Development has increasingly become concerned with improving the environment in which the world’s poor live, whether by upgrading the quality and range of services provided to them, or by improving their opportunities for engaging in productive activities. But as experience with these efforts accumulates, it has become increasingly obvious that if they are to be effective, they must be designed and implemented within a framework of understanding how the basic socioeconomic unit—the family functions among the poor. In the developing world, responsibilities and resources arc distributed among family members in ways which are quite different from those currently prevailing in Western developed countries. In many developing countries, for example, children make very important contributions from an early age to the family’s welfare through their labor and/or their cash earnings. Thus, in order for educational programs to be successful, parents must be provided with alternatives to these contributions. In many families too, resources, such as food, education, and health care are unevenly distributed—differentiations between male and female children, for example, are well documented. These patterns of distribution must be taken into account by planners attempting to meet the needs of specific beneficiaries.
Survival demands the productive work of women and children, whose contributions cannot be assessed in the absence of detailed data. Aggregate household income data are available for several countries but are not very useful since they tend to obscure important trends in the distribution of resources and opportunities within the family. There is evidence, for example, that higher total household incomes are not necessarily transformed into more food and better nutrition for each family member. The disaggregated income data available for some countries show how this type of information can provide a much clearer understanding of the link between who earns the income and what it is used for.
In defining low-income families in developing countries, coresidence of family members is not necessarily an important characteristic. Several studies show that men, women, and children are involved in different relations with kin who may or may not reside in the same household. In Ghana, for example, members of the same nuclear family may sleep in one household and eat in and contribute their labor to several households. Development policies, therefore, based on residential or conjugal family units take into account only a part of the network of relations between kin that are relevant for production, consumption, labor, and child rearing. These different patterns of family links have important implications for nutritional surveys and nutrition education, as well as for the understanding of the prevailing division and availability of labor in any one household.
Further, the organization of the family among the poor is characterized by flexibility. Families constrict or expand their membership to meet internal needs and changing socioeconomic conditions. In female-headed households in some Latin American and Caribbean societies, intermittent mating allows varying residence and kinship patterns, extensive bonds between different sets of kin, and the pooling of resources. This flexibility, which allows different sets of kin to cooperate, represents a viable adaptation to high unemployment. For the development planner it is therefore important to view the family as a flexible organizational unit of individuals who have different roles, privileges, rewards, options, and obligations.
Distribution within families
The criteria for distributing resources and responsibilities within families in developing countries do not coincide with the Western criteria of development needs and requirements. Nor do sex and age, especially among the poor, determine the roles people play in the family. Activities are not distributed so that women are consistently responsible for household and child care, while the men tend to agricultural and other productive activities. Women play important productive roles in developing countries, and a crucial role in the cultivation of all crops, subsistence and cash, in sub-Saharan Africa.
Children make important contributions as well. From the age of five or six, children of both sexes contribute to housework, child care, and production. In rural households in the Philippines, for example, children spend twice as much time as mothers and 80 per cent of the time spent by fathers working outside the home, and 50 per cent of the time spent by mothers and more than twice the time spent by fathers working inside the home. Children, particularly girls, in poor households must start work at a younger age and work for more hours than boys or children of both sexes in higher socioeconomic groups. These patterns have important consequences for boys’ and girls’ chances to attend and remain in school.
Despite the fact that all family members must work hard, scarce resources and immediate needs often result in unequal distribution of resources and access to services. Some family members are clearly valued over others in poor societies. One recent review of nutrition among low-income families concluded that, although the intake of calories and protein was often inadequate for all groups in low-income households, adults tended to fare better than children. Adult males, as traditional primary providers, and male children as potential providers, were generally fed more and often had greater access to education and health care than did other family members. Among adults and children, men and boys tended to have a better chance than women and girls of meeting a high percentage of their minimum calorie and protein requirements. A nutritional study of 17 villages in Morinda, Punjab (India) showed that nutrition was determined mainly by sex. By being breastfed longer and by being given more food after they were weaned, boys aged 6-24 months in all castes were better nourished than girls of the same age. The nutritional discrepancy was much larger in the lower, agricultural caste. Girls in the same caste were not only breast-fed for a shorter time period but also consumed less supplementary milk and less solid food, began consuming solid food later, and received less of each of the nutrients.
Multiple income earners
In developing countries, all members of low-income families are drawn into the struggle for survival, since no one source of income is adequate. Even when the rural poor have a small piece of land, they need to have additional nonfarm sources of income to protect themselves against such factors as bad crops and emergencies which might force them to sell land, livestock, or farming equipment. The International Labour Office report on Kenya estimated, for example, that in 1972 about 44 per cent of the peasant population needed additional income from nonfarm activities, and this 44 per cent did not include another 22 per cent of landless rural households. In Bangladesh, wives and children of small and marginal landowners work for wages in order to help the family accumulate resources and/or to be able to buy some more land. In the Philippines, wives’ market income averages about one third of that of husbands. These data, however, are based on households in different socioeconomic groups and tend to obscure the fact that in India and rural Bangladesh, for example, wives’ market income is twice as high in poor as in richer households and their income contributions are often as high as, if not higher than, those of their husbands.
Among the urban poor, male heads of households often earn less than 50 per cent of the total household income and secondary earners, such as women and children 15-19 years old, contribute the other half of the income—most often by working in the informal sector. In Belo Horizonte, Brazil, it was found, for example, that more than twice as many households would fall into the category of the poor if the additional income of secondary earners were not available.
This pattern of multiple incomes seems to explain why many more urban female-headed households are poorer than male-headed ones. In Belo Horizonte, for example, 81 per cent of female-headed households are either poor or low-income. In female-headed households, women are primary rather than secondary earners and there are only children as secondary earners, unless the households extend to include other working kin. The availability of nearby informal sector employment for women and 15-19 year-old children is an important factor in urban planning, for the urban poor cannot afford high transportation costs. The absence of an informal sector nearby explains why default rates have been high for relocation projects of urban slum dwellers in government housing at a distance from the city. The move out of the city and away from high-income neighborhoods eliminates for the poor, and especially for women, a variety of service and other informal sector jobs. Total household income therefore decreases significantly and many more households become impoverished. This has been documented for urban relocation schemes in Rio de Janeiro and New Delhi.
The family’s decision-making process is implicit in all its distribution of responsibilities and resources. The most powerful family member, the one who can orchestrate—if not dictate—decisions, is determined by prevailing cultural values, such as those attached to old age, sex, and the relative importance attached to men’s and women’s productive roles. In highly stratified and patriarchal societies, like those of North Africa, the Middle East, and most of South Asia, women are not expected to be self-supporting, have few opportunities to earn significant incomes, and cannot inherit land or other property. Even when women do earn, they are not usually able to translate their incomes into family power since they have to turn their earnings over to their husbands who control all family income. In these societies, a woman’s most reliable power base is her ability to reproduce—especially sons (a fact that has ominous implications for family planning).
Wives in poor households can have greater power if they are able to retain control of their earnings. A study of Bombay slum families found that married women working as domestics or small vendors used their income almost exclusively for family maintenance. Sixty per cent of them kept and controlled their income; the others had to turn part or all of it over to their husbands. What is not known is what determines whether or not these women are able to control their incomes.
In other societies where the patriarchal system is less rigid, where a woman’s active economic participation is common and accepted, or where matrilineal traditions exist, the decision-making process within the family can be markedly different. In Thailand, a country in which deliberate policies have led to a steady improvement in women’s status, rural women’s responsibility to feed and clothe the family is related to their ability to keep the money they earn and to have an important say in how it is spent.
In sub-Saharan Africa, where women play a major agricultural role, their power bases are broader, particularly in societies that are matrilineal. Women are usually allotted a piece of land on which to grow food for the family, they make all decisions concerning its cultivation, and have had the traditional right to sell the surplus of their crops and to control their earnings. This traditional economic responsibility of women to feed the family (and in many cases to pay school fees as well) and the ability to control the income they earn can be translated into some power in the family. Further, wives’ substantial labor contributions to their husbands’ land make men need their cooperation for the successful operation of the family’s farms. This economic interdependence between spouses gives wives a partner rather than a dependent status, which results in relatively egalitarian decision making in most farming matters.
Control of income explains why decisions regarding the quality and quantity of food for the family are not necessarily left up to women. They are tied, instead, to the spouse who is responsible for food and who has control over the income available for food. Where women have the responsibility of feeding the family throughout the year by cultivating subsistence crops and/or by marketing products, they do most or all of the shopping. Among the poor, however, in the Philippines or rural Bangladesh where the men are responsible for feeding the family and use their income or productive activities to do so, they are mainly responsible for buying the family’s food.
Disaggregated household income
Disaggregation of household income is important for two main reasons. First, it permits an assessment of how each family member’s income is used and how increases in each member’s income influence existing patterns of use and control of income, capital asset formation, consumption, nutritional patterns, and the distribution of labor and resources within the family.
Studies have documented, for example, that overall household income is not as significant a factor in the status of child nutrition as the income of the mother. Because women’s income is most often used to buy food, whether or not women are responsible for feeding the family, increases in this income tend to improve the quality and quantity of food, but increases in men’s income do not. Meanwhile, increases in men’s income tend to go to productive investments, consumer goods, or entertainment, and only occasionally to help out their wives with cash for food in case of emergencies. In rural Kerala (India), for example, it was found that increases in the maternal income of landless families were significantly associated with children’s nutritional status, this association being less strong during the slack (“hungry”) agricultural season when even aged grandmothers work and children do not receive adequate attention and well-prepared food. When, however, women did not work for wages, increments in husbands’ income did not seem to improve children’s nutrition.
The second main reason why disaggregated household income is important is that it allows planners to determine the impact of development programs on individual family members and on economic decision making within the family, which in turn affect availability of labor and food. Some rural development projects, for instance, in sub-Saharan countries, such as Benin, Cameroon, and Kenya, which helped intensify and commercialize cash crops, incorporated no provisions to safeguard women’s income, and this resulted in the curtailment of their income. Growing more cash crops demands more land and the best land as well as much greater labor contributions on the part of women. The increased labor demands on women made it difficult for them to cultivate their own small plot of land adequately and even more difficult to feed the family or produce a crop surplus. As a result, women became reluctant to contribute their labor and occasionally withdrew it, as they felt unhappy and rebellious as a result of their reduced status and their increased dependence on men. Such an increased dependence on men has serious demographic implications since women tend to want to have more children to help them out with the increased labor demands as well as to improve their status within the family.
Finally, disaggregated household income data show the significance of children’s (especially sons’) economic contributions to the family, beginning with adolescence and often totaling up to one third of the family income. The data may help us understand why poor families find it difficult to forego this income and send their children to school and why having many children under present conditions remains a rational choice for some poor families.
Understanding and taking family structure and dynamics into consideration is especially crucial in developing strategies aimed at alleviating poverty. Detailed data on the percentage and type of female-headed households can, for instance, pinpoint the need for programs designed to extend equal access to land, credit, agricultural and technical information and training, income-generating activities, housing, labor-saving technology, and other services and amenities to female- and male-headed households.
The most desirable long-term strategy is, of course, to increase the income available for food (or to make food cheaper), health care, and education so that goods and services are in sufficient supply and, therefore, their distribution need not be selective. Disaggregated income data for poor households will allow planners to determine whose income is used for food and whose income is used for capital asset formation. The fact that husbands’ and wives’ incomes are used in different ways can help in the development of strategies to increase both spouses’ incomes and thus meet the urgent needs of their families.
Another set of strategies can aim to change the bases for unequal resource distribution in the family. Preferential treatment of boys, for instance, can be counterbalanced by establishing the potential economic value of girls to their parents. Income-generating activities for adolescent unmarried women, especially activities that require literacy or primary education, can make a significant difference. Training and employment of young unmarried women as auxiliary health workers, teachers, aides, and agricultural extension auxiliaries can facilitate women’s access to health care, agricultural information and training, and education. It provides income before marriage, permits them to make a contribution to their family’s well-being, and often leads to a later age at marriage and therefore possibly to lower fertility. The eventual outcome may be a shift in the criteria for the distribution of resources based not on sex, but on need in the case of food and health care, and ability and interest in the case of education.
Whenever sociological as well as economic evaluations of development projects have been undertaken, they have showed how the neglect of one type of family structure (namely, female-headed households), or of family patterns and dynamics can be responsible for undesirable side effects or for a failure of measures to alleviate poverty. While more international agencies as well as private foundations are beginning to undertake such sociological evaluations and while this type of evaluation must become an essential component of all evaluations, the resulting information about behavior within the family is also grossly inadequate. Basic research is needed regarding family patterns and dynamics among the poor in developing countries so that we can obtain data to explain observed patterns of change and the effects of development efforts, and to guide planners in alleviating poverty.
Mayra Buvinic and Nadia H. Youssef, Women-Headed Households: The Ignored Factor in Development Planning, Report to AID/WID, International Center for Research on Women (Washington, 1978).
Mead Cain, et al., “Class, Patriarchy, and Women’s Work in Bangladesh,” Population and Development Review (September 1979), pp. 405-38.
R. E. Evenson, B. M. Popkin, and E. King-Quizon, “Nutrition, Work, and Demographic Behavior in Rural Philippine Households: A Synopsis of Several Laguna Household Studies,” New Haven: Yale University, Economic Growth Center (mimeo), 1979.
Constantina Safilios-Rothschild, “The Role of the Family in Development: A Neglected Aspect of Poverty,” background paper to the World Development Report, 1980, World Bank (January 1980).