Mail voting fraud: Data points to low risk and high benefits for voters | Brookings

The U.S. has a long history of mail voting. Large-scale use of mail ballots originated during the U.S. Civil War, when some soldiers were allowed to vote remotely after absentee voting laws were passed in their home states. Today, mail voting is widely used around the world, with more than 30 countries—including Switzerland, Germany, and South Korea—allowing voters to cast ballots by mail.

Mail ballots are widespread across the United States. For example, in 2024 alone, the U.S. Postal Service processed over 99.2 million mail ballots.

Several studies indicate that certain forms of mail voting can increase voter turnout. A 2009 study for the Pew Charitable Trusts found that no-excuse absentee voting increased voter participation by about 3 percentage points in comparison to states with excuse absentee voting, when controlling for other factors that may impact turnout.

First, we find that cases of fraud involving any form of mail ballots were very rare. Across the entire country, and utilizing a maximally inclusive estimate, between six and 46 cases of mail voting fraud were identified in each general election. To calculate the percentage of mail voting fraud in a given year, we divided the total number of mail voting fraud cases by the total number of mail votes cast for each general election. We find an average total mail voting fraud percentage across the 2016, 2018, 2020, and 2022 general elections of only 0.000043%, or about four cases of mail voting fraud out of every 10 million mail votes.

Although the database we utilized self-identified as “not…comprehensive,” we have reason to believe that this limited scope did not meaningfully distort our overall findings. The News21 database, which is among the most extensive databases of its kind currently publicly available, includes over 2,000 cases, of which we found 1,605 related to alleged voting fraud between 2000 and 2012. That averages to roughly 134 cases per year. Even if we assume this same case rate persisted for the four general elections examined, and assume the 134 cases of voting fraud were all mail voting fraud,8 that would still translate to only about 2.5 cases of mail voting fraud per 1,000,000 mail votes. This indicates that, even under assumptions that greatly inflate the frequency of mail voting fraud, the resulting probability of fraud remains negligible.