Reciprocity and Democratic Accountability

In this paper, we introduce reciprocity concerns in a political agency model with symmetric learning about politicians’ ability and moral hazard. Voters with reciprocity concerns are both prospective—that is, seek to select competent politicians—and retrospective—that is, reward fair actions and punish unfair ones. We focus on how electoral incentives induce politicians to exert effort (electoral control) and how voters remove incompetent politicians (electoral screening). We show that taking voters’ reciprocity concerns into account has important normative implications and can overturn results from standard models that neglect them: increasing transparency about the incumbent’s effort improves electoral control if and only if voters have sufficiently strong reciprocity concerns; increasing benefits from office improves electoral control if and only if voters have sufficiently low reciprocity concerns. Moreover, we show that reciprocity concerns can affect electoral screening, by affecting the competence threshold incumbents must clear to ensure reelection, generating incumbency advantages or disadvantages.