We extend the literature applying treatment effects methods to the evaluation of agro-environmental schemes (AES) in three ways. First, we derive the statistical assumptions underlying Difference-in-Difference matching as restrictions on an economic model. These restrictions are consistent with what we know about the practical implementation of the program. The model also provides guidance for selecting control variables. Second, we present the first disaggregated estimation of the effects of a nationwide AES program on environmentally-relevant practices for a nationally representative sample of farmers. Third, we test the robustness of our findings confronting them with a credible alternative identification strategy. Our results suggest that voluntary AES programs that seek to reduce nitrogen use and encourage crop diversification may have large windfall effects. In contrast, more ambitious AESs such as conversion to organic farming, which combine strong requirements with large payments, seem to have achieved their goals.