« APRW: Burden & Hillygus on Biased Voter Learning | Main | More Journals? »

28 February 2006

PPBW: Hainmueller & Kern on Incumbency

In the next session of PPBW, Jens Hainmueller and Holger Lutz Kern will present "Party Incumbency as a Source of Contamination in Mixed Electoral Systems." The paper uses a cool regression discontinuity design to estimate the effect of candidate incumbency on party vote shares in the German mixed-member system. Although the effect turns out to be modest in size, it is a nice demonstration of a potentially quite powerful technique. Here's what the abstract has to say:

We identify party incumbency as a new source of contamination in mixed electoral systems. We argue that incumbency not only boosts incumbents' SMD vote shares; it also has a positive effect on vote shares in the PR tier of mixed electoral systems. This effect has been overlooked in the contamination literature so far. We test our hypothesis with data from German federal elections and employ a regression-discontinuity (RD) design in order to overcome some of the obstacles to reliable causal inference in observational studies. The RD design exploits the random variation in party incumbency status that occurs when a district race is close and thus allows for causal inference under a weaker set of assumptions than the regression models commonly used in the electoral systems literature. We find that party incumbency results in a gain of about 1 to 1.5 percentage points in PR vote share, which is sufficient to potentially trigger significant shifts in Bundestag majorities.

Posted by Barry Burden at February 28, 2006 10:00 AM