This study examines the relationship between economic conditions and health, using fixed-effect models estimated with state-level data from 1972 to 1991. Health is measured by total and age-specific mortality rates, as well as specific causes of death. The findings show that unemployment rates are negatively and significantly correlated with total mortality and nine out of ten specific causes of death, with suicides being the exception. The procyclical variation in mortality is more pronounced for 20-44 year olds than for older individuals. Personal incomes are weakly and negatively associated with some types of mortality but are often statistically insignificant. The study suggests that health may improve during economic downturns due to cyclical variations in the time costs of medical care or healthy lifestyles, and that job-holding may have adverse effects on health.This study examines the relationship between economic conditions and health, using fixed-effect models estimated with state-level data from 1972 to 1991. Health is measured by total and age-specific mortality rates, as well as specific causes of death. The findings show that unemployment rates are negatively and significantly correlated with total mortality and nine out of ten specific causes of death, with suicides being the exception. The procyclical variation in mortality is more pronounced for 20-44 year olds than for older individuals. Personal incomes are weakly and negatively associated with some types of mortality but are often statistically insignificant. The study suggests that health may improve during economic downturns due to cyclical variations in the time costs of medical care or healthy lifestyles, and that job-holding may have adverse effects on health.