Fourth, education at motherhood may differ from completed education if women return to school after childbirth. This is important because a temporary reduction in schooling will not necessarily affect permanent household income, whereas a permanent reduction in schooling would be expected to, because of the labor market return to schooling. Schooling reductions are likely to be permanent if women find it difficult (even if desirable) to return to or complete school after having children. In the National Longitudinal Survey of Youth, school enrollment rates subsequent to birth are strongly inversely related to age at first birth. Even for those who do return to school, completion of schooling is less likely than for those not having children.
These patterns are consistent with the notion that hours spent caring for children crowd out hours required to attend or study for school and suggest that for the women in our sample, schooling differences we observe at motherhood may well become permanent schooling differences. Indeed, some of our results indicate income differences between women born before and after the cutoff date, suggesting that schooling reductions may be permanent. This observation suggests that there may heterogeneity by age in the effects of education on infant health, an issue we address below, but find little evidence to support. While temporary and permanent reductions in schooling may lead to different effects on income, both temporary and permanent reductions in schooling may affect learning and the ability to process information, the causal pathway emphasized by Glewwe.
The remainder of the paper is organized as follows. In Section II we describe the mechanisms by which education could affect fertility and infant health and briefly summarize the existing literature on the topic. In Section III, we discuss our identification strategy, as well as our approach to nonparametric estimation, model selection, and inference. After describing the data we use in Section IV, we present the results of our estimation in Section V. Section VI presents evidence on heterogeneous effects and discusses a variety of important interpretation issues. Section VII concludes. canadian health and care mall
Why Should Education Matter?
In broad terms, education may affect a woman’s fertility and child-investment choices through either income or learning (Michael 1973).5 Education increases a woman’s income stream through both the labor market and the mating market, the latter through assorta-tive mating. In addition to the income channel, education may improve a woman’s stock of knowledge regarding contraceptive technologies or healthy pregnancy behaviors, either because it augments her knowledge directly (i.e., educational curricula are important), or because it improves her ability to absorb and process information generally. We next describe each of these mechanisms in turn.