Fertility and labor supply in Mexico

Purpose The aim of the article is to estimate and compare wage elasticities of labor supply of Mexican married women with different numbers of children. In addition, we analyze the treatment effect of fertility on labor supply of these women. Design/methodology/approach To obtain the estimates, we use Integrated Public Use Microdata Series data on Mexico for the year 2020. First, we apply a least squares regression. Second, we use an instrumental variables estimator to address the endogeneity of wages. Third, we extend this estimator to ad-dress the issue of self-selection into employment by considering a Heckman model with instrumental variables. Finally, we generalize the latter model to account for endogenous switching based on the number of children. Findings We found a notable heterogeneity in the wage elasticities of labor supply among married women with different numbers of children. Specifically, after addressing the endogeneity of wages, fertility and employment, our analysis reveals that married women with children exhibit positive labor supply elasticities of significantly greater magnitude than those with no children. Furthermore, the elasticities show a tendency to increase with the number of children. Practical implications The fertility-induced heterogeneity in the wage elasticity of labor supply may have significant policy implications, particularly for optimizing wage tax rates to account for variations in women’s childbearing choices. Originality/value Our article contributes to the literature on the estimation of wage elasticities by considering their heterogeneity based on the number of children. To study this heterogeneity, we propose a novel three-step estimator, which allows us to address the endogeneity of wages, self-selection into employment and endogenous switching based on the number of children. In addition, for this model we derive the expressions for the average treatment effects and average treatment effect on the treated.