four policy domains: particular disadvantage. According to estimates in chapter 2, the average poor person selling 1. Using the WTO ministerial to launch a de- into globalized markets confronts barriers that velopment round of trade negotiations that are roughly twice as high as the typical worker would reduce global trade barriers. Those in industrial countries. In general, tariffs in bargains will only be enduring and have high-income countries on imports from devel- greatest development impact if industrial oping countries, though low, are four times countries are willing to reduce restrictions those collected from industrial countries (0.8 on products and services that poor coun- percent as opposed to 3.4 percent). Subsidies tries and poor people produce-particularly and other support to agriculture in the high- protection of agriculture (including subsi- income countries are particularly pernicious- dies), textiles, and clothing; and even re- and are now running roughly $1 billion a strictions on temporary movement of work- day-or more than six times all development ers. Similarly, developing countries can assistance. Distortions in tariff codes-excep- improve their own situation while at the tionally high tariffs on developing country same time winning concessions by liberaliz- products (tariff peaks), embedded incentives ing services, and lowering barriers to import against processing abroad (tariff escalation), competition. To be sure, a trade round also and tariffs that are far higher once specified im- involves issues of interest primarily to in- port ceilings are reached (tariff rate quotas)- dustrial countries. Nonetheless, a true de- and trade practices, such as frequent recourse velopment round would produce win-win to antidumping actions, are often more impor- gains for the entire national community, in- tant impediments that keep the poor from tak- cluding the world's poor. ing advantage of trading opportunities. 2. Engaging in global collective action to pro- Other costly asymmetries in trade-related mote trade outside the negotiating frame- agreements and practices can at times work at work of the WTO. Providing market ac- odds with development objectives. For exam- cess may not by itself be enough to elicit xii SUMMARY new trade from developing countries, par- would enhance the prospects of developing ticularly the poorest. Increasing multilateral countries. "aid for trade"-development assistance to promote trade infrastructure, adoption of Reshaping global trade architecture best practice standards and rules, and a for development would reduce healthy investment climate-could help. No world poverty- less important, global cooperation to im- Seizing the opportunity to reshape the global prove the environment and labor standards trade architecture for development would can most effectively be undertaken outside make an enormous difference to the world's the WTO. poor. Some 2.8 billion people today live on 3. Adopting pro-trade development policies of less than $2 a day. In the base-case long-term high-income countries unilaterally. First, if projection of this report, developing countries the high-income countries were to allow would grow at rates that reduce poverty to 2.2 low-income countries duty-free and quota- billion by 2015, effectively lifting some 600 free access to their markets, they would pro- million people above this poverty line. This vide a strong stimulus to trade that would would be an important achievement. help these poor countries overcome their past But better results are possible. This report lackluster trade performance. Second, high- simulated the effects of taking the mutually re- income countries could also demonstrate inforcing actions in all four policy domains-ef- good faith by reining in mushrooming an- fectively removing restrictions on trade and ser- tidumping cases. Third, increasing bilateral vices in combination with the "aid for trade" "aid for trade" can complement the multi- agenda and other companion policies that trans- lateral effort. late the trade impulse into rising incomes for the 4. Enacting new trade reform in developing poor. These exercises have methodological limi- countries. Developing countries individu- tations but are indicative of what's at stake. ally can improve their competitiveness Three headlines are worthy of note: First, through trade reforms that lower restrictive the pace of poverty-reducing globalization barriers, especially in services markets. In- would clearly be accelerated. This combination deed their own policies hold the largest po- of policies could spur new growth that will lift tential for policy-induced gains from trade. an additional 300 million people above the Trade reforms, especially those reinforced poverty line relative to the normal growth in with reforms in governance and in domes- the base case.2 Said differently, because of faster tic investment climates, can raise productiv- growth associated with trade integration, the ity and incomes, irrespective of policies of world would have 14 percent fewer people liv- other nations. ing in poverty in 2015 than in the base-case sce- nario. Faster integration through lowering bar- Other aspects of global trade architec- riers to merchandise trade would increase ture-for example, regional trading arrange- growth and provide some $1.5 trillion of addi- ments, standards, and world institutions with tional cumulative income to developing coun- effects on trade (such as the World Customs tries over the 2005-15 period.3 Liberalization Organization and so on)-are also important. of services in developing countries could pro- However, save for brief mention in chapter 6, vide even greater gains-perhaps as much as they fall outside the focus of this report. This four times larger than this amount. is for reasons of parsimony and because they Second, the effects on income distribution of have been covered in recent Bank reports.1 removing trade restrictions in the simulation Nonetheless, if the policies recommended in are broadly positive. The simulations show that these four areas were adopted, they would labor's share of national income would rise move the global trade architecture in way that throughout the developing world. And un- xiii GLOBAL ECONOMIC PROSPECTS skilled workers generally do better in most re- world economy should grow by 1.6 percent, gions. Finally, this scenario would bring down with the recrudescence of consumer spending in infant mortality more rapidly and contribute to the United States, prompted by lower interest improved child health throughout the develop- rates and fiscal stimulus, and renewed expan- ing world. sion in Europe in response to recent interest rate cuts and lower oil prices. High-income countries, still shackled by slow growth in the Chapter Highlights first half of 2002 but picking up in the second, are likely to grow at about 1.1 percent for the This report is dedicated to the year, up slightly from the anemic 0.9 percent in trade-for-development agenda 2001. Dynamism in major economies of the de- Realizing the promise of the new global initia- veloping world-particularly China and India tives to expand trade requires concerted effort and, to a lesser extent, Brazil and Mexico-will to move development to center stage in trade reinforce these positive trends. South Asia policy formulation. This report is dedicated to seems likely to become the fastest-growing that agenda. It begins with a review of global region, with growth at 5.5 percent, followed prospects and ways globalization links the closely by East Asia, at 4.9 percent. Other re- fates of industrial and developing countries. gions will not achieve these growth rates, but The report then considers issues in four broad all will predictably do better than in 2001. areas that are particularly important to devel- The recovery of the global economy is oping countries: merchandise trade, services, likely to transmit new growth to developing transport, and intellectual property rights. A countries through more robust trade demand. final chapter summarizes the forward-looking Although unlikely to reach the boom rates of policy agenda, and assesses the potential im- 2000, trade expansion seems likely to surpass pact of further global integration and more 4 percent in 2002, up considerably from the rapid growth for the standards of living in 2001 rate. poor countries everywhere. Risks to this forecast are unusually high. The terrorist violence in the United States in Global prospects September will have negative short-run conse- By the third quarter of 2001, the global econ- quences for the United States and the global omy was precariously close to recession. For economy, but could be even more severe than the first time in more than two decades, the these projections indicate if unforeseen events three major engines of the global economy- prove highly disruptive. These uncertainties the United States, Japan, and Europe-were with enormous downside risks overlay struc- slowing at the same time. With recession al- tural risks. U.S. consumers may be less respon- ready a fact in Japan and the probability of sive to interest rates than on previous occa- negative growth in the United States rising-in sions; foreign investors, concerned about the part attributable to the demand and supply high external current account deficit, may pre- shocks from the September terrorist attack- cipitate a sudden adjustment; European growth and Europe suddenly slowing, the global econ- may level off at a lower-than-expected plateau; omy has ceased supporting rapid growth in de- and Japan's structural reforms may falter and veloping countries. cause the dip in 2001 to carry over into the next Nonetheless, the outlook for 2002, though year. Thus, with the global economy in precar- subject to unusually high risks, is that the ious balance, unforeseen shocks from whatever global economy will begin to recover. Develop- source are magnified and could push the global ing countries are expected to grow by 3.7 per- economy into recession. cent if the external environment improves as This said, the long-term prospects for de- expected, up from 2.9 percent in 2001. The veloping countries remain bright. Fundamen- xiv SUMMARY tals-savings, population growth, and invest- Europe, or Canada. As estimated in chapter 6, ments in education-are favorable. Moreover, phasing out restrictions on agriculture would many of the policy distortions prevalent in produce dynamic gains that could well mean many developing countries during the 1980s higher incomes in 2015 by nearly $400 billion. have been progressively diminished during the The Agreement on Textiles and Clothing 1990s. Budget deficits have generally come (ATC), which replaced the Multi-Fiber Agree- down, reserve levels are higher relative to debt ment in the Uruguay Round, succeeded in in- levels, and economies are now more open. For tegrating these products into the WTO. How- these reasons, the growth rates in the base-case ever, the agreement provided a much delayed scenario of 3.6 percent for the 2005-15 period phaseout schedule that put off much of the are both technically feasible and realistic. market liberalization until the very end of the However, not all countries and regions process in 2005. And, because the implemen- bask in this bright long-term outlook. Non-oil tation of the ATC allows importers much lee- commodity exporters, countries with high debt way in selecting the products to be freed of levels, and countries with poor credit histories quotas, forgone export earnings for develop- will find themselves at a disadvantage in trade ing countries are sizable. Because high tariffs and financial markets. Sub-Saharan Africa in loom behind the quotas, market access will re- particular confronts enormous problems in main restricted even after the quotas have all of these dimensions-as well as the public been abolished in 2005. Removing these bar- health epidemic of AIDS (acquired immune de- riers would, we estimate, produce increases in ficiency syndrome). For these reasons, invigo- income of perhaps $120 billion by 2015. rating the global trade agenda, even in these These issues provide fertile areas where rec- times of uncertainty, is imperative. iprocal negotiations in a development round of the WTO could provide substantial benefits for Merchandise trade development. Developing countries would ben- Restrictions on agriculture and labor-intensive efit from reducing their own protection in these manufactures, notably textiles and clothing, sectors as part of negotiated reciprocal reduc- are particularly damaging to the world's poor. tions in high-income countries for agriculture Virtually all major agricultural commodities and labor-intensive manufactures. Beyond this, face barriers to trade on a scale that dwarfs high-income countries could also expand trade manufactured products. Barriers include high, by enlarging the scope for preferential access steeply escalating, and nontransparent tariffs; for poor countries. Existing schemes in high- tariff peaks; tariff rate quotas on maximum income countries have limited coverage and, low-tariff imports; and a plethora of domestic together with other impediments to trade, un- and export subsidies in high-income countries, dermine their otherwise positive effects. to say nothing about state enterprise trading that still survives in many developing coun- Services tries. Support to agricultural producers in Services are the fastest growing components of high-income countries runs in excess of $300 the global economy, and trade and foreign di- billion annually. During downturns-such as rect investment in services have grown faster the one the global economy is now experienc- than in goods over the past decade. In virtu- ing-these subsidies tend to increase and force ally every country the performance of the ser- a disproportionate share of the cyclical adjust- vices sectors can make the difference between ment onto producers in developing countries. rapid and sluggish growth. More efficient ser- Tariff peaks also work against the poor. Fully vices-in finance, telecommunications, domes- one-third of exports of the poorest developing tic transportation, and professional business countries face tariff peaks in at least one of the services-improve the performance of the four major markets, the United States, Japan, whole economy because they have broad link- XV GLOBAL ECONOMIC PROSPECTS age effects. Collectively, they are essential to age 1.5 percentage points faster than other increasing domestic productivity. countries over the past decade. Developing countries, in particular, are likely to benefit significantly from further do- Transport mestic liberalization and the elimination of bar- International transportation costs to move de- riers to their exports. In a range of services- veloping countries' exports to foreign markets from financial sector and business services to often are a far greater barrier to trade than telecommunications and retailing-restrictions tariffs. Both public policies and private prac- on foreign investment are still common, par- tices exercise a significant influence on costs. ticularly in developing countries. Even more Policies toward maritime transport, such as stringent restrictions affect the export of ser- cargo reservation policies and limitations on vices, such as professional and construction the provision of port services, often protect in- services, through the movement of persons-a efficient service providers and unduly restrain mode of supply in which many developing coun- competition. Competition-restricting practices tries have a comparative advantage. among shipping lines increase freight rates by As with merchandise trade, reforms in ser- up to 25 percent on selected routes. Increasing vices have to be managed carefully. The largest concentration in the market for port terminal gains come from eliminating barriers to entry services poses the risk that the benefits of lib- and new competition, but many developing eral government policies may not be passed on countries have been content only to change to consumers. ownership through privatization while retain- International air transport services, despite ing limits on entry that buttress monopolies, being at the heart of the globalization process, Privatization without competition can vitiate are one of the most protected from interna- well-intentioned reforms. Effective regulation tional competition. The current regime of bi- is also critical to the success of liberalization. lateral air service agreements largely denies ac- Even though governments can initiate reforms cess to efficient outside carriers-and inflates of services unilaterally, multilateral agreements export costs for developing countries. through the General Agreement on Trade in Countries themselves can take actions to im- Services (GATS) could help accelerate domestic prove management of their ports and reduce reform and improve access to foreign markets costly delays associated with inefficient cus- for developing countries. In parallel, global co- toms. In Brazil, for example, failure to deploy operation to expand trade could mobilize sup- efficient container services has kept costs up to port for developing countries at four levels: in more than twice international norms in cus- devising sound policy, strengthening the do- toms, warehousing, inland transport, and ports. mestic regulatory environment, enhancing their Recasting institutional arrangements to maxi- participation in the development of interna- mize competition in the provision of port ser- tional standards, and ensuring access to essen- vices could also drive improvements. Adopting tial services in the poorest areas. non-discriminatory policies of open access in The payoffs to success, however, are espe- international air transport can enhance the effi- cially high. Studies comparing reduction of ser- ciency of air services. At the same time, there is vices barriers to reductions in barriers to mer- a need to regulate private practices of transport chandise trade find that services liberalization service providers by competition policies, to en- can provide benefits up to four times higher. sure that the gains from liberalization are not Estimates suggest that, after controlling for captured by private firms. other determinants of growth, countries that A special responsibility for promoting com- fully liberalized trade and investment in fi- petitive international transport markets falls nance and telecommunications grew on aver- on the large industrial countries. These coun- xvi SUMMARY tries, with their strong regulatory capacity and Moreover, enforcing all property rights is of- history of antitrust enforcement, are well posi- ten a major problem needed to improve the in- tioned to enforce competition disciplines on vestment climate, so governments have to ask multinational transport operators. To date, they whether it makes more sense when measured have not done so. against objectives of poverty reduction to forgo Beyond this, multilateral negotiations on allocating scarce resources for enforcement of transport services under the GATS can sup- (say) land rights in agriculture-where returns port domestic reforms by unleashing greater to investments often benefit poor owners di- liberalization and by lending credibility to do- rectly-in order to enforce IPRs. mestic policies. The scope for creating binding Because economic advantages and capabil- multilateral disciplines on transport services is ity of enforcement tend to rise as countries be- large. Only little progress has been made in come more developed, and low-income coun- the past on maritime transport, and even less tries markets are of marginal importance to has been made on air transport. patent holders, there is a compelling logic to rebalance the TRIPS agreement to accommo- Intellectual property date the problems of low-income countries. Intellectual property rights (IPRs) are designed This could take three forms: It may make sense to balance the needs of society to encourage to recognize the validity of a phased imple- innovation and commercialization of new tech- mentation of TRIPs based upon development nologies, products, and artistic and literary capacity. Second, negotiating compulsory li- works, on the one hand, with needs to pro- censing provisions to allow poor countries with mote use of those items, on the other. Since the no production capability of their own to li- overwhelming bulk of intellectual property is cense producers in other countries for sale in created in the industrialized countries, the their markets would improve their competi- Uruguay Round TRIPS shifted the global rules tion access to critical development inputs. This governing intellectual property in favor of de- may provide small developing countries with veloped nations. If TRIPS were fully imple- greater flexibility in addressing public health mented, rent transfers to major technology- crises. Third, since industrial countries are the creating countries-particularly the United main up-front beneficiaries of IPRs, they may States, Germany, and France-in the form of find it in their interest to provide assistance to pharmaceutical patents, computer chip de- the poorest countries for the implementation signs, and other intellectual property, would of TRIPS. Beyond this, developing countries amount to more than $20 billion. can realize concrete benefits from TRIPs by en- To be sure, there are reasons to believe that couraging domestic intellectual property devel- the enforcement of IPRs is associated positively opment and its protection abroad. with growth. However, these benefits tend not to materialize until countries move into the Reshaping global trade architecture middle-income bracket. Therefore, many coun- for development tries, especially low-income countries, see these This report thus proposes actions to reshape potential benefits as elusive promises against global trade architecture to promote develop- which they have to weigh heavy, up-front costs ment in four policy domains: launching a de- of enforcement and administration. Adminis- velopment round of trade negotiations within tration and enforcement, together with higher the WTO, moving forward on the global co- prices for medicines, agricultural inputs, and operation agenda to expand trade outside the other key technological inputs, could readily WTO, enacting new policies in high-income absorb a significant portion of annual public countries to provide aid for trade, and adopt- expenditures in many low-income countries. ing trade reforms within developing countries. xvii GLOBAL ECONOMIC PROSPECTS Reshaping global trade architecture for development: The four-part policy agenda 1. Convening a development round in the WTO Market access Agriculture * Reduce applied tariffs, phase out tariff rate quotas, and bind tariffs at applied rates in both industrial and developing countries * Phase out export subsidies in high-income countries and commit to eliminate domestic support linked to production levels * Reduce tariff escalation and cut off tariff peaks Manufactures * Reduce applied rates further, and bind tariffs to levels that equal or are close to applied rates * Reduce tariff escalation and cut off tariff peaks * Accelerate implementation,of ATC quota eliminations and reduce tariffs in lines now covered by quotas * Negotiate tighter disciplines on antidumping and other forms of contingent protection Services * Liberalize entry of foreign services suppliers through elimination of restrictions on entry and promoting increased competition, with wider use.of GATS to bind nondiscriminatory access and lend credibility to domestic programs * Enhance scope of services provision through the temporary,movemrent of service providers (both skilled and unskilled) * Secure openness of e-commerce in services, through wider arid deeper GATS commitments on cross-border supply * Strengthen multilateral rules to deal with anticompetitive practices in services * Adopt a nondiscriminatory trading regime for air transport, including traffic rights, under GATS Implementation procedures and phasing * Adopt a phased implementation of TRIPS and other administrative-intensive agreements for low-income countries, based upon development capacity. * Establish a consensus that the TRIPS Agreement allows developing countries with no domestic production capacity to grant compulsory licenses to foreign firms * Convert "best endeavor" promises to binding commitments to'provide low-income countries with financial and technical assistance to implement WTO accords hproving WTO transparency and participation * Require WTO disclosure of databases; reports and their ful associated information; and analyses for particular decisions * Provide assistance to strengthen capacity of all members to participate effectively in negotiations 2. Global cooperation to support trade outside the WTO Provide "aid for trade" through stepped up development assistance * Expand "Integrated Framework" assistance to all low-income countries * Provide assistance to enhance the efficiency of the customs clearance process in developing countries, notably the good customs practices that are laid out in the revised Kyoto Convention (World Customs Organization) * Expand multilateral assistance to overcome country-specific bottlenecks to improving competitiveness and trading potential (for example, in finance, transportation infrastructure, education for low income workers, and public sector trade-related institutions) and to promote trade xviii SUMMARY * Fund mechanisms to help developing countries use intellectual property protection to their benefit by protecting intangible assets such as traditional knowledge, designs, music, and ethnobotanicals, and patent protection for industrial goods as well as improve enforcement of IPRs * Establish a global health fund to purchase licenses from developers of new medicines essential to treating debilitating diseases in poor countries Expand global efforts beyond tiade to improve erironment, aise labor standards, and adopt adequate produd standards outside the WT) * Expand global environmental cooperation with financing to improve environmental protection in developing countries, and create multilateral forum of environmental exchange * Strengthen international actions on labor standards through the International Labour Organisation (LO), with project collaboration from multilateral development banks * Create a Standards Development Facility to introduce science and other professional evidence into standard setting for products, with adequate representation from developing countries; and provide assistance to developing countries' standard setting bodies 3. Policies for high-income countries Market access * Grant to all low-income countries duty-free and quota-free access to markets of all countries of OECD * Reduce uncertainty of market access by harmonizing rules of origin, and by reducing threats of antidumping Expand bilate-al "aid for trade" * Provide financial and technical assistance to developing countries for "behind the border" trade-related invest- ments necessary to take advantage of market access * Improve policy coherence by establishing coordinating mechanisms between development policies and trade policies to ensure effective development outcomes * Assist developing countries to strengthen competition agencies and improve legislation, and require antitrust agencies to provide to developing countries information on third market effects of domestic mergers as well as pending cases of price-fixing and restrictive business practices; and review the anticompetitive consequences of antitrust exemptions in transport and other sectors that adversely affect development Domestic policies that facilitate adftstment of labor to ecoOnMic change s Review domestic policies to ensure displaced workers have adequate social support to deal with rapid changes in labor market conditions, including unemployment insurance, social safety nets (particularly health and pensions), and access to training and education 4. Policies for developing counties * Adopt program of trade reform, including phased lowering of border protection for goods and services as part of a poverty reduction strategy * As part of trade reform program, adopt companion policies to cushion any impact on the poor of adjustment to new trade incentives, and ensure investment responses; solicit foreign assistance when necessary to implement administrative requirements of programs * Spur development of industries essential to trade, such as transport, telecommunications, financial sector, and business services, particularly through introduction of regulatory policies that, where feasible, harness competition * Invest in upgrading public sector institutions related to trade, including customs, administration of drawback programs, and financial supervision agencies * Encourage domestic intellectual property development through TRIPS-consistent standards appropriate to country needs, and pursue protection of domestic intellectual property abroad * Ensure adequate macroeconomic policy framework to provide sound investment climate xix GLOBAL ECONOMIC PROSPECTS While this report focuses on global issues, fifth birthday to become productive citizens of chapter 6 indicates ways regional agreements, the world. Continuing down the path of properly designed, can be steppingstones to greater integration will not be easy, but if the promote new trade and deeper integration international community succeeds in doing so, that reinforces multilateral collective action, the world will undoubtedly be more prosper- The box summarizes specific measures that ous and stable. can produce faster economic integration. Removing barriers to trade and services, in conjunction with companion policies to fo- Notes ment a supply response, would give a strong 1. On regional trading arrangements, see World Bank growth impetus to the global economy and 2000. On standards, see World Bank 2001, chapter 3. long-run development. Chapter 6 quantifies 2. Trade liberalization has a relatively small impact these effects, if with the large margin of un- on the rate of growth, but has a large impact on the net number of poor lifted out of poverty. The reasons are certainty and qualifications that estimating threefold as described in chapter 6: First, under the techniques impose. If remaining restrictions on base-case scenario, growth-assuming population were merchandise trade were phased out in the held constant-will reduce the number of poor from 2006-10 period, economic growth in develop- 2.8 to 1.9 billion, but population growth will push that ing countries would be about 0.5 faster than in number back up to 2.2 billion in 2015. Hence compar- the base-case scenario-including services lib- ing the net change to the change associated with fast eralization would add significantly to the boost integration records an impressive increment. Second, growth has a disproportionate and positive effect on in growth. Much of the benefits come from poverty, and we have assumed a poverty elasticity with trade reforms in their own countries and in respect to growth of two, consistent with historical ex- other developing countries-and in that sense perience. Finally, trade liberalization changes the com- developing countries as a group control a con- position of production to incomes of the poor. siderable portion of their own trade destiny. In 3. This is the discounted present value in 2005 of some regions, these new trade policies could cumulative income gains over the decade to 2015. well make the difference between achieving their objectives (for poverty reduction, lower- ing maternal and child mortality, and improv- References ing educational attainment) and falling short Ganesan A.V. 2000. "Seattle and Beyond: Developing by a large margin. Country Perspectives" in Jeffrey Schott (ed.), The WTO After Seattle. Washington: Institute for In- The long-term promise of well-imple- trainlEoois ternational Economics. mented trade reform is therefore tangible: a World Bank. 2000. Trade Blocs. Washington, DC: world with a much higher standard of living, World Bank. hundreds of millions lifted out of poverty, and _. 2001. Global Economic Prospects 2001. a greater share of children living beyond their Washington, DC: World Bank. xx Abbreviations and Data Notes ACP African, Caribbean Pacific Group of States ASA Air Service Agreement ASEAN Association of Southeast Asian Nations ATC Agreement on Textile and Clothing CAP Common Agricultural Policy CAPAS Coordinated African Program of Assistance on Services CGE Computable General Equilibrium CEPR Consortium and the Centre for Economic Policy Research CEWAL Associated Central West Africa Lines CIS Commonwealth of Independent States CPI Consumer Price Index DEC Development Economics DFA Duty-free access EA East Asia EAC East African Community EAP East Asia and the Pacific EC European Community ECA East and Central Asia EDT Electronic data interchange EEC European Economic Community ERF Economic Research Forum EU European Union FDI Foreign Direct Investment FIAS Foreign Investment Advisory Service FTAA Foreign Trade Agency of the Americas GATS General Agreement on Trade in Services GATT General Agreement on Tariffs and Trade GDF Global Development Finance xxi GLOBAL ECONOMIC PROSPECTS GDP Gross domestic product GSP Generalized System of Preference GTAP Global Trade Analysis Project GEP Global Economic Prospects HIV/AIDS Human immunodeficiency virus/acquired immune deficiency syndrome ICAO International Civil Aviation Organization ICT Information and Communications Technology ICTB International Clothing and Textiles Bureau IDA International Development Agency IDB Inter-American Development Bank ILO International Labour Organization IMF International Monetary Fund IPRS Intellectual Property Rights IT Information Technology ITA Information Technology Agreement JPS Japanese patent system LAC Latin America and the Caribbean LATN Latin American Trade Network LDCs Least Developed Countries MENA Middle East and North Africa MERCUSOR Latin America Southern Cone trade bloc (Argentina, Brazil, Paraguay, and Uruguay) MFA Multi-fiber Agreement MFN Most favored nation MRAs Mutual Recognition Agreements NAFTA North American Free Trade Agreement NAPM National Association of Purchasers and Manufacturers NASSCOM National Association of Software and Service Companies NGO Non-governmental organization NTBs Non-tariff barriers PREM Poverty Reduction and Economic Management PRSP Poverty Reduction Strategy Papers PSE Producer support estimates ODS Ozone depleting substance OECD Organization for Economic Cooperation Development OPEC Organization of Petroleum Export Countries PBRS Plant breeders' rights QUAD U.S., Canada, European Union and Japan SOEs State-owned enterprises xxit ABBREVIATIONS AND DATA NOTES SSA Sub-Saharan Africa SAR South Asia Region TBT Technical Barriers to Trade Agreement TFP Total factor productivity T&C Textiles and clothing TFP Total factor productivity TRAI Telecommunication Regulatory Authority of India TRIPS Trade-Related Intellectual Property Rights TRQs Tariff Quotas UNCTAD United Nations Conference on Trade and Development UNDP United Nations Development Programme UR Uruguay Round URAA Uruguay Round Agreement on Agriculture USDA United States Department of Agriculture WCO World Customs Organization WDR World Development Report WHO World Health Organization WTO World Trade Organization WIPO World Intellectual Property 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. xxiii Prospects for Developing Countries: Coping with a Global Slowdown Global economy stalls dence quickly spread to Europe, first evident in The global economy, already balanced precari- equity markets but soon transmitted to business ously between recession and recovery in the and consumer demand. The phased contraction summer of 2001, received a sharp negative of U.S. and then European import demand, in shock with the terrorist attacks in the United combination with the reversal of incipient re- States on September 11. The probability of a covery in Japan, heralded an unprecedented de- more severe global slowdown has since in- celeration of world trade in 2001 that has ad- creased, but because of the difficulties of antic- versely affected developing countries. ipating the responses of businesses and con- Growth of global trade fell from record 13.3 sumers to these unprecedented events, together percent growth in 2000 to 1 percent in 2001. with the unpredictable ramifications of the at- Because nearly half of U.S. investment was in tacks, forecasts are subject to an unusually computers, electronics, telecommunications, high degree of uncertainty. Nonetheless, U.S. and other high technology products, the sharp consumer demand, instead of fueling a recov- contraction of investment hit East Asia's tech- ery in global demand in the fourth quarter, nology-heavy exports swiftly and hard. The now seems likely to decline. The September 11 U.S. downturn also rippled through Mexico events snuffed out the first signs, clearly dis- into the rest of Latin America. The downturn cernible in late summer, of an incipient re- spread to the Euro area where economies were bound in U.S. manufacturing production. As reaching the top of the business cycle in early perceived risks rose, stock markets fell around 2001. Conditions then worsened as the tech- the world, and new private lending to most de- nology slump cut into demand, rising oil prices veloping countries has effectively ceased. Trade cut into the purchasing power of consumers, flows, already depressed by slowing demand, and the profits of European companies in the are under new pressures from rising security- large U.S. market sagged. The European Cen- related costs, disruptions in normal air traffic, tral Bank, still fearful of future inflation associ- and further post-attack slackening in demand. ated with oil prices and the Euro's value, did The origins of the global downturn can be not immediately cut interest rates. In Japan, traced to the sudden decline in U.S. financial slumping exports cut the tenuous string pre- markets in mid-2000. This signaled the end to venting the economy from slipping back into the worldwide bubble in equity values and cre- recession as the government bumped up against ated over-capacity in global high-tech sectors. In ceilings on fiscal headroom. These forces, to- the ensuing slowdown of the U.S. economy, gether with weak commodity prices, reduced investment demand plummeted, and consumer growth in virtually all major regions of the de- confidence waned. Weakening investor confi- veloping world. 1 GLOBAL ECONOMIC PROSPECTS Those countries that still depend heavily on facing severe financial problems, is unlikely to commodity exports were particularly hard-hit. become a source of global growth in the short Many have experienced falling commodity run. Oil prices, averaging $25 a barrel in 2001, prices since 1997, prices that never recovered are likely to drift downward to an expected from the East Asia crisis. These countries were long-run equilibrium of $20 a barrel, underpin- unable to rebuild reserves and other buffers to ning growth in oil-importing countries. cushion this year's further terms-of-trade losses, and suffered declines in income. Sub-Saharan -creating a global environment better for Africa as a whole, for example, seems likely to developing countries in 2002-03 witness a decline in real per capita income of These developments in the high-income coun- 0.7 percent during 2001. tries, if they evolve as anticipated, will create a For the first time since 1974-1975, the moderately propitious external environment world's major economies are decelerating in for a rebound in developing countries in late tandem. With Japan in recession, Europe de- 2002, and stronger growth in 2003. Aggregate celerating, the United States dealing with the growth rates for the developing countries are aftermath of the attacks, and many develop- expected to fall from 5.5 percent in 2000 to 2.9 ing countries seeing their own growth slow, percent in 2001; if global recovery takes hold the downturn has now become global in as anticipated by mid-2002, growth in devel- scope. Global gross domestic product (GDP) oping countries would probably pick up to is projected to increase by a tenuous 1.3 per- 3.7 percent in 2002, and then rebound to over cent in 2001, down from 3.8 percent in 2000 5 percent in 2003. Trade and financial links, (table 1.1). which had transmitted weakening impulses to growth from the large high-income economies The most probable scenario is a recovery to developing countries in 2001, appear now in mid-2002- likely to reverse in 2002, and to do so sharply The global economy, in the most probable sce- in 2003. nario, will begin to recover in mid-2002, prob- Trade growth is likely to accelerate mod- ably starting in the United States and then estly next year to 4 percent, and then gain sub- spreading to Europe and elsewhere. With infla- stantial momentum in 2003 to exceed 10 per- tion in abeyance, U.S. monetary authorities cent. The last decade of trade growth has have progressively brought interest rates down created structural changes that now favor ex- 400 basis points over the course of the year pansion for many developing countries. Many through October and new tax cuts and spend- countries not only gained global market share ing increases provide additional stimulus for during the 1990s, but also diversified heavily this year and next. The European Central Bank into manufactures. This enabled them to es- cautiously started to bring interest rates down cape the volatility inherent in commodity trade in the third quarter of 2001, and, after Septem- and price movements. It also cushioned the ber 11, did so in tandem with other central 2001 shock, and should allow them to benefit banks around the world. Both the low inflation from high growth when the projected rebound and the improved structural policies in most in- in global demand begins to occur over 2002- dustrial countries have created an environment 03. However, those countries remaining de- in which technology-driven productivity growth pendent on commodity exports are experienc- can gain traction rather promptly, once the ing severe stress today, and can expect little re- cyclical downturn has reversed. Moreover, rapid lief from forecast developments over the next technological developments, high depreciation years. rates of investment goods, and just-in-time pro- Developments in global financial markets duction systems tend to generate relatively are also likely to favor renewed growth begin- rapid rebounds after downturns. Only Japan, ning in 2002, if on a more selective basis than PROSPECTS FOR DEVELOPING COUNTRIES 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) April2001 Current 20 Current forecasts Forecasts estimate 2000 2001 2002 2003 2001 2002 2003 Global conditions World trade (volume) 13.3 1.0 4.0 10.2 5.5 7.3 7.3 Inflation (consumer prices) G-7 OECD countriesab 1.9 1.8 1.4 1.5 1.8 1.7 1.8 United States 3.4 2.8 2.2 2.3 3.0 2.6 2.7 Commodity prices (nominal dollars) Commodity prices, except oil (dollars) -1.3 -8.9 1.6 8.1 -0.3 5.4 5.6 Oil price (dollars, weighted average), dollars/barrels 28.2 25.0 21.0 20.0 25.0 21.0 20.0 Oil price, percent change 56.2 -11.3 -16.0 -4.8 -11.4 -16.0 -4.8 Manufactures export unit value (dollars)c -2.0 -4.6 4.0 4.4 5.9 3.1 2.4 Interest rates LIBOR, 6 months (dollars, percent) 6.7 3.6 2.8 3.0 48 4.7 5.0 EURIBOR, 6 months (euro, percent) 4.5 4.1 3.3 3.3 4.3 4.2 4S World GDP (growth) 3.8 1.3 1.6 3.9 2.2 3.3 3.4 High-income countries 3.4 0.9 1.1 3.5 1.7 2.9 2.9 OECD countries 3.3 0.9 1.0 3.4 1.6 2.8 2.9 United States 4.1 1.1 1.0 3.9 112 3.3 3.2 Japan 1.5 -0.8 0.1 2.4 0.6 1.8 2.3 Euro Area 3.5 1.5 1.3 3.6 2.5 3.1 29 Non-OECD countries 6.3 0.6 3.2 5.7 4.1 4.9 5. Developing countries 5.5 2.9 3.7 5.2 4.2 4.9 4.9 East Asia and Pacific 7.5 4.6 4.9 6.8 5.5 6.0 6.1 Europe and Central Asia 6.3 2.1 3.0 4.2 2.3 4.2 4.1 Latin America and the Caribbean 3.8 0.9 2.5 4.5 3.7 4.4 4.4 Middle East and North Africa 3.9 3.4 2.9 3.6 3.9 3.5 3.6 South Asia 4.9 4.5 5.3 5.5 5.5 5.5 5.6 Sub-Saharan Africa 3.0 2.7 2.7 3.9 3.0 3.4 3.6 Memorandum items East Asia crisis-affected countriesd 7.1 2.3 3.4 5.4 3.7 5.1 5.2 Transition countries of ECA 6.1 4.0 3.1 3.8 4.1 3.8 3.8 Developing countries excluding ECA 5.3 3.1 3.8 5.4 4.2 5.1 5.1 excluding China and India 5.0 1.9 2.9 4.6 3.4 4.4 4.4 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 dollars. d Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand. Source: Economic Policy and Prospects Group, October 2001; and GDF 2001 projections of April 2001. in the past. Lower international interest rates flight to quality, and rising risk premiums eased the pressure on developing countries' globally. These have increased financial strains debt servicing, particularly in the most credit- in some highly indebted countries. Investors, worthy countries. However, this positive news with memories of financial crises in East Asia is likely to be offset in the short run with a and elsewhere firmly in mind, are more dis- 3 GLOBAL ECONOMIC PROSPECTS criminating, and that at least partially ex- and growth. Yet, with these favorable growth plains why financial stress in Turkey and Ar- trends in most regions, some will be left be- gentina have not produced more widespread hind, and may find it difficult to meet develop- contagion. Capital flows are thus likely to ment goals, such as a reduction in child mor- constrain growth in some countries, but will tality, without additional policy measures and probably reward good policies in other coun- external support. tries. As in the case of trade flows, only by To understand ways the global slowdown 2003 should growth of capital flows be ex- is affecting the prospects for developing coun- pected to show a significant acceleration. tries, the first three sections below discuss ele- ments of the global environment that shape Short-term risks are high- the outlook for developing countries: the syn- Even though the most probable scenario is for chronous slowdown in the high-income coun- recovery by mid-2002, risks to this global out- tries, then deceleration of trade, and diverse look are unusually high and depend largely on trends in financial markets. We then show the still unfolding ramifications of the terror- how these forces shape the short-term outlook ist violence in the United States. Moreover for developing countries, as well as the risks structural and policy risks persist. The U.S. that could undermine this outlook. A final sec- current account deficit remains large, and tion concludes with a discussion of long-term global financial markets could impose a dis- prospects and potential consequences for re- ruptive adjustment. Japanese domestic finan- ductions in poverty. cial strains, should improvements in policy not be forthcoming, may have a destabilizing effect on the global economy. Furthermore if A simultaneous downturn in the both monetary and fiscal policies in Europe industrial countries are insufficient to offset the worsening of mar- ket sentiment, the slump could be deeper and Downturn broadened- longer than in present projections; a longer For the third time in two decades GDP growth and more pronounced downturn in Europe in the industrial world slowed below 1 percent would be especially harmful for developing in 2001. By September, the downturn had not countries due to the region's strong trade and become as deep as during the early 1990s and financial linkages with all developing regions. the beginning of the 1980s. However, a worri- Finally, financial tensions in some developing some characteristic of the current downturn is countries may significantly delay the recovery that all three industrial regions are simultane- in more tightly linked groups of countries. ously in a downward phase of the business These risks argue strongly for continued cycle (figure 1.1). More pronounced weakness policies in the high-income countries that will in the Euro Area and recession in Japan mir- support growth, and for policies in developing rored meager GDP growth in the United States. countries that quicken the pace of structural While the end of the high-tech boom, the reforms to improve their investment climate. collapse of stock markets, high oil prices, and currency movements were factors that con- -but long-term prospects are favorable tributed to the downturn in all three regions, Even though today's environment is weak and the relative importance of those factors dif- unusually uncertain, the long-term growth po- fered across regions. Plunging sales of semi- tential of developing countries is promising. conductors and related products primarily af- This is because improved macroeconomic man- fected the United States and Japan, as growth agement, rising savings, increased openness, in Europe depended less on high-tech manu- and greater diversification create better incen- facturing. Also, the collapse of stock markets tives for investment, technological progress, has probably affected the United States and 4 PROSPECTS FOR DEVELOPING COUNTRIES Figure 1.1 Industrial production in the G-3 countries falls in 2000-01 (percent change, 3-month/3-month, saar) 12 Germany 0 United States -12 Japan -18 Jan. 2000 April 2000 July 2000 Oct. 2000 Jan. 2001 April 2001 July 2001 Note: The G-3 countries are United States, Japan and Germany. Source: National statistics; Economic Policy and Prospects Group calculations. Japan more severely: wealth effects are tradi- Quarterly GDP growth rates flirted with re- tionally more important in the United States, cession during the first three quarters of the and the fragile Japanese banking sector is vul- year. Manufacturing production had sagged- nerable to low equity prices. Inflationary pres- a result of declining domestic investment and sure coming from high oil prices and the strong the adverse effects on exports stemming from dollar were a main impediment in Europe, the strong dollar and weak global growth- where the European Central Bank tried to es- which in turn yielded burgeoning excess in- tablish a tradition of keeping inflation strictly ventory levels that required sharp curtailments under control and hesitated to ease monetary in output. policy. The terrorist attacks on September 11 ex- acerbated the deterioration of economic con- -as growth stalls in the United States- ditions. The direct loss of U.S. output in the The downturn started in mid-2000 in the immediate aftermath of the September 11th United States, with sharp corrections in the tragedy is estimated to be $25 to $35 billion. stock market ending a long period of large cap- This is about one day's GDP (1.5 percent of ital gains, especially in the high-tech sectors. quarterly output)-as business, financial mar- Just when an expected soft landing seemed at kets, and air transport effectively came to a halt hand, market sentiment worsened toward the as the events of the day unfolded. Financial mar- end of the year. Uncertainty about future prof- kets remained closed for four trading sessions, a its sharply reduced investment demand and full ban on commercial air travel lasted for a borrowing, particularly by high-tech firms. week, and disruptions to the countries' "nor- The resulting negative wealth effects also mal" order of business became widespread. sharpened the downturn in other sectors. - The collapse of air transport aggravated U.S. business investment continued to the direct output effects, because the sector is weaken sharply during 2001, exports and im- highly labor intensive, and air travel is an es- ports plummeted at double-digit rates, and sential input to other economic activities. Ac- jobless claims rose at a recession-like pace. cording to the World Bank's sectoral Linkage 5 GLOBAL ECONOMIC PROSPECTS Model, a 20 percent cut in air travel supply -and Europe, led by Germany, weakens during one month would reduce annual GDP The weakening in Europe has been unexpect- by more than 0.25 percentage points. Accom- edly sharp in the second and third quarters of modations and restaurants are among the sec- 2001 (figure 1.2). Germany was the first to feel tors that were affected severely, with "recre- the impact of collapsing investment demand in ational services" representing over 3 percent the United States and East Asia. Its exports to of national value added-some three times those regions amount to almost 4 percent of that of air transport. GDP, more than for any other European coun- The indirect effects of the attacks are pro- try. The malaise in German manufacturing is spectively larger, operating through a fall in depressing activity in other countries and other consumer confidence and lower equity mar- sectors. Moreover, it became increasingly clear ket prices. The aggressive monetary policy re- that Europe could not escape the sharp cyclical actions following the attacks could well con- global downturn in high-tech products. In par- tain the deterioration of market sentiment to ticular the telecom sector, which had invested a limited period; but is unlikely to prevent a heavily in infrastructure and licenses for third- strongly negative impact on economic activity generation mobile communications, had to ad- during the fourth quarter of 2001. just its profit expectations. Trade volumes are falling rapidly across the region, reflecting the -while Japan is in recession- sharp downturn in global demand, as well as Even though the economy is likely to have declining intra-European Union (EU) flows. slipped back into recession in 2001, the Japa- Earlier strengthening of the dollar, remaining nese government persevered with its plans for high levels of the oil price and the fall in equity badly needed reforms. It moved toward some markets-in addition to slowing world trade, fiscal consolidation and began disposing of contributed to this recent sharpening of the banks' nonperforming loans, acknowledging downturn in Europe. that short-term economic costs are the price The unexpected strengthening of the dollar for achieving long-term gains. Bad loans have during the spring, and high energy prices both continued to mount in the domestic banking added to inflationary pressures, and the Euro- system, rising by official estimates from 9.7 pean Central Bank (ECB) was hesitant to lower percent of GDP in fiscal 1997 to 12 percent of interest rates further, even when it became in- GDP in fiscal 2000.1 Loan-loss reserve cover- creasingly clear that such a step would be nec- age of "risk management loans" dropped from essary to stabilize the economy. But the ECB re- 46 percent in fiscal 1997 to 24 percent as of duced its policy interest rates by 25 basis points fiscal 2000. Revised reporting criteria now in late August, and lowered rates again as part commit commercial banks and other financial of the coordinated policy response to the Sep- institutions to "mark-to-market" equity held tember 11 events. as capital, revealing losses that have likely been Declines in equity markets and financing substantial over the last years, and requiring conditions in the United States appear also to additional scale-backs in the loan portfolio (see have affected the domestic investment of glob- box 1.1). With the Nikkei at very low levels, ally active European firms. The EU has contin- this adjustment could be particularly sharp. ued to be the major investor in U.S. equity and The economy will probably contract by fixed-income markets, as well as in mergers and 0.8 percent in 2001. With limited policy in- acquisitions (M&A) over the last years-ac- struments other than structural reforms, only counting for some 70 percent of net foreign pur- a recovery of exports can offset continued chases of U.S. corporate bonds, 90 percent of stagnation in consumption and fall-off in pub- equities, and over $125 billion in M&A during lic works spending, to set the stage for even- 2000.2 Hence fall-out from slowing export tual recovery in private investment. market growth, accumulated losses in U.S. fi- 6 PROSPECTS FOR DEVELOPING COUNTRIES Box 1.1 Japan and rhe developing countries D aring the 1970s and 1980s, East Asia's develop- r i ,c mg and newly industrialized countries found that links through Japan's goods trade, banking flows, direct investment, and official development as- vonsofounentowar sistance formed a cement that fostered robust ad- vances in trade, financial integration, and growth 125 within the region. During the 1990s Japan experi- enced a burgeoning problem of nonperforming loans 0 in its banking system and meager output growth of 1.4 percent per year. Yet, Japan's impact on the re- 7 _ gion during bad times appears to be as strong as during the good times of the earlier decades. so O East Asia and Pacific O Latin America and the Caribbean 20 0 Sub-Saharan Africa/Middle East and North Africa O Europe and Central Asia Japan's share in GDP "exponed in r'gians exports Japan (perce) 19 19 19 1 199 2000 World region 1990 2000 1990 2000 Source Bank for International Setdetents. Quarterly Report Jane 2003 and Annex tables. Developing E. Asial NIEs 18.8 13.8 5.2 5.5 Japanese lending to East Asia doubled between China 14.7 15.6 2.6 3.9 1990 and 1996, rising to 4.4 percent of regional GDP, Korea, Republic of 18.6 11.1 5.0 4.0 and as high as 21 percent of GDP in Thailand. A criti- singapore 88 7.s 12.6 to.8 cal factor contributing to the upsurge in international Oil exporters 23.6 24A 114 13.1 lending was domestic-the demise of the traditional North America 10.3 6.7 0.9 0.7 or "Main Bank" lending system within Japan itself, European Union 2.1 1.8 0. 0.6 introducing more competition in the capital markets (Hoshi 2001). The financial crisis of 1997-98 yielded Major Latin America 6.8 3.0 0.6 0.5 a swift decline in Japanese bank claims on the "Crisis- Other major 5" countries dropping by 30 percent between 1996 Odeveloping 10 6. 0. L and 1998, representing a withdrawal of some 2.6 per- cent of GDP for these countries-although for Thai- Note: I Calculated as the share of total exports in GDP for country land it was 8.3 percent of GDP. From 1998 through "i" ties the share of Japan in country "Ps" exports. 2000, Japanese claims on developing countries contin- -Group includes Algeria, Arab Republic ofEyt India, Wa 00rJpnsecamnoieelpncoutis o n Republic of Iran, Morocco, the Ru c o npt n o,a t ued to fall, by 53 percent to the Asia-Pacific region Sourcer IMF Direction of Trade Statistics, Japan ERISA, World and 12 percent for other developing regions. At pres- Bank data, Economic Policy and Prospects Group calculations. ent, the persistence of nonperforming loans in the portfolios of Japanese banks constrains its ability to Japan has become less important as a destination generate new lending. Indeed, domestic lending by for regional exports over the last decade, as evidenced Japanese banks has fallen recently at 6 to 7 percent by a 5 percentage point drop in export share for East annual rates. Financial institutions have reduced over- Asia, and sharper fall-offs for other developing regions. seas exposures in order to build up capital and in- This is indicative of relatively low growth of Japanese crease funding for loan-loss reserves (Mori and others import demand and a diversification of export markets 2001). So Japan's share in developing countries con- by Asian economies. However, the share of East Asian mercial bank financing has dropped by one-half in the GDP', and particularly that of China "exported" to last decade, lessening the importance to emerging Japan, has risen slightly over the last decade. markets of further withdrawal of Japanese loans dur- On the financial front, however, the scale-back ing the present downturn. of Japanese commercial bank lending to developing countries has continued well beyond the retrench- Note: Crisis-5 countries are tidonesia, Republic of Korea, ments of the immediate post-crisis period. Malaysia, the Philippines, and Thailand. 7 GLOBAL ECONOMIC PROSPECTS lowered interest rates by 400 basis points in a Figure 1.2 European industrial series of nine cuts over the course of 2001, car- production falls rying the Federal Funds rate to 2.5 percent by (growth at seasonally adjusted annualized rates) mid-October, its lowest level since the early 6 1960s. European rates have also fallen, but not 40 04-00 as swiftly. In addition, on both sides of the At- 40 lantic increased fiscal stimuli are expected to 2 augment consumption. However, the U.S. re- covery, while expected to begin in the second or o -third quarter of 2002, will be reflected more in 2003 annual growth numbers than in those for 2002. Nonetheless, the U.S. recovery is unlikely -4- to be as quick and strong as earlier thought, and the European recovery will likely lag one -6 or two quarters. European manufacturers have ___ been faced with large-scale unsold inventories, Germany France Italy Euro United and it will take several quarters for this in- Area Kingdom ventory cycle to unwind. With lower external Source: National agencies; Eurostat. demand than earlier anticipated, a buoyant export-led U.S. recovery has become less likely. U.S. consumer demand, put on hold after Sep- nancial markets, and unfavorable conditions tember 11, is expected to lead the recovery-if under which to mobilize new corporate affilia- somewhat delayed. In Japan, there is no effec- tions have come to exert increasingly adverse ef- tive scope for monetary easing through inter- fects on the European business climate. The fall est rate cuts. Financial problems will continue in equity prices also had a negative impact on to weigh heavily upon the Japanese economy, consumer confidence, although wealth effects though a return to moderate positive growth in are, on average, less important in Europe than 2003 is expected with revival of world trade. in the United States. Euro Area GDP growth is In the medium term, prospects for industrial expected to fall to 1.5 percent in 2001, follow- countries remain favorable. Low inflation and ing strong output gains of 3.5 percent in 2000. improved structural policies in most industrial countries have created an environment in Rebound next year still likely, but which the potential benefits from investment probably later than earlier expected in technology can be reaped once the cyclical Before September 11, there were early signs downturn has reversed. Oil prices averaging that a recovery in the manufacturing sector $25 a barrel are expected to return slowly to was underway in the United States. Figure 1.3 long-run equilibrium of under $20 a barrel, shows, for example, that manufacturing pro- further underpinning growth in the industrial duction, excluding high-tech production, had countries. And in Japan, the new government's reached a trough, confirming information com- tougher approach toward the "bad loan" prob- ing from the purchasing managers' index lem should help the economy move gradually (NAPM). Since the terrorist attacks, that recov- toward potential growth rates. ery has been postponed. But prudent use of the Under these assumptions, the forecast antic- levers of economic policy is likely to bring ipates continued low growth rates in the United about at least the beginnings of a rebound in States, Japan, and Europe for 2002, 1.0, 0.1, 2002 for the United States and Europe. The and 1.3 percent respectively, but with strong United States has aggressively loosened mone- acceleration of GDP growth advancing into tary policy, with the Federal Reserve having 2003 (figure 1.4). [Weaker annual GDP growth 8 PROSPECTS FOR DEVELOPING COUNTRIES Figure 1.3 U.S. NAPM and manufacturing industrial production excluding high tech, 1994-2001 (percent change saar, 3-month/3-month) (index, sa. 3-month/3-month) 10 60 NAPM 5 55 0 50G Manufacturing 5 IP excluding tech 45 -10 40 Feb. 1994 Feb. 1995 Feb. 1996 Feb. 1997 Feb. 1998 Feb. 1999 Feb.2000 Feb. 2001 Source: National Association of Purchasing Managers, Datastream. rates in 2002 are due in large measure to statis- tical effects related to the initial conditions for Figure 1.4 GDP growth in OECD blocs the year-a "carry-over" of sluggish or declining (percent) growth in the final quarters of 2001. A pick-up in 5 the momentum of output growth across the major 2000 * 2001 * 2002 O 2003 OECD blocs on a quarterly basis is anticipated to 4 2000 - 2 - 2002 - 2003 commence in the second and third quarters of 2002, in turn yielding positive "carry-over" ef- fects and resulting in higher annual GDP figures for 2003.] The delayed recovery of these major 2 locomotives of the global economy will trans- form the earlier-expected outlook for the devel- 1 oping countries. Trade and financial markets will be the two main channels through which these dynamics are transmitted. The next two sections look at the global environment through 1 the lens of trade and finance, and a following United States Japan Euro Area section explores in detail the outlook for devel- ctountres Source: National agencies; Economic Policy and Prospects oping countries. Group projections. Global environment: trade vance in 2000 to a crawl of 1 percent growth Downturn hits manufacturing exporters- in 2001. Import demand in all three industrial The industrial country-downturn has led to regions slowed sharply, with the steepest de- the sharpest deceleration of world trade on cline in U.S. imports (figure 1.5). U.S. invest- record, from an extraordinary 13.3 percent ad- ment in equipment declined by 4.5 percent 9 GLOBAL ECONOMIC PROSPECTS The terrorist attacks on September 11 re- Figure 1.5 Import growth across strained trade flows further, exacerbating the industrial centers sharp cyclical downturn. Security concerns (percent) translated into higher freight rates,3 and limited 15 air travel greatly hindered the shipment of per- U 2000 U 2001 ishables and high-tech products.4 The impact E 2002 ]20031 on trade in services was significantly larger, with 10 tourism and business travel sharply lower.s Growth of exports from developing coun- tries plummeted from over 19 percent in 2000 to 2 percent in 2001. East Asian export growth fell more sharply still, from 25 percent o ,to 0.5 percent (table 1.2). These effects are so pervasive because developing countries are now more than ever linked to global trade cy- r Ucles in manufacturing. Countries increased Industrial United Japan Euro. countries States Area trade as a share of their economies, and in- creased their share of the-global market during Source: World Bank data; Economic Policy andth19 s.ndogs,mayevlpgcu- Prospects Group projections. the 1990s. In doing so, many developing coun- tries diversified away from commodities to manufactures, and further, into high-tech prod- ucts. The share of manufactured goods in de- after growing by over 11 percent last year. veloping countries' exports increased from 60 With almost 30 percent of such investment to 80 percent over the last decade, while the being imported, and 40 percent of total invest- share of capital goods increased from 27 to 42 ment consisting of high-tech products, this was percent (figure 1.6). a major force behind the slowdown of world trade and the collapse of the global semicon- -and commodity exporters ductor market. Indeed, U.S. imports of capital Commodity exporters have been hard hit by goods dropped at an annual rate of 32 percent price declines. Although the number of coun- during the first half of the year. tries that are highly dependent on commodity Table 1.2 Merchandise export volumes, annual average percentage change 1999 2000 200.1 2002 2003 World 4.1 13.5 0.3 3.7 10.3 High-income countries 4.1 119 -0.2 3.3 10.5 United States 3.9 113 -2.1 4.6 11.9 Euro Area 3.2 12.2 21 3.0 11.0 Japan 4.0 12.9 -7.7 4.0 11.8 Low-middle-income countries 3.8 19.2 2.1 5.0 9.7 East Asia and Pacific 9.7 25.5 0.5 6.4 11.5 South Asia 7.5 12.3 3.0 7.0 7.6 Middle East and North Africa -2.6 8.0 2.1 4.1 5.8 Europe and Central Asia -2.6 18.9 6.1 2.8 8.4 Latin America and Caribbean -0.8 12.0 1.9 3.8 9.3 Sub-Saharan Africa 0.0 0.g 3.4 2-9 6.4 Memo: Developing x ECA 5.3 19.2 1.2 5.5 10.0 Source: World Bank data; and Economic Policy and Prospects Group projections. 10 PROSPECTS FOR DEVELOPING COUNTRIES Figure 1.6 Export shares for developing Figure 1.7 Distribution of countries by countries excluding transition economies share of primary commodities in total (percent of total exports) merchandise exports Percent of countries 100 - 80 80 70 E 1981-85_ 801 70 M 1996-99 60 60 40 40 g Natural resources 30 - 20 Energy 2 *Agriculture 20- CO Manufacturing 10- 0 1971 1976 1981 1986 1991 1996 0 T- <60 60-0 80-100 Source: U.N. Commodity trade statistics (COMTRADE). commodity share Source: Staff calculations using UN COMTRADE data from WITS system. exports is declining, for more than 10 percent of developing countries commodities exports account for over 80 percent of total merchan- World trade expected to rebound later dise exports (figure 1.7). in 2002- Non-oil commodity prices are projected As noted in Global Development Finance to fall by about 9 percent in 2001 following a (GDF 2001), earlier cyclical downturns in the 1.3 percent decline in 2000. Substantial in- world semiconductor industry have been brief, creases in the supply of commodities such as largely due to technological advance and rapid coffee, vegetable oils, and timber, and currency inventory liquidation. More generally, faster weakness of major exporters relative to the depreciation rates of capital goods appear to U.S. dollar,6 also contributed to the fall of com- have shortened the investment cycle, while modity prices. In the case of metals, the price lower inventory ratios have reduced the struc- declines come in spite of production cuts, par- tural importance of inventory cycles. Com- ticularly in aluminum, but also in copper and bined with the positive impact of lower inter- other metals.7 Using its market power, the Or- est rates on demand for durable goods, the ganization of Petroleum Exporting Countries changing technological characteristics made a (OPEC) was able to sustain, at least in the short recovery of manufacturing production in the run, prices around $25 a barrel. However, United States before the end of 2001 proba- slackening demand, especially after September ble-and the first signs of a rebound were vis- 11, clearly will exert downward pressure on ible before September 11. However, the terror- price. OPEC's reassurance that it will guarantee ist attacks have now rendered such a scenario sufficient supply quickly, eased market con- of quick recovery unlikely. Although the mech- cerns after September 11, but the possibility of anisms behind the recovery remain unchanged, future supply disruptions in the aftermath of a delayed upturn is more plausible now. the terrorist attacks has not disappeared, keep- As a recovery in the United States begins to ing uncertainty at exceptionally high levels in gather pace by mid-2002, increased investment the short run. and Information and Communications Technol- 11 GLOBAL ECONOMIC PROSPECTS ogy (ICT) equipment spending should feed through to import demand, setting the stage for Figure 1.8 Episodes of world growth a pick-up in world export growth, mainly man- slowdown and agricultural and mineral ifesting in robust 2003 annual growth rates export prices (table 1.2). Expected growth of 4 percent in (difference in growth from the previous two years- 2002 is followed by growth over 10 percent in percent) 2003 in this projection. The rebound is likely to 10 be fastest for East Asia, with growth rates of * Agrculture U Minerals 6.4 and 11.5 percent in the coming two years. o _1 The dynamics in the region reflect countries' specialization in high-tech products and the -4.7 3 pronounced boom-bust cycles rippling through -10 -8.0 East Asia since the 1997-98 crisis. For regions depending more on commodity exports, export -15.9 growth will be lower under these circum- 20 -17.s stances. For example, Latin American exports are expected to grow about 4 and 9.5 percent 30 -28.8 in 2002 and 2003, while Sub-Saharan Africa 1980-82 1990-92 1998-99 could see exports growing near 3 and 6.5 per- cent, lower than Africa's record growth of Sample of 31 non-oil commodity dependent SSA 2000, but still high relative to past trends. countries, classified as mineral or agricultural exporters (see GEP 2000, Chapter 4, fn. 28 on p. 129 for details of coverage). Source: Economic Policy and Prospects Group staff remain vulnerable estimates. After the sharp 9 percent fall in commodity prices in 2001, almost no rebound is expected for 2002, and only by 2003 are current losses 1996; Kose and Riezman 2000). Figure 1.8 in- likely to be made up. Market conditions con- dicates that the negative impact during down- tinue to put downward pressures on commod- swings on mineral exporters is generally larger ity prices in local currencies, and the modest than on agriculture exporters, reflecting the price rebound expected for 2002 is mainly greater sensitivity of minerals demand to in- based on expected currency movements. With dustrial production. manufactures export prices expected to in- Oil prices are expected to gradually decline crease by 4-4.5 percent per year, in light of to under $20 a barrel over the forecast period, anticipated depreciation of the dollar, com- due to rising supply competition from both modity exporters are likely to experience fur- non-OPEC producers and within OPEC itself. ther terms-of-trade losses. The decrease in the Earlier episodes, notably the second oil crisis import-purchasing power of exports as a result during the early 1980s, show that, in the short of relative price changes might constrain con- term, it is possible to evoke large price swings sumption and investment demand, for instance with small supply shocks, but that large sup- because of reduced availability of intermediate ply adjustments are needed to keep the price at manufactures or capital goods. Research on high levels for a more extended period (figure North-South business cycles suggests that up 1.9). The sharpening of the downturn after to 20 to 50 percent of output volatility for such September 11th and OPEC's policy reactions countries may be explained by business fluc- to avoid sharp price bikes in the wake of the tuations in developed countries (see, for ex- terrorist attacks, have accentuated the ex- ample, Deaton and Miller 1996; Kouparitsas pected downward trend. 12 PROSPECTS FOR DEVELOPING COUNTRIES Figure 1.9 OPEC output and crude oil price Thousands of tons dollars per barrel 1,400 40 Production (left scale) --35 1,200 30 --25 1,000 20 Nominal crude price - 15 800 (right scale) 10 5 600 isli0 1971 1975 1979 1983 1987 1991 1995 1999 Source: World Bank; OPEC Bulletin. Global environment: emerging markets. Therefore, the benefits of financial markets lower OECD interest rates have been more- than-offset by higher risk premiums particularly Increasing risk generally outweighs for poorer performing countries. interest rate reductions- This is in sharp contrast to the 1991-93 The weaknesses of demand in the industrial slowdown in industrial countries, another countries had translated into lower interest rates episode of slowing growth and interest rate before September 11, as authorities provided a cuts. Then, when U.S. interest rates fell by 450 more accommodating monetary stance. After basis points cumulatively over three years, the terrorist attacks, monetary authorities in the global gross capital market flows increased by industrial countries swiftly eased monetary con- 22 percent a year, albeit from very low levels ditions further. These policies can potentially born of the 1980s' debt crisis. Developing soften both the slowdown in the OECD region countries benefited from the increased liquidity itself and the transmission of the downturn to with a similar increase in inflows by 32 percent developing countries, enabling the latter to ben- a year. efit from lower international interest rates and a In 2001, the opposite occurred when the shift of capital flows away from industrial coun- Federal Reserve reduced policy rates by 400 tries. However, the financial channel of trans- basis points. After rising in 2000 for the first mitting growth dynamics from the major mar- time since the 1997-98 crisis-ridden period, kets to developing countries produced higher net capital flows to developing countries are spreads instead of larger capital flows in 2001, expected to decline in 2001. Gross flows from and contained a strong bias in favor of well- international capital markets, which increased performing, creditworthy countries. This re- by over 30 percent last year, are down by near flects the behavior of investors, who are now 20 percent (year on year-y/y) in the first half more risk-sensitive and discerning than they of this year (figure 1.10). Foreign direct in- were in the years before the 1997-98 crises in vestment (FDI) flows have continued to fall 13 GLOBAL ECONOMIC PROSPECTS Figure 1.10 Gross capital market flows to developing countries billions of dollars (monthly averages) 20 15 10 p.. ..- Equity * Loans * Bonds 5 1998 1999 2000 Q4'00 Q1 '01 Q2'01 July'01 Aug.'01 Source: Euromoney; Economic Policy and Prospects Group. (by 4 percent [y/y] in the first quarter) from nities in emerging markets after many develop- their peak in 1999, and official flows are not ing countries eased capital market restrictions. expected to post a significant rise in 2001. The spreads were relatively low and developing The difference between the current down- countries' share in global capital flows were in- turn and the recession a decade ago is the risk creasing (figure 1.11). This year, the spreads perception of private investors. Toward the end were high and even rising toward the middle of of the 1991-93 recession, investors were gener- the year, reflecting the heightened risk percep- ally optimistic about new investment opportu- tions of the markets. As a result, the already Figure 1.11 Emerging market spreads and share of global capital flows Basis points Percent 1,500 15 EMBI spreads (LHS) 1,200 12 900 9 600 6 Emerging market 300 shae (RHS) 3 1991 HI 1993 H1 1995 H1 1997 H1 1999 H1 2001 HI LHS: Left scale RHS: Right scale Source: Euromoney. 14 PROSPECTS FOR DEVELOPING COUNTRIES small share of developing countries in global had little success in accessing the market much capital flows has declined further. in the past year raising relatively large sums. For example, a private Russian corporation success- -but mainly because of financial tensions fully issued a bond-the first access in a year- in some large emerging markets- while large sums were raised by the Arab Re- Although aggregate net capital flows are on the public of Egypt ($1.5 billion), Malaysia ($1 decline this year, the reduction comes mainly billion), and Hungary ($0.9 billion). Moreover at the expense of a few large recipients, no- many small issuers with less-than investment tably Brazil, Argentina, and Turkey. Investors grade credit ratings either tapped the bond shunned the latter two countries on the worry market (Jamaica, Lebanon, Romania, and the of default on their public debt. Brazil, which Republica Bolivariana de Venezuela) or ob- received FDI flows of over $30 billion a year in tained syndicated loans (Chad, El Salvador, Es- 1999 and 2000-more than covering current tonia, Gabon, Peru, and a Russian bank that re- account deficits-has seen a marked fall-off in ceived a loan for the first time since the 1998 FDI to $15-20 billion in 2001. A worsening of crisis in that country). the external environment for Brazil-weak ex- These "windows of opportunity," however, port market growth, low commodity prices, alternated with periods when they were closed and rising spreads due to contagion from Ar- decisively. Investor sentiment deteriorated gentina-and a domestic energy crisis have sharply in July, with average spreads rising by limited the country's access to capital markets. nearly 200 basis points in response to adverse Market reactions in the first weeks after the developments in Argentina and Turkey. Al- terrorist attacks showed a similar pattern. though spreads also increased for several Cen- Spreads on Brazilian and Argentine bonds in- tral European and other Latin American coun- creased by more than 150 basis points, but the tries, the contagion was limited. The terrorist average increase for all other emerging econo- attacks and threats of further violence will mies was just 60 basis points, only slightly more likely yield a further decline in private flows to than the decline in international interest rates emerging markets in 2001, both capital mar- (LIBOR). ket flows and foreign direct investment. Capi- Capital market financing to developing tal market commitments could drop to some countries other than Argentina, Brazil, and $160 billion-a third below 2000 levels. FDI Turkey declined by 9 percent in the first half to major emerging markets had declined from of 2001, contrasted with a 20 percent decline $61 billion in the first half of 2000 to $56 bil- including these three countries. Bond issuance, lion in the first half of 2001, and the attacks that seems to have benefited from the lower have raised the likelihood of a further down- interest rate environment, grew by 75 percent turn: they have greatly increased the uncer- excluding the three countries, and by 35 per- tainty involved in traveling to supervise for- cent for all emerging markets. eign subsidiaries; raised the cost of globally Countries that usually had difficulty in ac- integrated supply chains due to higher insur- cessing the global bond market in the past few ance rates and enhanced security measures at years were successful during "windows of op- the border; and demonstrated that these sup- portunity" that opened up periodically in the ply chains are vulnerable to interruption. first half of 2001. These episodes usually oc- curred when interest rates fell sharply in the in- Capital flows are unlikely to recover dustrial countries-thereby increasing liquidity in 2002 in the market-and when difficult conditions in Although near-term gross capital market flows key developing countries abated. June 2001 was are notoriously difficult to forecast, the global one such month, when gross capital flows political and economic environment seems too surged to over $21 billion, with countries that uncertain for a rapid upturn in capital flows to 15 GLOBAL ECONOMIC PROSPECTS developing countries. In the short run, emerg- tourism collapsing, and capital-market risk ing markets' share of private flows-which in- aversion heightened. Only by mid-2002 will creased rapidly during the last decade-is likely recovery gain some underpinning, as the to drop sharply as private investors seek safe United States and the Euro Area emerge from havens. By 2003, however, the right mix of con- their slumps. Economic prospects for the Mid- tinued low interest rates, a rebound in world dle East and North Africa (MNA) region- trade and reduced risk perceptions could gener- which, after Sub-Saharan Africa (SSA) suffers ate a recovery of capital flows, and perhaps an the smallest reduction in near-term growth- increase in developing countries' share of global are expected to deteriorate further in 2002, as flows. Yet private market flows will probably oil prices continue to fall and other commodity be much more selective than in the past. In- prices drop relative to the cost of manufactures vestors, with memories of financial crises in imports. The East Asia and Pacific (EAP) re- East Asia and elsewhere firmly in mind, will re- gion is anticipated to be first among develop- main more discriminating. Capital flows are ing regions to show a recovery in exports in thus likely to reward good policies in some late-2001 and early 2002, as the group was the countries, but continue to constrain growth in first to suffer from the collapse in high-tech others. Moreover risks remain weighted to the trade flows. South Asia (SAS) is expected to downside, as default by a major emerging mar- experience a less pronounced cycle, as the re- ket would escalate risk premiums for the ma- gion is relatively less integrated with the global jority of developing countries. economy. But GDP growth at 4.5 percent in 2001 corresponds with the 1997 low regis- tered by the region. (See appendix 1 for more The outlook for developing detailed discussion of the regions). countries A lthough the slowdown in global economic Diverse impacts of industrial activity is being led by the industrial countries' slowdowns countries, aggregate growth for developing Historically, the effects of downturns on devel- countries is being adversely affected and ex- oping countries have been quite diverse (figure pected to weaken from 5.5 percent in 2000 1.12). Financial conditions are discriminating to 2.9 percent in 2001. Delayed recovery in the factors that potentially even reverse the sign of OECD area is likely to keep developing coun- the impact. In the early 1980s, the developing try growth in 2002 restrained to 3.7 percent. countries followed the industrial countries into But stronger recovery in the advanced econo- recession; after the second oil crisis the indus- mies by 2003 should ignite a rebound to 5.2 trial countries tightened monetary policy to percent growth in the developing world. How- bring inflation under control. Higher interest ever, there are considerable regional variations rates generated severe debt-service problems underlying these summary figures. for oil-importing developing countries that had The Europe and Central Asia (ECA), and accumulated foreign debts, and the downturn Latin America and the Caribbean (LAC) re- in the developing countries was almost as deep gions have been hardest hit by the deteriorat- as in the industrial world. Moreover, growth ing global environment in 2001-although opportunities in highly indebted countries were conditions in these regions have been clouded limited for a longer period. by difficult financial and economic develop- In the beginning of the 1990s, growth in de- ments in Turkey, Argentina, and Brazil, respec- veloping countries, excluding the transition tively. Latin America in particular will begin group, accelerated despite recessions in the 2002 within a challenging external environ- United States and Europe. It was again mone- ment, with the United States likely in recession, tary transmission that played an important commodity prices falling sharply, international role. With inflation under control, the indus- 16 PROSPECTS FOR DEVELOPING COUNTRIES Figure 1.12 Industrial- and developing-country GDP growth, 1981-2003 (percent change) Developing excluding transitionA 4 / 2 1981 1984 1987 1990 1993 1996 1999 2002 2003 Source: World Bank data; Economic Policy and Prospects Group projections. trial world pursued an accommodative and improvements in many developing countries more predictable monetary policy. Encouraged justify the expectation that they will return to by major reforms in developing countries- relatively high growth rates, once the global including the opening up of capital markets- economy recovers from the current slowdown. international capital diversified away from in- dustrial country markets and found its way Current downturn follows region-specific to those developing countries that undertook channels on developing country growth major reforms. This enabled many low- and A probable consequence of the simultaneous middle-income countries not only to escape the downturn in the industrial world is that a downturn, but to grow at significantly faster broad range of developing countries will face rates than the high-income countries-until the an abrupt ending to the strong recovery that East Asia crisis brought about a sudden rever- followed the financial crises. GDP growth is sal in capital flows. expected to drop by 2.6 percentage points in Although the current downturn resembles 2001, with serious downside risks, discussed that of 1991-93 rather than the 1982 episode, in the next section. Apart from an 11 percent- developing countries today are more adversely age point drop in export market growth, and affected by falling import demand in the in- sharp fall in non-oil commodities-implying dustrial countries. This is because trade link- substantial terms of trade losses-developing ages have become increasingly important. And countries face a decline in capital inflows of in the aftermath of the East Asia crisis, a sharp almost 20 percent and spreads are again on rerouting of capital flows from industrial the rise after a steady decline since the finan- countries to developing countries (as occurred cial crises (table 1.3). in the early 1990s) is less likely. On the other The regional impacts follow closely the ex- hand, developing countries are now better port patterns that vary significantly across equipped than 20 years ago to absorb negative regions, both the commodity composition of external shocks, benefiting from diversification exports, and the orientation of exports across and domestic reforms associated with integra- various markets in the industrial and develop- tion into the global economy. The structural ing worlds (figures 1.13 and 1.14). Countries 17 GLOBAL ECONOMIC PROSPECTS Table 1.3 All developing countries: key indicators (annual percent change unless indicated) 1990-99 2000 2001 2002 2003 Export market* 7.5 13.4 2.3 4.4 9.4 Merchandise export volume 7.4 19.2 2.1 5.0 9.7 Terms of trade (percent of GDP) -0.2 0.5 -0.5 -OS -0.4 International market spreads (avg. bp) 807.6 707.2 733.6 .. ... Gross capital market flows 15.2 28.8 -18.5 .. Real GDP 3.2 5.5 2.9 3.7 5.2 *merchandise import growth in destination countries, weighted by export shares of exporting countries. ... Not available Source: Economic Policy and Prospects Group. in East Asia-and to a lesser extent Latin prices put additional pressure on countries in America-with large manufacturing exports, Central Europe, Sub-Saharan Africa, and Latin were the first to feel the impact of the collapse America. of import demand in the United States and External debt-to-export ratios, and the com- Japan. East Asia's high-tech laden exports position of the debt (private or official) differ (about one-third of total shipments from the widely across regions (figure 1.15), and can region) were especially adversely affected as largely influence the severity of the impact. The demand for computers, telecommunications more difficult external environment is especially equipment, and other semiconductor-based worrisome for highly-indebted countries relying capital goods dissipated. Now, increasing on private capital flows, such as Argentina, weakness in Europe and declining commodity Brazil, Turkey, and Indonesia. At a regional Figure 1.13 Composition of developing-country exports (average export shares, 1998-2000-percent) Mfgrs Oil Non-oil commodities l Services 100 50p 0 I I I All East South Middle East Europe and Latin America Sub-Saharan Developing Asia Asia and Central Asia and the Africa North Africa Caribbean Source: U.N. COMTRADE database. 18 PROSPECTS FOR DEVELOPING COUNTRIES Figure 1.14 Major destinations for developing-country exports (average export shares, 1998-2000-percent) U Europe U United States U Japan 0 Intra-Region U Other 100 1~ 00 75 All East South Middle East Europe and Latin America Sub-Saharan Developing Asia Asia and Central Asia and the Africa North Africa Caribbean Source: International Monetary Fund; Direction of Trade Statistics. Figure 1.15 Total external debt in developing countries, 2000 (as a percent of exports-) 200 0 Short term E Official 150 Private 100 50 0- LMICs* East Asia Middle East Europe and South Latin America Sub-Saharan and Pacific and Central Asia Asia and the Africa North Africa Caribbean Note: *Low- and middle-income (developing) countries. *exports of goods and services plus workers' remittances. Source: Global Development Finance 2001, World Bank. 19 GLOBAL ECONOMIC PROSPECTS Figure 1.16 Emerging market stripped spreads, 1999-2001 (basis points over US$ Treasury bonds) 2,500 2,000 Europe and Central Asia 1,500 Latin America and the Caribbean 1,000 500 0 Jan. 1999 July 1999 Jan. 2000 July 2000 Jan. 2001 July 2001 Source: Euromoney; Economic Policy and Prospects Group calculations. level, Latin America has one of the highest debt- tries are likely to bounce back from their fall to to-exports ratios, and at the same time the 2.9 percent in 2001 to a projected 3.7 percent in largest share of private debt in total debt. As a 2002 and 5.2 percent in 2003. The dynamics of result, debt from private sources as a share of this cycle are likely to be much different from exports in Latin America is more than twice as earlier downturns in the global economy-and large as in any other region. This makes Latin to be heavily conditioned by structural changes America particularly vulnerable. Potentially the in trade and financial markets. region can benefit from lower international in- Although the initiating factor driving recov- terest rates, but is at the same time most sus- ery among developing countries will be the re- ceptible to reduced availability of funds for refi- sumption of growth in the industrial centers, nancing. This vulnerability translates into a this should be augmented by conditions specific larger increase in Latin American spreads than to developing regions and major countries. For in other regions (figure 1.16). The hetero- example, an important factor rekindling growth geneous debt picture across developing coun- in East Asia should be a recovery of intra-region tries requires diverse policy responses. While for trade, following the return of more buoyant de- countries with large reserves and low debt, mand conditions in the United States and Eu- some macroeconomic easing may be warranted rope. In Central and Eastern Europe progress to stimulate the domestic economy; others face toward accession to the EU, albeit more pro- harsh, but ineluctable, fiscal adjustments. tracted, should serve to underpin policy reforms and set the stage for more robust growth as Region-specific factors will complement Western Europe emerges from its slowdown. coming recovery The degree of near-term success of reforms in Developments in the world economy, if they the larger developing and transition econo- evolve as anticipated, will create a more favor- mies-Brazil, China, India, Indonesia, and the able external environment for renewal of Russian Federation-will also prove critical ele- growth in developing countries later in 2002. ments in the outlook, as these countries address Aggregate growth rates for the developing coun- issues ranging from fiscal and financial sector 20 PROSPECTS FOR DEVELOPING COUNTRIES reforms, privatization, or trade liberalization. Low inflation in most countries provides the China's imminent accession to the World Trade opportunity for further monetary easing. How- Organization could spur its growth and con- ever, strengthening of financial systems may tribute to further global integration. be equally as important as domestic stimulus at the moment. Under cautious assumptions East Asia and Pacific ... trade slump about trade recovery and domestic policies, leads to sharp growth slowdown... and assuming no clear-cut financial crises, the EAP experienced an unprecedented decelera- forecast implies growth rates of 5 and 6.8 per- tion of exports in 2001, as noted in previous cent in 2002-03 respectively. sections. Export growth plummeted 25 per- centage points, mainly following the collapse in Latin America ... global context and global high-tech markets and the fall in Japa- financial stress in Argentina exact a toll nese and U.S. import demand. Worsening of fi- on growth nancial conditions, reflected in higher spreads, Developments during 2001 proved much was limited to highly indebted countries such as more challenging than anticipated in the Indonesia and the Philippines. Relatively low spring of the year, and have turned yet more debt levels, current account surpluses, and large adverse following September 11. Slackening foreign reserves tended to insulate other coun- world trade and weakening commodity prices tries from contagion. contributed to slower growth in many Latin The sharp drop in manufacturing exports re- American countries. The region's merchandise sulted in significant output declines---close to export volume fell from a 12 percent advance recession levels in some of the small, open econ- in 2000 to about 2 percent, following a simi- omies. However, carryover effects from last year lar decline in export market growth. At the resulted in positive annual growth rates for same time the region endured terms-of-trade most countries, while continued strong, albeit losses, equivalent to 0.2 percent of GDP in somewhat less dynamic growth in China keeps 2001. For commodity-dependent Central regional growth near 4.5 percent in 2001. American countries, this measure dropped by The strength of next year's rebound will a full percentage point in the year. Caribbean primarily depend on the vigor of the trade re- countries were hit hard by a sharp reduction covery and on policy responses to the deterio- in tourism earnings after the terrorist attacks ration of the financial environment. Signs that in the United States. the high-tech markets have potentially passed More important than the deterioration of trough levels are becoming clearer, but it re- the trade environment were changing condi- mains uncertain whether the upswing in these tions in financial markets. Rapidly declining markets will be as strong as they were during U.S. interest rates provided a degree of relief the last decade. And slower U.S. growth in the for some highly indebted countries by reducing wake of September 11 developments will dampen interest payments on debt. But international near-term export prospects. capital markets were less forthcoming than an- Export growth during the coming two ticipated, and a number of countries had diffi- years is expected to register 6.5 and 11.5 per- culties in financing maturing debt. Argentina's cent respectively. This is only 1 to 2 percent- fiscal strains and Brazil's drought-induced age points above export market growth, while shortage of hydroelectric power were among in previous years the difference between ex- the domestic factors that created an inflamma- port performance and export market growth ble mixture with the worsening of the external was much larger, averaging 6 points over the environment. GDP growth in the region was 1990s, reflecting the high-tech specialization constrained to 0.9 percent-a decline of 2.9 of East Asia. In that sense, the current forecast percentage points from the 3.8 percent perfor- is a cautious view on near-term developments. mance of 2000. 21 GLOBAL ECONOMIC PROSPECTS GDP growth in 2002 accelerates to 2.5 per- and Hungary, leading to appreciation of cur- cent in the current projections. However, this rencies during the first half of the year. With fi- point estimate is surrounded by a high degree nancial stress in Turkey, Western Europe need- of uncertainty, as it assumes that those coun- ing time to recover, oil prices expected to tries under financial pressure in 2001 are able decline, and a possible hiatus in progress to- to avert further adverse developments. Latin ward EU accession,8 a sharp and quick recov- America's high debt and continued large fi- ery of economic activity in the region is un- nancing requirements will keep risk percep- likely. In the Commonwealth of Independent tions elevated, due to the region's strong re- States (CIS), GDP growth is expected to decel- liance on volatile private capital markets. erate further to 3.2 percent in 2002, due in Argentina's struggle to establish sound footing large part to the easing of oil prices. In the for its fiscal position and debt burden contin- absence of high oil prices, significant institu- ues to depress its immediate prospects. Many tional and structural impediments remain a con- large countries are facing elections in 2002 straint to achieving higher sustained rates of (Brazil and Colombia), with the potential for growth. GDP growth in the CIS is likely to domestic shocks remaining high. But macro- achieve a 3.5 percent pace in 2003. economic management in many of the larger In Central and Eastern Europe, the poten- countries has improved steadily over the tial for solid growth exists in the medium run. course of the 1990s, laying the groundwork Two key assumptions underlying this perfor- for higher sustainable growth in the medium mance are that the EU accession process stays, term. generally, on track-albeit with some transi- In 2003, rebounding world output and trade tory difficulties. And that Turkey is successful activity should be supportive of a substantial in reestablishing macroeconomic stability, recovery in Latin America, reflected in 4.5 per- paving the way for a recovery over the coming cent GDP growth, powered in part by a revival period, driven in part by strong export of export growth to over 9 percent. growth, due to some extent to the fall of the Turkish lire. A gradual recovery to stronger Europe and Central Asia . .. prospects performance in the region would yield growth worsened by EU downturn of 3 percent in 2002, rising to near 4.5 by The drop in export market growth from 12 2003. Contagion from the financial strains in percent in 2000 to 5 percent in 2001 was not Turkey has been limited thus far, with average as sharp for the ECA region as for others. This spreads following a downward trend. This de- mainly reflects that the slowdown in Western velopment has been supported as well by higher Europe, accounting for 50 percent of ECA's oil prices serving to ease some of the financial exports, started half-a-year later than in the tensions that caused the Russian crisis. None- United States and Japan. theless, financial risks remain substantial and Despite the smaller than average drop in could drastically change the outlook. exports (about 13 percentage points), the de- celeration of output growth was larger than in South Asia . .. less affected by any other region, from 6.3 percent in 2000 to global slowdown 2.1 percent in 2001. This reflects the combined SAS, less integrated into the global economy, outturns of a strong contraction in Turkey's is generally less affected by the deteriorating GDP after recent financial upheaval; a return global environment, although uncertainty is to more moderate growth rates in Russia-and exceptionally high because the region could be other hydrocarbon exporters-after an unusu- directly affected by the ramifications of the ter- ally robust expansion of near 8 percent in rorist attacks in the United States. GDP is pro- 2000; and tight monetary policy in several jected to slow only moderately, from 4.9 per- Central European countries, notably Poland cent in 2000 to 4.5 percent in 2001. Growth is 22 PROSPECTS FOR DEVELOPING COUNTRIES then expected to remain near a rate of 5.4 per- ineluctably implies future uncertainty for the cent over the short-term forecast period. This subcontinent. Structural problems also entail reflects not only a relatively low level of global risks. Financial strains in India and Pakistan, integration across much of the region, but also left unattended, would significantly reduce the some positive domestic factors within many of opportunities for the region to benefit fully in the countries. Agricultural sectors, at least out- coming years from a recovery in the global side Pakistan, are expected to perform better economy, and jeopardize the promise of re- than last year, when weather conditions were gional growth. The ending of a preferential highly unfavorable. However, the most signifi- trade agreement with the United States- cant contribution to growth will be the contin- which accounts for 30 percent of Bangladesh's ued buoyancy of India's large domestic service exports-may damage the Bangladeshi cloth- sector, which is expected to grow at about ing sector, but this will probably be more than 8 percent over the next few years. Of course, offset in 2002, when Bangladesh will receive the sharp deceleration in global trade has de- duty-free access to European markets for its pressed growth in regional manufacturing pro- garment exports. duction to only slightly above 1 percent, but because this sector accounts for approximately Middle East and North Africa . . high oil 25 percent of regional economic activity, the prices and reprieve from drought boosts aggregate impact has been somewhat muted. near-term growth Government subsidies have been used over Oil revenues-which account for almost two- the last several years to cushion the impact of thirds of the region's export revenues-pro- high oil prices and poor crop production on vide the MNA region with a short-term out- consumer prices-which in turn is being re- look that is better than other regions. In 2001, flected in growing fiscal and current account the oil-exporting countries benefited not only deficits. For both India and Pakistan, central from strong export revenues, but also im- government fiscal deficits-of 5.3 percent and proved fiscal positions and higher rates of in- 5.8 percent respectively-have become major vestment. Among the diversified exporters, impediments to an acceleration of growth. agricultural production and rural incomes re- India's fiscal deficit is considerably higher on a ceived a strong boost from a reprieve of long- consolidated basis, including regional govern- standing drought conditions in some parts ments. Pakistan's fiscal woes are being com- of northern Africa. Growth in Morocco, for pounded by its deteriorating current account example, is anticipated to accelerate strongly position, which reflects higher oil prices and in- after consecutive years of stagnant or declin- terest payments, as well as drought-affected de- ing output. clines in export revenues. This has come at a While higher oil prices and recovery from time of high levels of public debt and low for- droughts in 2001 have provided a fillip to eign reserves, and Pakistan was forced to turn growth, the delayed recovery in industrial coun- to the IMF to help finance the current account. tries will significantly reduce the external im- Pressure on the current account may not ease in petus to growth at end-2001 and into 2002. the short run, because a severe water constraint World demand has slowed, and oil prices are for irrigation is likely to significantly reduce cot- expected to fall to $21 per barrel by 2002. ton output, the main foreign exchange earner Growth in the oil-dominant economies will for Pakistan. Aside from economic develop- slow to 2.3 percent in 2002. The diversified ex- ments, Pakistan's position on the "front line" of porters are similarly affected. Demand is now military action against Afghanistan will carry slowing sharply in the EU-the dominant ex- substantial near and medium term implications. port market for the countries of the Maghreb- Risks are substantial for the region. Politi- and in the United States, of importance to sev- cal uncertainty hangs over Afghanistan, and eral countries in the Masbreq. Lower income 23 GLOBAL ECONOMIC PROSPECTS growth and the erosion of confidence after Sep- growth in 2002, with per capita incomes again tember 11th will also affect tourism and related flat. However, the recovery in developed coun- sectors in the region in a substantial fashion. tries is expected to gather pace over the coming For some of the diversified exporters, increasing year and set the stage for a strong rebound in levels of public debt make it relatively hard to 2003, with growth rising to 3.9 percent. Non- cope with the current deterioration of the exter- oil exporters should see a substantial improve- nal environment. Increasing spreads threaten ment in performance, as commodity markets to worsen debt dynamics for several countries, firm and prices stabilize or even rise modestly Growth in the diversified exporters should in real terms. By contrast, oil producers face improve in 2003, as external conditions be- sharply weaker export prices and declining come more favorable to exports and tourism- terms of trade. Significant new capacity, espe- although the improvement in the international cially in offshore development will help to context is expected to be gradual. And the offset the negative impacts by allowing produc- MNA region will face increasing competitive tion and exports to increase. However, spill- pressures as the countries of Central Europe overs to non-energy sectors will be limited and enjoy greater access to the EU. oil exporters' growth is expected to slow as the boom unwinds. Sub-Saharan Africa ... suffering from low commodity prices In 2001, the world economic slowdown tem- Risks to the outlook porarily derailed SSA gradual recovery from the late 1990s slump. Average growth across The specter of a sharper slowdown in the region slowed to 2.7 percent from 3.0 per- industrial countries haunts the outlook- cent in 2000, and with population growing at Risks to this forecast are unusually high. The 2.5 percent, per capita GDP barely increased at terrorist violence in the United States in Sep- all. Despite generally better supply conditions tember will undoubtedly have negative short- in commodity producing sectors, weak demand run consequences for the U.S. and global econ- in industrial countries held export volume omy. But it is difficult to predict the severity of growth to 3.4 percent. Services, including the adverse effects because the response of con- tourism also felt the impact, growing at 3.6 per- sumers and businesses-and even future poli- cent. A widespread deterioration in terms of cies-are unknown. These uncertainties overlay trade compounded the difficulties for many structural risks that in other contexts would be countries. Agricultural and mineral commodity more manageable. U.S. consumers may be less export prices plummeted due to a combination responsive to interest rates than on previous oc- of weaker demand and a delayed supply re- casions as the stock market falls, high consumer sponse to the price surge of 1995-97. Oil prices debt, and heightened insecurity may render also weakened, although oil exporters contin- them more cautious; or foreign investors might ued to benefit from buoyant terms of trade and become concerned about the persistently high strong foreign investment demand. Not surpris- U.S. current-account deficit, and impose an ingly, oil exporters average growth of 3.6 per- abrupt adjustment; these events in turn would cent significantly outperformed the rest of the delay the recovery of investment and its implied region, where growth registered 2.6 percent for demand for high-tech imports. The European the year. downturn may become more severe once mar- Given the steep decline in developed coun- ket sentiment deteriorates further, or monetary tries activity toward the end of 2001, and the policy does not ease sufficiently or have the ex- prospect of a sluggish recovery in the first half pected effects. Japan's structural reforms may of 2002, the near term outlook for SSA is pes- falter or exact a higher toll on economic perfor- simistic. The forecast anticipates 2.7 percent mance, and cause the dip in 2001 to last into 24 PROSPECTS FOR DEVELOPING COUNTRIES Table 1.4 First year effects of a 2 percent of GDP decline in investment in the United States, Europe, and Japan (percentage points impact on regional GDP) East Asia Latin America Sub-Saharan Europe and Middle East and and Pacific and the Caribbean Africa Central Asia North Africa South Asia -1.5 -0.5 -0.9 -0.4 -0.4 -0.3 Source: Economic Policy and Prospects Group. the next year. Thus, with the global economy sharp decline in dollar interest rates. The euro in precarious balance, unforeseen shocks from interest rate is less important than the dollar whatever source are magnified and could push rate in the international financial system. Fur- the global economy into recession. thermore European trade linkages with Africa, What would be the consequences for the de- Central Europe, and some Latin American veloping countries of a further slowdown in the countries are relatively strong. These features industrial regions? We model simulations of illustrate why the recent slowdown in Europe, synchronous downturns in the United States, on top of low or negative growth in the United Europe, and Japan. For each, the simulation States and Japan is especially worrisome for assumes a decline in domestic investment of developing countries. The negligible short- 2 percent of GDP, spread over four quarters, term impact of the U.S. shock on Latin Amer- which is counterbalanced by monetary policy ica as a whole is the outcome of very diverse reactions, which leads to a shift in the business country-specific effects, strongly negative for cycle. The monetary reactions follow historical Mexico and positive for Argentina. The latter patterns, with the United States quickly react- result is mainly driven by the weakening of the ing to the fall in output with an accumulated dollar in the simulations. However, this posi- drop of 2.5 percentage points in interest rates. tive impact appears to be only temporary. European policy, much more focused on infla- The medium-term dynamics of downturns tion, is less aggressive, and interest rates fall by in the industrial countries differ markedly. Fig- 1 percentage point. In Japan, the scope for fur- ure 1.17 shows the impact on global GDP in ther lowering of interest rates is minimal as the years after the shock. For example, the first rates are close to zero in the baseline. The result year after the U.S. shock, global GDP is 0.6 per- of the differences in policy reactions is a weak- cent lower than the baseline, and three years ening of the dollar both against the euro and after the shock it has returned to baseline levels. the yen. Table 1.4 shows the first year effects of So the impact is not a permanent change in the these simulated downturns on developing re- level of global GDP, but rather a shift in the gions. East Asia is hardest hit followed by Sub- timing of the business cycle. While the rebound Saharan Africa. Indeed, the negative effects of after a U.S. downturn is likely to be quicker as a European recession on developing countries a result of monetary easing, it may create ten- is generally larger than the impact of a U.S. or sions in the medium run because of the mone- Japanese recession. A Japanese recession has tary impulses to the rest of the world. Lower in- fewer repercussions, due in part to the smaller terest rates potentially boost domestic demand size of Japan in world trade (see box 1.1). in other countries to compensate for export The impact of a U.S. recession is mitigated losses. In the medium run this may result in by a strong monetary policy response and, inflationary tensions that give rise to a new more importantly, by the strong and wide- downturn. The dynamics after a Japanese crisis spread effects of that reaction. Latin American show a different picture. Because of limited op- countries in particular may benefit from a tions for monetary easing, it takes several years 25 GLOBAL ECONOMIC PROSPECTS Table 1.5 Short-term claims of Figure 1.17 Global dynamics of international banks outstanding in selected recessions in industrial countries developing regions (percentage points cumulated impact on global GDP) (as of end December 2000; percent of annual imports of debtor countries) 0.20 North Creditor Japan America Europe Total 0.00 , , , ,, , United Debtor region States Africa 2.7 4.6 33.6 40.9 -0.20 \East Asia and Pacific 3.8 1.5 11.9 17.2 Europe Europe and -0.40 Central Asia 0.9 2.0 26.4 293 Latin America and the -0.60 Caribbean 1.8 10.8 36.9 493 Source: BIS; Economic Policy and Prospects Group. -0.80 1 3 ;5 7 9 year after shock Table 1.5 shows the short-term exposure of in- dustrial countries' banks in selected developing Source: Economic Policy and Prospects Group. regions as a percentage of the region's annual import bill. These economies account for two- thirds of all short-term borrowing by develop- to rebound, with a more protracted, but less ing countries from international banks. pronounced, impact on the global cycle. The The data show that the claims of European analysis suggests that a European recession banks are six to seven times larger than the ex- gives rise to global dynamics somewhere in be- posure of U.S. and Japanese banks. Moreover tween the other two. Again we see that a Euro- lending by European banks is much more di- pean recession on top of a U:S. recession is po- versified, while Japanese banks are focused on tentially dangerous. It could lead to a more East Asia, and U.S. banks on Latin America. protracted rebound, while at the same time ag- Consequently, banking problems in Europe are gravating medium-term tensions. potentially more disruptive for a broad range of developing countries than similar problems in -while financial instability is Japan or the United States, albeit that the prob- another risk ability of European financial strain is much The previous analysis explored the effects of lower than of escalating tensions in Japan. The slower growth through the main international simulation results highlight the dominance of transmission mechanisms: trade and interna- European banks and the large negative impact tional interest rates. To illustrate the impor- of a withdrawal of their lending on GDP in all tance of financial flows (here, bank lending), the debtor countries considered (table 1.6). we simulated a temporary withdrawal of short- Apart from the magnitude of the shock, do- term bank lending from the three industrial mestic conditions in the borrowing countries regions, along with a general increase in emerg- determine the impact. The shock is more detri- ing market spreads and a country-specific in- mental if reserves are low or foreign debt is crease in spreads, depending on the reduced net high relative to exports. In general, the initial capital inflows in terms of import coverage. In reduction in foreign funds is absorbed in three four quarterly steps the short-term debt owed ways: by attracting alternative foreign capital to industrial-country banks is halved, after at the cost of higher interest rates; by improv- which the original debt is gradually restored. ing the current account at the cost of domestic 26 PROSPECTS FOR DEVELOPING COUNTRIES Table 1.6 Withdrawal of short-term ing of historic performance, with advances in per- lending by industrial-country banks to capita GDP rising to 3.6 percent over the period selected developing regions: the first-year from 2005-2015, two full percentage points impact on GDP higher than the experience of the 1990s. Uoied sates Ewpeao Japanese While the near-term outlook for developing boolks books countries is heavily influenced by the interna- East Asia -31 -1,4 -0.1 tional business cycle and global environment, Eastern Europe 0.0 -1.3 0.0 long-term development trends are more di- Sub-Saharan Africa -0.1 -2.1 -0.1 rectly the result of economic fundamentals- Latin America -0.4 -4 savings, investment, population growth, trade Source: Economic Policy and Prospects Group calculations and productivity improvements as well as pol- icy in the various regions. Analyzing these fac- tors allows us to study possible paths of devel- recession and by reducing foreign reserves at opment and ways developing countries might the cost of becoming more vulnerable to fu- interact with the global economy over a longer ture shocks. For example, the Republic of period.9 To do so, we create a long-term growth Korea would replace, in the simulation, a scenario and analyze its consequences for devel- quarter of the lost short-term capital by other opment.10 This scenario allows us to test the capital inflows, reducing its capital needs by realism of growth and poverty reduction objec- roughly 10 percent of the shock, and absorb- tives, given reasonable expectations about fun- ing the remaining 65 percent of the shock by damentals and current policies. The scenario selling foreign reserves. also provides a baseline against which to simu- For countries with a low initial level of for- late policy changes. In chapter 6 we return to eign reserves, such as Brazil, reducing reserves long-run analysis and use the baseline to ana- much further would not be a viable option. lyze the effects of global reductions of barriers Therefore, the withdrawal of short-term lend- to trade. ing would directly translate into a domestic credit crunch. For countries with high external Long-term growth: A baseline scenario debt, such as Indonesia, Argentina, or Brazil, Income growth in the developing countries an improvement of the current account is dif- under the baseline scenario for 2005-15 would ficult because higher spreads result in higher be 3.6 percent in per capita terms, more than debt service. This would put more weight on 1 percentage point above the per capita growth the contraction of the domestic economy. rate of the high-income countries and 2 per- centage points higher than during the decade of the 1990s (table 1.7). Developing countries are Long-term prospects: expected to benefit from reforms carried out growth andpoverty reduction over the past decade. The policy environment is D espite the weakening of growth and uncer- much improved, especially in major countries tain present context, long-term prospects re- in virtually all of the regions. Tariffs have come main relatively promising for developing coun- down sharply in the last decade, and as a result tries. Moreover, it should be noted that the rate of economies are more open, with trade ratios GDP growth established by developing countries 50 percent higher than a decade ago. Macro- even in the sluggish global year 2002 (3.7 per- economic policies are improved; because gov- cent), contrasts favorably with historic experi- ernment budget deficits are lower now than in ence-some 0.2 percentage points above the av- the late 1980s and median inflation rates have erage performance of the 1980s, and 0.5 points been halved. These efforts to improve poli- above outturns for the 1990s. And longer term cies constitute investment in better long-term prospects hold the promise of continued better- prospects, and will allow regions to take ad- 27 GLOBAL ECONOMIC PROSPECTS Table 1.7 Long-term prospects: forecast and scenario growth of world GDP per capita (annual average percentage change) Forecast Scenario Medium-term Long-term 19800 1990s 2000-04 2005-15 World total 1.4 1.2 1.4 2.1 High-income countries 2.4 1.8 1.8 2.5 OECD 2.5 1.8 1.8 2.4 United States 2.2 2.2 1.7 2.2 Japan 3.4 1.1 1.1 2.6 Euro Area 2.2 1.8 2.2 2.7 Non-OECD countries 3.7 3.8 2.3 4.2 Developing countries 1.5 1.6 2.8 3.6 East Asia and Pacific 6.1 6.0 4.8 5.4 Europe and Central Asia 2.7 -2.5 3.2 3.5 Latin America and the Caribbean -0.9 1.6 1.5 2.6 Middle East and North Africa -0.6 1.0 1.4 1.4 South Asia 3.5 3.3 3.5 4.0 Sub-Saharan Africa -1.2 -0.5 0.9 1.5 Memorandum items Transition countries of ECA 2.7 -3.2 3.7 3.5 Developing countries, excluding ECA 1.4 3.0 2.8 3.7 excluding China and India 0.8 0.4 1.8 2.7 Note: Aggregations are moving averages, reweighted annually after calculations of growth in constant prices. Source: Economic Policy and Prospects Group. vantage of underlying momentum on economic Internal factors and policies drive differing fundamentals-savings rates, educational in- regional performance. East Asian countries vestments, population growth rates, and im- would not be able to maintain the exceptionally provements in productivity. And the external high growth rates of the 1980s and early 1990s, environment is poised to provide a more sup- but, on the strength of high savings rates and portive long-term context that redounds to the productivity, would continue its 30-year pattern benefit of the developing countries. as the most rapidly growing region. China's The acceleration of growth in the develop- growth will naturally slow as its economy be- ing countries seems, on first sight, somewhat comes larger and more modern, but still has more spectacular than it actually is. First, ex- scope to growth at 6 to 7 percent annually over cluding the transition countries, the next five the period (see World Bank 1997: China 2020). years feature a decrease of 0.2 percentage The countries in East Asia-should their say- points relative to the 1990s, before accelerat- ings rate persist at high levels--could well see ing by 0.9 percentage points to a 3.7 percent capital accumulation account for nearly two- rate of per-capita growth. Second, about two- thirds of their overall growth rate, with about thirds of the acceleration is the result of so- 30 percent generated by productivity increases. called composition effects. Fast growing coun- South Asia would follow a fairly similar pat- tries such as China have now a larger weight tern, although with somewhat more contribu- in the total than they had ten years ago, in- tion from labor supply growth and less from creasing the growth rate of the developing technological improvement. countries on aggregate, even without increas- For Latin America, the scenario assumes ing its own rate of growth. that the latent, but undeniable, growth poten- 28 PROSPECTS FOR DEVELOPING COUNTRIES tial in Mexico, Brazil, and Argentina, as well as other economies, is progressively realized. Figure 1.18 Income and population Better fiscal management and greater hemi- shares spheric and global integration are likely to (income in billions of 1997 dollars using market provide powerful forces that reduce debt bur- exchange rates, population in millions-percent) dens which in the past have shackled growth 100 - on the one hand, and, on the other, can un- 90 leash new productivity. The contribution of 80 capital accumulation in Latin America will 70 be about twice the contribution of productiv- 60 ity, with the contribution from labor supply 50 growth less than 15 percent on average.11 5,170 The countries of Eastern Europe and Cen- 40 tral Asia are expected to grow quite rapidly, 30 continuing a recovery from the transition- 20 generated depression of the early part of the 10 -Mi1111 14,500 1990s. Further openness among Central and 0 I Eastern European countries to trade and fi- 2000 GDP 2015 GDP 2000 2015 nancial integration with the EU, together Population Population with the reestablishment of the conditions of U Low and middle E High income peace in Southeastern Europe, should provide a favorable context for these countries to Source: Economic Policy and Prospects Group and grow. Both Eastern Europe and Central Asia World Bank Data and Projections (Population). will rely extensively on productivity growth, with labor supply growth either stagnant or in decline. While the Middle East and North Africa productivity growth. Europe, with a lower ini- would improve rates of per capita growth rela- tial level of integration, has somewhat greater tive to the historic period, growth under this scope for productivity improvements as it baseline scenario is likely to make only modest adopts new technology and becomes more inte- impact on achieving the International Develop- grated. Japan is expected to emerge from its ment Goals (IDG). Consistent with relatively current restructuring with a more efficient sys- high population growth rates, growth in the tem of allocating capital, and this will allow it Middle East and North Africa will be ac- to regain a degree of growth momentum, even counted for by labor supply growth-between though its aging population profile adds more 25 and 40 percent. Productivity growth will drag to its growth. If this scenario for the in- have to be fairly sustained-between 35 and 60 dustrial countries is realized, capital flows to de- percent of total growth-to achieve the IDG veloping countries could well resume within the GDP targets. context of a high productivity, low inflation, The external environment is likely to be sup- low interest rate environment. portive. Even though today's environment is Between 2000 and 2015 world income exceptionally weak, new technologies and fur- would, according to the scenario, expand by ther economic integration could indeed produce 60 percent-some $18 trillion (in 1997 dol- higher productivity-led per capita growth in lars) (figure 1.18). Income in the low- and high-income countries of 2.5 percent versus 1.8 middle-income countries would almost dou- percent during the 1990s. The U.S. economy is ble, and account for over 37 percent of the in- expected to recover to a long-term trend that is crease in world output. Over 1 billion persons somewhat higher than in the medium-term fore- will be added to world population, reaching cast because of the effects of technology-driven some 7.1 billion. Over 97 percent of the pop- 29 GLOBAL ECONOMIC PROSPECTS ulation increase will occur in low- and middle- nario. Moreover, should growth in developing income countries, with the high-income coun- countries turn out to be less than the 3.6 per- tries expected to add only around 30 million cent per capita of the baseline scenario, the persons in total.12 world as a whole would not reach the target. Nonetheless, these projections would con- Poverty trends tinue the reduction of the number of people liv- Growth will substantially reduce the number ing in poverty that began roughly about 1980. of people living in poverty. With base case Up through the 1970's, long-term increases in growth, the total number of destitute poor, liv- population swamped the growth effects in the ing on less than $1 per day, would decline to global economy and the number of people living about 750 million persons in 2015, down below $1 day increased (figure 1.19).14 How- from 1.15 billion in 1999 (table 1.8). The ever, since 1980 faster growth, particularly in number of people living on $2 per day or less China and South Asia, has contributed for the would decline by 600 million, from 2.8 billion first time in recent history to a steady decline in to 2.2 billion. While this rate of poverty re- the number living in destitute poverty. These duction would be sufficiently robust to achieve new projections confirm that trend. the target of reducing poverty by one-half in The new projections represent slight reduc- 2015,13 not all regions would succeed. Sub- tions in out-year poverty relative to last year's Saharan Africa would be far from reaching the forecast. The changes largely reflect updated goal even under this favorable growth sce- information on poverty and income distribu- Table 1.8 Regional breakdown of poverty in developing countries Number of people living on Number of people living on less than $1 per day (millions) less than $2 per day (millions) Region 1990 1999 2015 1990 1999 2015 East Asia and Pacific 452 260 59 1,084 849 284 Excluding China 92 46 6 285 236 93 Europe and Central Asia 7 17 4 44 91 42 Latin America and the Caribbean 74 77 60 167 168 146 Middle East and North Africa 6 7 6 59 87 65 South Asia 495 490 279 976 1,098 1,098 Sub-Saharan Africa 242 300 345 388 484 597 Total 1,276 1,151 753 2,718 2,777 2,230 Excluding China 916 936 700 1,919 2,164 2,040 Head count index (percent) Head count index (percent) Region 1990 1999 2015 1990 1999 2015 East Asia and Pacific 27.6 14.2 2.8 66.1 46.2 13.5 Excluding China 18.5 7.9 0.9 57.3 40.4 13.3 Europe and Central Asia 1.6 3.6 0.8 9.6 19.3 8.7 Latin America and the Caribbean 16.8 15.1 9.7 38.1 33.1 23.4 Middle East and North Africa 2-4 23 1.5 24.8 29.9 16.7 South Asia 44.0 36.9 16.7 86.8 82.6 65.5 Sub-Saharan Africa 47.7 46.7 39.3 76.4 75.3 68.0 Total 29.0 22.7 12.3 61.7 54.7 36.3 Excluding China 28.1 24.5 14.8 58.8 56.5 43.0 Source: World Bank staff estimates. 30 PROSPECTS FOR DEVELOPING COUNTRIES * Revisions to the estimates of the relation- Figure 1.19 World poverty, 1820-1998 ship between economic growth and poverty (Number of people living on less than $1 day, millions) reduction based on the new surveys, sug- 1,400 gesting a stronger positive effect of growth Depression/ on poverty reduction in several countries.16 World War II The long-term projection underscores the 1,200 H aimportance of achieving fast growth and dis- Historical trend tributing the benefits of growth equitably. Even under this scenario, the harsh reality is that over 2.2 billion persons will be living below the $2 per day income level in the year 2015-some 36 percent of developing country population. Without macroeconomic stability, co00, improved governance and sustained structural 1820 1870 1910 1950 1970 1990 reforms, including for example, improvements Source: Dollar (2001), Bourguignon and Morrisson (2001), in the provision of public services and infra- and Chen and Ravallion (2000). structure, as well as enhancing the participa- tion - of the poor in growth, the pattern of growth that underlies the baseline scenario tion in several countries. The downward revi- will not be realized, and millions more people sion in out-year poverty is despite the adverse will remain in poverty. effect on poverty of somewhat lower long- A faster rate of growth is possible. This re- term growth projections-the key economic port explores the potential for more rapid determinant of poverty reduction. As discussed growth associated with an acceleration of earlier in this chapter, the downgrading of the global integration. Reducing barriers to trade economic growth forecast reflects the effects in merchandise and services can accelerate of lower growth in 2001 and 2002, as well as growth if adequately supported by domestic minor revisions in long-term prospects for policies and development assistance. These some countries.15 policies and their consequences, the focus of There are three factors influencing the this report, are analyzed further in Chapter 6. current poverty forecast, reflecting updated information: Under-5 mortality One of the most important international devel- * The current forecast incorporates 31 new opment goals for 2015 concerns infant and household surveys leading to a reduction in child mortality. Are these goals likely to be the assessment of the base-year level of achieved under the scenario of growth and ris- poverty, with South Asia accounting for ing incomes? To project under-5 mortality, we the greatest reduction, somewhat offset by link long-term growth and urbanization, higher poverty levels in Latin America and shown to be effective predictors of child and the Caribbean. maternal health status and expected outcomes * Overall projected population growth in de- in under-5 mortality.17 Figure 1.20 presents the veloping countries is slower, in large part baseline forecast and compares it with the offi- due to the impacts of HIV/AIDS. This tragic cial IDG target. The only region with a fore- epidemic particularly affects population cast achieving the goal is South Asia (SAS). growth in Sub Saharan Africa, which ac- Three of the other regions are likely within counts for a significant part of the reduction range of achieving the target: East Asia (EAP), in this year's poverty forecast. Middle East and North Africa (MNA), and 31 GLOBAL ECONOMIC PROSPECTS Figure 1.20 Under-5 mortality-hopes and aspirations Under-5 mortality rates-Sub-Saharan Africa Under-5 mortality rates-East Asia and Pacific 180 60 1 60 50 100 80 30 60 20 40 10 20 1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015 - Hist + Forecast - Target - Hist + Forecast - Target Under-5 mortality rates-South Asia Under-5 mortality rates-Middle East and North Africa 140 80 120 710 100 60 SO S -50 0 .40 1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015 - Hist + Forecast - Target - Hist + Forecast - Target Under-5 mortality rates-Europe and Central Asia Under-5 mortality rates-Latin America and the Caribbean 35 60 30 50 25 20 40 1 030 10 20 5 10 0 0,,. . . . . .,, ,,,.,,. 1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015 - Hist + Forecast - Target - Hist + Forecast - Target Source: Economic Policy and Prospects Group; World Bank/Latin America Region staff. 32 PROSPECTS FOR DEVELOPING COUNTRIES Latin America (LAC). Europe and Central Asia that was highlighted most recently by Ireland's "no" (ECA) is not likely to achieve its target based vote on the Treaty of Nice (which makes changes to the on this scenario. And Sub-Saharan Africa voting structure of the EU to accommodate a larger (SSA) is well off the target path. SSA has al- membership). 9. The main strategy in developing the baseline sce- ready deviated significantly over the last dec- nario is that GDP growth rates are given and the model ade, and the current forecast foresees little solves endogenously for a technology parameter consis- closing of the gap. tent with the GDP target. The following assumptions underlie the base-case scenario. The baseline assumes that population and labor supply growth is exogenous. Co1clusions The latter is proxied by the growth of the population aged between 15 and 65 years, and implicitly assumes B oth the short-term outlook and the long- that participation rates are constant. Agricultural pro- term analysis underscore the importance ductivity is fixed at 2.5 percent per year. Manufacturing of realizing accelerating growth. One way to productivity is assumed to be 2 percentage points create incentives to grow is for developing greater than services productivity. International trans- countries to deepen their participation in the portation margins decline by 1 percent per year in the global economy. And that is the subject of the baseline. Income elasticities are fixed at base year levels. of this report. This involves recalibrating the parameters of the con- sumer demand system between solution periods. Capi- tal accumulation is modeled as the previous period's depreciated capital stock augmented by the previous Notes period's level of investment (including net foreign in- 1. Japan Financial Services Agency, January 2001, vestment). The model tracks capital vintages with sub- and Japanese Banks 2000, Japanese Bankers Association. stitution elasticities, typically lower for installed capital 2. Source for data is U.S. Treasury. It should be than for new capital. Countries with higher rates of in- noted that the United Kingdom accounts for about 50 vestment will exhibit more flexibility over time. Foreign percent of portfolio investment flows, channeling Eu- capital flows are fixed at base year levels (ensuring at ropean and well as developing country and offshore least their sustainability). Fiscal policies are unchanged funds into U.S. assets. at their base year levels, except for direct taxation that 3. For example, major shipping lines increased adjusts to target a given fiscal deficit. freight rates to South Asia by 10 to 15 percent. 10. This is distinct from a forecast in the sense that 4. Notably, the share of U.S. export value shipped unforeseeable positive or negative shocks-from tech- by air has risen from 25 percent a decade ago to 40 nology, politics, or other sources-are virtually certain percent in 2001, with 75 percent of total high-tech ex- to occur, and will push up or down actual perfor- ports moving by air. mance, especially at the regional level. 5. For example, around 65 percent of holidays 11. While one of the key outcomes of the baseline booked for the Caribbean had been canceled in the scenario is the endogenously determined productivity, weeks after the attack. many other variables of interest are generated endoge- 6. For example, the Brazilian real depreciated 27 nously in the baseline-relative factor prices, real ex- percent relative to the dollar in the first half of 2001, change rates, terms of trade, bilateral trade flows, and contributing to the 12.9 percent decline in soybean the composition of demand and output. prices and 11.6 percent drop in vegetable oil prices in 12. The above analysis provides a framework for the same period. Indonesia's currency depreciated 21 assessing the world distribution of income and rates of percent (from December to April), which pushed dollar conversion, but says nothing about how this outcome prices for palm oil and natural rubber down. Thailand, is achieved or what policies are needed to help bring the world's largest rice exporter, had a year-to-date cur- about these outcomes or even improve them. The rency depreciation of 8 percent, contributing to a 9 analysis relies to a large extent on an applied general percent decline in rice prices. equilibrium model of the global economy. The model 7. In the U.S. Pacific Northwest, 1.6 million tons of reflects key facets of economic theory-the most im- aluminum capacity has been closed-equivalent to 7 portant being that supply equals demand. This frame- percent of world capacity-because of the electricity work ensures that all economic flows are fully consis- crisis in California. tent-at the individual, national, and global level. The 8. Support for eastward expansion of the EU has model abstracts from some real world phenomenon, waned markedly in existing member countries, a trend notably international financial flows are not modeled 33 GLOBAL ECONOMIC PROSPECTS explicitly, allowing differential rates of return across selves will be country-specific and related to the level of nations. Nor are there linkages between monetary phe- economic development. nomena and the real economy, for instance, when the model exhibits super-neutrality. 13. Reducing poverty by one-half in 2015 (com- pared with the 1990 level) is one of the key develop- References ment goals (see for example http://www.paris21.org/ Bourguignon, Franqois and Christian Morrisson. betterworld). 2001. "Inequality among World Citizens: 14. Historical discussion based on David Dollar 1820-1992." DELTA, Paris. Processed. (Available (2001), Bourguignon and Morrisson (2001) and Chen at http://www.delta.ens.fr/XIX/paper WD19.pdf). and Ravallion (2000). Chen, Shaohua and M. Ravallion. 2000. "How did the 15. The long-term per-capita consumption growth World's Poorest Fare in the 1990s?" World Bank forecast for developing countries has dropped from Policy Department Working Paper No. 2409. 3.5 percent to 3.4 percent (average per cent change per August. Washington, D.C. annum), though with variation across regions. Deaton, Angus, and R. Miller. 1996. "International 16. The household surveys and national income ac- Commodity Prices, Macroeconomic Performance counts on which forecasts are predicated are undergoing and Politics in Sub-Saharan Africa." Journal of continual review and methodological changes, espe- African Economies 5: 99-191. cially for large countries, which can have important ef- Dollar, David. 2001. "Globalization, Inequality, and fects on overall poverty assessments. For example, the Poverty since 1980". Processed. Development Re- poverty levels for India may well be revised downward search Group, The World Bank, Washington, DC. even further next year, once the new household survey Hoshi, Takeo. "What Happened to Japanese Banks?" has been thoroughly evaluated in conjunction with na- Monetary and Economic Studies, Institute for tional authorities. Monetary and Economic Studies, The Bank of 17. This is based on the pioneering work initi- Japan Discussion Paper Series No. 2000-E7:1-50, ated by Quentin Wodon of the World Bank's Latin March 2001. Tokyo. American Region. We link long-term GDP forecasts to Kose, M. Ayan, and R. Riezman. 2000. "Trade Shocks econometrically estimated relations with selected Inter- and Macroeconomic Fluctuations in Africa." national Development Goals (IDGs). The work under- Journal of Development Economics 65: 55-80. taken in the Latin American Region has used panel es- Kouparitsas, Michael A. 1996. "North South Business timation techniques to fit a relation between the IDGs Cycles." Federal Reserve Bank of Chicago Work- and GDP and the rate of urbanization, two important ing paper WP-96-9. determinants of health access. In order to allow for Mori, Naruki, S. Shiratsuka, and H. Taghuchi. 2001. varying elasticities (with respect to income and urban- "Policy Responses to the Post-Bubble Adjust- ization levels), the LAC estimation procedure used ments in Japan: A Tentative Review." Monetary spliced data. The forecast presented here uses a logistic and Economic Studies (Special Edition), February function that has continuously variable elasticities (al- 2001. Institute for Monetary and Economic Stud- though with constant signs). The estimation procedure ies, the Bank of Japan, Tokyo. is based on pooled data on a regional level. It is as- World Bank. 1997. China Engaged: Integration with sumed that the relation between per capita GDP growth the Global Economy. World Bank, Wash. DC. and the rate of urbanization is uniform across coun- - . 2001. Global Development Finance 2001. tries within a region, although the elasticities them- World Bank WDC. 34 Market Access and the World's Poor With the conclusion of the Uruguay and clothing, the opening of markets has been Round and a wave of unilateral re- slow, because the implementation of the World forms, barriers to trade have fallen Trade Organization (WTO) agreement allows substantially around the globe, spurring the importers the leeway to select the products growth of world trade. Developing countries as to be freed of quota restrictions. Thus despite a whole gained significant market share-about progress made by the agreement, foregone ex- 7 percent-in world nonenergy merchandise port earnings for developing countries are size- trade, thanks mainly to ambitious domestic pol- able. And due to still-high tariffs, market access icy reforms in the 1990s. will remain restricted even after the quotas have been abolished in 2005. Global protection hits the poor hardest However, progress in lowering barriers has A "development round" would help- lagged in two of the sectors with both the high- These issues provide fertile areas where recip- est protection and with the greatest impact on rocal negotiations in a "development round" poverty-agriculture and labor-intensive man- could provide substantial benefits for develop- ufactures (such as textiles and clothing). Agri- ment. Open trade in agriculture and labor- culture and other labor-intensive products mat- intensive manufactures would raise incomes ter to the world's poor because they represent among the world's poor. more than half of low-income countries' ex- ports, and about 70 percent of least-developed -but effectiveness requires cooperative countries' export revenues. policies to complement negotiations Developing countries themselves are part of Global cooperation beyond negotiations could the problem. Although South-South trade is a also expand trade. For example, most prefer- much smaller share of total trade, average tar- ential access schemes to high-income countries' iffs in manufactures are three times higher for markets only partly breach the walls of pro- trade among developing countries than for tection. Their limited coverage and various exports to high-income countries. Taken to- other impediments to trade undermine their gether and because of high protection for labor- otherwise positive effects. The recent European intensive products around the globe, the world's Union's "Everything But Arms" initiative miti- poor face tariffs that are, on average, roughly gates these problems by removing barriers on twice as high as those imposed on the nonpoor. exports from least-developed countries. Exten- Protection that affects the poor takes several sion of this initiative to the United States, forms, including tariff peaks, quotas for textiles Japan, Canada, and other higher-income coun- and clothing, tariff escalation, and agricultural tries would greatly stimulate the exports and subsidies in high-income countries. In textiles growth of the least-developed countries. 37 GLOBAL ECONOMIC PROSPECTS Beyond these, "aid for trade" can help developed countries continue to be dependent countries take advantage of new market open- on agriculture and labor-intensive manufac- ings. Providing assistance for countries to im- tures, which together account for about 70 plementing WTO-sponsored reforms, design- percent of LDC exports. ing programs that protect the poor during The expansion of trade volumes in these reform, and upgrading work skills will help sectors did not keep pace with world trade ensure that trade benefits the poor. Moreover, growth, which has undermined the growth domestic reforms and assistance to improve prospects of the LDCs and hindered the battle backbone services-such as transport, finance, to reduce poverty. South-South trade repre- and communications--can better link the poor sents about 30 percent of low-income coun- to the global marketplace. tries' nonenergy merchandise exports, and is more important than for middle-income coun- tries. Exports of low-income countries to other A changing landscape of developing countries increased rapidly, espe- merchandise trade cially in agriculture. In labor-intensive man- T he 1990s witnessed a boom in world ufactures, South-South trade is far more trade, with an average annual increase of important in textiles than it is in clothing, foot- 6.3 percent in the volume of global merchan- wear, and leather, both for low-income and for dise trade (1990-99)-outpacing global gross middle-income countries. domestic product (GDP) growth by an aver- Moreover across products, the increase in age 4 percent per year over the same time pe- developing countries' exports was uneven. In riod. Exports grew faster than output in every labor-intensive manufactures, developing coun- major region. tries' market share increased sharply and now surpasses that of high-income countries. By con- Developing countries gained market share trast, in agriculture, another labor-intensive sec- in world merchandise trade- tor, developing countries' market share rose The share of developing countries in global more modestly. This rise in market share was export markets rose by almost 7 percentage driven by South-South trade, with about one- points, to about 25 percent of world non- third of all developing countries' agricultural energy merchandise trade, primarily on the exports now directed to other developing coun- strength of superior performance in manufac- tries-up from just about 20 percent in the early turing (figure 2.1). However the details behind 1990s. The slow increase of developing coun- these headlines reveal divergent trends-with tries' share in world agricultural exports partly some sectors and some countries enjoying ex- reflects developing countries' export diversifica- ceptional growth, while others remained al- tion out of agriculture, and partly reflects sur- most stagnant. plus production from high-income countries. Exports of the poorest countries are even -but poor countries remained more concentrated in agriculture and labor- on the sidelines, dependent on slow intensive manufactures. Sub-Saharan African growing commodities and labor agricultural products provide about 60 per- intensive manufactures cent of export revenues, with little contribu- Developing countries as a whole improved tion from manufactures.1 their penetration of world markets, but the ex- port share of the 49 least-developed countries (LDCs) shrank from 3 percent in the 1950s to Labor-intensive exports can spur around 0.5 percent in the early 1980s, and has pro-poor growth hovered around this very low rate over the last Jn developing countries-in particular the two decades (UNCTAD 2001). The least- Apoorest where inexpensive labor is plentiful- 38 MARKET ACCESS AND THE WORLD'S POOR Figure 2.1 Changing global trade patterns Export growth outpaced growth of output -and developing countries increased their everywhere- share in the global market In percent of world exports for each product group (left axis) Average annual growth in volume terms, 1990-99, in percent In US$ billion (over bars) 7 60 Exports to developing countries 212 6 Exports 50 Exports to high-income countries 125 5 Output 40 193 4 30 126329 2 20 580 2 - - _ 1-0 -_ - -uA MIE ~. 0 9299 1 18 1991-92 1998-99 Manufactures Mining products Agricultural Totali non-energy Agricultir Laborintnsive (including oil) products merchandise maniacnures Source: World Bank staff calculations, based on WTO data. Source: World Bank staff calculations, based on U.N. Comtrade. Manufactured exports have boomed- -but poor countries did less well Percent of developing countries' exports In percent of world exports for each product group 100 60 90 50 Exports from LDC 80 Exports from low-income countries 70 40 Exports from 60 Oil and other Middle-income countries 6030 cute 50 20 Milli 50 20 10 1991-92 1998-99 991-92 1999 1991-92s99-99 0 ,, , ..,, . . ., , , , , . . . ,.. ,. . Total nan-energy Agriculture Labor-Intensive 1970 1975 1980 1985 1990 1995 merchandise m nes Source: World Bank staff calculations. Source: World Bank staff calculations. The shares of labor-intensive products in world -and poor countries remain dependent on exports declined- these sectors In percent of world nonenergy merchandise exports In percent of total nonenergy merchandise exports for each country group 14 80 Agricultural products 12 1991-92 60 Labor-intensive 10 1998-990 manufactures 109 * gg-g 50 B 40 6 30 4 20 2 10 0 Ibzi 0 Agriculture Clothing Textile Footwear & # # Leather Sub-Shaan LDC Low-income Middle-income High-income Africa Source: World Bank staff calculations, based on U.N. Comtrade. Source: World Bank staff calculations, based on U.N. Comtrade. 39 GLOBAL ECONOMIC PROSPECTS export-led growth can accelerate the reduction where the poor have a stake. Capital-intensive of poverty. Faster export growth can boost and import-substituting growth has generally income growth of the poor, first, by stimulating not been effective in alleviating poverty; agri- overall economic growth. On average, every cultural growth, where there is a low concen- additional percentage point of growth in house- tration of land ownership and labor-intensive hold consumption reduces the number of peo- technologies are used, has almost always helped ple living on less than $1 a day by an estimated to alleviate poverty (Gaiha 1993; Dart and 2 percent (World Bank 2000a). And among all Ravallion 1998). Exports of textiles and cloth- developing countries, successful integrating ing have also spurred labor-intensive growth in countries-the top third of developing coun- manufacturing, contributing to the reduction of tries ranked by an increase in trade-GDP ra- urban poverty, especially among women. tios-grew faster (Dollar forthcoming). During each of the past two decades, the developing -Agricultural exports can reduce countries that have had fast export growth- rural poverty leading to an increase in the share of nonenergy Rural poverty accounts for nearly 63 percent merchandise exports in GDP-have also had, of poverty worldwide, reaching 90 percent in on average, 1 percent higher real GDP growth China and Bangladesh, and between 65 and (figure 2.2). 90 percent in Sub-Saharan Africa (Khan 2000). But if growth is necessary to reduce poverty, Developing countries that have had more rapid the pattern of growth also matters. Export-led agricultural export growth have also tended to growth can reduce poverty more directly when have more rapid growth of agricultural GDP it fosters employment in labor-intensive sectors (figure 2.3). Thus increased agricultural ex- ports contribute to increased agricultural in- come growth and reduced rural poverty. The effects of trade growth on poverty Figure 2.2 A rising share of exports in wol emed ts e xpane t GDP is associated with faster growth would be muted if exports expanded at the Average real GDP growth per year, percent expense of domestic food production. But in most cases, increased exports of nonfood agri- 3.5 9cultural commodities (such as coffee, cocoa, or 1 cotton) provided hard currency to purchase in- puts for food crop production, which boosts 2.5 _ overall agricultural growth. In Vietnam, for ex- 2 _ample, nonfood crop production and U.S. dol- _m lar exports (primarily coffee) rose by about 15 1.5 percent per year from 1990 to 1998, following economic reforms. This boosted fertilizer use 1 and contributed to a nearly 50 percent rise in 0.5 _ food crop production over the same period. Agricultural GDP grew by 4.6 percent per year, 0 e I e and rural poverty fell to 45 percent in 1998- Decreasin G e Inrng epo down from 66 percent in 1993 (World Bank 2000b). Note: Sample includes only developing countries. Exports In Uganda, nonfood crop production refer to nonenergy merchandise exports. Data from national accounts in constant 1995 dollars. Number of surged following marketing liberalization in countries with increasing (decreasing) export share in the early 1990s. This surge was followed by a GDP: 1980-89: 23 (63), 1990-99: 57 (29). tripling of fertilizer use and a rise in food crop Source: World Bank staff calculations. production. Thus, increased exports (primar- ily coffee) boosted agricultural GDP growth 40 MARKET ACCESS AND THE WORLD'S POOR Figure 2.3 Increases in exports and agricultural production go hand-in-hand Agriculture GDP vs. export growth Food production vs. agriculture export growth Agricultural GDP growth (percent/year, 1980-98) Food production growth (percent/year, 1980-99) 8 7 6 66 e5 o* 5 * * I o i 4 I | 1l I 0 2 I2 0 -2 1 -10 -5 0 5 10 -10 -5 0 5 10 15 Agricultural export growth (percent/year, dollars 1980-98) Source: World Bank staff calculations, based on FAO data. to about 4.4 percent per year, and eventually tional arrangements that give duty-free and ef- contributed to increased exports of other crops, ficient access to inputs and foreign investors are creating a virtuous circle in agriculture. Rural no less important. Most success stories among poverty fell from 60 percent in 1992 to 39 developing countries confirm that booming ex- percent in 2000, and among the poorest quin- ports of textiles and clothing fostered broad tile of population primary school enrollment output growth (table 2.1). rose from 51 percent to 69 percent in the same Increased T&C exports are associated with period (World Bank 2001a). growth of local manufacturing, through de- Cross-country comparisons confirm that in- mand linkages and increased purchasing creases in agricultural exports rarely occur at power among workers. In all successful textile the expense of food crop production. Rather exporters the share of private investment in than competing for scarce resources, the two GDP considerably increased during export are positively correlated at the national level booms. However, to benefit from backward (figure 2.3). Thus increased agricultural ex- linkages, the domestic suppliers of the T&C ports and increased food production are a win- industry must be competitive and responsive.2 win combination for developing countries. For example, Pakistan seems to be a case of low industrial linkages and spillovers, as the Exports of textiles and clothing tend to increase in the share of manufacturing value reduce urban poverty- added in GDP was just about the same as that Many developing countries have become major of the T&C industry. Pakistan remained fo- exporters of textiles and clothing (T&C), but cused on protective policies to boost the cot- others, especially in Sub-Saharan Africa, have ton-processing sector on the back of abundant yet to take advantage of this card. Low-cost domestic cotton production (including, for ex- labor and a competitive exchange rate are two ample, export controls on cotton), rather than important enablers of T&C exports. Institu- relying on globally integrated production 41 GLOBAL ECONOMIC PROSPECTS Table 2.1 Major export booms in textiles and clothing and effects on economic performance and poverty Hangladeshl Madagascar Mauritius Pakistan Tunisia Period 1975-80 1980-95 1992-95 1996-99 1980-85 1986-92 1980-85 1986-90 1980-86 1987-93 Share of T&C in exports (%) 02 39.3 7.7 19.1 27.7 47.8 36.5 53.7 19.3 34.7 Real GDP growth per year (%) 3.3 4.7 1.2 3.6 2.2 7.4 7.4 6.0 3.9 4.3 Share of manufacturing value added in GDP (%) 16 16.4 14.3 15.0 16.9 23.9 15.6 16.9 13.5 16.6 Poverty rate2 Total Head Count S.5 53.1 60.2 - 19.5 106 49.6 33.9 19.9 14.1 Urban Head Count 50.2 35.0 - - - - 38.2 28.0 12.0 8.9 Note: lExports of clothing. 2According to the national definition-end of period. For each country, the second column, indicates T&C export booms. Source: World Bank staff calculations. using best-available materials. By contrast, in improve export performance. Poor workers Mauritius and Tunisia, T&C export growth and farmers will benefit from domestic trade was accompanied by broad manufacturing policy reform, thanks to faster export growth, growth. increased demand for their labor, and higher Growth in textiles and clothing also bol- wages and producer prices. But trade liberal- sters employment and real wages. In Mauri- ization can also affect the poor in the short tius, growing T&C exports had a tremendous term in complex ways that depend on country impact on unemployment, which was at 14 circumstances (see Winters 2000). These effects percent before the export boom and was are difficult to track because trade policy re- virtually absorbed by 1992. Eventually the form is often undertaken in tandem with other shortage of labor became a constraint on the major reforms that may also affect the poor- expansion of output, bringing about a nearly such as labor market reform, product market 50 percent wage increase. Part of Mauritius' deregulation, or public enterprise reform. T&C production moved thus to Madagascar. Trade policy does have important short- Because of faster employment and wage term effects on the poor through three chan- growth for low-skilled workers in all countries nels. Trade policy reform will, first, affect the with T&C export booms, poverty declined sig- poor by changing the prices of their consump- nificantly (table 2.1). As T&C manufactur- tion basket. Trade liberalization will shift rel- ing is typically concentrated in urban areas, ative prices, eventually increasing the prices of urban poverty was more drastically reduced. In traded relative to nontraded goods and reduc- Bangladesh, one of the poorest countries with ing the prices of imported relative to locally overwhelming rural population, total poverty produced goods. declined slowly; however during the export The overall impact on the poor through the boom in clothing, urban poverty dropped by price channel depends both on the composition almost one-third (World Bank 1999b). Thus of their consumption and on other reforms that to effectively contribute to the reduction of may concurrently affect prices (such as the poverty, trade liberalization in agriculture and phasing out of subsidies and price controls). On labor-intensive manufactures must go hand in balance, in net food exporting countries, while hand. poor farmers may gain from higher producer prices, poor urban dwellers may suffer from -but trade liberalization affects the poor higher food prices. In Ghana for example, in various ways while the rural sector gained from the reform Domestic trade policy reform lays the ground- and overall poverty was substantially reduced, work for better use of productive resources to living standards in Accra deteriorated in 42 MARKET ACCESS AND THE WORLD'S POOR 1988-92. By contrast, in net food importing 1995). A similar pre-reform pattern of protec- countries, poor consumers may benefit from tion was also found in Morocco (Currie and lower domestic prices of imported food. Expe- Harrison 1997). A more common pattern is rience from Haiti illustrates the complex impact that the politically connected sectors that re- of trade liberalization on the poor (box 2.1). ceive the most protection are the ones em- Second, wages and employment may not ploying workers with higher wages. always change as expected, depending on the A third channel is through the effects of pre-reform structure of protection. If for ex- government taxing and spending. Institutional ample, the protected sectors employ many of disruptions can mitigate the benefits of trade the poor, they may suffer in the transition to liberalization for the poor. The abolition of the an open trade regime. This is more likely in marketing boards for export crops sometimes middle-income countries, where sectors inten- led to abandoning key services that they often sive in unskilled labor are often protected as provided-such as research, quality monitor- they face stiff competition from low-cost pro- ing, maintenance of rural roads, and credit to ducers (Davis 1996; Wood 1997). For exam- small farmers (Winters 2000; World Bank ple, in Mexico, a country that implemented an 2000a). In Zambia, for example, the abolition ambitious trade policy reform program from of the marketing board led to abandoning the 1985 to 1988, the nominal tariff and import purchase of maize in remote areas. There is license coverage in apparel and footwear was also concern that, in the absence of reforms to among the highest in manufacturing (Revenga broaden the tax base, reduced government rev- Box 2.1 The aftermath of trade liberalization in agriculture: lessons from Haiti T he bold steps to liberalize trade in Haiti begin- poor, small farmers were forced out of rice produc- ning in 1986 that continued throughout the tion. On the other hand, many urban and rural poor 1990s have not produced rising incomes or reduc- were better off, because most of Haiti's poor were tions in poverty. Today with per capita income of not producers of rice, but rather consumers who had less than $500 and about two-thirds of the people been paying a high tax on a very basic food staple, living in rural areas, and more than 80 percent of curtailing their rice consumption. When rice became the rural population living below the poverty line, more affordable, national consumption doubled, and Haiti remains the poorest country in the Western most of the poor were better off. Hemisphere. Yet this is only part of a story that does not Prior to trade liberalization, the agricultural see- have a happy ending. Severe governance problems tor of Haiti was highly protected through tariffs (40 eroded macroeconomic stability, discouraged invest- to 50 percent) and subject to many nontariff barriers ment, and undermined the capacity of the govern- and import prohibitions. Reforms brought down tar- ment to provide meaningful assistance for the poor, iffs on rice, an important food staple, to 3 percent. much less build infrastructure and institutions to The real price of rice to consumers was reduced by support and sustain trade capacity. The eventual vir- 50 percent and imports of rice jumped from zero to tual collapse of the state has left Haiti mired in about half of domestic demand. Domestic rice pro- poverty. This underscores the lesson that trade policy duction also fell-by more than 40 percent corn- cannot substitute for good governance and a pro- pared with the 1985-90 levels. poor development strategy. Does this mean that the poor suffered from the trade reform as some have contended (Oxfam Inter- Sourc Oxfam International 2001; IMF Staff Reports for the national 2001)? Not necessarily. To be sure, some 1999 and 2000 Article IV Consultations, Worl Bank staff. 43 GLOBAL ECONOMIC PROSPECTS enues from trade-related taxes may trigger a adequate social protection, and ensure maxi- decrease in social expenditures targeted for the mum access to retraining opportunities. poor. For example, Tanzania trade policy re- form in the mid-1980s shifted income toward the largely untaxed small farmers, small enter- Market access barriers prises, and the informal sector, thus reducing limit export opportunities of the domestic tax ratio (Kanaan 2000). developing countries In sum the evidence is clearer on the long- he Uruguay round of trade negotiations term consequences for the poor: on average, T made a significant contribution toward in countries where outward orientation has in- lowering global barriers to merchandise trade creased, income growth of the poor has kept on two fronts: improving market access, thanks pace with mean income growth (figure 2.4).3 to the reduction of tariffs and quantitative And on average, globalizers have grown faster, restrictions on a number of products; and ex- thus witnessing accelerated reduction of poverty. tending multilateral disciplines to previously ex- By contrast, countries where outward orienta- cluded sectors-particularly agriculture, textiles, tion has decreased have seen slower growth, and clothing. As a result of multilateral trade ne- and the poor in those countries have also gotiations and unilateral reforms, average tariff fallen behind. Nonetheless the transition to rates have been halved-although they still re- these higher growth rates can take time and main high in South Asia and in the Middle East reform can impose short-term costs on the and North Africa (figure 2.5).4 Progress in low- poor in some countries. Hence, governments ering tariff barriers has been particularly im- have to design trade reforms carefully, provide portant for more skill-intensive manufactures, on which an increasing number of developing countries in East Asia, Eastern Europe and Cen- Figure 2.4 In globalizing economies the tral Asia, and Latin America rely for exports. poor participate in stronger growth For example, the largely duty-free trade in in- Percent of change per year formation technology products that came into 3.0 A force with the "Information Technology Agree- Average income growth ment" of the Uruguay Round strongly boosted 2.5 South-North trade in the information and com- munications technology sector. 2.0 Income growth of the poorest 20 percent of the population Tariff walls are high, especially in 1.5 labor-intensive products 1.0 _ _ However despite progress, labor-intensive products still remain extensively protected. 0.5 - Tariff protection for agricultural commodities is higher than for manufactures, both in indus- 0.0 - trial and in developing countries. But in high- Increasing Decreasing income countries the average tariff rate on outward orientation outward orientation agriculture is almost double the tariff for man- Note: Sample includes 129 time periods, based on ufactures. Applied tariffs on labor-intensive developing countries' household surveys. Outwardufcre.tifso lar-nniv orientation is measured by the share of trade flows manufactures also largely surpass the average (the sum of exports and imports) in GDP. Periods of for industrial goods. Compared to industrial increasing outward orientation are those when the share of trade flows in GDP increases by at least 0.5 products as a whole, labor-intensive manufac- percentage points per year. tures are again more protected in high-income Source: World Bank staff calculations. than in developing countries, by an estimated one-third. 44 MARKET ACCESS AND THE WORLD'S POOR Figure 2.5 Tariffs still impede trade In a decade, average tariffs have been almost -but tariff protection in agriculture is higher than in halved- manufactures. Average MFN tariffs (unweighted in percent) Average MFN tariffs in 1997-99 (unweighted in percent) 70South Asia 30 Agriculture Manufactures S6le 25 A Aguea at 50 an 1993-95 20 urope 40 hveoping amenca tnb -ae.can15 cent 30 De oing Mddle East E. Europe 20199799 t.!!,a and and 0 0 Source: World Bank, based on WTO data. Source: World Bank. Labor-intensive manufacturers are also -as widespread tariff peaks shelter agriculture and sheltered- labor-intensive manufactures in the Quad Average MFN tariffs in 1997-99 (unweighted, in percent) MFN tariff lines in tariff peak (in percent for each product group; 1999) 45 SotAo Mdl W80 40 4n m oote, 70 *EU Japan Oada] 35 No Afdut 60 Agricultural Products W 30 50 25 Textle and Lin Arnena 40 20 c Dli Doping nban 30 15 Enua E n 20 10 Av 10 - 0 AII goods Primary food Processed Textile and Footwear 0 food clothing Source: World Bank staff estimates, based on WTO data. Note: Tariff peaks are defined as tariffs greater than 15 percent. Source: World Bank, based on OECD tariff files. Tariffs escalate steeply in the Quad- -but tariff escalation is also common in developing especially in agriculture- countries Average unweighted tariffs in percent (1998-99) Average uniweighted tariffs inpercent (1998-99) 16 30 14 First stage Agricultural products Firs stage 12 Semi-processed 25 - Semi-processed 1 Fully processed I0 A Fully processed 10 -32 8 Industrial products 6 M Industrial products 4 -- 0 a m- m . @ fvb* 0 Developing High-income Developing High-income Source: World Bank, based on WTO data. Source: World Bank, based on WTO data. 45 GLOBAL ECONOMIC PROSPECTS Trade barriers on labor-intensive products In developing countries, too, the average tar- are commonly raised through tariff peaks (tar- iff for fully processed agricultural products and iffs exceeding 15 percent) on imports of "sensi- manufactures is higher than on unprocessed tive commodities." Imports at tariff peaks rep- products. The reduction of tariff peaks in the resent about 5 percent of total Quad (Canada, Quad and other countries would mitigate tariff European Union (EU), Japan, and United States) escalation. In the EU and Japan, for example, imports from developing countries, and more tariff peaks are more widespread on imports of than 11 percent of total Quad imports from processed food than on primary food imports LDCs (Hoekman, Ng, and Olarreaga 2001). (figure 2.5). Trade of products where more Within the Quad, tariff peaks are wide- processed exports from LDCs have a chance of spread but their pattern differs (figure 2.5). In breaking through would thus receive a boost. North America, tariff peaks are commonly found in industrial goods, particularly on im- Despite progress, trade in agriculture ports of textiles and clothing. By contrast, tar- remains heavily distorted iff peaks in the EU and Japan are common in The Uruguay Round Agreement on Agriculture agriculture-especially on imports of processed (URAA), which came into force in 1995, food, and tariff peaks on imports of footwear marked an important step in improving access are widespread across all Quad markets and to sheltered. agricultural markets in high- surpass those found in textile and clothing. In income countries. A wide range of nontariff developing countries, tariff peaks are prevalent barriers was abolished, including quantitative also, because applied tariffs are close to the tar- import restrictions, variable import levies, and iff peak threshold. discretionary import licensing. These barriers were converted to ordinary tariffs (tariffica- Tariff escalation is a major concern for tion). Existing and new tariffs were bound, and developing countries these bindings were subject to reduction. De- Tariffs often rise significantly with the level of veloping countries were allowed more compli- processing (tariff escalation) in many high- ance flexibility through longer implementation income and developing countries. Tariff esca- periods and lower reduction commitments. lation in high-income countries has the poten- Because international agricultural prices in tial of reducing demand for processed imports the base period for the URAA (1986-88) were from developing countries, hampering diver- way below high domestic prices supported by sification into higher-value added exports quotas, the conversion of quotas into tariff (Blackhurst, Enders, and Francois 1996). equivalents resulted in high rates of tariff pro- In high-income countries, tariffs escalate tection (OECD 2001a; World Bank forthcom- steeply, especially on agricultural products (fig- ing). Moreover tariff reduction commitments ure 2.5). In the Quad, tariffs on more processed involved a simple average across products, cre- agricultural commodities are comparatively ating much leeway to spread reductions un- higher in the EU and Japan, while in the United evenly, with lower cuts in more sensitive com- States there is evidence of reverse escalation be- modities. Hence scheduled tariff reductions over tween unprocessed and semiprocessed com- the URAA implementation period may not have modities. Though less prevalent, tariff escalation reduced protection enough to significantly im- also affects imports of industrial products-es- prove market access and boost agricultural pecially at the semiprocessed stage. Examples of trade (Diakosavvas 2001). such products, in which many developing coun- Tariff peaks in agriculture occur frequently tries have a comparative advantage, include tex- on processed products and temperate commodi- tiles and clothing; leather and leather products; ties. They are less common on unprocessed fruits wood, paper, and pulp; furniture; rubber prod- and vegetables and tropical commodities, which ucts; and metals. are not produced in high-income countries but 46 MARKET ACCESS AND THE WORLD'S POOR are major export crops of least-developed coun- Support is often rationalized on the non- tries. Thus tariff peaks could be seen as not tar- economic benefits of agriculture, which are geting developing countries in particular, since not properly valued by the market-such as such tariff peaks can be found where market environmental protection, food security, and shares of developing countries in Quad imports maintenance of rural communities (Winters are comparatively low. However, many devel- 1990; Maier and Shobavashi 2001). But exten- oping countries in temperate zones have the sive support may be counterproductive for potential of competing as lower-cost producers these goals because subsidies, in addition to ac- in temperate commodities. Hence besides pro- counting for the "multifunctionality" of agri- viding sizeable market price support to high- culture, have a number of side effects. For ex- income countries' producers, developing coun- ample, production-linked subsidies encourage tries' exporters may be displaced by high tariff environmentally unsustainable farming prac- peaks, especially in the EU where intra-EU trade tices, boosting the use of chemicals, fertilizer, is duty-free. Indeed, intra-EU trade in product and fuel in order to produce additional output groups with a high share of tariff-peak lines is beyond what competitive conditions would prevalent, at about 70 percent of EU countries' dictate. Agriculture now thus contributes about agricultural imports. By contrast, when tariff one-fifth of global greenhouse gases-50 per- peaks are less widespread, non-EU suppliers cent of methane and 70 percent of nitrous seem to have more opportunities. oxide-while high-income countries account Concerns about market access also arise for the major share of global agricultural green- from the poor performance of tariff quotas house gas emissions (OECD 1999b).7 Enhanc- (TRQs) introduced by the URAA with the aim ing the environmental performance of agricul- of securing a minimum level of market access.5 ture remains a challenge, but efforts should rely The average fill rates of TRQs have been low on appropriate incentive policies, tailored to and declining, from 67 percent in 1995 to 63 local environmental circumstances and de- percent in 1998, while about a quarter of tar- mands (OECD 2001d). iff quotas were filled to less than 20 percent. Production-related support in high-income Evidence as to whether the method of adminis- countries also distorts agricultural commod- tration of tariff-quota allocations may have an ity trade and affects developing countries. It influence on the fill rates is still unclear (WTO boosts production of agricultural commodi- 2001; OECD 2001a). But the low fill rate ties and reduces agricultural imports, thus dis- could reflect high "in-quota" rates; in some placing developing-country exports in high- Quad markets in-quota rates are above the av- income countries' markets. The case of the erage for agriculture (OECD 1999a). And, for U.S. subsidies to sugar producers illustrates specific products, over-quota rates skyrocket- both the pernicious impact of support on de- such as the EU 130 percent tariff for above- veloping countries' exporters, and the large quota bananas. costs borne by high-income countries' con- sumers and taxpayers (box 2.2). Support to agriculture is sizable The unwanted production surpluses are and growing dumped into world markets with the aid of At an estimated $245 billion in 2000-about export subsidies, depressing prices for many five times the level of international development temperate agricultural commodities (Burfisher assistance-support to agricultural producers in 2001). The case of growing EU exportable high-income countries remains sizeable (OECD surpluses of wheat illustrates the potential dis- 2001b). Total support to agriculture (as defined tortions to trade (box 2.3). The incidence is by the OECD) is even higher, at about $327 bil- generally negative for agricultural exporters- lion in 2000-or 1.3 percent of OECD coun- especially developing countries that export tries' GDP.6 temperate commodities or have the capacity of 47 GLOBAL ECONOMIC PROSPECTS Box 2.2 U.S. sugar policy and its impact on imports T he United States began to directly intervene to support agricultural commodity prices in 1933 Steady support to U.S. sugar producers with the introduction of the Agricultural Adjustment Producer support estimate Act. The Act has been modified many times, but is (PSE), in percent of gross Ratio of U.S. sugar price still the basis of most of the U.S. agricultural policy. term receipts to tree-market price Sugar is one of the most protected commodities so under U.S. policy (see figure). 3.5 The United States is the world's largest con- 70 U.S. sugar price sumer of sweeteners, with the equivalent of 142 60 pounds of raw sugar consumed per person per year. 50.5 The U.S. sugar industry is heavily subsidized, with about half of sugar producers' revenues coming from 40 PSE refined sugar equivalent government support. U.S. sugar producers have been o 1.5 protected from lower world market prices since the early 1980s, by successive farm legislations that pro- 20 1 vided price supports through restrictive import con- 10 PSE total agriculture 0.5 trols. On average, U.S. sugar producers have received 2.6 times the world market price for sugar since the o 1 1 1 1 1 1 2000 .i 90 (e i)1986 1988 1990 1992 1994 1996 1998 2000 mid-1980s (see figure). Apart from protecting sugar production, domes- Source: OECD; World Bank. tic support to sugar also provides higher than world market prices to corn syrup producers. This has en- couraged the development of an important High Fructose Corn Sweetener (HFCS) industry that now fell by one-half over the period, from 4.3 million supplies half of the country's sweetener consumption, tons in 1980 to slightly less than 2 million tons in especially in products such as soft drinks. HFCS pro- 1998. The sugar policy costs foreign sugar producers duction is now four times higher than in 1980, and an estimated $1.5 billion in lost sales. surpasses sugar production, which has increased by about 50 percent (according to the U.S. Department of Agriculture). Sugar imports by the United States Source: World Bank staff, based on Shtales and others 1999. becoming low-cost exporters. There are also, base-period level.8 Subsidies that were exempt however, benefits for net food-importing devel- from reduction commitments now account for oping countries from lower import prices for about 60 percent of total OECD-country agri- food (Freeman and others 2000). cultural support, even though some of these The URAA also covered trade-distorting subsidies may affect production and trade measures of support and export subsidies (box (OECD 2001a). 2.4). The value of support subject to reduction The overall level of support to producers-as commitments in OECD countries declined sig- measured by the OECD's producer support esti- nificantly, to about 65 percent of its level in mates (PSE)-has further increased since 1998, the base period. However during the imple- in response to the decline in world commodity mentation period, this was largely offset by in- prices, and now represents about 35 percent of creased support under measures exempt from gross farm receipts (figure 2.6).9 And because reduction commitments, so that in 1997 over- support is counter-cyclical, it insulates farmers all support was practically unchanged from its in high-income countries from changes in world 48 MARKET ACCESS AND THE WORLD'S POOR Box 2.3 Wheat production with CAP support A griculture was given a central role when the original European Economic Community (EEC) EU wheat production and net imports established the Common Agricultural Policy (CAP) Million tons in 1957. The basic market support set out in the 100 Treaty of Rome remains much the same today, de- spite successive reforms to the CAP since the early 80 1990s. The CAP was very successful at achieving its 60 goals of food self-sufficiency and stable producer prices. In fact, it was so successful that it encouraged 40 farmers to produce more than was needed, which caused intervention stocks to build and commodities 20 to be exported using export refunds. Wheat produc- 0 tion is a case in point. The first nine countries to join the EEC account -20 for nearly 90 percent of EU wheat production. These Net imports -40 countries have adjusted to the high and stable wheat 1960 1970 1980 1990 2000 prices established by the CAP, and they have re- sponded by increasing yields by 2.5 percent per year since 1970, compared to only 1 percent per year for the United States, the world's largest wheat exporter. Domestic support for wheat (as measured by the of the CAP was that lower cost producers were de- OECD's producer support estimates) remains size- prived of a market for their products. Argentina for able in the EU despite several reforms to the CAP. example, is a low cost producer that could supply From 52 percent of gross farm receipts on average in wheat to the EU. With more than 50 percent of its 1986-88, it declined only marginally to an estimated exports concentrated on agricultural products and 48 percent in 1998-2000. agro-processing manufactures, Argentina in particu- The impact of high wheat prices was not only to lar may be suffering from trade distorting subsidies increase production, but also to reduce demand and (see Nogues 2000). further contribute to the surpluses. Net exports of wheat surged to 22.8 million tons in 1992, and then declined somewhat due to CAP reform measures during the 1990s (see figure). One of the consequences Source: World Bank staff. prices and makes production less responsive to creasing support for specific "sensitive" com- swings in demand. As a result, world commod- modities. Many commodities of export interest ity prices become more volatile, and during for developing countries remain heavily subsi- downturns the burden of adjustment is shifted dized-such as, for example, rice and sugar, disproportionately to producers in developing where support covers as much as 80 and 45 per- countries who enjoy much lower levels of sup- cent of gross farm receipts (OECD 2001b). port (Tyers and Anderson 1992; Winters 1994). Over the past 15 years support to agriculture Export subsidies are particularly in high-income countries has declined only mar- damaging ginally as a share of gross farm receipts. The The effectiveness of URAA in disciplining ex- outlook is unclear, because the reduction com- port subsidies is also questionable. Because ex- mitments are sectorwide, allowing governments port subsidies in the 1986-88 base period were much leeway to target the reductions, while in- sizeable, the limited reduction commitments 49 GLOBAL ECONOMIC PROSPECTS Box 2.4 Bringing support to agriculture and export subsidies under multilateral rules: A long-awaited endeavor key objective of the URAA was to reduce trade-distorting support to agriculture, while Billions of dollars creating room for government policies to design ap- Soo propriate nondistorting support schemes, in response to country-specific circumstances. Three main 250 categories of support were distinguished 200 a) Trade-distorting support (often referred to as "amber box" measures), such as market price- 150 support through administered prices supported by restrictive trade measures and production- 100 related subsidies (based on output or on the use of inputs). 5o b) Support with no, or minimal, distorting effect on trade (often referred to as "green box" measures). 0 19 These may include a vast array of programs, such 13 Aggregate measure of support as decoupled income support measures; payments 0 Blue box B Green box covering services for research and development; pest and disease control, infrastructural services; domestic food aid; structural adjustment and re- gional assistance; and environmental programs. c) A category of direct payments under production- Export subsidies in agriculture allow countries to limiting programs--the so-called blue box mea- export production surpluses to the world market at sures-was also distinguished. prices below the high prices prevailing in their domes- tic markets. Export subsidies were about $7 billion on Reduction commitments were scheduled on average in 1995-98, of which 90 percent was granted trade-distorting support, expressed in terms of a "total by the EU. In the URAA high-inccme countries agreed aggregate measurement of support" (AMS). Under the to reduce base-period subsidized exports by 21 per- URAA, developed countries are required to reduce cent, in equal steps over six years--and to cut the cor- total base-period AMS by 20 percent over a period of responding budgetary outlays by 36 percent. Develop- six years. Developing countries with AMS commit- ing countries agreed to a 14 percent reduction in ments are subject to a 13 percent reduction over 10 subsidized export volumes over a 10-year period. years. Measures in the "green box"-and also, under certain conditions, in the "blue box"-have been exempt from URAA reduction commitments. Source OECD 2001a; WTO 2001; World Bank forthcoming. taken in the URAA leave broad margins for creased for many products of export interest to continued subsidization. Thus for a number of developing countries. For example, subsidized products, permitted subsidized exports during exports of wheat represented 25 percent of total URAA implementation were larger than actual wheat exports in 1998, up from 7 percent in subsidized exports in the first half of the 1990s 1995, while subsidized exports of sugar rose to (OECD 2001a; World Bank forthcoming). And 31 from 19 percent in the same period (Ingco the share of subsidized exports has even in- and Winters 2001). 50 MARKET ACCESS AND THE WORLD'S POOR Figure 2.6 Support to agriculture in the ... partly due to the fall in commodity Quad is growing ... prices Producer support Producer support estimate in percent In U.S billions Total food prices in estimate in percent of gross farm receipts (left axis) (over bars) constant dollars of gross farm receipts (weighted index; 1990=100) (total OECD countries) 70 115 45 52 56 O 11986-90 01991-94 110 Producer support [f 1995-97 estimate (right) 40 0 1998-2000 50 - 105 98 _ 105 10 .... ...-... 35 40 - 100 PSE trend . 30 - - 95 30 20 - - 4 90 Food prices (left) 25 10 - - 85 0 I I , 80 20 Japan EU Canada United 1986 1988 1990 1992 1994 1996 1998 2000 States Source: OECD. Source: World Bank; OECD. In the EU, the Agenda 2000 Common Agri- portant step to improve developing countries' cultural Policy (CAP) reform marked a step in access to high-income countries' markets, be- the right direction to reduce the need for ex- cause it became very difficult for the importers port subsidies by cutting the support prices for to introduce new quotas. Moreover, the ATC cereals, beef, and dairy, but it is unlikely to abolished voluntary export restraints in re- be sufficient to eliminate the EU exportable sponse to pressure from developing countries. surpluses in the years ahead. In addition to di- These measures were identical in form with the rect export subsidies, officially supported ex- MFA quotas. port credits have expanded during the 1998 However the effectiveness of ATC in freeing Asian financial crisis, and are largely used in up markets has been limited by two main short- the United States. By resulting in targeted cost comings. First, scheduled quota integration is discounts for buyers, export credits might have "back-loaded," with quota-free market access similar distorting effects on trade as direct ex- for nearly half of all imports due only at the end port subsidies. The URAA called for negotia- of the transition. Hence the transition is un- tion of export credit disciplines, which has not likely to be smooth for currently shielded pro- yet been achieved. ducers. This could disrupt the post-ATC regime by encouraging calls for higher tariff protection, Due to remaining restrictions on textiles or for more intensive use of contingent protec- and clothing, developing countries forego tion measures (box 2.6). And in textiles, after sizeable export earnings the Uruguay Round, the use of contingent pro- The Uruguay Round Agreement on Textiles tection measures has increased faster than in and Clothing (ATC) provides for the gradual other sectors. In 1998-99, initiations of an- phaseout of the multifiber arrangement (MFA) tidumping investigations in textiles represented country-specific quotas over a 10-year period, 11 percent of total, up from only about 5 per- ending in 2005 (box 2.5). The ATC was an im- cent, on average, in 1990-92 (WTO 2001). 51 GLOBAL ECONOMIC PROSPECTS Box 2.5 A primer on the agreement on textiles and clothing mhe Multifiber Arrangement (MFA) that entered by an additional 16 percent in the first step, 25 per- 1 into force in 1974 (like its predecessors the cent in the second, and 27 percent in the third. Short- and Long-term Cotton Arrangements between The ATC is being implemented in four steps. In 1961 and 1973) established rules for the imposition the first step, which took effect on January 1, 1995, of country-specific quotas, either through bilateral WTO members had to secure quoca-free market ac- agreements or unilateral actions. This conflicted with cess matching, at a minimum, 16 percent of the total the General Agreement on Tariffs and Trade (GATT) volume of their 1990 imports. In the second step, principle of nondiscrimination against trading part- which started on January 1, 1998, an additional 17 ners. As of 1995, only the United States, EU, percent of total 1990 imports had to be integrated, Canada, and Norway continued to use quotas to re- followed by an additional 18 percent in the third strict their imports of textiles. step, which commences on January 1, 2002. Finally, The return to GATT rules has two components: on January 1, 2005, quota-free access corresponding (1) a schedule for freeing textiles and clothing from to the remaining 49 percent of total 1990 imports import quotas (the "integration" component of the must be secured. ATC); (2) additional provisions for accelerated The choice of products to be integrated is left to growth of remaining non-integrated quotas (the so- the importing country, but they must cover at least called liberalization component of the ATC). Prod- one item from each of four major product groups: ucts remaining under restriction will be allowed an yarns and tops, fabrics, made-ups, and clothing. additional increase in quota growth rates-above the general 6 percent annual growth agreed under the MFA. Such products will have their quota increased Source: Based on ICTB 1999. Second, the ATC rules for the removal of Due to the slow pace of the liberalization, quotas are framed in terms of overall import potential benefits for developing countries are shares in textiles and clothing, rather than in being eroded, and foregone export earnings are terms of the number of quotas. This allows sizeable. For example, on current trends, the importing countries the leeway to select the share of intra-EU trade in textiles and clothing products to be freed of quota restrictions in could further decline from 49 to around 43 per- each step, which slows the pace of liberali- cent of total EU countries' imports by the ATC zation.10 Up to 2000, more than 33 percent expiration in 2005. Assuming a twice-as-fast of trade was integrated, fulfilling the mini- decline under a more ambitious liberalization, mum ATC requirements. But products that this share could drop by an additional 7 per- have been freed of quotas by the EU and the cent. Thus foregone export earnings for re- United States represent only small shares of strained developing countries in the EU could their total textile and clothing imports-about be as high as $10 billion a year. In the United 6 percent of 1995-97 imports for the United States, after the creation of the North American States and less than 5 percent for the EU Free Trade Agreement, restrained suppliers were (ITCB 1999). Moreover, the products of inter- displaced by booming textile and clothing ex- est to developing countries that were inte- ports from Mexico, which grew by about 35 grated tend to have low value added-such as percent per year. Despite these trade diversion tops, yarns, and fabrics-with clothing repre- effects, the sharp increase in Mexican exports senting only a small share of the total. illustrates the potential for other restrained low- 52 MARKET ACCESS AND THE WORLD'S POOR Box 2.6 Antidumping-and better alternatives import-competing firms are often tempted to resort to antidumping laws-which are permitted by Developing countries are victims and WTO rules-to allege unfair trade practices by for- players in the rising game of antidumping eign competitors. A firm is said to be dumping if its Antidumping cases initiated by developed and developing export price is less than either the price in its home countries, 1986-1998 market or the average cost of production. Antiduip- 220 ing laws enable nations to impose offsetting duties 200 Developed on imports found to be both dumping products on 180 the domestic market and causing "material injury" 160 to a domestic industry. The main users of these laws 140 1 206 were developed countries, but increasingly develop- 100 ing countries have taken recourse to these laws (see 0 Figure). Industrial and developing nations are 60 Developing equally targeted by antidumping actions. 40 In addition, some nations take action against 20 imports that they suspect may have been subsidized 0 by another government. These so-called countervail- 1986 1988 1990 1992 1994 1996 1998 ing duty cases are also allowed under WTO rules and, if an investigation reveals that allegedly subsi- dized imports have injured a domestic industry, then causing serious injury to a domestic industry. Be- a tariff can be placed on the products in question. cause the import protection is temporary, trading Both antidumping laws and countervailing duty laws partners know that their market access has not been are referred to as "unfair trade laws," reflecting the permanently reduced. By contrast, the antidumping view that dumping and subsidization tilt the com- and countervailing duty laws are often implemented mercial playing field towards foreign firms. However in such a way that the tariffs once imposed they are the more widespread resort to "unfair trade laws" is almost never withdrawn. Worse still, if nations be- diluting the gains from trade liberalization. lieve that the market access obtained during a trade Disrupting surges in imports can be far better negotiation are going to be permanently eroded by handled through the use of safeguard measufres. the use of the unfair trade laws, then they will be less These afford domestic firms the chance to adjust to inclined to start trade negotiations in the first place. greater competition from abroad, but do so only for a fixed period of time. WTO rules allow members to Sowrce: World Bank staff. impose temporary restrictions on imports that are cost producers to expand their exports, should be twice the estimated amount in the EU- market-access obstacles be removed. equivalent to about 12 percent of total devel- Evaluating the impact of MFA quota aboli- oping countries' textile and clothing exports. tion requires a model comprehensive enough Market access in textiles and clothing will to take into account the interplay between remain restricted even after the MFA-related suppliers, as well as the sectoral interactions quotas have been abolished, because tariff bar- of each economy (see also chapter 6, and riers are high. While 90 percent of total high- Kathuria and others (2001) for South Asia). income countries' imports of manufactures face Given the equally slow pace of liberalization in tariff rates below 10 percent, only about half North America, a rough estimate of foregone of textile and clothing imports face such low export earnings for developing countries could tariffs. Moreover 28 percent of total OECD 53 GLOBAL ECONOMIC PROSPECTS countries' imports of textile and clothing still face tariff peaks, down only marginally from Figure 2.7 Despite preferences, LDC 35 percent in the pre-Uruguay Round regime exports to the Quad often face high tariffs (OECD 2001c).11 In percent of LDC exports to each market (1996-99) 100 Preferential market access for developing EU country exports - Japan Preferential access schemes to high-income 80 - SUnited States countries' markets, such as the generalized sys- C tem of preference (GSP) and the LDC regimes, 60 aim to partly mitigate the effects of high most- favored nation (MFN) tariffs on export prod- 40 ucts of developing countries. The United Na- tion's 48 least-developed countries benefit from 20 LDC preferential access in all Quad countries, where 75 percent of their exports are sold.12 0 Even though on average, these preferential Low tariffs Medium tariffs Tariff peaks schemes look relatively generous (Hoekman <5% -15 >15% and others 2001), a number of factors erode Source: World Bank staff calculations. their effectiveness in reducing trade barriers faced by poor countries. First, preferences mainly apply to products that already face rel- atively low MFN tariffs (below 10 percent). The margins of preference on tariff peaks are port share in the market of the GSP-granting significantly lower-with the exception of the country. EU, where the LDC preference margin for tar- There is evidence that tariff preferences help iff peak products is about 70 percent. This the least-developed countries take advantage of margin is only 25 percent in Canada and 30 better export opportunities in Quad markets. In percent in Japan and the United States. Reflect- the post-Uruguay Round period, LDC exports ing the selectivity of preferences and the struc- to the EU that receive high preferences, have ture of LDC exports, high tariffs are thus com- grown by about 8 percent per year on average, mon in some Quad markets on products on outpacing growth of LDC exports that receive which LDC beneficiaries reveal some compara- medium or low preferences (figure 2.8). A sim- tive advantage (figure 2.7). ilar pattern is seen in Canada and Japan.13 Second, tariff preferences under GSP and LDC regimes can also be easily eroded by non- On balance, global tariffs penalize tariff measures, such as antidumping, safe- developing countries- guards, rules of origin, and graduation mecha- The post-Uruguay Round tariff structure penal- nisms. The case of the safeguard measures izes developing countries as a whole because applied by Japan on imports of Shiitake mush- their exports tend to be concentrated in prod- rooms from China illustrates this point (box ucts where market access is highly restricted. 2.7). Finally, the GSP (and LDC regime implic- Trade-weighted applied tariffs convey a sense of itly) have graduation mechanisms that are re- tariff incidence across countries and product lated to income and market shares. They are groups (figure 2.9).14 In manufactures, develop- time-bound and subject to (uncertain) renewal. ing country exporters face, on average, higher Countries graduate if they pass a certain per trade-weighted tariffs than other suppliers, both capita income threshold and if they expand in high-income and in developing countries' their exports of products beyond a certain im- markets. Tariff walls faced by developing- 54 MARKET ACCESS AND THE WORLD'S POOR Box 2.7 Mushroom wars Shiitake mushroom are among the most popular use inspection requirements as a veiled form of pro- mushrooms in Japan, where an estimated 30,000 tection. But since April 17, 2001, imports over 8,000 farmers grow them. Japanese trading companies tons face a tariff of 266 percent, while an amend- began in the early 1990s to encourage Chinese farm- ment to the tariff code imposed temporary emer- ers to use Japanese spores and modern cultivation gency import curbs. The import curbs will be imple- techniques to improve the quality of Chinese mush- mented for up to 200 days through November 8. It rooms to Japanese standards. The effort was a huge is the first time Japan has invoked import curb mea- success. In the humid mountainous climate of Fujian sures under the WTO's ordinary safeguard mecha- province, as well as in Shandong, farmers quickly nism designed to slow imports to allow a specific adapted and China became a global mushroom industry to adjust to heightened competition from giant. The same techniques were applied to other foreign suppliers. The Chinese government urged mushrooms, and exports climbed to $120 million Japanese officials to reconsider the action, to no per year. Farmers' incomes in Fujian and Shandong avail. China has retaliated by imposing punitive rose, and consumers in Japan seemed happy to get duties on several Japanese exports. It is likely that the shiitake mushrooms at one-third the price of the mushroom war is not over. domestically produced shiitakes. But Japan's shiitake farmers feared for their jobs Source: The Economist. February 8, 2001: The Guardian, April and sought protection from the government. The 19, 2001; Peoples' Daily, April 17, 2001; Financial Times, June government responded. Japan at first threatened to 20, 2001. country exporters of manufactures in develop- tries, high levels of tariff protection in the South ing countries' markets are about three times may also impede prospects for export-led higher than in high-income countries' markets. growth. Trade in agriculture may suffer more In agriculture, developing-country suppliers from high levels of protection in middle-income face lower trade-weighted tariffs than do other developing countries because markets in these exporters, both in high-income and in develop- countries are growing fast, reflecting fast popu- ing countries' markets. This is because trade lation and income growth. High tariff protec- preferences to some extent mitigate the impact tion in middle-income developing countries may of tariff protection on developing countries, also damage the export opportunities of low- while a large share of developing countries' income countries, especially in agriculture and exports is in tropical commodities, for which in textiles where the export market shares of tariff protection is relatively low. By contrast, low-income countries have increased rapidly. high-income countries' agricultural exports are mainly concentrated in temperate agricultural -and denies the world's poor access to commodities and dairy products, which face the global markets widespread tariff peaks. Because developing countries are home to the Because average applied tariffs in agricul- world's poor-56 percent of the world's pop- ture are higher in developing countries, South- ulation defined as those living on less than South trade of agricultural commodities faces $2 per day (World Bank 2000c)-high tariff higher trade-weighted tariffs than exports from barriers on developing countries' exports act the South to the North (South-North trade). as a roadblock to market access by the poor. With an increasing share of developing coun- Compared with the nonpoor of the world, tries' manufactured and agricultural exports poor people are more exposed to high penal- being directed toward other developing coun- ties of the global system of protection. 55 GLOBAL ECONOMIC PROSPECTS Figure 2.8 LDC exports can grow fast Figure 2.9 Opposite patterns of tariff when tariff preferences are significant incidence in manufactures and LDC export growth agriculture (1996-99, average per year, in percent) Trade-weighted tariffs (average 1998-99), in percent 30 35 25 3Agricultural commodities 25 -3 30 20 -E EU 5- 15- 25 El importer developing F 10 0 - - -,20 0 15 Manufactures -5 -- United States Japan 10 -10 -15 5 High Medium Low preference preference preference 0 Note: Rankings of LDC preference margins on GSP tariffs: o& >10 percent = high preference; 3-10 percent = medium 0 ON09 preference; <3 percent = low preference. Source: World Bank staff calculations, based on Note: Trade-weighted tariffs, using 1997 applied tariffs U.N. Comtrade. and trade weights: excluding intra-EU trade. Source: Based on IMF-WB 2001. The world's poor generally earn their living in the rural sector and other labor-intensive activities-such as light manufacturing, infor- Liberalizing trade to promote mal services, and construction. When these development products find their way to the world markets, emoval of trade barriers on labor-intensive they face high tariff barriers-such as those R products will generate shared benefits, faced by agricultural commodities and labor- both for high-income and developing countries intensive manufactures. Labor services face (these are quantified in chapter 6). Benefits for particular restrictions-for example, restric- developing countries would include greater ac- tions on temporary cross-border movements cess to high-income countries' agricultural and of workers for the provision of construction apparel markets and more buoyant demand in services (see chapter 3). industrial countries as a result of lower prices One way to quantify the incidence of pro- to consumers. Middle-income countries that tection-albeit in rough fashion-is to look at have access to international capital markets effective tariffs faced by the different income but still depend greatly on exports of protected groups in access to the world markets. Despite products (for example, Argentina), could pos- the existing preferential access schemes for de- sibly see a decline in the risk premia they face, veloping countries' exports, the world's poor because more buoyant growth of export rev- face tariffs that are more than twice as high as enues could make their balance of payments the nonpoor face (box 2.8). This fact is inde- less vulnerable to economic swings. With bet- pendent of their position in the relative scale ter access to global markets, domestic policy of poverty. Making world merchandise trade reforms become important to create export work for the world's poor would require bold opportunities and absorb the dynamic gains steps to remove this disparity. from trade. 56 MARKET ACCESS AND THE WORLD'S POOR Box 2.8 Calculating effective tariffs faced by the poor E ffective tariffs faced by people in different in- come groups convey a sense of uneven access to The poor face high tariffs the world markets (see figure). For people in each Effective tariff faced by income groups (in percent, 1997-98) income group, effective tariffs are calculated on the 16 basis of the trade-weighted tariffs faced by exports 14 of their home countries. For simplicity, it is assumed that all poor can be found in labor-intensive mer- chandise production, while the nonpoor earn their 10 - living across the whole array of economic activities. Thus calculations of effective tariffs on those living 8 on less than $2 per day are based on trade-weighted 6 - tariffs faced by countries for exports of agricultural 4 products and labor-intensive manufactures. Calcula- tions of effective tariffs on the nonpoor are based on 2 - trade-weighted tariffs faced by countries across ex- 0 - ports of all goods. Trade-weighted tariffs are calcu- Deeplypoor Poor $1-$2/day pdoor lated from using 1998 applied tariffs and trade weights. Effective tariffs faced by each income group are calculated as the sum of trade-weighted tariffs faced by the exports of different countries, using as India represents 27 and 40 percent of poor people in weights the share of each country's population in each of the two groups. The trade-weighted tariffs each income group (based on 1998 poverty data). on labor-intensive exports from China and India are Since by global standards even the relatively poor in multiplied by these population shares to determine all high-income countries have consumption greater the contribution of the two countries in the effective than $2 per day, the whole population of these coun- tariffs faced by each of the two groups of poor. But tries is in the nonpoor group. China and India also account for 19 and 4 percent, Due to their size, China and India are the two respectively, of the world's nonpoor. These popula- single countries that weigh more in these calcula- tion shares, along with the "all-inclusive" trade- tions. The trade-weighted tariff on exports of labor- weighted tariffs faced by China and India (estimated intensive products from China is 15.5 percent and at 8.3 and 8.5 percent) are used in the calculation of for India 15.1 percent. China accounts for 29 and effective tariffs faced by the nonpoor. 21 percent respectively of the world's poor and deeply poor (those living on less than $1 a day). Source: WorLd Bank staff caculations. Domestic policies to create ambitious, the pace of agricultural reforms export opportunities has been uneven both at the commodity and Developing countries have gone a long way to- country levels (Townsend 1999; Shepherd and ward removing many of the domestic obstacles Farolfi 1999; Akiyama and others 2001). to export-led growth. Tariffs are lower every- Moreover a number of structural impediments where, the anti-export bias embedded in the hamper export diversification into manufac- domestic trade regimes and sectoral policies tures in the poorest countries-especially in has been reduced, while more sound macro- Sub-Saharan Africa (Fosu and others 2001). economic policies have led to more competitive Deepening the reform process in two direc- exchange rates. However, while macroeco- tions is key to realizing the trade promise for nomic policy and trade policy reforms were growth and poverty reduction: (a) reducing 57 GLOBAL ECONOMIC PROSPECTS further tariff and non-tariff barriers to trade in should the preferential market access regimes a context of supportive policies that link the for specific commodities be discontinued. This poor to expanded market opportunities and would require increased donor support. At the cushion transitional costs for any displaced same time, policy should remove distortions, group; and (b) building trade capacity by up- with the aim of facilitating the redeployment grading "behind-the-border" institutions, rang- of labor and released resources from the in- ing from customs and ports to telecommunica- dustry that enjoyed support. Although labor tions and domestic transport. markets are inherently more flexible in devel- oping countries, distortions-linked, for ex- Reducing tariffs and other barriers to ample, to state enterprise employment-often trade can increase incomes, but obstruct labor markets and hold back the ad- adjustment costs cannot be ignored justment to reforms. Reducing tariffs and other trade barriers will not automatically lead to higher growth. Trade Building trade capacity by upgrading policy cannot substitute for a development "behind-the-border" institutions program. However trade reform is an impor- If a country's investment climate is poor and its tant component of a development strategy, and institutions and infrastructure are weak, sim- developing countries, with average tariff levels ply changing relative price incentives through three times that of the high-income countries, trade policy may do little to promote sustained have ample scope for capturing further gains growth. In several cases, as for example, in from trade reform. Most analyses suggest that Haiti (box 2.1), failure to respond to opportu- unilateral reduction in barriers can produce nities created by trade liberalization has been the greatest and quickest gains. Several coun- related to poor macroeconomic policies that tries have realized this and undertaken impor- have fed volatility and discouraged investment. tant domestic trade policy reforms-including Weakness of "behind-the-border" institutions Chile, China, and Costa Rica. can have a similar dampening effect, as occurs Improving integration into the world trad- in transport, utilities, and communications. ing system involves lowering trade barriers Improving regulation and competition in these and reforming domestic institutions in ways sectors would strengthen the export response that may initially hurt low-income consumers, by reducing the cost of exporting. In agricul- unskilled workers in sheltered industries, and ture, this is key to ensuring competitiveness previously shielded producers-especially sub- in rapidly expanding markets for high-valued sistence farmers in remote areas with deficient commodities where competition is stiff-such rural infrastructure. Producers of import-com- as, for example, fruits, vegetables, meats, and peting commodities that receive dispropor- cut flowers. tionate support may suffer from lower levels Effective duty drawback and indirect tax of protection, at the same time that poor con- rebate mechanisms, are important to over- sumers benefit from lower prices. come the anti-export bias often embedded in Even though the benefits from trade inte- trade regimes. Export finance is often a major gration would eventually outstrip the costs, constraint inhibiting exports in many low-in- deployment of temporary safety nets-such as come countries. Inadequacies may result from support to displaced producers and retrain- the overall weakness of the financial sector or ing-would help cushion the costs of disloca- may reflect difficulties in assessing creditwor- tion for specific groups, and would ensure that thiness of traders. While ensuring availability trade-led growth is pro-poor (World Bank of trade finance is a matter that needs to be 1997, box 2.9). Efforts would also be needed left to the private sector, any effort to expand to cushion the consequences for affected coun- exports and to promote increased opportuni- tries of the reorientation of export flows, ties for the poor in the export sector needs 58 MARKET ACCESS AND THE WORLD'S POOR Box 2.9 Designing appropriate safety nets to ensure trade reforms are pro-poor S ince segments of the poor may be hurt by trade Direct income support tends to be the most effi- liberalization in the short run, determining the in- cient type of social safety net. But proper manage- cidence of the tariff structure on the poor, and de- ment of means-tested programs of support requires ploying appropriately tailored safety nets is impor- important administrative capabilities, which poor tant to ensure that subsequent growth is inclusive, countries often do not possess. One approach, which and secure domestic support for reform. Deployment was employed successfully in Jordan, is to initially of safety nets raises two broad policy options: Em- provide a money payment to a wide range of house- ploy general social safety nets, or establish safety holds, and subsequently narrow the program to only nets targeted to those who are harmed by the trade low-income families. Because distinguishing the poor reform (World Bank, 2001c). from the non-poor may be difficult, workfare pro- grams may be more generally applicable, and have Country-wide safety nets seem more appropriate gasmyb oegnryapial,adhv t untsywie safynety netgas or ap-roate been proven effective under certain circumstances than special safety net programs for trade-related (Ravallion, 1999), as individuals can self-identify for problems. Fundamentally, it is difficult to justify these programs. safety net programs to poor people who suffer from The World Bank's Poverty Reduction Strategy trade reform and deny assistance to other poor pea- Sourcebook outlines best practices for deploying so- pie who suffer from unemployment from other dis- cial safety nets in event of dislocation. In addition, ruptions, such as technological change, or domestic the Bank is working with other donors through the demand shifts. As the main need for the poor during Integrated Framework studies to ensure that best a difficult transition period is likely to be food, one practices are tailored to local capabilities and institu- approach is a time limited food subsidy and distribu- tions. But there are no easy answers as liberalization tion program. However, targeting a food subsidy is affects the poor differently depending on country cir- difficult, and often subject to abuse, while the bene- cumstances. The Bank intends to further deepen its fits may also spillover to middle and upper income knowledge and provide policymakers with analytical groups. An alternative is an untargeted subsidy on tools needed to answer some key questions. inferior goods, as has been pursued in Egypt (Adams, 2000). Source- World Bank staff. complementary policies to help overcome they have to improve information on market credit bottlenecks. Appropriately managed opportunities; overcome problems of product matching grants can be an effective instrument and country brand; and meet concerns about to assist small firms to penetrate export mar- quality. Foreign partners and FDI can be help- kets. Product standards based on interna- ful in providing needed contacts and expertise. tional norms facilitate market linkages, and But local associations of exporters or produc- act as safety, health, quality, or environmental ers can also help. Cooperatives and similar safeguards. Developing countries face a diffi- ventures can help improve marketing while cult challenge in this area, as they need to es- ensuring that benefits from exports accrue to tablish efficient testing, certification, and lab- small poor farmers. However, transparency oratory accreditation requirements to attain and competition in these institutions is impor- sanitary, phytosanitary, and product stan- tant, or poor farmers may receive lower prices dards. Low-income developing countries need for their outputs. both technical and financial assistance to meet In agriculture, in particular, where the stakes this goal. Marketing of exports is a challeng- for poverty reduction are high, additional com- ing task for all low-income countries, because panion policies and institutions would be 59 GLOBAL ECONOMIC PROSPECTS needed to improve the supply response to mar- tives that extend existing preferential access to ket incentives. Some of these policies demand LDC or African countries, but they all fall considerable up-front mobilization of resources, short of a full coverage.16 and should be backed by donor support. Exam- A number of studies have found that the ples include stepped up investment in rural in- export growth gains for LDCs could be signif- frastructure, which is a key enabler of agricul- icant if all Quad markets granted duty-free ac- tural exports in developing countries. Securing cess to LDC (Hoekman and others 2001; lan- sufficient supply of credit at competitive condi- chovichina and others 2001; UNCTAD 2001). tions is important to encourage private sector The projected trade diversion and decline in investment into storage, transportation, and other developing countries' exports are negli- marketing of agricultural products. Increased gible, since LDCs represent only a small part investment in skills through education and of world merchandise trade, and other devel- training in rural areas is needed to bolster pro- oping countries' exports are more diversified. ductivity in agriculture, and to enhance the abil- According to a study (Hoekman and others ity of absorbing emerging technologies-espe- 2001), even if all Quad members were to grant cially those stemming from the biotechnology LDCs duty-free access for only tariff-peak revolution. items, non-oil LDC exports would increase by But other initiatives in agriculture would an estimated 11 percent, while other develop- need to improve the regulatory and policy en- ing countries' exports would decline only mar- vironment. Continued trade policy reforms ginally-by an estimated 0.1 percent. should redress the still remaining anti-export Extending duty-free market access to all bias in developing countries' agriculture. Re- LDC exporters would also help mitigate the forms of pricing policies should be stepped up, drawbacks of current preferential access because in a number of LDC producer prices schemes targeted on specific beneficiaries. are still compressed compared to border prices, These schemes often distort trade, because they thus limiting export incentives. displace low-cost producers elsewhere in the Efficient land policies and land tenure insti- developing world. The case of the EU prefer- tutions are also key to improving the function- ential regime for bananas illustrates this point ing of land markets, securing property rights (box 2.10). Moreover, preferential access to to farmland, and supporting the emergence of high-income country agricultural markets with more efficient farm structures. Enhancing land highly subsidized domestic prices provides a rights and transferability can increase a farm- premium over world prices to the countries er's ability to produce both for subsistence and receiving the preferences. This form of "aid- for income, improve their incentives to invest, through-trade" is not an efficient way of pro- and enhance their ability to obtain credit (see viding aid because it creates dependence and also Freeman and others 2000). is targeted to particular economic activities rather than to identified areas of need. High-income countries can help High-income countries can also provide Domestic policies in developing countries have "aid for trade." This could include increased a greater chance of success if high-income grant aid for trade policy analysis (such as in countries realize their interest in development the integrated framework program discussed success. One policy high-income countries can in chapter 6), technical assistance on imple- adopt is following the lead of the European mentation of standards, and aid for aspects of Union. Its "Everything but Arms" proposal development that affect the "soft infrastruc- grants duty-free and quota-free access in all ture" of the investment climate, such as gov- but 25 lines related to arms trade.15 Other ernance. No less important is disciplining the Quad countries have also announced initia- burgeoning recourse to contingent protection. 60 MARKET ACCESS AND THE WORLD'S POOR Box 2.10 The banana dispute: good intentions ... bad policies? B ananas are almost exclusively exported by devel- Apart from the income transfer from EU con- oping countries to high-income countries, with sumers to ACP and EU banana producers, a number the four dominant exporters (Ecuador, Costa Rica, of other effects are in place. High prices lower EU Colombia, and the Philippines) accounting for three- banana demand. If EU per capita banana consump- quarters of global exports. The two major importers, tion was the same as in the United States, total ba- the United States and the EU, cover approximately nana consumption in the EU would increase by more 60 percent of the world market, currently estimated than 10 percent. This additional demand would raise between $5 and $6 billion. world prices, while lower-cost banana producers During the 1990s, bananas were a source of would export more. The current quota/tariff combina- trade dispute, often termed the "transatlantic banana tion is an inefficient and expensive way to provide aid. trade war." The trade dispute reflects primarily EU Borell (1999), for example, reported that for every import policies. In 1997, 40 percent of the EU ba- dollar that ACP producers receive as aid through nana market was supplied by domestic production higher banana prices, EU consumers pay $5.30. and duty-free imports from African, Caribbean, and Ecuador, Guatemala, Mexico, and the United Pacific (ACP) countries, with the rest being imported States brought a complaint before the WTO in 1997. from non-ACP banana producers (the so-called "dol- The panel ruled against the EU banana import lar bananas," primarily from Latin America), who regime. The EU has recently reached an agreement were subject to quotas and tariffs. with the United States and Ecuador that allows more By restricting imports from non-ACP exporters, import licenses, based on historical allocation, until the EU import regime causes its domestic banana 2006 when ACP preferences will be retained only prices to be much higher than other markets-on through tariff protection. average, about two-thirds higher than in the United States. Furthermore, the regime de facto guarantees high prices to EU and ACP producers, which has been the political justification for such intervention. Source: World Bank staff. Open regionalism could promote competition, reducing transaction costs, and trade creation- reinforcing nondiscriminatory investment and Regional arrangements continue to proliferate, services policies (World Bank 2001b).17 and are likely to remain an enduring feature North-South regional agreements are more of the trade panorama. Smaller memberships likely to improve welfare than South-South make it easier to negotiate the increasingly im- arrangements, because they usually result in portant issues inherent in trade and regulatory lower trade barriers with less trade diversion, regimes, while small countries often can exer- and because the greater structural differences cise greater influence in regional arrangements. in North-South economies produce greater "Open regionalism" holds the potential to gains from trade creation (World Bank 2000d). stimulate global trade and improve the effi- Although South-South arrangements can be ciency of regional producers. But regional made to work, a number of regional integra- arrangements can also become a vehicle for tion agreements have had negative or ambigu- protection, trade diversion, and unintended in- ous effects on income. In particular, agree- efficiency. Key conditions to benefit from ex- ments between richer and poorer developing panded trade and investment include lowering countries are likely to generate losses for the common external trade barriers, stimulating poorer ones when their imports are diverted 61 GLOBAL ECONOMIC PROSPECTS toward the richer members whose firms are An agenda of multilateral trade policy op- not internationally competitive. tions to make merchandise trade work for the Reflecting large differences in costs between poor would need to respond to three main high-income and developing countries, North- challenges. South arrangements hold also the greatest promise for trade creation in agricultural prod- -reducing distortions in ucts and labor-intensive manufactures. By con- agricultural trade- trast, a regional approach-even on a South- Removal of distortions to agricultural trade re- South basis-seems promising in the regulation quires coordinated efforts in different direc- of services, when combined with a nondiscrim- tions. As a first priority, MFN-applied tariffs inatory approach to liberalization (Subraman- should be reduced, on average by half in high- ian and others 2000). Possible areas of cooper- income countries, and by one-third in develop- ation-by pooling resources and expertise and ing countries. Agricultural tariff peaks in high- by upgrading and harmonizing standards- income countries should be phased out. Tariffs would include domestic regulation in sectors should also become more transparent by limit- such as financial services, telecommunications, ing the use of specific and compound tariffs. power, and transport. The EU and Japan should take the lead because in these countries tariff peaks on agricultural -but multilateral policies hold the imports and specific tariffs are more prevalent. key to a sustained improvement in The size of tariff quotas should be increased market access- and the "in-quota" tariff rates should be elimi- The next trade round has the potential to im- nated to improve the very low tariff quota fill prove access for developing countries' mer- rates, and the over-quota tariffs should be con- chandise exports to high-income markets, par- siderably cut to expand market access. Re- ticularly in agriculture and labor-intensive moval of tariff peaks in the Quad will help re- manufactures, where the stakes for the reduc- duce the tariff escalation that hampers trade tion of poverty are high. The Quad countries- in more processed agricultural products and the United States, EU, Japan, and Canada- higher value-added manufactures. But multilat- would serve their interest in expanding trade eral surveillance should also be enhanced, to and development well if they put serious con- progressively eliminate tariff escalation in both cessions on the table in these areas. high-income and developing countries. Offering to link "aid for trade" to progress As a second step, agricultural tariffs should in reforms in developing countries would also be bound to levels close to MFN-applied rates, serve the interest of development well. The particularly in developing countries where trade round should also provide more interna- bound rates are very high. Binding of tariff tional aid and technical assistance in key sec- rates will improve the predictability of the tors, such as agriculture, where the poor coun- global tariff system. High bound tariffs create tries need to build trade capacity. ample scope for tariff protection to rise without Developing countries too (especially the infringing WTO commitments. Investors' risks middle-income countries) should join multilat- could thus increase, limiting the benefits from eral efforts to further liberalize merchandise more open trade. trade if they want to maximize the benefits The third step would require a much bolder from freer global markets. Because merchan- overhaul of the system of support provided by dise trade among developing countries is set to high-income countries to agriculture. More accelerate further, outpacing the growth of binding disciplines should be introduced on world trade, reducing trade barriers in devel- production-affecting support, also encompass- oping countries holds a key promise in in- ing subsidies that are currently exempt from creasing the development impact of trade. URAA reduction commitments ("Blue Box," 62 MARKET ACCESS AND THE WORLD'S POOR "Green Box," and "de minimis" measures). countries. Similarly, reducing escalation in tar- As a benchmark, the producer support esti- iff codes in the developed and developing mate in high-income (OECD-member) coun- countries alike will produce more efficient, usu- tries should be cut on average by half as a ally pro-poor growth. share of gross farm receipts. This should be coupled with an accelerated phaseout of ex- Notes port subsidies-especially in the EU, where 1. Moreover, in Sub-Saharan Africa agricultural their usage continues to be widespread. To exports are concentrated in five major crops (cocoa, level the playing field, officially supported ex- coffee, cotton, sugar, and tobacco), which, in 1990-95, port credits-which are more prevalent in the accounted for an estimated 62 percent of total agri- United States-should also be brought under cultural exports. Export concentration has hardly multilateral disciplines. changed over time, since these same five crops repre- sented 63 percent of total agricultural exports in the 1970s (Ingco and others 2001). -expanding access in labor-intensive 2. Export processing zones (EPZs) were often used manufactures- extensively-for example, in Tunisia, Bangladesh, and The phaseout of the remaining quantitative Mauritius-to overcome the anti-export bias of do- restrictions in textiles and clothing should be mestic trade policy regimes and support export-ori- stepped up, ahead of the ATC expiration in ented T&C industries. But their effectiveness in pro- 2005, because the removal of quotas has been moting spillovers to the rest of the economy has been questioned. By creating economic enclaves, EPZs often "back-loaded." But this will not be enough to impair backward linkages with the rest of the econ- improve market access for developing countries. omy, as the supply chain of exporting firms may rely Applied tariffs in textiles and clothing remain more on imported, duty-free, intermediate goods than excessively high and should be cut on average on local producers. Such impediments to local produc- by half in high-income countries, and by one- tion linkages could be seen, for example, in the case of third in developing countries. Tariff peaks Bangladesh (World Bank 1999a). u 3. The data on income distribution for the countries should be phased out-especially in the United included in the sample are from household surveys re- States and Canada where they are prevalent. ported in Deininger and Squire 1996, but in some coun- Tariffs on footwear should be reduced across all tries household surveys measure expenditures, while in Quad countries. others they measure income. When household surveys To build confidence that trade in textiles report expenditure (47 out of 129 observations), con- and clothing will be freed up in the post-ATC sumption growth of the poorest 20 percent of the pop- regime, the increasing trend in the use of con- ulation is compared to growth in average per capita consumption-otherwise it is compared to growth in per tingent protection (especially in textiles) capita GDP. Time periods span irregular intervals for should be halted. Reducing the wide discrep- each country-depending on the availability of house- ancies between bound and applied tariffs hold surveys of acceptable quality. To smooth out short- would help build trust, because these discrep- run fluctuations in income or expenditure, time periods ancies provide the scope for using tariffs as span at least 3 years, with an average duration of 6.8 safeguards or for balance-of-payments reasons years. The geographical breakdown of the sample is: East r fAsia-31; Latin America and the Caribbean-SO; Mid- (Laird forthcoming). And multilateral surveil- dle East and North Africa and Europe and Central lance should be enhanced, to eliminate tariff Asia-IS; South Asia-18; Sub-Saharan Africa-15. escalation on labor-intensive manufactures in 4. In developing countries, applied tariffs are, on both high-income and developing countries, average, about three times higher than in industrial countries, partly because developing countries rely more -and eliminating tariff peaks and on trade-related taxes to raise revenue. In developing escalation on all products countries, the coverage of nontariff barriers, including state trading monopolies, has also been considerably Eliminating tariff peaks that now discriminate reduced (Martin forthcoming). against labor-intensive products would add con- 5. Tariff quotas allow a lower-tier, or "in-quota," siderably to the export potential of developing rate to be set at low levels, with the second-tier, or 63 GLOBAL ECONOMIC PROSPECTS "over-quota," rate set at a higher level-close to the African, Caribbean, and Pacific, southern Mediter- level of protection enjoyed before the URAA. The dis- ranean and Eastern European countries in the Euro- tribution of TRQs among countries and product groups pean Union; and those received by Mexico, Israel, An- reflects the incidence of tariffication. More than one- dean countries, and the Caribbean Community in the quarter of all tariff quotas apply to fruits and vegetables, United States. with the next four more important groups being meat, 13. LDC exports to the United States show an op- cereals, dairy products, and oilseeds (WTO 2001). posite pattern. However, high-preference exports rep- 6. Support to agricultural producers refers to the resent less than 1 percent of LDC exports to the United producer support estimate (PSE) computed by the States, which suggests that the outcome could be sensi- OECD. The PSE is an indicator of the annual monetary tive also to other barriers-such as standards, rules of value of gross transfers from consumers and taxpayers origin, antidumping, safeguards, distribution, and so to farmers, measured at the farmgate level, arising from forth. policy measures that support agriculture. These trans- 14. Trade-weighted applied tariffs may to some ex- fers include both government subsidies to agriculture tent underestimate the degree of tariff protection, since (taxpayer transfers) and effective market price support highly restrictive tariffs, which are likely to reduce through trade policies to restrict imports (transfers from trade flows, receive small weights in the construction consumers). A wider indicator of support, calculated by of the index. the OECD, is the total support estimate (TSE). In addi- 15. However, for three sensitive agricultural prod- tion to transfers included in the PSE, the TSE also in- ucts-bananas, rice, and sugar-the liberalization will cludes an estimate of general services support provided be phased in during a rather lengthy transition period, to agriculture-for example, public research and devel- to be completed by 2009 for rice and sugar, and 2006 opment, agricultural schools, inspection services, and for bananas. infrastructure. 16. Japan's "99 percent Initiative on Industrial Tar- 7. To be sure, the global impact of subsidy reduction iffs" does not cover agricultural products where tariff in OECD countries on agricultural greenhouse gas emis- barriers are particularly high. The U.S. "African Growth sions is uncertain, as production could increase in coun- and Opportunity Pact" that grants duty-free and quota- tries with higher emission intensity per unit of output. free access to the United States does not include sensitive 8. In the EU and Japan, production-related total ag- agricultural products (for which there are tariff quotas), gregate measurement of support (AMS) and "blue box" nor apparel and clothing, which have their own prefer- measures still account for the majority of support (83 ential regime-including quotas and relatively restrictive .rules of origin. percent in the EU and 53 percent in Japan), while in the 17. gore , o rr United States support is now provided mainly underTarr Ugreni Statesuprts n pro d m(1997) estimate that Chile was able to profit from its "green box" measures (84 percent). free trade agreement with MERCOSUR due to the fact 9. Domestic support to agriculture is also high out- that it lowered its external uniform tariff from 11 to side the Quad. For example, in the Republic of Korea, at 6 percent. an estimated $17.3 billion on average during 1998- 2000, the PSE covered 66 percent of gross farm receipts, and was more than four times higher than in Canada References (OECD 2001b). Adams, R. 2000. "Self-targeted Subsidies: The Political 10. For example, in 1990 only about 58 percent of and Distributional Impact of the Egyptian Food EU imports were under quota restrictions, which left Subsidy System." Economic Development and Cul- much room to defer the "integration" of restrained tural Change, 49(1), 115-36. products. Akiyama, T., J. Baffes, D. Larson, and P. Varangis. 2001. 11. These high levels of protection also have a large Commodity Market Reforms: Lessons from Two cost for high-income countries' consumers. According Decades. Washington, D.C. to a study, in 1997, quotas and tariffs on textiles and Blackhurst, R., A. Enders, and J. Francois. 1996. "The clothing cost to EU consumers about $10 billion, while Uruguay Round and Market Access: Opportuni- the loss of production efficiency due to the sheltering of ties and Challenges for Developing Countries." In domestic production will have cost another $10 billion The Uruguay Round and the Developing Coun- (Francois, Glismann, and Spinanger 2000). Hence, each tries W Martin and L. A. Winters, editors. New job saved through the delayed freeing up of the EU tex- York: Cambridge University Press. tiles and clothing market costs an estimated $24,000 Borell, B. 1999. "Bananas: Straightening Out Bent Ideas per year in textiles and $35,000 per year in clothing. on Trade as Aid." Presented in the WTO Confer- 12. Other preferential schemes offer better access to ence on Agriculture and the New Trade Agenda in some developing countries, such as those received by the WTO 2000 Negotiations. October 1-2. Geneva. 64 MARKET ACCESS AND THE WORLD'S POOR Burfisher, M. 2001. "The Road Ahead: Agricultural Pol- Ingco, M., and L.A. Winters, (eds). 2001. Agricultural icy Reform in the WTO: Summary Report." Agri- Trade Liberalisation in a New Trade Round. Dis- culture Economic Report No. 797, United States cussion Paper No. 418. The World Bank, Septem- Department of Agriculture. Washington, D.C. ber, Washington, D.C. Currie, J., and A. Harrison. 1997. "Trade Reform and Ingco, M., T. Kandiero, and J. A. Mistiaen. 2001. "Sub- Labor Market Adjustment in Morocco." Journal Saharan Africa and the New WTO Trade Negoti- of Labor Economics 15(3): S44-S72. ations in Agriculture: Economic Analyses of Inter- Datt, G., and M. Ravallion. 1998. "Farm Productivity ests and Options." Presented at the World Bank and Rural Poverty in India." Journal of Develop- Conference on "Leveraging Trade, Global Market ment Studies (34): 62-58. integration, and the New WTO Negotiations for Davis, D.R. 1996. "Trade Liberalization and Income Development." Washington, D. C. July 23-24. Distribution." NBER Working Paper 5693. August. ITCB (International Textiles and Clothing Bureau). Deininger, K., and L. Squire 1996. "A New Data Set 1999. "Agreement on Textiles and Clothing: Eval- Measuring Income Inequality." The World Bank uation of Implementation." August. Geneva. Economic Review (10): 565-91. Kanaan, 0. 2000. "Tanzania's Experience with Trade Diakosavvas, D. 2001. "The Uruguay Round Agree- Liberalization." Finance & Development (37) June: ment on Agriculture in Practice: How Open are 30-33. OECD Markets?" Presented at the World Bank Kathuria, S., W. Martin, and A. Bhardwaj. 2001. "Im- Conference on Leveraging Trade, Global Market plications of MFA Abolition for South Asian Integration, and the New WTO Negotiations for Countries." World Bank, Washington, D.C. April. Development. July 23-24. Washington, D.C. Processed. Dollar, David. Forthcoming. "Post-1980 Globalizers." Khan, M. H. 2000. "Rural Poverty in Developing In "Globalization" Policy Research Report. World Countries." Finance & Development December. Bank, Washington, D.C. International Monetary Fund, Washington, D.C. Fosu, A. K., S. M. Nsouli, and A. Varoudakis. 2001. Laird, S. Forthcoming. "Market Access For Merchan- Policies to Promote Competitiveness in Manufac- dise: How Large Are the Barriers?" In "Trade Pol- turing in Sub-Saharan Africa. Development Centre icy for Developing Countries in a Global Economy: Seminars with the International Monetary Fund, A Handbook." Washington, D.C.: World Bank. and the African Economic Research Consortium. Maier, Leo, and M. Shobayashi. 2001. "Multifunc- OECD, Paris. tionality: Towards an Analytical Framework." Francois, J., H. Glisman, and D. Spinanger. 2000. "The Summary and Conclusions. April, Paris. Cost of EU Trade Protection in Textiles and Martin, W. Forthcoming. "Trade Policies, Developing Clothing." Working Paper No 997, Kiel Institute Countries, and Globalization." In "Globaliza- for World Economics, Kiel, Germany. August. tion" Policy Research Report. World Bank, Wash- Freeman, F, J. Melanie, 1. Roberts, D. Vanzetti, A. Tielu ington, D.C. and B. Beutre, 2000. "The Impact of Agricultural Nogues, J. J. 2000. "El Sistema Multilateral de Co- Trade Liberalization on Developing Countries." mercio y el Desbalance de la Rueda Uruguay." ABARE Research Report 2000.6. Canberra. December. Buenos Aires. Gaiha, R. 1993. "Design of Poverty Alleviation in OECD 1999a. "Post-Uruguay Round Tariff Regimes: Rural Areas." United Nations Food and Agricul- Achievements and Outlook." Paris. ture Organization, Rome. - 1999b. "National Climate Policies and the Harrison, G. W, T. F Rutherford, and D. G. Tarr. 1997. Kyoto Protocol." Paris "Trade Policy Options for Chile: A Quantitative - 2001a. "The Uruguay Round Agreement on Evaluation." Policy and Research Working Paper Agriculture: An Evaluation of Its Implementation No 1783. World Bank, Washington D.C. June. in OECD Countries." Paris. Hoekman, B., F. Ng, and M. Olarreaga. 2001. "Tariff - 2001b. "Agricultural Policies in OECD Coun- Peaks in the Quad and Least-Developed Country tries: Monitoring and Evaluation 2001." Com- Exports." February. World Bank, Washington, mittee for Agriculture. May. Paris. D.C. Processed. - 2001c. "Trade and Development Issues in Non- IMF (International Monetary Fund)-World Bank. OECD Countries." Trade Committee. April. Paris. 2001. "Market Access for Developing Countries - 2001d. "Improving the Environmental Perfor- Exports." Joint Paper. April, Washington, D.C. mance of Agriculture: Policy Options and Market lanchovichina, E., A. Mattoo, and M. Olarreaga. 2001. Approaches." Paris. "Unrestricted Market Access For Sub-Saharan Oxfam International. 2001. "Rigged Trade and Not Africa: How much Is It worth and Who Pays?" Much Aid: How Rich Countries Help to Keep the World Bank, Washington, D.C. February. Processed. Least-Developed Countries Poor." May. London. 65 GLOBAL ECONOMIC PROSPECTS Ravallion, M. 1999. "Protecting the Poor in Crisis." . 2000. "Trade and Poverty: Is there a Connec- PREM Note no. 12, The World Bank. Washing- tion?" In "Trade, Income Disparity and Poverty," ton, D.C. Geneva: World Trade Organization. Revenga, A. 1995. "Employment and Wage Effects Wood, A. 1997. "Openness and Wage Inequality in of Trade Liberalization: The Case of Mexican Developing Countries: The Latin American Chal- Manufacturing." Policy Research Working Paper lenge to East Asian Conventional Wisdom." The 1524. World Bank, Washington, D.C. World Bank Economic Review (11): 33-57. Sheales, T, S. Gordon, A. Hafi, and C. Toyne. 1999. World Bank. 1997. Global Economic Prospects and the "Sugar: International Policies Affecting Market Developing Countries 1997. Washington, D.C. Expansion." ABARE Research Report 99.14. - . 1999a. "Bangladesh Trade Liberalization: Its Canberra. Pace and Impacts." South Asia Region. Novem- Sheperd, A. W., and S. Farolfi. 1999. "Export Crop Lib- ber. Washington, D.C. eralization in Africa: A Review." Food and Agri- - . 1999b. "Bangladesh. From Counting the Poor culture Organization Agricultural Services Bulletin to Making the Poor Count." Washington, D.C. No. 135. Rome. . 2000a. World Development Report 200012001. Subramanian, A., E. Gelbard, and others. 2000. Washington, D.C. "Trade and Trade Policies in Eastern and South- - . 2000b. "Vietnam Development Report 2000: ern Africa." IMF Occasional Paper No. 196. Au- Attacking Poverty." Washington, D.C. gust. Washington, D.C. . 2000c. Global Economic Prospects and the Townsend, R. 1999. "Agricultural Incentives in Sub- Developing Countries 2001. Washington, D.C. Saharan Africa." World Bank Technical Paper . 2000d. Trade Blocs. Washington, D.C. No. 444. August. Washington, D.C. . 2001a. "Uganda Poverty Status Report 2001." Tyers, R., and Anderson, K. 1992. Disarray in World February. Washington, D.C. Draft. Food Markets: A Quantitative Assessment. Cam- - 2001b. "Leveraging Trade for Development." bridge University Press, Cambridge. Development Committee. April. Washington, D.C. UNCTAD (United Nations Conference on Trade and World Bank. 2001c. "Trade Policy and Poverty," in Development). 2001. "Duty and Quota Free Mar- Poverty Reduction Strategy Sourcebook. Wash- ket Access for LDCs: An Analysis of Quad Initia- ington, D.C. tives." London and Geneva. World Bank. forthcoming. Agriculture and the Global Winters, L. A. 1990. "The So-Called Non-Economic Trading Systems, (draft) M. Ingco (ed), May 2001, Objectives of Agricultural Support." OECD Eco- Washington, D.C. nomic Studies No. 13: 238-66. WTO (World Trade Organization). 2001. "Market Ac- Winters, L. A. 1994. "The LDC Perspective," in K. A. cess: Unfinished Business," Special Studies. May. Ingersent, A. J. Rayner, and R. C. Hine (eds.). Agri- Geneva. culture in the Uruguay Round, St. Martin's Press, New York, 157-181. 66 Trade in Services: Using Openness to Grow Services are vital for economic countries have given priority to a change in development- ownership through privatization, while retain- Services are the fastest growing sector of the ing limitations on new entry. Effective regula- global economy, and trade and foreign direct tion ranging from prudential regulation of fi- investment in services have grown faster than nancial services to procompetitive regulation in goods over the past decade. Developing of telecommunications is critical to the success countries have witnessed even faster growth of liberalization, but regulatory weaknesses rates, and their share in world services exports are too prevalent in developing economies. increased from 14 percent in 1985-89 to 18 Liberalization also frequently requires com- percent in 1995-98. Technological progress plementary policies to help improve access to has greatly enhanced the scope for trade in essential services for the poor. The experience conventional services, such as education and fi- of several countries has demonstrated that nance, and also created a host of new tradable universal service goals can be achieved in com- services, such as software development and In- petitive markets. ternet access. Moreover, liberalization in many countries is leading for the first time to the pri- Multilateral engagement can be an vate and foreign provision of services such as important catalyst for liberalization telecommunications, transport, and finance. Even though governments can initiate reforms In virtually every country, the performance of services individually, multilateral engage- of the services sectors can make the difference ment can help in two ways. First, negotiations between rapid and sluggish growth (box 3.1). under the General Agreement on Trade in Ser- Developing countries, in particular, are likely vices (GATS) could help accelerate domestic to benefit significantly from further domestic reform and improve access to foreign markets liberalization and the elimination of barriers for developing countries. However, for these to their exports. The income gains from a re- negotiations to be fruitful, both developed and duction in protection to services are estimated developing countries must recognize mutual to be multiples of those from trade liberaliza- interests in reciprocal liberalization. In particu- tion in goods. lar, developed countries must see the advan- tages of allowing the temporary movement of -but the benefits from liberalization are individual service providers. Developing coun- not automatic tries must see the advantages of multilateral Flawed reform programs can substantially re- agreements to increase competition, enhance duce gains. The largest gains come from elim- credibility of potential domestic reform, and inating barriers to entry, but many developing strengthen domestic regulation. Recognizing 69 GLOBAL ECONOMIC PROSPECTS Box 3.1 Why do services matter for develo,iment? n developing countries, the average share of ser- larly, transport services contribute to the efficient dis- vices in GDP increased from around 40 percent in tribution of goods within a cout try, and are particu- 1965 to around 50 percent in 1999, while in the larly important in influencing a country's ability to OECD countries the average share increased over the participate in global trade. Although these are the same period from 54 percent to over 60 percent. more prominent services, others are also crucial. Among the fastest growing sectors in many countries Business services such as accouning and legal ser- are services such as telecommunications, software, vices are important in reducing transaction costs- and finance. the high level of which is considered one of the most Efficient services not only provide a direct bene- significant impediments to economic growth in fit to consumers, but also help shape overall eco- Africa. Education and bealth services are necessary nomic performance. An efficient and well-regulated in building up the stock of human capital. Retail and financial sector leads to the efficient transformation wholesale services are a vital link between producers of savings to investment, ensuring that resources are and consumers, and influence the efficiency with deployed wherever they have the highest returns; and which resources are allocated to meet consumer facilitates better risk-sharing in the economy. Im- needs. Software development is the foundation of the proved efficiency in telecommunications generates modern knowledge-based economy. Environmental economywide benefits, because this service is a vital services contribute to sustainable development by intermediate input and also crucial to the dissemina- helping alleviate the negative impact of economic ac- tion and diffusion of knowledge. The spread of the tivity on the environment. Internet and the dynamism that it has lent to economies around the world is telling testimony to the importance of telecommunications services. Simi- Source: World Bank staff. these potential mutual gains will allow recip- chandise trade, services are often intangible, rocal "concessions" that benefit both. invisible, and perishable, and usually require In parallel, global cooperation is needed to simultaneous production and consumption.1 provide support for developing countries at The need in many cases for proximity between four levels: in devising sound policies, strength- the consumer and the producer implies that ening the domestic regulatory environment, one of them must move to make an interna- enhancing their participation in the develop- tional transaction possible. Since the conven- ment of international standards, and in ensur- tional definition of trade-where a product ing access to essential services in the poorest crosses the frontier-would miss out on a whole areas. range of international transactions, it is now customary to define "trade in services" to in- Surging trade and investment clude four modes of supply: in services * Mode one: cross-border supply, which is analogous to trade in goods; arises when Trade in services: four modes of supply a service crosses a national frontier, for Services include activities as disparate as trans- example, the purchase of software or in- port of goods and people, financial interme- surance by a consumer from a supplier diation, communications, distribution, hotels located abroad. and restaurants, education, health care, con- * Mode two: consumption abroad; occurs struction, and accounting. In contrast to mer- when the consumer travels to the terri- 70 TRADE IN SERVICES: USING OPENNESS TO GROW tory of service supplier, for example, to to $1.2 trillion in the year 1999, and now ac- purchase tourism, education, or health counts for a quarter of all cross-border trade.3 services. Developing countries as a group have wit- * Mode three: commercial presence; in- nessed an even more rapid (nearly four-fold) volves foreign direct investment, for ex- increase in their services exports, and a conse- ample, when a foreign bank or telecom- quent increase in their share in world service munications firm establishes a branch or trade from 14 percent in 1985-89 to 18 per- subsidiary in the territory of a country. cent in 1995-98 (figure 3.1 right). From a re- * Mode four: movement of individuals; oc- gional perspective, Europe and Central Asia curs when independent service providers (ECA) and East Asia and Pacific (EAP) in- or employees of a multinational firm tem- creased their services exports by a factor of six; porarily move to another country.2 South Asia (SAR) and Latin America and the Caribbean (LAC) kept up with world growth; Services have been among the fastest grow- and Sub-Saharan Africa (SSA) and the Middle ing components of world trade over the last East and North Africa (MNA) lagged behind. 15 years. Over the period 1985-99, the com- Even so, most trade in services still takes place pound annual growth rate for services exports between rich countries. on a balance-of-payments basis-which covers Over the last two decades, there has been a primarily cross-border supply and consump- significant decline in the relative importance of tion abroad-was over 9 percent per annum, transport services in total services exports- compared to 8.2 percent per annum for mer- from around one-third to around one-fifth of chandise (figure 3.1 left). As a result, services total exports-which may in part reflect a de- trade more than trebled its size in fifteen years cline in the relative price of transport services Figure 3.1 Trade in services has grown -and developing countries share in faster than trade in goods- world exports have increased, 1986-98 (compound growth, 1985=1) Percent 3.5 25.0 Services 022.5 F Goods 22.0 3.0 225 Services 20.6 20.0 __ 2.5 18.0 183 180 17.5- m 2.0 1 Goods 15.0 1.5 12.5 1.0, 10.07- 1985 1987 1989 1981 1993 1995 1997 1999 1986 1990 1995 1998 Note: Population estimate from a sample of 100 countries for Note: Population estimate from a sample of 100 countries. period 1985-98. Figure for 1999 is estimate from 69 countries. Source: IMF BoP Rev. 5, through SIMA; EPPG staff World trade defined as (X+IMI)12 calculations. Source: IMF BoP Rev. 5, through SIMA; EPPG staff calculations. 71 GLOBAL ECONOMIC PROSPECTS Figure 3.2 Transport has declined, while "Other" services "other" services have increased Percent of world total services exports 50 F 1980 ""Communications 40 1990 51% Construction 40 01998 7% insurance 5% 30 -e Financial Other 10% businesses 53% Personal, cultural, 20 -and recreational 3% Computer and Royalties and information license fees 5% 0 , In . 12% Transport Travel Other Note: Population estimate from a sample of 89 countries. Source: Trade Handbook, based on IMF BoP rev. 5. Source: IMF BoP Rev. 5, through SIMA; EPPG staff calculations. (figure 3.2). While the 1980s witnessed a is the dominant mode of supply in all sectors growth in the relative importance of travel, the except transport, and to a more limited extent 1990s witnessed a significant increase in the telecommunications.5 share of other commercial services. Detailed in- formation on this last category is not available -but the growth rates of FDI flows to for most countries. Estimates suggest that fi- developing countries are higher nancial services are probably the most impor- The limited evidence available suggests that the tant, followed by construction, communications, bulk of FDI stocks are in the Organisation for and computer and information services. Economic Co-operation Development (OECD) countries (figure 3.3). However, over the pe- Most FDI in services goes riod 1988-97, stocks in developing countries to OECD economies- have witnessed much faster rates of growth, A large amount of "trade" in services takes increasing ten-fold in EAP, seven-fold in SSA, place through an established presence, for ex- and five-fold in LAC, compared to a three-fold ample through foreign direct investment (FDI). increase in the OECD.6 The only exceptions The available evidence suggests that commer- are the three countries in SSA for which data cial presence has been the most dynamic mode are available, where the stock declined by a of services supply in recent years.' This may re- half. In all regions except SSA, the services sec- flect the fact that there has been far greater lib- tor now accounts for nearly half of the entire eralization of foreign investment than of cross- FDI stock-from 1988 levels of less than one- border supply of services, which was either fifth in LAC and less than one-third in SAS. already open or did not witness significant new The limited information on sectoral composi- opening. At the level of individual sectors, de- tion of FDI stocks suggests that nearly half the spite the growing use of information and com- stock in SAS is in financial services, whereas munications technology, commercial presence the stocks in EAP and LAC are more uniformly 72 TRADE IN SERVICES: USING OPENNESS TO GROW Figure 3.3 FDI in services is concentrated -but the growth rates are higher for in the OECD countries- many developing countries Billions of dollars Simple average compound annual growth rates 1,000 50 46.4 FE 1988] 911.4 900 E 1997 40 E All industry 38.1 800 4 31.7 700 28.3 30 2. 600 400 200 101 0 1000 -2.2 011.2 0.6 0.8 7.2 8.W' 9 -4.2 O0r i t a-10 Sub-Saharan South Latin East Asia OECD Sub-Saharan South Latin East Asia OECD Africa Asia America and Africa Asia America and and the Pacific and the Pacific Caribbean Caribbean Source: UNCTAD, FDYTNC database. distributed across finance, transport, storage ports valued at between $40 billion and $120 and communications, hotels and restaurants, billion (World Bank 1995). This mode of deliv- real estate, and other business services. ery is still free of explicit barriers, though regu- latory barriers may impede trade (see box 3.2). Developing countries are becoming players One of the most striking recent examples in exporting services of a developing-country service export success While some developing countries are increas- story is the Indian software industry. Indian ingly investing in other countries to export software exports grew from $225 million in services-for example, Malaysia in environmen- 1992-93 to $1.75 billion in 1997-98 (at an tal services, and South Africa in telecommunica- annual growth rate of approximately 50 per- tions-most supply services via cross border cent).7 A recent report projects annual rev- sales (for example, data processing), to visiting enues of $87 billion, 2.2 million jobs, and a foreign consumers (for example, tourism), and market capitalization of $225 billion for the through the movement abroad of individual Indian information technology (IT) sector by services providers (for example, professional the year 2008.8 By the same year, the IT sector services). Developments in information and could account for 35 percent of India's ex- communication technology have dramatically ports and attract $5 billion of FDI per year. increased the scope for cross-border exports of These figures are not implausible because services, ranging from software development in India still accounts for only half a percent of the Philippines to data processing in Barbados. the world software market, and there are still Rough estimates suggest that the size of the po- wide differences across countries in the cost tential market for developing-country exports of of software development and support. The av- long-distance services could be in the range of 1 erage cost per line of code in Germany (the to 5 percent of the total employment in services most expensive country) exceeds by more than in the seven richest economies-implying ex- four times that of India (the cheapest country) 73 GLOBAL ECONOMIC PROSPECTS Box 3.2 Whose regulations and for what purpose? Challenges in electronic commerce D omestic regulations that affect trade pose the It is of course true that reporting of personal credit main challenge to ensuring open conditions for histories is critical to consumer credit, and, even in electronic delivery of services. Two examples illus- theory, excessively, strict privacy laws could create trate how difficult it is to distinguish between regula- significant asymmetries of information and affect the tions that incidentally impede trade in the pursuit of efficiency of markets (Kitchenman, 1999). This is legitimate objectives and regulations that deliberately not to suggest that there might not be good reasons discriminate against foreign provision for the sake of to protect privacy. However, achieving diverse na- protection. tional objectives without creating unnecessary im- pediments to trade is ideally accomplished through a Privacy multilateral process in which developing countries An issue that could have a profound effect on elec- participate. tronic commerce is privacy. In late 1998, the Euro- pean Union issued a wide-ranging directive that aims Offshore financial services to safeguard the privacy of personal data of EU citi- Several Caribbean countries have become off-shore zens and prevent its misuse worldwide. It is backed financial services centers. However, in recent years, by the power to cut off data flows to countries that their tax and regulatory regimes have drawn fire and the EU judges not to have adequate data protection elicited increased scrutiny. For example, the Financial rules and enforcement. The directive caused frictions Stability Forum (FSF), which assesses conformity wirb with the US, which accused the EU of trying to im- international regulatory standards (including cross- pose laws beyond its own frontiers. A compromise border cooperation) placed many of the Caribbean was reached under which the US agreed to set up ar- offshore centers in the lowest category; the Financial rangements for the processing by companies of Action Task Force (FATF), which is concerned with personal data from the EU, but the issue has not protecting financial systems from money laundering been fully resolved, and criminal use, placed a number of Caribbean cen- The issue could have an impact on developing ters in its list of "non-cooperative jurisdictions," from countries exports of data processing services, and the standpoint of willingness to cooperate with the poses a difficult choice for these countries. If they FATF on the basis of a list of its own criteria; and the choose not to enact laws deemed adequate, they countries also attracted the attention of the OECD could be shut off from participation in this growing for tax practices deemed harmful. market. In the absence of such laws and given the While the regulatory objectives are legitimate, weakness of local legal systems, it might be difficult several concerns have been raised about these initia- for private firms in developing countries to emulate tives. First, most developing countries have not par- United States firms like Microsoft and credibly com- ticipated in the development of the standards that mit to meet the required high standards. are being applied. Second, the standards are not al- If they do enact stringent laws, it is unlikely that ways applied uniformly. For example, the FATF ap- they could be made specific to trade with particular plies the FATF 40 Criteria when conducting mutual jurisdictions, and so the result could be an economy- evaluations of its members, but uses a different stan- wide increase in the costs of doing business. For in- dard, the FATF 25 Criteria, to assess jurisdictions stance, if private sector estimates generated in the that are not FATF members. Third, in some cases the United States are to be believed, information sharing assessment processes are not transparent. For exam- saves the customers of 90 financial institutions ple, the FSF does not specify how a country classified (accounting for 30 percent of industry revenues), in a low category can improve standards and gradu- $17 billion a year ($195 per average customer ate to a higher category. And FATF deliberations de- household) and 320 million hours annually (4 hours termining "non-cooperative jurisdictions" are held in per average customer household) (Glassman, 2000). closed sessions. Finally, the evaluation processes are 74 TRADE IN SERVICES: USING OPENNESS TO GROW Box 3.2 (continued) in some instances not voluntary and involve a "name with the aim to help them address any underlying and shame" approach to induce compliance. weaknesses. Key in this is the Bank-Fund Compre- These issues have provoked continuing discus- hensive Financial Sector Assessment Programs and sions in the international financial institutions and the recent IMF-led program of voluntary off-shore fi- other fora, but much work needs to be done before nancial center assessments. Several Caribbean off- international consensus can be established. The Bank shore financial centers have endorsed these initiatives. and the Fund are assisting many jurisdictions to as- sess their compliance with international standards Source: Bank staf f. (figure 3.4). Against the background of a total market for software services worth about $58 Figure 3.4 Software is cheaper to billion in the United States, $42 billion in Eu- develop in India rope, and $10 billion in Japan, cost savings Cost per line of code (dollars) could well be substantial.9 Other gains from $25 trade liberalization include a more competitive 0 Development $22 market structure for software services, increased $20Support choice (because countries may develop a special $18 expertise for certain development or support services), and greater diffusion of knowledge. $15 The movement of service-supplying person- $10 $10 nel remains a crucial means of delivery. Even $10 though the share of on-shore services in total Indian software exports has been in continu- $5 $s ous decline (in 1988, the percentage of on-site development was almost as high as 90 per- $60 $6o 10 $10 cent), about 60 percent of Indian exports are India Italy Ireland United Germany still supplied through the temporary movement States of programmers to the client's site overseas.10 Source: Adapted from Rubin (1999). Barriers to mode four deprive both home and host country of benefits Many more developing countries could "ex- temporary movement should create fewer so- port" at least the significant labor component cial and political problems than immigration. of services such as construction, distribution, Today, many different barriers constrain the environmental, and transport with greater lib- movement of individuals. The most obvious eralization in the movement of individuals barriers are explicit quotas or economic needs (mode four). If the movement is temporary, tests-for example, requirements that employ- then we can be fairly confident that both the ers take timely and significant steps to recruit host and home country will gain. For export- and retain sufficient national workers in the ing countries, it is clear that both the financial specialty occupation and that no worker has and knowledge benefits would be greatest if been laid off for a certain period preceding and service suppliers return home after a certain following the filing of any work permit or visa period abroad.11 For importing countries, such application.12 Then the many formalities (for 75 GLOBAL ECONOMIC PROSPECTS example, to obtain a visa) make red tape re- extend coverage abroad, but only for limited pe- lated to FDI seem trivial by comparison. The riods (two or three months). This constraint is entry of foreigners can be impeded by non- significant because it tends to deter some elderly recognition of their professional qualifications, persons from traveling or retiring abroad. Those burdensome licensing requirements, or by the who do retire abroad are often forced to return imposition of discriminatory standards on them. home to obtain affordable medical care. If indi- The requirement of registration with, or mem- vidual concerns about the quality of care re- bership of, professional organizations can also ceived abroad are addressed, then the potential constitute an obstacle for a person wishing to impact of permitting portability could be sub- provide the service on a temporary basis. stantial. If only 3 percent of the 100 million el- derly persons living in OECD countries retired Health services could be an area of to developing countries, they could bring with comparative advantage- them possibly $30 to $50 billion annually in Health services are another area in which de- personal consumption and $10 to $15 billion veloping countries could become major ex- in medical expenditures (UNCTAD and WHO porters, either by attracting foreign patients to 1998). domestic hospitals and doctors, or by tem- porarily sending their health personnel abroad. In Cuba, the government's strategy is to convert Service reforms can promote Cuba into a world medical power. SERVIMED, efficiency and growth a trading company created by the government, iberalization of trade in services, accom- prepares health and tourism packages. During L panied by the reform of complementary 1995-96 25,000 patients and 1,500 students policies, can lead to sectoral and economy- went to Cuba for treatment and training re- wide improvements in performance. spectively, and income earned from sales of health services to foreigners was $25 million. At the sectoral level- Cost savings for patients and health insurers Removing barriers to trade in services in a par- can be significant. For example, the cost of ticular sector is likely to lead to lower prices, coronary bypass surgery could be as low as improved quality, and greater variety. As in the 70,000 to 100,000 Indian rupees in India, about case of trade in goods, restrictions on trade 5 percent of the cost in developed countries. reduce welfare because they create a wedge Similarly, the cost of a liver transplant in India is between domestic and foreign prices, leading to one-tenth of that in the United States (UNCTAD a loss to consumers that is greater than the and WHO 1998). increase in producers' surplus and government revenue.13 Several empirical sectoral studies -but will require greater portability of support this contention.14 Because many ser- insurance vices are inputs into production, the inefficient A major barrier to consumption abroad (mode supply of such services acts as a tax on produc- two) of medical services is the lack of portability tion and prevents the realization of significant of health insurance. For example, U.S. federal or gains in productivity. As countries reduce tariffs state government reimbursement of medical ex- and other barriers to trade, effective rates of penses is limited to licensed, certified facilities in protection for manufacturing industries may be- the United States or in a specific U.S. state. The come negative if they continue to be confronted lack of long-term portability of health coverage with input prices that are higher than they for retirees from OECD countries is also one of would be if services markets were competitive.15 the major constraints to trade. In the United A major benefit of liberalization is likely to States for instance, Medicare covers virtually no be access to a wider variety of services whose services delivered abroad. Other nations may production is subject to economies of scale. 76 TRADE IN SERVICES: USING OPENNESS TO GROW Consumers derive not only a direct benefit from services sectors, ranging from telecommunica- diversity in services such as restaurants and en- tions to professional services, are maintained tertainment, but also an indirect benefit because not only against foreign suppliers but also a wider variety of more specialized producer against new domestic suppliers. Full liberal- services, such as telecommunications and fi- ization can, therefore, lead to enhanced com- nance, can lower the costs of both goods and petition from both domestic and foreign sup- services production (Ethier 1982; Copeland pliers. Greater foreign factor participation and 2001). In such circumstances, smaller markets increased competition together imply a larger can be shown to have a strong interest in liber- scale of activity, and hence greater scope for alizing trade in producer services, since this can generating the special growth-enhancing ef- offset some of the incentives that firms have to fects.18 Even without scale effects, the import locate in larger markets (Markusen 1989).16 of foreign factors that characterizes services sector liberalization could still have positive -and economywide- effects because they are likely to bring tech- Estimates of benefits vary for individual coun- nology with them.19 If greater technology tries-from under 1 percent to over 50 percent transfer accompanies services liberalization- of gross domestic product (GDP)--depending either embodied in foreign direct investment on the initial levels of protection and the as- or disembodied-the growth effect will be sumed reduction in barriers.17 In simulations of stronger.20 global service trade liberalization, developed Econometric evidence-relatively strong for countries gain more in absolute terms-which the financial sector and less strong but nev- is not surprising given the relative size of their ertheless statistically significant for the tele- economies-but developing countries also see communications sector-suggests that openness significant increases in their GDP. One model in services influences long-run growth perfor- predicts gains of between 1.6 percent of GDP mance (figures 3.5 and 3.6). After controlling (for India) to 4.2 percent of GDP (for Thailand) for other determinants of growth, countries that if tariff-equivalents of protection were cut by fully liberalized the financial services sector one-third in all countries (Chadha and others grew, on average, about 1 percentage point 2000). The gains from liberalizing services may faster than other countries. An even greater im- be substantially greater than those from liberal- petus on growth was found to come from fully izing trade in goods (box 3.3), because current liberalizing both the telecommunications and levels of protection are higher and because lib- the financial services sectors. Estimates suggest eralization would also create spillover benefits that countries that fully liberalized both sectors from the required movement of capital and grew, on average, about 1.5 percentage points labor. For instance, one model finds that the faster than other countries. While these esti- welfare gains from a 50 percent cut in services mates indicate that there are substantial gains sector protection would be five times larger from liberalizing key services sectors, it would than those from nonservices sector trade liber- be wrong to infer that these gains can be real- alization (Robinson and others 1999). These ized by a mechanical opening up of services results are particularly striking because they are markets. derived from models that do not fully allow for the temporary movement of individual service A flawed reform program can undermine suppliers-potentially a major source of gain, the benefits of liberalization If privatization of state monopolies to private -with accelerator effects on growth owners (sometimes foreigners) is conducted Certain services industries clearly possess without concern for creating conditions of growth-generating characteristics (see box 3.1). competition, the result may be merely transfers Furthermore, barriers to entry in a number of of monopoly rents to private owners. Similarly, 77 GLOBAL ECONOMIC PROSPECTS Box 3.3 Welfare g4-a"ns from servIce hiberalLi:ation: The case of Tunisia T he implications of services liberalization for the Tunisian economy have been analyzed by Konan Percentage change in GDP resulting from and Maskus (2000) using a computable general equi- liberalization of selected service sectors librium model. Using actual data as the foundation, they analyze the effect of liberalizing six service sec- tors: communications, construction, transportation, 2.5 business and insurance, distribution, and finance. The Tunisian economy is relatively closed, and also 2.0 faces constraints on its exports through the move- ment of individuals. The model is developed so as to 1. consider three different modes of liberalization: "im- port" liberalization of cross-border trade and the 1.0 right of establishment by foreign investors, as well as increased "exports" through cross-border movement 0.5-- of natural persons. The main finding is that services liberalization could provide significant gains to Tunisia, with wel- 6 -, 61 fare gains equivalent to 7 percent of GDP. These are 0 twice as large as the gains the model predicts for Tunisia from its preferential agreement with the EU. The largest benefits come from the liberalization of Source: Konan and Maskus 2000. foreign investment in financial services, communica- tions, and transportation. Liberalization vitalizes the economy by eliminating inefficiency through in- creased international competition. Services are avail- able not only at lower prices but also in greater vari- an 6 ecent ausent fhe ode prec eties through an increase in the number of firms that as a ose the Tunsia- free t e would operate in Tunisia. More efficient financial, communications, and transportation sectors are also labor market is a consequence of the fact that ser- likely to attract foreign firms to other industries in vices liberalization induces foreign invement, so Tunisia. As more and more foreign firms start to op- that workers simply change employers within the erate in Tunisia, the number of varieties of goods same sector. Finally, if Tunisia were to obtain a and services made available to consumers and pro- 20 percent increase in overseas p-rmits for its guest ducers also increases, which further improves wel- workers in foreign markers, then there would be an fare. The possible cost in terms of restructuring the additional gain in welfare equivalent to 0.4 percent economy turns out to be small. For example, it is predicted that a mere 3 percent of the workforce would have to change sectors-a much lower figure Sotrce Konan and Maskus 2000. if increased entry into financial sectors is not in place, liberalization need not improve access accompanied by adequate prudential supervi- to essential services for the poor. Managing re- sion and full competition, insider-lending and forms of services markets therefore requires in- poor investment decisions may result. Also, if tegrating trade opening with a careful combi- policies to ensure universal service are not put nation of competition and regulation. 78 TRADE IN SERVICES: USING OPENNESS TO GROW Figure 3.5 Services liberalization indices: telecoms & financial services South Asia (3) East Asia and Pacific (5) Financial Sub-Saharan Africa/ Services Middle East and North Africa (17) Eastern Europe and Central Asia (3) Latin America and Caribbean (18) High Income (26) Sub-Saharan Africa/ Middle East and North Africa (42) Eastern Europe and Central Asia (3) South Asia (5) East Asia and Pacific (8) Telecoms Latin America and Caribbean (21) - High Income (21) 0.0 2.0 4.0 6.0 8.0 10.0 Less restrictive Note: The openness index for telecommunications captures the degree of competition, restrictions on ownership and the existence of an independent regulator (needed to enable competitive entry), and draws on an ITU-World Bank database for 1998. The index for financial services captures the restrictions on new entry, foreign ownership and capital mobility, and draws primarily upon commitments made by countries under the GATS, which are known to reflect closely actual policy, and data in the IMF's Annual Report on Exchange Arrangements and Exchange Restrictions. Source: Mattoo, Rathindran, and Subramanian 2001. South Africa's experience with liberalizing Figure 3.6 Greater liberalization telecommunications services is instructive. The in services is associated with more government recognized the need for a more rapid growth efficient supply of services. It decided to sell a Growth rate (adjusted for other factors) 30 percent equity stake of the public incum- .059 bent, Telkom, to a strategic investor and to A Linear prediction D grant the newly privatized entity a five-year IS monopoly period for fixed-line telephone ser- MYS Nm wD vices. It was hoped that market exclusivity TM E would facilitate rapid infrastructure rollout to ,M previously underserviced areas, but the pro- Cw DIE gram has had mixed results. Even though net- work growth picked up, Telkom did not meet its rollout obligations and sought to renegoti- ate the targets specified in its monopoly li- .024 aAGO cense. The cost of the fixed-line monopoly Composite services liberalization 8.5 was also reflected in Telkom's rising price-cost margin, with gains in productivity leading to Source: Mattoo, Rathindran, and Subramanian 2001. higher margins rather than lower prices (Hodge 1999). Finally, despite some improvement, la- 79 GLOBAL ECONOMIC PROSPECTS bor productivity was only a quarter that of lead- same time, financial regulation and supervision ing international operators, with the lack of were fragmented with responsibilities spread competition in the domestic market identified unclearly between the Bank of Korea and sev- as a major contributing factor. Continued re- eral parts of the Ministry of Finance. In addi- strictions on domestic and foreign entry ap- tion, Korea had a restrictive regime in terms of pear to have prevented the realization of the foreign bank entry. Until the 1997 crisis, the full benefits of competitive markets. Korean banking system was virtually closed to In addition to competition, the institutional foreign banks, in contrast to some other East and regulatory framework plays a critical role. Asian economies, such as Hong Kong (China), For example, in the 1990s financial reforms which was almost completely open for all fi- were introduced in many African countries, nancial services. This restrictive regime impeded but have been less successful than expected the development of the local institutions, and (World Bank 2001a). Some of the reasons for may have contributed to the large capital out- the disappointing results are directly related flows as foreign creditors refused to rollover to the financial system, while others pertain their loans. to the general economic environment. The re- structuring of state-owned banks was not suf- Liberalization could increase prices of ficient to change the behavior of the financial some services for the poor- institutions. Public authorities still pressured Opening up essential services to foreign or do- these institutions to lend money to loss-mak- mestic competition could have an adverse ef- ing public enterprises. Liberalization failed to fect on the poor-which is often cited as a rea- trigger competition in the banking sector and son for the persistence of public monopolies. governments were generally reluctant to close However, a more efficient solution is to have down distressed state banks. Furthermore, lib- regulations with a social purpose. eralization of interest rates in a setting charac- If a country is a relatively inefficient pro- terized by uncontrolled fiscal deficits had a per- ducer of a service, liberalization and the resul- nicious effect on domestic public debt, which in tant foreign competition are likely to lead to a turn led to larger deficits. Finally, and crucially, decline in domestic prices and improvement in there was a lack of adequate regulation and su- quality. But there is a twist. Frequently, the pervision mechanisms to monitor the function- prices before liberalization are not determined ing of the financial system. by the market but set administratively, and are The collapse of the Republic of Korea's econ- kept artificially low for certain categories of omy in 1997 also reveals the precariousness of end-users or types of services products. Thus financial liberalization in an imperfect policy rural borrowers may pay lower interest rates environment. Korea did liberalize its financial than urban borrowers, and prices of local tele- markets substantially, but it encouraged the phone calls and public transport may be kept development of a highly fragile financial struc- lower than the cost of provision.22 This struc- ture.21 By liberalizing short-term foreign bor- ture of prices is often sustained through cross- rowing, the Korean authorities made it possi- subsidization within public monopolies, or ble for the larger and better-known banks and through government financial support. conglomerates (chaebols) to assume heavy in- Liberalization threatens these arrangements. debtedness in short-term foreign currency debt. Elimination of restrictions on entry imply an Meanwhile, the second tier of large chaebols end to cross-subsidization, because it is no greatly increased their short-term indebtedness longer possible for firms to make extranormal in the domestic financial markets (funded indi- profits in certain market segments. New en- rectly through foreign borrowing of the banks). trants may focus on the most profitable mar- The funds borrowed were being invested in the ket segments ("cream-skimming"), such as over-expansion of productive capacity. At the urban areas, where network costs are lower 80 TRADE IN SERVICES: USING OPENNESS TO GROW and incomes higher. And privatization could have frequently employed surplus labor. For mean the end of government support. The re- example, Alexander and Estache (1999) find sult is that even though the sector becomes that the privatization of electricity distribution more efficient and average prices decline, the in Argentina led to a 40 percent reduction in prices for certain end-users may actually in- the workforce after privatization. crease or availability decline, or both. But there is also evidence that pessimism The evidence on the relationship between may not always be justified. For example, a competitive market structures and wider ac- number of developing countries have managed cess to services is mixed. In some cases, a pos- to maintain or even increase employment in itive relationship has been observed in services their liberalized telecommunications sectors. such as basic telecommunications, especially Since many developing countries have low tele- in countries where initial conditions are fee- densities (in the vicinity of five lines per 100 ble, as exemplified by a low teledensity or ser- people), roughly 70 percent of telecom invest- vice rationing (long waiting lists for obtaining ment in developing countries is directed to- connections). However, there is also evidence ward building wire line and mobile networks, that financial services liberalization in some which are labor intensive and hence help main- countries has had an adverse affect on access tain or raise employment levels. Petrazzini and to credit for rural areas and the poor.23 These Lovelock (1996) find in a study of 26 Latin point to the need to create mechanisms to en- American and Asian economies that telecom sure that the poor have adequate access to ser- markets with competition were the only ones vices in liberalized markets. that consistently increased employment levels, while two-thirds of the countries with monop- -and entail adjustment costs olies saw considerable declines in their telecom Different modes of supply have different ef- workforce.25 Nonetheless, reform programs will fects on factor markets. Cross-border trade generally require complementary policies to miti- and consumption abroad resemble goods trade gate any social and economic costs of adjust- in their implications. The impact of the move- ment in factor markets. ment of factors depends critically on whether the factors are substitutes or complements for domestic factor services. Given the structure of Domestic pohcy: emphasizing factor prices in poor countries, we would typi- competition and reguRation cally expect liberalization to lead to an inflow of capital and skilled workers. Such inflows Increasing competition is the first order would tend to be to the advantage of the un- of business skilled poor, increasing their employment op- Many developing countries have moved away portunities and wages.24 Interestingly, it has from public monopolies in sectors such as com- been shown that even when foreigners com- munications, financial, and transport services, pete with local skilled workers in a services sec- but are still reluctant to allow unrestricted new tor, the productivity boost to the sector from entry. Privatization does not axiomatically mean allowing foreigners access could lead to an in- greater competition. Restrictions on foreign crease in the demand for domestic skilled presence assume particular significance in the workers-the scale effect could outweigh the case of services where cross-border delivery is substitution effect (Markusen, Rutherford, and not possible, because consumer prices then de- Tarr 2000). Given these predictions, why are pend completely on the domestic market struc- workers in developing countries sometimes ture. Several studies have concluded that larger skeptical about the benefits of liberalization? welfare gains arise from an increase in competi- One concern is the possible reduction in em- tion than from a simple change in ownership ployment in formerly public monopolies that from public to private hands (Armstrong and 81 GLOBAL ECONOMIC PROSPECTS others 1994). Foreign investment clearly brings fixed costs are significant, countries in Latin benefits even in situations where it does not lead America that granted monopoly privileges to to enhanced competition. Foreign equity may telecom operators of six to ten years to the pri- relax a capital constraint, can help ensure that vatized state enterprises saw connections grow weak domestic firms are bolstered (for example, at one and a half times the rate achieved under via recapitalizing financial institutions), and state monopolies, but only half the rate in serve as a vehicle for transferring technology Chile, where the government retained the right and know-how, including improved manage- to issue competing licenses at any time (Welle- ment. However, if restrictions on competition nius 1997). A recent study of countries in Asia artificially inflate the returns on investment, the found that the largest increases in mainline net returns to the host country may be negative. penetration and productivity were witnessed Are there good reasons to limit entry? In in countries where a change of ownership was some cases, technical limitations may prevent accompanied by the introduction of competi- competition-such as those imposed by the tion and the strengthening of regulation (Fink scarcity of radio spectrum needed for the pro- and others 2001). vision of mobile telecommunications services, and scarcity of space for department stores or Efficient regulation: Making airports in a city. More generally, entry restric- competition work tions might be justified by the existence of sig- Regulation in services, as in goods, arises es- nificant economies of scale. For example, if sentially from market failure, which is attrib- there are substantial fixed costs of networks, utable to the problems of natural monopoly competitive entry could lead to inefficient net- and inadequate consumer information, and work duplication.26 However, entry restrictions from considerations of equity and protecting are increasingly hard to defend in principle, in the poor. the face of technological change and in the face The existence of natural monopoly or oli- of mounting evidence that competition works. gopoly is a feature of the so-called locational First of all, entry restrictions change the na- services. Such services require specialized dis- ture of interaction between incumbents and tribution networks: roads and rails for land may well make collusion more likely. Second, transport, cables and satellites for communica- such restrictions dampen the impact of compe- tions, and pipes for sewerage and energy dis- tition on productive efficiency. Third, the reg- tribution (UNCTAD; and World Bank 1994). ulator is usually not better placed than the Many countries have instituted indepen- competitive process to determine the optimal dent regulators for basic telecommunications number of firms in the market, especially given services to ensure that monopolistic suppliers the difficulty of obtaining information about do not undermine market access by charging the cost structure of firms and other sources of prohibitive rates for interconnection to their regulatory failure. Furthermore, technological established networks (see box 3.4).28 A simi- advances have significantly lowered network lar approach is being taken in a variety of costs in a unisector such as telecommunica- other network services, including transport tions, and vertical separation (for example, (terminals and infrastructure), and energy ser- through network unbundling) has widened the vices (distribution networks). scope for competitive entry (Smith 1995). Regulation of the interconnection price may Therefore inefficiencies introduced by duplica- not, however, be sufficient. Small markets may tion of networks may be small compared to not be able to create conditions for effective operational inefficiencies that can result from competition in the supplies of certain telecom- a lack of competitive pressure.27 For example, munications, transport, and financial services, even in telecommunications, a sector where even if they eliminate all barriers to entry-for 82 TRADE IN SERVICES: USING OPENNESS TO GROW Box 3.4 Challenges in implementing procompetitive regulation It is now widely recognized that in basic telecom- place a significant resoUrce burden. Apart from spec- municatiors procompetitive regulation is needed to trum monitoring equipment, computers, and pro- deliver effective competition and gains from liberal- grams, there is the cost of professional assistance for ization. But the experience of different countries re- activities such as interconnection, cost estimation, and veals a range of political and economic difficulties spectrum management. For example, the total cost that are only gradually being overcome, of government in Dominica is $41 million a year, In India a conflict between the department whereas the budget of the U.S. telecom regulator (the of telecommunications (DOT) and the regulatory Federal Communications Commission) runs to $210 a agency, Telecommunications Regulatory Authority of year It is estimated that even a bare-bones regulator) India (TRA1), as it was initially constituted, hampered authority is likely to cost in the region $2 million each progress toward an efficient telecom infrastructure, year, or 5 percent of Dominica's government budgem Underlying a number of these problems was the In response to these problems, in May 2000, DOTs joint role in awarding licenses for both basic St Lucia, Dominica, Grenada, St Vincent and the and cellular services while remaining as the main Grenadines, and St Kitts and Nevis set up, with telecommunications service provider. Absent an inde- World Bank support, the Eastern Caribbean Tele- pendent regulator, empowered to rebalance tariffs, communications Authority (ECTEL), the first re- enforce fair interconnection agreements, and ensure gional telecoimunications authority in the world. rapid, equitable issuance of radio spectrum, the bene- ECTEL is in the process of developing from a legal fits of a sector opened to allow private participation entity into a functioning institution. Although the and foreign investment were significantly limited. member countries will retain their sovercign power The government announced a new telecom- over licensing and regulation, ECTEL will provide munications policy on March 26, 1999 that ad- technical expertise, advice, and support for national dressed several of these key outstanding issues. The regulations. Apart from the economies of scale in es- DOT"s policymaking and service provision functions tablishing a common regulator, there are at least were separated, and the operations arm was corpora- three other advantages. It will promote the dcvelop- tized. TRAI was reconstituted in 2000, and its dis- ment of harmonized and transparent regulation in pute resolution powers are now vested in a new the region, allow for a greater degree of indepen- quasi-judicial agency. The authority announced a new dence (and hence credibility) in regulatory advice, telephone tariffs decision that will substantially re- and enhance bargaining power in negotiations with structure telephone service prices over a three-year incumbents and potential entrants. In fact, there is period, significantly improving incentives for local evidence that the creation of ECTEL, along with network investment. The regulator has also pro- other reforms, has already prompted a decline in the grammed an agenda of activity to address several prices of telecommunication services in the region. other important regulatory matters, such as intercon- One example is that the per-minute cost of a daytime nection arrangements; a numbering plan; quality of call to the United States has fallen between 24 and service; rules of business; and customer satisfaction. 42 percent in these countries. For smaller countries, a different problem arises: the creation and operation of an efficient regulatory So-urce: DeFreitas, K:nnv, and Schware 2001; and Wnrld Bank agency involves substantial fixed costs that could staff. two related reasons. First, with services, unlike ond, changing technologies may have reduced in the case of goods, national markets are often the optimal scale of operation as well as sunk segmented from the international market due costs in these sectors, but not enough for small to the infeasibility of cross-border delivery. Sec- markets to sustain competitive market struc- 83 GLOBAL ECONOMIC PROSPECTS tures. Some form of final price regulation may, the protectionist is an issue to which we return therefore, be unavoidable. In some cases, such in the final section of this chapter. regulation can be implemented at the national level although, in practice, many developing Regulation to ensure universal service countries today lack the means to do so. In Reform programs can accommodate universal other cases, the limited enforcement capacity of service obligations by imposing this require- small states strengthens the case for multilat- ment on new entrants in a nondiscriminatory eral initiatives.29 way. Thus such obligations were part of the license conditions for new entrants into the Regulation to remedy inadequate fixed network telephony and transport in sev- consumer information eral countries. However, subsidies have often In many intermediation and knowledge-based proved more successful than direct regulation services, consumers have difficulty securing full in ensuring universal access (Estache and oth- information about the quality of service they are ers 2001).31 In 1999, Peru adopted a universal buying (UNCTAD and World Bank 1994). Con- service levy of 1 percent to finance a fund ded- sumers cannot easily assess the competence of icated to providing universal access in remote professionals such as doctors and lawyers, the areas. Funds were allocated through a com- safety of transport services, or the soundness of petitive bidding process that encouraged oper- banks and insurance companies. When such in- ators to adopt the best technology and other formation is costly to obtain and disseminate, cost-saving practices at minimum subsidy. The and consumers have similar preferences about the Chilean government adopted a similar scheme relevant attributes of the service supplier, the reg- that permitted it to leverage over $2 million in ulation of entry and operations in a sector could public funds into $40 million in private in- increase social welfare. However, the establish- vestment; this resulted in installation of tele- ment of institutions competent to regulate well is phones in 1,000 localities at about 10 percent a serious challenge, as is revealed by the difficul- of the costs of direct public provision. House- ties in the financial sector-not only in a number hold ownership of a telephone in Chile in- of developing countries but also in the United creased from 16 to 74 percent from 1988 to States, Sweden, and Finland in the 1980s and 2000, and all but 1 percent of the remaining 1990s. The fact that regulatory inadequacies can- households were provided with public access not be quickly remedied raises the issue of how to telephones. different elements of reform-particularly pru- Public subsidies may also be directed to the dential strengthening and trade and investment consumer rather than the provider (Cowhey liberalization-are best sequenced (see box 3.5). and Klimenko 1999). Governments have ex- A separate problem is that domestic regula- perimented with various forms of vouchers, tions to deal with the market failure may them- from education to energy services. This last in- selves become impediments to competition and strument has at least three advantages: first, it trade, as a result of differences across jurisdic- can be targeted more directly to those who need tions in technical standards, prudential regula- the service and cannot afford it; second, it tions, and qualification requirements in profes- avoids the distortions that arise from artificially sional, financial, and numerous other services low pricing of services to ensure access; and fi- (see box 3.2). In many cases, the impact on nally, it is an instrument that does not discrim- trade is an incidental consequence of the pur- inate in any way between providers. Of course, suit of a legitimate objective, but in some cases no single approach will fit all sectors and coun- regulation can be a particularly attractive tries, and the appropriate model to ensure ser- means of protecting domestic suppliers from vice delivery to low-income groups will depend foreign competition.30 The issue of how multi- on local circumstances, particularly regulatory lateral trade rules might sift the legitimate from capacity. 84 TRADE IN SERVICES: USING OPENNESS TO GROW Box 3.5 Financial sector liberalization: the need for policy coherence Financial reform is especially complicated. It is pressures from the market as best international prac- useful to distinguish three types of financial liber- tices and experiences were introduced. alization and the scope of each. While the two reform processes (international- ization and domestic financial deregulation) are mu- * Domestic financial liberalization allows market tually reinforcing, they are not sufficient in them- forces to work by eliminating controls on lending selves. More than in other sectors, the gains and and deposit rates and on credit allocation and, costs of financial reform depend on the regulatory more generally, by reducing the role of the state in and supervisory framework, (Barth and others the domestic financial system. 2001). Experience shows that it is vital to strengthen * Capital account liberalization removes controls the supporting institutional framework in parallel on the movement of capital in and out of a with domestic deregulation and internationalization. country and restrictions on the convertibility of In the absence of such strengthening, foreign entry currency. may entail risks. Foreign bank entry can destabilize * Internationalization of financial services eliminates local banks by taking away the lowest risk busi- discrimination in treatment between foreign and ness-including large, exporting firms-leaving local domestic financial services providers, and removes banks to venture further out on the risk frontier. barriers to the cross-border provision of financial Also several countries, especially in Africa, discov- services. ered with the failure of banks-such as BCCI and Meridien-that a foreign name did not guarantee Internationalization has raised several fears: the safety and soundness even when these foreign banks threat to the survival of local banks and financial were operating in industrial economies or had some companies; the loss of monetary autonomy; and the ownership links with reputable foreign sources. increased volatility of capital flows. Many of these Having a supportive institutional framework is concerns do not relate just to internationalization of even more obvious when it comes to capital account financial services, but also to the processes of finan- liberalization. Experiences in the past, most recently cial deregulation and capital account liberalization. in Asia, have shown that achieving the potential But the extent of benefits and costs of international- gains, and avoiding the risks, of capital account liber- ization depends, to a great extent, on how it is alization depend to a great extent on whether domes- phased in with these other two types of financial re- tic institutions and prudential authorities have devel- form, and, in particular, the strengthening of pruden- oped sufficiently to ensure that foreign finance will be tial regulation and supervision. channeled in productive directions (Eichengreen forth- Many countries that have successful experiences coming). Recent experiences also shows the potential opening up to foreign financial firms (Brazil, Chile, benefits of foreign financial institutions in stabilizing Hungary, Ireland, Poland, Portugal, Spain, and others) capital flows. Several countries with significant for- also engaged in a process of domestic deregulation eign presence, such as Argentina and Mexico, bene- and, consequently, reaped substantial gains (World fited from the access of these institutions to foreign Bank 2001b). The experience of the countries acced- capital during periods of economic presence (Dages, ing to the EU suggests that internationalization and Goldberg, and Kinney 2000). More generally, studies domestic deregulation can be mutually reinforcing. show that diversity in ownership contributes to Increased foreign entry bolstered the financial sector greater stability of credit in times of crises (Barth and framework by creating a constituency for improved others 2000a; b); and La Porta and others 2000). In regulation and supervision, better disclosure rules, and so far as foreign presence leads to a stronger regula- improvements in the legal and regulatory framework tory and supervisory framework, it contributes to for the provision of financial services. It also added to making capital account liberalization and internation- the credibility of rules. These benefits of opening up to alization mutually reinforcing. foreign entry followed from both top-down actions on the part of government, as well as from bottom-up Source: WorLd Bank staff. 85 GLOBAL ECONOMIC PROSPECTS Multilateral engagement: to ensure that the gains from liberalization Buttressing domestic reforms are not undermined by inadequacies in policy In principle, a country can liberalize its mar- choice and regulation. kets and strengthen its regulatory institu- tions unilaterally, but four types of issues cre- Using the current round of GATS ate benefits from multilateral engagement. negotiations to deliver liberalization at First, liberalization may be constrained by do- home and access to markets abroad mestic opposition from those who benefit from The General Agreement on Trade in Services protection. Second, a country cannot on its (GATS) had a deliberately symmetric struc- own improve access for its exports to foreign ture, encompassing the movement of both markets. Third, a small country may not be capital and labor for services provision. In the- able to deal adequately with anticompetitive ory, developed and developing countries could practices by foreign suppliers. Finally, a coun- indeed bargain to exploit their modal compar- try may lack the expertise and resources to de- ative advantage: improved access for capital vise and implement optimal policy, especially from developed countries being exchanged for in the area of domestic regulation. improved temporary access for individual ser- The WTO is the natural forum to pit the vice providers from developing countries. In first two elements-opposition to reform at practice, all countries have been unwilling to home and barriers to access abroad-against grant greater access for foreign individuals each other constructively through the process (except for the limited class of skilled intra- of mercantilist negotiations. But there is also corporate transferees), and a tradeoff between a need for complementary multilateral efforts modes of delivery simply has not occurred Figure 3.7 WTO members have been reluctant to make market access commitments on the movement of natural persons (Mode 4) 100 90 80 70 60 50 40 - - - - - 30 20 10 0 DC LDC AC DC LDC AC DC LDC AC DC LDC AC Mode 1 Mode 2 Mode 3 Mode 4 Note: Calculated on the basis of a sample of 37 sectors deemed representative for various services sectors. Source: See WTO Document S/C/W/99, March 2, 1999. Legend The upper part of each bar represents partial commitments, the lower part full commitments. DC = Developed countries LDC = Developing and transition economies AC = Acceding countries 86 TRADE IN SERVICES: USING OPENNESS TO GROW (figure 3.7). Moreover, even the negotiating At home, policies that are believed in are links across services sectors and between ser- most likely to succeed. Developing countries vices and goods sectors do not seem to have themselves could take greater advantage of the been particularly fruitful. And so, since gov- opportunity offered by the GATS to lend credi- ernments could not demonstrate improved ac- bility to reform by committing to maintain cur- cess to foreign markets as a payoff for domes- rent levels of openness or to greater levels of fu- tic reform, GATS commitments reflect for the ture openness. In basic telecommunications, the most part the existing levels of unilaterally de- one sector where countries have been willing to termined policy-rather than liberalization make such commitments, there is evidence that achieved through a reciprocal exchange of the commitments have facilitated reform. "concessions." 32 Developing countries have much to gain This may change with time. With severe from stronger multilateral rules on domestic reg- shortages of skilled labor in the United States ulations. Such rules can play a role in promoting and Europe and the powerful constituency of and consolidating domestic regulatory reform, high-technology companies lobbying for re- as happened to some degree in the telecommu- laxation of visa limits, the prospects for seri- nications negotiations. The rules are also needed ous intermodal tradeoffs-such as obtaining to equip developing-country exporters to ad- temporary labor movement in return for al- dress regulatory barriers in foreign markets in lowing greater commercial presence for for- the form of burdensome licensing and qualifica- eign service providers-are now greater. The tion requirements for professionals, or restric- challenge is, first, to devise mechanisms that tive standards in electronic commerce. provide credible assurance that movement is It is desirable also to remedy the current temporary rather than a stepping-stone to mi- weaknesses in the application of the MFN gration; and second, to devise negotiating for- principle in the GATS. One obvious problem mulae that credibly link Mode 4 liberalization is the explicit departure from the MFN obli- to reductions in restrictions in other areas. gation through numerous MFN exemptions listed by countries. Less visible, but potentially Strengthening GATS rules more serious, is the possibility of implicit dis- and commitments crimination through preferential recognition In line with the WTO's central concern with agreements and allocation of quotas. Rules in securing market access, it would also be nat- these areas need to be clarified and strength- ural to use the GATS to enhance the credibil- ened to protect developing countries both ity of policy at home and security of access to from discrimination in their export markets markets abroad through legally binding com- and from pressure to grant particular foreign mitments; to ensure that domestic regulations suppliers privileged access to their markets- support trade liberalization; and to prevent dis- as, for instance, is reported to be happening in crimination between trading partners by ensur- the Chinese insurance market. ing effective application of the most-favored nation (MFN) principle.33 Dealing with anticompetitive practices First, the GATS could help secure access to Anticompetitive practices that fall outside the markets that are already open. Trade in elec- jurisdiction of national competition laws may tronically delivered products, in which more be important in sectors such as maritime, air and more developing countries are beginning transport, and communication services. The to participate, must continue to remain free current GATS provision in this area provides of explicit barriers-should such barriers ever only for information exchange and consulta- become feasible. It would be far more effective tion. Strengthened multilateral rules are needed to widen and deepen commitments under the to reassure small countries with weak enforce- GATS on cross-border trade (see box 3.6). ment capacity that the gains from liberaliza- 87 GLOBAL ECONOMIC PROSPECTS Box 3.6 Ensurmg barrier-free trade in electromically delivered products rade in electronically delivered prod- lucts, in which more and more develop- Commitments on cross-border supply in selected ing countries are beginning to participate, services sectors and which offers an increasingly viable alternative to the movement of individuals, Data DC 26 is today largely free of explicit barriers. The processing serviceS main concern should be preventing the in- teone services 2o as troduction of new barriers if they become - - technically feasible. What is the best route O anlie ntoival DC 26 to preventing the imposition of explicit re- Audiovisual oc 4 strictions, such as tariffs and quotas? (The services LOC 40 issue of regulatory barriers is discussed in Retailing DC 25 box 3.2.) services m- __ WTO Members have so far focused on Adult DC 18 prohibiting the imposition of customs du- education 1 ties on electronically delivered products. It Non-life DC 26 is ironic that considerable negotiation en- insurane LOC 48 ergy has been invested in prohibiting the Ac eptan D 5 rJ economically superior (and probably not L 55 feasible) instrument of protection whereas Lnd ofyp little attention has been devoted to inferior Trading in DC 26 [ (and possibly more feasible) instruments securities LOC 45 such as quotas and discriminatory internal Entertainment DC 17 taxation. In any case, since the bulk of such services LC 22 commerce concerns services, open trading News agency Dc 22 conditions are more effectively secured through deeper and wider commitments 0% 20% 30% 40% 50% 60% 70% 80% 90% 100% Full Partial under the GATS on cross-border trade re- *Number of countries with commitments. garding market access (which would pre- clude quantitative restrictions) and national Source: World Trade Organization. treatment (which would preclude all forms . of discriminatory taxation). There is considerable scope for an im- provement in such commitments. For instance in mitments would be for all Members to agree that no data processing, of the total WTO Membership of restrictions would be imposed on cross-border deliv- over 130, only 66 Members have made commit- ery, either of all services or of a bundle whose com- ments; and only around two-thirds of these commit- position could be negotiated. ments guarantee unrestricted market access. Many These commitments have additional value be- developing countries have not made sectoral commit- cause other GATS disciplines, fcr example, on do- ments, but the commitments of the few which have, mestic regulations, would only meaningfully kick in are frequently superior to those of developed coun- once these commitments are in place. For instance, if tries. It is particularly striking that in some of the there were excessively restrictive regulatory barriers core financial services, about a third of the develop- to cross-border trade in the core banking services in ing countries which have made commitments guaran- developed countries, it would be difficult today to tee unrestricted cross-border supply, whereas none of challenge them, since these countries have not even the 26 developed countries does so. Developing committed to provide market access and national countries have also been more forthcoming than de- treatment. veloped countries in audiovisual and entertainment services. One possible approach to improving com- Source: Matoo and Schuknecht 2000 88 TRADE IN SERVICES: USING OPENNESS TO GROW tion will not be appropriated by international sional services to procompetitive regulation in cartels. For instance, the United States and the a variety of network-based services-is critical EU could begin by ending the exemption of to realizing the benefits of services liberaliza- cooperative price-setting and related practices tion. We have also seen that devising and im- in maritime transport from the scope of their plementing such regulation is not easy, and competition law. Ending the exemption would that there are acute regulatory problems in enable a careful assessment by competition au- many developing countries. Regulatory insti- thorities of the social costs and benefits of these tutions can be costly and may require sophis- collusive arrangements. Competitive discipline ticated skills. To some extent such costs can be could also be strengthened by creating a right recovered through fees or regional coopera- for foreign consumers to challenge anticompet- tion-but external assistance could help en- itive practices by services firms in the national sure that adequate regulation is in place. Some courts of countries whose citizens own or con- technical assistance is already being provided, trol these firms-a variant of the precedent in but often on an ad hoc basis either bilaterally the WTO rules on intellectual property and gov- or through international organizations. More ernment procurement. systematic efforts-along the lines of the Inte- grated Framework for least-developed coun- Global cooperation to support tries-are needed to assess the needs of indi- liberalization vidual developing countries and to ensure that Beyond WTO negotiations multilateral sup- the most appropriate assistance is provided in port is needed at four levels: in devising sound key sectors. policy, strengthening the regulatory environ- Improvements in domestic standards and ment, enhancing developing country partici- qualifications are also needed in order to ex- pation in the development of international port services. For example, in the case of pro- standards, and ensuring access to essential ser- fessional services, low standards and dispari- vices in the poorest areas. ties in domestic training and examinations can While there is growing consensus on the become a major impediment to obtaining for- benefits of liberalization, there is less agree- eign recognition. Thus inadequacies in domes- ment on the precise route to liberalization. tic regulation can legitimize external barriers Certain issues have prompted differing strate- to trade. At the same time, developing coun- gies. Should all barriers to entry be eliminated tries need to participate more actively in the in sectors with significant economies of scale? development of international regulations and How far should trade and investment liberal- standards, especially in new areas such as elec- ization be conditioned on strengthened pruden- tronic commerce. Otherwise, standards could tial regulation? Developing countries in partic- evolve to reflect the concerns only of devel- ular could benefit from the experience of other oped countries and impede the participation countries on these issues-but the experiences of developing countries in services trade. with electricity in California and rail transport There will remain certain poor countries, or in Britain suggest that there is scope for learn- certain regions within poor countries, where ing in all countries. More work is needed at the improvements in services policy and regulation national and international levels to take stock will not be sufficient to ensure access to es- of individual and cross-country experience to sential services. The criterion for determining identify the areas where there are clear pre- whether assistance is needed could be the ab- scriptions for policy and those where there is a sence of private sector provision despite com- need for further research, and therefore for hu- prehensive policy reform. International as- mility in policy advice and formulation. sistance effectiveness could be maximized by Sound domestic regulation-ranging from allocating it in a manner similar to that used prudential regulation in financial and profes- domestically by countries such as Chile and 89 GLOBAL ECONOMIC PROSPECTS Peru to achieve universal service. For instance, majority-owned U.S. affiliates of foreign companies. A once a country (or a region within a country) comparison of the balance of payments and foreign af- filiates transactions reveals in broad terms the relative importance of sales through cross-border delivery and those provided by certain countries to bridge commercial presence. the digital divide-could be pooled and allo- 5. It must be borne in mind, though, that the rela- cated through international competitive tenders tive importance of trade by different modes in a par- to the firm that offers to provide the necessary ticular sector reflects the choices of economic agents infrastructure at least cost. Providing inter- given the constraints of both technological feasibility national assistance in meeting the costs of the and policy restrictions. could increase the 6. The FDI data are extremely thin, with data miss- ing for many countries and only available for three SSA benefits of, and facilitate, liberalization by en- countries. suring that the needs of the poor would be met. 7. See the National Association of Software and Service Companies (NASSCOM) Web site . These exports consist mainly of stan- Notes dardized coding and testing services. 8. This report was prepared by McKinsey and 1. There are, however, exceptions to each of these Company for NASSCOM. characteristics of services: a software program on a 9. These figures were computed from WTO 1998, diskette or an architect's design on paper are both tan- table 3. Data refer to 1997. gible and storable, many artistic performances are vis- 10. See http://www.nasscom.org. The dominance of ible, and automated cash-dispensing machines make on-shore delivery is due, among other things, to a re- face-to-face contact between producers and consumers duction in information asymmetries with regard to the unnecessary. These exceptions do not, however, detract performance of programmers, the need for continuous from the usefulness of the general definition of services client-developer interaction, and demands by Indian presented above. programmers to be sent abroad, in part to improve 2. This view of trade originated in Bhagwati 1984 their skills and expose themselves to international mar- and Sampson and Snape 1985, and has been formal- kets (see Heeks 1998). ized in the General Agreement on Trade in Services 11. With permanent movement, the gains to the (GATS). host country must be weighed against the possible cost 3. The invisibility and intangibility of most services to the home country in terms of "brain drain." Over imply that when they are delivered across borders, 50 percent of all migrating physicians come from de- their passage is not recorded by a customs official. Data veloping countries. In Ethiopia, for example, during on services are therefore unreliable and volatile. Fur- 1984-94, 55.6 percent of the pathology graduates from thermore, statisticians in most countries do not keep the Addis Ababa Faculty of Medicine left the country. track of the sales of services by foreign investors or for- In Ghana, of the 65 who graduated from the Medical eign individuals who stay for longer than a year. Despite School in 1985, only 22 had remained in the country by these difficulties, it is possible to put together a rough 1997. If these countries had adequate medical staff at picture of trade in services by drawing on three com- home, these figures would be less cause for concern. plementary sources. The International Monetary Fund 12. Other barriers to movement of natural persons (IMF) balance of payments statistics are the only ser- include double taxation, wage-matching requirements vices trade statistics available on a global basis, and (wages paid to foreign workers should be the similar to capture cross-border supply, consumption abroad (as those paid to nationals in that profession, eliminating part of the category "travel"), and some temporary the cost advantage for foreigners), and local training movement of service suppliers. The more limited United requirements (to replace foreign with national labor Nations Conference on Trade and Development (UNC- within a certain time frame). TAD) data on FDI in services capture the flows through 13. This is strictly true in static models without which commercial presence is established. Finally, the market imperfections-such as monopolistic market United States is the only country that has regularly col- structures, internal and external economies of scale, or lected data on the sales of services by foreign affiliates. other distortions. The presence of imperfections opens 4. The United States is the only country that has up a plethora of possibilities in which the effects of regularly compiled data on sales of services to foreign trade policies are typically indeterminate, depending persons by majority-owned foreign affiliates of U.S. on the prior distortion. companies, and on sales of services to U.S. persons by 14. See Hoekman and Braga (1997) for a review. 90 TRADE IN SERVICES: USING OPENNESS TO GROW 15. Consider, for instance, the case of the Arab Re- tage, will also lead to increased static efficiency. This public of Egypt, where the import-weighted tariff was will strengthen the growth impact of liberalization. 31 percent in 1997, and the average manufacturing- 19. For example, there is evidence to suggest that wide effective rate of protection (ERP) was much higher foreign bank entry qualitatively changed Turkish bank- at 70 percent (Hoekman and Djankov 1997). However, ing by introducing financial and operations planning services inputs used by Egyptian industry, including con- and improving the credit evaluation and marketing sys- struction, communications, financial, business, distribu- tem (Denizer). Foreign banks also took the lead in tion, transport, and storage, were more expensive than spreading electronic banking and introduced new tech- they might have been if competition had been allowed. nologies. They raised the human capital level of Turk- If it were assumed that prices were higher by, say, 15 ish workers through domestic training programs, and percent, then the average ERP for manufacturing would by sending local recruits to training centers abroad. not only be lower, but negative for several industries 20. Coe, Helpman, and Hoffmaister (1999) and (chemicals, crude petroleum, and other extractive in- Lumenga-Neso and others (2001) are among those dustries), implying that the tariff on intermediate goods, who present empirical evidence demonstrating the im- together with the implicit tariffs on services inputs, out- pact of technology diffusion-in their case through weighed the tariff protection on the final goods. trade in goods-on total factor productivity growth. In 16. If no trade in either goods or services is possi- principle, the same should hold true for technology ble, the production of final goods is cheaper in larger that is diffused through factor flows markets, because a larger market can support a greater 21. In terms of the financial instruments employed variety of services. If trade in only goods is possible (too much reliance on short-term bills), in terms of the (for instance if services must be supplied through a financial intermediaries that were unwittingly encour- local establishment), then goods production tends to aged (lightly regulated trust subsidiaries of the banks, agglomerate in the larger country. The large country and other newly established near-bank financial inter- gains from this as productivity increases since a larger mediaries), and in terms of market infrastructure de- final goods sector can support a wider variety of inter- velopment (failure to develop the institutions of the mediate goods production. For the same reason, the long-term capital market). See, for instance, Claessens smaller country can lose from goods trade as final goods and Glaessner (1999) production shrinks. However, if there is free cross- 22. Sometimes the object is to ensure access to all border trade in services, then all countries have access consumers at the same price, irrespective of the cost of to the full range of producer services. As a result, pro- provision (for example, in transport and postal ser- ductivity in final goods production increases in all vices). At other times, the object is ensure cheaper ac- countries, and so all countries gain from trade. cess for certain categories of users (for example, in fi- 17. The last few years have seen a profusion of nancial services). national and global computable general equilibrium 23. Mosely (1999) estimates the impact of financial models seeking to estimate the economywide effects of liberalization on access to rural credit in four African services liberalization. The models suffer from weak- countries Uganda, Kenya, Malawi, and Lesotho. Using nesses, particularly the inadequate treatment of differ- sample survey data, Mosely reports that between 1992 ent modes of supply, the poor data on the levels of pro- and 1997, the percentage of sampled households with tection in different services sectors, and an inability to access to rural credit rose in Kenya and Uganda from capture the regulatory institutional detail that is a key 13.1 percent and 9.2 percent to 25 percent and 21 per- determinant of the consequences of services liberaliza- cent respectively. However, in Malawi, there was in a tion. The models are, nevertheless, useful in providing decline in the corresponding number from 12 to 8 per- a rough idea of the costs of maintaining services barri- cent. Access to credit of the poorest 10 percent (by in- ers and the corresponding welfare gains from their come) remained unchanged in Uganda and Kenya, but removal. in the case of Malawi and Lesotho declined from 18. As pointed out by Rodriguez and Rodrik (1999), 1.9 and 2 percent to .9 and 1.9 percent respectively. there are two contradictory impulses on growth ema- Mosely's study also shows that financial reform by way nating from the scale effect described above. Protecting of financial innovation in rural areas and development a sector increases its size, leading to higher growth, but of financial institutions catering to the poor has strong it also creates a wedge between domestic and foreign and significant effects on improving access to rural prices imposing a production inefficiency that rises over credit and lowering poverty. But simply privatizing time exerting a negative impact on growth. The larger state micro-finance agencies has proven to be unsuc- the size of the protected sector the larger this impact. By cessful, as illustrated by the experience of Malawi. contrast, liberalization of the services sector, in which 24. The poor are likely to be unskilled, so the ques- the country is assumed to have a comparative disadvan- tion arises as to the services sectors in which they are 91 GLOBAL ECONOMIC PROSPECTS likely to be employed. Unfortunately, data on the skill censes were still being used by the few competitive op- composition of the workforce in services sectors are erators who had managed to survive. only available for some OECD countries and that at a 29. Studies of Argentina show that all income rather aggregate level. Still a certain pattern can be in- classes gain from services reforms but that the rich (and ferred. Construction, distribution and personal services the foreign investors) gain relatively more if the regula- tend to be unskilled-labor intensive, whereas commu- tor is weak and that the poor win relatively more if the nications, financial and business services tend to be regulator is effective in ensuring that the rents of the skilled-labor intensive. sector are shared with the rest of the economy (Chisari 25. In India, the incumbent operator-the depart- and Romero 1999; and FIEL 2000). The additional ment of telecom expanded its workforce over the gains from good regulation are estimated to be about 1996-2000 period. In the face of competition, it was 0.35 percent of GDP on an annual basis. forced to improve its marketing strategy, expand its 30. As UNCTAD and World Bank (1994) argue, network and opened up thousands of public call offices "Service providers are likely to prefer the higher in- all over India. comes that result from control of entry into their occu- 26. One such possibility is the case of "nonsustain- pation, or form restrictions on competition between ability" of natural monopoly. This could arise, for in- those who are admitted to it ... whenever regulation is stance, under some natural monopoly cost conditions, judged necessary, a major concern must be to ensure when there exist no prices that will not attract entry, that regulatory powers are not captured by the exist- even though single firm supply is efficient. Armstrong ing providers of a service and used to further their and others (1994, p. 106) conclude that, "Notwith- interests." standing the logical possibility of this happening, 31. In some cases, though, where the cost of raising we are doubtful whether it provides a good case for revenue is very high, the direct regulation route may be entry restrictions in the utility industries, which are preferable. not for the most part remotely contestable and where 32. Hoekman 1996. there is little evidence that cost conditions give rise to 33. For a detailed treatment see Mattoo 2000 and nonsustainability." forthcoming. 27. Interesting evidence in this context is available from the Indian telecommunications sector. Das (2000) estimates a frontier multi-product cost function of the References incumbent fixed-line operator, covering 25 years from Alexander, I., and A. Estache. 1999. "The Role of Reg- 1969 to 1994. The study finds the existence of very ulatory Reform and Growth: Lessons from Latin high economies of both scale and scope in the technol- America." Paper presented at TIPS Annual ogy used-the parameter estimates even suggest that Forum. Johannesburg. September. telecommunications in India is a natural monopoly. Armstrong, M., S. Cowan, and J. Vickers. 1994. Reg- However, the incumbent operator displays great ineffi- ulatory Reform: Economic Analysis and British ciency, leading to a 26 percent increase of the opera- Experience. Cambridge, Mass.: MIT Press. tor's cost of production. Based on these findings, Das Barth, J. R., G. Caprio Jr. and R. Levine. 2000a. "The concludes that India's market liberalization program, Regulation and Supervision of Banks Around the started in the mid-1990s, is justified, but he argues that World: A New Database," Policy Research Work- there may be a need to regulate entry in order to reduce ing Paper #2588, Development Research Group. unnecessary duplication of common costs. Moreover, Washington, D.C.: The World Bank. with continued improvements in technology, the fixed - . 2000b. "Banking Systems Around the Globe: costs of entrants are likely to fall, reducing losses of Do Regulation and Ownership Affect Perfor- scale economies and thus increasing the costs of entry mance and Stability?" Policy Research Working restrictions. Paper #2325, Development Research Group. 28. Several countries have found it difficult to create The World Bank, Washington, D.C.. an open, competitive telecommunications sector be- - . 2001. "Bank Regulations and Supervision: cause of a weak regulatory environment. Poland opened What Works Best?" Paper presented at the "13th up its telecommunications sector to private competition Annual World Bank Conference on Development as early as 1990. There was a rush to invest, and about Economics." April. 200 licenses were awarded in the first six years of the Bhagwati, J. N. 1984. "Splintering and Disembod- newly liberalized regime. The dominant state operator, iement of Services and Developing Nations." The operating in a weak regulatory system, limited access to World Economy 7: 133-44. its network and benefited from unequal terms for rev- Chadha, R., D. Brown, A. Deardorff, and R. Stern. enue sharing, however. By 1996, only 12 of the 200 li- 2000. "Computational Analysis of the Impact on 92 TRADE IN SERVICES: USING OPENNESS TO GROW India of the Uruguay Round and the Forthcoming Ethier, W. 1982. "National and International Returns WTO Trade Negotiations." Discussion Paper No. To Scale in the Modern Theory of International 459, School of Public Policy. The University of Trade." American Economic Review June 72: Michigan, Ann Arbor. 492-506. Chisari, 0., and C. Romero. 1999. "Winners and Fink, C., A. Mattoo, and R. Rathindran. 2001. "Lib- Losers from the Privatization and Regulation of eralizing Basic Telecommunications: The Asian Utilities: Lessons from a General Equilibrium Experience." Paper presented at a workshop on Model of Argentina." The World Bank Economic "Trade, Investment and Competition Policies in Review 13 (2). the Global Economy." Hamburg Institute of In- Claessens, S., and T. Glaessner. 1999. "International- ternational Economics. January. ization of Financial Services in Asia." Paper pre- FIEL (Fundacion de Investigaciones Economicas Lati- sented at the conference "Investment Liberaliza- noamericanas). 2000. "La Regulacion de la Com- tion and Financial Reform in the Asia-Pacific petencia y de los Servicios Publicos: Teoria y ex- Region," Sydney, Australia. August. periencia Argentina Reciente." Buenos Aires. Coe, D. T., E. Helpman, and A. W. Hoffmaister. 1997. Glassman, C. A. 2000. "Customer Benefits from Cur- "North-South R&D Spillovers." The Economic rent Information Sharing by Financial Services Journal 107 (440): 134-49. Companies." A study for the Financial Services Copeland, B. R. 2001. "Benefits and Costs of Trade Roundtable. December. and Investment Liberalization in Services: Impli- Heeks, R. 1998. "The Uneven Profile of Indian Soft- cations from Trade Theory." Paper prepared for ware Exports." Working Paper No 3, Institute for the Department of Foreign Affairs and Interna- Development Policy and Management. University tional Trade, Government of Canada. of Manchester, U.K. Cowhey, P., and M. M. Klimenko. 1999. "The WTO Hodge, J. 1999. "Liberalizing Communications Ser- Agreement and Telecommunication Policy Re- vices in South Africa." Trade and Industrial Pol- forms." Draft report for the World Bank. Gradu- icy Secretariat, Johannesburg. ate School of International Relations and Pacific Hoekman, B. 1996. "Assessing the General Agreement Studies, University of California at San Diego. on Trade in Services." In The Uruguay Round March. and the Developing Countries, W. Martin, and Dages, G. B., L. Goldberg, and D. Kinney. 2000. "For- L. A. Winters, editors. Cambridge, U.K.: Cam- eign and Domestic Bank Participation in Emerg- bridge University Press. .ng M sHoekman, B., and S. Djankov. 1997. "Effective Pro- ing Markets: Lessons from Mexsco and Argen-'' . E i tection and Investment Incentives in Egypt and tina." Economic Policy Review 6(3):17-36. Jra:Ipiain fFe rd ihErp. Federal Reserve Bank of New York. September. Jordan: Implications of Free Trade with Europe." Das, N. 2000. "Technology, Efficiency and Sustain- World Development 25: 281-91. i oHoekman, B., and C. Primo Braga. 1997. "Protection ability of Competitio n the Indian Telecommu- and Trade in Services: A Survey." Open Econom- nications Sector." In formation Economics and ics Review 8: 285-308. Policy 12: 133-54. oisRve :2S38 Picy 12: 133-4. Kitchenman, W. F 1999. "U.S. Credit Reporting: Per- DeFreitas, D., C. Kenny, and R. Schware. 2001. ceived' peeisOteg rvc Cocrsg h "Caribbean Cooperation: Rise of the Regional Tower Benefirs Outweigh Privacy Concerns." The Journl ofPolic, Reulatinwan Group. Regulator." Journal of Policy, Regulation and Konan, D., and K. E. Maskus. 2000. "Service Liberaliza- Strategy for Telecommunications Information and tion in WTO 2000: A Computable General Equi- Media 3: 195-201. librium Model of Tunisia." February. Processed. Denizer, Cevdet. 2000. "Foreign Entry in Turkey's La Porta, R., E Lopez de Silanes, and A. Shleifer. 2000. Banking Sector, 1980-1997." Chapter 14 in S. "Government Ownership of Banks." Harvard In- Claessens and M. Jansen, editors, International- stitute of Economic Research Discussion Paper ization of Financial Services: Issues and Lessons Series (U.S.). 1890:1-55. March. for Developing Countries. Boston: Kluwer. Lumenga-Neso, 0., M. Olarreaga, and M. Schiff. Eichengreen, B. Forthcoming. "Capital Account Liber- 2001. "On Indirect Trade-Related Research and alization: What do the Cross-Country Studies Tell Development Spillovers." Policy Research Work- Us?" World Bank Economic Review. ing Paper No. 2580. World Bank, Washington, Estache, A., Q. Wodon, and V. Foster. 2001. "Ac- D.C. counting for Poverty in Infrastructure Reform: Markusen, J. R. 1989. "Trade in Producers Services Learning from Latin America's Experience." and in Other Specialized Intermediate Inputs." World Bank, Washington, D.C. American Economic Review 79: 85-95. 93 GLOBAL ECONOMIC PROSPECTS Markusen, J., T. F. Rutherford, and D. Tarr. 2000. Cross-National Evidence." Discussion Paper No. "Foreign Direct Investment in Services and the 2143, Center for Economic Policy Research. May. Domestic Market for Expertise." Policy Research Rubin, Howard A. 1999. Global Software Economics. Working Paper No. 2413, World Bank, Washing- Hunter College, Department of Computer Sci- ton, D.C. ence, and Rubin Systems Inc. New York. Mattoo, A. 2000. "Developing Countries in the New Sampson, G., and R. Snape. 1985. "Identifying the Is- Round of GATS Negotiations: Towards a Pro- sues in Trade in Services." World Economy 8 Active Role." World Economy 23 (4): 471-489. (June): 175-82. . Forthcoming. "Shaping Future Rules for Trade Smith, P. 1995. "End of the Line for the Local Loop in Services: Lessons from the GATS." In T. Ito, Monopoly." Public Policy for the Private Sector, and A. 0. Krueger, Trade in Services. Cambridge, Note No. 63. December. World Bank, Washing- Mass.: NBER. ton, D.C. Mattoo, A., and L. Schuknecht. 2000. "Trade Policies UNCTAD (United Nations Conference on Trade and for Electronic Commerce." World Bank Policy Development) and World Bank. 1994. "Liberaliz- Research Working Paper No. 2380. World Bank, ing International Transactions in Services, A Washington, D.C. Handbook." New York and Geneva. Mattoo, A., R. Rathindran, and A. Subramanian. UNCTAD (United Nations Conference on Trade and 2001. "Measuring Services Trade Liberalization Development) and WHO (World Health Organi- and its Impact on Economic Growth: An Illus- zation). 1998. International Trade in Health Ser- tration." World Bank Policy Research Working vice-A Development Perspective. Simonetta Paper No. 2655. World Bank, Washington, D.C. Zarilli and Colette Kinnon, editors. Geneva: Mosely, P. 1999. "Micro-Macro Linkages in Financial World Health Organization. Markets: The Impact of Financial Liberalization Wellenius, B. 1997. "Telecommunications Reform- on Access to Rural Credit in Four African Coun- How to Succeed." Public Policy for the Private tries." Journal of International Development 11: Sector, Note No. 130. October. World Bank, 367-384. Washington, D.C. Petrazzini, B. A., and P. Lovelock. 1996. "Telecommuni- World Bank. 1995. Global Economic Prospects. Wash- cations in the Region: Comparative Case Studies." ington, D.C. Paper presented at the "International Institute for - . 2001a. Can Africa Claim the 21s' Century? Communication Telecommunications Forum." Washington, D.C. Sydney, Australia. April 22-23. 2001b. Finance for Growth: Policy Choices in a Robinson, S., Z. Wang and W Martin. 1999. "Capturing Volatile World. New York: Oxford University the Implications of Services Trade Liberalization." Press. Paper presented at the Second Annual Conference WTO (World Trade Organization). 1998. "General on Global Economic Analysis, GL Avernaes Con- Agreement on Trade in Services: Results of the ference Center, Ebberup, Denmark. June 20-22. Negotiations on Financial Services." March. Rodriguez, E, and D. Rodrik. 1999. "Trade Policy and Economic Growth: A Sceptic's Guide to the 94 Transport Services: Reducing Barriers to Trade High transport costs are a barrier same time, competition restraining practices to trade- among shipping lines and port terminal opera- The costs of international transport services tors around the world pose the risk that the are a crucial determinant of a developing coun- benefits of government reforms will be cap- try's export competitiveness. Shipping costs tured by private firms. International air trans- often represent a more binding constraint to port is one of the services sectors most pro- greater participation in international trade than tected from international competition. The tariffs and other trade barriers. Across econ- current regime of bilateral air service agree- omies, a doubling of shipping costs is associ- ments largely denies access to efficient outside ated with slower annual growth of more than carriers. International airline alliances, while one-half of a percentage point. Transport costs enhancing network efficiency, can also be detri- determine the potential access to foreign mar- mental if they impede effective competition. kets, which, in turn, explains up to 70 percent of variations in countries' gross domestic prod- Policy reform can lower costs- uct (GDP) per capita. In most countries, policy can make better use of existing transport resources and signifi- -reflecting geography and income- cantly improve the efficiency of services. At the Transport costs depend on a mixture of geo- domestic level, targeted infrastructure invest- graphic and economic circumstances. Adverse ments, regional cooperation on transportation, geographic locations and low-income levels- and trade facilitation initiatives can play an the latter being associated with poor infra- important role in improving the transport structure and low traffic volumes-pose an in- competitiveness of exporters. As discussed in herent challenge for many countries' trade and chapter 3, liberalizing services policy can pro- development prospects-at least in the short duce substantial cost reductions and widen the to medium term. availability and choice of services. The prepon- derance of anticompetitive practices by trans- -but also competitive forces in port service providers also demands the de- service markets velopment of efficiency-oriented competition Public trade barriers and private commercial policies. practices hamper the provision of international maritime and air transport services. Policies to- -and multilateral policies can be ward maritime transport, such as cargo reser- supportive of domestic reforms vation and limitations on the provision of port Multilateral negotiations on transport services services, often protect inefficient service pro- under the General Agreement on Trade in Ser- viders and unduly restrain competition. At the vices (GATS) Agreement have, so far, not un- 97 GLOBAL ECONOMIC PROSPECTS leashed substantial liberalization, nor have efficiently handle small shipments, firms are countries bound existing policies to gain cred- likely to maintain higher inventory holdings at ibility in their domestic reforms. Indeed, the every stage of the production chain. The costs negotiations on maritime transport were the of financing large inventories can be significant, only post-Uruguay Round services negotia- especially in countries with high real interest tions that completely failed. International air rates. Gausch and Kogan (2001) find that in- transport services are largely outside the scope ventory holdings in the manufacturing sector of the GATS. The new round of services ne- in developing countries are two to five times gotiations offers the possibility of creating a higher than in the United States, and estimate rules-based services regime for maritime trans- that cutting inventory levels in half could reduce port, as well as an opportunity to develop a unit costs of production by over 20 percent. At framework under which a multilateral regime the wholesale and retail levels, firms depend for air transport services could be phased in. greatly on high quality transport services in dis- Moreover, the multilateral trading system can tributing products to geographically dispersed play a useful role in developing procompeti- markets. For example, seamless transport ser- tive regulatory principles for the transport sec- vices were critical to Kodak's decision to inte- tor, and in fostering international cooperation grate once-separate national warehousing oper- on competition policy matters more generally. ations in the Mercosur countries into one trade bloc-wide operation located in Brazil, thus reap- ing economies of scale in distribution.1 High transport costs Long journeys have a similar effect. They penalize exports delay payments if goods are exported on a cost, insurance, and freight (c.i.f.) basis or im- High transport costs push down profits porters may demand a time discount if goods and wages are delivered free on board (f.o.b.). If products The efficiency of transport services greatly de- are perishable (such as food) or subject to fre- termines the ability of firms to compete in for- quent changes in consumer preferences (such eign markets. For a small economy-for which as high-fashion textiles), longer journeys lead world prices of traded goods are largely to additional losses in terms of a product's given-higher costs of transportation feed into shortened lifetime in the export market. Box import and export prices. To remain competi- 4.1 illustrates the complex logistical arrange- tive, exporting firms that face higher shipping ments that ensure the timely delivery of costs must pay lower wages to workers, accept Kenyan cut flowers to European consumers. lower returns on capital, or be more produc- One recent estimate, based on comparisons be- tive. The pressure on factor prices and produc- tween air and ocean freight rates for U.S. im- tivity is even higher for industries with a high ports, puts the per day cost for shipping de- share of imported inputs. In these cases, small lays at 0.8 percent of the value of trade for differences in transport costs can easily deter- manufactured products. Only a small fraction mine whether or not export ventures are at all of these costs can be attributed to the capital profitable. In developing countries, for labor- costs for the goods during the time they are on intensive manufacturing industries such as tex- board the ship.2 Delivery time is found to have tiles, high transport costs most likely translate a more pronounced effect for imports of inter- into lower wages, directly affecting the stan- mediate products (Hummels 2000), suggesting dard of living of workers and their dependents. that the fast delivery of goods is crucial for the The cost structures of firms are equally af- maintenance of multinational vertical product fected by the quality of transport services. If ser- chains. Quality aspects of transportation are vices are unreliable and infrequent, or if a coun- thus likely to be an important factor in the lo- try lacks third party logistics providers who cation decisions of multinational companies. 98 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE Box 4.1 The Kenyan-European cut-flower supply chain K enyan exports of cut flowers to Europe have firms and clearance and import agcnts, in order to grown remarkably in recent years, increasing by ensure supply continuity and gain greater control 217 percent in value from 1992-98. The competi- over production, distribution, and sales. tiveness of Kenyan cut flower exports stems from About 40 percent of Kenyan flowers enter Euro- favorable climatic conditions, the use of modern pean wholesale markets through one of the seven farming technology and skilled manpower, and their flower auctions in the Netherlands. Dutch auctions counter-seasonality to the patterns of production in trade, on average, 15 million flowers and potted Western Europe. Although a wide range of flower plants daily, with total sales amounting to $1.9 bil- varieties are cultivated in Kenya, the industry's lion in 1998. After the flowers are collected and growth in the 1990s was primarily due to expanded checked for quality, ripeness, grading, and packing, rose production-sparked by strong consumer de- selling takes place with the help of computerized mand and relatively high prices in Europe. "auction ciocks,' which provide information on the Cut flowers are highly perishable commodities, grower, product, quality, unit of currency, and mini- having a vase and marketable life ranging from a few mum purchase required. The financial transactions days to not more than two weeks. International are settled immediately following the auction process, flower trade demands cold storage and transporta- and flowers are then distributed to the buyers, who tion facilities, efficient inland and air-freight shipping repackage and box the flowers for further air or land arrangements, and mechanisms for rapid distribution transport. in the export markets. Prior to packing, harvested Aside from the Dutch auction system, importe-s flowers are placed in solutions to maintain post har- are directly sourcing cut flowers from Kcryan grow- vest quality, then graded, bunched and placed in cold ers for European supermarkets and traditional retail- storage. Refrigerated or insulated trucks carry the ers In the United Kingdom, for example, supermar- flowers to specialized freight handlers, which consoli- kets have contractual arrangements with Kenyan date consignments from various growers, palletize exporters (via import agents) and send daily orders them, record temperatures, and load them directly to growers, which form the basis for harvesting, pro- onto commercial or charter airlines. They also facili- cessing and shipping schedules. Through fully inte- tate customs, inspections, and proper documentation, grated supply chains, products can be harvested and which serves as the basis for claims should flowers on U.K. supermarket shelves within 24 hours from arrive in Europe at exceedingly higher temperatures. harvest. The final retail price in the United Kingdom Import functions at the European end (cutting, is more than four times the farm gate price in Kenya, rehydrating, and repacking) are typically handled by with the difference between the to prices accounted independent agents, who also provide a wider array for by freight charges, fees and commissions, retail of services including consultancy and product and margin, and value-added tax. marketing in Formation. Several large Kenyan pro- duces have established forward linkages with freight Suriv: Thorn and others 2000. Shipping costs often represent a greater iff barriers. Only a few developing countries- burden than tariffs- including, among others, Bangladesh, the Arab Transport costs are important relative to other Republic of Egypt, Lesotho, Mauritius, Mon- trade barriers. Figure 4.1 compares countries' golia, Nepal, Pakistan, and Sri Lanka-are transport cost incidence for exports to the more constrained by trade taxes than by ship- United States (the share of international ship- ping costs. For the majority of Sub-Saharan ping costs in the value of trade) and their tariff African countries, the tariff incidence typically incidence (the trade-weighted ad valorem duty amounts to less than 2 percent, while the trans- actually paid). For 168 out of 216 U.S. trading port cost incidence often exceeds 10 percent. partners, transport cost barriers outweigh tar- Most striking is the example of Benin, where 99 GLOBAL ECONOMIC PROSPECTS Indeed, for some product groups, restrictions Figure 4.1 Transport costs are often implied by standards or domestic regulations higher than tariffs represent a bigger obstacle to trade than import Nominal tariff taxes. Third, it is somewhat arbitrary to look 20 only at transport services and ignore the costs 18 0 450X of other producer services critical to the supply 16 0of foreign markets. High costs of communica- 14 tions, legal assistance, or export finance, for 1 2example, represent other sources of inefficien- 12 cies that may erode exporters' competitive- ness.s Finally, transport costs-as distinct from 8 -.*9* tariffs-cannot be brought down to zero. 6 * One recent estimate finds that a doubling 2 0 * 0 0 of the ad valorem freight rate leads, on aver- 2 y etage, to a fall in aggregate import values be- 0 .tween five- and six-fold.6 These are rough cal- 0 2 4 6 8 10 12 14 16 19 20 culations, however, and the effect is likely to Transport cost vary substantially across countries and indus- Note: Data refer to 1998. Five countries (Benin, Guinea, tries. Much depends on the degree to which Solomon Island, Togo, and Western Samoa) exhibit a higher shipping costs are directly passed on to transport cost incidence greater than 20 percent and are not shown. consumer prices. Another factor is the price Source: U.S. Bureau of Census, sensitivity of final demand and the degree to which imports from one location can be sub- stituted with imports from another location, or from domestic sources. If final demand is exports faced duties equivalent to 0.6 percent highly price sensitive, and goods from differ- of total exports, but shipping costs represented ent locations are good substitutes, small changes 22.7 percent of trade. Amjadi and Yeats (1995) in shipping costs can have a substantial effect confirm that freight rates for African exports on bilateral imports.7 to the United States are considerably higher than on similar goods originating in other -and restrain trade in services- countries--contributing to the region's lacklus- Transport costs also represent a barrier to ter trade performance over the last two or trade in services. Though difficult to quantify, three decades.3 this is important for developing countries that In interpreting the relative importance of rely heavily on tourism services as a source of transport costs and tariffs, several points foreign exchange (figure 4.2). Tourists are sen- should be kept in mind. First, the freight rate sitive to travel costs, especially where close calculations, based on c.i.f/f.o.b. comparisons, substitute destinations exist. Estimates vary understates the true door-to-door shipping substantially across locations, but a doubling cost, because only the international leg of the in travel costs may reduce tourism demand as transport journey is considered. The impor- much as eight-fold.8 More than 90 percent of tance of port and inland transportation costs tourists arrive in developing countries by air, vary substantially by country and exporter lo- underscoring the importance of efficient air cation, but can take up as much as two-thirds transport services for this export industry. of the total door-to-door costs (see below). Sec- For example, air transport costs in East and ond, the U.S. tariff schedule is lower compared Southern Africa are reported to be up to ten to other countries, and exporters face other times higher than for Florida, in the United policy-induced barriers to trade besides tariffs.4 States, limiting the pool of lower- and middle- 100 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE Figure 4.2 Tourism earnings in developing countries, 1998 Tanzania Jordan Croatia l Dominican Republic Belize Dominica Fiji Jamaica Malta Barbados Seychelles Aruba Maldives East Asia and Pacific Europe and Central Asia Latin America and the Caribbean O Travel services as a percent of GDP O Travel services as a percent of total exports Middle East and North Africa of goods and services Sub-Saharan Africa South Asia 0 10 20 30 40 50 60 70 80 90 Source: World Bank Development Indicators. income tourists able to afford a holiday in duce rents earned from the exports of primary these regions.9 The price of international pas- products, lowering an economy's savings avail- senger transport also dictates the extent to able for investments. They push up import which firms can afford business trips neces- prices of capital goods, directly reducing real in- sary to maintain ties with foreign companies vestments. Second, all things being equal, coun- and to gather information about market de- tries with higher transport costs are likely to de- mand in other countries. In addition, the mo- vote a smaller share of their output to trade. bility of businesspeople is key to the formation Those countries are also less likely to attract of multinational production networks, which export-oriented foreign direct investment (FDI). have emerged as a dynamic driver of world Since trade and FDI are key channels of inter- trade over the past decades. national knowledge diffusion, higher transport costs may lead an economy to be farther re- Transport costs affect growth rates- moved from the world technology frontier and Shipping costs can affect economic growth in slow its rate of productivity growth.10 Third, several ways. First, higher transport costs re- transport costs determine a country's selection 101 GLOBAL ECONOMIC PROSPECTS of trading partners. If export markets largely goods to the domestic and foreign markets, as consist of poor, slow-growing markets and determined by shipping costs.12 This measure there are significant costs (including transporta- of market access explains up to 70 percent of tion) of switching to new, richer, and faster- variations in countries' GDP per capita in growing markets, countries may be constrained 1996 (figure 4.3). Admittedly, the study lends in their growth potential. This dilemma may be strong causative weight to transport costs, as especially severe for small landlocked countries other factors explaining income variation- far away from major economic centers.11 notably capital accumulation-are taken, Controlling for a large number of socioeco- themselves, to be determined by market access. nomic, geographic, and institutional factors, At the same time, the inclusion of characteris- Radelet and Sachs (1998) find that developing tics of physical geography and social, political, countries with lower shipping costs experi- and institutional variables does not fundamen- enced more rapid growth of manufacturing tally alter the study's result. While more re- exports relative to GDP in the period from search is necessary to verify and refine these 1965 to 1990. In addition, when exploring the findings, they support the view that a country's relationship between shipping costs and over- development prospects are greatly affected by all economic growth across economies, the their economic geography, of which shipping study concludes that a doubling of the cost of costs are an important determinant. transportation is associated with slower an- As much as transport costs explain the lo- nual growth of slightly more than one-half of cation of production across countries, they are a percentage point. -and help to explain regional variations in icomeFigure 4.3 Potential market access in incomeexplamns vaniations in income Transport costs-as opposed to tariffs faced epi vaiation c . Income per capita (log) by exporters-vary widely across trading na- tions. The availability, price, and quality of transportation services therefore have strong 10 q V4v implications for what countries produce and 0 6 0 00 with whom they trade. In a theoretical analysis, Venables and 9 co Limio (1999) find that transport costs may _ _ 0_ _ A cause the world to be divided into "zones of 0 specialization." The more transport-intensive a 8 good is, the more likely it is that it is exported o 0o 0 0 by countries that exhibit lower shipping costs to the economic center. By contrast, exceedingly 7 v high shipping costs to a major economic center 4 can lead a country to be self-sufficient in a par- 6 ticular good-despite the fact that it may not 13 15 17 19 21 23 hold a comparative advantage in its production solely based on its factor endowments. Coun- tries with higher transport costs but identical Note: Countries' potential access to the domestic and x rforeign markets are estimated by a gravity equation, factor endowments also exhibit lower real in- whereby bilateral trade flows are explained by comes, as more resources are devoted to trans- characteristics of the importing and exporting countries, portation and the gains from trade are smaller. as well as bilateral transport costs. Redding and Venables (2001) estimate the Source: Redding and Venables 2001. potential access of a country's manufacturing 102 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE equally important in affecting the location of (South Africa) is $2,500 and goes up to $4,000 exporting firms within countries. As foreign for Vienna (Austria), $6,500 for Asunci6n trade barriers are removed, firms have an in- (Paraguay), $7,800 for Yerevan (Armenia), centive to move to regions with good access to $10,000 for Bujumbura (Burundi), and $13,000 foreign markets, such as border areas or port for Kathmandu (Nepal). The geographic dis- cities-especially if exports account for a large tance from Baltimore alone cannot explain these fraction of total sales. For example, closer ties dramatic price differences. Transport costs are between the United States and Mexico caused determined by factors that can be changed in the a rapid expansion of manufacturing employ- short run by policy, and those that cannot. This ment in northern Mexico at the expense of the section concentrates on the second set of de- Mexico City manufacturing belt.13 Agglomer- terminants. Despite advances in transport tech- ation forces may create a self-reinforcing pro- nology, a large number of developing coun- cess, whereby entire industries move toward tries continue to be challenged by geography in exporting centers, causing sharp regional in- terms of being landlocked or far away from equalities in production and income. The sever- the world's economic centers. In addition, poor ity of this process depends on the efficiency of physical infrastructure and thin traffic densities, internal transport systems-as illustrated in box typically associated with low-income econ- 4.2 on China.14 omies, represent additional impediments to Transport services thus matter for trade transport competitiveness (although policy can competitiveness. Even if tariff and nontariff alter these constraints in the longer term). Thus, barriers to trade were removed, cross-country high shipping costs undeniably represent a con- evidence suggests that the penalty of high ship- straining factor in the trade and development ping costs will continue to hold down growth prospects of many developing countries. rates and income of countries with poor inter- national transport links. Furthermore, ineffi- Advances in transport technology- cient internal transport systems can sharpen Innovations in transportation have been an im- economic inequalities within countries, with portant factor in the globalization of goods hinterland regions being disconnected from in- markets observed in the late twentieth century. ternational commerce. Two questions that im- An examination of ad valorem freight rates for mediately arise in this context are why some U.S. imports, for which detailed data are avail- countries pay more for transport services than able, suggests that the share of shipping costs others, and what governments can do to im- in the value of trade in 1998 was smaller for all prove the transport competitiveness of trading major commodity groups compared to 1938, firms. and for all but two goods classes compared to 1974 (see table 4.1).15 However, declining ad valorem freight rates may also be due to changes Why some countries pay more in the composition of trade or in unit values of for transport services: traded commodities, due, for example, to im- geography and income provements in the quality of goods. Ocean, air, road, and railway shipping have International transport costs each seen a different mix of technological and vary dramatically institutional innovations, with profound impli- Transport costs vary widely across countries. cations on how traded goods are shipped from According to the price quotes of one U.S. freight one location to another.16 Ocean shipping is a forwarder, it costs $1,000 to ship a 40-foot relatively mature industry, yet there have been container from Baltimore to Dar es Salaam, the important advances in maritime transport tech- largest port city in Tanzania (figure 4.4). Yet the nology over the past decades. Specialized ships price of shipping the same container to Durban have emerged for dry bulk commodities, oil, 103 GLOBAL ECONOMIC PROSPECTS Box 4.2 Inefficient internal transport system3 contribute to the concentration of China's export industries in coastal regions A remarkable feature of China's dramatic expan. time five times longer, than they would be in Europe sion in international trade over the past two or the United States. China's railways still charge decades has been the concentration of export-oriented what is, in effect, a penalty rate for moving containers. industries in coastal regions. The four main coastal Priority on the congested rail network is still given to provinces (Guangdong, Jiangsu, Fujian, and Shangkai) low-value bulk freight (mostly col), rather than to have been the main recipients of outward-oriented for- high-value freight, such as containers. eign investment, with the remaining portion going to Surveys based on major foreign shippers, shipping either other coastal provinces or regions adjoining lines, and freight forwarders based in the United States, coastal areas. The provinces in the central core-usu- Japan, and Hong Kong (China) irdicate that China's ally referred to as lagging provinces-barely benefited transport systems, particularly inland transport, are from the incoming investment. While dispersion of well below international standards. First, respondents export-oriented units have narrowed coastal income pointed to the lack of container freight stations, yards, disparities-with the south coast regions catching up and trucks in inland regions. Second, border proce- with the hitherto affluent east coast-the export boom dures were perceived to be cumbersome and time- has exacerbated the coastal-inland gap. Thus, while consuming, due to the many certilcation requirements China's economic reforms have been successful in rais- and duplication of documents-in part, a consequence ing living standards for a considerable share of the of the lack of coordination between the different gov- population, a large number of Chinese people in ernment agencies involved in the various modes of inland provinces still live below the poverty line. transport. Third, container-tracking capability was Another contributing factor to coastal agglomera- particularly poor, with shippers oten unaware of their tion has been various inefficiencies in China's internal containers' whereabouts. Shippers attributed this to transport systems. Transport infrastructure disparities poorly trained staff, the lack of a reliable recovery sys- between the coastal and inland provinces narrowed tem, and the poor accountability system in government considerably following policies aimed at promoting agencies. Fourth, the intermodal transport system was more regionally balanced economic development found to be poorly integrated, with no streamlined since 1990. However, indications of increasing inter- procedures to support the continuous movement of provincial trade between inland regions, and between containers between the coast and inland. inland and coastal regions, suggests that it is not the Another source of inefficiencies is the dominance availability of transport infrastructure per se that have of state-owned enterprises and the lack of competition precluded inland provinces from actively participating in transport service markets. Since pricing in many of in foreign trade. Rather the inadequacies associated the intermediate transport service activities is con- with transport services are the more binding constraint trolled, the companies have little incentive for aggres- to better integrating China's hinterland economy. sively pursuing cost-cutting methods. Due to a lack of The compositional shift of exports from low- competition, intermediate service providers represent value raw materials to high-value manufactured goods the interests of transport operators. Hence value-added has made transport increasingly suitable for container- service and reliability, hallmarks of winning business ization. Though there has been significant increase in confidence in a modern economy are not practiced by the volume of container traffic in China since 1990, most participants. Investment by coreign enterprises or the increase is largely confined to coastal regions, and joint ventures between foreign and domestic enterprises associated with the oceangoing leg of travel. Container in intermediate transport services is limited in inland re- traffic in inland areas is much less, with no significant gions. Though foreign investmen is not prohibited, change in the percentage of sea-borne containers tray- there are restrictions on investors' activities. eling beyond port cities and coastal provinces. Truck rates for moving a container 500 kilometers inland are Source: Atinc 1997; Graham and Wada 2001; Naughton 2001; estimated to be about three times more, and the trip and World Bank 1996. 104 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE Figure 4.4 Shipping a container from Baltimore, Maryland, around the world: Distance is only half the costs story Bujumbura Brasilia (Brazil) Ankara (Turkey) (Burundi) Mbabane (Swaziland) $4,000 $4,000 $10,000 $12,000 Asunci6n Ashkhbad Durban Lima (Peru) (Paraguay) (Turkmenistan) (South Africa) $4,000 $6,500 $13,000 $2,500 kilometers 4,500 5,500 6,500 7,500 8,500 9,500 10,500 11,500 12 500 13,500 14,500 Dar es Salaam (Tanzania) $1,000 La Paz (Bolivia) Niamey (Niger) Beijing (China) $6,500 $7,000 $13,000 Vienna (Austria) Yerevan Kathmandu $4,000 (Armenia) (Nepal) $7,800 $13,000 Note: Shipments refer to loosely packed freight and do not include insurance costs. Source: Lim5o and Venables 1999. chemicals, automobiles, forest products, and cent.17 However, evidence from major developed- other goods. Probably the most far-reaching de- country shipping routes suggests that the real velopment in maritime transport has been the price of ocean liner shipping has not declined growth of containerized cargo shipping, which over the past decades, while tanker and tramp has allowed investments in larger and faster shipping has arguably become cheaper (Figure ships. Today, more than 60 percent of global 4.5). Unfortunately, no information is available general cargo trade moved by sea is carried in to assess the development of real ocean freight containers. On trades between industrialized rates for developing country routes in past countries the percentage is just over 80 per- decades. Table 4.1 Ad valorem freight rates for U.S. imports: 1938,1974, and 1998 (As percent of total import values) All countries Developing countries 1938 1974 1998 1938 1974 1998 Foods 9.3 9.4 7.0 12.5 8.3 8.4 Agricultural raw material 7.5 11.4 6.5 10.3 14.3 10.2 Crude materials and ores 65.2 44.5 12.0 57.3 30.4 13.9 Fuels 14,3 7.7 7.8 21.5 13.0 9.3 Chemicals 10.4 14.3 3.3 6.7 16.0 6.4 Metals 10.0 7.7 5.2 10.2 6.8 5.5 Other manufactures 10.1 10.6 4.6 5.2 7.5 4.5 Note: Ad valorem freight rates are based on comparisons between f.o.b export and c.i.f. import values, as reported by U.S. customs. They therefore do not include inland transportation costs and charges incurred at the port of exportation. Source: Yeats 1981 for 1938 and 1974; and U.S. Bureau of Census for 1998. 105 GLOBAL ECONOMIC PROSPECTS Figure 4.5 Ocean freight rates, 1970-99 200 180 Liner trade - Tramp tanker 160 Tramp dry cargo 140" 120 100 60 40 20 0 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 1999 Note: The liner trade series is based on freight rates observed in Germany seaborne trade and deflated by the German consumer price index (CPI). Freight rates are typically quoted in U.S. dollars, but shippers often apply currency adjustments to compensate for fluctuating exchange rates. using the U.S. CPI, the overall trend in prices is very similar, although freight rates would decrease over the 1970-99 period due to higher U.S. inflation. The tramp tanker and tramp dry cargo (trip charter) series are deflated by the U.S. CPL, since charter prices are typically quoted in U.S. dollars and set in highly competitive markets. Source: UNCTAD Review of Maritime Transport (various issues), based on data from the German Federal Statistical Office (for liner trade) and Lloyd's ship manager (for tanker and dry cargo). -have boosted air transport- Air passenger transport has also experienced Air transport is still a relatively young industry a dramatic real price decline, which has led to a that has gained in significance only after the sharp increase in international air travel, grow- emergence of long-distance jet airliners in the ing at an average annual rate of 5.8 percent in late 1950s and the introduction of the wide- terms of passenger-kilometers since 1980. body jet in 1967. The liberalization of air trans- port services, starting domestically with the -and have improved the quality United States in 1978, provided an additional of services impetus to the industry's growth, as airlines Due to the introduction of faster ships and the were granted greater freedom in determining growth of air transport services, the average their routes and schedules, and service competi- time of cargo delivery has fallen sharply in the tion intensified. Since 1980 airlines' freight op- past decades-from an estimated 40 days in erating revenues per ton-kilometer have fallen 1950 to 14.3 days in 1998 in the case of U.S. by 55 percent in real terms. As air shipping imports.20 prices have fallen relative to prices for ocean Managerial innovations and closer integra- transport, the share of world trade shipped by tion of transport services into production, in- air has continuously grown over the past ventory, and distribution systems have been decades-from 7 percent in 1965 to 30 percent additional drivers of change in the interna- in 1998 in terms of value for U.S. imports.18 In tional transport industry. Just-in-time delivery terms of ton-miles shipped worldwide, air cargo of intermediate inputs, for example, has al- shipping has grown by almost 10 percent annu- lowed firms to outsource certain stages of pro- ally from 1970-96, compared to only 2.6 per- duction, cut inventories, and geographically cent growth for ocean shipping.19 disperse production. Better management of the 106 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE supply chain has enabled producers of perish- is thus not surprising that landlocked countries able commodities to compete in distant con- pay, on average, more for shipping exports and sumer markets. These managerial changes imports than coastal economies. Multiple stud- have, in turn, led many transport operators to ies have documented the "penalty" of being become multidimensional providers of logis- landlocked, and estimates usually put the addi- tics services-including packing and labeling, tional cost of transportation at more than 50 freight forwarding, insurance and banking, the percent of that paid by countries with maritime processing of border formalities, tracking of ports.24 For many shipments to landlocked shipments, and other services. The growth of countries this "penalty" is likely to be higher. these services has, in part, been propelled by For example, the price quotes for container the falling cost and increasing power of com- shipments from Baltimore reveal that the cost munications and computing, as logistics pri- of shipment to Durban (South Africa) is marily involves the processing of information. $2,500, whereas the cost to Mbabane (Swazi- land) via Durban comes to $12,000-a land- Geography continues to exert its locked "penalty" of 380 percent (figure 4.4). own tyranny- Aside from longer overland distances, traffic to Despite technological advances, however, geog- and from landlocked economies often suffers raphy continues to be an important determi- from higher transaction costs due to the com- nant of international variations in transport plexities of coordinating multimodal transport costs. The distance between the origin and journeys and the crossing of multiple borders. destination points of a transport journey di- It is thus not surprising to find that land- rectly affects the variable cost of shipping in locked countries have only 30 percent of the the form of fuel, wear and tear of vehicles, and trade volume of average coastal economies; the amount of time that goods are traveling. that none of the 15 developing countries with Due to the existence of fixed costs of trans- the fastest export growth is landlocked; and portation, however, the effect of distance on that all 15 of those economies are located ei- transport costs is less than proportionate, sug- ther directly on major shipping routes or close gesting that distance matters more where the to a major developed-country market.25 The costs of packaging, documentation, port ser- study by Redding and Venables (2001) pro- vices, and other distant-invariant activities are vides additional proof of the burden of geog- small.21 Typically, a 1 percent increase in dis- raphy: access to the coast raises per capita in- tance causes trade volumes to fall by slightly come by 64 percent, while halving the distance more than 1 percent-although this large ef- to all trading partners increases per capita fect is also due to factors other than transport income by 74 percent. While these figures pro- costs.22 Countries that share a common bor- vide a pessimistic view on the trade and devel- der are found, on average, to trade signifi- opment prospects of geographically disadvan- cantly more than countries without a common taged countries, in the long run new economic border, which can in many instances be attrib- centers emerge. High-income landlocked econ- uted to more integrated transport networks omies such as Switzerland, or the state of Col- and the existence of bilateral customs agree- orado in the United States demonstrate that ments that reduce transit times. such disadvantages need not be permanent. -especially for landlocked countries Infrastructure links the hinterland to The effect of distance depends greatly, how- the world- ever, on the mode of transport. By one esti- Transport infrastructure, encompassing road, mate, an additional kilometer of overland railway, and internal waterway networks, sea- transport adds seven times more to transport ports and airports, warehousing facilities, and costs than an additional kilometer by sea.23 It supporting communications systems, is a key 107 GLOBAL ECONOMIC PROSPECTS prerequisite to efficient transport services. When structures are found to be important determi- goods originate or terminate in remote regions, nants of bilateral trade flows.27 inland shipping accounts for a substantial share of the total door-to-door transport charge (fig- -and poor countries are at a ure 4.6). If internal transportation networks disadvantage are dense, remote regions are in a better posi- Poor infrastructure conditions are often the di- tion to supply foreign markets. In countries rect result of low income levels, as the resources with well-developed infrastructures, exporters available for infrastructure investments are lim- can typically choose among alternative modes ited. Nonetheless, governments play an impor- of transport (truck, railroad, or internal water- tant role in expanding the reach and improving way) and alternative seaports and airports to the quality of existing infrastructure. Invest- ship their goods abroad. Aside from greater ments in transport infrastructure often take a flexibility, increased modal and port choice significant share of developing countries' GDP.28 directly promotes competition and limits the Governments in many countries-in part dri- potential abuse of market power by transport ven by the need to cut public expenditures- operators serving chokepoints. Based on an have increasingly turned to the private sector index that captures the densities of coun- for financing such large investments. Successful tries' road, railway, and telecommunications involvement of private investors necessitates an networks, Limio and Venables (1999) confirm attractive investment climate, transparent and that better infrastructure translates into signif- carefully designed concession contracts, and icantly lower transport costs.26 Moreover, a a credible overall policy regime.29 Yet where higher infrastructure density in transit coun- commercial risks are too high, public sector in- tries reduces transport costs to landlocked econ- vestments are still required-especially in the omies. Both own and transit country infra- poorest countries that are typically not able to attract private investment. Governments also play a crucial role in infrastructure planning. Figure 4.6 Decomposing the costs of Road, railway, and port capacities need to ac- door-to-door shipments commodate projected growth in trade. The de- U.S. dollars. sign and construction of transport networks 3,000 need to be coordinated with neighboring coun- 3 ton spection tries, which is especially important for small 2,500 - Cargo handling and transfer and landlocked economies. O Documentation & forwarding * Inland transport 2,000 Economies of scale and scope There are large economies of scale in the provi- sion of shipping services. Greater transporta- tion flows allow service providers to operate 1,000 larger vehicles and to spread fixed route costs 500 over a larger number of shipments. The capac- ity of containerships operating on the major o 6vwvmL_ East-West trading routes is several times that of Containerized carpet Tea shipment those operating on North-South routes, where shipment from (8.4 tons) from Kathmandu (Nepal) Ka mganj (India) traffic density is substantially smaller. Control- to Munich (Germany) to Liverpool (UK) ling for other determinants of liner freight Note: Costs refer to the cheapest route of shipment. prices, shipments from the port of Lagos, Nige- Insurance and bank processing charges are excluded. rla, to southern California would be 24 percent Source: Subramanian and Arnold 2001. cheaper, if traffic on this route would be the same as from the port of Hong Kong (China) 108 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE to the same region.30 Furthermore, due to rela- ments, and others-add to the overall logistical tively low trading volumes, developing coun- costs of international shipments. Uncertainty tries often face longer travel times and less fre- about the enforceability of legal documents quent services, as ocean carriers require a larger (such as bills of lading or letters of credit) in- number of stops to fill vessels.31 creases the risks faced by importers and ex- At sufficiently large traffic volumes, trans- porters as well as transport operators. One port operators can reap economies of scope by study on Brazilian ports reports the average offering services on connected routes. Through costs per container related to administration hub-and-spokes systems, maritime and ocean and customs clearance at $1,727, which could transport operators have been able to cut costs, be reduced to $320 according to international while at the same time offering transport links best-practice estimates (figure 4.7).32 In many between a larger number of locations at higher cases, transport-related transaction costs do frequencies. The overwhelming share of inter- not even show up in the final freight bill, but continental ocean trade is today delivered by take up a firm's resources that could be used hub-and-spoke systems, through major ports more productively. such as Hong Kong (China), Los Angeles, Rot- Multiple changes of transport modes dur- terdam, or Singapore. By contrast, most ocean ing the transport journey create costs in the carriers serving the routes to and from West form of frequent reloading of goods, coor- Africa still operate under so-called multiple dination problems that result in shipment de- ports of call systems. However, given current lays, and the need to contract several trans- traffic levels, commodity mix, port infrastruc- port operators instead of a single door-to-door tures, and inland transportation systems, Pils- service provider-often exacerbated by legal son (1997) finds that the adoption of a hub- provisions preventing foreign multimodal oper- and-spoke system would not systematically yield substantial cost savings. Future growth in trade as well as infrastructure improvements Figure 4.7 Potential door-to-door cost may change this calculus. Yet the implementa- savings on containerized imports in tion of a hub-and-spoke system would still re- Brazil quire the willingness of the spoke countries to US. dollars accept lower traffic volumes to the benefit of 5,000 the hub port. O Administration/customs 4,500 - Bonded warehouse 11 Inland transport 4,000 - 0 Port charges Why some countries pay more: 3,500 - policy-driven factors 3,000 - G overnment policy can inadvertently in- 2,500 - flate transport costs. Most developing 2,000 - countries enact rules that detract from using existing transport resources efficiently. These 1,00 rules drive up transport-related transaction costs and often preserve monopolies in service 500 markets. 0 Actual costs Potential costs Reducing high transaction Note: Figures are based on the port of Santos. Insurance charges are excluded. Potential costs are estimated based costs in-country- on international best practice. High costs of transport-related transactions- such as frequent reloading of goods, customs clearance, fulfillment of documentation require- 109 GLOBAL ECONOMIC PROSPECTS ators from undertaking door-to-door contracts. related transaction costs. The development of Containerization has substantially reduced the the electronic data interchange (EDI) system, reloading costs of multimodal journeys, as goods for example, has substituted the traditional are packed once at the factory's door and un- paper documentation routines for customs packed at the importer's site. Indeed, con- clearance. Through the global positioning sys- tainerization has fostered the integration of tem, firms can monitor the location of vehicles transport service providers toward multi- and better time loading and reloading. The modal operations, which internalizes transac- Internet has opened new ways of organizing tion costs resulting from modal switches. Even transport movements, creating more flexible though containerization of general cargo has and efficient transport markets with reduced taken hold in many developing country-ports, uncertainty regarding the quality of shipments. containers are less frequently used for inland Yet use of these technologies is still primarily transport (especially in Africa)-obviating one confined to developed countries and large of the main cost-saving characteristics of con- ports. Lack of communications infrastructure tainer shipping.33 The main reasons for this and the necessary skills, as well as an inade- are long inland turnaround times for contain- quate legal framework for electronic signatures, ers, risks of loss or damage to containers, and often present obstacles to the dissemination of inadequate road infrastructure unsuited to con- transport-related information technology in the tainer loads. Limitations on the cross-border developing world. provision of trucking services create bottle- necks at the border, because goods have to be -requires coordinated government action reloaded onto different carriers. Different na- There is much that governments can do to re- tional standards regarding safety requirements, duce transport-related transaction costs, usu- vehicle sizes, railway gauges, or coupling and ally under the umbrella of so-called trade facil- braking systems similarly constrain the smooth itation initiatives. Such programs can result in cross-border movements of goods. significant reductions in direct and indirect Although official customs fees are typically shipping costs in relatively short time periods. only a small portion of overall transportation They are most effective if implemented in part- costs, inefficiency in customs procedures can re- nership with the private sector (box 4.3). An- sult in congestion and long queues at the bor- other role for governments is to create an ap- der. For example, at the key border crossing- propriate legal and regulatory framework for point between India and Bangladesh as many as multimodal transport, which often represents 1,500 trucks queue up on both sides of the bor- one of the most pressing constraints to the pro- der, and waiting times vary between one and vision of efficient door-to-door services. Coop- five days.34 Inefficiencies often are the result eration on standard-setting and the conclusion of understaffing, burdensome documentation of mutual-recognition agreements with neigh- requirements, poorly defined procedures, and boring countries can facilitate the cross-border the need to obtain approval from many offi- movement of goods by trucks. While countries cials. High trade protection typically results in should remain free to adopt their preferred reg- more complex customs requirements-for ex- ulatory standards, it is important to ensure ample, through the need to obtain import li- that such standards do not unnecessarily dis- censes before goods are shipped. Corruption is criminate against foreign-service providers (see endemic in many developing country ports chapter 3). and is more widespread the more opaque are the customs procedures, and the greater the From public monopoly to discretion of customs officials. private competition Advances in information technologies have For a long time the provision of many trans- created a large scope for reducing transport- port services was the domain of public monop- 110 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE Box 4.3 Lessons from customs reforms in Mexico J ust years before joining the North American Free lines were required to pay tariffs through commercial Trade Agreement, Mexico introduced a series of banks that opened branches in customs facilities, An customs reforms as part of its ongoing trade reform important component of the modernization program in 1989. An important part of the modernization ini- was targeting enforcement efforts on mainly "high tiative was reducing customs clearance time, through risk" consignments, while allowing the cargo of usu- risk management and selective testing of cargo, in ally compliant importers with minimum or no inspec- line with similar initiatives in North America and tion. Customs uses a random system that determines Australia. whether or not goods will be inspected. This system Prior to 1989, customs procedures were highly relies on data-including the country of origin, im- centralized, a reflection of the then inward-oriented porter or exporter, type of merchandise, tariff item bias of the economy. The directorate general of cus- number, and other variables--to determine whether a toms (DGC)-part of the ministry of finance-had particular passenger or consignment is to be in- unlimited authority over customs, with little account- spected. The system is not entirely random, because ability enforced on customs officials. The import- it uses information in its database to determine the export guidelines were either not published ahead of level of risk. Upon completion of customs formalities time, or subject to frequent and arbitrary revisions, in respect of a passenger or a consignment, the sys- Adjudications of customs-related disputes were time- tem is interrogated by pressing the appropriate but- consuming. The approval of customs brokers' licenses ton. In 90 percent of the cases the green light flashes was strictly regulated, thereby providing incentives indicating that no further formalities are required. for collusion between customs officials and customs The benefits of the reforms, reduction in cus- brokers. toms transit time and attendant reductions in the As part of the reform program, DGC was di- costs of interest, storage, and transport, as well as vested of all ancillary functions and the customs ad- lower broker fees, were estimated to be about 5 per- ministration was decentralized. To ensure trans- cent of the total value of the merchandise. parency, the rights and obligations of the traders were widely published. Traders under the new guide- Source: World Bank 1997. olies, and indeed state-owned enterprises tage fully applies to the provision of transport continue to be a dominant force in many coun- services, as it does to other traded commodi- tries' transport sectors. Public monopolies were ties. By opening up domestic markets to for- often justified by natural monopoly argu- eign competition, shippers can choose among a ments, such as in the case of port operations, broader spectrum of services and opt for ser- which require large infrastructure investments. vice operators with superior technologies or Prestige arguments (for example, the desire to lower operating costs. operate a national flag airline) and security concerns (self-sufficiency in times of war) af- Maritime transport- forded a justification for limiting the participa- Due to differences in commodity type as well tion of foreign service providers in domestic as to technological improvements in the ship- transport operations. Such arguments are be- ping industry-most importantly, container- coming increasingly harder to defend. Private ization-international maritime freight trans- entry and competitive market structures have port has developed specialized branches. Liner proved to be feasible for virtually all transport shipping, meaning maritime transport of com- services and, to a large extent, have led to effi- modities by regular lines, which publish in ad- ciency gains and lower prices for consumers. vance their calls in different harbors, is dis- Moreover, the principle of comparative advan- tinct from tramp shipping, which refers to 111 GLOBAL ECONOMIC PROSPECTS transport performed irregularly, depending on being applied mostly to routes between West momentary demand. Typically, liner carriers Africa and Europe (box 4.4). Nevertheless, transport commodities with a higher degree of countries ranging from Benin to India still industrial processing using containers, while have in place reservation policies that at least noncontainerized raw materials (such as crude nominally restrict the scope for trade. and refined oil, iron ore, grain, coal, or baux- ite) tend to be carried in tramp carriers. -and the practices of ocean carriers Competition-restraining practices in liner -is affected by government policies- transport take the form of cooperative agree- Tramp shipping is generally believed to be a ments among maritime carriers on technical highly competitive market that is, as a rule, or commercial matters. Carriers' cooperative free from restrictions.35 Prices are set in spot habits are deeply rooted in the history of mar- markets based on either time charter or voyage itime transportation. The first shipping cartels, charter contracts. In contrast, liner shipping covering the routes between the United King- has traditionally been subject both to trade re- dom and Calcutta, India, date back to 1875. strictions and private cooperation. The most By joining carrier agreements, shipping compa- important policy-imposed barriers applied to nies retain their juridical independence, but international maritime transport have been consent to common practices with the other various cargo reservation schemes. These re- members regarding pricing, traffic distribution, quire that part of the cargo carried in trade or vessel capacity utilization. One of the most with other states must be transported only by common types of agreement are liner confer- flagships (ships carrying a national flag) or ence agreements, which typically provide for ships interpreted as national by other criteria. the fixing of and adherence to uniform freight Cargo reservation takes various forms. It tariff rates and conditions of service. Liner con- can be imposed unilaterally if ships flying na- ferences also employ exclusive contracts and tional flags are given the exclusive right to other loyalty-inducing instruments to deter transport a specified share of the cargo passing entry of outside shipping lines.37 Another type through the country's ports. An alternative of carrier agreement includes cooperative work- form involves cargo sharing with trading part- ing and discussion agreements, which establish ners on the basis of bilateral or international exclusive or preferential working relationships agreements. A specific form of multilateral between shipping lines, and provide a forum cargo reservation scheme is the United Nations for information sharing but do not necessarily Conference on Trade and Development (UNC- engage in unique price setting. A more recent TAD) Liner Code of Conduct, which was con- form of private cooperation is strategic alliances, ceived to encourage the development of the which aim at closely integrating vessel opera- shipping industry of developing countries by tion activities and service networks. guaranteeing domestic lines a minimum (40 It has been frequently pointed out that in percent) share of traffic, and ensuring their par- recent years the power of liner conferences ticipation in international liner conferences.36 and other cooperative arrangements has been Cargo reservation schemes have probably eroded. In the 1990s efficient outside shipping declined in significance, as more and more lines were able to gain a significant share of the countries have phased them out. In addition, market on many routes. Moreover, more liberal the increased transfer of ships to open reg- regulations affecting international shipping istries to enable the ship owners to benefit have weakened the command of rate-fixing from more efficient cost conditions has further agreements. For example, the Ocean Shipping diluted the importance of cargo sharing. The Reform Act of 1998 in the United States intro- UNCTAD Liner Code, which was never ap- duced the confidentiality of key service contract plied on a large scale, is even less visible today, terms, allowing greater scope for price com- 112 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE Box 4.4 Maritime shipping in West Africa W est Africa has lagged behind the rest of Africa mal" services from CEWAL members. Second, a and the world in terms of growth of sea-borne special agreement with the Zairian Maritime Freight trade over the last two decades. Sea-borne transport Administration granted the conference the power to increased by an average annual growth rate of 1.2 prevent any intrusion of competition on its market percent since 1990. This compares to 1.6 percent for and allowed it to monitor sea-borne trade. Third, all developing countries in Africa and just over 3 the conference employed "fighting ships" to elimi- percent for total world sea-borne trade. Between nate competition from its most direct competitor 1980 and 2000, West Africa averaged about 4 per- and potential market entrants. "Fighting ships" cent of total goods loaded in the world's seaports. were identified as those vessels that sailed at dates For goods unloaded, the region averaged about 2 close to the sailings of its principal competitor. percent of world sea-borne trade. Special freight tariffs-identical or less than those West African nations adopted the UNCTAD offered by the competing line-were established Liner Code in the 1980s and rapidly expanded their for those ships. fleets, hoping to be in a position to take full advan- These agreements and practices enabled the con- tage of the code's cargo sharing formula. However ference to maintain a high market share, which con- most of the shipping lines based in West Africa either trasts with other Euro-African trades for which the collapsed or went bankrupt. Today the market share market share of the conferences is sometimes less of West African lines is very slim in the containerized than 60 percent. After the court hearing, members of trade to and from Europe, with five national lines the CEWAL conference had to amend the terms of mustering up about 6-7 percent of total capacity of- their loyalty contracts to prevent infringement of EC fered. They run a fleet of small and generally old ves- competition rules. Moreover fines were imposed on sels and offer exclusive service between their home several members of the conference. The CEWAL case countries and Europe, consequently limiting them- demonstrates the positive spillover of competition selves to a small cargo base. Low traffic levels in law enforcement by a large trading bloc, such as the West Africa restrict the number of regional carriers EU. Since the final decision by the European Court that can be sustained and limit market entry by com. of Justice in March 2000, liner transport prices on mercially oriented carriers, routes between northern Europe and West Africa Due to low volumes, there is also concern about have reportedly fallen. abusive practices by private operators. Interesting Notwithstanding the dominance of certain liner evidence on such practices was revealed in the Associ- conferences, there is growing competition from inde- ated Central West Africa Lines (CEWAL) liner confer- pendent service providers in specific port to port ence court case, which was initiated by the European markets, including niche operators (operating in spe- Commission (EQ against three liner shipping confer- cial, well defined market segments, sometimes with ences operating on routes between continental North special equipment) and operators without vessels Seaports and West Africa. Although European Union who rely on chartering space from liner companies- (EU) regulations provide for a block exemption for so-called non-vessel-owning common carriers. These liner conferences, the abuse of a conference's domi- corporations frequently keep at arm's length from nant position still falls under the realm of the compe- the large operators and can be quite successful tition rules provided in the EC's treaty of Rome. within their particular markets. In view of West The members of the CEWAL conference were Africa's stagnant trade volumes, however, there is found to have abused their collective dominant posi- continued need to closely monitor competitive condi- tion in several ways. First, the conference established tions in this critical trade-supporting industry a system of loyalty agreements, whereby loyal ship- pers received rebates on routes between Northern Europe and Zaire, while disloyal shippers were Sources: Audige 1995; European Union 1999 and 2000; Pilsson "blacklisted" and could no longer count on "nor- 1997; and WTO 2000. 113 GLOBAL ECONOMIC PROSPECTS petition. It is also important to recognize that with regard to port management and coor- private cooperation can bring benefits to con- dination, the provision of infrastructure, and sumers of shipping services-notably due to the supply of services. For example, under the improved network coordination, which can Landlord Port concept-which is becoming generate economies of scope and a wider choice widespread worldwide-the public Port Au- of services available to shippers. thority owns the basic infrastructure-land, Yet one recent study, which examines the access, and protection assets-and leases it impact of price-fixing and cooperative work- out to private operators on a long-term con- ing agreements on liner freight rates for U.S. cession basis. Under the Tool Port concept the imports, concludes that private practices con- Port Authority owns the infrastructure, the su- tinue to exercise a significant influence on perstructure, and heavy equipment, and rents liner freight rates, and that the hypothetical it to private operators, which carry out com- breakup of carrier agreements could lead to mercial operations under licenses. The Port cost savings of as much as 20 percent (see box Authority usually retains all regulatory func- 4.5). In practice, the extent to which liner tions in the case of landlord and tool ports. In freight rates are pushed up by private anti- only a few ports in the world has port land competitive practices is likely to differ across been sold to private operators, and all public routes. Developing country routes are arguably management and regulatory functions been more prone to such practices, since low overall transferred to the private sector.39 traffic volumes limit the number of competi- tors that can be commercially sustained (see -but smaller ports are at box 4.4). a disadvantage- The feasibility of competitive provision of port Seaport services are increasingly driven by services, especially cargo handling, depends on private capital and competition- several factors. The availability of port space In performing their function as the interface poses a constraint to the number and the de- between various modes of transport, seaports gree of specialization of port terminals. Sec- provide multiple services. The management of ond, traffic levels have to be sufficiently large, ships in ports requires a mixture of services re- such that it is feasible to operate several termi- lated to berthing, including pilotage, towing, nals at full capacity. Experience has shown that and tug assistance. Cargo handling is the most the operation of more than one container ter- important service in moving goods through minal only becomes viable if port traffic ex- seaports, accounting for 70 to 90 percent of ceeds 150,000 twenty-foot equivalent units total port charges. Other services related to (TEUs) per year. Third, competition between cargo manipulation include customs clearance, ports depends on geographic factors, the den- storage, and warehousing. Specialized agents sity and quality of inland transport networks, or consignees take on the paperwork and all and overall traffic volumes in the greater port matters related to the use of port facilities by region. In practice, competitive forces are a ship. Finally, there is a series of ancillary ser- likely to lead to a cost-efficient provision of vices to crew members and ships, including services only in large seaports and regional provisioning, fueling and watering, garbage hubs. For smaller ports, ex ante competition, collecting, and repair facilities.38 in the form of auctions where private firms bid The last decades have seen profound for the right to operate a terminal, can extract changes in the organization of seaports-the potential monopoly or oligopoly rents that general trend being toward increased private service providers expect to generate. Further- sector participation and greater competition more, it is necessary to accompany private port within and between ports. A variety of owner- participation with appropriate regulation over ship and operational structures have emerged tariffs charged by service providers. Indeed, the 114 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE Box 4.5 How important are public and private barriers to trade in maritime services? A recent study by Fink and others (2001) has The econometric results show that both public attempted to quantify the absolute and relative policy and private practices exercise a significant in- importance of public and private barriers to trade in fluence on liner transport prices. Of public restric- maritime services. Using data on liner transport tions, the cargo reservation policies that proliferated charges for U.S. imports, broken down to the six- in the 1970s and 1980s seem to be largely ineffec- digit Harmonized System commodity level, the study tual, but restrictiveness in the form of mandatory estimates a model that explains port-to-port liner port services significantly raises prices. Most striking prices with their standard determinants, ranging is the even more powerful effect that private carrier from distance to containerization, as well as various agreements have in keeping prices high. The table proxies for public and private restrictions that exist below presents the estimated reductions in transport across countries and on different routes. Public prices due to policy liberalization and the hypotheti- policy restrictions include cargo reservation and the cal breakup of private carrier agreements. While port extent to which certain port services, such as pilotage liberalization would lead to an average reduction in and towing, are mandatory for incoming ships. transport prices by 8 percent and cost savings of up Private restrictions considered are price-fixing and to $850 million, the breakup of private carrier agree- cooperative working agreements on routes between ments would cause prices to decline further by 20 U.S. trading partners and selected U.S. coastal percent and there would be additional cost savings districts. of up to $2 billion on U.S. routes. Breakup of C"adaiew effec cooperaive Breakup of of the breakup of Lberazationa working price-f-ing private canirr CWmUzItv of Port services greemfents agreements areeMents totl effect 1. Average percentage price reduction 8.27 5.29 15.73 20.05 26.37 2. Projected total savings for all U.S. imports (in millions of dollars) 850.4 544.1 1618.4 2063.0 2712.5 Note: The average percentage price reductions are computed from the sample of 59 countries included in the study, while the projected total savings apply to all U.S. trading partners. Given the functional form of the underlying regression equation, the individual effects do not sum to the cumulative effects. See Fink and others 2001 for additiunal explanatory notes. While the study provides important evidence how public and private restrictions directly affect on the forces constraining competition in maritime charges for port and auxiliary services. Second, due transport, several important questions fall outside to data availability, the study only considers liner the scope of the empirical analysis. First, the overall traffic to the United States, where recent reforms restrictiveness of the port services regime is only im- have increased the scope for price competition, perfectly captured by the extent to which certain potentially reducing the role of private carrier port services are mandatory for incoming ships. The agreements. Evidence for other routes involving efficiency of cargo handling-the most important developing countries is needed to evaluate how service in bringing moving goods through ports-is public and private barriers to competition affect not considered in the analysis. Moreover, the data maritime shipping. employed only capture inefficiencies in the provision of port services to the extent that they push up liner freight rates. More research is needed to evaluate Source: Fink and others 2001. 115 GLOBAL ECONOMIC PROSPECTS creation of regulatory capacity is an important estimate this small group now accounts for element in every port reform package-not about 40 percent of the world's annual con- only to monitor and set tariffs, but also to en- tainer liftings.41 While consolidation may bring sure the safety and quality of services supplied. benefits to port users, there is the danger that With few exceptions, such as Singapore, dominant operators may abuse their market public port monopolies are typically associated power-for example, by offering exclusive con- with inefficient and expensive services; experi- tracts to shipping lines if they use their world- ence has shown that liberalization programs wide facilities. Such practices may pose the risk can, in principle, greatly improve performance that the benefits from port liberalization are to (see box 4.6). Yet achieving successful liberal- some extent captured by foreign firms. ization is a complex task-even in developed countries. To attract long-term private inves- International air transport services are tors, the overall policy regime has to be credi- heavily restricted- ble and consistent over time. At the same time, International air transport is divided into sched- governments need to ensure that efficiency uled passenger, freight, and mail services, and gains are passed on to port users, which re- chartered services that depend on momentary quires carefully designed concession contracts demand. In 1998, scheduled services repre- ex ante and appropriate regulatory mechanisms sented 87 percent of revenues, of which the ex post. Thus a country with weak institutions, overwhelming share (88 percent) came from high overall economic uncertainty, a reputation the movement of passengers.42 International for policy reversal, and limited regulatory ca- airfreight transport can be further divided into pacity arguably faces a significantly more diffi- passenger belly-hold freight and dedicated cult task in managing the liberalization process. freight services. Passenger belly-hold freight is Another frequently encountered obstacle in re- typically cheaper, because freight rates are set forms is the adjustment to the labor force in at marginal cost, whereas dedicated freight port. Due to technological progress, port oper- services need to recover the full costs of oper- ations have, over the past decades, become ating the aircraft. more capital-intensive, such that moderniza- Trade in international air transport services tion typically requires the reduction of excess is heavily restricted by governments around labor. Forming consensus with workers on the the world-more so than international mari- design of reforms, retraining programs, and time transport. Market access of foreign pas- measures to soften the social impact of labor re- senger and cargo carriers is largely determined ductions can overcome some of the resistance through a complex system of bilateral air ser- of often-powerful port unions.40 vice agreement (ASAs), which typically desig- nate the airlines allowed to operate on bilateral -and powerful operators are emerging at routes, the number and frequency of flights the global level they operate, what types of aircraft they use, Opening port services to the participation of and how much they charge.43 ASAs also deter- foreign operators can bring special benefits, as mine the traffic rights of airlines operating on multinational companies often bring technol- bilateral routes, which are defined by so-called ogy, experience, and managerial know-how. freedoms of the air. Under third and fourth Large global port operators can also offer a freedom rights, airlines are allowed to carry loyal customer base, networking possibilities, traffic between their home countries and for- and access to finance. Yet a number of ob- eign countries. Fifth freedom rights permit an servers have voiced concerns about the rising airline of one country to carry traffic between global concentration of the industry. A rela- two other countries, provided the flight origi- tively small group of port operators has estab- nates or terminates in its own country. The lished a regional or worldwide presence; by one most liberal-yet rarely granted-traffic rights 116 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE Box 4.6 Lessons from reforming Argentina's ports A s part of its overall program of macroeconomic Buenos Aires. Private investment picked up signifi- stabilization, liberalization, and public sector cantly in the second half of the 1990s, leading to a reform, in the 1990s the government of Argentina substantial expansion in capacity. Productivity has initiated a comprehensive reform of the port sector. picked up sharply, significantly reducing operational The reform was a major success, in that it greatly costs and duration of stay in ports. Combined with improved the performance of Argentina's largest sea- more intense competition between port service ports, sustaining a rapid expansion in the volume of providers, this has resulted in a reduction in overall sea-borne trade, growing more than four times from container terminal handling prices. 249,000 in 1990 to 1,070 million twenty-foot equiv- alent units (TEUs) in 2000. p ed periormance n the port of Buenn A!rw Before 1990, Argentinean ports were character- inamor 1997 ized by institutional inadequacies, including a major corruption problem, inefficient cross-subsidization, Cargo (thousand tons) 4,000 9,500 and insufficient investment in the modernization of Containers (thousand TEUs) 300 1,023 Capacity (thousand TEUs) 400 1,300 the secton Tariffs charged by the publicly operated Cranes 3 13 ports were reportedly among the highest in the Productivity (tons per employee) 800 3,100 world. Total cargo moved in the ports fell by Average container time at port (days) 2.5 1.3 10 percent between 1970 and 1989, with the port Charges per container ($TEU) 450 120 of Buenos Aires alone experiencing a 52 percent reduction in traffic. Despite these impressive achievements, unre- The overall reform program consisted of a com- solved issues from the first wave of port reforms as bination of devolution of most port responsibilities well as changes in the competitive environment in to the provinces, private sector participation, and the sector, although not pressing, demand solutions promotion of service competition. Provinces were in the long run. While intraport competition is given the freedom to operate, concession, or close working effectively, the likelihood of future mergers ports, with the exception of large ports, for which between terminal operators at the port of Buenos the creation of independent autonomous companies Aires raises the risk of collusion. Improved monitor- was foreseen. In the case of the port of Puerto Nuevo ing and benchmarking mechanisms, as well as the (Buenos Aires), six terminals were competitively con- fine-tuning of price regulations, may be necessary to cessioned to the private sector, with a payment of a ensure that services continue to be provided on a leasing fee to the government for use of infrastruc- cost-efficient basis. Inefficient customs operations ture assets-following the landlord port model. To pose a key constraint toward further productivity improve the contestability of port operations, the gains in the sector and represent a priority for future government also established free entry into the sector reform. Finally, some aspects of Argentina's port by allowing any operator to build, manage, and op- policy, such as restrictions on the circulation of con- erate a port for public or private use. A new regula- tainers, are reported to restrain intermodal integra- tory agency (Autoridad Portuaria Nacional) was cre- tion. Addressing this issue in the context of the wider ated under the ministry of the economy. Finally, the policy framework on multimodal transport would restructuring process included a major labor reform contribute to a better performance of the transport that eliminated restrictive work regulations and soft- system nationwide. ened the social impact of labor reductions. The main economic effect of the overall reforms was to transform Argentinean ports from the most expensive ones in Latin America into the cheapest Source: Trujillo and Nombela 1999, and Trujillo and Estache ones-as illustrated in the table below for the port of 2001. 117 GLOBAL ECONOMIC PROSPECTS are seventh freedom rights, which allow an air- jority of private capital. In addition, govern- line of one country to operate flights between ments have become less willing to come to the two other countries without the flight originat- rescue of distressed national flag carriers. In- ing or terminating in its own country.44 deed, selected countries-notably in the devel- oping world-have allowed the bankruptcy -but bilateral arrangements are and closure of national carriers. While privati- becoming more liberal- zation is frequently driven by short-term fiscal Over time, ASAs have become increasingly needs, there is growing recognition that direct more liberal. For example, so-called Bermuda- or indirect subsidies to national flag carriers type agreements do not regulate capacity on distort the allocation of resources. The tight- each route, but leave it to the designated air- ening of competition policies in relation to lines to negotiate the number and frequency state aids has also contributed to a more com- of flights. "Open skies" agreements are an mercially oriented climate in which airlines even less restrictive type of ASA, which origi- operate today. nally emerged on selected routes to and from Besides restrictive bilateral agreements, the United States. Under a multiple open skies market access of foreign airlines is sometimes agreement, airlines can typically fly on all limited due to regulatory standards and re- routes between two countries without any ex quirements. While it is legitimate for more de- ante controls on capacity or fares, and are veloped countries to seek higher safety and granted unrestricted fifth freedom rights. Do- environmental standards, they can potentially mestic reforms, especially the entry of second have adverse effects on air services with devel- and third carriers to compete with the former oping countries, which should be taken into national flag carriers have also led to more in- account when adopting new standards (box tense competition on a considerable number 4.7). International cooperation on technical of international routes. In addition, unilateral standards, for example under the umbrella of and bilateral policies toward air cargo services the International Civil Aviation Organization are, in most countries, more liberal than pas- (ICAO), can play a useful role in forming con- senger services. Governments have often been sensus about what are legitimate safety or en- willing to authorize dedicated freight services vironmental concerns and what can be consid- when demand for services exceed what na- ered unnecessarily discriminatory. tional flag carriers could provide.45 Another noteworthy development is the -fostering consolidation- conclusion of liberal regional air service agree- A large number of studies have documented ments that, at least partially, attempt to over- the benefits of liberal international air service come the distortions introduced by bilateral markets in developed economies. In principle, preferences. These are often linked to regional competition between airlines has been shown trade agreements, such as in the case of the to result in overall lower prices, and an in- common aviation market in the EU or the An- creased range and quality of services.46 Little dean Pact open skies agreement. The "Ya- formal research has been conducted to evalu- moussoukro Declaration" adopted by African ate the effects of air service liberalization in countries provides for liberalization of air developing countries, but anecdotal evidence transport on the continent by 2002. The fore- points to significant inefficiencies as a result seen regime would replace bilateral air ser- of restrictive air service policies. At the same vices arrangements and eliminate all restric- time, the experience of developed countries tions in traffic rights up to the fifth freedom. has shown that liberalization may foster con- Privatization of state-owned airlines has solidation in the industry, as airlines seek to also progressed in the 1990s. More than 70 expand the reach of their networks to generate percent of airline companies now have a ma- hub-and-spoke economies. 118 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE Box 4.7 EU noise regulations and their potential effect on air service to Central Asian countries T he EU has been continuously framing regula - The noise level near airports is determined not tions to curb noise pollution within the Corn- only by the fleet mix serving the airport, but also by munity caused by older types of civil jet aircrafts. A the quantity of aircraft movements. A regulation that trade dispute between the United States and the EU bans such aircrafts in populated areas but not in has thus been brewing over the issue of hushkits- areas with low population density is an alternative retro-fitted noise muffling systems used extensively that can be applied for a limited period until the air- in U.S. carriers to bring older aircraft in conformity crafts are re-engineered in accordance with the envi- with ICAO standards. The hushkits law, effective ronmental standards. That could also imply channel- from April 2002, would ban all non-EU aircraft ing some of the air cargo through specialized remote with built-in hushkits that are not already flying in cargo airports. the EU. The negative implications for developing nations The 1L76, an aircraft with high cross-country are evident. Delays in urgent relief could have cata- carrying capacity will be prevented from operating strophic results. Cargo traffic between developing as a result of these regulations. According to one air countries and the EU would also be adversely af- transport operator, almost 90 percent of all humani- fected, as the costs of maintaining a fleet that is in tarian and disaster relief operations around the line with stricter noise regulations would increase. world are performed by the IL76. Such aircrafts play Taking into account the implications for developing a role in the advancement of emerging industries by countries when setting environmental laws and regu- carrying maintenance equipment and spare parts lations would make better development policy. worldwide. They have also been used widely in ser- vicing regions such as Kazakhstan, which lack suffi- Source: Council of the European Union 1999; and www. cient infrastructure to support Western aircrafts. coyneaircom. A regional market with limited traffic may -and increasing the relevance of only sustain a number of airlines that is smaller private practices. than the number of states in the region. Some A related concern stems from the emergence observers have, for example, pointed out that of a large number of airline alliances and code- consolidation would be a likely consequence if sharing agreements between airlines of differ- air services were further liberalized in Africa. ent countries. One of the main rationales of Consolidation may be in the consumer's best these arrangements has been to expand the interest, if economies of scale and scope result reach of existing networks in an environment in lower airfares and freight rates, yet it also where cross-border trade and direct invest- raises the danger that "spoke" routes with thin ments are restricted by bilateral ASAs.48 In traffic densities will become monopolized and addition, regulation and market structure in airfares increase once price and capacity con- industries upstream or downstream from air trols are removed.47 Achieving successful liber- services can strongly affect competitive condi- alization may require the regulation of prices tions for both passenger- and cargo-transport. and the imposition of service requirements on Chiefly, the allocation of landing and takeoff thin routes-at least temporarily until compe- slots at airports can be used to favor domestic tition has sufficiently intensified. incumbents and lead to a high concentration 119 GLOBAL ECONOMIC PROSPECTS of services in city-specific markets. Similarly, liberalization entails costs, in that market ac- passenger carriers need access to computer cess may be denied to the world's most effi- reservation systems, which are provided glob- cient airlines, unless those airlines fall under ally by only a small number of operators. the ambit of a bilateral agreement. Despite the spread of airline alliances, which has led to improved international network coordination, Unleashing competition in limitations on foreign ownership of airlines international transport: similarly prevent foreign airlines from fully policy implications integrating service networks and achieving economies of scale and scope. In the long term, Domestic policy action is needed- the goal should be to move toward a nondis- Government policies can play an important criminatory trade and investment regime in air role in improving the efficiency of interna- transport. Further liberalization at the domes- tional transport services. Creating a favorable tic level would contribute to an environment in climate for private investments, targeted pub- which such a regime would become feasible in lic infrastructure investments, and regional the future. cooperation on transport matters can serve to lessen constraints imposed by adverse geo- -as well as a strong regulatory and graphic or economic circumstances. As pointed competition policy framework out in chapter 3, the liberalization of service Liberalization needs to be accompanied by the markets should focus on the removal of entry development of appropriate regulatory mecha- barriers in the form of public monopolies or nisms. Regulatory intervention is necessary to specifically government policies that directly remedy market failures, to protect consumer limit competition. Such policy-imposed re- interest and the environment, and to ensure the strictions are present in a large number of safety of services supplied. Good regulation is countries and can apply to virtually all trans- often the key to successful liberalization. Al- port services, ranging from public shipping though there is no unique model of a good reg- lines, port monopolies, and national flag air ulator, experience has shown that clearly de- carriers, to controlled freight forwarding, and fined responsibilities, institutional and some agency and third party logistics markets.49 degree of financial independence, well-trained Cargo reservation in maritime transport, staff, and credibility in the market are impor- while still applied in a number of developing tant ingredients to the regulator's effectiveness. countries, has arguably become less relevant. Assistance from bilateral or multilateral donors Liberalization of port services is a much newer can be supportive, especially for newly created phenomenon, but has proved to be a success- agencies with limited resources. ful strategy in improving the performance of An adequate competition policy framework port operations in both developed and devel- is needed to address potentially anticompeti- oping economies. tive business practices by operators, and to en- Notwithstanding the recent progress to- sure that the gains from policy liberalization ward more commercially oriented and liberal are passed on to consumers of services. In prin- air service markets, the current system govern- ciple, greater scrutiny of private carrier agree- ing international air transport remains one ments by competition policy would not auto- that essentially grants preferential access to matically imply the breakup of all forms of airlines that reside at one end of an interna- private cooperation, but would require a static tional route. Even the most liberal bilateral and dynamic efficiency test as to whether car- open skies agreements and regional accords do rier agreements, alliances, and other private not grant seventh freedom rights. Preferential practices-whether in maritime or air trans- 120 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE port-seek to lower operational costs or work account the interests of foreigner consumers, to the detriment of consumers. and foreign persons usually do not have stand- ing in developed country courts. Yet effective application of competition A case can therefore be made to review policy may be difficult- competition regulations-including sectoral Many developing countries lack an adequate exemptions-in the major industrial countries national competition policy framework to deal in terms of their potential development impli- with private practices by transport operators. cations. This would not only make for better Although a large number of countries have overall development implications, but in many recently adopted antitrust laws, examination cases it could lead to better outcomes in de- and enforcement capabilities often remain veloped countries. weak and take time to develop. There are also significant extraterritoriality problems related Multilateral negotiations can be to the application of national laws to transport supportive of domestic reforms- services that are inherently international. Large Reform programs aimed at improving the per- states can probably tackle such practices uni- formance of transport services are primarily laterally, but small states with limited enforce- a challenge for domestic policy. Nonetheless, ment capacity are at a disadvantage. Coopera- multilateral agreements can help in several tion on antitrust matters (such as the collection ways to achieve good policy-as chapter 3 has of evidence) can help in pursuing multijurisdic- discussed in greater detail. Transport services tional practices, but, again, such cooperation fall under the scope of the General Agreement currently is most pronounced only among de- on Trade in Services (GATS), which was one veloped countries. of the outcomes of the Uruguay Round of trade negotiations. Measures listed in member -in part due to developed country countries' specific commitments include, for antitrust exemptions example, quotas such as cargo reservation poli- Undoubtedly antitrust scrutiny of international cies, foreign ownership limitations of service transport operators in big trading nations, such providers, requirements regarding the legal form as the United States and the EU, is likely to gen- of commercial presence, discriminatory taxes erate positive spillovers for developing coun- and subsidies, restrictions on the hiring of for- tries; yet such positive spillovers are likely to be eign crew members, and the terms of access to limited, for several reasons. First, the United port services and other essential facilities such as States, the EU, and other countries have histor- computer reservation systems. ically exempted-at least partially-shipping conferences from the realm of antitrust law, on -but little has been achieved so far- the grounds that they provide price stability Notwithstanding the broad coverage of the and limit uncertainty regarding available ton- GATS, relatively little has been achieved to date nage.50 In some countries, governments even on disciplining transport services by multilat- facilitate price-fixing by requiring ocean carri- eral trade rules. Take the case of maritime ers to officially file their rate and schedule in- transport services, where negotiations stretched formation. Similarly, the United States has ex- over a period of nearly ten years.51 Liberaliza- empted selected airline alliances from the realm tion was a central concern in the Uruguay of its antitrust law-justified by airlines' need Round, but at the end of the process only 39 to share scheduling and pricing information, WTO-Member countries were willing to offer which could be challenged under existing com- commitments, most with significant limitations. petition regulations. Second, developed coun- As in other sectors, such as telecommunications try competition laws typically do not take into and financial services, it was decided to extend 121 GLOBAL ECONOMIC PROSPECTS negotiations in this sector until the end of June broader and deeper exchange of commitments 1996. However, no agreement could be reached by Members. Although specific negotiating in- and negotiations were suspended. Thus even terests on transport services are likely to vary though the maritime transport sector is an inte- from country to country, there are general gral part of the GATS, it is not subject to the guiding principles that would arguably con- most favored nation (MFN) rule, and existing tribute to beneficial outcomes. First, develop- commitments are limited to those that certain ing countries are likely to gain credibility in Members have been willing to make unilater- their domestic reforms by binding existing ally. The suspension of the MFN obligation transport policies in a multilateral commit- was prompted by the difficulty in eliminating ment. Holding on to commitments that are MFN-inconsistent measures in the maritime "below" actual policy-for example, moti- sector. Examples of such measures are the vated by the desire to preserve future negoti- bilateral cargo-sharing arrangements under ating leverage-entails significant costs, in the UNCTAD Liner Code of Conduct, and cer- that investors may be deterred by the risk tain unilateral retaliatory actions-such as those of policy reversal. In maritime transport, the maintained by the United States-against trad- prospects for locking in existing policies have ing partners who are perceived to resort to re- arguably improved since the last round of ne- strictive foreign trade practices. gotiations, as unilateral liberalization in this Liberalization of air transport services sector has gathered steam, and a larger num- under the GATS has also been very limited. ber of countries appreciate that restrictions Current commitments only apply to three an- on maritime trade impose a significant cost on cillary services-aircraft repair and mainte- the whole economy.52 nance services, selling and marketing services, Second, developing countries should use and computer reservation services. The GATS the negotiating process to advance liberaliza- expressly excludes the core issue of air traffic tion of transport services-especially in sec- rights. Because the bilateral structure of the tors where there are powerful interest groups, international air service regime is fundamen- such as in port services, which resist reforms. tally at odds with the MFN principle of the At the same time, market access demands by World Trade Organization (WTO), exclusion trading partners need to be reconciled with was preferred to the possibility of scheduling a domestic reform priorities and overall devel- large number of MFN exemptions. Several de- opment objectives. This "balancing act" re- veloped countries-in part supported by their quires careful analysis prior to negotiations, airlines-also preferred to pursue the liberal- which should be supported by bilateral and ization of air services in a bilateral context. multilateral development agencies. The fact that these countries can obtain a rapid Third, and specifically regarding the cover- and timely resolution of disputes under the ex- age of air transport services under the GATS, a isting bilateral system contributed to the lack stronger multilateral framework for aviation of enthusiasm for a strong GATS framework. would, in principle, be desirable and could contribute to a more level playing field for -leaving the door open for mutually smaller countries. Realistically, application of beneficial negotiations in the new round the MFN principle to air transport-for exam- In 2000, new negotiations on services were ple, by substituting bilateral quotas with non- initiated, as called for in the GATS. If a broader discriminatory taxes-would require major new round were to be launched at the Minis- changes in the way the industry is currently terial Meeting in Doha, Qatar, in November governed, which seems unlikely in the short 2001, the scope for intersectoral bargaining to medium term. One way forward would be would substantially widen and encourage a to negotiate the inclusion of air cargo and so- 122 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE called express integrated cargo services, which daily capital would be 0.017 percentage ad valorem, are already relatively more liberal than air pas- roughly 47 times smaller than the measured cost. senger services. In the long term, multilateral 3. Amiadi and Yeats (1993) also show that African rules for these subsectors can create the mo- countries use a larger share of their foreign exchange earnings on net payments for transport services com- mentum for a more comprehensive treatment pared to other developing country regions. of air transport under the GATS. 4. Hummels (1999a) makes similar comparisons Finally, it may be beneficial to create mul- between freight rates and import tariffs for several tilateral disciplines on transport regulation and Latin American countries and, in many cases, finds measures that address anticompetitive business that tariffs do exceed transport costs, especially among practices. Such disciplines could unleash a manufactured goods. d e 5. Note that insurance services are included in the deeper exchange of liberalization commitments, definition of freight rates shown in figure 4.3. as countries would be more confident that mar- 6. See Limio and Venables 1999. Geraci and Prewo ket access concessions are not reversed by regu- (1977) estimate a similar elasticity of trade with respect latory barriers and that the gains from more lib- to shipping costs. eral policies are not captured by private parties. 7. Hummels (1999a) directly estimates the degree The Reference Paper on Regulatory Principles, of goods' substitutabilities, controlling for the trans- which is part of the 1997 GATS Agreement on port and tariff incidence on import prices. The study Basic Telecommunications, has demonstrated suggests an even larger trade-inhibiting effect of trans- port costs for individual product categories than the that multilateral disciplines can play a positive aggregate estimate by Limio and Venables 1999. role in this regard, without aiming at harmoniz- 8. For an overview of travel cost elasticity estimates ing regulatory standards or practices. The expe- of tourism demand, see Witt and Witt 1995. rience with these behind-the-border issues in the 9. See Christie and Crompton 2001. WTO is still young; further work is necessary to 10. A recent study on productivity spillovers in Or- evaluate possible options in the transport sector. ganisation for Economic Co-operation and Develop- ment countries finds that foreign research and develop- For example, extending nondiscrimination prin- ment (R&D) stocks in distant economies have a much ciples under the GATS to essential facilities in weaker effect on domestic total factor productivity than transportation, such as seaports and airports, or do R&D stocks in closer economies (Keller 2001). computer reservation systems, could make a 11. Indirect evidence for the role of export market positive contribution toward a secure trading choice on growth is provided by Vamvakidis 1998. regime for transport services. Competition dis- This study finds that the size of open neighbors' market could call for an end to exemptions of and their level of economic development has a positive c ieeffect on home country economic growth, although a particular sectors-such as air and maritime faster growth rate of the neighboring economy was transport-from domestic antitrust law. An- found to not provide any positive spillovers. other useful role the WTO might play in this re- 12. Since direct data on transport costs are unavail- gard is to uncover anticompetitive practices, for able, Redding and Venables (2001) use geographic dis- example in the context of the already existing tance and the existence of a common border to ap- trade policy reviews mechanism, or in the form proximate the effect of shipping costs. Estimations are performed for a group of 101 developed and develop- of dedicated competition assessments. Develop- ing economies, using 1994 bilateral trade data. ing countries that have limited resources avail- 13. See lanson 1998. able for this kind of analysis would likely be the 14. Interesting new work even suggests that trans- main beneficiaries. port costs---as an element of trade costs-help explain a variety of puzzles in the field of international macro- economics. Their role in explaining countries' home bias in consumption may be the most straightforward, Notes but Obstfeld and Rogoff (2000) also demonstrate that 1. See Lakshmanan 2001. trade costs can be an explanatory factor of why savings 2. If one assumed a 6.26 interest rate (the average in most countries are typically invested domestically, or U.S. Treasury Bill rate in the year of estimation), the even why exchange rates are excessively volatile. 123 GLOBAL ECONOMIC PROSPECTS 15. It should be pointed out that 1974 is an unfor- 31. See Hummels 2000. tunate year for comparisons, however, because freight 32. Admittedly, such best-practice estimates are rates were pushed up by the oil price shock in the pre- often crude and sometimes do not fully take into ac- ceding years. Based on similar data from New Zealand, count that practices or technologies employed abroad Hummels (1999b) finds that freight costs increased by may not be applicable at home. Moreover, the study is at least 30 percent between 1973 and 1974, such that based on the performance of Brazilian ports in 1997. the decline in ad valorem freight rates between 1974-98 Since then, port charges have been significantly reduced would be nearly eliminated. through the concessioning of private container termi- 16. Most of the discussion on ocean and air trans- nals to private operators. port is based on Hummels 1999b. This study provides 33. See PAlsson 1997. an excellent treatment of available evidence on the evo- 34. See Subramanian 2001. lution of international shipping costs. 35. See World Trade Organization 1998a. 17. See World Bank 2000. 36. The UNCTAD Liner Code was adopted in 1974 18. See Hummels 2000. and entered into force in 1983 through its ratification 19. These estimated growth rates are based on by more than 70 countries. Signatories are required to Hummels 1999b. divide the cargo transported according to the following 20. See Hummels 2000. Based on an estimated rule: 40 percent for ships belonging to the exporting daily ad valorem cost of 0.8 percent of the import country, 40 percent for ships belonging to the import- value, this study concludes that, "... the advent of rel- ing country, and 20 percent for ships belonging to atively fast shipping is equivalent to reducing tariffs other countries. from 32 to 11.4 percent." 37. Marin and Sicotte (2001) provide historical ev- 21. In the case of maritime transport, Fink and oth- idence of how the stock returns of ocean lines respond ers (2001) estimate that a 1 percent increase in distance to anticipated changes in the legal treatment of exclu- pushes up liner transport prices by 0.2 to 0.3 percent. sive contracts. Besides fixed transport costs, it is also possible that dif- 38. See Trujillo and Nombela 1999. ferences in the variable costs of shipping across ships 39. For a more detailed description of port owner- and routes cause freight rates to increase less than pro- ship and management structures, see World Bank 2001a. portionately with distance. 40. See World Bank 2001b. 22. See, for example, Rose 2000 or Limdo and Ven- 41. See World Bank 2000. ables 1999. 42. These shares were computed from operating 23. See Limio and Venables 1999. revenue data published in the Statistical Yearbook of 24. See, for example, Radelet and Sachs 1998, and the International Civil Aviation Organization. They Limio and Venables 1999. refer to the revenue of scheduled airlines, which in 25. See Radelet and Sachs 1998, and Limdo and 1996 accounted for more than 97 percent of all carrier Venables 1999. revenue. 26. Improving the infrastructure density index in 43. With some exceptions, chartered air services the export destination country by one standard devia- remain outside the scope of the bilateral ASAs. Their tion reduces transport costs by the equivalent of 6,500 authorization remains largely at the discretion of indi- kilometers by sea or 1,000 kilometers by land. vidual countries; airlines must satisfy the charter re- 27. Raising infrastructure density of the median quirements of both the origin and destination countries landlocked economy to the 25th percentile reduces the before commencing services. disadvantage of being landlocked by 12 percentage 44. First and second air freedoms grant the right to points; improving the infrastructure of the transit econ- fly over another country's territory or to land in an- omy reduces the disadvantage by a further 7 percent- other country for nontraffic purposes such as refueling age points. or maintenance. Sixth freedom rights are a combina- 28. For example, one study for Latin America esti- tion of two sets of third and fourth freedom rights- mates investment needs of $18 billion annually in Latin they allow an airline of one country to carry traffic be- America for 2000 to 2005, in order to bring road in- tween two other countries via its own country. frastructure to the upper-middle-income country aver- 45. See WTO 1998b. age of 2.32 kilometers per capita. See Fay 2000. 46. See, for example, Dresner and Tretheway 1992, 29. For a more detailed discussion of the role of the Gillen and others 1999, Gonenc and Nicoletti 2000, private sector in transport infrastructure investments, and Productivity Commission 1998. see Estache 1999. 47. For example, Brueckner and Spiller (1994) sim- 30. This estimate is based on the empirical model of ulate the effect of industry consolidation based on ocean liner shipping by Fink and others 2001. structural estimates of cost and demand parameters in 124 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE the U.S. domestic market. They find that a merger of No. 12. World Bank. [available at http://www. two carriers who share the same hub results in a fare worldbank.org/afr/wps/wpl2.htm] increase for passengers traveling on routes where pre- Council of the European Union. 1999. Official Journal viously only the merging carriers operated. Fares on L 115, 04/05/1999: 1-4. routes that remain competitive after the merger, how- Dresner, Martin E., and M. W. Tretheway. 1992. ever, fall, indicating that density gains compensate for "Modelling and Testing the Effect of Market the loss of competition. Because the merger leads to a Structure on Price: The Case of International Air substantial increase in total airline profit, its net wel- Transport." Journal of Transport Economics and fare effect is positive. Policy 26 (2): 171-84. 48. A recent study on global airline alliances con- European Union. 1999. "Community Legislation in cludes that, while the existing alliances are not stable Force." Document No. 393D0082. [available at enough to threaten competition of global airline mar- http://europa.eu.int/eur-lex/en/lif/dat/1993/en kets, individual alliances may be able to dominate cer- 393D0082.htmll tain hubs or even city pairs (Laaser and others 2000). - 2000. "Proceedings of the Court of Justice and 49. For example, one study on container transport in the Court of First Instance of the European Com- China identified ineffective competition for freight for- munities." [available at http://curia.eu.int/en/actl warding services-with 80 percent of the market being act00/0009en.htm] controlled by two state-owned enterprises-as a reason Estache, Antonio. 1999. "Privatization and Regulation for limited inland container use (World Bank 1996). of Transport Infrastructure in the 1990s: Suc- Similarly, entry restrictions in the provision of third cesses and Bugs to Fix for the Next Millennium." party logistics providers in Brazil are found to adversely World Bank, Washington, D.C. [available at affect Brazil's distribution economy (World Bank 1997). http://www.worldbank.org/wbi/regulation/pdfs/ 50. The exemptions from competition law in the 2248bugstofix.pdf] Processed. United States and the European Union are arguably Fay, Marianne. 2000. "Financing the Future: Infra- accompanied by a strong regulatory framework and structure Needs in Latin America, 2000-05." mechanisms that monitor competitive conditions in the World Bank, Washington, D.C. Processed. affected transport markets. At the same time, the inter- Fink, Carsten, A. Mattoo, and I. C. Neagu. 2001. ests of foreign consumers are either not or only margin- "Trade in International Maritime Services: How ally taken into account by authorities in these countries. Much Does Policy Matter?" Policy Research Work- 51. See Mattoo 2001. ing Paper No. 2522. World Bank, Washington, 52. See WTO 1998b. D.C. Gausch, Luis J., and J. Kogan. 2001. "Inventories in Developing Countries: Levels and Determinants, a Red Flag on Competitiveness and Growth." References World Bank, Washington, D.C. Processed. Amjadi, Azita, and A. J. Yeats. 1995. "Have Transport Geraci, Vincent J., and W. Prewo. 1977. "Bilateral Costs Contributed to the Relative Decline of Sub- Trade Flows and Transport Costs." Review of Saharan Exports?" Policy Research Working Pa- Economics and Statistics 59(1): 67-74. per No. 1559. World Bank, Washington, D.C. Gillen, David, R. Harris, and T. H. Oum. 1999. "A Atinc, Tamar Manuelyan. 1997. Sharing Rising In- Model for Measuring Economic Effects of Bilat- comes: Disparities in China. Washington, D.C.: eral Air Transport Liberalization." Working Pa- World Bank. per No. 99-08. Wilfrid Laurier University, Water- Audige, Michel. 1995. "Maritime Transport Serving loo, Ontario, Canada. Processed. West and Central African Countries: Trends and Gonenc, Rauf, and G. Nicoletti. 2000. "Regulation, Issues." Sub-Saharan Africa Transport Policy Pro- Market Structure and Performance in Air Passen- gram (SSATP), Working Paper No. 16. World ger Transportation." Economics Department Work- Bank, and Economic Commission for Africa, Wash- ing Paper No. 254. Organisation for Economic ington, D.C. Co-operation and Development, Paris. Brueckner, Jan K., and P. T. Spiller. 1994. "Economies Graham, Edward, and E. Wada. Forthcoming. "For- of Traffic Density in the Deregulated Airline In- eign Direct Investment in China: Effects on dustry." The Journal of Law and Economics 37 Growth and Economic Performance." In Peter (2): 379-415. Drysdale, editor. Achieving High Growth: Expe- Christie, lain T., and D. E. Crompton. 2001. "Tourism rience of Transitional Economies in East Asia. in Africa." Africa Region Working Paper Series Oxford: Oxford University Press. 125 GLOBAL ECONOMIC PROSPECTS Hanson, Gordon H. 1998. "North American Eco- Harvard Institute for International Development. nomic Integration and Industry Location." Ox- Processed. ford Review of Economic Policy 14 (2): 30-44. Redding, Stephen, and A. J. Venables. 2001. "Eco- Hummels, David. 1999a. "Toward a Geography of nomic Geography and International Inequality." Trade Costs." University of Chicago. Processed. Centre for Economic Performance Discussion Pa- -. 1999b. "Have International Transportation per No. 495. London School of Economics. Costs Declined?" University of Chicago. Processed. Rose, Andrew K. 2000. "One Money, One Market: - 2000. "Time as a Trade Barrier." Purdue Uni- The Effect of Common Currencies on Trade." versity, W. Lafayette, Ind. Economic Policy 15 (30): 7-46. International Civil Aviation Organization. Various is- Subramanian, Uma. 2001. "Transport, Logistics, and sues. Statistical Yearbook. Montreal, Canada. Trade Facilitation in the South Asia Subregion." Keller, Wolfgang. 2001. "The Geography and Chan- In T. R. Lakshmanan, Uma Subramanian, Wil- nels of Diffusion at the World's Technology Fron- liam P. Anderson, and Frannie A. L6autier, edi- tier." University of Texas. Processed. tors. Integration of Transport and Trade Facilita- Laaser, Claus-Friedrich, H. Sichelschmidt, R. Solt- tion Washington, D.C.: World Bank. wedel, and H. Wolf. 2000. "Global Strategic Al- Thoen, Ronaldt, Steven Jaffee, Catherine Dolan, and liances in Scheduled Air Transport-Implications Lucy Waithaka. 2000. "Equatorial Rose: The for Competition Policy." Kiel Discussion Paper Kenyan-European Cut Flower Supply Chain." No. 370. Kiel Institute of World Economics, [Available at http://wwwl.worldbank.org/wbiep/ Germany. trade/c_papers/Roses2KenyaSupplychain.pdf]. Lakshmanan, T.R. 2001. "Transport and Trade in Processed. Mercosur." In T. R. Lakshmanan, U. Subraman- Trujillo, Lourdes, and Antonio Estache. 2001. "Surfing ian, William P. Anderson, and F. A. L6autier, edi- a Wave of Fine Tuning Reforms in Argentina's tors, Integration of Transport and Trade Facilita- Ports." World Bank, Washington, D.C. tion. Washington, D.C.: World Bank. Trujillo, Lourdes, and Gustavo Nombela. 1999. "Pri- Limao, Nuno, and A. J. Venables. 1999. "Infrastruc- vatization and Regulation of the Seaport Indus- ture, Geographical Disadvantage, and Transport try." Policy Research Working Paper No. 2181, Costs." Policy Research Working Paper No. 2257. World Bank, Washington, D.C. World Bank. UNCTAD Review of Maritime Transport. Geneva. Marin, Pedro L., and R. Sicotte. (2001). "Exclusive U.S. Bureau of Census. Washington, D.C. Contracts and Market Power: Evidence from Vamvakidis, Athanasios. 1998. "Regional Integration Ocean Shipping." CEPR Discussion Paper No. and Economic Growth." The World Bank Eco- 2828. London, United Kingdom. nomic Review 12(2): 251-70. Mattoo, Aaditya. Forthcoming. "The Maritime Nego- Venables, Anothony J., and Nuno Limio. 1999. "Geo- tiations in the WTO." In Trade Policy for Devel- graphical Disadvantage: a Heckscher-Ohlin-von oping Countries in a Global Economy. Washing- Thunen Model of International Specialization." ton, D.C. World Bank. Policy Research Working Paper No. 2256. World Naughton, Barry. 2001. "Problems of Lagging Regions Bank, Washington, D.C. in China." Background Paper submitted to the Witt, Stephen E, and Christine A. Witt. 1995. "Fore- World Bank Beijing Resident Mission. casting Tourism Demand: A Review of Empirical Obstfeld, M., and K. Rogoff. 2000. "The Six Major Research." International Journal of Forecasting Puzzles in International Macroeconomics: Is 11: 447-75. There a Common Cause?" NBER Working Paper World Bank. 1996. "Container Transport Services and No. 7777. Cambridge, Mass. Trade: Framework for an Efficient Container PAlsson, Gylfi. 1997. "Containerized Maritime Trade Transport System." Report No. 15303-CHA. Between West Africa and Europe: Multiple Ports - 1997. "Multimodal Freight Transport: Se- of Call versus Hub-and-Spoke." World Bank, lected Regulatory Issues." Report No. 16361-BR. Washington, D.C. Processed. . 2000. "The Evolution of Ports in a Competitive Productivity Commission. 1998. "International Air World." Port Reform Toolkit, Module 2. [avail- Services." Inquiry Report No. 2. Australian Pro- able at http://www.worldbank.org/transport/ports/ ductivity Commission, Canberra, Australia. toolkit.htm]. Radelet, Steven, and J. Sachs. 1998. "Shipping Costs, - . 2001a. "Alternative Port Management Struc- Manufactured Exports, and Economic Growth." tures and Ownership Models." Port Reform Too]- 126 TRANSPORT SERVICES: REDUCING BARRIERS TO TRADE kit, Module 3. [available at http://www. worldbank. - . 1998b. "Air Transport Services." Background org/transport/ports/toolkit.htm]. Note by the Secretariat. (S/C/W/59). . 2001b. "Labor Reform." Port Reform Toolkit, - . 2000. "Communication from the European Module 7. [available at http://www.worldbank. Community and its Member States." (WT/ org/transport/ports/toolkit.htm]. WGTCP/W/140). World Trade Organization. 1998a. "Maritime Trans- Yeats, Alexander. 1981. Shipping and Development port Services." Background Note by the Secre- Policy: An Integrated Assessment. New York: tariat. (S/C/W/62). Praeger Scientific Publishers. 127 Intellectual Property: Balancing Incentives with Competitive Access Intellectual property rights can promote pectation that stronger IPRs would encourage development- additional technology transfer and innova- One of the most fundamental changes in global tion. However, the promise of long-term ben- commercial policy set out by the Uruguay efits seems uncertain and costly to achieve in Round of trade negotiations was the commit- many nations, especially the poorest coun- ment by all World Trade Organization (WTO) tries. In addition, the administrative costs and Members to adhere to the requirements of the problems with higher prices for medicines and Agreement on Trade-Related Intellectual Prop- key technological inputs loom large in the erty Rights (TRIPS). TRIPS defines minimum minds of policy makers in developing coun- standards of protection for intellectual property tries. Many are pushing for significant revi- rights (IPRs) and their enforcement. IPRs seek sion of the agreement. to balance the incentives necessary to encour- There are reasons to believe that the en- age future innovations (such as the ability to re- forcement of IPRs has a positive net impact coup the costs and risks of development, and on growth prospects. On the domestic level, still earn a profit) against the desire to provide growth is spurred by higher rates of innova- wide access to those products in a competitive tion-although this tends to be fairly insignifi- market. Because the overwhelming majority cant until countries move into the middle- of intellectual property-new inventions, pro- income bracket. Nonetheless, across the range prietary commercial information, digital enter- of income levels, IPRs are associated with tainment products, software, trade names, and greater trade and foreign direct investment the like-is created in the industrialized coun- (FDI) flows, which in turn translate into faster tries, TRIPS decidedly shifted the global rules rates of economic growth. of the game in favor of those countries. None- theless, TRIPS may lead to several long-run -so the poorest countries may require benefits for countries that take advantage of its assistance and time- standards in an appropriate and flexible man- The most appropriate level of IPRs enforce- ner, while complementing those standards with ment therefore varies by income level. In par- broader development and competition regimes. ticular, poorer countries-which are less able to absorb the associated costs, and least likely -but should be appropriate to local to benefit from domestic innovation-may capacities and benefits- find it advantageous to stage implementation Developing countries went along with the TRIPS of some aspects of IPRs. Since industrial coun- agreement for a variety of reasons, ranging from tries are the main beneficiaries of IPRs, and the hope of additional access to agricultural given the challenges facing developing coun- and apparel markets in rich nations, to an ex- tries, the former may find it in their interest to 129 GLOBAL ECONOMIC PROSPECTS provide assistance to the poorest countries for benefit from the introduction of new products the implementation of TRIPS. and technologies. This goal is achieved by the exclusive market positions afforded by IPRs. -and they also may require At the same time, society has an interest in special consideration in the case promoting widespread access to new products of essential medicines and information. Countries therefore limit the The least-developed countries face critical needs scope and duration of protected exclusivity in for access to new drugs and vaccines that may order to place goods into the public domain be developed for treating human immunode- after an adequate expected return has been ficiency virus/acquired immune deficiency syn- earned. There is an obvious tension between drome (HIV/AIDS), malaria, tuberculosis, and invention and dissemination. other diseases. Patent protection will raise in- Despite the inherent difficulty of measure- centives marginally for drug firms to invent such ment,1 a growing body of empirical work sug- treatments but could also support considerably gests that IPRs, as represented by legislated higher prices. A mechanism needs to be found patent rights, influence international eco- to reward innovation in this area while provid- nomic activity and growth performance.2 ing new medicines to poor countries at low cost. Like other economic policies, IPRs are cho- sen by governments in response to competing interests. Thus the strength of intellectual Intellectual property rights property protection depends on economic and and development social circumstances, which in turn affect per- ceptions of the appropriate tradeoff between Rationale invention and dissemination. Historically, At their most basic level, intellectual property countries have adopted stronger IPRs only rights exist to strike a balance between the when domestic interests in their favor became needs of society to encourage innovation and sufficiently strong to decide policy. This is fur- commercialization of new technologies, prod- ther supported by the wide variation in stan- ucts, and artistic and literary works, on the one dards across countries. The stronger the capa- hand, and to promote use of those items, on bilities of a nation's enterprises to develop the other. Intellectual property takes several distinctive products and new technologies, the forms (box 5.1). The need for intellectual prop- greater the preferences of consumers for qual- erty protection arises from the fundamental ity guarantees among similar products; the characteristics of information. It is often costly wider the markets in which artists wish to sell to develop new technologies and products, their music and literature, and the easier it is to requiring considerable investment in research misappropriate the returns to invention through and development (R&D) with uncertain pay- imitation, the more pronounced will be inter- offs. The investment extends further to the ests in protection. costs of bringing new ideas to the marketplace. These costs must be recovered through a Enforcement of rights increases temporary ability to set prices above marginal with income- costs of production. If an intellectual creation is Several stylized facts emerge from the litera- potentially valuable but easily copied and used ture about the level of development and IPRs. by others, there will be free riding by competi- First, countries with a high ratio of R&D in tive rivals. Such behavior would quickly drive gross domestic product (GDP) or a high pro- the price to marginal production cost and pre- portion of scientists and engineers in the labor vent the inventor from recouping investment force have markedly stronger patent rights costs, thereby discouraging innovation. Society than others. Clearly such countries desire to has a dynamic interest in limiting free riding to protect returns to inventive activity. 130 INTELLECTUAL PROPERTY Box 5.1 An overview of intellectual property rights A tthe broadest level, intellectual property has tra- works. The primary limitation on copyright protec- ditionally been divided into industrial property, tion stems from the fair-use doctrine, which defines or inventions and identifying marks that are useful conditions under which copying for noncommercial for industry and commerce, and artistic and literary purposes is permitted. property, or works of culture. This distinction re- TRIPS requires that computer programs be flected a perception that cultural creations differed protected, at least by copyrights, under the principle fundamentally from functional commercial inven- that software code is a literary expression. However, tions. However, this distinction has blurred consider- countries may vary in the degree to which reverse ably in the age of information technology and digital engineering of computer programs is permitted under products. the fair-use doctrine. There are four primary forms of industrial prop- Because computer programs may constitute a erty rights. First, a patent awards an inventor the commercially useful process, a number of developed right to prevent others from making, selling, or using countries permit firms to patent them. This policy is the protected product or process without authoriza- pushing patent protection more deeply into new tion for a fixed period of time within a country. In areas, including methods of doing business on the return, society requires that the application be pub- Internet. A similar evolution explains the tendency lished in sufficient detail to reveal how the technology toward awarding patents for biotechnological works, thereby increasing the stock of public knowl- research tools. edge. The minimum period of protection required For some technologies sui generis, or special, under TRIPS is 20 years from the date an application protection regimes exist. One is the design of inte- is filed. Many countries recognize utility models or grated computer circuits. These are more than literary petty patents, which award rights of shorter duration expressions, but the inventive step is often minimal, to small, incremental innovations requiring some suggesting a compromise between patent and copy- investment in design and development. right. Indeed, a 10-year protection term is provided A second form is industrial designs which and requires only novelty in expression. Another is protect the aesthetic aspects of a useful commercial plant breeders' rights (PBRs), which permit develop- article. TRIPS requires that designs be protected for ers of new, distinctive, and genetically stable seed va- a minimum of 10 years. rieties to control their marketing and use for a fixed A third mechanism includes trademarks and term. Many countries limit these rights by permitting service marks, which protect rights to use a distinc- an exception for farmers to use seeds for subsequent tive mark or name to identify a product, service, or replanting, and for researchers to study the seeds. firm. The fundamental objective of these marks is to Although not literally IPRs, a related area of reduce consumer search costs and remove consumer business regulation lies in defining the boundaries of confusion over product quality and origin. protection for proprietary trade secrets of rival firms. A related device is geographical indications, A production process or formula may be kept secret which certify that such products as wines, spirits, within the firm, but if a competitor learns the confi- and foodstuffs were made in a particular place and dential information through legitimate reverse engi- embody quality characteristics of that location. neering, the originator has no rights to exclude its Artistic, musical, and literary works are pro- use. Unfair competition includes such activities as tected by copyrights, which grant exclusive rights to industrial espionage, inducing employees to reveal the particular expression of the work for a period trade secrets, and encouraging defection of technical of time, typically the life of the creator plus 50 years employees to produce their own versions of a prod- (70 years in the United States and the European uct based on proprietary information. However, Union). Copyrights cover only expressions rather there is considerable variability in such definitions than ideas, and therefore provide thinner protection across countries. than patents. Rights extend to the duplication, dis- play, performance, translation, and adaptation of the Source: World Bank staff. 131 GLOBAL ECONOMIC PROSPECTS Second, the evidence suggests that interests IPRs can boost trade volumes- in encouraging low-cost imitation dominate Imports of goods and services can transfer and policy until countries move into a middle- diffuse technology. For example, imports of income range with domestic inventive and ab- capital goods and technical inputs could reduce sorptive capabilities.3 Only at high income production costs and raise productivity. An im- levels do patent rights become strongly pro- portant question is whether IPRs affect such tective. These findings may be explained by trade flows. Maskus and Penubarti (1995, the nature of technological development. 1997) estimated changes in imports of manu- Least-developed countries devote virtually no facturing goods and high-technology manufac- resources to innovation and have little intel- tures that could be induced by stronger patent lectual property to protect. As incomes and rights. A patent index from Rapp and Rozek technical capabilities grow to intermediate (1990) was increased by various amounts for levels, some adaptive innovation emerges but different countries to reflect roughly the com- competition flows primarily from imitation. mitments required by TRIPS. The anticipated Thus, the majority of economic interests pre- impacts on trade volumes depended on the ex- fer weak protection. As economies mature to tent of patent revisions, market size, and reduc- higher levels of technological capacity and de- tions in the imitation threats from complying mands shift toward higher-quality products, with TRIPS. Estimated effects on trade ranged domestic firms come to favor protective IPRs. from small impacts in the United States and Finally, the strength of IPRs shifts upward at Switzerland, which were not required to under- the highest income levels (Evenson and West- take much legal revision, to substantial in- phal 1997). Not only do legislated IPRs be- creases in imports in China, Thailand, Indone- come stronger, but enforcement and compli- sia, and Mexico, which must adopt stronger ance also rise with income levels. rights.4 Mexico updated its IPRs regime early because of commitments made under NAFTA. -and with greater openness of trade The study found significant impacts of Third, countries that are more open to trade IPRs change on import volumes of developing tend to have stronger patent rights. This result countries. For example, there was an antici- suggests that trade interacts positively with pated increase in manufactured imports into the demand for intellectual property protec- Mexico of $6.3 billion, amounting to 9.4 per- tion and, possibly, domestic innovative efforts. cent of its real manufactured imports in 1995. Finally, the size of an economy, as measured Thus, evidence suggests that the long-run im- by absolute GDP, has no detectable correla- pacts could be substantial. The estimated in- tion with patent rights. Thus, even in large de- crease in China's high-technology imports was veloping countries such as India and China it $2.8 billion, or just under 2 percent of its total may be some time before patent rights are ef- imports in 1995. Note that Coe, Helpman, fectively enforced. and Hoffmaister (1997) found that total fac- tor productivity (TFP) is enhanced in develop- IPRs and international economic activity ing nations through such imports. In principle In strengthening their IPRs regimes-either there could be a notable bonus to productivity unilaterally or through adherence to TRIPS- performance. developing countries may be able to attract However, most of the largest predicted im- greater inflows of technology. The three chan- pacts were in nations with strong imitation nels through which technology is transferred capacities, such as Argentina and Brazil. In across borders include international trade in contrast, India and Bangladesh would experi- goods and services, foreign direct investment, ence relatively weak, though positive, trade and contractual licensing of technologies. impacts.5 132 INTELLECTUAL PROPERTY -and attract FDI inflows and licenses Table 5.1 TROPS: who gains? A primary channel of technology transfer is Estimated changes in Payments for Technology and in FDI FDI. IPRs should have varying importance Flows for selected countries for full application of TRIPS (millions of 2000 dollars) across sectors with respect to encouraging FDI. Investment in low-technology goods and usue fa services should depend less on the strength of Paten U.So Royalties an IPRs and more on input costs and market op- Country IeMts F1 Assets License fees portunities. Investors with technologies that United States 19,083 aa n/a are costly to imitate also would pay little at- Germany 6,768 -1,180 100 tention to local IPRs. However, firms with eas- Switzerland 2,000 -102 0 ily copied products and technologies, such as France 3,326 n/a n/a Australia 1,07 -279 2 pharmaceuticals and software, would be quite Ireland 18 -267 14 concerned about the ability of the local IPRs New Zealand -2,204 -83 4 system to deter imitation. Firms considering Portugal -282 97 n/a Greece -7,746 51 n/a investing in a local R&D facility would pay Netherlands 241 -1,503 32 particular attention to protection of patents and Spain -4,716 -341 47 trade secrets (Mansfield 1994, 1995). Japan S,673 -2,533 783 United Kingdom 2,968 -1,369 29 Thus, the strength of IPRs and the ability to Canada -574 -2,396 69 enforce contracts could have important effects Panama n/a 309 1/a on decisions by multinational firms in certain Israel -3,879 6 0.6 onColombia n/a 1,190 n/a sectors on where to invest and whether to South Africa -11 25 11 transfer advanced technologies. Table 5.1 re- Rep. of Korea -15,333 270 388 ports results from the econometric estimation Mexico -2,50 3,4 148 pot eslsIndia -903 139 63 of a model of FDI and patent rights (Maskus Brazil -530 3,505 124 1998).6 Using the Ginarte-Park index, there Argentina n/a 721 64 was a negative elasticity of FDI assets with re- Chile n/a 1,6 rVa China -5,121 687 u3/a spect to patents in high-income economies, but Indonesia n/a 1,966 181 a strongly positive elasticity among developing Source: World Bank staff and Maskus (2000a). Figures for net economies. Applying these impacts to antici- parent rents update McCalman's (2001) coefficients applied to pated changes in patent laws from TRIPS gen- 1995 data. Calculations for the stock of FDI assets use coeffi- cients from an econometric analysis of the impacts of patent erates the estimated impacts on asset stocks in rights on patent applications, affiliate sales, exports, and affili- column 2. Reductions in asset stocks in Japan ate assets, using data over 1986-94 for the foreign operations of U.S. majority-owned manufacturing affiliates in several de- and Canada would amount to over $2 billion, veloped and developing countries. These coefficients were ap- for example.7 However, FDI assets would rise plied to 1994 asset stocks and updated to year 2000 dollars. Computations for royalties and license fees use coefficients significantly in Brazil, Mexico, Chile, and In- from an econometric analysis of the effects of patent rights donesia as a result of stronger patents. Indeed, on U.S. licensing volumes in manufacturing for 26 countries in 1985, 1990, and 1995. These coefficients were applied to the increase in the Mexican FDI assets would 1995 royalty fees and updated to year 2000 dollars. be 2.6 percent of the 1994 stock of U.S.-owned assets in that country, and in Brazil that would be 7.4 percent. Note that these figures related vestment decisions of U.S. multinational en- solely to U.S.-owned assets. If multinational terprises to their perceptions of the weak- firms headquartered in other developed nations nesses of IPRs in a sample of developing coun- were to react similarly, there would be even tries. They found that FDI is negatively larger increases in overall inward FDI stocks. affected by weak protection. Using firm-level Other studies of FDI and intellectual prop- data, Smarzynska (2001) discovered that for- erty protection bear mixed messages. Lee and eign investors considering operations in the Mansfield (1996) statistically related the in- countries of Eastern Europe and the Former 133 GLOBAL ECONOMIC PROSPECTS Soviet Union pay attention to patent rights. In age domestic innovation, product development, particular, investment in technology-intensive and technical change. It is possible to structure sectors is deterred by weak protection; in all JPR systems in ways that promote dynamic sectors weak protection discourages invest- competition through technology adaptation, ment in production 'facilities but does not learning, and follow-on innovation. However, deter investment in distribution. Smith (2001) many developing countries have regimes that also found that international FDI flows are favor imitation of foreign products and tech- positively related to IP protection. Using a dif- nologies and discourage domestic technical ferent econometric approach, however, Fink change. Indeed, inadequate IPRs can limit in- (1997) could not detect.a significant impact novation even at low levels of economic devel- of patent rights on various measures of FDI opment. This is because much invention and activity by U.S. orkGerman multinational en- product development are aimed at local mar- terprises. Thus, there remains statistical am- kets and could benefit from domestic protec- biguity about the nature of the relationships tion of patents, utility models, and trade secrets between IPRs and FDT, though most studies (see box 5.1). In the vast majority of cases, in- suggest it is positive. vention involves minor adaptations of existing Yang and Maskus (2001) studied technol- technologies and products. The cumulative im- ogy licensing. The figures in.the last column of pacts of these small inventions can be critical for table 5.1 update their results, of estimating the growth in knowledge and productive activity. impacts of international variations in patent An example is that protection for utility rights on the volume of unaffiliated royalties models (or "petty patents")-minor adap- and licensing fees (a measure of arm's length tations to existing technologies-improved technology transfer):paid to U.S. firms. Japan productivity in some countries (Evenson and had a large absolute -response, reflecting the Westphal 1997). In Brazil, utility models importance of licensing in-the Japanese econ- helped domestic producers gain a significant omy. However, large impacts were also dis- share of the farm machinery market by en- covered in the Republic ,of Korea, Mexico, couraging adaptation of foreign technologies Brazil, and Indonesia. Indeed the analysis sug- to local conditions. Utility models in the gested that licensing,wdlumes would double in Philippines encouraged successful adaptive in- Mexico and India, and would go up by a fac- vention of rice threshers. tor of nearly five in Indonesia. In another example, the Japanese patent sys- The findings discussed here are economet- tem (JPS) affected postwar Japanese technical ric predictions of long-run impacts of patent progress (Maskus and McDaniel 1999). The reforms on imports, FDI, and market-based JPS in place over the period 1960-93 was de- technology transfer. The figures are not defin- signed to encourage incremental and adaptive itive but do support the view that stronger innovation and diffusion of technical knowl- IPRs could have potentially significant and edge into the economy. It stimulated large num- positive impacts on the transfer of technology bers of utility model applications, which were to developing countries. This conclusion is based in part on published prior applications strongest for middle-income developing coun- for invention patents. In that study utility mod- tries. The results are less positive for the least- els had a strongly positive impact on real TFP developed economies, where the potential for growth over the period, because they were an market-power effects looms larger. important source of technical change and infor- mation diffusion. It is interesting to note that as IPRs and innovation in Japan has become a global leader in technology developing countries creation, its patent system has shifted away Developing nations also hope that stronger from encouraging diffusion and more toward intellectual property protection could encour- protecting fundamental technologies. 134 INTELLECTUAL PROPERTY If constructed well, IPRs also stimulate ac- At the same time, in many poor countries, quisition and dissemination of new informa- the effectiveness of all types of intellectual tion. Patent claims are published, allowing property instruments is held back by inade- rival firms to use the information in them to quate administration and enforcement proce- develop further inventions. A recent study on dures. These inadequacies may be due to cor- trademark use in Lebanon suggests that inno- rupt and inflated bureaucracies or weaknesses vation through product development and in the legal system at large-frequently affect- the entry of new firms is motivated in part by ing also the security of real and physical prop- trademark protection, even in poor nations erty rights. Hence, a weak overall governance (Maskus 2000b). Firms in the Lebanese ap- structure typically poses one of the biggest parel industry are capable of designing cloth- challenges to harnessing the positive contri- ing of high quality and style aimed at Middle bution IPRs can make to the development Eastern markets. Their efforts have been frus- process. trated by trademark infringement in Lebanon and in neighboring countries. Firms in the IPRs can boost growth prospects food products sector suffered from rivals pass- The analysis reviewed here suggests that se- ing off goods under their trademarks. The lecting appropriate IPRs systems could boost problem has restrained attempts to build mar- economic growth. History does not provide kets for Lebanese foods in the Middle East and strong guidance on this hypothesis. At differ- elsewhere. Related difficulties plagued innova- ent times and in different regions of the world, tive producers in the cosmetics, pharmaceuti- countries have realized high rates of growth cals, and other sectors. Thus, product develop- under varying degrees of IPRs protection. ment and enterprise growth have been stifled by Two recent empirical studies have consid- trademark infringement targeted largely at do- ered this question in a cross-country econo- mestic enterprises.8 metric framework. Gould and Gruben (1996) Copyright protection can induce investments related economic growth rates across many in creative activities and also stimulate innova- countries to a simple index of patent strength tion. Where protection is weak, such copyright and other variables. They found no strong di- industries as publishing, entertainment, and rect effects of patent rights on growth, but there software are dominated by counterfeiting rather was a significantly positive impact when those than domestic creation. Thus, lower-quality rights were interacted with a measure of open- copies are widely available, but the economy's ness to trade. The impact of stronger patent cultural and technological development may be laws in open economies was to raise growth hampered. For example, Lebanon has a small rates by 0.66 percent, on average. This suggests but vibrant film and television industry that that market liberalization and IPRs jointly in- could successfully export to neighboring econ- crease growth. omies if those countries adopted stronger copy- Park and Ginarte (1997) studied how IPRs right protection (Maskus 2000b). In the face of affect growth and investment. They found no difficulties in expanding their markets, Chinese direct relation between patent strength and software enterprises are now playing a role in growth, but there was a strong and positive promoting enforcement (Maskus, Dougherty, impact of patent rights on physical investment and Mertha 1998). Finally, work in such coun- and R&D spending, which in turn raised growth tries as Jamaica and Senegal shows that weak rates. copyrights and the absence of supporting insti- While these results are encouraging, the link tutions, such as professional collection societies, between IPRs and long-term economic growth significantly reduce incentives for local musi- remains poorly understood, and is likely to re- cians to record and market their compositions main controversial. More research is necessary (World Bank 2000). to provide better guidance to policymakers. 135 GLOBAL ECONOMIC PROSPECTS Costs of enforcing IPRs more productive activities. Indeed, in many While developing countries may enjoy poor countries, devoting more resources to the long-run gains from strengthening their protection of tangible property rights, such as systems, the transition to stronger protection land, could benefit poor people more directly involves short-run costs that are not trivial. than the protection of intellectual property. Three factors could help offset these costs. Administrative costs First, intellectual property offices may charge It is costly to develop the administrative and en- fees to defray their costs. Fees should be set forcement mechanisms necessary to support a to meet the innovation and commercialization modern system of intellectual property protec- needs of each country. Second, poor countries tion. Costs include upgrading offices for regis- may petition for technical and financial as- tering and examining patents and trademarks, sistance from industrial countries and from and for accepting deposits of plant materials; the World Intellectual Property Organization training examiners, judges, and lawyers; im- (WIPO) and the WTO. Unfortunately, the re- proving courts to manage intellectual property sources available are small in relation to the litigation; and training customs officers and un- underlying needs. Third, authorities may take dertaking border and domestic enforcement ac- advantage of cooperative international agree- tions. The United Nations Conference on Trade ments to reduce administrative costs. Mem- and Development (UNCTAD 1996) provided bership in the Patent Cooperation Treaty, for some estimates of the administrative costs of example, provides significant economies be- complying with TRIPS in various developing cause examiners may read the opinions made countries. In Chile, additional fixed costs from by major patent offices about novelty and in- this upgrade were estimated at $718,000 and dustrial applicability, rather than undertake such annual recurrent costs at $837,000. Egyptian technical examinations themselves. fixed costs would be perhaps $800,000, with additional annual training costs of around $1 Rent transfers million. Bangladesh anticipated one-time costs Patents are overwhelmingly owned by inven- of administrative TRIPS compliance (drafting tors in the industrialized countries. For example legislation) amounting to $250,000, and over in Mexico in 1996, only 389 patent applica- $1.1 million in annual costs for judicial work, tions came from domestic residents, while over equipment, and enforcement efforts. If training 30,000 came from foreign residents, mostly in costs were included it is likely that a compre- the United States and the EU. Brazil's domestic hensive upgrade of the IPRs regime in the poor- applications were just 8 percent of total appli- est countries could require an up-front expen- cations in that same year. In the poorest coun- diture of $1.5 to $2 million, plus recurrent costs. tries virtually no patents are granted to domes- Finger and Schuler (1999) report World Bank tic residents. As patent rights are strengthened, surveys finding that these costs could be far this relative imbalance could be reversed to higher. some degree, particularly in countries that de- Given other pressing needs in education, velop innovation systems and inventive enter- health, and policy reform it is questionable prises. However, inventors from developed coun- whether the least-developed countries would be tries are expected to apply for most patents for willing to absorb these costs, or indeed whether the foreseeable future. they would achieve much social payoff from As patents and trade secrets are better pro- investing in them. Moreover, note that poor tected, imitation costs rise and the ability of countries are extremely scarce in trained admin- patent holders to set higher prices and license istrators and judges, suggesting that one of the and royalty fees is enhanced. Thus, one impact largest costs would be to divert scarce profes- of TRIPS will be to transfer economic rents from sional and technical resources out of potentially technology importers to technology developers. 136 INTELLECTUAL PROPERTY Suggestive evidence is provided in table 5.1. tents or exclusive marketing rights. Nothing Firms own patents in various countries, the is more controversial in TRIPS. It is conceiv- values of which depend on local protection able that patent protection will increase incen- and market size. In an interesting study, Mc- tives for R&D into treatments for diseases of Calman (2001) used an econometric model to particular concern to poor countries. However compute the value of these patents in 1988. because purchasing power is so limited in the World Bank staff used his methods and regres- poorest countries, there is little reason to ex- sion coefficients to compute the values of in- pect a significant boost in such R&D. Accord- ternational patents among 28 nations in 1995, ingly, many developing countries see little po- using the Ginarte-Park patent index, patent tential benefit from introducing patents. applications, and GNP levels. Note that both In contrast, potential costs could be signif- patent applications and GNP had reached far icant. Pharmaceutical supplies in many devel- higher levels in the later year, thereby raising oping countries often come from domestic or the value of patent portfolios. To assess the imported generic competition. Such competi- change in patent rents associated with stronger tion for drugs on patents in the industrialized IP protection, the index for each country was countries helps sharply lower drug costs in de- increased to reflect obligations accepted in the veloping nations with active pharmaceutical TRIPS Agreement. industries. In the future, enterprises in these The figures in the first column of table 5.1 countries must wait until patent expiration be- show that overwhelmingly the United States fore they can compete with generic versions, would gain the most income in terms of static or else must produce under license to patent rent transfers, with a net inflow of some $19.1 holders. It should be noted that if firms choose billion per year. U.S.-headquartered firms not to register patents in certain countries, this owned numerous patents in many countries issue will not arise. that were required by TRIPS to strengthen There is some scope for stronger patents to their intellectual property protection, while encourage local firms to develop patentable U.S. law was subject to little change. Germany drugs themselves. Several Indian enterprises would earn an additional net income of $6.7 claim to be developing treatments that may billion on its patent portfolio. Many countries be patentable abroad, although they currently would experience a rising net outflow of pa- refuse to place them on the Indian market for tent rents because they tend to be net technol- fear of imitation.9 In most cases, however, ogy importers. Korea would register the larg- local enterprises will come under pressure to est net outward transfer of some $15.3 billion close down or form alliances with larger firms, because of the large rise in volume of patents resulting in a concentration of the industry. registered there. Developing countries also There is evidence that patents generate consid- would pay more on their patent stocks, with erably higher prices for protected drugs than China experiencing a net outward transfer of for copied and generic drugs (Lanjouw 1998; around $5.1 billion per year. These calcula- Fink 2001). Watal (1999) computed that sta- tions are static and ask only what the addi- tic price impacts of patent coverage in India tional income on existing patents would have could raise average patented drug prices by at been under TRIPS. They suggest that TRIPS least 26 percent from a 1994 base. could have a significant impact on net incomes In light of this possibility, developing coun- earned from foreign patents. tries need to gird themselves with policies that, while consistent with TRIPS, bear potential to Prices of patented drugs moderate the price impacts of new patents. Re- By January 1, 2005, developing countries must cent attempts by South Africa and Brazil to push provide patents for new pharmaceutical prod- the boundaries of TRIPS in this regard have ucts and most have already implemented pa- proven contentious, as discussed in box 5.2. 137 GLOBAL ECONOMIC PROSPECTS Box 5.2 Pharmaceutical policies and the limits of TRIPS In response to TRIPS, South Africa and Brazil re- licenses may be issued without observing even these cently introduced new laws bearing directly on the constraints in cases of national emergency. Finally, ability of those countries to react to price increases the price-control provisions of the South African that may emerge from patents. The greatest spur to amendments do not seem to be restrained by TRIPS, these attempts to limit patent rights came from a de- which does not address domestic health regulation. sire to procure AIDS drugs at affordable prices in order to manage that enormous health-care crisis. Brazilian Industrial Property Law Both laws are controversial. Brazil passed an industrial property law (Law No. 9,279) that came into effect in 1997. The law up- South African Medicines Law dated most aspects of Brazil's industrial property In November 1997 South Africa enacted significant regime to comply with TRIPS. It provides patents for amendments to the Medicine and Related Substances pharmaceutical products as recuired. However, it Control Act. The amendments permit the health permits the issuance of compulsory licenses in cases minister to revoke pharmaceutical patent rights in where patent holders choose tc supply the market South Africa if he deems the associated medicines to through imports rather than local production. That be too expensive. They further empower the minister is, Brazil's law does not recognize imports as a to order compulsory licensing if the patentee engages method for meeting its "working requirements" on in abusive practices, defined basically as a failure to the Brazilian market. The legis ation explicitly de- sell a drug in adequate amounts to meet demand, or fines "failure to be worked" as "failure to manufac- a refusal to license the product on reasonable terms ture or incomplete manufacture the product" or so that domestic firms may meet demand. They also "failure to make full use of the patented process." permit parallel importation (imports of original or While the Brazilian industrial property law refers to generic versions without the authorization of the all patents, its most aggressive use is aimed at trans- South African parent holder) of drugs, and allow the ferring production of AIDS drugs to domestic firms health minister to override regulatory decisions con- and government agencies in order to reduce their cerning the safety and registration of medicines. The prices below those on the U.S. and European mar- law requires pharmacists to employ generic substitu- kets. Media reports indicate that this active interven- tion (prescribe generic versions of patented drugs) tion has dramatically reduced treatment costs in unless the doctor or patient forbids it, sets limits on Brazil.10 In combination with prevention programs pharmacy markup rates, and bans in-kind induce- and effective methods for distribution and clinical ments from drug manufacturers to physicians. stays, the country has limited AIDS mortality to far While it may be a heavy dose of regulation, lower levels than those in Sub-Saharan Africa. South Africa's law is probably consistent with TRIPS It remains to be seen whether Brazil's insistence (Abbott 2000). While some legal scholars claim that on local production as a working requirement may patent rights necessarily extend to an ability to pre- be sustained within TRIPS. Because it applies to all clude parallel imports, the bulk of opinion is that patented items and not solely to medicines, the law Article Six of TRIPS provides full latitude for each may generate less sympathy among the WTO mem- country to choose its own policy on exhaustion. Be- bership than the South African law, despite its evi- yond this issue, Article 31 of TRIPS provides ample dent value as a threat to bring down prices. In nego- grounds under which compulsory licenses may be is- tiating TRIPS, patent advocates strongly favored an sued, subject to certain conditions (Watal 2001). In end to domestic production requirements, lending particular, licensing may be compelled where a support to the American view on their inconsistency. prospective user has failed to achieve a license from the patent holder on reasonable commercial terms within a reasonable period of time, so long as market-based compensation is paid. Compulsory Source: World Bank staff. 138 INTELLECTUAL PROPERTY Agricultural inputs apply standard intellectual property tools to its Under TRIPS, patents must be awarded to protection. Many such products and designs agricultural chemicals and biotechnological in- have found their way into international com- ventions, and effective protection must be pro- merce under protection in foreign countries, vided for plant breeders' rights (PBRs). Because however, as firms abroad copy and register them. farming is the mainstay of economic activity These problems point to a shortcoming in in many developing countries, policies that in- TRIPS. That agreement makes it clear that in- crease costs of key agricultural inputs could be ventions from genetic resources are patentable damaging. Plant strains bioengineered for pest- except in unusual circumstances. However, it and drought-resistance are of particular inter- is silent on the issue of how nations may reg- est to many developing countries. Note that ulate their extraction, an issue in which IPRs plant patents preclude the breeder's research are only one consideration. Similarly, it con- exemption and, unless explicitly allowed for in tains no provisions for defining and protecting the rules, also the farmer's privilege to retain rights in collective knowledge. It is important seeds for replanting. Experience from Latin for the global community to work out appro- America suggests that providing PBRs while priate mechanisms for ensuring the appropri- retaining this privilege does not much disad- ate valuation of resources and knowledge and vantage farmers (Maskus 2000a). for effecting payments that both conserve the materials and provide incentives for efficient Genetic materials and indigenous innovation. knowledge Because firms can attain patents in some in- dustrialized countries on products developed IPRs policies for promoting from plant and animal resources they find development anywhere, incentives exist to extract such ma- espite the significant costs, stronger intel- terials as sources for new drugs, food prod- D lectual property protection could produce ucts, and cosmetics. New patents in develop- gains in the long run through greater domestic ing countries will increase such incentives. innovative activity and cultural creation, prof- This "bioprospecting" raises several concerns. itable international exploitation of that activity, First, foreign patents have been awarded to enhanced structural transformation, and in- products and formulas that were already creased technology transfer. These gains are known in the source countries, or were simple more likely to materialize if countries adopt improvements, preventing those with the orig- standards and supporting policy regimes that inal know-how from marketing abroad (Duran promote competitive processes on their markets. and Michalopoulos 1999). Second, genetic ma- terials often do not bear adequate property IPRs standards at varying levels rights. Plants may be extracted from public of development- lands or from farms and villages that cannot TRIPS prevents countries from discriminating assert ownership or represent collective inter- between domestic and foreign firms in the ests. The resources may be acquired without treatment of IPRs. Beyond this basic stipula- compensation or attention to socially optimal tion, however, TRIPS contains considerable extraction rates. flexibility in implementing and enforcing stan- There is much know-how in developing dards that are conducive to development. One countries among tribes, villagers, and other col- important principle of a pro-competitive devel- lective units about how to produce foodstuffs, opment of IPRs policy is that the standards apparel designs, and artistic works. Because the adopted tilt the balance in favor of second- knowledge is a collective good, and therefore of coming rival firms. A second principle is that uncertain ownership, it has proven difficult to governments should not discourage inward 139 GLOBAL ECONOMIC PROSPECTS transfer of technology and should not suffo- tries to do more to encourage private technol- cate innovative efforts of domestic firms. The ogy transfer. The weakness of such action to essential goal is to move local entrepreneurs date remains a sore point leading some ob- from "free-riders" to "fair-followers" in Reich- servers to question the balance of interests in man's apt phrase.'1 TRIPS. Table 5.2 divides developing countries into three types and lists IPRs standards that are Administration likely to be most appropriate for each group.12 Administration and enforcement are costly. The first country type is low-income nations, Authorities in low-income nations could or the least-developed countries and some coun- achieve some gains by publicized raids and tries in transition, which have weak environ- consumer awareness programs. While such ac- ments for advanced invention but some capa- tions would face opposition among infringing bility at small-scale innovation and cultural enterprises, they would signal some commit- creation. The second is middle-income nations, ment to IPRs and also encourage domestic which have a strong imitative capacity and a creative interests to become more active. The reasonable degree of human capital. Such awareness itself may be the most valuable, and countries need to encourage technology adop- authorities could limit economic damages by tion and incremental innovation. The third is imposing moderate penalties for first offenses, high-income nations, which have a strong hu- with the severity of the fines rising with the ex- man capital stock and a growing capacity for tent of the piracy and the number of violations. innovation. It is evident that as countries be- Low-income countries cannot readily afford come more developed they may choose to patent examination offices and should rely on strengthen their IPRs. Table 5.2 is only a guide- patent registration instead. However, authori- line; individual countries may choose to pur- ties need to consult international patent offices sue their own standards as interests require. and databases to see if applications were de- This section analyzes possibilities for the low- nied elsewhere. Thus, developing countries income and middle-income nations. would benefit from the cost savings of using foreign sources of information, such as the -allowing poor countries the possibility Patent Cooperation Treaty. Countries could of exemptions also gain from adherence to regional examina- While countries must meet the general obliga- tion systems. Electronic access to international tions of TRIPS, there are some areas in which patent and trademark registries also cuts costs poor nations are afforded special status. of performing prior art examinations. As coun- Under Article 66, those least-developed coun- tries grow richer and technologically more so- tries experiencing difficulties in implementing phisticated, the patent system could move to- legislation may petition the TRIPS Council for ward domestic examinations. time extensions, and there is no specified limit Application and renewal fees for patents on the number of such petitions. While it is and trademarks may be set to cover the costs important to consider carefully the signals a of administering those regimes. It is sensible to delay would send to the global community, select fees in ways that promote desirable in- some countries may wish to take advantage of novation and use of IPRs. It is possible, for ex- it, particularly as regards the complex and con- ample, to set lower patent application fees for troversial subject of patents. small and medium enterprises than for large Both low-income and middle-income coun- firms. Patent renewal fees may rise over time tries would benefit from greater flows of tech- in order to encourage firms to let protection nical and financial assistance to develop, im- lapse on less-valuable inventions. This can be plement, and enforce IPRs. Poor developing an important means of pushing technologies countries also should push the developed coun- into the public domain. 140 INTELLECTUAL PROPERTY Table 5.2 TRIPS-consistent IPRs standards: options for developing countries Area of TRIPS LowIncome Middle-Income figh-Income General transition Consider Article 66 extensions in periods patents, trade secrets Assistance Push for technical ad financial Push for technical and financial assistance, including an assistance international fund Technology transfer Push for fulfillment of technology Consider providing technology transfer commitments transfer Administration Reduce piracy and counterfeiting Reduce piracy and counterfeiting Full enforcement Enforcement and through raids and awareness through raids and awareness customs Moderate fines and civil penalties Stronger fines and civil penalties Deterrent penalties Train customs officers for periodic Train customs officers for inspections mspections on demand Upgrade professionaihsm Judiciary No special IP court No special IP court Consider special IP court Training for Judges and attorneys Training for judges and attorneys Patents Registration system Registration or limited examination Examination system Administration system Rely on international grants data Rely on international grants data Consult international grants data Rapid and full disclosure Rapid and full disclosure FullJ disclosure Post-grant opposition Pre-grant opposition Pre-grant opposition Differential fees by applicant size Differential fees by applicant size More uniform fee Structure Rising renewal fees Rising renewal fees Rising renewal fees Standards and scope Fullest exemptions from parent Broad exemptions from patent Consider appropriate exemptions eligibility eligibility High inventive step using rigorous High inventive step Moderate inventive step international examinations Oral prior art considered Oral prior art considered Oral prior art considered Narrow claims Narrow claims Broader claims Narrow or no doctrine of equivalents Narrow doctrine of equivalents Broader doctrine of equivalents Permit experimental use Permit experimental use Permit experimental use Compulsory licenses National emergency use National emergency use National emergency use Public non-commercial use Public non-commercial use Antimonopoly tool Antimonopoly tool Antimonopoly tool Working requirements Permit imports to satisfy Permit imports to satisfy Limited working requirements Liberal definition of demand Utility models Recognize utility models Recognize utility models Industrial designs Recognize design rights Recognize design rights Recognize design rights Originality requirement Originality requirement Originality and novelty Supplement with copyrights Supplement with copyrights Supplement with copyrights Nonvoluntary licenses of right Non-voluntary licenses of right Nonvoluntary licenses of right Plant breeders' rights Provide PBRs Provide PBRs Consider patents Recognize farmers' privilege Recognize farmers' privilege Limited exemptions for farmers Permit breeders' exemption Permit breeders' exemption Permit breeders' exemption UPOV 1978 model with national UPOV 1991 model UPOV 1991 model or patents treatment Publc research and extension Public research and extension Extension services (co1inued) 141 GLOBAL ECONOMIC PROSPECTS Table 5.2 TRIPS-consistent IPRs standards: options for developing countries (continued) Area of TRIPS Low-income Middle-Income Igh4ieome Biotechnology Maintain exemptions from Maintain exemptions from Limited exemptions from patentability patentability patentability Strict standards for patent eligibility Weaker standards for patent eligibility Weaker standards for parent eligibility Narrow claims Broader claims Broader claims Contracts for efficient and equitable Contracts for efficient and equitable extraction extraction Integrated circuits TRIPS minimum standards TRIPS minimum standards TRIPS standards plus possible patents Trademarks Indefinite registration with rising Indefinite registration with rising Indefinite registration renewal fees renewal fees Registration contingent on use after Registration contingent on use after Registration contingent on use after 3 years 3-5 years 5 years Fair use of descriptive terms Fair use of descriptive terms Register service marks Register service marks Register service marks Define sector broadly for which Narrower definition Narrower definsition trademark is "well-known" Limits on protecting marks against dissimilar goods Protect domain names Protect domain names Protect domain names Geographical List generic and semi-generic names List generic and semi-generic names List genetic and semi- generic names Indications (GI) Registration system for indications to Registration system for indications to Registration system for indications to be protected be protected be protected Oppose or cancel registration of own Oppose or cancel registration of own Oppose or cancel registration of own GI abroad GI abroad GI abroad Push for common WTO list for wines Push for common WTO list for wines Push for common WTO list for wines and spirits and spirits and spirits Expand TRIPS protection for relevant Expand TRIPS protection for relevant Expand TRIPS protection for relevant products products products Copyrights Reduce piracy and raise awareness Reduce piracy TRIPS minimum period TRIPS minimum period Liberal fair use and compulsory Liberal fair use and compulsory Liberal fair use licenses for education, research licenses for education, research Reverse engineering in software Reverse engineering in software Permit patents under Non-voluntary licenses of right in Non-voluntary licenses of right in tight criteria software software Establish collection societies, Improve infrastructure contracts, infrastructure Identify copyrightable works Compliance with minimum standards Compliance with minimum standards Adopt WIPO treaties in WIPO treaties in WIPO treaties Require creativity for data Require creativity for data Require creativity for data compilations compilations compilations Trade secrets and Minimum definition of unlawful Moderate definition of unlawful Moderate definition of unlawful test data disclosure methods disclosure methods disclosure Limit employment restraints in hiring Limit employment restraints in hiring More permissive toward employment restraints High standard for defining" new High standard for defining" new chemical entity" chemical entity" No period for excluding prior Short period for excluding prior Longer period for excluding prior applicant's test data applicant's test data applicants test data Source: World Bank staff. 142 INTELLECTUAL PROPERTY Encouraging innovation owners to rival firms as a remedy for anticom- For reasons of promoting dynamic competi- petitive activity. tion, developing countries should require Protection for industrial designs can also rapid publication of patent applications (most promote innovation in developing countries. of which will have been published elsewhere Providing rights to registered designs with a in any case), with full disclosure of the techni- small novelty requirement, for a limited time cal processes involved in producing the inven- period, can promote product innovation. Such tions, and how to reduce them to commercial rights may be supplemented in two ways. First, practice. This should encourage local firms to designs may be protected under copyright law, invent around patents and use the disclosed even without registration. Second, countries knowledge to improve their manufacturing could experiment with systems in which, after methods. Countries with a registration system a shorter defined period of protection, rivals should permit active opposition after grants are able to acquire licenses to use the designs in are made, in order to invalidate inappropri- their own work. ately awarded patents. Those countries that Protection of plant varieties remains contro- undertake examination could permit pre-grant versial. When establishing PBRs, poor coun- opposition. tries would be advised to follow the UPOV Developing countries could permit oral 1978 model,13 providing the farmers' privilege prior art to defeat claims of novelty. They could and a wide exemption for rival breeders to use also provide a limited grace period in order to protected seeds to develop their own strains. maximize the inventions available in the public There is a role for public agencies to undertake domain to domestic firms. Authorities could research and disseminate new seed varieties. also preserve the rights of prior users of newly Middle-income economies are seeing develop- patented inventions to continue to use them ment of plant breeders, and there are potential with appropriate license fees. gains from protection. For patents, countries could set high stan- In biotechnology, lower-income economies dards for the inventive step, thereby prevent- may prefer to recognize narrow patent claims ing routine discoveries from being patented. and retain exemptions from patentability where Regarding patent scope, it is sensible to exer- allowed by TRIPS. Countries with stronger in- cise strict claims and discourage multiple claims dustries, such as China and Brazil, might award in patent applications. stronger protection in order to promote tech- Under limited circumstances governments nology transfer and domestic invention. may resort to compulsory licensing to promote Recognition of trademarks can promote do- the public interest in health, welfare, security, mestic enterprise development. In developing competition, and other grounds. Low-income countries it is often domestic entrepreneurs countries may wish to ensure that their patent who are frustrated in building their enterprises legislation and health regulations permit the because their marks are infringed by inferior issuance of compulsory licenses in patented products. This problem raises confusion on the medicines under sharply defined conditions. In part of consumers about the inherent quality addition to being consistent with the require- of commodities they wish to purchase. Thus, ments of TRIPS, compulsory licensing should recognition of trademarks can be an important be transparent and not arbitrary in order to development spur, even for poor countries. avoid discouraging entry of foreign firms and Geographical indications may be of particu- development of new technologies by domestic lar interest to numerous developing countries. firms. Compulsory licenses are available also Again, such indications reflect the quality char- as a primary restraint on monopolistic behav- acteristics of products coming from a particu- ior. Indeed, the United States has an extensive lar location. Because many developing nations record of compelling licensing from technology have a comparative advantage in agricultural 143 GLOBAL ECONOMIC PROSPECTS products and processed foods and beverages, ers may prefer to acquire trade secrets by pur- significant gains could be realized from regis- chasing licenses from the originator, thereby tration of such place names. This is one area in paying some share of the invention rents and which developing countries might be advised raising incentives for future inventive activity. to push for extended global standards. Trade secrets are also instrumental in encour- Cultural resources-including folkloric arts, aging technology transfer from abroad. designs, and traditional remedies-could be Poor nations would be advised to adopt the protected by a combination of copyright and least stringent regulations set out in the Paris trademark principles. The difficulty here is that Convention and perhaps also actively encour- such resources are often collective knowledge age technology transfer. Middle-income coun- and effectively in the public domain. Efforts are tries could establish more protective regimes, needed to work out appropriate standards for for example by imposing more stringent re- protecting such knowledge and the economic quirements on technical employees who are advantages that can be earned from it. induced to change employment. A distinction should be made between Governments have some obligation to pre- straightforward duplication of published and vent the public disclosure of confidential test recorded goods-also called piracy-and ac- data submitted for approval of medicines and cess to new information. While the former ac- agricultural chemicals for some period. Devel- tivities only yield short-run benefits, they do oping countries could establish a high standard little to enhance the technological capabilities for what constitutes a new chemical entity and of copying nations. deny such protection to simple reformulations Countries are free to determine the fair-use or repackaging. For those submissions meeting exceptions they will permit in the copyright the originality test, data need to be protected, area. Copyrighted materials may be made even though denying such information to rivals available on a limited and noncommercial basis would extend the time before generic competi- for use in teaching, research, libraries, muse- tion ensues. ums, and charitable organizations. Indeed, the preamble to the 1996 WIPO Copyright Treaty contains language promoting this balance of Other policies can support interests and encouraging nations to carry for- technological progress ward such limitations into the digital network 1 Jhile the standards sketched above are environment. V important in promoting competition TRIPS requires copyright protection for and innovation, simply adopting a stronger set data compilations. The EU has gone well be- of IPRs cannot be sufficient to ensure a posi- yond TRIPS' standards in specifying strong tive outcome. Intellectual property protection protection for databases even when their is but a component of broader business regu- compilation involves no creative step. Devel- lation, innovation promotion, and consumer oping countries should insist upon a demon- protection that must be conjoined in an effec- stration of creativity before recognizing such tive overall system.14 protection. Perhaps the most important complemen- Recognition of the need to protect confi- tary factor is a commitment to education, dential business information can also be pro- training, and skill development. The positive competitive. A natural lead-time is provided to role of educational attainment in economic the owners of trade secrets because rivals must growth is well established empirically. It is invest in learning the technical information plausible that a positive relationship exists be- they embody. This effort can contribute to the tween the strength of IPRs and the level (or technical knowledge capital of an economy growth) of human capital, given the results re- and encourage follow-on innovation. Follow- viewed earlier. 144 INTELLECTUAL PROPERTY Economies that are more open to trade developing countries may benefit from this and FDI experience a growth premium from change, at least in the long run, although there strengthening their IPRs relative to closed are bound to be significant short-run costs. economies. Competitive markets help limit the However in the short run, the developed coun- scope of intellectual property rights to their in- tries are likely to be the primary beneficiaries. tended function, which is to encourage invest- Moreover the introduction of global IPRs into ments in new products but not to prevent fair such areas as pharmaceutical products, agri- entry. In addition, a liberal stance on inward cultural inputs, biotechnology, environmental trade and FDI improves a country's access to technologies, and electronic databases has seri- available international technologies, intermedi- ous development consequences that merit care- ate inputs, and producer services. As discussed ful consideration. This situation suggests poli- earlier, IPRs are a factor that encourages in- cies in three general areas: ward FDI under appropriate conditions. Making IPRs stronger invites consideration 1. Collective international actions that can be of competition rules to discipline anticompeti- combined with the new protection regime tive practices. To abuse an intellectual property to help achieve important public goods right is to try to extend its exclusive use beyond 2. Ways developed countries can ease the tran- permissible limits. Claims that a rights holder sition burden for poor countries has engaged in anticompetitive behavior are 3. Approaches to IPRs that developing coun- complex, and resolving them requires signifi- tries could take in the "Development Round" cant judicial and legal expertise. Administrative costs may limit a country's ability to undertake International collective goods competition enforcement but the issue is suffi- The new global IPRs system could affect the ciently important to merit a high priority.15 willingness and ability of the international IPRs need to be supplemented by programs community to find effective solutions to a to promote national technical change. How- number of critical public-goods problems. ever, there are opportunity costs to the alloca- Consider three of the most important issues. tion of scarce budgetary resources to R&D First, the health status of impoverished programs. To the extent that investment in people in the least-developed countries con- product development is underprovided by the tinues to deteriorate. Beyond the debilitating private market, there is a rationale for public costs diseases impose on patients, medical sys- assistance. The limited R&D could be caused tems, and government budgets, it has spill- by such factors as an inadequate environment over effects on other countries through expo- for risk-taking, taxation systems that fail to sure to infection and reduced productivity. A recognize R&D as a business cost, and miss- role for public intervention exists in resolving ing information about technological opportu- the crisis. nities. Policies could aim to relax such re- By requiring countries to provide patents straints. This could be particularly important for new pharmaceutical products, TRIPS sets for small- and medium-size enterprises, which up incentives that may work at cross-pur- remain the source of much innovation in de- poses. By slowing down generic competition, veloping countries. patents could raise prices of new drugs in de- veloping countries and reduce the ability of patients to acquire drugs at reasonable cost. Multilateral actions and IPRs in a At the same time, the promise of wider and development round stronger patent protection could raise incen- T he TRIPS Agreement ushered in a new tives for private pharmaceutical firms to en- global regime for protecting intellectual gage in more R&D into the diseases of poverty. property. There are numerous means by which There is little private research undertaken in 145 GLOBAL ECONOMIC PROSPECTS such diseases (Sachs and others 1999). This sit- to demonstrate that they have attained the uation stems from both the absence of patent approval of local villages before going bio- protection and the extremely low purchasing prospecting or removing resources. power of patients in poor countries. TRIPS A third issue is how TRIPS affects incentives affords a solution to the former problem but to develop new transgenic crops through not to the latter. Consequently, TRIPS could biotechnological research. Widespread intro- raise costs without providing much incentive duction of new crops raises concerns about for innovation. biodiversity. The rapid increase in output of ge- Effectively addressing the diseases endemic netically modified plants attests to their advan- to poor countries requires separation of the tages in terms of enhanced disease resistance, dynamic incentives for R&D from the need reduced use of chemical inputs, and higher for widespread distribution at low cost. yields. It also suggests that traditional varieties Any comprehensive solution to the prob- could be pushed out of the market. IPRs pro- lem requires significant increases in foreign vide incentives for producing better crops but assistance from industrialized countries and ultimately might limit consumer choice. financial support from multilateral organi- It makes little economic sense to retard in- zations and private donors. These resources centives for developing new plants and food would be used for two purposes. One is to products by restricting exploitation of IPRs provide an incentive to firms to engage in beyond their usual limitations. A more prom- R&D into new and effective vaccines and ising and direct approach would be labeling medicines. This incentive could involve pur- programs that permit consumers to express chasing targeted drugs at negotiated prices or preferences for traditional crops and provide paying royalties for licenses that permit desig- market incentives to sustain their production. nated countries to produce and distribute them. Further if the disappearance of plant varieties By their recent actions in the area of HIV/ were seen as potentially damaging in environ- AIDS drugs, pharmaceutical firms have indi- mental terms, an argument would exist for cated a willingness to sell medicines cheaply, domestic and international public agencies to provided that exports back to developed stockpile such strains for purposes of keeping countries, where prices would be higher, are them alive as a form of social insurance. prevented. The other task is to fund the devel- To some extent the global IPRs system is opment of effective health-care delivery sys- inconsistent with public interests in resource tems in poor countries. conservation and biodiversity. For example, A second issue relates to incentives set up by the United Nations Convention on Biological TRIPS to extract biogenetic resources from de- Diversity stipulates that countries have sover- veloping countries. In principle contracts could eign rights over biological resources, while be devised to manage extraction of genetic ma- TRIPS recognizes private rights to own mi- terials. However it is not easy to determine ap- croorganisms and microbiological processes. propriate royalties when the resources are de- Developing countries that are the sources of veloped in areas without clear rights in natural genetic resources and natural plant strains property. Ownership may be collective within a need to assess their interests in revising TRIPS village or even undefined. to deal with this inconsistency. If Article 27 of Thus contracts need to be developed that TRIPS (dealing with patents in life forms and pay attention to both private incentives and protection for plant varieties) is revised, many public objectives. A role for governments arises developing countries should push for a resolu- here to ensure equitable and efficient sharing tion of the concept of resource rights and col- of the economic rents to IPRs earned on prod- lective ownership, along with the obligations ucts from extraction of domestic resources. of firms that extract resources. Thus for ex- For example, some countries now require firms ample, countries could push to forbid patents 146 INTELLECTUAL PROPERTY on plant-based products obtained from ma- on offer now, it is insufficient for the major job terials in international germplasm banks and of reforming IPRs administration. The current other deposit institutions. approach, whereby grants are made to such In many of these new areas, the legal and organizations as WIPO and UNCTAD for un- technical expertise needed to design carefully dertaking specific projects, is inadequate given balanced intellectual property and related reg- various bureaucratic constraints. ulations is likely to exceed the capacities of A valid justification for expanding assis- least-developed countries and even middle- tance is found in the asymmetric costs and income countries. Multilateral assistance can benefits from TRIPS. Intellectual property de- play an important role in ensuring that poli- velopers in rich countries stand to be the pri- cies promote development and in complement- mary gainers from the new systems, while ing direct funding for research on technologies there is little promise of gains for poor coun- addressing poor country needs. tries, at least for a considerable period of time. It could also be a wise investment in promot- Policy options for developed countries ing compliance with TRIPS and enforcement on TRIPS of IPRs, which might otherwise emerge only Technology-exporting countries have a strong slowly. Thus, developed countries could con- interest in sustaining TRIPS. Because of systemic vert their "best efforts" promises to binding difficulties among developing countries in ad- commitments, with benefits on both sides. justing to the new obligations and concerns Finally, the most important action devel- about its implications, industrialized nations oped countries could take to affirm confidence could consider several options to make the agree- in TRIPS is to meet and expand their obliga- ment more directly supportive of development. tions to provide greater market access for the First, in recognition of extreme budgetary exports of developing countries. Especially im- and institutional difficulties, least-developed portant would be new attempts to reduce bar- countries should be afforded latitude in exercis- riers to agricultural trade, which would greatly ing delays in implementation of TRIPS, espe- benefit many developing nations. Moreover, cially in the technically complex and controver- agricultural liberalization would raise the in- sial areas of pharmaceutical patents and plant centives of firms in developing countries to in- protection. Similarly, noncompliance problems vest in new agricultural technologies protected should not be the subject of dispute resolution by IPRs, thereby cementing faith in TRIPS. unless they constitute willful departures from basic TRIPS obligations. Developing countries and TRIPS reform Second, it should be recognized that devel- The interests of developing countries in alter- oping countries need to have lower and more ing or extending TRIPS vary greatly because, flexible IPRs standards than do their devel- in part, they have different levels of income oped counterparts. TRIPS provides such flexi- and technological sophistication. To rebalance bility in many areas and the developing coun- the agreement in some measure toward the in- tries should be afforded the opportunity to terests of the poorest countries, while allowing operate at the lower limits if it is in their de- for the quite diverse circumstances of coun- velopment interests to do so. tries, would help promote development. Third, developed countries could go a long First, extending the transition periods beyond way toward raising enthusiasm for TRIPS if 2005 for the least-developed countries would they would actively implement their "best ef- ease their administrative burdens. Although they forts" commitments to encourage technology have a limited opt-out procedure as discussed transfer to the least-developed countries and to earlier, a general recognition by the WTO mem- provide technical and financial assistance for bership of needs for extensions could be benefi- developing countries. While some assistance is cial in avoiding disputes. Such extensions should 147 GLOBAL ECONOMIC PROSPECTS be accompanied by serious commitments to be necessary to permit small, poor, countries work toward ultimate implementation. the right to import from foreign producers of- Second, the low-income and middle-income fering low-cost or generic products prior to countries should weigh carefully the introduc- patent expiration. Such a provision would pro- tion into TRIPS of significant new protection vide greater flexibility in addressing public for IPRs that would reduce their access to in- health crises. Even if such licenses may not ac- formation and technology. Extending patents in tually be granted, the option itself would likely biotechnology to additional life forms and to increase the bargaining power of governments plant variety protection could impose signifi- with regard to pharmaceutical multinationals. cant costs on developing countries, as would Fifth, many developing countries are inter- any attempt to globalize the highly protective ested in establishing new forms of IPRs over database systems in place in the European collective and traditional knowledge. Such Union or under contemplation in the United knowledge covers literary creations, such as States. Another form of protection to weigh oral histories, artistic works, music, designs, carefully is patents for software and methods pharmaceutical preparations, and methods of for doing business. Similarly, erecting global re- production. It is difficult to protect these items straints on parallel trade might have adverse with traditional IPRs precisely because they are potential competitive effects on future prices. traditional (and therefore not novel) and col- On the other hand, many developing countries lectively known, without easily assigned prop- have economic interests in extending protection erty rights. Thus, development of new rights, for geographical indications to their food prod- combining elements of trademarks, copyrights, ucts and handicrafts. This may help to ensure and trade secrets along with sui generis recog- that valuable geographic indications do not be- nition of traditional practices, could be benefi- come generic terms. Further, there are sound cial. A global principle that patents are not reasons for introducing the WIPO Copyright available for items that had been known to the and Phonograms Treaties into TRIPS obliga- public by means of oral tradition or written tions, so long as they retain flexibility for estab- description also would be beneficial for poor lishing liberal fair use of Internet transmissions. countries. Coordinated public efforts may be Third, despite proposals to remove from pa- required to catalogue these pieces of tradi- tent eligibility those drugs that are on, or will tional information. be on, the WHO "Essential Drugs" list, it is As these final comments suggest, IPRs unlikely that such discrimination by product evolve dynamically over time to meet the needs would be acceptable and, moreover, it could of inventors and creators in market economies. significantly reduce incentives to develop crit- The TRIPS Agreement significantly increased ical new drugs. A better alternative, discussed the requirements for protecting intellectual above, is to use public funds to purchase drugs property incumbent upon nations that wish to or licenses. So long as the financial offers be part of the global trading system. While cover anticipated R&D costs the incentives to promising some eventual benefits, the new develop new drugs would improve. regime is asymmetric in its likely effects across Fourth, current TRIPS rules may not allow countries. Low-income economies may expect governments to grant a compulsory license to to incur net costs for some time, suggesting foreign firms, and may not permit firms pro- that patience and assistance are needed, along ducing under compulsory licenses to export with programs to limit potentially negative ef- much of their production.16 This situation fects in such areas as new medicines. The pic- threatens to raise the costs of drugs in countries ture in middle-income economies is more com- where domestic production capacities cannot plex as they feature a mix of interests between ensure adequate supply of essential medicines. intellectual property developers, users, and im- A revision of the Agreement in this regard may itators. Experience with the negotiation and 148 INTELLECTUAL PROPERTY implementation of TRIPS should improve the costly investments in marketing and distribution chan- ability of developing countries to participate nels; enterprises that achieved it found their trademarks effectively in the further evolution of interna- applied to counterfeit products. Such products were of lower quality and damaged the reputation of the legiti- tional norms. mate enterprise. This problem deterred enterprise de- velopment and prevented interregional marketing. 9. The Economist, June 22, 2001. Notes 10. New York Times, "Look at Brazil," January 28, 1. It is difficult to quantify the strength of IPRs be- 2001. cause they are rules concerning conditions of dynamic 11. See Reichman 1996/1997, which provides the competition rather than taxes or subsidies applied to basis for some of the analysis in this section. See also particular sectors. Moreover, those rules have different Watal 2001. impacts under different economic circumstances. 12. Evenson and Westphal (1997) provide a more 2. This material is summarized from Maskus 2000a. nuanced categorization of countries but provide little 3. Controlling for other influences, there is a qua- concrete guidance regarding IPRs. dratic (U-shaped) statistical relationship between the 13. UPOV refers to a series of revisions of a treaty strength of patent rights and real per capita GDP. for the protection of plant varieties, which is known by Specifically, patent rights become weaker as incomes its French acronym. The 1978 revision serves as a model grow to a level of approximately $2,000 per capita in for developing countries, but is not now available for ac- 1985 international dollars ($3,000 today assuming an cession. The 1991 version provides stronger protection average growth rate of 2.5 percent), then become in- for breeders and is available for membership. creasingly stronger as countries get richer. 14. Maskus 2000a provides extensive discussion. 4. China has largely met TRIPS requirements in its 15. The papers in Anderson and Gallini 1998 pro- legislation in anticipation of joining the WTO. vide an excellent and comprehensive overview. 5. Smith (1999) found a similar outcome. 16. The European Union submitted a paper to the 6. The figures in column 2 of table 5.1 use coeffi- WTO TRIPS Council arguing that such licenses are ac- cients developed in a four-equation simultaneous deci- ceptable under the Agreement ("Paper Submitted by sion framework, which incorporated the impacts of the EU to the TRIPS Council for the Special Discussion patent rights on patent applications, affiliate sales, ex- on Intellectual Property and Access to Medicines," 20 ports, and affiliate assets. The model was estimated June 2001, IP/C/W/280), but legal opinion is divided. with data from 1986 to 1994 for the foreign operations of U.S. majority-owned manufacturing affiliates in sev- eral developed and developing countries. The assets References equation had a negative coefficient on patent rights, Abbott, Frederick M. 2000. "The TRIPS-Legality of suggesting that, on average, across countries stronger Measures Taken to Address Public Health Crises: patents would diminish the local asset stock. However, Responding to USTR-State-Industry Positions there was a large positive coefficient on patents inter- that Undermine the WTO." Manuscript. Florida acted with an indicator variable for developing coun- State University, Talahassee, Fla. tries, resulting in a positive and significant net impact Anderson, Robert, and N. T. Gallini. 1998. Competi- in those nations. This result likely means that at low tion Policy and Intellectual Property Rights in the protection levels internalization decisions encourage Knowledge-Based Economy. Calgary: University FDI as patents get stronger. However, as protection ex- of Calgary Press. ceeds some level there emerges a substitution effect fa- Coe, David T, E. Helpman, and A. W Hoffmaister. voring licensing over investment. 1997. "North-South R&D Spillovers." The Eco- 7. One possible explanation for this negative impact nomic Journal 107: 134-49. is that firms may exploit their IPRs in richer countries Duran, Esperanza, and C. Michalopoulos. 1999. "In- relatively more through arm's length licensing relation- tellectual Property Rights and Developing Coun- ships. Indeed, economic theory suggests that as IPRs are tries in the WTO 'Millenium Round'." Journal of strengthened, firms would choose to substitute licensing World Intellectual Property. November. contracts for FDI (Horstmann and Markusen 1987). Evenson, Robert E., and L. E. Westphal. 1997. "Tech- 8. Similar problems exist in China (Maskus, nological Change and Technology Strategy." In Dougherty, and Mertha 1998). Interviews suggested Jere Behrman and T. N. Srinivasan, editors, that trademark infringement negatively affected inno- Handbook of Development Economics: Volume vative Chinese enterprises. Numerous cases were cited 3A. Amsterdam: Elsevier North-Holland. of difficulties facing Chinese producers of consumer Finger, J. Michael, and P. Schuler. 1999. "Implementa- goods. Establishing brand recognition in China requires tion of Uruguay Round Commitments: The De- 149 GLOBAL ECONOMIC PROSPECTS velopment Challenge." Manuscript. World Bank, Maskus, Keith E., and C. McDaniel. 1999. "Impacts of Washington, D.C. the Japanese Patent System on Productivity Fink, Carsten. 1997. "Intellectual Property Rights and Growth." Japan and the World Economy 11: U.S. and German International Transactions in 557-74. Manufacturing Industries." Manuscript. World Maskus, Keith E., and M. Penubarti. 1995. "How Bank, Washington, D.C. Trade-Related Are Intellectual Property Rights?" _. 2001. "Patent Protection, Transnational Cor- Journal of International Economics 39: 227-48. porations, and Market Structure: A Simulation - . 1997. "Patents and International Trade: An Study of the Indian Pharmaceutical Industry." Empirical Study." In K. E. Maskus, P. Hooper, Journal of Industry, Competition, and Trade 1: E. E. Leamer, and J. D. Richardson, editors, Quiet 101-21. Pioneering: The International Economic Legacy Ginarte, Juan Carlos, and W. G. Park. 1997. "Determi- of Robert M. Stern. Ann Arbor: University of nants of Patent Rights: a Cross-National Study." Michigan Press. 95-118. Research Policy 26: 283-301. McCalman, Philip. 2001. "Reaping What You Sow: Gould, David M., and W. C. Gruben. 1996. "The Role An Empirical Analysis of International Patent of Intellectual Property Rights in Economic Harmonization." Journal of International Eco- Growth." Journal of Development Economics nomics 55: 161-86. 48: 323-50. Park, Walter G., and J. C. Ginarte. 1997. "Intellectual Horstmann, Ignatius, and J. R. Markusen. 1987. "Li- Property Rights and Economic Growth." Con- censing Versus Direct Investment: A Model of In- temporary Economic Policy 15: 51-61. ternalization by the Multinational Enterprise." Rapp, Richard T., and R. P. Rozek. 1990. "Benefits Canadian Journal of Economics 20: 464-81. and Costs of Intellectual Property Protection in Lanjouw, Jean 0. 1998. "The Introduction of Phar- Developing Countries." Journal of World Trade maceutical Product Patents in India: Heartless 24: 75-102. Exploitation of the Poor and Suffering?" NBER Reichman, J. H. 1996/1997. "From Free-riders to Fair- Exoition ofer and6, Sfferi n" N R followers: Global Competition Under the TRIPS working paper 6366, Cambridge, Mass. Areet"NwYr nvriyJunlo n Lee, J.-Y, and E. Mansfield. 1996. "Intellectual Prop- Agreement." New York University Journal of In- erty Protection and U.S. Foreign Direct Invest- Sachs, Jeffrey, M. Kreme, and A. Hamondi. 1999. ment." Review of Economics and Statistics 78: "The Case for Vaccine Purchase Fund." Harvard 181-86. University, Cambridge, Mass. Manuscript. Mansfield, Edwin. 1994. "Intellectual Property Protec- Smarzynska, Beata. 2001. "Composition of Foreign Di- tion, Foreign Direct Investment, and Technology rect Investment and Protection of Intellectual Prop- Transfer" International Finance Corporation Dis- erty Rights: Evidence from Transition Economies." cussion Paper 19. Washington, D.C. Manuscript. World Bank, Washington, D.C. - 1995. "Intellectual Property Protection, Direct Smith, Pamela J. 1999. "Are Weak Patent Rights a Bar- Investment and Technology Transfer: Germany, rier to U.S. Exports?" Journal of International Japan, and the United States." International Fi- Economics 48: 151-77. nance Corporation Discussion Paper 27. Wash- - . 2001. "How Do Foreign Patent Rights Affect ington, D.C. U.S. Exports, Affiliate Sales and Licenses?" Jour- Maskus, Keith E. 1998. "The International Regulation nal of International Economics 55: 411-39. of Intellectual Property." Weltwirtschaftliches Ar- UNCTAD (United Nations Conference on Trade and chiv 134: 186-208. Development). 1996. The TRIPS Agreement and 2000a. Intellectual Property Rights in the Developing Countries. Geneva. Global Economy. Washington DC: Institute for Watal, Jayashree. 1999. "Introducing Product Patents International Economics in the Indian Pharmaceutical Sector: Implications - 2000b. "Strengthening Intellectual Property for Prices and Welfare." World Competition 20: Rights in Lebanon." In B. Hoekman, and J. E. 5-21. Zarrouk, editors, Catching Up with the Competi- - . 2001. Intellectual Property Rights in the WTO tion. Ann Arbor: University of Michigan Press. and Developing Countries. London: Kluwer Law Maskus, Keith E., S. M. Dougherty, and A. Mertha. International. 1998. "Intellectual Property Rights and Economic World Bank. 2000. Workshop on "Developing the Development in China." Manuscript prepared for Music Industry in Africa." June. Washington, D.C. the Southwest China Regional Conference on In- Yang, Guifang, and K. E. Maskus. 2001. "Intellectual tellectual Property Rights and Economic Develop- Property Rights and Licensing: an Empirical Inves- ment. Chongqing. September. tigation." Weltwirtschaftliches Archiv 137: 58-79. 150 Envisioning Alternative Futures: Reshaping Global Trade Architecture for Development Deepening global trade and investment least one of the Quad countries (United States, integration holds the promise of more EU, Japan, and Canada). Besides merchandise rapid increases in standards of living trade barriers, restrictions on global trade in around the world, particularly in developing services also have impeded development. On countries. Greater openness and expanded the one hand, the lack of progress in the high- trade, partly attributable to the Uruguay income countries to grant access on temporary Round, contributed to new opportunities for movement of workers (mode 4 under General growth. Trade and incomes of developing coun- Agreement on Trade in Services-GATS) has tries grew during the 1990s at twice the rate foreclosed a potential source of earnings for de- of the previous decade, and those developing veloping countries. On the other, restrictions countries that deepened their integration with that developing countries place on foreign di- the global economy have seen their incomes rise rect investment (FDI) in services industries have at more than three times the pace of those that left unrealized their own full productivity po- did not (Collier and Dollar forthcoming). tential. Moreover, costs of transporting devel- The challenge ahead is to expand those op- oping-country exports are higher because of portunities and ensure that the poorest coun- quasi-cartel restrictions, which, when added to tries and poorest people benefit. Today devel- the "behind the border" under-investments in oping countries' exports confront higher levels ports, customs efficiency, and domestic infra- of border protection than those of developed structure, drive up the landed price of exports countries. The average poor person selling into and reduce volume. globalized markets confronts barriers that are Trade can only realize its potential if devel- twice as high as the typical worker in developed oped and developing countries alike take ac- countries (chapter 2). Said differently, products tion to reshape the global trade architecture to that the world's poor produce are more likely promote development. This chapter discusses to be subject to high tariffs, quotas, disadvan- in summary form the key policy foundations tageous subsidies, and antidumping claims of a new trade global architecture for develop- than are those produced by the better-off. Al- ment, and then shows how a phased program though only partly because of disadvantageous putting in place those policies might affect external circumstances, the 49 least-developed the long-term growth prospects of developing countries have fared particularly badly during countries. Our conclusion: a reshaped global the last decade. Thirty percent of exports from architecture can have dramatically positive ef- least-developed countries face tariff peaks in at fects on the lives of the world's poor. 153 GLOBAL ECONOMIC PROSPECTS Reshaping global trade be less cumbersome to negotiate than multilat- architecture for development eral reforms. Smaller memberships may also TJ ile the global trade architecture is make it easier to negotiate the increasingly im- VVlikely to evolve only slowly, the discus- portant issues inherent in regulatory regimes, a sion among world leaders on a future trade sharp contrast with complicated multilateral round can forge the first underpinnings. This negotiations involving more than 100 coun- report has focused on four policy domains: tries. Also, small countries can exercise greater influence in regional arrangements. * Policies to ignite a successful development Regional arrangements, properly designed, round in the World Trade Organization have the potential to stimulate global trade (WTO) that would produce tangible and through improving the efficiency, and hence the durable benefits for developing countries competitiveness, of regional producers and ex- * Policies for global cooperation outside the panding demand for inputs from nonregional WTO necessary to expand trade on a sus- sources. But regional agreements behind trade tainable basis, and to promote development barriers may artificially shift import supply * Policies of high-income countries to ensure from external countries to countries within the continued global growth and to facilitate trade area, and this may lead to reduced effi- trade expansion through provision of access ciency for participants if displaced external and aid suppliers would provide goods at lower cost. * Domestic policies that developing countries This trade diversion may disadvantage global might undertake to promote trade-led de- export competitiveness in much the same way velopment-with or without the help of the that national barriers do. "Rules of origin" international community arrangements in some regional agreements can raise costs and stifle local industry. This is also This report has not addressed other aspects true of mutual recognition agreements that of global trade architecture that have been may shield regional partners behind discrimi- taken up in previous Bank reports and numer- natory testing and certification protocols or re- ous other studies. These include issues such gional standards. Smaller countries with less as standards and environment as well as the technical capacity to evaluate these schemes workings of trade-related global institutions may find themselves at a net disadvantage, and (such as the World Intellectual Property Orga- be better off with first-best unilateral trade nization, World Customs Organization, and reform. International Air-Transport Association). Sim- Whether a particular agreement improves ilarly, we have not dealt with another element national incomes depends on its design, and of trade architecture-regional trading arrange- on the trading partners involved. Key design ments-which are particularly germane to the tests include whether regional arrangements objectives of this report, hence the digression involve lowering common external trade bar- below. riers, whether they stimulate increased compe- tition, and whether they reduce transaction A digression: regional arrangements costs and extend to nondiscriminatory invest- Regional arrangements to expand trade con- ment and services policies-all elements cen- tinue to proliferate. Governments, now more tral to "open regionalism." The World Bank receptive to openness than in previous periods, Policy Research Report Trade Blocs (World have sought to expand existing trade by lock- Bank 2000a) concludes that North-South re- ing in increased market access with trading gional agreements are more likely to improve partners-most often neighbors. Moreover, re- welfare than South-South agreements, simply gional arrangements are attractive because they because experience shows they usually result can increase the credibility of reforms and may in lower trade barriers with less trade diver- 154 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Figure 6.1 Regional integration agreements are proliferating Number of WTO notifications of regional integration agreements 18 15 12 9 6 3 3 . Ill.. ....l. . .Iii 0***.. 1957 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 Source: World Trade Organization. -and now span the globe. Selected regional integration agreements EU Austria Italy Belgium Luxembourg Denmark Netherlands EA Finland rtugal SEU Fie Icela,, Gen L U trstein W United Kingdom Norway II Euro-Ma NAFTA ro-ME Canada Moro eidoS Tunisia ASA Untd0a ASEAN CARICOM CEM i Ara lei Wbbean Cam r o Uni ra a Cep nt I A rica ep. I e oobia Lao Meic MUn Myar~ib Venezuela, Anan P Equator I a i Phil R.. dea B iGabon inSingp Ctbla ted c Thaila_ LAIA MERCO UR u Vietna , Per Argenti a Argentina Venez ia, Brazil / NigAE Bolivia R.B. Parag6ay ne I APEC Brazil rguay 0 Aus ia New Zea nd Chile Chil Papu Guinea Colombia BruI Peru Ecuador Canada 1 ines Mexico Chile ian Fe tdlon Paraguay China igapor Peru Indonesia Thailand Uruguay Japan United States Venezuela, R.B. de Korea, Venezuela, R.B. de Rep. of Vietnam Malaysia Mexico Source: World Bank staff. 155 GLOBAL ECONOMIC PROSPECTS sion, and because the greater structural differ- intensive manufactures are critical. Both the ences in North-South economies usually pro- high-income countries, and even the middle- duce greater potential gains from trade crea- income countries, will have to reduce their tion.2 The EU arrangements under the 1992 levels of protection in agriculture. In manu- Single Market Program are a clear case in factures, political commitment is necessary to which the analysis shows income-increasing phase out the quotas of the Agreement on Tex- effects. The North American Free Trade Agree- tiles and Clothing (ATC) in 2004 and reduce ment (NAFTA) also appears to have had a the high levels of tariff protection that would positive impact on its Members, particularly otherwise impede access once the quotas are Mexico. ended. These efforts should be accelerated. It Regional arrangements are likely to remain also means a commitment from the high- an enduring feature of the trade panorama. To income countries (HIC) to reduce tariff escala- realize possible benefits of trade and invest- tion and tariff peaks that now discourage the ment expansion, arrangements have to be de- creation of new industrial activities in devel- signed in a way that they become stepping- oping countries. Trade in services can be ex- stones to greater openness and development, panded-opening new vistas of productivity rather than a vehicle for protection and unin- gains for developing and developed countries tended inefficiency. An important component alike-if countries permitted more movement of making them steppingstones rather than of temporary workers and reduced anticompet- stumbling blocs to greater openness is for itive and discriminatory restrictions on foreign the countries involved to have low protection investment. Electronic commerce (e-commerce) against non-Member countries. For example, deserves greater attention under GATS to pro- Harrison, Rutherford, and Tarr (1997) esti- vide maximum competition. Using GATS to mate that Chile was able to gain from its free eliminate anticompetitive aspects of the private trade agreement with Mercosur due to the fact carrier agreements in maritime transport and to that it lowered its external uniform tariff from engender new competition in air transport 11 to 6 percent. Regional agreements can fa- could lower the costs of delivering developing- cilitate the deep integration-reduction of bor- country exports to foreign shores. der barriers, promotion of cross-investments, Antidumping, recourse to other forms of adoption of common regulations, and even unilateral contingent protection, and overly cultural and political exchange-in ways that stringent produce standards have dampened reinforce and enhance multilateral efforts. the access that developing countries have to the Nonetheless, the world market is bigger world's major markets. Whether it is shiitake than the market next door, so for all their co- mushrooms entering Japan, steel entering the ordination difficulties, multilateral efforts to United States, or products entering the EU, expand market access can have greater impact raising barriers to trade to protect domestic on development. For these reasons this report markets has too often hurt development. Ap- has focused on a four-part agenda: a develop- plications of contingent protection are not lim- ment round, global cooperation to expand ited to developed countries. Middle-income trade, policies of high-income countries, and countries have increasingly sought refuge from policies of developing countries (box 6.1). the competitive pressures of their neighbors. One immediate measure would build confi- A development round: policies in the dence and show convincing movement on the WTO to expand trade opportunities for ATC: an agreement to limit the use of anti- the world's poor dumping on trade in textiles and clothing that Market access. For the world's 2.8 billion will be liberalized as negotiated in the Uruguay poor, reducing barriers to agricultural products, Round. Over the longer term, the use of anti- textiles, clothing, apparel, and other labor- dumping ought to be phased out (Finger 156 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT 1998). Other "safeguard" instruments might countries must understand them, participate be disciplined by giving standing to users of the fully in their formulation, and buy into them. goods concerned in the decision process. Two elements would serve that end. First, Implementation issues. No less important transparency is vital for ownership and im- than market access is tailoring implementation plementation. Enhancing the transparency of of existing and new agreements to the local WTO operations and improving access to and capabilities of developing countries. Develop- dissemination of WTO databases, reports, and ing countries, given power asymmetries, have information (for example, data underlying na- an interest in avoiding a two-track multilateral tional trade policy reviews) would broaden system that relegates them to a particular posi- the basis for participation of developing coun- tion; however, implementing global agreements tries to engage in the policy formation process can be better calibrated to domestic capacities. (Francois 2001).3 For example, the administrative costs of imple- Second, a determinant of ownership of menting Uruguay Round agreements on The agreements is the ability of countries to par- Agreement on Trade-Related Intellectual Prop- ticipate in the WTO process. Many countries erty Rights (TRIPS) and customs procedures have inadequate representation in Geneva, can run into the tens of millions of dollars, and impeding active engagement in negotiations. could easily swamp the investment budgets of Although options have been identified to ex- many poor countries (see Finger and Schuler pand representation in Geneva at relatively low 2000). As developing countries have empha- cost, expertise is still in short supply.4 Funding sized in several recent meetings (e.g., LDC3, could be made available to allow low-income Abuja and Zanzibar Trade Ministers' con- countries to finance the cost of hiring experts ferences), these implementation concerns are that can undertake the required analyses paramount if a negotiation round is to promote (Winters 2001). The annual cost of such an as- development. sistance program to least-developed countries Moreover the benefits they would receive in could be in the $10 million range.5 terms of greater access to low-cost technology are, relative to the implementation costs, ques- Global actions outside the WTO tionable. New trade rules recognizing these to expand trade: beyond negotiation constraints would allow flexibility and provide to cooperation for transition periods linked to development Expanding trading opportunities for the capacities. To be effective, implementation world's poor requires going beyond negotia- would have to be linked to a long promised fi- tions in the WTO to cooperation in other pol- nancing facility that would provide techni- icy domains. Two sets of complementary poli- cal assistance to implementation, and the high- cies are particularly important. income countries could convey their seriousness Increasing multilateral development assis- by agreeing to bind this commitment. Note that tance to expand trade can help countries take all of these issues could be decided during ne- advantage of existing global markets, respond gotiations, and none need hold a new round to global and domestic trade policy reforms, hostage to prior action. and link the poor to new opportunities. Multi- lateral cooperation among bilateral donors can WTO transparency and participation. Be- provide "aid for trade." One important exam- yond these elements, the convening of a round ple: The EU has taken the lead in providing gen- of talks is likely to promote development only erous assistance to the Integrated Framework if agreements enjoy full ownership among (IF), a program designed to analyze obstacles to WTO Members. For agreements to realize trade for least-developed countries and provide their potential mutual benefits, major con- assistance in overcoming them (see box 6.2). A stituencies in both developing and developed similar approach could usefully be applied to 157 GLOBAL ECONOMIC PROSPECTS Box 6.1 Reshaping global trade architecture for development: The four-part policy agenda 1. Convening a development round in the WTO Market access Agriculture * Reduce applied tariffs, phase out tariff rate quotas, and bind tariffs at applied rates in both developed and developing countries * Phase out export subsidies in high-income countries and commit to eliminate domestic support linked to production levels * Reduce tariff escalation and cut off tariff peaks Manufactures * Reduce applied rates further, and bind tariffs to levels that equal or are close to applied rates * Reduce tariff escalation and cut off tariff peaks * Accelerate implementation of ATC quota eliminations and reduce tariffs in lines now covered by quotas * Negotiate tighter disciplines on antidumping and other forms of contingent protection Services * Liberalize entry of foreign services suppliers through elimination of restrictions on entry and promoting increased competition, with wider use of GATS to bind nondiscriminatory access and lend credibility to domestic programs * Enhance scope of services provision through the temporary movement of service providers (both skilled and unskilled) * Secure openness of e-commerce in services, through wider and deeper GATS commitments on cross-border supply * Strengthen multilateral rules to deal with anticompetitive practices in services * Adopt a nondiscriminatory trading regime for air transport, including traffic rights, under GATS Implementation procedures and phasing * Adopt a phased implementation of TRIPS and other administrative-intensive agreements for low-income countries, based upon development capacity. * Establish a consensus that the TRIPS Agreement allows developing countries with no domestic production capacity to grant compulsory licenses to foreign firms * Convert "best endeavor" promises to binding commitments to provide low-income countries with financial and technical assistance to implement WTO accords Improving WTO transparency and participation * Require WTO disclosure of databases; reports and their full associated information; and analyses for particular decisions * Provide assistance to strengthen capacity of all members to participate effectively in negotiations 2. Global cooperation to support trade outside the WTO Provide "aid for trade" through stepped up development assistance * Expand "Integrated Framework" assistance to all low-income countries * Provide assistance to enhance the efficiency of the customs clearance process in developing ccuntries, notably the good customs practices that are laid out in the revised Kyoto Convention (World Customs O:ganization) * Expand multilateral assistance to overcome country-specific bottlenecks to improving competitiveness and trading potential (for example, in finance, transportation infrastructure, education for low income workers, and public sector trade-related institutions) and to promote trade 158 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Box 6.1 (continued) * Fund mechanisms to help developing countries use intellectual property protection to their benefit by protecting intangible assets such as traditional knowledge, designs, music, and ethnobotanicals, and patent protection for industrial goods as well as improve enforcement of IPRs * Establish a global health fund to purchase licenses from developers of new medicines essential to treating debilitating diseases in poor countries Expand global efforts beyond trade to improve the environment, raise labor standards, and adopt adequate product standards outside the WTO * Expand global environmental cooperation with financing to improve environmental protection in developing countries, and create multilateral forum of environmental exchange * Strengthen international actions on labor standards through the International Labour Organisation (ILO), with project collaboration from multilateral development banks * Create a Standards Development Facility to introduce science and other professional evidence into standard setting for products, with adequate representation from developing countries and provide assistance to developing countries' standard setting bodies 3. Policies for high-income countries Market access * Grant to all low-income countries duty-free and quota-free access to markets of all countries of OECD * Reduce uncertainty of market access by harmonizing rules of origin, and by reducing threats of antidumping Expand bilateral "aid for trade" * Provide financial and technical assistance to developing countries for "behind the border" trade-related invest- ments necessary to take advantage of market access * Improve policy coherence by establishing coordinating mechanisms between development policies and trade policies to ensure effective development outcomes * Assist developing countries to strengthen competition agencies and improve legislation, and require antitrust agencies to provide to developing countries information on third market effects of domestic mergers as well as pending cases of price-fixing and restrictive business practices; and review the anticompetitive consequences of antitrust exemptions in transport and other sectors that adversely affect development Domestic policies that facilitate adfustment of labor to economic change * Review domestic policies to ensure displaced workers have adequate social support to deal with rapid changes in labor market conditions, including unemployment insurance, social safety nets (particularly health and pensions), and access to training and education 4. Policies for developing countries * Adopt program of trade reform, including phased lowering of border protection for goods and services as part of a poverty reduction strategy * As part of the trade reform program, adopt companion policies to cushion any impact on the poor of adjustment to new trade incentives, and ensure investment responses; solicit foreign assistance when necessary to implement administrative requirements of programs * Spur development of industries essential to trade, such as transport, telecommunications, financial sector, and business services, particularly through introduction of regulatory policies that, where feasible, harness competition * Invest in upgrading public sector institutions related to trade, including customs, administration of drawback programs, and financial supervision agencies * Encourage domestic intellectual property development through TRIPS-consistent standards appropriate to country needs, and pursue protection of domestic intellectual property abroad * Ensure adequate macroeconomic policy framework to provide sound investment climate 159 GLOBAL ECONOMIC PROSPECTS Box 6.2. The recently renovated integrated framework T he Integrated Framework (IF) is a program set up ing the reform transition as options the government by bilateral donors to increase the effectiveness of might consider in preparing its poverty reduction trade-related technical assistance to the least-developed strategy papers (PRSP). The process starts with analy- countries. The IF was established in 1996; participat- sis: how trade might fit into national development ing agencies include the WTO, the International strategies, followed by assistance in the design and fi- Monetary Fund, the International Trade Center, nancing of projects (drawing on cross-country experi- United Nations Development Programme, United Na- ence). An interagency task force was formed during tions Conference on Trade and Development, and the 2000, and a trust fund has recently been established World Bank. Its purpose is to analyze options for to fund the "integration studies" and technical assis- trade-led integration, determine the relative payoff to tance that can be built into the country assistance trade-related reform, and work with local counter- strategies as appropriate. parts to design appropriate policy reform packages that both promote growth and protect the poor dur- Source. World Bank staff. other low-income countries, and this too will re- nars, and conferences-often in collaboration quire resources. "Aid for trade" could also help with global nongovernmental organizations speed adoption of best practices in customs ad- (NGOs). However, the potential for collective ministration as contained in the revised Kyoto action in the environment has barely been Convention and administered by the World scratched. The phenomenal success of the Mon- Customs Organization. It could help with fi- treal Protocol in reducing ozone-depleting sub- nancing of infrastructure related to trade (ports, stances (ODS), with the felicitous reversal of transport, and related services), logistics, trade trends toward an ever larger ozone hole over facilitation, and trade promotion; in many cases the Antarctic, is worthy of study and emulation. bottlenecks in one or another area impedes ex- The Global Environmental Facility to reduce port expansion from a particular country. Over greenhouse gases has also had some success, if the medium term, development assistance de- somewhat more limited. Besides administering voted to education can help upgrade schools to bilateral trust funds to reduce ODS and green- increase the productivity of poor workers. Fi- house gases, the World Bank now finances en- nally, if specific assistance were available it vironmental projects worth several billion dol- might be possible to help stakeholders in devel- lars in developing countries, as do the regional oping countries use intellectual property pro- multilateral development banks. Much more tection to their benefit by protecting intangible could be done. These positive efforts should re- assets such as traditional knowledge, designs, place efforts to use negative instruments such as music, and ethnobotanicals as well as patent trade sanctions and recourse under the WTO, protection for industrial goods. which are likely to be ineffectual and even coun- A second set of policies outside the WTO is terproductive (see box 6.3). to expand global efforts to improve the envi- Finally, more has to be done to make pa- ronment, raise labor standards, and adopt ade- tented drugs available in times of health crises, quate product standards. Environmental pro- such as acquired immune deficiency syndrome tection agencies from all over the world are (AIDS), in a way consistent with incentives to already engaged in a broad range of bilateral invest in research and development. One op- collaborations-joint studies, exchanges, semi- tion: developed country governments, interna- 160 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Box 6.3. Environmental standards and trade E nvironmental standards are at the forefront of NAFTA, and conclude that lax environmental regu- the public debate on trade. The WTO's Technical lations do not create a comparative advantage in Barriers to Trade Agreement and the Agreement an Mexico. Xu (1999), using a gravity model to investi- Sanitary and Phytosanitary Standards both include gate whether differences in environmental regula- some references to environmental protection and tions have affected bilateral trade between a sample trade, although to date there have been few formal of developed and developing countries in pollution- disputes brought before the WTO. intensive goods, found no evidence that countries The links between trade and the environment with stricter environmental standards lower their are complex.6 One effect is that trade can raise scales total exports of pollution-intensive goods. In sum, of production. These effects will be positive because the evidence on the specific linkages between envi- the amount of resources that used to produce the ronmental regulation and trade is mixed. same level of output will decline. However, if trade So what policy tools and institutions are best induces a change in output composition, it is possi- suited to promoting higher levels of environmental ble that dirty industries (even at larger scales) may protection? Trade sanctions to support environmental increase, and clean industries contract, counteracting protection can restrict developing-country market ac- the effects of scale. Trade may also permit greater ac- cess. Indeed they may be counterproductive: since en- cess to more advanced and cleaner technology. The vironmental regulations tend to improve as incomes net effect depends on the change in output mix and rise, policies that restrict trade and restrict growth technology that occurs with trade-induced growth. also undermine a driver of environmental improve- What are the trade consequences of environmen- ment (see World Bank 2001). Second, sanctions pe- tal regulation? One hypothesis is that pollution- nalize whole industries, the clean firms, as well as the intensive industries take flight to countries with lax polluters in an industry. Third, many polluters pro- environmental standards. However, there is limited duce for the local market and are unaffected by sanc- evidence to date to support this hypothesis. For ex- tions. Finally, domestic pollution and environmental ample, Pearson (1987) and Leonard (1988).7 protection can be controlled most effectively they are A second analytical approach considers the envi- targeted at the source-through taxes and other do- ronment as a factor of production, such as capital mestic policy instruments. A more productive ap- and labor. The idea is that countries with lax envi- proach is to establish policy coordination among ronmental regulations (for example, environmental countries. This would allow for joint regulation of abundance) tend to specialize in pollution-intensive common watershed and air basin controls in areas of goods. Here, too, the evidence is ambiguous. For transborder pollution, and for development assis- example, Tobey (1990), looking at five pollution- tance to transfer clean technology and environmental intensive industries in 23 countries, found that envi- aid to strengthen environmental protection over time. ronmental regulations have caused trade patterns to Global environmental agreements (such as the Mon- deviate from the predictions of the model. Wilson, treal Protocol that bans certain ozone-depleting Sewadeh, and Otsuki (2001), in a study of 24 coun- chemicals) and others, if based on sound cost benefit tries with five different pollution-intensive industries, analysis, can raise environmental quality over time. found that stringency of environmental regulation re- Voluntary ecolabeling programs also can provide in- duces net exports of the five pollution-intensive in- centives for environmental protection. dustries. On the other hand, Grossman and Krueger (1993) investigate the environmental impact of Source- Wilson 2001. tional organizations, and foundations could es- countries. These licenses would contribute to tablish a global health fund that could be used an adequate return on research and develop- in part to purchase licenses from developers of ment costs in order to promote new drug de- new drugs and vaccines that are essential for velopment and also permit distribution of the treatment of debilitating diseases in poor drugs to patients at low cost. 161 GLOBAL ECONOMIC PROSPECTS Similarly, collective action to improve labor mote their development. This could be done standards could also contribute to poverty re- if all high-income countries were to emulate duction. Some actions are primarily develop- the EU's "Everything but Arms" preferential mental in scope, such as providing educational scheme. This would provide an impetus to subsidies to ensure that children can attend LDC exports that could increase their revenues school and do not have to enter the workplace by more than 10 percent and the trade from (see Indonesia's highly successful "Stay in Sub-Saharan Africa by some 14 percent. School Program"). Other actions have to do Broadening this access for the 49 least-devel- with the propagation of core labor standards. oped countries to the 70 low-income countries Leadership of these activities are-and should would provide an important impetus to trade- continue to be vested-in the ILO, with pro- led development in those countries that need it ject collaboration from multilateral develop- the most. The effects in trade diversion would ment banks (see box 6.4). be minimal, and the benefits important to the Product standards are becoming increas- low-income countries. If high-income countries ingly important in international trade to pro- were to reduce antidumping threats, the effects tect consumers. However, standard setting can would be even greater. quickly become a ruse for protecting domestic Resources are essential to creating a supply producers. One solution is to create a Stan- response to incentives created through market dards Development Facility to introduce sci- access. Much can be accomplished from debt- ence and other professional evidence into stan- based multilateral flows, but some portion ul- dard setting for products, with adequate timately falls on bilateral developmental assis- representation from developing countries. This tance, often as grants, that can fill in the gaps. Facility could also collaborate with govern- If the high-income countries really wish to see ments to provide unbiased assistance to devel- developing countries become more vigorous oping countries' standard setting-bodies (see participants in global trade, they must make box 6.5). additional efforts to augment extant programs Countries could also undertake a program with trade-related assistance. Bilateral grant of collective action on government procure- aid can help with many aspects of trade facili- ment. The World Bank's Development Gateway tation-customs reform, disseminating techni- may be a vehicle to help countries implement cal standards, and trade law reform, to name a transparent and competitive processes in gov- few areas. ernment procurement of goods and services, an Technical assistance can be as important area where the multilateral development banks as financial assistance. One area where high- have accumulated vast experience. Agreeing on income countries could help immeasurably is key principles, procedures, and policies, supple- competition policy. Simply requiring antitrust mented with provision of technical and finan- authorities to present the structural effects of cial assistance to implement them could go far mergers and acquisitions in home markets on toward encouraging trade, engendering compe- third country markets to be publicly available tition, and augmenting efficiency. would aid authorities in developing countries to enforce competition policy in their own Policies of high-income countries jurisdictions. Moreover as analyzed in chapter A major objective of a new round of trade 4, conducting a regular review of antitrust ex- talks whose rationale is to promote develop- emptions and their adverse consequences for ment must be to lower the barriers to trade in developing countries could be helpful, partic- goods that the world's poor produce and to the ularly in international transport. services they can provide. An important first It would be a mistake to infer that policies in step would be to reduce barriers to trade with developed countries should be designed solely to the low-income countries as an effort to pro- promote trade in developing countries. No less 162 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Box 6.4 Improving labor standards in a way that works Improving labor standards is a fundamental aspect the growth in wages that expanded trade would oth- of development. However many developing na- erwise bring. Historically, the growth rate of wages tions have resisted efforts to include labor standards has been twice as rapid in the developing countries in world trade agreements. Proponents argue that that increased their trade participation in the world trade sanctions should be used to enforce labor stan- economy as compared with those that did not (Col- dards and to raise wages, while developing countries lier and Dollar forthcoming). Moreover, their wage fear that international labor standards could become growth has been even faster than in the rich coun- masks for protection. Their inclusion in the WTO tries, Depriving poor nations of export opportunities threatens a main comparative advantage of develop- in the name of raising wages is fatuous. ing countries. There are other problems with trade sanctions. Numerous studies have shown that low labor Trade sanctions penalize whole countries and indus- standards that affect working conditions do not tries when violators are firms-and often they are grant a competitive edge to developing countries. firms that do not export. Firms serving the domestic According to a 1996 OECD study, countries with market usually have worse labor standards than lower core labor standards (that is, the elimination export industries (Aggarwal 1995). Wages and work- of exploitative child labor, abolition of forced labor, ing conditions in export processing zones, for exam- nondiscrimination in employment, freedom of associ- ple, tend to be higher than the average for the do- ation, and the right to collective bargaining) do not mestic economy. Trade sanctions would in effect have an improved export performance. The study target the better performing export firms. Second, finds no correlation between real wage growth and trade sanctions are an inherently unequal instrument: the degree of respect for freedom of association. they are likely to be imposed only by developed On the contrary, it supports the view that higher countries against developing countries. Finally, trade national income levels and open-market reforms are sanctions can hurt the very people they are intended both associated with improved labor standards. to help. For example, in Bangladesh, children dis- placed from garment factories due to the fear of tWVvNMofageabeteen 19 0and f90s sanctions found alternative employment in activities so with even lower standards, such as street vending and prostitution (Panangariya 1999). Fortunately, the international community has more effective instruments to promote better labor so -standards. A main purpose of the ILO is to promul- gate good labor practices and legislations, and it, rather than the WTO, is far better positioned to lead 10 - -_ international efforts. Governments should be encour- aged to monitor and enforce their own legislation by, if necessary, imposing fines on enterprises that violate 0 g- core labor standards (Elliot 2001). Revenues from a0 rgmmarized auh Post-es these fines could be channeled back into enforcement GWlbae programs and investments to upgrade labor condi- tions. This has several advantages over trade sanc- Soume:coCier and Dogar 2001. tions: violators are punished rather than all firms; rev- enues stay in the country and are used to improve At the same time, trade sanctions to improve standards rather than imposing income losses on labor conditions are likely to be counterproductive, countries; and improvements occur in a manner con- By limiting trade between nations, sanctions shackle sistent with indigenous social values and mores rather 163 GLOBAL ECONOMIC PROSPECTS Box 6.4 (conanued) than according to the dictates of people in rich coun- forcement, and call public and inte-national attention tries. If violations are pervasive and egregious and in- to the most egregious violations (Cereffi and others ternational sanctions are needed, withholding develop- 2001). The international communi-y can help develop- ment assistance is potentially more effective (Torres ing countries improve wages and working conditions, 1996). The role of NGOs is important, too. The ag- but can do so better through the ILO than through gressive campaigns of NGOs have called attention to the WTO. firm violations around the world, and these can help promulgate stricter codes of conduct, encourage en- Source: World Bank staff. important are policies at home to help domestic mains large, and so nominal tariff averages may workers adjust to sudden changes in labor mar- understate the resulting economic distortion. In ket conditions. Since it is impossible to separate the small number of countries where nontariff out trade-related dislocation from technology- barriers continue to be used, elimination of related or "other"-related dislocations, these such instruments should be a priority. Their policies should focus on providing support and conversion into tariffs will generally generate flexibility to workers as they adjust to whatever revenues. As South-South trade is becoming in- forms of shocks to the labor market. creasingly important, developing countries can help themselves through lowering barriers that Domestic policies of developing countries impede access to their own markets. Governments in developing countries do not However, to be effective, reduction in bor- have to wait for international negotiations, der barriers must be accompanied by other other international collective actions, or poli- policies and institutional improvements in the cies in high-income countries to revamp trade investment climate, so that the potentially policies in a way that promotes development. powerful instrument of trade reform results in Country policies still hold the potential for the improved productivity and growth. Weaving greatest gains from trade for most countries. reforms that lower border protection together For this reason, countries and economies as with reforms to elicit a supply response and diverse as Chile, China, Hong Kong (China), promote propoor growth is more complicated and Singapore, as well as Costa Rica and than first-generation reforms. Openness, in Uganda have chosen to reduce tariffs unilater- combination with sound macroeconomic, fi- ally and to use multilateral agreement to legit- nancial, and governance policies, is one deter- imate and lock in the resulting more-efficient minant of sustained rapid growth, which has a price incentives for investors. direct and positive relation to increases in the Many developing countries still have high incomes of the poor (see World Bank 2000b; levels of protection that implicitly tax their and Dollar and Kraay 2001). export and growth potential. Border trade bar- Trade liberalization affects the poor differ- riers continue to be high in three regions- ently depending on the country (see World Africa, the Middle East, and South Asia. Av- Bank 2000b: 49 ff). The immediate effects of erage (unweighted) tariffs in these regions are trade reform on the poor depend (among other 20 percent or higher, nearly double the 10 per- things) on the initial nature of protection, the cent average now found in East Asia, Latin structure of production, the effects of reforms America, and Europe and Central Asia. More- on relative prices, and whether reforms increase over in many countries, tariff dispersion re- the demand for labor (the basic asset of the 164 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Box 6.5 Standards development facility: coordinated action to bridge the standards gap Product standards are a critical part of trade in the and integration of developing nations into the world .I 21st century. These include product standards trading system, a new commitment is needed. Action and sanitary and phytosanitary standards necessary to bridge the rapidly widening divide between devel- for market access in agriculture. Standards are di- oped and developing countries. A two-part strategy rectly linked to poverty reduction and human wel- to meet this challenge over a 10 year period. fare through health, safety, and other channels. The First, the G-8 leaders in Genoa in July 2001 development challenge posed by standards and bor- committed to better coordinate trade-related assis- der barriers are particularly important to future tance to "provide bilateral assistance on technical trade prospects of the least developed nations (World standards" and stressed the "paramount importance Bank 2001). of food safety." A new Standards Development Facil- The costs of barriers in standards are likely ity (SDF) could be created to move these commit- much higher than tariffs to global trade (Maskus and ments forward. The program, to be coordinated in Wilson 2001). Moreover, testing, and certification cooperation with the WTO, the Bank, and other requirements in global market remain a serious oh- multilateral institutions and bilateral donors, would stacle for developing countries seeking to expand ex- develop the framework for a long-term assistance ports. The OECD estimates that standards alone rep- plan to (1) expand access to standards for developing resent between 2 and 10% of final product costs countries, and (2) facilitate modernization of stan- (OECD 1999). Changes in product standards can dards infrastructures. This work could start with the have serious repercussions for developing country ex- PRSP countries and build upon pilot programs in porters. World Bank research and operational experi- coordination with the G-8. ence indicates that standards are today one of the The second goal of the SDF would center on fundamental "behind the border" barriers to poverty promoting trade expansion through regulatory re- alleviation through trade (Wilson 2000). form and removal of technical barriers in discrimina- Most developing countries do not have the re- tory standards, testing, and certification regimes. sources to apply standards. In Guatemala, for exam- This work is in the long-term economic benefit of ple, the total budget for standards in 2000 totals both the developed and developing countries. A $119,000 (Hufbauer, Kotschwar, Wilson 2001). This framework to promote the wider use of "supplier's represents a small fraction of the total government declaration of conformity" to regulatory require- budget. The World Bank's experience with standards ments should be developed. A systematic review of in the 1990s shows that investments of $ 3.5m (Viet- products subject to mandatory government testing nam) $155 million (Turkey), and $5 million in Mo- and certification that can be moved to declaration of rocco, for example, had to be undertaken to simply conformity status should be undertaken. A multilat- begin the process of modernization. The Bank along eral "Global Conformity Agreement" (GCA) could with other multilateral institutions and bilateral then be developed based on this list for international donors support standards-related projects in Africa negotiation and agreement (Wilson 2001). Funding and elsewhere. The international community needs, to support standards modernization and capacity however, a coordinated and sustained effort to 1ev- building in the least developed countries, as outlined erage work now done in an ad hoc fashion. A new above, must be part of this goal. commitment to action is needed that compliments the WTO agenda and trade negotiations. In order to secure the benefits that market-driven standards offer Source. World Bank staff. poor). For example in cases where the poor pri- ative prices, but when the poor work mainly marily produce for export or rely on imports in import-competing sectors, trade liberaliza- for consumption, lowering tariff barriers can tion can cause dislocation that adversely affects improve their situation through changes in rel- them. That trade reforms can produce increases 165 GLOBAL ECONOMIC PROSPECTS in income and, in the long term, offset negative around 3.6 percent on per capita terms, about effects offers little solace to those poor suffering 1.1 percentage points above the per capita transitional costs. For these reasons, trade re- growth rate of the high-income countries, with form programs have to identify the effects of re- the highest growth rates anticipated in Asia. forms on the poor, design targeted compensa- The countries of Eastern Europe and Central tion where possible, and build propoor social Asia were expected to grow quite rapidly in the protection into Poverty Reduction Strategies of next decade, while Africa and the Middle East low-income countries and development pro- would revert to modest growth rates. This per- grams of middle-income countries. For least- formance over the 2000-15 period expands in- developed countries, because much analytical come by nearly 60 percent-some $18 trillion and capacity-building work remains to be un- (in 1997 dollars). dertaken, donors have agreed to adopt an In- The baseline scenario establishes a path of tegrated Framework for the Least-Developed growth against which to assess the effects of Countries (see box 6.2). eliminating trade barriers. It is important to bear in mind that the baseline incorporates only those changes to the global trading regime Envisioning alternative futures up through 1997 carried forward to 2015. The M aking these changes in global trade key policy change to be simulated is the phased architecture requires political leadership elimination of all import tariffs, export subsi- around the globe and within countries. Policy- dies, and domestic production subsidies. These makers will ask: Are the benefits worth the are modeled to begin in 2005 and last through political effort? How will it affect poverty and 2010. Said differently, in each year between income distribution? 2005 and 2010, restrictions are reduced by Answering these questions in a quantitative one-sixth from their initial levels. The struc- sense poses challenges. Trade is only one of tural transformation therefore starts in 2005, many factors affecting the long-term prospects and it has five years to complete (2011-15) of developing countries. Chapter 1 presented a after full removal. In reality, of course, this type discussion of the long-term growth dynamics of policy change would not come about only of developing countries in the world economy. through multilateral negotiations; but a devel- This constitutes a baseline view about the likely opment round, together with regional agree- evolution of developing countries, based upon ments8 and unilateral domestic policy reforms, best guesses about generally stable parame- could well advance policy toward this frame- ters-savings, investment, population growth, work by 2010. trade, and productivity growth. To distinguish Two versions of the trade reform scenario are the effects of changes in trade policies, we then presented. In the first version it is assumed that simulate the removal of trade restrictions dis- trade reform has no impact on productivity. cussed in the foregoing chapters, and analyze These are the static gains.9 For the most part, their development consequences as measured the source of these gains comes from reducing against the baseline scenario. Although econo- the economic inefficiencies linked to trade pol- mists' ability to measure these changes is lim- icy distortions. These may have some dynamic ited (for reasons discussed below), the effort is impacts as they change savings and investment intended to give us some idea of relative mag- outcomes. The gains are sometimes counter- nitudes of effects. acted, partially or even completely, by changes in a nation's terms of trade.o (Annex 1 below Assessing the effects of trade openness: provides summary information on the design of a fast-integration scenario the simulations and the underlying model.) In chapter 1 recall that we had indicated that The second version entails dynamic gains. income growth in low- and middle-income It assumes that productivity is a function of countries under the baseline scenario will reach the degree of openness of the economy. While 166 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT much more work needs to be done in this area, The introduction of a link between open- the evidence available to date suggests a clear ness and productivity increases the static gains link between openness and productivity. This described above by a factor of over two, with has been implemented in the model assuming a the global gains jumping to over $830 billion direct relation between productivity growth using our base-case parameters. As a percent- and a measure of openness. Productivity age of the global gains in 2015, developing growth is linked to the export-output ratio countries do much better, improving their using a constant elasticity function.1' There are share from 52 percent to 65 percent. The gains several potential channels through which this as a percent of initial income rises to almost 5 mechanism operates. As firms' exports grow percent for developing countries, and repre- and they increase their penetration of world sents significantly higher gains than observed markets, they learn new technologies (through in the static version. The relative gains will be comparison with their competitors' products); highest in countries observing the greatest rise they improve production processes to match in their export-to-output levels. Typically, these international standards (such as safety, health, will be countries with either high initial tariffs, packaging, style, and others); and they can inducing a large shift in both imports and benefit from scale economies as they produce exports; or countries facing large tariffs and for a larger market.12 There are other channels able to increase market penetration; or both.15 through which openness could impact on pro- These results are broadly comparable in simi- ductivity that are not incorporated here. The lar studies (see box 6.6) key channels are imports of technology-laden intermediate imports or capital goods, or both. Agriculture provides the greatest Impacts on real incomes. Measured in static opportunity terms, world income in 2015 would be $355 The gains from further opening of the global billion more with trade liberalization than in economy can be decomposed in a number of di- the baseline (figure 6.2).13,14 Measured in dy- rections. Table 6.1 illustrates the sources of the namic terms, this would translate into cumu- income gains from two angles-regional and lative additional income of $1.5 trillion to de- sectoral. Along the regional angle it shows that veloping countries over the 2005-15 period the greater source of income gain among de- (valued at net present value in 2005). veloping countries is from their own reforms. About 48 percent of total gains accrue to the Thus full trade reform by developing countries high-income countries, with Western Europe generates an income gain of $121 billion (in garnering the highest static gains at $83 billion. the static) simulation, some two-thirds of their This largely reflects its highly distorted agricul- gains from global trade reform. Needless to say, tural policies, which not only are costly for full market access by the high-income countries European consumers and taxpayers, but also leads to gains for developing countries of nearly place a burden on more competitive farmers $125 billion if productivity gains are taken into outside the EU, who face lower world prices account. due to these distortions. Developing countries The sectoral decomposition is similarly illu- as a whole would benefit from a rise in real in- minating. Reflecting the high distortions in comes of 1.6 percent in the final year 2015, agriculture, the largest gains from merchandise compared to baseline income (figure 6.2). trade liberalization are to be realized from There are wide variations in the income gains eliminating all forms of agricultural protection. across developing regions reflecting two oppos- In both the static and dynamic simulations, ing forces. On the one hand, removal of tariffs agricultural reform in itself accounts for 70 leads to greater economic efficiency and higher percent of the global gains. Free market access growth. By contrast, terms-of-trade losses can in the high-income countries could result in partially counteract the gains from efficiency gains to developing countries of as much as improvements. $100 billion. On a lesser scale, but nonetheless 167 GLOBAL ECONOMIC PROSPECTS Figure 6.2 Developing countries could reap income gains of over $500 billion -which implies up to a 5 percent boost from full trade liberalization in incomes (billions of 1997 dollars, additional income in 2015 (percent of baseline income in 2015) compared with baseline) 900 5 S Static 0 Dynamic - Static O Dynamic 800 - 700 4 600 3 500 400 2 300 200 100 O0 0 Low- and middle- High-income World total Low- and middle- High-income World total income countries countries income countries countries Note: Static gains refer to the results holding productivity constant. Dynamic gains allow productivity to respond to sector-specific export-to-output ratios. Source: World Bank model simulations. far from trivial, elimination of existing protec- export-to-output ratio). The second is the share tion on textiles, clothing, and footwear would of total sectoral productivity affected by open- generate global income gains ranging from $40 ness in the baseline simulation.21 The estimate billion, in the case of fixed productivity, to al- of the gains in the baseline simulation corre- most $190 billion with endogenous productiv- sponds to a productivity elasticity of one and a ity. Both of these sectors, as reflected in chapter share of 40 percent, that is, $832 billion. The 2, harbor the larger share of the working poor first column represents the static gains-an in developing countries. elasticity of 0. As one would anticipate, the gains increase as both parameters rise. Since Results are sensitive to assumptions trade reform typically increases the export-to- While there is little doubt regarding the pro- output ratio, an increase in the responsiveness ductivity-enhancing impacts of greater open- of productivity to this ratio will increase the ness, there has been relatively little economet- openness-related productivity results. Similarly, ric analysis at either the macro or micro level the larger the share of productivity accounted to determine more precisely the magnitude of for by the openness factor, the greater will be the relation. In light of this uncertainty, table the impact on growth.22 6.2 illustrates some potential range of the global impacts of full trade liberalization with Service sector liberalization varying assumptions regarding the key para- The liberalization scenario described so far meters in the relation between openness and has concerned only the goods sectors. How- productivity. ever, chapter 3 of this volume clearly illus- Two parameters determine the relation. The trates the importance of liberalizing the ser- first is the responsiveness of sectoral productiv- vice sectors. With details to follow below, we ity to sectoral openness (as measured by the conservatively demonstrate that, for develop- 168 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Box 6.6 The complexities of measuring openness and growth S tudies of the relation between openness and trade openness can affect growth. While the early growth have followed two main lines. One line of models essentially estimated the static inefficiency research has been to estimate econometrically the re- losses from imposing tariffs and other trade barriers, lation between openness and growth using cross- more recent models have extended the analysis along country time series data and panel estimation tech- four main research paths-dynamic accumulation of niques (see, for example, Sachs and Warner 1995, static gains, allowing for imperfect competition and and Dollar and Kraay 2001). These studies conclude scale economies, endogenous growth in which pro- that there is indeed a strong link between openness ductivity is responsive to trade openness, and en- and growth.16 A second line of research has been the dogenous capital flows models in which capital is re- development of increasingly sophisticated general sponsive to trade opening.17 All of these magnify the equilibrium models. Model and data development measured static efficiency gains by a factor of two to have focused on the various channels through which four, depending on the methodology. (a gains are in billions of dollars) Type of Gain for Gain for Study Nature of reform smulation Base year industriat countries developing countries World total GKVa Full pre-UR trade Dynamic 1992 290 160 450 liberalization AFHHMb Full pre-UR trade Steady-state 1995 146 108 254 liberalization GEP Full trade Dynamic 1997 171 184 355 liberalization (from 1997 base) Dynamic 1997 293 539 832 w/productivity DFSd Full trade Dynamic 1995 450 760 1210 liberalization w/productivity (from 1995 base) DFATI Full trade Static 1995 750 liberalization (from 1995 base) including services BDS' Full post-UR trade Static 1995 1490 370 1860 liberalization including services Notes: a. Goldin, Knudsen and van der Mensbrugghe (1993). b. Anderson, Francois, Hertel, Hoekman and Martin (2000). c. Global Economic Prospects (2002). d. Dessus, Fukasaka, and Safadi (1999). e. Department of foreign Affairs and Trade, Australia (1999). f. Brown, Deardorff and Stern (2001). (ctid) 169 GLOBAL ECONOMIC PROSPECTS Box 6.6 (continued) The box table summarizes several studies using Only two of the studies cite here have some applied general equilibrium models.18 The differ- form of endogenous growth-the GEP and Dessus ences in results can be essentially explained by three and others (DFS 1999) studies, Tae DFS model as- factors: (a) the base from which the reform is simu- sumes a relation between aggreg.te openness (as lated, together with its assumptions about initial lev- measured by the ratio of exports plus imports to els of trade barriers (for example, pre- or post- gross domestic product-GDP). -The GEP model as- Uruguay Round); (b) whether productivity is fixed or sumes that the economic response to opening is responsive to trade opening, and (c) whether service sector-specific. As a result, in the GEP study, produc- trade liberalization is included. tivity increases are limited to the agriculture and The first three studies--Goldin and others (GKV manufacturing sectors, thereby capping to some ex- 1993), Anderson and others (AFHHM 2000), and tent the additional gains from introducing the open- Global Economic Prospects 2002 (GEP) (that is, the ness productivity link.19 study you are reading), use similar model specifica- Finally, the table in this box shows the impacts tions and estimate the long-term efficiency gains of including service trade liberalization in the com- from full merchandise trade liberalization. The ag- plete package.20 The Departmen of Foreign Affairs gregate gains from reform are roughly similar across and Trade (DFAT 1999) study slows a gain of $750 these three studies, and the differences can largely be billion, and a tripling of the gains of merchandise explained by the nature of the reform scenario. The trade reform (compared with AFHHM); and the first estimates the pre-Uruguay Round (UR) impacts Brown and others (BDS-2001) study shows a mas- of full trade reform, the second the post-UR impacts, sive gain of nearly $1.9 trillion. The study incorpo- and the third the impacts of trade reform before rates scale economies and imperfect competition that complete implementation of the UR. could readily explain the differences in results. ing countries, the income gain from service studies, not surprisingly, show the tremendous liberalization amounts to multiples of the gains potential gains from liberalizing the services from merchandise trade reform. There are sev- sectors, gains that are multiples of merchan- eral reasons for this. First, services play a dise trade liberalization. growing role in all economies as they develop, Rather than relying on imprecise estimates both from the point of view of the consumer, of the barriers to services delivery, results pre- as well as their importance as inputs into an sented below provide a very conservative esti- efficient modern economy. Second, liberali- mate of the potential gains using the same zation of services has lagged far behind liber- model used to assess merchandise trade liberal- alization of goods, where average tariffs are ization, though in a simpler context. The frame- today generally low. work is completely static using the 1997 base.23 However quantification of services sectors' The services sectors were disaggregated into six trade barriers and other forms of protection is categories:24 (1) trade and transportation; (2) still more art than science. Even the more communications; (3) financial services (includ- straightforward accounting of bilateral flows ing banking and insurance); (4) other private and the value of sales of foreign affiliates in services, including legal, accounting, accommo- the services sectors is sketchy, at best. Several dation, and restaurants; (5) public services; and efforts have been undertaken to measure the (6) housing. The scenario assumes that reforms barriers and assess the impacts of their re- are undertaken in four of these sectors-ex- moval. Two global studies are cited in box cluding public services and housing. 6.6, and the Tunisian case is developed more The barriers in the services sectors, as im- thoroughly in box 3.3 in chapter 3. These plemented in this framework, take three 170 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Table 6.1 Agriculture accounts for the bulk of the gains from merchandise trade liberalization (billions of 1997 dollars, additional income in 2015 as compared with baseline income) Simlations with fiaed pmduetivity Agricultre Tetle and food and clothing All other sectors Total Liberalizing region: Benefiting region: High-income High-income 73 -3 -25 49 Low- and middle-income 31 19 26 75 Total 104 16 1 124 Low- and middle-income High-income 23 20 78 118 Low- and middle-income 114 7 -5 121 Total 136 27 73 239 All regions High-income 106 17 50 171 Low- and middle-income 142 24 20 184 Total 248 41 70 355 Sinulations with endogenous productivity High-income High-income 144 -10 12 149 Low- and middle-income 99 20 7 124 Total 243 10 20 273 Low- and middle-income High-income 53 78 22 151 Low- and middle-income 294 104 21 424 Total 346 182 43 575 All regions High-income 196 66 35 293 Low- and middle-income 390 123 27 539 Total 587 189 62 832 Source: World Bank model simulations. Table 6.2 Global gains are sensitive to forms.25 The first is a cost penalty measuring productivity-openness linkages the relative inefficiency of firms operating as (billions of 1997 dollars) monopolies or otherwise protected from com- Ilasiiy petition. The second is a price markup over (Share percent) 0.0 0.5 10 1s average cost, representing the ability of firms to price to market in the absence of competi- 20 355 435 178 674 tion (be it domestic or foreign). The third cap- 40 355 515 832 1,026 60 355 596 933 1,174 tures barriers to cross-border trade.26 The bar- 80 3ss 677 1,031 1,34o riers were set at conservative levels. Both the Note: Sectoral productivity is decomposed into two factors. cost and trade penalties were set at 10 percent, The first is sensitive to an openness indicator represented by and the markup was also fixed at 10 percent. the sectoral ratio of exports to output. The second is a resid- ual determined by other factors. In the baseline simulation, (By comparison, in the Tunisian study cited in the trade-sensitive factor is calibrated so that its share in de- chapter 3 [Konan and Maskus 2000], barriers termining total sectoral productivity is fixed. The sensitivity analysis shows how the gains vary with respect to this share, in these same sectors varied from 3 to 200 per- respectively 20, 40, 60, and 80 percent. The table also shows the sensitivity of the aggregate gains with respect to the elas- cent, with most ranging from 30 to 50 per- ticity linking trade openness with productivity. cent.) The results presented below are limited Source: World Bank model simulations. to services liberalization in developing coun- 171 GLOBAL ECONOMIC PROSPECTS tries only; that is, there is no assessment of the wages and returns to capital. In many ways the impact of high-income country reform on de- markup is similar to a producer tax.29 In that veloping countries. sense, one would not anticipate that a reduc- The results are telling. So-called joint re- tion in the markup would lead to the same form, where all three instruments are relaxed boost as an improvement in efficiency. 30 More- simultaneously, yields an incremental income over, similar to a tax, elimination of a markup gain for developing countries of nearly $900 can also produce perverse results if it leads to billion, some 4.5 times greater than their gain increasing losses due to other inefficiencies (so- from global merchandise trade liberalization called second-best effects). Finally, the impacts alone, or $190 billion (table 6.3).27 In total, of the trade barrier instrument are also signifi- this represents a 9.4 percent income gain com- cantly smaller than the efficiency gains. In part pared to base levels.28 this reflects the low level of penetration of cross- Table 6.3 also reveals the decomposition of border trade in services. The long-run potential the "joint reform" into its three components. It would be much larger. is clear that reducing the cost penalty has the Table 6.3 also reports the decomposition of greatest impact. It is equivalent to shifting the the gains by category of service. The source of production possibilities frontier outwards (in the largest impact is the trade and transporta- the four service sectors). Though the markup tion sector, which accounts for roughly double allows for some decline in the producer price, of the aggregate gains. This largely corresponds it will be attenuated to some extent by higher to this sector's share in demand (compared to Table 6.3 Services liberalization generates substantial windfall gains for developing countries Goods trade Cost Markup Trade Joint alone penalty removal penalty reform Total Static income gain for developing countries (SI97 billion) Reforming sector Merchandise trade 190.0 All four service sectors 800.4 27.5 34.4 883.5 1073.4 Trade and transportation 443.0 7.9 26.0 477.7 667.6 Communications 39.0 1.1 1.3 41.4 231.3 Financial services 96.1 8.1 4.0 108.5 298.4 Other private services 209.0 6.4 231 235.6 425.5 Static income gain for developing countries (percent of base income) Merchandise trade 1.7 All four service sectors 7.0 0.2 0.5 7.7 9.4 Trade and transportation 3.9 0.1 0.2 4.2 5.8 Communications 0.3 0.0 0.0 0.4 2.0 Financial services 0.8 011 0.0 0.9 2.6 Other private services 1.8 0.1 0.2 2.1 3.7 Note: Though the results come from a comparative static simulation with a 1997 base, in order to make them comparable with pre- vious results, they have been scaled by the projected income of 2015, For example, the merchandise trade gain of $190 billion is equivalent to $84.2 billion when scaled to 1997 income. The results as a percent of base income are invariant to the choice of re- porting year. The first column represents the gains from merchandise trade liberalization only (in the comparative static framework). The next four columns represent the incremental gains from services liberalization, that is, those gains on top of the gains from mer- chandise trade liberalization. The "Cost penalty" column reports the impacts of a 10 percent increase in efficiency in the four private services sectors. The "Markup removal" column reports the impacts of the removal of a 10 percent markup in the same four sectors. The "Trade penalty" column reports the impacts of reducing the trade penalty parameter by 10 percent. And the fifth column, "Joint reform," represents the incremental impact of implementing all three reforms simultaneously. The final column represents the total gains: "Joint reform" added to merchandise trade liberalization. The instruments are only applied in developing countries. Source: World Bank model simulations. 172 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT the other three service sectors). For example, the big number, policy makers, businesses and on average in developing countries, input of the general public are often concerned about trade and transport services accounts for 6.6 the more direct impacts to specific segments of and 7.4 percent of output in the manufacturing society-who is likely to benefit and who may and services sectors, respectively. And trade and be hurt. The next sections shed some light on transport accounts for nearly 22 percent of pri- the more detailed economic impacts of mer- vate consumption. Other private services, the chandise trade reform-leaving the effects of next largest sector, has only 3.2 percent cost any service sector liberalization in abeyance. share in manufacturing and a 7 percent share in Four headlines are noteworthy. First, trade private consumption.31 reform tends to improve income distribution Despite the somewhat tentative nature of toward greater equality. Second, it leads to these results, they clearly illustrate the impor- sharp reductions in poverty. Third, the major- tance of services liberalization for developing ity of economic sectors tend to expand in the economies. They also illustrate the need for wake of reform. Finally, there is a significant significantly more research in the area of ser- expansion in trade, particularly in agriculture vices-both in fundamental data gathering, as and textiles, two of the most protected sectors well as in improving our knowledge of the in the global economy. economic mechanisms through which protec- Income distribution. The current frame- tion in the services sectors operate. work, although aggregate in nature, can eluci- date some of the underlying factors determin- Consequences for inequality and poverty ing income distribution, notably factor returns While all too frequently the focus of trade re- and structural changes.32 Table 6.4 presents forms is on the aggregate economic impact, i.e. the final year-impacts on real factor returns Table 6.4 Labor's share of national income rises substantially (percent change in real factor returns in 2015 as compared with baseline) With exogenous productivity With endogenous productivity Capital Unskilled Skilled Capital Unskilled Skilled returns wages wages returns wages wages High-income countries United States 0.1 0.5 0.4 0.4 0.5 1.0 Western Europe 0.1 0.7 2.6 0.7 0.7 3.1 Japan 1.2 1.5 2.7 2.2 1.9 3.4 Other high-income OECD countries 0.8 3.1 0.6 0.7 2.3 1.3 Newly industrialized economies 0.5 4.1 2.9 -0.4 4.1 3.8 Low- and middle-income countries Sub-Saharan Africa 0.8 6.9 4.5 3.4 5.4 6.8 East Asia and Pacific 4.6 6.2 7.8 9.3 11.2 15.0 South Asia 1.7 6.0 3.4 3.7 5.7 5.8 Eastern Europe and Central Asia 1.4 5.4 4.3 3.3 5.3 6.7 Middle East and North Africa 8.0 4.1 12.5 10.9 6.1 17.0 Latin America and the Caribbean 0.7 5.3 2.5 1.4 4.8 4.3 Rest of the world 0.8 3.3 2.2 3.2 2.7 4.2 Memorandum items High-income countries 0.2 1.0 1.6 0.7 1.1 2.2 Low- and middle-income countries 2.7 5.7 5.6 5.1 7.4 9.6 World total 1.0 2.3 2.5 2.1 2.8 3.8 Note: Nominal factor prices deflated by economywide CPI. Source: World Bank model simulations. 173 GLOBAL ECONOMIC PROSPECTS (as percentage changes from the baseline). Un- Structural transformation. Removal of skilled wages improve more than skilled wages, trade barriers has multiple structural implica- and capital returns in all of the developing re- tions-changes in the composition of produc- gions, except for East Asia and the Middle East tion, changes in trade-to-output ratios, and so and North Africa, and in some cases quite sub- on. While in aggregate these changes are highly stantially, for example in South Asia and Latin positive, they could cause significant displace- America. This suggests quite strongly that pro- ment, and potentially some losers. One of the tection has largely been detrimental to unskilled reasons trade reforms are difficult to imple- workers, including, of course, those working ment is that the potential losers are easy to in agriculture.33 With endogenous productivity identify (and quick to alert and influence poli- the relative gains of unskilled workers is some- cymakers) whereas the gains are more diffused, what dampened. Since the additional produc- and devoid of organized partisanship. Even if tivity is only labor-augmenting, capital becomes the losses are small compared to the gains, the relatively scarcer with endogenous productivity, political weight of protected sectors can, in thereby raising its relative return, and could many cases, impede improvements in policies. potentially reverse the trend toward improved Figure 6.4 reflects the aggregate losses in value income distribution, although this will in part added compared with the net aggregate gains. depend on the share of capital income in aggre- For most regions, the negative displacement is gate income. small relative to the aggregate gains. One of Poverty. Rising unskilled wages, as pre- the exceptions is Western Europe, where the sented above, are likely to lead to a decrease in value added losses, particularly in agriculture poverty. When coupled with changes in the and food processing, are much larger than the price of the poor people's consumption bas- overall gains. Sub-Saharan Africa, South Asia, ket, the reduction in poverty could be quite and the Middle East and North Africa regions substantial. Figure 6.3 presents the "food and also face relatively high negative displacement clothing" wage for unskilled workers in devel- compared with the aggregate gains. These are oping countries.34 The largest increase in real the regions with the highest distortions, and unskilled wages occurs in the Middle East and therefore are subject to the greatest structural North Africa region, but all developing re- transformation. However, on average for de- gions benefit from a substantial rise. The changes veloping countries, the displacement represents in the real wages of unskilled workers (de- only 23 percent of the total gains. flated by the food and clothing index) can be Trade. In the baseline scenario and in the ab- applied to the forecasts of poverty headcounts sence of any trade policy change, aggregate for the year 2015. Assuming an elasticity of world trade of goods and services would rise two, a standard assumption for these types of above $11.2 trillion (table 6.5). Market pene- analyses, figure 6.3 shows the implication on tration of developing economies in high-income poverty of the rise in unskilled real wages. countries would rise to 32 percent, a rise of 5 Overall dire poverty (those living on less than percentage points from its level in 1997. Under $1 per day) would fall by over [110] million un- the openness scenario, world trade would in- der these assumptions, some [15] percent below crease by an additional $1.9 trillion in 2015, an the baseline forecast for 2015. Sub-Saharan increase of 17 percent from baseline levels. Africa would account for over one-half of the Developing-country market penetration would improvement. Poverty would decline by over rise to 37 percent in the high-income countries, 320 million persons based on the $2 per day reflecting an increase of 26 percent in the value criteria, with the largest absolute improve- of exports from developing countries to the ments in Sub-Saharan Africa and South Asia. high-income countries. More impressively, This represents a 15 percent decline in poverty South-South trade would jump 59 percent, an globally. increase in value by over $700 billion. 174 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Figure 6.3 Unskilled wages rise substantially relative to cost of living- Percent change from baseline in 2015 18 16 14 12 10 8 6 4 2 0, Sub-Saharan East Asia South Asia Eastern Middle East Latin America Africa and Pacific Europe and and and the Central Asia North Africa Caribbean Note: Figure represents changes in unskilled wages deflated by the food and clothing CPI. Source: World Bank model simulations. -implying a substantial reduction in poverty Number of persons (in millions) compared to baseline in 2015 120 E $1 per day O $2 per day 100 80 60 40 20 0 F Sub-Saharan East Asia South Asia Eastern Middle East Latin America Africa and Pacific Europe and and and the Central Asia North Africa Caribbean Note: Figure represents potential reductions in number of persons living in poverty. A poverty elasticity of 2 with respect to the improvement in unskilled real wages is applied to the baseline poverty scenario for 2015. Source: World Bank model simulations. The sectoral composition of the change in nearly $180 billion. Reflecting tariff escala- trade is equally revealing (figure 6.5). Except tion in the food-processing sector, developing- for energy and the nontradable sectors (con- country exports jump 139 percent. Note that struction and services), developing-country high-income exports of food processing also exports in all sectors increase sharply, particu- expand considerably. larly in percentage terms from baseline levels. This outcome reflects several factors. First, Agricultural exports expand by $200 billion, tariffs in this sector are high around the world, and textile, clothing, and footwear exports by so industrial-country exporters are able to take 175 GLOBAL ECONOMIC PROSPECTS Figure 6.4 Reform has costs, but they are largely outweighed by the gains Change in billions of 1997 dollars compared to baseline 600 0 Displacement 0 Value added 500 400 300 200 100 Sub-Saharan East Asia South Asia Eastern Middle East Latin America High-income Africa and Pacific Europe and and and the countries Central Asia North Africa Caribbean Note: The displacement bars represent the sum of the difference in value-added in sectors with declining value added. The value- added bars represent the aggregate change in value added. The figures are from the endogenous productivity scenario. Source: World Bank model simulations. Table 6.5 Developing countries increase the costs of inputs for food processors, making their market share them more competitive internationally. (trillions of 1997 dollars in 2015) Exports of other manufactured products by Low- and developing countries represent the largest ab- Importing region High-income middle-income World solute increase. A significant portion of the in- Trade fows in baseline seni crease represents an increase in South-South trade where barriers to manufactured imports Exporting region are high in the baseline, compared with barri- High-income 5.1 2.4 7.6 ers to trade in the industrial countries for these Low- and middle-income 2.4 1.2 3.6 same products. World total 7.6 3.6 11.2 Trade flows with endogenous productivity Conclusions High-income 5.1 3.0 8.1 Launching a development round, moving for- Low-and I9ward on the global cooperation agenda to ex- middle-income 3.0 1.9 5.0 World total .2 4.9 13.1 pand trade, enacting policies in high-income countries to promote trade-led development, Source: World Bank model simulations, and enacting trade reforms within developing countries are all momentous tasks. But the long-term promise is tangible: $2.8 trillion in advantage of new opportunities. Second, with additional global income, $1.5 trillion of ad- decline in protection in their own markets, ditional income for developing countries, re- producers shift toward producing for export ductions in global poverty by an additional markets. And third, the decline in agricultural 320 million people, and fewer infants dying protection in high-income countries reduces before their fifth birthday. This, in turn, would 176 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Figure 6.5 World trade booms, particularly in food and agriculture (billions of 1997 dollars in 2015) 700 600 | = High-income countries r_- Low- and middle-income countries 3 500 400 300 200 - 87 r11 48 52 100 1 4 0 6 00 Agriculture Energy Processed Textiles, clothing, Other Non-tradables foods and footwear manufacturing Note: Number above columns represents percent increase in exports from baseline level. These represent the results of openness with endogenous productivity. Source: World Bank model simulations. Box 6.7 World Bank programs: activities to support trade-led pro-poor growth T he Bank, usually in partnership with other enti- and the European Union agreements, is especially ties, is working to help developing countries important. create-and take advantage of-new trade opportu- Finally, and most important, at the country nities. It is doing so in three policy domains: global, level, work on traditional border barriers remains a regional, and national, priority, particularly for countries in South Asia, the At the global level, developing countries, more Middle East, and Africa. At the same time, virtually important in size and sophistication than ever before, all countries are paying increased attention to "be- are now pivotal to the success of the world trading hind the border" issues-for example, investment system. Their interests have to be taken into account regulations, transportation infrastructure, trade facil- if any new multilateral trade negotiations are to be itation, telecommunications, and business services- successful and if the multilateral system is to be to ensure that producers can take full advantage of strengthened. The Bank's objective is to help devel- the opportunities globalization presents. This part of oping countries use the system of multilateral rules the new trade agenda may partially overlap with ex- to expand their trade and development. In particular, tant sectoral reform initiatives, and in these cases, a the Bank is focusing intensively on the barriers facing challenger is to ensure consistency between trade- least developed countries (LDCs) in using trade to related objectives and the other objectives of sectoral promote development. reforms. In all cases, the Bank's goal is to help gov- Regional arrangements are becoming increas- ernments design and implement pro-poor reform ingly important for trade policymakers in the de- programs that can leverage trade into faster growth veloping world. The Bank is focusing on analyzing and poverty reduction. Of particular importance is their effects, on helping governments to shape the Integrated Framework effort, a multilateral ini- arrangements so that they expand trade and be- tiative designed to help least developed countries re- come steppingstones to more effective multilateral spond to market opportunities and accelerate their participation, and on advising prospective mem- integration into the multilateral system. bers about costs and benefits. Understanding the In 2000, the Bank presented 46 projects to its effects of the largest arrangements, such as the Board with trade components, and was undertaking 35 proposed Free Trade Arrangement of the Americas studies in addition to the IF work-to advise clients. 177 GLOBAL ECONOMIC PROSPECTS contribute to a more sustainable standard of tainty regarding their levels. Systematic sensi- living around the globe-and a more stable tivity analysis is desirable to determine the ex- world community. tent to which the impact analysis is robust to changes in these parameters. This in itself is far from a trivial task, given the thousands of pa- Annex 1 rameters these models typically employ. A pplied general equilibrium (AGE) model- The closure rules pertain to the actions of ing, in some form, has been the tool of certain agents that are not modeled explicitly, choice for trade economists to analyze the im- or are exogenous to the model. There are three pacts of multilateral trade reforms for over two key closure rules in the simulations undertaken decades, starting with analyses of the Tokyo for this study.35 First it is assumed that gov- Round in the late 1970s and early 1980s (Cline ernment expenditures are fixed in real terms. and others 1978; Deardorff and Stern 1981; In the baseline scenario, they grow at the same and Whalley 1985). Their development took rate as real GDP; in policy simulations they are off with the rising accessibility of computing unchanged from their baseline levels. Govern- power and improved software, and have be- ment revenues are raised to achieve a targeted come increasingly more sophisticated, integrat- level for the fiscal deficit. The latter is held ing aspects such as dynamics, market structure fixed at its base level in order to avoid sustain- (for example monopolistic competition), and fi- ability concerns.36 The direct tax schedule ad- nancial flows. AGE models proved to be influ- justs to insure fiscal balance equilibrium. In the ential in the last round of multilateral negotia- case of trade reform, this implies that the re- tions, which culminated in the Uruguay Round duction in import tax revenues is replaced by Agreement signed in Marrakech in 1994 (see direct taxes (to the extent that revenues from Martin and Winters 1996, for example). other sources of taxation are not significantly AGE models capture the detailed interac- altered). tions across the many agents of an economy- The second closure rule concerns invest- producers, consumers, public entities, investors, ment. Investment is assumed to be savings- importers, and exporters. Despite their level of driven, for instance, there is no interest rate detail, they nonetheless represent a stylized rep- mechanism that equilibrates the savings supply resentation of a true economy. For example, and investment demand schedules. Foreign the version of the model used for this volume saving can add to or subtract from domestic represents economic activity by only 20 goods saving. Trade reform may have little impact on and services. A detailed domestic model may overall domestic savings to the extent that it have 100 to 200 sectors. would do little to modify consumers' choice The results of the model depend on two key between current and future consumption. How- sets of parameters and the so-called closure ever to the extent that the price of investment rules. The first set is the dynamic parameters- goods decline (due to the removal of tariffs on population and labor force growth rates, edu- capital goods), investment could rise substan- cation, savings behavior, and technological tially with positive long-term payoffs. In other progress (or productivity). The second set of words, the amount of investment per dollar key parameters includes the behavioral and saved has very positive dynamic impacts if tar- technological parameters of the economic iffs impose a high cost on capital goods. agents. How do consumers respond to price The final closure rule concerns foreign cap- changes? How do household budgets change as ital flows. In the absence of endogenous deter- incomes rise? How flexible is production? Can mination of foreign capital flows across coun- labor substitute for capital, or vice versa? While tries, these are assumed to be exogenous in any many of these parameters are econometrically given time period.37 Thus policy shocks are estimated, there is still a great deal of uncer- transmitted to a fixed trade balance, the re- 178 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT verse side of a fixed capital account balance, regional integration agreements have been formed that The typical impact of this closure rule in a have had negative or ambiguous effects on income. The trade reform scenario is a real depreciation. Trade Blocs report (World Bank 2000a) found that The removal of tariffs generates an increase in South-South agreements between richer and poorer de- veloping countries are likely to generate losses for the import demand. Given the fixed trade balance, poorer ones when the poorer members import products this must be met by a rise in exports, achieved from the richer members, whose firms are not interna- through a real depreciation. The extent of the tionally competitive. For example, in the 1960s, Kenya depreciation will depend on the levels of the had a more developed manufacturing sector than trade elasticities (import and export). This sim- Uganda and Tanzania, and when the three formed the plification of foreign capital flows implies fore- East African Community (EAC), the latter two lost tariff revenue by importing from Kenya at the high protected closing an important channel for growth, for price rather than at the lower world price, with transfers instance, the increase in foreign direct invest- going from them to Kenya. This asymmetry proved un- ment (FDI) in the aftermath of trade reform. sustainable and resulted in the demise of the EAC. Empirically, this channel has proven to be 3. Of particular importance is that the results of ne- quite important as witnessed by Portugal and gotiations are made publicly available in user-friendly Spain with their entry into the EU, or by Mex- form. For example, data on tariff bindings are not made ico when it joined the North American Free available in a database format, preventing analysts from Tre Aundertaking cross-country research. This is important Trade Agreement. China has also witnessed a because it impedes efforts to estimate the magnitude boom in direct foreign investment in anticipa- and incidence of costs of protection. It is a truism that tion of its accession to the W7TO. Some of the to reduce protection and resist protectionist pressures, potential benefits of increased FDI are cap- those that lose (pay) need to be aware of the costs of tured by the scenarios with endogenous pro- such policies. The suppliers of, and the clients for, such ductivity growth. analysis and information are not only governments, but The version of the model used for this also civil society (think tanks) and the constituencies in individual countries that are affected by policy. To do analysis decomposes the world economy into this, they need easy access to the relevant data. 15 regions and 20 economic activities. The 4. For example, Blackhurst, Lyakurwa, and Oyejide model is calibrated to the latest release of the (2000) propose that governments transfer national rep- Global Trade Analysis Program (GTAP) data- resentatives from United Nations agencies in New set with a 1997 base year.38 The model is York to the WTO to intensify cooperation by members solved forward as a series of linked sequential of regional integration arrangements. equilibria, where population and labor force 5. However, synergies could be realized through networking and collaboration between advisers. For growth rates are given, capital accumulation is example, the new Global Development Gateway that is based on the previous period's level of invest- being established by the World Bank in cooperation ment, and productivity is calibrated to a target with numerous public and private sector partners GDP growth rate. could provide a powerful vehicle for building a trade After a plausible baseline simulation is de- community and sharing expertise. Such a portal could veloped, policy shock scenarios are undertaken also be used to assist governments (and NGOs) seeking where parameters calibrated in the baseline to identify experts and determine what has already been done in specific countries or on specific issues. simulation are taken as given (for example, 6. For example, Esty (1994) argues that if taxes or productivity parameters) and GDP growth is an other measures compensate for environmental conse- outcome. Thus, in the absence of any change in quences, trade will result in more efficient use of re- the exogenous environment, the policy shock sources, spur innovation, and lower costs of environ- scenario should reproduce the baseline. mental protection everywhere. Ekins and others (1994) add that if commodities for export are produced with serious damage to the environment, then trade may ag- Notes gravate environmental problems. DeBellevue (1994) 1. See World Bank 2001, chapter 3. and Ropke (1994) share the same view. Another chan- 2. This is not to say that South-South arrangements nel of interaction between trade and the environment cannot be made to work. However, many South-South concerns transboundary problems where pollution 179 GLOBAL ECONOMIC PROSPECTS spills over from one country to another. Esty (1994) ar- high-income members, and 2020 for developing-coun- gues that when a country suffers transboundary harm try members. Expansion of the European Union toward due to exports of pollution-intensive product, the im- the east and south would also eliminate barriers across position of trade restrictions on the import of the cul- a broad number of partners. There are also numerous pable product can be justified. other proposed agreements, many of them bilateral. 7. Pearson (1987) asserts that there is no evidence 9. There are some dynamic gains coming from to establish that lax environmental regulations would changes in investment and structure. be captured by foreign investors as opposed to local 10. Depending on the market power of a country's firms. Leonard (1988) argues that environmental regu- trading partners, and its own market power and the lations do not alter plants' location decisions. He pre- size of the shock, the terms-of-trade impact could be sents case studies of foreign direct investment in Ire- significant. A potentially critical situation is a country land, Mexico, and Romania to examine trade data and that only imports highly differentiated goods from a investment statistics, and concludes that the data do small set of importers and exports an homogeneous not support an industry flight hypothesis. Smarzynska good on world markets. Other factors also influence and Wei (2001) consider the corruption level of the changes in the terms of trade. For example, the re- host country and use firm level datasets for 25 transi- moval of agricultural subsidies by high-income coun- tion economies to examine support for the industry tries is likely to be beneficial for exporters of these flight hypothesis. They find limited evidence to support commodities from developing countries, because they the assertion that firms move to countries with less would profit from a rise in the world price of these strict environmental regulations. commodities. Levinson (1996) uses industry abatement costs, 11. The following functional form is used: business taxes, wages, energy costs, and roads to mea- sure environmental performance and studies the effects (E of these factors on the probability that a new industry X, plant would open in a certain state. The results reveal little evidence that environmental regulations hinder es- where y is the growth in sectoral productivity due to tablishment of new plants. In contrast, Lucas, Wheeler, the change in openness (added to an exogenous growth and Hettige (1990); Mani, Pargal, and Huq (1997); factor), X is a calibrated parameter, E and X represent and List and Co (1999) find some evidence to support respectively sectoral export and output, and 9 is the anduelasticity. The parameter X has been calibrated so that the industry flight hypothesis. Lucas, Wheeler, and I-et-I tige (1990) find that toxic intensity has increased more (on average) openness determines roughly 40 percent rapidly in developing countries than in industrial coun- of productivity growth in the baseline simulation, and tries. They conclude that stringent environmental regu- the elasticity has been set to 1. lations in the OECD countries have caused relocation 12. Scale economies could, of course, be modeled of pollution-intensive industries. According to Mani, explicitly. Pargal, and Huq (1997), environmental spending in 13. Aggregate income gains or losses summarize the India has a positive impact on plant location. However, extent to which trade distortions are hindering growth they conclude that environmental regulations are not a prospects and the ability of economies to use the gains to help those whose income could decline. Figure 6.2 significant factor in determining plant location, because .mae these i core odele ecooe. costs involved with environmental regulations are not Themfie prese t agrte mpactseinoeso lare nouh o eced ohe cotsof oig bsis. The figure presents the aggregate impacts in terms of large enough to exceed other costs of doing business. List and Co (1999) study the relationship between In- the outcome in the final year of the simulation (2015). cation decisions and environmental regulations. They The results are presented in nominal value terms (ex- use state-level data from 1986-93; their results show pressed in 1997 prices), as well as relative to baseline in- . come. Finally, it shows the results of both scenarios- that a 10 percent increase in the median state's (West wit. and itho te-sest proctivity. Virginia) regulatory expenditures per manufacturer de- Real icoi smarie byoHiciaivln creases the probability of attracting a new firm by 3.9 .Raio is rep e th icetat consum- percent for the median state. They conclude that envi- . ers would be willing to forgo to achieve post-reform ronmental stringency and the location decision of a well-being (uP) compared to baseline well-being (ub) at new firm are inversely related. baseline prices (pb): 8. The proposed Free Trade Agreement for the Americas has a quick timeline for the elimination of most trade barriers in the Western Hemisphere. Asia- Pacific Economic Cooperation (APEC) has proposed a where E represents the expenditure function to achieve 2010 deadline for eliminating trade barriers among its utility level u given a vector of prices p (the b super- 180 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT script represents baseline levels, and p the post-reform Nordstrom 1996; and lanchovichina and McDougall levels). The model uses the extended linear expenditure 2000. system (ELES), which incorporates savings in the con- Imperfect competition and scale economies. Relax- sumer's utility function. See Lluch (1973) and Howe ing the assumption of constant returns to scale tech- (1975). The ELES expenditure function is easy to eval- nology and allowing for imperfect competition can uate at each point in time. (Unlike the OECD treatment lead to additional sources of gain from trade openness. of EV, we use baseline prices in each year rather than The ability to increase market size allows firms to base year prices. See Burniaux and others 1993). The spread fixed costs over greater output, thereby reduc- discounted real income uses the following formula: ing average costs-and greater competition from other firms can reduce price markups. Both effects can 2015 2015 significantly enhance the gains from openness. See for OE 3t-2004~~ p(at-2004)yd CEV= I(-04Ea pT-t example Harris 1984, Delorme and van der Mens- brugghe 1990; Harrison, Rutherford, and Tarr 1996; where CEV is the cumulative measure of real income Francois, McDonald, and Nordstrom 1996; and (as a percent of baseline income), 3 is the discount fac- Brown, Deardorff, and Stern 1992. tor (equal to 1/(1+r) where r is the subjective discount Endogenous growth (or productivity). Openness rate), Yd is real disposable income, and EVa is adjusted does not occur in a vacuum. As countries open their equivalent variation. The adjustment to EV extracts borders to new products and capital goods, local firms the component measuring the contribution of house- can take advantage of new technologies, foreign re- hold saving, since this represents future consumption. search and development, and other innovations to sig- Without the adjustment, the EV measure would be nificantly enhance their productivity. Greater market double counting. The saving component is included in access of local exporters also can generate productivity the EV evaluation for the terminal year. Similar to the externalities by gaining more knowledge of foreign OECD, a subjective discount rate of 1.5 percent is as- markets and processes and improving production to sumed in the cumulative expressions. match international norms and standards. See, for ex- 14. All nominal dollar figures are in 1997 prices; ample, de Melo and Robinson 1990; Rutherford and the model does not incorporate nominal inflation. The Tarr 2001; Diao, Roe, and Yeldan 1999; and Dessus, price anchor of the model is an export price index of Fukasaku, and Safadi 1999. manufactures from the OECD high-income countries, Endogenous capital flows. While many trade mod- similar in concept to the World Bank's Manufactured els typically abstract from incorporating endogenously Export Unit Value index. It is set to one in the base and determined capital flows, there is significant empirical all subsequent years. evidence that the gains from international capital mo- 15. Most of the action occurs in the agricultural and bility are quantitatively important. There are two manufacturing sectors because this version of the model channels through which capital flows influence does not incorporate significant barriers in services. growth. The first is the direct channel leading to capi- 16. Rodriguez and Rodrik (1999), among others, tal deepening (although this requires care in evaluating have criticized these studies on methodological grounds; the long-term gains, since eventually this generates a they have also criticized those who use them to advo- stream of income repatriated back to the foreign own- cate simplistic policy conclusions. Nonetheless, the pre- ers). The second channel is through productivity since ponderance of evidence points rather consistently it is often the case that the incoming capital embodies to the fact that countries with more open trade and new and improved technologies. See, for example, financial regimes, complemented with other appropri- McKibbin and Sachs 1991; Collado, Roland-Holst, ate macroeconomic and social policies, have improved and van der Mensbrugghe 1995; McKibbin and growth performance. Wilcoxen 1999; Hertel 1997; and lanchovichina and 17. These four ideas are described summarily: McDougall 2000. Dynamics. The main channels are two-fold. First, 18. Comparisons of model results are notoriously higher incomes lead to higher savings and thus greater difficult to make. Models can differ in numerous ways, capital accumulation. The second channel is that tariffs dimensionality (for instance, regions, sectors), data- are often imposed on investment goods. Their removal bases (notably policy instruments, such as tariff levels), leads to a rise in real investment, because per dollar of closure rules, time horizon, functional specification, saving a buyer can purchase more investment goods. and elasticities (such as supply, income, trade, and so Baldwin (1992) estimates that these dynamic gains on), and market structure (both goods and primary could triple the static efficiency gains. See, for example factors). Moreover, studies do not necessarily report Burniaux and van der Mensbrugghe 1994; Harrison, the same indicator as a measure of the gains from Rutherford, and Tarr 1996; Francois, McDonald, and trade. The choices are various: real GDP, real income, 181 GLOBAL ECONOMIC PROSPECTS some measure of welfare such as equivalent variation, 23. Thus, the results represent two different eco- real absorption, and so on. And the units of measure- nomic equilibria abstracting from any dynamic effects ment are not always identical. The indicators could be of changes in investment or saving, or both, and other reported in different base year dollars, or as some cu- structural transformation. mulated discounted value, or as a percentage of some 24. In the merchandise trade liberalization scenar- base year indicator. Some noteworthy attempts to com- ios, the service sectors were aggregated into a single pare model results include the Martin and Winters account. 1996 volume on the Uruguay Round simulations, and 25. See van der Mensbrugghe 2001 for further the OECD 1993 and 1998 and IPPC 2001 studies com- details. paring model results of the potential economic conse- 26. The latter is implemented as a trade penalty, quences of mitigating climate change. similar to an import tariff, but with no direct revenue 19. The sensitivity of these results to the openness implications. Formally, the model implements a version or productivity relationship is discussed below. of the so-called iceberg model. For example, if the 20. Modeling of services trade liberalization is still penalty parameter is set to 0.9, this implies that, of 100 in its infancy. First, simply assessing the trade (and in- units that are exported, only 90 units actually arrive at vestment) barriers quantitatively is a much more diffi- destination. cult task than developing tariff data on goods trade. 27. For the purposes of comparison, the income Second, the nature of the barriers is harder to specify gains-as measured in dollar terms-were scaled to and implement in a model. It is currently a very active projected 2015 income levels. This has no impact on area of research. the relative gains. 21. For example, assume total sectoral productivity 28. The spillover effects of this scenario for high- in the baseline is 2.5 percent. If the share affected by income countries are marginal. openness is 40 percent, total productivity is the sum of 29. The difference being that the revenues gener- two components-1 percent determined by the open- ated by the markup typically accrue to firms and not ness factor (for instance, 40 percent of 2.5) and the the government. residual 1.5 percent determined by other factors. In 30. In results not reported, the impact of the policy simulations, the trade openness indicator only markup is highly nonlinear. Elimination of a 20 percent affects the 1 percent in this example. Thus if openness markup-i.e., a doubling of the initial markup-gener- increases by 10 percent, and the elasticity is 1, produc- ated an incremental income gain of $106 billion, some tivity will increase to 2.6 percent (=1.5 + 1.1). four times the impact of eliminating a 10 percent 22. These results are within the range found in the markup. The model results were generally linear with few comparable studies available. For example Dessus respect to the other two instruments. and others (1999) estimate a macro relationship be- 31. The shares are based on GTAP data. tween openness-as measured by the export plus im- 32. The World Bank has an active research program port to GDP ratio-and per capita GDP growth using a to improve analysis of the openness distribution linkages. panel dataset. Their preferred elasticity is 0.09-that is, It involves developing both data and methodologies to an increase in the trade-to-GDP ratio of 10 percent incorporate multiple representative households directly leads to a rise in per capita GDP of 0.9 percent. As a into AGE models (see, for example, Hertel, Preckel, and rough approximation, the elasticity of 1 used above in Cranfield 2000) as well as to inject the results of the AGE the base simulation implies an elasticity of 0.4 for total simulations into much more detailed microsimulation sectoral productivity with respect to openness-some models based on country-specific household surveys, four times higher than the 0.09 used in the Dessus and typically with thousands of households. More on this others (1999) study. However, their endogenous pro- research program is available at: http://wwwl.world- ductivity applies economywide, that is, including ser- bank.org/wbiep/trade/povertyconf.html. vices. If the 0.4 elasticity is multiplied by the agriculture 33. Skilled and unskilled workers are assumed to be and manufacturing share of the economy, somewhere imperfect substitutes for one another. An alternative between 30 and 60 percent, the economywide impact specification would be to have skilled workers as com- falls to somewhere between 0.12 and 0.24. A second plements to capital and the two together an imperfect factor to consider is that productivity is only labor-aug- substitute for unskilled labor. The distributional out- menting. Correcting for the labor share in the economy, comes would change, but presumably would favor say, around 50 percent, the final impact on aggregate even more unskilled labor in most developing regions. productivity falls between 0.06 and 0.12. This is There is an active debate about the role of trade open- roughly in the range of the elasticity of Dessus and oth- ness on relative wages. The standard theoretical argu- ers (1999) and explains in part the differences in the es- ment suggests that returns to the relatively abundant timates of the two studies. factor, unskilled labor in the case of developing coun- 182 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT tries, should rise with trade openness. However, the re- http://www.spp.umich.edu/rsie/model/analytics.ht verse has been observed in some developing countries. ml for model specification.) Part of the explanation comes from an increase in FDI, - . 2001. "CGE Modeling and Analysis of Multi- which is assumed to require skilled labor. Another par- lateral and Regional Negotiating Options." Re- tial explanation is that openness frequently was under- search Seminar in International Economics Dis- taken simultaneously with other reforms, such as pri- cussion Paper No. 468. School of Public Policy, vatizations. The latter has often led to shedding of The University of Michigan, Ann Arbor. (Avail- labor to rationalize operations. able at http://www.spp.umich.edu/rsie/working- 34. The figure reflects the results of the scenario papers/papers4Sl-475/r468.pdf.) with endogenous productivity. It differs only slightly Burniaux, Jean-Marc, and D. van der Mensbrugghe. from the results of the scenario with exogenous pro- 1994. "The RUNS Global Trade Model." Eco- ductivity. nomic & Financial Modelling Autumn/Winter: 35. See van der Mensbrugghe 2001 for complete 161-282. model specification. Burniaux, Jean-Marc, G. Nicoletti, and J. Oliveira- 36. The GTAP dataset merges public savings with Martins. 1993. "GREEN: A Global Model for the household sector, and thus the public balance is zero Quantifying the Costs of Policies to Curb CO2 in all countries in the base year. The balancing items are Emissions." OECD Economic Studies: Special net transfers between the government and households. Issue on the Economic Costs of Reducing CO2 37. Although exogenous in any given time period, Emissions 19 Winter: 49-92. Paris. the capital account could vary over time (subject to the Cline, William R., N. Kawanabe, T. 0. M. Kronsj6, constraint that it must sum to zero globally). To avoid and T. Williams. 1978. Trade Negotiations in the sustainability issues, the capital account is assumed to Tokyo Round: A Quantitative Assessment. Wash- be fixed at its base year level. Alternatively, it could ington, D.C.: The Brookings Institution. converge toward zero over the time period. Collado, Juan Carlos, D. Roland-Holst, and D. van der 38. GTAP, based at Purdue University, has developed Mensbrugghe. 1995. "Latin America Employment the most widely used dataset for global trade analysis Prospects in a More Liberal Trading Environ- (See Hertel 1997 or GTAP's Web site: www.gtap.org). ment." In Social Tensions, Job Creation and Eco- The latest release (version 5) incorporates data for 66 nomic Policy in Latin America, D. Turnham, C. countries or regions, and 57 sectors. Foy, and G. Larrain, editors. Paris: Organisation for Economic Co-operation and Development. Collier, Paul, and D. Dollar. Forthcoming. Globali- References zation: Facts, Fears, and an Agenda for Action. Aggarwal, Mita. 1995. "International Trade, Labor World Bank Policy Research Report, Washington, Standards, and Labor Market Conditions: An D.C. Evaluation of the Linkages." Office of Economics de Melo, Jaime, and S. Robinson. 1990. "Productivity Working Paper 95-06-C. U.S. International Trade and Externalities: Models of Export-Led Growth." Commission, Washington, D.C. PRE Working Papers WPS 387. World Bank, Wash- Anderson, Kym, J. Francois, T. Hertel, B. Hoekman, ington, D.C. and W. Martin. 2000. "Potential gains from trade Deardorff, Alan V., and R. M. Stern. 1981. "A Disag- reform in the new millennium." Paper presented gregated Model of World Production and Trade: at the Third Annual Conference on Global Eco- An Estimate of the Impact of the Tokyo Round." nomic Analysis, held at Monash University, June Journal of Policy Modeling 3: 127-152. 27-30. (Available at http://www.monash.edu.au DeBellevue, E. B., E. Hitzel, K. Cline, J. A. Benitez, J. policy/conf/65Anderson.pdf.) Ramos-Miranda, and 0. Segura.. 1994. "The Baldwin, Richard E.. 1992. "Measurable Dynamic North American Free Trade Agreement: An Eco- Gains from Trade." Journal of Political Econ- logical Economic Synthesis for the United States omy, 100 (1), pp. 162-74, February. and Mexico." Ecological Economics 9: 53-71. Blackhurst, Richard, Bill Lyakurwa, and Ademola Oye- Delorme, Franqois, and D. van der Mensbrugghe. 1990. jide. 2000. "Options for Improving Africa's Par- "Assessing the Role of Scale Economies and Im- ticipation in the WTO". World Economy. 23(4), perfect Competition in the Context of Agricultural pp. 491-510. April. Trade Liberalization: A Canadian Case Study." Brown, Drusilla K., A. V. Deardorff, and R. M. Stern. OECD Economic Studies No. 13 (Winter 1989- 1992. "A North American Free Trade Agreement: 90): 205-36. Analytical Issues and a Computational Assess- DFAT (Department of Foreign Affairs and Trade). 1999. ment." The World Economy 15: 15-29. (See also Global Trade Reform: Maintaining Momentum. 183 GLOBAL ECONOMIC PROSPECTS Canberra, Australia. (Available at http://www. Harrison, Glenn W, T. F. Rutherford, and D. Tarr. dfat.gov.aultrade/negotiations/gtr_2000.pdf.) 1996. "Quantifying the Uruguay Round." In W. Dessus, S6bastien, K. Fukasaku, and R. Safadi. 1999. Martin and L. A. Winters, editors, The Uruguay "Multilateral Tariff Liberalisation and the Devel- Round and the Developing Countries. Cam- oping Countries." OECD Development Centre bridge, U.K.: Cambridge University Press. Policy Brief No. 18, Paris. (Available at http://www. Harrison, Glenn W., Thomas E Rutherford and David oecd.org/dev/ENGLISH/publication/Policy-B/ G. Tarr. 1997. "Trade policy options for Chile: a pbl8a.pdf.) quantitative evaluation". World Bank Policy Re- Diao, Xinshen, T. Roe, and E. Yeldan. 1999. "Strategic search Working Paper, No. 1783, Washington, DC. policies and growth: an applied model of R&D- Hertel, Thomas W., editor. 1997. Global Trade Analy- driven endogenous growth." Journal of Develop- sis: Modeling and Applications. New York: Cam- ment Economics 60: 343-80. bridge University Press. Dollar, David, and A. Kraay. 2001. "Growth is Good for Hertel, Thomas W., P. V. Preckel, and J. A. L. Cran- the Poor." World Bank Policy Department Work- field. 2000. "Multilateral Trade Liberalization ing Paper No. 2587. April. Washington, D.C. and Poverty Reduction." Paper presented at the Ekins, P., C. Folke, and R. Costanza. 1994. "Trade, Conference on Poverty and the International Environment and Development: The Issues in Per- Economy, jointly sponsored by World Bank and spective." Ecological Economics 9: 1-12. the Parliamentary Commission on Swedish Policy Elliot, Kimberly Ann. 2001. "Finding Our Way on for Global Development (Globkom), October Trade and Labor Standards." International Eco- 20-21, 2000, Stockholm, Sweden. (Available at nomics Policy Briefs. Institute of International http://wwwl.worldbank.org/wbiep/trade/papers/ Economics, Washington D.C. cranfield.pdf.) Esty, D. C.. 1994.. "Greening the GATT: Trade, Envi- Howe, Howard. 1975. "Development of the Extended Linear Expenditure System from Simple Savings As- ronment and the Future." Institute for Interna-supin"Eroen cnmcRvew6 tionl Eonomcs,Wasingtn, .C.sumptions." European Economic Review 6: tional Economics, Washington, D.C. 351 Finger, J. Michael. 1998. "GATT Experience with Safe- 305-10. guars: akig Ecnomc ad Plitial ens of lanchovichina, Elena, and R. McDougall. 2000. "The- guards: Making Economic and Political Sense of oretical Structure of Dynamic GTAP." GTAP the Possibilities that the GATT Allows to Restrict Technical Paper No. 17. Center for Global Trade Imports." World Bank Policy Research Working Analysis, Department of Agricultural Economics, Paper, No. 2000, Washington, DC. Purdue University, West Lafayette, Ind. (Available Finger, J. Michael, and Philip Schuler. 2000. "Imple- at http://www.agecon.purdue.edulgtap/resources/ mentation of Uruguay Round commitments: the download/160.pdf.) development challenge". World Economy, 23(4), IPCC (Intergovernmental Panel on Climate Change). pp. 511-25. April. 2001. "Climate Change 2001: Mitigation." Con- Francois, Joseph F, B. McDonald, and H. Nordstrom. tribution of Working Group III to the Third As- 1996. "The Uruguay Round: A Numerically sessment Report of the Intergovernmental Panel Based Qualitative Assessment." In W. Martin and on Climate Change (IPCC). Edited by B. Metz, L. A. Winters, editors, The Uruguay Round and 0. Davidson, R. Swart, and J. Pan. Cambridge the Developing Countries. Cambridge, U.K.: University Press, Cambridge, U.K. (A technical Cambridge University Press. summary is available at http://www.ipcc.ch/pub/ Gereffi, Garri, R. Garcia-Johnson, E. Sasser. 2001. wg3TARtechsum.pdf.) "The NGO-Industrial Complex." Foreign Policy Leonard, H. J. 1988. Pollution and the Struggle for July-August. World Product. Cambridge, UK.: Cambridge Uni- Goldin, Ian, 0. Knudsen, and D. van der Mensbrugghe. versity Press. 1993. Trade Liberalization: Global Economic Im- Levinson, Arik. 1996. "Environmental Regulations plications. Paris: OECD/The World Bank. and Manufacturers Location Choices, Evidence Grossman, Gene M., Krueger, A. B. 1993. "Environ- from the Census of Manufacturers." Journal of mental Impacts of a North American Free Trade Public Economics 62: 5-29. Agreement." In The Mexico-U.S. Free Trade Agree- List, John A., and C. Y. Co. 1999. "The Effects of En- ment, Peter M. Garber, editor. Cambridge and vironmental Regulations on Foreign Direct In- London: MIT Press. P.13-56. vestment." Journal of Environmental Economics Harris, Richard. 1984. "Applied General Equilibrium and Management 40: 1-20. Analysis of Small Open Economies with Scale Lluch, Constantino. 1973. "The Extended Linear Ex- Economies and Imperfect Competition." The Amer- penditure System." European Economic Review ican Economic Review 74 (5-December): 1016-32. 4: 21-32. 184 RESHAPING GLOBAL TRADE ARCHITECTURE FOR DEVELOPMENT Lucas, Robert E. B., D. Wheeler, and H. Hettige. 1990. Ropke, I. 1994. "Trade, Development and Sustainability: "Economic Development, Environmental Regula- A Critical Assessment of the 'Free Trade Dogma"' tion and the International Migration of Toxic In- Ecological Economics 9: 13-22. dustrial Pollution." Paper presented at the sympo- Rutherford, Thomas E, and D. G. Tarr. 1998. sium on International Trade and the Environment, -. Forthcoming. "Trade Liberalization, Product Washington D.C. Variety and Growth in a Small Open Economy: Mani, Muthukumara, S. Pargal, and M. Huq. 1997. A Quantitative Assessment." Journal of Interna- "Does Environmental Regulation Matter? Deter- tional Economics. minants of the Location of New Manufacturing Sachs, Jeffrey D., and A. M. Warner. 1995. "Economic Plants in India." Policy Research Working Paper Reform and the Process of Global Integration." No. 1718. World Bank, Washington, D.C. Brookings Papers on Economic Activity (1): Martin, Will, and L. A. Winters, editors. 1996. The 1-118. Uruguay Round and the Developing Countries. Smarzynska, Beata, and S.-J. Wei. 2001. "Pollution Cambridge, U.K.: Cambridge University Press. Havens and the Location of Foreign Direct in- McKibbin, Warrick J., and J. D. Sachs. 1991. Global vestment: Dirty Secret or Popular Myth?" World Linkages: Macroeconomic Interdependence and Bank, Washington, D.C. Processed. Cooperation in the World Economy. Washington, Tobey, James A. 1990. "The Effects of Domestic Envi- D.C.: The Brookings Institution. ronmental Policies." Kyklos 43 (2): 191-209. McKibbin, Warrick J., and P. J. Wilcoxen. 1999. "The Torres, Raymond. 1996. "Labor Standards and Inter- Theoretical and Empirical Structure of the G-Cubed national Trade" The OECD Observer No. 202. Model." Economic Modeling 16 (1):123-48. van der Mensbrugghe, Dominique. 2001. "LINKAGE OECD (Organisation for Economic Co-operation and Technical Reference Document." World Bank, Development). 1993. "OECD Economic Studies: Washington, D.C. Processed. (Available at http:// Special Issue on the Economic Costs of Reducing www.worldbank.org/prospects/pubs/TechRef.pdf) CO2 Emissions No. 19." Winter 1992. Paris. Wells, P., and M. Jetter. 1992. The Global Consumer. -. 1996. "Trade, Employment and Labor Stan- London: Victor Golancz. dards: A Study of Core Workers' Rights and In- Whalley, John. 1985. Trade Liberalization among ternational Trade." Paris. Major World Trading Areas. Cambridge, Mass.: . 1998. "Economic Modeling of Climate MIT Press. Change." OECD Workshop Report. September Wilson, John S. 2001. "Environmental Standards and 17-18, 1998. Paris. (Summary available at http:// Trade: A Primer." World Bank, Washington, D.C. www.oecd.org/dev/news/Environment/summary. Processed. pdf.) Wilson, John S., M. Sewadeh, and T. Otsuki. 2001. Pananagariya, Arvind. 1999. "Labor Standards in the "Dirty Exports and Environmental Regulation: WTO and Developing Countries: Trading Rights Do Standards Matter?" World Bank, Washing- at Risk." University of Maryland, College Park. ton, D.C. Processed. Pearson, C. 1987. Multinational Corporation, the En- World Bank. 2000. Trade Blocs. World Bank Policy vironment and Development. Washington, D.C.: Research Report. World Bank/Oxford University World Resources Institute. Press. Oxford, UK. Rodriguez, Francisco, and D. Rodrik. 1999. "Trade Xu, Xinpeng. 1999. International Trade and Environ- Policy and Economic Growth: A Skeptic's Guide mental Regulation: A Dynamic Perspective. Com- to Cross-National Evidence." NBER Working mack, NY. Nova Science Publishers, Inc. Paper W7081. NBER, Cambridge, Mass. (Avail- able at http://papers.nber.org/papaers/w7081.) 185 Appendix 1 Regional Economic Prospects East Asia and Pacific Region and services exports. Overall, the global eco- nomic impacts of the terrorist attacks are Recent developments likely to have pushed back the prospects for a HE YEAR 2001 WAS SHAPING UP AS A DIF- recovery in world trade and in East Asian ex- ficult year for East Asia and Pacific ports by six to nine months. (EAP) even before the September 11 ter- rorist attacks on the United States. The unex- Near-term outlook pectedly sharp cyclical downturn in the world Growth in the developing East Asia region is economy during the year had centered on a expected to fall to 4.6 percent in 2001 from recession in the global high-technology (high- 7.5 percent in 2000, and to recover only tech) sector, resulting in a plunge in the exports mildly to around 5 percent in 2002. These of the many East Asian economies that have be- would be the region's second weakest years come important suppliers of components and for growth since 1990, bar only the near-zero- finished products for world high-tech markets. growth year of financial crisis, 1998. Most of East Asian exports were also especially hurt by the slowdown in growth is concentrated in the the fact that the slowdown in global demand "Crisis 5" countries (that is, the five countries has been steepest in the region's largest external that were most affected by the financial crisis markets, the United States and Japan, which to- of 1997-98-Indonesia, Korea, Malaysia, the gether buy almost 40 percent of regional ex- Philippines, and Thailand). Aggregate growth ports. By July or August economies with a high in this group is expected to slow to only 2-3 reliance on high-tech, such as the Republic of percent in 2001 from around 7 percent in Korea, Malaysia, the Philippines, Singapore, 2000. Growth in 2001 will still reach around and Taiwan (China), were seeing U.S. dollar ex- 7 percent in China, which contains two-thirds port declines of around 20-25 percent on year of the region's poor (at the $2 a day poverty earlier levels. There was, nevertheless, some line). With a more diversified export basket comfort in the fact that, apart from the Phil- than some other countries in the region, ippines, the main impact of the high-tech re- China's export growth, while slowing sharply cession was falling on high income or upper- to an average 2 percent pace in June-August middle-income countries with low poverty. 2001, has at least avoided the huge 20-35 per- In the wake of the terrorist attacks, the re- cent declines seen elsewhere. Growth has also gion's export downturn is likely to become been bolstered by a robust fiscal stimulus pack- broader based, as falling consumer confidence age to offset the export slowdown. in the United States and around the world Other transition countries, such as Cam- dampens demand for the region's consumer bodia and Vietnam, which rely on low- or 187 GLOBAL ECONOMIC PROSPECTS Figure A1.1 Industrial production in East Asia and world semiconductor sales volumes 3-month moving average, percent y/y 25 40 20 30 15 20 10 East Asia East Asia 10 5 excluding China 0 0 -10 -5 World semiconductor -20 sales volume [right scale] -10 ----30 Jan. 1999 July 1999 Jan. 2000 July 2000 Jan. 2001 July 2001 Source: Semiconductor Industry Association through Bloomberg and World Bank. medium-tech exports, have also been less af- lower world growth after the attacks. Some of fected by the high-tech recession, and showed the smaller economies of the region that rely on continued buoyancy in domestic investment in commodity exports, such as Mongolia, Papua the first part of the year. Growth prospects for New Guinea, Fiji, and the island economies the region in 2002 will depend significantly on will be hurt by lower non-oil primary com- the timing and scope of world recovery. It is modity prices. As regards other sectoral effects probable that the steep cuts in interest rates, of the attacks, airline travel, tourism, and in- income tax cuts, and post-attack emergency surance are likely to be the worst affected, spending in the United States, as well as policy while inputs for military materials, information measures in other industrial countries, will technology (IT) infrastructure, and telecommu- lead to a rebound in the second half of 2002, nications may benefit. As a result, East Asia, strengthening to a more full-blown global re- with its reliance on high-tech exports, may be covery in 2003. less badly affected than other regions. How- Prospects for the region will also be affected ever, selected Pacific islands and countries such by specific sectoral trends in the wake of the as Thailand will feel a more significant effect of terrorist attacks. Oil prices have been volatile the pullback in world tourism. after the attacks and may have an upward bias Among other factors affecting near-term for the rest of the year because of the risk of prospects, it is notable that, despite serious military action in the Middle East, with a emerging capital market crises in Turkey and mixed impact on the region. If major disrup- Argentina, there were few signs of a general- tions are avoided, weak world growth will tend ized contagion effect or pullback of private to push oil prices lower in 2002. However, flows to the region in the first part of the year. countries that rely on worker remittances could Gross capital market flows to the region of be hurt by political turmoil in the Middle East, about $31.5 billion in January-July 2001 were as well as from weaker growth in Asia. Remit- only slightly lower than during the same pe- tances to the Philippines are already down. riod in 2000. This overall stability reflected Non-oil commodity prices have weakened in improvements in crisis countries' external bal- 2001, and are likely to weaken further with ance sheets in the last several years, including a 188 REGIONAL ECONOMIC PROSPECTS buildup of foreign reserves and reductions in liabilities-mean that in most countries higher short-term debt. Exchange rates, while volatile, spending can only be sustainable for a limited were not much different in early September time. Indeed, very high public debt levels will from the start of the year, while the majority of essentially preclude greater fiscal stimulus in equity markets had actually risen modestly Indonesia and the Philippines. Given these con- over this period. Capital market stresses were straints, a temporary increase in spending- concentrated on the Philippines and Indonesia, where possible-is best seen as a means of ad- reflecting political uncertainties earlier in the dressing specific social or sectoral objectives, year, as well as concerns about high public and as a complementary policy that allows debt. After the Sept. 11 attacks, secondary mar- countries to continue to make progress on dif- ket spreads for Indonesia and the Philippines ficult structural policies such as corporate re- widened. Equity prices fell sharply in most structuring, even in the current weak economic countries in the region. To some extent, the re- climate. gion will share in a more widespread investor The impact of this year's slowdown on pullback from emerging markets. Corporate re- poverty will be mitigated by the fact that the structuring and privatization efforts may be steepest declines in growth are in the high- hampered by reduced foreign investor interest. income, newly industrialized economies (NIEs- Among near-term policy responses to the including Hong Kong (China), Singapore, and slowdown, a number of countries have in- Taiwan (China) and in the richer Crisis-5 coun- creased fiscal expenditures somewhat to help tries, which have relatively low poverty rates. smooth the impact of the export shock, includ- Still, with less growth, this year's downturn in ing China, Korea, Malaysia, and Thailand. East Asia will stall the pace at which income Such expenditures can be especially helpful poverty in the region falls, while the risk of a rise when carefully targeted to address social pro- in poverty has also increased. According to cal- tection, infrastructure, or other particular sec- culations by the Bank's East Asia and Pacific Re- toral needs that may be warranted in a sharply gion, the proportion of people living below the slowing economy. However, concerns about $2-a-day line may edge down from an estimated relatively high or growing public debt-espe- 47 percent in 2000 to a forecast 46 percent in cially when measured inclusive of contingent 2001. Given continued robust growth in China Table A1.1 East Asia and Pacific forecast summary (percent per year) Baseline forecast Growth rates/ratios 1991-2000 1999 2000 2001 2002 2003 2004-2010 Real GDP growth 7.2 7.0 7.5 4.6 4.9 6.8 6.2 Consumption per capita 5.4 6.0 6.8 5.5 5.7 5.9 6.0 GDP per capita 6.0 5.9 6.4 3.6 4.0 5.9 5.4 Population 1.2 1.1 1.0 0.9 0.9 0.9 0.8 Inflationa 5.4 0.0 3.4 7.1 6.7 5.3 3.7 Gross Domestic Investment/GDPb 34.1 29.0 30.0 30.4 30.7 30.7 33.7 Central Gvt Budget Balance/GDP -1.0 -2.5 -2.2 -2.0 -2.4 -2.3 3. 1 Export Volumec 13.0 7.7 23.7 0.4 6.2 11.3 7.3 Current Account/GDP 05 4.3 3.3 1.4 0.0 0.5 -0.8 Memorandum Items GDP growth: EAP excl. China 5.3 6.9 7.1 2.3 3.4 5.4 5.0 a. Local currency GDP deflator; median. b. Investment ratio measure in real terms. c. Goods and non-factor services. Source: World Bank baseline forecast, October, 2001. 189 GLOBAL ECONOMIC PROSPECTS and other transition countries, which contain The region is not without its vulnerabili- the large majority of the region's poor, the main ties, as evidenced by the financial crisis of the source of slower region-wide poverty reduction late 1990s and the economic slowdown that in 2001 is likely to be the sharp slowdown in started at the end of 2000. The financial crisis growth in Indonesia, the Philippines, and Thai- revealed in stark terms the deficiencies of the land, which contain most of the rest of the re- region's banking and financial institutions, gion's poor. and the lack of sufficient regulatory oversight In a longer-term perspective, it is notable to compensate for those deficiencies. In the af- that the pace of poverty reduction in the re- termath of the crisis, many of the countries in gion has slowed dramatically, something that, the region have undertaken a significant over- persisting over time, cannot help but have deep haul of both the financial and the regulatory social, political, and policy implications. Be- institutions, but the legacy of the crisis persists tween 1990 and 1996 the regional poverty in many of the countries. Economic recovery rate at $2-a-day fell from 67 to 49 percent, but and current account surpluses have provided from 1996 to 2000 it fell only 2 percentage some breathing room, but as the current slow- points more. The less numerically significant down indicates, the region's authorities need reason is the financial crisis and slow recovery to pursue financial reform, in particular to in Indonesia, the Philippines, and Thailand. boost financial intermediation to ensure that The other is slower income growth in China's the most productive investments get funded. rural areas-where most of China's poor live-even as urban income growth has gone Risks from strength to strength. Thus the drama of A key issue for policymakers in the region is to East Asian poverty reduction will largely de- position their countries in order to be able to pend on how countries address disparities in take full advantage of the global recovery rural-urban and intra-regional growth, as well when it arrives. Medium-term structural re- as the structural and institutional improve- forms that strengthen the fundamental under- ments needed to bolster growth overall. pinnings of development are likely to have a more significant impact on growth and Long-term prospects poverty reduction than possible short-term Despite these near-term weaknesses, the long- gains from fiscal stimulus. At the same time, term prospects for East Asia remain broadly this year's largely unexpected global downturn positive. Average annual growth rates could has shown the weakness in the strategy of sim- exceed 6 percent in the 2004-10 period. Most ply trying to "grow out" of the problems left of the countries in the region are committed to over from the financial crisis of 1997-98. In- strengthening the underlying determinants of deed, in the wake of the September 11 attacks, strong and sustained growth-improvements higher uncertainty and risk may become a in education, enhancing the rule of law, pro- more prevalent feature of international affairs moting high domestic savings (including pru- for some time. Structural reforms should then dential fiscal policies), and openness to trade also help make the region's economies more and investment. As demonstrated over the last robust in riding through a more uncertain and three decades, the region's economies have volatile external environment. Among struc- been able to scale the technology ladder and tural issues facing the region, the importance significantly close the production and income of renewed attention to corporate and finan- gap compared to the most industrialized na- cial restructuring; trade reform; and institu- tions. China's entry into the World Trade Or- tional and governance reforms are worth par- ganization (WTO) is a particularly notable ticular note. If the region is able to implement event that has positive trade and productivity contemplated reforms in these areas, it will im- implications for the whole region. prove the climate favoring new investment 190 REGIONAL ECONOMIC PROSPECTS (foreign and domestic) and technological prog- Even so, its external financial position remains ress, opening the way to realizing its long-term strong, with a small current account deficit, potential. very low external debt, and $45 billion in for- eign exchange reserves. Public finances have weakened in Bangladesh over recent years, South Asia with the public sector deficit reaching about 8-9 percent of GDP in 2001. In this context Recent developments this country's external position has become outh Asia, one of the world's poorest re- increasingly vulnerable. In contrast, Pakistan gions, remains a relatively closed econ- and Sri Lanka have been successful in reducing omy, despite progress toward trade liber- their fiscal imbalances. In the context of an alization in the 1990s. Imports and exports IMF program, Pakistan's fiscal deficit has de- are a much smaller share of GDP than in Latin clined to 5.2 percent of GDP but financial America or East Asia, and tariffs are among stress remains, with a large public debt, low re- the highest in the world. The region is also rel- serves, and a large financing gap. Sri Lanka atively closed to private capital flows. How- also reached an agreement with the IMF in ever, dependence on official flows is large for 2001 and is undertaking macroeconomic and some countries. structural reforms. Notwithstanding chronically high fiscal deficits, the region was able to grow at re- Near term outlook spectable rates over the 1990s. However, in With the deterioration in the external environ- 2001 the global slowdown adversely affected ment, GDP growth in 2001 and 2002 in South growth across the region and GDP growth fell Asia is projected to be 4.5 percent and 5.3 per- from 4.9 percent in 2000 to 4.5 percent in cent, respectively. Agricultural production is 2001. Because of the region's narrow tax base expected to increase in the second half of 2001 and reliance on custom imports, slower as a result of favorable weather conditions. growth had an immediate adverse fiscal im- Normal-and in some cases excessive-mon- pact resulting in additional pressure on al- soon rain has filled water reservoirs and water ready overstretched fiscal positions. tables throughout the region. The tragic events of September 11 focused However, private investment will be influ- attention on South Asia because the military enced negatively by the heightened degree of response created special risks to the countries uncertainty in the wake of the events of Sep- of the region, especially in Pakistan. For exam- tember 11. The external environment will pro- ple, freight rates to and from Pakistan have vide little or negative stimulus, and the need been increased 10 to 15 percent by major ship- for fiscal consolidation will further dampen ping lines.' Importers in other countries, fear- aggregate demand. ing supply disruptions, have canceled orders The global downturn in growth in 2001-02 for goods from Pakistan.2 These developments will have some deleterious effects on the re- have put additional pressure on Pakistan's gion, perhaps more limited than in other re- already vulnerable external position-with a gions. The downturn in export market growth, heavy external debt of $38 billion that absorbs from 13 percent in 2000 to an average of 3.3 more than 40 percent of Pakistan's exports percent in 2001-02, will be mirrored by fall- earnings and external financing needs that have ing import demand stemming from slower risen to $3.4 billion. growth domestically. Some countries in the re- Large fiscal deficits have been a persistent gion are depreciating their currencies to pro- challenge in most of the region. India's fis- mote exports and increase competitiveness, cal deficit is 10.5 percent on a consolidated which should have a positive effect on the trade basis, including central and state governments. balance. Given improved weather conditions, 191 GLOBAL ECONOMIC PROSPECTS Figure A1.2 Industrial production in South Asia (percentage) 35 25 15 A Bangladesh 5 -5 `1P n India Pakistan D* -2 5 1 1 1 1 1 1 1 1 t I I I I I I I I I I I I I I I I II I I I Source: IMF, International Financial Statistics. and as oil prices fall, trade balances will bene- to garment exports of the region. TDA-2000 fit as well. provides duty-and-quota-free access, under cer- The region may struggle in its clothing sec- tain conditions, to the U.S. market for textile tor in the near term. The Trade Development and apparel products to 72 countries in Sub- Act-2000 (TDA-2000), passed in January Saharan Africa and the Caribbean. Bangladesh 2000 by the U.S. Congress, poses a challenge and India both have reported sharply lower Table A1.2 South Asia forecast summary (percent per year) Baseline forecast Growth rateslratios I ~ -2,1 IfHlu ,"I 01 101 'a r Real GDP growth 5. 4. . Consumption per capita 2.6 6.1 1.6 3.6 3.0 3.0 3.0 GDP per capita 3.3 3.9 3.0 2.8 3.6 3.8 4.0 Population 1.9 1.9 1.9 1.7 1.7 1.6 1.4 Inflationa .1 4.6 5.8 6.1 7.3 7.3 6.5 Gross Domestic Investment/GDPb 22,8 22.6 24.3 24.8 25.4 25.6 28.9 Central Gvt Budget Balance/GDP -8.6 -4.0 -5.7 -4.8 -4.7 -4.5 -4.1 Export Volumec 9.3 1.8 7.5 6.0 8.8 9.2 7.9 Current Account/GDP -1A -0.8 -0.3 -0.1 -0.5 -0.6 -0.8 Memorandum Items GDP growth: SAS excl. India 4.2 3.6 3.9 4.7 4.9 5.2 5.2 a. Local currency GDP deflator; median. b. Investment ratio measure in real terms. c. Goods and non-factor services. Source: World Bank baseline forecast, October, 2001. 192 REGIONAL ECONOMIC PROSPECTS monthly garment exports this year compared easing of tariff and non-tariff barriers and im- to last year. For example, Bangladesh reported port substitution policies, providing greater op- 15 percent lower garment exports in February portunities for trade integration with the global 2001 compared to a year ago.3 An expansion economy, particularly for the smaller countries of TDA-2000 to include the South Asia re- within the region. gion, which is just as poor as those currently included in TDA-2000, would redress the im- Risks balance and improve growth prospects of the Besides political risks in the short run, other region. risks to the forecast stem from the major chal- lenges that countries in the region face in the Long term prospects consolidation of their fiscal positions and debt Long term growth in South Asia should aver- levels. In India, sharp reductions in the fiscal age about 5.4 percent, similar to the projec- deficit may prove difficult in the short run. As tions of GEP 2001, and near the average a result, the recent fall in the rate of inflation growth rates of the 1990s. This reflects ex- is not expected to last beyond 2001. pectation of reforms continuing to be imple- Design and implementation of tax reform mented at a gradual pace. Lower population measures will be difficult. A gradual increase in growth and structural reform in the next tax revenue collection in all countries in the re- decade will possibly lead to per capita growth gion is required. Changes in the incidence of close to 4 percent per year. taxation will be necessary to decrease the re- Potential output growth in the region has in- liance on trade taxes and broaden the tax base creased, given the improvement in human cap- to stabilize revenue collections over time. More ital indicators in recent years, with higher liter- discipline will be required in fiscal expendi- acy rates and school enrollments and lower tures to ensure fiscal sustainability, while being infant mortality rates. Additionally, the high careful to maintain expenditures that are es- skill levels of Indian workers with training in sential for development programs. For exam- technology sectors, a boom area of growth, ple, reducing subsidies has been singled out will ensure that the highly productive invest- as a target in expenditure reform programs. ment in these sectors will continue in the long Broadening the tax base away from trade is term. As scheduled privatization and reform of also a part of the trade liberalization strategy state-owned enterprises occur, private invest- that will ensure that exporters have access to ment will account for a greater share of do- cheaper inputs and consequently become more mestic investment, with the concomitant bene- competitive in global markets. Sustainable fis- fits flowing from higher productivity of private cal revenues and a responsible expenditure investment compared to that of public invest- program will be required in several countries ment. Additionally, privatization will encour- to counter financial vulnerability. Countries age foreign investment and the associated with healthy debt levels should also act to en- spillovers to the domestic economies. Trade lib- sure sustainable fiscal positions to prevent a eralization is also expected to continue with the decline into unsustainable debt levels. 193 GLOBAL ECONOMIC PROSPECTS Latin America and the Caribbean slower economic activity in the region as a whole-exerting downward pressure on most Recent developments regional currencies. Growth out-turns for most countries in Large external financing requirements as a the region in 2001 were much worse share of GDP coupled with fiscal deficits and than anticipated in the spring of the high public debt reduced the ability for counter- year. Adverse developments in the external en- cyclical fiscal and monetary policies in some vironment and in domestic conditions in some countries. Despite falling U.S. interest rates, countries were the primary reasons behind the which reduce dollar debt-service payments- sharp reduction in the region's GDP growth, the depreciating exchange rates, slowing eco- from 3.8 percent in 2000 to an estimated 0.9 nomic activity, and rising domestic interest percent in 2001, about 2.8 percentage points rates (needed to maintain investor confidence) lower than anticipated in the spring. The placed additional pressure on fiscal balances, growth slowdown was most acute in the "Big limiting the scope for automatic stabilizers Three" (Argentina, Brazil, and Mexico), reflect- to function properly. Indeed, some countries, ing the increasing impact of the global, and par- such as Brazil, had to tighten both fiscal and ticularly the U.S. slowdown; economic difficul- monetary policies in an effort to offset the ties in Argentina; and the energy crisis in Brazil. combined negative effects of an energy crisis, a Uncertainties linked to the electoral process in large reduction in FDI inflows, and contagion Argentina this year and in Bolivia, Brazil, from the Argentine crisis, which resulted in a Colombia, Costa Rica, and Ecuador next year sharp reduction in capital market flows as sec- contributed to falling investment rates in a ondary market spreads rose and remained high number of countries. Weaker growth in Ar- in the wake of the September 11 terrorist at- gentina and Brazil, along with a worsening of tacks (figure A1.3). Although these policies the external environment, contributed to a de- tended to keep inflation under control, they re- celeration of growth in other South American suIted in a sharp growth deceleration and ex- countries, while a collapse of commodity prices acerbated the already high level of unemploy- (especially for coffee and semiconductor prices) ment throughout the region (16 percent in and a severe drought lowered growth rates in Argentina, for example). Central America. In contrast, Ecuador and Repii- In Argentina, recovery from the deep reces- blica Bolivariana de Venezuela did better than sion in 1999 has proven elusive, with each up- in 2000 due to relatively high oil prices. turn in economic activity usurped by political Rapid deterioration in global activity con- stalemate on reforms, a weakening of fiscal ac- tributed to a sharp decline in export revenues. counts during the first half of the year, and Excluding Mexico, dollar exports from the re- volatile capital flows-reflecting investor un- gion grew by an average of about 8 percent certainty about solvency of public debt. GDP (year over year or y/y) in the first half of the growth has remained in negative territory since year-down from over 15 percent in 2000. the third quarter of last year. The authorities With the exception of Brazil (where dollar ex- undertook a number of initiatives to bolster ports grew by 11.5 percent y/y), most countries investor confidence-including a $29.5 billion had exports growing below 4 percent. In Mex- debt swap, a severe fiscal adjustment aimed ico, the decline was even more dramatic, from at zero deficit, and a package of tax reforms an average of 22.6 percent growth in 2000 to aimed at improving competitiveness of Argen- zero (0 percent) by June 2001. Moreover, capi- tine firms. tal market commitments to the region weak- Negative fallout from Argentina was most ened markedly (that is, they fell by 21 percent acute in Brazil, and other Mercosur partners- (y/y) in the first half of the year)-reflecting slowing capital flows, especially FDI inflows; the deteriorating conditions in Argentina and increasing yield spreads; and contributing to a 194 REGIONAL ECONOMIC PROSPECTS Figure A1.3 Secondary market spreads for selected LAC countries, 2000-2001 (basis points above U.S. Treasuries) 1,600 1,600 Argentina 1,200 1,200 800 800 F" Colomb i 400 400 Mexico 0- 0 Jan. 2000 July 2000 Jan. 2001 July 2001 Source: JP Morgan's EMBI Global indices through Bloomberg. weakening of the real. Brazil's drought-induced mid-year elections kept investment rates low energy crisis and falloff in FDI, in conjunction and restrained consumer spending, resulting in with a depreciating currency, put upward pres- growth slowing to below 1 percent from over sure on inflation and set in motion tighter fis- 3 percent last year. Growth in Colombia also cal and monetary policies. Targets for the pri- weakened compared with 2000, due to lower mary (before interest payments) fiscal surplus coffee prices and lower-than-expected invest- were raised, and policy interest rates rose by ment caused by rising uncertainty (including 325 basis points in the four months to August, the electoral cycle, legal infrastructure, and the contributing to a slowdown in growth. In con- guerrilla war). In contrast, growth in the oil trast, Mexico suffered little contagion from the exporters in the Andean region held up well crisis in Argentina, with both the currency and due to high, although declining, oil revenues. equity markets rising strongly-reflecting the Growth in Venezuela was sustained by large- continuing positive impact of North America scale public expenditure, while growth in Free Trade Agreement (NAFTA) membership Ecuador accelerated from the low or negative on FDI. Nonetheless, output growth con- growth in 1998-2000 with the construction of tracted rapidly-in line with the sharp slow- a new oil pipeline. down in U.S. activity, reaching zero (y/y) by the In Central America, growth in 2001 was second quarter of 2001 after averaging nearly lower by about 1.7 percentage points compared 7 percent in 2000. with 2000, due primarily to a weakening of Slowing economic activity in Argentina and economic activity in Mexico and in the United Brazil along with falling copper prices affected States, a collapse of coffee prices, and a major growth negatively in Chile, but with lesser ex- drought, which severely affected Honduras and ternal financing concerns than in other coun- Nicaragua. The sharp fall in semiconductor tries, the authorities were able to reduce inter- prices hurt Costa Rica particularly hard as est rates. GDP growth slowed in 2001 but by semiconductors account for nearly two-fifths of much less than in many other countries. In their exports, resulting in merchandise exports Peru, political uncertainties in the run-up to declining by 21 percent (y/y) in the first half of 195 GLOBAL ECONOMIC PROSPECTS Table A1.3 Latin America and the Caribbean forecast summary (percent per year) Baseline forecast Growth rates/ratios 1991-2000 1999 2000 2001 2002 2003 2004-2010 Real GDP growth 3.3 0.1 3.8 0.9 2.5 4.5 3.9 Consumption per capita 1.5 -1.7 2.2 1.0 1.6 2.0 2.5 GDP per capita 1,6 -1.5 2.2 -0.7 1.0 3.0 2.6 Population 1.7 1.6 1.6 1.6 1.5 1.4 1.3 Inflationa 12.6 5.8 6.9 7.9 6.3 6.0 5.0 Gross Domestic Investment/GDPb 21.8 19.4 19.7 19.4 20.1 20.7 23.5 Central Gvt Budget Balance/GDP '-3.5 -4.4 -2.9 -3.2 -3.0 -2.5 -1.7 Export Volumec 8.4 5.7 9.7 2.6 4.2 9.5 6.7 Current Account/GDP -2.8 -3.1 -2.4 -2.8 -3.3 -3.2 -2.2 Memorandum Items GDP growth: LAC excl. Brazil 3.8 -0.4 3.4 0.5 2.3 4.7 3.7 Central America 4.4 4.3 2.7 1.0 2.2 4.0 3.8 Caribbean 34 5.7 5.5 1.4 3.0 4.2 4.0 a. Local currency GDP deflator; median. b. Investment ratio measure in real terms. c. Goods and non-factor services. Source: World Bank baseline forecast, October, 2001. the year. Caribbean countries also saw a reduc- in the United States, and world output in the tion in growth rates due to declining tourism latter part of 2002 and into 2003. revenues, especially in the latter part of the year. Revisions to external conditions, as well as domestic considerations, will impact the Near-term outlook growth prospects for countries differently. The The region's growth prospects for 2002 have expected delay in the U.S. recovery will have dimmed in light of a significant worsening of the most significant trade impacts in Mexico the external environment over the past six and the Central American and Caribbean months, and especially since the September 11 countries. For many of these countries, export- terrorist attacks in the United States. The re- processing zone (maquilas) exports destined gion's GDP is now expected to grow by 2.5 mainly for the North American market are a percent in 2002-1.9 percentage points lower significant proportion of total exports (30 per- than the spring forecast-provided that those cent of net exports in Costa Rica and El Sal- countries currently under financial stress are vador, for example). Moreover, remittances are able to avoid debt-service defaults. The delay in also likely to decline at a time when many Cen- the U.S. recovery, weak global output and trade tral American countries are facing weak coffee growth, a continuation of soft non-oil com- prices (after a four-year decline) and the effects modity prices and falling oil prices, and the of a severe drought in 2001. Weakness in labor likelihood of reduced capital flows to develop- and equity markets in the United States and in- ing countries underpin the moderate growth re- creased risk aversion to air travel will adversely covery for next year. (As a consequence, there impact tourism receipts-which are extremely is great uncertainty surrounding the forecast important for Caribbean countries. Prelimi- with more negative or positive responses of nary estimates indicate that loss of tourism rev- consumers and investors possible.) In 2003, enues could reduce these countries' GDP by GDP is expected to grow by 4.5 percent, re- 1.5 to 5 percent with potentially damaging so- flecting the expected rapid growth momentum cial impact in light of high unemployment in 196 REGIONAL ECONOMIC PROSPECTS the region. Argentina and Brazil are likely to strong trade ties to Argentina and Brazil, limit- be more hurt from disturbances in capital mar- ing their growth prospects to the fortunes of kets (if they were to be prolonged) than from those countries. direct trade impacts, due to weaker global ac- tivity. This reflects their high public and private Long-term prospects debt and large current account deficits (nearly Per capita GDP growth over the long term 3 percent of GDP for Argentina, about 5 per- (2004-10) is projected to average 2.6 percent a cent for Brazil). Although lower U.S. interest year, a full percentage point higher than what rates will help to alleviate debt-service pay- the region achieved in the 1990s. Key factors ments, risk perceptions have remained ele- supporting the cautious optimism for growth in vated-partly due to the market view that debt the 2000s compared with the 1990s include im- restructuring for Argentina may be necessary, provements in: (a) human capital (health, edu- as occurred in Ecuador in 1999-and have cation, and literacy indicators have all im- kept capital market flows subdued, reducing proved over the course of the 1990s, although the ability of these countries to roll over debt. much remains to be done in this area); (b) In Argentina, these factors are likely to keep macroeconomic management leading to greater the recovery modest. In Brazil, contagion from domestic macroeconomic stability (inflation events in Argentina (despite a $15.58 billion rates have fallen over the 1990s, for example, IMF-led package) is reducing the room for although they are still more volatile than in countercyclical policies. In addition, presiden- other regions); (c) investment climate attracting tial elections due next year could be another FDI; and (d) progress on deepening trade inte- factor restraining a return of investor confi- gration with the regional and global economies. dence and the acceleration of growth. FDI as a share of region-wide GDP rose As oil prices soften in 2002-03, the adjust- from less than 1 percent at the beginning of the ment that oil exporters will have to undergo 1990s to nearly 4 percent in 2000, with a sig- will be difficult and growth-restraining. Vene- nificant share going into telecommunications; zuela, for example, used buoyant oil revenues this represents benefits to the economy that are to finance growth in 2001, resulting in the non- likely to accrue in the next decade. Regulation financial-public sector's fiscal balance shifting and supervision of financial sectors have been from a surplus of 2.9 percent in 2000 to a strengthened, and trade regimes have been lib- deficit of about 3.1 percent of GDP. In con- eralized, with trade doubling as a proportion of trast, Ecuador may avoid a contraction in GDP over the last 10 years. These develop- growth in 2002, because oil revenues may re- ments have contributed to a large rise in total main high with expanded output partially off- factor productivity, from negative growth in the setting the expected decline in oil prices. 1980s to nearly 1 percent a year in the 1990s. Colombia's prospects hinge increasingly on fis- In the 2004-10 period, TFP growth is expected cal deficit reduction and on progress in the to remain in the 1 to 2 percent range, while im- peace process, but growth prospects will re- provements in the investment climate-includ- main subdued with the expected weak oil and ing strengthening the financial sectors through coffee prices. In Peru, the new administration better supervision and regulation-could con- will face tension between containing the fiscal tribute another 1 percentage point to regional deficit and reactivating growth quickly to re- growth. duce the danger of popular discontent, which could lead to political and social instability. Risks However, the combination of weak metals The region remains vulnerable in a number of prices, delayed FDI flows, and limited access to areas however. First, national saving rates re- capital markets could delay the economic re- main low in many countries, resulting in a per- bound. Bolivia, Paraguay, and Uruguay all have sistent dependence on foreign savings (of about 197 GLOBAL ECONOMIC PROSPECTS 3 percent of GDP)-typically from volatile pri- interest rates and severe economic disruption in vate capital markets. These markets have the wake of the financial crisis, which erupted demonstrated their power in delivering severe in late 2000 and early 2001. Second, there has external shocks to developing countries, and been a pronounced moderation of growth in the the region has had to endure at least two such Russian Federation, Poland, and the former Yu- episodes in the 1990s (Mexico in 1995, and goslav Republic of Macedonia (FYR Macedo- Brazil in 1999). The case of Argentina is still nia). In the Russian Federation, the impetus developing, and will obviously impact risk per- behind exceptionally strong growth of over 8 ceptions in the region for some time. percent in 2000 (generated from a combination Second, the prevalence of large debt over- of high oil prices and import substitution, driven hangs (both in the public and the private sec- by devaluation) is receding. In Poland high in- tors) in countries throughout the region re- terest rates, aimed at containing inflation, have quires rollover on a continuing basis. Although stymied demand. In FYR Macedonia, the mili- the region's debt-to-GNP ratio is in line with tary conflict with the Albanian rebels, which the average of developing countries, the debt- began in March 2001, has clearly begun to take to-exports ratio is very high. This exposes some its toll on the budget and on economic activity. countries to exogenous shocks emanating from Third, the slowdown in global demand in 2001, global capital markets, which are at times inde- particularly in the European Union (EU), has pendent of domestic considerations. had a negative impact on growth in the ECA re- Third, trade integration is incomplete with gion, in contrast to 2000 when external demand ratios of trade-to-GDP remaining low by inter- acted as a strong engine for growth. national standards (Chile, Mexico, and small Countervailing some of these negative pres- economies are exceptions), and diversification sures on regionwide growth, domestic demand of exports is still limited-many countries are has strengthened in a number of countries (such still commodity dependent. as the Czech Republic, Hungary, Romania, and Finally, the region still lags behind its po- the Slovak Republic). Similarly, strong growth tential in financial deepening (which could in domestic demand, particularly in private con- help raise national saving rates), infrastruc- sumption, stimulated by an increased money ture, and quality of institutions-areas which, supply through large hard currency inflows, if improved, can propel high and sustainable among other factors, is providing a buffer to growth rates. Many countries in the region the slowdown in the Russian Federation. Within have made strides in addressing some of these the Commonwealth of Independent States (CIS) areas and, should investor sentiment toward subregion, strengthened domestic demand in emerging markets improve significantly, the the Russian Federation in 2001 has translated region could grow at a faster pace than in the into a significant firming of import demand and current forecast. has provided a boost to growth in a number of countries that export to the Russian Federation (for instance, Ukraine). In contrast, export sec- Europe and Central Asia tors in a few countries with significant revenues from Turkey, (for instance, Bulgaria and Geor- Recent developments gia) are expected to be impacted by the plunge R eal gross domestic product (GDP) in Turkish import demand. growth for the Europe and Central Asia For most countries in ECA, current account (ECA) region is projected to decelerate deficits are forecast to stay at 2000 levels or to markedly in 2001 to about 2.1 percent, down deteriorate in 2001, although they should re- from 6.3 percent in 2000. This rapid slowdown main manageable. In the few cases where there is dominated by three main factors. First, in are current account surpluses, they are ex- Turkey domestic demand collapsed due to high pected to narrow. In some countries the cur- 198 REGIONAL ECONOMIC PROSPECTS Figure A1.4 Russian imports and partner exports in 1998-2001 (3-month moving average, y/y percent change (of US$ merchandise trade) 125 75 25 -75 -125 -175 N qN5 N N N'b N N Russia - Imports CEECs (excl. Turkey) - Exports -6- Other CIS - Exports Turkey - Exports Source: IMF. rent account deficits are already quite large or other countries (for instance, Hungary and are growing rapidly relative to GDP (such as in Poland) subsequently came under considerable Poland and Romania). For countries such as downward pressure during July 2001, when Poland, with an already high current account international investors became more bearish deficit, the EU slowdown will be felt more di- on emerging market instruments. In contrast, rectly, although the sharp deceleration in do- the Russian ruble has remained relatively firm mestic demand there will reduce imports. and generally appreciated in real terms over While the Russian Federation is expected to the year, bolstered, in particular, by a large cur- post a large surplus again for 2001, it will be rent account surplus. Elsewhere in ECA, due in significantly below the record $46 billion sur- part to fixed currency regimes and inflation plus in 2000. In Turkey the current account is differentials, the Bulgarian (currency board) expected to post a sizeable surplus due to a and Baltic (pegged) currencies have continued sharp contraction in imports and strengthening to appreciate. exports stimulated by the massive devaluation Inflationary pressures in the ECA region on of the Turkish lire subsequent to the abandon- the whole were relatively contained in 2001, ment of the crawling-peg regime in February with the general rate of increase either declin- 2001. ing somewhat or remaining flat. Turkey, with Real foreign exchange rates throughout the the consumer price index running at about 55 region remained on a broadly stable path over percent in 2001, is an important exception. the first half of 2001. The most notable excep- Until domestic markets stabilize there, height- tion is the sharp devaluation of the Turkish lire ened uncertainty will contribute to higher in- of about 60 percent in nominal terms, or about flationary pressures, as will the hefty increase 30 percent in real terms, as of August 2001, in the cost of imports that will likely generate year over year (y/y). The currencies of some significant pass-through effects. Driven in most 199 GLOBAL ECONOMIC PROSPECTS Table A1.4 Europe and Central Asia forecast summary (percent per year) Baseline forecast Growth rates/ratios 1991-2000 1999 2000 2001 2002 2003 2004-2010 Real GDP growth -2.3 1.8 6.3 2.1 3.0 4.2 3.6 Consumption per capita -3.5 -2.9 4.2 3.8 3.0 2.9 3.9 GDP per capita -2.5 1.7 6.1 1.9 2.9 4.1 3.5 Population 0.2 0.2 0.1 0.1 0.1 0.1 0.1 Inflationa 347.1 7.3 7.5 7.5 5.9 5.4 4.3 Gross Domestic Investment/GDPb 23.6 18.0 19.0 19.4 19.5 19.7 24.3 Central Gvt Budget Balance/GDP -19.0 -10.5 -7.4 -7.5 -7.2 -6.4 -4.8 Export Volumec 0.5 -1.4 11.1 8.5 2.8 8.3 5.9 Current Account/GDP -0.6 0.0 1.9 1.2 -0.4 0.0 -1.4 Memorandum Items GDP growth: Transition countries -3.1 3.3 6.1 4.0 3.1 3.8 3.4 Central and Eastern Europe 0.8 2.3 3.9 2.8 2.9 4.3 4.3 CIS countries -5.2 4.1 7.8 4.9 3.2 3.5 2.6 a. Local currency GDP deflator; median. b. Investment ratio measure in real terms. c. Goods and non-factor services. Source: World Bank baseline forecast, October, 2001. cases by an accommodating fiscal stance, in- hydrocarbon exporters of the CIS, this scenario flation remains at double-digit levels in a hand- implies a further slowdown in growth in 2002. ful of other ECA countries, for example in Be- For the ECA region oil-importers, the decline larus, Romania, Tajikistan, and Uzbekistan. For in the energy bill is expected to partially offset the region's oil-importing countries, the recent the negative impacts of a less favorable external pass-through impact of higher energy prices has environment. If indeed Turkey stabilizes and begun to diminish. In contrast, strengthening begins to recover in 2002, which is an assump- domestic demand in a number of ECA countries tion underlying our forecast, it will lift aggre- could lead to higher inflationary pressures. gate growth for the region. Throughout the region, access to foreign pri- Near-term outlook vate capital (including foreign direct invest- The severity and duration of the current slow- ment, FDI) is expected to remain more difficult down in the EU, along with policy responses over the near-term, due to increased aversion in the transition countries, will be important to emerging markets by international inves- factors for near-term prospects. In the EU, a tors. Correspondingly, domestic and foreign in- recovery is not expected until the second half vestment in the ECA economies is expected to of 2002, and much stronger external demand decelerate through 2002, in part reflecting an- from the EU is not anticipated until 2003. This ticipated delays in privatization programs. is especially significant for the Central and Tourism, an important source of foreign cur- Eastern European countries (CEECs), because rency in a number of ECA countries (such as their economies have become well integrated Croatia and Turkey), is also projected to slow with the EU. Another important near-term as- markedly. sumption is that the combination of slowing In sum, over the near term (2002-03), world energy demand and an accommodating growth is expected to stabilize at close to 3.5 stance by the Organization for Petroleum Ex- percent for the region as a whole. At the sub- porting Countries (OPEC) will likely translate region level, we are forecasting a pattern of into lower nominal and real oil prices. For the diverging growth becoming manifest in 2003. 200 REGIONAL ECONOMIC PROSPECTS For the CEECs, aside from anticipated stronger tinue to boost FDI into the subregion, although external demand in 2003, the EU accession as privatization programs wind down, this is process is expected to stimulate a continuation expected to diminish somewhat. These flows of reforms and to further boost growth. In con- have largely financed the subregion's shortfall trast, CIS growth is expected to slow in 2002 in domestic savings. Domestic savings rates are and to remain generally flat in 2003 as energy forecast to increase over the forecast horizon as prices stabilize at lower levels, and the boost FDI inflows moderate, but they are not ex- from high oil rents winds down in a policy en- pected to increase sufficiently to close the gap vironment of gradual reforms. As a conse- over the forecast horizon. This potential imbal- quence, import demand from the Russia Feder- ance between savings and investment exposes ation is expected to decline, which is in turn the CEECs to the risk that investment demand expected to result in lower export volumes for will be bridled by inadequate domestic savings the smaller CIS countries. or by a sudden drop in foreign inflows. Never- theless, prospects are broadly positive as most Long-term prospects of the countries of the subregion have achieved Over the coming decade through 2010, GDP a significant degree of stability and realignment growth for the ECA region is forecast to aver- of institutions and markets over the last decade age close to 4 percent, in contrast with the 2.3 and are on a path to continue the process. The percent region-wide average rate of contraction CEEC subregion growth forecast of just over 4 witnessed during 1991-2000, the first decade percent over the long term, albeit not insignifi- of transition. From a region-wide perspective, cant, suggests only slow convergence with EU the main drivers of higher growth are an im- per capita income levels. proved policy environment and a greater de- As with the CEECs, high educational attain- gree of macroeconomic stability leading to ment provides a strong positive contribution to higher investment and savings rates as a share growth potential in the CIS. However, invest- of GDP. Growth for the CEEC subregion is ex- ment in human capital in the region has de- pected to average above 4 percent during the clined substantially following the breakup period 2001-10, up significantly from close to of the Soviet Union and in the wake of the 1 percent posted during 1991-2000. Growth in 1997-98 crisis. Should a turnaround in the in- the CIS subregion is expected to average some- vestment in human capital not materialize, an what below 3 percent, also a marked increase important positive dynamic of the subregion's compared to the sharp contraction of about 5 growth picture will deteriorate further. The re- percent annually registered during 1991-2000. cent surge in growth in the CIS subregion of hy- In the CEECs, during the second decade of drocarbon exporters has created an important transition, a number of factors are contributing opportunity to introduce reforms more actively. to the anticipation of stronger growth perfor- The Russian Federation is an example of where mance, including rising investment as a share of this process has begun, especially during 2001. GDP and continued restructuring of the capital However, there the implementation process is base. Broad-based reforms and a well-educated just being initiated, and much remains uncer- labor force have been-and are expected to re- tain. Significant institutional and structural im- main-important factors contributing to fruit- pediments remain constraints to growth. Con- ful returns on rising investment. sequently for the CIS countries as a group-and Almost all of the CEECs are EU accession in contrast to the CEECs-investment as a candidates, and have significantly benefited share of GDP is expected to remain at relatively from the EU accession process, which has pro- low ratios, after having declined during the vided an incentive to address underlying struc- 1990s. Also, considerable excess capacity re- tural and institutional impediments to growth. mains, though much of it could be obsolete, so The EU accession process is expected to con- investment demand could kick in sooner if the 201 GLOBAL ECONOMIC PROSPECTS economy picks up. If good policy reforms are sion process could slow the reform process and introduced more aggressively, then the CIS econ- undermine long-term growth prospects within omies could shift to a higher growth path. the CEECs. Potential output could be increased if re- Risks form programs in the CIS were to move for- Over the near- to medium-term, risks to the ward more aggressively than anticipated. In forecast are predominantly on the downside. the case of the Russian Federation, this would Within the region the main risks include the generate positive demand dynamics through- possibilities of a deepening of the crisis in out the CIS and in Turkey. Depending on dy- Turkey or a sharper economic slowdown in the namics both internal and external to the re- Russian Federation, or both. In either case, gion, there is the upside risk that the EU growth prospects in smaller economies of the accession process will regain stronger positive region would also decline. Another internal risk momentum and proceed more smoothly and factor is an escalation of political tensions and more rapidly than currently envisioned. No- instability in the Balkans. The September 11 tably, the recent terrorist crisis could act as a terrorist attacks have increased both external catalyst to strengthen political resolve in both and internal risks. With regard to the former, the EU and applicant countries to move for- there is the possibility of greater risk aversion ward with the accession process. Turkey, the to emerging markets and capital flight. Regard- Russian Federation, and Central Asian coun- ing the latter, a risk of increased political uncer- tries might also benefit from strengthened po- tainty is an important factor, especially in the litical backing from the west and a possible countries of Central Asia, due to the heightened increase in official assistance as a reward for conflict and instability in Afghanistan. There supporting U.S.-led strikes into Afghanistan. could also be an influx of refugees to the ECA countries bordering Afghanistan, namely Tajik- istan, Turkmenistan, and Uzbekistan. Sub-Saharan Africa Other external risks are mainly associated with the EU, both in terms of its growth Recent developments prospects and with public support for the ac- rowth in Sub-Saharan Africa (SSA) cession process. A stronger and more pro- slowed to 2.7 percent in 2001 from 3 tracted decline in external demand from the EU percent in 2000, interrupting a pro- would add pressure to external balances and gressive recovery from the slowdown of the likely reduce growth outcomes, particularly in late 1990s. With population growing at 2.4 the CEECs. An important aspect of an ex- percent, the rise in per capita GDP was mini- tended slowdown in the EU is the timing-that mal. The slowdown was widespread through- is, coinciding with important EU accession ne- out the region, in East, West, and Southern gotiations-because it will likely reduce ma- Africa, and in both oil and non-oil commodity neuverability for both candidate countries and exporters. existing members. Correspondingly, support The primary cause was the slowdown in for the EU accession process (both within the developed countries. In the face of weaker existing EU countries3 and within prospective demand from the United States and the Euro member countries) has been diminishing. This Area, merchandise exports managed just 3.4 could become a higher risk over the near term percent growth in volume terms compared to because more difficult issues-such as the free 8.8 percent in 2000. Services exports, including movement of labor and capital, agriculture, tourism, were also affected, growing by 3.6 and the distribution of structural funds-are percent. Commodity prices remained well now shifting to the front burner in enlargement below levels of the late 1990s, including those negotiations. Extensive delays in the EU acces- that rebounded from recent lows. Beverage 202 REGIONAL ECONOMIC PROSPECTS Figure A1.5 Real GDP growth of SSA oil and non-oil exporters (percent) 4.5 4.5 05 Oil E Non-o0il 4 3.5 - 2.5 2- 1.5- 0.5 - 0 2000 2001 2002 2003 Source: Economic Policy and Prospects Group. producers were particularly hard hit, with cof- that the need for food aid will be unchanged fee prices down over 25 percent from 2000 and from last year at around 2.7 million tonnes cocoa prices-although they were up around (FAO 2001). Weather also contributed to a 12 10 percent in 2001-only 75 percent of the av- percent reduction in the cocoa crop in West erage for 1995-2000. While oil prices eased Africa after the bumper harvest of 1999-2000, back from their peak of nearly $30 a barrel in according to the International Cocoa Organi- mid-2000 they remained strong, and oil ex- zation (African Business, July/August 2001). porters outperformed the region as a whole, In the political sphere, some progress to- growing at an average of 3.6 percent for the ward stability was achieved in the Democratic year, compared to 2.6 percent for non-oil ex- Republic of Congo, Guinea, and Sierra Leone, porters. Oil constitutes less than a third of SSA but peace seemed as elusive as ever in Angola, exports, however, and net energy exports are Liberia, and the Sudan, and Zimbabwe's crisis only 5 percent of GDP. Thus on balance, recent intensified with the approach of elections in world commodity market trends represented a spring 2002. Countries in conflict or experi- major drag on growth and incomes. encing severe governance problems4 recorded Apart from the external environment, de- the worst performances, growing at -0.4 per- velopments within the region painted a mixed cent in 2001. On the plus side, robust growth picture. Better weather boosted agricultural pro- continued in a number of countries, including duction and household incomes in a number Ethiopia, Madagascar, Mozambique, and of countries in East and Southern Africa, in- Uganda, reflecting better policy and economic cluding Ethiopia, Kenya, Mozambique, and management. Finally, 19 countries reached de- Tanzania. However, localized drought condi- cision points under the enhanced Heavily In- tions persisted in these and many other coun- debted Poor Countries Initiative, cutting debt tries. In Southern Africa, food production fell servicing costs by a third, and relaxing balance by as much as 25 percent, due to both adverse of payments and budgetary pressures. weather conditions and civil disturbance. In South Africa, the region's largest econ- Overall, the Food and Agriculture Organiza- omy, a robust recovery in the second half of tion of the United Nations (FAO) estimates 2000 dissipated in the first half of 2001 as in- 203 GLOBAL ECONOMIC PROSPECTS adequate rains led to a disappointing maize har- ity prices should firm on average, even though vest. The impact spilled over from agriculture non-oil exporters' terms of trade deteriorate into manufacturing and, on the demand side, slightly because of higher import prices. The into consumer spending, and growth slowed to modest improvement in the external environ- 2.4 percent. Both public and private investment ment will raise non-oil exporters' growth to 2.7 remained strong, as did productivity growth, percent from 2.6 percent 2001. For the SSA re- although the investment rate at only 16 percent gion as a whole in 2003, the forecast antici- of GDP remains well below the level needed pates a strong acceleration in export volume to support adequate employment growth. The growth to 6.4 percent, pushing GDP growth rand came under strong selling pressure in the to 3.9 percent. With decent rains, the actual second half of the year as a result of ongoing outcome might be even better. Nevertheless, uncertainty about emerging markets generally terms-of-trade weakness is expected to persist and the situation in Zimbabwe specifically. through the forecast period, especially for oil In Nigeria, the energy sector registered exporters, as oil prices fall further to below $20 strong gains, thanks to both oil and natural gas a barrel. revenues and to keen investor interest, particu- Despite weak energy prices, substantial in- larly in the offshore sector. However, it is in- vestment in oil exporters promises to sustain creasingly evident that progress on reforms to real growth in oil sectors in the medium term. date has had little impact on the non-oil econ- Nigeria has struggled recently to meet OPEC omy. A one-year, $1 billion standby credit from quotas, but plans to increase capacity signifi- the IMF was extended from August to October cantly over the next few years and a second liq- despite the government's failure to meet impor- uid natural gas train at Bonny Island will boost tant conditionalities, but especially with the ap- production by 50 percent beginning in 2002. proach of elections in late 2002, the future of Meanwhile, recent offshore discoveries could the reform process is uncertain, substantially raise medium-term production and exports for non-OPEC Angola and Equa- Near-term outlook torial Guinea. Even in the near term, ex- While many idiosyncratic factors will bear on ploration and development activity-including near-term performance, the slowdown in indus- the Chad-Cameroon pipeline project-is help- trial countries during 2001 and sluggish recov- ing to offset terms-of-trade losses, keeping real ery in the first half of 2002 virtually guarantee growth higher than otherwise would have been a poor out-turn for the coming year. Weak de- the case. For non-oil exporters, faster world mand will continue to depress export prices growth will tighten the supply demand balance and volumes. However, as recovery consoli- in primary commodity markets allowing export dates in OECD trade partners, demand for the prices and terms of trade to strengthen. In ad- region's exports will strengthen setting the stage dition to the rebound in the world economy for stronger gains in 2003. For the region as a generally, export prospects will also benefit whole, merchandise exports are expected to from a number of specific trade initiatives, in- grow by only 2.9 percent in 2002, while terms cluding the United States' Africa Growth and of trade fall by 6.2 percent, equivalent to 1.8 Opportunities Act (AGOA), the EU's "Any- percent of GDP. The subdued external perfor- thing but Arms" initiative, and the EU-South mance will hold GDP growth to 2.7 percent for Africa Free Trade Agreement. Early evidence a second year, again leaving per capita incomes from the first half of 2001 shows that 13 SSA flat. Oil prices are expected to fall to $21 a bar- countries benefited from $3 billion of exports rel in 2002, implying steep terms-of-trade losses under AGOA preferences (USTR 2001). None- for oil exporters of 4.1 percent of GDP; their theless, SSA's medium term performance will real growth will average 3.1 percent, down from remain subdued as a result of inelastic export 3.6 percent in 2001. However, other commod- demands and a lack of diversification. 204 REGIONAL ECONOMIC PROSPECTS Table A1.5 Sub-Saharan Africa forecast summary (percent per year) Baseline forecast Growth rates/ratios 1991-2000 1999 2000 2001 2002 2003 2004-2010 Real GDP growth 2.2 2.5 3.0 2.7 2.7 3.9 3.7 Consumption per capita -0.6 0.0 0.4 0.2 0.5 0.9 1.1 GDP per capita -0.4 0.0 0.5 0.3 0.3 1.6 1.5 Population 2.6 2.4 25 2.4 2.4 2.3 2.2 Inflation' 9.7 5.3 6.3 6.0 5.0 4.5 4.1 Gross Domestic Investment/GDPb 17.4 17.1 17.2 17.5 17.6 17.8 18.4 Central Gvt Budget Balance/GDP -7.4 -8.1 -2.2 -3.4 -3.3 -3.2 -2.8 Export Volumec 4.3 3.0 7.0 3.4 2.4 7.6 6.3 Current Account/GDP -2.1 -1.2 -1.5 -1.0 -2.4 -2.0 -1.8 Memorandum Items GDP growth: SSA excl. South Africa 2.6 3.0 2.9 3.0 2.8 4.0 4.2 Oil exporters 2.7 2.6 3.8 3.6 3.1 3.4 3.6 CFA countries 2.6 2.4 2.7 2.4 2.9 3.6 3.8 a. Local currency GDP deflator; median. b. Investment ratio measure in real terms. c. Goods and non-factor services. Source: World Bank baseline forecast, October, 2001. Long-term prospects vate foreign capital inflows will limit investment Over the long term, the expectation is for a rates to an average of below 19 percent of GDP. continuation of the trend toward better eco- Although up from barely 17 percent currently, nomic policies and management and a broadly this is far from what is needed. As a result, cap- favorable external environment. Internal mar- ital accumulation will contribute less than 1 ket reforms, deregulation, and privatization percent annually to growth-not even a quarter have raised productivity and improved incen- of the rate anticipated for East Asia. Low rates tives, and encouraged nontraditional exports of human capital investment and slow progress such as fish and horticulture at a time when on rebuilding infrastructure will hold produc- prospects for many traditional crops are poor. tivity growth to around the same rate. Notably a number of well-managed reformers Despite the somewhat pessimistic outlook, if have sustained high growth even through diffi- the forecast is accurate the coming decade will cult external conditions. In the baseline sce- see the region's best sustained performance nario, which assumes a continuation of current since the 1960s. There are manifold reasons for productivity trends, output growth averages SSA's historically poor performance-disease, 3.7 percent from 2004-10. With population civil strife, poor governance, inauspicious cli- growth falling to 2.2 percent, real per capita mate, low savings and investment, and falling income growth will average 1.5 percent, reach- terms of trade. Some of these conditions are ing $640 in real (1995 dollars) terms by 2010. unlikely to change any time soon, but for oth- For many countries, export diversification and ers there are signs of real improvement. Politi- favorable price trends will sustain performance cal and economic reforms have gained pace well above the regional average. since the mid-1980s, and are contributing to This performance will fall short of what is higher standards of governance and economic needed to achieve the international develop- management. Private sector growth and in- ment goals, and SSA will continue to lag behind creasing regional integration are helping to other regions in the developing world. Low do- boost efficiency and rationalize production. mestic savings combined with only modest pri- Greater openness and debt relief are relaxing 205 GLOBAL ECONOMIC PROSPECTS balance of payments constraints, easing import bilities. Government revenues have also bene- restrictions, and over time encouraging more fited from high oil prices. Many oil exporters foreign investment interest. But even as some achieved balanced budgets or surpluses in countries notch up high growth rates, overall 2000, and some of that momentum has con- performance will continue to be constrained by tinued in 2001. Governments did spend more the devastating effects of HIV/AIDS, slow prog- than previously budgeted from their revenue ress on governance in some countries, and the windfall but most were relatively restrained, limited availability of resources to rehabilitate given the expenditure profiles of earlier windfall productive capacity and infrastructure. gains. For example, the Saudi government re- ceived 58 percent more revenue than budgeted in 2000 but only spent approximately 10 per- Middle East and North Africa cent more than planned, with much of the extra spending being used to pay domestic arrears. Recent Developments The diversified economies grew 3.4 percent evelopments in the Middle East and in 2000, lower than their historical average. North Africa were strongly positive in Drought conditions in Morocco, Tunisia, and 2000, with a rare convergence of si- the Levant contributed to lower production and multaneous increases in oil prices and export agricultural exports, despite high export market volumes contributing to stronger-than-antici- growth in 2000. Additionally, domestic condi- pated growth of 3.9 percent. Growth in 2001 tion in the Arab Republic of Egypt deteriorated will be lower at 3.4 percent, as declines in significantly as the budget and current account OPEC export quotas affect oil production and deficit increased, placing pressures on interest increasingly weak growth in industrial coun- rates, exchange rates, and domestic investment. tries affects demand for goods and services GDP growth in the diversified exporters will from the region. Short-term prospects have rise to 4.4 percent in 2001. Morocco, which weakened considerably since September 11 in had a partial recovery from drought this year, the face of a slowdown in external demand, accounts for the increase in GDP growth. Addi- with economic recovery in Europe and the tionally, stronger oil prices and a relief from United States delayed into 2002. drought are behind increased growth in Syria. A The oil exporters have reaped the benefits of weaker external environment, particularly in higher demand and disciplined adherence to Europe, has affected trade prospects with fall- OPEC quotas, boosting both export volume ing growth expected in most countries as export and revenue growth in 2000, with GDP growth market growth fell from 13 percent in 2000 of over 4 percent in several countries. Export to 1.9 percent in 2001. Workers' remittances, volume growth is weaker in 2001 because tourism, and services receipts will be similarly OPEC quotas were reduced throughout the affected. year in an effort to target supply around a price of $25 a barrel, with growth falling to 3.1 per- Near-term outlook cent. The boost in revenue has fostered income Looking forward, GDP growth in the region is gains and led to strong growth in domestic de- expected to fall to 2.9 percent in 2002 and to mand through stronger consumption and im- recover to 3.6 in 2003. The sharper downturn port growth. Current account surpluses rose to in industrial countries and the delayed recovery 14.9 percent of GDP in 2000 and 8.4 percent in into mid-2002 will reduce the external impetus 2001. Domestic interest rates fell, and there to growth. Slower world demand growth will was an increase in investment in the oil and keep oil prices at the lower end of the OPEC non-oil sectors, with several countries also ben- price band (around $21 a barrel) and produc- efiting from higher foreign investment. Oil ex- tion and income growth will be adversely af- porters have had few problems refinancing lia- fected. The diversified exporters face lower 206 REGIONAL ECONOMIC PROSPECTS Figure A1.6 Tourism and workers remittances as a share of GDP in 2000 (percent) 25 20.6 U Workers Remittances U Tourism 20 15 13.5 10 7.0 6.9 5.5 5 52.9 3.6 Egypt, Arab Jordan Morocco Tunisia Rep. Source: IMF Balance of Payments. trading partner-import growth, and adverse bilization fund will be used to finance deficits impacts on tourism from lower external income and retire debt. However, if fiscal policy be- growth. Growth will probably fall in 2002 to comes too expansionary, it will be difficult to 4.2 percent, but will recover along with the oil maintain the lower interest rates and inflation exporters in 2003, if, as anticipated, Europe that have been apparent recently. and other trading partners gather momentum. The windfall gains have provided opportu- The momentum of growth in the oil ex- nities for several oil exporters, particularly the porters will slow in 2001-02, as weaker global Islamic Republic of Iran and Algeria, to amor- growth affects energy demand and OPEC keeps tize external debt and retire domestic debt. Oil a tight rein on oil production quotas. Produc- exporters in the Gulf have built up foreign tion and export volumes in 2002 are expected reserves and had few problems maintaining to be lower than 2001 levels, thus ensuring that their fixed exchange rates. The Islamic Repub- export volumes and GDP growth will decline lic of Iran appears on-track to unify its ex- from 2000 and 2001 rates. As the oil price falls change rate regime in the 2002/03 fiscal year; to $21 a barrel in 2002, the terms-of-trade as a result, it may face a dose of imported in- gains made over the last several years will de- flationary pressures in the forecast period. For cline, and current account surpluses, which most countries, however, inflationary pres- reached 14.9 percent of GDP in 2000, will fall sures should remain low. Interest rates in oil to 1.7 percent of GDP in 2002. Similarly, gov- exporters have been falling, along with rapid ernment balances will show some deterioration, growth in liquidity; therefore, there will be both as oil revenues fall and governments im- continued support to growth from domestic plement new expenditures. The Islamic Repub- demand as demand for oil softens. lic of Iran is locked into a balanced budget rule, Short-term prospects in diversified export- and, with the conservative oil price assumptions ers are mixed. Growth is expected to average used for budget purposes, should retain fiscal around 4.3 percent in 2002-03. Morocco and balances. Algeria is expected to increase fiscal the Syrian Arab Republic are expected to re- expenditures greatly in 2001-02, but the oil sta- cover from the debilitating droughts of recent 207 GLOBAL ECONOMIC PROSPECTS Table A1.6 Middle East and North Africa forecast summary (percent per year) Baseline forecast Growth rates/ratios 1991-2000 1999 2000 2001 2002 2003 2004-2010 Real GDP growth 3.2 2.2 3.9 3.4 2.9 3.6 3.3 Consumption per capita 0.4 0.3 3.1 1.7 0.7 0.8 0.9 GDP per capita 1.0 0.3 1.9 1.5 1.0 1.6 1.4 Population 2.2 1.9 2.0 1.9 1.9 1.9 1.9 Inflation $.2 3.5 3.4 4.5 4.5 4.0 4.0 Gross Domestic Investment/GDPb 22.7 22.4 23.4 23.9 24.2 24.3 25.4 Central Gvt Budget Balance/GDP -1.6 -2.7 -2.5 -3.0 -2.9 -2.6 -2.2 Export Volumec 5.8 13.1 6.2 3.0 4.0 5.7 4.6 Current Account/GDP -1.9 -1.0 8.1 4.7 0.8 -0.9 -2.3 Memorandum Items GDP growth: Oil exporters 2.6 1.8 3.3 2.6 2.3 3.3 2.7 Diversified exporters 4.0 33 3.4 4.4 4.2 4.3 4.3 a. Local currency GDP deflator; median. b. Investment ratio measure in real terms. c. Goods and non-factor services. Source: World Bank baseline forecast, October, 2001. years, and agricultural production and exports not fare well in the near term. The majority will provide support for growth in the near of tourists come from Europe and, with term. This will offset, to some extent, the ex- low-income growth in Europe into 2002, and pected external slowdown in demand in 2001- confidence eroded because of the events of 02. The slower activity in the European econ- September 11, tourism will suffer. Political omy in 2001-02 will adversely affect all the di- uncertainty may also contribute to a decline in versified exporters, with merchandise export tourism, particularly in the Levant and in Egypt. growth falling from 7 percent in 2000 to 1.9 This can already be seen in Egypt, where after percent in 2001, before recovering to 4.9 per- an almost 15 percent rise in tourist arrivals in cent in 2002. Jordan is enjoying a broad-based 2000, arrivals fell by 8.1 percent in April 2001 increase in activity, but deterioration in the ex- and 8.5 percent in May over the same period in ternal environment (tourism, remittances, and the previous year. Jordan's tourism receipts fell capital flows) could dampen growth next year. by 3.6 percent in the first half of 2001 com- Current account deficits, which widened in re- pared to a year ago. Tunisia and Morocco will cent years in drought-stricken countries, will suffer less from the effects of the political in- remain higher than previously anticipated due stability in the Levant. Indeed, Tunisia should to lower export volume growth. Fiscal policy in continue to experience some growth, and Mo- the drought-stricken countries has by necessity rocco is investing heavily in tourist infrastruc- been somewhat expansionary to counter de- ture. Remittances will remain stagnant or grow clines in agricultural incomes. Several countries very slowly as growth slows in the near term in that have signed EU Association Agreements oil exporters, and as income growth is damp- (such as Morocco and Tunisia) have lowered ened in Western Europe, the main source of re- or eliminated customs duties that were a source mittances for the region. of revenue, placing upward pressures on fiscal deficits. Consequently, the public sector will Long-term prospects continue to play a large role in growth. Long-term prospects in the Middle East and Tourism and workers' remittances, two of North Africa are less positive than in most other the mainstays for diversified exporters, will developing regions. Growth for the oil ex- 208 REGIONAL ECONOMIC PROSPECTS porters in the long term is expected to average political stability. Current events in Israel and 2.7 percent; in the diversified exporters growth the West Bank and Gaza have affected tourism, is expected to average 4.3 percent. In each case, not just in these areas but in the entire Levant growth is only slightly higher than the average and in Egypt. While countries such as Tunisia for the 1990s. Growth in 2004-10 for the re- and Morocco are making concerted efforts to gion is expected to average 3.3 percent, similar improve accommodations and service, they still to the average of the 1990s and lower than the face fierce competition from other destinations average of 3.5 percent for 2000-03. The rea- that have lower levels of political conflict and sons for the lack of acceleration of growth in better facilities and services. Remittances by the forecast period include the real long-term nationals working in the Gulf countries and in decline in oil and other commodity prices ex- Europe are also an important source of income, pected in the next 10 years; the high level of. but remittances from the Gulf are not expected vulnerability of countries in the region to com- to continue growing rapidly. The Gulf coun- modity price and other external shocks; and the tries have begun programs to increase the num- low level of attractiveness of the region to for- bers of their own nationals in their domestic eign investment outside commodity sectors. workforce, and this will certainly be at the ex- The main external sources of growth and in- pense of non-nationals. With slower growth come in the region come from commodity ex- expected in the oil exporters in the long term, ports, tourism, and workers' remittances. Each remittances are also expected to decline or to of these sectors is highly vulnerable. In the long- grow very slowly. term, energy prices are not expected to in- crease-in fact, the real crude oil price in 2010 is projected to be approximately half its 2000 Notes level. On the supply side, it is expected that non- 1. Dawn Internet Edition, Pakistan, September 20, OPEC supply will grow in the next several 2001, http://www.dawn.com. years, indicating that OPEC production and ex- 2. The News International, Pakistan, September 20, yeart s, dcin tht oPeC pod inand pies 2001, http://www.jang.com.pk ports must decline in order to maintain prices 3. Support for eastward expansion of the EU has above $20 a barrel, given expected demand con- waned markedly in existing member countries, which ditions. Without further diversification in oil- was highlighted most recently by Ireland's June 2001 exporting countries, many of which receive up "No" vote on the Treaty of Nice (which makes changes to 95 percent of export revenues from hydro- to the voting structure of the EU to accommodate carbons, the external impetus for growth seen expansion). since 2000 will diminish. The agricultural and 4. Angola, Burundi, C6te d'Ivoire, Democratic Re- public of Congo, Guinea-Bissau, Kenya, Sierra Leone, mineral exports of the diversified exporters and Zimbabwe. No reliable data are available for should fare better as prices increase in the next Liberia or Somalia. decade, but the recovery will be slight, and will come from the extremely low levels seen in re- cent years. In terms of agricultural exports, the References scope for increasing penetration of markets is FAO (Food and Agriculture Organization of the United limited by the restrictions remaining on agricul- Nations). 2001. "Food Supply Situation and Crop ture in the initial Association Agreements signed Prospects in Sub-Saharan Africa." No. 2 (August). with the EU by the Mediterranean countries. USITC (United States International Trade Commis- Tourism receipts are an important source of sion). 2001. "Sub-Saharan Africa: Major U.S. Im- port Suppliers Under the Generalized System of revenue for many of the Mediterranean coun- Preferences, and the African Growth and Oppor- tries but, as can be seen in the current context, tunity Act, Year-to-Date." (http://reportweb.usitc. they are highly vulnerable to issues concerning gov/africa/trade_data.html) 209 Appendix 2 Global Commodity Price Prospects Commodity prices declined in 2001, from 1997 to 2000 despite a decline of 53 per- however oil prices remain high relative cent in nominal prices over the same period. to non-oil prices (figure A2.1). Agricul- Currency devaluations, relative to the U.S. tural prices have yet to begin a sustained re- dollar, have also depressed prices of some com- covery from the declines that began just before modities--especially in countries with weak the Asia crisis in mid-1997, due to continued currencies that are also major commodity ex- large supply increases, weak demand, and cur- porters, such as Brazil, Indonesia, and Thai- rency devaluations of major commodity ex- land. For example, Brazil's currency has deval- porters relative to the dollar. Metals and min- ued about 50 percent relative to the dollar erals prices made a modest recovery from the since 1997, and this has led to lower dollar lows reached in 1999, but have since returned prices for its major agricultural exports-cof- to near those lows, due mostly to weak de- fee, soybeans, and sugar. Indonesia, a major mand. Oil prices rose sharply from their 1998 exporter of natural rubber and vegetable oils, lows due to cuts by OPEC producers, but have has seen its currency devalue 70 percent rela- weakened in the past year due to weakness in tive to the dollar since 1997. Thailand, the the global economy and, most recently, the ter- largest rice exporter, has seen its currency de- rorist attacks in the United States. value 30 percent relative to the dollar since One of the main reasons for the divergence 1997, which has sent rice prices lower. of oil and non-oil commodities is that com- The current decline in non-oil commodity modity producers have responded very differ- prices has been more severe than the two de- ently to the price declines following the Asia cri- clines of the 1980s. There are strong similari- sis. Cuts in crude oil production and exports by ties in all three periods, however (figure A2.2). OPEC producers starting in 1999 sent oil prices The current decline began in May 1996, and higher, while metals and minerals prices got a prices fell by 30 percent in 38 months, com- boost from cuts in mine and smelter output. pared to a decline of 27 percent in 32 months However producers of agricultural commodi- from the February 1980 peak, and a 23 per- ties were slow to adjust to low prices; this has cent decline in 37 months from the June 1988 contributed to continuing price weakness. Some peak. In all three cases, commodity prices agricultural commodities are still facing large reached their initial lows after about three year-to-year production increases despite the years and then rallied before returning to their nearly 32 percent decline in agricultural com- previous lows. Then the patterns diverge, with modity prices from 1997 to 2001. Global coffee prices declining for another year in one case production, for example, increased 21 percent and rising in the other. 211 GLOBAL ECONOMIC PROSPECTS Figure A2.1 Commodity price trends January 1997=100 160 120 80 Jan. 1997 Jan. 1998 Jan. 1999 Jan. 2000 Jan. 2001 Sept. 2001 - Agriculture - Crude oil - Metals and minerals Source: World Bank. percent increase in 2002. Thereafter prices are Figure A2.2 Non-energy commodity expected to rebound rapidly as extreme low prices prices curtail supplies and prices rise 8.1 per- Index (price peak= 00) Feb. 1980 cent in 2003. The increases are expected to be 110 -June 1988 below the recoveries of the 1980s (figure A2.2) May 1996 because of large surplus production capacity 100 Arelative to demand that exists in many com- modities; improvements in technology that have lowered production costs; and policies in so tfl many OECD countries that have insulated pro- ducers from declines in global prices. Agricul- 80 ,tural prices are projected to rise 1 percent in 2002 and 8.8 percent in 2003, while metals 70 and minerals prices are projected to rise 3.2 Forecast and 7.2 percent, respectively, during 2002 and 60 2003. Beyond 2003, we expect nominal non- 12 24 36 48 60 72 84 96 oil prices to continue to increase about 5 per- Months cent per year through 2005. Specific commod- ity price projections are contained in tables A2.12 and A2.13 for selected years to 2005, 2010, and 2015. Projected nominal and real The recent terrorist attacks and resulting commodity indices are given in table A2.14. economic slowdown is expected to delay the Oil prices are expected to fall to $21 a bar- recovery in non-oil commodity prices until the rel in 2002 compared to $25 a barrel in 2001. latter half of 2002 and result in a modest 1.6 However, the recent terrorist attacks in the 212 GLOBAL COMMODITY PRICE PROSPECTS United States have added substantial risk to real non-oil prices in the early 1970s and the the outlook, and prices will likely be more peak of oil prices in 1980, real prices of both volatile than previously expected. Prices are have declined by about two-thirds. expected to settle in the $18-20 range over the The structural decline in agricultural com- balance of the decade as recent high prices modity prices relative to manufactures appears stimulate new production capacity. to be the direct consequence of more rapid Over the forecast period to 2015, real non-oil productivity gains (see box A2.1) Such gains commodity prices are projected to remain about have been fueled by rising yields, improved constant relative to 2001 levels as nominal policies in developing countries, and invest- prices recover from current severely depressed ments in infrastructure and irrigation. Metals levels. In contrast, real oil prices are expected to and minerals costs have also declined due to fall 40 percent over the same period as prices re- improvements in technology, better manage- treat from current high levels. This divergent ment, and better policies. Demand growth for forecast for non-oil and oil prices reflects the ex- commodities has slowed in response to slower treme divergence in current prices rather than a population growth and declining income elas- fundamental difference in the long-term out- ticities. These trends are expected to continue look. The trend of real commodity prices of the and lead to continued declines in real com- last century are expected to continue, with both modity prices over the longer term. oil and non-oil prices declining relative to man- ufactures prices. During the twentieth century, non-oil commodity prices fell about 1 percent Agriculture per year relative to the prices of manufactures; gricultural commodity prices have been oil prices fell even more rapidly until the early the weakest component of commodity 1970s when OPEC's market power emerged prices, down 33 percent in 2000 compared to and supplies were curtailed. Since the peak of their 1995 highs. Box A2.1 Total factor productivity growth M artin and Mitra (2001),l in a cross-country Total tactor productfvity growth In agricutture study of nearly 50 countries for the period and manufacturing 1967-92, estimated total factor productivity (TFP) (Percen) growth for agriculture at between 2.3 percent and 2.9 ApicWt- Ma&i-9 percent per year (depending on the econometric speci- fication used) compared to 1.1 percent to 1.9 percent Overall TFP 2.3 to 2.9 1.1 to 1.9 for manufactures. The TFP growth was found to be Developed countries 3.4 to 3.5 1.9 to 3-3 Developing countries 1.8 to 2.6 0.6 to 0.9 faster in developed countries than in developing coun- Low incom countries 1.4 to 2.0 0.2 to 0,9 tries, for both agriculture and manufacturing, and Middle income countries 1.8 to 2.9 0.8 to 1.0 growth was faster in middle-income than low-income developing countries. The difference in TFP growth between agriculture and manufactures was most strik- role in accounting for the decline of agricultural prices ing for low-income developing countries, where the relative to manufactures. range of TFP estimates was 1.4 to 2.0 for agriculture compared to 0.2 to 0,9 for manufactures (table A2.1). Martin, W. and . Mitra (2001). "Productivity Growth in Agricul. Thus the greater gain in total factor productivity of ture verus Manufacturing." Economic Developoent and Cultural agriculture relative to manufactures has played a large Change, vol. 49, no. 2, January, pp. 403-422. 213 GLOBAL ECONOMIC PROSPECTS Beverages Colombia, C6te d'Ivoire, Indonesia, and Mex- The World Bank's monthly index of nominal ico are all expected to have large crops. beverage prices (comprised of the export value Low prices have been met with several re- weighted average of coffee, cocoa, and tea cent attempts to curtail exports by the Associ- prices) has declined 71 percent since the 1997 ation of Coffee Producing Countries (ACPC). highs, due mostly to steep declines in coffee So far these attempts have been ineffective and prices. all efforts have ended in failure. Current efforts Prior to 1998-99, coffee production and appear to lack an effective mechanism to con- consumption were relatively equal, with little trol coffee exports and have not yet inspired overall increase in either since the late 1980s. much market response. In addition, withhold- Since 1998-99, production has increased about ing stocks without reducing supplies encour- 20 percent, and arabica and robusta coffee ages sales outside of the agreement and under- prices have declined 66 and 63 percent, respec- mines the agreement. tively, from 1997 to the first nine months of The recent decline in coffee prices has been 2001. Despite these dramatic price declines, due primarily to a surge in supplies, but the production is expected to increase for the equally important longer-term problem for fourth consecutive year (see table A2.1). Prices coffee producers is weak demand. Per capita are not expected to recover until this imbalance consumption in Europe and the United States, is resolved. It is possible that coffee prices have which accounts for nearly 90 percent of inter- permanently shifted lower to accommodate in- national demand, has been declining. In the creased production by efficient producers. United States, for example, per capita coffee Cocoa and tea prices have not seen the sharp consumption has been declining since 1970, declines observed in coffee because supplies while per capita consumption of soft drinks have not increased as significantly. Cocoa con- has more than doubled. Unless tastes change, sumption has grown at a fairly steady 3 percent coffee producers will probably need to adjust per year over the past two decades, while global to slow-perhaps stagnant-demand growth. tea consumption has grown at a more modest 1 A significant recovery of coffee prices is not percent per year (see table A2.2). expected soon unless there are major supply disruptions due to droughts or frosts, which oc- Coffee curred in 1994 and 1997. We project a modest Brazil, the largest coffee producer with about recovery in robusta prices beginning in 2002 30 percent of the world's total, is expected to and arabica prices in 2003 (table A2.12-13 for have a near-record crop, while Vietnam, the specific price forecasts), but we also recognize second largest producer, is expected to have a the risk that prices could drift lower until sup- record crop. Other major producers such as plies are sharply reduced. Over the longer-term, Table A2.1 Coffee production (million bags) 197-98 1998-99 1999- 2000 2000-01 2001-02 Brazil 22.8 35.6 30.8 34.1 33.7 Vietnam 6.9 7.5 11.0 113 12.5 Colombia 12.2 10.9 9.3 11.5 11.4 Indonesia 7.8 7.0 6.5 7.3 6.3 Mexico 5.1 5.0 6.2 5.5 5.5 C6te d'Ivoire 33 2.2 5.7 4.3 4,7 World 96.4 108.4 113.7 117.0 11771 Source: USDA; and International Coffee Organization (ICO). 214 GLOBAL COMMODITY PRICE PROSPECTS Table A2.2 Beverages' global balance Annual growth rates (percent) 1970 1980 1990 1999 2000 2001 1970-80 1980-90 1990-2000 Coffee (Thousand bags) Production 64,161 86,174 88,849 113,723 117,001 117,739 2.11 1.36 1.20 Consumption 71,536 79,100 96,300 98,000 103,290 105,340 1.01 1.97 0.22 Exports 54,186 60,996 76,163 92,338 87,502 96,095 0.78 2.41 1.06 1970 1980 1990 1998 1999 2000 1970-80 1980-90 1990-2000 Cocoa (Thousand tons) Production 1,554 1,695 2,506 2,884- 3,032 2,809 0.46, 4.62 1.82 Grindings 1,418 1,556 2,335 2,785 2,911 2,977 0.16- 4.48 2.38 Stocks 497 675 1,791 1,231 1,321 1,125 2.38 13.89 3.95 Tea (Thousand tons) Production 1,286 1,848 2,526 2,963 2,847 2,895 4.09 2.87 1.24 Exports 752 859 1,099 1,2% 1,272 1,309 2.35 2.39 1.62 Notes: The 2001 figures for coffee are preliminary forecasts. Time reference for coffee and cocoa are based'on crop year shown under the year that production begins: October to September for cocoa,.and April to March for. coffee. For tea, time is calendar year. Source: International Cocoa Organization; International Tea Committee;.FAO; USDA; and World Bank.. real coffee prices are expected to recover, but countries have seen slower- demand growth, remain well below historical highs of the 1970s partly due to the recent economic slowdown. or recent highs of the 1990s. By 2015, real ara- Prices are projected to average a little over bica and robusta prices are projected to in- $1.00/kg in 2001 and about $1.10/kg in 2002. crease 54 and 74 percent, respectively, from By 2015, real prices are projected to increase 2001 levels, but they would still be only half of 21 percent from 2001 but still be 20 percent be- their 1990s peaks. low their 1998 highs. Cocoa Tea Following the three-decade low in February The three-auction average tea price fell 17 per- 2000, cocoa prices recovered somewhat dur- cent in the first nine months of 2001, com- ing the first nine months of 2001 to average pared to 2000, due mostly to increased pro- $1.01/kg compared to $0.91/kg in 2000. The duction by the major exporters (India, Kenya, partial price recovery was largely due to pro- and Sri Lanka). In addition, currency devalua- duction cutbacks and export disruptions in tions. in Sri Lanka relative to the U.S. dollar Cbte d'Ivoire (due to political instability), and contributed to the dollar price declines. Since Ghana. The 2000-01 cocoa crop is expected the high in 1997, nominal tea prices are down to be down 7.3 percent from the 1999-2000 about 21 percent. record crop, and more in line with the average Tea prices have been held up by several production levels of the early 1990s. years of poor harvests in some exporting Demand for cocoa is expected to grow by countries, combined with strong demand in 2.3 percent this season, just a little slower than the Middle East and the Russian Federation, the 1990-2000 average of 2.4 percent, but far following high export earnings from crude oil. below the 1980-90 average of 4.6 percent (table However, prices are expected to decline as A2.2). Demand from Eastern Europe and the supplies increase and demand weakens along former Soviet Union (FSU) has grown by more with the expected decline in crude oil' prices. than 10 percent per year, while East Asian We project tea prices to decline about 1 per- 215 GLOBAL ECONOMIC PROSPECTS cent in 2002, but there is potential for larger Table A2.3 Soybean production declines because of a possible disruption in (millions of tons) trade to the Middle East and Central Asia fol- unse4 lowing recent events. Year Argenana rczi state World The growth of global tea exports has 1990 1s 1s.8 52.4 104.1 slowed significantly during the 1990s com- 1995 12.4 24.2 59.2 124.9 pared to previous decades (table A2.2), and 2000 26.0 37.5 75.4 1712.1 this has not been offset by more rapid growth 2001 25.s 38.0 79.9 177.2 in domestic demand in major producing coun- Source: USDA. tries, such as India. Thus, we project real prices to decline 14 percent by 2015 relative to 2001 as exporters intensify their push to increase and exporters (Argentina, Brazil, and the output and demand growth remains weak. United States), which together account for 80 percent of global production (table A2.3). Food Since 1990, palm oil production has more than Despite considerable volatility in the compo- doubled (table A2.4), with the large increases nents of the food price index, the overall index coming from Indonesia and Malaysia. of nominal food prices has remained relatively constant since 1999, but is down nearly 32 per- Table A2.4 Palm oil production cent since peaking in 1996. Prices are expected (millions of tons) to increase about 1 percent in 2002 and then Year Indonesia Malaysia World begin to recover more rapidly as the global economy rebounds from the current slowdown, 1990 2.41 6.10 11.03 and agricultural commodity prices recover from 1995 4.22 7.81 15.22 current lows. By 2015, real food prices are ex- 2000 6.95 10.84 2177 2001 7.35 11JS 23.01 pected to return to long-run trends, down 13 2001 7_35_ 1_.55_ 23.01 percent relative to 2001 levels. Source: Oil World. Fats and oils Prices of most fats and oils are expected to Fats and oils prices have taken a beating, increase in 2002 and 2003, but remain well down 8.1 percent in the first nine months of below 1999 highs. Once the current imbalance 2001 compared to 2000, and down 40 percent is resolved, price prospects improve due to the since 1997. The declines are due generally to strong demand growth expected in China and increased supplies and currency devaluations India. Real fats and oil prices are projected to of major producers versus the dollar. Global increase 12 percent from 2001 to 2015 as fats and oils production in 2001-02 (October prices recover from current lows. to September) is expected to increase about 2 percent from the 2000-01 level, which is well Grains below the trend growth of about 3.5 percent The USDA's projection for the new season per year, but follows large increases in recent (2001-02) is for significant declines in ending- years that have left the market oversupplied. stocks of grain (table A2.5), and this should The increase has been greatest in the two cause most grain prices to increase in 2001 and largest vegetable oils-soybean and palm- 2002 after reaching lows in 1999 or 2000. which account for 23 and 19 percent of total Maize prices appear to have bottomed out in fats and oils, respectively. 2000 and are expected to increase about 2 World soybean production has grown by percent in 2001 and 7 percent in 2002. Wheat 5.1 percent per year over the past decade, with prices, which hit bottom in 1999, are projected growth centered in the three major producers to increase 10 percent in 2001 and 4 percent in 216 GLOBAL COMMODITY PRICE PROSPECTS Table A2.5 Global grain stocks to use 2.6 percent per year during the 1970s to 1.8 (percentages) percent during the 1980s, and to 1 percent Miz Ric Wheat Tota Graias during the 1990s (table A2.6). 1997-98 25.5 33.3 29.3 26.9 Sugar 1998-99 29.3 34,3 29.8 28.5 1999-2000 28.4 35.8 28.3 27.8 World sugar production has exceeded con- 2000-01 25.9 34.2 26.8 26.1 sumption in 8 of the past 10 seasons, causing 20esited) 23.0 31.4 22.4 23.1 the ending stocks-to-use ratio to reach 0.27 in 1990s low 22.6 31.4 25.2 23.2 the 2000-01 marketing season-the highest since 1985. World sugar consumption has Source: USDA. Data for 2001-02 are the USDA's August 2001 estimate. grown by 3 percent per year during the last decade (table A2.6). Sugar prices had recov- ered from the sharp drop following the Asia 2002. Rice prices, which are still falling, are ex- crisis, but have since declined due to large sup- pected to decline 16 percent in 2001, and then plies (figure A2.4). increase 9 percent in 2002. In real terms, maize, Brazil, which is the largest sugar exporter rice, and wheat prices are projected to increase with about one-quarter of world exports in 14, 17, and 13 percent by 2015 relative to their 2000-01, more than doubled production from lows during 1999-2001. 1990-91 to 2000-01 and increased exports Substantial surplus production capacity ex- from 1.5 to 11.3 million tons. Australia and ists because yields have continued to grow Thailand increased production by 50 and 70 along historical trends, while the area devoted percent, respectively, from 1990-91 to 1997- to grain production has fallen (figure A2.3). 98 when prices were attractive, but have cut Despite these reductions in land use, real prices production as prices have declined. have declined by half since 1980. The growth Sugar prices are expected to fall about 11 of global grain consumption has slowed from percent in 2002 in response to large supplies and weak demand, and then increase 12 per- cent in 2003. However, prices are expected to remain relatively weak for the next several Figure A2.3 World grain area and yields years, with fluctuations depending on the year- to-year balance of production and consump- 750 3.50 tion. Over the longer term, real prices are ex- -3.00 pected to trend lower as production continues to outpace consumption and stocks periodically 700 -2.50 build. Relative to the 1999 lows, real prices are projected to increase 49 percent by 2015. W OYields -2.00 - 1.50 Raw materials eso The index of agricultural raw materials prices (comprised of tropical hardwoods, cotton, -0.so and natural rubber) declined sharply during the Asia crisis and then stabilized. Recently 1960 1970 1980 1990 2000 prices have again declined, and are now about 40 percent below their 1997 nominal levels Note: Area is in million hectares (left side) and yields aresd in tons per hectare (right side). (1gure A2.5). Prices are expected to reach a low in 2001 and then increase modestly dur- ing the next several years. By 2005, nominal prices are projected to rise 28 percent relative 217 GLOBAL ECONOMIC PROSPECTS Table A2.6 Foods' global balance Annual growth rates (percent) 1970 19s0 1990 1998 1999 2000 1970-80 1980-90 1990-2000 Grains (Million tons) Production 1,079 1,430 1,769 1,888 1,887 1,840 2.88 1.55 1.04 Consumption 1,114 1,450 1,717 1,857 1,890 1,876 2.58 1.78 1.02 Exports 109 215 203 225 241 227 6.35 0.13 0.94 Stocks 193 309 490 528 525 489 7.24 3.83 0.56 Soybeans (Thousand tons) Production 44,269 80,873 104,093 159,819 159,659 172,107 6.84 1.87 5.08 Consumption 47,988 84,017 103,643 159,567 159,839 171,486 6.53 2.04 4.99 Exports 12,572 24,514 24,488 38,945 47,231 52,686 5.24 0.80 2.88 Stocks 3,599 11,538 12,992 14,297 14,338 14,209 13.83 0.66 0.20 Sugar [Thousand tons (ra equivalent)] Production 70,919 84,742 109,393 143,388 133,634 136,882 2.80 1.59 3.26 Consumption 65,395 91,062 106,802 138,168 127,499 129,449 3.30 1.40 3.00 Exports 21,931 27,571 34,078 41,933 36,742 39,911 3.26 0.83 3.12 Stocks 19,614 19,494 19,309 28,178 31,639 35,225 3.96 0.77 4.52 1970 1980 1990 1999 2000 2001 1970-80 1980-90 1990-2000 Fats and oils (Million tons Production 39.78 58.09 80.84 113.50 117.48 119.84 3.68 3.54 3.70 Consumption 39.82 $6.80 80.87 112.20 117.54 121.29 3.55 3.69 3.64 Exports 8.83 17.763 26.89 35.13 37.82 39.37 7.05 4.19 3.39 Stocks 5.18 9.25 12 .15 14.04 14.00 12.80 7.09 2.44 0.69 Note: Time reference for grains, soybeans, and sugar are based on marketing years, shown under the year in which production begins, and varies by country and commodity; for fats and oils time is crop year beginning September. Source: USDA; and Oil World. Figure A2.4 Nominal sugar price U.S. cents per kilogram 35 30 15 [ iiI~ I I I 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Source: World Bank. 218 GLOBAL COMMODITY PRICE PROSPECTS term, real prices are expected to rise only Figure A2.5 Agricultural raw materials modestly from current low levels. By 2015, price index real prices are projected to increase 14 percent Index, 1990= 100 relative to 2001. 130 Natural rubber Natural rubber prices have contributed to the 110 recent weakness in raw materials prices by de- clining 11 percent during the first three quar- ters of 2001 compared to 2000. This price weakness occurred despite a nearly 9 percent increase in global demand (partly in response to the second Firestone tire recall). The three top producers and exporters of 70 I,,,II ,,,,,, n,,,,, |,,,,, natural rubber-Indonesia, Malaysia, and Thai- Jan. Jan. Jan. Jan. Jan. land-have, in principle, agreed to establish a 1997 1998 1999 2000 2001 buffer stock with the objective of cutting pro- Source: World Bank. duction by 4 percent annually starting in 2002 until a full price recovery is realized. While the details of the buffer stock scheme have not yet to 2001, and real prices are projected to rise been outlined, the trilateral organization, Tri- 17 percent by 2015. partite Rubber Corporation (TRC), is expected to act soon. The historical record of managed Cotton supply cut mechanisms is poor, but because Cotton prices declined almost 14 percent in the TRC consists of the three members who to- first three quarters of 2001 compared to gether account for about two-thirds of global 2000-in response to a 6 percent increase in output, the outcome may be different. global production in the 2001-02 season. The Natural rubber prices are expected to de- surge in production (compared to recent his- cline 11 percent in 2001 and then begin to torical growth of about 0.8 percent a year) was recover in 2002-rising 25 percent by 2005. largely due to a 7 percent increase in global Real prices are expected to increase 16 percent area planted in cotton, which was in response by 2015 relative to 2001. to the relative attractiveness of cotton prices compared to other annual crops. China, India, Tropical timber and the United States accounted for three- Asian meranti log prices fell 14 percent during quarters of the total production increase. the first three quarters of 2001 compared to Cotton demand has been stagnant for most 2000, due to weak demand in Japan and the of the past decade and is unlikely to quickly strong dollar relative to the Japanese yen. absorb recent production increases. Cotton's African sapelli log prices fell 5 percent over the share of total fiber consumption exceeded 80 same period due to reduced supplies because of percent in 1950, but fell to 50 percent by restrictions and bans on log exports from 1980, and reached a low of 40 percent in re- Cameroon and other African countries. The cent years. Consumption is only expected to weakness of the euro against the dollar and the increase 1 percent in 2001-02; consequently instability of meranti prices encouraged Euro- stocks are expected to rise significantly. There- pean buying in the African market. fore, the widely used Cotlook A Index is pro- As growth in the global economy slows, de- jected to average $1.06/kg in 2001 and then mand in the tropical timber industry continues decrease to $1.02/kg in 2002. Over the longer to weaken, and prices are expected to follow 219 GLOBAL ECONOMIC PROSPECTS Table A2.7 Raw materials' global balance Annual growth rates (percent) 1970 1980 1990 199 2li00l 2001 1970-SO 1980-90 1990-2000 Cotton (thousand tons) Production 11,740 13,832 18,970 18,841 19,360 20,800 1.22 3.09 0.84 Consumption 12,173 14,215 18,576 19,784 19,700 19,930 1.11 3.10 0.21 Exports 3,875 4,414 5,081 6,102 5,770 5,900 0.93 2.79 0.49 Stocks 4,605 4,895 6,645 8,802 8,580 9,460 1.71 2.83 1.38 1970 1980 1990 1998 1999 2000 1970-80 1980-90 1990-2000 Natural rubber (thousand tns) Production 3,140 3,820 5,080 6,820 6,800 6,880 1.78 3.17 3.08 Consumption 3,090 3,770 5,190 6,540 6,660 7,260 1.58 3.16 3.25 Net Exports 2,820 3,280 3,950 4,690 4,660 5,000 1.26 2.07 1.84 Stocks 1,440 1,480 1,500 2,300 2,530 2,150 0.60 0.23 3.71 1970 1980 1990 1997 1998 1999 1970-80 1980-90 1990-1999 Tropical lumber (thousan bi Y,neters) Logs, production 210 262 300 311 289 299 1.47 1.71 0.45 Logs, imports 36.1 42.2 25.1 17.9 14.6 18.9 0.18 5.10 5.36 Sawnwood, production 98.5 115.8 131.8 115.0 108.3 108.2 1.17 1.74 1.99 Sawnwood, imports 7.1 13.2 16.1 21.2 19.5 21.6 4.95 2.57 3.33 Plywood, production 33.4 39.4 48.2 56.1 47.6 52.0 1.17 2.02 0.46 Plywood, imports 4.9 6.0 14.9 19.5 18.3 18.3 0.69 9.10 3.60 Note: The 2001 figures for cotton are preliminary forecasts. Time reference for cotton is based on crop year shown under the production year beginning August; for rubber and tropical timber, time refers to calendar year. Source: International Cotton Advisory Committee; International Rubber Study Group; FAO; and World Bank. demand. Log imports to Japan are expected to timber prices remain one of the few commodi- fall about 6 percent in 2001 compared to the ties with trend real price increase due to supply previous year, according to industry estimates. constraints. Real meranti logs and sawnwood China, which is the largest global log importer, prices are projected to increase 37 and 27 per- has continued rapid import growth and this cent, respectively, from 2001 to 2015 while has partially offset weak Japanese imports. sapelli log prices are projected to increase 9 However, the combination of the strong dollar, percent by 2015 compared to 2001. The slower economic growth, and the abundance slower projected growth of sapelli log prices of softwoods that can substitute for hardwood reflects the smaller price decline compared to in some uses, should lead to lower prices in meranti log prices during the Asia crisis. 2001. The recovery of timber prices beyond 2001 will be closely linked to the global eco- nomic recovery expected in 2003 and to a Fertilizers weakening of the dollar. We project timber he fertilizer industry is burdened by sur- prices to remain unchanged in 2002 and to re- TL plus capacity and weak demand, but prices cover in 2003. By 2005, nominal prices of mer- appear to be near their lows. The situation anti logs are projected to increase 43 percent varies by fertilizer type, with nitrogen (urea) relative to 2001; sapelli logs are projected to fertilizer prices recovering in 2000 after falling increase 13 percent; and meranti sawnwood is for four years; phosphate (TSP) fertilizer prices projected to increase 34 percent. still declining but near expected lows; and Over the longer-term, real timber prices are potash (MOP) prices holding steady due to ag- projected to recover from current levels, as gressive production cutbacks (figure A2.6). 220 GLOBAL COMMODITY PRICE PROSPECTS Figure A2.6 Fertilizer prices Dollars per ton 250 200Ura A - TSP 150 100 MOP \ J 50 Jan. 1990 Jan. 1992 Jan. 1994 Jan. 1996 Jan. 1998 Jan. 2000 Source: World Bank. The industry is still adjusting to the sharp domestic demand-which led to export declines in consumption in former Soviet and growth from Eastern Europe of 4 percent per Eastern European countries following the col- year since 1993. These exports displaced tradi- lapse of the Soviet Union. Prices had been tional exporters, and depressed prices of nitro- heavily subsidized under state control and fer- gen and phosphate fertilizers. Global consump- tilizer use was high, but subsidies were cut and tion fell about 17 percent from the high in 1988 consumption fell sharply after the collapse of to the low in 1993 and has only recently recov- the Soviet Union. This left many countries ered to near the 1988 peak (figure A2.7). (such as the Russian Federation and Ukraine) The fertilizer industry has had to contend with large production capacity and reduced with several other changes in recent years, in- Figure A2.7 World fertilizer consumption and exports Millions of tons 160 120 Consumption 80 40 Exports 1970 1975 1980 1985 1990 1995 Source: FAO. 221 GLOBAL ECONOMIC PROSPECTS Table A2.8 Fertilizer global balance (million tons) Annual growth rates (percent) 19170 198 199o 137 1938 19,99 1970-SO 1980-90 1990-99 Nitrogen Production 33.30 62.78 82.26 87.60 88.48 90.85 6.53 3.12 1.11 Consumption 31.76 60.78 77.14 80.12 82.62 85.53 6.86 2.60 1.15 Exports 6.77 13.15 19.48 23.24 23.95 24.58 7.23 5.10 2.62 Phosphate Production 2204 34.51 39.35 32.81 32.99 32.65 3.72 1.70 2.05 Consumption 21.12 3170 35.90 33.34 33.17 33.15 3.85 1.39 .88 Exports 292 7.51 10.50 12.24 12,54 12.90 8.37 5.01 2.31 Potash Production 17.59 27.46 26.82 26.16 24.98 25.42 3.97 0.03 0.59 Consumption 16.43 24.24 24.68 22.63 22.36 22.68 3.93 0.05 0.94 Exports 9.45 16.72 19,82 22.52 22.13 22.63 4.89 0.73 1.48 Note: All data are in marketing years. Source: FAO. cluding weak grain prices since 1996; high tions have also helped to rationalize surplus natural gas prices in the United States and capacity within the industry. However global Europe in the past two years; reduced fertilizer output continues to exceed demand and in- use in the EU because of environmental con- ventories have risen. Stocks of most metals cerns and lower commodity intervention prices; have risen by more than 60 percent this year, and increased domestic fertilizer production in with aluminum and copper stocks more than major importing countries such as China. doubling (see figure A2.9). The slow recovery of agricultural commod- The negative impact on the global economy ity prices and weakness in the global economy from the terrorist attacks of September 11 will suggest that prices may remain near current lev- result in lower demand for most metals and els for several years or begin a modest recovery. minerals, higher inventories, and lower prices. Over the longer term, nitrogen prices are pro- Further closure of high-cost production is likely, jected to rise as production capacity is rational- and this may help underpin prices somewhat. ized and demand increases; phosphate prices However the recovery in prices will likely be de- are expected to remain about constant follow- layed well into 2002, and will largely be deter- ing recent declines; and potash prices are ex- mined by the timing and the strength of the re- pected to decline as surplus capacity continues. bound in global economic activity. Higher prices will also bring forth new ca- pacity and the restart of idle facilities, and prices Metals and Minerals will eventually recede. Real prices are expected T he index of metals and minerals prices fell to decline in the longer term, as production costs 15 percent during the first nine months of continue to fall due to new technologies and im- 2001, with copper prices down 23 percent (see proved managerial practices (see figure A2.10). figure A2.8). Production cutbacks have helped slow the price decline, most notably in alu- Aluminum minum where significant capacity has been Aluminum prices have fallen 14 percent this shut in the United States' Pacific Northwest- year, while London Metal Exchange (LME) in- and to a lesser extent in Brazil-because of ventories have risen 124 percent. Prices have electric power problems. Mergers and acquisi- been partly supported by large reductions in 222 GLOBAL COMMODITY PRICE PROSPECTS Figure A2.8 Aluminum and copper prices Dollars per ton 2,700 2,300 1,900 1,500 um 1,100 1 1-7 7 7 11 '11 11 11 II I I I III II Jan. 1996 Jan. 1997 Jan. 1998 Jan. 1999 Jan.2000 Jan. 2001 Source: London Metal Exchange. Figure A2.9 Aluminum and copper stocks Thousands of metric tons 1,000 800 Aluminum 000 400 200 Copper 0 II1I1111111 I r r i l r r f elrII I I I I I lI I lI I I llII I I IlIllI lI Jan. 1996 Jan. 1997 Jan. 1998 Jan. 1999 Jan.2000 Jan. 2001 Source: London Metal Exchange. production in the United States because of the producers to stay off-line for up to two years or electricity crisis on the West Coast and produc- face high power prices when new contracts tion curtailments in Brazil and Canada due to went into effect on October 1, 2001. The BPA hydropower shortages. announced that load reductions by utilities and About 1.6 million tons of capacity in the U.S. industries helped reduce the rate increase to 46 Pacific Northwest has been idled because of percent (approximately $34/MWh) compared electric power shortages in the region. The Bon- to possible rate increases of 250 percent, and neville Power Authority (BPA) asked aluminum spot power rates that were several times that 223 GLOBAL ECONOMIC PROSPECTS Figure A2.10 Metals and minerals prices Price index (1990 = 100) 160 120 Forecast 140 120 Real 100 80A V /A 60 o Nominal 40 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Source: Platts and World Bank. amount. As compensation for not taking power 9 percent. Demand has been weak elsewhere, and curtailing production, aluminum smelters with the notable exception of China, partly will receive an average rate of $20/MWh. due to its infrastructure programs. Mean- Despite production cutbacks, the global while, world production rose 4 percent in the market is expected to retain a small surplus first half of the year. this year, before moving into a deficit in 2002, With recovery of demand in 2002, the mar- but this will partly depend on any structural ket balance is expected to slip into deficit, impact to demand following the September since only moderate growth in production is terrorist attacks. Prices are expected to re- expected. Prices could rebound sharply as the cover during the next economic cycle, but real next cycle commences, which could also pro- long-term prices are expected to decline. New vide upward momentum to other metals prices. low-cost capacity is coming on-stream, but In the longer term, increases in new low-cost profitable new investments will continue to re- capacity are expected, and real prices are ex- quire low-cost power supplies. pected to decline. Copper Nickel Copper prices declined 23 percent in the first Nickel prices have fallen 31 percent this year nine months of 2001, due to weak demand because supply has significantly exceeded de- and rising stocks. mand. LME inventories have risen by 73 per- LME inventories have more than doubled cent, but are still quite low compared with lev- this year, and are only 13 percent below the els in recent years (see table A2.9). Production highs in early 2000. World consumption fell in the first seven months increased by 2.7 per- 2 percent during the first six months, due to cent, with Canada, Colombia, and New Cale- the slump in economic activity. In the United donia recording large gains. However first-half States, the construction sector has been buoy- world consumption dropped 10 percent, with ant, but weakness in the auto and technology demand in Japan and the United States down sectors has resulted in total demand falling sharply, while China provided the one bright 224 GLOBAL COMMODITY PRICE PROSPECTS Table A2.9 Metals and minerals global balance (thousand tons) Annual growth rates (percent) 1970 1980 1990 1998 1999 2000 1970-80 1980-90 1990-2000 Aluminum Production 10,2S7 16,027 19,362 22,648 23,705 24,495 3.2 1.9 2.2 Consumption 9,996 14,771 19,244 21,842 23,505 24,905 3.2 1.8 2.2 LME ending stocks n.a. 68 311 636 775 322 n.a. 0.3 0.4 Copper Production 7,583 9,242 10,809 14,14$ 14,455 14,788 1.9 1.1 3.5 Consumption 7,294 9,400 10,780 13,364 14,094 15,099 2.5 1.0 3.3 LME ending stocks 72 123 179 592 790 357 7.4 5.6 15.7 Nickel Production 0 717 842 999 1,073 1,140 n.a. 1.6 3.1 Consumption 0 742 858 1,042 1,028 1,107 n.a, 1.5 2.6 LME ending stocks (tons) 2,130 4,554 4,344 65,964 46,962 9,678 n.a. 0.5 8.3 n.a. = not available. Source: World Bureau of Metal Statistics; London Metal Exchange and World Bank. spot of growth. Stainless steel production has prices surged toward $300/toz as some in- declined owing to the slowdown in economic vestors turned to gold as a safe haven. Once activity, which lowered demand and prices for calm returns to world markets, gold prices nickel (and zinc). The market is tilting into sur- should revert toward previous levels, as gold plus, and a small surplus is expected to endure demand will be adversely affected by higher in 2002 and 2003 as production increases. prices and the slowing global economy. Gold demand has been sluggish this year, falling 3 Gold percent in the second quarter, in part because Gold was the one major metal to rise sharply of the higher U.S. dollar gold price. Central immediately following the September terrorist Bank sales continue (see table A2.10), with attacks. After averaging $267/toz this year, the U.K. government about to complete its Table A2.10 Gold global balance (tons) Taos (percent p.a.) 1991 1994 1995 1996 1997 1998 1999 2000 1991-2000 Jewelry 2,358 2,618 2,791 2,851 3,349 3,156 3,149 3,185 3.4 Other fabrication 518 457 503 484 560 569 595 564 0.9 Bar hoarding 252 231 306 182 325 173 240 211 2.0 Other n.a. n.a. 6 n.a. n.a. 208 170 n.a. Total demand 3,128 3,305 3,606 3,518 4,234 4,106 4,154 3,971 4.0 Mine production 2,159 2,279 2,274 2,361 2,479 2,538 2,568 2,576 2.0 Net official sales 111 81 173 279 626 374 464 471 17.4 Old gold scrap 482 617 625 640 628 1,097 616 607 2.6 Net hedging 66 163 535 142 504 97 506 n.a. Other 310 173 95 297 316 0.7 Total supply 3,128 3,305 3,606 3,518 4,234 4,106 4,154 3,971 2.4 n.a. = not available Source: Gold Field Minerals Service; and World Bank. 225 GLOBAL ECONOMIC PROSPECTS planned series of auctions of 395 tons in early recently chosen band of $22-28 a barrel for its 2002. Gold prices are expected to remain basket of crudes. Due to the seasonality of oil under $300/toz over the forecast period, gen- demand, OPEC must both raise and lower pro- erally trading in a relatively narrow range. As duction during the year to stabilize prices (see has been the case for some time, higher prices figure A2.12). With non-OPEC supply increas- will stimulate new supplies, encourage pro- ing, it will be more and more difficult to coun- ducer sales, and lessen demand, while low terbalance the downward pressure on prices. prices will reduce investment and encourage The terrorist attacks on the United States consumption. Mine production is expected to on September 11, 2001, have accentuated this continue to increase moderately, as new low- picture, while at the same time uncertainty is cost operations come on-stream. An important exceptionally large. Following the attacks, oil determinant of prices will be the decision by prices slumped below $23 a barrel due to ex- Central Banks whether to further stem official pectations of weak oil demand, little immediate gold sales when the Washington Agreement ex- threat to oil supplies, and no action by OPEC to pires in 2004. reduce production and prop up sagging prices. However should there be a significant supply disruption-either from military attacks, sanc- Petroleum tions, or reactions from oil producers (for in- S ince the rebound in oil prices that began in stance, from Iraq)-oil prices could rise sharply. early 1999-propelled by a large cutback OPEC announced immediately after the at- in OPEC production and sharp decline in in- tacks that it would raise production if necessary ventories (see figure A2.11)-prices have held to help prevent oil prices from spiking higher. firm primarily because of OPEC production re- Given surplus capacity of around four million straint. Ten OPEC countries (excluding Iraq, barrels a day, the organization could easily which remains outside the quota system while make up for a loss of, say, Iraq's exports of under U.N. sanctions) are taking pre-emptive around two million barrels a day. At its meeting production decisions to keep prices within their at end-September 2001, the organization de- Figure A2.11 Oil price and OECD stocks Dollars a barrel Millions of barrels 35- 2,850 - 2,800 30 -2,750 - 2,700 2565 Price 2,650 20 A Y r, V2,600 J 2,550 15 2,500 SStocks -2,450 10 - -, 2,400 Jan.1996 Jan.1997 Jan.1998 Jan.1999 Jan.2000 Jan.2001 Source: World Bank and International Energy Agency. 226 GLOBAL COMMODITY PRICE PROSPECTS Figure A2.12 OPEC production and quotas Millions of barrels a day 30 EC OPEC-1 0 quota - OPEC- 10 production - OPEC production 28 26 24 2 2 . . . . . ..I I . . . . . . . Jan.1996 Jan.1997 Jan.1998 Jan.1999 Jan.2000 Jan.2001 Note: OPEC-10 excludes Iraq. Source: International Energy Agency. cided not to cut production, despite the fact that sions may result in both the over- and under- oil prices were starting to fall below the lower shooting of prices. end of its range. The organization felt com- In the longer term, if OPEC is successful in pelled not to raise prices at this time because of keeping prices above $25 a barrel, the impact on the impact on the weakening global economy, demand, and particularly on competing supplies, and to show support for the allied coalition. will increasingly exert downward pressure on In 2002, the requirements for OPEC oil are prices. While higher prices in 1999-2000 were projected to be lower than in 2001, due to min- achieved relatively easily with little apparent im- imal growth in global oil demand and contin- pact on demand, supply, and economic activity, ued rise in non-OPEC supplies. Consequently, long-term responses are likely to be much higher OPEC will need to lower production to keep and could thwart part-and possibly much--of prices within its band. OPEC is expected to the growth in demand for OPEC crude. strive to maintain prices within the lower end To the degree that higher oil prices are of its range. However, in the present political deemed to be temporary, there will be little and economic environment, it is expected to structural change to oil demand. But if high fall short because of weak oil demand, higher prices are perceived to be "permanent," it will inventories, and overproduction by some mem- accelerate advances in conservation and sub- ber countries. stitution away from oil. High prices have al- Once some form of normalcy returns to the ready generated policy responses, such as the political and economic climate, and a global re- new U.S. energy policy, and increasing envi- covery commences, OPEC is expected to con- ronmental pressures will also tend to restrain tinue its policy of adjusting output to keep in- oil consumption over time. High prices will ventories lean and to maintain prices within its also stimulate development of conventional band. However, this requires OPEC to micro- and unconventional oil supplies, and make al- manage the market and to anticipate seasonal ternative energy supplies more competitive. changes in demand for its crude. Given the There are no apparent resource constraints far many uncertainties affecting underlying levels into the future, and oil consumption has only of oil demand and supply, its production deci- risen moderately over the past 20 years (see 227 GLOBAL ECONOMIC PROSPECTS Table A2.11 Petroleum global balance (million barrels per day) Million barels pr day Annual growth rates (percent) 1970 1980 1990 1000 2001 2002 1970-80 1980-90 1990-2000 OECD 34.0 41.5 41.5 47.8 47.8 47.8 2.0 0.0 1.4 FSU 5,0 8,9 8.4 3.6 3.7 3.7 5.9 0.6 8.1 Other nonOECD 6.8 12.3 16.1 24.4 24.6 24.9 6.1 2.7 4.3 Total consumption 45.7 62.6 66.0 75.9 76.1 76.4 3.2 0.5 1.4 OPEC 23.5 27.2 24.5 30.8 30.3 29.3 1.5 1.0 2.3 FSU 7.1 12.1 11.5 7.9 8.5 8.8 5.4 0.5 3.6 Other nonOPEC 17.4 24.6 30.9 38.0 38.0 38.7 3.5 2.3 2.1 Total production 48.0 63.9 66.9 76.7 76.8 76.8 2.9 0.5 1.4 Stock change, misc. 2.3 1.3 0.9 0.8 0.7 0.4 Source: British Petroleum; International Energy Agency; and World Bank. table A2.11). In addition, new areas continue expected to decline from $25 a barrel in 2001 to be developed (e.g., deep water offshore and to $21 a barrel in 2002, and fall below $20 a the Caspian Sea), development costs continue barrel by mid-decade (see figure A2.13). A risk to fall (shifting supply curves outward), and to the forecast is if OPEC takes strong, con- the large profits being generated will lead to certed action on production levels over the higher investment. In addition, OPEC coun- next few years to keep prices at or above $25 tries are increasing capacity, and will add to a barrel. If successful, it will add to the grow- the supply competition in the coming years. ing pressures on world demand and competing Due to rising supply competition and supplies, and prices would still be expected to below-trend oil demand growth, oil prices are fall below $20 a barrel by mid-decade. Figure A2.13 Crude oil prices Dollars a barrel 50 Forecast 40 1990 dollars 30 / 20 t l ' 10 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Source: World Bank. 228 GLOBAL COMMODITY PRICE PROSPECTS Table A2.12 Commodity prices and price projections in current dollars Actual Projections Commodity Unir 1970 1980 1990 2000 2001 2002 2003 2005 2010 2015 Energy Coal, U.S. $/mt n.a. 43.10 41.67 33.06 44.00 38.00 36.00 34.00 35.00 36.00 Crude oil, average $/bi 1.21 36.87 22.88 28.23 25.00 21.00 20.00 18.00 19.00 21.00 Natural gas, Europe $/mmbta n.a. 3.40 2.55 3.86 4.00 3.30 3.10 2.75 2.75 3.00 Natural gas, U.S. $/mmbtu 0.17 1.55 1.70 4.31 3.95 2.50 2.60 2.75 3.00 3.25 Nonenergy Commodities Agriculture Beverages Cocoa cents/kg 67.5 260.4 126.7 90.6 105.0 110.0 120.0 140.0 157.0 168.0 Coffee, other milds cents/kg 114.7 346.6 197.2 192.0 136.7 138.9 154.3 209.4 265.0 280.0 Coffee, robusta cents/kg 91.4 324.3 118.2 91.3 61.7 63.3 70.6 88.2 132.0 142.6 Tea, auctions (3) average cents/kg 83.5 165.9 205.8 187.6 162.0 160.0 168.0 180.0 182.0 184.0 Food Fats and oils Coconut oil $/nit 397.2 673.8 336.5 450.3 315.0 365.0 430.0 600.0 645.0 670.0 Copra /nt 224.8 452.7 230.7 304.8 200.0 350.0 400.0 450.0 480.0 500.0 Groundnut oil $/mt 378.6 858.8 963.7 713.7 675,0 725.0 775.0 820.0 850.0 875.0 Palm oil $/nt 260.1 583.7 289.8 310.3 290.0 330.0 360.0 400.0 450.0 475.0 Soybean meal $/mr 102.6 262.4 200.2 189.2 180.0 183.0 190.0 215.0 235.0 245.0 Soybean oil $/mt 286.3 597.6 447.3 338.1 357.0 385.0 395.0 425.0 460.0 505.0 Soybeans in1t 116.9 296.2 246.8 211.8 200.0 205.0 210.0 235.0 260.0 270.0 Grains Maize $1mr 58.4 125.3 109.3 88.5 90.0 96.0 108.0 122.0 125.0 130.0 Rice, Thai, 5 percent $/mr 126.3 410.7 270.9 202.4 170.0 185.0 205.0 235.0 260.0 270.0 Sorghum $/mte 51.8 128.9 103.9 88,0 95.0 91.8 103.3 116.6 119.5 123.5 Wheat, U.S., HRW $/mt 54.9 172.7 135.5 114.1 125.0 130.0 138.0 150.0 155.0 160.0 Other food Bananas, U.S. S/mt 166.1 377.3 540.9 424.0 610.0 523.6 523.6 529.1 568.0 590.0 Beef, U.S. cents/kg 130.4 276.0 256.3 193.2 207.0 202.8 202,8 213.9 220.0 230.0 Oranges $/mt 168.0 400.2 531.1 363.2 630.0 625.0 550.0 450.0 475.0 500.0 Shrimp, Mexican cents/kg n.a. 1,152 1,069 1,513 1,575 1,550 1,600 1,660 1,690 1,720 Sugar, world cents/kg 8.2 63.16 27.67 18.04 18.80 16.75 18.70 22.00 24.00 26.00 Agricultural raw materials Timber Logs, Cameroon $cum 43.0 251.7 343.5 275.4 265.0 265.0 275.0 300.0 338.0 385.0 Logs, Malaysia $/cum 43.1 195.5 177.2 190.0 162.0 162.0 190.0 232.0 260.0 295.0 Sawnwood, Malaysia $/cum 175.0 396.0 533.0 594.7 485.0 485.0 570.0 650.0 720.0 820.0 Other raw materials Cotton cents/kg 67.6 206.2 181.9 130.2 105.8 102.1 114.6 132.3 149.9 159.6 Rubber, RSS1, Malaysia cents/kg 40.7 142.5 86.5 69.1 61.7 63.9 72.8 77.2 88.0 95.1 Tobacco $/mt 1,076 2,276 3,392 2,976 3,011 3,080 3,150 3,250 3,300 3,450 Fertilizers DAP $/mr 54.0 222.2 171.4 154.2 147.0 155.0 165.0 180.0 190.0 200.0 Phosphate rock S/mt 11.00 46.71 40.50 43.75 41.75 41.00 42.00 43.00 46.00 48.00 Potassium chloride $/mi 32.0 115.7 98.1 122.5 119.0 120.0 121.5 125.0 127.0 130.0 TSP $/mit 43.0 180.3 131.8 137.7 125.0 126.0 127.0 138.0 145.0 165.0 Urea, E. Europe, bagged /mt 48.0 222.1 130.7 112.1 105.3 110.0 120.0 140.0 145.0 150.0 Metals and minerals Aluminum $/nt 556 1,456 1,639 1,549 1,440 1,500 1,650 1,800 1,850 1,900 Copper $/nt 1,416 2,182 2,661 1,813 1,575 1,625 1,800 2,000 2,100 2,200 Gold $/toz 36.0 607.9 383.5 279.0 275.0 280.0 275.0 275.0 300.0 300.0 Iron ore, Carajas cents/dmntu 9.84 28.09 32.50 28.79 30.03 30.50 31.00 32.00 33.00 33.00 Lead ceits/kg 30.3 90.6 81.1 45.4 47.0 50.0 55.0 60.0 64.0 64.5 Nickel $/mt 2,846 6,519 8,864 8,638 5,900 6,100 6,200 6,400 6,500 6,600 Silver cenits/toz 177.0 2,064 482.0 499.9 450.0 475.0 500.0 520.0 550.0 550.0 Tin cents/kg 367.3 1,677 608.5 543.6 440.0 465.0 485.0 525.0 540.0 550.0 Zinc cents/kg 29.6 76.1 151.4 112.8 89.0 90.0 95.0 100.0 110.0 120.0 n.a. = Not available. Note: Projections as of October 12, 2001. Source: World Bank, Economic Policy and Prospects Group. 229 GLOBAL ECONOMIC PROSPECTS Table A2.13 Commodity prices and price projections in constant 1990 dollars Actual Projections Commodity fUnit 1970 1980 1990 2000 2001 2002 200 2005 2010 2015 Energy Coal, U.S. $1rnt n.a. 54.71 41.67 33.94 47.33 39.31 35.68 31.97 30.48 29.17 Crude oil, average $/bb] 4.31 46.80 22.88 28.98 26,89 21.73 19.82 16,92 16.54 17,02 Natural gas, Europe $/mbts n.a. 4.32 2.55 3.96 4.30 3.41 3.07 2.59 2.39 2.43 Natural gas, U.S. $1mtbts 0.61 1.97 1.70 4.42 4.25 2.59 2,58 2.59 2.61 2.63 Nonenergy Commodities Agriculture Beverages Cocoa cents/kg 240.6 330.5 126.7 93.0 113.0 113.8 118.9 131.6 136.7 136.1 Coffee, other milds cents/kg 408.8 440.0 197.2 197.1 147.0 143.7 153.0 196,9 230.8 226.9 Coffee, robusta cents/kg 325.7 411.7 118.2 93.7 66.4 66.1 69.9 82.9 114.9 115.6 Tea, auctions (3) average cents/kg 297.7 210.6 205.8 192.6 174.3 165.5 166.5 169.2 158.5 149.1 Food Fats and oils Coconut oil $1mt 1416.0 855.3 336.5 462.3 3389 377.6 426.2 564.1 561.7 542.9 Copra $/mr 801.6 574.7 230.7 312.9 215.2 362.1 396.4 423.1 418.0 405.1 Groundnut oil $/mt 1349.5 1090.1 963.7 732.6 726.1 750,1 768.1 771.0 740.2 709.0 Palm oil S1mt 927.1 740.9 289.8 318.5 312.0 341.4 356.8 376.1 391.9 384.9 Soybean meal s/mt 365.7 333.1 200.2 194.2 193.6 189.3 188.3 202.1 204.6 198.5 Soybean oil $/mt 1020.8 758.6 447.3 347.1 384.0 398.3 391.5 399.6 400.6 409.2 Soybeans S/mt 416.8 376.0 246.8 217.5 215.2 212.1 208.1 221,0 226.4 218.8 Grains Maize $/mt 208.2 159.0 109.3 90.9 96.8 99.3 107.0 114.7 108.9 105.3 Rice, Thai, 5 percent $/mt 450.3 521.4 270.9 207.8 182.9 191.4 203.2 221.0 226.4 218.8 Sorghum $/mt 184.7 163.6 103.9 90.3 102.2 95.0 102.3 109.7 104.1 100.1 Wheat, U.S., HRW $/mt 195.7 219.3 135.5 117.1 134.5 134.5 136.8 141.0 135.0 129.6 Other food Bananas, U.S. S/mr 592.1 478.9 540.9 435.3 656.2 541.7 518.9 497.5 494.6 478.0 Beef, U.S. cents/kg 465.0 350.3 256.3 198.4 222.7 209.8 201.0 201.1 191.6 186.4 Oranges $/me 599.1 508.0 531.1 372.9 677.7 646.6 545.1 423.1 413.6 405.1 Shrimp, Mexican cents/kg n.a. 1,462 1,069 1,553 1,694 1,604 1,586 1,561 1,472 1,394 Sugar, world cents/kg 29.32 80.17 27.67 18.5 20.2 17.3 18.5 20.7 20.9 21.1 Agricultural raw materials Timber Logs, Cameroon $/cum 153.3 319.5 343.5 282.8 285.1 274,2 272.6 282.1 294.3 311.9 Logs, Malaysia $/cum 153.8 248.2 177.2 195.0 174.3 167.6 188.3 218.1 226.4 239.0 Sawnwood, Malaysia $/cum 623.9 502.7 533.0 610.5 521.7 501.8 564.9 611.1 627.0 664.4 Other raw materials Cotton cents/kg 241.1 261.7 181.9 133.7 113.8 105.6 113.6 124.4 130.5 129.3 Rubber, RSS1, Malaysia cents/kg 145.2 180.8 86.5 71.0 66.4 66.1 72.1 72.6 76.6 77.0 Tobacco $/mt 3,836 2,889 3,392 3,055 3,239 3,186 3,122 3,056 2,874 2,795 Fertilizers DAP $/mt 192.5 282.1 171.4 158.3 158.1 160.4 163.5 169.2 165.5 162.1 Phosphate rock $/mt 39.2 59.3 40.5 44.9 44.9 42.4 41.6 40.4 40.1 38.9 Potassium chloride $/mt 114.1 146.9 98.1 125.8 128.0 124.2 120.4 117.5 110.6 105.3 TSP s/mt 153.3 228.8 131.8 141.4 134.5 130.4 125.9 129.8 126.3 133.7 Urea, E. Europe, bagged $/Imt 171.1 282.0 130.7 115.1 113.3 113.8 118.9 131.6 126.3 121.5 Metals and minerals Aluminum $mr 1,982 1,848 1,639 1,590 1,549 1,552 1,635 1,692 1,611 1,539 Copper $/mt 5,047 2,770 2,661 1,862 1,694 1,681 1,784 1,880 1,829 1,783 Gold $/toz 128.1 771.6 383.5 286.5 295.8 284.5 272.6 Z58.6 261.2 243.1 Iron ore cents/dmtu 35.1 35.7 32.5 29.6 32.3 31.6 30.7 30.1 28.7 26.7 Lead cents/kg 108.0 115.0 81.1 46.6 50.6 51.7 54.5 56.4 55.7 52.3 Nickel $/mt 10,147 8,275 8,864 8,867 6,347 6,311 6,145 6,017 5,660 5,348 Silver centsloz 631.0 2619.4 482.0 513.2 484.1 491.4 495.5 488.9 478.9 445.6 Tin cents/kg 1309.6 2129.3 608.5 558.0 473.3 481.1 480.7 493.6 470.2 445.6 Zinc cents/kg 105.5 96.6 151.4 115.8 95.7 93.1 94.2 94.0 95.8 97.2 n.a. = Not available. Note: Projections as of October 12, 2001. Source: World Bank, Economic Policy and Prospects Group. 230 GLOBAL COMMODITY PRICE PROSPECTS Table A2.14 Weighted indices of commodity prices and inflation Actual Projectionsa Index 1970 1980 1990 2000 2001 2002 2003 2005 2010 2015 Current dollars Petroleum 5.3 161.2 100.0 123.4 109.3 91.8 87.4 78.7 83.0 91.8 Nonenergy commoditiesb 43.8 125.5 100.0 86,9 79.1 80.4 86.9 97.4 106.8 109.5 Agriculture 45.8 138.1 100.0 87.7 80.1 80.9 88.0 100.3 112.3 114.8 Beverages 56.9 181.4 100.0 88.4 71.8 73.4 80.5 101.4 123.6 130.8 Food 46.7 139.3 100.0 84.5 86.5 87.2 91.6 100.4 107.4 100.4 Fats and oils 64.4 148.7 100.0 96.2 89.5 95.2 100.9 115.1 126.6 132.8 Grains 46.7 134.3 100.0 79.5 77.4 82.0 90,0 101.0 106.7 110.6 Other food 32.2 134.3 100.0 77.7 89.1 83.7 84.8 88.0 92.0 68.1 Raw materials 36A 104.6 100.0 91.4 78.1 78.4 89.1 99.6 110.4 121.7 Timber 31.8 79.0 100.0 111.0 91.0 91.0 107.0 123.1 136.6 155.5 Other raw materials 39.6 122.0 100.0 78.0 69.3 69.8 76.8 83.5 92.5 98.6 Fertilizers 30.4 128.9 100.0 105.8 97.9 97.7 99.0 105.2 111.3 122.7 Metals and minerals 40.4 94.2 100.0 83.0 74.9 77.3 82.9 89.3 92.6 95.2 Constant 1990 dollarse Petroleum 18.9 204 6 100.0 126.7 117.5 95.0 86.6 74.0 72.3 74.4 Nonenergy commodities 156.3 159.3 100.0 89.2 85.1 83.1 86.1 91.5 93.0 88.7 Agriculture 163.3 175.3 100.0 90.0 86.2 83.7 87.3 94.3 97.8 93.1 Beverages 202.8 230.3 100.0 90.7 77.2 75.9 79.8 95.3 107.7 106.0 Food 166.5 176.8 100.0 86.7 93.0 90.2 90.8 94.4 93.5 81.3 Fats and oils 229.5 188,7 100.0 98.8 96.2 98.5 100.0 108.2 110.3 107.6 Grains 166.6 170.5 100.0 81.6 83.3 84.8 89.2 94.9 92.9 89.6 Other food 114.9 170.5 100.0 79.8 95.9 86.5 84.0 82.7 80.1 55.2 Raw materials 129.8 132.7 100.0 93.8 84.1 81.1 88.3 93.6 96.1 98.6 Timber 113.3 100.3 100.0 113.9 97.9 94.2 106.0 115.7 118.9 126.0 Other raw materials 141.1 154.9 100.0 80.0 74.6 72.2 76.2 78.5 80.5 79.9 Fertilizers 10.3 163.6 100.0 108.6 105.3 101.0 98.2 98.9 96.9 99.4 Metals and minerals 143.9 119.6 100.0 85.2 80.5 80.0 82.1 84.0 80.7 77.1 Inflation indices, 1990=100d MUV indexe 28.05 78.78 100.00 97.41 92.96 96.66 100.90 106.36 114.84 123.42 Percent change per annum 10.88 2.41 -0.26 -4.56 3.98 4.38 2.67 1.55 1.45 U.S. GDP deflator 33.59 65.93 100.00 123.73 126.58 128.86 131.43 137.28 153.06 170.65 Percent change per annum 6.98 4.25 2.15 2.30 1.80 2.00 2.20 2.20 2.20 a. Commodity price projections as of October 12, 2001. b. The World Bank primary commodity price indices 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 indices expressed as ratios to the nonenergy index are as follows in percent: agriculture 69.1, fer- tilizers 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. c. Computed from unrounded data and deflated by the MUV index. d. Inflation indices for 2001-10 are projections as of October 3, 2001. MUV for 2000 is an estimate. Growth rates for years 1980, 1990, 2000, 2005, 2010, and 2015 refer to compound annual rate of change between adjacent end-point years; all others are annual growth rates from the previous year. e. 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. Source: World Bank, Economic Policy and Prospects Group; Historical U.S. GDP deflator; U.S. Department of Commerce. 231 Appendix 3 Global Economic Indicators GLOBAL ECONOMIC PROSPECTS Table A3.1 Growth of real GDP, 1971-2010 (GDP in 1995 prices and exchange rates-average annual percentage growth) .2000 GDP billios of Estimate Forecast US. dollars) 1971-80 1981-90 1991-00 2000 2001 2001-2010 World 31,981. 3.8 3.2 2.6 3.8 13 2.9 High-income economies 25,599 3.4 3.1 2.4 3.4 0.9 2.5 Industrial 24,811 3.4 3.0 2.4 3.3 0.9 2.4 G-7 21,028 3.4 3.1 2.3 3.2 0.7 2.4 United States 9,873 3.1 3.2 3.2 4.1 1.1 2.7 Japan 4,752 4.5 4.0 1.4 1.5 -0.8 2.0 G-4 Europe 5,693 2.9 2.4 1.8 3.1 1.4 2.3 Germanya 1,872 2.7 2.2 1.8 3.1 0.6 1.9 Euro Area 6,077 3.3 2.5 2.1 3.5 1.5 2.4 Non-G7 Industrial 3,783 3.2 2.7 2.8 3.9 1.9 2,8 Other high-income 788 7.7 5.2 5.2 6.3 0.6 3.9 Asian NIEs 571 9.5 7.4 6.1 7.8 0.4 4.2 Low- and middle-income economies 6,401 5.4 3.5 3.2 5.5 2.9 4.5 Excluding ECA 5,365 5.5 3.5 4.8 5.3 3.1 4.7 Asia 2,595 5.4 7.2 6.8 6.9 4.6 5.8 East Asia and Pacific 1,982 6.6 7.8 7.2 7.5 4.6 6.0 China 1,080 5.3 9.2 10.1 8.0 7.2 ... Korea, Rep. of 457 7.6 9.1 6.1 8.8 2.5 .. Indonesia 153 7.9 6.4 4.2 5,2 3.6 South Asia 612 3.1 5.8 5.2 4.9 4.5 5.3 India 479 3.0 5.9 5.6 5.2 4.5 Latin America and the Caribbean 1,949 5.9 1.1 3.3 3.8 0.9 3.5 Brazil 588 8.5 1.5 2.7 4.4 1.4 ... Mexico 584 6.7 1.8 3.5 6,9 0.6 ... Argentina 285 3.0 -1.5 4.6 -0.5 -1.9 ... Europe and Central Asia 993 5.2 3.5 -2.3 6.3 2.1 3.4 Russian Federationb 247 5.2 4.7 -5.2 8.3 4,7 ... Turkey 200 4.2 5.2 3.6 7,1 -7.4 ... Poland 158 5.0 -0.1 3.7 4.2 1.4 Middle East and North Africa 549 6.6 2.4 3.2 3.9 3.4 3.3 Saudi Arabia 149 10.3 0.4 2.3 4.0 1.5 Iran, Islamic Rep. 134 1.8 2.7 4.1 5.2 41 ... Egypt, Arab Rep. 96 6.6 5.5 4.4 5.1 4.3 Sub-Saharan Africa 315 3.3 1.7 2.2 3.0 2.7 3.6 Republic of South Africa 126 3.5 1.3 1.7 31 2.4 - Nigeria 41 4.7 1.1 2.5 2.8 3.1 a. Data prior to 1991 covers West Germany. b. Data prior to 1992 covers the former Soviet Union. Note: This table comprises a sample of 145 countries representing 99 percent of world GDP. Source: World Bank data and staff estimates. Figure A3.1 Real GDP growth, 1991-2010 Prcent 8.0 E3 1991- 200 60 [] 201-201 40 2.0 High-income East Asia South Asia Latin America Europe and Middle East Sub-Saharan economies and Pacific and the Central Asia and North Africa Caribbean Africa 234 GLOBAL ECONOMIC INDICATORS Table A3.2 Growth of real per capita GDP, 1971-2010 (GDP in 1995 prices and exchange rates-average annual percentage growth) 2000 GIDPI per capita (Cuantt Estimiate Forecast US. dollars) 1971-80 1981-90 1991-00 2000 2001 2001-2010 World 5,530 1.9 1.4 1.2 2.4 0.0 1.8 High-income economies 28,751 2.6 2.4 1.8 2.8 0.5 2.2 Industrial 29,395 2.6 2.5 1.8 2.8 0.5 2.2 G-7 30,421 2.6 2.5 1.7 2.7 0.3 2.1 United States 35,840 2.1 2.2 2.2 3.2 0.3 2.0 Japan 37,520 3.3 3.4 1.1 1.5 -0,9 2.0 G-4 Europe 22,048 2.6 2.1 1.5 30 1.3 2.3 Germany' 22,821 2.6 2.1 1.5 3.1 0.7 2.1 Euro Area 20,084 2.8 2.2 1.8 3.3 1.4 2.5 Non-G7 Industrial 24,754 2.4 2.1 2.3 3 1.6 2.7 Other high-income 17,010 5.2 3.4 3.7 4.7 -0.7 2.8 Asian NIEs 17,595 7.2 5.9 4.7 6.4 -0.7 3.4 Low- and middle-income economies 1,301 3.2 1.5 1.6 4.0 1.5 3.2 Excluding ECA 1,211 3.2 1.4 3.1 3.6 1.5 3,3 Asia 861 3.2 5.3 5.2 S.4 3.2 4.7 East Asia and Pacific 1,175 4.5 6.1 6.0 6.4 3.6 5.1 China 857 3.4 7.6 9.0 7.0 6.4 ... Korea, Rep. of 9,684 5.7 7.8 5.1 7.9 1.7 Indonesia 729 5.4 4.4 2.5 3.5 2.1 South Asia 461 0.6 3.5 3.3 3.0 2.8 3.8 India 472 0.7 3.6 3.7 3.3 2.8 . Latin America and the Caribbean 3,889 3.4 -0.9 1.6 2.2 -0.7 2.1 Brazil 3,450 5.9 -0.4 1.3 3.0 0.2 ... Mexico 5,900 3.6 -0.3 1.7 5.2 -1.3 Argentina 7,703 1.3 -2.9 3.2 -1.7 -3.2 .. Europe and Central Asia 2,174 4.1 2.6 -2.5 6.1 1.9 3.3 Russian Federationb 1,693 4.2 3.8 -5.2 8,6 5.0 Turkey 3,062 1.8 2.8 2.0 5.5 -8.7 Poland 4,071 4.1 -0.8 3.5 4.1 1.4 Middle East and North Africa 1,948 3.6 -0.6 1.0 1.9 1s 14 Saudi Arabia 6,714 5.1 -4.8 -1.1 0.7 -1.5 Iran, Islamic Rep. 1,415 -1.4 -0.7 2.4 3.5 2.5 ... Egypt, Arab Rep. 1,508 4.4 2.9 2.4 3.5 2.7 ... Sub-Saharan Africa 484 0.5 -1.2 -0.4 0.5 0.3 1.3 Republic of South Africa 2,942 1.2 -1.2 -0.3 1.4 1.0 Nigeria 325 1.7 -1.9 -0.3 0.4 0.3 ,.. a. Data prior to 1991 covers West Germany. b. Data prior to 1992 covers the former Soviet Union. Note: This table comprises a sample of 145 countries representing 99 percent of world GDP. Source: World Bank data and staff estimates. Figure A3.2 Real per-capita GDP growth, 1991-2010 Percent 6.0 O 1991-2000 o 2001-2010 4.0 2.0 -2.0 High-income East Asia South Asia Latin Amenca Europe and Middle East Sub-Saharan economies and Pacific and the Central Asia and North Africa Caribbean Africa 235 GLOBAL ECONOMIC PROSPECTS Table A3.3 Inflation: GDP deflators, 1971-2010 (percentage changea) Estimate Forecast 1971-80 1981-90 1991-00 2000 2001 2001-2010 World 9.0 5.8 4.1 2.2 2.8 1.9 High-income economies 8.8 52 2.3 1.2 1.7 1.2 Industrial 8.6 4.6 1.9 1.2 1.8 1.1 G-7 8.3 4.2 1.8 0.8 1.5 1.1 United States 7.1 4.2 2.2 2.3 2.7 1.8 Japan 7.8 1.9 0.1 -1.7 -0.6 -0.4 G-4 Europe 9.7 5.7 2.7 0.7 1.5 1.4 Germanyb 5.3 2.8 2.9 -0.4 1.0 2.4 Euro Area 9.4 6.0 2.9 1.2 1.8 1.2 Non-G7 Industrial 10.4 6.9 2.8 3.1 2.9 0.8 Other high-income 19.3 32.1 15.7 -0.2 1.3 3.1 Asian NIEs 9.5 4.7 2.4 -2.0 1.2 3.0 Low- and middle-income economies 9.8 8.3 11.8 6.4 6.7 4.6 Excluding ECA 11.5 9.8 8.9 5.8 6.4 4.5 Asia 11.0 7.0 7.0 4.8 7.0 5.0 East Asia and Pacific 11.0 5.6 3.4 3.4 7.1 4.5 China 1.8 5.5 6.3 1.0 5.1 ... Korea, Rep. of 20.8 7.1 4.8 -1.5 11.7 Indonesia 20.6 8.8 15.0 9.4 10.5 South Asia 11.9 8.9 8.1 5.8 6.1 6.6 India 8.9 8.3 8.4 6.9 3.3 ... Latin America and the Caribbean 14.6 19.3 12.6 6.9 7.9 6.4 Brazil 39.7 330.8 205.6 7.2 11.1 ... Mexico 18.1 63.7 18.1 11.8 10.9 ... Argentina 117.7 439.5 10.2 1.1 -0.5 ... Europe and Central Asia 0.3 2.4 347.1 7.5 7.5 4.9 Russian Federationc 0.3 2.3 179.1 41.1 18.1 Turkey 32.6 46.3 72.1 53.3 53.2 ... Poland 4.5 72.5 23.5 7.6 7.7 Middle East and North Africa 11.8 7.7 5.2 3.4 4.5 4.1 Saudi Arabia 23.8 -3.1 1.2 2.5 4.5 Iran, Islamic Rep. 20.2 15.6 25.6 22.5 20.7 ... Egypt, Arab Rep. 11.0 13.1 8.4 4.0 5.0 ... Sub-Saharan Africa 10.3 9.4 9.7 6.3 6.0 4.4 Republic of South Africa 13.3 15.1 9.9 6.9 6.9 ... Nigeria 13.4 16.6 28.8 26.1 18.5 ... Note: Deflators are in local currency units: 1995=100. a. High-income group inlation rates are GDP-weighted averages of local currency inflation. Low- and middle-income groups are medians. World is GDP-weighted average of the two groups. b. Data prior to 1991 covers West Germany. c. Data prior to 1992 covers the former Soviet Union. Note: This table comprises a sample of 145 countries representing 99 percent of world GDP. Source: World Bank data and staff estimates. Figure A3.3 GDP inflation, 1991-2010 Percent 250 20.0 liE 1991-2000] 0 2001-2010 150 10.0 High scome East Asia South Asia Latin America Europe and Middle East Sub-Sharan economies and Pacific and the Central Asia and North Africa Caribbean Africa 236 GLOBAL ECONOMIC INDICATORS Table A3.4 Current account balances, 1971-2010 (percentage of GDP) 2000 cureant account (billions of Estimate Forecast US. dollatrs) 1971-80 1981-90 1991-00 2000 2001 2001-2010 World -197 -0.2 -0.6 -0.3 -0.6 -(0.7 -0.8 High-income economies -276 0.0 -0.3 -0.1 -1.1 -09 -0.6 Industrial -328 -0.2 -0.5 -0.2 -1.3 -1.0 -0.6 G-7 -335 -0.1 -0.4 -0.3 -1.6 -1.3 -0.9 United States -445 0.0 -1.9 -1.9 -4.5 -3.9 -2.3 Japan 117 0.2 2.3 2.4 2.5 2.2 2.1 G-4 Europe -25 0.1 0.3 0.0 -0.4 0.1 -0.7 Germanya -21 0.5 2.6 -0.7 -1.1 0.1 -1.0 Euro Area -10 -0.1 0.3 0.3 -0.2 0.3 -0.1 Non-G7 Industrial 7 -1.1 -0.8 0.3 0.2 0.7 0.7 Other high-income 52 5.9 3.9 3.2 6.6 4.1 1.2 Asian NIEs 35 1.2 0.4 4.2 6.1 4.8 3.9 Low- and middle-income economies 79 -0.5 -1.2 -1.2 1.2 0.1 -1.3 Excluding ECA 60 -0.5 -1.7 -1.4 1.1 -0.1 -1.4 Asia 64 -1.1 -1.3 0.0 2.5 0.9 -0.5 East Asia and Pacific 73 -1.4 -1.0 0.5 3.3 1.4 -0.4 China 21 0.1 0.2 1.5 1.3 0.2 Korea, Rep. of 11 -6.9 0.7 0.9 2.4 1.2 Indonesia 8 -2.3 -3.1 -0.4 5.5 1.0 South Asia - -0.5 -2.0 -1.4 -0.3 4.5 -0.7 India -1 0.2 -1.7 -1.1 -0.2 -0.6 Latin America and the Caribbean -47 -2.8 -1.7 -2.8 -2,4 -2.8 -2,9 Brazil -25 -4.3 -1.6 -2.2 -4.2 -4.8 Mexico -17 -4.0 -0.6 -3.7 -3.0 -28 Argentina -10 -0.4 -2.1 -3.2 -3.5 -2.9 Europe and Central Asia 19 -0.6 -0.2 -0.5 1.9 1.2 -1.2 Russian Federationb 46 2.1 3.5 4.7 18.7 10.5 Turkey -10 -2.1 -1.4 -1.1 -4,9 3.0 Poland -10 -3.0 -6.5 -2.8 -6.3 -5.1 . Middle East and North Africa 41 7.4 -1.6 -1.9 8.1 4.7 -1.1 Saudi Arabia 25 21.7 -7.2 -5.8 16.6 9.2 Iran, Islamic Rep. 9 6.7 -0.4 0.8 10.1 7.6 Egypt, Arab Rep. -1 -5.2 -3.4 1.5 -4.2 -.0.2 Sub-Saharan Africa -5 -1.7 -2.6 -2.1 -1.5 -1.0 -1.8 Republic of South Africa -1 -1.1 0.6 -0.2 -0.4 0.4 Nigeria 2 1.6 -.0.7 1.0 4.6 5.2 a. Data prior to 1991 covers West Germany. b. Data prior to 1992 covers the former Soviet Union. Note: This table comprises a sample of 145 countries representing 99 percent of world GDP. Source: World Bank data and staff estimates. Figure A3.4 Ratio of current account balance to GDP, 1991-2010 Percent 2.0 0 1991-2000 1.0 0 2001-aulu -1.0 -I -2.0 -3.0 -4.0 High-income East Asia South Asia Latin America Europe and Middle East Sub-Saharan economies and Pacific and the Central Asia and North Africa Caribbean Africa 237 GLOBAL ECONOMIC PROSPECTS Table A3.5 Exports of goods, 1999 (percent) Merehandise Average Effective Merchandise Average Effective Merchandise Average Effective exports annual market exports annual market exports annual market (US$ growth growth (USs growth growth (USS growth growth milomg) 1990-99 1990-99 mihowns) 1990-99 1990-99 miillions) 1990-99 1990-99a World *.i7.31i 5.9 1.0 Latin America and Middle Ei and the Caribbean iconrinued) North Nric3 icontinuedi All deWloping U .u,Ca. 2,304 - l fI. ni., i ' 2 4 i.i ctrys. 1,374,752 7.1 7.0 Venezuela 20,819 4.6 8.5 Yemen, Rep. 2,478 -3.5 8.1 Asia 621,080 12.0 7.5 Europe and Sub-Saharan Central Asia 253,930 2.6 4.9 Africa 81,154 2.7 6.6 East Asia and Armenia 247 -1.5 1.8 Angola 5,009 5.2 7.7 Pacific 564,636 12.3 7.6 Azerbaijan 1,025 5.6 2.1 Botswana 2,671 2.2 0.0 China 194,931 16.0 7.4 Belarus 5,646 3.1 0.2 Cameroon 1,675 -1.0 6.8 Fiji 538 4.0 6.2 Bulgaria 4,006 -10.0 5.3 Cote d'Ivoire 4,104 5.5 6.5 Indonesia 48,666 8.4 7.1 Czech Rep. 26,242 8.6 6.8 Ethiopia 467 3.6 6.7 Korea, Rep. 143,537 12.3 8.5 Estonia 2,4S3 11.3 2.5 Gabon 3,459 6.3 7.4 Malaysia 84,052 10.8 7.3 Georgia 330 9.0 2.6 Ghana 2,117 11.2 6.5 Myanmar 1,125 15.5 8.0 Hungary 21,846 4.0 5.4 Kenya 1,740 6.9 5.3 Papua New Kazakhstan 5,989 1.9 1.3 Madagascar 507 4.6 5.7 Guinea 1,927 4.2 6.4 Kyrgyz Rep. 463 9.0 -0.1 Nigeria 12,876 5.0 71 Philippines 34,210 12.4 7.3 Lativia 1,889 6.8 2.2 Senegal 1,027 2.2 5.9 Thailand 56,775 9.1 7.3 Lithuania 3,147 3.1 2.7 South Africa 28,624 1.0 6.4 Vietnam 8,779 23.4 7.0 Moldova 469 -4.9 -0.5 Sudan 780 5.9 6.1 Poland 26,347 7.5 5.5 Zambia 864 -2.2 5.4 South Asia 56,444 9.2 6.7 Romania 8,503 -1.9 6.2 Zimbabwe 1,879 4.4 5.6 Bangladesh 5,458 14.6 6.6 Russian India 37,528 9.8 6.7 Federation 75,900 -1.5 5.2 High-income Nepal 709 13.7 7.1 Slovak Rep. 10,201 5.4 6.5 economies 4,221,189 5.6 7.0 Pakistan 8,164 4.8 6.9 Tajikistan 689 -1.4 3.$ Sri Lanka 4,586 9.0 6.6 TFYR Industrial 3,735,447 5.5 6.9 Macedonia 1,192 -2,0 5.5 Latin America a u Turkmenistan 1,223 -5.1 4.1 G-7 2,726,116 5.5 7.2 the Caribbean 302,952 5.5 8.1 Turkey 29,326 9.1 5.3 Canada 245,832 7.6 7.7 Argentina 23,333 5.8 8.1 Ukraine 13,189 4.2 2.0 France 334,103 6.0 6.4 Bolivia 1,051 4.4 9.3 Uzbekistan 1,976 0.9 2.1 Germany 544,281 5.7 6.2 Brazil 48,011 0.1 8.5 Italy 232,082 3.5 6.7 Chile 15,555 3.8 7.8 Middle East an Japan 402,647 2.7 8.4 Colombia 12,030 4.4 7.8 North Africa 115,635 3.9 6.7 United Costa Rica 6,668 12.9 7.6 Algeria 11,514 1.3 6.5 Kingdom 268,921 5.3 6.6 Dominican Rep. 5,137 14.6 7.9 Bahrain 4,140 2.7 6. United States 698,250 7.4 7.6 Ecuador 4,451 5.6 8.0 Egypt, Arab El Salvador 2,500 11.3 8.9 Rep. 5,236 4.9 6.3 Other Guatemala 2,781 6.2 8.8 Iran, Islamic industrial 1,009,331 5.7 6.2 Jamaica 1,490 1.5 6.8 Rep. 15,271 2.0 6.3 Australia 56,048 6.3 6.8 Mexico 136,392 12.0 7.9 Iraq 10,838 5.5 Austria 64,424 6.7 6.3 Panama 5,300 6.1 8.1 Jordan 1,832 8.4 6.2 Belgiumb 154,069 5.1 6.3 Paraguay 2,707 4.4 10.3 Morocco 7,509 5.2 5.9 Denmark 49,823 2.7 6.5 Peru 6,112 0.7 7.4 Oman 7,218 5.8 7.6 Finland 41,983 5.0 5.7 Trinidad and Saudi Arabia 50,757 5.6 7.0 Greece 5,249 -2.8 5.7 Tobago 2,402 4.8 8.1 Syrian Arab Re 3,806 2.9 5.1 Iceland 2,009 0.9 5.9 238 GLOBAL ECONOMIC INDICATORS Table A3.5 Exports of goods, 1999 (continued) (percent) Merchandise Average Effective Merchandise Average Effective Merchandise Average Effective exports annual market exports annual market exports annual market (US$ growth growth (US$ growth growth (US$ growth growth millins) 1990-99 1990-994 milions) 1990-99 199 milkon) 1990-99 1990-99' Other industrial (continued) Other high- Other high-incoe (continued) Ireland 71,375 11.4 5.9 income 485,742 6.2 8.1 Kuwait 12,276 0.5 7.1 Netherlands 199,026 6.5 6.0 Brunei 2,552 1.6 6.3 Singapore 115,639 5.7 8.4 New Zealand 12,594 2.7 6.6 Hong Kong, Taiwan, China 121,129 5.3 7.8 Norway 45,651 4.2 6.0 China 173,865 8.3 8,6 United Arab Spain 113,490 8.7 6.3 Israel 25,564 6.8 6.8 Emirates 27,645 4.7 5.7 Sweden 84,849 6.0 5.9 Switzerland 83,320 2.3 6.6 .. 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, 1999 Percent 40.0 30.0 World 20 0 Industrial Sub-Saharan East Asia South Asia Latin America Europe and Middle East economies Africa and Pacific and the Central Asia and North Caribbean Africa Figure A3.5b Average annual growth rate of export volumes, 1990-99 Percent 14.0 12.0 10.0 World 8.0 6.0 0.0 Industrnal Sub-Saharan East Asia South Asia Latin America Europe and Middle East economies Africa and Pacific and the Central Asia and North Caribbean Africa 239 GLOBAL ECONOMIC PROSPECTS Table A3.6 Imports of goods, 1999 (percent) Merchandise Average Merchan- Merchandise Average Merchan- Merchandlse Average Merchan- imports annual dise imports annual dise imports annual dise (US$ growth Imports/ (US$ growth Imports/ fUS$ growth Imports/ millions) 1990-99 GDP millions) 1990-99 GDP ndions) 1990-99 GDP World 5,604,722 6.2 18.1 Latin America a Middle East and the Caribbean Comnued) North Africa (c nued) All developing Uruguay 3,173 8.8 15.3 Tunisia 8,014 5.2 38.3 ctrys. 1,Z70,637 6.2 19.7 Venezuela 13,213 4.7 12.8 Yemen, Rep. 2,120 -4.9 31.3 Asia 521,113 8.2 19.0 Europe and Sub-Saharan Central Asia 265,272 2.5 30.6 Africa 76,453 3.5 24.3 East Asia and Armenia 721 0.1 39.1 Angola 2,167 3.3 35.0 Pacific 451,721 8.6 20.8 Azerbaijan 1,433 3.9 31.8 Botswana 1,996 2.9 39-5 China 165,699 10.5 16.7 Belarus 6,216 -0.6 22.6 Cameroon 1,218 -0.1 13.3 Fiji 653 2.1 37.9 Bulgaria 5,087 -7.3 42.0 C6te d'Ivoire 2,513 2.1 22.4 Indonesia 24,003 5.0 16.8 Czech Rep. 28,073 7.3 52.9 Ethiopia 1,387 4.5 21.5 Korea, Rep. 119,631 8.1 29.5 Estonia 3,331 2.4 64.9 Gabon 1,514 5.9 34.8 Malaysia 61,404 9.7 77.7 Georgia 863 10.2 31.7 Ghana 3,228 10.9 41.5 Myanmar 2,160 19.6 62 Hungary 24,037 7.8 49.8 Kenya 2,570 1.5 24.1 Papua New Kazakhstan 5,645 -3.1 33.5 Madagascar 518 4.0 13.9 Guinea 1,071 -3.1 29.9 Kyrgyz Rep. 547 -1.6 43.8 Nigeria 8,588 7.2 24.5 Philippines 29,252 9.4 38.4 Lativia 2,916 3.0 43.8 Senegal 1,373 1.9 28.9 Thailand 47,847 4.8 39.2 Lithuania 4,551 1.6 42.7 South Africa 24,474 2.5 18.8 Vietnam 12,383 16.0 42.0 Moldova 597 -3.7 50.9 Sudan 1,256 1.8 12.9 Poland 40,727 11.9 26.2 Zambia 756 -1.4 24.4 South Asia 69,392 5.5 12.1 Romania 9,595 0.3 27.3 Zimbabwe 1,712 1.3 30.5 Bangladesh 7,420 7.2 16.1 Russian India 45,556 6.0 10.1 Federation 39,600 -5.0 21.0 High-income Nepal 1,590 8.9 34.6 Slovak Rep. 11,310 8.9 57.6 economies 4,334,085 6.2 17.7 Pakistan 9,533 1.2 16.4 Tajikistan 663 -2.5 612 Sri Lanka 5,293 8.6 33.7 TFYR Industrial 3,856,210 6.0 16.3 Macedonia 1,602 3.0 46.4 Latin America ai Turkmenistan 1,210 0.3 36.6 G-7 2,866,532 6.2 13.9 the Caribbean 310,945 10.8 14.8 Turkey 39,773 9.9 22.6 Canada 219,994 6.4 33.5 Argentina 24,103 18.1 8.5 Ukraine 12,945 -4.3 42.1 France 304,819 4.7 21.2 Bolivia 1,539 5.6 18.5 Uzbekistan 2,621 -3.8 14.2 Germany 485,257 6.9 23.0 Brazil 49,272 9.3 9,3 Italy 218,111 4.0 18.2 Chile 13,951 6.4 20.6 Middle East an Japan 280,190 4.9 6.2 Colombia 10,254 7.4 11.8 North Africa 96,853 1.8 19.4 United Kingdo 311,261 4.8 21.6 Costa Rica 6,008 13.3 38.6 Algeria 8,746 -1.0 18.3 United States 1,046,900 7.8 11.3 Dominican Rep. 8,041 14.6 46.2 Bahrain 3,468 1.4 51.0 Ecuador 2,786 3.6 14.7 Egypt, Arab Other El Salvador 3,859 11.2 31.0 Rep. 15,165 4.0 17,0 industrial 989,677 5.4 29.2 Guatemala 4,226 9.9 23.2 Iran, Islamic Australia 65,828 5.5 16.6 Jamaica 2,628 4.4 39.3 Rep. 12,483 -2.4 11.3 Austria 67,749 6.8 32.3 Mexico 141,973 14.5 18.9 Iraq 5,516 21.7 19.9 Belgiuma 146,814 4.5 66.2 Panama 6,715 7.8 703 Jordan 3,292 4.6 40.8 Denmark 43,135 3.7 24.5 Paraguay 3,041 9.8 39.3 Morocco 9,957 5.6 28.4 Finland 29,815 1.0 23.1 Peru 6,729 9.9 13.0 Oman 4,300 5.2 22.0 Greece 21,700 3.4 17.3 Trinidad and Saudi Arabia 25,717 1.8 18.5 Iceland 2,317 4.7 263 Tobago 2,558 9.5 32,7 Syrian Arab Re 3,590 5.6 22.5 Ireland 47,252 9.0 50.6 240 GLOBAL ECONOMIC INDICATORS Table A3.6 Imports of goods, 1999 (continued) (percent) Merchandise Average Merchan- Merchandise Average Merclan- Merchadise Average Merchan- imports annual dise hports annual dise hoports annual dise (US$ growth Imports/ (US$ growth Imports/ (S$ growth Imports/ Millions) 1990-99 GDP indions) 1990-99 GDP * n# 0 1990-99 GDP Other industrial Other high- Other high- (continued) 989,677 5.4 29.2 income 477,875 8.2 64.7 income (contin med) Netherlands 181,152 7.0 45.5 Brunei 1,328 2.2 48.4 Singapore 104,337 7.3 122.8 New Zealand 13,029 4.4 23.8 Hong Kong, Taiwan, Norway 35,532 2.6 23.3 China 179,861 9.2 113.7 China 110,585 8.6 38.5 Spain 144,882 7.7 23.7 Israel 29,972 7.1 29,7 United Arab Sweden 67,658 4.0 28.1 Kuwait 6,705 -0.7 2.6 Emirates 33,239 12.0 68.7 Switzerland 83,602 3.8 32.3 1 1 1 1 .. Not available a. Includes Luxembourg Source: See technical notes. Figure A3.6a Merchandise imports as a share of GDP, 1999 Peircent 40.0 200t 10.0 Industrial Sub-Saharan East Asia South Asia Latin America Euope and Middle East economies Africa and Pacific and the Central Asia and North Caribbean Africa Figure A3.6b Average annual growth rate of import volumes, 1990-99 Percent 12.0 World Industrial Sub-Saharan East Asia South Asia Latin America Europe and Middle East economies Africa and Pacific and the Central Asia and North Canbbean Africa 241 GLOBAL ECONOMIC PROSPECTS Table A3.7 Direction of merchandise trade, 1999a (percentage of world trade) High-income importers Low- and middle-income importers Latin Middle America All Sub- East Europe East and low- Other All Other All Saha- Asia and and the and United indus- inds- high hgh- ran and South Central North Carib- middle- Source of exports States EU-15 Japan trial trial income income Africa Pacific Asia Asia Africa bean income world High-income econ. 12.6 29.7 3.0 7.0 $2.4 5.5 57.9 0.9 6.5 0.7 3.2 1.5 4.3 16.9 74.8 Industrial 10.6 28.2 22 6.7 47.7 4.3 52.0 0.8 4.1 0.4 3.0 1.4 4.1 13.8 65.8 United States ... 2.7 1.1 3.6 7.4 12 8.6 0.1 1.1 0.1 0.2 0.3 2.7 4.5 13.1 EU-15 3,6 22.3 0.7 2.4 29.0 1,4 30.4 0.5 0.9 0.2 2.7 0.9 0.9 6.2 36.5 Japan 2.5 1.3 ... 0.4 4.3 1.4 5.7 0.1 1.6 0.1 0.1 0.1 0.3 2.3 8.0 Other industrial 4.4 1.9 0.4 0.3 7.0 0.3 7.4 0.0 0.4 0.1 0.1 0.1 0.1 0.8 8.2 Other high-incomeb 2.1 1.5 0.8 0.3 4,7 1.2 5.9 0.1 2.4 0.2 0.1 0.1 0.2 3.1 9.0 Low- and middle- income economies 6.7 6.0 2.1 0.9 15,7 3.0 18,7 0.5 2.1 0.5 1.6 0.5 1.3 6.5 25.2 Sub-Saharan Africa 0.2 0.4 0.0 0.0 0.7 0.1 0.8 0.2 0.1 0.0 0.0 0.0 0.0 0.4 1.2 East Asia and Pacific 2.4 1.6 1,6 0.4 6.0 2.3 8.4 0.1 1.4 0.2 0.2 0.2 0.3 2.4 10.8 South Asia 0.3 0.3 0.1 0.0 0.6 0.1 0.8 0.0 0.1 0.0 0.0 0.0 0.0 0.3 1.0 Europe and Central Asia 0.3 2,3 0.1 0.1 2.7 0.1 2.9 0.0 0.2 0.0 1.2 0.1 0.0 1.6 4.5 Middle East and North Africa 0.2 0.7 0.3 0.0 1.2 0.2 1.5 0.0 0.3 0.1 0.1 0.1 0.0 0.7 2.1 Latin America and Caribbean 3.4 0.7 0.1 0.2 4.4 0.1 4.4 0.0 0.1 0.0 0.1 0.1 0.8 1.1 3.6 World 19.4 35.7 5.2 7.9 68.1 8.5 76.6 1.3 8.6 1.2 4.8 2.0 5.6 23.4 100.0 a. Expressed as a share (percent) of total world exports. World merchandise exports in 1999 amounted to some $5,135 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. 242 GLOBAL ECON-OMIC INDICATORS Table A3.8 Growth of current dollar merchandise trade, by direction 1990--99a (average annual percentage growth) Righ-income importers Low- and middle-income importers Latin Middle America All Sub- East Europe East and low- Other All Other All Saha- Asia and and the and Uneitd Indus- indus- high- high- ran and South Central North Carib- middle- Source of exports States EU-15 Japan trial trial income income Africa Pacific Asia Asia Africa bean income World High-income econ. 6.6 3.8 3.0 5.0 4,5 7.7 4.8 2.0 8.5 2.8 9.3 3.5 10.6 9.0 5.4 Industrial 6.7 3.6 .4 $.1 4.3 7.0 4.3 1.6 7.2 0.8 9.1 3.5 10.6 8.6 5.0 United States ... 4.2 2.6 7.1 5.2 7.1 5.4 4.1 7.6 1.3 2.1 4.7 11.6 9.4 6.5 EU-15 7.1 3.6 3.5 15 4.A 79 4.1 1.4 7.6 -0.1 11.0 3.3 9.8 9.5 4.6 Japan 33 2.8 ... 0.7 2.9 6.5 3.6 -0.7 7.1 0.2 -3.2 2.6 7.9 6.3 4.2 Other industrial 9.0 3.1 03 4.3 6.2 5.0 61 3.1 5.8 5.6 0.7 3.6 5.4 5.9 5.9 Other high-incomeb 6.2 7.6 4.9 44 6.2 10.7 7.0 6.3 11.3 8.1 16.6 4.2 10.5 10.7 8.1 Low- and middle- income economies 12.3 103 6.5 10.3 10.4 10.3 104 13.5 18.2 13.3 11.9 7.5 13.2 16.2 11.1 Sub-Saharan Africa 3.7 39 6,7 14.0 4.3 23.6 5.0 14.2 24.2 23.7 3.1 10.3 18.5 17.3 7.5 East Asia and Pacific 11.7 12.6 7.4 11.8 10.6 10.6 10.6 13.7 19.9 13.5 8.7 10.9 20.9 18.0 11.7 South Asia 9.2 7.0 -.7 8.4 7.0 11.7 7.6 21.4 10.5 11.4 9.7 7.3 27.9 12.8 8.5 Europe and Central Asia 13.5 9.4 -3.1 10.5 9.3 151 9.5 6.1 7.9 2.3 4.9 2.3 7.9 12.3 7.6 Middle East and North Africa 1.2 4.5 .0 -3.4 3.8 5.4 4.0 11.8 20.5 10.4 -0.3 3.5 6.3 11.6 5.5 Latin America and Caribbean 15.3 2.9 1.0 7.8 11.1 41 10.9 4.7 5.9 12.7 -1.0 6.5 11.4 10.1 10.5 World 8.2 4.3 4.1 5.4 5.4 8.5 5.7 4.6 10.0 5.6 7.5 4.2 11.1 10.1 6.3 Note: Growth rates are compound averages. Source: IMF, Direction of Trade Statistics. 243 GLOBAL ECONOMIC PROSPECTS Table A3.9 Structure of long-term debt, 1999 Share of long-term debt (%): concessional debt; nonconcessional debt at variable interest rates; nonconcessional debt at fixed interest rates Non-concessional Non-concessional Concessional Variable Fixed Concessional Variable Fixed All developing countries 19.0 43.4 37.6 Europe and Central Asia continued) Bulgaria 2.6 78.6 18.8 Asia 28.1 39.0 33.0 Czech Republic 1.2 38.6 60.2 East Asia and Pacific 19.3 43.9 36.7 Estonia 1.7 16.6 81.7 China 21.7 26.7 51.6 Georgia 61.8 9.0 29.2 Indonesia 26 1 62.8 11.1 Hungary 19 14.4 83.7 Korea, Rep. 0.8 55.4 43.8 Kazakhstan 8.1 40.7 51.2 Malaysia 65 61.9 31.6 Kyrgyz Republic 35.1 16.3 28.6 Myanmar 88.2 0.0 11.8 Latvia 4,7 36.4 58.9 Papua New Guinea 38.2 11.3 50.5 Lithuania 4.3 29.6 66.1 Philippines 30.0 31.6 38.4 Moldova 31.7 49.7 18.6 Thailand 14.0 45.0 41.0 Poland 15.1 45.7 39.2 Vietnam 27.4 12.3 60.3 Romania 4.6 38.5 56.9 Russian Federation 0.3 42.3 57.4 South Asia 58.5 21.7 19.9 Slovak Republic 5.1 24.3 70.6 Bangladesh 99.0 0.1 1.0 Tajikistan 74,0 8.7 17.3 India 49.5 23.9 26.6 Turkmenistan 13.3 70.0 16.7 Nepal 99.3 0.0 0.7 Turkey 7.2 47.2 45.7 Pakistan 60.5 28.0 11.5 Ukraine 30.3 31.4 38.1 Sri Lanka 90.0 5.3 4.6 Uzbekistan 19.4 44.2 36.4 Latin America and Middle East and the Caribbean 4.6 59.2 36.2 North Africa 31.9 30.5 31.6 Argentina 1.8 48.8 49.3 Algeria 12.7 51.0 36.3 Bolivia 62.9 22.3 14.8 Egypt, Arab Rep. 86.3 4.2 9.5 Brazil 0.8 77.7 21.5 Jordan 53.7 24.3 22.0 Chile 1.2 52.3 46.5 Morocco 32.0 34.3 33.6 Colombia 3.2 61.6 35.2 Oman 33.3 31.8 34.8 Costa Rica 19.4 27.2 53.4 Syrian Arab Republic 92.8 0.0 7.2 Dominican Republic 41.6 30.9 27.5 Tunisia 25.1 22.0 52.9 Ecuador 15.4 56.0 28.6 Yemen, Rep. 91.6 2.1 6.2 El Salvador 39.5 36.9 23.6 Guatemala 40.9 29.4 29.7 Sub-Saharan Africa 51.5 14.1 34.5 Jamaica 32.5 27.4 40.1 Angola 29.8 10.4 59.9 Mexico 1.0 49.1 50.0 Botswana 60.6 9.5 29.9 Panama 6.4 47.5 46.1 C6te d'Ivoire 39.1 46.6 14.2 Paraguay 53.5 20.8 25.7 Cameroon 53.6 11.8 34.6 Peru 15.4 51.3 33.4 Ethiopia (excludes Eritrea) 89.3 0.3 10.3 Trinidad and Tobago 0.7 44.1 55.3 Gabon 24.9 11.5 63.5 Uruguay 4,5 50.3 45.2 Ghana 81.6 4.6 13.7 Venezuela 0.2 72.3 27.5 Kenya 73.9 7.7 18.4 Madagascar 72.6 5.2 22.2 Europe and Nigeria 7.6 20.2 72.2 Central Asia 6.5 42.2 51.4 Senegal 80.7 10.2 9.1 Armenia 69.0 18.2 12.8 Sudan 49.8 17.8 32.5 Azerbaijan 50.4 29.7 19.9 Zambia 773 8.7 14.0 Belarus 9.8 47.4 42.8 Zimbabwe 45.1 21.7 33.2 Note: Nonconcessional debt data are available only for countries which report to the World Bank's Debtor Reporting System. Non-concessional debt contains estimates of private non-guaranteed in addition to public and publicly guaranteed debt. For aggregate figures, missing values are assumed to have the same average value as the available data. 244 GLOBAL ECONOMIC INDICATORS Figure A3.9a Structure of long-term debt, by group, 1999 Percent 100 80 -- -- - - - 60 40 - 60 Severely Moderately Severely Moderately Other indebted indebted indebted indebted low-income low-income middle-income middle-income Figure A3.9b Structure of long-term debt, by region, 1999 Percent 0 60 - 40- H ---- -5 - 2 - 20 El E Sub-Saharan East Asia South Asia Latin America Europe and Middle East Africa and Pacific and the Central Asia and North Caribbean Africa 245 GLOBAL ECONOMIC PROSPECTS Table A3.10 Long-term net resource flows to developing countries, 1999 (millions of U. S. dollars) Private Official Official Percentage Debt flows development Total of GDP Total act PD1 Portfoo Total assistance Other All developing ctrys 264,900 4.37 219,214 -649 185,408 34,456 45,686 40,725 4,960 Asia 72,475 2.93 53,235 -28,322 $9,111 22,445 19,241 13,796 5,444 East Asia and Pacific 65,318 3.45 51,062 -26,112 56,041 21,133 14,257 9,533 4,724 China 42,670 4.31 40,632 -1,854 38,753 3,732 2,038 1,798 240 Indonesia -4,928 -3.49 -8,416 -6,944 -2,74S 1,273 3,487 1,757 1,731 Korea, Rep. 9,758 2.40 6,409 -15,350 9,333 12,426 3,349 1,514 1,835 Malaysia 3,616 4.58 3,247 1,173 1,553 522 369 -111 480 Myanmar 245 .. 203 -14 216 0 42 43 -1 Papua New Guinea 568 15.85 499 -30 297 232 69 101 -32 Philippines 4,955 6.46 4,915 3,920 573 422 40 887 -847 Thailand 4,700 3.85 2,471 -6,269 6,213 2,527 2,229 893 1,336 Vietnam 1,924 6.71 828 -781 1,609 0 1,096 1,086 11 South Asia 7,157 1.23 2,173 -2,209 3,070 1,312 4,984 4,264 720 Bangladesh 1,247 2.71 198 15 179 4 1,049 1,041 8 India 3,351 0.75 1,813 -1,658 2,169 1,302 1,538 1,367 171 Nepal 227 4.54 -8 -13 4 0 235 237 -2 Pakistan 1,145 1.97 53 -478 530 0 1,092 541 551 Sri Lanka 366 2.33 109 -74 177 6 258 259 -1 Latin America and the Caribbean 116,526 6.41 111,302 17,058 90,352 3,893 5,224 3,335 1,889 Argentina 33,041 11.65 32,296 7,963 23,929 404 744 -230 974 Bolivia 1,354 16.27 1,017 0 1,016 O 338 353 -15 Brazil 23,515 4.44 22,793 -11,828 32,659 1,961 722 -18 741 Chile 11,782 17.41 11,851 2,612 9,221 18 -69 7 -75 Colombia 4,708 5.43 3,635 2,502 1,109 25 1,073 152 920 Costa Rica 822 5.28 924 255 669 0 -102 -26 -76 Dominican Rep. 1,542 8.86 1,404 66 1,338 0 138 61 77 Ecuador 1,167 6.15 944 254 690 0 223 135 89 El Salvador 508 4.07 360 129 231 0 148 77 71 Guatemala 455 2.50 98 -57 155 0 357 199 157 Jamaica 345 5.17 425 -99 524 0 -80 -49 -31 Mexico 25,106 5.23 26,781 13,865 11,786 1,129 -1,674 -53 -1,621 Nicaragua 931 42.08 382 82 300 0 549 579 -30 Panama 635 6.64 620 597 22 0 15 -6 22 Paraguay 206 2.67 109 38 72 0 97 25 72 Peru 3,858 7.43 3,140 883 1,969 289 718 184 534 Trinidad and Tobago 652 9.49 713 80 633 0 -61 5 -66 Uruguay 326 1.56 66 -163 229 0 260 -1 261 Venezuela 3,063 2.96 3,130 -124 3,187 67 -67 3 -70 Europe and Central Asia 52,483 5.97 43,164 13,080 26,534 3,550 9,319 8,311 1,008 Armenia I 237 12.87 122 0 122 0 115 119 -4 Azerbaijan 783 17.36 396 86 $10 0 187 184 3 Belarus 367 1.37 394 169 225 0 -27 4 -31 Bulgaria 1,457 11.75 1,112 204 806 102 346 187 159 Czech Republic 4,936 9.31 4,837 -756 5,093 500 99 119 -20 Estonia 631 12.31 569 72 305 191 62 46 16 Georgia 198 7.28 86 4 82 0 112 148 -36 Hungary I 5,169 10.67 4,961 2,418 1,950 592 208 86 122 Kazakhstan 1,729 10.25 1,477 -110 1,587 0 252 112 140 Kyrgyz Republic 211 16.90 -16 -52 36 0 227 194 33 Latvia 650 9.75 303 -45 348 0 347 65 282 246 GLOBAL ECONOMIC INDICATORS Table A3.10 Long-term net resource flows to developing countries, 1999 (continued) (millions of U. S. dollars) Private Official Official Percentage Debt flows development Total of GDP Total net PI Porfolio Total assistance Other Europe and Central Asia (co ttined) Lithuania 1,320 12.38 1,148 661 487 0 172 70 102 Moldova 114 9.73 11 -22 34 0 103 55 48 Poland 10,499 6.77 10,452 2,461 7,270 721 47 294 -246 Romania 920 2.61 714 -327 1,041 0 206 64 142 Russian Federation 5,058 2.62 3,780 -173 3,309 644 1,278 1,040 238 Slovak Rep. 575 2.92 281 -73 354 0 294 71 223 Tajikistan 84 7.77 10 -14 24 0 74 74 0 Turkmenistan 50 1.51 -54 -134 80 0 104 94 10 Turkey 8,127 4.40 8,667 7,084 783 800 -540 -131 -409 Ukraine 701 2.28 371 -125 496 0 330 47 283 Uzbekistan 1,052 6.16 658 545 113 0 395 294 101 Middle East and North Africa 2,456 0.42 1,064 -1,066 1,461 669 1,392 3,418 -2,026 Algeria -1,797 -3.77 -1,486 -1,496 7 3 -311 71 -382 Egypt, Arab Rep. 1,814 2.03 1,558 -57 1,065 550 255 460 -205 Iran, Islamic Rep. -2,605 -2.62 -1,385 -1,470 85 0 -1,220 24 -1,244 Jordan 501 6.21 112 -57 158 11 389 366 23 Morocco -26 -0.08 -118 -212 3 91 91 302 -211 Oman -390 -2.61 -413 -484 60 11 23 31 -9 Syrian Arab Rep. 55 2.12 87 -4 91 0 -32 7 -39 Tunisia 968 4.62 739 389 350 0 228 165 63 Yemen, Rep. 155 2.28 -150 0 -150 0 305 323 -17 Sub-Saharan Africa 20,960 6.55 10,449 -1,399 7,949 3,899 10,511 11,865 -1,354 Angola 2,690 43.43 2,373 -98 Z,471 0 317 323 -6 Botswana 14 0.28 36 -1 37 0 -22 9 -31 Cameroon 206 2.24 -13 -53 40 0 218 313 -95 C6te d'Ivoire 147 1.31 74 -283 350 8 73 291 -219 Ethiopia 579 8.97 78 -12 90 0 501 505 -4 Gabon 41 0.95 209 9 200 0 -167 14 -181 Ghana 416 5.36 -16 -52 17 19 432 473 -41 Kenya 13 0.12 -51 -70 14 5 64 227 -163 Madagascar 299 8.04 52 -6 S8 0 247 267 -19 Nigeria 633 1.81 860 -146 1,005 2 -227 126 -353 Senegal 379 7.97 54 -6 60 0 325 353 -29 South Africa 4,778 3.67 4,533 -698 1,376 3,855 245 244 1 Sudan 581 5.98 371 0 371 0 211 216 -6 Zambia 466 15.01 151 -12 163 0 315 400 -85 Zimbabwe 287 5.12 70 7 59 4 217 274 -57 Source: World Bank data. Figure A3.10 Distribution of long-term net resource flows, 1999 Percent 100 80s - - - _ _ - . -__ - - 80 - 40 Sub-Saharan East Asia South Asia Latin America Europe and Middle East Africa and Pacific and the Central Asia and North Caribbean Africa 247 Technical Notes The principal sources for the data in this ap- of Payments database is the principal source pendix are the World Bank's central databases. for data through 1999; in some cases these data The cut-off date for data updates was August have been supplemented by UNCTAD and UN 31, 2001; revisions/releases since that time have Comtrade databases or by World Bank staff es- not been incorporated. timates. Trade figures for countries of the for- Regional aggregates are based on the clas- mer Soviet Union reflect the total of non-CIS sification of economies by income group and and intra-CIS exports and imports. region, following the Bank's standard defini- Tables A3.7 and A3.8. Growth rates are tions (see country classification tables that fol- compound averages and are computed for cur- low). Debt and finance data refer to the 137 rent dollar measures of trade. countries that report to the Bank's Debtor Re- Table A3.9. Long-term debt covers public porting System (see the World Bank's Global and publicly guaranteed external debt but ex- Development Finance 2001). Small economies cludes IMF credits. Concessional debt is debt have generally been omitted from the tables with an original grant element of 25 percent but are included in the regional totals. or more. Nonconcessional variable interest Current price data are reported in U.S. rate debt includes all public and publicly guar- dollars. The cut-off date for data updates was anteed long-term debt with an original grant August 31, 2001; revisions/releases since that element of less than 25 percent whose terms time have not been incorporated. depend on movements of a key market rate. This item conveys information about the bor- Notes on tables rower's exposure to changes in international Tables A3.1 through A3.4. Projections are interest rates. For complete definitions, see consistent with those highlighted in Chapter 1 Global Development Finance 2001. and Appendix 1. Table A3.1 0. Long-term net resource flows Tables A3.S and A3.6. Merchandise ex- are the sum of net resource flows on long-term ports and imports exclude trade in services. Im- debt (excluding IMF) plus non-debt-creating ports are reported on a c.i.f. basis. Growth flows. Foreign direct investment refers to the net rates are based on constant price data, which inflows of investment from abroad. Portfolio are derived from current values deflated by rel- equity flows are the sum of country funds, de- evant price indexes. Effective market growth is pository receipts, and direct purchases of shares the export-weighted import growth rate of the by foreign investors. For complete definition, country's trading partners. The IMF's Balance see Global Development Finance 2001. 248 Classification of Economies GLOBAL ECONOMIC PROSPECTS Table 1 Classification of economies by income and region, July 2001 Europe and Middle East Sub-Saharan Africa 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 Central Asia Europe East Africa Ameencas Low- Angola Benin Cambodia Afghanistan Armsenia Yemen, Rep Haiti income Burundi Burkina Faso Indonesia Bangladesh Azerbaijan Nicaragua Comoros Cameroon Korea, Dem. Bhutan Georgia Congo, Dem. Central African Rep. India Kyrgyz Rep. Republic Lao PDR Nepal Republic Eritrea Chad Mongolia Pakistan Moldova Ethiopia Congo, Rep. Myanmar Tajikistan Kenya Cte d'Ivoire Solomon Ukraine Lesotho Gambia, The Islands Uzbekistan Madagascar Ghana Vietnam Malawi Guinea Mozambique Guinea- Rwanda Bissaut Somalia Liberia Sudan Mali Tanzania Mauritania Uganda Niger Zambia Nigeria Zimbabwe S e Tom6 and Principe Senegal Sierra Leone Togo Middle- Lower Namibia Cape Verde China Maldives Albania Iran, Islamic Algeria Belize income Swaziland Equatorial Fiji Sri Lanka Belarus Rep. Djibouti Bolivia Guinea Kiribati Bosnia and Iraq Egypt, Arab Colombia Marshall Herzegovina Jordan Rep. Cuba Islands Bulgaria Syrian Arab Morocco Dominican Micronesia, Kazakhstan Republic Tunisia Republic Fed. Sts. Latvia West Bank Ecuador Papua New Lithuania and Gaza El Salvador Guinea Macedonia, Guatemala Philippines FYR Guyana Samoa Romania Honduras Thailand Russian Jamaica Tonga Federation Paraguay Vanuatu Turkmenistan Peru Yugoslavia, St. Vincent Fed. Rep. and the Grenadines Suriname Upper Botswana Gabon American Croatia Isle of Man Bahrain Libya Antigua and Mauritius Samoa Czech Turkey Lebanon Barbuda Mayone Korea, Rep. Republic Oman Argentina Seychelles Malaysia Estonia Saudi Arabia Brazil South Africa Palau Hungary Chile Poland Costa Rica Slovak Dominica Republic Grenada Mexico Panama Puerto Rico St. Kitts and Nevis St. Lucia Trinidad and Tobago Uruguay Venezuela, RB Subtotal 155 25 23 23 8 26 2 10 6 32 250 REGIONAL ECONOMIC PROSPECTS Table 1 Classification of economies by income and region, July 2001 (continued) Europe and Middle East Sub"aharan Aftica Asia Central Asia and North Africa East and Easter Income Southern West East Asia South Europe and Rest of Middle North group Subgroup Africa Africa and Pacific Asia Ceotral Asia Europe East Africa Americas High- OECD Australia Austria Canada income Japan Belgium United New Zealand Denmark States Finland Franeb Germany Greece Iceland Ireland Italy Luxembourg Neterlands Norway Portugal Spain Sweden Switzerland United Kingdom Non-OECD Brunei Slovenia Andorra Israel Malta Aruba French Chiannel Kuwait Bahamas, Polynesia islands Qatar The Guam Cyprus United Arab Barbados Hong Kong, Faeroe Emirates Bermuda Chinac Islands Cayman Macao, Greenland Islands Chinad Liechtenstein Netherlands New Monaco Antilles Caledonia San Marino Virgin N. Mariana Islands Islands (U.S.) Singapore Taiwan, China Total 208 25 23 35 8 27 28 14 7 41 a. Former Yugoslav Republic of Macedonia. b. The French overseas departments French Guiana, Guadeloupe, Martinique, and Reunion are included in France. c. On 1 July 1997 China resumed its exercise of sovereignty over Hong Kong. d. 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- income (GNI) 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 2000 GNI 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. 251 GLOBAL ECONOMIC PROSPECTS Table 2 Classification of economies by income and indebtedness, July 2001 Income Sub- Not classified group group Severely indebted Moderately indebted Lea indebted by indebtedness Low- Afghanistan Mali Armenia Azerbaijan Uzbekistan income Angola Mauritania Bangladesh Bhutan Bein Myanmar Burkina Faso Eritrea Burundi Nicaragua Cambodia India Cameroon Niger Chad Korea, Dem. Rep. Central Nigeria Gambia, The Lesotho African Pakistan Georgia Nepal Republic Rwanda Ghana Solomon Islands Comoros Sao Tom6 Haiti Tajikistan Congo, Dem. and Principe Kenya Ukraine Rep. Sierra Leone Moldova Congo, Rep. Somalia Mongolia C6te d'Ivaire Sudan Mozambique Ethiopia Tanzania Senegal Guinea Uganda Togo Guiinca-Bissau Zambia Turkmenistan Indonesia Vietnam Kyrgyz Republic Yemen, Rep. Lao PDR Zimbabwe Liberia Madagascar Malawi Middle- Lower Bolivia Algeria Albania Lithuania Marshall Islands income Bosnia and Belize Belarus Macedonia, Micronesia, Fed. Sts. of Herzegovina Colombia Cape Verde FYRI West Bank and Gaza Bulgaria Ecuador China Maldives Cuba Honduras Costa Rica Namibia Guyana Jamaica Djibouti Paraguay Iraq Morocco Dominican Romania Jordan Papua New Republic Sri Lanka Peru Guinea Egypt, Arab Suriname Syrian Arab Philippines Rep. Swaziland Republic Russian El Salvador Tonga Federation Equatorial Vanuatu Samoa Guinea Yugoslavia, St. Vincent and Fiji Fed. Rep. the Grenadines Guatemala Thailand Iran, Islamic Tunisia Rep. Turkey Kazakhstan Kiribati Latvia Upper Argentina Chile Antigua and Oman American Samoa Brazil Estonia Barbuda Poland Isle of Man Gabon Hungary Bahrain Saudi Arabia Mayotte Lebanon Botswana Seychelles Palau Malaysia Croatia Slovak Puerto Rico Mauritius Czech Republic Panama Republic South Africa Uruguay Dominica St. Kitts and Venezuela, RB Grenada Nuvis Korea, Rep. St. Lucia Libya Trinidad and Mexico Tobago 252 REGIONAL ECONOMIC PROSPECTS Table 2 Classification of economies by income and indebtedness, July 2001 (continued) Income Sub- Not classified group group Severely indebted Moderately indebted Less indebted by indebtedness High- OECD Australia Japan income Austria Luxembourg Belgium Netherlands Canada New Zealand Denmark Norway Finland Portugal Franceb Spain Germany Sweden Greece Switzerland Iceland United Ireland Kingdom Italy United States Non- Andorra Kuwait OECD Aruba Liechtenstein Bahamas, Macao, The Chinac Barbados Malta Bermuda Monaco Brunei Netherlands Cayman Antilles Islands New Caledonia Channel N. Mariana Islands Islands Cyprus Qatar Faeroe San Marino Islands Singapore French Slovenia Polynesia Taiwan, China Greenland United Arab Guam Emirates Hong Kong, Virgin Chinad Islands (U.S.) Israel Total 208 46 43 57 62 a. Former Yugoslav Republic of Macedonia. b. The French overseas departments French Guiana, Guadeloupe, Martinique, and R6union are included in France. c. On 20 December 1999 China resumed its exercise of sovereignty over Macao. d. 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, 2000 GNI 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 1997-99) are above critical levels: debt to $2,996-9,265; and high income, $9,266 or more. GNI (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 GNI 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. 253 Nicholas Stern, Senior Vice President and Chief Economist The global economy is slipping precariously toward recession. In 2001 the three biggest economies of tle world-the United States,Japan, and the European Union-entered the first simultaneous downturn since 1982, and were then hit with the added shock of the terrorist attacks in the United States.While the most probable scenario is for a recovery beginning in 2002, today's slow growth of global trade and weakening of financial flows to all but the most creditworthy countries have shackled growth in developing countries. Increasing trade would help developing countries over the medium term.To accelerate the global integtation that has been a hallmark of rapidly growing developing economies for the last two decades, the international community must work together to expand market access-and must help developing countries respond to new trading opportunities. This report presents a four-part policy agenda to do just that. This twelfth annual edition of Global Economic Prospects a analyzes prospects for the global economy and the implications for developing countries and poverty reduction g reviews policy issues in agriculture and textile trade, services, international transport, and intellectual policy rights a presents a four-part policy agenda: launching of a development round of WTO talks, new global cooperation to expand trade outside the WTO, new policies of high-income countries to expand trade, and enacting trade reforms in developing countries * examines trade in services and finds that reduction in barriers to entry would provide large income gains M presents findings that suggest developing countries could increase their incomes by a cumulative $1.5 trillion over 2005-15 if all countries progressively enact the proposed trade reforms and, as a consequence, lift an additional 300 million people out of poverty by 2015. Global Economic Prospects 2002 provides essential information for those concerned with development shaping today's global economy. 1818 H StreMt. NW. Washington, D.C. 20433, U.S.A. Telephone: 202 417 1234 Facsimile: 202 477 6391 Internet: www.worldhank.org E-mail: feedback@worldhank.org For more information see: wvw.worldbank.org/prospecrs 14 9 96 9 780821 349960 ISBN-0-8213- 996-1