Does Ranked-Choice Voting Reduce Racial Polarization? Evidence from Agent-Based Modeling and Bay Area Mayoral Elections
44 Pages Posted: 1 Apr 2021 Last revised: 5 Apr 2022
Date Written: April 1, 2022
Abstract
Does changing electoral systems from first-past-the-post (FPTP) to ranked-choice voting (RCV) induce moderation in ethnic party competition and reduce the level of intergroup polarization? Despite the importance of this question in the literature on electoral systems, party competition, and conflict management and rich qualitative studies in racially and ethnically divided societies, previous research has lacked formal and empirical evidence as to whether and why switching from FPTP to RCV yields moderation both at the elite and mass levels. This article develops a neo-Downsian spatial model of ethnic party competition under FPTP and RCV as well as an agent-based model augmented by the spatial model. By computationally simulating the policy adjustment by ethnic parties, we find that RCV induces more moderation than FPTP does while also finding that ethnic parties moderate more when they seek to maximize up to their third-choice probabilities as opposed to maximizing only up to their second-choice vote shares. To study moderation among the electorate, we also collect more than 5.5 million ranked preferences and 13,635 precinct-level data from Bay Area mayoral elections from 1990 to 2020. After estimating the level of racially polarized voting with rank clustering and ecological inference models, we show that switching from FPTP to RCV did not alter the level of racially polarized voting for the overall electorate and all pairs of racial and ethnic groups we study
Keywords: Ranked-choice voting, electoral engineering, ethnic conflict, rank data, agent-based model
Suggested Citation: Suggested Citation