The Examiner said that Brooks “did not provide evidence to support the claim” but instead pointed to a speech he made on Jan. 7 that laid out his calculations. In the address on the House floor, he declared: “Noncitizens overwhelmingly voted for Joe Biden in exchange for the promised amnesty and citizenship and, in so doing, helped steal the election from Donald Trump, Republican candidates, and American citizens all across America.”
The speech is actually rather detailed in showing Brooks’s math. But it relies on dubious assertions, some of which we have fact-checked before, to come up with these numbers.
Brooks lists what he calls “exhibits” to make his case. But he cherry-picks lines and data. For instance, in Exhibit B, he quotes from the 2005 Commission on Federal Election Reform, headed by former president Jimmy Carter and former secretary of state James A. Baker III.
The report did indeed say that “noncitizens have registered to vote in several recent elections.” But the numbers it cited were relatively small — a few hundred here and there in scattered elections. The issue is not that there may be some instances of fraudulent voting, but whether it is so widespread that it would change an election result. Most reputable studies have found that in virtually all cases, any vote fraud by noncitizens is infinitesimal.
(Carter has been so annoyed at how Republicans have twisted the findings in the report, especially on voting by mail, that he issued a statement in March saying he believes “voting by mail can be conducted in a manner that ensures election integrity.”)
Similarly, in Exhibit C, Brooks cites a Government Accountability Office report as saying that “up to 3 percent of people on voter registration lists are not U.S. citizens.” But when you dig into the report, you see that it gives three examples of the percentage of noncitizens found in jury pools (which are drawn from voter registration lists): one-third of 1 percent, less than 1 percent and 1 to 3 percent. Brooks simply cites the highest number. Moreover, the courts have supplemented the jury pool with other lists, such as driver’s licenses, which do not require citizenship, so the fact that noncitizens appear on the master jury list may not indicate that they are registered to vote in those jurisdictions.
The core of Brooks’s calculation rests with Exhibit D, a citation of a 2014 paper that has been highly controversial ever since a summary was published in The Washington Post’s Monkey Cage blog. Brooks claims that report surveyed 339 noncitizens, of whom “8% admitted voting in American elections.” This calculation is derived from the number of people in a 2008 survey cited in the study who said that they had voted.
But the article said “our best guess, based upon extrapolations from the portion of the sample with a verified vote, is that 6.4 percent of non-citizens voted in 2008 and 2.2 percent of non-citizens voted in 2010.”
The study relied on an opt-in Internet survey designed for citizens only, which is one reason it quickly came under fire from other political scientists.
In any case, the article currently has this note at the top: “The post occasioned three rebuttals (here, here, and here) as well as a response from the authors. Subsequently, another peer-reviewed article argued that the findings reported in this post (and affiliated article) were biased and that the authors’ data do not provide evidence of noncitizen voting in U.S. elections.”
When Trump and his allies started citing the study in 2016, Old Dominion University, where the primary author Jesse Rickman is a professor, published an article headlined: “ODU Voting Study Swept up in the Fake News Vortex.” Rickman told Vice he was so tired of having the study misquoted that he wished he hadn’t written it.
And yet, here it is again, being cited on the floor of the House.
Then, in Exhibit E and Exhibit F, Brooks cites estimates for the undocumented population of 11 million and 22 million. The latter comes from a 2018 Yale University study and is not widely accepted by immigration experts. But that’s how Brooks gets his high-end number.
The Department of Homeland Security has an annual estimate of how many undocumented immigrants there are in the United States, including how many are under the age of 18 (about 1 million people). Brooks fails to account for that. DHS says that in 2018, there were 10.3 million undocumented adults in the United States.
Finally, in Exhibit G, Brooks assumes Biden has a 60-point advantage (80 percent for Biden, 20 percent for Trump) in noncitizen voting, again because of the disputed ODU study. Recall that the sample size was rather small. Moreover, DHS indicates that about 70 percent of undocumented immigrants are Hispanic — and Trump actually made gains in Hispanic support in 2020. It’s still possible Biden would have received a majority of votes from noncitizens, but after crunching the numbers in exit polls, it appears Biden would have about a 40-point advantage.
To recap: Brooks claimed that Biden received 900,000 to 1.7 million votes from noncitizens. That is based on assuming 11 million to 22 million noncitizens, 8 percent of whom cast ballots, at a rate of 80 percent for Biden.
But, as we have shown, there are about 10.3 million undocumented immigrants old enough to vote. The percentage who in theory are illegally voting range from one-third of 1 percent to as much as 6.4 percent (though that’s dubious). The Biden advantage is likely about 40 percentage points.
If you accept the disputed ODU study — which most experts do not — that means between 227,000 and 659,000 votes might have been cast by undocumented immigrants. Using the numbers in the GAO report, you end up with a range of as little as 31,000 or as much as 309,000.
The Biden advantage would then yield between 12,000 to 264,000 votes from noncitizens — out of nearly 160 million votes cast. This range is being rather generous to Brooks, as the 6.4 percent figure is not widely accepted by political scientists.
The biggest problem for Brooks is that Biden’s popular vote margin was over 7 million. Even if Brooks’s wildly inflated numbers were correct, Biden easily would have won the most votes.
Of course, the electoral college determines the winner. But most undocumented immigrants reside in states that were not especially competitive, such as California, Texas, Florida, New York, New Jersey and Illinois, according to DHS.
The margins in Georgia and Arizona were close enough that only if you accept 6.4 percent, which we regard as an outlier, might the vote of undocumented immigrants have made a difference, given the number of such immigrants in those states. But Biden still would have won the electoral college without Arizona and Georgia. And, as we keep repeating, the preponderance of evidence is that relatively few noncitizens vote in federal elections.
A spokesman for Brooks did not respond to a request for comment.
The Pinocchio Test
Brooks claimed that if only lawful votes by eligible Americans were counted, Trump would have been reelected. His own faulty math does not prove that, and he relies on dubious calculations and a disputed study to make his case. Brooks earns Four Pinocchios.
Send us facts to check by filling out this form
Sign up for The Fact Checker weekly newsletter
The Fact Checker is a verified signatory to the International Fact-Checking Network code of principles