Chapter 4. Trade liberalization and food security in developing countries^[45]

4.1 Introduction

Developing countries face a number of risks associated with trade. Perhaps the best known is declining terms of trade, as the world prices of the primary commodities they export tend to fall over time relative to the price of the manufactures they import. A related problem is the volatility of world prices for the primary (especially agricultural) commodities they export. Furthermore, these prices are determined in markets beyond the influence of individual poor countries and typically affected by factors beyond their control. Related to this are supply side risks, especially the sensitivity of output to climatic variability. Droughts and excess rain creating flooding can cause serious damage to agricultural output.

An example is the decline in the value of SSA exports: “Of 47 African countries, 39 are dependent on a mere two primary commodities for over 50 percent of export earnings and the substantial drop in commodity prices in 1998 encompassed the entire range of African exports”. Food and tropical beverage prices fell by 13.6 percent and prices of agricultural raw material by 10.8 percent in 1998^[46].

The terms of trade faced by SSA countries deteriorated by 9 percent, a loss of real income equivalent to 2.6 percent of GDP, between 1997 and 1998^[47]. The trend is continuing: coffee prices in 2002 fell to less than a third of their 1997 level. This is perhaps the most extreme example of a general trend, but highlights how severe the risk is. Uganda is an apposite example of a country that implemented the trade and economic reforms requested of it in the 1990s, reaping the benefits of economic growth, increased coffee production and revenue, and reduced poverty. The country could soon see many of the gains undermined, if not wiped about, by a decline of world prices that is beyond its control. On a positive note, if this is associated with an increase in the relative return to food crops, farmers will substitute.

A new type of risk is emerging in the face of increasingly integrated global markets (one facet of globalization). This can be represented by distinguishing comparative from competitive advantage^[48]. Comparative advantage captures the potential provided by a county’s resource endowment to derive gains from trade. Competitive advantage considers why certain producers, in particular multinational firms, are able to exploit the rents from comparative advantage. Trade in agricultural commodities is dominated by large, typically multinational, companies that are present in all or critical stages of the commodity chain. At one extreme is contract farming where corporations control production^[49], at the other are supermarkets that control purchasing, and often multinationals control the distribution chain between production and final sale. The risk arises because small producers, and even some large producers in small countries, are the weakest link in the chain.

In addition, most developing countries are price-takers in the majority of international markets in which their nationals trade, but their activities are concentrated in a small number of markets. They cannot influence world market prices (mainly because of the small relative size of their market contribution), but at the same time are severely affected by changes in world market prices, especially when these changes are dramatic or unexpected. A related issue here is the increasing tendency for large multi-national companies to capture the benefits of comparative advantage by virtue of their monopsony position.

The distributional impact of trade reform can also be a critical issue in poor countries. The impact of trade liberalization on poorer groups within society may well be lost if only the aggregate implications are considered.

4.2 Trade policy and developing countries

Trade theory tells us that developing countries, since they tend to be endowed with land, labour and natural resources (rather than with capital and technology), should have a comparative advantage in agriculture. At the same time, the conventional view among trade economists at least, has been that the policy bias against agriculture in developing countries has often been severe^[50]. Trade policies, by protecting manufacturing and taxing (implicitly or explicitly) agriculture, have contributed to this distortion. Misguided agricultural, fiscal and investment policies have also contributed to the bias. Consequently, it is argued, trade reforms alone will be insufficient to remove this bias against agriculture^[51].

It follows that even if a government wishes to support and promote certain sectors, restrictive or protectionist trade policies may not be the optimal way to achieve such aims. Protection of a sector, typically manufacturing, through trade barriers increases prices of the output of that sector and increases profits of producers, by conferring rents rather than encouraging efficiency. Resources are allocated to the protected sector but evaluated according to world relative prices, these “importables” sectors are less efficient than “exportables” sectors. In other words, protection encourages the allocation of resources into sectors in which a country does not have a comparative advantage. Of course, this perspective ignores the “fallacy of composition” argument and the evidence that many developing countries (especially those exporting to world beverage markets) suffered from lower export earnings as a result of reducing agricultural export taxes: world supply increased and commodity prices fell.

In many developing countries, according to this view, protection has encouraged excess resources into inefficient manufacturing and insufficient resources into potentially efficient agriculture. This bias is exacerbated by policies that tax and discriminate against agriculture. Furthermore, protection reduces the quantity and variety of imports and increases the price of importables, therefore reducing consumer welfare. Tariffs and non-tariff barriers also encourage unproductive activities (rent-seeking), tax avoidance and evasion. These also contribute to inefficiency in the economy.

Assuming that a country initially has relatively high levels of protection, a satisfactory definition of trade liberalization is any set of reforms that reduces the bias against the production of exportables^[52]. The objective is to bring relative prices for importables and exportables in a country closer to relative world prices for the relevant commodities. While this argument only holds if world prices are good indicators of comparative advantage, frequently this is not the case. However, the theoretical logic is explored further, making this rather heroic assumption.

Import liberalization - the removal of quantitative restrictions, reduction and simplification of tariffs - contributes by reducing the price of importables (which, due to protection, were above the relative world level). This confers two types of general benefit.

• First, it promotes greater efficiency as it encourages the reallocation of domestic resources, away from relatively inefficient production of importables towards increased production of exportables. Export promotion may be required to ensure investment in and expansion of the exportables sector (so that agents respond to the altered relative incentives).

• Second, it reduces the price and increases the variety of imported goods available to consumers. In this way trade expands the consumption possibilities of a country.

Not all countries will benefit equally, however, and some countries may not benefit at all while, at least in the short-run, liberalization is likely to impose costs on some developing countries^[53]. The import-substituting producers that are most inefficient or are unable to increase efficiency will be unable to compete with imports and may close down. Export production may not increase fast enough to absorb resources released. Although this applies for the whole economy, consideration is restricted to the agriculture sector.

4.3 Some evidence from Sub-Saharan Africa

Many developing countries, especially in SSA, have liberalized trade policies since the 1980s^[54]. While there is fairly convincing cross-country evidence that exports are associated with growth, the evidence that liberalization increases growth is much weaker^[55].

Much of the export growth (where it occurred) in the 1990s was indeed agriculture-led. However, the evidence for SSA is quite mixed (see Table 4.1). Uganda is one of the few cases were incentives were improved for both food and cash crop producers. This did not automatically translate into increased value of exports, largely because world prices are beyond the control of small-country exporters.

Table 4.1 Impact of trade liberalization, Africa

Note: Coding is ‘+’ where effect was positive, ‘-’ negative and ‘mix’ indicates mixed evidence (trade policy reforms impacted differently across sectors/countries or over time).

Source: Adapted from Morrissey 2002, Table 20.3.

Often, the anticipated benefits from trade liberalization do not materialise because only limited or partial reforms are actually implemented, i.e. there is no significant increase in incentives for exportables^[56]. This is especially true of many SSA countries. Furthermore, even when significant trade reforms are implemented, important constraints remain. Many countries face natural barriers to trade arising from geographical remoteness, especially if land-locked, and high transaction costs. In Uganda, for example, transport costs represent an implicit tax equivalent to 24 percent of value added of coffee exports^[57].

Several reasons explain the only limited agricultural export supply response. Unilateral trade reforms do not affect the price received by exporters (multilateral liberalization may affect world prices). Devaluation of the exchange rate increases the domestic currency value of a given world price, therefore increases incentives to exporters. There is evidence that farmers do respond to relative (crop) prices, and in particular they will shift into food production if prices increase relative to export crops^[58]. However, their ability to increase production and exports to respond to increased incentives will be constrained by farming practices, limited access to inputs, credit and new technologies^[59]. Poor infrastructure and natural barriers act as a tax, often very high, on exports^[60]. Delays in implementing institutional reforms have been suggested as one factor limiting export supply response in Uganda^[61].

Another reason why one may not observe an increase in exports (by value if not by volume) is the fallacy of composition, whereby the simultaneous attempt by many countries to expand exports of the same commodity results in a decline in the world price. This is one of the problems faced by countries that are dependent for export earnings on a few primary commodities. While trade liberalization is beneficial, both through improving incentives to exports and providing gains to consumers, it is not a guarantee of economic growth, nor even of growth in exports.

4.4 Trade and food security

There are two different approaches to problems regarding the relationship between trade, specifically imports of food, and food security. The first is to argue that it is unimportant that a country be able to grow the food it needs, all that is necessary is that it should able to acquire the food it needs, i.e. to export goods to earn enough to pay for food imports. This has been defined above as self-reliance. Others argue that countries should be self-sufficient so that they meet their food needs fully from domestic production. This may imply supporting, if not protecting, farmers. Not all countries can expect to be self-sufficient in food. Some countries may not even be able to be self-reliant, if they have very limited export opportunities and high food needs relative to local production (e.g. many small island economies). Thus, governments should not begin by choosing a strategy of self-sufficiency or self-reliance. Rather, they should start by establishing an efficient (undistorted) agriculture sector and identify the extent to which this meets food needs.

Many developing countries start with a bias against agriculture, so agriculture sector reform complemented by trade reform will be necessary. Agriculture sector reforms are intended to increase productivity. In general, these will increase farm incomes or profit margins, and allow prices (especially of foods) to be reduced (at least in real terms). In this sense, agriculture sector reforms confer widespread benefits. Trade reform has mixed benefits. Import liberalization (easier access at lower prices) benefits those using imported inputs. This may include producers - farmers using imported fertilizer - or consumers (e.g. lower prices for food). However, it increases competition against those competing with imports, and this may include food producers. Commercial farmers, for example, may benefit from cheaper imported inputs but face increased competition from cheaper food. Measures that favour exporters are generally beneficial, but from the self-sufficiency perspective a problem arises if farmers substitute from food to cash crops. However, if the cash crops earn the revenue to import food, it is consistent with self-reliance. Assuming that appropriate reforms have been implemented so that policies are not biased against agriculture, countries can find themselves in a number of situations. Four cases are presented here.

• Some countries with efficient agricultural producers will be net food exporters. Food security will not be an issue, but they will be concerned with open access to foreign markets.

• Some countries will be naturally self-sufficient. At prevailing domestic prices, which should be equivalent to true world prices (see below), domestic producers are capable of meeting local food needs at least in normal years. In good years they could export food, or stock food as insurance against a bad year.

• Some countries will not be self-sufficient, but will have export earnings that allow them to meet food import needs, i.e. they are self-reliant. They may be exposed to risk if they are export-dependent on primary commodities. Therefore, it is preferable that export earnings are from a diversified, especially manufacturing, portfolio.

• Some countries will be naturally food insecure. Strictly speaking, it is only in respect of such countries that the issue of an active food security policy arises. A costly option is to provide subsidies to farmers, and this may not be viable. A number of countries are likely to remain dependent on food aid, or aid that can be used to finance food imports.

An important issue to address in the context of food security is whether imports are at true world prices or are in fact subsidized. The policies of many developed countries (and some developing countries) ensure that many temperate foods are sold on world markets at subsidized prices. Although this is a benefit to net food importers, it represents a clear disadvantage to developing countries that are aiming for self-sufficiency (it). They cannot afford, and in many cases are discouraged by the World Bank and/or the IMF, to subsidize domestic producers. If they permit food imports tariff-free, this amounts to unfair competition against domestic producers. In cases where subsidized imports compete with local production, it could be appropriate to levy a tariff equivalent to the subsidy. In fact, the distortion in world prices is, in many cases, substantially due to the massive transfers to agriculture and very high import tariffs in many developed countries, the combination of which has a much greater distortional effect on world prices than export subsidies.

It would be wrong to preclude this policy option (anti-dumping duties are a similar response, but are administratively more complex). This would be preferable, on efficiency and cost grounds, to direct subsidies to local farmers. Often, however, imports do not compete with local products, because of product differentiation and/or market segmentation. The best solution is to ensure there are minimal, or no, biases against agriculture domestically. Once this is ensured, one can then assess if imports are indeed subsidized and if they compete with local products. A countervailing tariff is justifiable if the answer to both is yes.

The very poorest countries are typically predominantly rural. The combined effects of changes in prices and domestic policies will affect farmers in different ways, depending on how the relative margins on the crops they produce are affected. One possibility is that relative incentives to agricultural producers will alter in favour of food crops, especially if urban food subsidies are removed (as these are often engineered by reducing the price paid to farmers). Supply response should lead to an increase in output, and a corresponding increase in farm incomes (as farmers shift to more profitable crops). There should be a positive direct effect on rural employment, although this may require an increase in aggregate output. This depends on how mobile factors are within agriculture, but in general both land and labour should be quite mobile between crops. Agricultural reforms that improve factor mobility (such as improved access to credit and functioning markets for land) or productivity can play an important role here.

These effects relate to substitution possibilities between crops (a crucial feature of supply response). The effect on aggregate output is less clear, and depends crucially on the scope for adopting new technologies. A benign scenario would assume farmers can gain access to new technology, increase yields and profitability, and so increase food output and exports, whilst allowing lower domestic food prices. If the real price of food were reduced, both rural and urban poverty could be reduced. A less optimistic scenario would be where farmers’ ability to increase yields and profitability were constrained. If increased output (or increased import competition) reduced food prices, less efficient farmers may suffer from reduced real incomes. The overall impact is impossible to predict, as it depends on features specific to the farmers and country in question, in particular the pattern of production and the severity of constraints to substitution and expanding yields. Nevertheless, if agriculture sector reforms are implemented, the potential impact of trade liberalization on farmers is beneficial.

Thus, a number of steps are necessary to evaluate the link between trade policy and food security, and such an evaluation needs to be specific to the country. First, one should account for any policies that discriminate against domestic agriculture relative to other sectors, and where appropriate these biases should be reduced or eliminated. Second, having done this one can classify the country under one of the types listed above. Countries that are net exporters or naturally self-sufficient can expect to benefit from trade liberalization. Countries that are inherently food insecure will need some assistance, and will face increased import costs if multilateral liberalization leads to higher food prices. Some countries will be borderline self-sufficient, and these are the most likely to be adversely affected by subsidized imports (and should be permitted to take countervailing actions). Finally, some countries may be borderline self-reliant, if export earnings are volatile.

4.5 Conclusion

It is apparent that the relationship between trade policy and food security is not a simple matter. As a general principle, if domestic policy towards agriculture provides adequate support and incentives for farmers, trade barriers should not be used as an instrument of protection. If rich countries subsidize food exports, the presumption should be that this is good for consumers in food importing countries. The poor in particular can benefit from lower food prices. One needs to assess carefully if imports actually represent unfair competition with domestic producers. Often, independently of the price of imports, domestic production is insufficient to meet demand. For example, the United Republic of Tanzania retains a relatively high average tariff of about 20 percent on food imports, and this was unchanged in the late 1990s. The value of imports almost trebled between 1997 and 1998, but this was because of a shortfall in domestic production and not because of trade liberalization. Furthermore, the commodities imported are often not directly competing with local production (there is market segmentation), so an increase in (cheap) imports does not necessarily pose a threat to local farmers.