Recent Stagnation of Married Women’s Labor Supply: A Life-Cycle Structural Model

This study investigates the effects of changes in various determinants of labor supply on the dramatic changes from the older (the 1950s and earlier) cohorts to the younger (the 1960s and later) cohorts in life-cycle labor supply behavior, and ultimately provides a model-based quantitative explanation of the recent decline in the aggregate labor supply of married women. On the basis of the Current Population Survey data, it first documents that, while lifecycle labor supply profiles are non-overlapping and bell-shaped for the older cohorts, they are roughly flat for the younger cohorts, and from the mid-thirties of the life-cycle, the younger cohorts continue to supply less labor than the 1950s cohort does. Then in a life-cycle model of women’s labor supply, the behavioral changes are explained by a combination of changes in various labor supply determinants, with the opportunity cost of childbearing (as represented by returns to work experience and the rate of human capital depreciation during a nonworking period) being the dominant contributor. In particular, relative to the older cohorts, the higher opportunity cost for the younger cohorts makes them supply more labor at the early stage of the life-cycle, delay childbearing to a later stage, and upon childbearing, stay out of the labor force, other things being constant. A calibration of the model demonstrates that the aggregate labor supply of married women would increase by 1.96 percentage points from 2000 to 2010 if there were no changes between the older and the younger cohorts in the labor supply determinants; however changing the determinants for the same period actually results in a reduction of aggregate labor supply by 1.36 percentage points. Of the 3.32 percentage points of the pseudo-reduction of married women’s aggregate labor supply (difference between the hypothetical 1.96 percentage point increase and the actual 1.36 percentage point reduction), 67 percent is explained by the increased opportunity cost for the younger cohorts, and the rest is accounted for by a combination of changes in the tax code, business cycle conditions, and preferences, among others. JEL Classifications: D1, J13, J22, J24