Raymo, James, Andrew Halpern-Manners, and John Warren
Working paper no. 2013-05
Abstract
Our understanding of career influences on mortality is limited by reliance on relatively old data, use of surveys that contain only partial employment histories, and lack of consensus about how to best summarize detailed life history data. In this article, we address these limitations using data from the Wisconsin Longitudinal Study (WLS), a unique data source that contains mortality information through age 72 for one of the first cohorts exposed to growth in unstable employment and “bad jobs” for much of their adult lives. Results of parametric survival models show that less favorable employment histories are associated with a higher risk of death, but the nature of this relationship differs for men and women. The relationship is indirect for men and thus provides no support for the life course hypothesis that trajectories of employment experience should matter above and beyond employment characteristics and other correlates of mortality observed in late mid-life. For women, however, we find that inconsistent labor force participation across mid-life and loss of access to employer-provided health insurance in midcareer remain associated with a relatively high risk of mortality, net of temporally proximate correlates of death. These gender differences are unexpected in light of the theorized centrality of employment in men’s lives and provide an empirical basis for further efforts to advance our understanding of the pathways through which employment histories shape later-life well-being.