Contents - Previous - Next

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

Effects of food policy on intrahousehold food distribution in Bangladesh

Najma Rizvi
International Centre for Diarrhoeal Disease Research, Dhaka, Bangladesh


Bangladesh, a country of more than 90 million people living in an area of about 55,000 square miles, is a food deficit country. Therefore, achieving self-sufficiency in food has always been a national concern. The need for having a food policy has been recognized for a long time, but it was not until the Second Five Year Plan that a food policy known as the Food Security Plan was developed by the Ministry of Food in August 1980. The main goal of this food security plan is to ensure people the minimum desirable level of consumption.

The present paper will discuss the major characteristics of Bangladesh's food policy and show its effects on household food availability and the food distribution pattern. The approach of both economists and anthropologists and their different perspectives in studying food-related behaviour will be discussed in reference to food policy formulation, and the need for having a broader perspective in food policy formulation will be emphasized.


The Food Security Plan was formulated on the basis of per capita food need and the gap between production and need, estimated to be 1.5 million tons of rice and wheat. The major thrust of the food policy rests on increasing cereal production from 13.4 million tons in 1979/80 to 20 million tons in 1984/85. This is to be achieved by expanding the high-yielding variety (HYV) rice acreage from 3.08 million to 7.9 million acres and wheat acreage from 0.99 million to 2.50 million acres during this same period. Since increasing HYV rice and wheat acreage/ production will require greater availability of chemical fertilizers and irrigation facilities, the Government has made attempts to ensure adequate and prompt supply of these agricultural inputs at competitive prices.

In addition to stimulating increased cereal production, the Food Security Plan involves procurement of food at a price that gives incentive to farmers and an open market sale of rice (also wheat) during lean periods to alleviate market pressure. The resale of procured grain in the open market is believed to bring down the price level, making the grain easily available to the great majority of landless customers. Along with the open market sale of procured grain, the Government plans to reduce its subsidy to the statutory ration system operating in the city so that it becomes less attractive to the well-to-do population who have been the major recipients of this rationing. The withdrawal of rice from statutory rationing will increase government stock for open market sale and make food (mainly wheat) available for a public distribution system aimed at the welfare of the poorer class.

Other strategies used to help feed the poor include (a) modified rationing, (b) distribution of relief foods for the disadvantaged, and (c) food for work. Modified rationing is available to rural households only. The existing tax classification is used as the criterion for selecting households. The rural households belonging to groups A and B, who pay no tax or pay a yearly union tax of up to 3 takes (20 takes = US$1), are eligible to draw rice, sugar, and oil, whereas groups C and D, paying 3 to 8 takes or more, can buy only sugar and oil from the ration. The price of rice and wheat is approximately 6.50 takes per seer (1 kilogram), and wheat flour, including milling costs, comes to about 4.35 takes per seer. Wheat is more frequently available than rice.

While foods distributed through modified ration have to be purchased, relief foods are free, mainly to households headed by women and widows who have little or no other financial support.

Food for work, another arm of the public food distribution system, which receives food under Title III (PL 480), pays food as wages for work related to development projects such as building roads, dams, and other public works. A worker generally receives 15 seers of wheat for a week's work, i.e., 40 hours of heavy manual labour. Like other wage work, food for work is temporary and therefore does not ensure a continuous supply of wheat.

TABLE 1. Rice, Wheat, and Pulse Acreage (in thousands of acres) and Production ( in thousands of tons, in Bangladesh, 1971 - 1981.

  Rice Wheat Pulses
Acreage Production Production* Acreage Production
1971/72 22,975 9,775 - 888 281
1972/73 23,795 9,930 - 777 222
1973/74 24,410 11,721 - 704 209
1974/75 24,196 11,108 - 765 223
1975/76 23,974 10,700 - 754 220
1976/77 25,525 12,561 93 823 230
1977/78 24,420 11,567 310 835 238
1978/79 24,779 12,763 437 838 226
1979/80 25,105 12,539 729 819 214
1980/81 25,474 13,662 1,350 804 208

* No data on acreage available
Sources: Statistical Yearbook of Bangladesh, 1981; Administrative records of the Ministry of Food and Monthly Statistical Bulletin

TABLE 2. Per Capita Food Intake by Source. 1962-1964 and 1975-1976




g per person
per day
% of total
g per person
per day
% of total
Cereals 545.8 62 523.0 65 - 4.3
Animal protein 56.5 6 44.0 - 5 - 22.1
Vegetables and pulses 283.7 32 240.3 30 - 15.3

TABLE 3. Per Capita Nutrient Intake in Bangladesh, 1962-1964 and 1975-1976

  1962-1964 1975-1976
Calories (kcal) 2,301 2,094
Protein (g) 57.9 58.5
Fat (g) 15.8 12.2
Vitamin A (IU) 1,870 730

To summarize, the Food Security Plan operates under the assumption that by increasing food production and price support policy and public food distribution it can meet the food needs of its people. Given such a policy, it is important to see what effect it has on household food availability and on the distribution system.


Food available to a household is dependent on what the household grows, its purchasing power, and food preferences. The great majority of Bangladesh villagers are landless (more than 51 per cent, and in the study area more than 60 per cent, neither have nor own land to grow their major staple) and therefore suffer from an inadequate food supply in terms of both quantity and quality. Although in recent years there has been an increase in cereal production (table 1), this increase has not necessarily been translated into feeding those who suffer from chronic food deficiency. This situation is not unique to Bangladesh. Fleuret and Fleuret (1), Linowitz et al. (2), and Teller and co-workers (3) report that much of the world population's food and nutritional status remains far from satisfactory. According to the Nutritional Survey of 1975 - 1976,60 per cent of rural households were deficient in calories and 30 per cent were deficient in protein. Tables 2 and 3 compare per capita food and nutrient intakes for 1962- 1964 and 1975 - 1976 in Bangladesh. The Nutrition Survey of 1981 - 1982 (yet to be published) estimates a 10 per cent decrease in per capita food intake since 1976.


Little is known about the distribution of food within the family. The presence of inequitable food distribution in developing countries has been pointed out by a few authors (4,5). In a review article published in the FAO Newsletter of 1972, den Hartog gave an overview of inequality in food distribution. In discussing this topic, some authors have cited sex bias as an important factor in the aetiology of malnutrition among women and children. While anthropologists have mainly been concerned with studying food preferences and restrictions, nutritionists and occasionally anthropologists have reported preferential treatment of adult males in intra-household food distribution.

The Food Security Plan of Bangladesh has not taken into account the inequality present in the community and at the household level between different age and sex groups. The presence of a vulnerable group is recognized, but such a definition does not include women and children as especially disadvantaged groups who suffer from serious calorie and other nutrient deficiencies. Before discussing the impact of food policy on intra-household food distribution, it would be useful to examine the factors affecting food intake at the intra-household level.


The type and quantity of food available to a household with no major food base depends on income and budget allocation. An in-depth, careful examination of the income and food purchase pattern in a group of 25 rural households has revealed that a low-income rural household in Bangladesh spends over 90 per cent of its income on food only, and when the price of fuel is included, the household runs on a deficit budget (6). So, the accusation often made that poor households do not know how to allocate income seems unfounded. The purchase of rice and other alternative cereals constitutes the major expenditure in poor households. The lack of employment opportunities, the low rate of wages in bonded labour, and the limited supply of food grain received in food for work programmes reduce total food availability and naturally affect food distribution and consumption patterns of household members.

Contrary to present belief that the size of the household affects the food intake of the individual, we found that size per se does not seem to affect food allocation. A small household is not necessarily supplied with adequate food, nor does a large household necessarily have an inadequate supply. Household size becomes important when it is viewed in relation to income.


Intra-household food distribution cannot be understood without referring to the different status of males and females in the context of cultural norms in Bangladesh. The distinction between male and female is articulated through a set of behaviours that include preferential treatment of the husband and other adult males and sacrifice by the wife/mother in the distribution of food to members of the household. For young children zero to five years of age, being male or female does not seem to affect food allocation. Because of the belief that early introduction of solid food (rice and other adult foods) might cause gastrointestinal problems and a "pot belly," children below the age of two years do not consume any appreciable amount of family food. Mothers do not make any distinction between male and female children during breast feeding or serving of family food. However, once a young girl is expected to learn the social norms dictating her inferior position, the mother tries to instill in her the value of the sacrifice necessary for her to become an ideal wife and mother.

From the time of puberty, a girl is expected to be a near perfect approximation of an idealized wife and mother. It is from this time that a female child becomes least demanding and consequently receives a smaller allocation of food. It needs to be emphasized here that, while this custom of considering women as the epitome of sacrifice is emphasized in all socio-economic groups, adherence is strongest in the low-income group where a limited supply of food leads to greater inequality.

The intra-household food distribution pattern cannot be understood in terms of economic or cultural factors alone. The income or the purchasing power that determines total food availability and variety or lack of it cannot be ignored in any discussion of food allocation at the household level. Similarly, a knowledge of cultural factors relating to food beliefs, food preferences, and cultural norms and values is also necessary.


The Food Security Plan was designed to meet the food needs of people at a minimum desirable level of consumption. Since per capita figures have been used in estimating food requirements, the inequality present in the distribution of food at the inter- and intra-household level has received very little attention. The economists who planned the Food Security Plan are aware of the inequality in distribution of land and income, but in formulating food policy this was not given due consideration. Rather, they operated under the assumption that increasing cereal production and encouraging the open market sale of cereal grain, coupled with a public food distribution system aimed at meeting the needs of the poorer classes, would be able to alleviate the problems of the poor and ensure a minimum level of consumption.

The impact of food policy on meeting the food needs of poor people has been minimal. Increased food production has not resulted in closing the food gap between rich and poor. On the other hand, increased production has enabled well-to-do landowners to hold off sale at the post-harvest season and release the grain later for sale at a higher price. Government procurement and resale have not been too successful either, as it has been difficult for the Government to procure enough rice and sell it at a price affordable by the poor.

Recent study has shown that the modified ration system has been operating at a level that has no substantial impact on household food supplies. The supply was inadequate; finding a private ration dealer, which demands a huge investment, was difficult, delaying distribution; the information system announcing that the food supply was ready for distribution was poor; and the time and money spent on transportation and the wages lost through travelling to the pick-up point did not make it a profitable venture. Moreover, the grain was often supplied at inadequate intervals. Theoretically, a modified ration should be distributed monthly to the eligible households, but in practice distribution has been highly irregular. The majority of households reported getting the ration once in three months, or even less frequently. Distribution of relief foods was also far from adequate. The food for work programme demands heavy manual labour and pays only 15 seers' worth of wheat for a week's labour, which is even less than the daily wage rate prevailing in the area. Moreover, such work is purely temporary; it may last only one or two months.

As is evident, the various strategies used in meeting the needs of the poor have been far from satisfactory. As pointed out earlier, intrahousehold food distribution is affected by income and food availability as well as cultural factors. It is, therefore, axiomatic that food policy having limited effect in ensuring minimum food supply will have no positive effect on household food distribution.

On the other hand, limited food availability has further reduced the share of food for women, including pregnant and lactating mothers. Determination of food needs has not taken the increased requirements of this vulnerable group into account, nor has any attention been paid to the food requirements of children. Food needs have been calculated only for calories; therefore, the requirements for protein and other nutrients were not considered. In view of this, the name "Food Security Plan" may be more justified than national food and nutrition policy. In a recent communication with a senior economist, I learned that planners are aware of the lack of attention to the caloric and nutrient needs of women and children, and they intend to use nutrition survey data in revising their food policy.

Increased cereal production has led to an increase in supplies on the market, but has had a negative effect on the household food supply of underemployed landless day labourers. Moreover, with the recent expansion of rice mills in rural areas of Bangladesh, unhusked paddy is no longer sold in village markets at a low price because the paddy is bought by the mill owners as soon as the crop becomes available for sale. This means that low-income landless households can no longer take advantage of the harvesting season and buy paddy at a lower price and process it at home. The displacement of women from home-processing of rice has also reduced family income, which, in turn, has a negative effect on household purchasing power.

The lack of pulse production and the transfer of pulse acreage into HYV rice has resulted in a continuous decrease in the supply of pulses and a rise in prices. The market price of a seer of pulse varies between 9 and 14 takes, which has greatly reduced the quantity consumed. It no longer forms a regular accompaniment to the rice-based diet of Bangalees. In poor homes, when the fish supply was insufficient, women used to sacrifice their share of fish and eat rice with pulses and other inexpensive greens, so that protein deprivation for women and girls was not as great as it is today.

The trade policies, particularly the export of frozen shrimp and transport of fish from the rural areas to the city markets, has reduced the supply of fish in village markets and in turn raised the price to a level unaffordable to the poor.

The unrestricted importation of baby formula and milk powder has led to the desire to use such foods. This practice is responsible for undermining the confidence of mothers in the food value of breast milk. Nutrition intervention programmes that involved distribution of donated milk powder and other baby foods (pureed vegetables, fruits, and gelatinous, starchy desserts) did little to improve the nutritional status of children, but did lead the mother to believe that such foods were of superior quality. Poor mothers in urban slums over-diluted the milk powder, to the detriment of their infants.

TABLE 4. Food Intake of Mothers in a Village (per Day), Assessed by 24-Hour Dietary Recall Method, Compared with Requirements

  Intake Requirement*
Energy (kcal) 1,062.6 1,900
Protein (g) 23.8 42.6

N = 10
* Requirements do not include extra calories and protein needed during pregnancy and lactation.

Although inequitable allocation of food resources at the household level cannot be attributed to insufficient food supply alone, there is no doubt that food scarcity leads to increased sex discrimination in who eats more, as reported in a number of studies (7 - 10). The food policy in Bangladesh focused on increasing food production, which gives token attention to inequality in the distribution of resources and operates with little or no in-depth knowledge of household food behaviour, and thus has had little success in alleviating the serious food and nutritional deprivation of the poor, particularly women. A quantitative assessment of the food intake of mothers in landless homes revealed that both calorie and protein intake were grossly inadequate (table 4).


To formulate an effective food policy, the collaborative input of economists, anthropologists, and nutritionists is needed. The different perspectives of these disciplines can help to diagnose the extent and severity of the problem at national, community, and household levels, and thus help formulate an effective food and nutrition policy.

However, it must be recognized by economists and policy makers that an economic data base derived mainly from macro studies needs to be supplemented with in-depth anthropological studies of household food behaviour. The complex set of factors affecting household food behaviour can be unraveled only through in-depth observation and intimate contact with household members, particularly mothers. The rationale used by women in giving up their share of quality food and limiting their intake of staples in times of scarcity will otherwise be ignored. Nor will economists/policy-makers know the agony suffered by women when there is no staple (rice/wheat flour) to feed their families. Although men are the major providers of food, it is the women who have to worry about what to cook for the next meal and how to serve their husbands and children. The responsibility for borrowing food from neighbors and kinfolk rests on them, and it is they who suffer from the embarrassment of such tasks. While doing my field work in Bangladesh, on many occasions I heard women saying, "It is getting dark, time for my husband to get back from work, and the children are asking for food, but I don't have any rice for the next meal. I've already borrowed yesterday from 8.1 don't know whom to approach next. I'm really embarrassed to do it."

I hope this paper has shown why collaborative efforts are essential in formulating an effective nutrition policy. In a country like Bangladesh, and many other Third World countries, where economists enjoy a privileged position in the formulation of food and other policies, it may not be easy for anthropologists to convince authorities/planners about the necessity of integrating economists', anthropologists', and nutritionists' approaches and perspectives in policy formulation. The attitude prevailing in many countries, that only economists have the expertise for policy formulation, need not discourage anthropologists from underscoring the contribution they can make in various types of planning and policy formulation. It is time for anthropologists to move away from the stance of neutrality and make an effort to become involved in alleviating the suffering of people.


1. P. Fleuret and A. Fleuret, "Nutrition Consumption and Agricultural Change," Human Organization 39 (3): 250 ( 1 980).

2. S. M. Linowitz and the Presidential Commission Members, "Preliminary Report of the Presidential Commission on World Hunger" (Washington, D.C., 1979).

3. C. R. Teller, B. Sibrian, C. Talavera, V. Brent, J. Del Canto, and L. Saenz, "Population and Nutrition: Implications of Socio-demographic Trends and Differentials for Food and Nutrition Policy in Central America and Panama," Ecol. Food Nutr., 8:95 (1979).

4. S. A. Taha, "Household Food Consumption in Five Villages in the Sudan," Ecol. Food Nutr., 7 13711978).

5. L. C. Chen, E. Huq, and S. D'Souza, "Sex Bias in the Family Allocation of Food and Health Care in Rural Bangladesh," Pop. Dev. Rev., 7:55 (1981).

6. N. Rizvi, "Rural and Urban Food Behavior in Bangladesh: An Anthropological Perspective to the Problem of Malnutrition" (Ph.D. dissertation, University of California, Los Angeles, 1979).

7. J. F. Levinson, Morinda: An Economic Analysis of Malnutrition among Young Children in Rural India, Cornell/MIT International Nutrition Policy Series (Cambridge, Mass., USA, 1974).

8. M. G. M. Rowland, A. A. Paul, A. M. Prentice, R. A. E. Barrell, and R. G. Whitehead, "Seasonal Aspects of Factors Relating to Infant Growth in a Rural Gambian Village" (paper presented at a conference on Seasonal Aspects of Rural Poverty, University of Sussex, Brighton, England, 3-6 July 1978).

9. A. S. Carloni, "Sex Disparities in Distribution of Food within Rural Households," Food and Nutr. (FAO), 7: 3 (1981).

10. N. Rizvi, "Socioeconomic and Cultural Factors Affecting Intra-household Food Distribution in Bangladesh" (paper presented at the American Anthropological Meeting, Los Angeles, 1981).

Contents - Previous - Next