Patterns of Electoral Participation in Local and National Elections: Evidence from a Large-Scale Survey in Fourteen Countries of Central and Eastern Europe

Local elections ale usually classified as yet another type of „second-order elections”. However, in contrast to the more extensively studied “second-order” elections, that is, regional and European ones, we still know surprisingly little about the municipal elections. It is so even when it comes to the most fundamental issues, such as electoral participation. The model of second-order elections offers only a black-box assumption that voter turnout in local elections should be lower than in first-order (usually parliamentary) elections, but a closer look at the patterns of electoral participation in local elections suggests that it is not a universally observed regularity. Local elections seem to be more appealing to citizens living in smaller communities, because of a stronger social pressure to vote, friends-and-neighbors mobilization or an increased probability to cast a decisive vote. Consequently, the electorates casting a ballot at the elections of local councilors differ from those electing national MPs, and it is likely that these differences are systematic rather than random. In our paper we try to shed some light on the above puzzles by examining evidence from a survey of over 20,000 individuals from over 1,000 municipalities located in 14 countries of Central and Eastern Europe. We compare predictors of the respondents’ intentions to vote (or not to vote) in national and local elections, looking both at individual and contextual factors determining electoral participation. We attempt to offer a more nuanced, fine-grained picture of patterns of participation and abstention in local elections. Our preliminary analyses focus on the selectivity of electoral mobilization and its potential contingency upon the size of a local community. We test whether or not the contextual effect of a municipality size on voter turnout in local elections tends to persist even if the individual-level predictors of electoral participation are accounted for.