WPS6463 Policy Research Working Paper 6463 Developing Country Trade Policies and Market Access Issues 1990–2012 Constantine Michalopoulos Francis Ng The World Bank Development Research Group Trade and Integration Team May 2013 Policy Research Working Paper 6463 Abstract The study presents a comprehensive review of developing nation and thus pose a potential threat to the multilateral country trade policies and market access issues as they system and a potential stimulus to further multilateral evolved over the period 1990–2012. The main findings collaboration. Fifth, sanitary and phytosanitary and are, first, that applied tariffs as well as traditional core technical barriers to trade are being increasingly used non-tariff measures have declined significantly over by both developed and developing countries but their time in both developed and developing countries. protective intent is difficult to gauge. There is a need for Second, the instruments of protection used by developed increased vigilance, transparency, and reporting to ensure and developing countries are becoming increasingly that they are not being used as a means of protection of similar: trade remedies, especially anti-dumping are economic interests. Sixth, the service sectors are the most the instruments of choice for all except low-income promising area where efforts for further liberalization developing countries. Third, agriculture is the main are needed and may produce significant benefits. And sector where developing countries face access problems in seventh, far less additional protection has been put in OECD markets. Fourth, regional and other preferential place following the 2008 financial crisis compared with trade agreements are both a result and a cause of the lack what had been feared or what had happened in earlier of progress in multilateral trade negotiations. They violate crises. the basic World Trade Organization tenet of most favored This paper is a product of the Trade and Integration Team, Development Research Group. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at fng@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team DEVELOPING COUNTRY TRADE POLICIES AND MARKET ACCESS ISSUES: 1990-2012 Constantine Michalopoulos and Francis Ng Keywords: Trade policies, merchandise and service trade, market access issues, tariffs and non- tariff measures, trade remedies, developing countries JEL Classification: F01, F10, F13, F15, F43 Sector Board: PREM Constantine Michalopoulos is a former director at the World Bank and at present a visiting scholar at the School of Advanced International Studies of Johns Hopkins University. Email: c.michalopoulos@comcast.net. Francis Ng is a senior economist of Trade and International Integration Team (DECTI) of Development Research Group in the World Bank. Email: fng@worldbank.org. The authors would like to thank Mariem Malouche and Jose-Daniel Reyes of International Trade Department (PRMTR) for their help in obtaining information from the UNCTAD/World Bank NTM database about non-tariff measures in a number of developing countries. The valuable comments and useful suggestions from Chad Bown and anonymous referees on an earlier version of this paper are also greatly appreciated. The authors are responsible for any remaining errors. The findings, interpretations, and conclusions expressed herein are those of the authors and do not necessarily reflect the views of the World Bank and its affiliated organizations. DEVELOPING COUNTRY TRADE POLICIES AND MARKET ACCESS ISSUES: 1990-2012 I. INTRODUCTION The remarkable expansion of developing countries’ trade in the period 1980-2010 and especially in the last two decades (Michalopoulos and Ng 2013) was fueled in part by their introduction of more outward looking trade regimes, involving less domestic protection as well as by favorable developments in the access they enjoyed in developed country markets. Supply and logistics constraints continued to plague many low-income countries and LDCs. But the spread of off-shore production and the increased role played by value chains played an important role in the growth of many countries’ trade. This paper has four parts: the first examines the evolution of developing countries’ own policies affecting merchandise trade; the second focuses on developed countries and the third examines policies affecting trade in services in both groups of countries. The last part discusses protectionism in both developed and developing countries since 2008. No effort will be made to discuss other policies that affect trade performance, including exchange rate or macro-economic policies or domestic subsidies and taxes. We will also not review the huge theoretical and empirical literature on the relationship between trade and economic growth at the macro or micro level. The debate about the appropriate trade policy appears to have become increasingly academic as developing countries are revealing their preferences through their practices: very few countries have actually increased protection in recent periods; and the ones that did (Argentina) appear to have done so on a temporary basis for balance of payments reasons. 2 II. DEVELOPING COUNTRY POLICIES The analysis in this section documents the great progress made by many developing country WTO members in liberalizing their trade regimes since the early 1990s (see also Drabek and Laird 1998; Finger and Schuknecht 1999). The liberalization has had several dimensions: (1) applied tariffs have been lowered; (2) the overall use of formal non-tariff barriers to trade has decreased in many countries; (3) services have been liberalized in many sectors; and (4) in general, the incidence of government intervention in trade has declined. The analysis also throws light on a number of outstanding issues in the reform agenda and some new challenges which have emerged. For example, unilateral liberalization has increased the gaps between tariff bindings and applied tariffs, increasing the uncertainty to exporters wishing to access these countries' markets as well as the opportunities for resurgent protectionism. Also, while the overall use of non-tariff measures has declined, the use of trade remedies, such as anti-dumping, as well as TBT and SPS controls on imports is on the increase. The analysis of trade policies is based in part on information from 50 developing countries for which Trade Policy Reviews (TPRs) had been conducted under the GATT and the WTO 1. The group includes 15 economies in Latin America and the Caribbean, 17 in Asia, 13 in Sub-Saharan Africa and five in Europe, Middle East and North Africa. 2 They account for over 95 percent of the total trade of the developing members of the WTO. Their policies and problems can be taken to represent the developing countries as a whole. For 40 of these countries the TPRs go back to 1990 and give us a perspective on the evolution of their policies over more than twenty years. 3 3 A. TRADE POLICIES THAT AFFECT MERCHANDISE IMPORTS 1. Tariffs The simple average applied MFN tariff level and the standard deviation in the applied tariff level for the latest year available, as well as the average level of binding, the average difference between applied and Uruguay Round bound rates and the proportion of tariff lines unbound for the 50 developing countries in the sample are presented in Table 1. The table also contains a column that shows the average applied rates for the 1990s. The average applied rates in the 50 countries range from zero in Hong Kong SAR, China and Singapore, to 5-10 percent in many countries in Latin America, to 10-20 percent in most of the rest of the countries except Tunisia, which has an average of over 20 percent. Tariff rates vary substantially within each country with overall standard deviations in excess of ten for several countries such as Egypt, India, Korea and Tunisia, and similarly high coefficients of variation. Perhaps the most interesting aspect of the table is the comparison of the applied rates over time. Only three countries (Benin, Brazil and Senegal) had a higher average applied rate in the 1990s than in the 2000s and that by a small margin. Hong Kong SAR, China and Singapore with zero rates—showed no change. The remaining 45 developing countries showed a decline in their average applied rates, frequently by more than half. The simple average applied tariff rate for the 50 countries in the sample was 17 percent in the 1990s and 9.1 percent in the 2000s. 4 4 Table 1: Bound and Applied Tariffs in Developing Countries Latest Bound Applied Rate (%) Standard Coeff. of Tariff % of Country Year Rate (%) Mid 1990s Latest Yr Deviation Variation Margin unbound Argentina 2011 31.9 14.4 9.8 7.2 0.7 22.1 0.0 Bangladesh 2008 167.1 26.7 14.0 9.3 0.7 153.2 84.3 Benin 2011 28.3 12.3 13.3 7.1 0.5 15.0 60.9 Bolivia 2011 40.0 9.6 9.1 8.7 1.0 30.9 0.0 Brazil 2011 31.4 13.2 13.5 8.4 0.6 17.9 0.0 Cameroon 2011 79.9 21.0 18.9 9.9 0.5 60.9 86.7 Chile 2010 25.1 11.0 4.9 2.4 0.5 20.3 0.0 China 2011 9.9 22.0 7.8 6.9 0.9 2.1 0.0 Colombia 2011 42.8 12.2 6.8 5.4 0.8 36.0 0.0 Costa Rica 2009 43.0 9.7 4.7 6.9 1.5 38.3 0.0 Cote d'Ivoire 2011 11.1 21.0 12.9 6.9 0.5 -1.9 66.6 Dominican Republic 2010 34.9 20.0 8.3 9.0 1.1 26.6 0.0 Egypt, Arab Rep. 2009 36.8 24.3 12.3 130.0 10.5 24.4 0.7 El Salvador 2010 36.7 10.0 5.0 7.3 1.5 31.6 0.0 Fiji 2011 40.1 12.4 11.8 11.0 0.9 28.4 48.9 Ghana 2009 92.5 17.5 13.0 7.3 0.6 79.5 85.6 Honduras 2009 32.4 9.6 6.4 6.9 1.1 26.0 0.0 Hong Kong, China 2011 0.0 0.0 0.0 0.0 .. 0.0 53.6 India 2009 49.4 37.0 10.1 14.5 1.4 39.3 25.6 Indonesia 2011 37.2 10.8 5.0 7.8 1.6 32.3 3.4 Jamaica 2011 49.6 21.1 8.4 11.2 1.3 41.3 0.0 Kenya 2011 95.1 32.1 12.3 11.2 0.9 82.8 85.2 Korea, Rep. 2010 15.8 15.0 10.1 43.4 4.3 5.7 5.4 Kuwait 2009 100.0 6.0 4.1 2.0 0.5 96.0 0.1 Malawi 2011 74.7 25.5 12.3 10.8 0.9 62.4 68.0 Malaysia 2009 14.6 8.9 5.3 9.8 1.9 9.3 15.9 Mauritius 2011 94.4 32.7 1.2 4.8 4.0 93.2 82.2 Mexico 2010 35.0 12.6 7.3 8.7 1.2 27.7 0.0 Morocco 2009 41.3 23.5 9.1 13.9 1.5 32.1 0.0 Mozambique 2010 97.5 15.6 7.7 7.5 1.0 89.8 86.4 Nepal 2011 26.0 14.8 12.3 9.1 0.7 13.7 0.6 Nigeria 2010 118.3 23.0 10.9 8.2 0.8 107.4 80.8 Pakistan 2009 60.0 50.1 14.7 12.3 0.8 45.3 1.3 Paraguay 2011 33.4 11.7 8.4 7.4 0.9 25.0 0.0 Peru 2011 30.1 16.2 3.1 4.1 1.3 27.0 0.0 Philippines 2010 25.6 19.8 5.3 7.1 1.3 20.4 33.2 Saudi Arabia 2009 10.7 13.0 4.0 2.1 0.5 6.7 0.0 Senegal 2011 30.0 12.3 13.3 6.9 0.5 16.7 0.0 Singapore 2010 7.0 0.0 0.0 0.0 .. 7.0 30.0 South Africa 2011 19.2 15.0 7.0 10.7 1.5 12.3 4.0 Sri Lanka 2011 29.9 20.0 9.0 12.3 1.4 21.0 61.6 Taiwan, China 2011 6.0 7.8 4.2 7.2 1.7 1.8 0.0 Thailand 2009 25.7 21.0 10.5 12.2 1.2 15.3 24.6 Tunisia 2008 57.2 29.7 21.8 17.8 0.8 35.4 41.6 Uganda 2011 73.0 16.8 12.1 11.6 1.0 60.9 84.2 Uruguay 2011 31.5 12.7 9.5 7.9 0.8 22.0 0.0 Venezuela 2011 36.5 12.8 11.8 8.2 0.7 24.7 0.0 Vietnam 2010 11.5 13.0 7.1 11.0 1.5 4.4 0.0 Zambia 2011 106.0 14.7 10.6 10.3 1.0 95.4 83.2 Zimbabwe 2007 89.8 17.2 14.7 16.2 1.1 75.1 77.6 Average 46.3 17.0 9.1 11.6 1.3 37.2 27.7 Notes: Bound and applied tariffs are based on simple average in mid 1990s and the latest year data available. For applied tariff rates in mid 1990s, detailed data available in each countrty can be found in Michalopoulos 2001, Table 4-1. Coefficient of variation is calculated as standard deviation divided by applied rate. Tariff margin is computed as the difference between bound and the latest applied tariffs. % of unbound is the share of tariff lines unbound. Sources: UNCTAD TRAINS (applied tariffs) and WTO IDB (bound tariffs) databases. 5 The table also shows significant variations in the proportion of total tariff lines developing countries have bound in the Uruguay Round: while all WTO Members have bound all their agricultural tariff lines, many developing country members have bound only a small proportion of the lines in the rest of their tariff schedules. There is an apparent regional pattern. In Latin America all the countries analyzed have bound virtually all their tariff lines. But in Africa and Asia many countries have bound only a small proportion of tariffs outside agriculture. In Hong Kong SAR, China and Singapore, which are committed to zero applied tariff rates, 54 percent and 30 percent, respectively, of the tariff schedule is unbound. Their practice appears to be motivated primarily by a desire to use the portion of the unbound tariff as a bargaining chip in the multilateral trade negotiations. In other countries, for example, India, Nigeria, Pakistan, where an equal or even higher proportion of the tariff schedules are unbound, there may be a mixture of motivations including the desire to maintain the freedom to increase protection as needed, for development or other objectives. Table 1 also illustrates the large differences, on average, between Uruguay Round bound and applied rates in most developing countries. Developing countries have bound their tariffs at substantially higher rates than those they apply, if they have bound them at all. For example, Brazil has bound its whole tariff schedule but at ceiling rates of 31 percent. Sometimes (for example, Nigeria) the differences are in excess of 100 percent. For countries which have bound the whole tariff schedule (Latin America, and a few others, for example, Morocco) the average difference between applied and bound rates is 30 percentage points. In some cases, such as, India, Nigeria and Pakistan, countries have bound a small portion of their tariff schedule and have used ceiling bindings with high average rates for that part which has been bound. In a very few 6 countries, for example, Ghana and Zimbabwe, a number of applied tariffs were higher than the bindings (TPRs, Ghana 2008, Zimbabwe 2010). Ceiling bindings, just like unbound rates, allow flexibility in developing country policy, when governments feel the need to increase protection. Since 2008, a number of developing countries, notably Brazil and Indonesia but also Fiji and Sri Lanka increased their MFN tariffs (see Section V). However, ceiling bindings also carry significant risks. First, domestically, they are an invitation to particular interest groups to exert pressure on governments to increase protection. Second, for foreign suppliers they reduce predictability and increase uncertainty in terms of the market access barriers they will be facing. This in turn impedes the activity of private agents, especially where investments are marked by a degree of irreversibility, and could result in reduced inflows of foreign financing. At the same time, their widespread existence undermines developing countries’ arguments that their development requires intrinsically higher levels of protection than those agreed in the WTO. The amount of actual policy flexibility ceiling bindings offer is much less than the formal difference between the MFN and the bound rates: some of the bound rates are so high as to be redundant and economically meaningless, in the sense that no imports would occur, even if the rates were substantially lower. Similarly, some of the applied rates are based on preferential agreements and thus cannot be increased without retaliation or other adverse trade repercussions. About a third of the difference between bound and MFN rates is thus not available policy space for low and middle-income developing countries (Foletti et al. 2009). Perhaps the greatest 7 usefulness of ceiling bindings is as part of developing country bargaining in multilateral trade negotiations. Table 2 provides the same information as Table 1, but distinguishes between 'Agriculture' (HS 1-24) and 'Manufactures' (HS 25-97). The Table shows that for most countries average applied tariffs on agricultural products are higher than tariffs for the rest of the product groups - which include raw materials, fuels as well as manufactures. But there are many exceptions, especially in Latin America (Argentina, Brazil, and Chile) as well as South Africa and Malaysia. The exceptions are even fewer in the case of bound tariffs with only Malaysia and a number of countries which have chosen ceiling bindings at the same rates for both agricultural and other products. Developing countries’ average applied tariff rates are much higher than for developed countries both for agriculture and manufactures. Interestingly however, the variation in the applied tariff rate structure of the developing countries in the sample is lower than that from that of developed countries (see Table 2). This is true both in manufactures where there are many tariff peaks (see Section III) and in agriculture where the main reason was the developed countries’ tariffication of non-tariff barriers in agriculture following the Uruguay Round. Finally, Table 3 shows simple averages for applied and bound tariffs as well as for differences between the two for different developing country income groups and regions for agricultural and manufactured products and for total trade. The averages contained in this table should be used with caution for reasons discussed earlier (see footnote 4). It is important nonetheless, to note the pattern that both average bound and applied tariffs tend to vary inversely with per capita income: the poorer the country, the higher the tariffs. This holds for all sectors and 8 Table 2: Bound and Applied Tariffs by Sector in Developing Countries Latest Bound Applied Rate (%) Standard Coeff. of Tariff Country Year Sector (HS-2) Rate (%) Mid 1990s Latest Year Deviation Variation Margin Argentina 2011 Agriculture 32.5 11.1 8.4 7.1 0.8 24.0 Manufactures 31.8 14.7 9.9 7.3 0.7 22.0 Bangladesh 2008 Agriculture 192.3 30.0 18.5 8.9 0.5 173.8 Manufactures 39.6 27.0 13.6 9.2 0.7 26.0 Benin 2011 Agriculture 61.8 17.0 15.9 7.1 0.4 45.9 Manufactures 11.3 13.7 12.9 7.1 0.5 -1.6 Bolivia 2011 Agriculture 40.0 10.0 8.2 7.2 0.9 31.8 Manufactures 40.0 9.6 9.2 8.8 1.0 30.8 Brazil 2011 Agriculture 35.6 11.7 9.9 7.0 0.7 25.7 Manufactures 30.8 15.4 13.7 8.4 0.6 17.1 Cameroon 2011 Agriculture 80.0 23.9 24.4 9.9 0.4 55.6 Manufactures 62.0 17.6 18.5 9.8 0.5 43.5 Chile 2010 Agriculture 26.1 11.0 4.4 2.7 0.6 21.7 Manufactures 25.0 11.0 4.9 2.3 0.5 20.1 China 2011 Agriculture 15.8 34.9 13.4 10.9 0.8 2.4 Manufactures 9.1 20.8 7.4 6.2 0.8 1.6 Colombia 2011 Agriculture 91.5 14.5 9.9 9.1 0.9 81.6 Manufactures 35.4 12.1 6.6 5.0 0.8 28.8 Costa Rica 2009 Agriculture 43.2 15.6 7.8 13.6 1.7 35.4 Manufactures 43.0 11.4 4.5 5.6 1.2 38.5 Cote d'Ivoire 2011 Agriculture 14.9 21.5 16.3 6.5 0.4 -1.3 Manufactures 8.6 19.2 12.5 6.9 0.5 -4.0 Dominican Republic 2010 Agriculture 39.6 14.8 13.9 9.8 0.7 25.8 Manufactures 34.2 8.9 7.8 8.8 1.1 26.4 Egypt, Arab Rep. 2009 Agriculture 98.9 35.9 57.1 428.0 7.5 41.7 Manufactures 27.5 23.6 9.1 61.0 6.7 18.4 El Salvador 2010 Agriculture 42.8 14.5 10.7 13.8 1.3 32.0 Manufactures 35.7 9.6 4.6 6.1 1.3 31.1 Fiji 2011 Agriculture 40.5 14.3 15.8 12.0 0.8 24.6 Manufactures 40.0 12.4 11.3 10.7 0.9 28.7 Ghana 2009 Agriculture 97.2 20.6 18.1 4.9 0.3 79.1 Manufactures 37.2 13.1 12.5 7.2 0.6 24.7 Honduras 2009 Agriculture 32.1 14.5 11.7 11.3 1.0 20.4 Manufactures 32.5 9.1 5.9 6.1 1.0 26.5 Hong Kong SAR, China 2011 Agriculture 0.0 0.0 0.0 0.0 .. 0.0 Manufactures 0.0 0.0 0.0 0.0 .. 0.0 India 2009 Agriculture 113.6 34.7 34.6 38.8 1.1 79.0 Manufactures 34.4 28.6 8.5 6.6 0.8 25.9 Indonesia 2011 Agriculture 47.4 15.4 4.2 20.5 4.9 43.2 Manufactures 35.6 10.4 5.0 5.9 1.2 30.6 Jamaica 2011 Agriculture 97.3 24.5 16.1 14.7 0.9 81.2 Manufactures 42.5 20.5 7.7 10.4 1.4 34.8 Kenya 2011 Agriculture 100.0 35.2 18.7 12.9 0.7 81.3 Manufactures 56.5 31.0 11.8 11.0 0.9 44.6 Korea, Rep. 2010 Agriculture 51.9 49.0 44.9 124.4 2.8 7.0 Manufactures 10.2 7.7 6.7 4.6 0.7 3.5 Kuwait 2009 Agriculture 100.0 .. 2.9 2.4 0.8 97.1 Manufactures 100.0 .. 4.2 1.9 0.5 95.8 Malawi 2011 Agriculture 121.3 24.5 16.5 10.5 0.6 104.8 Manufactures 42.2 25.6 11.9 10.7 0.9 30.3 Malaysia 2009 Agriculture 12.4 4.8 2.2 5.9 2.8 10.3 Manufactures 14.9 8.7 5.6 10.1 1.8 9.3 Mauritius 2011 Agriculture 119.6 28.8 0.7 4.1 5.5 118.9 Manufactures 23.3 35.2 1.2 4.9 3.9 22.0 Mexico 2010 Agriculture 35.9 12.2 14.0 18.8 1.3 22.0 Manufactures 34.9 12.4 6.9 7.8 1.1 28.0 Morocco 2009 Agriculture 54.4 33.3 26.2 27.9 1.1 28.3 Manufactures 39.2 19.7 8.1 11.2 1.4 31.1 9 Mozambique 2010 Agriculture 100.0 26.1 10.1 8.9 0.9 89.9 Manufactures 15.5 15.9 7.4 7.2 1.0 8.1 Nepal 2011 Agriculture 41.5 21.1 14.6 8.6 0.6 26.9 Manufactures 23.6 21.8 12.0 9.1 0.8 11.6 Nigeria 2010 Agriculture 150.0 32.8 14.0 7.7 0.5 136.0 Manufactures 48.9 22.3 10.6 8.2 0.8 38.3 Pakistan 2009 Agriculture 95.6 43.3 19.1 16.6 0.9 76.5 Manufactures 54.8 50.6 14.4 11.9 0.8 40.4 Paraguay 2011 Agriculture 33.1 13.1 7.7 7.9 1.0 25.5 Manufactures 33.5 11.5 8.5 7.4 0.9 25.0 Peru 2011 Agriculture 30.8 16.8 2.6 3.1 1.2 28.2 Manufactures 30.0 16.2 3.2 4.2 1.3 26.9 Philippines 2010 Agriculture 35.3 27.2 8.6 10.7 1.2 26.7 Manufactures 23.3 19.1 5.0 6.6 1.3 18.3 Saudi Arabia 2009 Agriculture 12.3 11.8 3.0 2.6 0.9 9.3 Manufactures 10.5 12.4 4.1 2.0 0.5 6.4 Senegal 2011 Agriculture 29.8 15.6 15.7 7.0 0.4 14.2 Manufactures 30.0 13.7 13.1 6.9 0.5 16.9 Singapore 2010 Agriculture 9.5 0.0 0.0 0.0 .. 9.5 Manufactures 6.4 0.0 0.0 0.0 .. 6.4 South Africa 2011 Agriculture 41.5 10.7 6.4 10.0 1.6 35.1 Manufactures 15.7 15.4 7.0 10.8 1.5 8.7 Sri Lanka 2011 Agriculture 49.8 29.9 21.8 14.9 0.7 28.0 Manufactures 19.8 18.9 8.2 11.4 1.4 11.6 Taiwan, China 2011 Agriculture 14.0 19.4 8.7 13.3 1.5 5.4 Manufactures 4.8 6.9 3.9 6.2 1.6 0.9 Thailand 2009 Agriculture 35.9 33.9 23.0 18.9 0.8 12.9 Manufactures 24.1 20.3 9.7 11.1 1.1 14.5 Tunisia 2008 Agriculture 116.0 33.4 38.6 19.9 0.5 77.4 Manufactures 40.1 29.5 21.0 16.6 0.8 19.1 Uganda 2011 Agriculture 77.4 21.7 18.1 14.1 0.8 59.4 Manufactures 50.9 16.2 11.6 11.2 1.0 39.4 Uruguay 2011 Agriculture 34.0 12.1 6.2 7.0 1.1 27.9 Manufactures 31.2 13.1 9.7 7.9 0.8 21.5 Venezuela 2011 Agriculture 55.2 13.7 13.9 10.5 0.8 41.4 Manufactures 33.6 12.7 11.7 8.0 0.7 21.9 Vietnam 2010 Agriculture 19.0 30.0 13.5 14.9 1.1 5.5 Manufactures 10.3 13.3 6.6 10.4 1.6 3.8 Zambia 2011 Agriculture 123.3 18.7 9.4 11.1 1.2 113.9 Manufactures 43.5 15.1 10.6 10.2 1.0 32.9 Zimbabwe 2007 Agriculture 144.5 38.5 23.3 28.5 1.2 121.2 Manufactures 10.4 40.8 14.2 14.3 1.0 -3.8 Average: All above Agriculture 61.1 21.3 14.4 21.5 1.3 46.7 developing countries Manufactures 31.7 17.2 9.0 8.9 1.1 22.7 Memo: Average: Developed Agriculture 17.3 9.3 6.2 12.2 2.3 11.1 countries Manufactures 6.6 5.5 3.2 4.5 1.9 3.4 Note: The classification of sector is defined as agricultural products HS 1-24 and manufacturing products HS 25-97. Sources: UNCTAD TRAINS (applied tariffs) and WTO IDB (bound tariffs) databases. groups with the exception of applied tariffs in agriculture, where there is little difference between the average for the low and middle-income developing countries. Similarly, the average differences in the margins between applied and bound tariffs tend to be highest in the low-income countries and lowest in the highest-income ones. 10 Table 3: Average Tariffs in Sectoral Product Categories by Income Group and Region Bound Rate Applied Rate Tariff Margin Country Group (No. of Countries) All goods Agric gds Manuf gds All goods Agric gds Manuf gds All goods Agric gds Manuf gds BY INCOME LEVEL Developed economies (21) 8.5 17.3 6.6 3.4 6.2 3.2 5.1 11.1 3.4 Developing economies (135) 43.2 59.3 32.8 9.8 13.6 9.5 33.4 45.7 23.3 High income Developing (18) 26.9 37.9 25.2 7.8 9.9 7.7 19.1 28.0 17.5 Upper Middle income (39) 37.0 57.6 31.1 9.5 12.4 9.2 27.5 45.1 21.9 Lower Middle income (46) 41.6 57.1 33.0 9.8 15.5 9.3 31.8 41.7 23.7 Low income (32) 62.1 76.2 38.6 11.5 14.5 11.3 50.5 61.7 27.3 Least Developed countries (42) 61.2 76.2 41.8 12.0 15.3 11.7 49.1 60.9 30.0 World (156) 39.0 54.3 29.6 9.0 12.6 8.7 30.1 41.6 20.9 BY REGION (Developing countries) Asia (36) 32.4 44.0 23.2 8.1 13.3 7.7 24.3 30.7 15.5 Latin America & Caribbean (34) 42.0 61.1 38.7 9.7 12.3 9.4 32.3 48.8 29.3 Europe and Middle East & N. Africa (21) 28.9 43.8 26.1 8.2 13.6 7.8 20.7 30.2 18.3 Sub-Saharan Africa (44) 57.4 74.5 37.0 12.2 14.9 11.9 45.2 59.6 25.0 Memo: BRICS (5) 27.5 51.6 22.5 9.2 14.3 8.8 18.3 37.3 13.6 OPEC (12) 47.1 55.9 37.7 8.5 10.4 8.4 38.6 45.5 29.4 WTO (50) 46.3 61.7 30.8 9.1 14.5 8.7 37.2 47.3 22.1 Sources: UNCTAD TRAINS (applied tariffs) and WTO IDB (bound tariffs) databases. There are a few points to note in the regional breakdown as well. First, the few Sub-Saharan African countries that have bound tariffs have done so at levels on average higher than in the other regions. But this is due primarily because of higher bindings on agricultural products. On manufactures the highest average rates are in Latin America. Also, the simple average bound tariffs on agricultural products in Africa and Latin America tend to be much higher than in Asia. For many developing countries tariff escalation is also an issue, as other developing countries are becoming more important as markets. Escalation is measured by calculating the average tariff rates applied to three groups of products, raw materials, intermediate products and final goods, which are defined at the BEC 2-digit level (Board Economic Categories). Table 4 summarizes the information. The table shows that by and large, and for most sectors escalation 11 seems to be inversely related to per capita income. With few exceptions, low-income countries and Sub-Sahara Africa tend to have the highest degree of escalation. In some cases however, (LDCs parts and accessories relative to capital goods) there is reverse escalation. Among sectors, automobiles and food processing have the highest escalation. By comparison, tariff escalation in developed countries, while still present in some sectors, is minimal compared to that of developing countries. Table 4: Tariff Escalation in Developed and Developing Countries by Income Group and Region in 2010 (unweighted average in %) Food & Bev products Industrial products Capital goods Transport equipment Consumer goods Country Group Final gds FoodprimFood proc Final gds Ind prim Ind proc All Cap Cap gds Pts & acc All Transp Auto Others Pts & acc All Cons Durable Semi-Dur Non-Dur BY INCOME LEVEL Developed economies 7.3 6.2 7.9 3.0 2.0 3.0 1.7 1.7 1.8 2.8 2.9 3.8 2.5 6.0 3.8 7.1 5.3 Developing economies 14.9 13.1 15.7 8.7 6.7 8.8 6.2 6.0 6.6 11.0 21.7 9.9 10.3 14.8 14.1 16.4 13.0 High income developing econ. 10.6 11.6 10.4 7.4 6.4 7.5 7.4 7.4 7.5 8.5 13.0 6.6 8.5 9.2 9.1 9.6 8.5 Upper middle income econ. 13.8 12.4 14.3 7.6 5.0 7.7 5.7 5.6 5.7 11.5 26.3 11.0 10.8 16.1 14.0 17.9 14.2 Lower middle income econ. 17.1 14.0 18.5 8.7 7.4 8.8 5.5 5.2 6.0 10.9 22.2 9.9 10.0 14.9 14.2 16.7 13.0 Low income economies 15.5 13.4 16.3 10.8 8.1 10.9 7.3 6.9 8.0 11.6 20.1 10.1 11.0 16.2 16.8 17.9 13.9 Of which: Least developed co. 16.4 14.6 17.0 11.7 9.8 11.7 8.0 7.6 8.9 12.4 20.8 10.7 11.7 16.6 17.1 18.1 14.3 BY REGION World 13.9 12.2 14.6 7.9 6.1 8.0 5.6 5.4 5.9 9.9 19.1 9.0 9.2 13.6 12.7 15.1 11.9 Developing economies 14.9 13.1 15.7 8.7 6.7 8.8 6.2 6.0 6.6 11.0 21.7 9.9 10.3 14.8 14.1 16.4 13.0 Asia 14.5 13.9 14.9 7.4 6.7 7.4 5.4 5.1 5.8 11.1 28.0 10.3 10.1 11.4 10.3 12.6 10.2 Latin America & Caribbean 13.7 12.8 14.0 7.9 5.0 8.1 6.2 6.3 5.9 10.6 21.4 9.9 10.0 15.2 15.0 16.4 13.5 Europe, Middle East & N. Africa 15.4 10.2 17.8 6.9 5.7 7.0 5.6 5.4 5.9 8.7 12.9 8.1 8.4 12.8 11.6 14.5 11.2 Sub-Saharan Africa 15.9 14.1 16.5 11.2 8.6 11.3 7.2 6.7 8.1 12.2 21.8 10.4 11.5 18.2 17.4 20.3 15.6 Memo: BRICS-5 15.9 12.3 17.5 7.7 4.4 7.9 6.1 6.2 6.0 10.2 34.1 13.1 8.1 13.6 10.0 15.4 12.2 OPEC-12 12.1 9.8 13.1 7.7 5.4 7.8 5.8 5.8 5.7 8.2 19.0 8.7 7.5 15.0 13.0 17.2 13.3 WTO-50 16.4 13.5 17.7 8.2 5.4 8.3 5.2 4.9 5.6 10.7 23.7 10.4 9.8 14.7 13.1 16.4 12.9 Note: The product categories are defined as BEC on food & beverages in Section 1 where food primary (111) and food processed (112); industrial products in Section 2 where industrial primary (221) and industrial processed (222); capital goods in Section 4 where capital goods (441) and parts & accessories (442); transport equipment in Section 5 where automobiles (551), other transport equipment (552), and parts & accessories (553); consumer goods in Section 6 where durable goods (661), semi-durable (662), and non-durable (663). Data are based on the latest year available in applied rate from country. Sources: UNCTAD TRAINS databases. 12 2. Non-Tariff Measures The analysis of non-tariff measures (NTMs) has three main dimensions: • The relative importance of the different policy measures employed by all developing countries in the sample, as measured by the frequency of their use. • The main product categories whose importation is affected by non-tariff measures across the countries in the sample; • The overall use of non-tariff measures by developing countries to control imports over the period 1990-2012, as measured by the overall frequency of application of such measures. The analysis relies on frequency ratios as indicators of the existence and scope of protective measures on different products by various countries. The advantages and limitations of frequency ratios as indicators of protection are well understood (Nogues et al. 1986; OECD 1997). These ratios are indicators of the extent to which countries resort to particular measures and the proportion of total products in terms of tariff lines or product groups that are affected by such measures, irrespective of the value of the products actually imported. They do not necessarily capture the protective effect of the measures taken. The protective effect of a prohibition of the importation of a product into for example, Thailand is going to be completely different from the application of a variable levy in Uruguay or the use of a non-automatic license by India. The frequency ratios are presented here in order to give overall impressions of the trade regimes in place in individual countries, and the various measures used by different countries on different products -- not to measure the actual protection provided to each product or product group. A detailed discussion of the estimating procedures followed and their limitations is presented in Appendix B. It is important to bear in mind these limitations and use the estimates of the prevalence of non-tariff measures with caution. These indicators are more useful in tracing the 13 evolution of trade regimes within each country over time, than for making inter-country comparisons, especially when the differences in indicator values are small. The non-tariff measures include non-automatic import licensing (and approvals), import prohibitions (partial or total), quotas, tariff quotas, variable levies and/or minimum pricing, and import monitoring. Frequency ratios were calculated for each measure as well as a total for each country. The Relative Importance of Different Kinds of Non-Tariff Measures The relative importance of different kinds of non-tariff measures employed by developing countries over time is shown in Table 5. The Table shows the product coverage of each NTM employed by each developing country relative to 97 product categories at the HS 2-digit level or for some countries and measures at the HS 6-digit level. 5 Thus, for example, the line for Argentina shows that non-automatic licensing affected products in 3 percent of the 97 product categories during the GATT period (1992). The data reveal that for developing countries non-automatic import licensing (including various forms of administrative approvals) continues to be the measure that affects by far the greatest number of products imported with prohibitions of various kinds ranking second. An effort was made to exclude from consideration in these calculations the large number of products which are subject to non-automatic licensing to ensure public health, safety, environmental and other standards. These are frequently justified by reference to GATT Article XX. Even so, during the early period considered (1990-1994) non-automatic licensing affected large proportions of developing country imports: non-automatic licensing was present in more about a quarter of the product categories. Several countries, such as Bangladesh, India, Nigeria and Pakistan have 14 indicated that the licensing and prohibition measures which they used were due to balance-of- payments difficulties. Table 5: Non-Tariff Measures in Developing Countries Non-Auto Tariff Variable Country Year Licensing Prohibition Quotas Quotas* Other Levies Surcharges Argentina 2008 85 * 34 * 3* 1999 1 1 1 1992 3 2 Bangladesh 2012 17 1992 34 43 Benin 2010 1 1997 1 Bolivia 2008 12 * 4* Brazil 2009 1 1996 11 11 1 1 1993 10 7 100** Cameroon 2007 3 1 1995 8 Chile 2009 1 3 1997 1 4 1991 1 5 China 2012 1 1 Colombia 2012 19 * 4* 3 6 1996 6 1 6 1990 55 7 3 Costa Rica 1995 6 6 Côte d'Ivoire 2012 1 1995 31 5 Dominican Rep. 2008 0 1996 5 1 1** Egypt 2011 42 * 1993 14 53 El Salvador 1996 5 1 1 Fiji 2009 4 1997 5 Ghana 1992 3 Hong Kong SAR, China 1990 16** India 2011 4 4** 4 1998 94 3 1993 99 1 Indonesia 2011 26 * 1998 31 1991 53 5 3 1 Jamaica 2011 4 Kenya 2011 37 * 10 * 1994 87 Korea Rep. 2008 1 2 0.1 1996 1 25 1992 32 3 26 Kuwait 2012 1 Malawi 2010 4 15 Malaysia 2010 6 3 1997 20 14 2 7 1993 55 4 2 Mauritius 2009 14 * 1995 9 7 Mexico 2008 2 1997 6 1 7 1993 28 2 2 Morocco 2011 2* 2* 1996 13 1 1990 51 24 Mozambique 2009 3 Nepal 2012 1 Nigeria 2011 17 1998 2 9 1991 14 Pakistan 1995 17 1 Paraguay 2008 1* 1* 1 Peru 2008 1* 2 1994 6 Philippines 2012 5 1 1993 7 4 1 Senegal 2011 2* 1* 1994 10 5 Singapore 1996 1 1992 1 South Africa 2009 1 5 3 1998 5 3 1993 36 Sri Lanka 2010 4 1995 23 Taiwan, China 2010 1 Thailand 2007 3 8 1995 11 6 1 12 1991 36 2 2** 3 Tunisia 2011 22 * 1994 54 2 Uganda 1995 3 Uruguay 2012 3 1998 4** 1992 1 1 31 Venezuela 2008 24 * 3* 1* 1996 2 3 13 Zambia 1996 1 Zimbabwe 2011 100*** 1995 23 Notes: Frequency ratio of percentage in HS 2-digit products affected by each NTM. All countries report several product categories—usually 15-30% of tariff lines as subject to licensing and/ or prohibitions for health, sanitation, safety or defense reasons. Only licensing and/or prohibitions for economic reasons are reported here. * in percent at HS 6-digit level; ** import monitoring; *** exchange control Sources: GATT TPR (1989-1994); WTO TPR (1995-2012); UNCTAD/World Bank NTM database. 16 No effort was made in this study to determine the consistency of these licensing arrangements or for that matter, any other member policies with GATT provisions. The limitations of WTO’s system of voluntary notifications are well understood (Bachetta et al. 2012). A significant part of the total membership has not provided information on the licensing procedures they use or notifications of quantitative restrictions they employ. Most of the members notifying the use of quantitative restrictions did so under GATT Article XX. Three members, India, Korea Rep. and Philippines also justified restrictions under the balance-of-payments provisions of Article XVIIIb. While caution is needed in interpreting the findings on the evolution of policy over time because the sample of countries is partly different in the three sub-periods examined, the data strongly suggest that the utilization of all 'core' non-tariff measures with clearly protective effect, such as licensing, prohibitions, quotas, and variable levies/administrative pricing has declined over time. This is shown by the decreasing frequency ratios over time for the large majority of countries for which data are available. The only measure clearly showing an increase is the use of tariff quotas which is permitted under the Uruguay Round Agreement in agriculture. The Main Products Subject to NTMs The TPRs show that agricultural products (Group I-IV or HS 1-24) were the most subjected to overall controls especially in the earlier period. The number of countries imposing these controls has substantially declined in the period 1995-1998, following the tariffication in agriculture under the Uruguay Round Agreements (See Michalopoulos 2001, Table 4.6). In addition to agriculture, mineral products, in particular fuels (HS-27), rubber products (HS-40), machinery, especially Electrical Machinery (HS-85), and precious stones and metals continued to 17 be subject to controls, especially through licensing in a significant number of countries in all periods (see also Michalopoulos 1999b, Table A-3). 6 The Uruguay Round Agreement on agriculture has been praised for subjecting the agricultural sector to the same trade disciplines that apply to other sectors. However, it has been criticized for not resulting in significant liberalization of developed country policies. At the time the agreements were being negotiated, the main developing country concerns focused on the market access issues -- and the benefits that some of the key agricultural exporters would obtain, as well on the possible adverse impact reduced export subsidies in developed countries would have on net food importing developing countries. 7 Little attention was focused, until the decade of the 2000s, on the level of support and the kinds of measures appropriate for developing countries to implement in pursuit of their agricultural development which is essential to the elimination of poverty, itself very often mainly a rural phenomenon. The Uruguay Round Agreement on Agriculture contains provisions which permit developing countries to increase their support to agriculture (and of poor consumers) through means not available to developed countries. For example, direct and indirect investment and input subsidies to poor farmers are excluded from the calculation of aggregate measures of support (AMS); reduction of the support commitments by developing countries may take ten years to implement while LDCs are totally exempt; and food subsidies to urban and rural poor are excluded from the calculation of AMS. However a number of critics have pointed to the 'unfairness' of the agriculture agreement as it still permits greater support levels for developed countries (which in the past have given a great deal of assistance to their agricultural sector) than developing countries which penalized agriculture in the base period (Das, 1998). Also, the Special 18 Safeguard Mechanism in the agreement was designed in such a way as to be of limited use to developing countries that had not tariffied their agricultural protection. Following the launching of the Doha Round, developing countries focused without success on developing a viable Special Safeguard Mechanism that would cover all of their agricultural production, emphasizing special concerns that arise to the welfare of their poor farmers in the case of an import surge. In practice, the most common cause of import surges has been shortfalls in developing country domestic production. Developing countries which are keen to reduce the variability of food prices both to their poor consumers as well to their farmers have resorted to an extensive use of variable price supports and export taxes on agricultural commodities. Trade Remedies Trade remedies include anti-dumping, countervailing and safeguard actions. In principle, such actions are consistent with WTO provisions. The legal basis and procedures for the imposition of trade remedies in each instance are as various as are remedies, which usually do not involve quantitative restrictions but changes in duties and charges to address the problem appropriately in each case. In the case of anti-dumping and countervailing duties, these remedies are intended to correct for distortions that occur when exporters are obtaining subsidies and engaging in discriminatory pricing practices which result in injury to domestic producers. In the case of safeguards, the issue is simply injury to domestic producers, even if no unfair trade practices are involved. Evidence regarding the frequency with which developing countries have taken trade remedy actions is presented in Table 6. As it has been argued that the mere initiation of an investigation on anti-dumping action tends to have a restraining effect on imports -- irrespective of the outcome 19 of the case (Finger 1993; OECD, 1997), the main indicator used in this table is the number of anti-dumping investigations that had been initiated each year going back to 1978. The table shows that the percentage of actions by the developing countries and against them changed spectacularly over the period of the last 30 plus years. Until 1986 no developing country had initiated an anti- dumping action—although they were subjected to some anti-dumping actions by the developed countries. Over time the situation was totally reversed: developing countries initiated the majority of the actions and were the subject of most of the anti-dumping actions-both by other developing countries and the developed countries (see Appendix Table B-4). 2002 was the peak year for anti-dumping investigations by developing countries (206), while 2001 was the peak year for investigations against them (227). Since that time anti-dumping investigations appear to have declined. The increased use of trade remedies by developing countries in recent periods has been extensively documented by Bown (see Bown, 2011 and Bown, 2013). Developing countries especially Argentina, China and India have increased both the average annual imposition of remedies, and the overall stock of products subject to such measures especially anti-dumping, and in particular against other developing countries. 8 20 Table 6: Antidumping Investigations in Developed and Developing Countries, 1978-2011 By Reporting Member (no. of cases) By Exporting Member (no. of cases) Developed Developing Developed Developed Developing Developed Year Members Members as % of Total Members Members as % of Total GATT: 1978 6 0 100 5 1 83 1979 58 0 100 54 4 93 1980 72 0 100 66 6 92 1981 94 0 100 77 17 82 1982 221 0 100 174 47 79 1983 178 0 100 130 48 73 1984 159 0 100 119 40 75 1985 206 0 100 131 75 64 1986 156 5 97 105 56 65 1987 101 19 84 86 34 72 1988 112 12 90 78 46 63 1989 87 9 91 55 41 57 1990 149 18 89 80 87 48 1991 209 19 92 139 89 61 1992 260 65 80 167 158 51 1993 149 157 49 140 166 46 1994 138 91 60 96 133 42 WTO 1995 78 79 50 71 86 45 1996 79 147 35 95 131 42 1997 127 119 52 125 121 51 1998 90 176 34 135 131 51 1999 176 182 49 154 204 43 2000 138 160 46 130 168 44 2001 179 193 48 145 227 39 2002 109 206 35 127 188 40 2003 86 148 37 90 144 38 2004 113 107 51 61 159 28 2005 62 139 31 44 157 22 2006 71 133 35 49 155 24 2007 61 104 37 38 127 23 2008 75 138 35 48 165 23 2009 64 145 31 48 161 23 2010 38 133 22 66 105 39 2011 63 92 41 44 111 28 Sources: GATT and WTO antidumping database. Table 7 shows the number of anti-dumping, countervailing and safeguard measures taken by the 50 countries in our sample cumulatively over the period 1995-2011 as well the number products affected by these actions. The last four columns show the number of trade remedy 21 measures in place in each of country in mid-2011. The first thing to note about the table is that anti-dumping is by far the most popular instrument dwarfing countervailing or safeguard actions. Indeed out of 50 countries 24 initiated anti-dumping measures, 15 safeguards and only nine took countervailing actions. As of mid-2011, 19 developing countries had trade remedy measures in place involving at least some anti-dumping, and anti-dumping accounted for over 90 percent of the actions they took. Among developing countries, anti-dumping is, for the most part, a middle and higher- income developing country practice. Argentina, Brazil, China, India, Korea Rep., Mexico and South Africa account for the bulk of developing country actions. Pakistan is the country with lowest per capita income ($1,190 in 2011) that has an anti-dumping action. India is the country that has taken the most measures over the period. It is also the country with the most measures in place in 2011. Mexico is the country that has used anti-dumping actions affecting by far the largest number of products over time (1,685), estimated at 19 percent of the tariff lines at the HS-6 digit level. 9 The main reason for this is that it applied anti-dumping measures across broad categories of products – such as, all the tariff lines in several whole HS2 product groups, for instance, textiles against China. But a lot of these measures have expired recently. 22 Table 7: Trade Remedy Measures 1995-2011 and in Force 2011 by Developing Country Antidumping, 1995-2011 Countervailing, 1995-2011 Safeguards, 1995-2011 Total AD+CV+SG, 1995-2011 Measures in Force, 2011 * No. of No. of No. of No. of No. of No. of No. of No. of Counter- Total Country (No. of Countries) Measures Products ** Measures Products ** Measures Products ** Measures Products ** Antidumping vailing Safeguards AD+CV+SG Argentina 206 1,075 4 32 4 82 214 1,189 96 0 0 96 Brazil 119 541 7 53 2 23 128 617 75 1 1 77 Chile 9 35 2 85 7 158 18 278 1 0 0 1 China 151 378 4 40 1 223 156 641 115 2 0 117 Colombia 24 196 .. .. 1 7 25 203 12 .. 0 12 Costa Rica 3 10 1 2 .. .. 4 12 2 0 .. 2 Dominican Republic 1 1 .. .. 2 12 3 13 1 .. 2 3 Egypt 53 69 .. .. 4 25 57 94 15 .. 0 15 India 478 1,465 .. .. 13 108 491 1,573 206 .. 2 208 Indonesia 42 288 .. .. 10 60 52 348 16 .. 9 25 Jamaica 4 15 .. .. .. .. 4 15 .. .. .. .. Korea, Republic of 72 336 .. .. 2 17 74 353 33 .. 0 33 Malaysia 25 59 .. .. .. .. 25 59 6 .. .. 6 Mexico 85 1,685 8 65 .. .. 93 1,750 38 0 .. 38 Morocco .. .. .. .. 2 23 2 23 .. .. 0 0 Pakistan 36 294 .. 20 .. .. 36 314 33 .. .. 33 Paraguay 2 2 .. .. .. .. 2 2 .. .. .. .. Peru 49 379 5 35 .. .. 54 414 21 2 .. 23 Philippines 11 51 .. .. 7 18 18 69 1 .. 5 6 Singapore 2 2 .. .. .. .. 2 2 .. .. .. .. South Africa 128 460 5 20 1 2 134 482 33 1 0 34 Taipei, Chinese 15 172 .. .. .. .. 15 172 6 .. .. 6 Thailand 34 318 .. .. 1 1 35 319 23 .. 1 24 Uruguay 1 12 .. .. .. .. 1 12 .. .. .. .. Venezuela 25 94 1 8 5 109 31 211 .. .. 0 0 Memo Items: Other Developing (3) 9 37 .. 38 16 87 25 162 1 .. 3 4 Developed (13) 1,017 20,963 133 11,460 44 1,728 1,194 34,151 604 74 11 689 All countries (41) 2,601 28,937 170 11,860 122 2,683 2,893 43,480 1,338 80 34 1,452 Notes: .. = 0 or data not available * Number of measures in force based on notifications to WTO as of June 30, 2011. ** Total products affected by AD, CV, and SG are based on number of products on raw data files at HS 6-, 8-, 9-, 10-digit level or mixed together in reporting countries. Sources: WTO Antidumping database and Trade Profiles data; and Bown, C (2012), Temporary Trade Barriers (TTB) database (product data). The Overall Use of Non-Tariff Measures The overall use of non-tariff measures by developing countries during the GATT and WTO sub-periods is presented in Table 8. Column one shows country per capita income in 2011. Column two shows an openness index for each country calculated as the ratio of exports plus imports divided by GDP also in 2011. Column three shows the presence of tariff peaks in the latest tariff schedule. Column four shows the total number of non-tariff measures in place as of June 2011. The other columns show total frequency ratios (Tfm ) for 'core' non-tariff measures (see Appendix B for a discussion of the meaning of total frequency ratios) for different periods. 23 Core measures are defined as those that involve quantitative restrictions or price controls on imports, such as non-automatic licensing of any kind, prohibitions, quotas and tariff quotas as well as variable import levies and administrative/minimum pricing. 10 As a particular product category may be affected by more than one NTM, duplicative measures have been excluded in this calculation. The first point to note from the table is that in the GATT part of the period covered 1990- 1994, the values of the total frequency ratios for NTMs were extremely high for several countries - covering more than 50 percent of products in such countries as Bangladesh, Colombia, Egypt, Kenya, India, Indonesia, Malaysia, Morocco and Tunisia. Without doubt NTMs at that time were an important feature of many developing countries in all regions. Subsequently, the data strongly suggest that the total frequency ratios have been substantially reduced. There are exceptions: for example both Argentina and Zimbabwe have introduced extensive foreign currency controls which de facto substantially limit trade. Tariff quotas have increased since the earlier period as a consequence of the implementation of the Uruguay Round provisions in agriculture. Second, as with tariffs, there is a tendency for the total frequency ratios of non-tariff measures to be greater in countries with lower levels of per capita income and lower degrees of openness. Third, the most important recent development is the very large increase during the period 1999-2011 in the use of trade remedies, especially anti-dumping, compared to the period 1990- 1998. While strictly comparable data are not available, the tariff line frequencies of anti-dumping measures in the earlier period rarely exceeded one percent (with Mexico being the exception), while in the later period the number of products affected by trade remedies in countries like 24 Table 8: Trade Openness and Barriers in Developing Countries GDP Openness Tariff Peak /a Share of Nonoil Share of Nonoil No. of Msre Ratio of Core NTMs (in %) /e Per Capita in 2011 (as % of total Imp in AD, CV, Imp in AD, CV, in Force on Core NTMs Core NTMs TQ /f Core NTMs Country in 2011 (US$) (Trade/GDP) tariff lines) SG in 2007 /b SG in 2011 /b AD, CV, SG /d 1990-95 1995-98 2005-11 2005-11 /g Kuwait 62,664 86 0.0 .. .. 0 .. .. .. .. Singapore 46,241 386 0.0 .. .. 0 1 2 .. .. Hong Kong, China 34,457 440 0.0 .. .. 0 15 0 .. .. Korea, Rep. 22,424 102 6.4 0.5 0.4 33 50 25 2 1 Saudi Arabia 20,540 97 0.0 .. .. 0 .. .. .. .. Taiwan, China 20,139 140 0.0 0.1 0.2 6 .. .. 1 .. Chile 14,394 70 0.0 0.1 0.1 1 5 5 1 3 Uruguay 13,866 53 29.0 .. .. 0 3 0 .. 3 Brazil 12,594 23 27.5 1.8 1.7 77 17 2 0 Argentina 10,941 40 21.6 1.8 2.5 96 3 2 .. 100 Venezuela 10,810 46 44.0 .. .. 0 .. 18 .. 28 * Mexico 10,064 62 5.8 0.7 0.3 38 28 13 1 .. Malaysia 9,656 177 15.9 0.3 0.2 6 56 19 3 6 Mauritius 8,797 116 6.0 .. .. 0 .. 17 .. 1 Costa Rica 8,676 78 0.6 .. .. 2 .. 31 1 .. South Africa 8,070 55 22.6 1.0 0.3 34 37 8 5 3 Colombia 7,067 34 17.3 0.4 0.5 12 .. 6 .. 32 * Peru 6,009 48 0.0 1.9 1.3 23 6 .. .. 3 Jamaica 5,562 71 42.9 .. .. 0 .. .. .. .. Dominican Rep. 5,530 56 30.4 .. .. 3 .. 6 0 .. China 5,445 55 13.3 3.0 3.2 117 55 10 3 6 Thailand 4,972 135 18.9 1.1 1.2 24 37 18 8 3 Fiji 4,391 117 36.8 .. .. 0 .. 5 .. 4 Tunisia 4,297 103 57.0 .. .. 0 54 ... .. ... El Salvador 3,702 69 1.7 .. .. 0 .. 5 .. .. Paraguay 3,635 112 20.2 .. .. 0 .. .. .. 3* Indonesia /c 3,495 47 7.4 0.3 1.2 25 54 31 .. 26 * Morocco 3,054 76 23.6 .. .. 0 58 13 .. .. Sri Lanka 2,835 52 42.6 .. .. 0 .. 23 .. 4 Egypt, Arab Rep. 2,781 47 18.4 .. .. 15 57 .. .. .. Bolivia 2,421 76 17.5 .. .. 0 0 .. .. 16 * Philippines 2,370 71 5.0 0.1 0.1 6 12 .. .. 6 Honduras 2,226 108 0.5 .. .. 0 .. .. .. .. Ghana 1,570 71 40.5 .. .. 0 3 .. .. .. India /c 1,489 50 6.3 3.2 6.3 208 99 93 .. 12 Nigeria 1,452 69 35.0 .. .. 0 14 11 .. 17 Zambia 1,425 79 48.6 .. .. 0 .. 1 .. .. Vietnam 1,411 165 23.1 .. .. 0 .. .. .. .. Cameroon 1,271 61 54.1 .. .. 0 8 .. .. .. Cote d'Ivoire 1,195 77 46.8 .. .. 0 .. .. 1 1 Pakistan 1,194 32 44.5 1.1 1.4 33 18 .. .. .. Senegal 1,119 68 50.1 .. .. 0 10 .. .. .. Kenya 808 65 37.7 .. .. 0 87 .. .. 47 * Benin 802 42 50.3 .. .. 0 .. 2 .. 1 Zimbabwe 776 126 32.9 .. .. 0 23 .. .. 100 Bangladesh 735 43 38.0 .. .. 0 54 .. 17 Nepal 619 46 45.3 .. .. 0 .. .. .. 1 Mozambique 535 71 23.8 .. .. 0 .. .. .. 3 Uganda 487 58 37.7 .. .. 0 .. 3 .. .. Malawi 371 78 33.9 .. .. 0 .. .. .. 4 Notes: /a Number of tariff lines in percentage at HS 6-digit level with applied rate at 15% or above in 2011 or latest data available. /b Trade-weighted share of nonoil imports in % subject to antidumping (AD), countervailing (CV), Safeguads (SG) and China-specific transitional safeguards. /c Trade-wgted shares of nonoil imports in subject to AD only in 2011 for Indonesia and India are 0.8 and 5.8% respectively. All others are the same as total AD,CV,SG. /d Number of measures in force by notifications to WTO in antidumping, countervailing duties, and safeguards as of 2011 /e Frequency ratio in percentage of core non-tariff measures (NTMs) in HS 2-digit product categories. /f Tariff quota (TQ) data is computed in HS 6-digit products due to data available in WTO TPR reports. /g * Other core NTMS are based on UNCTAD/World Bank NTM database, where figures are computed at HS 6-digit level. Sources: World Bank WDI database (GDPPC, GDP and trade data), UNCTAD TRAINS database (tariff data), WTO Trade profiles and TPR reports (various issues); and Bown, Chad (2013), "Emerging Economies and the Emergence of South-South Protectionism," Journal of World Trade, 47(1): 1-44. 25 Argentina, India, China, Korea Rep. and Chinese Taipei are likely to be much higher. In some cases, e.g. India, liberalization of the trade regime has taken the form of reducing licensing and increasing the use of trade remedies. Overall, Table 8 suggests that high-income developing countries tend to have less overall tariff and non-tariff restrictions to trade; low-income countries have higher tariff barriers but limited use of trade remedies; while middle-income countries have significant of trade remedies and their tariff structure is somewhat in between that of the low and high-income developing countries. The results are consistent with the findings of Kee et al (2009) regarding the relationship between per capita income and total restrictiveness of the import regime. 11 Finally in some lower-income economies the greatest non-tariff measures have little to do with formal barriers. The latest TPR (2012) of the East African Community (EAC) is instructive in this respect. The EAC has no formal NTMs either with respect to the rest of the world or for internal trade. Yet in December 2008 the EAC Secretariat identified 35 NTMs that impeded Trade, of which 12 involved customs and administrative procedures, 11 business licensing and roadblocks and nine TBT and SPS (WT/TPR/S/271, p.8). 12 3. SPS and TBT The establishment of the SPS and TBT agreements has had a variety of implications for developing countries. Practically all developing-country Members of the WTO have institutions that deal with the maintenance of technical standards and enforce the health, sanitary and phytosanitary regulations, as agreed. However, there is serious doubt, about the capacity of many of these institutions to implement regulations. For example, most developing countries have legislation on national standards that are based on those of the International Standards 26 Organization (ISO), of which many are also members. But in some cases, for example, Zambia, there appear to be no government testing facilities. Most countries report that they have concluded numerous formal bilateral mutual recognition agreements on technical standards. But it is not clear whether these agreements provide adequate information to producers on the standards applicable in countries where they may consider marketing their products. The situation regarding sanitary and phytosanitary measures appears to be quite similar. Countries have established regulations requiring sanitary or phytosanitary certificates for imported food, drugs and agricultural and veterinary products. Again, the standards used are supposed to be based on those laid down in international agreements such as the Codex Alimentarius. But it is difficult to judge the adequacy of the regulations or the effectiveness of the institutions entrusted with their implementation. The TPRs suggest that the situation in many low- income countries is far from satisfactory. For example, in Malawi, the SPS legislation is reportedly outdated and the country also lacks adequate facilities to test products for import or for export. Similarly, in Tanzania, “Due to these shortcomings, Tanzania has suffered some outbreaks of plant pests and diseases (including cassava green mite, large grain borer in maize, banana wilt disease and a variety of fruit fly) that infests citrus and mango-growing areas due to staff inadequacies� (Tanzania TPR 2012, p.A4-416). 13 In recent periods, concern has been expressed by many that the TBT and SPS provisions are used by countries to camouflage protection (Gourdon and Nicita, 2012). This may well be the case in some situations. But it is difficult to discover protective intent when there are legitimate reasons to introduce more effective TBT and SPS controls. And the problem in some developing countries may be not that there too many controls but not enough. 27 B. MEASURES THAT AFFECT EXPORTS There are clear links between a country’s trade policies that affect its exports and those that affect its imports: for example, measures to control exports of raw materials that are used as inputs by domestic industries distort resource allocation in much the same way as import protection measures for that industry. Similarly, the imposition of measures that restrict the quantity and increase the domestic price of imports may adversely affect the profitability of exports and a country to take offsetting measures in favor of exporters. 14 Thus, it is not surprising to find that over the period reviewed, developing countries liberalized their policies affecting exports in much the same direction as they liberalized policies affecting imports. Developing countries’ policies towards exports are characterized by the following broad tendency: they have tended to impose controls and taxes on their exports of primary products and foodstuffs while providing incentives and subsidies to their exports of manufactures. The control and taxation of exports of primary products have had two purposes: to capture some of the rents from the production and sale of raw materials; and to provide incentives for industrialization by taxing exports of raw material and other inputs, thereby making them available to domestic industries at lower than world prices. For foodstuffs, the main justification is the promotion of food security. Manufacturing or non-traditional exports (some of which may involve processed agricultural or agriculture- related products) are being backed by incentives because they are believed to contribute to long term growth and development, and because it is felt that, without government assistance exporters will have difficulty breaking into foreign markets due to externalities of various kinds. Also, some of the so-called incentives amount to no more than a government effort to offset the disincentives to non-traditional exports created by the import regime. 28 1. Measures Which Tend to Tax or Regulate Exports Table 9 shows the frequency of developing-country use of various kinds of policy measures to tax or regulate their exports. The table is constructed much like those on imports and shows snapshots of the frequency of use of different measures by various countries in different time periods. The frequencies are defined in terms of the proportion of total product groups at the HS 2 level (and some at HS 6) which may contain products that are taxed or regulated by different measures in different countries. The measures include export taxes or similar levies, minimum export prices, non-automatic export licensing (or approval), export prohibitions (total or partial) and export quotas. The last column shows whether the country also uses export subsidies. Just as with imports, export taxes are the most commonly used measure, although licensing used to be very important in earlier periods. The variation among countries, especially during the earlier periods is rather large, with some countries, such as Brazil, Indonesia, Malaysia controlling more than 20 percent of export product groups through licensing or other approval procedures, while many others limited such licensing to the implementation of health and safety standards and environmental obligations. 15 There is no obvious pattern with respect to export taxes: their use has increased in some cases (Brazil, Sri Lanka) but decreased in others (India, Kenya). The TPRs suggest that often a combination of measures - sometimes a quantitative measure, such as a license in combination with a price linked measure, such as an export levy -- would be used. A rather crude adding up of the number of countries and measures that are imposed by product group showed that the most frequently regulated/taxed product groups of exports by developing countries in 1990-1998, at the HS 2-digit level, were live animals, coffee, cocoa and tea, fuels, hides and skins, wood products, and cotton (Michalopoulos,2001 p.70). Much the same products appeared to be regulated in the period since 2000. 29 Table 9: Developing Country Policies on Exports Export Taxes Minimum Export Export Export Export Country Year & Levies Export Prices Licensing Prohibitions Quotas Subsidies Argentina 2008 1 1 1999 2 x 1992 1 x Bangladesh 2012 3 x 1992 3 1 20 9 Benin 2010 4 1997 3 1 2 Bolivia 2005 3 1999 1 Brazil 2009 5 1996 3 1 1 3 x 1993 1 1 29 3 x Cameroon 2007 2 3 1995 6 3 1 2 Chile 1997 x 1991 x China 2012 4* xx Colombia 2012 3 1996 3 3 2 1 1 x 1990 1 2 1 x Costa Rica 2007 2 1995 4 1 1 x Côte d'Ivoire 2012 5 5 1995 5 9 2 3 Dominican Rep. 2008 2 1996 1 Egypt 2011 2* 2* 2* 1993 1 1 4 3 x El Salvador 2010 x 1996 1 1 x Fiji 2009 3 3 1997 2 7 Ghana 2008 2 1992 1 24 3 1 Honduras 2010 1 1 x Hong Kong SAR, China 1990 1 India 2011 9 3 3* 4 4 x 1998 13 1 9 3 5 x 1993 1 4 10 2 4 x Indonesia 2007 5 4 1998 5 19 x 1991 6 33 6 3 x Jamaica 2011 4 x Kenya 2011 4* 18 * 1992 10 16 Korea Rep. 2008 2 x 1996 4 x 1992 42 Kuwait 2012 1 7 Malaysia 2010 5* 1 1997 10 35 30 1993 9 15 Malawi 2010 3 10 2 Mauritius 2008 1 x 1995 1 4 3 x Mexico 2008 9 1997 14 7 x 1993 3 24 9 x Morocco 2011 2* 1996 2 2 2 x 1990 2 5 x Mozambique 2009 5 Nepal 2012 1 Nigeria 2011 1 6 1998 1 2 5 1 1991 2 5 Pakistan 2008 10 4 x 1995 21 3 20 13 x Paraguay 2005 2 1 1 1997 6 1 1 Peru 2007 3 1994 1 Philippines 2012 1 2 3 1 1993 26 15 Senegal 2009 1 3 1994 1 3 4 x Singapore 1995 2 1992 1 6 South Africa 2009 1 1998 1 8 1 x 1993 1 9 2 x Sri Lanka 2010 14 3 1995 6 1 7 1 Thailand 2007 2 7 1995 4 8 4 1 1991 3 2 11 x Tunisia 2011 2* 1994 1 1 1 Taiwan, China 2010 1 Uganda 2012 3 1995 1 1 4 x Uruguay 2012 3 x 1995 4 1992 3 Venezuela 2002 10 x 1996 1 x Zambia 2009 3 1996 1 Zimbabwe 2011 12 8 8 1995 9 Notes: Frequency ratio of percentage in HS 2-digit products affected by each measure. xx no export subsidies but extensive domestic supports. * computed at the HS 6- or 8-digit level. Sources: GATT TPRs 1990-1995; WTO TPRs, 1995-2012; and UNCTAD/World Bank NTM database. 31 2. Export Incentives Practically all the countries reviewed have a variety of policies and institutional support measures for non-traditional exports. In some cases explicit export subsidies have been introduced—but their use is declining. Table 9 shows that about 17 percent of the 49 countries for which there are data provided export subsidies in the period 2005-2011, while 40 percent of the total gave subsidies in the 1990’s. The experience in a number of countries with export subsidies suggests that they are frequently needed to offset incentives to sell to the domestic market that is created by protection. But when protection is relatively low and the exchange rate not overvalued, subsidies tend to be an expensive way of promoting exports whose main impact is to provide additional profits to established exporters. More frequently the measures involve policies aimed at off-setting the impact of import controls as well as institutional support through the provision of export financing -- sometimes on concessional terms -- and insurance, marketing and the establishment of export processing zones (EPZs) or similar arrangements of temporary admission aimed at export promotion. In most cases the measures are not product specific but apply to broad categories of products -- usually defined as non-traditional or manufacturing exports. Sometimes, the incentives are available only to certain groups of producing/exporting units, such as small and medium sized enterprises. For these reasons it is not possible to undertake an analysis at the tariff line or product group level as has been done for other policies affecting imports or exports; rather, it is only possible to note the presence or absence of a particular kind of program and its main characteristics. While some TPRs have identified problems or constraints affecting the effectiveness of programs this has not been the general rule. Indeed a study of the effectiveness of export promotion agencies concluded 32 that they have had on the whole a positive impact on exports; but that there are diminishing returns to size—i.e. in export promotion small is beautiful (Lederman et al 2009). Practically all countries have introduced some type of duty drawback system. 16 In many of the countries the system is intended to cover both duties and other border taxes as well as domestic taxes such as VAT. In an increasing number of cases it also covers taxes on domestically produced inputs. Frequently, in order to facilitate administration, the actual tax rebate mechanism involves the reimbursement or credit for a certain percentage of the firm’s overall tax liability, rather than a rebate of specific duties or taxes. As a consequence, some of these programs may contain an element of implicit export subsidy. On the other hand, this is one area where several of the TPRs reported problems and difficulties with delays in payment and rebates which result de facto in penalizing exporters relative to their overseas competitors (World Bank, 2000). With few exceptions countries reported the existence of one or more export processing zones or temporary admission schemes. The importance of these zones varies tremendously: In Korea they amounted to $4.6 billion of exports in 2008 compared to $11 million in Nigeria in 2011. While EPZs are still large in countries like Costa Rica and India, they are declining in importance in a number of countries, for example Mauritius and the Dominican Republic, where they had played an important role in earlier periods. In Mauritius the government announced in 2008 that it was going to create an island wide-EPZ, at the same time unilaterally eliminating tariffs on a more that 1200 tariff lines (Mauritius TPR 2008). 33 C. REGIONAL PREFERENTIAL ARRANGEMENTS In tandem with global integration, many developing countries have made intensive integration efforts at a regional level. Two types of measure can be distinguished in this regard: regional PTAs with other developing countries, and similar PTAs with developed countries. Examples of the former include the MERCOSUR customs union which involves several countries in South America, and the SADC a free trade area among 14 developing countries in South and East Africa. Examples of North-South regional integration include North America Free Trade Area (NAFTA) as well as the various regional agreements between the EU and different groups of ACP countries (WTO, 2011). In the last decade there has been a large increase of both North- South and South-South PTAs. The total number of PTAs in place is estimated at 360 with over two-thirds of the agreements being among developing countries. Mongolia is the only WTO member which has not declared that it participates in a PTA. Chile is the developing country with the largest number of PTAs with 26. It is estimated that, as a result, 95 percent of Chilean trade is with PTA partners. But not all trade between PTA partners involves preferences. While more than 50 percent of world trade is among PTA partners only 16 percent of world trade involves PTA preferences. This is because for a lot of world trade the MFN tariffs are zero. Although developing countries are involved in a lot of PTAs the total trade involved is relatively small: almost 50 percent of all trade among PTA partners involves the EU and another 13 percent involves NAFTA. The bulk of the PTAs involved preferential tariffs on merchandise trade; but there are also a few involving only services; and a growing number which involve both goods and services as well as other aspects of ‘deeper integration’ such as cooperation in competition policy, investment and protection of intellectual property rights. 34 The preferential margins on tariffs on average are small, only about two percent—as MFN tariffs have come down in many countries; and developing countries enjoy other unilateral preferences in developed country markets. Few tariff lines have preferential margins of more than 5 percentage points. However, for some countries, for example Mauritius, preferential margins in EU markets are high as a number of its products e.g. textiles and sugar are subject to substantial barriers. There are reports that in a number of preferential arrangements in Sub-Sahara Africa (WAMU TPR, 2011), non-tariff barriers in intra- regional trade may de facto negate the preferences accorded to regional partners. The 2000s saw a veritable explosion of PTAs: in the 1990s on average WTO members were parties to two PTAs. In 2011 the average number of PTAs per WTO member rose to 13. Developing countries, especially in Africa, have committed to participate in a number of overlapping PTAs giving rise to the familiar ‘spaghetti bowl’ image. A huge literature has emerged to analyze the causes of this phenomenon and its implications for developing countries and the trading system. Some of the reasons for the increase, especially of PTAs with deeper integration commitments in Asia had little to do with the traditional preference motive and more to do with off-shoring and the establishment or safeguarding production networks. In other cases it may be economically advantageous to join a PTA or become part of a new one to offset the disadvantages of being excluded. In some other cases, such as PTAs involving developed with developing countries, the motivation for deeper integration may have been the desire to liberalize beyond what was possible to agree in the context of the WTO negotiations—with the anticipation of multi-lateralizing agreements at a later date, as the political economy of liberalization makes it easier to move forward through a PTA rather than through multilateral negotiation. In still other 35 situations, the main motivation for increasing regional integration may stem from political/security considerations. The trend towards more PTAs by developing countries raises questions about the effects on the participants as well as third countries. There are also issues as to consistency with broader multilateral efforts to liberalize trade under the WTO. First, regarding the effects on non-participants, there is some evidence that trade diversion effects in some cases, for example MERCOSUR, can be adverse and substantial. (Chang and Winters, 2002) But as more and more PTAs involve commitments toward greater integration, the traditional trade creation/ trade diversion analysis appears to be less relevant and more attention needs to be devoted to determining the impact of other PTA provisions. In this area, one of the important developing country issues has to do with commitments they have undertaken in PTAs with developed countries to impose less flexible intellectual property rights regimes than required under the TRIPS agreement (Nkomo, 2011). Also, a number of ACP countries have failed to implement PTAs concluded with EU in part because of concerns about the ‘deeper integration’ provisions of these agreements. The question of consistency with the multilateral trading system has two aspects: a narrow (and less interesting) one which has to do with the consistency of the multitude of PTAs with Article 24 of the GATT and the Enabling Clause Provisions; and the broader one of whether the spread of PTAs benefits or hurts global welfare and whether it undermines future efforts to liberalize trade on a multilateral basis. WTO has established a Committee to review the very large number of PTAs which have been notified to it in order to determine their consistency with WTO provisions. More than 400 PTAs have been notified and many dozens have been reviewed; but the Committee has not been 36 able to conclude definitively that any of these agreements are consistent with Article 24 provisions and the Enabling Clause. The problem derives from the fact that existing WTO provisions are broad enough to give considerable latitude of interpretation. And the Committee’s work is bogged down by an inability to agree on any specific tightening in the interpretation of the existing provisions. Similar problems have been encountered in the Doha Round discussions of the issue. The lack of consensus results primarily from the desire of many members, including developing countries to have as much flexibility as possible in the conclusion of such agreements. There is thus a danger that the multiplicity of agreements will yield welfare reducing results both for the participants and for the rest of the world (Panagaryia, 2000). There is also the danger that the existing preferential arrangements will result in a reluctance to engage in multilateral MFN reductions on account of many countries losing preferences. This concern has been voiced by a number of mostly low-income developing countries in the context of the Doha Round. But it is doubtful that this concern has been decisive in the failure to reach agreement on MFN tariff reduction. Indeed, it could be argued that some of the PTAs were put in place precisely because of the lack of progress in the Doha Round negotiations. Indeed PTAs among developing countries may be considered useful training grounds for those among them wishing to ‘deepen’ their integration in other spheres of economic co-operation beyond trade and thereby may strengthen their future participation in global co-operative efforts. III. DEVELOPED COUNTRY POLICIES ON MERCHANDISE IMPORTS Since the 1980s the role of developed countries’ markets in determining demand for developing countries’ exports has shrunk, as the role of the emerging economies increased, especially in the decade of the 2000s. However, developments in developed country markets 37 continued to be an important factor that shaped developing country trade performance. On the demand side, developed countries’ GDP increased at a rate slightly lower than the long term growth trends in these countries – just under 2.5 percent for 1990-1999 and 1.7 percent for 2000- 2010 in compared to 3.1 percent in the 1980s and 3.6 percent in the 1970s (OECD 2000, World Bank 2012). On the other hand, the market access conditions for developing countries improved, partly as a result of standstills and subsequent liberalization linked to the Uruguay Round agreements and reductions in tariffs following the Information Technology Agreement, and the termination the Agreement on Textiles and Clothing and partly as a result of numerous preferential arrangements concluded between developed and developing countries. The implementation of the Uruguay Round agreements took a number of years to complete 17 and there was some retrogression following the severe economic crisis of 2008. But there is little doubt that access to developed country markets was much better at the end of the period than at the beginning. Access continued to be impeded by high trade barriers in certain sectors, such as agriculture; and while progress was made on tariff escalation, the problem persisted in certain sectors, for example textiles and leather products. More barriers are also appearing connected to the implementation of the TBT and SPS agreements. Finally, as part of the Uruguay Round agreements, WTO members committed themselves to take several explicit additional steps in favor of developing countries in anti-dumping, whose implementation is important to review. 18 A. TARIFFS 1. Overall Averages The effects of the Uruguay Round agreements on the MFN tariffs of the developed countries have been extensively considered in other studies whose main findings will be 38 summarized here (Martin and Winters, 1996; UNCTAD/WTO, 1997; OECD, 1997). Broadly speaking, MFN tariffs on imports of manufactures into the major industrial countries markets were reduced by an average of 40 percent from a trade weighted average of 6.3 percent before the Round to 3.8 percent after the Round, with the reductions to be phased in over five years and the first installment to be put in place on January 1, 1995. 19 Countries have reduced their tariff rates accordingly since then. Moreover, in 1997, following the Information Technology agreement, the duty on a number of products in this sector was reduced to zero on an MFN basis (Finger and Schuknecht, 1999). Canada undertook a further unilateral liberalization of about 1,800 lines of MFN tariffs in 2008-2009 (WTO TPR, Canada 2011). Applied rates are even lower, averaging 3.2 percent reflecting various preferential arrangements (Table 3). MFN tariffs on products other than agriculture range, for example, from an average of 0.8 percent in Japan to 3.7 percent in the US and 4.0 percent in the EU (UNCTAD/WTO, 1997, Annex Table 1). But the rates are higher on products of interest to developing countries such as textiles and shoes (Hertel and Martin, 1999). At the same time, for the Quad countries (Canada, EU, Japan, US), one third of all MFN tariff lines are duty free. The tariffication of various measures of support and protection in the agricultural sector resulted in substantial increases in the initial tariffs on a wide range of agricultural products in some major markets. Thus, the average applied MFN rate for agricultural commodities (production weighted) in 1996 ranged from 7.9 percent in the U.S to 10.7 percent in the EU (OECD, 1997, Table 3.1). Subsequently, developed countries were to reduce agricultural tariffs by 36 percent across the board, at the same time as access for agricultural products would be enhanced by reductions in domestic support measures. 39 The problem was that developed (and developing) countries chose to bind their tariffs at higher rates than the actual tariff equivalents during the years just before the conclusion of the Uruguay Round agreements (1989-1993). For example the final bindings for the EU were almost two-thirds higher than the tariff equivalents for 1989-1993 (Binswager and Lutz, 2000 p.9) and for the U.S., they were more than three quarters higher. Binding the tariffs at such high levels allowed countries to vary actual tariff rates according to the results they wished to achieve in protecting their domestic markets -- much as the EU used to do with the variable levies, which have been prohibited since the Uruguay Round. The result of this so called ‘dirty tariffication’ has not been improved market access merely that protection has become more transparent. Similarly, commitments to reduce export subsidies were made relative to a base period in the early 1990s when these subsidies were at very high levels. During the 2000s aggregate measures of support (AMS) to agriculture as well as export subsidies by the two major developed WTO members, the EU and the US, fell significantly from 1990 levels both because of budgetary constraints and because of rising prices for many agricultural commodities (Blandford, Josling and Orden 2011; Gorted, Ruis and Ingco 2003; WTO/TPR/OV/W/6, June 2012). But the overall structure of protection negotiated in the Uruguay Round agreements has remained intact- despite strong efforts by developing countries during the Doha Round discussions to obtain better treatment in agricultural products such as cotton of critical importance to some of the poorest and LDCs. 20 2. Preferences The actual tariff rates applied to imports from individual developing countries tend to be even lower than the abovementioned MFN rates, however. There are two main reasons for this: because the generalized system of preferences (GSP) further reduced tariffs on selected 40 commodities and countries; and because of the existence of preferential arrangements for particular countries in specific developed-country markets, for example, the preferences afforded to the ACP countries in the EU market, the ones enjoyed by Mexico as part of NAFTA and the Caribbean and Central American countries’ preferential treatment in the US market as a consequence of the Caribbean Basin Initiative and the AGOA initiative affecting low-income countries in Africa In addition, individual developed countries have granted even more preferential treatment to LDC imports following the WTO High Level Meeting on Trade-related Measures for LDCs in 1997 (see below). The overall margin of unilateral non-reciprocal preferences for developing countries as a group is of the order of 3.5 percentage points, (Hoekman et al 2009) but varies significantly for different groups and in different markets from 6.6 points for LDCs in the EU to 0.5 for low- income countries in the US. The various GSP programs have narrower coverage and lower margins of preferences than regional non-reciprocal programs (see Table A-1). Among the regional preferential programs the ones in Latin America and the Caribbean contain larger preference margins than the ones in Sub-Sahara Africa. Given the complexity of the various preferential systems in place, it is difficult to assess their overall effect on the average tariffs applied and most importantly on the exports of eligible developing countries. An earlier study suggested that the GSP remains a valuable tool for promoting developing country exports' (UNCTAD/WTO 1997, p.9). More recent estimates suggest that the regional non-reciprocal programs are much more important than the GSP for individual countries. 41 A number of these programs had been especially beneficial for particular countries and products because of quotas in various export markets, for example bananas and sugar in the EU and textiles in the US. The reduction of MFN rates, the spread of reciprocal preferential arrangements, the liberalization of the EU banana and sugar regimes in 2009, the termination of the Agreement on Textiles and Clothing in 2005 and the increasing limitations on product and country eligibility imposed by the 'graduation' features of various programs, have all resulted in reducing the value of the non-reciprocal preferential programs offered by developed countries. Still, the programs can be useful to individual developing countries and individual products. It is estimated (Hoekman et al 2009, p.20) that “US preferences are equal to 5 percent or more of dutiable exports for 27 countries, while in the EU case the value of non-reciprocal preferences exceeds 6 percent of dutiable exports for 16 countries.� Countries benefiting in particular include Malawi (tobacco), and Lesotho (textiles). Notwithstanding such benefits, the GSP as well as other unilateral preferential schemes may offer the wrong types of 'dependency' incentives to developing countries and retard rather than promote their integration into the international trading system (Finger and Winters, 1998; Srinivasan, 1998). 3. Peaks Despite the overall decline in the average applied MFN tariffs since the Uruguay Round, there are a number of sectors and product groups in various developed countries where tariffs are substantially higher, thus limiting market access. The Uruguay Round agreements increased the tariff dispersion in some developed country markets and decreased it in others (OECD 1997, p.18; Daly and Kuwahara 1998, p.223). But in all countries there are many products and product groups in which the average MFN applied tariff level exceeds 12 percent, or roughly three times the 42 overall average MFN applied tariff level of developed countries. These products and groups can be defined as having tariff 'peaks'. They exist in both agriculture and manufactures in a number of developed country markets. But the very high rates typically have been the consequence of tariffication in agriculture. 21 The main products in which such tariff peaks can be observed in various markets are the following: • Major staples, such as meat, sugar, milk and dairy products and cereals, where the tariffication of quantitative restrictions has resulted in tariff rates frequently exceeding 100 percent and ranging, for example up to 550 percent for rice in Japan. 22 • Cotton and tobacco: for similar reasons similarly high rates are applied to these products, with the rate on tobacco reaching 350 percent in the US. • Fruit and vegetables, including 180 percent for above-quota bananas in the EU, and 550 percent and 132 percent for shelled groundnuts in Japan and the US respectively. • Processed food products, including fruit juice, canned meat, peanut butter, and confectionery with rates exceeding 30 percent in several markets and rising to 230 percent for grape juice in the EU. • Textiles and clothing: tariff rates are in the 12-30 percent range for a large number of products in Canada, the EU and the US. • Footwear and leather products, with tariff peaks in excess of 35 percent for 10 percent of products in Japan (rising to 160 percent for shoes) and 17 percent of products of Australia (Smeets and Fournier 1998). • Selected automotive and transport sector products (trucks in the US and the EU, ships and boats in Canada) with rates exceeding 20 percent. 43 The GSP and other preferential schemes operated by the various developed countries tend to reduce, in some cases significantly, the tariff rates applicable to imports of these products from developing countries. However, in almost all cases where tariff peaks are present, the sensitivity of the domestic industry to imports has resulted in the exclusion of various products from the schemes or in some type of limitation being imposed either on the amount that can be imported under the preferential rates or the countries that are eligible. For example, the US excludes most textile products from its GSP scheme, the EU limits preferential margins and imposes country/sector quotas. Quota limitations also exist for non-traditional suppliers of various fruits and vegetables to the EU market and in Japan's market for leather goods and footwear. Some of the more recent schemes favoring LDCs however, do provide virtually total duty free and quota free (DFQF) access to a number of developed country markets. 4. Least Developed Countries The international community has made special efforts to address the market access problems faced by the LDCs. Following the 1997 WTO high level meeting on the LDCs developed countries (and some developing countries) have made additional preferential market- access commitments to LDCs. The developed country commitments were legally permissible under the GATT Enabling Clause of 1979. Those of the developing countries were formalized through a ten year waiver under the WTO decision of 15 June 1999 (WT/L/304) and extended for another ten years in 2009. At present, LDCs enjoy essentially duty-free-quota-free (DFQF) market access to Australia, the EU, New Zealand, Switzerland and Turkey. Canada offers a DFQF scheme that covers 99 percent of tariff lines, Japan 98 percent and the US somewhat less, with 34 Sub-Saharan African LDCs having a duty free but quota limited access to 96 percent of tariff lines while Asian LDCs 44 much less—as the program excludes textiles. Brazil, China, India and Korea Rep. also operate LDC preference schemes, which are considerably more restrictive than those offered by developed economies (Laird, 2012). The average tariff rates LDCs face in developed country markets generally average about 2 percent, about half those facing other developing countries. Tariffs in other developing country markets against LDC products are much higher, 14.5 percent (WTO 1997). The low average tariff rates in developed country markets reflect primarily the fact that the bulk of LDC exports are primary commodities where tariffs are low or non-existent and less the extensive preferences they obtain in these markets—which on average offer no more than few percentage points of advantage. On the other hand, the dutiable products in which LDCs tend to have comparative advantage, namely agriculture, textiles, and footwear are subject to tariff peaks and tariff escalation. For a number of products (beef, asparagus, cigarettes, processed wood, clothing and footwear) are amounting to about 10 percent of their total exports, tariffs in the Quad, Australia, Norway and Switzerland ranged between 5 and 15 percent (WTO 1997). In these cases DFQF can be meaningful as preference margins may be substantial. But for LDCs to exploit these they need to have relatively simple qualification requirements regarding rules of origin. The requirements for such rules can be prohibitive if they require substantial value added in the country benefitting from the preferences. LDCs typically do not have enough industrial capacity or the investment climate to attract the upstream suppliers needed to produce sufficient domestic value added or to transform inputs into a processed product (Francois and Hoekman et al 2009, p.7). The evidence from DFQF programs which offer both a significant 45 margin of preference and relatively liberal rules of origin (Canada, Australia and US- AGOA) show that eligible LDCs have expanded their exports significantly following the establishment of the programs (Francois and Hoekman 2011, p.9). In sum, market access problems for LDCs are concentrated in a few products and tend to be greater in developing than in developed country markets. In the course of the Doha Round negotiations a variety of proposals have been made aimed at enhancing the existing preferential arrangements benefiting the LDCs. Some focus on expanding the coverage in the US and major developing country markets. Others emphasize the importance of simplifying rules of origin. In all cases a key prerequisite is the need to address LDCs supply side constraints that limit their capacity of using fully existing preferential arrangements. 5. Escalation Tariff escalation is a matter of concern for developing countries in the context of market access because it tends to increase the rate of effective protection at higher stages of processing, thereby making market access more difficult for finished manufactured products, which in turn can adversely affect on developing countries' industrialization efforts. As a consequence of the Uruguay Round agreements the degree of overall escalation has decreased. Also, escalation is more pervasive in developing countries than in developed (see Table 4). But it continues to be a matter of concern for many developing countries and their exports of specific products, such as processed foods (wheat flour, orange juice, vegetable oils, dairy products), clothing, leather and wood products (Lindbland 1997; WTO 1998c; UNCTAD 1997a). These results should be interpreted with caution because data limitations as well as the continued existence of quantitative restrictions in some of these product chains make it difficult to calculate the effective rates of protection. But they do suggest that tariff escalation -- as with tariff peaks -- in certain products, 46 though reduced by the Uruguay Round agreements, continues to be an area of concern in respect to market access for developing-country exports. B. NON-TARIFF MEASURES Since the Uruguay Round and until about 2009 it is probably fair to say that the pervasiveness of core non-tariff measures in developed country trade regimes had fallen to its lowest point in more than 50 years. By core non-tariff barriers, we mean the use of non-automatic licensing, quotas and tariff quotas and voluntary export restraints as well as price control measures such as variable charges, minimum prices or voluntary export price restraints. Since 2009 there has been some resurgence of protectionism, but much less than had been feared, given the depth of the economic recession in developed countries. The key issues for the future relate to the use of trade remedies i.e. anti-dumping, countervailing and safeguards—all of which are permitted under the WTO agreements, as well as the expanding use of SPS and TBT measures which have an impact on trade. As noted earlier, there are considerable limitations in the data that can be used for assessing coverage and frequency ratios of the incidence of non-tariff barriers in developed or developing countries over time. 23 But the available data suggest that these barriers have declined over time. Voluntary export restraints were phased out by the end of 1998—although there have been some reports of their “informal� re-emergence in recent periods. The termination of the Agreement on Textiles and Clothing in 2005 as well as the end in 2008 of the special quantitative restrictions applied to imports from China as part of its WTO accession ended the last vestiges of formal ‘legal’ quantitative restraints in manufactures. 24 47 Non-tariff measures have been drastically reduced in agriculture, where only tariff quotas exist in a number of products in some of the major developed markets. These range from a few e tariff lines in the US and Australia up to 1.5 percent in Japan and 4.9 percent in the EU (see Table 10). 25 At the same time, the use non-automatic licensing has been reduced in all major developed country markets, and is now focusing primarily on restraints linked to the maintenance of sanitary and phytosanitary standards, technical standards or protection for the environment. Concerns however, have been raised as to whether the implementation of WTO agreements in these areas has been motivated by protection of domestic industries and a number of disputes have arisen regarding the compatibility of certain measures implemented by some countries with these agreements. The reduction of NTMs in agriculture after the conclusion of the Uruguay Round does not seem to have resulted in significant improvements in market access for developing countries for a variety of reasons. First, there were problems with the way tariffication’ was implemented, as discussed earlier. Second, the reductions of AMS have not affected the significant supports to products of interest to developing countries such as cotton, sugar and dairy products. Third, despite the commitments to reduce export subsidies, these were maintained at such high levels in 1990s as to continue to undermine incentives provided to developing country producers. Examples of the adverse effects of EU export subsidies on developing-country producers abound: subsidies in dairy products have hurt production in a large range of countries including Brazil, Jamaica and Tanzania; subsidies on tomato concentrate impact especially West Africa countries like Burkina Faso, Mali and Senegal; support for beef has undermined efforts to raise livestock production in some of the same countries; and EU beef has come to dominate the markets of Benin and Ivory Coast for which Burkina Faso and Mali were once been important suppliers. 48 These subsidies were significantly reduced in the 2000s primarily due to budgetary pressures within the EU. Still overall, there was far less ‘real’ improvement in the agricultural sector than developing countries anticipated. C. TRADE REMEDIES Over the last thirty years, as the incidence of quotas and other explicitly protective quantitative controls has diminished, the importance of trade remedies—anti-dumping, countervailing and safeguards has risen in both developed and developing countries. However, there is no evidence that their use has increased in absolute terms during the most recent financial crisis relative to what happened in 1992-1993: during those two years there were 326 and 299 total anti-dumping investigations respectively, while the peak number of investigations during the most recent crisis was 215 in 2008. In the US, which over the years has instigated the largest number of total anti-dumping investigations and still has some measures in place which were instituted more than 40 years ago, only 1.1 percent of total imports are affected by trade remedies (Table 10). At the same time, as discussed earlier the evidence is extremely clear that the share of developing countries in the total remedies has increased dramatically. Table 10 also contains some scattered information that permits us to gauge the evolution of trade remedies in major developed countries over time. In the US, the tariff line frequencies of anti-dumping, countervailing and safeguards have declined from five percent in 1996 to 1.1 percent in 2011. For the EU over a comparable period, but measured in total number of cases, it appears that there has been little change in the totals. It is also interesting to look at the situation in 2003. At this intermediate point, the number of anti-dumping measures in place for both the US and the EU as well as Canada was higher than in 2011. Japan is the only country which 49 unambiguously increased its use of trade remedies over this period. But its total use of these instruments is minimal. Table 10: Non-Tariff Measures (NTMs) and Trade Remedies by Selected Major Developed Country, 1996-2011 Australia Canada ** EU ** Japan Norway Switzerland United States ** Trade Measure 1996 2011 1996 2011 1996 2011 1996 2011 1996 2011 1996 2011 1996 2011 NTMs (in % of total tariff lines) Licencing, Non-Automatic 0 X 0 X 0.8 X 1.3 X 2.6 X 0 X 0 X Tariff Quota 0 * 0 2.2 0 4.9 0.6 1.8 0 0.4 0.2 3.5 0 * AD/CV and VEPRs 0.4 .. 0.7 .. 0.2 .. 0 .. 0 0 0 0 5 1.1 Trade Remedies (no. of measures in Force) Antidumping .. 22 .. 33 143 140 0 6 0 0 0 0 .. 253 Countervailing .. 3 .. 9 2 11 0 0 0 0 0 0 .. 50 Safeguards .. 0 .. 0 0 0 0 0 0 0 0 0 .. 0 Total AD/CV/SG .. 25 .. 42 145 151 0 6 0 0 0 0 .. 303 Australia Canada ** EU ** Japan Norway Switzerland United States ** Trade Measure 2007 2011 2007 2011 2007 2011 2007 2011 2007 2011 2007 2011 2007 2011 Trade-Weighted Share of Non-oil Imports (in %) *** Antidumping .. 0.4 .. 0.7 .. 1.7 .. 0.1 0 0 0 0 .. 3.9 Total AD/CV/SG 0.5 0.5 0.7 0.7 3.8 1.8 1.1 0.1 0 0 0 0 3.3 4.0 Notes: X = licensing for health, sanitary, safety, environmental and/or defense. * = less than 1%. ** = antidumping measures in effect in 2003 for Canada, 91; EU, 156; and US, 279. *** = Data are taken from Bown, C. (2012), Emerging Economies and the Emergence of South-South Protectioniem, PRD #6162, World Bank. (Table 1 and Appendix Table A.1) .. = data not available Sources: WTO Trade Profiles database, WTO TPRs, and OECD, 1997 (table 5.1); and Bown (2012). The lower panel of Table 10 shows the changes that have been in the use of trade remedies by developed countries over the period 2007-2011. The trade weighted share has increased in this period for the US primarily because of the use of additional countervailing measures, but has decreased both for the EU and Japan. 26 The overall conclusion from this analysis of the frequency of use of trade remedies by developed countries is that, while still substantial, it is certainly not more extensive than in the nineties. Combined with the earlier discussion of the declining trends in the use of other core non- 50 tariff measures, it would appear that developed country markets in 2011 were definitely no less and probably more open than they were in the mid-1990s. The major product groups which are the object of anti-dumping investigations are much the same for developed as for developing countries: Table 11 shows that Basic Metals, and Chemicals, are the two sectors which account for the bulk of anti-dumping investigations, between 40 and 50 percent, of all anti-dumping investigations in both developed and developing countries. Over the last thirty years, little has changed in this respect which may simply reflect generic pricing practices by firms operating in these sectors. Table 11: Sectoral Distribution of Antidumping Initiations by Developed and Developing Countries 1987-1994 1995-2001 2002-2011 Developed Countries Developing Countries Developed Countries Developing Countries Developed Countries Developing Countries HS Section No. of Cases % Share No. of Cases % Share No. of Cases % Share No. of Cases % Share No. of Cases % Share No. of Cases % Share I Animals and products 8 0.6 17 2.2 11 1.3 13 1.2 10 1.6 20 1.4 II Vegetable products 11 0.8 18 2.3 12 1.4 16 1.5 8 1.2 20 1.4 III Fats, oils and waxes 4 0.3 13 1.7 2 0.2 2 0.2 3 0.5 7 0.5 IV Prepared foodstuff and tobacco 52 3.7 9 1.2 17 2.0 20 1.9 5 0.8 13 0.9 V Mineral products 41 2.9 11 1.4 20 2.3 26 2.5 9 1.5 18 1.2 VI Chemical products 210 15.0 157 20.2 140 16.4 183 17.1 154 25.0 348 23.6 VII Plastics and rubber articles 147 10.5 102 13.1 105 12.2 130 12.2 81 13.2 197 13.4 VIII Hides, skins and leather products 8 0.6 1 0.1 1 0.1 2 0.2 0 0.1 2 0.1 IX Wood and pulp 16 1.1 8 1.0 13 1.5 15 1.4 18 3.0 45 3.0 X Paper, paperboard and articles 65 4.6 47 6.0 42 4.9 48 4.5 36 5.8 82 5.6 XI Textiles and articles 95 6.8 55 7.1 61 7.1 77 7.2 43 7.0 122 8.3 XII Footwear and headgear 24 1.7 9 1.2 8 0.9 10 0.9 4 0.6 10 0.7 XIII Glass, stone and ceramic 59 4.2 15 1.9 26 3.0 31 2.9 29 4.7 67 4.6 XIV Pearls and precious stones 0 0.0 0 0.0 0 0.0 1 0.1 0 0.0 0 0.0 XV Base metals and articles 374 26.7 178 22.9 274 32.1 342 32.0 148 24.1 339 23.0 XVI Machinery and electrical equipment 212 15.2 83 10.7 82 9.6 103 9.6 44 7.1 120 8.2 XVII Vehicles, aircraft and vessels 27 1.9 7 0.9 9 1.0 11 1.0 6 0.9 15 1.0 XVIII Instruments, clocks and recorders 19 1.4 21 2.7 12 1.4 13 1.2 6 1.0 17 1.1 XX Other miscellaneous manufactures 27 1.9 27 3.5 20 2.3 25 2.3 12 1.9 29 2.0 Total -- above all sectors 1,399 100.0 778 100.0 855 100.0 1,068 100.0 615 100.0 1,472 100.0 Note: Exclude arms sector XIX Source: WTO antidumping database. 51 It is important to recall that during the Uruguay Round and in the context of the Agreement on the Implementation of GATT Article VI on anti-dumping, the developed countries committed themselves, to give special regard to the special situation of developing county Members when considering anti-dumping measures under the Agreement. Article 15 urges the use of the constructive remedies provided for in the agreement before applying anti-dumping duties that will affect the essential interests of developing-country members. There is no guidance regarding how this article is to be implemented, particularly in respect of the way in which 'the special situation of developing country members is to be taken into account'. The article does not appear to commit developed countries to do anything more than make use of the 'constructive remedies' mentioned in the agreement before applying anti-dumping duties. But presumably, they would have to do this in all cases, not only those involving developing countries. In practice the results have been mixed. Most of the developed country anti-dumping investigations and definitive measures against developing countries have been directed at higher and middle-income developing countries – (often the same countries that have themselves made increasing use of anti-dumping measures themselves) -- and non-WTO Members (see below). China and Chinese Taipei in particular have been a target both before and after their accession to the WTO, both by developed and other developing countries. Bangladesh is the only LDC to have been subjected to anti-dumping investigations and definitive measures on three occasions, in 1992 (see Miranda et al, 1998). Table 12 shows that developing countries are more frequently the target of anti-dumping actions by both developed and by other developing countries. Developed countries, on the other 52 hand, are the least targeted group both by other developed countries and developing countries. In particular, the share of overall investigations in which developing countries are affected is much higher than the share of developing countries in world exports (see Table 12). The ratio of the two shares, Rad, (the share of total anti-dumping investigations against a particular country or group relative to its share of world exports) should be interpreted with care because the proportion of measures may not accurately reflect the actual amounts of exports affected. Nevertheless, it would be difficult to conclude from the data in Table 12 anything other than that the developing countries have sustained a disproportionate amount of anti-dumping measures over the last twenty years. Rad investigations for developing countries had a value of 1.5 compared to a Rad of 0.6 for the developed countries for the period 1987-1994. The value of the ratio progressively increased for developing countries in the period 1995-2001 and 2002-2011, while it decreased for developed countries. This means that developing countries, over the last twenty five years, were more than twice as likely to have their imports affected by an anti- dumping investigation (relative to their share in international trade) than developed countries (see also Table B-4). 53 Table 12: Antidumping Measures by Affected Country 1987-1994 1995-2001 2002-2011 AD Worl d Exp Ra ti o of AD Worl d Exp Ra ti o of AD Worl d Exp Ra ti o of Affected Country / Group Ms re (%) Sha re (%) Ra d Ms re (%) Sha re (%) Ra d Ms re (%) Sha re (%) Ra d WTO Members 74 88 0.8 63 89 0.7 93 98 0.9 Devel oped countri es 41 66 0.6 33 68 0.5 23 62 0.4 Devel opi ng countri es 33 22 1.5 30 21 1.4 70 36 1.9 Devel opi ng co. excl . Chi na .. .. .. .. .. .. 39 27 1.5 Non-WTO members 26 12 2.2 37 11 3.4 7 2 3.5 Memo Items Chi na ** .. .. .. 17 3 5.7 31 9 3.4 Chi nes e Ta i pei ** .. .. .. 5 2 2.5 6 1 6.0 Former FSU .. .. .. 10 2 5.0 5 3 1.7 TOTAL: Al l countri es 100 100 100 100 100 100 Notes : .. = da ta not a va i l a bl e. ** Chi na a nd Chi nes e Ta i pei beca me WTO members a fter 2001 a ccordi ngl y. Ra d = the ra ti o of s ha re i n tota l a nti dumpi ng i nves ti ga ti ons rel a ti ve to the s ha re i n worl d exports . Sources : WTO a nti dumpi ng da ta ba s e a nd Worl d Ba nk WDI da ta ba s e (tra de da ta ). The Table also shows the very large incidence of anti-dumping actions taken against China, Chinese Taipei and non-WTO members, especially so called 'non-market economies' mostly countries other than the Baltics that emerged from the former Soviet Union, in particular Russia, Ukraine, Belarus and Kazakhstan which have seen their exports being targeted for anti-dumping measures at a far greater rate than their share in world trade. Indeed, controlling for the value of total exports over the last three decades anti-dumping measures were at least four times more likely to be directed against a product from a non-market economy which is not a WTO Member than a product from a developed market economy. These conclusions are entirely consistent with the Bown’s findings (Bown, 2013, p21) regarding the changing use of trade remedies against different countries over time. He shows a decreasing but still substantial use of trade remedies against countries like Ukraine, Russia and Kazakhstan between 2001 and 2011 and an increasing use against China and Korea. 54 China’s and Chinese Taipei’s situation is even more pronounced as it has been a target of anti-dumping measures much more often than other countries or its share in world trade would justify. The situation has not changed after they became WTO members, unlike a country like Ukraine which had been targeted a great deal before its accession and much less afterward. Table 12, suggests that countries taking anti-dumping action may feel less constrained when taking action against non-WTO Member countries. And with respect to so called non-market economies, there is evidence indicating that the procedures used tend to be more opaque and may well lead to a greater incidence of definitive findings than those against other economies (Michalopoulos and Winters 1997). What can be concluded from this analysis of the prevalence of anti-dumping actions against developing countries as an indicator of market access to developed- country markets? There appears to have been be an improvement in recent years, because in general fewer anti-dumping actions have been taken by developed countries; but the developing countries’ share of these actions, taking account of the value of their exports, has not changed and tends to be disproportionately high. This reflects an insufficient response by the developed countries to their commitment 'to give special regard to the special situation of developing countries'. Similarly, there is no evidence that the Article 15 on using constructive remedies before applying anti- dumping duties on imports from developing countries has been employed. D. SPS AND TBT The implementation of the SPS and TBT agreements has raised two sets of issues for developing countries: First, standards along with testing and certification represent between 2 and 10 percent of overall product costs (Maskus et al 2005). Thus, they impose a burden on 55 developing country exports -- even when the standards are used for legitimate reasons and the countries are able to meet them. Second, while the two agreements may make developed countries actions more transparent and force a degree of accountability they may also serve to legitimize developed country non-tariff measures that unnecessarily restrain developing countries’ trade. The pioneering work of Otsuki, et al. (2001 and 2002), estimated that a 1998 EC regulation that imposed more stringent controls on the maximum level for certain types of aflatoxin, (a toxic substance) found in foodstuffs and animal feed, than those required by the Codex Alimentarius, would cost close to $700 million in lost revenue to African exporters of groundnuts. Wilson et al (2003) found that the stringency of antibiotics regulation in developed countries significantly reduced bovine meat exports for Argentina, Brazil and South Africa. And several studies have concluded that SPS regulations inhibit developing-country agriculture and food exports (Cadot and Malouche 2012, p.6). Early complaints by developing countries that developed countries may be using the SPS agreement as disguised protection had focused on meat (Burkina Faso), fresh fruits and vegetables (Kenya), canned tuna (Papua New Guinea), fish (Uganda). In 2001-2012 more than one thousand notifications of new SPS measures and a slightly larger number of TBT notifications were submitted to the WTO. About 60 percent of the SPS notifications and the bulk of the TBT notifications came from developing countries. As the overall developing-country share of world trade as well of SPS and TBT notifications is increasing, it is clear that this may not be a North-South problem but a general NTM issue. WTO members have an opportunity to complain about the introduction of new SPS and TBT measures in the respective WTO Committees by raising specific trade concerns (STCs). The 56 STCs submitted to the TBT Committee per annum more than quadrupled (from 17 to 76) in the period 2000-2011. 14 new STCs were raised in the SPS Committee in 2011 about half by developing countries. Some of these concerns were left unresolved and have led to cases going to WTOs dispute settlement although the number of cases in dispute settlement which specifically refer to SPS and TBT is relatively small (WTO 2012, p.111). In light of the increased use of SPS and TBT measures, it has been argued that all of the measures imposed on these grounds could be considered as NTMs similar to quantitative restrictions and trade remedies (Gourdon and Nicita, 2012). If such a definition were used, then SPS and TBT measures would constitute the bulk of developed (and developing country) NTMs. On average countries apply TBT measures on approximately 30 of products and trade and SPS on 15 percent of products and trade. Taken together these measures would cover almost half of a typical country’s trade. And they are likely to be much more extensive than core non-tariff barriers such as quotas, trade remedies and non-automatic licenses. But it would not be useful to lump together TBT and SPS measures together with other non- tariff barriers for two reasons: first, it is quite clear that there are very legitimate reasons for introducing such measures for health and safety reasons; and when they are abused there is a process for seeking redress—even though this process poses some constraints for developing countries. Second, it is possible that the establishment of standards can actually benefit developing country exporters. A study of Chinese exports of agricultural products showed that the adoption of more stringent standards had a positive effect as the signaling of enhanced food safety and quality in foreign markets apparently surpassed compliance costs (Manzelsdorf, Perez and Wilson 2012). Similarly a recent study by Xiong and Beghin (2011) found that the predictions by Otsuki et al. of a dire effect of the introduction of EU regulations on aflatoxin content on West 57 African groundnut exports did not materialize. Instead, they posit that the main constraints on these exports came from the supply side and that in the case of edible groundnuts the regulations may have actually stimulated trade by enhancing the safety of the products in the eyes of EU consumers. Thus, while the impact of SPS and TBT regulations on developing country exports is quite complex and they cannot all be considered as NTMs used for protection of domestic industry, these regulations certainly have the potential of doing so. The problem is that in defining these as NTMs it is necessary to identify intent—which is inherently difficult to demonstrate, as well as resolve a variety of complex scientific issues. The WTO undertook an exhaustive analysis of NTMs and especially SPS and TBT measures in its 2012 annual trade report. Its conclusions are important to report here. It states (p.143) “The existing empirical literature finds that at the aggregate level TBT/SPS measures may not be associated with lower trade�, (also p.153) “TBT/SPS measures do not unambiguously increase or decrease trade. In general TBT/SPS measures have positive effects for more technologically advanced sectors, but negative effects on trade in fresh and processed goods.� Given these difficulties, it is important not to lump TBT and SPS measures with other non-trade barriers which quite unambiguously restrict trade such as quotas or trade remedies. To do so will result in ‘crying wolf’ too often and missing true increases in protection. The only recommendation is for continued vigilance, monitoring and transparency through timely reporting and discussion in the WTO and other relevant international bodies entrusted with the setting of these standards. 58 IV. SERVICES A. Conceptual and Measurement Issues There are no comparable international data that permit a systematic examination of the state of policies regarding restrictions on trade in services in developing or developed countries. The reasons are well known: because many of the modes of delivery of services are intangible, there are no barriers that take the form of tariffs. Barriers take the form of quantitative restrictions -- sometimes involving complete bans, and many times government regulation. Some of the latter may be explicitly discriminatory against imports of services or foreign providers; others may be of a general regulatory nature that applies equally to national and foreign services providers. And to date no general description of the barriers to trade in services that cuts across all four modes of delivery (cross border supply, consumption abroad, commercial presence, temporary entry) and all sectors has been attempted -- and may not be feasible with the present state of data collection. Some general impressions can be obtained as to what the actual policy situation tends to be by using a recently developed World Bank data base whose results are summarized in Table 13. This database describes a new initiative to collect comparable information on services trade policies for 103 countries, five sectors (telecommunications, finance, transportation, retail and professional services) and the key modes of service supply. It contains richly textured policy information as well as a preliminary quantification of policy measures across a range of service sectors and the relevant modes of service delivery (Borchert, Gootiiz and Mattoo, 2012). The first impression from the existing data is that developed countries have liberalized far more than developing countries, especially in the financial, telecom and retail services sectors. The differences are smaller in the transport sector and even less in the professional services—which 59 frequently requires the movement of natural persons. The resultant database reveals interesting patterns in policy. Across regions, some of the fastest growing countries in Asia and the oil-rich Gulf states have the most restrictive policies in services, whereas some of the poorest countries are remarkably open. Across sectors, professional and transportation services are among the most protected in both industrial and developing countries, while retail, telecommunications and even finance tend to be more open. Service protection in developing countries can have serious implications on efficiency and growth: protection causes high cost expensive service inputs which result both in waste of resources and adverse effects on export performance of both goods and other services. A protected and inefficient banking sector can be as damaging to profitability of exportables as the imposition of high tariffs on imported components used in the production of these same exportables. Among developing countries, there is strong evidence that countries in Latin America and Caribbean have liberalized far more than developing-country regions in all sectors. In retail and transport their liberalization is at least as great as that of developed countries. There are few differences in the degree of service liberalization by sector in other groups of developing countries classified by region or income level. Some low-income developing countries, such as Ghana and Nigeria have relatively liberal regimes while several economies in Asia with fast growing export sectors such as China, India, Indonesia, Malaysia and the Philippines have restrictive service sectors (see Table 13). In telecommunications, most countries have introduced some competition by private providers in both mobile and fixed services, although foreign providers are often limited in Asia. 60 State ownership of banking is being reduced, but the allocation of licenses “remains opaque and highly discretionary� (Borchert, Gootiiz and Mattoo 2011, p121). As to the trend over time, it is clear that developing countries are moving, in general, towards more liberal regimes in services. A large number of developing countries made forward liberalization commitments in the Uruguay Round agreements on Financial Services and Telecommunications with many of the commitments to start in 2000. Later year TPRs confirm the liberalizing trend for most countries. 61 Table 13: Sectoral Service Trade Restrictiveness Indices (STRI) in Developing Countries STRI (0 = open; 100 = close) Country Overall Financial Telecom Retail Transport Professional Argentina 17 10 0 0 22 49 Bangladesh 44 46 63 25 63 35 Benin .. .. .. .. .. .. Bolivia 14 19 25 0 10 21 Brazil 23 36 0 0 10 58 Cameroon 26 21 0 25 30 43 Chile 23 22 25 25 19 27 China 37 35 50 25 19 66 Colombia 18 25 50 0 4 34 Costa Rica 29 29 38 0 30 60 Cote d'Ivoire 26 14 25 0 31 68 Dominican Rep. 12 3 0 0 22 33 Egypt 52 43 25 50 50 82 El Salvador .. .. .. .. .. .. Fiji .. .. .. .. .. .. Ghana 18 25 25 0 6 44 Honduras 21 7 50 0 35 34 Hong Kong, China .. .. .. .. .. .. India 66 48 50 75 62 88 Indonesia 50 23 25 50 66 76 Jamaica .. .. .. .. .. .. Kenya 30 23 25 0 31 73 Korea, Rep. 23 2 50 0 21 66 Kuwait 52 42 75 50 50 57 Malawi 34 35 50 25 32 38 Malaysia 46 45 25 25 55 73 Mauritius 17 9 0 0 31 42 Mexico 30 15 38 0 62 43 Morocco 21 14 25 0 28 46 Mozambique 19 17 75 0 6 30 Nepal 43 23 50 25 56 76 Nigeria 27 26 25 25 24 36 Pakistan 28 49 13 0 25 48 Paraguay 16 22 38 0 5 25 Peru 16 42 0 0 3 28 Philippines 54 45 50 50 44 80 Saudi Arabia 43 46 25 25 39 72 Senegal 19 16 25 0 25 37 Singapore .. .. .. .. .. .. South Africa 35 20 25 25 41 62 Sri Lanka 38 24 50 25 46 57 Taiwan, China .. .. .. .. .. .. Thailand 48 49 50 25 47 74 Tunisia 45 33 25 25 55 79 Uganda 35 28 25 50 21 38 Uruguay 28 45 63 0 41 11 Venezuela 35 23 25 25 32 69 Viet Nam 42 41 50 50 39 32 Zambia 21 8 75 0 10 44 Zimbabwe 64 56 63 75 68 60 Average 32 28 35 19 33 51 Memo Items: EU-20 26 4 0 25 37 54 United States 18 21 0 0 8 54 Note: Data is based on the latest year available from 106 countries in the database. Source: World Bank, DECTI Service Trade Restrictions database. 62 Following the Uruguay Round agreements, the situation regarding the formal commitments of developing countries to maintain a liberal services trade regime parallels the situation with respect to tariffs: in both cases the actual policies are far short of the commitments. Except in the case of services the difference is often greater. Table 14 shows that actual policies in developing countries in all major regional groupings restrictive as they may be, are still far more liberal than their commitments—just like applied tariffs are far lower than bindings. As with tariffs, developing countries argue that this is needed to maintain policy flexibility – and the possibility of pursuing other social objectives. But the cost, as with tariffs, is the increased risk of policy change which often inhibits foreign investment. Table 14: Services Trade Liberalization Commitments by Income Group and Region Overa l l Offer Overa l l Bi ndi ng Tota l Servi ces Fi na nci a l Improvement Ga p Li bera l i za ti on Servi ces Actua l (UR Commi t - As % of UR (UR Commi t - As % of Actua l Commi tments (No. of commi t Country Group (No. of Co.) Pol i ci es Doha Offer) Commi tments Actua l Pol i ci es ) Pol i ci es a t UR a s % of ma x.) INCOME GROUP Devel oped countri es (27) 18.0 6.6 18% 18.9 105% 57.8% .. Devel opi ng countri es (65) 29.5 5.5 8% 42.6 144% 16.5% 24.0% Hi gh i ncome Devel opi ng (7) 38.7 3.7 5% 30.1 78% 18.9% 23.4% Upper mi ddl e i ncome (21) 29.2 8.7 12% 40.5 139% 25.5% 26.4% Lower mi ddl e i ncome (18) 27.8 -2.5 -4% 43.4 156% 14.4% 27.8% Low i ncome (19) 28.2 .. .. 48.8 173% 14.7% 27.6% Of whi ch: LDCs (13) 27.8 .. .. 57.1 206% 3.0% 15.0% Worl d 21.0 6.4 13% 27.2 130% 21.7% .. REGION OECD (20) 14.0 9.4 46% 13.1 88% .. .. Ea s t As i a (7) 36.8 1.7 3% 26.0 71% .. .. South As i a (3) 36.0 15.5 18% 47.8 133% .. .. La ti n Ameri ca (9) 16.6 6.3 10% 48.2 290% .. .. Other Europe & C. As i a (5) 10.6 5.4 19% 9.8 93% .. .. Mi ddl e Ea s t & N. Afri ca (5) 36.9 3.8 7% 21.2 57% .. .. Sub-Sa ha ra n Afri ca (6) 16.9 0.7 1% 53.6 318% .. .. Sources : Worl d Ba nk Servi ces Tra de Res tri cti ons da ta ba s e a nd Gooti ze & Ma ttoo (2009), Servi ces i n Doha : Wha t's on the Ta bl e? Worl d Ba nk PRD Worki ng Pa per, no. 4903. 63 Looking at the pattern of commitments in the UR by the range of commitment and degree of development, a similar pattern develops to that observed in trade policy on goods: the degree of liberalization appears to increase with the level of income -- as shown in Table 14. Lower-income countries appear to have committed to much less liberalization than higher-income ones. The same conclusion can be drawn by looking at the commitments of the fifty developing countries whose trade regimes were analyzed more systematically. Indeed, the pattern in Table 14 has a striking parallel to the pattern in Table 3 above, with the lower the developing country income, the lower the number of commitments and hence the highest the remaining protection. (See also Borchert, Goortiiz and Mattoo 2011, p.123.) The basic justification low-income countries make for not liberalizing their service sector is the same infant industry argument used for so very long in the areas of merchandise trade. There are obvious dangers and limits to such a strategy as many developing countries have realized in the areas of goods. These dangers have to be seriously evaluated by low-income developing countries which continue to protect their service sectors. On the other hand the table shows that this relationship between the extent of liberalization and the level of development did not hold true for the financial services sector which includes banking and insurance. In this case there is basically no pattern discernible, with most groups of countries liberalizing about a quarter of the maximum possible -- if one weighs partial restrictions in each of the modes of supply. Again the LDCs made the fewest commitments. Only nine of the twenty nine LDC made any commitments. But those which did, on average made greater liberalizing commitments than other developing countries. Earlier analysis (Mattoo 1998) suggests that in the area of commitments on commercial presence for financial services, Latin American countries tend to limit the number of suppliers, while Asian economies limit either solely the percentage of equity or both equity and the number 64 of suppliers; and they are both more restrictive than African countries in this respect. This is actually an encouraging sign for African countries, especially if the pattern is repeated for other sectors, because of the implications it has for attracting FDI more generally. There are no systematic investigations of developed country policy regarding trade in services and the implications of restrictions they impose on potential exports from developing countries. Existing data suggest that their overall policies on service imports are more liberal than those of the developing countries and that the gap between actual policies and Uruguay Round commitments is smaller. But, there is one mode of delivery, the movement on natural persons, in which they tend to be more restrictive than developing countries regardless of the sector. Liberalization in this area would be of considerable benefit to developing countries because of the latter’s comparative advantage of providing labor intensive services across all modes of supply. In addition, there is one sector, maritime services, where international competition is extremely limited and there are serious constraints deriving from national legislation that actively discriminate against foreign suppliers. While some progress has been made regarding the movement of qualified professionals to work abroad, developed country restrictions inhibit increased service earnings for developing countries through this mode of supply. The commitments on trade in services have tended to emphasize measures regulating commercial presence -- which is important for foreign direct investment, rather than 'mode four' involving movement of natural persons. There are various kinds of restrictions involving, quantitative restrictions on the number of persons that are provided with visas annually. There are also restrictions deriving from the existence of qualification and licensing requirements of professional organizations formally discriminating against foreign qualified professionals. Still other barriers involve wage matching 65 requirements such as those requiring wages paid to foreigners to be the same as those to nationals or providing for local training. Also, the commitments made regarding the movement of natural persons have primarily involved intra-corporate transferees and business visitors and, to a lesser extent, independent professionals, including those providing services within a service contract. One of the limitations imposed by a large number of countries is an economic needs test (ENT). This typically involves judgments by government agencies based on non-transparent criteria, as to market conditions, availability of local service providers and so on, regarding which foreign service providers to permit and which not. Indeed of the 54 countries which have made commitments subject to a needs test, only three have stated criteria for the ENT. Frequently, the result is to nullify access commitments involving 'Mode Four' supply of services. There are many sectors in which developing countries have a comparative advantage, usually based on labor costs, and which can benefit from developed country liberalization, in particular, software development and construction services. The former is an area where developing country exports, for example from India, have been expanding very rapidly in recent periods. A lot of these exports involve on-shore delivery because of the importance of the need for continuous contact between client and programmer. While technological and managerial innovations may result in the share of on-shore delivery to decline, and the increases to come mainly through cross border trade, the increases in software demand from developing countries may be so high as to require continued liberalization of existing limitations in 'Mode Four' restrictions. Similarly, several developing countries, especially in Asia, have the capacity of exporting construction services based on their comparative advantage in labor intensive activities which are constrained by developed country restrictions on 'movement of natural persons'. 66 In maritime services there are extensive government restrictions in cabotage and cargo handling. At the same time oceans continue to be populated by cartels known as shipping conferences. These cartels set prices and pursue other collusive activities in the substantial portion of maritime services they control and they are often exempted from antitrust law in developed countries (Francois and Wooton, 1999). Their impact in raising transport costs to poorer developing countries, especially to low volume, high distance destinations -- in Africa and poorer island economies can be even more important than further tariff liberalization: shipping margins on merchandise trade in Sub-Saharan Africa exceed 6 percent compared with OECD tariffs (after preferences are taken into account) of less than 2 percent (Francois and Wooton, 1999). Liberalization in this sector, which would lead to increased competition and reduced margin may be of great importance to many of the small economies members of the WTO. V. THE WORLD SINCE 2008 In October 2008, with the world gripped by financial panic WTO Director General Lamy decided that the WTO Secretariat would monitor trade restrictive measures taken in response to the crisis using the WTO Trade Policy Review Mechanism. In November 2008, just after the US presidential election, leaders of the then new G-20, fearing a repetition of the experience of the great depression and its transmission across frontiers by trade restricting beggar-thy-neighbor policies, promised to refrain from imposing new import barriers, export restrictions and WTO- inconsistent measures. Subsequently the G-20 also promised to minimize the negative impact of stimulus measures on trade and asked the WTO to monitor and publicize the extent to which they adhered to their promises. 67 In this manner a new reporting system of trade measures taken by both developed and developing has emerged. The WTO has produced a number of reports on new trade restrictive as well as trade liberalizing measures undertaken by WTO members worldwide. Initially, quarterly reports were prepared. In more recent periods they have been produced on a semiannual basis. The sources for the reports are official notifications as well as press stories and third party complaints. The Secretariat attempts to verify each measure and reports, both those that it has been able to verify-- which are the vast majority-- and those it has not (See WTO 2009 and WTO 2012). In parallel with the WTO a private data base has been developed by Global Trade Alert (GTA), an independent group of experienced analysts who have published regular reports on trade measures: They classify them in three groups: Green Alerts are measures that increase liberalization or improve transparency; Amber Alerts are measures which have the potential of being discriminatory against trade; and Red Alerts which involve measures that discriminate against or restrict trade (Evenett 2009 and 2012). The novel feature of both sets of reports is that they extend the definition of trade measures far beyond the traditional border actions involving tariffs or other restraints. Both include for example measures that aim to provide general economic stimulus or strengthen the financial system especially as such measures may discriminate against foreign suppliers or banks. GTA also includes on occasion, but not always, TBT or SPS measures as well as migration measures. In both cases the result may be restrictive—but it is unclear whether the action was justified by a health or other hazard or in the case of migration whether they should be included at all. Similarly giving the same weight to a German ‘Rescue aid scheme for the 68 manufacturing industry’ amounting to a grand total of 5 million Euros for six months and to the US Trade Adjustment Assistance Act which affects tens of millions of workers for years makes a mockery of any quantitative analysis of this information. At the same time there are very few reports on actions that inhibit trade in services. Both sets of reports suffer from a shortcoming which is also part of their strength: in an effort to be all inclusive they contain a number of notifications which are temporary in nature: for example, as part of the WTO agreement on agriculture, many countries impose seasonal or temporary tariffs on imports of various fruits and vegetables as well as other foodstuffs, which they subsequently reduce: all these show up as trade restrictive or liberalizing measures-although in fact nothing has changed in the underlying policy or regime. The other shortcoming is that they both focus on changes in policy, not on the overall policy set—which can be obtained, albeit with some delay only through each country’s TPRs. The number of measures counted by the two sets of reports, as well as their conclusions tend to be similar, though their tone is very different: The GTA reports tend to be far more alarmist while those of the WTO are much more measured (see Wolfe 2011). Since the actual situation has evolved much more in keeping with the WTO reports and because they contain an effort to verify reported measures, they are to be preferred and their findings are used as a basis of the remaining discussion. 27 Over the whole period 2008-2012 more than 1000 verified trade measures were reported by the WTO and the GTA. Of these more than 60 percent were trade-restrictive. When these measures are cumulated over time they are estimated to impact about 4 percent of world trade. It is impossible to say how much of an impact. Also, note that this 4 percent has not been offset by 69 the amounts of trade that has benefited from liberalizing measures. For example, a number of the reported ‘measures’ took the form of termination of anti-dumping measures previously in place. The actual measures taken in the early part of the period did not appear to have a major impact on trade “to date we have not observed large scale increases in the level of discrimination by major trading states� (Evenett, Hoekman and Cattaneo 2009, p.5). Similarly Messerlin writing in late 2009 cautioned that “evoking a rise in protectionism in April 2009 was premature—still is� (Messelrin 2009, p.25). He goes on to argue that raising the fear of protectionism when few protectionist measures have been taken makes it harder to resist true protectionist measures when they actually materialize. Another writer (Wolf 2011, p.13) concluded that the great collapse of trade in 2008 was not due to any failings in the trading system and ‘the amount of trade affected in relation to actual trade flows was trivial’. Dadush et al (2011) have similar views. Finally, it appears that the trend, at least until the end of 2012 was towards greater rather than less liberalization: in the period mid-May 2012 to mid-October 2012, there were more liberalizing measures reported than restrictive ones. This does not mean that all is well in the trading system: a number of countries continued to take additional trade restrictive steps: Argentina has been the chief offender in recent periods resulting in a number of new complaints in the DSM. And in the United States there is an anti-dumping measure – which was supposed to be ‘temporary’ which has been in place for over forty years. As in previous years trade remedies are the main restrictive measures reported. In the period between 2007 and 2009, there was an increase of about 25 percent in remedies. The amount of trade affected was only 0.3 of total world trade. And it was directed on developing countries by developing countries; and the focus was on China (Bowen 2010, p.81). Even here the 70 news was not all bad: In the last two years, anti-dumping investigations have fallen back to levels much lower than before. There were a total of 155 AD investigations reported in 2011, which was the lowest number since 1989! And in the period May-Oct 2012 there were 46 new investigations and other trade restrictive actions but an almost equal number 44 liberalizing ones (WTO 2012). Similarly, there were export restraining actions and a significant number of tariff reductions, especially of capital goods and intermediates by Brazil. But there was no evidence that developed countries were more responsible for trade restrictive measures than developing ones. In sum, the increasing attention devoted to trade policy is likely to increase awareness and transparency which is a good thing not only in developed but increasingly in developing economies. But care should be taken to focus more explicit attention to specific obviously restrictive measures. Enlarging the scope of the definition of non-tariff measures is not likely to be helpful unless detailed information about the motivation for the measures and the manner in which they are implemented is available. There is no way to inoculate the population against the protectionist virus which infects many with the notion that a specific measure would help “us’ against “them� while undermining the health of the whole economy. And in keeping with the theme of the whole study, it is important to look with care at explicit practices in the service sector. It is the most important sector in developed economies and the most rapidly expanding sector of developing ones. Yet the information about recent measures taken by governments that limit competition with foreign providers is very limited. As this is being written, the future trade environment is clouded by the obvious slowdown in trade during 2012 and weak economic prospects in major economies, including especially the EU but also China and India. So far the experience with increased protection has not been too bad. Whether it will continue in the future remains to be seen. 71 The other, somewhat disquieting trends, is the increased emphasis being placed on new preferential arrangements especially among major economies as opposed to a serious effort to conclude the Doha multilateral agreement in the WTO. Three major agreements are at various negotiating stages in the first quarter of 2013: (a) a preferential agreement among 46 countries, mostly OECD members but with the participation of a number of developing countries from Asia and Latin America; (b) the Trans-Pacific Partnership (TPP), a preferential trade agreement with major participation from North and South America, Australia, and East Asia, including probably Japan; and an EU-US trade and investment agreement. VI. CONCLUSIONS Several policy conclusions have emerged in the last decade that deserve international attention. First, the traditional developed-developing country issues are becoming far less important by comparison to issues that arise from the emergence of China as a very important low cost supplier of products worldwide. Second, applied tariffs as well as traditional core non-tariff measures such as non- automatic licensing have declined significantly over time as instruments of protection in both developed and developing countries. But low-income countries tend to have more restrictive merchandise trade regimes. Third, the means of protection used by developed and developing countries—especially, middle and high-income ones are becoming increasingly similar: trade remedies, especially anti- dumping are becoming the instruments of choice for all except for low-income developing countries. 72 Fourth, low-income countries have to give the highest priority to strengthening their institutional capacity and address supply constraints. With the exception of agriculture, the main problems they face in international markets are protection in other developing countries rather than in the OECD countries. Fifth, SPS and TBT are being increasingly used by both developed and developing countries but their protective intent is difficult to gauge. This implies the need for increased vigilance, transparency and reporting to ensure that they are not being used as a means of protection of economic interests as opposed to their original intent. Sixth, it is quite obvious that liberalization in services has lagged much behind liberalization in goods; and that commitments countries have been willing to make in this sector are even less than in goods. This suggests that the service sectors are the most promising area where efforts for further liberalization may produce results. Seventh, regional and other partial preferential arrangements are both a cause and a result of the lack of progress in multilateral trade negotiations. But whatever their origin and motivation they violate the basic WTO tenet of MFN and thus pose both a potential threat to the multilateral system and a potential stimulus to further multilateral collaboration. 73 APPENDIX A COUNTRY GROUPINGS There is no formal ‘developing country’ definition in any of the major international organizations such as the World Bank or the World Trade Organization. The former uses for statistical purposes a per capita income grouping which does not distinguish between developed and developing countries which is used in part in this analysis. The WTO has no official breakdown of developed versus developing countries. For operational purposes ‘developing’ countries use the principle of self selection. The breakdown between developed and developing countries used in this analysis follows roughly the breakdown used by the WTO for statistical purposes with a few changes to be noted below. Developed countries in our analysis include 47 countries in all of Europe (including Belarus, Kazakhstan, Russia and Ukraine, but not Armenia, Azerbaijan, Georgia, or Moldova), Australia, Canada, Israel, Japan, New Zealand, Turkey and the US. This is pretty close to the WTO definition with the exception that South Africa, which the WTO classifies as ‘developed’ in our case is in the developing country group—while Turkey, classified by WTO as developing is in our analysis with the developed—as it is applying for association with the EU. Also, Armenia, Georgia and the Kyrgyz Republic classify themselves in the WTO as ‘transition’ economies- a category that had been used in the past but which is of doubtful usefulness in this analysis. All three countries are classified as ‘developing’ as is Moldova, Tajikistan, Turkmenistan and Uzbekistan. All remaining countries and territories are considered developing. For merchandise trade, the analysis has data for 145 countries. 46 are in Sub-Sahara Africa, 42 in Asia, 35 in Latin 74 America and Caribbean, and 22 in Europe, Middle East and North Africa. The latter region includes the five North Africa countries (Morocco, Algeria, Tunisia, Libya and Egypt) and stretches all the way East to include Iraq and Iran (but not Afghanistan -- which is in Asia). It also includes Armenia, Azerbaijan, Georgia and Moldova. Far less service data are available for developing countries. In this case our analysis includes information for 132 developing countries, 46 in Sub-Sahara Africa, 33 in Latin America and the Caribbean, 33 in Asia, and 20 in Europe, Middle East and North Africa. OPEC consists of 12 members as follows: Algeria, Angola, Ecuador, Libya, Nigeria, Iran, Iraq, Venezuela, Kuwait, Qatar, Saudi Arabia, United Arab Emirates. The income level analysis uses the same definition for developed countries as above. Developing and transition economies are then grouped into five categories using basically the World Bank definitions of groupings and per capita income in 2012 for 192 economies/countries, except that the Least Developed countries (LDCs) which are the 48 countries in the UN list are shown as a separate category; Low-income countries -- those with per capita income less than $1,025 (except the LDCs); lower middles-income, $1,026-4,035; upper middle-income, $4,036- 12,475; and high-income, $12,476 or more. (See details in Appendix Table A.1.) For merchandise trade sectoral breakdown data are available for 161 countries from UN COMTRADE with 42 LDCs (6 LDCs are missing data); but information for service sectors is available for 173 countries in 2010, only 46 for developed countries and 127 for developing economies. Similarly, the low-income group includes 22 countries for merchandise trade but 32 for services. The number of developing countries in the other groups is as follows: lower middle- income: 41 for merchandise trade and 44 for services; upper middle-income, 34 and 37 and; high- income, 17 and 14 respectively. 75 Appendix Table A.1: Classification of Country Groups Developed High Income Upper Middle Income Lower Middle Income Low Income Least Developed WTO TPR Developing OPEC Member Economies (48) Developing Econ (20) Developing Econ (39) Developing Econ (54) Developing Econ (34) Economies (48) Economies (50) Counties (12) Albania Aruba Algeria Angola Afghanistan Afghanistan Argentina Algeria Andorra Bahamas, The Antigua and Barbuda Armenia Bangladesh Angola Bangladesh Angola Australia Bahrain Argentina Belize Benin Bangladesh Benin Ecuador Austria Barbados Azerbaijan Bhutan Burkina Faso Benin Bolivia Iran, Islamic Rep. Belarus Bermuda Botswana Bolivia Burundi Bhutan Brazil Iraq Belgium Brunei Darussalam Brazil Cameroon Cambodia Burkina Faso Cameroon Kuwait Bosnia and Herzegovina Equatorial Guinea Chile Cape Verde Central African Rep. Burundi Chile Libya Bulgaria French Polynesia China Congo, Rep. Chad Cambodia China Nigeria Canada Hong Kong SAR, China Colombia Cote d'Ivoire Comoros Central African Rep. Colombia Qatar Croatia Korea, Rep. Costa Rica Djibouti Congo, Dem. Rep. Chad Costa Rica Saudi Arabia Cyprus Kuwait Cuba Egypt, Arab Rep. Eritrea Comoros Cote d'Ivoire United Arab Emirates Czech Republic Macao SAR, China Dominica El Salvador Ethiopia Congo, Dem. Rep. Dominican Rep. Venezuela, RB Denmark New Caledonia Dominican Rep. Fiji Gambia, The Djibouti Egypt, Arab Rep. Estonia Oman Ecuador Georgia Guinea Equatorial Guinea El Salvador Finland Qatar Gabon Ghana Guinea-Bissau Eritrea Fiji France Saudi Arabia Grenada Guatemala Haiti Ethiopia Ghana Germany Singapore Iran, Islamic Rep. Guyana Kenya Gambia, The Honduras Greece Taiwan, China Jamaica Honduras Kyrgyz Republic Guinea Hong Kong SAR, China Hungary Trinidad and Tobago Jordan India Liberia Guinea-Bissau India Iceland United Arab Emirates Lebanon Indonesia Madagascar Haiti Indonesia Ireland Libya Iraq Malawi Kiribati Jamaica Israel Malaysia Kiribati Mali Lao PDR Kenya Italy Maldives Kosovo Mozambique Lesotho Korea, Rep. Japan Mauritius Lao PDR Myanmar Liberia Kuwait Kazakhstan Mexico Lesotho Nepal Madagascar Malawi Latvia Namibia Marshall Islands Niger Malawi Malaysia Lithuania Palau Mauritania Rwanda Mali Mauritius Luxembourg Panama Micronesia, Fed. Sts. Sierra Leone Mauritania Mexico Macedonia, FYR Peru Moldova Somalia Mozambique Morocco Malta Seychelles Mongolia Tajikistan Myanmar Mozambique Montenegro South Africa Morocco Tanzania Nepal Nepal Netherlands St. Kitts and Nevis Nicaragua Togo Niger Nigeria New Zealand St. Lucia Nigeria Uganda Rwanda Pakistan Norway St. Vincent & GrenadinesPakistan Zimbabwe Samoa Paraguay Poland Suriname Papua New Guinea Sao Tome & Principe Peru Portugal Thailand Paraguay Senegal Philippines Romania Tunisia Philippines Sierra Leone Saudi Arabia Russian Federation Uruguay Samoa Solomon Islands Senegal Serbia Venezuela, RB Sao Tome & Principe Somalia Singapore Slovak Republic Senegal Sudan South Africa Slovenia Solomon Islands Tanzania Sri Lanka Spain Sri Lanka Timor-Leste Taiwan, China Sweden Sudan Togo Thailand Switzerland Swaziland Tuvalu Tunisia Turkey Syrian Arab Rep. Uganda Uganda Ukraine Timor-Leste Vanuatu Uruguay United Kingdom Tonga Yemen, Rep. Venezuela, RB United States Turkmenistan Zambia Vietnam Tuvalu Zambia Uzbekistan Zimbabwe Vanuatu Vietnam Yemen, Rep. Zambia Note: The classification of income country groups is based on GNIPC in World Bank WDI 2012 that high income is $12,476 or more, upper middle income is $4,036-12,475, lower middle income is $1,026-4,035, and low income is $1,025 or less. 76 APPENDIX B METHODOLOGY IN ESTIMATING FREQUENCY RATIOS The TPR data permitted us to estimate frequencies, in the applications of these non-tariff measures at the Harmonized System (HS) 2-digit level involving 97 product categories. Thus, the frequency ratios (f) calculated from the TPR relate to the proportion of HS2 product categories out of the total which is affected by a particular measure. The weakness of this indicator is that it gives equal weight to the presence of a measure in a country that could affect only one or a few lines in an HS2 category, for example HS72, Iron and Steel, with the presence of the same measure in another country which affects a large number of tariff lines or, for example all steel products. Formally, let Nqm be a non-tariff measure imposed by country m on a product or group of products Qm. Then the frequency ratio is measured as fnm = ΣNqm / ΣQm Where Qm is the total number of products, measured in total tariff lines or product groups. Thus, for the calculation of (f), using the HS2 product breakdown employed in most of the analysis for the Qm =97. Where tariff line information was available, as for example in the case of tariff quotas for the calculation of (f'), a standard HS 6- digit tariff line classification of approximately 5200 lines was used. It may appear at first glance, that f would always be larger than f'. This is not the case, however. The two different frequency ratios show different aspects of a country's trade regime: if a specific non-tariff measure involves a large number of tariff lines concentrated in one or two groups of products, f may be smaller than f'; the converse will be the case if a particular measure 77 applies to a few products in a large number of groups. A simple example from one of the countries, Thailand, in which tariff line and broader category measures are available for the same year, can be used to illustrate this point: In 1997 Thailand applied non-automatic licensing on a total of 25 product categories, involving 713 tariff lines. In this case, f = ΣNqm / ΣQm = 25 / 97 = 26 % while f’ = 713 / 5200 = 14 %. For the same year, Thailand's prohibitions were concentrated in 6 product categories involving 613 tariff lines. In this case, the computations are as: f = 6 / 97 = 6% while f’= 613 / 5220 = 12 %. Appendix Table B.1: Average Preferential Tariffs (GSP) of Major Sectors in OECD Markets by Exporter in 2010 Importer (OECD Markets) GSP Rate (Simple Average of %) Canada European Union Japan United States Country Group All gds Agric gds Indus gds All gds Agric gds Indus gds All gds Agric gds Indus gds All gds Agric gds Indus gds BY INCOME LEVEL Developed economies 1.9 1.9 2.0 0.5 2.3 0.4 2.8 4.5 2.0 0.8 1.0 0.7 Developing economies 0.2 0.3 0.2 0.7 1.5 0.6 2.0 3.9 1.4 0.7 2.0 0.6 High income Developing 0.1 0.2 0.0 1.2 3.6 1.1 2.5 7.2 2.4 0.6 0.3 0.6 Upper Middle income 0.0 0.1 0.0 0.8 1.8 0.7 2.8 5.7 1.9 0.5 0.9 0.3 Lower Middle income 0.0 0.0 0.0 0.7 1.3 0.7 2.3 3.8 1.4 0.9 2.1 0.7 Low income 0.0 0.0 0.0 0.3 0.3 0.2 0.4 0.5 0.4 0.9 4.1 0.6 Of which: LDCs 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.2 0.4 0.7 2.4 0.4 BY REGION World 0.2 0.3 0.2 0.7 1.5 0.6 2.0 3.8 1.4 0.7 2.0 0.6 Developing economies 0.0 0.0 0.0 0.7 1.4 0.6 2.0 3.8 1.3 0.7 2.1 0.6 Asia 0.0 0.0 0.0 1.6 2.1 1.6 1.7 3.0 0.9 1.0 3.9 0.9 Latin America & Caribbean 0.0 0.1 0.0 0.4 0.9 0.4 3.3 4.7 2.3 0.6 1.0 0.5 Europe, Middle East & N. Africa 0.0 0.0 0.0 1.0 3.8 0.9 2.4 5.9 2.0 0.8 1.1 0.7 Sub-Saharan Africa 0.0 0.0 0.0 0.1 0.4 0.1 1.1 2.8 0.8 0.6 2.6 0.3 Memo: No. of total countries/econ. received pref. tariffs 69 57 69 146 144 145 125 88 114 153 111 153 Note: Empty space indicates data not available or no GSP rate. Source: Based on UNCTAD TRAINS databases through WITS. 78 Appendix Table B.2: Antidumping Initiations and Measures by Developing Country, 1995-2011 Antidumping by Exporting Country No. of Cases No. of Measures Total ** Country (No. of Countries) Total 1995-01 2002-06 2007-11 Total 1995-01 2002-06 2007-11 Product Argentina 35 13 14 8 19 8 5 6 1,075 Bangladesh 2 1 0 1 2 1 0 1 .. Brazil 114 64 28 22 82 47 23 12 541 Chile 29 19 6 4 18 9 7 2 35 China 853 265 281 307 630 182 201 247 378 Colombia 5 4 1 0 2 2 0 0 196 Costa Rica 2 2 0 0 .. .. .. .. 10 Côte d'Ivoire .. .. .. .. .. .. .. .. .. Dominican Republic 3 1 0 2 1 0 1 0 1 Egypt 12 10 2 0 5 4 0 1 69 El Salvador 3 0 0 3 .. .. .. .. .. Honduras 1 1 0 0 1 1 0 0 3 Hong Kong, China 28 16 6 6 16 8 4 4 .. India 155 70 58 27 94 39 37 18 1,465 Indonesia 165 79 51 35 101 33 40 28 288 Jamaica .. .. .. .. .. .. .. .. 15 Kenya 2 0 0 2 .. .. .. .. .. Korea, Republic of 284 148 87 49 171 77 66 28 336 Kuwait 2 0 0 2 1 0 0 1 .. Malawi 1 1 0 0 1 1 0 0 .. Malaysia 104 36 38 30 65 21 22 22 59 Mexico 55 28 12 15 32 16 10 6 1,685 Morocco .. .. .. .. .. .. .. .. .. Mozambique 1 1 0 0 .. .. .. .. .. Nepal 2 2 0 0 2 0 2 0 .. Nigeria 1 0 1 0 1 0 1 0 .. Pakistan 13 5 5 3 6 3 3 0 294 Paraguay 2 1 0 1 3 2 0 1 2 Peru 4 2 0 2 1 0 0 1 379 Philippines 12 4 7 1 7 2 5 0 51 Saudi Arabia 24 10 5 9 10 4 2 4 .. Singapore 46 23 19 4 32 6 17 9 2 South Africa 60 36 18 6 40 19 15 6 460 Sri Lanka 5 0 1 4 2 0 1 1 .. Taipei, Chinese 211 96 75 40 140 57 49 34 172 Thailand 164 72 49 43 109 41 36 32 318 Tunisia .. .. .. .. .. .. .. .. .. Uruguay 4 2 1 1 3 1 1 1 12 Venezuela 20 14 4 2 12 8 4 0 94 Viet Nam 30 3 15 12 22 3 9 10 .. Zimbabwe 2 2 0 0 1 1 0 0 .. Total above countries (41) 2,456 1,031 784 641 1,632 596 561 475 7,940 Memo Items Developed countries (45) 1,470 855 371 244 931 510 293 128 21,016 Other developing co. (21) 84 37 19 28 38 11 21 6 37 All countries (107) 4,010 1,923 1,174 913 2,601 1,117 875 609 28,993 Note: ** Total products affected by antidumping are based on number of products on raw data files at HS 6-, 8-, 9-, 10-digit level or mixed together in the exporting or reporting countries. Sources: WTO Antidumping database and Bown, C (2012), Temporary Trade Barriers (TTB) database (product data). 79 Appendix Table B.3: Antidumping, Countervailing and Safeguard Measures by Reporting Developed Country, 1995-2011 Antidumping Countervailing Safeguard Total AD+CV+SG No. of Msre Total Affected No. of Msre Total Affected No. of Msre Total Affected No. of Msre Total Affected Country 1995-2011 Products ** 1995-2011 Products ** 1995-2011 Products ** 1995-2011 Products ** Australia 89 857 4 34 .. .. 93 891 Bulgaria .. .. .. .. 2 35 2 35 Canada 96 3,947 17 417 .. .. 113 4,364 Czech Republic 1 .. .. .. 5 90 6 90 European Union 282 2,879 30 238 3 518 315 3,635 Hungary .. .. .. .. 3 57 3 57 Japan 7 10 1 3 .. .. 8 13 Latvia 2 .. 1 1 2 108 5 109 Lithuania 7 .. .. .. 1 1 8 1 New Zealand 24 216 4 5 .. .. 28 221 Poland 9 .. .. .. 4 171 13 171 Slovak Republic .. .. .. .. 2 4 2 4 United States 305 12,055 73 10,757 6 644 384 23,456 Memo Items: 0 0 Israel 22 105 2 2 1 4 25 111 Turkey 145 894 1 3 13 53 159 950 Ukraine 28 .. .. .. 2 43 30 43 Note: ** Total products affected by AD, CV, and SG are based on number of products on raw data files at HS 6-, 8-, 9-, 10-digit level or mixed together in reporting countries. Sources: WTO Antidumping database and Bown, C (2012), Temporary Trade Barriers (TTB) database (product data). 80 Appendix Table B.4: Antidumping Investigations by Initiating v.s. Exporting Country, 1995-2011 By --> Al l Devel oped Devel opi ng Aga i ns t Countri es (46) Countri es (15) Countri es (31) Numbers of Antidumping Initiations Al l Countri es (102) 4,010 1,609 2,401 Devel oped Countri es (45) 1,470 589 881 Devel opi ng Countri es (57) 2,540 1,020 1,520 Of whi ch: Chi na 853 360 493 Percentages of Antidumping Initiations by Affected Co. Al l Countri es (102) 100.0 40.1 59.9 Devel oped Countri es (45) 100.0 40.1 59.9 Devel opi ng Countri es (57) 100.0 40.2 59.8 Of whi ch: Chi na 100.0 42.2 57.8 Percentages of Antidumping Initiations by Exporter Al l Countri es (102) 100.0 100.0 100.0 Devel oped Countri es (45) 36.7 36.6 36.7 Devel opi ng Countri es (57) 63.3 63.4 63.3 Of whi ch: Chi na 21.3 22.4 20.5 Percentage of Export Share by Exporter Al l Countri es (102) 100.0 100.0 100.0 Devel oped Countri es (45) 69.2 75.4 53.3 Devel opi ng Countri es (57) 30.8 24.6 46.7 Of whi ch: Chi na 4.4 3.6 6.4 Ratio of Antidumping Initiations to Export Share Al l Countri es (102) 1.0 1.0 1.0 Devel oped Countri es (45) 0.5 0.5 0.7 Devel opi ng Countri es (57) 2.1 2.6 1.4 Of whi ch: Chi na 4.8 6.2 3.2 Sources : WTO a nti dumpi ng da ta ba s e a nd UN COMTRADE da ta (tra de). 81 ENDNOTES 1 With the exception of Vietnam. 2 For a list of the countries see Appendix A 3 The GATT/WTO published findings of TPRs, conducted on a regular basis since the beginning of the 1990s, are a useful source of information regarding the evolution of developing country policies (Cadot, Malouche and Saez 2012, p.12). The main objective of the review mechanism is to 'contribute to improved adherence by all WTO Members to rules, disciplines and commitments under the Multilateral Trade Agreements by achieving greater transparency and understanding of the trade policies and practices of members' (WTO, 1995, p.434). The reviews contain detailed information on policies and institutions affecting both imports and exports on a consistent basis over time. All this information has been reviewed and discussed by the countries in question and the WTO Members (and previously, the GATT Contracting Parties) and can therefore be considered accurate and authoritative. While various aspects of the reviews could be strengthened, (Keesing, 1998), there is little dispute over the accuracy of the information they provide. 4 The simple arithmetic average should be interpreted with care: the data refer to different years in each of the decades. Some rates may have also increased since the data were collected. 5 These estimates have been obtained from the UNCTAD/World Bank NTM data base. 6 Table 11 suggests that the product groups which is the focus of anti-dumping actions, both in developed and developing countries tend to be somewhat similar and include especially basic metals and chemicals. 7 It should be recalled however, as part of the URA and previous negotiations, there were significant reductions in tariffs on horticultural and floricultural products of interest to developing countries. 8 Bown’s country classification differs from the one presented in this paper as he treats Turkey as a developing country and Korea and Taiwan, China as developed thus making overall comparisons between ‘developed’ and ‘developing’ countries’ averages difficult. 9 Based on the UNCTAD/World Bank NTM data base. It was estimated at 28 percent at the HS 2-digit level in Michalopoulos (2001). 10 See Nogues et al, 1986 and OECD, 1997 on the definition of 'core' non-tariff measures. 11 The results of this methodologically innovative study are based on earlier data, tariffs 2001- 2004 and NTBs in the 1990s; unfortunately the NTB data used are not reliable (see Michalopoulos 2001). 12 Given that EAC has few formal NTMs, the NTMs reported for Kenya in Table 8 may well be of the administrative kind reported in the TPR. 82 13 “The Plant Health Service (PHS) has 165 inspectors based at 28 entry points (out of 51). However, most entry post are poorly equipped for pest inspection, staff often do not have specific training in phytosanitary matters and have little or no direct communication with PHS headquarters. Hence pest surveillance capacity is severely constrained.� 14 The broad theoretical point is covered in the so called Lerner symmetry theorem on the equivalence of import and export taxes (Lerner, 1936). 15 Just like the estimates of non-automatic licensing affecting imports, the analysis has attempted to exclude licensing which the authorities state they undertake in order to meet health, environmental and safety standards, national security reasons. 16 The only economies in which such a program was not reported to exist were Hong Kong SAR, China (which has no import duties), three African countries (Benin, Cameroon, Ivory Coast) and the Dominican Republic, where the export incentives focus on the EPZ. 17 Reportedly Japan did not finish implementing its commitments until 2009 (Japan TPR, 2011). 18 See Article 15 of the UR Agreement on the Procedures for Implementation of GATT Article VI. 19 The EU advanced from January 1, 1997 to January 1, 1996 its schedule of implementation of the third stage of tariff reductions for most non-agricultural products as part of its compensation for the EU enlargement through the accession of Austria, Finland and Sweden (WTO, TPR, European Union, 1997, p.15). 20 With the exception of the ‘Peace Clause�, under Article 13, , a provision which expired in 2003 and which had constrained members from taking action under the DSM in connection with their commitments on domestic support, market access and export subsidies in agriculture 21 Some writers speak of ‘mega-tariffs’ for tariff lines in excess of 100 percent. According to one report, in agriculture alone the US has 24 mega-tariffs, the EU 141 and Japan 142 22 These are estimates reported in UNCTAD/WTO, 1997 and UNCTAD, 1997. It is difficult to estimate accurately the ad valorem tariff equivalents for many agricultural commodities where tariffication involves the adoption of specific duties often combined with quotas. 23 There is a major effort currently undertaken by the World Bank in co-operation with UNCTAD and WTO. The UNCTAD TRAINS data base has been in place for some time but it is not considered reliable and complete. 24 The special safeguards against China are supposed to expire in 2013. 25 See TPRs for EU (2004), US (2004), Canada (2004). 26 Caution should be used in reaching conclusions based on the evolution over time of the trade weighted share of trade remedies excluding oil. This share depends very much on the dependence of a country on oil imports—which in the US has declined over this period. 27 The GTA reports suffer from some additional shortcomings noted in (Wolf, 2011). 83 REFERENCES Binswagner, H. and E. Lutz. 2000. Agricultural Trade Barriers, Trade Negotiations and the Interests of Developing Countries, UNCTAD X, TD(X)/RT.1/8. Bachetta M., J.Richtering and R. Santana. 2012. “How Much Light Do WTO Notifications Shed on NTMs?� in O. Cadot and M. Malouche, eds., Non-Tariff Measures: A Fresh Look at Trade Policy’s New Frontier, CEPR and World Bank: Washington DC. Borchert, I, B. Gootiiz and A. Mattoo, 2011. “Services in Doha: What’s on the Table?� in W. Martin and A. Mattoo, Unfinished Business? The WTO’s Doha Agenda, CEPR and World Bank: Washington DC. Borchert, I, B. Gootiiz and A. Mattoo, 2012, Policy Barriers to International Trade in Services: Evidence from a New Database,� Policy Research Working Paper, no. 6109, World Bank: Washington DC. Blandford, D., T. Josling, and D. Orden, 2011, WTO Disciplines on Agricultural Support, Cambridge University Press. Bowen, C. and H. L. Kee. 2011. “Developing Countries, New Trade Barriers and the Global Economic Crisis,� in M. Haddad and B. Sheppard, eds., Managing Openness, World Bank: Washington DC. Bown, C., 2011, “Taking Stock of Antidumping, Safeguards and Countervailing Duties, 1990- 2009,� The World Economy, 34(12): 1955-98. (TTB database is available at http://econ.worldbank/ttbd/) Bown, C., 2013, “Emerging Economies and the Emergence of South-South Protectionism,� Journal of World Trade, 47(1): 1-44. (Also Bown, C., 2012, PRD Working Paper, # 6162, World Bank) Cadot, O. and M. Malouche. 2012. “Overview,� in O. Cadot and M. Malouche, (eds.), Non-Tariff Measures: A Fresh Look at Trade Policy’s New Frontier, CEPR and World Bank: Washington D.C. Cadot, O., M. Malouche and S. Saez, 2012. Streamlining Non-Tariff Measures: A Toolkit for Policy Makers, World Bank: Washington DC. Chang, W. and L. A. Winters, 2002. “How Regional Blocks Affect Excluded Countries: the Price Effects of Mercosur� American Economic Review, 92(4): 889-904. Dadush, U., S.Ali and R.E. Odell, 2011, Is Protectionism Dying? Carnegie Endowment: Washington DC. Daly, M. and H. Kuwahara. 1998. ‘The Impact of the Uruguay Round on Tariff and Non-Tariff Barriers to Trade in the Quad,� The World Economy, 21: 207-234. Das, J. 1998. The WTO Agreements: Deficiencies, Imbalances and Required Changes, Penang: Third World Network. 84 Drabek, Z. and S. Laird. 1998. “The New Liberalism: Trade Policy Developments in Emerging Markets,� Journal of World Trade, 5(3):241-269. Evenatt, S., B. Hoekman, and O. Catteneo, 2009, Effective Crisis and Response to Openness: Implications for the Trade System, (eds), CEPR: London. Evenett, S. J. 2009. Global Trade Alert 1st Report, CEPR: London. Evenett, S. J. 2012.�Summary,� Debacle: the 11th GTA Report on Protectionism, http//www.global trade alert. Finger, J. M. 1993. Anti-dumping: How It Works and Who Gets Hurt, Ann Arbor: Michigan University Press. Finger, J. M. and L. A. Winters. 1998. “What Can the WTO Do for Developing Countries,� in A. O. Krueger (ed.), The WTO as an International Organization, Chicago: The University of Chicago Press. Finger, J. M. and L. Schuknecht. 1999. Market Access Advances and Retreats since the Uruguay Round, Policy Research Working Paper, No. 2232, World Bank: Washington DC. Foletti, L. et al .2009. “Smoke in the Water�, in S. Evenett, B.Hoekman and O. Cattaneo eds.Effective Crisis Response and Openness: Implications for the Trading System, CEPR and World Bank: Washington DC. Francois, J. et al. 2011. “Opening Markets for the Poorest Countries: Assessing the Effects of Duty-Free, Quota-Free to the G20,� mimeo, World Bank: Washington DC. Francois, J. and I. Wooton. 1999. “Trade in International Transport Services: the Roof Competition,� mimeo, August. Gorter, H. D., L. Ruiz and M. Ingco. 2003. “Export Competition Policies,� in M. Ingco and J. Nash (eds.), Agriculture and the WTO, World Bank: Washington DC. Gourdon, J. and A. Nicita, 2012. “NTMs: Interpreting the New Data,� in O. Cadot and M. Malouche, (eds.), Non-Tariff Measures: A Fresh Look at Trade Policy’s New Frontier, CEPR and World Bank: Washington DC. Hoekman, B., W. Martin and C. A. Primo Braga, 20009, “Quantifying the Value of Preferences and Potential Erosion Losses,� in B. Hoekman, W. Martin and C.A. Primo Braga, eds. Trade Preference Erosion: Measurement and Policy Response, Palgrave Macmillan and World Bank: Washington DC. Kee, H L, A. Nicita, and M. Olarreaga, 2009, “Estimating Trade Restrictiveness Indices,� Economic Journal, vol. 119, 172-199. Keesing, D. B. 1998. Improving Trade Policy Reviews in the World Trade Organization, Institute for International Economics: Washington DC. Laird, S. 2012. A Review of Trade Preference Schemes for the World’s Poorest Countries, ICTSD, Geneva. Lederman, D., M. Olarreaga, and L. Payton, 2009, Export Promotion Agencies Revisited. Policy Research Working Paper, no. 5125, World Bank: Washington DC. 85 Lindbland, J. 1997. The Impact of the Uruguay Round on Tariff Escalation in Agricultural Products, ESCP, No. 3, FAO: Rome. Mangellsdorf, A., A. Perez, and J. S. Wilson. 2012. “Do Better Standards Facilitate Exports? Evidence from China,� in O. Cadot and M. Malouch, eds. Non- Tariff Measures: A Fresh Look at Trade Policy’s New Frontier, CEPR and World Bank: Washington DC. Martin, W. and L. A. Winters. 1996. “The Uruguay Round: A Milestone for the Developing Countries,� in W. Martin and L. A. Winters (eds.), The Uruguay Round and the Developing Countries, Cambridge: Cambridge University Press. Maskus, K, O. Tsunehiro, and J. S. Wilson, 2005, The Cost of Compliance with Product Standards for Firms in Developing Countries: An Econometric Study, World Bank: Washington DC. Mattoo, A. 1998. Financial Services and the WTO, Policy Research Working Paper, No. 2184, World Bank: Washington DC. Messerlin, P. 2009. “A Significant Slippage in Protectionism? Not Yet,� in S.J. Evenett, B. Hoekman, and O. Cattaneo (eds.), Effective Crisis Response and Openness, CEPR and World Bank: Washington DC. Michalopoulos, C. 1999. Trade Policy and Market Access Issues for Developing Countries, Policy Research Working Paper, No. 2214, World Bank: Washington DC. Michalopoulos, C. 2001. Developing Countries in the WTO, Palgrave: Houndsmills. Michalopoulos, C. and F. Ng. 2013. “Trends in Developing Country Trade, 1980-2011� Policy Research Working Paper, No. 6334, World Bank: Washington, DC. Michalopoulos, C. and L. A. Winters. 1997. “Summary and Overview,� in P. D. Ehrenhaft et al, Policies on Imports from Economies in Transition, Studies of Economies in Transformation, No. 22, World Bank: Washington DC. Miranda, J., R. Torres and M. Ruiz. 1998. “The International Use of Antidumping: 1987-1997,� Journal of World Trade, 32(5): 5-71. Nkomo, M. 2011. “The Under-Utilization of TRIPS Flexibilities by Developing Countries: The Case of Africa,� in WTO-WIPO, Colloquium for Teachers of Intellectual Property, WTO: Geneva. Nogues, J., A. Olechowski and L. A. Winters. 1986. “The Extent of Non-Tariff Barriers to Industrial Countries' Imports,� World Bank Economic Review, 1(1): 181-199. Otsuki, T., J. Wilson and M. Sewadah. 2001. “What Price Precaution? Europe’s Harmonization of Aflatoxin Regulations and African Groundnut Exports,� European Review of Agricultural Economics, 28(3): 263-84. OECD. 1997. Indicators of Tariff and Non-Tariff Barrier, Paris: OECD. Panagariya,A. 2000. “Preferential Trade Liberalization: the Traditional Theory and New Developments�, Journal of Economic Literature,38 (2). Smeets, M. and C. Fournier, 1998, Trade and Tariffs in Leather and Leather Products, WTO: Geneva. 86 Srinivasan, T. N. 1998. Developing Countries and the Multilateral Trading System, Oxford: Westview Press. UNCTAD. 1997a. Post-Uruguay Round Tariff Environment for Developing Country Exports, UNCTAD/WTO Joint Study, TD/B/COM.1/14, October. ________. 1997b. The Uruguay Round and Its Follow-up: Building a Positive Development Agenda, Geneva: United Nations. Wilson, J., T. Otsuki and B. Majumdstar. 2003. “Balancing Food Safety and Risk�, Journal of International Trade and Economic Development, (12)4: 377-402. World Bank. 2000. Global Economic Prospects, World Bank: Washington DC. World Bank. 2012. World Development Indicators, World Bank: Washington DC. WTO. 1995. The Results of the Uruguay Round of Multilateral Trade Negotiations: The Legal Texts, WTO, Geneva. WTO.1997. ‘Market Access for Least Developed Countries’, High Level meeting on Integrated Initiatives for Least Developed Countries WT/LDC/HL/14, November. WTO. 2009. Report to the TPRB from the Director General on the financial and economic crisis And trade related developments, WT/TPR/OV/W/1, WTO: Geneva. WTO. 2011. “The WTO and Preferential Trade: from Co-existence to Coherence�, World Trade Report, WTO: Geneva. WTO. 2012a. Report on G-20 Trade Measures, October 31, WTO: Geneva. WTO. 2012b. “Trade and Public Policies: A closer Look at Non-Tariff Measures in the 21st Century,� World Trade Report, WTO: Geneva. Wolf, R. 2011. Did the Protectionism Dog Bark? Policy Report, No. 1, Entwined: Stockholm. Xion, B. and J. Beghin. 2011. Does European Aflatoxin Regulation Hurt Groundnut Exporters from Africa? Iowa State University Working Paper, No. 11026, February. 87