2004 Poverty Research Small Grants
Funded research
The Impact of the Earned Income Tax Credit on the Labor Supply of Married Couples: Structural Estimation and Business Cycle Interactions
Bradley T. Heim, Assistant Professor of Economics, Duke University
Description
Although a large body of evidence has accumulated on the effect of the Earned Income Tax Credit on the labor supply of single women; there exist relatively few studies documenting effects of the EITC on the labor supply behavior of married couples. This is true despite the fact that, in 2000 for example, a substantial 25.4% of EITC recipients were married couples with children who received checks averaging $1766. In addition, the studies that do exist only estimate the effects of the EITC on the participation margin, assume the traditional (husband chooses first) behavioral model, and often do not identify structural parameters that can be used to examine the effects of further changes to the EITC.
In this project, a structural model of married couples' labor supply will be estimated using a method developed in Heim (2003). This method estimates the unitary model of family labor supply, while accounting for heterogeneity in tastes for work and measurement error in hours of work, and still allowing for a continuous hours choice of both the husband and wife. The estimation method will be extended to allow for the incorporation of the nonconvex budget constraints generated by the EITC. In addition, repeated cross section data from 1985-2003 waves of the March Current Population Survey will be used, so that both cross sectional and policy change induced variation can be exploited to identify preference parameters.
First, using all years from the sample, the structural model will be estimated, so as to examine the effects of the EITC on the extensive and intensive labor supply margins, and to compare the effects under the unitary behavioral model estimated here to those from other papers to those used a different behavioral model and different sources of variation. Next, the model will be estimated using subsets of years that are at similar stages of the business cycle. These estimates will provide evidence on the interactions between the business cycle and the effectiveness of the EITC. Finally, policy simulations will be performed to evaluate the effects of further expansions in the Earned Income Tax Credit, and to examine the extent to which these effects depend on the state of the macroeconomy.
