Unemployment and mortality: evidence from the PSID

We use micro-data to investigate the relationship between unemployment and mortality in the United States using Logistic regression on a sample of over 16,000 individuals. We consider baselines from 1984 to 1993 and investigate mortality up to ten years from the baseline. We show that poor local labor market conditions are associated with higher mortality risk for working-aged men and, specifically, that a one percentage point increase in the unemployment rate increases their probability of dying within one year of baseline by 6%. There is little to no such relationship for people with weaker labor force attachments such as women or the elderly. Our results contribute to a growing body of work that suggests that poor economic conditions pose health risks and illustrate an important contrast with studies based on aggregate data.