This paper investigates the relationship between agricultural subsidies and the outflow of labor from agriculture. We use new and more representative subsidy indicators than have been used before and panel data from 215 EU regions over the period 2004-2014. The data allow to correct better for sample selection bias than previous empirical studies. We find that, on average, CAP subsidies reduce the outflow of labor from agriculture, but the effect is entirely due to decoupled payments and rural development payments. Coupled payments have no impact on reducing labor outflow from agriculture, i.e. on preserving farm employment. Acknowledgement :