The Endurance of Identity-Based Voting: Evidence from the United States and Comparative Democracies

TL;DR

This paper shows that identity-based voting, especially racial identity in the US, remains dominant across democracies, with digital mobilization amplifying this trend and issue-based voting remaining limited.

Contribution

It provides empirical evidence from 20 democracies that identity-based voting persists globally, challenging the notion that issue voting can replace identity politics.

Findings

Race is the strongest predictor of party affiliation in the US.

Identity-based voting remains prevalent across different democratic systems.

Digital mobilization amplifies identity-based voting trends.

Abstract

This study demonstrates the persistent dominance of identity based voting across democratic systems, using the United States as a primary case and comparative analyses of 19 other democracies as counterfactuals. Drawing solely on election data from the Roper Center (1976 through recent cycles), we employ OLS regression, ANOVA, and correlation tests to show that race remains the strongest predictor of party affiliation in the US (p < 0.001), with White voters favoring Republicans and Black voters consistently supporting Democrats (85% since 1988). Income, education, and gender exemplified by gaps like 10 points in 2020 further shape voting patterns, yet racial identity predominates. Comparative evidence from majoritarian (e.g., India), proportional (e.g., Germany through 2025), and hybrid (e.g., South Korea with a 25 point gender gap) systems reveals no democracy where issue based voting fully supplants identity based voting. Digital mobilization amplifies this trend globally. These findings underscore identity enduring role in electoral behavior, challenging assumptions of policy driven democratic choice.