Concurrent elections and turnout: Causal estimates from a German quasi-experiment

Conventional wisdom suggests that scheduling elections concurrently with other contests increases turnout. However, causal empirical evidence in favor of this claim is scarce, at best. Moreover, it is a priori unclear whether theory would predict a higher turnout if two low-office elections are combined instead of a combination of a low-office election and a major electoral race. Using a quasi-experiment and a difference-in-differences estimation approach at the German municipality-level, I provide evidence that combining two low office elections significantly increases turnout. (C) 2016 Elsevier B.V. All rights reserved.