Selection, investment, and women's relative wages over time

In theory, growing wage inequality within gender should cause women to invest more in their market productivity and should differentially pull able women into the workforce. Our paper uses Heckman's two-step estimator and identification at infinity on repeated Current Population Survey cross sections to calculate relative wage series for women since 1970 that hold constant the composition of skills. We find that selection into the female full-time full-year workforce shifted from negative in the 1970s to positive in the 1990s, and that the majority of the apparent narrowing of the gender wage gap reflects changes in female workforce composition. We find the same types of composition changes by measuring husbands' wages and National Longitudinal Survey IQ data as proxies for unobserved skills. Our findings help to explain why growing wage equality between genders coincided with growing inequality within gender