Political parties not only aggregate the policy preferences of their supporters, but also have the ability to shape those preferences. Experimental evidence demonstrates that, when parties stake out positions on policy issues, partisans become more likely to adopt these positions, whether out of blind loyalty or because they infer that party endorsements signal options consistent with their interests or values. It is equally clear, however, that partisans do not always follow their party’s lead. The authors investigate the impact of three party-level traits on partisan cue taking: longevity, incumbency, and ideological clarity. As parties age, voters may become more certain of both the party’s reputation and their own allegiance. Governing parties must take action and respond to events, increasing the likelihood of compromise and failure, and therefore may dilute their reputation and disappoint followers. Incumbency aside, some parties exhibit greater ambiguity in their ideological position than other parties, undermining voter certainty about the meaning of cues. The authors test these hypotheses with experiments conducted in three multiparty democracies (Poland, Hungary, and Great Britain). They find that partisans more strongly follow their party’s lead when that party is older, in the opposition, or has developed a more consistent ideological image. However, the impact of longevity vanishes when the other factors are taken into account. Underscoring the importance of voter (un)certainty, ideologically coherent opposition parties have the greatest capacity to shape the policy views of followers.

No abstract

In a laboratory experiment, we explore the effects of group identities on the principal-agent relationship between voters and representatives. In an adverse selection framework with observable effort, voters can choose to condition their reelection choices on representatives' effort alone, beliefs about representatives' competence, or both of those jointly. We show that inducing social identities increases the weight of representatives' effort in voters' reelection decisions. Further, when voters and representatives share a social identity, representatives tend to invest less effort and their effort is independent of their competence. In contrast, "out-group" representatives compensate for lower competence with higher effort and reduce effort when voters are likely to perceive them as competent. Voters often adopt laxer retention standards for representatives who are fellow group members and are responsive to evidence of other-regardingness from out-group representatives, but some voters actively resist treating representatives with shared identity more favorably and "overcorrect" as a consequence.

In a laboratory setting, we explore strategic discrimination in principal-agent relationships, which arises from mutually re-enforcing expectations of identity-contingent choices. Our experimental design isolates the influence of the strategic environment from effects of other sources of discrimination, including statistical differences between sub-populations and outright prejudice. We find that, in a strategic setting, principals who reward agents based on outcomes more readily attribute high performance to effort when they share the agent's group identity. No such bias exists either for principals whose reward decisions are outcome-independent or for principals in a non-strategic environment. Agents in the strategic setting tend to anticipate higher demands from out-group principals, and condition their effort choice on that expectation. Because they under-appreciate this conditionality, principals tend to underestimate the effort from out-group agents.

While scholars and pundits alike have expressed concern regarding the increasingly "tribal" nature of political identities, there has been little analysis of how this social polarization impacts political selection. In this paper, we incorporate social identity into a principal-agent model of political representation and characterize the impact of social polarization on voting behavior. We show that identity has an instrumental impact on voting, as voters anticipate that political representatives' ex post policy decisions have an in-group bias. We also conduct a laboratory experiment to test the main predictions of the theory. In contrast to existing work that suggests social polarization may have a positive impact by increasing participation, we show that social polarization causes political representatives to take policy decisions that diverge from the social optimum, and voters to select candidates with lower average quality.