This chapter provides an analytical framework within which to assess the linkages between trade reform (whether unilateral, plurilateral or multilateral) and food security. Problems arise at both ends of the link: with food security and with trade reform.
Food security problems:
food security applies to individuals, but trade policy is implemented at state level; i.e. the analysis must be multi-tier;
the impact on any individuals food security of a change in trade policy will be mediated through a host of domestic, social, economic and institutional structures that will produce different outcomes depending upon the individuals status; i.e. the analysis must be multi-faceted.
it is the relative as much as the absolute treatment of agriculture that is important, and so the analysis must cover not only trade policies focused on agriculture but also those on the non - agricultural sectors as well; i.e. the analysis must be multi-sectoral;
policy changes in other countries have an effect on the impact of changes in the focus country; i.e. the analysis must be multi-country;
trade flows are not only affected by trade policy narrowly defined, but exchange rate policy (not to mention international shocks) will also affect the relative prices of agricultural products; i.e. the analysis must be multi-instrument.
In addition, the process of policy making may often be so opaque as to hide the details of proposed trade policy change until it is too late to undertake ex ante analysis, as was the case in the Uruguay Round (UR). The data required for analysis, either ex ante or ex post, is often lacking also.
The aim of this chapter is to describe and explain the type of data that is required for impact assessment and the decisions that need to be made. Some of this data will be available in advance of the final negotiating decisions on trade policy change. In other cases, steps can be taken to undertake the collection of the data once negotiations have advanced sufficiently for the details of change to be known. In either case, the identification in advance of the type of information that is needed and the prior collection of whatever relevant data is available will not only facilitate speedy impact analysis once the details of trade policy decisions are known but, equally important, may inform the negotiating process.
The international dimension of entitlements
The approach developed by Sen is helpful, since it emphasises the different ways in which individuals can acquire food. Each can be affected directly and indirectly by trade policy change.
The entitlement relationships Sen identifies enable individuals to acquire food in one of four ways.
They may produce it for themselves (production-based entitlement),. which may be particularly important, for example, for small farmers who aim for self-sufficiency. It can be affected by policies altering the demand and supply of factors used in production, some of which will relate to international trade.
They may sell or barter physical assets (trade-based entitlement). Many farmers will augment their own production by exchanging either a surplus of some crops or a on-food product. The amount of food they can acquire will be influenced by policies that affect the level and variability of prices for food relative to what they are able to exchange.
They may sell their labour power (labour-based entitlement). Rural landless labourers and urban employees all need to buy or barter food in the market. Their food security is determined by the level and location of employment opportunities which, in turn, may be altered by trade policy.
They may receive informal gifts from individuals and formal transfers from government (transfer-based entitlement). These are important for those lacking other adequate means. Formal transfers such as food aid may be influenced by multilateral trade agreements.
Trade policy can affect these entitlements and hence food security in two ways:
directly, by introducing change to prices and to the policies (of both domestic and foreign governments) that impact on entitlements (e.g. by altering food prices, the availability of food aid, or the access of exports to foreign markets);
indirectly, by making more or less feasible some of the policies that are considered desirable to promote or protect entitlements (e.g. by altering the legality or practicality of input subsidies).
While changes to international trade policy will affect food security both directly and indirectly, the actual impact on individuals will be heavily influenced by domestic mediation. This is provided by the institutions and structures that stand, metaphorically, between the individual and the external market and translate border prices into retail prices, which influence wages and endowments facing the household, and also determine the individuals relationship to the household.
The impact of liberalizing a countrys international trade policy on food security is likely to be different from (and probably smaller than) that of liberalizing its domestic trade policy. However, the effects of the former are difficult to disentangle from those of the latter. Liberalization normally occurs at the same time as other changes, which may push in the same or a different direction (for example, in the case of Zambia, the collapse of the copper industry which had funded a substantial subsidization of agriculture). The expected results from an international trade policy change may not occur if there are either institutional or physical factors that influence domestic trade. Domestic monopolies or oligopolies (whether or not they are state-owned or private) will mute the transmission of any price signals. If rural infrastructure is degraded the transactions costs of domestic trade may exceed the price signal.
Moreover, governments rarely act in an entirely consistent manner, and so it may often be difficult to decide how much liberalization has actually occurred. For example, applied as opposed to bound tariff rates are often not known in detail by food security analysts, and still less is the impact of tariff-like taxes such as special duties that apply in practice, though not in principle, only to imports. If applied tariffs are set well below the bound rate but then are frequently raised and lowered, the effect of such uncertainty on traders actions may outweigh in whole or part the effects of the initial liberalization.
Relative sectoral effects
The direct effects of a shift from a less liberal to a more liberal international trade regime will be to alter relative prices, principally between:
tradables and non-tradables;
sectors (those in which policy has changed most vis-à-vis the others);
products within sectors (for example, between goods that are for export and those that compete with imports on the domestic market).
The net effect of trade policy changes that impact differentially on the various sectors of the economy may be very different from what might be deduced by looking just at the changes directed at agriculture. Trade protection to manufacturing that is greater than to agriculture represents a bias against the latter even if it receives high absolute protection. By the same token, trade policy changes could result in an increase in the relative bias against agriculture even if the average level of trade protection is reduced.
Whatever the net effect, these shifts will produce indirect geographical and social changes. Some social/gender groups will be affected more than others (positively or negatively) if they are more associated with one group of products than with others. Even a shift between the production of maize for domestic consumption and horticulture for export will affect employment, and the effects of a shift between either (or both) of these and clothing production may be even more far reaching. There will be similar differential effects on geographical regions (both within and between countries).
An individual will be affected both by what their national government does and by the consequences of changes introduced by other countries. In the case of sub-Saharan Africa (SSA), for example, it is the WTO policy changes made by foreign governments that may have the greatest effect on the price of agricultural goods people buy and sell. One implication is that governments cannot avoid these effects simply by not participating in the WTO talks.
For example, the food security of some vulnerable groups in developing countries will almost certainly have been affected in the past by Northern agricultural subsidies. The OECDs total agricultural producer support in 2001 was estimated to be equivalent to 31 percent of total farm receipts. This compares with a figure of 38 percent in 1986-88 (the base period for the Uruguay Round subsidy cuts). This decrease is smaller than the 20 percent cut in aggregate producer subsidies to which the industrialized WTO members committed themselves in the Uruguay Round.
Such subsidies have depressed the world price of various temperate agricultural goods, of which cereals are probably the most important. In 2001 the OECD PSE stood at 36 percent for wheat and as much as 81 percent for rice. In a static sense, this has made it easier to supply food-deficit countries with imported cereals. Whether this static gain has been offset by a dynamic loss (because it has dulled incentives for domestic agricultural growth) is a moot point. But, to the extent that they exist, both static and dynamic effects will change if Northern agriculture is liberalized.
The initial effect is expected to be an increase in world prices, especially for cereals. In due course, full liberalization will allow countries like Australia, New Zealand, Canada, Argentina and Brazil to increase production, which will moderate any rise in world prices. But full liberalization will not occur at one fell swoop. Crab-like, partial liberalization (see next section) could easily result in a combination of changes that raise world prices.
The effect of multilateral liberalization on developing country exports will vary considerably according to the product and the market. The broad effect is that a reduction of import tariffs by developed or developing countries should increase their demand for imports, including those from developing countries. But how this affects particular developing countries depends both upon the current trade regime for the products they export and their capacity to alter supply in response to increased demand.
A particularly important influence during the period before full liberalization is the position of any given country in the trade preference hierarchy of its actual and potential export markets. For example, many developing country exports will not benefit initially from EU liberalization and, indeed, may suffer. This is partly because traditional commodity exports (such as beverages) already face very low tariffs worldwide and so will be unaffected one way or the other. In addition, those agricultural exports that do face heavy protectionism (such as sugar, tobacco and horticulture) have preferential access to the European market, which absorbs the great bulk of exports.
To the extent that the EU liberalizes its agricultural regime, it will adversely affect exporters from countries receiving heavy preferences either by increasing competition from less favoured suppliers or by reducing prices in the protected export market. SSA will be most adversely affected since it is highly preferred in the EU, which is its main market. By contrast, there will not be such adverse effects in South Asia or those parts of Latin America (such as Mercosur), which receive no such preferences in the EU. But South Asia will be affected by the phase out of the Multifibre Agreement which will have differential effects for India and Sri Lanka (wholly positive), Pakistan (ambiguous) and Bangladesh (possibly negative). These, in turn, will affect the labour entitlements of clothing workers and, hence, their food security. The phase out of the MFA will also adversely affect SSA - magnifying the negative effects on food security. Only if SSA states could increase their output sufficiently to take advantage of liberalization in non-EU industrialized country markets where multilateral liberalization may genuinely improve access, could they, too, could benefit.
In addition to the effects of liberalization on the level of prices, there may also be effects on price volatility. The practice of some OECD (and other) states of insulating their farmers and consumers from the price effects of world market shocks has been to transfer onto the rest of the world the burden of adjusting to price volatility. Empirical investigation has suggested that in 30 countries it takes five years or longer to transmit half of any world price shock onto their domestic market and a further 30 countries appear to be virtually isolated from international price signals. To the extent that this continues the food security of individuals will be adversely affected by price volatility in those countries lacking the means (policy, implementing, or financial) to provide any buffer.
Trade liberalization will alter the domestic price of tradables - but so will other policy changes (that may be reinforcing or offsetting) as well as external shocks. The most important of these is the exchange rate. A recent study of four central European states showed an imperfect (and sometimes negative) correlation between border and domestic producer price changes, largely as a result of offsetting exchange rate changes. Much of the past bias against agriculture has been produced by overvalued exchange rates that make imports artificially cheap.
Whilst there is broad agreement that full, global liberalization of agriculture could be expected to have favourable aggregate effects for developing countries, there is also a consensus that the Doha Round will not fully liberalize trade. It is also not the only liberalization act in town. The combination of partial multilateral liberalization and parallel change in other fora may produce complex results that are not easily predictable.
The WTO Agreement on Agriculture (AoA)
Although the negotiations on the AoA formally began in 2000, there has been remarkably little progress on defining the precise rule changes and liberalization that are front runners. In the absence of such guidance the rules most likely to have effects on food security are those on:
tariffs, which could affect the price of imported food (with differential effects on consumers and producers) and also government revenue (which would impact on many policies);
domestic subsidies, which could alter the feasibility of policies to enhance labour entitlements (e.g. input credits or market development);
export subsidies, which could affect the feasibility of transfer and safety net policies that use imported food, since cuts will tend to increase commercial import prices and may reduce the availability of food aid.
A particular cause for concern is that the Round might result in reductions in developing country tariffs that are more rapid than the removal of production and export subsidies in developed states. As long as developed country subsidies remain, prices in the world market will not reflect production costs.
It may be economically undesirable for developing countries to liberalize further their domestic markets on items from other countries that are sold at below the cost of production. Whilst this might favour transfer and also exchange entitlements (for producers of non-competitive goods) it would reduce the production entitlements of farmers.
In the absence of world prices that give the right signals, there will be a role for government action to promote production that would be self-sustaining in a situation of efficient markets. Yet in cases where a policy depends upon government expenditure, any reduction of tariffs (as a consequence inter alia of a WTO accord) is likely to make this more difficult. This will persist during an adjustment period (which could be quite lengthy) until alternative sources of government revenue are introduced. In addition, policies that require government to spend directly in the agricultural sector (for example to support marketing or assist labour-absorbing SMEs) could be affected by future changes either on the allowable areas of domestic subsidy or on the total permitted volume of subsidy.
Those policies that provide vulnerable groups with subsidized food may be affected as they will often be supported by concessional imports. The supply of these in turn will be influenced by the negotiations on export subsidies (particularly if these go beyond dealing with direct subsidies to cover cross-subsidy from the protected domestic market).
The reduction, or even elimination, of export subsidies is a high priority for the Cairns Group and some others in the new Round. Developing country policies that involve the provision of cheap food to enhance the trade-based or transfer entitlements of the food insecure, and which rely upon imports to deliver some or all of these supplies, are likely to be less feasible if the substantial subsidies still given by two or three major exporters are removed.
Agricultural liberalization may also occur in fora other than the WTO. Two EU initiatives illustrate the ways in which these may affect entitlements.
The combination of the EUs domestic reform of its Common Agricultural Policy (CAP) and its new trade policies towards LDCs (under the Everything But Arms initiative) and its free trade area partners may reduce the trade entitlements of preferred exporters without providing any opportunity to offset such losses. The extreme case is given by sugar which will be fully liberalized for LDCs by 2009. This will open the way for greatly increased imports from LDCs but at prices close to world market levels, displacing the high priced imports from current suppliers; non-LDC suppliers will be constrained in their ability to respond by the tariff quotas that apply under the EU-ACP Sugar Protocol and the prohibitively high Most Favoured Nation (MFN) tariffs that apply outside it.
The other EU initiative is to seek the replacement of the non-reciprocal Cotonou trade regime with reciprocal Economic Partnership Agreements (EPAs). These will require ACP states to remove tariffs and charges concerning about 86 percent of their imports from the EU. The negotiating mandate handed down by the member states to the European Commission makes reference to the possibility of a safeguard clause (which would be needed by ACP states to offset EU agricultural subsidies) but only in a weak form. The result of unrestricted imports of subsidized EU agricultural products would be a loss of trade entitlements by farmers.
Structural adjustment programmes (SAPs) often include the unilateral liberalization of international trade policy (which will have similar effects to multilaterally agreed liberalization under the WTO), but the package also usually includes a range of other elements, including reforms that affect the prices of things people buy and sell. Typical examples of these are domestically oriented trade-related policies (such as price controls, parastatal policy, etc.), non-trade international policies (such as the exchange rate and foreign exchange convertibility) and the creation/destruction of institutional infrastructure. In addition, altered government policies may affect the availability of physical infrastructure and supply of transport.
The arguments presented in this chapter are summarized in Figure 5.1. It is based upon the flow charts in McCulloch et al. 2001, designed to map the link between liberalization and poverty.
The figure applies only to changes to any particular states national trade policy. The overall impact on food security will depend upon both national and foreign trade policy change. The greatest challenge for developing countries may be that both sets of change are likely to occur at the same time. This would produce substantial overall change to the production, trade, labour and transfer entitlements of many potentially food insecure people.
Figure 5.1 Flowchart for policy-makers on national trade policy and food security
What trade liberalization is proposed?
Steps needed: 1. Find out for each good (agricultural and non - agricultural): - what is the world price? - what is the domestic price? - what is the current tariff (or tariff equivalent)? - what is the proposed tariff (or other restriction)? 2. From this, identify the goods that may experience the biggest potential change in price
Are these goods that are consumed or produced by vulnerable groups?
Steps needed: 1. List the extent to which the goods supply entitlements to potentially food insecure groups. 2. Identify those goods that supply entitlements that are likely to experience a large price change
Is the change in the border price likely to be passed on to poor consumers, or to the markets in which poor producers and labourers operate?
Steps needed: 1. Identify the reasons why the price change may not be passed on. 2. Formulate structural and institutional reforms to improve price transmission where this will enhance entitlements. 3. Design complementary policies to offset price transmission where this will protect entitlements
Will new trade rules affect the feasibility of complementary government action (by reducing revenue or restricting subsidies)?
Steps needed (if effects are adverse): 1. Put in place measures to diversify revenue sources if trade taxes are important. 2. Consider if there are alternative ways of achieving the ends sought that might avoid WTO restrictions.
Direct effects on food security likely to be limited
Source: McCulloch, N, L A Winters, & Cirera, X. 2001. Trade Liberalization and Poverty: A Handbook. London: CEPR and DFID.
 This chapter is based on
a paper by Christopher Stevens, Linking trade and food security presented
at the FAO Expert Consultation on Trade and Food Security: Conceptualizing the
Linkages. 11 - 12 July 2002, Rome.|
 Sen. 1981. op cit.
 Stevens, C., Greenhill, R., Kennan, J. & Devereux, S. 2000. The WTO Agreement on Agriculture and food security, Economic Paper 42. London: Commonwealth Secretariat.
 This impact is neatly illustrated in McCulloch, N., Winters, L. A. & Cirera, X. 2001. Trade Liberalization and Poverty: A Handbook. London: CEPR and DFID, Figure 4.2.
 Anderson, K. 2002. Economywide Dimensions of Trade Policy and Reform. In Hoekman et al. eds. Development, Trade, and the WTO: A Handbook. Washington DC: The World Bank.
 OECD 2002. Agricultural Policies in OECD Countries - Monitoring and Evaluation 2002. Paris. Annex Table 2.
 OECD 2002: Annex Table 3.
 Valdés, A & Foster, W. 2002. Reflections on the Policy Implications of Agricultural Price Distortions and Price Transmission for Producers in Developing and Transition Economies. OECD Global Forum on Agriculture: Agricultural Trade Reform, Adjustment and Poverty, 23-24 May 2002.
 Valdés, A. 2000. Measures of agricultural support in transition economies: 1994-1997. In A. Valdés, ed. Agricultural Support Policies in Transition Economies, World Bank Technical paper 470, Europe and Central Asia Environmentally and Socially Sustainable Development Series.
 The Cairns Group is a coalition of 17 agricultural exporting countries who together account for one-third of the worlds agricultural exports (Argentina, Australia, Bolivia, Brazil, Canada, Chile, Colombia, Costa Rica, Guatemala, Indonesia, Malaysia, New Zealand, Paraguay, the Philippines, South Africa, Thailand and Uruguay). The Cairns Groups objectives include deep cuts to all tariffs (including tariff peaks) and removal of tariff escalation, the elimination of all trade-distorting domestic subsidies; the elimination of export subsidies and clear rules to prevent circumvention of export subsidy commitments.
 An ACP-EU Partnership Agreement signed in Cotonou, Benin, in June 2000, which includes a financial protocol covering aid programmes for member states through the European Development Fund (EDF). The agreement follows on from four successive Lomé Conventions.
 McCulloch, N., Winters, L. A. & Cirera, X. 2001. Trade Liberalization and Poverty: A Handbook. London: CEPR and DFID.