The productivity impacts of the CAP payments, and especially of the Rural Development measures, are still uncertain and mixed effects from a diversity of empirical approaches have been found in the literature. This paper analyses the effects of the CAP on the productivity of the farms in the EU28 by adopting an alternative approach based on a system of equations derived from a non-nested production function with constant elasticity of substitution. The econometric strategy consists of a simultaneous equation model with GMM estimator which allows dealing with potential endogeneity bias. Results show that the majority of Pillar I and II payments have a positive impact on the productivity of the farming sector, with the noticeable exceptions of agro-environmental schemes and other rural development subsidies. However, the productivity impact of CAP payments is sensitive to the composition of the sample and it is different for different groups of Member States.