Contents - Previous - Next

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

Incorporating nutritional goals into the design of international agricultural research

Per Pinstrup-Andersen
International Food Policy Research Institute (IFPRI), Washington, D.C., USA


International agricultural research has been successful in facilitating rapid expansions in food production and productivity in many developing countries. Although some have questioned whether the poor have obtained their fair share of the benefits, it is now documented beyond reasonable doubt that urban as well as rural poor in general would have been considerably worse off without the "Green Revolution" (1). This is not to argue that research aimed at expanded food production and improved productivity has been or should be considered an ideal substitute for other policies to improve the lot of the poor. Increasing rural income along with more and lower cost food are only some of the elements of a successful strategy to improve living standards among the poor. Without other essential elements, particularly appropriate government policies, many of the poor may be by-passed and some may be worse off as a consequence of the changes resulting from agricultural research.

A review of a large number of studies of the equity effects of the Green Revolution clearly shows that existing distribution of ownership of productive resources, access to modern inputs, output market structure, and presence or absence of various related policies along with commodity priorities in research, technology characteristics, and other elements of research strategies are major determinants of the extent to which poor people gain or lose from the introduction of new agricultural technology (1). Thus, although agricultural production research per se should not be expected to eliminate an existing skewed income distribution and associated problems such as poverty and malnutrition, its impact (positive and negative) on the poor and their nutritional status can be very significant and is likely to depend to some extent on decisions made by the research community. It is for this reason that explicit consideration of nutrition implications in decision making on agricultural research is important to enhance positive effects and avoid negative ones.

Direct nutrition intervention such as feeding programmes for preschool children may be effective in reducing nutritional deficiencies. However, to be effective they are usually very costly. Furthermore, because they frequently do not become self-sustaining over time, their effectiveness may depend on a continued outlay of funds. Thus, while such programmes may provide a partial solution to the nutritional problem in the short-run, a long-term self sustaining solution must be sought through broader development efforts that will eventually reduce and, it is hoped, eliminate the needs for direct nutrition intervention programmes. Agricultural research and related policies offer great opportunities for such long-term nutritional improvements, principally through higher incomes to low-income producers and agricultural workers and/or lower unit costs of production and thus lower food prices to low-income consumers.

To realize these opportunities fully, nutritional issues must be explicitly considered when agricultural research is planned, and due consideration must be given to potential nutrition effects of alternative research priorities and technology characteristics. If this does not occur, positive nutrition effects may not be fully realized and negative effects may be overlooked or ignored. This does not suggest that nutritional goals should take priority over all other goals of agricultural research, but that they should be explicitly considered along with other goals. If the nutritional effects of alternative research strategies are estimated, it is possible that positive nutrition effects may be enhanced and negative ones reduced without unacceptable losses in the achievement of other goals.

Positive effects of modifications in research strategies may exceed the effects of direct intervention programmes while negative effects may cancel them. Thus, ignoring the potential and real nutrition impact of modifying agricultural research and related projects and policies while promoting direct nutrition intervention programmes makes little sense even for a short time.

The purpose of this paper is not to assess the nutritional impact of international agricultural research. Instead, the paper focuses on (i) identifying ways in which the nutritional impact of such research may be improved, and (ii) summarizing current efforts by the international agricultural research centres to consider the nutrition effects of their work.

The paper is divided into four parts. A brief overview of the linkages between agricultural research and human nutrition is presented first. Emphais is on identifying the factors that may be influenced by the research community to improve the effects on nutrition. Then follows a discussion of how the research community may incorporate nutritional concerns in its decision-making. The third section summarizes some of the current activities by the international agricultural research centres in this general area, and the last section proposes additional activities that the centres might wish to consider.


Agricultural production research influences human nutrition through its impact on:

• Incomes acquired by households at risk of having malnourished or undernourished members (here after called "malnourished households").
• The prices they have to pay for food commodities.
• The nature of the production systems among semi subsistence farmers.
• Risk and fluctuations in food production, storage, prices, and incomes.
• The nutrient composition of the foods available to malnourished households.
• Household income composition, intra-household income and budget control, and women's time allocation.

Although the reasons for existing protein-calorie deficiencies differ among countries and population groups, low household incomes, insufficient food availability, and high food prices are likely to be primary ones. Changes in any of these three factors are likely to influence food consumption. From a nutritional point of view, only changes in food consumption by households within which some or all members are currently malnourished, or where the risk of malnutrition is significant, are of interest. Thus, changes in food supplies affect the nutritional status only to the extent that the food consumption of malnourished or at risk individuals is affected. The degree to which expanded food production is translated into expanded food consumption by the malnourished varies greatly, depending on the crop or livestock species for which production is expanded, the nature of the technology that brought about the expansion, and who produces the increase. Thus, using total production expansion as a proxy for nutritional impact is likely to be misleading. What is important from a nutritional point of view is not how much more will be produced but how food consumption by the malnourished will be affected. Protein-calorie undernutrition frequently exists alongside a more or less plentiful food supply. But whether a plentiful food supply exists is irrelevant to households without access to adequate food (5).

Incomes and Food Prices

Changes in the incomes of households with malnourished members and the food prices facing these households influence their ability to obtain food and may also change relative cost of the various food commodities as well as the cost of food relative to other goods competing for the household budget.

Food consumption by the poor-particularly the urban poor and those rural poor who do not produce part or all of the food they consume-is very sensitive to changes in prices of individual food commodities. Thus, changes in food prices and their fluctuations over time are of particular interest from a nutritional point of view. High price levels and severe price fluctuations are much more harmful to the poor than to the better-off consumers. On the other hand, low price levels may have severe negative effects on the nutrition of the rural poor who depend on food production for their incomes whether they are producers or farm labourers. Furthermore, the long term nutrition effects of low food prices may be severe if future food production and employment are depressed.

The impact of expansions in food production on food prices, and, thus, on urban and some rural malnutrition depends on a number of factors, particularly foreign trade and price policies. Countries that shift from being net importers to net exporters may experience initial price reductions reflecting transaction costs. If free trade is maintained and domestic food prices reflect international ones, production expansions of traceable commodities are unlikely to result in significant price reductions except perhaps for short-term local price falls caused by ineffective marketing and transportation facilities. In those cases, the nutrition effect on the urban poor may be negligible. On the other hand, if domestic prices are insulated from international ones- as is frequently the case-and are permitted to adjust to reflect the additional domestic supply, then the nutritional status of the urban poor and those rural poor who do not derive their incomes from the production of the commodities experiencing price falls may be significantly improved because their real incomes increase and food becomes less expensive relative to other goods. The impact on the rural poor who derive their incomes from the production of such commodities will be positive in the former and negative in the latter case.

Whether the agricultural research community may influence foreign trade and price policies is not the issue. Rather, what is important in the context of this paper is that agricultural research may facilitate a reduction in the unit cost of production, thus generating an economic surplus that may be captured by the producer, landless labourer, the consumer-in the way of lower food prices- or shared in some proportion among the three groups. The distribution of this surplus will be determined largely by policies outside the reach of the research community, although research decisions regarding commodity priorities and technology characteristics will also have an impact. What is of an even greater importance from a nutrition perspective is the distribution of benefits between malnourished and well-nourished in each of the three sectors, i.e., consumers, laborers, and producers. Again, decisions on commodity priorities in research and technology characteristics along with policies are likely to influence the distribution of benefits and thus human nutrition.

The Production System, Risk and Fluctuations

A considerable proportion of existing malnutrition is found among households that produce most or all of the food they consume-the semi-subsistence farm households. For those households, changes in cropping systems that influence the amounts and kinds of food produced and the fluctuations in food availability during the year may be much more important than changes in food prices. Agricultural research may change crop mix and the cropping patterns to meet the nutritional requirements of the semi-subsistence farm family more satisfactorily. It may also result in a worsening of the nutritional situation of these families.

Traditional systems have been developed, modified, and adapted over a long period of time to best meet farmers' goals, including implicit or explicit nutritional goals, within existing ecological conditions and constraints external to the farm. Agricultural research may remove some of these constraints and thus greatly improve the living standards of farm families, including their nutritional status. However, to assure success in such efforts, it is important to understand fully why farmers use existing systems and what the repercussions would be if the systems were changed (6).

Research and policies resulting in a shift from mixed cropping systems to mono-cropping should be carefully watched for possible negative nutritional effects. This is particularly important for activities that promote substitution of a cash crop for a mixture of food crops traditionally produced for home consumption. This is not to argue that the nutritional effects of such projects are usually or always negative. The additional incomes generated within the farm household and the associated expansion of food purchase may fully compensate for the loss in the consumption of own-produce and provide for additional real incomes that may be used for nutritional improvements or other desired welfare gains. But increasing risk associated with mono-cropping, local price increases for traditional foods caused by the change from semi-subsistence food production to cash crop production, and inefficient food marketing systems and a number of other factors, e.g., characteristics of intra-household income control, may result in a nutritional impact smaller than expected or even in a negative one (7).

A large part of existing calorie protein deficiencies, particularly among the rural poor, appears to be a result of fluctuations-seasonal or irregular ones-in food prices, incomes of the malnourished, and food availability. Depending on its nature, agricultural research may contribute to this problem, e.g., through substitution of mono-cropping for mixed cropping in situations where existing policies and rural infrastructure are unable to deal effectively with such a change, or it may alleviate the problem by improving production or storage systems or making available improved crop carieties, e.g., those that permit a more appropriate crop rotation, pest-and disease-resistant varieties, and varieties with more stable yields under difficult and variable climates.

Nutrient Composition

Agricultural research may also influence human nutrition through improvements in the nutrient composition of a particular crop. Improvements in the protein quality of maize and sorghum are cases in point. in the past, the nutritional implications of agricultural research were frequently assessed on the basis of the extent to which the nutritional composition of a given food commodity was improved. At one extreme, there was an underlying goal of making the nutritional characteristics of an individual food commodity match as closely as possible the nutritional needs of people, as if each commodity did not enter into a multiple commodity diet. Such emphasis clearly resulted in misleading guidance to agricultural research. What is important is not whether expanded nutrient intake by the malnourished is a result of: (i) improved nutrient composition of a particular food commodity, or (ii) more nutrients from a larger quantity or a more appropriate dietary mix of various foods. What counts is expanded intakes from the diet as a whole. In some cases, this is best accomplished by modifying the nutritional composition of particular commodities; in others it is more efficiently done by facilitating the availability of more and cheaper foods that will add up to an adequate diet.

Intra-household Factors

Agricultural research may influence human nutrition through changes in household income composition, e.g., from own production to cash, intra-household income and budget control, e.g., changing the share controlled by women as new technology is introduced in food production on semi-subsistence farms, and women's time allocation, e.g., changing the demand for women's time in food production and thus changing time available for food preparation and child care.

The impact on these factors - and thus nutritional status-may vary among agricultural research strategies. Thus, they should be explicitly considered in research planning. However, current understanding of the relevant processes is still very deficient and additional research is needed to assist in clarifying how agricultural research planning may best deal with this issue.


In view of the linkages between agricultural research and human nutrition discussed above, what can international agricultural research centres do to assure that nutritional goals are considered along with other goals in the planning of their strategy and activities? There are four areas of decision-making where nutritional concerns might be incorporated:

• Establishment of commodity priorities.
• Specification of desired changes in commodity characteristics
• Specification of desired technology characteristics.
• Choice of production systems to be researched.

Each of these four areas is briefly discussed below, with emphasis on the principal nutrition-related issues, how nutritional goals might be explicitly considered, and the type of information needed. Examples of current efforts by the international agricultural research centres in each of these areas will be discussed in a subsequent section.

Commodity Priorities

The key issue here is how total calorie and nutrient intakes by the malnourished would change as a consequence of the introduction of yield-expanding or yield stabilizing technology for a particular crop or livestock species. This would be determined primarily by (i) the relative importance of the crop or livestock species in the total diet of the malnourished and their reaction to changes in the price of the commodity; (ii) the extent to which farm households with malnourished members produce the crop or livestock species; (III) the extent to which additional employment is created, and (iv) existing price and trade policies.

If existing malnutrition is primarily found among those who do not participate in food production, i.e., urban and some rural consumers, then nutritional considerations would suggest that research emphasis be placed on commodities that take up a large share of these consumers' food budget, and for which price reduction would lead to expanded consumption of the particular commodity as well as substitutions among foods towards relatively large expansions in total intakes of calories and/or deficient nutrients.

If, on the other hand, nutritional deficiencies are primarily found among low-income agricultural producers, nutritional considerations would suggest research emphasis on commodities that would generate more incomes for these households, reduce risks and seasonal fluctuations, and/or make more food available to them from their own production. Because international agricultural research by its nature serves a number of countries, and the relative importance of consumer vs. producer malnutrition differs among countries, commodity priorities in international agricultural research cannot usually be limited to one or the other. Ideally, from a nutritional point of view, emphasis would be on commodities that take a large share of the budget of households with malnourished members for whom a price reduction would result in a large increase in total calorie and protein intakes. Commodities that occupy a large share of the resources (land and labor) owned or controlled by producing households with malnourished members should be considered, as should those that generate employment and incomes for the landless poor.

While emphasis on the "nutritionally ideal" commodity combination may be unattainable because of conflicts with desires to achieve other goals or for other reasons, attempts to move towards such an emphasis may nevertheless be feasible. The success of such attempts will depend on availability of information on which foods households with malnourished members consume and produce, how they would adjust their food intakes in reaction to changes in prices of individual food commodities and incomes, and on commodity-related employment of rural landless poor, and the likely impact of technological change.

Existing information provides some general idea of what poor people eat. However, the information is very deficient at best, and partly for that reason, use of such information in planning agricultural research policy has been an exception to the general rule of using average consumption figures for the population as a whole. But average figures are not very useful for estimating nutrition effects because they do not adequately reflect consumption by the malnourished.

Similarly, reported household reactions to price and income changes as measured by the price and income elasticities of demand, usually reflect the population as a whole. Because the concern is for households with malnourished members, the estimates must be relevant for these households. In societies with a very skewed income distribution and considerable malnutrition, average estimates are not likely to represent the behaviour of households with malnourished members. Thus, the estimates must be based on income group.

Reliable estimates specifically for groups of households with malnourished members are of recent origin and their use in food policy design has been very limited indeed. During the last few years, however, there has been a considerable increase in research efforts to estimate demand characteristics by income stratum, and there are now reliable estimates for a small number of countries.

Although each research institute is usually charged with a certain commodity portfolio, it is nevertheless possible to alter the relative budget allocation to each of these commodities. Furthermore, relative commodity emphasis within the research system as a whole is an important consideration for which additional information would be useful. Efforts by the international agricultural research institutes to obtain relevant information in this general area, and proposed additional activities, are discussed in subsequent sections.

Commodity Characteristics

As stated earlier, the nutritionally important question is whether total intakes of digestible nutrients by an individual are sufficient and not whether the particular nutrient originates from one food or another. Furthermore, in most cases, foods currently consumed by the malnourished are capable of providing an adequate diet if consumed in appropriate quantities and dietary combinations. Thus, improving nutritional characteristics of foods currently consumed by the malnourished is not essential to eliminate malnutrition. This does not mean, however, that agricultural research should ignore potential nutritional gains from improving traditional foods. A modest research input into the testing of promising breeding and selection material for such things as protein content, quality, and digestibility, and oil content may facilitate a nutritionally more appropriate choice of the material without unacceptable reductions in the achievement of other goals such as improved yields and yield stability. Such testing is especially important for commodities that show large genetic variation in a particular nutrition-related characteristic critical for those who traditionally consume the commodity.

Increasing the energy density of foods with large genetic variation in oil content and that traditionally are principal calorie sources for small children is a case in point. Improving the protein quality, quantity, or digestibility of food that presents large genetic variation in these characteristics and that is an important component of diets with a specific protein deficiency is another exam pie, provided that the additional protein is expected to be available at a lower price than the cheapest alternative source, i.e., expanding consumption of another food.

In addition to nutrient composition, attention should be given to characteristics that influence the acceptability of a particular commodity among low-income consumers, e.g., cooking quality, texture, and colour.

Five types of information are required for decision making in this area. First, it is important to know whether a particular food commodity plays an important role in diets of people suffering from nutritional problems other than a combined calorie-protein deficiency that could be corrected simply by expanding consumption of the particular food. Such knowledge should result from information regarding consumption patterns by the malnourished, as discussed under commodity priorities. Second, information is needed on consumer preferences regarding the characteristics of a particular commodity, e.g., cooking quality, colour, and texture, and how these preferences vary among income groups. Third, information of the genetic variation in the particular nutrition related characteristics is important. Fourth, actual testing will provide breeding and selection guidelines, and fifth, implications of changing commodity characteristics for other parts of the food system should be anticipated and included in the estimation of nutritional effects, e.g., spoilage problems associated with higher oil content.

Technology Characteristics

Research decisions determine or influence the nature of the resulting technology that, in turn, influences human nutrition in at least four ways. First, the nature of the technology is an important determinant of how much more is produced of a particular commodity and at what cost. The nutrition implications of these matters were discussed under commodity priorities. Second, the extent to which rural households with malnourished members gain from new technology depends on whether the technology is suited for the production environments controlled by these households and the effects of the technology on employment. Thus, identification of the production environments most commonly controlled by these households along with explicit specification of the production environments best suited for particular technology designs would be useful in efforts to incorporate nutrition concerns in decision-making that influences the nature of the technology to be developed (1, pp 124-132).

Third, because seasonal and irregular fluctuations in food availability, prices, and incomes are important contributors to malnutrition in many rural areas, new technology that facilitates a reduction in such fluctuations would be preferable from a nutritional point of view. Past and current international agricultural research places high priority on reduced lodging losses, improved pest resistance, better adaptation to improved cropping patterns, and other factors that increase yields while reducing the risk of partial or total crop failure and diminishing seasonal fluctuations, thus serving nutritional as well as other goals.

Finally, the interaction between technology characteristics and household decision-making and labour allocation may be an important consideration from a nutritional point of view. A particularly important element of this issue is the role played by women in low-income farm households in decisions regarding technology adoption and labour allocation and how technological change affects time allocation by women to other nutrition-related activities such as child care-including breast-feeding- and food preparation. These issues have traditionally been ignored in decisions on technology design.

Although relevant research on this topic is of recent origin and still very limited, it is clear that women play an important role in decisions regarding technology adoption and labour allocation on low-income farms and ignoring this role may lead to technology misspecification (8-10). There is also some evidence of the trade-off between increased demand for women's time in agriculture and reduced labour allocation to child care and food preparation (11). The most important issues appear to be whether the low income farm household is able to capture the benefits from new technology only if household women allocate more time to the crop, whether these women are willing and able to allocate such additional time, and what the nutritional implications are of such change in time allocation, possibly away from child care, cooking, food and water gathering and care of livestock. More research is needed on these issues because of their importance in decision making on technology design and public policies.

The impact of technological change on the distribution of household budget control between men and women and the resulting effects on food consumption and nutrition is another important issue deserving additional attention by the research community.

Production Systems

From a nutritional perspective, the key issue to be considered in decisions related to research on production is how existing production systems may be changed in order that malnourished members of farm and rural labourer households may expand their Intakes of deficient dietary elements, e.g., calories, protein, vitamins, during critical periods, whether from own-household production or purchase. Many of the issues raised in the above discussion of commodity priorities and technology characteristics, e.g., production environments, risk, seasonal fluctuations and household decision-making issues, are also relevant to decisions on production systems.

The impact of alternative changes in existing systems on the ability of the household to meet nutritional requirements from own-production should be assessed. This is not to argue that research on production systems should focus solely on subsistence food production. Rather, the focus should be on an appropriate mix of cash and subsistence crops to meet nutritional needs (from own production and/or food purchase) while seeking general expansions in real household incomes. Local marketing inefficiencies, inability of the household to deal effectively with rapid changes from semi-subsistence to a cash economy, and related factors may render a traditional economic criterion ineffective in assuring nutritionally appropriate changes in existing production systems. The question is not whether but how rural transformation should be promoted. Arguments by some that rural transformation based on technological change is generally disadvantageous for the rural poor conflicts with available empirical evidence. However, a great deal can be done to avoid hardships on the poor in those cases where they do occur and to enhance positive effects. Agricultural ret search can do some, public policies and institutional changes must do most.


This section reports on selected past and current activities undertaken by the international agricultural research centres that facilitate the explicit consideration of nutritional issues in research planning. The discussion is meant as an illustration of the kinds of activities undertaken by the centres rather than an exhaustive report of all such activities.

Commodity Priorities

A number of centres have undertaken research to improve existing knowledge of what the poor eat and how they react to changes in incomes and prices of individual foods. In an on-going study of seven countries, the International Potato Center (CIP) in Peru examines the role of potatoes in the diet of various population groups under a variety of agroecological, cultural, and socio-economic conditions (12, 13). A proposal has been developed for an expansion of this work to include other root and tuber crops, possibly as a collaborative effort among various centres (14). The ultimate goal of this work is ".... to improve the ability of planners to determine national food supply needs and prioritize areas of food production, marketing and consumption research" (14, p. 3). As an outcome of the second social science planning conference of CIP held in September 1981, the evaluation committee endorsed ".... an increasing emphasis on demand and consumption studies so as to: (i) develop basic information on actual and potential use of potatoes as food for poor people; and (ii) assist CIP policy-makers in identifying potential target countries and beneficiary groups within countries" (conclusions and recommendations for the Second Social Science Planning Conference, CIP, Lima, Peru, September 7 11, 1981, draft, p. 3).

On the basis of existing data, the International Maize and Wheat Improvement Center (CIMMYT) in Mexico undertook an analysis of the role of maize and wheat in the diet of the poor and how this role is affected by various policy measures (15).

Using household consumption data collected by Estudios Conjuntos sobre Integración Económica Latinoamericana (ECIEL), the International Center of Tropical Agriculture (CIAT) estimated the relative importance of various foods and food groups in the diets of the poor in a number of Latin American cities as well as the related income elasticities ( 16). Similar work that also included estimates of price elasticities for the malnourished was completed for the city of Cali, Colombia, on the basis of primary data (17). These estimates were then used to illustrate how commodity priorities in agricultural research might be affected if improved nutrition were an explicit goal. More in-depth consumption analyses of individual CIAT commodities, e.g., cassava, have also been undertaken (18). Food consumption data are used to assist in the development of ClAT's long-term plans (19).

Estimation of income and price elasticities specific to the poor and/or malnourished has also been undertaken by the International Crops Research Institute for the Semiarid Tropics (ICRISAT) for India (20) and by IFPRI for Brazil (21). These studies illustrate the utility of such elasticities for estimating the nutrition effects of alternative commodity priorities in agricultural research and policies.

One additional piece of research in the area of commodity priorities should be mentioned. This is the ICRISAT Study of the impact of shifts from pulses to wheat in India on the production of protein and essential amino acids per unit of land (3). This is an important contribution because it effectively refutes the popular belief that such substitution, which resulted from the availability of yield-expanding technology for wheat but not for pulses, reduced total protein and amino acid availability.

Commodity Characteristics

Most research contra activities in this area relate to the protein content (quantity and/or quality) and digestibility, including toxic factors, in individual crops. CIMMYT's research on quality protein maize is well known and needs no further elaboration. Many of the centres undertake analyses of protein quantity and quality in breeding and selection materials, e.g., ICRISAT in sorghum, pearl millet, chick peas, and pigeon peas (22), and the International Center for Agricultural Research in Dry Areas (ICARDA) on wheat, barley, and various grain legumes (23, 24). Most of these analyses are based on chemical testing.

The International Rice Research Institute (IRRI) has tested for protein quality of rice in children and rats (25). Furthermore, CIP, in collaboration with the Instituto de Investigación Nutricional (IIN), has carried out studies of potato protein quality and digestibility in children (26). IIN has also undertaken evaluation of protein quality and digestibility of sorghum varieties in infants (27). However, biological evaluation appears to be declining among the centres.

The emphasis on improved protein content (quantity and quality) as a crop breeding and selection goal in agricultural research presents an interesting evolution. Up until the early 1970s it was generally believed that the principal nutrition problem was one of protein deficiency or deficiency of certain essential amino acid in the diet. This, together with promises of dramatic improvements in the protein quality in grains, as exemplified by the Opaque-2 gene in maize and high-lysine lines of sorghum, led to pressures on the international agricultural research centres to place emphasis on improved protein content as a breeding and selection goal even, if necessary, at the expense of yield advances. It was during this period that the high-lysine maize programme was initiated at CIMMYT. It was also during this period that a Protein Quality Planning Conference at CIP recommended that ".... data derived from chemical and biological evaluations of nutrition quality are to provide the basis for selecting parental materials", although not at the expense of yield increase (26, p. 1).

As the scientific nutrition community provided more evidence in support of the argument that the principal nutrition problem was one of a combined calorie-protein deficiency, and that in most cases both would be alleviated by greater food consumption under existing dietary compositions, while increased protein intakes or improved protein quality in the absence of expanded energy intakes would do little to alleviate the problem, the pressure lessened. ICRISAT contributed greatly to a clarification of this issue through its work on protein and amino acid consumption and the implications of a yield vs. protein focus of plant breeding (2). Work sponsored by CIMMYT (28) also made an important contribution in this regard

Partly as a consequence of the changing view of the role of protein in alleviating malnutrition, and partly because of other difficulties associated with high-quality protein material, e.g., poor consumer acceptance and storage problems, as illustrated by CIAT research (29), the international agricultural research centres have de-emphasized protein, particularly in cases of an apparent conflict between more or better protein and higher yields.

Recent testing for nutrient content includes studies by IRRI on the effect of sulphur deficiencies in soils on the content of s-containing amino acids in rice (25), evaluation by the International Institute of Tropical Agriculture (IITA) of N content in sweet potato leaves and roots (30), and exploratory research by CIMMYT to improve nutritional quality aspects other than protein in maize, e.g., oil content (31). Because maize is an important component of the diet of small children in many countries and low-energy density of small children's diets is a significant problem, higher oil content may be of considerable nutritional importance.

Findings from ICRISAT village-level studies show that certain vitamins and minerals are limiting factors in the diets of the semi-arid tropics of India, and Ryan (personal communication) concludes that there may be some scope for selection of breeding lines for improved vitamin and mineral content. This raises an interesting perspective that deserves further examination. While on the one hand, care should be taken not to introduce an excessive number of breeding and selection criteria, some exploratory work would certainly be worthwhile to assess whether widespread vitamin and mineral deficiencies could be effectively dealt with through breeding and selection without adverse effects on the achievement of other goals such as yield expansions.

Technology Characteristics and Production Systems

The centres have completed a large number of analyses of small farmer production systems, including adoption behaviour, and the expected (ex ante studies) or actual (ex post studies) impact of new technology. This work has assisted in the design of technology suited for small farmers. Similarly, a large number of studies have been undertaken on the employment effects of various types of new technology. Findings from some of this work are summarized by Pinstrup-Andersen (1). A summary of centre work in the area of production systems may be found in a report by the Consultative Group on International Agricultural Research (32).

While the above work provides important guidelines regarding the nutritional effects of alternative commodity priorities, technology characteristics, and production systems through its impact on incomes of low-income farmers and rural workers, its effectiveness would be greatly enhanced if it were linked with effects on food consumption by malnourished households. Efforts to establish such links are undertaken by CIMMYT as exemplified by a recently completed study in Ecuador (33).

Nutritional considerations are an integral part of cropping systems research at IITA and have been incorporated into village studies at ICRISAT through a collaborative study with the National Institute of Nutrition in India and Andra Pradesh Agricultural University (B. Okigbo and J. Ryan, personnel communications). Furthermore, IFPRI is currently undertaking research to estimate the impact of the adoption of new technology on household food consumption and nutritional status of children in regions of three countries (34). Other IFPRI research on household food acquisition behaviour and the nutrition effects of selected food policies is expected to improve existing knowledge of the processes that determine the nutrition effects of various programmes and policies, including agricultural research. However, more work is urgently needed on the link between agricultural research and food consumption by the malnourished in agricultural households, to assist in the design of appropriate changes in production systems and technology as well as in public policies, in order to enhance positive nutritional effects and avoid negative ones.

Other Activities

Nutrition-related contra activities not discussed above include: (i) workshops on nutrition-agriculture interfaces held at ICRISAT and ICARDA under joint sponsorship with the United Nations University, (ii) CIMMYT's establishment of, and periodic interaction with, a panel of nutritionists, (iii) research on the effect of nutritional status on wages and labour participation (35), (iv) an IRRI conference on "Women in Rice Farming Systems" planned for September 1983, and (v) research by IFPRI on the nutrition effects of various policies (36-39).


A number of activities are currently being undertaken by the international research centres to assure that nutritional considerations are reflected in research planning. Additions to these activities might further improve the nutritional effects of agricultural research whether national or international. Some such additional activities would be:

1. The preparation of a document for publication that outlines and discusses nutrition-related activities by each centre and how the goal of improved nutrition does or does not enter into research priorities and planning. Such a document would contribute to a better understanding of the complexity of the centre research, the extent to which the research aims at the poor and, in turn, assist in improving the information base for future allocations of funds to the CGIAR system. It should also advance a useful dialogue among the various sectors and disciplines, e.g., agricultural scientists, nutritionists, anthropologists, and economists, in an effort to further improve the approaches to the integration of nutritional goals into agricultural research planning. Possible components of such a document were prepared by Ryan for ICRISAT, Poats for CIP, and Okigbo for IITA as background material for this paper.

2. Assist interested national research institutions and other public entities in collecting and/or utilizing nutrition-related information and incorporating nutrition concerns into research and policy decision-making. It appears that the centres are doing considerably more in this area than most national agricultural research institutions, and transfers of ideas and methodologies that have been developed and/or tested by the centres, along with technical assistance, are likely to be of mutual benefit.

3. Undertake a study of what households with malnourished members eat (rural and urban) and produce (rural only), and how these households are likely to change their consumption of calories and nutrients in response to price and income changes resulting from technological change in each of the principal commodities. The potential utility of knowledge made available by such a study was discussed earlier and will not be repeated here. Many of the centres have undertaken studies in this area. However, most of them are limited to the commodities being researched by the particular centre. Furthermore, they tend to be limited to one or a few countries or regions within countries. There is a need for an integrated study with participation from all centres that would draw on existing data, whether available from the centre's own work or data collection by others, e.g., consumption surveys undertaken in many countries during the last 10 years, to provide estimates of the global importance of each of the major foods in the diets, the budgets and incomes of households with malnourished members, and how food intakes by these households would be affected by technological change and price policies for each of the principal commodities. Such information would be useful in future decisions by each of the centres and by the system as a whole on commodity priorities, technology design, and production systems.

4. Undertake one or more studies of the nutrition effects of past research and resulting technological change in selected crops and countries. Except for work by Ryan and Asokan (31, very little solid empirical research has been done on these effects. As a consequence, the topic is subject to considerable speculation and unfounded accusations.

5. Undertake or sponsor studies of the role of women in technology adoption, labour availability, and household food acquisition behaviour. Initiation of such studies might await the findings of the IRRI conference on the subject.

6. Incorporate nutritional considerations explicitly into research on production systems as one of the criteria for an appropriate system.

7. Maintain close contact with the nutrition community for the purposes of: (i) periodic review of on-going and planned research to identify ways in which nutritional improvements may be enhanced without unacceptable effects on the achievement of other goals, (ii) informing centre staff of advances in the nutrition field relevant to agricultural research, and (iii) contracting or collaborating with nutrition institutions on research or other activities needed by the centres but for which they have no comparative advantage. The most appropriate approach for maintaining contact with the nutrition community may vary among centres, the CIMMYT nutrition panel being one example. Irrespective of the approach followed, it is important for each centre to have on its staff at least one person with a basic understanding of, and an appreciation for, nutrition-related issues in order to facilitate communication between centre staff and management and the nutrition community.

8. Maintain close contact with other international institutions such as FAO, WHO, SCN, and INCAP on nutrition-related activities that are important to the centres but for which they have no comparative advantage to undertake themselves.

9. Continue on-going testing of breeding and selection material for nutrient composition and digestibility in order to identify ways of improving commodity characteristics without unacceptable effects on the achievement of other goals.

10. Undertake exploratory evaluation of breeding and selection materials to assess the feasibility of using the content of certain vitamins or minerals as breeding and selection criteria for certain crops.


This paper was prepared at the request of the ACC/SCN Working Group on Nutrition in Agriculture and Rural Development. Acknowledgement is due to a number of people who provided information on nutrition-related activities currently undertaken by the International Agricultural Research Center and/or commented in an earlier draft, including George H. Beaton, Ricardo Bressani, Peter Brumby, Derek Byerlee, Doris Howes Calloway, Bede N. Okigbo, Susan V. Poats, James G. Ryan, John Sanders, Peter Oram, Ammar Siamwalla, John Mellor, Harold Alderman, John Lynam, Robert W. Herdt, Robert Tripp, and Don L. Winkelmann. Errors and omissions are solely the responsibility of author.


1. P. Pinstrup-Andersen, Agricultural Research and Technology in Economic Development (Logrnan, London and New York, 1982).

2. J.G. Ryan, R. Sheldrake, and S.P. Yadav, Human Nutritional Needs and Crop Breeding Objectives in the Semi-Arid Tropics: A Further Note (ICRISAT, Economics Program, Occasional Paper, Hyderabad, India, January 8,1974.

3. J.G. Ryan and M. Asokan, "Effect of Green Revolution in Wheat on Production of Pulses and Nutrients in India." Indian J. Agric. Econ., 32: 8 (1977).

4. K. Somel, "The Green Revolution: Achievement and Implications." Paper presented at the ICARDA/UNU Workshopon the Interfaces between Agriculture, Food Science, and Human Nutrition in the Middle East, Aleppo, Syria, 21-25 February, 1982.

5. A. Sen, Poverty and Famines-An Essay on Entitlement and Deprivation (Clarendon Press, Oxford, England, 1981),

6. B. Okigbo, "Introducing Nutritional Considerations into Research and Training in Farming Systems." Paper presented at the ACC/SCN Symposium on Introducing Nutritional Considerations Into Agricultural and Rural Development, Rome, March 2,1981.

7. P. Pinstrup-Andersen, Export Crop Production and Malnutrition (Institute of Nutrition, University of North Carolina, Occasional Paper Series, Vol. 11, No. 10, February, 1983. [Also IFPRI reprint]).

8. I. Tinker, New Technologies for Food Chain Activities: The Imperative of Equity for Women. Office of Women in Development US/AID, Washington, D.C. (1979).

9. C. Jones, 'Women's Labor Allocation and Irrigated Rice Production in North Cameroon." Paper presented at the IAAE Conference, Jakarta, Indonesia, August-September 1 982.

10. J.I. Guyer, "Household Budgets and Women's Incomes." Working Paper No. 28, African Studies Center, Boston University, Boston, Massachusetts (1980).

11. B.M. Popkin, "Time Allocation of the Mother and Child: Nutrition." Ecol. Food Nutr., 9 (1): ( 1980).

12. S.V. Posts and G.T. Castillo, "Beyond the Farmer: Potato Consumption in the Tropics." Paper presented at the CIP Decennial Anniversary, February 22-26, Lima, Peru, 1982.

13. S.V. Poets, "Potato Consumption in the Kingdom of Bhutan." CIP Draft Paper, Lima, Pen, October, 1982,

14. S.V. Poats, "National Food Policy Implications of Root and Tuber Consumption in Developing Countries. A Preliminary Proposal," CIP, Lima, Peru, September, 1982.

15. D. Byerlee and L. Harrington, "New Wheat Varieties, Poor Producers and Poor Consumers." Paper presented at the 18th International Conference of Agricultural Economics, Jakarta, Indonesia, September, 1982.

16. D. Pachico and J.K. Lynam, "Food Supply and Malnutrition in Latin America." In: Latin American Agriculture: Trends in CIAT Commodities (CIAT Internat. Document Econ. 1.6, Cali, Colombia, May, 1981).

17. P. Pinstrup-Andersen, N. Ruiz de Londono, and E. Hoover, "The Impact of Increasing Food Supply on Human Nutrition: Implications for Commodity Priorities in Agricultural Research and Policy." Am. J. Agric. Econ., 58:131(1976).

18. D. Pachico and J. K. Lynam, "Cassava Production, Marketing and Utilization." In: Latin American Agriculture: Trends In CIAT Commodities (CIAT Internat. Document Econ. 1.6, Cali, Colombia, May, 1981).

19. Centro Internacional de Agriculture Tropical, CIAT in the 1980s: A Long-Range Plan for the Centro International da Agricultura Tropical, Series 12E-5, November, 1981.

20. K.N. Murty and R. Radhakrishna, "Agricultural Prices, Income Distribution, and Demand Patterns in a Low Income Country. " Paper presented at the First International Working Conference on Computer Applications In Food Production and Agricultural Engineering Havana, Cuba, October 26-30, 1981,

21. C. Williamson-Gray, Food Consumption Parameters for Brazil and Their Application to Food Policy, IFPRI Research Report No. 32 (International Food Policy Research Institute, Washington, D.C., September, 1982).

22. P. Jambunathan, U. Singh, and V. Subramanian, "Grain Quality of Sorghum, Pearl Millet, Pigeonpea, and Chickpea." Paper presented at the Workshop on Interface between Agriculture, Nutrition, and Food Sciences, ICRISAT/National institute of Nutrition, Hyderabad, India, November 10-12, 1981,

23. M. Mekni and M. Nachit, "Branding for Nutritional Quality in Cereals." Paper presented at the ICARDA/UNU Workshop on the Interfaces between Agriculture, Food Science, and Human Nutrition in the Middle East, Aleppo, Syria, February 21-25,1982.

24. P.C. Williams, "Regional Research on Processing and Nutritional Quality of Cereals and Legumes at ICARDA." Paper presented at the ICARDA/UNU Workshop on the Interfaces between Agriculture Food Science, and Human Nutrition in the Middle East, Aleppo, Syria, February 21-25, 1982.

25. International Rice Research Institute, Research Highlights for 1981. (IRRI, Los Baños, Philippines, 1982).

26. S.V. Poats, Potato Nutrition and Consumption Research at the International Potato Canter (CIP, Lima, Peru, Mimeo, restricted circulation, 1982).

27. W.C. MacLean and G.G. Graham, Instituto de Investigación Nutricional, Evaluation of the Protein Quality and Digestibility of Four Varieties of Sorghum in the Diets of Infants. Progress Report submitted to the Agency for International Development, Office of Nutrition, US/AID, Washington, D.C.(1979).

28. P.R. Payne, "Nutritional Criteria for Breading and Selection of Crops: With Special Reference to Protein Quality." Plant Foods for Man, 2: 95 (1976).

29. P. Pinstrup-Andersen, The Feasibility of Introducing Opaque 2 Maize for Human Consumption in Colombia (Technical Bulletin No. 1, Centro Internacional de Agricultura Tropical, Cali, Colombia, 1971).

30. International Institute of Tropical Agriculture (IITA).Annual Report for 1980 (I ITA, Ibadan, Nigeria,1981),

31. International Maize end Wheat Improvement Center (CIMMYT), Movies 1981 (CIMMYT, Londres, Mexico,1982).

32. Consultative Group on International Agricultural Research CGIAR), Farming Systems Research At the International Agricultural Research Centers (CGIAR. New York, September,1978).

33. R. Tripp, "Including Dietary Concerns in On-Farm Research: An Example from Imbabura, Ecuador." (CIMMYT Working Paper, Londres, Mexico, February, 1982).

34. International Food Policy Research Institute (IFPRI),Report 1981 (IFPRI, Washington, D.C., 1982).

35. J.G. Ryan, "Wage Functions for Daily Labor Market Participants in Rural South India" (ICRISAT, Economics Program, Progress Report 38, Hyderabad, April, 1982).

36. H. Alderman, J. von Braun, and S. Ahmed Sakr, "Egypt's Food Subsidy and Rationing System: A Description" (Research Report No. 34, International Food Policy Research Institute, Washington, D.C., October, 1982).

37. R. Ahmed, "Foodgrain Supply, Distribution and Consumption Policies within a Dual Pricing Mechanism: A Case Study of Bangladesh" (Research Report No. 8, International Food Policy Research Institute, Washington, D.C., 1979).

38. J.D. Gavan and US. Chandrasekera, "The Impact of Public Foodgrain Distribution on Food Consumption and Welfare in Sri Lanka" (Research Report No. 13, International Food Policy Research Institute Washington, D.C., 1979).

39. S. Kumar, "Impact of Subsidized Rice on Food Consumption and Nutrition in Kerala" (Research Report No. 5, International Food Policy Research Institute, Washington, D.C., 1979).

Contents - Previous - Next