Women's Work, Cohort Change, and Health

Rapid changes in women's labor force participation, access to good jobs, and changing work-family pressures have altered the landscape of work and family life. We use logit negative binomial hurdle models to examine whether these countervailing trends have affected the physical health of women across four birth cohorts. Longitudinal data are used to compare successive cohorts of U.S. women when they are between the ages 44 and 50. While the health of women overall did not change across cohorts, we find an increase in health problems among employed women, explained by increases in the ability of women with physical limitations to become and remain employed. Health problems among housewives decline across cohorts, resulting in better health among housewives than among employed women in the most recent cohort. These findings provide further evidence of the importance of selection processes in understanding health effects of roles, and they highlight the need for greater attention to the health effects of unpaid work.