| Rashmi Barua
Boston University
Effect of kindergarten entrance age on academic performance and maternal labor supply
This study evaluates the effect of delaying kindergarten entry on maternal labor market outcomes. Employing data from the National Longitudinal Survey of Youth (NLSY), it addresses the simultaneity of women's decisions regarding labor force participation and the optimal kindergarten entrance age of their children. The identification strategy accounts for the endogeneity between parents' decision to hold back their children for an additional year and maternal labor supply. It also corrects for the selection bias due to unobserved heterogeneity across families. This study is an improvement upon existing literature that attempts to deal with the endogeneity of entrance ages. The scope of this paper is wider than previous studies because the author estimates the long term impact on labor market outcomes instead of a one year impact. From a methodological standpoint also it is an improvement because the author corrects for specification issues by instrumenting for actual delay using a binary instrument derived from variation in month of birth and state kindergarten entrance age laws. In addition, Quantile Treatment Effects (QTE) are used to estimate the effect of delaying entry to the distribution of income. If parents, who cannot afford market childcare costs, are less likely to red shirt, then the labor supply responses to a policy change that increases the minimum entrance age would affect the mothers of the most financially constrained families. The binary nature of the endogenous variable and instrumental variable will allow for estimates, using QTE methods, of the effect of delaying school entry on the entire income distribution of mothers.
In interpreting the results it should be kept in mind that when treatment effects are heterogeneous, instrumental variable estimates fail to capture the average treatment effect on the entire population. This empirical strategy will provide consistent estimates of the Local Average Treatment Effect (LATE) of the group of "compliers" in the spirit of Angrist and Imbens (1994). LATE estimators are policy relevant since they capture the effects of externally induced changes on those whose status can be manipulated by certain policy variables.
Back to Funded Dissertation Grants Page |