Unmasking Accountability

Abstract

As local conditions come to reflect extralocal forces, signals of government competence grow more obscure. Yet we know relatively little about how voters evaluate incumbent performance in the context of interdependence. We use a series of simulated voting tasks to examine three theoretical possibilities: blind retrospection, rational discounting, and benchmarking. Across five experiments requiring “voters” to judge performance in a setting that obscures incumbent competence, we find consistent evidence of benchmarking—subjects rewarded incumbents, capable or otherwise, who outperformed a peer. Benchmarking was evident in information processing, information seeking, and both hard and easy tasks. The disposition to benchmark was also generally robust to the availability of information that clarified incumbent competence. Our findings advance the study of performance voting, especially its underlying mechanisms, and raise questions about the availability of performance information across domains of government action.

Citation

Hart, A. & Matthews, J.S. (2022). Unmasking accountability: Judging performance in an interdependent world. Journal of Politics. https://doi.org/10.1086/716971