Steve Rosenfeld at The Fulcrum reports on a forthcoming paper on turnout and mail-in voting:

States that mailed a ballot to every registered voter in 2020’s presidential election saw voter turnout increase by an average of 5.6 percent, and turnout was even higher among infrequent voters, according to the first peer-reviewed academic study of 2020 mail voting.

The turnout increase was slightly larger than previous studies of mail-based voting conducted before 2020’s election, where 71 million-plus voters, including voters in swing states, turned to mailed-out ballots in response to the COVID-19 pandemic.

Less convenient versions of mail-based voting, such as local officials sending all voters an application to fill out and submit before receiving a ballot, or a state relaxing that application’s criteria but still requiring voters to apply, had smaller impacts on lifting voter turnout.

“We tested these different approaches,” said Eric McGhee, a co-author of the new study and a Public Policy Institute of California senior fellow. “And of those three approaches, we found that the only consistent effect, either on turnout or on the composition of the electorate, was that mailing everybody a ballot elevates turnout.”

McGhee’s paper, “Vote-by-Mail Policy and the 2020 Presidential Election,” which will be published later this year by the scholarly journal Research and Politics, was co-authored by PPIC’s Jennifer Paluch and the University of Southern California’s Mindy Romero.

The research, which surveyed county-level turnout from 1992 through 2020, buoys claims by voter advocates that getting mailed-out ballots directly into voters’ hands—especially in states with automatic voter registration—could be pivotal in 2022’s midterm elections, when historic patterns suggest November’s turnout may fall to two-thirds of the presidential election.