Li Xia *
Abstract：In this study I use data from the National Survey of Family Growth to estimate the impact of children on mothers' labor supply and earnings. Because it is highly likely that some factors that are both associated with a woman's labor market outcome and her fertility outcome (such as if she has children or not) are not included in the data, in a regression analysis of the data, a fertility outcome variable will be endogenous. To address this issue, I use a variable indicating if a woman's first pregnancy ended in a miscarriage or not as the instrumental variable and use the instrumental variable estimation technique to estimate the coefficient of the fertility outcome variable in the labor market outcome regression. I argue that the variable I use as the instrument is a valid one because evidence shows that the occurrence of a miscarriage is closely related with a woman's fertility outcome, and miscarriage largely happens randomly according to relevant medical studies. The estimation result shows that there is a negative and modest impact of children on mothers' labor supply and earnings. The magnitude of my estimates is comparable with that in the studies that estimate the impact of children with instrumental variables based on other natural experiments during a woman's reproductive history.
Keywords: Female Labor Supply, Fertility, Instrumental Variable, Natural Experiment.
* Correspondence address: School of Economics, Shanghai University of Finance and Economics, 111 Wuchuan Road, Shanghai 200433. Email address: firstname.lastname@example.org. This paper is a chapter of my Ph.D. dissertation at Yale University. I am grateful to Professors Joseph Altonji, Paul Schultz, Hanming Fang, and Ebonya Washington for their help and encouragement. I would like to thank participants of labor workshops at Yale and 2008 International Symposium on Labor Economics at Xiamen University, particularly Professor Steven Stern, for their helpful comments and discussion. All remaining errors are mine.