Disproportionate Effects of Preferential Voting Under Proportional Representation: Evidence from the Czech Republic

Preferential voting in open-list and flexible-list proportional representation systems is usually framed as a supplementary mechanism that allows voters to correct party-controlled candidate rankings while preserving overall predictability. However, preferential voting in the two latest parliamentary elections in the Czech Republic suggests that under certain conditions, preferential votes may produce outcomes that are difficult to anticipate and that substantially reshape the allocation of seats within party and pre-electoral coalition lists. Based on the Czech case, this paper aims at explaining when and why preferential voting can become a source of electoral unpredictability rather than a marginal corrective tool, even within electoral rules that have remained unchanged over time. Empirically, the paper utilises candidate-level electoral data from parliamentary elections in the Czech Republic since 1996 to examine which candidate characteristics are associated with the distribution of preferential votes. A regression analysis employs variables such as candidacy within pre-electoral coalitions, gender, age, incumbency status, and ballot position, examining how voters allocate preferential votes within party lists and whether the relevance of these factors is stable or has changed over time. Building on this, the paper interprets the findings as indicative of broader conditions under which preferential voting may play a disproportionate role in electoral results. The candidate-level patterns are situated within a wider set of structural and contextual considerations, including pre-electoral coalitions, growing demands for descriptive representation, the increasing personalization of electoral competition, and targeted efforts to encourage preferential voting. Additionally, as pre-election polling typically does not incorporate preferential considerations, voters are unable to anticipate the majority effect of preferential voting. Together, these conditions help to clarify how relatively small and potentially coordinated groups of voters may cause outsized influence, allowing preferential voting to function as a "wild card" in the Czech Republic and raising questions about whether similar dynamics may plausibly arise in other countries with other electoral rules and institutional contexts.