May 2011

IZA DP No. 5729: A Panel Data Analysis of Racial/Ethnic Differences in Married Women's Labor Supply

We study differences in life-cycle labor supply among white, black, and Hispanic women, focusing on the interaction between race/ethnicity, education, and fertility. We use panel data that capture women's labor market and fertility histories and an econometric model that accounts for the endogeneity of labor market and fertility decisions, the heterogeneity of the effects of children and their correlation with the fertility decisions, and the correlation of sequential labor market decisions. Our results show an intricate connection between race/ethnicity, education, and fertility as determinants of women's life-cycle labor supply. For all levels of education, white women have fewer children, have the first birth later in life, and space subsequent births more closely together. The level of labor market involvement before the first birth is highest for white women and lowest for Hispanic women, but children reverse the relationship between race/ethnicity and level of labor market involvement. The negative effects of children are largest for white women and smallest for Hispanic women, and as a result, among women with two children, black and Hispanic women work more than white women. Racial/ethnic differences in fertility decisions, pre-natal labor supply, and labor supply responsiveness to children decline with the level education. Educational differences contribute to the racial/ethnic differentials in labor supply. White women have the highest levels of education and Hispanic women have the lowest levels of education. Other things equal, women with higher education have fewer children, have the first birth later in life, space subsequent births more closely together, work more before the birth of the first child, but face larger negative effects of children on their level of labor market involvement.