Chapter 3. Trade liberalization and food security: conceptual links^[42]

3.1 Introduction

Food security is traditionally discussed in terms of either food self-sufficiency or food self-reliance. The former requires production of food in the quantities consumed domestically, while the latter requires domestic availability. Self-sufficiency rules out imports as a major source of supply while self-reliance has no such restriction. Some commentators do not regard self-sufficiency as an economically sound alternative, given the much greater worldwide capacity to produce food than to consume it, the few restrictions on the exports of food items in countries with excess capacity, and the availability of international transport. Instead, what countries need, it is argued, is sufficient capacity to generate the foreign exchange necessary to import whatever quantities they consume over and above what it is efficient to produce, based on comparative advantage.

Accepting food self-reliance as the means to achieve food security, it is possible to ask how the liberalization of trade in agriculture including food will impact on developing countries. To answer this question, it is necessary to distinguish between importers and exporters of the products and between liberalization in the developed and developing countries. If the object is to study the impact on the poor, much finer analysis is required since the effects must be decomposed at the national level into effects on the poor and non-poor.

3.2 Trade liberalization and food security

Using FAO and World Bank data, Valdés and McCalla^[43] classify 148 developing countries according to a variety of criteria. For the purpose at hand, the following two classifications are the most useful:

• Classification according to income: the World Bank divides these countries into Low Income Countries (LIC), Lower Middle Income Countries (LMIC) and Upper Middle Income Countries (UMIC). Based on 1996 income levels, countries with per capita incomes of US$785 or less are in the first group, those with incomes between US$785 and US$3 125 in the second, and those between US$3 125 and US$9 655 in the third.

• Classification according to net trade status in food and agriculture: countries are divided into Net Food Importing (NFIM) and Net Food Exporting (NFEX) on the one hand, and Net Agricultural Importing (NAIM) and Net Agricultural Exporting (NAEX) on the other.

Table 3.1 shows that of the 148 developing countries, 63 are LICs, 52 LMICs and 33 UMICs. As the bulk of the world’s poor are in the LICs it is important to pay special attention to the countries in that group. As many as 48 out of 63 LICs are net importers of food. Even among the LMICs, 35 out of 52 are net food importers. It is clear that any realistic analysis of trade liberalization must address the question as to how food importing countries and the poor living there will be impacted by agricultural liberalization.

Table 3.1 Countries classified according to income status and food trade position, 1995-1997(number of countries)

Source: Valdés and McCalla, 1999.

Table 3.2 classifies the three groups of countries according to their net position in agriculture as a whole. More LICs appear as agricultural exporters (33) than as food exporters (15). Taking agriculture as a whole, therefore, export interests seem to dominate. The overall picture differs less for LMICs and UMICs when compared according to their trade position in food versus agriculture as a whole.

Table 3.2 Countries classified according to income status and agricultural trade, 1995-1997 (number of countries)

Source: Valdés and McCalla, 1999.

Valdés and McCalla also report that of the UN classification of 46 Least Developed Countries (LDCs), as many as 45 are net food importers. Again, considering agriculture as a whole, the number of net exporters rises to 15.

Table 3.3 shows the extent of overlap between importers of food and of agriculture, and exporters of the two sets of items. Not surprisingly, the largest numbers concentrate along the diagonal: 83 countries are net importers of food and of agriculture, while 41 countries are net exporters of both. This still leaves a large number of countries (22) that are net food importers and net agricultural exporters. Fourteen of these 22 countries are LDCs.

Table 3.3 Countries classified according to food trade and agricultural trade, 1995-1997 (number of countries)

Source: Valdés and McCalla, 1999.

It is clear from trade patterns described in the previous section that the effect of liberalization by the developed countries is bound to be quite uneven on developing countries. Of the 46 least developed countries, 31 are net importers of both food and agriculture. These countries are likely to be hurt by the developed country liberalization, which must raise agricultural prices. On the other hand, the bulk of the benefits will accrue to the relatively well-to-do developing countries in Latin America and Asia and the United States.

Thus, there is a case for transferring some of the agricultural subsidies currently given to farmers in the OECD countries to net importers of food and agriculture in the developing world. It must also be acknowledged that unqualified assertions by many, including the heads of some multilateral institutions, that subsidies and other interventions in agriculture in the OECD countries are hurting the poor countries are not grounded in facts. While there remains a strong case for the removal of agricultural protection and export subsidies on efficiency grounds, the claim that the change will bring net gains to the least developed countries as a whole is at best questionable and at worst outright wrong.

3.3 The implications of trade liberalization in developed countries

Trade policy reform involves a combination of:

• domestic support measures;

• export subsidies;

• tariffs.

In each case, there are complications that must be taken into account. This is illustrated below starting with price supports.

The removal of domestic price support on, say, wheat, will lower output of wheat and raise its price in the world markets. Wheat-exporting developing countries will benefit and wheat-importing countries that continue to be importers after the removal of the support will lose and those that switch from being importers to exports may benefit or lose.

In some cases, however, the support may be given to induce farmers not to cultivate some proportion of their land. In this case, the withdrawal of support could expand output, lower the price and have exactly the opposite effect: importers will benefit, exporters that remain exporters will lose and exporters who switch to being importers may benefit or lose. The critical question one must ask, therefore, is whether the removal of the support will increase or reduce the output of the supported product.

In the same vein, a reduction in tariffs by the developed importing countries will increase the world price of the product, benefiting exporters, hurting importers and leading to an ambiguous effect on those turning from importers to exporters. But this standard analysis is complicated by the presence of trade preferences. The reduction in the tariff cuts into the preference margin of the beneficiary countries and lowers the profitability of their exports. Liberalization can potentially hurt these exporters.

Finally, under normal circumstances, the reduction in export subsidies raises the world price of the product, benefiting developing country exporters, hurting importers and yielding ambiguous effect on those turning from being importers to exporters. Again, if the export subsidies were being countervailed, the net impact of the two measures is likely to be a transfer of the export subsidy from the exporting country government to the importing country government in the form of duty, without a significant effect on prices and output. The removal of the export subsidy will also result in the removal of the countervailing duty and the world supply will be unchanged. This is the case with export subsidies in general, although with the removal of targeted export subsidies the affect may be less predictable.

In all these cases, it is possible to consider one intervention at a time. But in practice these interventions have been used simultaneously in agriculture. An especially important case arises where a country is a potential importer of a product but domestic support measures, tariffs and export subsidies are combined in such a way as to turn it into its exporter. This case, characterizing some of the European Union (EU) policies, is illustrated in Figure 3.1.

Figure 3.1 The Common Agricultural Policy of the EU

In this figure, the demand and supply curves of a potential import of the EU are shown. Suppose the world price in the absence of any intervention by the EU is P1, with the EU importing the difference between the demand and supply at that price. Suppose next that the EU sets the internal price at P, which generates an excess supply of the product. To eliminate this excess, it gives an export subsidy. Because the EU now exports rather than imports the product, the world price falls below P1 to, say, P2. The export subsidy that supports the exports is P2P. At the same time, to prevent imports from taking advantage of the higher internal price, an import tariff equal to P2P or higher is also imposed.

It is evident that the removal of the three restrictions together in this case will have the effect of raising the price from P2 to P1. If knowledge of the complete picture was not available, and each measure was looked at in isolation, the evaluation would be quite different. In the standard analysis, each measure by itself lowers the world price but they are not cumulative in this example. It is the export subsidy that holds the world price down to P*. The tariff by itself only plays the role of eliminating the arbitrage between the world price and the internal EU price. It does not contribute to a reduction in the world price.

3.4 Trade liberalization by developing countries

On the assumption that developing countries are individually small, orthodox theory leads to the expectation of efficiency gains from their own liberalization. The main departure from theory may be that the adjustment process would be slower and therefore adjustment costs higher in countries primarily dependent on agriculture. The ability of other sectors to absorb workers released by such liberalization may be limited in the short-run, a limitation that would be compounded if there are unemployed resources elsewhere in the economy. This is a strong argument for adjustment assistance, which should, nevertheless, be temporary. This is because eventual adjustment is essential if liberalization is to produce efficiency gains.

Some countries also impose restrictions on exports. From a national welfare perspective this may not always be advantageous, but in practice export taxes may be a major source of government revenue, and one of the few that can be easily collected. It should also be remembered that export taxes can be used as an instrument for keeping domestic food prices relatively low, and in these circumstances their removal could have a negative impact on food security. Finally, following second best arguments, it may in the interests of a country exporting raw materials, to tax its exports when its processed exports in turn face tariff barriers. Exports do not affect food security adversely even if the objective is self-sufficiency rather than self-reliance. In all likelihood, they will enhance food security by increasing the country’s ability to import food items it does not produce in adequate quantity.

At the same time, with respect to exports in which the country has market power such as cocoa and coffee, an economic case can be made for a modest export tax to exploit market power^[44]. But countries need to be careful here since these taxes have frequently created internal bureaucracies in the form of marketing boards that make exporting costly. Bureaucracy can also result in a substantial part of tax revenues being used up in so-called rent-seeking activities.

3.5 Conclusion: the impact on the poor

Translating the impact at the national level into impact on the poor remains a tough task and room for error here is great. The best approach would be to identify the major sources of income of the poor and ask how liberalization would impact these sources. For example, for a large chunk of the poor population, the principal source of income is likely to be labour, and the question is how liberalization will impact the real wage. Some of the poor may own small amounts of land, and thus earn a part of their income from producing and selling agricultural products. In this case, how the profitability of what they produce is impacted must be taken into account.

Theory will not give an unambiguous answer even after the question is thus narrowed down, unless a specific model to determine the likely impact on the factor prices is chosen. For example, if we rely on the two-factor, Heckscher-Ohlin model and think of agriculture as being labour-intensive, the rise in agricultural prices will lead to a rise in the real wage and hence a reduction in poverty. But if we think in terms of the specific-factors model, where land is specific to agriculture and capital to industry, the rise in the price of agricultural products will increase the wage in terms of industrial products but reduce it in terms of agricultural products. In so far as the poor spend the bulk of their income on agricultural products, they are likely to be worse off.

Precisely how developing countries and the poor will be impacted by trade liberalization in agriculture under the Doha Round is a complex issue. The presumption that such liberalization will broadly benefit the poor countries, implicit in the allegations that agricultural subsidies in the rich countries hurt the poor in developing countries, is unlikely to be supported by closer scrutiny in its unqualified form. In so far as such liberalization will raise food prices and the poor spend a disproportionately large amount of their income on food items, the opposite is entirely possible. The lion’s share of benefits of such liberalization is likely to accrue to potential exporters of these products, which happen to be relatively richer developing countries concentrated in Asia and Latin America. As such the case for agricultural liberalization must be made more on the grounds that the current system is hugely inefficient, resulting in very substantial deadweight losses and transfers to the relatively rich farmers in the OECD countries. Redirection of even a small fraction of these subsidies towards the poor in the Third World will go a long way towards alleviating poverty.