The ‘voter fraud’ fraud

Many states have moved toward voting by mail for the 2020 elections due to pandemic concerns, leaving only seven states lacking this option for all voters. Members of Congress have called for national legislation for a vote-by-mail option for federal elections this year, which would cover the remaining states. President Trump and some other Republicans have resisted, arguing that mail voting risks election fraud. There’s little empirical evidence to back up this fraud claim, but there have been enough instances of absentee ballot fraud over the years to make it worth a look.

Evidence for the pro-vote-by-mail side may come from an unlikely source: A database of fraud cases maintained by a conservative think tank that raises alarms over voter fraud and is decidedly not in the pro-mail ballot camp. Its data suggests that mail ballot related fraud is actually more common in states that restrict absentee voting than in other states.

The Heritage Foundation is an established conservative think tank. It has long raised the alarm about the perceived dangers of voter fraud, most notably as a justification for strict voter photo identification laws for in-person voting. But they have also spoken out against mail-in voting, suggesting, among several complaints, that it raises an unacceptable risk of fraud.

The photo ID arguments lack merit. They have been shown to disenfranchise large numbers of people. And the only kind of voter fraud that it could prevent – in-person “voter impersonation” fraud, in which one voter shows up at the polls claiming to be a different person – is so rare as to be virtually nonexistent . 

But not so with absentee ballot fraud. Although voter fraud of any kind is rare, absentee ballot fraud is the most common type. A high-profile instance in North Carolina last year involved “ballot harvesting,” in which unscrupulous operatives collected absentee ballots and filled them out themselves. A similar case occurred the year before in a small town in Mississippi. Well-publicized cases of such fraud have also been reported in Texas over the years.

These cases are certainly the exception and not the rule. A comprehensive study of election fraud cases by Arizona State University’s News21 journalism project found only 491 cases of absentee ballot fraud out of literally billions of votes cast in U.S. elections since 2000. 

But, rare or not, if we were to move to vote-by-mail, would that significantly increase the risk of voter fraud?

Probably not, according to the Heritage Foundation’s data. The think tank compiles a database of reported instances of voter fraud or election fraud. It lists 1,277 “proven instances of voter fraud” in the 50 states over decades, dating back to 1979. The database caveats that it does not purport to be an “exhaustive or comprehensive list.” But given its repeatedly expressed concerns about voter fraud in general and mail ballot fraud in particular, it seems unlikely it would leave out many reported instances of that kind of fraud.

The database includes many categories of the types of voter fraud involved:  Registration fraud, voter impersonation fraud, illegal “assistance” at the polls, etc. But only some are relevant to the specific concerns about mailed ballots: “Fraudulent use of absentee ballots” and “vote buying.”

If widespread use of mail ballots truly did engender fraud, one would expect that the instance of such kinds of fraud would be lower in the 16 states that strictly regulated absentee balloting as opposed to the 28 states, which allowed anyone to vote absentee (“no excuse” absentee voting states) or the five states that automatically mailed ballots to all voters for them to mail back or drop off (“vote-by-mail” states). 

Instead, the opposite is true. An examination of the Heritage Foundation database for the period 2000-2020 shows that reported instances of such fraud per capita are actually higher in “strict” states than either “no excuse” states or complete “vote by mail” states.

For these types of fraud, within the 29 “no-excuse” absentee states, there was one reported fraud case for every 2.4 million persons. This compared favorably to one such case for every 1.6 million persons in “vote by mail” states, and even more favorably than the strict states, with one fraud case for every 740,000 persons. Although mail ballot fraud was by no means frequent in any of the states, it was actually more common in the states that took the stricter, more Trump-favored approach.

It is true that pure vote-by-mail states had a slightly higher rate than no-excuse states, potentially aiding the argument that more mail ballots leads to more fraud. But the vote-by-mail rate was only 1.5 times that of the no-excuse rate, while the strict states’ rate was twice that of vote-by-mail, and over three times that of no-excuse states. This doesn’t seem consistent with the notion that liberalizing mail voting increases fraud.

The 2000-2020 period was chosen to address the risk that the database was more “spotty” the further back in time one went. But the same relationship among the three types of states holds true when you go all the way back to 1979, the database’s earliest reported instance of fraud.

Obviously, this doesn’t prove causation. It doesn’t seem plausible that restricting access to absentee ballots increases mail ballot fraud. But the data, taken from a leader in the “mail balloting leads to fraud” camp, certainly seems to undercut the assertion that increasing access to mail ballots will significantly increase the rate of voter fraud. 

This should not be surprising. States have long experience with checking the provenance of absentee ballots. They check to make sure that the voters sending in the ballot are properly registered, for one thing. And they compare the signature on the absentee ballot with the voter signature on file, for another.

Nor is it the case that there is no time to switch to mail balloting for the November election. About nine states have already done so this year, with more likely on the way. Ohio even switched to mail voting for its presidential primary in the past month, with only a few weeks’ lead time.

There are many good reasons to move to mail balloting in this pandemic year. There don’t seem to be any good reasons not to. And vote fraud certainly isn’t one of them.

Steve Mulroy is a University of Memphis law professor and author of the book “Rethinking US Election Law.” Memphis Law student Hayden Phillips assisted with database research for this article.