"I WON THE ELECTION," he declared, adding that the bureau would soon analyze the remaining 32 million voting records and that the number of illegitimate votes would "explode."
The report Trump was referring to was real, but highly unusual. The seven-page document posted on the Census Bureau website did not list any authors.
While the use of the "Census Bureau" label implies an air of seriousness and professionalism, Misty Heggeness, a former senior adviser for the Census Bureau, told Democracy Docket that the lack of clear authorship signified that it was "suspect of any integrity or rigor" and said it would be hard to trust unless career bureau staff put their name on it.
Sure enough, NPR reported on Wednesday that "work on the analysis was not conducted by career civil servants at the Census Bureau, and the team behind it included individuals affiliated with the America First Policy Institute," a Trump-aligned think tank that has helped to turn the president's election denial narrative into Republican policy.
The group has helped to push for aggressive voting rights restrictions in states, and its director for "election integrity" policy, Thomas Lane, was involved in the effort to create a slate of fake electors in Arizona, one of several states where Trump claimed to have won in 2020.
Earlier this year, according to CNN, Trump "installed a group of partisan analysts" from AFPI inside the Census Bureau. But their identities have not been publicly disclosed, and other bureau officials who spoke to NPR did not identify which AFPI officials worked on the new report "because they fear retaliation at work."
Bureau officials told CNN that the lack of listed authors suggests that the career employees were "sidelined."
Experts who reviewed the report have identified several "fatal" flaws that make its conclusions difficult to take seriously.
The authors of the report used a commercial voter database and matched its records to those from the Internal Revenue Service, Social Security Administration, State Department, and Department of Homeland Security, and matched noncitizen records to those identified as voters in 2020.
As Yunior Rivas and Jim Saksa of Democracy Docket explained, there are numerous problems with this approach:
Record matching... is fallible. The Census Bureau previously published research in 2014 specifically devoted to measuring false matches in the same identification system...
Census researchers have previously warned that the likelihood of successfully linking administrative records can vary across demographic groups, potentially introducing bias when researchers analyze only records they can confidently match.
That limitation is particularly significant when the result in question is relatively small compared with the enormous volume of records being searched: A mistaken link between an eligible citizen’s voter record and another person’s immigration record could produce a false positive.
The report also does not address another obvious danger: mistaking a naturalized citizen for a noncitizen due to an older immigration record... If evidence of a person’s citizenship is missing or fails to link correctly, the methodology can treat the existing noncitizen record as evidence of the person’s status at the time they voted. The report does not estimate how frequently such a misclassification could occur.
Aaron Reichlin-Melnick, a fellow at the American Immigration Council noted that the system used to find noncitizen voters is "riddled with false positives." According to the Texas Tribune, when states like Texas and Missouri used the tool, it resulted in hundreds of voters being flagged incorrectly.
But even if the Census Bureau report were factually bulletproof, it would be far short of what Trump needs to claim that he "won" the election. The number of supposedly illegitimate votes is not anywhere near enough to swing the result in any of the states where Trump lost to former President Joe Biden, who won the national popular vote by more than 7 million.
The report claims to have found 1,100 noncitizen votes in Arizona, which Biden won by 10,457; 400 noncitizen votes in Georgia, which Biden won by 11,779; and 500 noncitizen votes in Wisconsin, which Biden won by 20,682. Since voting is secret, there is also no basis to assume that all of these votes went to Biden.
Commerce Secretary Howard Lutnick, whose department contains the Census Bureau, has said the report was "just the beginning" and that the next one "will likely show tens of thousands of additional illegal voters.”
But the first report says it analyzed the vast majority of those who voted in 2020. Assuming that the remaining 32 million votes had similar match rates, it would only generate about 6,000 more supposedly illegitimate votes.
Even if every single one of those votes went for Biden, which can't be proved, it would still not be enough to swing a single state to Trump, let alone enough for him to declare victory.
Trump used this report to call for Congress to pass his "SAVE America Act," which would abolish most forms of mail-in voting and introduce new proof of citizenship requirements that voting rights experts have warned would lead millions of eligible voters to be wrongly kicked off the rolls.
He has also signed an executive order directing states to verify their voter lists against a federal citizenship verification system that has been found to erroneously flag many eligible voters.
Amanda Carpenter, an editor at the nonpartisan group Protect Democracy, said the Census Bureau report was "nothing but junk data science" and that Lutnick "is abusing the Census Bureau data to generate sensational headlines that Trump wants."
"He has been searching for these massive pots of fraudulent votes for years," Carpenter said. "It is like they are looking for the Loch Ness Monster. It does not exist, but they keep searching, and hoping, and lying and pointing to something shadowy that you kind of have to squint real hard at to pretend is there, but it isn't."
