For a Federal Reserve operating with a data-driven approach to monetary policy, what happens when the data is wrong?

Chairman Jerome Powell has been receiving rising pressure from President Donald Trump — and more recently, other Fed officials — to cut rates, but there's another issue that Powell's worried about: the quality of economic data collected by the Bureau of Labor Statistics.

Economists have been raising concerns about this topic for the last few months, and Powell voiced his own concerns on Tuesday during his testimony to Congress.

"I wouldn't say that I'm concerned about the data today, although there has been a very mild degradation of the scope of the surveys," Powell said when Rep. Sam Liccardo asked him about his thoughts on data quality. "But I would say the direction of travel is something I'm concerned about."

"It's really important not just for the Fed, but for Congress and for businesses, frankly, to know what really is going on in the economy," Powell continued. "I don't like to see the kind of stories I'm reading and the idea being that the data is going to become more volatile and less reliable. That'll make it more difficult for the private sector and for you and for us.

That means inflation, employment, and other economic measures that the Fed and other institutions depend on to determine policy might be less accurate than they were in the past.

DOGE cuts on government funding could be a culprit. BLS was not spared from budget cuts earlier this year, and a proposal in Trump's Big Beautiful Bill is aiming to further reduce the agency's budget by $56 million.

Staffing shortages at the BLS have resulted in reduced data availability. BLS teams calculate CPI numbers by collecting price quotes across 75 urban areas in 200 item categories. On June 16, BLS announced it had suspended data collection in Buffalo, NY. This follows the agency's April suspension of CPI data collection in Lincoln, Nebraska, and Provo, Utah.

BLS said the number of imputations, or estimated values, in CPI data increased in April due to these changes, but affirmed that these exclusions have an overall "minimal impact" on inflation data.

Torsten Sløk, Apollo's chief economist, pointed out that imputations have increased significantly in the last few months, reducing data quality. Usually, around 10% of the CPI values are imputed when data is not available. However, the May percentage is estimated to be triple the average, at 30%.

"In other words, almost a third of the prices going into the CPI at the moment are guesses based on other data collections in the CPI," Sløk wrote in a note last week.

This could be leading to more frequent revisions of economic data, experts say, and the labor market is another area of scrutiny.

The May jobs report showed 139,000 new jobs created, but both Peter Berezin, chief global strategist at BCA Research, and Samuel Tombs, chief US economist at Pantheon Macroeconomics, believe the final number could be revised down to around 100,000.

Increased imputations can mask new developments in the economy, making inflation and jobs data look more optimistic than they actually are.

In Tombs' opinion, the jobs numbers are excluding a large swath of the economy: small businesses, which are filing late as they struggle with tariff impacts.

"When there's a gap in the data, [the BLS] just interpolates from past trends, but if the trend itself is weakening, once you actually get the data that you previously had interpolated, usually you find that it's weaker than your original estimate had suggested," Berezin told Business Insider.

"The response rate to these surveys is very low, so the Fed has to do a lot of guesswork. And in a weakening economy, usually you're going to be guessing too high on payrolls, rather than too low" Berezin added.