While the party politics literature commonly assumes the existence of stable parties, party splits are relatively common in both established and new democracies. The effect of these splits on the electoral behaviour of voters, the electoral fortunes of individual parties and the development of party systems varies a lot. More specifically, the supporters of the original party before a split may switch to one, both or none of the successor parties. While this variation has important implications for electoral outcomes and the shape of party systems, the party politics literature does not systematically account for it. This paper seeks to fill this gap by studying how votes are redistributed between successor parties after a split. For example, when do the supporters of the “parent” party continue to vote for it even after it splits, when do they shift their support to a splinter party and when do they switch to third parties? In order to answer these questions, the paper draws on the established theories of voting behaviour. In order to test the theoretical expectations derived from these theories, the paper draws on the new previously unused dataset recording around 300 splits in 24 established and new European democracies in the post-war period. In addition, the analysis of the survey data will explore the micro-level implications of the theoretical expectations. By considering how the change in party supply affects voter choice, the paper provides an important contribution to the literature on voting behaviour. Furthermore, its findings also provide important insights into the effect of party instability on party systems.
