Dissertation Project

My dissertation project examines how voting costs influence political participation by focusing on an under-researched population: residentially mobile individuals. Residential mobility is a nearly universal phenomenon, as almost everyone moves at least once during their life. However, political science research has not thoroughly explored how relocation affects voter behavior, beyond identifying negative correlations between mobility and turnout.

Further, there are underutilized opportunities to leverage residential mobility to enhance our understanding of ways in which voting costs affect political participation. The administrative, social, and informational costs incurred by a mover vary depending on the attributes of their move. For example, those who move between counties will likely face higher administrative costs than those who do not, as the National Voter Registration Act of 1993 and other state-level policies make it easier for within-county movers to re-register to vote than cross-county movers.

To address these gaps in the literature, I introduce a novel method for measuring residential mobility using commercial records of real estate deed transactions. Using these data, I capitalize on plausibly exogenous variations in the geographic and political boundaries crossed during a move. I employ modern methods for causal inference with observational data to identify the extent to which heightened administrative, social, and informational costs affect a mover’s decision to cast a ballot.

I find that turnout is very sensitive to administrative costs, but not social and informational costs. While the sample in this work consists of movers, the findings have broader implications for political behavior research and election administration best practices. If election administrators and lawmakers aim to encourage political participation in their communities, the most effective use of resources would be to reduce the administrative burdens associated with voter registration.

Writing

Please click each title below for a link to the text

Lost in Transition: Participatory Disparities Between Residentially Mobile Voters

In Progress

Movers are less likely to participate in electoral politics than those who are residentially stable. Many studies of residential mobility on turnout view moving as a binary treatment: a person either relocates or does not. However, the degree to which moves increase voting costs depends on the political and administrative boundaries crossed during the course of a move. I examine how different types of moves influence political participation. To this end, I present a taxonomy that categorizes moves, providing a framework for understanding the increased voting costs associated with each move type. Additionally, I propose a novel method for measuring residential mobility using commercial records of real estate transactions. I find that moves in which an individual changes election jurisdictions (often at the county level) substantially reduce voter turnout. These cross-jurisdiction movers are approximately 10 percentage points less likely to vote than within-jurisdiction movers. I argued that this effect is primarily due to the heightened re-registration costs for cross-jurisdiction movers. I find no consistent evidence that increased informational and social costs substantially reduce turnout among the residentially mobile.

Intra-County Moves and Fail-Safe Voting Under the National Voter Registration Act of 1993

Under Review

Residential mobility can greatly impact the accuracy of voter registration lists and impede voters' ability to cast ballots. Section 8 of the National Voter Registration Act of 1993 offers a partial remedy to these challenges through fail-safe voting. This provision allows registered voters who move within their election jurisdiction without updating their address to correct their registration and vote on Election Day. Thus, fail-safe voting acts as a county-level form of portable voter registration. Despite their potential to improve list accuracy and expand voter eligibility, these fail-safe provisions are not thoroughly explored in the election administration literature. By analyzing data from North Carolina's voter files, I aim to address this gap by estimating the number of registrants who use fail-safe voting and identifying those most likely to do so. In each of the 2020, 2022, and 2024 elections, approximately 18,000 to 24,000 individuals were fail-safe voters. Highly mobile voters, such as younger and minority individuals, are more likely to rely on fail-safe voting. Fail-safe voting effectively corrects registration addresses and ensures individuals can vote at the correct location.

Automatic Voter Registration and List Maintenance

w/ Marc Meredith

2023 - Published in Pandemic at the Polls

This chapter argues that the primary benefit of Automatic Voter Registration (AVR) policies is not the enfranchisement of new voters, but rather voter list maintenance. When states enact AVR, there are significant increases in the number of voter registration address updates. This leads to cleaner voter rolls and greater voter eligibility for residentially mobile individuals. Additionally, this chapter discusses the utility of AVR as a tool to conduct registration when in-person registration opportunities are limited, such as during the 2020 COVID-19 pandemic.