Incumbency Effects in Open-List PR Elections

Most analyses of the incumbency advantage focus on the case of majoritarian electoral systems. This paper analyzes the incumbency advantage in the context of open-list proportional representation. While some authors have found evidence of an incumbency advantage when relying on ‘controlling on observables’ identification strategies (e.g. Dettmann et al. 2017), researchers using regression discontinuity designs (RDD) have found no evidence of an incumbency advantage (Golden and Picci 2015) under open-list PR. This paper discusses certain challenges for analyzing incumbency effects in open-list PR elections and highlights how the results are affected by different identification strategies. In addition, it is argued that the incumbency effect in PR settings has two dimensions. First, incumbency can improve a candidate's intra-party position as reflected by a better list placement in the next election (intra-party dimension). Second, voters might prefer voting for incumbents (vote-earning dimension). The empirical analysis focuses on the election results of more than 18,000 candidates in the open-list PR system in Poland and shows that incumbency has a strong impact on the intra-party dimension, but not on the vote-earning dimension.