سرمایه های انسانی، عشق واقعی و نقش جنسیتی : سرنوشت رابطه جنسی؟

We assume optimal specialization (in the nuclear family) involves one partner doing ‘homework’ and the other ‘wagework’. Basic human capital for one or the other is imparted to individuals through early childhood training. Mates are chosen on the basis of noneconomic criteria-‘true love’. Because love strikes randomly after basic human capital is acquired, a coordination problem arises in choosing complementary specializations. Even with no discrimination, there exist equilibria characterized by gender roles – perfect correlation between sex and family roles. There may also exist other equilibria in which the correlation between sex and family roles is imperfect and individuals are trained by aptitude. We rank the equilibria when they coexist and demonstrate that affirmative action policies can increase average welfare by eliminating some equilibria. The analysis is conducted in a framework which characterizes individuals by training, by aptitude, and by family role in a 2 × 2 × 2 model. In this framework, we formalize the equality of the sexes as well as the relative importance of nature versus nurture and various types of training.

Only recently have economic theorists seriously attempted to understand the economic causes and implications of gender roles in society1. Understanding gender roles is difficult for standard economic theory because it requires modeling the organization of the family and the interaction of the household with the marketplace. The approach taken in recent papers has been to model the family as consisting of individuals each maximizing their own utility within a thick blanket of externalities.2 In this context gender roles have been explained as emerging from market asymmetries which yield bargaining asymmetries. The battle in the economic market and its implication for the battle between the sexes has produced some insightful and plausible explanations for patterns of family and social organization. We believe that this methodological individualist approach is a fruitful one for the economic analysis of the family. However, in this paper we step back from methodological individualism and take a different approach, one in which the battles in the market and within the household are put aside. We do this to ask what we believe is a more basic question: Might gender roles emerge in a society in which markets and households treat the sexes equally? And the subsequent question: If gender roles emerge would they be efficient? To this end we build a model that concentrates on what we consider to be the standard building blocks of a modern society. Societies consist of families in which adults divide the responsibilities of earning income and raising children. The family is organized in the way to best facilitate these tasks. This depends on the aptitude and training of the adults. The formation of families is a bit of a mystery, one we model as depending on criteria that are not economic, and which gives rise to a fundamental coordination problem. In particular, our basic model has four key ingredients. First, we restrict attention to the nuclear family consisting of two adults, one male and one female, and an unspecified number of children. Second, we assume that specialization within the nuclear family is efficient. The most efficient specialization involves one adult working in the home, raising children, and the other adult working in the marketplace. Third, early training of children by parents can be done in only one of the two spheres of activity. Children can be trained in skills and attitudes best suited to either the home or the marketplace; these skill sets are mutually exclusive. Four, families are formed randomly, on the basis of noneconomic criteria which we flippantly call ‘true love’. We discuss the rationale for our specific assumptions and how relaxing them affects the results in the body of the paper. These four ingredients give rise to a coordination problem in choosing complementary specializations. Family welfare is maximized if complementary human capital investments have been made before marriage. If partners are known in advance, then it is possible to make the appropriate investments. However, individuals do not know whom they are going to marry, and choose mates on the basis of ‘true love’. Since true love strikes randomly, there is a coordination problem. When aptitudes are homogeneous, the ex ante optimal investment decision involves individuals of one sex acquiring the specialization believed to best complement the specialization chosen by the majority of the members of the opposite sex. This leads to two stable Nash equilibria with gender specialization: either all men do wagework and all women homework, or the opposite. For individuals ‘sex is destiny’ within each equilibrium, though the coexistence of the equilibria implies that sex is not destiny for society. With homogeneous aptitudes both these ‘gender role’ equilibria are efficient. However, with heterogeneous aptitudes for the occupations, gender specialization may be inefficient. For example, if a young boy demonstrates an aptitude for homework, his parents may still choose to train him for wagework when all girls are being trained for homework. Gender specialization is still an equilibrium as long as the distributions of aptitudes are not too skewed. This depends on both the importance of aptitude in the occupations and the difference in aptitudes between the sexes. In addition to gender specialization, there exists an equilibrium in which aptitude determines training. In this training-by-aptitude equilibrium, some children of each sex are trained for wagework while others are trained for homework. This yields an imperfect correlation between biology and training for economic roles. The correlation between biology and actual economic activity is also imperfect, and gender roles are entirely absent when the distribution of aptitudes is the same for both the sexes. For general distributions, the training-by-aptitude equilibrium exists over some range of parameters as long as training and aptitude reinforce each other. We show that under a large set of circumstances the training-according-to-sex equilibria and the training-by-aptitude equilibrium are the only pure strategy equilibria. We then compare the average payoffs between the two types of equilibria. Which equilibrium dominates depends on the relative importance to families of having complementary training as opposed to having members who are trained in their aptitude. We demonstrate that affirmative action policies can increase average welfare by eliminating some equilibria. We are aware of no other paper where this is so. Several recent papers examine models of gender roles. In Echevarria and Merlo (1995) matching is by ‘true love’ and individuals acquire education before marriage. Unlike our model, there is only one type of training and sex is destiny because pregnancy imposes an absolute disadvantage on women’s participation in the paid labour force. François (1998) and Danziger and Katz (1996) demonstrate that a sexual division of labour within the household can be a feature of equilibrium when agents have identical abilities. The division of labour within the household supports (and is supported by) discrimination within the paid labour force. These papers rely on either asymmetric information or some explicit sexual discrimination. We have full information and no discrimination in our model. Further, we model more than one type of aptitude and examine training by aptitude. The paper proceeds to develop the prototype model in Section 2. In Section 3 the model is extended to heterogeneous aptitudes for different types of work. Section 4 analyzes welfare, policy, and evolution. Finally, Section 5 concludes by summarizing our main results and discussing possible extensions.