This paper argues that the Agreement on Agriculture does not fully recognize that the role of agriculture in enhancing food security in developing countries differs significantly from that in developed countries. Greater flexibility than is currently provided by the Agreement is therefore required for developing countries, particularly the food-insecure ones, in order for them to pursue agricultural modernization and meet their food security objectives.
1. The role of domestic agriculture in enhancing food security in developing countries
In 1996-98, 826 million people were estimated to be undernourished, of whom about 792 million were in developing countries. Although the number fell by 40 million between 1980-82 and 1995-97, this improvement was uneven, being attributable to a reduction of 100 million in 37 countries, whilst in the remaining countries the numbers increased by 60 million. Although food availability for direct human consumption grew by 19 percent between 1960 and 1994-1996, there is considerable variation among countries. For example, per capita agricultural production in LDCs has been on a downward trend over the past 40 years, whereas it has increased by 40 percent in the developing countries as a whole.[5] From 1980 to 1996, it declined in 29 of the 42 cereal-producing LDCs.
Analyses of the major factors underlying these trends (as exemplified by 14 case studies) have led FAO to state that “significant progress in promoting economic growth, reducing poverty and enhancing food security cannot be achieved in most cases without developing more fully the potential capacity of the agricultural sector and enhancing its contribution to overall economic development”.[6] This view is supported by a number of recent studies[7] which have demonstrated that in many poorer rural areas, increasing the productivity of agriculture often has the greatest potential for poverty-reducing growth, either through direct income benefits and indirect expenditure linkages from the production of tradable commodities or through consumer benefits in the production of non-tradables.[8] Therefore, indicators of food security are often directly influenced by the relative success of rural development initiatives in terms both of increasing the availability and stability of national food supplies and of enhancing individual access to food.
The liberalization of trade is justified in part by the incentives it is expected to provide to producers in developing countries, thereby improving prospects for rural development. However, for some developing countries there is unlikely to be an immediate benefit because of supply constraints and the inability of producers to take advantage of new trading opportunities as well as to compete in the domestic market with imports, some of which still benefit from subsidies and other non-competitive practices.
2. Policy measures for enhancing food security
Development experience over the last 50 years has amply demonstrated that vast rural poverty and food insecurity in developing countries have largely been the result of development strategies that overlooked the importance of the agricultural sector, particularly as regards the production of staple foods.[9] Experience has also shown that enhancing food security in developing countries requires a well-designed and adaptable package of policies that address in an integrated manner the supply, distribution and consumption aspects of the food chain.
The policy options available to contemporary poor countries are constrained by a number of factors including: (a) limited resources for public spending programmes; (b) the dilemma between remunerative prices for producers and prices that a large number of poor households can afford, thus making the option of border protection less attractive, despite high bound tariffs,[10] (c) seriously limited foreign exchange availability, leading to pressure to boost production of export crops. The channelling of resources to export sectors may not, however, always guarantee adequate foreign exchange for food imports.[11] Hence, from the food security point of view, import substitution is often a better option than export promotion.
What follows is a brief description of some specific policy measures that countries have used to enhance the food security of households or individuals, both directly (through programmes to boost food consumption and that provide targeted assistance) and indirectly via their farm programmes and trade policy. The most important of these policies are those that, given price support, affect input use and are concerned with boosting food consumption.
Price support
Price support is provided either by border measures alone or border measures in combination with domestic price arrangements (e.g. procurement prices, minimum guaranteed prices, target prices). Sometimes a border policy (e.g. a limitation on exports) that lowers farm prices in an effort to keep domestic food prices at a level that consumers can afford is partially offset by administered prices, allowing farm prices to rise somewhat. Usually, however, the border measure (e.g. tariffs) complements administered prices, and in consequence consumer prices may be higher than the corresponding world prices.
Depending on the combination of border and domestic price policies, various groups in society can find their food security affected differently. Thus, arrangements that lead to higher domestic market prices benefit farmers and hit consumers. The impact can be mitigated through targeted domestic food aid policies (e.g. rationing, fair price shops, income transfers to the poor) but only at a considerable cost to the budget, which is scarcely possible in poorer, food-insecure countries. Maintaining world prices for the consumer but paying subsidies to the farmer places a similar strain on the government budget, but which can to some extent be offset in the case of imported foodstuffs by the revenue raised by import duties.
The situation for export products is similar to that for imported goods. Exports have often been taxed in the past to help keep down domestic prices (e.g. Thailand’s rice policy in the 1970s and Indonesia’s palm oil variable export tax), but this is usually at the expense of the welfare of farmers who in many cases may be poorer than the urban consumer. By contrast, policies that boost exports through subsidies can raise farm incomes but at the expense of both the government budget and consumers. Export subsidies can be used to promote market access and are also a useful option for disposing of an unusual surplus in good harvest years.[12] Although their use is relatively widespread in developed countries, most developing countries do not resort to export subsidies.
Despite all the complications for food security of pursuing an active price policy, it must be noted that for imported food products a combination of tariffs to sustain farm prices and the use of the revenue derived therefrom to fund domestic food aid policies can be compatible with the WTO rules and efficient in terms of reducing rural poverty and food insecurity. Whether, however, the revenue so raised would be adequate to tackle the extent of domestic consumer food insecurity in the poorer countries is a matter of doubt; most likely extra resources would be required to strike the right balance. All other policies discussed below come with a price tag that may be prohibitive in many cases.
Input subsidies
Input subsidies are often used to offset low producer prices. Typically, they are used for items such as fertilisers, seeds and fuel. However, if not targeted, their use can bypass the poor producer and result in uneconomic transfers to better-resourced farmers. The effectiveness of targeted subsidies is demonstrated by developments in maize production in Malawi when output prices were liberalized in the early 1990s and fertiliser subsidies reintroduced. The result was a 70 percent increase in maize production by the small-scale sector and a substantial rise in fertiliser use.[13]
Facilitating the purchase of fixed capital stock (such as imported machinery and on-farm storage facilities) can both enhance productivity and reduce post-harvest losses. Examples include Zambia’s Agricultural Investment Programme that provides (with donor assistance) matching grants to groups of farmers for building on-farm infrastructure.
Whilst infrastructure investment is currently exempt from AMS inclusion, it often provides a form of subsidy to producers. In Sri Lanka, for example, the construction and maintenance of irrigation facilities are provided free of charge to farmers. It has been estimated that in 1995 irrigation subsidies comprised 3 percent of the total value of paddy production. In Egypt the adoption of advanced irrigation technology is seen as being essential to increasing water use efficiency.[14] Investment in infrastructure can also be for the purposes of conservation. For example, in South Africa, subsidized or low interest-loans have been provided under the Conservation of Agricultural Resources Act.
Intervention to correct for institutional failure
Where markets do not function well - for example, when there is poor market information, a monopolist-monopsonist situation or political interference - the expected supply response may not occur. In such circumstances, the promotion and/or strengthening of institutions governing access by poorer producers to input and output markets may be highly rewarding at a relatively low cost to the government. Strengthening credit institutions by influencing the incentives faced by market intermediaries and financing of their costs can result in the provision of more reliable seasonal finance.[15] Land tenure and rural finance institutions are often quite important, the former in allowing access to more productive land, investment in land, better decisions about land use, and greater consumption linkages, and the latter in promoting investment, particularly in seasonal inputs.[16] Overall, the provision of improved institutions, regulatory frameworks and training are possible but they may initially require subsidies.
Food consumption policies
One of the most widely used food security policies over the years has been assistance to public food stocks. To be compatible with the AoA, volumes and accumulation must correspond to predetermined targets related solely to food security. Food purchases by governments must be made at current market prices and sales from stocks must be at not less than the current domestic market price of the product and quality in question. For developing countries stocks may be acquired and released at administered prices, provided that the difference between the administered and external reference price is accounted for in the AMS calculation. In reality this does not constitute any special treatment, since any country can pursue such a policy provided that the associated expenditure is counted in its AMS.
Such schemes attract additional costs to the government in terms of accumulation, storage, price monitoring and distribution. The promotion of private storage and marketing by, for example, reducing barriers to entry, improving infrastructure and market information may provide a mechanism for passing on such costs.
Though widely used in the past, these policies have come under critical scrutiny more recently because of their relatively high costs. But in situations where physical availability of food is at risk, such stocks, in conjunction with trade policies, can have a valuable role to play.
The provision of food, or of means to allow eligible recipients to buy food, at market or subsidized prices is not subject to WTO disciplines, largely because the measures involved are not prejudicial to exporters, since they are trade-enhancing. Although leakage can be reduced via targeting, this is generally associated with higher administrative costs. The efficiency of consumer subsidies often depends upon the type of food chosen. For example, in Brazil $1 spent on subsidizing bread transferred about $0.18 to low-income consumers, whereas the same amount spent on legumes transferred about $0.39.[17] Subsidizing inferior grades can enhance the effectiveness of a food subsidy.
There are a number of other policies used to provide food safety nets in developing countries. Examples include targeted employment programmes, or food-for-work, which can also create assets such as rural access roads, storage facilities and irrigation infrastructure. However, the cost is often high and there is a need to take into account the scarcity of financial resources of developing countries in setting the criteria for such programmes.
3. Meeting food security objectives: distortion versus effectiveness
The preceding discussion indicates that a variety of policy instruments can be used to increase production in support of the food security objectives of developing countries. The focus then falls on identifying mechanisms for raising support for domestic production in an efficient manner under existing domestic constraints, while at the same time preserving the option of using more costly measures if greater resources become available in the future.
The major criteria against which to assess alternative policy measures are efficiency, effectiveness, cost and equity. Efficiency relates to changes in the volume of trade resulting from any change in domestic production and any impact on the level and variability of the domestic or world market price. Effectiveness relates to the impact on food security at the national and household levels in the short run, and the extent to which it promotes rural development more generally and hence food security in the long run. The costs of different policies may be evaluated in terms of the burden on government and on consumers and producers. The impact on equity considers the extent to which the various measures target poor producers and consumers.
Most policies will have some impact on trade volumes either directly or by enhancing incentives to domestic producers. However, in determining the significance of the relative effect of policies on trade, it is necessary to consider the “size” of a country, or group of countries, in world markets. It is assumed that the food-insecure countries, including LDCs, will in general be too small to have a distorting impact on most agricultural commodity markets. Where that is not the case recourse could validly be made to the type of market share cap envisaged in the Agreement on Subsidies and Countervailing Measures, Article 27.
In addition to the type of capping of market shares suggested above, there is the question whether developing countries, other than the food-insecure countries and the LDCs, with competitive large-scale agricultural sectors that produce significant shares of their exports or import substitutes, should be excluded from exemptions made on the basis of food security. Several typologies of developing countries in terms of the impact of trade liberalization have been developed, but none investigates the structure of agriculture within these countries.[18] As demonstrated above, the contribution of agriculture to rural development and hence long-term food security is significant and therefore it can be argued that exemptions should be made not on the basis of a country’s net trade position or agrarian structure, but of its current level of development as characterised by the extent of its food insecurity.
In terms of their impact on food security, price policies can have mixed effects, as discussed above, depending on whether those suffering from food insecurity are mainly net producers or net consumers of food, given the often high budgetary outlays required to reconcile different interests. Policy measures that promote productivity improvements with no detrimental effect on consumers are therefore likely to be preferred. Subsidies on inputs, if targeted to specific crops, can be less costly but may distort resource use. Such distortions may, however, be weighed against the benefits from increasing the supply of a crop which contributes substantially to food security. Interventions to correct for market failures appear to provide low-cost opportunities for enhancing household and national food security in both the short and the longer term, particularly when they remove distortions that discourage production. Border measures and market price support are likely to be relatively inequitable in that trade policy favours marketed commodities and therefore the more prosperous farmers. Subsidies on non-tradables and on inputs have a more equitable impact on food security at the micro level.
In sum, measures used to strengthen institutions governing access to inputs, which may be associated with the provision of input subsidies, are likely to provide the most cost-effective, and least distortionary, way of enhancing food security. Policies implemented to support the production of domestic food staples are likely to be preferable to those focusing on exportables. Where required, measures which provide enhanced incentives to producers via the output price are likely to be more effectively implemented via border measures than by subsidies on output prices, which are more complex to administer. However, the greater impact of producer incentives on consumer prices may limit their use unless offset by targeted consumption subsidies. A judicious combination of these measures may be the most appropriate approach to achieve the best result in terms of efficiency, effectiveness, cost and equity.
4. Increasing policy flexibility of the developing countries
The preceding analysis makes a prima facie case for the wider use of policies to promote agriculture in developing countries so as to increase their food security. How far would such policies be compatible with a renegotiated AoA?
A number of recent proposals from developed countries call for a tightening of exemption criteria. The proposal of the United States is for simple differentiation of support between measures that are market-distorting and those that are not and for the establishment of criteria for the latter to ensure that they are targeted, transparent and truly not distortive of trade. Norway has suggested that policies involving expenditure to be included in the AMS calculation be divided into two categories: (i) support to production for the domestic market, which would be subject to less stringent reduction commitments and (ii) support to export-oriented production (incorporated in the calculation of AMS), which should be subject to further reductions. By contrast, most developing countries, notably India, have called for exemptions that are more responsive to specific developing country needs.[19]
Although a classification of measures along the lines proposed by the United States or Norway may provide more clarity and less scope for violation within the existing AoA rules, a case can be made for a more flexible mechanism for developing countries that recognises fundamental differences among them in their food security objective. Considering that not all developing countries are faced with the problem of food security, while all of them can be presumed to need to foster their rural development, two different types of policy response at the WTO negotiations may be considered:
(a) Food-insecure developing countries could be granted the same exemptions as are already granted to LDCs, thus allowing them to increase their AMS for commodities critical to food security, unless that results in a share of the world market for a particular commodity above a certain level, in line with Article 27 of the Agreement on Subsidies and Countervailing Measures.(b) All other developing countries could be allowed to raise the existing de minimis level (perhaps limited to commodities critical to food security) to an agreed new level and extend input and investment subsidies to all farmers producing commodities critical to food security. Also, tariff bindings for food crops could be renegotiated without their having to make corresponding concessions elsewhere; they should also be granted access to the SSG for food commodities.
