Published online by Cambridge University Press: 12 April 2016
In this paper I borrow Roger Myerson’s game-theoretic model on incentives for the success of democracy, adapt the model to an experimental setting, derive testable predictions from it, and test those predictions in the laboratory. Results show that (i) corrupt leaders are replaced more often in federal democracies than in centralized ones; (ii) in the centralized case subjects converged to an equilibrium in which leaders act responsively and are reelected; (iii) honest presidents are reelected at the same rate in both institutional designs, and (iv) the president acted less responsively in the decentralized case. I also run robustness checks and test for the effect of the language used in the experiment. The data show that the results are robust with regards to experimental choices.
Rodolpho Bernabel, PhD Candidate, Department of Politics, New York University, 19 West 4th Street, New York, NY 10012 ([email protected]). This study was presented at the Midwest Political Science Association Conference, the American Political Science Association Annual Meeting, the University of Texas Graduate Conference in Public Law, and graduate student workshops at New York University and Yale University, where the author received valuable feedback. The author would like to thank Rebecca Morton, Sanford Gordon, Christopher Dawes, Tiberiu Dragu, Eric Dickson, Gregory Huber, Dominik Duell, Kai Ou, the editor, and two anonymous readers for their invaluable help, comments, and suggestions. All remaining errors are the author’s own responsibility. The author also greatly appreciate the funding provided by the Center for Experimental Social Science at New York University. To view supplementary material for this article, please visit http://dx.doi.org/10.1017/psrm.2016.17