35929 Trade Note March 30, 2006 WTO's DOHA COTTON INITIATIVE: HOW WILL IT AFFECT DEVELOPING COUNTRIES? For many developing countries, Trade Ministerial meeting in especially in Africa and Central December 2005 are implemented as Asia, cotton is an important cash part of the DDA? crop (Figure 1). It is receiving The attention of late because four poor The Cotton Initiative under the World Bank cotton-exporting West African WTO's DDA has not only the trade Group countries (Benin, Burkina Faso, policy reform component, but also a www.worldbank.org Chad and Mali) have demanded that development component aimed at cotton subsidy and import tariff boosting the international removal be part of the World Trade competitiveness of cotton producers Organization's Doha Development in low-income (especially West Agenda (DDA). Cotton subsidies African) countries. One prospective are mostly provided by governments way to do that is for governments of in high-income countries, and part those countries to encourage of the US cotton subsidy program adaptation and allow adoption of new has been ruled illegal following a varieties of cotton emerging from the WTO dispute settlement case biotechnology revolution, the brought by Brazil. Hence some affordability of which will be greater reform can be expected soon, in the absence of cotton market International especially if the DDA is to live up to distortions. How do the estimated Trade its name of being a development gains from cotton subsidy and tariff Department round. reform compare with the prospective gain from wider adoption by How much would Sub-Saharan developing countries of genetically Africa and Central Africa gain from modified (GM) cotton? And how removal of all cotton subsidies and much greater would be the gains to By Kym Anderson tariffs? How would the welfare of cotton-producing developing and Ernesto those cotton-importing developing countries from GM cotton adoption if Valenzuela countries with export interests in global cotton markets were not textiles and clothing be affected by distorted by subsidies and tariffs? such reform? What would be the These notes summarize relative contributions of domestic Background: The Global Cotton recent research on global supports, export subsidies and Market trade issues. They reflect solely the views of the author import tariffs to the global gains Cotton production is highly and not necessarily the views from removal of those measures? concentrated in a few countries: as of of the World Bank Group or And how would the gains from full 2005/06, nearly half is produced by its Executive Directors. reform compare with the gains that just China and the United States, and could be expected if and when (a) that rises to more than two-thirds the US complies with its WTO when India and Pakistan are added obligations as laid out in the WTO's and to more than three-quarters when Trade Note 27 dispute settlement Panel and Brazil and Uzbekistan are included. Appellate Body reports (WTO Also highly concentrated are exports 2004b, 2005a) and (b) the partial of cotton lint, with the US, Australia, reforms proposed in the Hong Kong Uzbekistan and Brazil accounting for TRADE NOTE March 30, 2006 almost two-thirds of the world's exports, while the What is striking about the welfare effects is their cotton-four in West Africa and the other four distribution among developing countries (Table 1 countries in Central Asia bring that total to almost and Figure 2). Especially noteworthy is the four-fifths. Cotton production is also concentrated relatively large benefit bestowed on Sub-Saharan in the sense that a number of low-income countries Africa, of $147 million per year. About two-fifths depend heavily on cotton for earning foreign of that would go to the Cotton-4 and another one- exchange (Figure 1). Cotton usage, on the other fifth to other West African countries. This is Figure 1: Net income from cotton production as % of GDP, and cotton driven by an estimated increase in Sub-Saharan exports as % of total exports African cotton output and net farm income of (both relative to the global average in 2001) nearly one-third, and in the real value of the region's cotton exports of more than 50 percent. By contrast, cotton output and exports would fall by one-quarter in the United States and would halve in the EU (middle columns of Table 1). That would raise Sub-Saharan Africa's share of global cotton exports from 12 to 17 percent, and the share of all developing countries from 52 to 72 percent. The region's welfare gain of $147 million per year is no less than one-fifth of the estimated gain for the region from the freeing of all goods markets globally, according to our GTAP model results. It is therefore not surprising that some African trade Source: GTAP database (www.gtap.org) negotiators have threatened to walk out of the hand, is distributed across countries roughly in WTO's Doha round of talks if substantial reforms proportion to their volumes of textile production. to cotton policies are not included in the final Because of high domestic usage by exporters of Doha agreement ­ in which case the global cost of textiles and clothing in developing Asian countries not reforming cotton would be many times greater (and Mexico because of its preferential access to than implied in Table 1. the US and Canadian markets under NAFTA), even relatively large cotton producers such as Figure 2: Welfare gain from cotton tariff and subsidy reform as a percent of GDP, as a multiple of the share for the world as a whole China, Pakistan and India export only a small 142 Other Sub-Sahara Africa fraction of their crop, in contrast to Sub-Saharan Uganda Africa where textile production is relatively minor Mozambique (columns 2 and 3 of Table 1). Other Southern Africa Australia Full Liberalization The Global Cost of Cotton Subsidies and Tariffs C. Asia and S.E. Europe Doha partial reform US-only partial reform Using the GTAP database and model of the global Sri Lanka economy, Anderson and Valenzuela (2006) Other South Asia estimate that the removal of all cotton subsidies China Zambia and import tariffs would boost global economic Other South East Asia welfare by $283 million per year, and would raise Brazil the price of cotton in international markets by an 1 11 21 31 41 51 61 71 81 average of 13 percent. (These estimates do not Source: Authors' GTAP model simulation results take into account gains from productivity that the ending of subsidies might stimulate). The price Export subsidy removal would contribute almost rise ensures that all cotton-exporting countries none of the global benefits from reform, and would benefit, while net importers of cotton would cotton tariff removal would account for only one- be worse off, as shown in the right-hand columns ninth of the global gain, with the other eight-ninths of Table 1. due to cutting domestic support programs. This latter result contrasts markedly with that for the removal of all agricultural subsidies and tariffs (to 2 TRADE NOTE March 30, 2006 Table 1: Cotton net farm income and net export positions in 2001, and impact of removing cotton subsidies and tariffsa on cotton output, exports, net farm income and economic welfare (percent and 2001 US$m) Index of Net exportsb ($b) of: Change Change in Change in Welfare change ($m): cotton in cotton cotton net value of that are production textiles output farm cotton due to special- cotton and volume income exports TOTAL terms of izationc clothing (%) (%) (%) trade change High-income countries 0.3 1.0 -92.0 -20 -15 -18 465 275 Australia 3.8 1.1 -2.6 25 22 38 137 125 United States 0.6 2.2 -60.7 -25 -18 -29 429 443 EU25 0.1 -1.0 -28.8 -54 -53 -49 14 -109 Japan 0.0 -0.4 -14.1 1 2 62 -24 -49 Korea-Taiwan 0.1 -0.7 22.5 12 7 34 -61 -84 Developing countries 3.8 -1.0 92.0 6 4 46 -182 -275 E. Europe & C. Asia 4.3 0.3 7.4 7 3 36 -14 -36 Turkey 11.6 -0.4 8.7 2 2 37 -86 -80 Other ECA 2.1 0.7 -1.3 10 9 35 72 44 East Asia 3.0 -1.4 60.4 2 2 72 -83 -127 China 4.0 -0.1 41.9 2 2 76 50 45 South Asia 14.5 -1.0 24.5 2 1 55 -96 -99 Bangladesh 14.2 -0.3 3.8 8 5 68 -11 -21 India 13.7 -0.6 11.9 -1 0 31 -85 -79 Pakistan 29.9 -0.1 6.8 5 3 61 -7 -5 M. East &North Africa 2.5 0.4 -3.3 6 6 37 19 26 Sub-Saharan Africa 5.8 1.1 -1.8 32 31 55 147 113 South Africa 0.3 -0.0 -0.2 19 21 47 -1 -2 Mozambique 6.1 0.0 -0.0 19 18 29 2 1 Zambia 11.6 0.0 0.0 4 4 11 0 0 Uganda 6.8 0.0 -0.0 27 26 45 4 3 Other Sthn & E. Africa 7.5 0.2 0.7 21 20 46 17 14 Nigeria 2.2 0.0 -0.7 23 21 47 -1 0 Other Sub-Sah. Africa 12.6 0.8 -1.6 39 37 60 126 97 Latin America & Car. 1.1 -0.4 4.8 11 9 54 -155 -152 Argentina 1.1 0.1 -0.4 14 11 66 7 6 Brazil 1.5 0.1 -0.0 10 10 58 13 12 Mexico 0.8 -0.5 4.0 13 11 42 -128 -136 World 1.0 0.0 0.0 -1 -2 8 283 0 aRemoval of those distortions left after the phase-out of the quotas at the end of 2004 bExports minus imports, both valued at f.o.b. prices as in the GTAP database 6.05 cCotton's national share in GDP relative to the global share. In the GTAP database the sector is `plant-based fibers' and so includes such products as flax (important only for Bangladesh in the above countries) Source: Authors' GTAP model simulation results 3 TRADE NOTE March 30, 2006 which cotton is a tiny contributor), whereby tariff program and the export credit guarantees were removal accounts for a huge 93 percent of the prohibited export subsidies and domestic-content global benefits and domestic support programs subsidies. In response, the US Congress agreed to only 5 percent (Anderson, Martin and Valenzuela repeal the two parts of its Step 2 program in 2006. 2005). If US expenditure on cotton support is reduced by the full amount of the Step 2 payments, this would Turning to the impacts on cotton farmers' incomes be equivalent in 2000-02 to a one-seventh of such reform (Figure 3), we estimate they would reduction in the aggregate subsidy to US cotton decline by one-sixth in the United States and by production. The complaining country (Brazil) may just over half in the EU. In virtually all other expect a reduction also in US cotton producer regions, however, they are estimated to rise. subsidies, which in 2000-2002 averaged $3.0 Crucially, they would rise by a huge 30 percent in billion while in 1992 (the limit year under the Sub-Saharan Africa and around 40 percent in West Uruguay Round Agreement on Subsidies and Africa in particular ­ more than three-quarters of Countervailing Measures) they were just $2.0 which is due to cuts to domestic support programs. billion. To simulate a US reform that would fully Figure 3: Percentage change in cotton farm income from reform of comply with the WTO rulings, we ran a scenario cotton tariffs and subsidies, as a multiple of the percentage change for in which not only the Step 2 program is removed developing countries as a whole but also domestic producer subsidies are cut by Other Sub-Sahara Africa one-third, from $3 billion to $2 billion. Uganda Nigeria Other Southern Africa The WTO's Hong Kong Trade Ministerial meeting Malawi of the DDA in December 2005 went further: Mozambique Full Liberalization members agreed that all cotton export subsidies be Sri Lanka Doha partial reform US-only partial reform VietNam eliminated during 2006, that least-developed Argentina countries get duty free access for their cotton Brazil exports to high-income countries by the time C. Asia and S.E. Europe Other Latin America implementation of the DDA commences, and that Malasya domestic cotton subsidies be reduced faster and more ambitiously than other agricultural domestic Indonesia 1 2 3 4 5 6 7 8 9 support programs during DDA implementation. To Source: Authors' GTAP model simulation results see how far that would go towards yielding the Effects of Partial Reform of Cotton Subsidies potential gains to low-income countries from full and Tariffs reform as reported above, we ran another partial While the full reform results presented above are liberalization scenario in which we eliminated not likely to materialize in the immediate future, those two sets of trade measures and cut domestic they provide a useful benchmark against which to cotton supports by one-third in all high-income compare the estimated effects of partial reforms. countries (not just in the US). Consider two partial reform scenarios: liberalization in the United States alone, as a Among the impacts of these partial reform possible response to the outcome of the WTO simulations are the following: dispute settlement case brought against it by · The national welfare gains and boost to cotton Brazil; and a broader liberalization consistent with farmers' incomes from partial reform are still what was agreed at the Hong Kong Trade concentrated in Sub-Saharan Africa and Ministerial in December 2005 as part of the Doha Central Asia, although less so than under full Development Agenda (DDA). reform (Figures 2 and 3) · Sub-Saharan Africa's cotton output and How much cotton reform can be expected in the exports would rise only one-quarter as much United States as a result of the US being found, by (and Central Asia's 40 percent as much) under the WTO's dispute settlement Panel and Appellate full reform as under the Doha partial reform Body, to be not in compliance with its WTO scenario. obligations? The WTO ruled that the Step 2 4 TRADE NOTE March 30, 2006 · Compared with what Sub-Saharan Africa can The estimate of the global benefits from full GM expect from Doha cotton reform, US-only cotton adoption by developing countries is eight partial reform would generate only around times larger than the above estimate of the global three-fifths of the estimated net welfare and gain from complete removal of all cotton subsidies net cotton income effects, and two-fifths of the and tariffs, and twelve times larger than the global export effects. gain from the Doha partial cotton reform · The average price of cotton in international simulation. The differences are less marked for markets is estimated to rise by 12.9 percent Sub-Saharan Africa, but even so its estimated gain under full reform, but by just 4.4 and 3.2 from adopting GM cotton varieties is well above percent in the Doha and US-only scenarios, that from full removal of all trade-distorting cotton respectively. policies and around six times that from the Doha partial reform simulation considered above. What Impact Would GM Cotton Adoption Have on the Gains from Trade Reform? If all distortions to cotton markets were removed, The Cotton Initiative involves two parts: in the gains to developing countries in the absence of addition to trade and subsidy reform, importance distortionary cotton policies would be slightly has been attached to boosting the international greater (12 percent so in the case of Sub-Saharan competitiveness of cotton production in low- Africa. Were these two reforms (GM catch-up and income countries. One prospective way to do that subsidy removal) to occur simultaneously, they is for governments of those countries to allow the would reinforce each other in Sub-Saharan Africa adoption of new varieties of cotton emerging from as each expands the region's cotton production the biotechnology revolution, following the and exports and so makes the gain from the other example of the early adopting countries (the change larger: the gain to Sub-Saharan Africa United States, Australia, China and South Africa). would then be $370 million. Also, by comparing To simulate the economic effect of global Figures 2 and 4 it is clear that, while some cotton- adoption of GM cotton varieties, Anderson, importing developing countries lose from subsidy Valenzuela and Jackson (2006) assume total factor reform on its own, they gain when it is combined productivity in cotton production, net of any with the spread of the productivity enhancing GM higher cost of GM seed, would rise by 5 percent in cotton varieties. This is a clear example of all adopting countries except India and Sub- complementarity between the trade and Saharan Africa other than South Africa (who's development components of the Doha Cotton yields are well below half the global average) Initiative. where 15 percent is assumed. Figure 4: Welfare change from the combination of cotton tariff If all other countries adopt GM cotton, the value of and subsidy reform and post-2001 GM cotton adoption, as a percent of GDP, as a multiple of the percentage change for the cotton output in the four early-adopting countries world as a whole would fall in response to the output expansion in Other Sub-Sahara Africa newly adopting regions, but global welfare would India jump $2.3 billion. Asia's developing countries that Uganda are net importers of cotton would gain even if they Pakistan OtherSthnAFR grow little or no cotton, because the international Full Liberalization Zambia price of that crucial input into their textile industry Mozambique GMO adoption after Full Liberalization would be lower by an average of 4.1 percent in C. Asia and S.E. Europe this scenario. The gains to Central Asia, Sub- Turkey Saharan Africa and South Asia are estimated to be SACU (w/o S. Africa) Nigeria ten, thirteen and twenty-three times greater than MENA the global gains when expressed as a percentage of Other South Asia regional GDP. South Asia's are especially large -8 -4 0 4 8 12 16 20 24 28 32 36 because it is a large producer and user of cotton. Source: Authors' GTAP model simulation results 5 TRADE NOTE March 30, 2006 Conclusions adoption of new genetically modified (GM) cotton How far the DDA will go towards generating the varieties are one obvious way of contributing ­ benefits of full liberalization will hinge very and that is within the powers of developing heavily on the extent to which the US and to a countries themselves and so does not need to wait lesser extent EU governments are willing to cut until that Doha round concludes. While the above their applied domestic subsidies to cotton results suggest that developing country welfare production. But potentially Doha partial reform would be enhanced by far more from allowing GM could deliver roughly twice the gains to cotton- cotton adoption than by the removal of all cotton exporting developing countries as the reform that subsidies and tariffs, cotton subsidy reductions ­ in the absence of the DDA ­ the US might be would enhance the capacity of poor farmers in expected to do anyway to bring its cotton support low-income countries to purchase the more- programs into conformity with its WTO expensive GM cotton seeds and make the obligations. Meanwhile, there are other ways in necessary adjustments to their farming practices, which incomes of cotton farmers in developing thereby increase the prospects of realizing the countries can be enhanced. Adaptation and potential gains from GM adoption. Further Reading Anderson, K., W. Martin and E. Valenzuela (2005), "Why Market Access is the Most Important of Agriculture's `Three Pillars' in the Doha Negotiations", Trade Note No. 26, World Bank, Washington DC, Dec. See www.worldbank.org/trade/wto Anderson, K. and E. Valenzuela (2006), "The WTO's Doha Cotton Initiative: A Tale of Two Issues", mimeo, World Bank, Washington DC, March. See www.worldbank.org/trade/wto Anderson, K., E. Valenzuela and L.A. Jackson (2006), `GM Cotton Adoption, Recent and Prospective: A Global CGE Analysis of Economic Impacts', mimeo, World Bank, Washington DC, March. See www.worldbank.org/trade/wto Anderson, K., W. Martin and D. van der Mensbrugghe (2005), `Would Multilateral Trade Reform Benefit Sub- Saharan Africa?' World Bank Policy Research Working Paper 3616, Washington DC, June. See www.worldbank.org/trade/wto (Forthcoming in Journal of African Economies 15, 2006.) Baffes, J. (2005), "The `Cotton Problem' ", World Bank Research Observer 20(1): 109-43, Spring. Sumner, D.A. (2006), "Reducing Cotton Subsidies: The DDA Cotton Initiative", Ch. 10 in K. Anderson and W. Martin (eds.), Agricultural Trade Reform and the Doha Development Agenda, London: Palgrave Macmillan (co-published with the World Bank). WTO (2004c), "Agriculture: The Cotton Sub-Committee", Geneva: World Trade Organization, November 19. See www.wto.org/english/tratop_e/agric_e/cotton_subcommittee_e.htm 6