Did the U.S. government overestimate job growth by more than 1 million jobs this spring? That’s the claim after the Philadelphia Federal Reserve Bank released a December report, which makes early revisions to Bureau of Labor Statistics data for the second quarter of 2022.

“In the aggregate, 10,500 net new jobs were added during the period rather than the 1,121,500 jobs estimated by the sum of the states; the U.S. CES estimated net growth of 1,047,000 jobs for the period,” the report read.

To do the math, that puts the job growth data off by about 1 to 1.1 million jobs from March to the end of June.

That paragraph caught the eye of at least one person in Congress. Sen. Rick Scott (R-FL) called the finding “outrageous” in a tweet and accused the Biden administration of lying to the American people about the economy.

WRONG BY A MILLION JOBS. THIS IS OUTRAGEOUS.@JoeBiden's admin has been lying to the American people about our economy to prop up his failed agenda & I won't stand for it. I'm requesting an immediate meeting with the head of @BLS_gov. WE NEED ANSWERS NOW! pic.twitter.com/OKwsiYSoER

— Rick Scott (@SenRickScott) December 16, 2022

Because jobs data informs how economic and policy leaders run the economy, the potential discrepancy could have tremendous consequences.

“[There were] 2.7 million people hired in the first half of the year, it doesn’t make sense that the economy would be in a recession with this kind of thing happening,” Federal Reserve Chair Jerome Powell said during a press conference in July.

The supposed strong labor market has fueled the Fed’s path to higher interest rates to fight inflation, which has tightened money supply to levels not seen since 2007. But if the job market is not what it seems, then the Fed is crunching a much more vulnerable economy, which already saw negative economic growth the first two quarters of the year.

Revisions are normal in any economic data, but not by 1 million. I presented the Philadelphia Fed revision to the BLS, which largely refuted the findings.

“Using a simple methodology and a series of assumptions, the Philadelphia Federal Reserve Bank benchmarked to a less complete version of the [Quarterly Census of Employment and Wages] data than is available to BLS,” BLS economist Emma Sillman responded in an email. “BLS does not believe that the monthly [Current Employment Statistics] data dramatically overestimated employment growth.”

Sillman said the preliminary revisions run by the department are well within normal ranges. The official national BLS revision will be published Feb. 3, 2023, with state-level data out March 13, 2023.

In bringing the BLS response back to the Philadelphia Fed, the bank pointed me to its article, stating, “By making timelier revisions, we produce more accurate estimates of states’ job growth.”

The bank revision comes roughly five months after the quarter and updates on a quarterly basis, rather than waiting on the BLS to update the data once per year.

So where do we go from here? Did the U.S. economy add 10,500 jobs or more than 1 million jobs in the second quarter? Officially, that answer won’t be reached until February, when the BLS posts its annual revision. Even though the Philadelphia Fed and the BLS are a world apart in current estimates, the Philadelphia Fed acknowledged the accuracy of final BLS revisions.

“Once the BLS completes its more comprehensive annual revisions — with a more sophisticated methodology and better data access than we can deploy — we accept their benchmarked data as more accurate than our fourth and final Early Benchmark,” the Philadelphia Fed said.