A recent political brouhaha in Northern Utah regarding election security has been followed up by local news coverage about how election security measures prevent fraudulent voting in Utah’s vote by mail (VBM) program.

Despite the long history of VBM in America and Utah, questions about the security of election policies like VBM continue to be politically relevant both nationally and locally. The debate about VBM and the politics around it have tended to focus on anecdotal evidence, despite the body of empirical research on the topic.

Does VBM help one political party over another? How does VBM impact voter turnout? Is fraudulent voting a problem with VBM? Empirical research has evaluated each of these questions.

Partisan impact

A common convention in election politics is the idea that higher turnout benefits Democratic candidates and lower turnout benefits Republican candidates. The logical conclusion is that VBM would benefit Democratic candidates the most, since its convenience expands the number of voters casting ballots.

The research offers little evidence that VBM generates a partisan advantage for either side.

A 2020 study published in the journal Science Advances examined 40 million individual voting records from Washington and Utah spanning 30 years, during which time both states incrementally implemented universal VBM programs. The researchers sought to determine whether the implementation of universal VBM gave an advantage to one political party’s candidates.

One of the study’s authors commented:

We ran dozens of analyses and every single time we found no impact in partisan vote shares. So whether you’re advocating for vote-by-mail because you think it’s going to be really good for your party or advocating against it because you think it’s going to be bad for your party, you’re probably wasting your time.

Another 2020 study published in the journal Proceedings of the National Academy of Sciences (PNAS) came to similar conclusions. Examining data from 1996 to 2018 from the three states (Washington, Utah and California) that implemented universal VBM over that time, the researchers analyzed whether universal VBM increased the share of votes for one of the two major political parties over the other.

After examining the evidence, the authors “conclude that VBM does not have meaningful partisan effects on election outcomes. … Universal VBM does not appear to tilt turnout toward the Democratic party, nor does it appear to affect election outcomes meaningfully.”

The results of these studies on the partisan impact of VBM may, in part, be explained by broader scholarly work on voter turnout published in a book in 2020. The book “refutes the widely held convention that high turnout in national elections advantages Democratic candidates while low turnout helps Republicans.”

Looking at five decades’ worth of election data from presidential, gubernatorial, and U.S. Senate and House elections, the authors find no nationwide correlation between voter turnout and the partisan share of the vote over time or for specific offices. Rather, the authors find:

In some states, across the decades, gubernatorial and senatorial contests show a pro-Democratic bias to turnout; in others an increase in turnout helps Republicans. The pattern repeats for House elections during the 1970s, 1980s, 1990s, 2000s, and through the 2010s. … [I]t is the participation and abstention of easily influenced, less engaged citizens—peripheral voters—that move the outcome between the parties. These voters are the most influenced when the short-term forces of the election—differential candidate appeal, issues, scandals, and so forth—help the parties. Since these influences advantage Republicans as often as Democrats, the oscillation in turnout that coincides with pro-GOP and pro-Democratic forces leaves turnout rates inconsequential overall.

Since the data show that voter turnout lacks a systematic partisan impact, it stands to reason that VBM would similarly lack such an impact through its effect on voter turnout.

Voter turnout

The results from research on VBM and voter turnout have been mixed. A 2021 study from scholars at Stanford and UCLA examined turnout in the 2020 election, when many states significantly expanded VBM programs due to the pandemic. They noted that “states newly implementing no-excuse absentee voting for 2020 did not see dramatically larger increases in turnout than states that did not.” They concluded that pandemic-driven expansions of VBM “mobilized few voters” and that “voter interest appears to be far more important in driving turnout.”

The 2020 PNAS study found that “universal vote-by-mail modestly increases overall average turnout rates.” The increase is estimated to be about 2 percentage points.

There is some evidence in particular states of a significant boost in turnout from VBM. A 2020 study of universal VBM in Colorado estimates that it increased 2014 voter turnout by 8 percentage points. A 2018 study of universal VBM in Utah estimates that it increased 2016 voter turnout in counties that chose to implement it by 5 to 7 points compared with counties that stuck with a traditional voting system relying on polling places. Both studies concluded that VBM has the effect of increasing turnout among groups of voters that traditionally have low turnout rates (e.g., young voters, blue-collar workers, voters with less formal education, and voters of color).

A report from MIT’s Election Data + Science Lab summarized the research on VBM and voter turnout by concluding that “the safest conclusion to draw is that extending VBM options increases turnout modestly in midterm and presidential elections but may increase turnout more in primaries, local elections, and special elections.” Meanwhile, a report from the Stanford Institute for Economic Policy Research adds an additional wrinkle:

We do find that universal vote-by-mail has big effects on changing how people vote, with many more people mailing in their ballots now that they have the default option to do so. The main effect of universal vote-by-mail, prior to COVID, is not to change who votes, but to change how people vote.

In summary, the research seems to indicate that VBM can significantly increase voter turnout in targeted ways (e.g., among specific groups of voters or in a particular local geography). As the geographic area in question expands, the impact of VBM on voter turnout tends to shrink. Access to VBM, however, seems to lead to significant changes in how people vote: When given the option, many voters choose to vote by mail.

Fraudulent voting

The MIT report mentioned two features of VBM that drive concerns over fraudulent voting: (1) the possibility of voter coercion when a ballot is filled out away from a voting center, and (2) the security of mailing a ballot versus voting at a polling place. According to the MIT report, “even many scholars who argue that fraud is generally rare agree that fraud with VBM voting seems to be more frequent than with in-person voting.”

However, despite the theoretical possibility of fraud under VBM, significant evidence of such fraud has not materialized in the research. The Stanford Institute for Economic Policy Research report stated that “political scientists and election administration experts have evaluated claims related to voter fraud for many years, repeatedly finding little evidence to support such claims.”

A 2021 article published by PNAS evaluated specific claims of voter fraud in 2020 from the campaign of former President Donald Trump. After statistically analyzing voter data relevant to each claim, the authors concluded that none of the claims are substantiated by the data.

Conclusion

Because public confidence in the integrity of voting systems is an essential foundation of our republic and critical to the health of our democracy, different systems and methods of voting must continually be evaluated and compared. Fraudulent voting is a possibility in every election, no matter when and how people choose to vote, and that potential threat must be continually met and turned back in every election cycle.

Fortunately, when it comes to vote by mail in Utah this is exactly what is occurring – there are more than 20 layers of security currently embedded in Utah’s vote-by-mail program (with additional layers being proposed and considered by election officials). Despite the political controversy brewing out of norther Utah regarding VBM, the scholarly research finds that VBM is neutral in partisan impact, modest in its effect on turnout, and secure in practice.