In an effort to provide support for his fact-free outbursts, President Trump appointed a commission to examine voter fraud.
It turns out that the biggest fraud is the commission itself.
Any reasonable examination of the issue would have to grapple intelligently with three core questions:
• Is voter fraud a significant problem?
• Will the proposed remedies alleviate the alleged problem?
• Do the proposed remedies strike the right balance between reducing alleged fraud and protecting the right to vote?
Trump’s Commission has failed on every score.
Start with whether there is really a problem.
Kansas Secretary of State Kris Kobach, the Commission’s Vice Chair, claimed to know of at least 100 cases of double voting in his state alone.
Hold to the side for a moment whether 100 cases out of 1.1 million votes (9 100,000ths of a percent) is “a lot.”
In the end, Kobach’s claim was demonstrably fraudulent. He has successfully prosecuted just nine cases. These “successes” included provisional ballots that were never counted.
But the Commission’s fact problems go well beyond individual cases.
The methodology they use to assert large numbers of double votes is fundamentally flawed.
Kobach set up his own organization, Interstate Voter Registration Crosscheck Program, to ferret out voters casting ballots in more than one place.
They flag any two voters with the same name and birthdate as potential double voters.
Kobach seems unfamiliar with an issue known to every Stats 101 student—the birthday paradox.
Since there are only 366 possible birthdays, if you select a random group of 367 people there is a 100% chance that two of them share the same birthday. With just 70 people though, the probability of two individuals sharing the same DOB reaches 99.9 percent.
Some fancy math would extend this to names and birthdates. But it’s no surprise that nearly 2 million registrants on the national voter file share the same name and birthdate. As a group of academic researchers found, that is about what you expect by chance alone.
Using Kobach’s flawed assumptions, Crosscheck claimed Iowa has over 100,000 double voters.
The academics who performed the analysis referred to above, were able to use the last four digits of Social Security numbers in Iowa to further reduce the matching error.
They found only three cases in which registration records with the same names, and common SSN4s, were both used to cast a ballot. And access to the full Social Security number may have demonstrated these three pairs were really different people.
Once again, sizeable fraudulent claims by Kobach. Precious little factual evidence behind them.
When it comes to the second and third criteria listed above, the Commission’s standards are even lower.
How would requiring an ID stop double voting? It wouldn’t.
How about purging those with the same name and birthdates in different jurisdictions? Such a “reform” would remove about 200 valid registrations for every one registration that could be used to cast a double vote elsewhere.
Which brings us to the third core issue outlined above — the balance between reducing fraud and depriving legitimate voters their constitutional right.
Requiring IDs to vote denies some people their rights. Twenty-one million Americans do not have a government issued ID.
Purging those identified as registered in two places will deprivers 200 people their right to vote for every double vote it may stop.
Americans have a clear sense of the appropriate balance here. A poll we conducted with North Star Research for the Bipartisan Policy Center found 61 percent said it was more important to make “sure that every individual who has the right to vote is allowed to exercise that right,” while 35 percent prioritized, “Making sure that no one commits voter fraud and harms the rights of legitimate voters.”
By refusing to deal honestly with this, or any of the real questions at stake, the Kobach Commission is perpetrating a massive fraud.
Mellman is President of The Mellman Group and has helped elect 30 U.S. Senators, 12 Governors and dozens of House Members. Mellman served as pollster to Senate Democratic Leaders for over 20 years and as President of the American Association of Political Consultants.