WASHINGTON, D.C.: The U.S. labor market, long seen as a pillar of economic resilience, turns out to have been far weaker than initially reported over the past year, according to new government data that is already rattling investors and raising pressure on the Federal Reserve.

The Labor Department said that employers added 911,000 fewer jobs in the 12 months ending in March 2025 than earlier estimated. The preliminary revision, released annually, suggests hiring momentum was far softer heading into this year's trade-driven slowdown. Final figures will be published in February 2026.

The downward revision affected a wide range of industries. Leisure and hospitality added 176,000 fewer jobs than first reported, professional and business services were 158,000 lower, and retail employment was trimmed by 126,000.

The news followed last week's dismal August jobs report, showing just 22,000 positions added. It is the latest sign that President Donald Trump's sweeping and unpredictable import tariffs may be discouraging companies from expanding payrolls.

"This paints a much weaker portrait of the job market than initially thought," said Sal Guatieri, senior economist at BMO Capital Markets. He warned that the data suggest the economy entered the current trade war with less momentum than believed and has slowed even further, with job growth since March averaging only 53,000 a month.

Trump has repeatedly clashed with Labor Department statisticians. As a candidate in 2024, he accused officials of "rigging" revisions that showed 818,000 fewer jobs, although the final figures turned out less severe. In July, angered by another round of downward revisions, he fired the top government economist overseeing the payroll data.

The latest weak data is expected to heighten calls for the Fed to cut interest rates at its September 16 and 17 meeting. ING economist James Knightley predicted further cuts in October and December, arguing that early downturns typically lead the Labor Department to overstate hiring by new firms while undercounting business closures.

"These revisions suggest that job momentum is being lost from an even weaker position than originally thought," Knightley wrote. "It also reinforces the belief that even the poor numbers seen in 2025 are probably overstating the health of the employment market."

Economists say part of the problem is declining survey responses from employers. The monthly payroll figures come from 121,000 employers representing 631,000 worksites, but revisions rely on more complete quarterly job rolls submitted to state unemployment tax offices. Those filings showed fewer workers on payrolls than federal surveys suggested, especially from non-responding firms.

The government acknowledged that the discrepancy highlights limits in its data collection at a time when accurate measures of economic strength are under intense political and financial scrutiny.