Contents - Previous - Next

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

The significance of intra-household food distribution patterns in food programmes

Judit Katona-Apte


Nutrition surveys reveal that within the same income level and household structure some households contain malnourished members while others do not. Many causes have been attributed to this phenomenon, such as differences in the following: kin and social network, purchasing power, food preferences, demand for women's time, and so forth. All of these factors influence the household concerning its ability to acquire food, but they do not provide explanations for what happens to the food once it reaches the household. However, knowledge about the differential allocation of food within the household could provide insights as to which individuals benefit and to what degree. Just as little is known about the flow of food once it reaches the household, there is also no information available on how food programmes might affect it. Yet, food programmes that focus on the household are designed with specific goals in mind. Awareness of the differences in intrahousehold food distribution practices could provide insights into why the desired impact may not be achieved Knowledge of what the intra-household food distribution practices are could provide assistance in the planning and effective implementation of more programmes.

This paper presents and discusses a model of food flow within the household, explains the pathway of nutritional effect and the conveyance of nutritional benefits from food programmes, and concludes with an explanation of the interactions among them.


Figure 1 is a model of food flow to and within the household. Household here is defined as a group of individuals who share a domestic unit that pools its resources, especially for the purpose of sharing food. The model starts with the collective ability of the household to acquire food and lists some of the factors affecting it as well as those affecting the choice of foods acquired. These factors, however, will not be discussed. Instead, I will concentrate on factors involved in the pathway between acquired food and individual allotment, labelled as intrahousehold food distribution factors in figure 1. While the initial point in the model is the household, the final outcome is individual health and well-being, which is the ultimate aim of all programmes. Here, too, some of the factors that may interfere with the most positive outcome are noted but not discussed.

Knowledge of the factors that influence intra-household food distribution would aid in explaining the nutritional status quo and would provide useful information for the design and implementation of food programmes. Figure 1 shows the major factors believed to influence the amount of food each household member receives: the nature of the distribution system, variations in patterns of distribution, role and function of the person responsible for distribution, and other factors such as food preferences of household members, seasonal variations, and the influence of special events. While the factors themselves are not part of the model because of the importance only of their effects, it might still be worthwhile to discuss them briefly.

Nature of the Distribution System

There are several possible systems for dispensing food; for example, it could be distributed according to sex, age, and status, a common practice in many societies. In this system males usually get served first (1, 2) followed either by women or more likely children, as in India (2) or Peru (3). Non-family members of the household, such as hired labourers and servants are usually served last.

Another system is one in which wage-earners get a disproportionately larger share of the household's food supply than other members. Or, the decision-maker could distribute it according to how he/she perceives each member's need, which may or may not have any relationship to actual need. On the other hand, food may be dispensed in similar proportion to everyone's actual need. This is possible because activity levels adjust to energy intake; therefore, those with reduced energy intake are likely to expend less energy in activity and thus will have a decreased food need.

FIG. 1. Proposed Model of Food Flow in the Household

Households with marginal food availability in which wage earners are given preference use a system that is responsive to the best interest of the group, even at the expense of the vulnerable members, i.e. infants and pregnant/lactating women. If infants and women are not as healthy as wage earners, the effects on the household are not as devastating as when the wage earner's working ability is impaired (4).

Neither in all societies nor at all times are individual households able to have their food together. Work schedules, school attendance, and so forth may make it necessary that some household members eat at different times from others. In such situations the food available for each member may not be the same. For example, those who are present when the food is first available may have an advantage over those who eat later. Some foods, especially animal products, are known to spoil more easily than grain products. Therefore, those who leave early or get home late, such as children who attend school, may not get any of those foods that are prepared while they are away.

It is also possible that the food is not dispensed according to any system. In some households food is served and consumed from a common utensil as in parts of the Middle East, in which case cultural rules of manner operate; it might be perfectly acceptable for fast eaters to have an advantage over slow ones. It is also not uncommon for food to be put on the table and for all to partake of it at the same time. In this situation, not only speed but the size of the eating utensils has advantages; this lack of system could be detrimental to the food intake of young children. In some households food may be prepared and put aside during the day for all to eat from as they get hungry. This is common practice in the southern United States, where a container of cooked food is left on the stove all day for household members to consume as they wish (author's own research in rural Mississippi and rural North Carolina). In this case young children are usually fed by an older female household member.

Variations in Patterns of Distribution

Not all the foods consumed in the household may be dispensed according to the same system. Staple foods may be plentiful and therefore distributed freely, while other more nutrient-dense foods may be dispensed with greater care. In some societies a disproportionate share of meat may go to the wage earner, but dairy products or fruits may be fed mainly to children. Another common variation in the distributional pattern is one where the household consumes one formal meal a day together, with cooked food dispensed in a particular way, but the rest of the time household members consume whatever is available without being specifically allocated any particular amounts of food. In this situation children "snack" frequently and it may be a child's level of aggressiveness and the size of his/her palm that has the greatest impact on how much of certain foods he/she receives.

Role and Function of the Person Responsible for Distribution

The person who decides how the available food is to be distributed may be the dispenser, though the decision maker and the dispenser are not necessarily the same person. Usually the individual who dispenses the food is the female head of household, but the decision on who gets how much of each food may be made by another person such as the male head of household. Whether the decision-maker is a female or a male appears to make a difference in the amount of food allocated to children; women tend to be more aware of children's food needs than men are.

Dispensing is an important function because the person who dispenses the food has some control over its distribution. All foods may not be dispensed by the same person, or the same individual may not do it at all times. Note, for example, that on Sundays and other festive occasions in the United States it is customary for the male head of household to carve and distribute the meat while females dispense other foods at the same table.

Other Factors

There are many other factors that individually or collectively have an effect on how the food is distributed within a specific household. For example, the food preferences of household members could be taken into account. Some foods are liked more by certain family members than others, and the person responsible for deciding what foods will be served can control the amounts consumed by keeping individual food preferences in mind.

Seasonal variations could also play a role. Both the type and the amount of food available may be affected by the change in seasons and may have an effect on the way food is distributed in the household. In addition, activity levels and work schedules may also change with seasons, thereby affecting the food dispensing pattern.

Festive occasions could change everyday food distribution patterns. As nutrient-dense foods tend to be more abundant at such events, all household members may receive adequate amounts of them. Since many nutrients can be stored, some of the results of everyday inequalities of food distribution are alleviated in societies with frequent special occasions.


Food aid programme planners have presumed that the food provided in the context of programmes was consumed totally in addition to the recipients" usual diet. More recent data, however, indicate that food programmes usually reach households that, due to inadequate resources, have an insufficient food supply. This means that the food provided is not necessarily consumed totally in addition to foods usually available, but replaces some of the foods normally purchased (5-7).

Figure 2 is a simplified illustration of the pathway of nutritional effect from food programmes. There are two possible options for the food to exert an effect (8). The direct pathway implies that the food is consumed in addition to the usual diet or to what would be available without any food aid. The nutritional effect results from the actual consumption of the food, either by the household (arrow to acquired food) or by the individual (arrow to individual allotment).

The other option is for the recipients to substitute the food provided by the programme for the same or similar foods they usually purchase. The nutritional impact then is an income-mediated one and depends on how the recipients choose to spend the additional income, i.e., what proportion will be spent on food and what foods will be purchased. There is reason to believe that the foods purchased as a result of increased income are more nutrient dense than the ones being replaced (9,10).

The nutritional effect of most food programmes is achieved through both pathways at the same time: The food provided is consumed -direct pathway - making it unnecessary to purchase as much as before, thus freeing up some income for more and better foods-indirect pathway. However, if all the food received is sold, or is identical to what is usually purchased, then only the indirect pathway is operating; and if all of it is consumed in addition to the usual diet, then only the direct pathway is in effect. The latter is most likely to be the case in emergency or disaster relief-type programmes.

FIG. 2. Pathways of Nutritional Effect from Food Programmes. (1) Effect from income transfer potential of programme. (2) Effect from substitution of food with higher nutritional value than that in present diet. (3) Effect by same means as in pathway 2 but with programme food targeted to specific individuals in the household.



The means through which nutritional benefits are conveyed can be classified into two types of programmes: targeted and untargeted. Targeting is an attempt to increase the energy and/or nutrient intake of specific members of the household who are perceived to be at greater risk of nutritional deficiency than other individuals, i.e., infants, young children, pregnant women, and lactating mothers.

In fact, programme planners invest both labour and expense in attempts to reach these individuals, though it is "not clear why so much effort is spent on actually trying to reach the individuals within these households unless there is some concern that household decision making does not reflect the same desire to reach these individuals as that which motivates government . . . " (4, p. 18, italics mine). Untargeted programmes are those from which specific nutritional impacts are not expected to result.

Many different permutations and definitions of the variety of food programmes, both targeted and untargeted, are possible. However, for the purposes intended here, the following classification will be appropriate to illustrate the relevant points to be discussed later.

Untargeted Programmes

Take-home food packages. These are provided to recipients for consumption by the household at a future time. Examples of this type of programme are food-for-work projects in developing countries or the United States commodity programme for Indian reservations.

Entitlement programmes. These programmes specify criteria for participation and either distribute coupons that can be exchanged for food items, such as in food stamp programmes, or provide specific amounts of specific foods or categories of foods, as in "ration" or "fair price" shops.

The pathway of nutritional effect in untargeted programmes is mainly indirect, since the nutritional benefit depends on how many extra or new foods are brought into the household. There may also be a direct nutritional benefit if the distributed food is nutritionally superior to the food it replaces. The magnitude of effect depends on how the food is distributed in the household. If all of it goes to individuals whose nutritional needs are already met, then the effect is nullified.

Targeted Programmes

On-site feeding of designated beneficiary. In this type of programme the food is prepared and consumed on the premises of a child-care facility, health clinic, civic centre, and so forth under the auspicies of an agency. Examples of such programmes are school feeding, soup kitchens, and institutional feeding.

Take-home food packages. Food packages can be targeted either to specific classes of individuals such as infants or to whole households. The food, if properly used, is supposed to achieve a stated nutritional impact such as increased growth, eradication of anaemia, decreased incidence of infections, and so forth.

Entitlement programmes. These are similar to the above, but instead of a food package the recipient is provided with vouchers or coupons to be exchanged for specific types of foods to be given to designated members of the household. The supplementary feeding programme for women, infants, and children (WIC) in the United States anti many milk programmes in other countries are examples.

In targeted programmes the pathway of nutritional effect is meant to be direct, and the targeted individual's nutritional status is expected to improve as a result of the food provided. However, the extent of the effect depends on a number of factors related to the distribution of food within the household, as will be seen in the next section. Targeted programmes also have indirect effects because the targeted individual rarely consumes the intended food totally in addition to what the household provides for him/ her. Therefore there is some displacement of foods that would go to the target in the absence of a programme, and these displaced foods do not have to be provided for in the household budget.

It is important to note that the classificatory categories listed above are not mutually exclusive. It is possible, for example, (a) to have a targeted entitlement programme in which the food is provided on site, as in the United States school lunch programme, (b) to combine on-site feeding five days a week in a maternal and child health centre with a targeted take-home food package for weekends, or (c) to design a take-home food package with both targeted and untargeted components.

But who in the household benefits and to what degree as a result of food programmes? This question cannot be answered until more is known about the intra-household distribution of the food supply. The rest of this paper will concentrate on illustrating the importance of acquiring knowledge about intra-household distribution of food in the context of programmes.

Problems of Conveying Nutritional Benefits from Food Programmes within the Household

This section will elaborate on some of the problems that could interfere with the magnitude of effect from food programmes. Both the pathway of nutritional effect and the conveyance of nutritional benefits have been discussed. The problems that may result from not being aware of household level activities regarding food allocation for different types of programmes will be discussed below.

Take-home food package

These programmes are most popular in situations where food is provided either as an incentive for some other purpose-such as for people to bring their children to a health centre for routine care, or to attend nutrition education classes, or to reforest their land-or as part wages for work performed. There are usually no choices in either type or amount of food received, though the content of the food package may change periodically. At times unfamiliar foods are used such as soy-fortified analogues, staples, dried milk powder, lyophilized products, and so forth. The effect of such programmes is usually indirect, but in certain situations it could also be direct, as when soy-fortified flours or other high-protein foods replace foods of lesser nutritional quality, for example, when soy-fortified corn meal replaces regular corn meal or non-fat dried milk replaces tea in a household's diet.

The indirect effect of this programme is increased purchasing power. However, several questions arise regarding the outcome of such an innovation. How does it affect each household member's nutritional intake? How is the new food supply allocated among household members? If, for example, more of the expensive but less nutritious staple is purchased as a result of the programme, such as more rice but less corn, who gets how much of each with what final effect? if the extra income is spent on more nutritious foods, such as animal products or fruits, how is that divided within the household? Some household members may not have improved diets in spite of the programme, while others could suddenly meet and exceed their needs.

Entitlement programmes

Entitlement programmes have become more popular in recent years. They operate on the principle that persons or households who meet certain criteria are entitled to a minimum amount of food or the ability to acquire such food. These programmes have an indirect effect, as in food stamps, since the aid is provided as cash, coupons, ration cards, and so forth, to be exchanged for food. Often there is an additional direct effect if such a programme is an on-site feeding one.

Without knowledge of intra-household food distribution patterns it is impossible to predict what changes in household allocation of resources such programmes might initiate in terms of food, and therefore impossible to arrive at conclusions about the possible nutritional consequences to household members.

On-site feeding of designated beneficiaries

There are many variations of this type of programme. The purpose is usually to increase the amount of food or specific nutrient(s) the proposed beneficiary should consume. The effect of such programmes is both direct and indirect.

Unfortunately, there is little information on how the household reacts to the fact that one or more of its members is consuming a free meal or free food away from home. How is this taken into account? For example, does a child who receives milk in school then receive no milk at home? Or only less milk? Or the same as before? Or, if there was no milk given at home to begin with, does the child receive less of other nutrient-dense foods?

If a pregnant woman receives a meal at a health facility, does that replace the meal she would have consumed at home, or does it also displace other meals and/or foods? It is possible that instead of a person's total intake increasing as a result of the feeding programme, it could decrease because of the adjustment made in the household.

Targeted take-home food package

These programmes provide a food package with the stipulation that a specific part, or possibly the whole package, should be fed to the targeted individuals only. Many maternal and child health programmes are this type. It has both direct and indirect effects: consuming the targeted food should increase the energy and/or nutrient intake of the proposed beneficiaries and should release income that had been spent on feeding these individuals. In the case of an infant food supplement or milk this could be quite a large amount of money that is now released for spending on other foods and goods.

However, the results could be quite different from those intended, depending on what is happening in the household (11, 12). For example, the goal of the programme could be to increase protein intake for the preschool-age child using an expensive blended food such as corn-soy-milk (CSM). However, if the distributed food is shared by all members of the household, only a small portion may reach the targeted child. In that case the programme becomes an expensive and inefficient attempt at increasing a targeted individual's protein intake.

If the programme is targeted at the household and its purpose is to increase protein consumption for the total household, then the use of high protein foods in the package is an appropriate method for achieving this. Depending on the intra-household allocation of food resources, however, the results might be different from those intended. If non-fat dried milk, for example, is distributed to increase a household's protein intake, the milk may be considered acceptable only to children (13) and fed to them alone. Not only does the programme not achieve its purpose because all members of the household do not partake of the food, but it could even be dangerous if Young children consume large quantities of food with such a high protein-to-energy ratio.

Targeted entitlement programmes

These programmes specify the household member and the food that he/she is entitled to. The WIC programmes mentioned earlier are an example of targeted entitlement. This programme provides vouchers to be exchanged for specific nutrient-dense foods with some choice of brand names, flavours, types, and so forth within a category; for example, dairy products could be chosen as either milk or cheese, fruit juice could be cranberry, orange, grape, and so forth. The effect of this programme is meant to be direct since the special nutrient-dense foods are aimed at the preschool-age group and at pregnant women. However, lack of knowledge of what happens in the household of recipients makes it difficult to establish how the targeted food is consumed. For instance, older children tend to be more vocal about their food preferences than younger ones, and therefore more of the nutrient-dense foods may be brought into the house for their benefit.


The focus of this paper has been on the phenomenon of intrahousehold food distribution and its numerous aspects because it was felt that such an input has been consistently absent from the design, implementation, and evaluation of food programmes. The questions that need to be answered regarding intra-household food distribution were discussed along with some hypothetical answers to illustrate how such information would affect the outcome of food programmes. Combined with an awareness of the differences in the pathway of nutritional effect and of the different modes of conveying nutritional benefits, information regarding intrahousehold food distribution could have significant implications for the planning of food programmes. Table 1 summarizes the interaction among all the relevant components of the theoretical framework proposed in the paper. Many other aspects, such as how the resources are allocated for food versus other items, or how much of the income transfer resulting from food programmes is spent on food, though important, were considered to be outside the scope of this paper.

Acquiring the information is not as expensive or time consuming as it may appear. Some data are already available by way of ethnographic information gathered by anthropologists in specific cultures on such aspects as order of dispensing foods (2, 14, 15); foods appropriate to specific sex, physiological state, or status (16, 17); foods prepared for special events (18, 19); and so forth. In addition to any available information, a social scientist could collect the specific data needed in a short time at little cost. The cost-effectiveness of such an undertaking is self-evident.

TABLE 1. Interaction between the Conveyance of Nutritional Benefits, Pathway of Nutritional Effect, and Intra-household Behaviour

Conveyance of
Nutritional Benefits
Pathway of
Summary of Questions on Changes in
Household with Programme Goal in Mind
Untargeted programmes
All programmes Indirect How are extra and/or new foods distributed?
Take-home packages Direct Which members receive how much of what foods from those distributed?
Targeted programmes
On-site feeding Direct What proportions of targeted individuals' nutritional needs are met?
Take-home packages    
individual is beneficiary Direct How much of intended food does targeted individual receive?
household is beneficiary Direct Do all members partake in proportion to their needs?
Entitlement programmes Direct Which foods does target receive from those intended and in what amounts?
All programmes Indirect Which foods are displaced from those usually allocated to the beneficiary and in what amounts?

The policy implications of the above are clear. When information is available on what actually happens with regard to food distribution at the household level, more efficient programmes in terms of conveying the desired nutritional benefits at the least possible cost can be planned, implemented, and evaluated.


1. C. Chen, E. Huq, and S. D'Souza. "Sex Bias in the Family Allocation of Food and Health Care in Rural Bangladesh," Population Dev. Rev., 155 (1981).

2. J. Katona-Apte, "The Relevance of Nourishment to the Reproductive Cycle of the Female in India," in Dana Raphael, ea., Being Female: Reproduction, Power, and Change (Mouton, the Hague, 1975).

3. L. Huenemann and C. Collazos, "Nutrition and Care of Young Children in Peru," J. Amer. Diet. Assoc., 30: 559119541

4. Per Pinstrup-Andersen, "Food Policy, Household Behavior and Nutrition,'' (paper prepared for the fourth biennial meeting of the Agricultural Economics Society of Southeast Asia, Singapore, 1981).

5. Mary Ann Anderson, "CARE Preschool Nutrition Project," (CARE, New York, 19771.

6. M. G. Herrera, J. O. Mora, N. Christiansen, N. Ortiz, J. Clement, L. Vuori, D. Weber, B. de Paredes, and M. Wagner, "Effects of Nutritional Supplementation and Early Education on Physical and Cognitive Development," in R. R. Turner and F. Reese, eds., Life-Span Developmental Psychology: Interventions (Academic Press, New York, 19801.

7, J. O. Mora, J. Clement, N. Christiansen, J. Suescun, M. Wagner, and M. G. Herrera, "Nutritional Supplementation and the Outcome of Pregnancy: II I. Prenatal and Neonatal Mortality," Nutr. Rep. Int., 18:167 (1978).

8. S. Reutlinger and J. Katona-Apte, "The Nutritional Impact of Food Aid: Criteria for the Selection of Cost-Effective Foods," World Bank Report no. ARU 12. (World Bank, Washington, D.C. 1983).

9. M. Baertle, E. Morales, G. Verastegui, and G. Graham, "Diet Supplementation for Entire Communities," Amer. J. C/in. Nutr., 23:207 (1970).

10. C. Overholt, S. Sellers, J. Mora, B. de Paredes, and M. Herrera, "The Effects of Nutritional Supplementation on the Diets of Low-Income Families at Risk of Malnutrition," Amer. J. Clin. Nutr., 36: 1153 (1982).

11. George H. Beaton and M. Ghassemi, "Supplementary Feeding Programs for Young Children in Developing Countries," Amer. J. Clin. Nutr., 35: 864 (1982).

12. O. Mora, B. de Paredes, M. Wagner, N. Christiansen, and M.G. Herrera, "Nutritional Supplementation and the Outcome of Pregnancy," Amer. J. Clin. Nutr., 32:455 (1979).

13. M. Flores, B. Garcia, Z. Flores, and M. Y. Lara, "Annual Patterns of Family and Children's Diet in Three Guatemalan Indian Communities," Brit. J. Nutr., 18: 281 (1964).

14. Russell Sage Foundation Culture Program, Project for Cultural and Class Variables Underlying Food Habits, "Gastronomic Categories" (Russell Sage Foundation, New York, manuscript).

15. D. R. Gross and B. A. Underwood, "Technological Change and Caloric Costs: Sisal Agriculture in Northeastern Brazil," Amer. Anthrop., 73:724 (1971).

16. J. Katona-Apte, "The Socio-cultural Aspects of Food Avoidance in a Low-Income Population in Tamilnad, South India," J. Trop. Ped. and Environ. Child Hlth., Apr. 1977, p. 83.

17. R. W. Grant, "Nutrition and Health of Gold Coast Children: 2. Care and Physical Status of Children," J. Amer., Diet. Assoc., 31: 694 (1955).

18. J. Katona-Apte, "Dietary Aspects of Acculturation: Meals, Feasts, and Fasts in a Minority Community in South Asia," In Margaret Arnott, ea., Gastronomy: The Anthropology of Food Habits (Mouton, The Hague 1976).

19. Richard Lee, "Eating Christmas in the Kalahari," Natural History, Dec. 1969, p. 14.

Contents - Previous - Next