A Comparison of Salience and Proximity Models of Voting Evidence from Belgium and Switzerland

Voters’ issue perceptions recently have gained prominence as determinants of electoral behavior. Theories of issue voting follow either a salience-based or proximity-based approach. Salience theories emphasize issue reputations, asserting that electoral support is determined by parties’ ownership of issues – shifts in electoral support are caused by shifts in the public salience of issues over time, causing parties with strong reputations on salient issues to gain electorally. Spatial approaches emphasize issue positions, and posit that voters prefer parties with issue positions that are similar to their own: shifts in electoral support are a function of parties strategically altering their positions on issues to maximize electoral support. Yet, work comparing both approaches as explanations of voters’ electoral support is rare, let alone work that does so in a comparative design. Given that proximity-based models are more cognitively demanding, we expect that the politically most sophisticated voters would rely on it, whereas salience-based models are likely to have a stronger impact on less sophisticated voters. We present novel evidence from recent elections in Belgium and Switzerland that allows us to model individual voters’ electoral choices, accounting for key issue perceptions on a variety of issues. Our data allow to directly test salience (issue ownership) and spatial models (issue congruence) of electoral behavior.