Beiträge zur Jahrestagung des Vereins für Socialpolitik 2012: Neue Wege und Herausforderungen für den Arbeitsmarkt des 21. Jahrhunderts - Session: Microeconometrics and Statistics D09-V1
In this paper incumbency effects in Federal Elections and Mayoral Elections in Germany are estimated using a quasi-experimental design which allows for causal inference under a set of rather mild assumptions. Relying on nonparametric and parametric estimation procedures and exploiting a recently developed bandwidth selection criteria, incumbency effects for the two major parties in Germany cause an increase in vote share of 1.4%-1.7% in Federal Elections. Analyzing Mayoral Elections, the causal effect of incumbency is about ten times larger with an increase in vote share of 14%-17%. Both results are robust with respect to the inclusion of further covariates. These huge differences might possibly be explained by the differences in visibility and popularity of mayors compared to candidates in Federal Elections.