We estimate the effects of employer downsizing on older workers' health outcomes using different approaches to control for endogeneity and sample selection. With the exception of the instrumental variables approach, which provides large imprecise estimates, our results suggest that employer downsizing increases the probability that older workers rate their health as fair or poor; increases the risk of showing symptoms of clinical depression; and increases the risk of being diagnosed with stroke, arthritis, and psychiatric or emotional problems. We find weaker evidence that downsizing increases the risk of showing high levels of C-reactive protein (CRP), a measure of general inflammation. We find that downsizing affects health by increasing job insecurity and stress, but that its effects remain statistically significant after controlling for these pathways, suggesting that other mechanisms such as diminished morale and general demotivation also affect worker health. Our findings suggest that employers ought to consider actions to offset the detrimental health effects of reducing personnel on their remaining (older) workers.