Reaches and Limits of Nationalization in U.S. Politics

• In Event: Nationalization and Other Long-Term Trends in Voting

Abstract

A growing literature in political science focuses on the “nationalization” of U.S. politics, referring to the extent to which national political actors and issues influence state and local political activity. While research in the area has grown to cover many different offices, levels of government, and time periods, we have lacked a unified picture of how nationalized these contests are relative to each other. In this paper, I attempt to provide this picture by adopting a latent trait approach to quantifying nationalization. In doing so, I model electoral results in any given contest as a function of a geographic unit’s underlying ideology rather than Presidential vote shares themselves, thus interrogating the deeper question of the nationalization literature regarding the degree to which the dimensions of politics are similar across all contests. I apply the approach to an expanded sample of contests: presidential, senatorial, and gubernatorial contests since the early 1900s, all statewide contests since 2015, and a sample of local contests from the largest U.S. counties. The result is a fuller, unified picture of downballot nationalization in the U.S., with potential consequences for the quality of representation in low-information contests.

Author

• Derek Holliday, University of California, Los Angeles