This paper presents empirical evidence of a positive effect of council size on government spending using a data set of 2,056 municipalities in the German state of Bavaria over a period of 21 years. We apply a regression discontinuity design to avoid an endogeneity bias. In particular, we exploit...