This paper tracks factors contributing to the ups and downs in women's employment from 1970 to 2010 using regression decompositions focusing on whether changes are due to shifts in the means (composition of women) or due to shifts in coefficients (inclinations of women to work for pay). Compositional shifts in education exerted a positive effect on women's employment across all decades, while shifts in the composition of other family income, particularly at the highest deciles, depressed married women's employment over the 1990s contributing to the slowdown in this decade. A positive coefficient effect of education was found in all decades, except the 1990s, when the effect was negative, depressing women's employment. Further, positive coefficient results for other family income at the highest deciles bolstered married women's employment over the 1990s. Models are run separately for married and single women demonstrating the varying results of other family income by marital status. This research was supported in part by an Upjohn Institute Early Career Research Award.
women's employment other family income stalled gender revolution regression decomposition