One major recent change in family/household structure that has attracted much comment has been the reported rise in the proportion of households headed by women. The reasons for this increase, like its rate and magnitude, are diverse, but it is a trend that has been noted for many different countries in the world at varying patterns of economic development.
It is evident that the number of female-headed households is related to marriage strategies, property and inheritance transfers, as well as the intersection between production systems and the reproduction of labour. This means that it is unwise to treat female headship as a unitary phenomenon. The definition of headship itself complicates the picture because of the variable relationship between economic provision, decision making and structures of power and authority. Women, for example, are rarely classified as heads even when they are the major economic providers if there is a male over 15 years in the household, while men are frequently designated as the head even when they are not the major provider.
One significant cause of the rise in female-headed households in developing countries is labour migration.2 Out-migration is on the increase as disparities between rural and urban locations, and between countries, become more marked. While significant and growing numbers of women migrate, the general growth in migration figures is reflected in the number of women left to care for children and maintain household reproduction without the help of a spouse. However, it is important to distinguish between those households where male labour migration has resulted in female headship and those where women are involved in polygynous marriage or have been abandoned, divorced, separated or widowed. It is equally crucial to note that many households will pass through a phase of female headship during their developmental cycle, which may be because of migration or because of divorce and subsequent remarriage or because of a subsequent marriage by the husband. The analysis of female headship thus needs to be closely tied to an examination of life cycles, marital strategies and labour deployment.
A recent study of family welfare in Ghana disaggregated the available data and looked at different types of female-headed household. Households headed by married women were found to be best off and those headed by widows worst, with the households of divorced women in an intermediate position. However, it is the case that larger households containing adults of both sexes have improved access to cash income as well as lower dependency ratios.3 It is also evident that women’s overall access to income and labour is improved through co-residence with men, particularly spouses, because women suffer discrimination with regard to their access to land, capital, education and credit.
On average, the study found that female-headed households in Ghana are no worse off than male-headed households and are slightly less likely to be found in the lowest quartile of the income distribution. Thus, an increase in the proportion of female-headed households does not necessarily indicate a growing concentration of poverty among women, but it does suggest their increasing primary economic responsibility and their growing vulnerability. Some women may have no choice except to become a household head— particularly those who are widowed, divorced or become lone mothers when very young—but it should not be assumed that membership of a female-headed household is always a disadvantage for women or necessarily deleterious for child welfare (Lloyd and Gage-Brandon 1993:118).
Data from the Ghana Living Standards Survey document the critical role played by women’s work in all households with resident children. Male household heads do almost no domestic work, even though such work has important productive value. Women work more hours in total than men, but fewer market hours. Women’s access to the cash economy makes an important contribution to the economic standing of households with children. The adjusted consumption levels of household members are highest in households where women have a primary work role either as co-head with their husband, or as primary head of their own household.
The straightforward assumption that poverty is always associated with female-headed households is dangerous, both because it leaves the causes and nature of poverty unexamined and because it rests on a prior implication that children will be consistently worse-off in such households because they represent incomplete families. There are a number of points to be made here. First, the overall findings from Ghana are supported by data from elsewhere that suggest that resources under the control of women are more likely to be devoted to children than are resources in the hands of men (Dwyer and Bruce 1988; Haaga and Mason 1987; Kennedy 1992; Buvinic et al. 1992). Thomas (1990) found that income in the hands of Brazilian women increased the health and survival chances of their children, and that it had an effect on child health almost twenty times greater than income controlled by the father. Nutrition data from the Northern Province of Zambia show that children under 5 years old in femaleheaded households are less likely to be malnourished than children in slightly better-off households where both parents are resident. The reasons for this have to do with women’s improved access to child care and to networks of sharing within female-headed households (Moore and Vaughan 1994). The available data suggest that the income that poorer women earn can lead to higher health and social benefits than the income men earn (World Bank 1993).
To argue that the position of female-headed households is complex and internally differentiated is not to deny the validity of data from around the world that show them to be disadvantaged with respect to property, capital, income and credit. Many such households exist in the context of nation states that are rolling back their boundaries and pushing more ‘social care’ into the arena of the family. This phenomenon provides a particularly graphic demonstration of the way in which women are expected to carry a disproportionate share of the costs of child care and social reproduction, and to do so often from a diminished resource base. Furthermore, the distribution of the costs of social reproduction—caring for children, the elderly and the sick—are inequitable within family/household units. As Nancy Folbre argues, the distribution of income and labour time within families is an important determinant of economic growth and welfare, and yet it has remained largely unmeasured and unexamined because of the persistent tendency to analyse families as undifferentiated and altruistic units (Folbre 1983, 1986, 1991). In all societies, the family contributes a very large share of the time and money devoted to social reproduction, that is the production and maintenance of ‘human capital’ (Folbre 1991:3-4). The unequal distribution of income and labour within the family means that women carry a disproportionate burden of the costs of the reproduction of that capital.
Women’s particular responsibility for the reproduction of human capital is often reflected in the way they are held to be primarily responsible for child welfare and for any inter-generational transfer of disadvantage. Much of the recent research on single mothers points out that these women’s own lack of resources, including poor education, contributes to increased levels of child mortality and delinquency, and decreasing levels of educational attainment and life opportunity for their children. The easy elision between female heads of households, teenage pregnancies and dysfunctional families works to make these linkages seem obvious and pre-given. Aggregate figures and generalized categories, such as ‘teenage mother’ and ‘lone parent’, exacerbate this tendency. Premature parenthood, for example, does appear to be increasing in many developing countries (Population Reference Bureau 1992), and when it is associated with low educational attainment, low rates of marriage, low wages and low levels of property inheritance and transfer, it will also be associated with poverty (Yeboah 1993). But to speak of women in such circumstances as being responsible for the inter-generational transfer of poverty and/or disadvantage to their children is more than disingenuous. For one thing, it implies that the individual is to be held accountable and that she is somehow at fault for not bringing her children up in a ‘proper’ family. This places the responsibility firmly on the individual for her failure to achieve economic and social security, and effectively prevents a thorough analysis of the causes and consequences of poverty. Teenage pregnancy does not cause poverty, however strongly it may be correlated with it under certain circumstances.
Focusing on women in the case of premature mothers and lone parents reveals the extent to which women are held to be responsible for child welfare in a way in which men are not. In fact, the reported rise in teenage pregnancies in some contexts may indicate that it is the young men who are refusing to marry. The reasons for this are diverse, but perhaps the most significant factor is that under conditions of economic decline, and where family labour cannot contribute directly to production, the cost of children has become too great. This is particularly the case where male employment opportunities and wage levels are also in decline. Increasing numbers of men are finding that they cannot support families and that marriage acts as a net drain on their own meagre resources.4 Even among middle-class families, it has become evident that some men are using their greater bargaining power within the household to renegotiate the distribution of responsibilities so that women shoulder a greater proportion of the costs of child-rearing and welfare. This means that, where women are able to earn an income, they may find that they are forced to take on the cost of the running of the household, while the husband retains his own income for other purposes (Dwyer and Bruce 1988).