Strictly Sincere Best Responses under Approval Voting and Arbitrary Preferences

University of Zurich, Department of Economics, Working Paper No. 302

22 Pages Posted: 4 Nov 2018

See all articles by Carlos Alós-Ferrer

Carlos Alós-Ferrer

University of Zurich - Department of Economics

Johannes Buckenmaier

University of Zurich - Department of Economics

Date Written: October 2018

Abstract

Approval voting allows voters to support as many candidates as they wish. One advantage of the method is that voters have weak or no incentives to vote insincerely. However, the exact meaning of this statement depends on how the voters' preferences over candidates are extended to sets. We show that, under a combination of standard, well-established assumptions on the extended preferences, voters will always have a strictly sincere best response (that is, a best response ballot such that every approved candidate is strictly preferred to every disapproved one) given the ballots of other voters. The result holds for arbitrary preferences over candidates, allowing for indifferences but covering the extreme cases of dichotomous or strict preferences. As a corollary, we show that the classical strategy-proofness result for the case of dichotomous preferences on alternatives (Brams and Fishburn, 1978) holds for a larger class of preferences on sets than originally assumed.

Keywords: Approval voting, manipulation, preferences among sets, strict sincerity

JEL Classification: C72, D71, D72

Suggested Citation

Alos-Ferrer, Carlos and Buckenmaier, Johannes, Strictly Sincere Best Responses under Approval Voting and Arbitrary Preferences (October 2018). University of Zurich, Department of Economics, Working Paper No. 302, Available at SSRN: https://ssrn.com/abstract=3265187 or http://dx.doi.org/10.2139/ssrn.3265187

Carlos Alos-Ferrer (Contact Author)

University of Zurich - Department of Economics ( email )

Zürich
Switzerland

Johannes Buckenmaier

University of Zurich - Department of Economics ( email )

Zürich
Switzerland

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
32
Abstract Views
327
PlumX Metrics