Previous PageTable Of ContentsNext Page

Agricultural trade, trade policies and the global food system

9.1 Introduction

For centuries countries have relied on trade in agricultural and food commodities to supplement and complement their domestic production. The uneven distribution of land resources and the influence of climatic zones on the ability to raise plants and animals have led to trade between and within continents. Historical patterns of settlement and colonization contributed to the definition of trade patterns and to the emergence of an infrastructure to support such trade. More recently, transnational firms with global production and distribution systems have taken over from post-colonial trade structures as a paradigm for the organization of world agricultural trade. Changes in consumer taste have encouraged the emergence of global markets and added to the significance of trade. Few countries could survive the elimination of agricultural trade without a considerable drop in national income, and none could do so without considerable reduction in consumer choice and well-being.

The projections presented in Chapter 3 suggest that the role of trade in meeting global food demand will further increase over the next 30 years. Developed countries will provide a growing share of developing countries’ food needs, and in return will continue to import larger quantities of other agricultural products, notably tropical beverages, rubber and fibres. However, the developing countries are not a homogenous trading bloc. While the group as a whole will increase its net exports of tropical products and import more and more temperate-zone commodities, within the group there will remain important net exporters of temperate-zone commodities.

Like all projections, the global trade outlook presented in this chapter is based on numerous assumptions about the likely evolution of policies that will affect trade flows, as well as basic trends in income, population and productivity. A principal premise of the quantitative projections is the continuation of existing policies related to the support and protection of agriculture, including policy changes that will be implemented in the future, for example, the EU's Everything but Arms Initiative (EBA) that foresees a liberal import regime for rice and sugar from the least developed countries (LDCs) in the future. If policies differ substantially in the future, so will the outcomes. If, for instance, the reform process that began under the Uruguay Round Agreement on Agriculture were to achieve a fundamental reform of the sector, and if there were significant reductions in production-enhancing subsidies and protection in industrial countries, this could have an impact on predicted trading patterns. And if policy reforms extended beyond the developed countries and led to the removal of the remaining bias against agriculture in the policy of several developing countries, this could mobilize resources to enhance productivity and stimulate development of the rural economy.1 As a consequence, much of the qualitative discussion in the chapter is an attempt to indicate how policies might develop over the next three decades.

This chapter begins with a discussion (Section 9.2) of the evolution of patterns in global agricultural trade. The analysis reviews the changing share of agricultural trade relative to manufactures in world trade and identifies the rapidly changing role of agricultural trade in the developing countries. This is followed by a discussion of the agricultural trade balances for developing countries by major commodity categories. It shows changes in the net trade balances of these product categories. It also provides an overview of complementarity and competition in global agricultural trade and how, where, and to what extent policies have affected the current trading patterns between developing and developed countries.

Section 9.3 examines the trade policy environment, focusing on the Agreement on Agriculture. It assesses the progress made to create a “fair and market-oriented” trading system for agriculture, and looks at other trade policy developments in agriculture, particularly the role of preferential and regional trade agreements. Section 9.4 addresses the prospects for and likely impacts of further reforms in the current and future rounds of negotiations, with particular attention to the ways in which the new emphasis on non-trade concerns will influence trade liberalization. This is followed by an analysis of how different policy options could affect and alter the baseline projections to 2030. The focus is on the stake of developing countries in international agricultural trade policy reform. The chapter concludes (Section 9.5) by looking at the emerging issues that will affect the overall trade policy environment and how they may influence agricultural trade in the coming three decades.

9.2 Long-term trends in the pattern of global agricultural trade

9.2.1 From agricultural exports to manufactures exports

The last 50 years have witnessed an impressive growth in international trade. The volume of global merchandise trade has increased 17-fold, more than three times faster than the growth in world economic output.

A number of factors contributed to this growth. Average import tariffs on manufactures, for instance, fell from 40 to 4 percent over the four decades of trade negotiations under the General Agreement on Tariffs and Trade (GATT) (Abreu, 1996). Non-policy factors have also been important, including a reduction in transport costs and new transportation facilities as well as cheaper and more efficient communications. Moreover, growth in manufactures trade has been spurred by the rapid expansion of intra-industry and intrafirm trade, exploiting a division of labour within companies operating across various countries or continents. Much of this trade is an exchange of components or semi-processed products. Finally, manufactures benefited from a virtuous circle whereby the gains from trade translated into higher incomes and in turn fuelled growth in trade.

Agricultural trade has also grown during the last 50 years, but only at about the rate of global economic output. Notable among the factors that contributed to this relatively slow growth in trade was the failure to include agriculture fully in the multilateral trade negotiations under GATT that were so successful in reducing industrial tariffs. As a result, agricultural tariffs are as high now, on average, as industrial tariffs were in 1950. The effects of high border protection have been compounded by domestic support policies in many developed countries and in some developing countries by policies that promoted import substitution at the expense of international trade.

Growth in agricultural exports from developing countries was also held back by the limited absorption capacity of their export markets. The major part of their agricultural exports was destined for largely saturated developed country markets without much responsiveness in demand. Tropical products such as coffee, cocoa, tea or rubber were most severely affected by these limitations. Rising output from developing countries met inelastic demand in developed countries and resulted in a persistent downward pressure on prices. In fact, lower prices cancelled out much of the gains in export volumes with the result that export earnings increased only moderately.

Moreover there has been very little intra-industry or intrafirm trade in food and agricultural products. This reflects mainly the nature of agricultural trade, which is often largely determined by different agro-ecological conditions. But intra-industry trade has also been held back by traditional trade and investment barriers that made international sourcing more difficult than for manufactures. When and where these barriers have declined, the exchange of processed and semi-processed agricultural products has increased considerably and brought about levels of intra-industry/intrafirm trade close to levels observed for non-agricultural products (see Chapter 10). Much of this trade has been stimulated by the activities of global food companies and traders, but has also involved retailers and small food exporters exploiting niche markets.

The result of the rapid export growth in manufactures and slow growth in agricultural exports was a dramatic decline in the relative importance of agricultural exports. The share of developing countries’ agricultural exports in their overall exports fell from nearly 50 percent at the beginning of the 1960s to barely more than 5 percent by 20002 (Figure 9.1). Even for the group of the 49 LDCs, where agriculture is often the largest sector of the economy, the share of agricultural exports declined from more than 65 percent in the early 1960s to less than 15 percent by 2000 (Figure 9.1). Whereas the low shares for developing countries are, among many other factors, also a reflection of protectionist policies in OECD countries, OECD policies have probably contributed to keeping shares high for the LDCs. The numerous preference agreements in which they participate offer the LDCs higher export prices and encourage them to export more than otherwise.

Figure 9.1
The agricultural trade balance and share of agricultural exports

9.2.2 From net exporters to net importers of agricultural commodities

Together with the overall decline in the share of agriculture in international trade, the structure of agricultural trade has changed markedly. One manifestation of this change is the balance in food trade between developed and developing countries. In 1961/63 developing countries as a whole had an overall agricultural trade surplus of US$6.7 billion, but this gradually disappeared so that by the end of the 1990s trade was broadly in balance, with periodic minor surpluses and deficits. The outlook to 2030 suggests that the agricultural trade deficit of developing countries will widen markedly, reaching an overall net import level of US$31 billion. Net imports of food will increase to about US$50 billion (in US$ of 1997/99, for details see Table 9.1).

The 49 LDCs have been in the forefront of this shift: their agricultural trade deficit has increased so rapidly that, already by the end of the 1990s, imports were more than twice as high as exports (Figure 9.2). The outlook to 2030 suggests that this trend will show no sign of abating. The agricultural trade deficit of the group of LDCs is expected to widen further and will increase overall by a factor of four over the next 30 years.3

Figure 9.2
Least developed countries (LDCs) have become major net importers of agricultural products

As already discussed in Chapter 3, the evolution of the overall net agricultural trade balance per se may not necessarily represent a deterioration of the overall economic situation in a developing country. For some countries, a growing agricultural trade deficit may simply reflect rapid overall development. This is, for instance, the case of countries such as the Republic of Korea in which the growing agricultural deficit has gone hand in hand with high rates of overall development and growing food consumption. Likewise, rising imports of vegetable oil in China primarily reflect an improved ability to meet domestic food needs through imports. A declining agricultural trade balance is, however, a negative developmental outcome in countries that continue to depend to a high degree on export earnings from agriculture or that have to divert scarce foreign exchange resources (eventually building up unsustainable foreign debt) to pay for growing food imports. It is an even more negative indicator where such food imports are not associated with rising food consumption per capita but are necessary just to sustain minimum levels of food consumption.

Commodity dependence. Notwithstanding the declining importance of agricultural exports for developing countries as a whole, some developing countries still rely heavily on agricultural exports for their foreign exchange earnings. In more than 40 developing countries, the proceeds from exports of a single agricultural commodity such as coffee, cocoa or sugar account for more than 20 percent of total merchandise export revenue and more than 50 percent of total agricultural export revenue (Figure 9.3). In Burundi, for example, coffee exports alone accounted for 75 percent of the country's foreign exchange earnings in 1997/99. Half of these countries are located in sub-Saharan Africa, and three-quarters are LDCs and/or small islands. The heavy reliance on a few crops is often a reflection of the fact that many of these economies are very small.

Figure 9.3
Dependence on agricultural export earnings by commodity, 1997/99
Share of export earnings in total mechandise exports (percentage)

Dependence on a few commodities brought about numerous problems for these countries during the period of low prices in 1999/2001. Particularly low world prices for coffee and sugar reduced their overall foreign exchange availability, lowered rural wages, increased rural poverty and thus underlined the importance of undistorted agricultural trade for overall economic development. The sluggish demand for primary agricultural commodities and the recurring conditions of boom and bust in their exports created problems for commodity-dependent economies. Unstable commodity prices and export earnings are known to make development planning more difficult and to generate adverse short-term effects on income, investment and employment.

9.2.3 Projected shifts in the agricultural trade balance of developing countries

The outlook is that developing countries will become significant net importers, with a trade deficit of almost US$35 billion by 2030 (Table 9.1).4 The following section focuses on the structural changes within agricultural trade by commodity group, to identify the main factors that have brought about these shifts in trade patterns and the likelihood of their continuation.

Table 9.1: Trade flows between developing and developed countries


Net trade of developing countries
(negative values denote net imports)

Cumulative increase

OECD support

Commodity category









Billion US$ (current)

Billion US$
(in US$ of 1997/99)



Total agriculture








Total food








1. Temperate-zone








Cereals (excluding rice)
















Coarse grains








































2. Competing
















Vegetable oils and oilseeds








Fruit, vegetables and citrus
























Cotton lint
















3. Tropical
















































4. All other commodities








Other study commodities








Commodities not covered in this study








1 Based on projected growth in quantities (as in Chapter 3), applied to the 1997/99 trade values from FAOSTAT, rounded numbers; the projected trade balances in values are implicitly expressed in constant US$ of 1997/99, while the historical values are in current US$. It follows that the implied rates of change over time are not comparable between past and future. For such comparisons refer to Chapter 3.
2 "Guesstimates".
3 Pro-rated according to shares in values of production.
n. a.=not available.

The first category includes temperate-zone commodities (wheat, coarse grains and livestock products) of which the developed countries produce the bulk of world-exportable surpluses. These commodities are also highly protected in many countries and the world market price is often influenced by heavily subsidized export surpluses. Developing countries are already significant importers of these commodities and the projections suggest that by 2030 their net imports will have increased further, e.g. more than 2.5-fold for wheat and coarse grains (Table 3.5) and almost fivefold for meat (Table 3.14).

One of the most important changes that affected the overall agricultural trade balance of developing countries was the rapid growth in imports of temperate-zone commodities. The net imports in this product category increased by a factor of 13 over the last 40 years, rising from a minor deficit of US$1.7 billion in 1961/63 to US$24 billion in 1997/99 (Table 9.1). These figures are in current US$.

A number of factors contributed to this shift. First, developing countries found it increasingly difficult to compete with subsidized surpluses of temperate-zone commodities disposed of by OECD countries. These subsidies hindered exports from all developing countries, although some countries such as Argentina, Brazil and Uruguay managed to remain net exporters. Subsidies and subsequent surplus disposal put downward pressure on international prices, and held back export volumes and earnings for temperate-zone commodities. Second, the developing countries themselves added to the growing trade deficit by taxing their own agriculture directly through low procurement prices and quotas or indirectly through overvalued exchange rates and high protection rates for non-agricultural goods. In many cases, these policies had been in place before OECD surplus disposal policies became a significant factor in world agricultural trade. Third, overall economic development contributed to higher imports of temperate-zone commodities. With higher incomes and more people to feed, demand for temperate-zone commodities in developing countries increased fast, too fast for domestic production to keep pace. The rise in import requirements was particularly pronounced in countries where agro-ecological constraints restrained agricultural production growth and where urbanization and income growth brought about rapid increases in demand (notably in the Near East/North Africa region).5

Within the group of temperate-zone commodities, there is a discernible shift in trade from cereals to livestock products, notably to meats. This is the result of various mutually reinforcing developments that took place in parallel in developed and developing countries. In developed countries, technological and organizational progress in livestock production (vertical integration, etc.) outpaced productivity gains in cereal production. This made it more profitable to convert cereals into meat domestically and export meats rather than cereals (OECD, 1998). In developing countries, income growth, particularly in Asia, has contributed to a shift in consumption patterns from grain-based diets to livestock-based diets with rapidly rising per capita consumption levels (Chapter 2). Moreover, alongside overall economic development and urbanization, some developing countries that recorded higher income growth also created the infrastructure facilities (e.g. cold chains) that were needed to handle livestock product imports.

The long-term outlook presented in Chapter 3 suggests that the shift from cereals to the meat trade will continue. While cereals (excluding rice) will still account for most of the increase in absolute terms, meats are expected to exhibit the strongest relative growth. Developing countries’ imports of livestock products will increase very rapidly, although starting from relatively low trade levels in the base year 1997/99. Also developing countries' overall imports of temperate-zone products will continue to rise, albeit at a lower pace. By the year 2030, developing countries are expected to increase their net imports of temperate-zone commodities to about US$61 billion (in 1997/99 US$, Table 9.1; for changes in quantities see Chapter 3, Tables 3.5, 3.14).

One of the crucial issues in this context is whether and to what extent developing countries would be able to expand their production and exports if policy distortions, particularly those imposed by OECD countries, were to be removed. Would the removal of distortions change developing countries' net trade position? Numerous studies suggest that this is unlikely to be the case. Instead, there is wide agreement that the limiting effect of agro-ecological constraints often outweighs the effects of policy distortions. In fact, a removal of OECD subsidies would largely result in a shift in market shares from subsidized to unsubsidized producers within the group of OECD countries. Only a few developing countries with additional production capacity for temperate-zone commodities, such as Argentina, Brazil, Uruguay or Thailand would expand their net export positions (see, for instance, FAO, 2000e, p. 27-28). The majority of developing countries, however, would not shift from net importers to net exporters. Where the responsiveness of agricultural supply is particularly low (as in many LDCs) and where non-agricultural protection remains high, countries will experience higher food import bills and a deterioration of their terms of trade.

The second category includes primarily competing commodities, i.e. those commodities that are produced in both north and south, even though they may originate from different primary products (sugar from beets or cane, oil from several oilcrops), or commodities for which competition is limited to certain parts of the year (fruit and vegetables). Overall, there is considerable competition for export shares in these markets. Subsidies in OECD countries often offset the comparative advantage of producers in developing countries.

Developing countries’ export interests are adversely affected by OECD policy distortions affecting competing products. Many developing countries have a comparative advantage in producing these commodities, either because their production is labour intensive (fruit and vegetables) and/or because they are strongly favoured by the agro-ecological conditions of tropical or subtropical regions (tropical fruit, sugar and rice). Developing countries’ net exports in this product category amounted to about US$6 billion in 1997/99, about twice as much as in the early 1960s (in current US$, Table 9.1). At a net export level of US$8.4 billion, fruit, vegetables and citrus accounted for the largest portion under this heading, and exhibited the highest growth over the past 40 years.

Yet export growth might have been even more rapid had it not been for policy distortions, particularly OECD subsidies that totalled about US$111 billion in 1998/2000 (Table 9.1). Fruit and vegetables, together with rice, accounted for nearly three-quarters of this OECD subsidy. Developing countries, at least in aggregate, are likely to benefit from a cut in OECD subsidies and an increase in access to developed countries’ markets.

The third category encompasses tropical commodities that are mainly produced in developing countries, but primarily consumed in OECD countries. These are mostly tropical products such as coffee, cocoa or rubber for which developing countries have been increasing output substantially over the past decades. Developed countries’ import markets for these commodities have become increasingly saturated. Demand has become inelastic, and prices are subject to a secular decline. Since developed countries do not produce these commodities in significant quantities, they do not support or protect these markets.

Developing countries have been rather successful in expanding production and exports of tropical products. Overall, net exports of tropical products have increased by a factor of five, from about US$3.8 billion in 1961/63 to about US$19.2 billion by the late 1990s at current prices (Table 9.1). Export growth will continue over the next 30 years, and in 2030 could be higher by some 36 percent (in volume terms).

The single most important commodity in this group is coffee. Net exports of coffee account for about half of total net exports of tropical products. They increased by a factor of 5.5 since the early 1960s (in current prices), largely matching the growth of overall net exports of tropical products. Production of coffee (green beans) increased by 60 percent from 1961/63 to 1997/99 and, as domestic consumption in developing countries remained low, much of the additional produce went into export markets. Coffee exports rose rapidly over much of the 1960s and 1970s. But as developed countries’ markets for coffee became increasingly saturated, growth in export volumes was no longer matched by a similar growth in export revenues. From 1974/76 to 1998/2000, volumes of coffee exports doubled while export revenues increased by merely 15 percent (Figure 9.4).

In addition to these demand-related reasons for a slowdown in export growth, growth in production and trade was also affected by policies. In developing countries, the export taxes of large exporting countries resulted in periods of attractive prices and lured more competitive countries into these markets. In developed countries, tariff escalation on more processed products kept a
lid on exports of processed coffee and thus contributed to the slow growth in export earnings. Over and above tariff escalation, OECD trade and support policies play no significant role in these markets. Overall OECD countries spent less than US$1 billion, i.e. less than 0.5 percent of total producer support to subsidize the production of these commodities in 1998/2000 (Table 9.1).

The country that made the most substantive inroads into the international coffee market was Viet Nam. Over the 1990s, the country’s coffee exports rose by a factor of ten in terms of quantities and a factor of five in terms of revenues (Figure 9.4). Together with rapidly rising rice production and exports, Viet Nam’s coffee boom provided a boost to the country’s overall rural economy. Fertilizer use increased by about 50 percent and agricultural GDP grew by about 4.5 percent p.a. The coffee boom also boosted food production and thus contributed to a reduction in hunger and poverty. Rural poverty fell from 66 percent in 1993 to 45 percent in 1998 (World Bank, 2001c) and from 1990/92 to 1997/99, the share of undernourished persons declined by 8 percentage points or 4 million people (FAO, 2001a).

Figure 9.4
Coffee exports, world and Viet Nam

The coffee production and export boom in Viet Nam, however, also augmented the downward pressure on international prices. By the end of 2001, international prices for green coffee had declined to a 30-year low. Low coffee prices had serious impacts for many rural poor in those areas where production was growing less rapidly. Wages for pickers, for instance, fell along with prices for coffee and reached levels below US$1-2 per day in many African and Latin American countries. Oxfam has documented such impacts on wages and rural poverty in a number of case studies for the United Republic of Tanzania and Mexico (Oxfam, 2001). Moreover, the low international prices also affected average coffee qualities. It rendered some of the previous efforts to create high-quality coffee segments unprofitable (for instance in Colombia) and, as much of the production in the newly emerging coffee producers such as Viet Nam was of inferior quality, it lowered both average prices and average qualities.

The effects of the coffee boom confirm a number of general problems that plague producers and markets for tropical products. First, where rapidly rising supply is primarily exported to largely saturated markets, prices tend to decline rapidly and sharply without necessarily resulting in the expected contraction in production. The asymmetry in the supply response is because many of these crops are perennials, grown on a multiyear basis with a relatively large share of total investment requirements locked in at the beginning of the investment periods. To increase the returns on the fixed investment, investors have to reduce the variable costs of production. In plantation production systems, this means lower wages for rural workers (pickers); in smallholder systems, it means a considerable decline in family incomes. Second, while the promotion of production in new low-cost environments (Asia) can be an interesting agricultural development strategy in a given country, the same policy pursued simultaneously in many countries can nullify or reverse the advantage for the original beneficiary. Markets for tropical agricultural products are particularly prone to this fallacy of composition problem (see also Section 3.6 in Chapter 3 for a discussion of problems with tropical commodities).

9.3 The trade policy environment for agriculture

The environment in which agricultural trade policy is formed includes both the domestic political balance between competing ideas and interests and the international climate for trade rules and negotiations. This section examines current issues in the trade policy environment for agriculture and the extent to which a continuation of present trends is likely.

9.3.1 Global trends in trade policy reform

At the completion of the Uruguay Round of multilateral trade negotiations, the Agreement on Agriculture was hailed as an important first step towards the fundamental reform of the international trading system for agriculture. Since then, however, many countries have been disappointed by the modest benefits deriving from it. Indeed, some observers argue that the Agreement on Agriculture may have “institutionalized” the production- and trade-distorting policies of the OECD countries without addressing the fundamental concerns of developing countries (Green and Priyadarshi, 2002). Negotiations on the continuation of the reform process began in March 2000, as mandated under Article 20 of the Agreement on Agriculture, but the progress that can be achieved will depend, in part, on the experience of World Trade Organization (WTO) Members with the reform process thus far.

The Doha Round of multilateral trade negotiations. Article 20 of the Agreement on Agriculture declares Members’ “long-term objective of substantial progressive reductions in support and protection resulting in fundamental reform” and pledges to continue the reform process, “taking into account:

The negotiations on agriculture were subsumed in a broader round of multilateral trade negotiations launched by WTO at its Fourth Ministerial Conference held in Doha, Qatar in November 2001 (WTO, 2001e). 6 The round will consider agriculture as part of a “single undertaking”, and the negotiations are scheduled to conclude not later than 1 January 2005. Modalities for further commitments on agriculture are to be established no later than 31 March 2003 and comprehensive draft schedules based on these modalities should be submitted no later than the date of the Fifth Session of the WTO Ministerial Conference to be held in September 2003 in Cancún, Mexico.

During the first phase of the negotiations on agriculture, which ended in March 2001, WTO Members submitted their proposals and other papers reflecting their positions and concerns. A total of 44 negotiating proposals were submitted, sponsored by 125 countries either individually or in groups. The proposals address the range of issues mandated for the negotiations, including the three pillars of the Agreement on Agriculture (market access, domestic support and export competition), as well as the cross-cutting issues of SDT for various groups of countries and non-trade concerns ranging from food security to animal welfare. They span a broad range of possible outcomes, from the rapid elimination of all trade-distorting barriers and subsidies to a slowing or even reversal of the reforms undertaken in the Uruguay Round.

The Ministers made a commitment to undertake negotiations on agriculture that, without pre-judging the outcome, would aim at substantial improvements in market access, the reduction of all forms of export subsidies with a view to phasing them out, and substantial reductions in trade-distorting domestic support. It was agreed that SDT should be applied to developing countries to reflect their development needs, including food security and rural development. Non-trade concerns will also be taken into account in the negotiations. Given the wide divergence of the initial positions, radical reform seems unlikely in the near term. Nevertheless, successive rounds of gradual liberalization implemented over the next 30 years could result in “fundamental reform of the sector” within the timespan of this study.

Market access. Bound tariffs remain high in the OECD countries, especially for temperate-zone basic food commodities and products that compete with them. Developing countries typically also have high bound tariffs on agricultural goods, at 68 percent on average, but their applied tariffs are often much lower. Some countries bound tariffs for certain sensitive staple products at very low levels, often because they had previously relied on other means to control imports, such as quota exemptions under GATT Article XII for balance of payments purposes (Green and Priyadarshi, 2002). Many agricultural goods do, however, enter industrial countries duty free or at low tariffs. OECD tariffs on unprocessed tropical products are typically quite low, but tariff escalation is a problem in some of these commodities. In the EU and Japan, for example, bound tariffs on coffee escalate from zero percent on green beans to 7.5 and 12 percent, respectively, on roasted beans; while in India, this escalation goes from 100 to 150 percent (WTO, 1994).

Tariff schedules also reflect the prevalence of tariff rate quotas (TRQs),7 with both in-quota and over-quota tariffs listed, and the proliferation of complex duties that combine elements of specific and ad valorem duties (UNCTAD, 1999b). Thirty-eight WTO Members have tariff quota commitments in their schedules, for a total of 1379 individual quotas. But TRQs have done little to improve market access. Many countries allocated the quotas to traditional suppliers and counted pre-existing preferential access quotas as part of their minimum access commitments.8 New access volumes created by TRQs were typically less than 2 percent of world trade, and TRQ utilization rates or fill rates – the amount of trade that actually takes place relative to the TRQ level – have averaged about 60-65 percent (see OECD, 2001g). Unlike simple tariffs, TRQs generate market rents that may be captured by various groups (producer, exporting government, importing government, trader) depending on the administrative mechanism and the degree of market competition (ABARE, 1999). Thus, even if new market access is created, the producer may not capture the benefits, and there may be vested interests arguing against further expansion of these quotas.

Thirty-eight Members have the right to invoke special agricultural safeguards (SSGs) on a total of 6072 individual tariff items. During the implementation of the Agreement on Agriculture, a total of 649 tariff items have been subject to special safeguard notification, with more than half of the price-based actions being taken by the United States, and more than half of the volume-based actions taken by the EU (WTO, 2000a).9 Most developing countries do not have access to SSGs because they did not use the tariffication procedure. Their only practical means of stabilizing domestic markets in the face of import surges or fluctuating world prices is by varying their applied tariffs (within their bound rates), although such “price bands” could yet be challenged under the rules of WTO.

Domestic support. The Agreement on Agriculture included rules and disciplines on domestic support in recognition of the potential these policies have to distort trade. The specific provisions aim to ease trade conflicts between developed countries, to reform policies that resulted in overproduction in the past, and to ensure that commitments on market access and export competition are not undermined through domestic support measures.

Although the Agreement on Agriculture began the process of disciplining domestic support measures, the rules thus far have done little to restrain the subsidies provided by OECD countries. Furthermore, the types of policies available to developing countries under the Agreement may not be appropriate to the conditions of their agricultural sectors or sufficient to enable them to overcome the handicaps they face in international markets (Green and Priyadarshi, 2002).

The major portion of domestic support expenditures is provided by three WTO Members: the EU, Japan and the United States (Table 9.2). Although several OECD countries have reformulated their agricultural policies towards less distorting instruments, the overall level of support to OECD agriculture has not declined since the Agreement on Agriculture came into force. Support to agricultural producers in OECD countries, as measured by the producer support estimate (PSE)10 increased relative to the Uruguay Round base period to US$258 billion in 1998-2000 (OECD, 2001e). The aggregate measurement of support (AMS) limits have not been constraining thus far: few developed countries used more than 80 percent of their commitment level in 1996. Only four WTO Members have reported that they are using or have used blue box11 policies (the EU, Japan, Norway and the Slovak Republic).

Table 9.2: Domestic support expenditures 1996, US$ million



% of AMS commitment used

Measures exempt from reduction commitments

Total expenditures

De minimis

Blue box

Green box


























India (1995)



















Korea, Rep.







9 354




New Zealand
























South Africa
















United States








Source: WTO (2001a) and FAO (2000e).

The maximum permitted domestic support in most developing countries12 is set by the de minimis level and the Agreement on Agriculture provisions for product-specific support. This is because only a few developing countries reported AMS figures and only 12 set them at levels above the de minimis. In theory, domestic support could total 20 percent of the total value of agricultural production (10 percent product-specific plus 10 percent non-product-specific support). In fact, very few developing countries have provided support to agriculture (exempt and non-exempt) in excess of 2-3 percent of the value of production, and many have reduced support since the Agreement on Agriculture base period because of budgetary and administrative constraints. Indeed some countries still tax the agricultural sector or specific commodities, although they are not allowed to offset such negative product-specific supports with positive non-product-specific support.13

Export competition. The third main pillar of the Agreement on Agriculture dealt with export competition. Although the original GATT 1947 prohibited export subsidies in most sectors, an exception had been made for primary products, including agriculture. The Agreement on Agriculture sought to redress this omission by establishing disciplines on the use of export subsidies.

Only eight of the 25 WTO Members that have export subsidy commitments are developing countries, as defined by WTO (Brazil, Colombia, Cyprus, Indonesia, Mexico, South Africa, Turkey and Uruguay), and only one (Colombia) reported using export subsidies in 1998. The majority of direct export subsidies are used by the EU, which in 1998 accounted for more than 90 percent of all direct export subsidies under the Agreement on Agriculture (Table 9.3).

Table 9.3: Export subsidy use (million US$)



% of commitment



















New Zealand






South Africa






Source: WTO (2001a).

The terms of the Agreement on Agriculture also commit Members to negotiate disciplines on the use of export credit guarantees or food aid shipments, which might be used to circumvent the disciplines on direct subsidies. The United States is by far the biggest user of officially supported export credit guarantees, having provided 46 percent of all export credits used from 1995 to 1998. Three other exporters (Australia, the EU and Canada) were responsible for almost all the remainder. The subsidy element in export credits is fairly small compared with direct export subsidies (OECD, 2000b).

Food aid has also received attention as a possible means of circumventing disciplines on export subsidies. The Agreement on Agriculture stipulates only that food aid should be provided in accordance with FAO Principles on Surplus Disposal and, to the extent possible, fully in grant form. While most food aid is made in grant form, some donors provide food aid in kind, and there has been an apparent countercyclical relationship between such aid and international commodity prices, suggesting that aid is increased when prices are low and withdrawn when prices rise.

Non-trade concerns. The Agreement on Agriculture allows significant scope for governments to pursue important “non-trade” concerns – such as food security, environmental protection, rural development and poverty alleviation – through the use of domestic support measures that are exempt under the green box (Agreement on Agriculture, Annex 2), the de minimis or SDT provisions for developing countries (Agreement on Agriculture, Article 6). Furthermore, Article 20 of the Agreement requires that non-trade concerns be taken into account in the mandated negotiations on the continuation of the reform process in agriculture.

However, many countries feel that the Agreement on Agriculture does not provide sufficient policy flexibility for the pursuit of their non-trade concerns. They argue that agriculture has specific characteristics not shared by other sectors, and should not be subject to the same types of discipline on the use of subsidies and border protection.

Thirty-eight countries submitted a note in the ongoing WTO negotiations in which they address the “specific and multifunctional” characteristics of agriculture such as its contribution to rural development, food security, environment and cultural diversity (WTO, 2000b).14 Several formal negotiating proposals have elaborated on these themes. Norway, for example, used the non-trade concerns argument to justify its proposal that certain basic food commodities be exempt from further market access commitments and that production-distorting supports for commodities destined for the domestic market be subject to less stringent AMS disciplines (WTO, 2001a). Similarly, Japan uses the multifunctionality of agriculture to justify its call for discretion in setting tariffs and providing domestic support (WTO, 2000c).

Most WTO Members have agreed that agriculture is multifunctional, although several have argued that it is not unique in this regard. The Members have also agreed that each country has the right to pursue its non-trade concerns. The question being debated in the WTO is whether “trade-distorting” subsidies, or other subsidies not currently exempt from disciplines, are needed in order to help agriculture perform its many roles.

Where Members differ sharply is in their views regarding the appropriate policy response to non-trade concerns. Many countries, particularly the members of the Cairns Group, have argued that non-trade concerns do not justify the use of production- or trade-distorting support and protection. South Africa, for example, insists that the non-trade concerns of some countries should not become trade concerns for others (WTO, 2000d). Furthermore, many developing countries have sought to distinguish their concerns regarding food security and sustainable development from the non-trade concerns of developed countries. India in particular has argued that the food security and livelihood concerns of low-income countries with large agrarian populations “should not be confused or equated with the non-trade concerns advocated under ‘Multifunctionality of Agriculture’ by a few developed countries” (WTO, 2001b).

Developing countries and the Agreement on Agriculture. Clearly, the Agreement on Agriculture does not in itself help developing countries to strengthen their agriculture. It expects essentially the same type of commitments by developing countries, including the least developed, as by developed countries. Measures for SDT for developing countries were generally in the form of lower reduction commitments or longer implementation periods that did not recognize the fundamental differences between the agricultural sectors of developed and developing countries. Many SDT provisions were of little use to developing countries. A longer time period to reduce subsidies on domestic support or exports is meaningless for a country that does not provide such subsidies. But, by binding subsidies at their de minimis or base levels, the future ability of developing countries to invest in the agricultural sector may have been constrained.

Another issue of interest to many developing countries is the full implementation of the Marrakesh Decision on Measures Concerning the Possible Negative Effects of the Reform Programme on Least Developed and Net Food-Importing Developing Countries. This issue will be addressed in the current negotiations on agriculture. Trade Ministers have approved the recommendations of the Committee on Agriculture that the delivery of food aid to Least Developed Countries (LDCs) be fully in grant form and, to the extent possible, that levels of food aid be maintained during periods of rising world prices. The Ministers have also endorsed the recommendation that a multilateral facility be explored to assist LDCs and Net Food-Importing Developing Countries (NFIDCs) with short-term difficulties in financing normal levels of commercial imports, including the feasibility of creating a revolving fund (FAO, 2001j).

9.3.2 Regional and preferential agreements

Although the WTO rules apply generally to international trade, almost all WTO Members also participate in regional trade agreements. Moreover, many developing countries are granted preferential access into industrial markets for tropical products. The terms under which trade takes place within these agreements is important to the development of agriculture and the future of the food system. Thus a discussion of the long-term outlook for agriculture must include the role of regional trade agreements (RTAs) and the future of non-reciprocal preferential trade agreements (PTAs).

Regional trade agreements. The past decade has seen a proliferation of regional agreements involving agriculture, and this trend is likely to continue and intensify, particularly if the WTO multilateral negotiations stall. At the same time, PTAs that have so far characterized much of the trade between developing and developed countries are set to change significantly.

Nearly all WTO Members are parties to at least one RTA, some belonging to ten or more. Since 1948, more than 225 RTAs have been notified to GATT/WTO, of which about 150 are considered active, and more than two-thirds of these have been formed since the WTO entered into force in 1995 (WTO, 2001d). About 60 percent of these new agreements were formed by countries in Europe: among countries of Central and Eastern Europe, between those countries and the EU or between the EU and countries in other regions such as North Africa. In addition, many others, particularly in Africa, have been declared but not yet notified to WTO.

Countries have many reasons for joining RTAs. The economic benefits can be significant if lower-cost imports from the partner country displace higher-cost domestic goods (trade creation) or if access to a larger market allows producers to achieve greater economies of scale. Regional agreements may also stimulate foreign investment and technology transfer among members. RTAs may provide a forum in which trade liberalization can be pursued at a different pace, faster or slower, than in the multilateral system. But the benefits may extend beyond pure economics. For example, regional integration may be a useful strategy for improving regional security, managing immigration flows or locking in domestic reforms.

Against these potential gains, regionalism is not without costs. The classic trade diversion effect occurs when lower-cost imports from non-members are displaced by higher-cost products from a member, raising consumer costs and exacerbating inefficient production patterns. In addition to diverting trade away from efficient suppliers, complicated rules of origin and sourcing requirements may create difficulties for members themselves. Other costs are largely related to the administrative burdens associated with negotiating and operating RTAs. These can be particularly severe for small countries with scarce negotiating capacity and for countries involved in numerous agreements.

Is regionalism a threat to the multilateral system or a response to its failure? As noted above, about two-thirds of the RTAs that are currently in force have been notified since the creation of WTO, and a large share of them are association agreements with the EU. Since agriculture is typically given special treatment in these agreements (i.e. less integration and longer transition times than other sectors), it is difficult to argue that they are responding to a failure of the Agreement on Agriculture to liberalize fast enough. For many of these RTAs, it is likely that the agricultural sector was less important than other sectors and that non-economic considerations have greater importance than the potential economic benefits.

Other RTAs, such as the North American Free Trade Agreement (NAFTA) among Canada, Mexico and the United States, and the Mercado Común del Sur (Mercosur) among Argentina, Brazil, Paraguay and Uruguay, have liberalized agriculture at a considerably faster pace than in the Agreement on Agriculture, albeit more slowly than other sectors. Some of the agreements that are currently being negotiated, such as the Free Trade Area of the Americas and the Caribbean, and the African Union, envision a more aggressive pace of agricultural liberalization and integration than the multilateral system seems capable of delivering. If the WTO negotiations stall, it seems likely that these countries would put greater efforts into their regional integration plans.

Bilateral, regional and plurilateral institutions also play a role in global food regulation (Josling, Roberts and Orden, 2002). There is the potential for bilateral or regional standards to become de facto international norms without the full participation of all trading countries. When nations try to harmonize their respective technical regulations to permit the free movement of goods within a region, their external trading partners may face new technical requirements for gaining entry to the unified market. These external regulatory changes, or even proposed regulatory changes, can lead to market disruptions for the private sector, which in turn can produce trade conflicts for the public sector to resolve. New regional trade alliances, as well as the enlargement and deeper integration of older alliances, have been factors in the increase in technical barriers brought to the attention of policy-makers by exporters.

Regional agreements can, however, also point the way towards global standards. An early example of the regional “test bed” function was the provision in the United States-Canada Free Trade Agreement for the mutual recognition of testing and inspection facilities for livestock destined for sales across the mutual border. An early version of what eventually became the Sanitary and Phytosanitary (SPS) Agreement was inserted into NAFTA.15 The three partners agreed to incorporate the use of scientific evidence in domestic regulations and to set up a mechanism for examining cases where one country (typically the exporter of a particular product) considered that the importer was using sanitary and phytosanitary regulations to protect domestic production. The main impetus for such a provision was the view that market access negotiated under NAFTA could be severely jeopardized if arbitrary health and safety rules were to be allowed to go unchallenged.

A more recent example of a bilateral agreement to handle food regulatory issues is that between Australia and New Zealand. These two countries regulate food safety jointly through the Australia New Zealand Food Authority (ANZFA) and administer standards through the Australia New Zealand Food Standards Council (ANZFSC). Such a close working agreement is made easier, and more necessary, by the existence of a fairly complete free trade area between the two countries, the Australia-New Zealand Closer Economic Relations Agreement (CER). As two exporters of food products, relatively remote from other landmasses, both countries are anxious to keep out plant pests and animal diseases where possible and to maintain their reputation as suppliers of disease-free, quality foods (Josling, Roberts and Orden, 2002).

At the other end of the spectrum of bilateral approaches to food regulations is the attempt to use the transatlantic partnership between the United States and the EU. Several agreements of potential significance have been negotiated, although not without considerable time and effort. These tend to be agreements for the mutual recognition of testing and certification, rather than the recognition of the standards of the transatlantic partner. On the most contentious issue, however, that of the introduction of genetically modified (GM) foods, fundamental differences in approach still remain unsolved (Patterson and Josling, 2001).

Among the more ambitious regional initiatives in the area of plurilateral coordination of food regulations is the one discussed in the Asia-Pacific Economic Cooperation Council (APEC) process. APEC debated establishing an APEC Food System, which would have included both food safety and trade liberalization instruments. The food safety part of the programme has yielded a framework for mutual recognition agreements (MRAs) in the area of conformity assessments. Building on the SPS Agreement and the CODEX guidelines for such MRAs, the APEC Food MRA system allows countries to negotiate agreements that will facilitate trade in foodstuffs in the Asia-Pacific region. Whether this is the start of a more substantial cooperation in the food regulations area will in part determine the attraction of such plurilateral schemes.

Preferential agreements. Developing countries often depend on PTAs for access to the protected developed country markets in Europe, North America and Japan, particularly so for agricultural exports. But some of the existing and proposed preferential schemes may be incompatible with the terms of WTO or with regional integration agreements. Even when compatibility is not a concern, the value of these preferences is likely to shrink as general tariffs decline. Thus, many of the developing countries that currently depend on trade preferences may face difficult adjustments in the coming decades (Tangermann, 2001). Four non-reciprocal preferential arrangements are of particular relevance: the Generalized System of Preferences (GSP) under WTO, the ACP-EU Cotonou Agreement, the United States Trade and Development Act of 2000, and the Everything But Arms Initiative to provide duty-free and quota-free market access to the EU for the products of LDCs.

Generalized System of Preferences. The broadest of the existing non-reciprocal preferential arrangements is the GSP. Within the GSP, the preference-granting country has unilateral discretion over product coverage, preference margins and beneficiary countries. Many developing countries have complained that the product coverage of GSP schemes is not consistent from year to year, making it difficult for them to attract the necessary investment to develop the supply-side capacity to take advantage of the preferences. Others have noted that the GSP schemes granted by different countries vary considerably, making it difficult for exporters to know what tariffs their products will face in various markets.

Several proposals in the WTO negotiations on agriculture address these problems, calling for preferences to be made more transparent, stable and reliable. Some countries have argued that, in addition to the two main categories of countries that currently receive preferences under GSP (developing and least developed countries), other subgroups of countries should be eligible for non-reciprocal preferences. These subgroups comprise countries that are particularly disadvantaged in global markets by factors such as their size or geographical remoteness.

ACP-EU Cotonou Agreement. The EU historically granted trade preferences to the ACP countries under the Lomé Conventions that, in their later years, operated under a waiver from the GATT/WTO principle of non-discrimination that expired in 2000. The Lomé preferences have been extended for seven years under the ACP-EU Cotonou Agreement (which has also been given a WTO waiver), during which time the parties will negotiate new WTO-compatible trade arrangements, to come into force on 1 January 2008. These negotiations are envisaged to result in regional economic partnership agreements (REPAs), in effect, fully reciprocal free-trade areas with the EU. Agriculture will play a major role in these new agreements.

As the REPAs will have to be fully consistent with GATT Article XXIV and include “substantially all trade”, it will be more difficult for the EU to restrict access for agricultural products from REPA partners, although some quantitative constraints can presumably continue in cases where domestic production is also curtailed. One would expect pressure from developing countries that are not signatories to the Cotonou Agreement to make sure that those countries negotiating REPAs do not perpetuate preferential access for particular commodities that would be inconsistent with WTO rules. But, on the other hand, within the REPAs there may well be scope for agricultural policies that benefit ACP countries as they continue to supply the EU market.

United States Trade and Development Act of 2000. The US Trade and Development Act of 2000 extends certain trade benefits to selected groups of developing countries, including those covered by the Caribbean Basin Initiative and the Africa Growth and Opportunity Act. The US Trade and Development Act of 2000 is somewhat less comprehensive than the Cotonou Agreement, and some of its provisions could prove problematic in the context of regional economic agreements among developing countries. The main difficulty will likely relate to eligibility requirements and rules of origin.

For the African initiative, eligibility for the programme is to be determined annually by the United States Government on the basis of numerous criteria spelled out in the legislation. Rules of origin are quite detailed and strict, and penalties for transshipment are severe. Generally speaking, eligible products may receive preferential treatment only if they are produced almost entirely in one or more beneficiary countries or the United States. Origin rules for certain textile products, for example, allow no more than 7 percent of the fibre content by weight to originate outside the United States or one of the beneficiary countries, while for other products, no more than 25 percent of the final value may originate from outside. No provision is made for the regional agreements that may exist between eligible and non-eligible countries (OAU/AEC, 2000).

Everything But Arms Initiative (EBA) and LDCs. The conditions of access for LDCs appear to be one area where trade liberalization is making some headway. Agricultural products are included in the schemes that have been discussed in this regard. In 2001 the EU granted duty-free and quota-free market access, under the EBA, to all products originating from the 49 LDCs except armaments and, after 2006-2009, bananas, rice and sugar. Since the EU announcement, some other countries have declared their intention to extend similar preferential access for LDCs. Several proposals have been tabled in the WTO agriculture negotiations that would make this a permanent obligation of developed country members, a commitment that was taken up by all WTO Members in the Doha Ministerial Statement. These preferences could mark a significant improvement in the market access enjoyed by LDCs, although some have questioned the degree to which developing countries as a group would benefit, since the economic gains for LDCs under this initiative would come largely at the expense of other developing countries, some of them only slightly better off than the beneficiaries. Of course, this holds for all preferential schemes that cover some, but not all, developing countries (including those mentioned above).


1 As discussed in Chapter 10, many countries ran policies that were implicitly or explicitly biased against their agricultural sector. While some of this bias appears to have been removed in the course of domestic policy reforms and structural adjustment programmes, the patterns of government investment still favour urban areas in many countries.
2 Both for agriculture and total merchandise trade, gross trade data include intradeveloping country trade.
3 The increase in the trade deficit is expressed in constant international dollars. For the commodities covered in this study, it is projected to widen from about US$4 billion in 1997/99 to US$8.5 billion in 2015 and US$16.6 billion in 2030, respectively. The country composition of the group of LDCs has been assumed to remain unchanged.
4 All projected trade balances in Table 9.1 are expressed in constant (1997/99) US$, but the historical data are in current US$. Therefore, the rates of increase of the historical period cannot be compared with those for the projections.
5 The rapid overall increase in temperate-zone commodities over the last 40 years masks the fact that import growth varied considerably during this period. Imports rose particularly fast during the 1970s and early 1980s when the oil boom afforded many developing countries, particularly in North Africa and the Near East, with the foreign exchange earnings needed to increase imports of products such as cereals, meats and dairy products, for which domestic production capacity was limited. Together with the decline in oil prices, however, growth in food imports slowed down to considerably lower rates in the late 1980s and 1990s.
6 The subject areas for negotiations will include the following: agriculture, services, market access for non-agricultural products, aspects of TRIPS, antidumping, subsidies and countervailing measures, dispute settlement, trade and environment, trade and investment, trade and competition policy, government procurement and trade facilitation.
7 A TRQ is a specified quantity that can be imported at tariff rates (in-quota rates) well below the “normal” (over-quota) ones. The TRQ quantities and in-quota rates should be set at levels allowing for imports no smaller than the market access opportunities that existed in the Agreement on Agriculture base year 1986-88 or in any case to allow for a “minimum access”, whichever is higher.
8 Examples are sugar imports in the EU and sugar and beef imports in the United States.
9 Two product categories were responsible for more than half of all SSG actions: dairy and sugar and sugar confectionery.
10 Defined by the OECD as the annual monetary value of gross transfers from consumers and taxpayers to support agricultural producers, measured at farmgate level, arising from policy measures that support agriculture, regardless of their nature, objectives or impacts on farm production or income. The PSE includes the effects of border measures and thus captures both support and protection.
11 Direct payments under production-limiting programmes. They belong to the “blue box”, i.e. they are not subject to reduction commitments if they meet certain criteria.
12 For support measures included in the AMS and subject to reduction commitments.
13 Many developing countries reported that all their support to agriculture met the green box and/or SDT criteria for exemption, without specifying the measures or providing budgetary outlays. The conformity of these policies with the Agreement on Agriculture will remain uncertain unless assessed, perhaps in the context of dispute settlement.
14 The economic debate surrounding the multifunctionality of agriculture is generally framed in terms of public goods and externalities. Economists generally agree that agriculture can be a source of public goods. These are goods that are non-rival (one individual’s consumption of the good does not reduce the quantity available to others) and non-exclusive (once the good becomes available it is not possible to prevent someone from consuming it). The non-rival and non-exclusive nature of public goods means that they are not properly priced in a market system. Agriculture may also be a source of positive externalities. Externalities impose social costs or generate social benefits jointly with the production of a good, beyond the private costs or benefits deriving from the good. Positive externalities are consumable but are not priced in the market. For public goods and positive externalities associated with agriculture, it can be argued that producers should be rewarded in order to ensure a sufficient supply of the desired goods (Blandford, 2001).
15 The negotiations were progressing in parallel in the period 1991-93, with the NAFTA negotiations a little ahead. However, much of the work in the GATT negotiations on the SPS Agreement had been conducted in a subcommittee of the agriculture negotiating group and had made progress before the NAFTA process was initiated.

Previous PageTop Of PageNext Page