Association for Public Policy Analysis & Management

Economic theory predicts ambiguous effects of wages on childbearing decisions. Identifying such effects empirically is complicated by the likely endogeneity of wages. First, wages may be correlated with other (unobserved) determinants of fertility. Second, women who plan to have more children may choose lower paying jobs that are easier to combine with large families. And third, having children may have a negative causal impact on women’s subsequent careers and wage paths, e.g. if employers perceive mothers to be less competent and/or invested in work. All these three mechanisms would lead to a negative relationship between wages and fertility, even in the absence of any causal relationship.

This paper focuses on effects of minimum wage policies rather than individual observed wages. To fix ideas, I first present a simple economic model of fertility and employment. The model captures the impact of childcare costs by making the money cost of children dependent on labor force participation. The intuition of the model is simple: higher childcare costs raise the reservation wage of mothers, leading some mothers who face low market wages to exit the labor force altogether. Women in this situation may be better off not working, instead staying home to raise large families. In this model, a wage increase predicts both higher labor force participation rate and a reduction in family size.

The empirical analysis identifies effects on fertility by leveraging plausibly exogenous variation in state level minimum wage policies. This variation will allow us to estimate panel models of fertility using detailed information from the Vital Statistics on all births occurring between 1983 and 2015. A fundamental identifying assumption in these models is that the timing of policies is uncorrelated with other unobserved drivers of fertility. To test this empirically, I implement an event study framework, analyzing how fertility patterns change around the time of minimum wage increases. These models indicate parallel pre-trends, which lend support to the parallel trends assumption.

The estimated panel models indicate that higher minimum wages significantly affect fertility. The preferred specification finds that a ten percent increase in the minimum wage reduced second order or higher birth rates by 1.2%. Reassuringly, the models find no effect on first or higher order births to women with at least a bachelor's degree, a placebo sample who is unlikely to be exposed to the minimum wage. More generally, a subsample analysis finds that the estimated fertility effects line up well with estimated effects on labor market outcomes: the largest effects are found for groups that also have large effects on earnings and employment.