Chapter 1. Food security and trade: an overview

1.1 Introduction

This chapter seeks to link the concerns of developing countries with respect to trade, food security and economic policy in the context of the Doha Round negotiations. It defines the changing conceptual basis of food security and presents some indicators and estimates of trends in aggregate food security status. This is followed by a review of approaches to food security at the household level, and of frameworks for investigating the wide range of factors influencing food security status at this level of disaggregation.

The second part of the chapter is concerned with the theoretical underpinnings of trade policies and strategies. It reviews the conventional or orthodox approach to trade theory and its predictions, as well as some of the criticisms of this approach and of the supporting evidence. Some of the associated risks for developing countries of further trade liberalization are reviewed.

The final section links the two issues of food security and of trade liberalization, highlighting the implications at the national and household levels.

1.2 Food security at the national and household levels

Food security is a multi-faceted concept, variously defined and interpreted. At one end of the spectrum food security implies the availability of adequate supplies at a global and national level; at the other end, the concern is with adequate nutrition and well-being.

In this section, issues surrounding food security at the national level are investigated first, before a review of approaches to household food security. The question of food security in intra-household relationships will not be pursued here.

Food security at the national level^[1]

The extent of global food insecurity

Recent initiatives aimed at improving the food security situation of the poor - most notably the World Food Summit (WFS) - have been stimulated by the fact that although food availability for direct human consumption grew by 19 percent between 1960 and 1994-96, to 2 720 kcal/day (against an estimated minimum daily energy requirement of 2 200 kcal/day), availability is still very uneven. In sub-Saharan Africa (SSA) calorific intake is still only 2 150 kcal/day compared to 2 050kcal/day thirty years earlier. In contrast, the average calorie consumption in South Asia rose from 2 000 kcal/day to 2 350 kcal/day in the same period.

However, during the 1990s per capita growth of world agricultural production slowed. World cereal output^[2], for example, fell from a peak of 342 kg per person in the mid 1980s to 311 kg per person in 1993-95, although it has since risen to 323 kg per person in 1996-98^[3].

The results of such statistics are evident in the fact that in 1995-97, 820 million were estimated by the FAO to be undernourished, with 790 million living in developing countries. Although the number of undernourished people in developing countries actually fell by 40 million between 1980/82 and 1995/97, this improvement was also uneven, being attributable to a reduction of 100 million in 37 countries, whilst in the remaining countries the numbers increased by 60 million. In addition, the fall in absolute numbers is too low to achieve the WFS goal of reducing the numbers of undernourished by half by 2015, since this would require an additional reduction of 20 million undernourished individuals each year until that date^[4]..

Food security indicators

The Committee on World Food Security, a body set up in 1975 by the UN World Food Conference to oversee developments in food security, adopted in the early 1980s the recognition of food security as a tripartite concept, reflecting the criteria of availability, access and stability. Similarly, the OECD suggests that food security has three dimensions: availability, access and utilization, although this source indicates that there is a tendency to characterize it in terms of availability. Chapter 2 discusses these concepts in more detail.

Attempts to capture trends in variables that are likely to reflect food security^[5], can be broadly categorized into two interrelated sets: those that directly measure shortfalls in consumption requirements, and those that concern the potential to meet such shortfalls.

The United States Department of Agriculture (USDA)^[6] evaluates two aspects of food security, availability and distribution, both of which capture the extent of the shortfall, and analyse predicted trends through to 2009. The most recent study covers 67 countries that have been, or are, potential food aid recipients. Two key indicators are used: first, the Status Quo gap, which measures the difference between projected food supplies (calculated as domestic production plus commercial imports minus non-food uses) and a base period (1995-97) per capita consumption^[7]; and second, the Nutrition gap, which is the difference between projected food supplies and the amount of food needed to support minimum per capita nutritional standards.

The Status Quo indicator provides a safety net criterion, whilst the Nutrition gap indicator gives a comparison of relative well-being. In some regions, the size of food gaps is quite small relative to commercial imports, meaning that if imports grew at a slightly higher rate the projected gaps could close (for example in North Africa and in Latin America and the Caribbean). In Asia^[8] however, the ratio of the nutrition gap to commercial imports is about 20 percent and in SSA it is projected to be 229 percent. It is highly unlikely that the gap can be filled. Food imports would need to grow by 10 percent per year in SSA and 4.7 percent in Asia to fill this gap by 2009.

At a more aggregate level, the FAO Committee on Food Security reviews a set of six indicators derived from observations of the global cereals market. Although these indicators (see Box 1.1) are confined to cereals, the contention is that they shed light on the global food situation due to the weight of cereals in the overall food basket and thus overcome the difficulty of aggregating over food commodities in calculations of the total food supply and of food imports.

A key difficulty in interpreting these indicators is that they make no reference to the ability of a country to meet increased import requirements. For some countries the availability of foreign exchange will be a binding constraint. Financial constraints can, however, limit the role of imports in filling the shortfall between production and consumption in many countries. Low commodity prices, for example, may limit export earning potential. The USDA^[9] notes that the ratio of foreign exchange availability to food imports is not one to one, but is higher, meaning that a 1.3 to 2 percent increase in foreign exchange availability is associated with a 1 percent growth in food imports.

The second set of food security indicators relates to indicators of changes in world markets, which in turn indicate the potential to meet food shortfalls. Two key primary indicators are world food price stability and world food price levels.

These affect both the ability to finance imports via export earnings and changes in the food import bill, themselves potential indicators of changes in the food security situation.

The European Commission, for example, suggests that the instability of world markets is mainly transferred to each country via the import price of cereals^[10]. The aggregate impact on a country therefore depends on cereal imports as a share of total imports, the price elasticity of imports and the capacity to finance imports via export earnings. Vanzetti concludes that the linking of domestic and world markets that would occur under a free trade regime with no government stocks would reduce the variability of the world price of grain by one-third. However, he cautions that any analysis of the instability of food consumption needs to distinguish between instability due to fluctuations in national production and instability of unit import costs, i.e. world prices^[11].

A recent study by Valdés and McCalla^[12] calculates an indicator of Food Import Capacity as the ratio of the food import value to the total export value (excluding services). The authors find that this indicator is relatively large for small Island Developing Countries (0.70), and for a number of sub-Saharan African countries including Gambia (1.99), Lesotho (0.85) and Mozambique (0.94). However, the ratio is much smaller for larger economies such as India (0.05) and Argentina (0.04).

In contrast to the above, Paarlberg^[13] argues against using primary indicators of changes in international grain markets as indicators of food security, because most food insecure countries still depend only lightly on imports of grain from the world market. Paarlberg states that importing countries often do better overall when world grain prices are high, because prices often rise under conditions of rapid international growth. Evidence for this claim is that during the “world food crisis” of 1973/74, when the real export price of wheat increased by 103 percent and of maize by 58 percent, and when food reserves dropped to the equivalent of 33 days of global consumption requirements, there was no decline in overall consumption levels. Indeed, in most countries per capita cereal consumption was steady or even expanded. The increased prices in 1995/96 also failed to produce any notable decline in consumption. Paarlberg states that between 1994/95 and 1995/96 wheat export prices increased from US$157 to US$216 per tonne and global stock levels fell by 14.1 percent, but import levels were sustained. By contrast, the 1980s that were characterized by low world market prices and severe food crises were also marked by global recession.

However, Paarlberg does acknowledge that some poor countries have come to rely on food imports to a greater extent during the last three decades. Sub-Saharan Africa’s reliance on imports (import dependence calculated as the share of food import costs to total import costs), including food aid and commercial imports was 13.6 percent in 1993 (up from 10 percent twenty years earlier).

In the light of this discussion, it is apparent that potential indicators should reflect changes in the food import requirements of developing countries, and in their ability to finance any increase in the import bill (see also Chapter 4). They should also be able to capture the effect of the gap between an increase in the import bill and any increase in domestic production (and potentially exports) as a result of a world price increase.

Two indicators may prove useful in distinguishing the impact of a weak supply response in agriculture in some developing countries:

• changes in the ratio of cereal import requirements (derived in value terms as consumption requirements minus domestic production) to total agricultural export earnings;

• changes in the ratio of cereal import requirements (derived in value terms as consumption requirements minus domestic production) to total merchandise.

In economies where the agriculture sector is less flexible than in other sectors facing improved incentives, one would expect the first indicator to increase at a greater rate than the second. In assessing the potential for increased export earnings from agriculture, it is also important to determine changes not only in the total value of agricultural and merchandise trade, but in their shares of total exports and in the diversification of the export portfolio.

This brief review of potential indicators points to the fact that those capturing the ability to finance import requirements, by for example export earnings, are likely to be more robust indicators of food security than either those based on the primary indicators of price levels or price instability, or those based upon trends in stocks and flows in global cereal markets.

Household food security

The ability to ensure adequate food security hinges on the ability to identify vulnerable households. Chapter 2 reviews many of the links between food availability and nutrition. Here we focus on the broad picture. Vulnerability refers to the full range of factors that place people at risk of becoming food insecure. The degree of vulnerability of an individual, household or group of persons is determined by their exposure to the risk factors and their ability to cope with or withstand stressful situations. Generally, vulnerable households will constitute three groups:

• those which would be vulnerable under any circumstances: for example, where the adults are unable to provide an adequate livelihood for the household for reasons of disability, illness, age or some other characteristic;

• those whose resource endowment is inadequate to provide sufficient income from any available source;

• those whose characteristics and resources render them potentially vulnerable in the context of social and economic shocks: e.g. those who find it hard to adapt to sudden changes in economic activity brought about by economic policy. A significant increase in the consumer price of staple foods might be an example.

Although no definition of ‘vulnerable’ is complete, a useful starting point is estimates of income. It can be assumed that the first two categories will be relatively poor both in terms of income and assets, and it is also likely that the third category will have a fragile resource base and other characteristics which make its income sources uncertain. An appropriate proxy, therefore, in identifying vulnerable households, is how poor is a particular household measured against some established criterion or ‘poverty-line’.

Having defined who the poor are, the second step is to identify their household characteristics:

• location: rural/urban; small village/large village; remote province/near to capital city etc.;

• composition: size, age and dependency ratios; male/female head;

• sources of income: production, employment, trade, remittances and other transfers.

A frequent problem in delineating those sections of the population most vulnerable, or at risk from changes in policy direction, is the lack of baseline data regarding household income and consumption patterns.

The principal concern of this volume is the way that trade liberalization impinges on food consumption through food availability, food access, and the stability of food supplies. The notion of household entitlement to food, derived from the work of Amartya Sen^[14], is now widely used to investigate issues related to both food security and nutrition. Chapter 2 refers to this approach, while Chapter 5 elaborates the concepts as they relate to policy variables. The word “entitlement” refers to the various means through which households avail themselves of food, whether through household production, or through other income-generating activities such as the sale of labour or participation in trading. A number of these activities may be pursued by the same member of each household, or by different members. In addition, transfers from sources external to the household, i.e. from the state or friends and relatives, will also add to household entitlement.

Entitlement can also be perceived as the household’s ability to express effective demand for food. It presupposes the availability of food, since for demand to be effective it must be capable of being transformed into consumption. This applies as much to food grown for household consumption as to that purchased with income generated through other activities or from transfers. The former entails a decision to retain part or the whole of the output of productive activity, as opposed to selling it and purchasing food or non-food commodities. Demand is expressed in these decisions.

Household activity or transfers do not directly result in access to food, for there are a number of intervening stages that mediate the process. Both governments and agencies concerned to augment household food security intervene in order to mediate between potential and reality.

In the first place, the resource endowment of the household will determine its capacity to produce or to trade. Events such as civil unrest or climatic disasters can seriously deplete households’ resource potential, and increase the likelihood of structural food insecurity. If what might have appeared as a transitory problem is not to become chronic, the replenishment of productive capability should be a necessary part of programmes aimed at reversing this process. Physical resources by themselves, however, may be inadequate, and the upgrading or changing of the range of skills possessed by household members may be a necessary component of any programme. Consequently, training in new agricultural techniques, or in the necessary skills required by local industries or trades, can form an integral component of food security interventions.

For many poor households, particularly those whose resource base been eroded by drought, additional resources are the primary requisite if their productive base is to be restored. Recognition of this is apparent in the increasing emphasis on development programmes by governments, agencies and donors alike. For other households, both rural and urban, access to productive resources may be less relevant. These will seek, according to their location and particular skills, to generate entitlement to food through trade or direct employment.

The promotion of income-generating activities, both local employment opportunities and self-employment (particularly those associated with the rural informal sector), forms a second essential approach to food security.

Moreover, in circumstances where both the outcome of productive activity is always uncertain and the purchasing power of cash-generating activities is subject to sudden and dramatic shifts, it is both probable and desirable that households will seek to diversify their occupations. This may be either through the principal income earner undertaking a variety of activities, or through different household members generating income or produce from a variety of tasks. Here again, policies designed to promote food security might also simultaneously address resource and skill constraints.

Apart from the choice (insofar as one exists) between producing food or non-food crops, farm households also make decisions about whether to retain or sell the food they produce. To some extent, these decisions are dictated by the existence, non-and efficiency of marketing infrastructures and of household storage facilities. Where either of these is inadequate, inopportune selling in unfavourable markets can have a detrimental effect on food security. Inadequate storage facilities will, in most circumstances, lead to heavy storage losses, significantly affecting the seasonal availability of food.

The provision of marketing infrastructures is essential not only for traded income, derived through both farm and non-farm activity, and food and non-food production. Its absence in rural areas will also impede the transfer of essential food and non-food commodities, and so reduce the incentive for household economic activity.

Finally, transfers from the state or individuals can augment entitlement to food. Typically, these latter sources of entitlement take the form of cash payments or gifts, although in-kind payments and remittances are also a common occurrence. In the latter case, the household is faced with the previously discussed choice of sale or retention. In both cases there is likely to be a basket of essential cash purchases that households will wish to undertake, and cash remaining can be used to purchase food. The actual mix of food and non-food essentials that are purchased will be determined by both availability and price, with both absolute and relative consumer prices being a crucial determinant of household food security.

It is important to recognize, however, that access to food through any of these entitlement endowments contributes only to the availability of food to the household. It does not ensure efficient utilization and says nothing regarding intra-family distribution, both of which can have a profound effect on nutritional status regardless of food availability.

1.3 The gains from trade: theoretical perspectives

The arguments for trade liberalization are strong, and typically inform policy advice to governments from international institutions. These arguments are premised on Ricardian “conventional” or “neo-classical” trade theory, and in particular the theory of comparative advantage using general equilibrium models. These deal with resource allocation in the whole economy under the stylized conditions of perfect competition.

The theory argues that differences in productivity and opportunity costs of production between countries form the underlying reasons why it is advantageous for countries to engage in trade. Many reasons explain why such differences occur. Climate is of obvious importance for agriculture as is the availability of extensive arable land and abundant water supply. The availability of other natural resources, such as large and easily accessible mineral deposits, and differential access to productive technologies give rise to varying labour productivities.

The Heckscher-Ohlin (H-O) theorem provides the most widely accepted explanation of the pattern of trade, based on countries’ differing factor endowments and the factor requirements of different kinds of goods. Chapters 3 and 4 draw their conclusions with respect to the advantages of trade liberalization from the H-O model.

The theory states that trade occurs because the cost of labour relative to that of capital is lower in the labour-abundant country, which means that the price ratio of labour-intensive goods to capital-intensive goods is lower in the labour-abundant country than in the capital-abundant country.

This provides a basis for comparative advantage and when trade begins each country exports commodities that use the relatively abundant factors and imports those that use scarce factors more intensively. This is the equivalent of exporting labour for capital, in the case of the labour-abundant country, but as factors are not mobile internationally, commodities have to move instead.

This model is sometimes referred to as the factor proportions or factor endowment model. An adequate non-technical representation of comparative advantage is that countries should produce those products that use relatively intensively the factors with which the country is relatively well endowed. A logical consequence of trade, therefore, is a process of eventual factor price equalization leading, for example, to real wages (as well as other factor prices) becoming the same across trading countries. It also implies that, other things being equal, the labour-abundant country exports labour-intensive goods, whilst the capital-abundant country exports capital-intensive goods. Arguably, this process could play an important role in poverty reduction in labour-abundant developing countries, by bidding up the price of labour and thus raising workers’ incomes.

Advocates of free trade also argue that, under competitive free market conditions (the stylized conditions of perfect competition) trade maximizes potential economic welfare internationally, by creating a situation where no country can be made better off without another being made worse off. It is a situation where those that gain from trade could fully compensate those that lose and still be better off: the total gain will be greater than the total loss. With free trade a point would be reached where more of each traded good is produced, such that everyone will gain if suitable redistribution is made.

There are a number of important qualifications to these predictions of the model, however, that must be held in mind. First, the consequences described are dependent on the assumption of competitive markets (a level playing field). In the absence of these, countries may be better off intervening to restrict free trade. Second, countries will not necessarily gain equally from trade: the relative gains will depend on the terms of trade. Thirdly, there are no mechanisms in place to ensure that losers in the world market will be compensated by those that benefit, so the gains remain potential. Fourthly, the issue of redistribution also applies within countries, where there will also be gainers and losers from trade. Finally, any comparative static solution described by the conventional theory assumes that all external costs are internalized, including environmental externalities, a subject of some contemporary debate. Although this theory is the basis of modern “orthodox” trade economics, this does not mean that it is accepted without questioning: there are gaps in the theory’s coverage and question marks over some of its predictions.

A large number of empirical studies have considered the extent to which the hypotheses of conventional theory are supported by empirical observation. These have usually tested the “factor proportion” prediction of the model by comparing the factor intensities of imports and exports. While empirical observations suggest that factor proportions alone cannot explain the pattern of international trade, the theory does seem to provide a partial explanation of trade flows between developing and industrialized countries. In addition, a variety of extensions to the model have been developed to take account of any empirical shortcomings, and to cater for such factors as externalities and the absence of perfect competition.

In sum, the conventional theory uses a simplified model of the world in order to generate a logically consistent theory of the effects of free trade liberalization. While the simplifications may be questioned, the theory has proven to have considerable analytical power, and to produce clear, testable predictions. For these reasons, it remains the dominant framework of analysis for the policy decisions of governments and international organizations.

The theoretical approach outlined above underpins the policy advice concerning trade liberalization given to governments by international institutions, as well as the approach adopted by the WTO Agreement on Agriculture. It is relevant, however, to review the assumptions that underpin the model: it assumes perfect competition, where no country or firm is able to influence prices, where there are no economies of scale and where products are homogeneous. It also assumes that second-best situations have been recognized and acted upon, and that externalities have been internalized.

An important question emerges from this theoretical review, however. If free trade could potentially raise economic welfare in the world as a whole and even in all trading nations, why are border intervention policies so commonly used by governments to restrict free trade?

Trade theory literature provides three main explanations for this apparent anomaly. First, the case of the “optimum tariff” shows that in certain circumstances a country can gain more from imposing a tariff than from free trade (assuming other countries do not retaliate). Such gains are at the expense of losses by trading partners (a zero-sum game, in other words). However, the optimum tariff argument mainly applies to large countries, which can use tariffs to influence their terms of trade in world markets. It does not generally inform developing countries’ portfolios of potential policies, unless they are part of a larger trading bloc.

A more interesting reason for protection in the context of this study is the infant industry perspective. Where an industry has large economies of scale, firms may need protection to allow them time to grow before competing head-on with more established firms overseas. This assumes that an underlying comparative advantage in the particular product exists. This remains an important justification of protectionism in developing countries, especially for manufacturing industries. It may equally concern the food and agriculture sector where the argument can be applied to primary processing industries, in the context of a development strategy involving an export shift from raw-materials to processed products.

A final explanation concerns political imperatives, including the influence of groups which gain from protection, and the importance of revenue from border measures for developing country governments, where other tax bases are not strong. While the latter may be a short-term expedient that is difficult, in some cases, to substitute for without reducing government spending or increasing borrowing, the importance of non-trade concerns such as food security and rural viability are often put forward as powerful imperatives for protecting domestic agriculture.

The political economy of trade policy, suggests that the glaring gap between theory (of gains from free trade) and reality (of widespread protectionism) can be largely explained by the political and economic forces that come into play when the assumptions of perfect competition and frictionless exchange do not hold. In other words, where there is not a level playing field.

1.4 The impact of trade liberalization in developing countries

As has been demonstrated above, the arguments that openness to trade contributes to economic growth and that this can, in turn, be beneficial for poverty reduction and food security, are well grounded in conventional economic theory and have been supported by a number of empirical studies. However, some commentators caution that in studying the correlation between more trade and higher economic growth, researchers need to be careful about implying causality.

At the same time, however, the potential gains from trade liberalization are not guaranteed and will not necessarily be reflected in improved food security status of all groups within society. In particular, there are likely to be significant differences between the impacts on small scale and commercial farmers, rural non-farm producers and urban consumers both within and across countries. These need to be considered in identifying the food security implications of trade liberalization.

The apparent lack of success in stimulating development in many rural economies following economic and trade policy reform programmes has resulted in a wide-ranging debate that has recently broadened to consider the impact of not only domestic structural adjustment programmes, but also of globalization forces, including the global trade reform agenda.

A recent World Bank report^[15] reviews the evidence as to whether globalization supports poverty reduction and concludes that whilst a category of “new globalizers” are benefiting from greater integration into the world economy, a significant group are becoming more marginalized. The degree of openness to trade has been proposed as one potential reason for this divergence. Diaz-Bonilla and Reca^[16] for example, find a positive correlation between trade openness and economic growth. Sachs and Warner^[17] suggest that openness explains in part, the different export performance of Asia, Africa and Latin American Countries (LAC) in processed and high value added agricultural goods. However, they also note that diverse performances can be attributed to a broad range of factors, including for example: differentiated population dynamics, climate patterns, degrees of technical development and domestic policy sets.

The main theme of a paper by Rodrik^[18] concurs with this qualification, suggesting that “there is no convincing evidence that trade liberalization is predictably associated with subsequent economic growth” and that studies that suggest that there is evidence are “misattributing macroeconomic phenomena to trade policy”. Rodrik finds that the only “systematic relationship is that countries reduce barriers as they get richer”, concluding that initial economic growth was generated when trade was protected.

A similar standpoint is presented by SAPRIN^[19], which argues that liberalization has resulted in growth in imports exceeding growth in exports, and that this increased exposure to imports is associated with a reduction in domestic productive capacity and in the purchasing power of consumers. The authors also suggest that an absence of domestic market reform can result in reduced competitive advantage as trade reform proceeds, because the costs of production increase relative to those in countries that have successfully implemented domestic reform programmes. This may be reflected in trade patterns.

Whilst theory may suggest that the liberalization of trade policies will result in net benefits to the liberalizing country, and whilst there may be a growing collection of empirical studies to support the theory, it is also clear from the preceding discussion that the benefits of liberalization will not necessarily be achieved, and even where they are, some groups of individuals within some countries are likely to be disadvantaged. In a concise and convincing paper, Winters^[20] argues that although he believes that trade liberalization aids economic growth, it “may have some adverse consequences for some - including some poor people - that should be avoided or ameliorated to the greatest extent possible”. He suggests that rather than using this as a reason for resisting reform, it should “stimulate the search for complementary policies to minimize adverse consequences and reduce the hurt that they cause”.

It is clear that there is no clear consensus that liberalization results in economic growth, despite a number of major research programmes investigating this relationship. It is therefore important to understand the types of reform that have had the greatest impact on economic growth in each country.

1.5 The relation between trade reform and food security

For many developing countries, especially the poorest, the relationship between trade reform and food security is likely to provide the foundation of one of the most critical debates of the Doha Round of international trade negotiations.

The international dimension is significant, since trade policy influences both global food availability (in the case of a major importer or exporter), and national food availability (through both imports and production). The effect on food imports will be mediated by any implications of trade policy for foreign exchange earnings.

Trade policy will also have implications for food security through the link with incomes and expenditures. Any change in the trade regime will have a direct effect on both rural and urban incomes, and employment, and through these on income distribution. In addition, there will be an effect on government revenues through, for example, a change in the level of revenue from import levies.

Both national food availability and government revenues impact at the household level, affecting household access to food directly and indirectly through household incomes.

Food security and trade liberalization

Trade liberalization implies a change in the relative prices of traded and non-traded goods and factors in a previously protected sector or economy. The change in relative prices will induce changes in the allocation of resources to different activities and hence changes in both subsectoral and aggregate levels of production. In turn, changes in income levels (which are expected to increase in aggregate as resources are used more efficiently) have the potential both to reduce poverty levels and in doing so, to improve the food security status by increasing the access of the poor to food.

In the short-run, agricultural sectors in poor economies are often not well placed to benefit from trade liberalization even when this has had a significant impact on both income levels. This is because of the inflexible structure of production and trade in this sector, often manifested in limited market access and weak institutional development, as well as limited capacity to respond to improved incentives. However, food importers are affected in the short-term via higher import bills. As a result, there is often a hiatus during which the food security situation worsens.

The strategy employed by individual countries to improve their food security status is one of the key factors in understanding the relationship between trade liberalization and food security. Two broad options have generally been followed by countries attempting to achieve adequate levels of food security: food self-sufficiency and food self-reliance.

• Food self-sufficiency, or the provision of a level of food supplies from national resources above that implied by free trade, represents a strategy followed by a wide range of countries. While this approach implies the provision of sufficient domestic production to meet a substantial part of consumption requirements, it does not necessarily imply that all households in the country have access to all the food they require. In a number of countries which are net food exporters, substantial numbers of households are suffering from malnutrition.

• A strategy of food self-reliance reflects a set of policies where the sources of food are determined by international trade patterns and the benefits and risks associated with it. This strategy has become more common as global trade has become more liberal. It is even argued that improved food security, as well as efficiency gains, may be achieved more satisfactorily, even in countries where agriculture remains a major contributor to GDP, by shifting resources into the production of non - food export crops and importing staple food requirements.

The success of these broad options will depend, inter alia, on the ability of producers to react to price incentives (particularly important), or of countries to use income gains for improved efficiency of resource allocation in order to procure food on the international market. The distinction can also be used at the household level to motivate an understanding of individuals’ entitlements to food.

1.6 Conclusion: some key questions

Achieving food security means ensuring that sufficient food is available, that supplies are relatively stable and that everyone can obtain food. At least at the household level, if not at the national level, food security can be interpreted as being determined, inter alia, by purchasing power. Changes in the latter, in turn, are conditional on economic growth and the distribution of income and resources. Put very simply, this implies that, for many developing countries, food security and equity are two sides of the same coin.

The main issues of contention regarding trade liberalization are summarized in the following paragraphs.

• Many commentators suggest that a positive correlation exists between more open trade regimes and economic growth. Although the evidence for this may be questionable, it can be argued that a production structure based on current comparative advantage is more likely to be efficient in terms of resource allocation. The case is less clear-cut with regard to the distributional consequences of this form of economic priming, however. For example, positive employment outcomes and consequences for the poorer strata of society may not be axiomatic.

• More open trade implies a change in the structure of production. It is important to know what is produced, with what resources and by whom; and how adaptable local resources are, particularly labour, both in terms of skills and location. If further trade liberalization leads to a contraction of the agricultural sector, it may be neither easy nor quick for investment and employment opportunities to be created elsewhere in the economy.

• It is argued that a more open trade regime reduces the supply variability of staple foods. This is certainly likely in the context of stable and predictable international markets, where reliance on domestic stocks to stabilize domestic consumer (and producer) prices may be an expensive alternative. If, however, the open trade context is less stable and predictable than under protection, then supply variability will increase.

• A further advantage of agricultural trade liberalization, it is argued, stems from the possibility of lower domestic food prices. If this does not occur there would be no reason for protection in the first place. Any effect will depend partly on the transmission elasticities between international and domestic prices. Even where food prices do fall, however, this is not necessarily a straightforward advantage. The outcome depends on the location and employment of the food insecure, i.e. of the poorest strata of society. If many of the poorest households are dependent directly or indirectly on agricultural production for their main income, the overall effect on food security may be negative.

• On the other hand, it can be argued that the tax revenue from additional imports can be used to finance the adjustment of those disadvantaged. Of course additional tax revenue will only accrue if the increase in imports more than compensates for the lower tariff leading to the rise in imports. This is a reflection of the elasticity of import demand.

• It may well be the poorer, small-scale farmers who are using more labour-intensive techniques, who are more able to adapt to changing demand patterns. In any event, it is important to know the location of the poorest households, what their income earning possibilities are, and the constraints they may face in adjusting to changing economic opportunities. Where there may be short-run difficulties in responding to the changing context it may be incumbent on government to introduce the liberalization in a gradual and sequential fashion.

Obviously each case needs to be considered on its merits, and often it may be sensible to eschew a firm commitment either to freer agricultural trade or to greater self-sufficiency. The discussion concerning the advantages and disadvantages of a more open position does highlight a number of questions for policy makers. Before reviewing these, however, it is worthwhile considering the trade-off between growth and equity.

The ideology of liberalization which has dominated conventional wisdom in recent years as far as economic policy is concerned, makes an implicit assumption regarding the welfare effects of growth: i.e. that liberalization will engender economic growth which will in turn lead to enhanced economic welfare. This subsumes any argument regarding who is made better off by the growth which ensues, it being assumed that or in the long-term no-one will be made worse off. Once this assumption is accepted (as it is within most structural adjustment strategies) it becomes straightforward to model economic policy around a growth strategy, with distributional issues relegated to the role of short-term ameliorative measures: the provision of a social and economic safety net.

The problems of providing effective ameliorative measures in the context of growth orientated strategies, however, have often proved intractable, hence persistent high levels of under-employment and food insecurity in many high-growth economies. This is not to argue that growth should not be a priority, but to suggest that it might be helpful if economic policy was shaped around meeting short- and medium-term social and distributional priorities, with trade and investment strategies moulded around these more primary objectives.

Any new accord on international agricultural trade will imply a number of questions that governments might wish to take into account during the negotiation process.

• Is the goal of future policy more concerned with self-sufficiency or food security?

  Implicit in this question is the observation that self-sufficiency does not necessarily imply that all households in the country have access to all the food they require.. Implicit also is the notion that, in some circumstances, food security may be better achieved, even in countries where agriculture remains a major contributor to GDP, by shifting resources into the production of non - food export crops and importing staple foods.

• How would domestic production be affected by a more liberal import regime?

  To what extent would border prices be transmitted internally, and what would be the likely producer response? Would there be a restructuring of agricultural production? Would many farmers’ livelihoods be threatened? Are these more likely to be small-scale or large-scale farmers?

• What are the employment implications of a more open trade regime?

  Important here is not only the future capacity of the agricultural sector to absorb labour, but also the likelihood that any labour shed from the sector could be absorbed elsewhere. The limited nature of employment prospects outside agriculture might be a valid reason for maintaining a larger agricultural sector than might be the case in the context of further trade liberalization.

• Food security for whom: who would be affected and how?

  It is axiomatic that a more open trading system will have different implications for food security depending on whether those households that are considered to be vulnerable in terms of access to food are net consumers or net producers of food. Notwithstanding the implications for household food security of any changes in production and employment, therefore, the relative impact on urban and rural food security and the relative weights attached to any changes in each need to be considered.

• What complementary and/or compensatory policies might be required?

  Although there is a body of evidence, both theoretical and empirical, to support the case that liberalization does result in net benefits, these benefits are not guaranteed (if for example, implemented where markets are not functioning well) or may be associated with negative impacts on some groups in society. Successful liberalization has often been associated with complementary policies that are aimed at facilitating the process of adjustment, and/or with compensatory policies that minimize the negative impacts on potentially disadvantaged groups.