University of Wisconsin - Madison Institute for Research on Poverty Discussion

The economic model of fertility emphasizes two important "roles of increased human capital investments in women on fertility: (1) an income effect which tends to increase the demand for children and (2) a price effect which tends to decrease the demand for children because of the higher opportunity cost of childcare in terms of foregone labor force opportunities. Previous empirical estimates of these effects have focused on human "capital investment in the form of formal schooling, and have found that the price effect dominates. However, in developing countries a broader definition of human capital may be appropriate because of widespread nutrition and health deficiencies. We estimate the Easterlin, Pollak and Wachter (EPW) extended model of fertility, which includes endogenous tastes and biological factors, with a broader representation of women's human capital that encompasses health, nutrition and work experience in addition to schooling. We also explore the implication of data incompleteness in two senses: (1) selectivity in providing the neces~ary data and (2) incomplete fertility in" that many women in our sample may have further children. Our estimates suggest that selectivity, which is ignored in previous studies, is a significant factor. They also suggest that the form of control for incomplete fertility is important, but that a multiplicative control for duration of exposure has the same implications as the more sophisticated Boulier and Rosensweig procedure. In more substantive respects, we find that income has a positive effect and that increased pre?icted earnings for the women has a negative effect with our broader definition of human capital. We also find that human capital investments in nutrition and health have additional effects, beyond those incorporated in the opportunity cQsts of labor market earnings. For nutrition the addit ional impact is to reinforce the opportunity cost dimension of more human capital by lowering fertility. But improved health increases family size, apparently because of iilcreased fecundity, lengthened productive spans and increased energy, whi~h provides support for the EPW enphasis in biological factors. MOreover .our estimate for the health and nutrition effects of fertility are more robust than are those for the traditional schooling variable. Therefore we conclude that fertility analysts and family planning, health and nutrition policy analysis in similar developing countries should be expanded to incorporate a broader definition of human capital toinclude health and nutrition status.