Arizona’s dealing with a record-breaking crush of early ballots.
Does that mean we should get ready for record-breaking fraud?
Not if you believe the existing studies of mail-in ballots, in use for years in many states.
On the other hand, it probably means a nail-biting wait to find out how the election turned out.
Seven days before the election, some 66 million Americans had already voted, compared to 58 million in 2016. Democrats make up 41% of those early voters and Republicans 34%. The share of young voters had doubled while the share of white voters has declined.
However, roughly twice as large a share of Republicans say they plan to vote in person on Election Day compared to Democrats.
President Donald Trump has made an issue of the national shift to mail-in ballots, accelerated this year by the effort by many states to prevent people from crowding into polling places in the midst of a pandemic.
Trump has predicted widespread fraud, especially in states that send mail-in ballots to all voters — whether they’ve applied for a ballot or not.
Four states routinely send out mail-in ballots to all voters in all elections, including Washington, Oregon, Nevada and Utah. Four others will send all voters mail-in ballots this year due to the pandemic, including California, Nevada, New Jersey and Vermont.
Most states provide mail-in ballots for anyone who asks for one, including Arizona.
In Arizona, people remain on the early ballot list once they’ve applied. As a result, about 78% of voters have received mail-in ballots this year.
A handful of states allow for mail-in balloting only if a voter offers a reason they can’t vote in person, including Texas, Louisiana, Mississippi, Tennessee and Indiana.
Elections officials in most states say any kind of ballot fraud is rare in the U.S. and elaborate safeguards remain in place. Voters who use mail-in ballots sign the ballots on the outside and elections officials check that signature with the one on file in the register’s office before they count the vote. If the voter forgets to sign or the signature doesn’t match, elections officials can use a bar code and printed address label to contact the voter to “cure” the ballot.
One of the most comprehensive studies of voter fraud involved Oregon’s mail-in balloting system, where all voters receive a mail-in ballot in all elections. The state has mailed 100 million ballots since 2000 and so far documented a dozen cases of fraud, according to a study by the Brennan Center for Justice.
That mirrors other studies showing a low incidence of voter fraud in the U.S. One attempt to track all allegations of voter fraud between 2000 and 2014 found just 43 cases out of a billion votes cast nationally, according to the study by Loyola Law School scholar Justin Levitt.
Nonetheless, the avalanche of mail-in ballots could cause two significant problems — delayed results and discarding challenged ballots.
The delay in the vote count stems from the extra trouble it takes to confirm mail-in ballots before they’re run through the voting machine. Some states don’t even start verifying the mail-in ballots until after Election Day — significantly delaying final results. Other states — like Arizona — verify signatures before Election Day then run those early ballots through the vote counting machines along with the ballots cast at polling places. However, even it still takes more time to count the mail-in ballots turned in at the polls or placed in drop boxes a day or two before the election. Election workers must verify the signatures before counting those votes after Election Day.
As a result, if any contest is close, elections officials might not have a decisive, final vote total until a week or more after Election Day.
But there’s one more problem with mail-in ballots.
A small percentage of ballots are rejected due to some problem, like a missing or mismatched signature, failure to find the voter on the voting roles or other issues. Moreover, mail-in votes that arrive by mail after Election Day won’t get counted.
In 2016, about 1% of the early ballots sent through the mail were rejected.
Ironically, the mail-in ballots of Democrats and minority groups are more likely to be discarded than votes of Republicans and white voters, according to a study by the University of Florida political scientist Daniel Smith.
The study analyzed millions of postal ballots cast in Florida in 2016 and 2018. The votes of Hispanic and Black voters were twice as likely to be rejected as the votes of whites. The votes of people aged 18-21 were four times as likely to get rejected as voters aged 45-64.
So if the overall rejection rate’s about 1%, this means 2% of minority votes and 4% of the votes of young people may get rejected — a significant total, especially for Democrats.
However, those high rejection rates could merely reflect inexperience in filling out and signing the mail-in ballots. Voters who voted in person in the previous three elections and then shifted to a mail-in ballot for the first time were three times as likely to see their ballots rejected, even when controlling for age, race and political party.
Overall, Democrats were more likely to have their ballots rejected as Republicans — and are also more likely to vote by mail.
So why have Republicans often opposed vote-by-mail and Democrats favored it?
That has to do with the impact on turnout.
States that shifted to vote-by-mail between 1992 and 2018 saw a 2% or 3% increase in turnout, according to a study by Brigham Young University political scientist Michael Barber.
Another study in Colorado found a whopping 9% increase in turnout in 2014 when the state shifted to an all-mail ballot format.
Still, it’s unclear how the massive shift nationally to mail-in ballots will play out this year. So far, only 28% of Democrats say they plan to vote in person on Election Day compared to 65% of Republicans.