Do coalitions really cause larger government expenditures? : mixed evidence from a regression discontinuity design

This paper measures the causal effect of coalition vs. single-party governments on fiscal policies using a data set of 396 municipalities in the German state of North Rhine-Westphalia in the time period 1985-2004. Using a regression discontinuity design to take the endogeneity of the type of government into account, we exploit a discontinuity that comes through the change from a coalition to a single-party government at 50% of the seat share of the strongest party. Our results point to a significant effect of the type of government on personnel expenditures, while we do not find significant results for material spending and investment expenditures. These results differ substantially from simple OLS estimates. -- Legislative organization ; regression discontinuity design ; local fiscal policy ; coalition governments ; government spending ; panel data