Appointed Versus Elected Mayors and Incentives to Pork-Barrel : Quasi-Experimental Evidence from Germany

Do incentives and policy choices of public officials depend on whether they are appointed by an elected body or directly elected by voters? I investigate this question using the example of state grants for highly visible municipal investment projects. To attract these grants, mayors must prepare and submit applications to the state government. My identification strategy exploits a natural experiment in a German state where mayor elections were gradually introduced. The difference-in-differences estimation results show that elected mayors attract 7 to 8% more investment grants from the state tier in election years, while for appointed mayors there is no cycle. Results based on hand-collected data for individual mayors exclude alternative transmission channels such as changes in (self-)selection of mayors or partisan alignment in grant allocation