Declining fertility rates and increasing life expectancy necessitate a higher labor participation rate among older people in order to sustain pension systems and boost economic growth. At the same time, researchers have only recently begun to pay attention to the health effects of a longer working life, with rather mixed results thus far. Utilizing panel data from eleven European countries, and two distinct identification strategies to deal with endogeneity, we provide new evidence of the health effects of retirement.In contrast to prior research, we analyze both the impact of being retired and the effect of spending longer time in retirement. Using spouses characteristics as instruments, while taking precautions to ensure validity, we find a robust, negative impact of being retired and spending longer time in retirement on selfassessed, general, mental and physical health.In addition, we show that the impact on selfassessed health remains similar in models using instruments from previous research while also including individual- and time-fixed effects to remove time-invariant unobserved heterogeneity between individuals as well as common health shocks.Overall, the results suggest that this innovation and the fact that we take lagged effects into account explain the differences in comparison to prior multi-country research using these instruments. While the short-term health impact of retirement in Europe remains uncertain, the medium- to long-term effects appear to be negative and economically large.