House Oversight Committee Chairman James Comer said Monday that his committee has received information showing the Census Bureau's 2020 count was grossly wrong, and that Republicans paid for it with congressional seats they should have held. The miscalculation, Comer argues, wasn't random noise. It was patterned, and it broke one way.
The bureau inflated population figures in Democratic-leaning urban areas while undercounting suburban communities that have been among the fastest-growing Republican strongholds in the country. The result, according to Comer, was a congressional map built on bad numbers.
Comer put a number on it. Appearing on the "Just The News, No Noise" TV show, he said:
"The basic count was miscounted grossly in the last census, to the tune of costing the Republicans anywhere from four to five congressional seat."
That is not an abstraction. Four to five seats is the difference between a governing majority and a white-knuckle legislature where a single defection can sink any bill on any given day. Comer made exactly that connection.
"Now we've got a one-vote majority in the House of Representatives. If we had a five-vote majority, where we didn't have to worry that much about Thomas Massie, then we would have probably been able to have passed a lot more of President Trump's agenda."
Every Republican voter who watched the House majority strain to pass legislation over the past two years has felt this. Comer is now saying it had a structural cause.
The method matters here. Comer pointed directly to the pandemic as the opening for the errors:
"The fact that we lost [is] because the Census Bureau was biased. If you go back to when the census was taken, the last time it was during COVID, no one knocked on anybody's doors because of COVID. So they ended up estimating."
When you stop counting and start guessing, your priors drive the numbers. And the bureau's priors, Comer charges, ran left.
"They overestimated Democrats in the Democratic areas, which are the urban, rundown communities across America, and they undercounted all the suburban areas, which are the Republicans [fastest] growing areas."
This tracks with the bureau's own Post-Enumeration Survey data. The Census Bureau's Post-Enumeration Survey, which measures census count accuracy, revealed significant errors in 14 states, and those errors predominantly benefited Democratic-leaning states in the allocation of congressional seats and electoral college votes. No states were found to have had such errors in the 2010 census.
The directional skew was documented. The Post-Enumeration Survey suggests there were overcounts in New York, Massachusetts, Rhode Island, Hawaii, Delaware, Minnesota, Utah, and Ohio, and undercounts in Texas, Florida, Mississippi, Arkansas, Tennessee, and Illinois. Six of the eight states benefiting from overcounts tended to favor Democrats in awarding congressional apportionment and electoral college votes, while five of the six states penalized by undercounts tended to vote for Republicans.
That pattern does not prove intent. But it does make "random error" a very hard sell.
Comer said California, New York, and Rhode Island should have lost more House seats than they did, and that Florida and Texas should have gained at least one new member of Congress because of an exodus during the pandemic.
According to Just the News, Florida and Texas were the destinations. Millions of Americans fled high-tax, high-regulation blue states during COVID, reshaping the Sun Belt's political and economic footprint. Those residents moved. The census count didn't fully follow them. Because of the 2020 Census failure to produce accurate state-by-state counts, Colorado wrongly gained a seat, Rhode Island and Minnesota kept seats they should have lost, and Texas and Florida were not awarded seats they should have gained.
Rhode Island kept a seat it should have surrendered. Texas and Florida were shorted representation they had earned. Let that sink in.
Comer's committee has been pressing the bureau for answers since at least September 2024. The committee had already taken multiple steps to investigate what went wrong during the 2020 Census and the erroneous apportionment that followed, but the bureau was less than fully cooperative with the committee during President Biden's administration.
An institution that produced a politically skewed result and then stonewalled oversight. That is the pattern conservatives have come to recognize from the federal bureaucracy, and the Census Bureau apparently decided it was no exception.
The bureau's partnerships with certain organizations in 2020 also cast doubt on the reliability of the results. The bureau enlisted many partner organizations to facilitate aspects of the enumeration process, and statements made by leaders of some of those organizations raised serious questions about their ability to remain non-partisan and unbiased in their involvement with the census.
The damage from 2020 is locked in. This allocation of congressional seats and corresponding electoral college votes cannot be changed until the 2030 Census. The seats are gone for this decade. The question now is whether the same errors get made again.
Comer was direct about where his focus sits now:
"What we're focused on, I'm focusing on is making sure we get an accurate count, because that will be a significant improvement ... Just get an accurate count. We didn't have that the last time, and it was off by a lot."
In a letter to Acting Director of the Bureau, Comer highlighted the committee's ongoing need to conduct oversight and requested documents and information on measures the bureau is taking in its testing and preparation of the 2030 Census to avoid repeating these errors.
Comer also acknowledged that the separate debate over whether to count illegal immigrants in census apportionment is its own issue. "You know, the whole issue of whether you count illegals or not, that's a whole other issue," he said. His immediate focus is simpler, and harder to argue against: count the people who are actually there.
Comer has noted that the Census Bureau needs to be apolitical, and that politically biased unelected and unchecked federal workers at agencies are a serious concern; employees at the Census Bureau, who should be nonpartisan, should be prioritizing an accurate count.
A census that counts ghost populations in shrinking cities while missing the Americans pouring into growing suburbs is not a minor technical glitch. It is a structural distortion of representation, and it has consequences that last a decade. Comer is right to push hard before 2030 arrives. The next count has to be different. The cost of getting it wrong again is already on the record.
