New government data released Tuesday reveals that the U.S. economy employed 911,000 fewer people than previously reported as of March 2025, signaling that the labor market had been slowing long before the summer months. The revisions cover the period from April 2024 through March 2025, reducing the average monthly job gains during that year from roughly 147,000 to about 71,000. Earlier data had suggested the economy added around 1.76 million jobs over that span; the new figures are less than half that, marking one of the largest downward adjustments in recent memory.
The revision has immediately fueled criticism of the government’s data collection methods. The White House and allies of former President Trump highlighted the weaker baseline, arguing that the president inherited a more fragile economy than previously understood. Vice President JD Vance commented, “It’s difficult to overstate how useless BLS data had become.” The new data also raised political pressure on Federal Reserve Chair Jerome Powell. White House press secretary Karoline Leavitt stated that Powell “has officially run out of excuses and must cut rates now.”
Economists echoed concerns about the labor market and monetary policy. PNC Financial Services Group chief economist Gus Faucher told Yahoo Finance that the revisions show job growth was “not as strong as the Fed thought,” suggesting interest rates are too high and may need to be reduced to support employment in the near term.
The largest downward revisions were concentrated in leisure and hospitality, which lost 176,000 jobs compared with previous estimates, followed by professional and business services with 158,000 fewer positions. Almost all of the revisions occurred in the private sector, which saw 880,000 fewer jobs, while government employment was revised down by 31,000. The magnitude of the revisions exceeded many economists’ expectations, who had forecast adjustments of several hundred thousand jobs. A final, definitive revision for this period is scheduled for February 2026.
These benchmark revisions reflect a routine process in which the Bureau of Labor Statistics (BLS) updates employment estimates based on more concrete data sources such as quarterly state unemployment insurance filings, rather than relying solely on surveys. In recent years, revisions have frequently been larger than average due to declining survey response rates, and they have become a flash point in political debates. For instance, last year’s preliminary annual revision—released during the presidential campaign—showed 818,000 fewer jobs than initially reported, drawing intense scrutiny.
The political context has further intensified after President Trump publicly accused the BLS of producing “phony” numbers and fired the agency’s commissioner. Trump’s pick for a new BLS commissioner, E.J. Antoni of the Heritage Foundation, is expected to face a confirmation hearing before the Senate’s labor committee in the coming months. Antoni has been a polarizing figure, with Trump and aides citing the large revisions in recent years as justification for rethinking survey-based data collection methods. Tuesday’s release is likely to strengthen those calls for change.
Beyond technical implications, the revised data is being leveraged politically. The White House has used the numbers to highlight challenges inherited from the previous administration, while Trump and his allies have criticized the Fed’s monetary policy, arguing that Powell should have acted sooner to cut rates and that the 2% inflation target is overly rigid. On the morning of the release, Trump cited a market analyst’s view that rate cuts should have started in 2021, framing the current economic slowdown as a consequence of delayed action.
The downward revisions underscore broader questions about the accuracy and reliability of U.S. labor market data. They also reinforce concerns that job growth has been slowing for longer than initially believed, suggesting that policymakers may need to consider additional support for the labor market in the months ahead.