Contents - Next

This is the old United Nations University website. Visit the new site at

Hunger, health and society

National agricultural policies and world hunger
Functional consequences of marginal malnutrition among agricultural workers in Guatemala
Social, economic, health, and environmental determinants of nutritional status
Ethnographic studies of the effects of food availability and infant feeding practices
The duration of breast-feeding adequacy in a rural area of Bangladesh
The need and rationale for monitoring and evaluation of nutrition programmes

National agricultural policies and world hunger

Miguel Urrutia
Development Studies Division, The United Nations University, Tokyo, Japan


The Rome Declaration on Hunger states that it is indeed possible to end world hunger by the year 2000 (1). It would seem that there is a fairly wide consensus that such an objective is feasible in the sense that technically it should be possible to produce sufficient food for six billion people. However, unless major policy innovations take place in many countries, it is likely that an unacceptably large proportion of the world population will not have access to an adequate diet in the foreseeable future.

An FAO background paper on the world food situation states that, in the last analysis, the reasons such limited progress has been made up to now towards abolishing hunger lie in a failure of political will in both developing and developed countries. The report then goes on to say that declarations and resolutions have to be translated into concrete policies if they are not to remain mere words, and that the only option finally open to political leaders is to turn to such policies as agrarian reform and redistribution of purchasing power to procure food, but these, so far, have been more often postponed or avoided than tried (2).

I believe that international agencies too often blame lack of political will on the part of national governments for postponement of the solution to pressing global problems, as if the technocrats in national and international agencies had available the concrete recommendations for overcoming such problems. This is not realistic, Quite a few countries have tried land reform, and while in some countries hunger has diminished (although if not accompanied by rapid industrialization, it has seldom been eliminated), in many countries that have carried out land reforms generally applauded by international experts, malnutrition has not been reduced. Figueroa has shown, for example, how inadequate agricultural pricing policies in Peru more than cancelled out the positive income distribution effect of land reform in the 1960s (3). In some cases, a well executed land reform may indeed diminish malnutrition and in the medium-run increase food production, but quite often attempts at land reform are inadequately thought out and can have negative effects on food supplies.

It is easy to talk about redistribution of purchasing power, but it is very difficult to design effective policies to bring about such redistribution. The International Food Policy Research Institute has analysed a series of national food subsidy schemes that have been meant to redistribute purchasing power in favour of malnourished families, and the conclusion was that such subsidies often do not decrease malnutrition among the very poor. A case in point is the analysis of Brazilian wheat subsidies, in which Cheryl Williamson Gray concludes that not only is actual calorie intake likely to decline when wheat prices fall because of the substitution of wheat for higher calorie foods, but calorie intake may well fall farthest for the most severely malnourished sector of the urban population (4).

In many countries, policy-makers cynically maintain policies that benefit politically powerful groups and harm the nutritionally vulnerable, but there are also cases of governments genuinely concerned with the welfare of underprivileged groups. My conviction, based in part on personal experience, is that when such governments want to embark on ambitious attempts to combat hunger, they cannot find concrete and detailed blueprints for action. They can find only rather general recommendations, with little detail as to how they can actually be implemented and at what cost.

There is also agreement now that an effective attack on the hunger problem must include co-ordinated action in many fields, but for that very reason it is imperative that researchers present policymakers with detailed policy recommendations that can lead to both greater food production and adequate levels of food consumption by poor families. No one policy will achieve both objectives. This ideal can only be realized on the basis of a rather detailed analysis of food production functions, consumption, and cultural practices with respect to food for urban and rural families of different income levels, and the role of agricultural production in balance of payments and management of inflation. This type of general equilibrium analysis is always difficult, and it has not been carried out in many nations. If a real effort is not made to clarify policy options, it is likely that an excessive proportion of the word's six billion people will continue to be hungry at the turn of the century.


Some countries have been able quite successfully to promote rapid increases in agricultural production. The Philippines, Kenya, and India are cases in point. But, as Amartya Sen has said, "India's socalled 'self-sufficiency' in food goes hand in hand with keeping a quarter-perhaps a third-of the rural population . . . suffering from regular hunger and systematic malnourishment" (5). Unless food production strategies are co-ordinated with policies to give the poor access to food, adequate nutrition will not be achieved.

At the same time, a developing country is unlikely to achieve and maintain adequate nutritional levels if its agricultural production is stagnating. Global food supplies might also be inadequate if a great number of developing countries have low growth rates in their agricultural sector. Unfortunately, this is at present the case

There are declining rates of per capita food production in sub-Saharan Africa (6). One and a quarter billion people live in countries where average food supplies per capita are below average energy requirements, and over one billion of these live in countries whose average supplies of calories per capita fell or stagnated in the 1970s (2, p. 4).

There is no one set of policies that will promote agricultural and food production in these countries, and specific strategies must be designed in each case. Some specific recommendations may be applicable to many local situations, including the necessity of strengthening agricultural research capabilities and extension services and of giving farmers access to credit facilities when they need them. All reports speak of improving marketing facilities, but this is an area where concrete proposals are hard to come by. But most important, food production can only increase if farm gate prices are attractive to the farmer. However, unfortunately, high food prices do not contribute to diminishing malnutrition.


The major difficulty faced by policy-makers in developing countries intent on alleviating hunger is the nature of price and income elasticities of demand for food. Average price and income elasticities of demand for food are quite low, so that any increase in food supplies might decrease the net income of farmers because of decreasing farm gate prices. (A low price elasticity of demand means that even if prices decrease substantially, people will not increase their consumption by much. An increase in food availability will, therefore, bring about relatively large price decreases and therefore a fall in incomes of producers. A low income elasticity of demand also means that, as family incomes increase, consumption of a commodity will not grow very much. Thus, increases in production will not be easily absorbed by the higher demand generated by national income growth.) Since a large number of the families below the poverty line are rural farmers, increasing production may lead to deteriorating income among the poor.

There are three ways out of this dilemma. The first is to export the increased production, but this assumes that it is always possible to export food crops. The reality is that the international market for many crops, from plantains to cassava, potatoes, and even rice, is quite badly organized, and most countries have agricultural self-sufficiency policies that make for thin international markets for these basic foods, with the exception of the major grains. Furthermore, food habits create area-specific demands for certain foods, and transport costs limit the export potential of other foods, such as potatoes. (The difficulty of exporting food products is one of the reasons modern sector farms in many developing countries switch to industrial crops that do have international demand and can be exported. In Latin America, for example, modern farms produce exportable crops such as cotton, soya, or sugar, while the traditional farmers are the ones who produce the food. In other words, modern agriculture tends to produce crops for which, because of international trade, it is not so obvious that local increases in production will lead to lower prices.)

In summary, except for grains, the export option is not realistic. For many food items, assuming a closed economy at the national or even the regional level might be more realistic. In that case, there are really only two ways out of the dilemma created by low price and income elasticities of demand for food.

The first of these is to achieve rapid productivity gains, which may lead to higher farm incomes despite decreasing real farm gate prices. In the case of subsistence farmers who consume a large part of their own production, the productivity gain will automatically be transformed into a welfare gain. Increases in productivity can be achieved through technological progress or through a decrease in the rural labour force, or both. Historically, income distribution has only started to improve and absolute poverty has begun to be eliminated when the agricultural labour force ceased to grow in absolute terms. The solution to rural poverty is therefore closely correlated to non agricultural employment growth and decreasing rates of demographic growth in rural areas. Achieving these two objectives is what development is all about, and it depends on the whole development strategy

The second way out of the dilemma is to take advantage of the fact that the low average price and income elasticities of demand are the result of almost unitary elasticities for the very poor and very low elasticities for the rich. (Unitary elasticities mean that a 1 per cent increase in income, or a 1 per cent decrease in price, generates a 1 per cent increase in consumption.) This statistical fact is quite logical. A poor family will spend most of any additional income on food, and decreases in food prices will increase its food purchases substantially. In the case of rich households, changes in income or food prices affect food consumption very little.

A developing country that wants to promote food production and at the same time increase the income of poor farmers can increase food demand by transferring food purchasing power to the poor, and thus avoid the negative nutrition and incentive effects of lower farm prices on poor rural families. Because small farmers produce in part for their own consumption, giving them capability for greater production will increase both the supply and demand for food, because one expects these families to have high demand elasticities for food. But any increase in marketed food production must be matched with purchasing power transfers to landless farmers and the urban poor in order to increase food demand and thereby avoid a substantial fall in farm prices.

These income transfers are hard to set up and administer. Targeting food subsidies or income transfers to landless rural workers is particularly difficult, but even in the urban areas food subsidies or food coupons often benefit the well-off more than the poor. Designing effective and efficient delivery systems is a real challenge both conceptually and administratively. Few countries can show successful subsidy systems. This is an area where policy analysis can make a real contribution to human welfare.

Realistically, the food subsidy or income transfer schemes must be quite well targeted in order to keep government expenditures on this component of development strategy at a reasonable level. Most counries today have a serious fiscal problem, and it is unreasonable to expect them to maintain massive government expenditures on food subsidies or income transfers for a long time in order to improve the nutritional status of their populations.

There is no question, however, that in countries that have adopted massive food transfer schemes, the social indicators show a very clear increase in human welfare. Countries like Sri Lanka and China have been able to reduce vastly the incidence of regular malnourishment and hunger The fact that the average Sri Lankan or Chinese now lives more than a decade and a half longer than the average Indian is probably largely the result of state programmes of food distribution in the first two countries. The cost of the programmes in both China and Sri Lanka, however, became a serious budget burden and may have had undesirable effects on economic growth (7). An attempt can be made, however, to design less costly and better targeted nutrition interventions.

On economic and humanitarian grounds, it is probably best to design systems for increasing the food intakes of pregnant and lactating mothers and children of less two years. These are usually the groups most vulnerable to malnutrition, and both within the family and in society at large they often have little bargaining power and tend to lose the battle for food. Pregnant and lactating mothers often have to drop out of the labour force or decrease their work effort, and this decreases their command over food at precisely the moment when malnutrition will have particularly harmful effects on them and their babies. Avoiding food deficiencies among these vulnerable groups is probably one of the best long-term investments a nation can make.

Food subsidies to these sectors of the population may not entail excessive costs. Entitlements such as coupons can be delivered by the health system when mothers go for pregnancy check-ups or when they bring their children for mandatory immunization. Delivery of entitlements (coupons, food stamps, etc.) is also an incentive for families to obtain immunization for their children, a result that will have deep health and nutrition effects. UNU research has shown that the combination of malnutrition and an infectious disease is often fatal for Third World chiIdren.

Food acceptable to babies and young children often will not be attractive to other family members, and this will avoid the leakages that can make nutrition subsidies so expensive. It may be that subsidies for food that is more expensive per calorie (some manufactured foods) may improve the nutrition of vulnerable groups at a lower total cost than subsidies for cheap calorie foods that nontarget groups will tend to consume. Manufacturing these foods might also promote decentralized agro-industries.

In summary, in countries where poor farmers produce a large proportion of the food consumed, an attempt to increase food production may have to be carried out in parallel with government subsidies and transfers that increase the food purchasing power of poor urban families and rural labourers. Unless this is done, increased production may affect food prices adversely, increase rural poverty, and lead to future decreases in food production. In this model, there is clearly no place for foreign food aid in countries that have a potential for increasing agricultural production. Such aid would be justified only if products not grown locally are used and if there is not a high elasticity of substitution between food aid products and locally grown food.

The desired set of food pricing policies and purchasing power transfers will depend on the specific situation of each country. The right price and distribution policy will depend on whether small or commercial farmers produce a certain commodity and on whether the poor rural producers are more or less numerous than the poor urban consumers. The specific consumption habits of a population will also have to be taken into account.

What I wish to convey is the conviction that policy-makers do not have available to them at present well thought-out strategies for eliminating malnutrition in their societies. Researchers must do the hard work and design such strategies in detail for each country, or at least for categories of countries with similar conditions. A recent survey of the literature by Sahn and Scrimshaw unearthed only one or two attempts at macro-food-policy planning within a general equilibrium framework (7, p. 13). Much work still needs to be done before we can blame the lack of progress in alleviating malnutrition exclusively on lack of political will.


An adequate level of food availability for the population in the year 2000 will also depend on the composition of food demand. Available agricultural land will meet a smaller proportion of total human energy demand if people in high-income countries demand diets with a high proportion of meat. Grain converted to meat loses 75 to 90 per cent in calories and 65 to 90 per cent of its protein (8).

Furthermore, for health reasons, governments might be encouraged to follow policies that discourage the consumption of some foods, including meat. Analysts believe that the low proportion of meat consumed in Japan because of its high price might help explain the longevity in that country. Currently, the average life span of men is 74.22 years and that of woman 79.66 years. According to Hajime Mizuno, a medical affairs commentator, a well-balanced diet is one of the chief causes of the high life expectancy in Japan.

Thus, the pressure by the United States for Japan to import more meat might not only contribute to decreasing the world supply of calories but also actually be harmful for the Japanese consumer. In the same way, developing countries could contemplate taxing meat consumption, particularly if meat is usually produced by large farmers and if cattle ranches have low ratios of man-years of labour per hectare.

Taxes and subsidies on certain food items can also improve nutrition. The already quoted research in Brazil has shown that in that country subsidies that will lower prices of rice benefit the poor much more than subsidies to decrease the prices of wheat or milk. Taxing soft drinks may also improve nutrition, particularly if tax revenues are spent on generating income transfers to nutritionally vulnerable persons.


There is no question, that, in order to eliminate hunger by the year 2000, national agricultural and food policies will have to be carefully designed. At the same time, analysis should be carried out of the impact of various food export and import policies on international agricultural prices. Once these policies are designed, it is crucial to be able to implement them. Research results will help create a constituency for better policies, but even if there is political support for the adoption of what appear to be rational policy options, it will still be necessary to reform old structures and create new institutions in order to execute these policies effectively.


1. "Rome Declaration on Hunger," adopted by the World Food Day Colloquium, Rome, 16 October 1982 (FAO, Rome).

2. Food and Agriculture Organization, "World Food Situation and Issues," background paper for the World Food Day Colloquium, Rome, 14-16 October 1982 (FAO, Rome), PP. 10-11.

3. A. Figueroa, ``Política de precios agropecuarios e ingresos rurales en el Peru," Coyuntura Economica (Fedesarrollo, Bogotá, Colombia), 10 (1): 125 (April 1980).

4. C. W. Gray, "Food Consumption Parameters for Brazil and Their Application to Food Policy," IFPRI Research Report 32 (lnternational Food Policy Research Institute, Washington, D.C., USA, 1982).

5. A. Sen, "Unequal Access to Food: Causes, Conflicts, Options," Future, Fourth Quarter 1982, p. 11.

6. J. W. Mellor, "Food and the Structure of Economic Growth: Its Relevance to North-South Relations" (International Food Policy Research Institute, Washington, D.C., USA, 1982).

7. D. E. Sahn and N. S. Scrimshaw, "Nutrition Interventions and the Process of Economic Development," Food and Nutr.. Bull., 5 (1): 2 (1983).

8. International Bank for Reconstruction and Development, World Devolopment Report (World Bank, Washington, D,C,, USA, 1982), p. 42.

Contents - Next