Some states, including Missouri, require voters to provide a valid excuse in order to receive an absentee ballot in the mail as an alternative to in-person voting. Other states allow voters to request a ballot without any excuse, or automatically mail all voters a ballot. Due to the COVID-19 pandemic, many states, including Missouri, expanded the availability of absentee voting for the 2020 primary and presidential elections. As a result, the use of mail-in ballots nearly doubled with respect to previous election years. Studies have shown that allowing all registered voters to request a mail-in ballot without any excuse requirement can result in a moderate increase in voter turnout. There have been no indications that mail-in ballots are associated with large increases in the share of votes obtained by one party or with increases in voter fraud.
Every state in the U.S. has some form of absentee voting (sometimes called “mail-in voting” or “vote by mail”), which allows voters to cast ballots through the mail rather than voting in person at a polling location. There are three state-level approaches to absentee voting eligibility (Figure 1). Sixteen states, including Missouri, Arkansas, and Kentucky, require a valid “excuse” as determined by the state to request an absentee ballot. The types of excuses required vary by state and include old age, disability, absence due to active service in the military, and religious practice. Missouri requires one of the following excuses to request an absentee ballot: absence on election day from the jurisdiction in which the voter is registered, incapacity or confinement due to illness or physical disability, religion, employment as an election authority, incarceration, or participation in the address confidentiality program.1
Missouri is also one of three states that requires a notarized signature on the absentee ballot envelope.2 Missouri voters using absentee ballots must also provide a copy of valid identification if they registered by mail and have not voted in person.1 Some states allow voters the chance to rectify a ballot if it fails the signature verification process, but Missouri does not require local election authorities to allow this type of ballot “curing”. (See our Science Note on this topic here.)
Twenty-six states, including Iowa, Kansas, and Oklahoma, and Washington, D.C. allow voters to request a mail-in ballot without any excuse requirement (this is referred to as “no-excuse absentee voting”). Eight states automatically mail ballots to all registered voters for an election (“universal mail-in voting”)
During the COVID-19 pandemic, several states expanded eligibility for absentee voting. Thirteen out of the 16 states which typically require an excuse to request an absentee ballot, including Missouri, accepted concerns over COVID-19 as a legitimate excuse.3 Of the states that were not already implementing universal mail-in voting, 15 states mailed absentee ballot applications and 8 states mailed absentee ballots to all registered voters.
In the 2016 and 2018 general elections, 24.9% and 27.4% of votes, respectively, were cast as absentee ballots throughout the United States.4 In the 2020 presidential election this increased to 46% of votes cast as absentee ballots. There was a large difference in absentee ballot usage based on presidential choice: 58% of all votes for Joe Biden were absentee ballots, while 32% of people who voted for Donald Trump did the same.5 However, this split is unique to the 2020 election; in 2016 for instance, 23% of Republicans voted absentee, as well as 23% of Democrats.6 Voters over 65 were more likely to vote by mail than other age groups, as were Asian American and White voters over other racial and ethnic groups.
In the 2020 presidential election, preliminary research finds states that implemented no-excuse absentee voting for the first time saw similar increases in voter turnout to those that did not implement similar changes.7 This indicates that, for the 2020 election, no-excuse absentee voting had little effect on voter turnout.
Absentee voting eligibility rules in Texas provided a natural experiment to explore the effects of no-excuse absentee eligibility on turnout. In Texas, voters over the age of 65 have access to no-excuse absentee voting; voters who are 64 years old do not, but are expected to possess similar characteristics and voting attitudes as 65 year olds. In 2020, 65 year old voters turned out at almost an identical rate as 64 year olds, despite being able to vote absentee without an excuse.7 The rate of 65 year old voters in Texas who did vote absentee increased by about 9.5% compared to 64 year olds, but in-person voting decreased by about 9.5% as well. Data from Texas for previous general elections found that no-excuse absentee voting increased voter turnout from 1.2% to 2.9% for 65 year olds, with higher percentages in midterm election years, suggesting that access to absentee voting may increase voter turnout when voter attention is low.
In some states, mail-in ballots are automatically sent to every registered voter in the state. Universal mail-in voting has been estimated to increase voter turnout by 2–8%.8-10 The increase in voter turnout is smaller for those who are already likely to vote, and larger for voters who have a less active voting history.10 In California, the adoption of universal mail-in voting in some small precincts (less than 250 people) decreased voter turnout by about 1% overall, but increased turnout of 18-24 year olds by 2%.11 This was attributed in part to poor communication and a lack of “get out the vote” efforts during the transition to mail-in voting, resulting in many ballots being rejected due to late arrival.
Evidence from Utah, California, and Washington indicate that universal mail-in voting does not differentially affect either party’s turnout or total vote share.9 In another study on voters in Texas during the 2020 election, the Democratic share of turnout for 65 year olds who had access to no-excuse absentee voting was 0.22% higher than 64 year olds, for whom no-excuse absentee voting was unavailable.7 These studies indicate that mail-in voting does not provide any strong partisan advantage.
Recent studies and preliminary research have found that there is no evidence for increased voter fraud resulting from the expanded use of absentee ballots.12,13 States with and without universal mail-in voting were found to have similar voter fraud rates; in either case, fraud rates were lower than one improper vote per million voters. Additionally, analysis of absentee ballots in Washington from 2011–2018 found that only 14 votes out of a total 4.5 million cast during that period appear to have been improperly cast using deceased individuals’ information.