The American jobs market has been running in a much lower gear than previously thought, according to a preliminary report released Tuesday.
The US economy added about 911,000 fewer jobs than initially estimated for the year ending in March, the Bureau of Labor Statistics report suggests. If this were to hold – the final annual benchmark revision will be reported in February 2026 – it would be the largest annual revision to US jobs data on record.
Recent employment data, including the August jobs report, show a US economy with job growth that’s stalling out.
“Overall, this shows that the jobs market is in an even more precarious position than we thought,” Sarah House, senior economist with Wells Fargo, told CNN in an interview. “This doesn’t tell us anything explicitly about hiring over the past few months, but it does suggest that we’re already on a much weaker setting heading into what’s been a rougher stretch since.”
The expected, but significantly large preliminary revision, comes as the BLS is in the throes of upheaval and the labor market is showing signs of cracking. President Donald Trump last month fired BLS commissioner Erika McEntarfer claiming, without evidence, that she “rigged” the weak July jobs report to include larger-than-typical downward revisions.
Trump administration officials excoriated the BLS after Tuesday’s large estimated revision, claiming that it was a symptom of bigger problems at the agency.
White House Press Secretary Karoline Leavitt said Tuesday’s preliminary revision shows that Trump “was right” in his claims that “Biden’s economy was a disaster and the BLS is broken.”
“This is exactly why we need new leadership to restore trust and confidence in the BLS’s data on behalf of the financial markets, businesses, policymakers, and families that rely on this data to make major decisions,” Leavitt said in a statement. “Much like the BLS has failed the American people, so has [Federal Reserve chair] Jerome ‘Too Late’ Powell – who has officially run out of excuses and must cut the rates now.”
BLS officials did not immediately respond to a request for comment.
BLS officials have, for more than a decade, sounded alarm bells about being too underfunded and too understaffed to implement the necessary practices to modernize data collection, analysis and reporting. In recent months, the agency has cited staffing challenges as the reason for reduced collections on critical inflation data.
A Labor Department official said the department is committed to modernizing the BLS’ operations and working around funding cuts, but the official said the BLS has been uncommunicative.
White House economic advisers have been conducting a review and assessment of BLS data quality and operations in preparation for a report that is expected to be sharply critical of the agency, according to three people familiar with the matter. The report has not yet been completed.
DOL official said the department is committed to modernizing the BLS’ operations and working around funding cuts, but the official said the BLS has been uncommunicative.
But Tuesday’s revisions release is the first step in BLS’ annual benchmark review of jobs data, a process that has taken place in some shape or form going back 90 years. And, economists say, the outsized revisions are in part a continued spillover from a global pandemic that threw the US economy on its head as well an indication of a labor market that’s in a time of transition.
Every year, the BLS seeks to provide a near-complete employment count by squaring past jobs data from business surveys (more timely but not as accurate) with comprehensive unemployment insurance quarterly tax filings (highly accurate but significantly lagged in timing).
The preliminary benchmark revision of -911,000 came in on the high end of economists’ estimates and accounts for about a 0.6% share of overall employment. The annual benchmark revisions during the past 10 years had an absolute average of 0.2% of total nonfarm employment, BLS data shows.
Economists said Tuesday that the massive revision was probably attributable in part to the pandemic and subsequent economic environment throwing out of whack the so-called birth-death model, a longstanding statistical tool that’s used to measure business and job creation.
Prior to Tuesday’s release, economists predicted that a large downward revision was likely due to three primary factors: weaker-than-inferred job creation at new firms; sampling errors resulting from declining survey response rates; and, to some extent, adjustments for asylum-seekers and other undocumented workers.
The overestimation of new firms versus those that have gone out of business is “part of the seasonal residuality problem that continues to bedevil data-based estimates that have not yet been properly adjusted following pandemic era shocks,” Joe Brusuelas, RSM chief economist, wrote in a note to investors on Tuesday. “This is quite like challenges in adjusting data following the years following the Great Financial Crisis. The BLS birth-death model has well know issues in capturing declines in business formation and bankruptcies especially around mid-cycle corrections and end of business cycle dynamics.”
One-quarter of the preliminary estimated revision was in the trade, transportation and utilities sector, which had an estimated downward revision of 226,000 jobs, or 0.8% of employment.
However, the information sector (which encompasses technology firms) looks to be revised down by the biggest share on a percentage basis: The preliminary estimates show a downward revision of 67,000 jobs, or 2.3%.
“It was pretty broad declines across industries like financial activities, professional and business services and things that aren’t really closely tied to undocumented workers,” Wells Fargo’s House said. “I think that suggests there are bigger things than just slower growth in the labor supply at play.”
Federal data is fluid and frequently subject to change, as more detailed and accurate information becomes readily available. The BLS’ monthly jobs report is meant to provide a higher-frequency look at employment trends, but that timeliness comes with a cost to accuracy.
To arrive at the monthly payroll estimates, the BLS surveys about 120,000 US employers, accounting for 600,000 work sites (roughly one-third of employment) and is based upon survey responses from employers across a wide swath of industries. Those respondents are given three opportunities to report their payroll gains and losses for any given month.
Every year, the BLS conducts a revision to the data from its monthly survey of businesses’ payrolls, then it benchmarks the March employment level to those measured by the Quarterly Census of Employment and Wages program.
The QCEW provides a more comprehensive read on the number of businesses, employees and wages at the state, regional and county level because it derives that data from quarterly tax reports submitted by businesses to their states. Given that process, the QCEW comes with a significant lag: The data for the first quarter of this year also was released Tuesday.
The annual revision estimates are not adjusted for seasonality; however, if spread out through the year ended in March, the revision would lower the average monthly job gains by nearly 76,000 positions between April 2024 and March 2025. As it stands now, job growth during that period was 146,500 per month, seasonally adjusted data show. (When the final benchmark is released, revisions are made to past seasonally adjusted and not seasonally adjusted data).
If finalized, this downward revision would bring that to about 70,500 per month, BLS data shows. The final benchmark revision and its monthly distribution won’t be known until February 2026. The largest annual revision on record came in 2009, with a downward revision of 902,000 jobs, or 0.7% of overall employment.
The final annual benchmarked seasonally adjusted data for the 12 months ending in March 2024 was a negative 589,000, or 0.4% of overall employment (-598,000 not adjusted for seasonality), which was the largest downward revision since March 2009’s negative revision of 902,000 and just above the negative 514,000 revision in March 2020 (for the year ended March 2019) that occurred during Trump’s first term.
The recent years’ massive increase in immigration helped to ease substantial labor shortages that occurred as the US economy quickly ramped up following the pandemic. However, that very dynamic wasn’t the easiest to measure and estimate.
Since the QCEW data is drawn from tax filings for unemployment insurance, it’s likely to exclude undocumented workers who are not eligible for benefits. Whereas the monthly jobs reports are drawn from surveys of businesses who relay back the number of people on their payroll.
“It can include people of every status,” Ron Hetrick, labor economist at Lightcast, told CNN in an interview last week. “If you believe an asylee is going to be granted, if you think they’re going to have a visa or they have a visa, and if they get that visa revoked (later on), you’re reporting them on your payroll (at the time).”
Tuesday’s release comes at a time when economic data is frequently weaponized and the BLS has found itself in the crosshairs.
Secretary of Labor Lori Chavez-DeRemer posted on X Tuesday that Tuesday’s preliminary revision “gives the American people even more reason to doubt the integrity of data being published” by the BLS.
“Leaders at the bureau failed to improve their practices during the Biden administration, utilizing outdated methods that rendered a once reliable system completely ineffective and calling into question the motivation behind their inaction,” Chavez-DeRemer claimed, adding that Trump and his administration “are putting a stop to years of neglect.”
Chavez-DeRemer said the Labor Department is committed to finding solutions to the large BLS revisions, including modernization that could improve jobs data transparency “and deliver more accurate and timely data for American businesses and workers.”
The latest comments inject further uncertainty into an agency that provides critical data that underpins the US economy.
McEntarfer’s unprecedented firing and Trump’s subsequent nomination of conservative Heritage Foundation economist E.J. Antoni stoked backlash from economists and statisticians from both sides of the aisle last month – particularly when comments surfaced that Antoni had previously suggested suspending the market-moving jobs report “until it is corrected.”
Federal economic data is considered the gold standard because of its longstanding reliability, quality, comprehensiveness, history and transparency.
However, that critical statistical infrastructure has long been at risk of crumbling. Economists, statisticians and policymakers have sounded alarm bells for years, stating that federal data is in a precarious state, because of decreased funding, survey response rates and public trust.
The BLS’ staffing is down about 20% since February and one-third of its leadership roles are unfilled, Erica Groshen, former BLS commissioner and co-chair of the Friends of the Bureau of Labor Statistics organization. Amid the shortages and an ongoing hiring freeze, the agency has scaled back on data collections, including those that feed into closely watched inflation reports.
Budget proposals from the Trump administration call for further funding and employment reductions.
“They are scrambling, reassigning people, retraining people, doubling up and cutting some granularity in some programs in order to cope,” Groshen told CNN. “They need to be restored to full staffing just to be able to produce (reports), and then they need extra funding for the modernization that’s too long neglected.”
This story has been updated with additional context and developments.
