Local Voters Have Their Reasons. Mapping Voting Motives in Municipal Elections in Belgium

How the electorate determines its vote is one of the most studied but still much contested questions in political science. Contemporary assertions point to the interaction between predisposition and information affected by long and short term (contextual) factors. Given this complex decisional environment, attention should be given to how voters ultimately interpret their own electoral choice. This refers to the subjective perception and expression of the vote. Existing research shows this is a mixture of knowledge and judgement, attitude and conviction, perception and intuition. To determine their choice, voters appear to be led by a political map with cognitive and affective directions. Some fairly general and deeply engrained, others focused on specific events preceding the elections. This approach has often been used to let voters reflect in their own words about their electoral choice. These are the voting motives often described in national electoral research. There has been much less concern with such motives at the local level although two main tendencies can be discerned. Some consider local as second order national elections. In this line of thinking, voters will mainly hold national motives (rewarding or punishing nationally governing parties, expressively voting for smaller and/or newer national political parties, etc.). Others emphasize place-bound patterns and dynamics in voting motives (acquaintance with specific candidates, local issues, assessing the locally governing majority, etc.). Still, more research is needed beyond those theoretical presumptions to empirically determine motives for the local vote. Therefore, this paper aims to address two questions: 1. Which types of motives do voters attribute to their choice in local elections? 2. What explains similarities and differences in (1)? To answer these, we draw on the data of the Belgian Local Elections Study 2018. Designed as a non-predictive exit poll conducted in a stratified sample of 45 municipalities on election day, the study contains nationally representative data of a bit less than 4000 individual voters. In the related survey, respondents were invited to answer an open-ended question on their voting motives for the municipal elections. The content of these answers has been coded through IRaMuTEQ software and analyzed by the Alceste-method. This produced a number of classes linguistically capturing voting motives. The classes could be summarized along two dimensions. The first is relatively general and juxtaposes continuity to change. Trust in and support for a list of the departing majority or distrust and support for a list of the opposition divide the concomitant voter groups. The desire for change also appears to be stronger in municipalities confronted with a political scandal. The second is more specific and concerns personal notoriety and proximity vis-à-vis program and substance. It distinguishes voters in rural municipalities from those in more urban ones. Also, those higher educated indicate more often programmatic or substantial motives. The map thus has little national directions. Voting in local elections does not appear as of second order. Hence, local voters predominantly have general and specific place-bound reasons.