Sustainability of high food import dependence

Contents - Previous - Next

The fact that FDCs are predominantly agriculture-based economies facing particular difficulties to finance the most essential of imports food - indicates serious general development problems. The problems of food-import dependence are worsening in these countries. This implies increasing costs for FDCs' treasuries and also in terms of external dependence in general, of development opportunities missed because of the need to limit imports of productive capital goods and of a precarious base for food security.

A recent study conducted for FAO attempted to quantify the impact of the factors affecting food imports in these countries. The results of this study are revealing, as much for what they do not show as for what they do. They confirm the importance of changes in import prices, purchasing capacity of exports and, more than any other factor, domestic production performances, in determining food imports. The results, however, did not always conform to expectations or, in spite of the use of sophisticated econometric techniques, did so leaving margins of uncertainty. In addition, such important factors as changes in per caput incomes and foreign exchange reserves did not appear to influence food imports. These sometimes puzzling or inconclusive results indicate the existence of other factors and influences that require further case-by-case investigation, but they also point to another general conclusion. It has been observed that many FDCs purchase far more food than their export sector can afford. This is so because the external account gap is covered primarily through financial or food assistance, remittances and other, more or less reliable, sources of income. Thus, it is hardly surprising that in many cases food imports should not depend on export performances, in the short term at least. Large but irregular flows of resources from non-export origins may also help to explain why food imports should not appear to react much to such fundamentals as changes in import prices, domestic income or international reserves.

Such situations present obvious risks. The sustainability of non-trade forms of financing can be questioned to the extent that they largely depend on discretionary donor dispositions or other similarly uncertain sources of financing. If allowed to continue, the combination of high and growing dependence on food imports and precarious or unreliable sources of financing may create explosive situations.

The variable that appears to have the greatest impact on food imports in FDCs is domestic food production. This unsurprising finding has important implications. On the one hand, it suggests that these countries have found the resources to import more food in periods of domestic production shortfalls, even though international market conditions or other variables affecting import decisions may have been negative. This is, again, largely related to the availability of external assistance in various forms. On the other hand, positive changes in domestic supply conditions resulted in commensurate reductions in food purchases, underlining the importance of domestic agriculture in alleviating external welfare and import dependence.

Agricultural specialization has not enabled FDCs to develop a strong agricultural export base or expand their domestic food supply adequately. Yet, for many FDCs there are currently few alternatives to agriculture-based development, and it is likely to be a long time before these countries are able to diversify their overall economic structures significantly and gain competitiveness in other economic sectors. Conversely, several FDCs may have a considerable potential for developing agriculture efficiently and thus reducing import dependence and/or enhancing export revenues. The observed move towards diversification of agricultural exports in many FDCs is an encouraging feature in this context. The most effective way of achieving food security, financial autonomy and maximization of developmental opportunities may, therefore, still be to enhance the productive potential of domestic agriculture.

POLICY REFORM AND THE CONSUMER

Since the early 1980s, policy reforms initiated in many countries have been biased in favour of greater market orientation and a more open economy. Such reforms have had a substantial impact on food consumers, by directly and indirectly affecting the factors that determine food demand, as well as on food producers because of changes in agricultural policies and policy measures. The following discussion examines how and why these policy reforms have come about, and the effects they are expected to have had on consumers.

Most, although certainly not all, analyses of the effects of policy reform in the economic literature have attempted to quantify the impacts of the agricultural market and trade liberalization on the producer and on the net welfare of the country as a whole. Rather less attention has been given to the taxpayer and particularly little to the consumer. As a result, this review presents a conceptual overview that is supported to the extent possible by the limited empirical findings available.

Food consumers in the policy process

Why should the food consumer have received comparatively so little attention? In the first place because, to the extent that macroeconomists analyse the impacts of macropolicy measures on real sectors of the economy, including agriculture, they tend to be concerned with whether the country as a whole gains or loses. It is often not politically astute to expose the gains (far less the losses) of individual groups and, indeed, the necessary data are often not available in the quantity and quality required to perform such analyses. Furthermore, much of the economic analysis is driven by or directed towards ministries of agriculture and trade which, by and large, consider their clientele to be respectively agricultural producers and exporters (including agricultural exporters).

Another reason is that although - or perhaps because - every person in society is a food consumer, food consumers for the most part are not generally organized to bring systematic political pressure to bear on policies that run counter to their best interests. An increasing number of countries have consumer associations that focus on providing better information to consumers, directing their lobbying power accordingly. At the same time, many of the higher-income countries have seen the formation of consumer advocacy groups around specific issues such as tighter meat and slaughterhouse inspection, banning the use of growth hormones, increasing the nutritional levels in feeding programmes or the more general issue of food safety. Given that, as income increases the proportion of household income spent on food declines and consumption patterns shift in favour of preferred higher value items in terms of both perceived quality (for example, paying a premium for meat and eggs produced in more "humane" production systems) and quantity, the interests of different income groups diverge and, arguably, consumer groups tend to reflect disproportionately the concerns of the better-off and more politically articulate consumers.

In contrast, the issues that have, usually informally, united consumers (especially urban consumers) in the lower-income countries have been food shortages, high food prices and sudden large increases in food prices. Consumers' concerns have been manifested in demonstrations or sometimes riots and governments have occasionally even fallen as a result. Many developing countries have therefore had policies to subsidize food prices for urban consumers. Since the ministries of agriculture and trade tend to be the focus of any consumer unrest, who better at least to demonstrate concern about the impacts of the policies they promulgate on all affected groups, including the consumer?

Whither policy reform?

In the developing countries, government policies have led to pervasive price distortions and hence misallocation of resources. Vollrath reports that empirical findings indicate that "the trade, macroeconomic and sector-specific pricing policies adopted in the developing countries since the early 1950s have given rise to the following incentive biases: against the production of tradable goods and in favour of non-tradables; within the tradable goods sector, against exports compared with import-competing goods; within the export sector, against agricultural products compared with manufactured goods; and within agriculture, against export compared with food crops". The correction of these policy distortions has necessitated painful adjustments.

Pressures for policy reform during the past 15 years have four main sources. First, a large number of countries that had taken on very high levels of external debt in the 1 970s suddenly found, in the early 1 980s, that their economic circumstances were reversed and they could no longer sustain their heavy debt burdens. Belt-tightening and policy reform became the order of the day. Such structural adjustment usually called for: the reduction of government spending; the reduction of government intervention in or its complete withdrawal from agricultural input and output markets; the phasing-out of subsidies; the privatization of the functions of former parastatals; and the devaluation of overvalued exchange rates, i.e. in general, "getting the prices right".

Second, beginning in the first half of the 1 980s, a number of countries, particularly in the developed world, found that the budgetary costs of supporting their farmers in the style to which they had become accustomed had become untenable. Much of the policy reform that has taken place in these countries has been prompted by the desire to limit government budget exposure. Reform has consisted of more precise definition and directing of aid to beneficiary groups, lower support levels, decoupling support payments from current production and reducing the presence of the government.

Third, the late 1980s and early 1990s saw the breakdown of the command and control economies of Central and Eastern Europe and the republics of the former USSR and the beginning of a difficult and painful process of transition towards more open market-oriented economies. In the early stages, the transition process disrupted traditional trading patterns, drastically reduced real per caput incomes, reduced demand and disrupted production, processing and distribution. It is only recently that some countries of Central and Eastern Europe have begun to recover from this macroeconomic and sector-specific pricing policies adopted in the developing countries since the early 1950s have given rise to the following incentive biases: against the production of tradable goods and in favour of non-tradables; within the tradable goods sector, against exports compared with import-competing goods; within the export sector, against agricultural products compared with manufactured goods; and within agriculture, against export compared with food crops. The correction of these policy distortions has necessitated painful adjustments.

Pressures for policy reform during the past 15 years have four main sources. First, a large number of countries that had taken on very high levels of external debt in the 1 970s suddenly found, in the early 1 980s, that their economic circumstances were reversed and they could no longer sustain their heavy debt burdens. Belt-tightening and policy reform became the order of the day. Such structural adjustment usually called for: the reduction of government spending; the reduction of government intervention in or its complete withdrawal from agricultural input and output markets; the phasing-out of subsidies; the privatization of the functions of former parastatals; and the devaluation of overvalued exchange rates, i.e. in general, "getting the prices right".

Second, beginning in the first half of the 1 980s, a number of countries, particularly in the developed world, found that the budgetary costs of supporting their farmers in the style to which they had become accustomed had become untenable. Much of the policy reform that has taken place in these countries has been prompted by the desire to limit government budget exposure. Reform has consisted of more precise definition and directing of aid to beneficiary groups, lower support levels, decoupling support payments from current production and reducing the presence of the government.

Third, the late 1 980s and early 1 990s saw the breakdown of the command and control economies of Central and Eastern Europe and the republics of the former USSR and the beginning of a difficult and painful process of transition towards more open market-oriented economies. In the early stages, the transition process disrupted traditional trading patterns, drastically reduced real per caput incomes, reduced demand and disrupted production, processing and distribution. It is only recently that some countries of Central and Eastern Europe have begun to recover from this initial plunge in economic well-being. As part of the transition process, even the most basic legal and economic institutions had to be created to allow and encourage the private sector to operate as part of a functioning market economy.

Finally, even countries that do not fall into any of the above categories have been caught up in the general moves to reduce the presence of government in those areas where the private sector can perform better and more efficiently and to liberalize markets and trade.

The last 15 years have seen a serious questioning of the very role of government in a market-driven economy. From this questioning and from necessity, policy reform in the direction of open market-orientation has taken place to varying degrees in almost every country. An important outcome is that food prices are increasingly being determined by market forces, i.e. by the interaction of the relationship between demand and supply.

The effects of policy reform on food consumers depend, therefore, on the extent to which the policy measures abandoned and those adopted affect the determinants of demand and of consumer supply. It is, however, difficult to predict the probable outcomes of changes in individual variables because the complex of policy reforms under structural adjustment programmes, in the transition economies and as a result of the GATT agreement (and any combination of the above), have a variety of effects on both supply and demand functions. The overall outcome for consumer prices and total quantity purchased will depend on the direction and magnitude of the shifts in the demand and supply curves and also on the price elasticities of demand and supply.

In addition, the adjustment period following the introduction of major policy reforms can be long. To free prices so that they can be determined by market forces means the "marketization" of all aspects of the production and marketing system. Experience has shown that, initially, liberalization and privatization of the processing and distribution sectors often result in serious slowdowns of production and little competitive challenge. The reason for the former is that the withdrawal of the state from these parts of the marketing chain can rarely be substituted immediately by hitherto underdeveloped private institutions. The lack of competitive challenge comes when a government or parastatal monopoly is replaced by a private one - such a change in ownership merely changes the recipient of monopoly profits. The outcome is that during the adjustment period the transmission of price signals through the marketing chain is severely hindered in both directions. In other words, the signals about consumer demand that are sent at the retail level are not being properly transmitted through the intermediate levels of processing, distribution and wholesaling to farmers so that they can alter their production decisions accordingly. Neither are the signals about the supply of agricultural and food products, whether they originate with the farmer or elsewhere in the marketing system, being adequately sent through the marketing chain to the consumer. Consequently, the pressures to reduce consumer prices and respond to changing consumer demand through increases in the efficiency and effectiveness of marketing may well, in the short to medium term, have no effect on the retail price increases caused by subsidy removal. Many governments have yet to put appropriate policies in place to address these problems.

In fact, the whole question of consumer supply - i.e. the relationship between quantities of different food items supplied at the retail level and retail prices - is particularly complex because it depends on a whole range of variables among which policy measures tend to figure prominently. Macroeconomic policies (especially those relating to the exchange rate and exchange controls), trade policies, fiscal policies, agricultural support measures (and the way in which they are implemented) and any other policies that affect the development of the food processing and distribution industries all affect consumer supply (i.e. supply at the retail level). Furthermore, the extent to which the retail supply of food depends on the supply of domestically produced agricultural and processed products is extremely variable both across countries and within the same country at different times and for different items. This degree of complexity means that, even if one could be reasonably sure about the effects of specific policy reforms on the domestic agricultural supply function, the relationship between farmgate and consumer supply is too dependent on such a wide range of other policy measures to be determined a priori.

Food consumption and the demand for food

Farmers produce food because consumers purchase it. This may seem obvious, but it is a perspective that tends to be overlooked. High priority is given to all aspects of food and agricultural production while relatively little attention is paid to the final users. Part of the reason probably lies in a frequent failure to distinguish between consumption and demand.

Consumption refers only to a physical process and can therefore be measured in physical units. In contrast, demand is an economic concept; a demand function describes the relationship between the price of a commodity and the quantity demanded at each price, all other things being equal. Changes in consumption can be caused either by movements along a given demand curve - a response to changing prices - or by shifts in a demand curve, i.e. at any given price a different quantity is now demanded because of such factors as changing income levels (see Box 2). The demand for food at the consumer level leads, by way of the demand for processed foods and added "marketing" services (primary and secondary processing, packaging and distribution), to a derived demand for agricultural products at the farm level - and that is the demand function faced by the farmer, or would be if government policies did not distort the market-signalling mechanism.

A number of the policy reforms that have been implemented have directly affected the total demand for food and the composition of the food basket purchased, through their effects on household incomes and food prices, both absolute and relative. Structural adjustment programmes of the traditional type have generally led to decreasing real wages in the short to medium term, often accompanied by higher unemployment and therefore fewer wage-earners per household. This phenomenon has also occurred in the countries in transition from centrally planned to market economies. The declines in real household income have been substantial and the speed of recovery can be very slow.

A number of policy reforms are likely to affect consumer prices of food items. The removal of food subsidies increases food prices. In some countries, only imported staple foods, largely purchased by urban consumers, were subsidized. In other countries, the effective consumer food subsidy arose from massive subsidies to the agricultural sector and the removal of these subsidies has increased the prices of all food items for rural and urban consumers. Overvalued exchange rates impose a tax on exports and subsidize imports. Correcting an overvalued exchange rate will cause the price of tradable goods to rise relative to the price of non-tradables. Exports become cheaper in the foreign currency so quantity demanded in the export market will increase, putting upward pressure on the domestic price of the export good. Similarly, the domestic price of import goods will rise, thus curtailing the domestic quantity demanded of imported items. Thus the prices of tradable food items will rise whether they are exportable or imports. Some countries have used direct agricultural export taxes to curtail exports and to maintain lower domestic agricultural prices, thereby, provided these lower prices are passed on, lowering consumer prices. Removal of such export taxes will increase consumer food prices. In contrast, a reduction in import tariffs and relaxations of import controls will tend to exert a downward pressure on prices.

A fall in monetary and real incomes shifts the demand curve so that at any given price less food is now purchased. For those countries that had been subsidizing urban or indeed all consumer prices, the removal of such subsidies and other policy reforms that lead to increased food prices also reduce the quantity purchased. Both declines in real income and changes in relative prices (not only between different food items but also between food and nonfood items) lead to changes in the composition of the household food basket.

BOX 2
FACTORS AFFECTING THE DEMAND FOR FOOD

The demand for a food commodity is a function of several variables: the price of the item in question, the prices of complementary and substitutable items, income, demographic variables and tastes or preferences. In the short to medium term, the major determinants are prices and incomes and these are the most likely to be immediately affected by changes in government policies. In this context, it should be noted that a change in the price of a commodity has two effects, the income effect and the substitution effect. The substitution effect is such that a negative change in the price of the commodity always leads to a positive change in the quantity demanded. The income effect, however, depends on whether the commodity is "normal" or not. If it is normal, the positive change in income implicit in the fall of the commodity's price will lead to a positive change in the quantity demanded and therefore reinforce the substitution effect. If the commodity is "inferior", the income effect will be negative and therefore partially offset the substitution effect since it will work in the opposite direction. However, for an inferior good, the overall effect is still that a price fall (or rise) will lead to an increase (or decrease) in the quantity demanded. In contrast, the effect of a change in income, with no change in the price of the commodity, depends on whether the good is normal - in which case an increase in income increases the quantity demanded - or inferior - an increase in income decreases the quantity demanded.

The household's demand for different food items also depends on a variety of demographic factors, including the number and age of household members and the age of the principal food purchaser. The age of household members affects demand in two ways. First, children and elderly people eat less on average. Second, children have different consumption patterns from adults. The effects of age of the principal food purchaser might purely be a reflection of changing needs over a lifetime, irrespective of when that lifetime began, but there could also be a "vintage" effect, which refers to the changing tastes of each generation of food purchasers. In addition, the size of the household may exert an independent effect on demand the "household scale effect".

Tastes or preferences include such factors as seasonal variations in consumption patterns for reasons other than seasonal price variability, religious and social taboos and lack of familiarity with particular foods.

Associated with demand functions are a number of different "elasticity" measures, each of which shows the responsiveness of demand to changes in a particular variable. An elasticity coefficient can be interpreted as the percentage change in quantity demanded in response to a 1 percent change in the relevant variable all other things being equal. The most important elasticity coefficients are:

The cross-price elasticity can be positive or negative, depending on whether the two commodities are substitutes or complements:

There are two measures of income elasticity: the income elasticity of expenditure on the commodity and the income elasticity of quantity purchased. Strictly speaking, these measures should be identical if the commodity is precisely defined since they are calculated with all the other variables assumed constant. However, this is rarely the case in practice. Take, for example, rice. Evidence from countries in West Africa and Asia suggests that the local varieties of rice command a premium price over imported varieties. Thus the total expenditure on rice could increase as real incomes rise without necessarily increasing the total quantity purchased. Conversely, the total quantity purchased could increase without an increase in total expenditure as households suffering a decline in real incomes substitute the cheaper commodity for the more expensive.

The empirical evidence strongly suggests that, in countries of any income level, lower-income households have higher price and income elasticities of demand for food, which means that the poorer groups will be more severely affected in terms of the total quantity of food purchased and the nutritional quality. The impact will be exacerbated if lower-income households face higher unit food prices than do higher-income households. This can occur if, for example, the household has insufficient income reserves to be able to buy in bulk or if transport costs to cheaper retail outlets are high. The consumer responds to reduced incomes and higher prices by trying to maintain food intake through increasing the proportion of household income spent on food and altering consumption patterns in favour of relatively cheaper commodities. In addition, attempts will be made to augment supplies through interhousehold transfers (e.g. obtaining food from rural relatives) and where possible increasing own production in the urban and pert-urban areas, which can in certain circumstances lead to environmental and health problems.

Policy reform and food prices

How great an impact policy reform has on consumer prices depends critically on the economic and crop-specific situation at the time of price liberalization and on the speed with which reforms are implemented. Reusse noted that the countries that had implemented major reforms in terms of desubsidization, deregulation and exchange-rate adjustment in the period 1985-87 had done so in favourable circumstances, "characterized by declining dollar and interest rates, falling petroleum prices and low international price levels for staple foods, especially grains and livestock products, together with good domestic crop results. The combination of these factors kept the immediate effect on urban consumer prices moderate." Whether countries undergoing reform since 1987 have done so in relatively less favourable conditions is a question that can only be answered by carrying out comparative analytical studies. Certainly the effects of policy reforms in the countries in transition from centrally planned economies have had a substantial negative impact on consumer prices, but in those countries the effective consumer subsidy arose from heavy subsidies to agriculture as a whole rather than by specific subsidies on imported staples.

To the extent that policy reforms in certain countries have succeeded in their objective of shifting the burden of agricultural support from the consumer to the taxpayer, one would expect lower-income consumers to have benefited more than higher-income groups from any resulting price falls because of the progressivity of income tax. However, if the costs are partly being met through the imposition of a value-added tax on food, the effect is highly regressive and hits hardest the poorest sections of the community. It is highly questionable how far consumer prices have actually fallen in response to such policy reforms. There seems to be a degree of "stickiness" in retail food prices - except, perhaps, for highly seasonal fresh produce such as fruit and vegetables which may reflect the use of market power by processors, distributors or retailers. Furthermore, the proportion of the farmgate price of an agricultural commodity in the retail price of a food item is lower the more value is added in the marketing system. For example, in the United States the value of the wheat in a loaf of bread that costs US$1.20 is about 6 cents ($0.06). The recent reductions in support prices of changes in the Common Agricultural Policy of the EC have yet to be felt through noticeable changes in consumer prices in the member states. The effects of entry for new member countries will depend on the level and form of existing agricultural support. Finland and Austria, for example, which had much higher levels of agricultural protection, expected to see quite substantial reductions in consumer prices as a result of EU membership. On the other hand, the consumers of the Central and Eastern European economies in transition have already had to adjust to the removal or reduction of food subsidies: EU membership for them will lead to yet higher food prices on entry and, in addition, food may be subject to value-added tax.

The inclination of governments to focus attention on their vociferous urban consumer constituency has tended to obscure the effects of policy reform on the agricultural household as food consumer as well as food producer. A large proportion of the population in most developing countries, including many of the poor and malnourished, resides in semi-subsistence farm households. The way in which such households respond to changes in food prices is rather less straightforward. A widely used tool of empirical analysis is an agricultural household model that integrates the farm family's production and consumption decisions. When a food commodity is produced partly for consumption and partly for sale, the price increase could have a positive impact on farm profits and hence household income so that the effect of price increases is completely offset. Thus if the profit effect is large and the income elasticity of demand for the commodity high, a price increase could lead the farm household to consume more of the product. The implication for nutrition is that higher food prices may actually lead to an improvement in the nutrition of farm household members, because of the effects of profits on income. Even if the consumption of the commodity for which the price has increased declines, the increased profits and income can be used to buy increased amounts of other foodstuffs, the result being an improvement in nutrient intake.... Lower food prices raise the welfare and probably the nutrition of urban households, at least in the short run. However, higher prices may improve the welfare and nutrition of agricultural households, and possibly even of rural non-farm families whose income depends on agriculture, such as agricultural labourers.

Protecting vulnerable groups of consumers

In order to alleviate the negative impacts of policy reforms, including the removal of general food subsidies, on the poorest and most vulnerable sections of society, a number of governments have tried to direct support more effectively to poorer consumers. Broadly speaking, support can be directed to a specific geographical area or income level or by using a "self-targeting" commodity (discussed below). However, major problems with such support systems - be they through subsidies such as food stamps or through some form of rationing - are increased administrative and information costs and incentives to abuse the system. In a poorly designed system, the budgetary costs of support could be as high as a general subsidy or, if substantially lower, could well mean that an unacceptably high proportion of the target group is not being assisted. Whether this is in fact true or not requires rigorous assessment on a case-by-case basis which, by and large, seems not to have been carried out.

There is a further serious problem with such support schemes and that is the very weak information base on which decisions about how to design them are being made. This is far more complex than the recurrent complaint by analysts about the scarcity of reliable data in developing countries. Problems have been identified with the methodologies used and also with the assumptions about the sort of data that are needed to yield results that are accurate enough to be of real use to policy-makers.

Recent work in Kenya and the Philippines investigated the reliability of food quantity information collected from household expenditure surveys. The conclusions are that the food expenditure data on which most demand analysis has rested systematically overestimate the income elasticities for food staples. There are two main reasons for this upward bias. First, food quantities or food expenditures are not measured independently of income if, as is often the case, total expenditures are used as the proxy for incomes. (Indeed, for very low-income consumers whose total expenditure on food is typically as much as two-thirds of total income, this is intuitively a reasonable approach.) Second, food transfers from high-income to low-income households are underrecorded. Such transfers occur, for example, as guest and hired worker meals.

The policy consequences of what might at first sight appear to be a minor technical question are profound. For food staples projections of aggregate demand that are based on assumptions of aggregate income growth and use such upwardly biased income elasticities will overstate the actual requirements. Furthermore, the benefits of various income-generating policies on household food security have consequently also been overstated. It is therefore probable that insufficient policy emphasis has been given to intrahousehold issues, micronutrient consumption and non-nutrient health issues. The use of 24-hour food recall data that detail food intake rather than availability seems likely to give unbiased estimates of food demand and hence provide better information to policy-makers.

Intrahousehold issues undoubtedly require further attention if vulnerable groups are to receive effective protection. One issue is the allocation of food among the family members, which, if inappropriate, can exacerbate the effects of an inadequate household food supply or leave some individuals malnourished even when the total food availability is apparently adequate. Studies show substantial variations of intrahousehold food allocation from country to country. In some cases, the allocation pattern discriminates against women and girls, while in others the bias is in favour of adults over children. In yet others, there is both a sex and age bias, with the male household head receiving a disproportionate share. If this type of bias is detected from studies of intrahousehold food allocation patterns, it might be possible to use the information to design special intervention programmes aimed at protecting the vulnerable groups of consumers so identified. Such special intervention programmes could include supplementary feeding to certain groups, such as pregnant and lactating women and small children, or the provision of school meals or milk either to all children of school age or perhaps just at girls' schools.

A related issue is the decision-making process within the household which is often correlated with the generation of income. The evidence suggests that the more control women have over income, the better is the nutritional status, particularly of children. Thus concern has been expressed about policy measures and reforms that encourage the commercialization of smallholder agriculture and the production of cash crops because men tend to take control over the income so generated whereas in traditional agriculture, especially in Africa, women have exerted a great deal of control over the food produced for home consumption as well as the cash from any sales. This would tend to suggest that even if the benefits of income-generating policies on household food security have been overstated in general, policy measures that can assist women to gain control over some cash income are likely to be of some benefit in improving child nutrition and health. Possible policy measures to this end could include programmes to support the partial commercialization of activities that are more likely to be carried out by women, such as soap-making and kitchen gardening, subject always to there being an adequate demand for such products.

Increasing attention is being given to introducing "self-targeting" subsidies and it is in this area that inadequate information can have potentially serious consequences. The idea behind the self-targeting subsidy is that the subsidy should be attached to a commodity that is consumed disproportionately or solely by low-income groups. In order to identify an appropriate candidate for self-targeting, it is necessary to disaggregate food consumption parameters to obtain estimates of own-price, cross-price and income elasticities for the different commodities and for different income groups. Presumably, the disaggregation ought also to differentiate between urban and rural consumers as they have greatly differing consumption patterns. Of particular interest has been the identification of "inferior" food items. In this context, the word inferior is a technical term that refers to a commodity for which the quantity consumed falls as income increases and which is therefore purchased mostly by lower-income groups. Subsidizing such a commodity would cause its consumption to rise, thereby improving the calorific intake of the poorest consumers and, because the fall in the price of a commodity also has the effect of increasing the consumer's income, the subsidy would enable more of other, preferred food items to be purchased.

A review of attempts to disaggregate food consumption parameters as a basis for designing directed nutritional interventions was carried out in 1985. The findings demonstrate a clear need to improve and standardize survey techniques and then to standardize methodologies and models. The conclusion is that the disaggregation approach could provide useful information to policy-makers, allowing them to design nutrition programmes that would more effectively direct a government's limited resources to the poor. One might also expect that such information could better identify who is really "poor". However, the approach would not necessarily lead to the identification of a suitable candidate for a self-targeting subsidy. After all the expense and effort of determining disaggregated consumption parameters, it is possible that no ideal commodity may be found. For example, in Brazil no commodity was found that is eaten by the poor and not by the rich. Although some commodities, such as low-quality rice, were found to make better vehicles than others, targeting nutritional subsidies in Brazil is far from ideal.

If a commodity is identified but is insufficiently inferior, then the costs of the subsidy could turn out to be extremely high. The point is also made that even if a suitable inferior commodity can be identified, subsidizing it will not by itself solve all of a country's nutritional problems. Thus a self-targeting scheme based on one commodity could at best form part of a larger strategy that could include a whole variety of additional measures, such as special intervention programmes, buffer stock schemes and food stamps (or some other form of income transfer), to improve household food security and nutrition.

Zambia attempted to introduce a self-targeting subsidy for maize in late 1986 by subsidizing only the relatively inferior roller meal. At the same time, breakfast meal, which was the preferred form, was to be sold at an economic price, which involved a price increase of 120 percent. The scheme went badly wrong, leading to civil disturbances, and had to be abandoned. It has been argued that the problems lay not with the scheme itself so much as in the manner of its introduction. It appears to have been instituted without sufficient attention to problems of supply and the relative demands for the different grades of maize-meal which would ensue in the context of the new price structure. As a result the rapid disappearance of roller meal from the retail outlets, and the sharp rise in the price of breakfast meal, meant that the majority of consumers were faced with a de facto overnight abolition of the maize-meal subsidy, with no opportunity to adapt their consumption patterns.

Given this turn of events, Zambia adopted a different approach by first introducing a rationing programme restricted to urban consumers and later supplying coupons only to eligible households, eligibility being determined by household income level for those in formal employment while, for those not in formal employment, the number of dependents for whom coupons can be claimed was limited.

The degree to which any country has ever really been successful at directing support has been called into question. A review in 1988 looked at ten countries with large food subsidy programmes that had been attempting major policy reform. It found that, in the eight years since 1980, none had been able to increase the degree to which food subsidies were directed to the absolute poor; in addition, there were no indications that the efficiency of the programmes had been improved either. However, seven years on, a thorough analysis of the results in some of the countries that have experimented more or less successfully with directed consumer subsidy schemes could provide useful guidance in designing systems that are both more efficient and more effective under the conditions in which policy reform is now being implemented.

In conclusion

Policy reforms affect consumers directly because they have an impact on the factors that affect food demand, which in the short to medium term are largely the absolute and relative prices of different food items and household incomes. Whatever the problems with data and methodology, there is substantial evidence that in any country the lower-income groups are more sensitive to changes in these variables and that the very poor (however poverty is defined) are particularly vulnerable in the short and medium term to changes that are expected to be of benefit to them in the long term. However, the true responses of the most vulnerable groups of consumers to changes in household incomes and food prices are very far from being intuitively obvious. By carefully analysing consumer behaviour that reflects the differing patterns of demand response in lower-income households, governments will be able to design and implement policy interventions that can go a long way towards protecting those most affected by broader policy reforms. What such policy interventions should be depends on a wide range of factors, including how much a particular government can afford. So far no final conclusions have been reached with discussions being able to indicate only those areas where policy measures might be used without jeopardizing the ultimate objectives of the reform programme. Another point to be made is that, while protecting the lowest income groups may justifyably be a high priority in policy terms, governments should recognize that "vulnerable" groups of consumers are not necessarily restricted only to the poorest households. Factors that are not easy to deal with in the framework of consumer demand analysis, such as intrahousehold allocation of food resources and decision-making power, need much more study, as do the differences between urban and rural consumers and the consumption responses of households that are simultaneously food consumers and food producers.

Malnutrition is not only a problem for the poor, although they are undoubtedly most at risk. Changes in tastes, in preferences for different food items and in the food basket purchased have normally been evolutionary rather than revolutionary. However, it may be that radical policy reforms change relative food prices, as well as reduce incomes, so quickly and drastically that the resulting disruption of traditional consumption patterns gives governments sufficient cause for concern to justify intervention.

Identifying issues and setting out a conceptual framework is only the first step, however. Countries that are undergoing or planning to undergo major policy reform need to improve their knowledge and understanding of the effects of such reform on consumers, especially on vulnerable groups, whose prior identification must not be taken for granted. "The generation of such information should be pursued in two ways: i) analyses to improve current understanding of how specific policies frequently found in adjustment programmes and combinations of such policies affect incomes, food consumption, nutritional status and health of specific groups of low-income people; and ii) continuous monitoring and surveillance of changes in the welfare of various groups of low-income people using indicators such as nutritional status, mortality, household food sufficiency, incomes and prices during periods of adjustment."" International agencies such as FAO and UNICEF, in collaboration with research institutes and universities, could play an important role in assisting governments to generate and analyse the necessary information.


Contents - Previous - Next