#79: Mail-In Vote Fraud
President Trump, Fox News, the Heritage Foundation (“HF”) and others repeatedly claim that there is “massive” mail-in vote fraud in Presidential elections, and that fraud is significantly greater for mail-in voting than in-person or “absentee” (excuse-required, mail-in) voting. When pressed for proof, they cite individual anecdotes of vote fraud, most of which do not even involve mail-in voting.
There are many studies that have repeatedly concluded there is very minimal vote fraud. See, e.g. https://www.brennancenter.org/our-work/analysis-opinion/false-narrative-vote-mail-fraud,https://www.npr.org/2020/06/04/864899178/why-is-voting-by-mail-suddenly-controversial-heres-what-you-need-to-know, https://www.brookings.edu/blog/fixgov/2020/06/02/low-rates-of-fraud-in-vote-by-mail-states-show-the-benefits-outweigh-the-risks/.
But rather than face claims of data bias, let’s look at HF’s own vote fraud data. First, here’s what you need to know about the HF data:
— It includes ALL types of vote fraud, not just mail-in vote fraud, and as to mail-in vote fraud, it appears to be lumped in the category of absentee voting fraud.
— It includes vote fraud from ALL elections, local, county, state, and both federal non-presidential elections and federal presidential elections. (It also includes fraud data from U.S. territories that do not have electoral votes and thus can’t determine presidential election outcomes.)
— It covers a 38-year period (1982–2020).
— It reports the data by conviction year, not election year.
So, during this 38-year period, how many instances of all types of vote fraud does the HF study identify in ALL elections? Just 1290 — out of billions of votes cast in those 38 years, or less than one case per million votes cast. And what if we just included mail-in vote fraud? That number is a mere 208 — less than one per 10 million votes cast in those 38 years.
Even those numbers are overstated since the HF data includes ALL elections, not just federal elections, and not just federal Presidential elections. Clearly, HF’s data provide ABSOLUTELY NO support for the contention that there is massive vote fraud in the United States at all, and certainly not in U.S. Presidential elections. Indeed, the data establish the EXACT OPPOSITE: there is miniscule vote fraud of any type in the U.S. and even less fraud involving mail-in voting in U.S. Presidential elections. Further, neither the 1290 cases of all fraud, or 208 cases of mail-in vote fraud, even if all had occurred in a single presidential election — not all elections in 38 years — would have determined the election outcome in any state in the 2016 election or in any Presidential election during that 38-year period.
Let’s look further. I examined each of the HF-reported incidents of fraud by state — again, still including all elections over that 38-year period, examining it by type of mail-in voting allowed: the 5 states with universal mail-in voting, the 28 states with no-excuse mail-in voting, and the remaining states that require an excuse (so-called “absentee” voting). Using the total number of presidential votes cast in 2016, I calculated a per million mail-in vote fraud rate for each state. (Again, this will substantially overstate the fraud rate because the number of fraud cases includes all elections in the 38-year period of the study, not just the 2016 Presidential election.) The per-million absentee vote fraud rate was 1.02 per million in universal mail in voting states, 1.02 per million in no-excuse mail-in voting states, but almost three times that rate (2.84 per million) in the remaining states!
Three other points:
— Although the data does not identify the number of votes affected by each instance of fraud, I looked at the case descriptions for each of the absentee vote fraud cases, focusing on those where the description did not clearly identify the fraud as occurring in a state or local election. The vast majority involved single-vote cases, e.g. an absentee voter voting in each of the voter’s 2 states of residence. But, again, because the analysis above includes all 38 years of data, and all elections, compared against only one-year of presidential election votes, the conclusions above still substantially overstate the fraud rates reported above.
— I also looked at the annual absentee vote fraud cases in the 5 states who have all-mail in voting to see if there has been any escalation of such fraud since those states began all-mail in voting. There wasn’t.
— Recently, President Trump has claimed (again without facts) that Florida’s mail-in-voting is less subject to fraud than most or all other states. That simply is false. Florida’s rate of mail-in vote fraud is greater than 27 states and the District of Columbia, based on the HF study.
The conclusion: Claims of massive fraud from mail-in voting have absolutely no factual basis. Indeed, fraud rates in states with no-excuse mail-in voting, including those with universal mail-in voting, are about 3 times less than the average fraud rate in the remaining states.
Data source: https://www.heritage.org/voterfraud.
Tomorrow: What You Need to Know about the 2020 Democratic Convention
