Social Roles, Human Capital, and the Intrahousehold Division of Labor:
Evidence From Pakistan

In 1965 Becker became the first economist to formally articulate the roles that comparative advantage and learning by doing (that is, gaining experience in a specific task) may play in the intrahousehold division of labor. In particular he argued that if one member of the household must stay at home to take care of domestic chores, economic efficiency dictated that it should be the one with the lowest expected wage and the highest productivity in domestic chores. This simple but powerful observation has sparked a voluminous empirical literature, the general conclusion of which is that job market participation indeed responds to the human capital characteristics of individual household members. Although most of this literature relates to developed economies, applications to developing economies have yielded similar conclusions.

Noneconomists have also proposed explanations about the intrahousehold division of labor. These explanations emphasize the role of customs and social norms and argue that the sex and status of individuals determine the tasks that society assigns to them. This paper tests whether social norms, human capital, and learning by doing can account for the observed division of labor within rural Pakistani households.

Earlier Studies of Labor Allocation

In much empirical work, testing the role of social norms has been hindered by the fact that data from developed countries are available for households almost always with, at most, one working adult male and female. Furthermore, the existence of markets for utilities, food preparation, child care, and the like drastically reduce the number of tasks undertaken by households. In such small households with few tasks to perform, the prospect for intrahousehold division of labor is limited. The situation is quite different in developing countries where households are large, children actively participate in productive activities, and households provide much of their own food, fuel, water, and child care, in addition to pursuing a multitude of income-generating activities. In such households with multiple tasks and participants, sufficient room exists for specialization and scope abounds for social norms to fashion what individuals do. These households are also likely to be more organizationally complex. In contrast to small nuclear households, large households offer more room for delegation of responsibilities, thereby creating incentive and information problems similar to those encountered in firms.

". . .the allocation of tasks within households is not solely driven by comparative advantage considerations. Rather, households seem to operate as hierarchies with socially specified roles partially determined by gender and family status."

One strand of work on time allocation has examined the choice between household and market-oriented activities. These choices have been shown to depend on the age of the female adult, her education, household demographic composition, family wealth, and distance to schools, town, or market centers. Another issue that has dominated the literature on time allocation in South Asia is the extent to which social norms, particularly patriarchy and the norm of female seclusion or purdah, dominate economic factors that affect division of labor. This literature, however, is not fully conclusive because sufficiently detailed data and convincing instruments for social roles are lacking. This paper tests an alternative methodology to investigate whether human capital, learning by doing, and socially defined roles affect the division of labor inside the household. Results largely confirm previous work but considerably refine current understanding of the factors influencing intrahousehold task allocation.

What Influences the Division of Household Labor?

Results show that tasks get allocated within a household only partly as a result of comparative advantage considerations reflected in the differences in human capital among household members. Better educated individuals and households with better educated members participate more actively in non- farm work, a finding in line with earlier research conducted by the authors of this study. Better educated household members also do fewer household chores and enjoy more leisure. This is true for both males and females, suggesting that schooling raises the intrahousehold bargaining power and implicit welfare weight of women.

Other dimensions of human capital such as age, a proxy for experience, and height, a proxy for past nutritional status, have the expected effect on intrahousehold allocation, with taller and older members taking on chores that are more physically demanding and require traveling away from the household.

After controlling for individual human capital and task-specific experience, the study finds overwhelming evidence of division of labor by gender and family status. Activities are organized into gender-specific spheres of influence. Males are responsible for "market" work (farming, herding, and other income-generating activities); females, for home production activities. This pattern is not peculiar to rural Pakistan and has been observed in many other societies as well. The allocation of activities corresponds closely to the dichotomy between the "productive" and "reproductive" roles often assigned to men and women in traditional societies.

Task allocation within each gender group varies systematically with family status. The head of household and his or her spouse provide most of the labor for most activities; other adult members of the household work less. In agreement with popular perceptions in Pakistan, results also show that daughters-in-law work systematically harder than daughters of comparable age, build, and education. These findings reflect the fact that daughters-in-law either have less bargaining power or take a long-term interest in the well-being of the household where they will spend the rest of their life.

Taken together, these results indicate that the allocation of tasks within households is not solely driven by comparative advantage considerations. Rather, households seem to operate as hierarchies with socially specified roles partially determined by gender and family status. The correlation found between financial control and labor effort is consistent with moral hazard considerations, long-term commitment to the household, and internal bargaining power. Individuals with more control over household finances and with a longer-term stake in the household work harder.

Finally, the study finds some evidence of long-lasting returns to learning by doing in nonfarm activities and farm management, but not in household chores. The constant reallocation of household chores among women suggests that they are easy to learn. The idea that women get locked into these tasks because they learned them as little girls can, therefore, be ruled out. If lock-in is present, it is in off-farm activities where males dominate and returns to schooling are high.

The analysis throughout has regarded household composition as a given, but the evidence collected suggests that household composition affects what individuals do and how hard they work. To the extent that households form to maximize the gains from being together, the findings indicate that two fundamental forces shape household formation: gains from specialization, which favor larger households, and incentive issues, which penalize them. If gains from specialization are large, households should, on average, be larger. This seems to be the case in the area studied: the wide variety of tasks that large rural Pakistani households undertake leaves much room for a precise division of labor, which also helps mitigating monitoring and shirking problems. In addition, large households benefit from returns to scale in household chores and can more easily let some members fully specialize in nonfarm work. What appears to keep household size in check are problems of shirking and monitoring, with the household head and his or her spouse working harder than other members except daughters-in-law. In this respect, one cannot but notice the formal similarity between rural Pakistani households and firms: both solve internal organization problems via a complex hierarchical structure.

What remains unclear from this work is how households are formed over time. For instance, do individuals with a nonfarm occupation join already existing households? Or do larger households let some of their members specialize in less subsistence-oriented activities such as nonfarm work? These issues deserve more research.

Recent FCND Discussion Papers

To order a discussion paper, please fill out an online order form, email IFPRI-FCN@cgiar.org or send requests to Food Consumption and Nutrition Division, IFPRI, 2033 K Street, N.W., Washington, D.C. 20006, U.S.A.