Disentangling the effect of the economy on vote choice: a survey experiment

It is often observed that voters punish governments for poor economic performance. This is commonly explained with reference to economic voting, i.e. effects of the economy on vote choice which work through competence evaluations. It is often overlooked that the impact of the economy on the vote can also be studied in a spatial voting framework: economic downturns may change voters’ policy preferences, which in turn affect the ideological proximity to political parties and therefore may trigger a change in vote choice.

Which of these two explanations explains vote choice best? We identify two problems. First, most research only tests one explanation. Second, most studies analyze cross-sectional observational data. This is a problem, because economic perceptions, which are often used to analyze the effect of the economy, are potentially endogenous, i.e. related to the same variables explaining vote choice. As such, it is impossible to disentangle to which extent voters change vote choice because they change their evaluation of the candidates or because they change policy preference. For this reason, we propose a survey experiment conducted in Denmark and UK in which we combine a frame giving poor economic forecasts with (a) messages stressing different policy implications and (b) blame attribution to the current government. By manipulating issue salience, policy considerations and competence evaluations we can disentangle the effect of the economy of different components of vote choice. In addition, we evaluate the moderating of effect of political knowledge, partisanship, employment status and personality traits.

Co-author: Dr. Paul Marx, University of Southern Denmark