Polarization, Abstention, and the Median Voter Theorem

TL;DR

This paper challenges the median voter theorem by exploring how polarization, abstention, and third-party options can lead candidates away from centrist positions, especially under realistic voter preference distributions.

Contribution

It introduces a model analyzing three mechanisms—abstention, third-party influence, and bimodal preferences—that can cause candidates to polarize rather than converge to the center.

Findings

Candidates may fail to converge to the center under certain voter distributions.

Polarization can increase due to third-party and bimodal voter preferences.

Abstention and ideological extremes undermine the median voter theorem's predictions.

Abstract

The median voter theorem has long been the default model of voter behavior and candidate choice. While contemporary work on the distribution of political opinion has emphasized polarization and an increasing gap between the "left" and the "right" in democracies, the median voter theorem presents a model of anti-polarization: competing candidates move to the center of the ideological distribution to maximize vote share, regardless of the underlying ideological distribution of voters. These anti-polar results, however, largely depend on the "singled-peakedness" of voter preferences, an assumption that is rapidly loosing relevance in the age of polarization. This article presents a model of voter choice that examines three potential mechanisms that can undermine this finding: a relative cost of voting that deters voters who are sufficiently indifferent to both candidates, ideologically motivated third-party alternatives that attract extreme voters, and a bimodal distribution of voter ideology. Under reasonable sets of conditions and empirically observed voter opinion distributions, these mechanisms can be sufficient to cause strategically-minded candidates to fail to converge to the center, or to even become more polarized than their electorate.