Prejudice Based on a Non-Partisan Identity: How Brexit Identities Shape Political Behaviour in the UK

More voters in the UK now hold a Brexit identity rather than a party identity: they identify as Leaver or Remainer rather than with one of its political parties. The conflict over the British membership in the EU created a division that cuts across party lines even several years after the EU referendum. Do these new political identities generate prejudice akin to party identities? And what is their role for political behaviour in comparison to party identities? To answer these questions, we test a set of hypotheses derived from a typology of prejudice developed by Allport (1954). To examine our hypotheses empirically, we conducted a set of six pre-registered online survey experiments with a diverse (non-student) sample of more than 900 respondents. Our results show that Brexit identities are at least as important for political behaviour as party identities. We do not find that a common party identity trumps a shared Brexit identity. In cases where we find that one identity exerts a greater bias than the other, it is a shared Brexit identity. Some voters are more positively inclined towards information from a politician who shares their Brexit position, even if that MP does not share their party affiliation. This is remarkable given the strong bias that party affiliations tend to create in the absence of a conflict such as Brexit. Moreover, we find that Brexit identities bias citizens’ judgments when it comes to democratic norm violations. For instance, when presented with text that criticises their own Brexit position in a hateful manner, only 13 percent of respondents supported its publication. However, when the text criticised the other side, 49 percent of respondents supported the publication of the text. This shows how Brexit identities bias reactions in statistically significant and substantively important ways: voters do not approve hateful speech when it harms their side, but they are much more likely to approve it when it harms those whose Brexit position they do not share. When asking respondents to make judgments on infringements of campaign laws, we also found statistically significant biases in citizens’ reactions. After reading information about their own Brexit side having broken campaign laws, 54 percent of respondents supported a fine that was issued. When the text described campaign laws broken by the Brexit side they did not share, 77 percent of respondents supported the fine. We did not however find a similarly grave bias when it comes to approval of violence that harms political competitors. We discuss the findings in light of the debate on the disruptive role of affective polarisation and how its behavioural implications cause problems for the functioning of liberal democracies.