Contents - Previous - Next

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

Food and nutrition policy

The impact of agricultural and food-supply policies on nutrition and health status
The demand for higher-status food and nutrition in rural India: the experience of Matar Taluka. I. Basic data and Interrelationship of variables
The effects of food price and income policies on the nutrition of low-income groups: a Philippine case-study

The impact of agricultural and food-supply policies on nutrition and health status


In July 1982, the Food, Nutrition, and Poverty Programme of the United Nations University (UNU), with the support of the United Nations Development Programme (UNDP), joined in a collaborative research effort with the International Food Policy Research Institute (IFPRI) and the UN Research Institute for Social Development (UNRISD) to provide information to policymakers and planners concerning measures that would increase food security for vulnerable groups in developing countries. The project was to focus on how the nutritional and health status of at-risk populations would be affected by agricultural and food-supply policies. It examined four different types of food-security policies: (a) those intended generally to increase agricultural output; (b) those directed at modifying crop mix or variety, i.e. attempts to "breed" or genetically engineer better nutrition; (c) those designed to expand available food supplies by improving the quality or efficiency of storage, transportation, processing, and distribution; and (d) those directed at the fundamental political and socio-cultural obstacles that frequently impede efforts to improve the circumstances of the poor and powerless.


At the end of the project, in February 1985, a workshop was held at the Rockefeller Foundation's International Conference Center in Bellagio, Italy. Members of the project's task force of expert advisers and the 13 principal investigators discussed and improved each individual research project report and summarized the overall findings of the UNU project. The cluster concerned with the effects of seasonal fluctuations on nutrition and health status appeared in the previous issue. In the present issue we are publishing a second cluster of two articles concerned with food prices, food selection, and nutrition.

The most significant points emerging from the workshop discussions are summarized below.

  1. The results of the FAD/WHO expert committee recommendations concerning modifications in international protein-energy requirements were analysed in terms of their implications for policy. Certain changes in methodological approach to the problem were suggested. The issue of adaptation to available protein and energy was identified as a major concern, including physiological, social, and economic adaptation. From a policy perspective, the decision by a government to make more protein and/or calories available in the aggregate will have profound impacts, but these will most often be manifested at the individual or family level. It is also critically important to take account of market forces (i.e. effective economic demand for protein and calories). Thus, each household will have different requirements based on levels of physical activity, social requirements, and available income, and all three factors must be considered together in order to reach an overall assessment of the malnutrition problem and the nature of the policy response.
  2. The view was expressed that the international nutrition community - both life scientists and social scientists - needs to devote more adequate attention to developing or identifying real linkages with agricultural and food policies that can have a positive impact on the circumstances of poor people.
  3. A concern was raised about the tendency to offer nutritional advice for agricultural and food policy in terms of Western-style diets. The example of recommendations for the consumption of milk and other dairy products was given. Sometimes such recommendations are made because certain foods are considered to have higher status.
  4. In addition to the project's focus on agriculture-nutrition linkages at the individual, household, and national levels, the international research community needs to give adequate attention to the adverse effects of international pricing patterns for agricultural products. Since the international market is the primary source of income in many developing countries and since the price of agricultural products has not kept pace with other necessities (e.g. agricultural inputs and energy), rural peasant farmers tend to fall further and further behind economically and, therefore, nutritionally.
  5. It was noted that there has been a tendency for research to examine only the negative impacts of cash-cropping on nutrition and health status. Yet the UNU projects and other research have produced evidence that, under some circumstances, cash-cropping can have a positive impact on basic human needs.
  6. There was substantial interest in opportunities presented by small-scale food production at the household and/or village levels. Much of the discussion focused on the potential for expanding home-garden-type approaches, both for participation in the cash economy and for individual/family consumption.
  7. It was acknowledged that the decision-making arena for nutrition and health issues is very different from that for agriculture and food policy. In many cases, officials from the two areas do not approach problems, or policymaking, from the same mind-set. It is essential, therefore, that the concerns and perceptions of the "client" be taken into account when proposing policy recommendations that will promote the development of linkages between agriculture and nutrition. This obvious need has largely been ignored up to the present.

There was general agreement on one central point at the culminating workshop in Bellagio: the subject-matter is both enormous and extraordinarily complex. Until the advent of the UNU project, these areas of concern had received little systematic scholarly attention. In this respect, the results reported here represent merely a useful beginning to an effort to explore the possible linkages between agriculture and food-supply policies and the nutrition and health status of vulnerable groups, with the goal of increasing long-term food security.

Mitchell B. Wallerstein,
National Academy of Sciences
Washington, D.C., USA.

The demand for higher-status food and nutrition in rural India: the experience of Matar Taluka. I. Basic data and Interrelationship of variables

C. H. Shah
Gujarat Institute of Area Planning, Ahmedabad, India

This article and the one that follows exceed normal length limits for the Food and Nutrition Bulletin, but are published as one of the clusters arising from the workshop described in the preceding article. This study, which we are publishing in two parts, provides evidence that, under some circumstances at least, the poor are more likely to spend additional income on more costly or prestigious foods than are those who are somewhat better off. While this may appear to nutritionists to be irrational behaviour, it is a fact that must be recognized by those responsible for nutrition education and food policy. There is hearsay evidence, and some published material, shoving that this phenomenon is by no means limited to India. Clearly, it requires formal documentation elsewhere.


To effectively assess and remedy food shortages or surpluses in a society, all foods cannot be treated as equivalent in nutritional value, nor can nutritionally similar foods be treated as equivalent per unit cost. Moreover, since foods vary in acceptability to consumers, a shortage of preferred foods may occur when overall food supplies are abundant. Taste or individual preference plays an important role in food choice.

The role of individual taste in food choice is important not only in estimating food supply adequacy, but also in evaluating the accuracy of some variables as indicators of poverty in a society. Taste may lead individuals to reject low-cost nutritionally advantageous foods in favour of ones that are not cost-efficient in supplying calories and other nutrients. Therefore, if the intake of calories or other nutrients is used as an indicator of poverty in a culture, poverty may be overestimated.

Although no two individuals have identical tastes, food preferences are not random. They are influenced by social and economic factors that determine whether foods are categorized as status goods or not. The major aims of this study are to define and identify status foods and to identify factors influencing their consumption. We will also analyse the relationships between the consumption of status foods and the intake of calories and other nutrients.


In brief, the concept of a status food was developed in three steps. A detailed description of this procedure is given in subsequent sections.

Step 1

We ran a linear programming (LP) exercise using as constraints the requirements for an adult male for calories, protein, vitamins, and minerals. We used individual items of food as activities, i.e. processes for deriving nutrients from different items of food, and market prices of food items as costs. Food items designated as "basic foods" were those that provided the required nutrients at the lowest cost.

We used the requirements for nutrients defined by the Indian Council of Medical Research. The nutrient contents of each food item were provided by the National Institute of Nutrition (Hyderabad, India).

Step 2

When the cost of achieving the desired level of nutrient intake had been minimized, the food items selected made up the consumption basket. The levels of nutrients were calculated on the basis of the quantities of food items in the minimum cost basket. The dual or opposite exercise was run to maximize nutrient levels for a given amount of expenditure, Shadow prices suggested by the LP exercise represented the real cost if choices were made to conform with the minimum cost of nutrients (i.e. to maximize the nutritional output of given expenditures). Since choices made by consumers did not conform to the LP solution the shadow prices differed from the market prices. From the dual of the LP exercise we derived shadow prices for calories, vitamins, and minerals. Multiplying each nutrient contained in the individual item of food by the respective shadow price, we calculated the total nutrient worth of each food item. Since the shadow prices reflected the minimum cost of individual nutrients in terms of the prices of the basic foods, the nutrient value of each non-basic item of food thus obtained would be below its market price.

Step 3

We took the difference between the nutritional value of individual items of food (as determined in step 2) and their respective market prices, regarding this difference as the "preference cost"; that is, it represented the willingness of the given community of consumers to buy, at a price, additional value (i.e. subjective utilities) to satisfy tastes or preferences. Step 3 measured the approximate preference value of a food separately from its nutritional value.

TABLE 1. Definitions

Household: A household consists of persons staying together under one common roof and having a common kitchen. In rural India a household usually contains people other than those in the immediate nuclear family: more than one generation is often found living together, and farm labourers, especially those employed on a yearly contract basis, share food with the family of the employer. We have not included farm servants as part of a household.

Standard adult unit: A male 20 to 30 years old is taken as the standard adult unit. Conversion ratios are used to represent other classes of individuals in standard units (table 2). These ratios are based on information from FAO.

Calorie requirements: Considering a body weight of 55 kg and medium to light activity for rural adult males, calorie requirements are fixed at 2,360 keel per day per adult male. This level of kcal gives about 1.4 basic metabolic rate (BMR) and is about 84 percent of the 2,800 keel recommended by the National Institute of Nutrition, India, for an adult male engaged continuously in medium activity.

Nutrient requirements: Requirements for nutrients other than calories, that is, protein, vitamins, and minerals, are for a standard adult male. We have not adjusted their levels downward as we have the levels for calories.

Occupations: Households are classified on the basis of the occupations of their heads. The three main categories of occupations are cultivator, labourer, and non-farm occupation. Cultivators are subdivided into five subclasses on the basis of farm size. People with non-farm occupations are subdivided into three classes: (a) those engaged in trade, money-lending, business, a profession, the transportation and processing industries, etc.; (b) those engaged in non-farm occupations, whether self-employed or employees, such as cobblers, barbers, blacksmiths, carpenters, potters, weavers, basket-makers, sweepers, sanitary workers, and messengers; and (c) a miscellaneous group consisting of minors (as heads of household, widows, pensioners, and those without specific occupations.

On the basis of the preference values, individual items were ranked in descending order. One-third of the total were classified as highly preferred; those in middle ranks were classified as medium preference foods; and those in the lowest ranks were classified as low-preference foods. All of the foods with preference values, however, ranked above the basic food items. In this report the high preference foods are referred to as status foods.


The pattern of food consumption habits was examined; for example, we identified the shares of the food budget allotted to each of the four categories of foods for consumers in different expenditure decile groups, and juxtaposed these results with a pattern of nutrient intake by the same docile groups. The exercise was repeated for calorie-deficient households (D), whose calorie intake per day per adult unit was below the required level, and for calorie non-deficient households (ND), whose calorie intake per adult unit was equal to or above the required level. Table 1 provides definitions of important terms used in this article, including household and adult unit calorie and nutrient requirements. Table 2 gives the conversion ratios used to obtain standard units to represent classes of individuals other than standard adult males.

Regression analyses were used to examine the influences of social and economic factors on expenditures for food and on nutrient intakes. This test was carried out for all households as an aggregate and then for the subsets of households that were calorie-deficient (CD) and calorie non-deficient (ND). Separate regressions were run for the two major categories of households.

TABLE 2. Conversion ratios for consumer groups using adult males as standards

Age Conversion factors
Males Females
Less than 1 0.43 0 43
1-3 0.54 0.54
4-6 0.72 0.72
7-9 0.87 0.87
10-12 1.03 0.93
13- 15 0.97 0.80
16-19 1.02 0.75
20-39 1.00 0.71
4049 0.95 0.68
50-59 0.90 0.64
60-69 0.80 0.51
70+ 0.70 0.50

TABLE 3. Constraints used for linear programming

Nutrients Unit Requirements per day per adult male
Calories kcal 2,800 (2,360)a
Protein g 55
Calcium mg 400
Iron mg 20
Carotene IU 3,000
Thiamine mg 1.4
Riboflavin me 1.5
Niacin mg 19
Vitamin C (ascorbic acid) mg 50
Phosphorus IU 1,400

a. Used for solution 2.

TABLE 4. Linear programming solutions

Commodities in minimum cost diet Kg per adult male per day
Solution 1 Solution 2
Bajra 0.275 0.355
Kodra 0.568 0.301
Fresh vegetables 0.074 0.0B8
Fresh fruits 0.032 0.028
Peanuts 0.001 0.018
Total 0.949 0.792
Minimum cost (rupees) 1.22 1.11


The data were collected in a field survey conducted in Matar Taluka in the Kheda district of the state of Gujarat, India, from 1974 to 1975. From 28 villages, 1,106 households were selected on the basis of two-stage sampling. We intended to collect data in three rounds covering the three main seasons, monsoon, winter, and summer-from July to October, November to February, and March to June respectively. In several instances, however, the second and third rounds had to be combined because of a delay in canvassing the second round. In each round data were collected for consumption during the previous month. The average over the three months was extrapolated to obtain yearly consumption figures for different items of food.

Data were collected on the quantities of most items. For items treated as equivalent within a class, such as fruits, vegetables, and spices, the information on quantities was related to the expenditures incurred. Quantities of food items were multiplied by a common price obtained from the local markets. Such data were expected to yield interpersonal quantity-intake differences.

The data were not detailed enough to reflect qualitative differences in food items. Especially for rice, the quality of grain consumed by upper-income-level households and those in the lower-income brackets could differ markedly. Other cereals did not differ much in quality. Since people consumed mainly locally produced grains at the household level, the range of quality, even for rice, would be smaller than for the supplies available from the market. For other grains the range would be insignificantly small. However, ignoring quality differences for individual cereals would understate the role of taste.


The purpose of the linear programme was to identity foods that would meet minimum nutritional requirements at low cost. Two different sets of solutions were obtained using 19 individual items of food (activities) supplying different nutrients and calories and two levels of calorie requirements-2,800 and 2,360 per day per adult male- as well as other nutrient requirements as constraints (table 3). Common prices obtained from the market were used to calculate the cost of nutrients. Solution 2 was obtained using the 2,360-calorie requirement per adult male. The two solutions are presented in table 4.

The two LP solutions yielded the same set of food items for a minimum cost diet (table 4). Bajra is pearl millet and kodra is also a millet, but kodra is not as well accepted locally as bajra. Bajra is a traditional food item consumed daily, and until recently it constituted a major part of the total quantity of grain intake. Bajra is gradually yielding its place of importance to rice and wheat.

The food items identified by the LP solutions are mainly traditional and popular among low-income groups. Their nutrient values differ, with wheat and rice having a higher protein content than bajra and kodra. The fresh vegetable category listed in table 4 consisted mainly of leafy vegetables that were locally produced The upper-income groups bought the superior varieties of vegetables more often than the lower-income groups. The fresh fruits included mainly local berries and guava. The total intake of food was less than one kilogram per day per adult in both the solutions. Thus, the LP solutions suggested that the weight and bulk of the diet necessary to meet nutritional requirements were reasonable and not beyond the ingestion capacity of the human stomach.

TABLE 5. Market prices, nutritional value, and shadow prices of non-basic food items

Items Market
Ranks on basis
of shadow prices
Rice 2.49 1.18 1.31 7
Bavta 1.37 1.17 0.20 12
Jowar 1.31 1.17 0.14 14
Wheat 1.66 1.50 0.16 13
Pulses 2.63 1.71 0.91 9
Milk 2.08 1.69 0.39 11
Ghee (fried butter) 20.36 2.28 18.09 1
Hydrogenated oil 9.09 2.30 6.82 2
Edible oil
(Peanut oil)
5.65 2.30 3.38 5
Raw sugar (gur) 2.49 1.47 1.02 8
Sugar 3.05 1.08 1.97 6
Mutton/meat 5.24 1.60 3.64 3
Eggs 4.16 1.19 2.97 4
Potatoes/onions 1.50 0.64 0.86 10

None of the animal products, not even milk, showed up in the LP solutions; even legumes, the major vegetarian source of protein and a part of the daily diet in most cases, did not show up. Foods that are popular and commonly accepted as protein-rich had relatively high market prices compared to their nutritional values; in other words, they were expensive sources of nutrients.


Contents - Previous - Next