Mandatory mail-in voting hurts neither Democratic nor Republican candidates

Mandatory mail-in voting leads to a slight uptick in voter turnout — for both Democrats and Republicans.

That’s the conclusion that researchers came to after analyzing more than 40 million individual voting records from Utah and Washington — two states that have switched from in-person voting to almost exclusively mail-in voting over several years — as well as nearly 30 years of nationwide county-level voting data.

The finding, published August 26 in Science Advances, suggests the current political zeitgeist that mail-in voting benefits one party over another is false, says political scientist Michael Barber of Brigham Young University in Provo, Utah. When voters cast ballots by mail, he says, “neither party is hurt.”

Although Utah and Washington do not reflect voting patterns in the country as a whole, the authors note, because Washington leans blue and Utah red, the two states demonstrate how mail-in voting could affect voter turnout by party. 

With the country in the midst of a pandemic, policy makers and public health experts have proposed voluntary mail-in voting as one way for Americans to safely vote in November. And many states have made such voting easier. A recent Washington Post analysis shows that over 80 percent of voters in the country can now cast their ballots by mail.

But some Republican voters especially are wary of the process. A Gallup poll from May 2020 found that 83 percent of Democratic respondents were OK with their state allowing all residents to vote by mail, but only 40 percent of Republicans were. In a separate question, 76 percent of Republicans said mail-in voting would lead to more fraud compared with 27 percent of Democrats. 

Barber and John Holbein, a political scientist and public policy expert at the University of Virginia in Charlottesville, combed through multiple datasets to understand how mail-in voting has influenced voter turnout overall and by party in recent decades. The researchers first identified counties in six states that adopted more universal mail-in voting procedures from 1992 to 2018, namely mailing all constituents a ballot before Election Day and limiting or eliminating in-person voting. Those states were Washington, Oregon, California, Utah, Colorado and Nebraska.

Using census data, the researchers identified the total number of eligible voters in the counties that adopted mail-in voting and those nationwide that did not. The team then compared voter turnout across counties by looking at election data showing how many eligible voters actually cast a ballot in each midterm or presidential election during that quarter century. Barber and Holbein also compared turnout within a given county before and after the implementation of a vote-by-mail policy. Another dataset, Dave Leip’s Atlas of Elections, enabled the researchers to evaluate each party’s share of votes per county in midterm and presidential elections.

The researchers then drilled down into individual voting patterns for Washington and Utah. Altogether, the researchers evaluated more than 40 million voting records from 2012 to 2018 in Utah and from 2002 to 2016 in Washington.

The individual data from Washington and Utah showed voter turnout increased slightly for Democrat, Republican and independent voters. And the county-level analysis likewise showed that mail-in voting led to a 1.8 to 2.9 percentage point increase in voter turnout.

Partisan vote shares in counties that adopted mail-in voting tilted 0.7 percentage points in favor of the Democratic Party, the team found. But with the margin of error in the results between –0.7 and 2 percentage points, that increase was not statistically significant and could potentially go in either direction in any given election, Barber says. As it was, only 1.5 percent of analyzed counties had an electoral margin as narrow as 0.7 percentage points between parties anyway, the authors say.

The authors used “best statistical practices” by corroborating their nationwide, county-level results with the individual results from Washington and Utah, says Donald Green, a political scientist at Columbia University. But “I’m not sure that Republican leaders would be willing to concede 0.7 percentage points,” he says.  

Broad geographic analyses of this sort can obscure smaller communities, adds political scientist Jean Schroedel of Claremont Graduate University in California. Her work shows that mail-in voting disenfranchises Native Americans on reservations, many of whom lack regular mail access and have a long-standing distrust of nontribal government entities. “Native people don’t trust voting — full stop — but they really don’t trust voting by mail,” Schroedel says.

Even if states sort out how to protect the vote of vulnerable community members, such as keeping some physical polling places open, the findings may do little to allay other recent concerns over mail-in voting, Barber acknowledges. Those include fears over the United States Post Office’s ability to keep pace with such a huge influx of mail and the possibility of mail-in ballots getting thrown out for having an allegedly faulty signature or arriving late.  

“I don’t know what’s going to happen in 2020 with vote by mail,” Barber says. “This has gotten so unnecessarily messy.”