In the last three decades several countries around the world have transferred authority from their national to their regional governments. However, not all their regions have been empowered to the same degree and important differences can be observed between and within countries. Why do some regions obtain more power than others? Current literature argues that variation in the
redistribution of power and resources between regions is introduced by demand. Yet these explanations are conditional on the presence of strong regionalist parties or territorial cleavages. This article proposes instead a theory that links the government’s risk of future electoral defeat with heterogeneous decentralisation, and tests its effects using data from 15 European countries and 141 regions. The results provide evidence that parties in government protect
themselves against the risk of electoral defeat by selectively targeting decentralisation towards regions in which they are politically strong. The findings challenge previous research that overestimates the importance of regionalist parties while overlooking differences between regions.