New data from the US Census Bureau showing startling population losses in US cities has led some demographers to question the reliability of the figures after the Bureau acknowledged that it was missing substantial numbers of minority populations during its count ten-year.

Figures released last month showed that 62 of the 100 largest US counties lost population over a one-year period between July 2020 and July 2021.

At least ten of the country’s largest metropolitan areas lost more than one percent of their population during this period, which would constitute a historic exodus. Data shows that San Francisco lost 6% of its population, while more than one in 50 residents left New York, Boston and Washington, DC.

But to some who study demographics, the numbers suggest the possibility that Trump-era decisions to restrict numbers have hurt these major cities, all of which have large communities of color.

“It’s in the big cities, like San Francisco, Los Angeles, where the hard-to-count people live. So if you conduct the census in such a way that you severely, and I would say deliberately, undercount hard-to-count communities, you’ll get counties with low populations in those cities,” said Melissa Michelson, a demographer at Menlo. College in the Bay Area. “I think that’s part of what the Trump administration was doing, was undervaluing people of color, undervaluing immigrants, undervaluing mixed-status households.”

The Census Bureau said last month its 10-year count missed only a small number of Americans when it counted 331 million people living in the United States. But he was much more likely to underestimate black Americans, Latinos and American Indians, and Alaska Natives.

Data from a survey conducted by the Bureau after each count showed that the black population was undercounted by 3.3%, while the Bureau missed 5% of the Hispanic or Latino population. Both numbers were much higher than the undercount that occurred a decade ago.

“It was a bigger undercount than what we’ve seen in the past,” said Mark Hugo Lopez, director of race and ethnicity research at the Pew Research Center. “We are still waiting to get more data on this to determine its magnitude.”

On the other hand, the Bureau overcounted non-Hispanic whites by 1.6%, nearly double the last census’s overcount rate.

“The post-census survey results demonstrated that the Bureau replicated the problems of this decade that it observed over several decades through 2020, namely the persistent undercount of communities of color,” said Tom Wolf, deputy director of the Brennan Center. for Justice’s Democracy Program and a census expert. “We are now several decades into the same story.”

The undercount has significant ramifications for federal government programs that use census data to distribute billions of dollars in federal aid. Missing a few people – or in the case of some cities, tens of thousands – can cost those governments dearly for a decade to come.

Cities and counties that receive this money can dispute census results if they feel their population has been understated. Twenty-two local governments have challenged census counts so far this year, from Wichita, Kan., to Bennett, Iowa, Tumwater, Wash., to Jonesboro, Ga.

Detroit last week became the largest city in the nation to challenge the census count. In a letter, Mayor Mike Duggan (D) said an undercount would have “disastrous financial consequences for the city.”

“With every count, you will find that there are cities or counties or even states that will dispute the count,” Lopez said. “The amount these places receive from the federal government for various services is impacted.”

In an interview with The New York Times published Thursday, U.S. Census Bureau Director Robert Santos said his agency was exploring a number of ways to correct the count, including using administrative records generated by other agencies. government and even data collected by private companies.

“We’re exploring ways to identify commercial sector data, acquire it, and see how useful it is,” Santos said. “The only danger, of course, is that if you become too dependent on trade data, you may be one IPO or CEO change away from not having access to it.”

Wolf of the Brennan Center said the data produced by the Bureau remains unique in its breadth and scope, even though the Bureau missed counting more than 18 million people.

“When you have significantly high misses, you start to have additional data quality issues,” he said. “These questions of fitness for use must be answered in relation to the alternatives. The alternatives do not exist, so it is difficult to delve into this question.

Some demographers have looked for evidence of an urban exodus, either before the pandemic or following subsequent lockdowns and the rise of remote working. Most warn that the numbers released so far won’t yet paint a full picture of how Americans have responded to the pandemic, though population declines are an enticing data point.

“Perhaps we should all take a step back from these dramatic statements about the changes, and maybe ask ourselves if we’re analyzing fake data,” Michelson said. “Without an accurate count, it’s hard to know what’s really going on.”