Raymo, James, and Miho Iwasawa
Working paper no. 2004-17
Abstract
Results from cross-national research suggest that increasing economic independence for women contributes to lower rates of marriage only in societies characterized by relatively asymmetric gender division of domestic and market labor. We develop and evaluate an alternative theoretical scenario in which the same results could be explained by changes in marriage market composition and women’s continued dependence upon men’s economic resources. Using data from four recent surveys, we examine how changes in Japanese marriage rates reflect changes in the general propensity to marry, the likelihood of particular pairings, and composition of the marriage market with respect to age and educational attainment. Construction of counterfactual marriage rates and synthetic cohort marriage trajectories indicates that changes in marriage market composition due to relative improvements in women’s educational attainment have contributed to lower rates of marriage among highly educated women and higher rates of marriage among less educated women. Net of this compositional effect, decline in the general propensity to marry among highly educated women has actually been smaller than among high school graduates. Explicit consideration of change in the feasibility of marriage thus results in a fundamentally different theoretical interpretation of decline in the relative risk of marriage for highly educated women.