Policy or Affect? The Effects of Party-Voters Issue Congruence on Affective Polarization

Is affective polarization simply a reflection of ideological polarization, or is this relation more complex? Recent research is divided as to how ideological and affective polarization are related. Though a clear association has been established, it remains unclear to what extent affective polarization can be explained by its ideological counterpart, or even which is the direction of causal flows. Even further, it is unclear whether it is the actual policy differences that drive affective polarization. Indeed, some contend that partisan identity is the principal mechanism of affective polarization, and that policy preferences merely influence affective polarization because they signal partisan identity. Still, in this paper, we argue that in multiparty systems, the nature of the relation between ideological and affective polarization is largely determined by issue-specific preferences. Notably, likes and dislikes towards out-parties are not simply determined by left-right ideological distance, but rather by issue-specific congruence and incongruence, which vary for different out-parties. That is to say, likes and dislikes about an out-party A rest on (in)congruence on a specific issue position, which is not necessarily relevant to explain likes and dislikes for out-party B. To test this argument, we rely on the RepResent Belgian Panel (RBP) of 2019, an original panel survey which is complemented with party positions extracted from with a Voting Advice Application (VAA). This dataset allows to measure affective polarization and citizens-party issue congruence on 18 policy positions. We can thus look at the effect of issue-specific ideological polarization on affective polarization. Additionally, this design allows us to detect the effect of actual ideological polarization rather than perceived ideological polarization, since party positions are determined by their manifestos rather than by citizens’ perceptions. We find that congruence on salient issues, such as immigration, is an important predictor of affective polarization, while congruence on less salient issues is less important. We conclude that the salience of issues signals partisanship more evidently, and therefore fuels affective polarization.