Retrospective pledge voting and mistrusting citizens: Evidence for the electoral punishment of pledge breakage from a survey experiment

Do voters punish government parties for breaking an election pledge? How do preexisting opinions affect retrospective pledge voting (RPV)? This study hypothesizes that pledge breaking entails a fundamental loss of confidence resulting in a lower probability to vote for a government party, and that mistrusting citizens punish pledge breakage more severely. A representative survey experiment (N ¿ 10,000) with a one-dimensional treatment, that is carried out in Germany, provides evidence for these hypotheses. These findings have major implications for our understanding of political representation and the voter–party-relation. The results of this paper stress the importance of promissory representation and provide evidence that the electorate keeps tabs on government parties, but they also hint at the crucial role of individual predispositions.