Causal Effects of Universal Mail-In Voting on Voter Turnout and Composition

• In Event: Constraints on Election Administration

Abstract

Five States—Colorado, Hawaii, Oregon, Washington, and Utah—routinely conduct their elections almost entirely by mail. Four more states—California, Nevada, New Jersey, and Vermont—sent mail ballots to all registered voters for the November election this year. The impact of universal mail-in voting on voter turnout, voter composition, and ultimately partisan vote share, however, is not well understood. Since counties are important units in terms of election administration—as it is often the case that county election officials decide the number of polling stations, set precinct boundaries, and enforce voter identification laws— existing work analyzing cross-county differences are unlikely to discover the true effect of universal mail-in voting.
In this paper, we exploit a state law that requires election officials to send mail ballots to registered voters in some but not other congressional and legislative districts within a large county to estimate the causal effect. In order to estimate the causal effect, we employ a geographic regression discontinuity design. To control for these confounding effects, we compare election cycles before and after universal mail-in voting was introduced to voters in certain congressional and legislative districts and compares permanent absentee voters who by definition are not affected by the policy and the rest of the electorate.
The paper proceeds as follows. First, we look at the aggregate voter turnout, voter composition, and partisan vote shares by congressional and legislative districts, before and after the introduction of universal mail-in voting, and by permanent absentee status. In order to implement the geographic regression discontinuity design, we then geo-locate voters based on individual-level administrative records. With the geo-locations at hand, we proceed to estimate the effect of sending all registered voters mail-in ballots on voter turnout, voter composition, and partisan vote shares. Finally, we discuss the political implications as states move to expand or restrict access to mail-in voting in future elections.

Authors

• R. Michael Alvarez, California Institute of Technology

• Yimeng Li, California Institute of Technology