2008 World Development Report , The World Bank - LCR series - Agriculture for LCSAR Agriculture and Agriculture and Rural Development Rural Development Development 69393 Agricultural trade policy for rural development While primary agriculture accounts for only about 6 percent of in light of the food price crisis, it is more critical than ever that total GDP in Latin America, agricultural and food exports com- countries follow sensible unilateral trade policies to ensure that bined represent well over 20 percent of the region’s total mer- the trade system does not backslide while waiting for the con- chandise exports.1 This share of exports is considerably higher clusion of the Doha Round. in some countries, such as Nicaragua (85 percent), Argentina (48 percent), Ecuador (43 percent), Chile (34 percent), and Bra- zil (32 percent). Table 1 shows that the region as a whole has Table 1: Index of Comparative Advantage in a strong “revealed comparative advantage� in trade of agricul- Agriculture and Processed Food (World = 1.0) tural and food products. Even in Mexico, which does not have a revealed comparative advantage in agricultural trade, exports Period Latin American as a share of agricultural GDP has roughly doubled (to over 30 country/subregion 1985-89 1995-99 2000-04 percent) since the early 1990s, largely due to the huge growth in high-value fruit and vegetable exports under NAFTA. The cur- Argentina 4.4 4.9 5.4 rent situation of high prices and short supplies in agricultural markets presents an excellent opportunity for the Latin America Brazil 2.4 3.2 3.6 and Caribbean region to help keep the world fed while reaping Chile 2.3 3.4 3.9 significant economic benefits. Colombia 3.6 3.2 2.6 Because trade clearly has the potential to play a major role in Dominican Republic 3.2 1.2 4.7 increasing agricultural growth and reducing rural poverty in Latin America and the Caribbean (LCR), creating external and Ecuador 3.2 5.5 4.9 domestic policy environments that support agricultural export Mexico 0.9 0.7 0.6 is important to both sectoral and national interests. In this light, the 2008 World Development Report (WDR) examines the po- Nicaragua 6.1 7.4 9.5 tential gains from global trade liberalization (and how the Doha Round negotiations could best reach a “pro-development� out- Other countries 1.7 2.5 n.a. come) as well as domestic policies and investments to better Caribbean 0.9 1.5 n.a. integrate the agricultural sector into the global economy. While negotiations in the WTO broke down in July 2008, the parties Central America 5.2 5.4 5 have declared their intentions to continue to try to conclude South America 1 1.6 1.6 an agreement, and as World Bank president Zoellick has noted, “There was a good package of results left on the table. It would All Latin America 2.1 2.2 2.2 be a mistake for the world economy and harmful for develop- Note: The index of comparative advantage is the share of agriculture ing countries not to retrieve it.� In the mean time, and especially and processed food in the merchandise exports of a country (or region) relative to the share of world agriculture and processed food in world 1 Figures in this paragraph and in Table 1 come from Sandri, D., E. Va- merchandise exports. Index value greater than 1 indicates compara- tive advantage in production of agricultural and food products. Source: lenzuela, and K. Anderson (2007), “Economic and Trade Indicators for Sandri, Valenzuela, and Anderson (2007). Latin American,� Agricultural Distortions Working Paper 19, World Bank, Washington DC, December. Available at www.worldbank.org/agdistortions. 1 Multilateral Trade Policy Reform Figure 1. Estimated Real International Using global computable general equilibrium models, the Commodity Price Increases following World Development Report 2008 finds that the costs of current Complete Trade Liberalization (%) agricultural trade policies are high for both industrialized and developing countries, so reform offers significant potential wel- fare gains. It is estimated that by 2015 the annual global costs of Cotton 20.8 current trade tariffs and subsidies would be between $100 bil- lion and $300 billion, about two-thirds of which would be from Oilseeds 15.1 agricultural policies.2 This far exceeds the share of agriculture and processed food in global GDP (6 percent) or international Dairy products 11.9 trade (9 percent). In both agricultural and nonagricultural trade it is estimated that half the tariff and subsidy costs to develop- Coarse grains 7.0 ing countries are caused by their own policies and half by the policies of industrialized countries. In the agriculture sector, the Wheat 5.0 policies of the industrialized countries currently cost develop- ing countries about $17 billion per year—roughly five times the Processed meat 4.3 amount of total overseas development assistance to agricul- ture. 3 Rice 4.2 It is estimated that full trade liberalization would increase in- ternational commodity prices by an average of 5.5 percent for Fruit and vegetables 2.8 primary agricultural products and 1.3 percent for processed foods (Figure 1).4 Developing countries would increase their Other crops 2.6 rate of annual growth in agricultural output from 3.9 percent to 4.2 percent and their share of world agricultural exports from Sugar 2.5 54 percent to 65 percent. This is an 8 percent increase in the growth rate and a 4.3 percent increase in agricultural output Livestock 2.5 over 10 years. The gains would be greatest in Latin America and the Caribbean, where the annual rate of growth in agricultural Vegetable oil and fats 1.9 output would increase by 2 percentage points (Figure 2), due mainly to the region’s strong comparative advantage in these products (Table 1). Figure 2. Impact of Global Trade Reform on Agricultural Output Growth Represents the difference between 2005 baseline without liberalization and the estimated average annual agricultural growth rate in 2015 under full liberalization 2.5 2.0 Annual agricultural output growth (%) 2.0 1.5 1.0 0.3 0.4 0.5 0.0 0.0 0.0 -0.5 -0.2 -0.1 -1.0 -1.5 -2.0 -1.7 Developed Developing SSA SA EAP MENA ECA LAC countries countries Source: Derived from Anderson, Martin, and van der Mensbrugghe (2006). Abbreviations: SSA—Sub-Saharan Africa, SA—South Asia, EAP—East Asia and Pacific, MENA—Middle East and North Africa, ECA—Europe and Central Asia, LCR—Latin America and the Caribbean. 2 Anderson and Martin 2005; Bouët 2006a; Polaski 2006. 3 The $17 billion cost is a conversion to 2005 GDP and prices of the static share of the $26 billion in 2015 estimated by Anderson, Martin, and van der Mensbrugghe (2006). 4 Anderson, Martin, and van der Mensbrugghe 2006. 2 While full liberalization is an unlikely outcome of the nego- Domestic Agricultural Trade Policies tiations, the Doha Round is an opportunity to realize at least part of the potential gains of full trade liberalization. While all A large-scale study in the late 1980s concluded that agricultural three “pillars� of the agricultural negotiations are important— policies in most developing countries discriminated heavily market access (mainly tariffs), export subsidies, and domestic against the agriculture sector.7 The study found that without subsidies—tariff reduction is expected to have the greatest government trade and pricing policies (including those carried impact on global welfare and poverty reduction.5 A “develop- out via state-owned marketing enterprises) farmers would have ment friendly� Doha agreement would (a) significantly reduce received higher prices. More importantly, in many countries tariff bindings, (b) lower the industrialized countries’ subsidies this anti-agricultural bias was magnified by policies that either for the commodities that matter most for developing countries, granted high protection to other sectors or lowered returns to such as cotton, and (c) tightly limit exemptions under “sensitive exports in general (such fixing exchange rates to overvalue cur- products� for industrialized countries and “special products� for rency). developing countries. Multilateral trade reform is also possible in a regional or bilateral Figure 3: Relative Rates of Assistance for context. Since trade between developing countries is a growing Asia, Africa, and Latin America (1965 to 2004) share of their overall trade, improving the access of develop- 10 ing countries to one another’s markets through South–South regional agreements can have a significant effect. On average, 0 countries in Latin America are already involved in seven re- 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 -10 gional trade agreements, which often lower not only tariffs but also barriers associated with regulations, standards, and border -20 Percent crossings. However, trade agreements with low barriers between -30 members but high barriers for nonmembers may increase trade within the group but actually reduce the members’ total trade. -40 A recent World Bank review of regional agreements concluded LAC -50 that the ones most likely to increase national incomes are those Africa with low external “most-favored nation� tariffs, few sectoral and -60 Asia product exemptions, nonrestrictive rule-of-origin tests, mea- sures to facilitate trade, rules on investment and intellectual Five-year weighted averages with value of production at undistorted property that are appropriate to the development context, and prices as weights. RRAs for China (for 1965-81) have been extrapolat- implementation schedules put into effect on time.6 ed back assuming they were the same as the average for years 1982-89. Source: Chapters 2-9 of this volume and Chapter 1 of Anderson (2008). A recent study revisited this earlier work and found that current policies in much of the developing world have greatly improved incentives for agricultural production.8 Figure 3 shows that the average “relative rate of assistance� for agriculture in Latin America—that is, the extent to which government interven- tions improve or deteriorate returns in agricultural production relative to other activities—improved from negative 40 percent in the 1970s to almost zero in 2000–04.9 These averages ob- scure important variations between countries, but the overall conclusion, that policies have improved a great deal, is correct. An appropriate trade policy for countries in Latin America, which generally have relatively low trade barriers, is to continue pursuing greater integration into world markets. But to take full advantage of these policies and also reduce the burden on those who lose from liberalization, a host of “behind the border� measures must be addressed. It is important to examine the dis- 5 Anderson, Martin, and Valenzuela 2006; Hertel and Keeney 2005. tinct demographic and geographic characteristics of adversely 6 World Bank 2004b. affected groups and analyze the magnitude of their potential 7 Krueger, A.O., M. Schiff, and A. Valdés (1988), “Agricultural Incentives in losses and gains. Transitional support might include grants to Developing Countries: Measuring the Effect of Sectoral and Economy- facilitate production shifts (as was done in Turkey)10 or cash wide Policies,� World Bank Economic Review 2(3): 255-272. Krueger, A., M. Schiff, and A. Valdés (1991), The Political Economy of Ag- transfers and social safety nets (as in Mexico).11 ricultural Pricing Policy, Volume 1: Latin America, Baltimore: Johns Hop- kins University Press for the World Bank. 9 The LCR countries included in this study were Argentina, Brazil, Chile, 8 Anderson, K. (ed.) (2008), Distortions to Agricultural Incentives: A Glob- Colombia, Dominican Republic, Ecuador, Mexico, and Nicaragua. al Perspective, London: Palgrave Macmillan; and Washington DC: World 10 World Bank 2004a. Bank (forthcoming). 11 Ashraf, McMillan, and Zwane 2005. 3 However, cash transfers to compensate for losses are insufficient ness and generate spillover benefits for domestic consumers. to induce a supply response. Targeted investments, such as in- Although there will inevitably be winners and losers, this view frastructure investments and extension services, are needed to suggests that enhanced capacity to comply with stricter stan- improve productivity or education and to facilitate transition.12 dards can provide sustainability and profitability in the long In particular, the supply response of smallholders depends on term. rural infrastructure (irrigation, roads, transport, power, and tele- communications), markets, finance, research, and extension.13 Suppliers rarely face all-or-nothing choices when making chang- es and investments to conform to emerging standards and they Public spending has often been diverted away from these need to weigh the costs and advantages for different products much needed long-term investments and into various types and market segments.19 In some cases, there may be better op- of agricultural input subsidies, price supports, or other direct portunities to serve domestic, regional, or industrial markets payments. One recent study in LCR showed that shifting public that impose less stringent standards or allow more time to im- spending away from these subsidies and into true public goods plement them. In any case, addressing the export challenges of can increase rural incomes, lower poverty rates, and reduce en- SPS standards requires joint public and private efforts. vironmental impacts.14 It found that a reallocation of just 10% of the subsidy expenditures to supplying public goods instead may increase per capita agriculture income by about 5%. One increasingly important area of public expenditure in LCR is support for coping with sanitary and phytosanitary (SPS) standards in trade, as discussed in Chapter 5 of the WDR. In re- cent years many countries have tightened their SPS standards, expanded their coverage, or introduced rules to ensure fair competition, reduce information costs to consumers (organic foods), and promote competition based on quality.15 In parallel, the private sector has expanded its own standards and supplier protocols.16 There is concern that developing countries—and especially smaller countries, enterprises, and farmers—will be excluded from export markets because they LCRk the admin- istrative and technical capacity to comply with SPS standards or cannot afford the costs of compliance. Anecdotal cases and research lend some support for this “standards as barriers� per- spective.17 But an alternative view highlights the opportuni- ties in the evolving standards environment and the scope for capitalizing on them.18 Common standards across international markets can reduce transaction costs. Standards can also pro- vide incentives for modernizing supply chains and help clarify References the proper risk management functions of government. Greater Adapted from the World Development Report 2008 Agriculture for De- attention to good practices may improve export competitive- velopment with additional information from regional studies and anec- dotal evidences. 12 Coady, Dorosh, and Minten, forthcoming. 13 Binswanger 1989; Schiff and Montenegro 1997. About the Authors 14 “Should governments stop subsidies to private goods? Evidence from John Nash is the Lead Economist with the World Bank’s Latin America rural Latin America�, Ramón López, Gregmar I. Galinato, Journal of Pub- lic Economics 91 (2007) 1071–1094. and the Caribbean Region Agricultural Sustainable Development De- 15 Buzby, Frenzen, and Rasco 2001; Henson 2006. partment. 16 Unnevehr 2003. 17 Henson and Caswell 1999; Jha 2002; OECD 2003; Wilson and Abiola 2003. 18 Jaffee and Henson 2004; World Bank 2005b. Latin America and the Caribbean Region 19 World Bank 2005c. The 2008 World Development Report (WDR) “Agriculture for Development� characterizes agriculture as vital development tool for achieving the Millennium Development Goal that calls for halving by 2015 the share of people suffering from extreme poverty and hunger. The report provides guidance to governments and the international community on designing and implementing agriculture- for-development agendas that can make a difference in the lives of hundreds of millions of rural poor. This brief is part of a series prepared by LCSAR that summarizes and interprets the principal messages of the WDR 2008 and discusses region-specific implica- tions for Latin America and the Caribbean (LCR). The series comprises the following topics: (i) Agricultural Innovations in Science and Technology, (ii) Value Chain Development and Integration of Small Farmers, (iii) Agricultural Trade Policy, (iv) Land Administration and Access, and (v) Territorial Development. 4