90290 Market Access for Developing Countries’ Exports Prepared by the Staffs of the IMF and the World Bank April 27, 2001 Table of Contents Page List of Abbreviations ............................................................................................................. 3 Executive Summary............................................................................................................... 4 I. Introduction ....................................................................................................................... 8 II. Changes in the Pattern of World Trade.............................................................................. 9 III. Protection in Merchandise Trade ................................................................................... 14 A. Tariffs ................................................................................................................. 15 B. Agricultural Subsidies and Nontariff Measures (NTMs) ....................................... 25 C. Trade Preferences................................................................................................ 31 D. Conclusions......................................................................................................... 34 IV. Trade In Services .......................................................................................................... 37 A. Barriers to Trade in Services ............................................................................... 39 B. Measuring Barriers to Trade in Services .............................................................. 39 C. Conclusions......................................................................................................... 43 V. The Gains From Liberalization........................................................................................ 43 A. Gains from Liberalization in Merchandise Trade .................................................. 43 B. Gains from Liberalization in Services................................................................... 47 C. Conclusions......................................................................................................... 48 Figures 1. Growth of Merchandise Exports, 1970–2000.............................................................. 10 2. Developing Countries: Share of Exports Going to Other Developing Countries, 1965–98 ................................................................................................... 10 3. World: Product Composition of Merchandise Exports, 1965–98 ................................. 10 4. Developing Countries: Composition of Merchandise Exports, 1965–98....................... 12 5 Share of Commercial Services in Total Exports of Goods and Services, 1980–97........ 12 6. Sub-Saharan Africa: Composition of Merchandise Exports, 1965–95 .......................... 13 7. Commodity Composition and Export Growth, 1980–99.............................................. 14 8. Simple Average MFN Tariffs, 2000 ............................................................................ 19 9. Tariff Escalation: Average Applied Tariffs by Stage of Production.............................. 24 -2- 10. Initiations of Antidumping Investigations, 1987–99..................................................... 29 11. Relationship Between Income Level and GATS Commitments.................................... 41 12. Structure of WTO Members’ Commitments by Sector, June 2000............................... 41 Text Tables 1. Uruguay Round Tariff Concessions............................................................................. 16 2. Post-Uruguay Round Average Tariff Rates ................................................................. 18 3. Quad Countries: Imports and Tariff Peaks, 1999......................................................... 21 4. Select Countries: Import Tariff Maximum and Tariff Peaks......................................... 22 5. Industrial Country Imports from Developing Countries: Changes in Tariff Escalation..................................................................................................... 23 6. Trade-Weighted Average Tariffs................................................................................. 25 7. Initiations of Antidumping Investigations, 1995–99..................................................... 29 8. Tariff Peaks and Preferential Duty Rates in the Quad, 1999......................................... 33 9. Welfare Gains from Post-Uruguay Round Trade Liberalization ................................... 45 10. Welfare Effects of Trade Liberalization on Developing Countries................................ 46 Text Boxes 1. Indicators of Tariff Protection..................................................................................... 15 2. Agriculture: Tariffication and Tariff Rate Quotas (TRQs) ........................................... 17 3. Standards in International Trade ................................................................................. 30 4. The Cost of Customs and Border Procedures.............................................................. 32 5. Recent Market Access Initiatives for LDCs................................................................. 35 6. What’s Different About Services? ............................................................................... 38 7. Barriers to Trade in Services....................................................................................... 40 Appendix Tables 1. Shares of Manufactures in Total Merchandise Exports................................................ 50 2. Post-Uruguay Round Bound Tariffs on Agricultural and Industrial Products............... 51 Appendices II. The Implementation of the Agreement on Textiles and Clothing (ATC)....................... 52 III. The GATS Rules on Measures Affecting Services Trade............................................. 53 Bibliography........................................................................................................................ 55 -3- List of Abbreviations ACP African, Caribbean and Pacific Countries AGOA African Growth and Opportunity Act AMS Aggregate Measure of Support APEC Asian-Pacific Economic Cooperation ATC Agreement on Textiles and Clothing CAP Common Agricultural Policy CGE Computable General Equilibrium DFAT Department of Foreign Affairs and Trade EC European Community EFTA European Free Trade Association EU European Union FDI Foreign Direct Investment GATS General Agreement on Trade in Services GATT General Agreement on Tariffs and Trade GDP Gross Domestic Product GEP Global Economic Prospects Report (World Bank) GNP Gross National Product GSP Generalized System of Preferences GSTP Generalized System of Tariff Preferences GTAP Global Trade Analysis Project LDCs Least Developed Countries MFA Multifiber Arrangement MFN Most-Favored-Nation NTMs Nontariff Measures OECD Organization For Economic Cooperation and Development OPEC Organization of the Petroleum Exporting Countries QRs Quantitative Restrictions SACU South Africa Customs Union SPS Sanitary and Phytosanitary Measures SSA Sub-Sahara Africa TBT Technical Barriers to Trade TRQs Tariff Rate Quotas URAA Uruguay Round Agreement on Agriculture WTO World Trade Organization -4- EXECUTIVE SUMMARY Integration into the world economy has proven a powerful instrument for countries to promote economic growth, development, and poverty reduction. Trade has been an engine of growth for the past fifty years, owing in part to eight successive rounds of multilateral trade liberalization, as well as unilateral and regional trade liberalization. The growing integration of the world economy has raised living standards and brought increased opportunity to many parts of the globe. Many developing countries have shared in this prosperity. As a group, developing countries have become much more important in world trade, and their trade relationships have changed markedly from the traditional north-south pattern. Developing countries now account for one-third of world trade, up from about a quarter in the early 1970s, and many have substantially increased their exports of manufactures and services relative to traditional commodity exports. The share of manufactures in developing country exports has risen to 80 percent; moreover, trade between developing countries has grown rapidly, with 40 percent of their exports now going to other developing countries. However, the process of integration has been uneven. Progress has been very impressive for a number of developing countries in Asia and, to a lesser extent, in Latin America, which have become successful participants in global trade and attracted the bulk of foreign direct investment in developing countries. This includes China and India since they initiated reforms and some of the middle- and higher-income countries in Asia that were themselves poor in the 1970s. But progress has been less rapid for many other countries, particularly in Africa and the Middle East. The poorest countries have seen their share of world trade decline substantially and risk further marginalization. These poorer countries comprise about 75 developing and transition economies, including virtually all of the least developed countries and the heavily indebted poor countries. In contrast to the successful integrators, they depend disproportionately on production and exports of traditional commodities. The reasons for their marginalization are complex, including deep-seated structural problems, weak policy frameworks and institutions, and protection at home and abroad. These trends point to the need to liberalize trade further, particularly in those areas of importance to poorer developing countries. These areas include not only traditional ones as such as textiles, clothing, and agriculture, but also manufacturing and services, involving access to both industrial and developing country markets. Although border protection, including tariff and nontariff measures, has declined substantially over the past three decades, it remains significant in both industrial and developing countries, particularly in areas such as agriculture and labor-intensive industrial products where developing countries have comparative advantage. • Industrial countries maintain high protection in agriculture through an array of very high tariffs, tariff peaks and escalation, and restrictive tariff quotas. Average tariff -5- protection in agriculture is about nine times higher than in manufacturing. In addition, agricultural subsidies in industrial countries undermine developing countries’ agricultural sectors and exports by depressing world prices and pre-empting markets. • In industrial countries, border protection in manufacturing is generally low, but it remains high on labor-intensive products of interest to developing countries. For example, applied tariffs on textiles and clothing are three times the average in manufacturing, and the bulk of restrictive quotas under the Multifiber Arrangement will not be phased out until 2005. Other labor-intensive manufactures are also disproportionately affected by tariff peaks (tariffs above 15 percent) and escalation (tariffs that increase with the level of processing) which inhibit the diversification of exports toward higher value-added products. • Many developing countries themselves have high tariffs. On average, their applied tariffs on industrial products are three to four times as high as those of industrial countries, and they exhibit the same characteristics of tariff peaks and escalation. Tariffs on agriculture are even higher (60 percent for bound rates and 18 percent for applied rates) than those on industrial products. • Nontraditional impediments to trade are harder to quantify and assess, but they are becoming more significant as border protection declines. Antidumping activity is on the rise in both industrial and developing countries; the need to conform to technical, sanitary, and phytosanitary standards imposes costs on exporters that can exceed benefits to consumers; and complex and inefficient customs procedures can deter trade. For a variety of reasons, preferential access schemes for poorer countries have not proven very effective in increasing market access for these countries. Such schemes often exclude or provide less generous benefits for highly protected products of most interest to the poorest; and they are often complex, nontransparent, and subject to various exemptions and conditions (including noneconomic ones) that unilaterally limit benefits or terminate them once significant market access is achieved. Compared with trade in goods, the liberalization of trade in services is at a very early stage, and substantial barriers to trade in services exist in both industrial and developing countries. Market access in services is affected by a wider array of government decisions, regulations, and controls than those applied at the border (such as ownership restrictions, restrictions on the number of suppliers, or professional qualifications). • In general, industrial countries have opened up their services sectors more than developing countries; however, barriers in industrial countries are most restrictive for services that require the temporary movement of service providers across borders, an area of interest to developing countries. Improved access in this area -6- would permit developing countries to export a significant labor component of services in areas such as construction, transportation, distribution, health, and software. • Developing countries themselves would gain substantially from liberalizing key “backbone” services—transport, financial services, and telecommunications—within a regulatory framework that fosters competition; inefficiencies in these sectors often add more to export costs than foreign trade barriers. The above review points to priority areas for further trade liberalization—in both industrial and developing countries—if trade is to realize its potential for economic growth and development. Greater efforts by industrial countries, and the international community more broadly, are called for to remove the trade barriers facing developing countries, particularly the poorest countries. Speedier liberalization of agriculture and textiles and clothing are prime candidates, along with the elimination of tariff peaks and escalation in agriculture and manufacturing. In turn, developing countries would benefit from sustained efforts to improve their own economic policies and further reduce their trade barriers. This would both strengthen their own economies and enhance market access for other developing countries. • The potential gains from eliminating remaining trade barriers are considerable. Estimates of the static welfare gains from eliminating all barriers to merchandise trade range from US$250 billion to US$550 billion per annum. About one-third of these gains would accrue to developing countries, more than twice the annual flow of aid to these countries. While the modeling of trade in services is in its infancy, gains from liberalization of similar magnitudes are considered possible. • Because their economies are more highly protected, and because they face higher barriers, developing countries gain more from global trade liberalization as a percentage of their GDP than industrial countries. However, industrial countries command a larger share of the total gain because of the size of their economies. If only static gains are considered, agricultural liberalization generates the largest welfare gains, because agriculture is most highly protected in both industrial and developing countries. Once dynamic effects are considered, liberalization of manufacturing generates the largest gains in both industrial and developing countries. • The main benefits of global trade liberalization for industrial countries as a group come from the liberalization of their own agricultural markets. Developing countries as a group also gain most from removing their own barriers to trade, with the gains derived about equally from liberalization of manufacturing and agriculture. In each other’s markets, both groups benefit most from liberalization of manufactures, because of the bulk of trade between industrial and developing countries is in manufactured products. The group of low-income countries gains most from agricultural liberalization in industrial countries because of the greater relative -7- importance of agriculture in their economies. They also gain substantially from liberalization of both agriculture and manufactures in developing country markets. Special efforts to enhance market access for the poorest developing countries would enable them to harness the benefits of trade for development and poverty reduction. In particular, improving preferential access for the poorest countries by providing duty- and quota-free access to world markets would provide significant benefits to these countries at little cost to the rest of the world. The recent market-opening initiatives by the EU and some other countries are important and welcome steps in this regard. To be effective, such access should be made permanent, extended to all goods, and accompanied by simple, transparent rules of origin. This would give the poorest countries the confidence to persist with difficult domestic reforms and ensure effective use of debt relief and aid flows. -8- I. INTRODUCTION 1. Integration into the world economy is a powerful instrument for economic growth and poverty reduction. Trade liberalization and export development have therefore been prominent elements of reform in industrial and developing countries. Average tariffs across all countries are significantly lower today than in the late 1970s, and the incidence of quantitative restrictions (QRs) and other nontariff measures (NTMs) has been greatly reduced. Improvement in economic incentives as a result of lower trade barriers has helped many countries achieve higher growth, and has paved the way for greater participation in the rules- based multilateral trading system. 2. The Uruguay Round of trade negotiations made a significant contribution to the world trading system on two fronts: improving market access (reduction of tariff and NTMs) and bringing international trade rules to sectors previously excluded or subject to weak rules (particularly agriculture, textiles and clothing, and services). In market access, the tariff cuts covered a greater percent of world trade than previous rounds and it is estimated that they will lower import costs by US$50 billion per year (Finger and Schuknecht, 2001).1 3. Despite these achievements, significant barriers to market access remain in world trade. In agriculture, for example, the rules for tariffication of NTMs under the Uruguay Round allowed for significant increases in tariffs, so they remain high even after the agreed reductions have been implemented, and only limited progress has been made in reducing trade-distorting export subsidies. In manufacturing, the rules for the elimination of the Multifiber Arrangement (MFA) under the Uruguay Round allow for a postponement of most of the liberalization in textiles and clothing until 2005. In both agriculture and manufacturing, tariff peaks and escalation persist and affect disproportionately the exports of interest to developing countries. In addition, some less transparent barriers are becoming more prevalent, especially antidumping and requirements related to technical and health standards. The growing number of regional trading arrangements and preference schemes has also led to increasing discrimination in international trade, complex rules of origin, and other administrative procedures that can hamper trade. 4. This paper examines the landscape of remaining protection in world trade, with particular emphasis on the trade barriers that developing country exports face in both industrial and developing country markets. Section II begins by examining some key changes in patterns of trade over the past thirty years. Section III analyzes the patterns of protection in merchandise trade and discusses the incidence of tariff and nontariff protection on exports from industrial, developing, and the least developed countries (LDCs). It also reviews the effects of preferential schemes for poorer country exports into industrial country markets. Section IV summarizes the state of knowledge of barriers to trade in services. 1 On the basis of tariff cuts applied to 1997 world imports. -9- Section V reviews estimates of the costs of protection and the gains that would accrue to different regions and groups of countries from further trade liberalization. II. CHANGES IN THE PATTERN OF WORLD TRADE 5. One of the most important developments in world trade over the past thirty years has been the growing participation of developing countries.2 While the value of global merchandise trade expanded on average by 10 percent per year from 1970 to 1999, outpacing the growth of output, developing countries’ exports grew at 12 percent per year and their share of total trade expanded from about one-quarter to one-third (Figure 1). Emerging market countries in Asia and, to a lesser extent, Latin America were the main contributors to this impressive performance. This includes China and India since they initiated reforms and some of the middle- and higher-income countries in Asia that were themselves poor in the 1970s. These countries were also the main recipients of the dramatic increase in foreign direct investment in developing countries that occurred in the late 1980s and 1990s as advances in transportation and communication facilitated the geographic dispersion of production processes and a rapid expansion of intra-industry trade (Deardorff, 2001). 6. This increase in trade was not limited to providing exports for industrial country markets. Over the period, developing countries became an important market for their own exports as well. The share of developing country exports to other developing countries increased from 17 percent in the mid-1960s to around 40 percent in 1995-98 (Figure 2). 7. A second important change has been the dramatic increase in manufactures in world merchandise trade (Figure 3). From 1965 to 1980, the share of manufactures in total merchandise exports oscillated around 60 percent. Since 1980, however, it has substantially increased, to reach 85 percent in 1998. In contrast, the share of agricultural products declined almost continually over the period, and the share of mineral products, which rose from 1965 to 1981 reflecting changes in oil prices, has declined since then. 2 The definition of developing countries in this paper is broadly consistent with those countries likely to elect for developing country treatment in the World Trade Organization (WTO). The group therefore includes Hong Kong SAR, the Republic of Korea, and Singapore, which were clearly developing economies in 1965, but are now classified by the World Bank as high- income. - 10 - Figure 1. Growth of Merchandise Exports, 1970-20001 2600 2100 Other Developing Countries index (1980=100) 1600 World 1100 600 Sub-Saharan Africa Least Developed Countries 100 1970 1975 1980 1985 1990 1995 2000 Source: IMF World Economic Outlook (WEO). 1 Excluding oil exports. Figure 2. Developing Countries: Share of Exports Going to Other Developing Countries, 1965-98 45 Minerals 40 Agriculture Manufactures 35 Total 30 percent 25 20 15 10 5 0 1965 1970 1975 1980 1985 1990 1995 1998 Source: Global Trade Analysis Project (GTAP) database, version 5. Figure 3. World: Product Composition of Merchandise Exports, 1965-98 100 90 Minerals 80 70 Agriculture 60 percent 50 40 Manufactures 30 20 10 0 1965 1970 1975 1980 1985 1990 1995 1998 Source: GTAP database, version 5. - 11 - 8. Developing countries also led in this change, transforming the composition of their exports from commodities to manufactures much more rapidly than the world as a whole (Figure 4). The share of manufacturing in developing country exports grew dramatically, from roughly 15 percent in 1965 to over 80 percent by 1998, while that of agricultural commodities fell sharply. The share of minerals has fluctuated, with a rise in the 1970s associated with the OPEC increase in oil prices, followed by a decline after the early 1980s, when oil prices declined. 9. Added to this change in the pattern of world trade has been the increase in trade in commercial services (Figure 5). 3 Developing countries also participated in this change, with the share of services in their total exports of goods and services rising rapidly from less than 10 percent in the early 1980s to 17 percent in 1998. 10. Not all developing countries have participated in these trends. From 1970 to 1999, the LDCs experienced much lower trade growth (5.7 percent) than the world average (10 percent), and saw their share of global trade decline from 1.9 percent to 0.5 percent (Figure 1). Sub-Saharan African (SSA) countries have performed slightly better, with trade growth of 7.9 percent over the period, but still with a decline in their share of world trade from 3.7 percent to 1.4 percent. 11. Many of these countries continue to be dependent on traditional commodity exports. An examination of shares of manufactures in developing country exports by individual countries and groups finds countries and regions where the share of manufactures remains low (Appendix Table 1). Some of these countries are successful reform-oriented ones such as Chile and Uganda that specialize in natural resource-based products. But a large number of countries in SSA, the Andean region, and the Middle East and North Africa have been less successful in integrating into the world economy. The reasons for this marginalization are complex, including deep-seated structural problems, weak policy frameworks and institutions, and protection at home and abroad. 3 The services exports represented here are somewhat narrower than the concept of trade in services used in the WTO’s General Agreement on Trade in Services (GATS). Karsenty (2000) estimates that this category of services now accounts for roughly 60 percent of the total exports of services covered by the GATS (see Section IV). - 12 - Figure 4. Developing Countries: Composition of Merchandise Exports, 1965-98 100 90 80 Minerals 70 60 percent 50 40 Agriculture 30 20 10 Manufactures 0 1965 1970 1975 1980 1985 1990 1995 1998 Source: GTAP database, version 5. Figure 5. Share of Commerical Services in Total Exports of Goods and Services, 1980-97 25 20 Industrial Countries percent 15 Developing Countries Sub-Saharan Africa 10 5 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Source: World Bank; World Development Indicators (2001). - 13 - 12. For regions such as SSA, there is concern about continuing dependence on commodity exports. An examination of changes in the composition of exports from SSA shows that, even for these countries, there has been a consistent but less dramatic upward trend in the share of manufactures exports (Figure 6). Nevertheless, three-fourths of SSA’s exports are still concentrated in primary commodities. While this explains part of the decline in SSA’s share of world trade, more than a third of the decline results from the loss of market shares in the goods that SSA produces and exports, rather than from the relatively slow growth of those commodity exports themselves (Figure 7). Figure 6. Sub-Saharan Africa: Composition of Merchandise Exports, 1965-95 100 90 80 70 Minerals 60 percent 50 40 30 Agriculture 20 10 Manufactures 0 1965 1970 1975 1980 1985 1990 1995 Source: GTAP database, version 5. - 14 - Figure 7. Commodity Composition and Export Growth, 1980-991 400 World 350 300 250 World Exports Weighted by index (1980=100) the Commodity Composition 200 of 1980 SSA Exports 150 Sub-Saharan Africa (SSA) 100 50 0 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Source: IMF staff estimates. 1 Excluding oil exports. 13. These changes in trade patterns have important implications for trade policies. The shift in trade from primary commodities to manufactures clearly focuses attention on the need to continue to lower protective barriers affecting manufactures in the importing countries. In addition, it highlights the importance of reducing trade barriers—on both imports and exports—in exporting countries, since protection is a tax on exports. The increase in the importance of services trade similarly focuses attention on the barriers facing developing country exports of services, and the regulatory and infrastructure barriers inhibiting the successful supply of exportable services. Finally, the marginalization of a group of low-income countries from the growth of world trade motivates the search for policies that would enhance market access for these countries and support their own efforts to strengthen their policy frameworks. III. PROTECTION IN MERCHANDISE TRADE 14. This section reviews the patterns of protection following the outcome of the Uruguay Round Agreement to present a picture of where protection remains the highest. It first reviews tariff protection in manufactures and agriculture, including the existence of tariff peaks and tariff escalation, with an emphasis on how these features of the - 15 - tariff regime affect the exports of the developing countries. It then reviews the remaining patterns of protection in NTMs, including traditional NTMs and contingent trade measures, such as antidumping. The costs associated with product standards and complicated customs procedures are also considered. A. Tariffs Overall structure of post-Uruguay Round tariffs 15. As a result of the Uruguay Round and previous trade negotiations, a large proportion of the tariffs of WTO members were bound (Box 1).4 In terms of 1989 imports, 97 percent of all merchandise trade is covered by bound tariffs, including all agricultural trade (Table 1). About 91 percent of manufactured imports is covered by bound tariffs, but only 80 percent of imports in textiles and clothing. In terms of tariff lines, a significant number of lines in manufacturing remain unbound. For most industrial countries, the share of tariff lines bound is above 97 percent, but among developing countries, there are still many, particularly in Asia and Africa, where the scope of bindings is much more limited (WTO, 2001). Box 1. Indicators of Tariff Protection There are a number of different indicators of tariff protection. Unless indicated otherwise, the analysis in this section is based on most-favored-nation (MFN) tariffs, which exclude trade preferences. In addition, a distinction is made between bound and applied tariff rates. Bound rates represent a country’s commitments in the WTO and thus are an upper bound on the level of protection. In many cases, countries have applied rates that are below bound rates, and in some cases (particularly in textiles and clothing in developing countries) they have not bound their tariffs. To the extent that applied and bound rates differ, or rates are not bound, this imparts a greater degree of uncertainty regarding market access. Thus, although applied rates may be a better indicator of the actual level of trade restrictions, bound rates are a better indication of a country’s market access commitments. Another important distinction is the way in which tariff rates are aggregated across countries and tariff lines to present summary measures. The simple (unweighted) average tariff is used as an indicator of the average level of restrictiveness, unless otherwise indicated. Trade-weighted average tariffs are used to measure the incidence of tariffs on trade from different regions, and for different product groups, since they measure the barriers faced by trade flows that actually take place. The trade-weighted average understates the degree of restrictiveness, since very high tariffs receive very small (or zero) weights, thus biasing this measure downward. 4 Binding a tariff in the WTO establishes the maximum tariff which can be applied on imports from other WTO members. This binding reduces uncertainty in the tariff regime, even in cases where the applied tariff is below the bound tariff and there is room for tariff increases. - 16 - Table 1. Uruguay Round Tariff Concessions Bindings (in percent of Average Tariff Rate 1989 imports) Bound Applied All merchandise 97 7 4 Industrial goods 91 6 4 Textiles and clothing 80 12 10 1 Agricultural products 100 32 25 Source: Finger, Ingco, and Reincke (1996). 1 Excludes processed agricultural products. 16. Full implementation of the Uruguay Round is estimated to result in a relatively low bound simple average tariff of 7 percent across all merchandise trade and all WTO members.5 However, this low figure covers significant differences across products and countries. For manufactures, the bound average tariff is 6 percent, while for textiles and clothing it is twice as high, at 12 percent, and for agriculture, it is more than five times as high, at 32 percent. 17. The high tariffs applicable to agriculture resulted in large part from the process of tariffication (replacing QRs and other NTMs with tariffs) as part of the Uruguay Round Agreement on Agriculture (URAA) and ceiling bindings offered by many developing countries. Tariffication improved the transparency of market access conditions, but most observers agree that the URAA will not result in a significant reduction in protection in agriculture, because quotas were converted into high tariff rates and tariff rate quotas which restrict market access (Box 2 and Appendix Table 2).6 18. On average, post-Uruguay Round bound rates are significantly higher than applied rates (7 percent versus 4 percent), with the largest difference in agriculture (32 percent versus 25 percent). Thus, especially in agriculture, there is considerable scope for applied tariff protection to rise and still be consistent with Uruguay Round commitments. 5 Estimates of the bound rates are based on the Uruguay Round schedules of concessions. The estimated applied rates post-Uruguay Round take into account the reductions in pre-Uruguay Round rates required by countries’ Uruguay Round commitments (see Finger and Schuknecht, 1999; Finger, Ingco, and Reincke, 1996). 6 Tariff rate quotas allow a certain quantity of imports to enter under a low tariff (in-quota tariff) and then apply a higher tariff to imports above the quota (out-of-quota tariff). - 17 - Since industrial countries generally set applied rates close to bound levels, the difference between bound and applied rates is due mainly to such differences in developing countries. Box 2. Agriculture: Tariffication and Tariff Rate Quotas (TRQs)1 The Uruguay Round Agreement on Agriculture (URAA) changed the form of protection in agricultural markets. Before its implementation most barriers to trade in agriculture were in the form of NTMs which were nontransparent in their application and effects. A major achievement of the Round was to require members of the WTO to replace NTMs with tariffs and to bind them against future increases. In addition, the URAA required members to reduce tariffs by 36 percent on average for industrial countries and 24 percent on average for developing countries. The implementation period was from 1995 to 2000 for industrial countries and from 1995 to 2004 for developing countries. The intent of the URAA was to replace NTMs with bound tariffs that provided protection equivalent to the NTMs, and then to reduce these bound tariffs by 36 percent (24 percent) for industrial countries (developing countries). In practice, however, the process allowed some countries to set tariff bindings that permitted increased protection. In addition, tariffs in some countries were expressed in absolute or specific terms as opposed to ad valorem terms, which make them far less transparent and their economic effects more difficult to quantify. For example, in the United States and the EU, approximately 44 percent of agricultural tariff lines have specific tariffs. The tariff equivalents of existing NTMs were established as the price gap between protected domestic prices and the average of reference world prices from 1986 to 1988. Since 1986–88 were years of exceptionally low world prices for many commodities, the tariff equivalents were exceptionally high. In addition, some countries interpreted the tariffication procedure in such a way as to result in even more generous levels of tariff equivalents (so-called “dirty tariffication”) (WTO, 1998). Many developing countries chose to use a special provision that allowed them to declare a ceiling bound rate of tariffs for the entire agricultural sector instead of tariff reductions. As part of the tariffication process, tariff rate quotas (TRQs) were introduced to provide minimum levels of market access after high tariffs were imposed. TRQs allow a certain quantity of imports to enter a market under a low tariff (in-quota tariff), and then apply a higher tariff to imports above the quota (out- of-quota tariff). The difference between in-quota and out-of-quota tariffs is frequently large. For example, in those OECD countries that apply TRQs, in-quota rates on agricultural products average 36 percent, while out-of-quota rates average 120 percent. Although only 37 WTO member countries use tariff quotas, they constitute a significant source of protection for agricultural markets. Most TRQs are concentrated in a few products, including fruits and vegetables, meat, cereals, dairy products, and oilseeds. Many of these products (especially vegetables and meat) are also subject to the highest levels of protection from regular tariffs. The six WTO members with the highest number of TRQs are located in Europe: Norway, Poland, Iceland, the EU, Bulgaria, and Hungary, but countries in other regions, including Colombia, the Republic of Korea, South Africa, the United States, and República Bolivariana de Venezuela, are also significant users. Because in-quota rates are high and many quotas are not administered in a transparent manner, many have been under-filled—the average fill rate in 1995 was only 66 percent, and it declined to 62 percent in 1998. 1 This box draws on WTO (2001), which provides a detailed discussion of the URAA and its implementation. - 18 - 19. For industrial countries, the post-Uruguay Round bound simple average tariff rate across all commodities is 4 percent; for developing countries, it is considerably higher— 25 percent (Table 2). Large differences in average tariff rates persist between industrial and developing countries across all products, with developing countries generally having higher tariffs. The smallest difference is in textiles and clothing, and the largest in agriculture. A similar pattern is evident in applied rates, except for agriculture, where industrial countries have applied rates (27 percent) that are significantly higher than those of developing countries (18 percent). Table 2. Post-Uruguay Round Average Tariff Rates (in percent) Bound Applied Industrial Developing Industrial Developing Countries Countries Countries Countries All products 4 25 3 13 Industrial goods 4 20 3 13 Textiles and clothing 11 24 8 21 1 Agricultural products 27 52 27 18 Source: Finger, Ingco, and Reincke (1996). 1 Excludes processed agricultural products. 20. Applied tariff rates in 2000 vary considerably across country groupings (Figure 8). The highest average tariff protection is in SSA countries (19 percent) and in the Middle East and North Africa (18 percent). Among broad country groupings, it is notable that the average tariff of LDCs (18 percent) is higher than that of other developing countries (15 percent) and well above that of industrial countries (5 percent). - 19 - Figure 8. Simple Average MFN Tariffs, 2000 20 18 Average of Least Developed Countries (LDCs) 16 14 World Average 12 percent 10 Average of Industrial Countries 8 6 4 2 0 Europe Baltics, Russia, and Western Hemisphere Asia Middle East and North Sub-Saharan Africa other countries of the Africa former Soviet Union Source: IMF Trade Policy Information Database (TPID). Tariff peaks and escalation7 21. This aggregate picture masks considerable variation across products and countries, resulting in pockets of high protection even in the most open economies. For example, there are products in manufacturing in industrial and developing countries that have tariffs above 50 percent, and in agriculture above 100 percent. In addition, there is a significant degree of tariff escalation in many product chains. 7 Tariff peaks are defined as tariffs of 15 percent or higher, or about three times the average tariff level in industrial countries. Tariff escalation is said to exist when tariffs rise as the level of processing of a product increases. This increases the effective rate of protection for downstream activities. For example, when the tariff on chocolate is higher than that on cocoa beans, tariff escalation occurs in the processing chain giving the factors of production (capital and labor) greater returns than they would otherwise receive under a uniform tariff regime, and distorting the patterns of production and trade. - 20 - 22. The prevalence of tariff peaks in the Quad countries (Canada, the EU, Japan, and the United States) is illustrated in Table 3.8 In the United States and Canada, most tariff peaks are in manufactures (more than 85 percent), particularly textiles and clothing, footwear, glass and glassware, and electrical parts. In the EU and Japan, peaks are concentrated in agriculture and food products, particularly dairy products, vegetables, coffee, tea, cereals, sugar, cocoa, and tobacco, although they also occur in footwear. 23. Products subject to tariff peaks in the Quad represent about 5 percent of total 1999 imports from developing countries; however, the LDCs are more adversely affected than other developing countries. Products subject to tariff peaks represent more than 11 percent of their exports to the Quad (Hoekman, Ng, and Olarreaga, 2000). For example, in Canada and the United States, the largest import category subject to tariff peaks is textiles and clothing, where more than 90 percent of LDC exports to these countries is concentrated. In the case of the EU and Japan, tariff peak products occur in agriculture, food products, and footwear. In short, tariff peaks in the Quad are concentrated in labor-intensive products of significant export interest to developing countries, and particularly LDCs.9 24. The tariff structure of developing countries also contains significant tariff peaks comparable to those of the Quad countries (Table 4). This fact, along with the significant trade among developing countries and their relatively high average tariff rates, indicates that a nontrivial share of the barriers facing developing country exports is to be found in other developing country markets. 8 The Quad is an informal consultative group in the WTO. 9 Some LDCs receive preferential access to industrial country markets in those goods where tariff peaks exist, which mitigates their effect to some extent. However, in general, preferences for tariff peak products are significantly lower than for other products (see Section III.C). - 21 - Table 3. Quad Countries: Imports and Tariff Peaks, 1999 1 USA EU15 2 Japan Canada All Quad 3 Number of tariff peak products (MFN>=15%) All products 311 317 233 732 1077 3 Agricultural products 48 290 178 85 364 3 Industrial products 263 27 55 647 713 3 Tariff peak products as percentage of all tariff lines 6.1 6.2 4.6 14.3 … Average MFN tariff rates (unweighted) Tariff peak products 20.8 40.3 27.8 30.5 28.0 All products 5.0 7.4 4.3 8.3 6.2 Maximum tariff rate (in percent) 121.0 251.9 170.5 342.7 221.5 Imports of tariff peak products (in billions of US$) 42.1 27.1 15.8 8.7 92.8 of which: All preferential and GSP countries 26.3 16.5 4.8 7.6 55.2 Least developed countries 0.9 0.3 0.03 0.09 1.3 ACP developing countries (excluding LDCs) … 2.7 … … … Imports of tariff peak products (as percent of all imports) 4.6 3.4 4.9 4.6 4.2 of which: All preferential and GSP countries 6.6 4.9 2.8 4.8 5.2 Least developed countries 15.0 2.8 2.6 30.2 11.4 ACP developing countries … 15.0 … … … Import revenue collection in tariff peak products (in billions of US$) 5.4 8.9 6.3 1.6 22.2 of which: All preferential and GSP countries 4.6 4.3 1.4 0.7 11 Least developed countries 0.2 0.03 0.001 0.02 0.2 ACP developing countries … 0.57 … … … Source: Hoekman, Ng, and Olarreaga (2000). 1 Applied tariff rates. 2 Excludes all intra-EU trade in world totals. 3 Number of nonoverlapping categories. - 22 - Table 4. Select Countries: Import Tariff Maximum and Tariff Peaks1 Tariff peaks Year Maximum average 2 (in percent) Argentina 2000 33 19 Brazil 1998 49 20 Egypt 1997 164 43 India 1997 260 36 Indonesia 1999 200 22 Korea 1999 937 76 Thailand 2000 80 31 Mexico 1999 260 20 Canada 1999 343 31 EU 1999 252 41 Japan 1999 179 28 United States 1999 121 21 Source: World Bank staff estimates. 1 Tariff peaks are defined as those tariffs 15 percent or higher. 2 Applied ad valorem tariffs. Excludes specific tariffs. 25. In addition to tariff peaks, there is a considerable degree of tariff escalation in both industrial and developing country protection. Tariff escalation is a major concern for developing countries. By reducing demand for more processed imports from developing countries, tariff escalation hampers the expansion of their processing industries, and hence the means of accumulating skills and capital, and export diversification. Also, the concentration of exports in less processed commodities often results in slower export growth (because of slower growth in demand for these products in industrial and high-income developing countries), low value-added in production, and greater exposure to the risk of commodity price volatility. 26. While the Uruguay Round reduced tariff escalation for bound rates, the problem of tariff escalation persists (Table 5). Moreover, reductions in tariff escalation are by no means uniform. Large variations exist among different production chains and among different importing countries. For example, in agriculture, after Uruguay Round liberalization, the number of production chains with tariff escalation will be 50 percent above the number with de-escalation (Lindland, 1997). - 23 - Table 5. Industrial Country Imports from Developing Countries: Changes in Tariff Escalation (billions of U.S. dollars and percentage) Bound Tariff Rates Imports (US$b) Share of each Pre-UR Post-UR Absolute stage reduction All industrial products Raw materials 36.7 22 2.1 0.8 1.3 Semi-manufactures 36.5 21 5.4 2.8 2.6 Finished products 96.5 57 9.1 6.2 2.9 All tropical industrial products Raw materials 5.1 35 0.1 0.0 0.1 Semi-manufactures 4.3 30 6.3 3.4 2.9 Finished products 4.9 34 6.6 2.4 4.2 Natural resource-based products Raw materials 14.6 44 3.1 2.0 1.1 Semi-manufactures 13.3 40 3.5 2.0 1.5 Finished products 5.5 17 7.9 5.9 2.0 Source: Blackhurst, Enders, and Francois (1996). 27. Tariff escalation affects a range of products, and its practice is not confined to industrial countries (Figure 9). It is commonly used in many developing countries to promote manufacturing activities. For example, among 18 major processing chains of developing country exports to industrial countries, the majority suffers from tariff escalation.10 Among major manufactured products, tariff escalation is most prevalent in textiles and clothing, leather and leather products and across a wide range of countries (both industrial and developing). Rubber products, wood, pulp, paper and furniture, and metals are also frequently subject to tariff escalation. These are all products in which many developing countries have comparative advantage. The incidence of tariff measures 28. The incidence of applied tariffs across different country groups can be compared using trade-weighted tariffs (Table 6).11 In general, all countries face higher tariffs in developing country markets (12.6 percent on average). However, there is greater differentiation in the incidence of tariffs in industrial country markets: in manufacturing, tariffs faced by developing countries (3.4 percent on average) are 70 percent higher than 10 These are mainly agricultural processing chains: coffee, tea, cocoa, spices, vegetable plaiting materials, oil seeds and vegetable oils, tobacco, manioc roots and tubers, tropical fruits, tropical nuts, tropical woods, rubber, jute, fishery, forestry, hides and skins, paper and paperboard, and iron and steel. 11 The results reported here are broadly consistent with preliminary results from the UNCTAD/World Bank study of market access barriers facing developing country exporters. - 24 - Figure 9. Tariff Escalation: Average Applied Tariffs by Stage of Production United States: MFN Tariffs, 2000 Canada: MFN Tariffs, 2000 20 20 15 15 average tariff average tariff 10 10 5 5 0 0 Agricultural products Textiles and clothing Other products Agricultural products Textiles and clothing Other products EU: MFN Tariffs, 2000 Japan: MFN Tariffs, 2000 20 20 15 15 average tariff average tariff 10 10 5 5 0 0 Agricultural products Textiles and clothing Other products Agricultural products Textiles and clothing Other products WAEMU Countries: MFN Tariffs, 1999 MERCOSUR Countries: MFN Tariffs, 2000 25 25 20 20 15 average tariff 15 average tariff 10 10 5 5 0 0 Agricultural products Textiles and clothing Other products Agricultural products Textiles and clothing Other products Raw Materials Semi-processed goods Finished goods Source: IMF staff estimates. - 25 - those faced by industrial countries (2 percent on average). In developing country markets, LDCs face tariffs that are about 25 percent higher than those faced by industrial countries. In agriculture, there are no significant differences in the incidence of tariffs in developing and industrial country markets. However, the level of agricultural tariffs is very high compared to tariffs on manufactured goods: about ten times higher for industrial countries and two and a half times higher for developing countries. It is also striking that it is industrial country agricultural exports, not those from developing countries, that face the highest tariffs (32 percent) in other industrial countries and in developing country markets (31 percent). Table 6. Trade-Weighted Average Tariffs Importer Tariff rate Relative to world average Industrial Developing Industrial Developing Exporter World countries countries World countries countries All goods World 8.1 4.4 12.6 1.0 1.0 1.0 Industrial countries 8.5 4.2 12.8 1.0 1.0 1.0 Developing countries 7.7 4.6 12.3 0.9 1.0 1.0 of which: LDC 6.3 5.0 11.0 0.8 1.2 0.9 Manufactures World 6.7 2.6 11.4 1.0 1.0 1.0 Industrial countries 6.6 2.0 11.1 1.0 0.8 1.0 Developing countries 6.8 3.4 12.0 1.0 1.3 1.1 of which: LDC 5.2 3.5 13.7 0.8 1.3 1.2 Agricultural goods World 27.4 26.3 28.4 1.0 1.0 1.0 Industrial countries 31.6 32.0 31.2 1.2 1.2 1.1 Developing countries 23.2 21.9 25.0 0.8 0.8 0.9 of which: LDC 16.4 16.2 16.9 0.6 0.6 0.6 Source: Preliminary GTAP 5, prerelease 2, using 1997 applied tariff and trade weights, excluding intra-EU trade. 29. Information relative to world averages shows how the incidence of tariffs affects LDC exports. For merchandise trade as a whole, LDCs face tariffs on average 20 percent higher than the world into industrial country markets, and 10 percent lower than the world into developing country markets. In manufacturing, however, LDCs face higher tariffs than the world average in both industrial country markets (30 percent higher) and in developing country markets (20 percent higher) . In agriculture, by contrast, they face much lower tariffs, 40 percent below the world averages in both industrial and developing country markets. Even so, LDCs face tariffs of 16–17 percent on their agricultural exports. B. Agricultural Subsidies and Nontariff Measures (NTMs) 30. The Uruguay Round provided important disciplines related to agricultural subsidies and NTMs of protection. In agriculture, in addition to the tariffication of NTMs, it extended - 26 - disciplines to domestic policies that subsidize agricultural production and exports and distort trade. These policies undermine developing countries’ agricultural sectors and exports by depressing world prices and pre-empting markets. In manufacturing it prohibited the use of voluntary export restraints which were commonly used before the Round and concluded the Agreement on Textiles and Clothing (ATC), which will phase out MFA quotas by 2005. Nevertheless the extension of disciplines in agriculture had only a modest effect on the extent of agricultural support measures or export subsidies, and in manufacturing significant NTMs still exist, particularly in textiles and clothing. Agricultural subsidies 31. The URAA extended disciplines to domestic support policies and direct export subsidies. Regarding domestic agricultural support policies, a key aspect of the Agreement is the distinction between domestic policies that distort trade and those that do not. This distinction is reflected in the agreement through the designation of an “amber box” for policies deemed to have the largest effect on production and trade and a “green box” for policies that have minimal effect on trade.12 “Green box” policies are exempt from reduction requirements. For “amber box” policies, member countries committed to reduce the ceilings for total support as measured by the Aggregate Measure of Support (AMS) by 20 percent over the 1995–2000 implementation period (13 percent for developing countries over 1995– 2004).13 32. Commitments to reduce AMS were made by 45 out of 140 members of the WTO under the Uruguay Round or as a result of WTO accessions. However, in most cases the ceilings of support under these commitments were well above actual levels and thus not constraining (WTO, 2001). For example, for OECD countries, the estimate of total support for agriculture increased from US$329 billion in 1997 to US$361 billion in 1999 (OECD, 2000a). The EU, Japan, and the United States account for about 90 percent of total domestic support. Not all this support is trade distorting, because disciplines under the URAA have resulted in some “re-instrumentation” of domestic support programs away from the most trade-distorting measures toward the less distorting ones. Nevertheless, some analysts have concluded that the benefits from the URAA have been modest, because of the aggregate (i.e., nonproduct-specific) nature of the reduction commitments and because even some of the measures considered in the “green box” as nondistortionary may not be entirely production or trade-neutral (OECD, 2001). 12 “Amber box” policies include commodity-specific subsidies and administered product prices. Among these there are so-called "blue-box" policies which provide support linked to certain production-limiting programs; these policies are exempt from reduction requirements. 13 Most developing countries had no AMS for most commodities in the base period and are subject to a de minimis limit of 10 percent. - 27 - 33. Regarding direct export subsidies, key elements of the URAA were the commitments of members to limit and reduce the volume and value of export subsidies and the prohibition of new export subsidies.14 Commitments to reduce export subsidies affect mainly OECD countries and particularly the EU, which accounts for 90 percent of OECD export subsidies (OECD, 2001). Overall, the URAA has contained and reduced direct export subsidies, even though disciplines were weakened by the possibility to redistribute the value of subsidies or the volume of subsidized exports between years and to aggregate products within a commodity group in their commitments. After implementation of commitments, use of export subsidies notified by WTO members, amounting to about US$13 billion annually, will remain significant, with the EU continuing to account for the lion’s share (WTO, 2001, Table III.10). Traditional NTMs 34. Traditional NTMs, like quotas, voluntary export restraints, and nonautomatic licensing, have been significantly curtailed by the Uruguay Round Agreements. Nevertheless, NTMs remain a market access issue. Among developing countries, restrictive licensing is the most common NTM, and products most often subject to control include fuel and mineral products, rubber products, machinery, and precious stones and metals. 35. In industrial countries, NTMs are most prevalent in textiles and clothing. The ATC provided for a phased elimination of quotas inherited from the MFA over a period of 10 years; however, the ATC will not lead to a substantial increase in market access for developing countries during the transition period because of the way it has been implemented, with most industrial countries backloading their commitments (Appendix II). As a result, halfway through the implementation of the ATC, only a few quotas have been liberalized, and it has been estimated that by 2004 the 11 principal developing country exporters will still face quota restrictions on over 80 percent of their exports (Spinanger, 1999). 36. Because the phase-out of MFA quotas is backloaded, developing countries are concerned that some political resistance to liberalization could increase in the final stage of the transition period. Although the ATC cannot be extended, some other forms of protection could substitute for quotas. In addition to tariffs, which will remain high after MFA quotas are phased out, contingent measures such as antidumping or technical barriers could become more common as a means of protection against imports from developing countries. 14 Each country agreed to reduce the volume of subsidized exports by 21 percent over six years (14 percent over 10 years for developing countries) and reduce the value of export subsidies by 36 percent (24 percent over 10 years for developing countries). - 28 - Other NTMs 37. The decline in traditional measures of protection has gone hand-in-hand with a rise in the use of contingent trade measures such as antidumping. Administrative requirements such as standards and customs procedures can also hinder imports. The restrictiveness of these measures is difficult to assess, especially since not all are trade- distorting. On the contrary, they may be used to maintain fair competition and safeguard health and security. They may nevertheless become barriers if used in a nontransparent manner, or if abused. Antidumping, countervailing duties, and safeguards 38. Trade remedies permitted under the WTO agreements include antidumping measures, countervailing duties against “actionable” subsidies, and safeguard measures to protect against “serious” injury from import surges. Among these, antidumping has become the trade remedy of choice for both industrial and developing countries. In 1958, only 37 antidumping decrees were in force across all GATT contracting parties. By the 1990s, antidumping had become the industrial countries’ preferred safeguard instrument and gained increased popularity among developing countries (Figure 10). 39. During 1995–99, over 1,200 antidumping investigations were initiated (Table 7). Industrial countries initiated 25 percent of their cases against other industrial countries and 49 percent against (mostly higher-income) developing countries (excluding transition economies), while developing countries brought 35 percent of their cases against industrial countries and 36 percent against other developing countries. Thus developing countries appear to be the major object of antidumping cases from both industrial and other developing countries. Measured in terms of how intensely a country’s exports are targeted by foreign antidumping cases, transition economy exporters are more adversely affected than developing countries (Finger, Ng, and Wangchuk, 2000). 40. Antidumping measures can be important obstacles to market access in particular products for both developing and industrial countries. 15 However, the exports of LDCs are little affected. First, their exports are not concentrated in the types of products that elicit antidumping protection (generally more advanced or more processed manufactures). Second, LDCs do not generally export in the volumes required to threaten, or cause injury to, domestic producers in industrial and relatively advanced developing countries. Product standards 41. Product standards, or rules governing characteristics of goods, are essential for the effective functioning of markets and provide an important support for the multilateral trade system. Through the Agreement on Technical Barriers to Trade (TBT) 15 As to sectoral distribution, producers of base metals, chemicals, machinery, electrical equipment, plastics and textiles frequently seek antidumping measures. - 29 - and the Agreement on the Application of Sanitary and Phytosanitary Measures (SPS), the Uruguay Round attempted to strengthen international rules governing product standards in order to minimize their use for protectionist purposes and create a level playing field. The TBT relates to all products and measures, while the SPS covers sanitary standards for food and phytosanitary standards for animals and plants. Figure 10. Initiations of Antidumping Investigations, 1987-99 300 250 200 number of investigations Industrial Countries 150 100 50 Developing Countries 0 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Source: WTO. Table 7. Initiations of Antidumping Investigations, 1995-99 Affected countries Industrial Developing Transition Initiating country countries countries countries Total Number of investigations 382 502 334 1218 Industrial countries 126 244 129 499 Developing countries 252 258 201 711 Transition countries 4 0 4 8 Percent of investigations 31 41 27 100 Industrial countries 25 49 26 100 Developing countries 35 36 28 100 Transition countries 50 0 50 100 Source: WTO (2001), Table II.8. - 30 - 42. A source of concern regarding standards is the capacity of poor countries to meet increasingly complex health and technical standards, which they do not play a major part in developing, or to bring disputes when standards are used to discriminate against their exports. For example, health and safety standards usually require testing and conformity assessments for all producers, as well as upgrading production methods to meet quality standards; the associated costs will be higher for exporters if they must conform to standards different from those in their home markets (Box 3). Box 3. Standards in International Trade Product standards can ensure quality, safety, and other public objectives, facilitate the mix and match of parts and components, and lower entry barriers by lowering inspection and testing. A significant portion of internationally traded goods is subject to product standards, including 60 percent of U.S. exports and 75 percent of intra-EU trade. However, product standards can also raise difficult issues because of the manner in which they are formulated or negotiated, and the costs and know-how required to implement and dispute them. As border protection has declined, there is a concern that product standards have been increasingly used to limit market access (adding unnecessarily to the costs of trade) and competition from imports. Standards also raise entry costs for exporters by increasing one-time costs of product redesign and the creation of an administrative system for compliance. In addition, standards impose recurrent costs of maintaining quality control, testing, and certification, and result in costs to meet precise technical regulations and carry out conformity assessments (whether a product conforms to a regulatory requirement). Indeed, conformity assessments are perceived as presenting the largest potential costs for exporters among technical standards (World Bank, 2001). Revisions to standards can have important implications for exporters. In the area of health standards, for example, the EU is harmonizing standards for levels of aflatoxin, a substance that may cause liver cancer, in food products.1 The new standard, which is more stringent than those suggested by international guidelines, would lower health risks by approximately 1.4 cancer deaths per billion people per year. Few developing countries have the technology to evaluate the danger of aflatoxins, nor do they have the capabilities and the scientific analysis to address the new EU standard. A study by Otsuki, Wilson, and Sewadeh estimated that implementation of the new EU aflatoxin standards could reduce African exports of cereals, dried fruits, and nuts to Europe by 64 percent, or US$670 million per year. Source: Finger, Ng, and Wangchuk (2000); Otsuki, Wilson, and Sewadeh (2000); and World Bank (2001). 1 Aflatoxins are a group of structurally related toxic compounds which contaminate certain foods and result in the production of acute liver carcinogens in the human body. The most predominant and most toxic of the major four categories of aflatoxins is identified in corn and corn products, groundnuts and groundnuts products, cottonseed, milk, and tree nuts such as Brazil nuts, pecans, pistachio nuts, and walnuts. 43. One indicator of the relevance of standards in restricting trade is the increasing number of trade disputes over standards and technical barriers that have been initiated during the past six years. By end-2000, the WTO Dispute Settlement Mechanism had considered 27 disputes that referenced either the SPS or the TBT, 11 of them focusing on food safety regulations and 5 on technical regulations tied to customs requirements. Most of these complaints were brought to the WTO by industrial countries against other industrial countries; out of 27, only 6 were brought by developing countries against industrial countries. No low-income country, other than India, has brought cases to the WTO under the TBT or - 31 - the SPS.16 Another indication of the importance of standards and their effects on trade is the fact that most countries at the 1999 WTO Ministerial Conference emphasized the need to address standards, in particular how they are set, in the context of a new round of trade talks (Wilson, 1999). Customs/red tape 44. Customs procedures are increasingly regarded as a source of concern for the international business community and a source of conflict in the WTO. The concern can be attributed to various factors, including the need to create new tariff codes for products subject to antidumping duties and the numerous and more detailed rules of origin associated with preferential trading arrangements (bilateral and regional trading arrangements and nonreciprocal preferences). Industrial countries regard customs procedures in developing countries and emerging markets as not having evolved sufficiently to adjust to the rapidly growing volume of trade of the last decade. These views are endorsed by service providers (e.g., express mail couriers). Developing countries, on the other hand, complain about the “spaghetti bowl” of rules of origin included in the customs procedures of industrial countries (in particular, those which are hubs of regional agreements). 45. The Uruguay Round Agreements that have implications for customs procedures are those on Customs Valuation, Import Licensing Procedures, Preshipment Inspection, Rules of Origin, Technical Barriers to Trade and Sanitary and Phytosanitary Measures. The main obstacles related to customs procedures include excessive documentation requirements, lack of automation and low use of information technology, lack of transparency in import and export requirements, corruption, lack of use of risk assessment techniques and audit-based controls and lack of cooperation among customs and other government agencies (Box 4). Customs procedures have emerged as one of the recurrent themes in WTO dispute settlement cases. Cases have involved customs classification, duty collection, Generalized System of Preferences (GSP) coverage, quota management and import measures. Industrial and developing countries, both as complainants and defendants, have participated in these dispute settlement cases. C. Trade Preferences 46. An argument often used to allay concerns about the lack of market access of developing countries to industrial country markets is that there are preferential schemes, such as the GSP, that are especially designed to help these countries sell their products in industrial country markets.17 There is evidence, however, that the preferential margins under these schemes are much smaller than they at first appear, and these margins are shrinking. 16 This may reflect the fact that pursuing a case in a manner consistent with WTO rules is expensive and resource-intensive. 17 There is also the Generalized System of Tariff Preferences (GSTP), under which developing countries extend preferences to each other. - 32 - Box 4. The Cost of Customs and Border Procedures Numerous and complex customs procedures and requirements generate substantial costs. According to one source, the average customs clearance transaction in developing countries involves 40 documents, 200 data elements, some 30 of which are requested at least 30 times, and 60 to 70 percent of which must be re-keyed at least once. Excessive control and inefficiency of customs clearance procedures, combined with the monopoly of service providers in the port or other key entry points, have been widely observed in many parts of the developing world. Resulting costs can exceed tariffs in many cases. Documentary red tape in customs procedures has been estimated to increase the cost of imports substantially, by around 7-10 percent of world trade. For example, in the mid-1990s, the average customs clearance transaction in countries of the Middle East and North Africa (e.g., Egypt, Jordan, Lebanon) required 25 to 30 stages, and took from one day to several weeks (Roy, 1998). Valuation procedures are a major uncertainty for importers, as customs generally expects under-invoicing, unless there are incentives for capital flight, in which case over-invoicing is common. It is the practice in some Middle East and North Africa countries that customs officers question every invoice in order to charge penalties or collect “rewards” (for instance, in Jordan, the law rewards customs officials who detect invoice misreporting and charge penalties to importers). Problems facing transport operators when crossing borders also impede trade. Transportation by road is an important cross-border mode of passenger and freight transport in many countries (in particular for the landlocked countries of Africa and Latin America). According to the International Road Transport Union, the most restrictive cross-border transportation can be found in some parts of Europe and the Middle East and North Africa (WTO Document, S/C/W/60). The cost of delays at the border that affect international trade is estimated to be about 6 percent of transport time in some countries of Central and Eastern Europe. Complaints from transporters also include the excessive charges that impair the freedom of transit contained in Article V of the General Agreement on Tariffs and Trade (GATT). Source: Hoekman and Kostecki (2001); Roy (2001); Staples (1998); and WTO (S/C/W/60). 47. Frequently, developing countries gain preferential access to industrial country markets through unilateral schemes such as the GSP or through bilateral agreements, such as the Lomé Convention (Cotonou Agreement). About 170 developing countries benefit from GSP (or better) preferences in one or more of the Quad Countries.18 The margins of preference provided by the GSP schemes and bilateral trade agreements to beneficiaries are frequently significant. In the EU, Japan, and the United States, the GSP results in a margin of preference of 50 percent or more with respect to the MFN rate, while for Canada the margin is 25 percent (Table 8). For LDCs, the preferences are even greater in Japan and the United States (60 percent) and in Canada (45 percent). 18 Hoekman, Ng, and Olarreaga (2000). In 1971, the European Community was the first member of the Quad to grant GSP preferences to developing countries. The EU GSP has a tier of countries that benefit from the so-called “Super GSP.” From the mid-1980s, this improved GSP, which provides access terms that are as favorable as Lomé, was granted only to LDCs. In 1990, the EU extended it to the countries in the Andean Pact—Bolivia, Colombia, Ecuador, and Peru, and in 1991 it was also provided to the countries of Central America. - 33 - Table 8. Tariff Peaks and Preferential Duty Rates in the Quad, 1999 Average preferential tariff rate Number of Preferential Trade Agreements/GSP countries Tariff peak products All Goods United States MFN rate 20.8 5.0 Preferential rates Canada 1 0.6 0.1 Mexico 1 1.6 0.3 Israel 1 0.6 0.1 Andean countries1 4 14.0 1.7 Caribbean Community2 22 13.5 1.6 GSP-only beneficiaries3 80 16.0 2.4 Least developed countries4 38 14.4 1.8 European Union MFN rate 40.3 7.4 Preferential rates Eastern Europe and Middle East5 30 20.1 1.8 GSP-only beneficiaries6 42 19.8 3.6 ACP countries that are least developed 7 37 11.9 0.8 Other ACP countries8 32 12.4 0.9 Other least developed countries9 11 12.6 0.9 Japan MFN rate 27.8 4.3 Preferential rates GSP-only beneficiaries10 127 22.7 2.3 Least developed countries11 42 19.0 1.7 Canada MFN rate 30.5 8.3 Preferential rates United States 1 7.1 1.6 Australia 1 28.2 7.8 New Zealand 1 28.2 7.8 Mexico 1 15.9 3.1 Chile 1 12.2 2.4 Israel 1 11.8 2.5 Caribbean countries12 18 23.3 4.3 GSP-only beneficiaries13 108 28.2 6.2 Least developed countries14 47 22.8 4.4 Source: Hoekman, Ng, Olarreaga (2000). 1 Includes Bolivia, Colombia, Ecuador, and Peru under Andean Trade Preference Act. 2 Includes 20 Caribbean countries under the Caribbean Basin Economic Recovery Act plus the Bahamas and Nicaragua. 3 Includes 80 developing countries or territories under the GSP scheme, excluding 29 other developing economies. 4 Based on UN 49 least developed countries, but excludes 10 countries. 5 Includes countries with reciprocal and nonreciprocal trade agreements with the EU. 6 Most developing countries in Latin America and Asia; excludes Hong Kong SAR, Korea and Singapore (non-GSP nations). 7 Includes 37 ACP Countries that are least developed under Lomé Convention. 8 Includes 32 ACP countries that are under Lomé Convention, but not in group of least developed countries. 9 Includes 11 Least Developed Countries that are not under ACP. 10 127 countries; excludes Albania, Bosnia, Estonia, Latvia, Lebanon, Lithuania, Macedonia, Moldova, Vietnam, Yugoslavia. 11 Excludes 3 LDCs: Comoros, Djibouti and Tuvalu. 3 other LDCs (Congo DR, Kiribati and Zambia) are included in the GSP group. 12 Includes 18 Caribbean countries or territories under the Commonwealth Caribbean Countries Tariff. 13 Excludes 8 developing countries: Albania, Aruba, Bosnia & Herzegovina, Macedonia FYR, Mongolia, Oman, Saudi Arabia, Yugoslavia. 14 Excludes Myanmar. - 34 - 48. However, while these margins seem generous, they mainly apply to products that already face relatively low tariffs, between 4 and 8 percent on average. With the exception of the EU, the margins of preference on the much higher peak tariffs are significantly lower.19 In Canada, the margin of preference on developing countries’ tariff peak products is 8 percent (of the MFN tariff); in Japan, 18 percent; and in the United States, 23 percent. For LDCs, the preference margin for tariff peak products is 25 percent in Canada, and 30 percent in Japan and the United States. 20 49. All the GSP schemes have graduation mechanisms that are related to income and market share and restrictive rules of origin. These schemes are time bound and subject to (uncertain) renewal, often on an annual basis. Countries graduate if they pass a certain per capita income threshold. In addition, preferences are removed for particular products if they exceed a certain import share in the market of the GSP-granting country. Some recipient countries have noted instances where preference-granting countries have attached “noneconomic” conditions to renewals. Invariably, countries graduate in stages with the more developed beneficiaries having some of their products removed before they graduate from the program. Some analysts argue that these limitations undermine the benefits provided by the GSP since they increase uncertainties about market access (Stevens and Kennan, 2000). 50. The Director-General of the WTO has called for the extension of duty- and quota- free access for the exports of LDCs, including as part of a number of “confidence-building” measures to enhance LDC participation in international trade. In response, a number of countries have announced significant market-opening measures during the past year (Box 5). In this regard, the EU recently granted the LDCs free access for all products but arms, which will provide significant benefits given the importance of the EU market for these countries.21 D. Conclusions 51. The Uruguay Round resulted in greater security of the trading environment by increasing the overall number of tariff bindings. Unbound tariffs are concentrated in manufactured goods, often in developing countries. In general, industrial countries set their applied rates close to bound rates; however, for developing countries, bound rates are twice as high as applied rates for all products, and three times as high for agriculture. This imparts a significant degree of uncertainty to their tariff regimes. 19 In the EU, the preference margin on peak products is approximately 51 percent, which compares favorably with the GSP preference margin for all products. The margin for LDCs is about 70 percent. 20 As an illustration, a 30 percent preference on a tariff of 15 percent would still result in exports facing a tariff of 10.5 percent. 21 Liberalization is being phased in for bananas, rice, and sugar (Box 5). - 35 - Box 5. Recent Market Access Initiatives for LDCs1 The following is a summary of “notifications” by WTO members and the recent EU “Everything But Arms” Proposal adopted by the EU Council on February 27, 2001. 1. Canada: Canada notified the WTO on December 7, 2000 that 570 additional tariff lines would be accorded duty- free treatment, or about 90 percent of product categories from the LDCs. However, Canada still maintains high tariffs on a number of products, including cane sugar, wine, textiles, clothing, and footwear. 2. The European Union: The EU Council adopted the “Everything But Arms” proposal that extends to the LDCs duty- and quota-free access for 919 tariff lines, leaving 25 lines related to arms trade. For three sensitive agricultural products—bananas, rice, and sugar—it was agreed that liberalization would be phased in. Duties on fresh bananas will be reduced by 20 percent annually starting on January 1, 2002, and eliminated at the latest on January 1, 2006. Duties on rice will be reduced by 20 percent on September 1, 2006, by 50 percent on September 1, 2007, and by 80 percent on September 1, 2008, and eliminated at the latest by September 1, 2009. Duties on sugar will be reduced by 20 percent on July 1, 2006, by 50 percent on July 1, 2007, and by 80 percent on July 1, 2008, and eliminated at the latest on July 1, 2009. To “compensate for the delay in liberalization for these products,” the EU will extend immediate market access to the LDCs, through the creation of duty-free quotas for sugar and rice, based initially on the best figures for LDC exports during the 1990s, plus 15 percent. These will increase by 15 percent each year during the interim period. The European Commission has pledged that it will monitor imports of rice, bananas, and sugar, and apply safeguard measures if necessary to prevent damaging import surges. There will also be monitoring to verify respect for rules of origin, as well as anti-fraud measures. 3. Japan: On February 6, 2001 Japan notified the WTO that a new system will be put in place that enables Japan to designate separately products originating in LDCs for duty-free and quota-free treatment. The required amendments to laws for this were passed at end-March 2001 and the system became effective April 1, 2001. The new arrangement increased the number of eligible LDCs from 42 to 46 and introduced an expanded list of products to be given preferential treatment, which included both industrial/agricultural food products, with an addition of about 360 industrial products that previously were not on the GSP list (referred to as “99 percent initiative on industrial tariffs”). The Japanese initiative mainly focuses on industrial products (led by the “99 percent initiative on industrial tariffs”) and does not cover a large part of agricultural and food products (e.g., rice), some of which are subject to very high tariffs. 4. Republic of Korea: On April 28, 2000 the Republic of Korea notified the WTO that it would remove tariffs on some 80 items originating from LDCs. 5. New Zealand: In November 2000 New Zealand notified its decision to extend duty- and quota-free access to its market for all products from LDCs with effect from July 1, 2001. 6. Norway: In November 2000 Norway notified its GSP system which accords duty-free treatment to all industrial and agricultural imports from LDCs, with the exception of flour, grains, and feeding stuffs. These products originating in a least developed country are entitled to a 30 percent reduction of the applied customs duty. In April 2001, following the EU’s lead, Norway announced that it would provide duty- and quota-free access for all products from LDCs from July 1, 2002; this proposal has to be adopted by the national legislature. It goes further than the EU’s proposal by granting earlier access for all products except weapons. 1 Initiatives during 2000–01 are covered. Some developing countries (e.g., Hong Kong SAR, Hungary, and the Slovak Republic) already provide duty- and quota-free access to LDCs. - 36 - Box 5. Recent Market Access Initiatives for LDCs (continued) 7. United States: In October 2000 the United States announced that 34 SSA countries were designated as beneficiaries under the African Growth and Opportunity Act (AGOA) and would receive benefits for 1835 tariff lines as from December 2000. AGOA extends duty-free and quota-free treatment to imports of virtually all products as long as they meet AGOA’s rule of origin requirements and are imported directly from a beneficiary SSA country. Exceptions include fabrics and yarns not imported as part of a finished apparel product, and products determined by the U.S. Government to be import sensitive (e.g., all agricultural commodities that are subject to a tariff-rate quota). Apparel and clothing have their own preferential regime that includes some flexibility in the rules of origin to utilize regional fabrics (i.e., not made in the United States). Notably, SSA countries that are LDCs (only 6 SSA countries are not LDCs) can export apparel wholly assembled in their countries, irrespective of the origin of the fabric, for a period of four years. 52. In manufacturing, average tariff protection is generally low, but tariff peaks and escalation in sensitive products (textiles and clothing, agriculture, food products, wood products, and pulp and paper) disproportionately affect the products exported by developing countries. This biases incentives for developing country exporters, and particularly those of LDCs, towards products with lower value added and inhibits the diversification of exports. 53. At the same time, developing countries maintain much higher average tariffs in manufacturing (by a factor of three or four) than industrial countries. More advanced developing country markets are among the fastest growth markets for manufactured exports, and the fact that they are relatively more protected further disadvantages developing country and LDC exports.22 54. Tariff protection in agriculture is much higher than in manufacturing. The simple average tariff on agricultural products in industrial countries is nine times that for manufactured products. For developing countries it is one and a half times higher. Tariff protection in agriculture also exhibits tariff peaks and escalation which disproportionately affect developing countries, particularly LDCs. In addition there are tariff rate quotas on certain products, such as fresh fruits and vegetables, that restrict market access for developing country exporters. 55. Although the Uruguay Round imposed disciplines on agricultural subsidies and the use of NTMs, they continue to distort and restrict trade in selected areas. Total support to agriculture has not been reduced, although there has been a shift towards less trade-distorting support programs and a reduction in direct export subsidies. Even so, export subsidies will remain significant at the end of the Uruguay Round implementation period. 22 These are also the markets where LDC exporters of labor-intensive light manufactures are likely to face the most competition from domestic producers. - 37 - 56. Traditional NTMs are concentrated in textiles and clothing, and so far, implementation of the Uruguay Round has not resulted in significant increases in market access. Once quotas are dismantled, developing country exporters will continue to face high tariffs and escalation. 57. Nontraditional NTMs are harder to quantify and assess, but their use is becoming more prevalent as traditional forms of protection are reduced under WTO disciplines. Antidumping activity is on the rise; this concerns essentially high-income developing and transition countries, not LDCs. The need to conform to technical, sanitary, and phytosanitary standards imposes costs on exporters that can exceed benefits to consumers. This makes it difficult for the poorer countries to shift exports towards higher value-added manufactures. Complex and/or inefficient customs procedures can also deter trade. 58. To some extent, the restrictive effects of tariff and nontariff measures are mitigated by preferential access schemes for poorer countries. However, these schemes are less generous than they appear because they often apply to products that already face low tariffs and they are subject to various eligibility criteria and conditions. IV. TRADE IN SERVICES 59. During the past two decades, production of services has become the main economic activity of most economies. At present the share of services in value added ranges from around 38 percent in low-income countries to more than 65 percent in high-income countries. In parallel to the growing importance of services in domestic production, there has been an expansion of the share of services in world trade.23 Since 1980, the share of commercial services in world exports of goods and services has grown from 15 percent to almost 20 percent.24 Based on data collected from balance of payments statistics, which underestimate services trade, the value of service trade was over US$2,200 billion in 1998. Developing countries as a whole account for a small but growing share of this trade (10 to 36 percent of this trade, depending on the mode of delivery; Box 6). Although no developing country ranks among the most prominent exporters or importers of services, trade in services is an important source of foreign exchange and export earnings for many developing countries. 23 These are primarily commercial services which include shipment (freight and insurance), other transportation, travel, and other private services such as brokerage, communications, leasing, management, retailing, nonmerchandise insurance, processing and repair, professional and technical services, and subscriptions to periodicals. 24 Maurer and others (2001). These figures underestimate the true share of services in total world trade. For instance, conventional balance of payments statistics fail to cover trade in services embodied in goods or production and sales of foreign affiliates. - 38 - Box 6: What’s Different About Services? There are several fundamental differences between goods and services. Unlike goods, many services are often seen as intangible, invisible and perishable, requiring simultaneous production and consumption. The proximity aspect of many services transactions has crucial implications for the nature of services trade. The need for proximity creates a need for factor mobility. Trade in services is thus inextricably linked to the movement of factors of production: capital (in the form of foreign direct investment) and labor. The General Agreement on Trade in Services (GATS) defines trade in services (in Article I) to include four modes of supply: • Cross-border supply (Mode 1): services supplied from the territory of one Member into the territory of another. An example is software services supplied by a supplier in one country through mail or electronic means to consumers in another country. • Consumption abroad (Mode 2): services supplied in the territory of one Member to the consumers of another. Examples are tourism or education services consumed in another country. Also covered are activities such as ship repair abroad, where only the property of the consumer moves. • Commercial presence (Mode 3): services supplied by one member through commercial presence in the territory of another. An example is an insurance company owned by citizens of one country establishing a branch in another country to provide services. • Presence of natural persons (Mode 4): services supplied by nationals of one Member in the territory of another. Examples are a doctor of one country supplying through his physical presence services in another country, or the services supplied by an on-site engineer. 60. In 1998, about 40 percent of trade in services occurred via cross-border trade; another 40 percent occurred via commercial presence; whereas services trade involving the temporary movement of labor accounted for only 1.5 percent of the total. Despite some gains by certain developing countries (mainly in Asia), industrial countries account for the greatest share of trade in services. More than 40 percent of services trade is carried out among the major traders: the United States, the EU, and Japan. Only a few developing countries rank among the biggest 16 exporters or importers of services (Karsenty, 2000). However, the picture radically changes if one looks at the importance of services exports in total exports of goods and services of particular countries (an indication of specialization). For example, for cross- border trade, all but one of the top 16 countries most specialized in services (i.e., countries for which services account for the largest share of total exports) are developing countries. The same is true for travel services and for the movement of persons, approximated by the compensation to employees. Exports of travel services, for instance, contribute to as much as 89 percent of export earnings of many small-island developing countries. Similarly, compensation to employees is a high source of export earning for several developing countries. 61. Unlike trade in goods, the liberalization of trade in services is at an early stage. The General Agreement on Trade in Services (GATS) established under the WTO provides a comprehensive framework for the liberalization of services (Appendix III). Initial - 39 - commitments under the Uruguay Round did not result in significant liberalization as most countries bound their existing regimes, although this enhanced the security of market access. Thus, in general, services trade is subject to substantial barriers in industrial and especially developing countries. Areas of particular interest to developing countries are those involving the movement of persons (Mode 4) where liberalization has been least. This includes the labor component of construction, distribution, and transport services; health services; and software services. A. Barriers to Trade in Services 62. Because of the nature of trade in services, trade restrictive measures in services differ in important ways from measures in goods (Box 7). First, border taxes equivalent to tariffs are difficult to impose on services imports because they are often not delivered across borders. Quotas, and other QRs, on the other hand, are pervasive. Second, services trade can be greatly affected by numerous internal policies that discriminate against foreign producers. These include measures that directly provide a cost advantage to domestic producers, such as subsidies, and other measures that impose a cost or create a competitive disadvantage for foreign producers (e.g., internal direct or indirect tax instruments). Third, because of the simultaneous nature of production and consumption of many services, restrictive measures that affect the movement of factors (labor or capital (FDI)) must also be considered as part of impediments to trade in services. B. Measuring Barriers to Trade in Services 63. Owing to the nature of trade in services and the impediments to such trade, measuring the barriers to services trade is a complicated task. Much like the measurement of NTMs for trade in goods, it is hard to come up with a simple measure that is comparable across modes of trade, sectors, and countries. Attempts to measure service trade barriers to market access have centered on (i) descriptive statistics or (ii) construction of “indices” of restrictiveness of service trade for various subsectors or countries. These measures generally show a strong positive correlation between the level of income of a country and the degree of openness to trade in services, as well as wide variations among sectors and modes of delivery (Figure 11).25 64. Most attempts to measure the barriers to market access in services start from the current schedule of commitments of WTO members, either in comparison with other members and/or with the highest possible number of such commitments (the most ideal or closest to unimpeded access). These are often referred to as frequency measures.26 It is important to 25 WTO (2001), Chapter IV. Data quoted in subsequent paragraphs of this section are taken from this study. 26 Appendix III summarizes the operations of the GATS and provides an explanation of frequency measures. - 40 - note that the absence of a commitment in a particular sector by a country in the GATS does not necessarily imply that the sector is not open to trade—it is similar to having unbound tariffs in the case of goods or binding tariffs at levels that are well above the applied rates. The outcome of the services negotiations resulted mainly in undertaking commitments based on the status quo, rather than increased liberalization. In general, industrial countries have made far more commitments than developing countries. 65. Among the 11 broadly defined services sectors by the WTO, some are far more open than others. The tourism sector, for example, has drawn the highest number of bindings: more than 90 percent of WTO members have included at least one subsector of tourism in their schedules (Figure 12). Financial and business services rank next, while health and education rank last with 48 and 46 entries, respectively. - 41 - Box 7. Barriers to Trade in Services The following types of barriers can be distinguished: (i) quotas, local content, and prohibitions; (ii) price- based instruments; (iii) standards, licensing, and procurement; and (iv) discriminatory access to distribution networks.1 • Quantitative restriction (QR) type policies are commonly applied to service providers and affect all four modes of services trade. On cross-border trade, they are most evident in the transport sectors. Foreign providers are either completely shut out (i.e., a zero quota) of certain segments, such as cabotage, or only provided limited access, as in international transport. In many countries, there are outright prohibitions directed against foreign providers of services such as domestic transportation, basic telecommunications, and legal, insurance, education, surveying, and investment advising services. On consumption abroad, quotas are sometimes implemented through foreign exchange restrictions, e.g., the ability of citizens to consume services, such as tourism and education, abroad is curtailed by limits on foreign exchange entitlements. On commercial presence, quotas are imposed on the number of foreign suppliers who are allowed to establish in sectors like telecommunications and banking. Quotas on foreign participation also take the form of restrictions on foreign equity ownership in individual enterprises. Finally, quotas are perhaps most stringent in the case of movement of service-providing personnel, and affect trade not only in professional services, but also in a variety of services that are intensive in the use of unskilled or semi- skilled labor such as construction. • Price-based barriers may take the form of visa fees and entry or exit taxes, discriminatory airline landing fees, and port taxes. Tariffs can be significant barriers to trade in goods that embody services (e.g., films, television programs, computer software) or goods that are used in producing services (e.g., computers, telecommunications equipment, advertising material). Further, many service sectors are subject to government-sanctioned or monitored price controls, examples include air transportation, financial services, and telecommunications. Government subsidies are commonly used in service sectors such as construction, communications, and road and rail transport. • Licensing or certification requirements may be imposed on foreign providers of professional and business services. In the absence of recognition measures, such requirements can discourage or prohibit foreign participation in the provision of services. Environmental standards may also affect service providers, particularly in transportation and tourism. Government procurement policies are often designed to favor domestic over foreign providers of services as well as goods by means of preference margins and outright prohibitions. • Discriminatory access to distribution and communications systems exist in many countries such as telecommunications, air transport, advertising, insurance, and dealer network. 1 Hoekman and Braga (1997) and Stern (2001). - 42 - Figure 11. Relationship Between Income Level and GATS Commitments 140 y = 0.0036x + 20.181 2 R = 0.6303 120 100 number of sub-sectors committed 80 60 40 Uruguay Round Participants 20 New WTO Member 0 0 5000 10000 15000 20000 25000 30000 1998 GNP per capita USD (purchasing power parity basis - PPP) Source: WTO (2001). Figure 12. Structure of WTO Members' Commitments by Sector, June 2000 Tourism 128 Financial 106 Business 103 Communication 99 Transport 84 Construction 74 Recreation 63 Environmental 54 Distribution 52 Health 48 Education 46 0 20 40 60 80 100 120 140 number of countries Source: WTO (2001). - 43 - 66. According to descriptive statistics, bindings undertaken under consumption abroad (Mode 2) are significantly more liberal than those for other modes, and bindings under presence of natural persons (Mode 4) are the least liberal. About 50 percent of consumption abroad (Mode 2) entries are without limitations, whereas there exist virtually no unlimited commitments under temporary movement of persons (Mode 4). This may explain the extremely small share of services trade delivered in the presence of natural persons (Mode 4) in total world trade in services mentioned above, although in absolute terms such trade is not insignificant, estimated at US$33 billion in 1998 (Maurer and others, 2001). 67. Quantitative measures of service trade barriers use other information, in addition to information in the schedule of commitments by countries, to obtain an index of the restrictiveness of services trade for various subsectors or countries. These measures confirm the findings of the descriptive statistics: transportation, storage, and communication services are estimated to have the highest trade barriers, and construction services the lowest.27 Communications and financial services have been found to be the most subject to FDI restrictions in APEC, while business, distribution, environmental, and recreational services are least restricted (Hardin and Holmes, 1997). These findings are consistent with a recent review of the literature suggesting that barriers to competition are highest in transportation, finance, and telecommunications. Notably, these are the basic “backbone” inputs that are crucial to the ability of enterprises to compete internationally. Policies towards these sectors appear to be more restrictive in developing countries than in high-income nations, which has implications for their ability to compete in global markets. 68. Liberalization of services, especially key infrastructure services like telecommunications, transport, and financial services, could be very beneficial to developing countries; inefficiencies in these sectors often add more to export costs than foreign trade barriers. Such liberalization could foster foreign investment in these sectors and an important gain could be a more efficient and robust financial sector (Tamirisa and others, 2000). However, successful domestic liberalization requires emphasis on introducing competition, not just changing ownership, and must be accompanied by an appropriate domestic regulatory framework that provides for legitimate social goals (such as universal service) and adequate prudential standards. According to recent studies in Latin America, for instance, countries that granted monopoly privileges of six to ten years in the provision of telecommunication services to the privatized state enterprises saw connections grow at 1.5 times the rate achieved under state monopolies but only half the rate in Chile, where the government retained the right to issue competing licenses at any time (Mattoo, 2000b). Another good example is provided by recent research that shows how increased competition 27 Hoekman (1995; 1996) uses frequency ratios (country commitments in relation to the maximum possible number of unrestricted commitments) and some benchmark tariff equivalents for individual services subsectors to assign an overall tariff value to each country and sector. Hoekman’s measures are designed to indicate the relative degree of restriction and do not indicate absolute ad valorem tariff equivalents. - 44 - in international maritime services could lead to a substantial reduction in transport costs (Fink and others, 2001). C. Conclusions 69. Despite the difficulties of measuring barriers to trade in services, it is apparent that services trade is subject to substantial barriers, both in industrial and in developing countries. In general, industrial countries have undertaken more commitments regarding access in their services sector than developing countries. Barriers in services trade vary in restrictiveness from one sector and/or mode of delivery to another, but are most restrictive for services which require the temporary cross-border movement of persons, an area of particular interest to developing countries. 70. There are likely to be significant gains worldwide if restrictions on services exports from developing countries are eliminated. With greater liberalization—particularly in the temporary movement of natural persons—many more developing countries could “export” a significant labor component of services such as construction, distribution, health and environmental services, transport, and software services. Developing countries themselves would also gain considerably from liberalizing their own services sectors (particularly telecommunications, transport, and financial services) within the context of an appropriate regulatory framework fostering competition. V. THE GAINS FROM LIBERALIZATION A. Gains from Liberalization in Merchandise Trade 71. Most estimates of potential gains from further liberalization of merchandise trade are calculated using computable general equilibrium (CGE) models.28 Because the effects of trade reform are economy-wide and multilateral liberalization has global implications, CGE models are appropriate tools for quantifying these effects. The Global Trade Analysis Project (GTAP) database has been most commonly used for these CGE studies because of its wide coverage of commodities and countries as well as frequent updating.29 The results of these models reflect two general principles derived from trade theory: 28 See Anderson and others (2000); Anderson and others (1999); Hertel (2000); Dessus and others (1999); Australian Department of Foreign Affairs and Trade (DFAT) (1999); Bradford (2000), Brown and others (2001), Dee and Hanslow (2000). For an overview of these studies, see Madani (2001). Assessments of the results of general equilibrium models can be found in Francois (1999) and Whalley (1999). 29 The GTAP database is part of a project initiated in 1992 to develop a database for multicountry CGE analysis of international economic issues. It is coordinated at Purdue University and is documented in McDougall and others (1998). - 45 - • On average, countries benefit most from liberalizing the most protected sectors of their own economies; these welfare gains increase roughly with the square of the initial restrictiveness of their trade barriers. • In the short term, the static gains from liberalization, as measured by changes in producer and consumer surplus or similar indicators, are usually a small fraction of GDP—given the current status of barriers. Over the medium and long term, as factors of production adjust to international prices and new investment takes place, much higher gains result. These gains are derived from increases in productivity and economies of scale that result from new investments. 72. Given that trade barriers remain high in some sectors (e.g., agriculture, and textiles and clothing) in industrial countries, and they are generally high in most sectors in developing countries, there are substantial benefits to be gained from further liberalization of merchandise trade. Estimates for the static medium-term welfare gains from liberalizing all trade range from US$250 billion to US$550 billion; one-third to two-fifths of these gains would accrue to developing countries, well in excess of annual aid flows to these countries. The lower estimates are generally from models of perfect competition and constant returns to scale (such as the standard GTAP model), while the higher estimates are from models that incorporate economies of scale and monopolistic competition. Estimates that attempt to incorporate dynamic productivity effects and services can be much larger. 73. Because their economies are more highly protected, most studies find that developing countries gain more as a percentage of their GDP/GNP from liberalization than industrial countries. In absolute terms, however, industrial countries command a larger share of the global gain. According to some estimates, in the short run, some countries may suffer welfare losses from liberalization. In the medium and long term, however, all developing countries gain as the dynamic effects of trade liberalization counterbalance the short-term losses (Dessus and others, 1999). 74. If only static gains are considered, agricultural liberalization generates the largest welfare gains. This is because the level of protection is much higher in agriculture for both industrial and developing countries. Once the dynamic effects are taken into account, manufacturing liberalization generates larger gains than agricultural liberalization because of the greater scope for productivity growth and innovations from trade in manufactures.30 75. A study by Anderson and others (2000) estimates the gains from global trade liberalization for major country groups and commodities (Table 9). The results show that each country group gains the most from reducing its own trade barriers. Industrial countries gain most from liberalization of their most protected sector: agriculture. An important component of agricultural protection in industrial countries is in the form of producer 30 Yang (1997); Dessus and others (1999). - 46 - subsidies which, by depressing international agricultural prices, subsidize net importers of these products, particularly among the developing countries. In the short run, the elimination of these subsidies will have a negative effect on the welfare of net food importers as international agricultural prices rise. This effect partly offsets the gains to net food exporters; hence the overall benefits for developing countries of agricultural liberalization in industrial countries are relatively small (US$12 billion).31 Indeed, as a group, developing countries gain more from industrial country liberalization in manufactures. Table 9. Welfare Gains from Post-Uruguay Round Trade Liberalization 1 (in billions of 1995 US$) Type of goods Other Textiles Liberalizing Agricultural primary and Other region Benefiting region goods goods clothing Manufactures Total Industrial countries Industrial countries 110.5 0.0 -5.7 -8.1 96.6 Developing countries 11.6 0.1 9.0 22.3 43.1 All countries 122.1 0.0 3.3 14.2 139.7 Developing countries Industrial countries 11.2 0.2 10.5 27.7 49.6 Developing countries 31.4 2.5 3.6 27.6 65.1 All countries 42.6 2.7 14.1 55.3 114.7 All countries Industrial countries 121.7 0.1 4.8 19.6 146.2 Developing countries 43.0 2.7 12.6 49.9 108.1 All countries 164.7 2.8 17.4 69.5 254.3 Source: Anderson and others (2000). 1 Gains after full implemention of Uruguay Round in 2005 from elimination of all trade barriers. 76. Developing countries also gain most from their own liberalization, divided about equally between agriculture and manufacturing since they have relatively high levels of protection in both sectors. Industrial countries gain most from developing country liberalization in manufactures since the bulk of their trade with developing countries is in these products. 31 Some Caribbean Island producers of bananas and sugar, which benefit from EU preferential access, also stand to lose as a result of agricultural liberalization. - 47 - 77. Within the developing countries, a distinction can be made between high-income manufacturing exporters and low-income agriculture and commodity exporters. The low- income countries (a group that includes the LDCs) gain most from agricultural liberalization in industrial countries because of the greater relative importance of agriculture in their economies (Table 10). They also gain substantially from liberalization of both agriculture and industrial goods in developing country markets. In contrast, the larger and more advanced developing countries (“Other developing countries”) gain most from liberalization in industrial goods because of the greater importance of manufacturing in their production and exports. Table 10. Welfare Effects of Trade Liberalization on Developing Countries (in billions of 1995 US$) Agricultural goods Industrial goods Industrial Developing Industrial Developing countries countries countries countries All developing countries 11.6 31.4 31.5 33.7 of which: Low-income countries 4.8 4.4 1.7 4.5 South African Customs Union 0.7 0.4 0.1 0.1 Other Sub-Saharan African countries 1.5 1.0 0.2 0.5 Rest of World 1 2.6 3.0 1.3 3.9 Other developing countries 6.8 27.0 29.8 29.2 Source: Anderson and others (2000). 1 Includes Afghanistan , Andorra, Bangladesh , Bhutan , Bosnia & Herzegovina, Brunei Darussalam, Cambodia, Croatia, Cyprus, Fiji, Kiribati, Democratic People's Republic of Korea, Lao P.D.R., Liechtenstein, Macedonia FYR, Maldives , Malta, Monaco, Mongolia, Myanmar, Nauru, Nepal , Pakistan, Papua New Guinea, Samoa , San Marino, Solomon Islands, Sri Lanka, Tonga, Tuvalu, Vanuatu, Vietnam, Federal Republic of Yugoslavia. (Least developed countries are in italics.) 78. Most CGE studies focus on the impact of multilateral liberalization. There are, however, specific sectors and trade policies that are of special interest to developing countries. MFA quotas in industrial countries on textile and clothing imports from developing countries impose substantial costs for both developing and industrial countries. Total removal of MFA quotas is estimated to improve developing country welfare by US$13–22 billion.32 Global trade in textiles and clothing may increase as much as 34-60 percent once MFA quotas are completely eliminated (Deardorff, 1994). As for the benefits for industrial countries of the removal of MFA quotas, a recent estimate puts the gain of MFA phase out at about US$14 billion per year for the EU, or approximately US$297 for a family of four (Francois and others, 2000). 32 Anderson and others (2000) and DFAT (1999). - 48 - 79. The EU’s Common Agricultural Policy (CAP) is also of special interest to developing countries, as well as to industrial countries. A recent study by Borrell and Hubbard (2000) estimates the cost of the CAP to be US$75 billion a year, of which one-third is borne by the rest of the world. Developing countries bear the main cost of these policies outside the EU (around US$20 billion per year) and hence would receive the greatest gains (outside the EU) from its reform. These results underestimate the potential benefits for developing countries of CAP reform, since CAP reform would encourage agricultural investment and hence long-term growth of agriculture in developing countries. 80. Particular importance has also been given to providing free market access to the poorest developing countries in industrial country markets. As mentioned in Section III, preferential schemes currently in place are still affected by the existence of tariff peaks and tariff escalation, and they are subject to various conditions that impair the security of market access. As an alternative, recent initiatives have proposed completely unrestricted access for the LDCs into industrial country markets. Hoekman and others (2000) estimate that full duty- and quota-free access for LDCs in the Quad markets would boost their exports by US$3 billion (11 percent), and that LDC exports of tariff peak products would increase by 75 percent. A related study by Ianchovichina and others (2000) shows that a subset of 37 mostly low-income SSA countries would be able to increase its non-oil exports by 14 percent, or US$2.5 billion, if Quad countries provided duty- and quota-free access to their markets. This would boost the income of these countries by around 1 percent. The same study estimates that the EU’s “Everything But Arms” initiative will increase the non-oil exports of SSA by 3 percent. In both cases, these increases in exports would occur at only a very small aggregate cost to industrial countries and to other developing countries, whose exports are much larger and more diverse. Even so, specific industries in individual countries could be adversely affected by such preferences, which underscores the need for complementary multilateral liberalization. B. Gains from Liberalization in Services 81. Restrictions on trade in services, as on trade in goods, reduce the level of real income. As in the case of goods, many services are crucial inputs into production and inefficient production of such services acts as a tax on the production of other goods and services. Services industries often exhibit economies of scale, and hence trade restrictions together with other government regulations tend to create monopolistic or oligopolistic market structures when domestic markets are small. Service liberalization thus increases competition and reduces costs, and the resulting welfare effects can be large. Indeed, some studies have shown that liberalizing core services such as communication and transportation in developing countries can result in cost savings that far outstrip the benefits of reductions in trade barriers in industrial country markets (Fink and others, 2001). 82. The static effect of removing barriers to trade in services is much like that of trade in goods: liberalization leads to a more efficient use of resources and is thus welfare enhancing. The dynamic (growth) effect of removing services trade barriers is likely to - 49 - be far more important than static gains because increased competition will accelerate productivity growth. Moreover, because services liberalization often entails the movement of factors of production, removing restrictions on services trade is likely to augment a country’s stock of both human and physical capital (through increased FDI) and technology that is embodied in or associated with such FDI.33 The positive impact of services trade liberalization on growth is supported by a recent study, which found a strong and statistically significant relationship between financial and telecommunication services liberalization and economic growth (Mattoo and others, 2001). 83. A number of studies have estimated the welfare effects of services trade liberalization, finding significant gains in both industrial and developing countries. One study (Brown and Stern, 2001) estimates that removing barriers to the establishment of foreign firms providing services (only partial liberalization including Mode 3 services— delivered through commercial presence) results in welfare gains of US$42 billion. A larger part of this gain comes from the indirect effect on capital formation of services liberalization than from the direct static effect of lowering barriers. Other studies estimate welfare gains of US$55 billion (Hertel, 2000), US$133 billion (Dee and Hanslow, 1999), and US$500 billion (DFAT, 1999). The substantial differences in the results of these studies are due to differences in model structure, liberalization scenarios, estimates of the barriers to services trade, and the way these barriers are modeled (OECD, 2000b). C. Conclusions 84. From the above review, the following priorities for trade liberalization can be derived: for industrial countries, the main benefits would come from liberalization of agriculture. This will also be of interest to low-income countries (especially LDCs) and others that are agricultural exporters. For most developing countries, the greatest gains come from liberalization of manufactures, although significant gains also result from liberalizing agriculture. In the longer term, the greatest gains from liberalization of trade will come in the manufacturing sector in both industrial and developing countries. The acceleration of the removal of MFA quotas and the reduction of tariff peaks and escalation in agriculture and manufacturing (including textiles and clothing) would facilitate the participation of developing countries (especially LDCs) in the expansion of world trade. In this connection, building on the EU initiative by providing duty- and quota-free access for the exports of the poorest countries to all markets would provide significant benefits to these countries at little aggregate cost to the rest of the world. As complementary multilateral liberalization proceeds, these preferences, and any associated distortions, would diminish. 33 However, if FDI comes simply because the returns to investment are artificially raised by restrictions on competition, much as in the case of immiserizing growth literature, the net returns to the host country may be negative. - 50 - 85. The empirical studies reviewed most likely underestimate the benefits of liberalization. Most welfare estimates include only short- or medium-term, static gains from trade liberalization. Even those that do consider dynamic effects are quite rudimentary and incomplete as these effects are difficult to identify and quantify. However, most analysts agree that the dynamic gains from trade liberalization—involving investment, the transfer of technology and knowledge, and favorable changes in the business environment—far outstrip the static gains. In addition, most studies only evaluate the effects of tariff cuts, since information about NTMs is more difficult to obtain and quantify. Perhaps the most difficult benefits to quantify are those that stem from the strengthening of rules governing international trade. For these reasons, the overall gains from trade liberalization are most likely substantially larger than the estimates cited above. 86. In the area of services, the studies reviewed are only indicative of potential gains of services trade because the modeling of services liberalization is still very much in its infancy. More comprehensive and reliable estimates of trade barriers, and consensus on the best way of quantifying trade barriers in services, as well as evolving modeling techniques, will substantially improve these estimates over time. 87. Finally, the studies reviewed do not consider the transitional costs that will inevitably occur as a result of structural change. In practice, these costs can be mitigated to some extent by the design and phasing of liberalization measures. Beyond this, adequate social assistance programs would be needed to support adjustment. The revenue implications of trade liberalization also need to be addressed through appropriate design and phasing of liberalization and measures to move toward broad-based domestic taxes (Ebrill and others, 1999). - 51 - APPENDIX I Appendix Table 1. Share of Manufactures in Total Merchandise Exports1 (in percent) Country /Region Share Japan 98.1 Taiwan Province of China 96.3 Singapore 96.0 Hong Kong SAR 95.9 Republic of Korea 94.4 Sweden 94.3 Finland 92.7 Austria 92.2 Italy 92.2 Germany 92.1 Portugal 91.4 China 90.7 Philippines 89.3 Bangladesh 89.3 Rest of Central European Association 89.0 Rest of South Asia 87.7 Switzerland 87.5 United States 86.8 Belgium/Luxembourg 86.3 Sri Lanka 85.9 United Kingdom 85.3 France 84.2 Malaysia 84.1 Thailand 83.2 Hungary 83.2 Mexico 81.7 Ireland 81.6 Spain 81.4 World Average 81.2 Turkey 79.8 Poland 77.4 India 76.8 Canada 76.5 Netherlands 73.6 Morocco 69.3 Denmark 68.4 Rest of World 62.3 Indonesia 62.1 Greece 61.8 Central America, Caribbean 61.0 Brazil 59.1 Vietnam 56.8 Uruguay 47.4 Rest of SACU 47.1 Baltics, Russia, and other countries of the former Soviet Union 44.6 Rest of North Africa 44.2 Argentina 39.7 República Bolivariana de Venezuela 37.7 New Zealand 36.3 Rest of EFTA 34.5 Colombia 33.6 Australia 32.1 Rest of Middle East 31.8 Zimbabwe 31.2 Other Southern Africa 25.5 Chile 24.3 Mozambique 19.6 Peru 18.3 Rest of Sub-Saharan Africa 17.3 Rest of South America 14.9 Rest of Andean Pact 14.1 Tanzania 11.8 Malawi 9.8 Zambia 9.7 Uganda 1.2 Source: GTAP 5 database. 1 This table presents information on the shares of manufactures in developing country exports for the 65 countries identified in the GTAP 5 database for the year 1997. - 52 - APPENDIX I Appendix Table 2. Post Uruguay Round Bound Tariffs on Agricultural and Industrial Products Agricultural products Industrial products Percent of Simple Standard Percent of duties Simple Standard Percent of 1 1 average deviation duties > 15% >100% average deviation duties > 15% North America Canada 4.6 4.7 1.2 0.0 5.3 5.2 7.2 United States 5.5 5.5 2.6 0.0 3.8 4.2 2.0 Latin America Argentina 32.8 5.9 97.3 0.0 30.6 6.3 98.5 Brazil 35.3 10.1 96.4 0.0 29.7 6.9 97.3 Colombia 88.3 32.3 100.0 22.9 36.1 9.8 100.0 Mexico 42.9 35.2 96.2 4.9 34.8 3.1 99.6 Venezuela 55.4 33.3 99.4 14.5 33.8 4.2 99.4 Western Europe EU-15 19.5 22.1 33.9 0.9 4.1 3.6 0.6 Iceland 48.4 85.7 58.9 8.1 10.0 12.1 30.0 Norway 123.7 147.6 61.1 44.2 3.4 5.5 0.2 Switzerland 51.1 96.1 16.5 6.9 1.9 3.4 0.3 Turkey 63.9 55.4 86.7 16.8 40.7 34.2 77.3 Eastern Europe Czech Republic 13.3 19.2 23.4 0.9 4.5 3.3 1.0 Hungary 22.2 19.1 47.5 0.6 6.8 4.0 1.4 Poland 52.8 44.6 74.7 8.1 10.6 5.2 12.9 Romania 98.6 84.0 97.5 39.6 34.4 7.1 98.9 Asia/Pacific Australia 3.3 4.6 3.0 0.0 10.6 10.8 15.9 Bangladesh 83.8 25.9 … 69.0 83.5 23.0 … India 124.3 51.0 99.4 44.7 59.0 33.5 97.9 Indonesia 47.2 21.7 98.3 3.2 38.6 8.6 98.9 Japan 11.7 12.6 17.5 0.3 3.6 3.8 0.6 Republic of Korea 62.2 108.8 74.1 10.1 11.4 9.0 18.9 Malaysia 13.6 20.5 18.9 1.5 16.4 10.7 41.8 New Zealand 8.7 7.3 12.6 0.1 13.8 14.7 33.9 Philippines 35.3 10.9 91.8 0.0 25.1 10.2 88.3 Sri Lanka 50.0 0.0 100.0 0.0 50.0 0.0 100.0 Thailand 34.6 22.1 91.4 2.1 28.4 9.6 89.1 Africa Tunisia 116.7 40.4 100.0 50.3 41.2 16.7 100.0 Source: WTO (2001). 1 Post-Uruguay Round bound rate. - 53 - APPENDIX II THE IMPLEMENTATION OF THE AGREEMENT ON TEXTILES AND CLOTHING (ATC) Under the ATC, WTO members that maintain import restrictions on textiles and clothing must progressively reduce the restrictions and integrate the products concerned into the body of GATT rules. This process is to be carried out progressively in four stages with all products integrated at the end of a ten-year transition period (2005). The first stage began on January 1, 1995 with the integration of products representing at least 16 percent of the volume of the member's total imports in 1990. At stage 2, on January 1, 1998 at least a further 17 percent was integrated. At stage 3, on January 1, 2002 not less than a further 18 percent will be integrated. Finally, at the end, on January 1, 2005 all remaining products (amounting to up to 49 percent of 1990 imports) will be integrated. While the legal commitments of the ATC have been met by most members, there has been criticism of the way the agreement has been implemented. For each stage of implementation, it is up to the discretion of the importing countries maintaining quotas to select the particular products that will be integrated to reach the above-mentioned thresholds. The only constraint is that the integration list must encompass products from each of the four groupings: (i) tops and yarns, (ii) fabrics, (iii) made-up textile products, and (iv) clothing. It is significant that in the higher value-added groupings, particularly clothing, the selection to date can be described as minimal. For example, for the United States, the percentage by volume of 1990 imports of clothing imports integrated during the first two stages amounts to only 3.9 percent.34 The situation is comparable for the EU and Canada. Another indication of the slow implementation of the integration process is that the number of quotas actually eliminated during the first two stages by the United States and the EU are 2 out of 750, and 14 out of 219, respectively (compared to the total number of MFA and other quotas at end-1994). As a result, although a total of 33 percent of trade by volume has been integrated by stage 2, the actual amount liberalized has been much lower. Less than 7 percent by volume and value per annum of restrained trade in the United States, and less than 5 percent by volume and value per annum of restrained trade in the EU, has been freed of quotas. This is a result of the nature of the product coverage, which allowed importers to “free” items not actually under restraint. 34 See AITIC (1999). - 54 - APPENDIX III THE GATS RULES ON MEASURES AFFECTING SERVICES TRADE35 The GATS rules operate at two levels. First, there is a set of general rules that apply across the board to measures affecting trade in services, and then there is a set of sector-specific commitments that determine the extent of liberalization undertaken by individual countries. The specific commitments on market access and national treatment are the core of the GATS, and the impact of the Agreement depends to a large extent on the commitments made by Members. Article XVI stipulates that measures restrictive of market access which a WTO Member cannot maintain or adopt, unless specified in its schedule, include limitations on: (a) the number of service suppliers; (b) the total value of services transactions or assets; (c) the total number of services operations or the total quantity of service output; (d) the total number of natural persons that may be employed in a particular sector; (e) specific types of legal entity through which a service can be supplied; and (f) foreign equity participation (e.g., maximum equity participation). With the exception of (e), the measures covered by Article XVI all take the form of QRs. Three aspects of Article XVI are important. First, the Article XVI list does not include all measures which could restrict market access. Perhaps most significantly, fiscal measures are not covered. Thus, a Member could maintain, without being obliged to schedule, a high nondiscriminatory tax on a particular service which severely limits market access. Secondly, Article XVI has been interpreted to cover both discriminatory and nondiscriminatory measures, that is, measures of the type “only five new foreign banks will be granted licenses” and also measures such as “only ten new [foreign and domestic] banks will be granted licenses.” Finally, the limitations must be read as “minimum guarantees” rather than “maximum quotas,” that is, a country which has promised to allow five foreign banks entry is free to grant entry to more than five. Article XVII:1 states the basic national treatment obligation: “In the sectors inscribed in its Schedule, and subject to any conditions and qualifications set out therein, each Member shall accord to services and service suppliers of any other Member, in respect of all measures affecting the supply of services, treatment no less favorable than that it accords to its own like services and service suppliers.” Unlike Article XVI, Article XVII provides no exhaustive list of measures inconsistent with national treatment. Nevertheless, Article XVII:2 makes it clear that limitations on national treatment cover cases of both de jure and de facto discrimination. The Explanatory Note 35 Mattoo (2000a). - 55 - APPENDIX III provides two examples of limitations on national treatment. If domestic suppliers of audiovisual services are given preference in the allocation of frequencies for transmission within the national territory, such a measure discriminates explicitly on the basis of origin of the service supplier and thus constitutes formal or de jure denial of national treatment. Alternatively, consider a measure stipulating that prior residency is required for the issuing of a license to supply a service. Although the measure does not formally distinguish service suppliers on the basis of national origin, it de facto offers less favorable treatment for foreign suppliers because they are less likely to be able to meet a prior residency requirement than like service suppliers of national origin. A Member's specific commitments can be seen as the outcome of a two-step decision. Each Member first decides which service sectors will be subject to the GATS market access and national treatment disciplines. It then decides what measures will be kept in place for that sector which violate market access and/or national treatment respectively. Commitments on both market access and national treatment have been specified by modes of supply. The GATS schedules of commitments are structured in the following manner. In the left hand column of the table are inscribed the service activities which are the subject of specific commitments. For each of the four modes of supply noted in column two, columns three and four state whether there are limitations on market access and national treatment respectively. The extreme right hand column provides for the additional commitments on other measures affecting trade in services. Entries in the schedule in a given sector with respect to a particular mode of supply fall into one of four categories: (i) Full commitment: “none” or “no limitations,” which implies that the Member does not seek in any way to limit market access or national treatment through measures inconsistent with Articles XVI or XVII; (ii) Commitment with limitations: the Member describes in detail the measures maintained which are inconsistent with market access or national treatment, and implicitly commits itself to take no other inconsistent measures; (iii) No commitment: “unbound” indicates that the Member remains free to maintain or introduce measures inconsistent with market access or national treatment; (iv) No commitment technically feasible: “unbound*” indicates that in the sector in question, a particular mode of supply cannot be used, for instance cross-border supply of hairdressing services. - 56 - Bibliography Agency for International Trade Information and Cooperation (AITIC), 1999, “Textiles and Clothing: Implications for Less-Advantaged Developing Countries,” (Geneva). Anderson, K., and Y. Hayami, 1986, The Political Economy of Agricultural Protection: East Asia in International Perspective (Sydney: Allen and Unwin) Anderson, K., B. Hoekman, and A. Strutt, 1999, “Agriculture and the WTO: Next Steps,” revision of a paper presented at the Second Annual Conference on Global Economic Analysis, Avernaes Conference Centre, Helnaes, Denmark, June. Anderson, K., J. Francois, J., T. Hertel, B. Hoekman, and W. Martin, 2000, “Potential Gains from Trade Reform in the New Millennium,” paper presented at the Third Annual Conference on Global Economic Analysis, Monash University, Mt. Eliza, Australia, June. Bagchi, S., 1994, ‘The Integration of the Textile Trade into GATT,” Journal of World Trade, Vol. 28 (June), pp. 31–42. Baldwin, R., 2000, “Regulatory Protectionism, Developing Nations and a Two-Tier World Trade System,” CEPR Discussion Paper 2574 (London: Centre for Economic Policy Research). Blackhurst, R., A. Enders, and J. Francois, 1996, “The Uruguay Round and Market Access: Opportunities and Challenges for Developing Countries,” in The Uruguay Round and the Developing Countries, ed. by W. Martin and A. Winters (New York: Cambridge University Press). Borrell, B. and Hubbard, L., 2000, “Global Economic Effects of the EU Common Agricultural Policy,” Economic Affairs, Vol. 20 (2), pp. 18–26. Bradford, S., 2000, “Paying the Price: The Welfare Effects of Protection in OECD Countries” (Provo, Utah: Brigham Young University). Brown, D., A. V. Deardorff, and R.M. Stern, 2001, “CGE Modeling and Analysis of Multilateral and Regional Negotiating Options,” Department of Economics Discussion Paper 468 (Ann Arbor: University of Michigan Press). Deardorff, A.V., 1994. “Economic Effects of Tariff and Quota Reductions,” in The New GATT: Implications for the United States, ed. by Susan M. Collin and Barry P. Bosworth (Washington: Brookings Institute). - 57 - ——, 2001, “Fragmentation Across Cones” in Fragmentation: New Production Patterns in the World Economy, ed. by Sven W. Arndt and Henryk Kierzkowski (New York: Oxford University Press). Dee, P. and K. Hanslow, 2000. “Multilateral Liberalization of Services Trade,” Staff Research Paper, Productivity Commission, Canberra, March. Department of Foreign Affairs and Trade (DFAT), Australia, 1999, Global Trade Reform: Maintaining Momentum (Canberra). Dessus, S., K. Fukasaku, and Safadi, R., 1999, “Multilateral Tariff Liberalization and the Developing Countries: A Global Policy Simulation” (Paris: Organization for Economic and Cooperation and Development). Ebrill, Liam P., Janet Stotsky, and Reint Gropp, 1999, Revenue Implications of Trade Liberalization, IMF Occasional Paper No. 180 (Washington: International Monetary Fund). Findley, C., and T. Warren, 2000, “Measuring Impediments to Trade in Services,” GATS 2000, New Directions in Services Liberalization, ed. by P. Sauvé and R.M. Stern (Washington, D.C.: Brookings Institution). Finger, J.M., 1993, Antidumping: How It Works and Who Gets Hurt (Ann Arbor: Michigan University Press). ——, and L. Schuknecht, 1999, “Market Access Advances and Retreats: The Uruguay Round and Beyond,” Policy Research Working Paper 2232, Development Research Group, Trade (Washington, D.C.: World Bank). Finger, J.M., F. Ng, and S. Wangchuk, 2000, “Antidumping as Safeguard Policy,” presentation at University of Michigan Conference on Issues and Options for the Multilateral, Regional and Bilateral Trade Policies of the United States and Japan, October. Fink, C., A. Mattoo, and I.C. Neagu, 2001, “Trade in International Maritime Services: How Much Does Policy Matter?”Available via Internet: http://www1.worldbank.org. Francois, J., 1998, “A Gravity Approach to Measuring Services Protection” (Rotterdam: Erasmus University). ——, 1999, “Assessing the Results of General Equilibrium Studies of Multilateral Trade Negotiations,” Tinbergen Institute and CEPR (Rotterdam: Erasmus University). - 58 - ——, H. Glismann, and D. Spinanger, 2000, “The Cost of EU Trade Protection in Textiles and Clothing,” Tinbergen Institute (Rotterdam: Erasmus University). General Agreement on Trade and Tariffs, 1993, “An Analysis of the Proposed Uruguay Round Agreements, with Particular Emphasis on Aspects of Interest to Developing Countries,” MTN.TNC/W/122, MTN.GNG/W/30, November. Gehlhar, M., 1998, “Reconciling Bilateral Trade Data for Use in GTAP,” in Global Trade Assistance and Protection: The GTAP 4 Data Base, ed. by R.A. McDougall, A. Elbehri, and T.P. Truong, Center for Global Trade Analysis (West Lafayette, Indiana: Purdue University). Gehlhar, M., T. Hertel, and W. Martin, 1994, “Economic Growth and the Changing Structure of Trade and Production in the Pacific Rim,” American Journal of Agricultural Economics, Vol. 76 (December), pp. 1101–10. Hardin, Alexis, and Leanne Holmes, 1997, “Services Trade and Foreign Direct Investment,” Staff Research Paper, Industry Commission (Canberra: Australian Government Publishing Services). Henson, S.J., R. J. Loader, A. Swinbank, M. Bredahl, and N. Lux, 1999, “Impact of Sanitary and Phytosanitary Measures on Developing Countries” (London: Department for International Development). Henson, S., and others, 2000,“Impact of Sanitary and Phytosanitary Measures on Developing Countries,” (Berkshire, England: University of Reading). Hertel, T. and W. Martin, W., 2000a, “Liberalising Agriculture and Manufactures in a Millennium Round: Implications for Developing Countries,” The World Economy, Vol. 23 (April), pp. 455–70. Hertel, T., 2000, “Potential Gains from Reducing Trade Barriers in Manufacturing, Services and Agriculture,” Federal Reserve Bank of St. Louis, July/August. Hertel, T., B. Hoekman, and W. Martin, 2001, “Developing Countries and a New Round of WTO Negotiations,” mimeo (Washington, D.C.: World Bank). Hoekman, Bernard, 1995, “Assessing the General Agreement on Trade in Services,” The Uruguay Round and the Developing Countries, ed. by Will Martin and L. Alan Winters, World Bank Discussion Paper No. 307 (Washington, D.C.: The World Bank). ––—, and Carlos A. Primo Braga, 1997, “Protection and Trade in Services: A Survey,” Open Economies Review, Vol. 8, pp. 285-308 - 59 - ––—, F. Ng, and M. Olarreaga, 2000, “Tariff Peaks in the Quad and Developing Countries’ Exports,” Development Research Group (Washington, D.C.: World Bank). Ianchovichina, E., A. Mattoo, and M. Olarreago, 2000, “Unrestricted Market Access for Sub- Saharan Africa: How Much is it Worth and Who Pays?” Development Research Group (Washington, D.C.: World Bank). Karsenty, G., 2000, “Assessing Trade in Services by Mode of Supply,” GATS 2000, New Directions in Services Liberalization, ed. by P. Sauvé and R. M. Stern (Washington, D.C.: Brookings Institution). Keesing, D. and M. Wolf, 1980, “Textile Quotas Against Developing Countries,” Thames Essay No. 23 (London: Trade Policy Research Centre). Kirmani, Naheed, and others, 1994, International Trade Policies: The Uruguay Round and Beyond, Volume I, Principal Issues, and Volume II, Background Papers, World Economic and Financial Surveys (Washington: International Monetary Fund). Lindland, J., 1997, “The Impact of the Uruguay Round on Tariff Escalation in Agricultural Products,” United Nations Food and Agriculture Organization (Rome). Madani, D., 2001, “Effect of Multilateral Liberalization—Assessing Results from a Select Number of CGE Models,” mimeo (Washington, D.C.: World Bank). Maskus, K., J. Wilson, and T. Otsuki, 2000, “Quantifying the Impact of Technical Barriers to Trade,” Policy Research Working Paper 2512, Development Research Group (Washington, D.C.: World Bank). Mattoo, Aaditya, 2000a, “Trade in Services: Economics and Law,” mimeo (World Bank: Washington, D.C.). ——, 2000b, “Developing Countries in the New Round of GATS Negotiations: Towards a Pro-Active Role,” The World Economy, Vol. 23 (April), pp. 471–89. Maurer, A., and P. Chauvet, 2001, “The Magnitude of Global-Trade-in-Services Flows,” in Trade Policy for Developing Countries in A Global Economy: A Handbook, preliminary draft, (Washington, D.C.: World Bank). McDougall, R.A., Elbehri, A. and Truong, T.P., eds., 1998. Global Trade Assistance and Protection: The GTAP 4 Database, Center for Global Trade Analysis (West Lafayette, Indiana: Purdue University). - 60 - Messerlin, P., and J. Zarrouk, 1999, “Trade Facilitation: Technical Regulations and Customs Procedures,” background paper presented the World Trade Organization and World Bank Conference on Developing Countries in a Millennium Round, September. Michalopoulos, C., 1999, “Trade Policy and Market Access Issues for Developing Countries: Implications for the Millennium Round,” Policy Research Working Group Paper 2214 (Washington, D.C.: World Bank). Nicita, A., and M. Olarreaga, 2000, “Exports and Information Spillovers,” World Bank Policy Research Working Paper 2474 (Washington, D.C.: World Bank). Organization for Economic Cooperation and Development, 1997, Market Access for the Least-Developed Countries: Where Are the Obstacles? (Paris). ——, 2000a, Agricultural Policies in OECD Countries (Paris). ——, 2000b, “Quantification of Costs to National Welfare from Barriers to Services Trade: A Literature Review,” TD/TC/WP(2000)24 (Paris). ——, 2001, Trade and Development Issues in Non-OECD Countries (Paris). Otsuki, T., J. Wilson, and M. Sewadeh, 2000, “Saving Two in a Billion: A Case Study to Quantify the Trade Effect of European Food Safety Standards on African Exports,” mimeo (Washington, D.C.: World Bank). Sharer, Robert, and others, 1998, Trade Liberalization in IMF-Supported Programs, World Economic and Financial Surveys (Washington: International Monetary Fund). Spinanger, D., 1999, “ Faking Liberalization and Finagling Protectionism: the ATC at its Best,” (Kiel: Kiel Institute of World Economics). Staples, B. R., 1998, Trade Facilitation, Global Trade Negotiations Home Page. Stephenson, S., 1997, “Standards and Conformity Assessment as Nontariff Barriers to Trade,” Policy Research Working Group Paper 1826 (Washington, D.C.: World Bank). Stern, R.M., 2001, “Quantifying Barriers to Trade in Services,” forthcoming. Tamirisa, Natalia, and others, 2000, “Trade Policy in Financial Services,” IMF Working Paper 00/31 (Washington: International Monetary Fund). - 61 - United Nations Economic and Social Comission for Asia and the Pacific, 1996, Implications for the Phasing-Out of the MFA for the ESCAP Region (Bangkok). United Nations Conference on Trade and Development, 1995, An Analysis of Trading Opportunities Resulting from the Uruguay Round in Selected Sectors: Agriculture, Textiles and Clothing, and Other Industrial Products (Geneva). Whalley, J., 1999, “Quantifying the Impact of WTO Negotiating Rounds: Lessons for the Millennium Round and the Developing Countries from the Uruguay Round Models,” (Coventry, England: University of Warwick). World Bank, World Development Indicators 2000 (Washington, D.C.: World Bank). ——, Global Economic Prospects, 2000 (Washington, D.C.: World Bank). ——, Global Economic Prospects and the Developing Countries, 2001 (Washington, D.C.: World Bank). World Trade Organization, 1998, Document S/C/W/60 (Geneva). ——, 2001, “Market Access: The WTO’s Unfinished Business: Post-Uruguay Round Inventory and Issues,” (Geneva). Yang, Y., 1996, “Safeguarding the Future of World Trade in Textiles and Clothing,” paper presented at the Symposium on Antidumping and Safeguard Systems, Taipei, February. ——, 1997, “A General Equilibrium Assessment of the Uruguay Round with Trade-Related Externalities,” in East Asian Trade After the Uruguay Round ed. by D. Robertson (Cambridge, England: Cambridge University Press). ——, W. Martin, W. and K. Yanagishima, 1997, “Evaluating the benefits of MFA Liberalization in the Uruguay Round,” in Global Trade Analysis, using the GTAP Model, ed. by T. Hertel (Cambridge, England: Cambridge University Press).