Changes in Voter Behavior Amidst Electoral Insecurity and Perceived Voter Fraud

Abstract

The expansion of voting access through no-excuse absentee voting, early in-person voting, or mandatory vote-by-mail (VBM) during the COVID-19 pandemic has grown increasingly polarized along party lines in the United States. Widespread claims of fraudulent, missing, or otherwise miscounted ballots may have further eroded voter confidence in the nontraditional forms of voting mentioned above. Given the extensive nature of misinformation targeting mail-in voting and absentee voting, it is unclear how voters, particularly those affiliated with the Republican party, have responded and/or changed their preferred voting method in practice. This paper evaluates the impact of widespread allegations of fraudulent voting methods on individual-level voter behavior preceding and following the 2020 U.S. election cycle. Recent data collected from the North Carolina State Board of Elections details comprehensive voter registration data and voting history of over 4 million North Carolina residents spanning primary and general U.S. elections from 2016 to 2022. Utilizing a series of cross-sectional dynamic difference-in-differences analyses, we find that Republican voters were significantly less likely to vote by mail relative to Democratic or Independent voters following widespread allegations of fraudulent voting in early 2020 by prominent GOP leaders. However, the same subset of Republican voters was more likely to vote early absentee. These effects are more persistent among nonwhite Republican voters and in Republican-majority counties. As a robustness check for pre-treatment trends, an additional D.I.D. model with a fake placebo treatment further bolsters results. Taken altogether, these findings suggest that post-pandemic voter restrictions laws targeting nontraditional methods of voting may lead to changes in voter turnout and/or vote share along party lines.