چرا زنان در آموزش و پرورش پیشرو هستند؟ نابرابری جنسیتی در سرمایه انسانی، بازار کار و تنظیم خانواده در فیلیپین

This paper shows mutually consistent evidence to support female advantage in education and disadvantage in labor markets observed in the Philippines. We set up a model that shows multiple Nash equilibria to explain schooling and labor market behaviors for females and males. Our evidence from unique sibling data of schooling and work history and from the Philippine Labor Force Survey support that family arrangement to tighten commitment between daughters and parents keeps a high level of schooling investments in daughters. Because wage penalty to females in labor markets means that education is relatively important as a determinant of their earnings, parental investments in their daughters’ education has larger impacts on the income of their daughters than on their sons. Parents expect larger income shared from better-educated adult daughters. In contrast, males stay in an equilibrium, with low levels of schooling investment and income sharing. Our results also imply that the above institutional arrangement is stronger among poor families.

Gender plays important roles in decisionmaking and resource allocations critical to economic development. Mothers’ human capital improves child health and education, which determines the well-being of the next generations (for example, Behrman and Rosenzweig, 2002 and Strauss and Thomas, 1995). Empowering women, who tend to allocate more resources into productive means is an important channel to improve economic performance in the long run (for example, Chattopadhyay and Duflo, 2004 and Pitt and Khandker, 1998). Yet gender inequality remains an urgent policy issue. In this paper we take up the Philippine case that women are more progressive in education than men but still suffer from lower wages in the labor market (after controlling other factors). Though we observe many countries where both education and labor markets favor males (Blau and Kahn, 1992 and King and Hill, 1998), we show that the Philippines represents a unique case in which women receive more schooling than men but penalty (discrimination) against women still exists in the labor market.1 Advantage in schooling and disadvantage in earnings coexist for women in the Philippines.2 In this paper, we argue that the above seemingly contradictory situation can be explained in a consistent way. That is, schooling investment in females has been an optimal response to the labor market discrimination. Table 1.1 shows the percentages of males and females who completed college education in different cohorts. The data come from the October 2009 round of the Philippine Labor Force Survey.3 It is shown that a larger fraction of females graduated from college than males did in all cohorts, and the gap widened recently among younger cohorts. This means that women attain more education in the country; that is, more general human capital is embodied in women than in men.Next we show female disadvantage in labor market earnings by estimating the standard wage equations (Table 1.2). In all cohorts, the female effect is significantly negative, which means that women experience penalty to their wages in the labor market, despite rather offsetting positive effects of more schooling attained among women. However, the female effect is becoming smaller in younger cohorts. Note that if more able women are likely to find jobs (therefore, their wages are observed), we expect upward bias in the above estimates, which implies that the true wage penalty (negative effect) is even larger. In Table 1.2, we did not control occupation and industry, so the female effect contains the effect of females’ endogenous choices of occupations and industries within a region. On average, the female effect is about 40 percent of the college premium (that is, income gain from completing college relative to high school) Discrimination against women implies that education plays a more important role in determining wages among women than among men. An additional year of schooling is more important among women because they experience unconditional penalty, unrelated to their human capital, in the labor market.5 Therefore, the labor market discrimination can motivate women to attain higher levels of education to have comparable earnings with men. Our intuition relies on the above observation. Moreover, it is possible that returns to children's schooling for parents can differ from those for children if children share their earnings with their parents. For example, parents and children co-reside and share resources, and parents often receive some remittances from children who live away. Through these family arrangements, a portion of the returns to schooling investments is accrued by parents. These arrangements can take the form of income sharing, insurance, support, and inheritance. In exchange for support from parents who invest in child schooling, children have an incentive to share schooling returns with their parents. Alternately, parents have an incentive to invest in their children if the children share returns with them. Empirical results of this paper support the above hypothesis for women. The institution of family is critical to determining child schooling. Filipino kinship patterns are generally categorized as belonging to a bilateral family system in which sons and daughters play equal roles in supporting their parents (Agree et al., 2002, Domingo and Asis, 1995, Natividad and Cruz, 1997 and Ofstedal et al., 1999). In some cases, daughters play more important roles than sons. This practice is quite common in Southeast Asia. Such observations of the intergenerational support system are consistent with tighter commitments between parents and daughters, which enhances parental investments in daughters, who are more likely to help parents. Other compelling hypotheses were proposed in the studies that explain female advantage in education in the context of the Philippines. Estudillo et al., 2001a and Estudillo et al., 2001b posit a hypothesis that parents equally care about the expected lifetime incomes of their sons and daughters.6 In the agrarian setting, sons are more likely to inherit farmland, but daughters do not. To compensate for this gap in inheritance, parents rather invest in daughters’ schooling.7 In the bilateral family system, we can also explain the above egalitarian behavior as the parents’ optimal action because diversification of investments in children is the best strategy, given that they do not know who principally will take care of them (that is, sons and daughters are equally likely or daughters are more likely to provide intergenerational support to them). In their hypothesis, the intergenerational support system and labor market issues are not incorporated. In our hypothesis, however, parents invest in children to maximize their own utilities in a game-theoretic setting. Parents can receive some income from their children, which constitutes returns to schooling for parents. Interestingly, Nash equilibria in this setting support two different situations: (1) high levels of investment and income sharing and (2) low levels of investment and income sharing. In the Philippines, the former case characterizes the situation for females, while the latter does for males. Our empirical analysis uses unique sibling data collected in a recent survey. In 2010, the survey was conducted in eight provinces covering 3481 households. For all siblings of our sample individuals, we gathered information on schooling, work histories, and family arrangement including income sharing and remittances. This is an ideal dataset for testing our hypothesis because we can wipe out unobserved household-specific components by using within-sibling variations. In addition, because birth control is not usually practiced in a poor rural environment, the gender composition of children is regarded as random. We estimate and compare female effects in schooling, earnings, income-sharing and co-residence equations. The paper is organized as follows: the next section lays out a simple model in which the multiplicity of Nash equilibria is generated to explain the stylized facts. Given the labor market imbalance, females tend to be in the equilibrium with higher levels of schooling investment and income sharing. Sections 3 and 4 describe our empirical framework and data, respectively. Section 5 discusses empirical findings in detail. Our evidence shows that, as Table 1.1 and Table 1.2 confirmed, females are likely to attain more education but experience penalty in labor market earnings. By receiving more education, females can have earnings that are more comparable to those of males. Second, females are likely to share more of their income with their parents. In this sense, for parents, more returns are accrued from investing in female schooling. Third, our results also imply that such a gendered strategy, with a greater role of females than males, is stronger among poorer families and can be weakened if income level rises. Section 6 concludes.

This paper shows mutually consistent evidence to support female advantage in education and disadvantage in labor markets. Family arrangement to tighten commitment between daughters and parents is an institution that keeps high levels of schooling investments in daughters. This is rational to parents because they can expect larger income shared from daughters who become educated in adulthood. For females relative to males, wage penalty in labor markets means that education is of relatively large importance as a determinant of their earnings. Thus, parental investment in daughters’ education has larger impacts on the income of daughters than sons. Our evidence from unique sibling data of schooling and work history, as well as the Philippine Labor Force Survey, supports the above hypothesis. Our results also suggest that the above institutional arrangement is stronger among poorer families, and can be weakened if income level rises. Whether the gendered institutional arrangement becomes strengthened or weakened along with economic development is an interesting issue, but beyond the scope of this paper.