The COVID-19 pandemic has threatened the American electoral system in unprecedented ways. In response to concerns about the safety of voting in person during the pandemic, many states have expanded opportunities for voters to cast their votes by mail. This expansion has triggered an extraordinary partisan debate over the legitimacy of the election, with President Trump and some supporters insisting that widespread vote-by-mail massively favors the Democratic Party and invites fraud.
Research rejects both of these claims.
Bringing data to evaluate these specific claims is important. The public needs to understand whether these claims are accurate, how our elections are administered, and whether the 2020 election will be legitimate. Since the spring, my research group and I have been collecting and analyzing data to study both of these claims.
There are three major types of voting by mail. The most intense version of the policy is one in which the state mails every registered voter a ballot—called “universal vote-by-mail.” In universal vote-by-mail states, voters typically have three options: They can mail their ballots in after filling them out; they can return the ballots in person; or they can decide to vote in person at a polling center.
Prior to 2020, only Colorado, Hawaii, Oregon, Utah, and Washington employed universal vote-by-mail, while California was in the process of phasing it in across counties. In response to COVID-19, three more states, Nevada, New Jersey, and Vermont, along with the District of Columbia, have implemented the policy, while California accelerated its ongoing implementation. Montana has also begun to phase in the practice.
Most other states have “no-excuse” absentee voting, in which any voter may choose to vote by mail by asking for and receiving a mail-in ballot.
A final group of states only allows absentee voting with a valid, legally delineated excuse— which means that the rate of absentee voting in these states is generally quite low.
Although President Trump has at times appeared to cast doubt on all of these forms of voting by mail, he has reserved his most bitter criticism for states that pursue universal vote-by-mail.
Let's turn first to the claim that universal vote-by-mail dramatically benefits Democrats. We focus on this issue because it affects the strategic incentives of the parties to support or oppose vote-by-mail, and because it can make the policy seem partisan to voters. That being said, whether or not vote-by-mail advantages one party or the other is, of course, not a principled reason to justify supporting or opposing the policy. If vote-by-mail facilitates democratic participation and prevents people from being disenfranchised during the pandemic, it may be a good policy even if it advantages one party or the other. Our goal is not to justify or oppose the policy but rather to understand some of its effects.
In our first study, published recently in the Proceedings of the National Academy of Sciences, we examined historical data from California, Utah, and Washington, where universal vote-by-mail was phased in over time, county by county (Thompson et al. 2020). This created a natural experiment, which we used to estimate the effects of universal vote-by-mail on overall rates of turnout, as well as on each party’s share of turnout and of votes.
We found that, in pre-COVID times, switching to universal vote-by-mail had only modest effects on turnout, increasing overall rates of turnout by approximately two percentage points. Because universal vote-by-mail has such modest effects on overall turnout, it’s not surprising that we also found that it conveyed no meaningful advantage for the Democratic Party. When counties switched to universal vote-by-mail, the Democratic share of turnout did not increase appreciably, and neither did the vote shares of Democratic candidates. Our largest estimate suggests that universal vote-by-mail could increase Democratic vote share by 0.7 percentage points---enough to swing a very close election, to be sure, but a very small advantage in most electoral contexts, and a much smaller effect than recent rhetoric might suggest.
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.
Although our historical analyses gave us reasons to think universal vote-by-mail wouldn’t have major partisan effects, there are two major differences between that study and the current situation.
First, the COVID-19 pandemic completely changes the electoral landscape. Fears of COVID may lead a wider group of voters to stay home in the absence of vote-by-mail options. In turn, this could make the effects of vote-by-mail during COVID much larger than in normal times and also more partisan. Since Democrats appear to be more enthusiastic about voting by mail than Republicans (Lockhart et al. 2020), vote-by-mail could advantage them during the pandemic in ways it didn’t before.
Second, our historical analysis focused on universal vote-by-mail, but most voters who are voting by mail in 2020 will be using absentee voting, in which they request an absentee ballot and mail it back.
For a more timely analysis, we next studied Texas’s July 14th runoff primary. This case provides us a unique opportunity to understand vote-by-mail during the pandemic, because the July 14th runoff occurred during a time of widespread COVID-19 cases in Texas and only people 65 years old and up were allowed to vote absentee without providing a valid excuse as defined by state law.
This age cutoff, first studied in a previous paper (Meredith and Endter 2016), provides us with a natural experiment of sorts that we can use to estimate the effects of expanding access to absentee voting during the pandemic.
By comparing the voting behavior of 65-year-old Texans with that of 64-year-olds in the runoff, we can evaluate how absentee voting changes turnout both overall and across the two parties, holding fixed many of the other factors in the election that both age groups experience, such as which candidates are running for office, etc.
To implement this approach, we obtained the official voter file from the state of Texas, which we combined with historical versions of the voter file for past elections. This dataset provides us with individual-level information on all voters in the state of Texas, including their birth date, whether they voted in the runoff primary, and, if so, the manner in which they voted (by mail, in person on election day, or early in person).
Figure 1 shows how absentee voting has dramatically increased during COVID-19. The plot shows the share of voters in Texas who voted absentee, across ages, before and during the pandemic. As the “Pre-2020” curve shows, in past elections, there is a noticeable jump at age 65. Almost no Texans under the age of 65 vote absentee, because the excuse requirements are quite burdensome. Absentee voting becomes more common among those older than 65. During COVID-19, as the 2020 curve in the plot shows, absentee voting has become much more popular. We find that 20 percent of 65-year-olds who voted in the runoff voted absentee, a large jump from past election cycles.
Clearly, voters appreciate the opportunity to vote absentee during the pandemic. But does it change turnout?
Figure 2 shows that the answer is no. Although we see in the plot that 2020 turnout is up across the board compared with past runoff primaries, we don’t see any jump at age 65. Even though 65-year-olds voted absentee in the July 14th runoff at much higher rates than in past runoffs, the 64-year-olds who couldn’t vote absentee also had a higher turnout rate. There is no obvious growth in the gap between 65- and 64-year-olds. Even during COVID-19, the main effect of vote-by-mail appears to be to encourage voters who would have otherwise voted in person to instead vote by mail, though it is possible that turnout is increasing for voters older than 65, something our approach cannot directly assess.
Finally, we examine the effects of absentee voting on partisan turnout. Not surprisingly, since absentee voting does not lead to any change in overall turnout, we do not find that it advantages one party over the other in 2020 during COVID-19.
Although we do find Democrats using absentee voting at much higher rates in the July 14th runoff, we find no evidence that they are expanding their turnout relative to Republicans, compared to previous runoff elections. Thus, while absentee voting is encouraging more Democrats than Republicans to switch to vote-by-mail during the pandemic, Republicans in Texas appear to have offset this phenomenon by simply voting more in person than their Democratic peers.
There are important reasons to be cautious in extrapolating these findings to November. First, our analysis focuses on a very particular Texas statewide primary runoff, and turnout behavior in that runoff might not predict turnout behavior in November, as a much broader set of more casual voters who would not pay attention to a primary runoff are likely to consider voting in a general election. Second, Texas has an unusually robust system of early in-person voting, which might offset some of the deterring effects of not being able to vote by mail. In states where voters do not have the option to vote early in person, not being able to vote by mail could reduce turnout and could advantage Republicans.
These concerns notwithstanding, the results suggest the general phenomenon of vote-by-mail mainly switching how voters vote, rather than who votes, continues even during the COVID-19 pandemic.
Having shown evidence against the claim that vote-by-mail dramatically helps the Democrats, we turn last to the assertion that vote-by-mail facilitates widespread voter fraud. Political scientists and election administration experts have evaluated claims related to voter fraud for many years, repeatedly finding little evidence to support such claims (see for example Goel et al. 2020; Hood and Gillespie 2012).
In ongoing research (Wu et al. 2020), we contribute to this body of work by evaluating a specific claim that people in President Trump’s orbit have made about universal vote-by-mail: That mail-in ballots sent to the residences of deceased individuals are fraudulently filled out, sent back in, and counted. To assess this claim, we collected official death records for the state of Washington, which we merged to the Washington state voter file in order to look for cases where a deceased individual is subsequently marked as having had a ballot counted in an election.
After accounting for potential false positives due to individuals who share the same name and live in the same county, out of roughly 4.5 million voters that we study, we find approximately 20 individuals who appear to have ballots cast in their name after they passed away, and even these cases may be clerical errors or cases where two individuals in the same county share the same name and birth date. This is a rate of 0.000004 fraudulent cases for every voter. In other words, this is a vanishingly rare problem, historically. While there may be other reasons for people to worry about the logistics of voting by mail, the casting of deceased individuals’ ballots is unlikely to be a relevant issue.
The COVID-19 pandemic has presented unique challenges to the 2020 election. Criticisms of voting by mail, spearheaded by the president, have added fuel to the fire. Our research on voting by mail suggests that the president’s two major claims about vote-by-mail—that it will massively help the Democrats and that it will encourage fraud—are unlikely to be correct.
That being said, there are important November-specific factors our research cannot address. The most important issue concerns the logistics of vote-by-mail. Historically, mail-in ballots are rejected at higher rates than in-person votes. Capacity issues in the face of an enormous surge in voting by mail could drive these rejection rates higher. And if Democrats cast more mail-in ballots than Republicans, as looks extremely likely, these higher rejection rates could mean that vote-by-mail paradoxically hurts Democrats.
As a result of these issues, the public conversation around vote-by-mail has shifted, with people beginning to focus on their entire spectrum of voting options rather than focusing only on voting by mail. While voting by mail may be the safest option for the most vulnerable voters in some states, other options may also make sense for many voters: Early in-person voting, when polls are likely to be less crowded; delivering absentee ballots by hand; or voting on election day in polling places that are taking CDC-recommended precautions to mitigate the spread of the virus.
Given the logistical challenges to voting by mail, and the probably small but real COVID-related risks of voting in person, there is no one-size-fits-all solution to voting this year. All voters should be empowered to make an informed choice concerning how they want to vote.
Beyond November, the disruption of 2020 is exposing critical gaps and long-running shortfalls in America’s election system. In 2019, prior to COVID-19, only 40 percent of American respondents in a Gallup survey reported that they were confident in the honesty of U.S. elections, a shockingly poor showing for the world’s richest country. Our inability to deliver a smooth election during COVID-19 is unlikely to help this already dismal view.
A widespread belief in the honesty of our elections is critical for maintaining the legitimacy of our democracy. Persistent doubts that those in power did not fairly win their mandate from voters, if left unchecked, threaten the stability of our country, the rule of law, and, ultimately, our prosperity and freedom. In the aftermath of one of the most difficult elections in American history, we should throw ourselves into the project of rebuilding our election infrastructure and restoring faith in our electoral system.
Goel, Sharad, Marc Meredith, Michael Morse, David Rothschild, and Houshmand Shirani-Mehr. 2020. “One Person, One Vote: Estimating the Prevalence of Double Voting in U.S. Presidential Elections.” American Political Science Review 114(2): 456–469.
Hood III, MV, and William Gillespie. 2012. “They Just Do Not Vote Like They Used To: A Methodology to Empirically Assess Election Fraud.” Social Science Quarterly 93(1): 76–94.
Lockhart, Mackenzie, Seth J. Hill, Jennifer Merolla, Mindy Romero, and Thad Kousser. 2020. “America’s Electorate Is Increasingly Polarized Along Partisan Lines About Voting By Mail During the COVID-19 Crisis.” Proceedings of the National Academy of Sciences 117(40): 24640-24642.
Meredith, Marc, and Zac Endter. 2016. “Aging into Absentee Voting: Evidence from Texas.” Working Paper.
Thompson, Daniel M., Jennifer Wu, Jesse Yoder, and Andrew B. Hall. 2020. “Universal Vote-by-Mail Has No Impact on Partisan Turnout or Vote Share.” Proceedings of the National Academy of Sciences 117(25): 14052–14056.
Wu, Jennifer, Chenoa Yorgason, Hanna Folsz, Cassandra Handan-Nader, Andrew Myers, Tobias Nowacki, Daniel M. Thompson, Jesse Yoder, and Andrew B. Hall. 2020. “Are Dead People Voting By Mail? Linking Death and Voter Records to Evaluate Voter Fraud.” Working Paper.
Yoder, Jesse, Sandy Handan-Nader, Andy Myers, Toby Nowacki, Daniel M. Thompson, Jennifer Wu, Chenoa Yorgason, and Andrew B. Hall. 2020. “How Does Expanding Absentee Voting Affect Participation During COVID-19? Quasi-Experimental Evidence from Texas.” Working Paper.