This study estimates the causal effect of working hours on health. We deal with the endogeneity of working hours through instrumental variables techniques. In particular, we exploit exogenous variation in working hours from statutory workweek regulations in the German public sector as an instrumental variable. Using panel data, we run two-stage least squares regressions controlling for individual-specific unobserved heterogeneity. We find adverse consequences of increasing working hours on subjective and several objective health measures. The effects are mainly driven by women and parents of minor children who generally face heavier constraints in organizing their workweek.