Determinants of Local Election Resource Distribution Nationwide

Abstract

Election administration resources like polling places and workers are integral to the voting process. Though scholars have learned much about the effects of election administration on outcomes such as voter turnout or confidence in elections, less is known about what determines the distribution of election administration resources in the first place. Previous studies have suggested a number of explanations for how election resources are distributed, including as a service to constituents, out of racial bias, due to administrative capacity and demand, and through partisan lenses. However, evidence in support of these theories tends to come from only a handful of counties or states and, potentially as a result, is mixed. As such, current literature does not speak well to how elections are administered in the United States as much as how they are administered in specific states or counties. Using the density of early voting and Election Day polling places and workers (number of places or workers per 1,000 voting age people) from 2012 to 2018, this study examines several explanations for what drives election administration resource distribution nationwide. Results suggest that administrative demand (e.g., voting age population size) is the most influential factor. Evidence also suggests that partisanship plays a role, but whether partisan effects reflect partisan preferences for election administration or to increase co-partisan turnout is unclear. Conflicting results are found regarding the constituent service hypothesis, while evidence is found suggesting a lack of racial bias. These results suggest that, nationally, election administration resources are distributed to places where they are needed, but that partisan lenses may partially shape that distribution while racial biases may be artifacts of specific localities.