SPY•| Year | Preliminary benchmark revision (NSA) | Final benchmark revision (NSA) | Final benchmark revision (SA) |
|---|---|---|---|
| 2026 | -79 | ||
| 2025 | -911 | -862 | -898 |
| 2024 | -818 | -598 | -589 |
| 2023 | -306 | -187 | -266 |
| 2022 | 462 | 506 | 568 |
| 2021 | -166 | -7 | 374 |
| 2020 | -173 | -121 | -250 |
| 2019 | -501 | -505 | -514 |
| 2018 | 43 | -16 | -1 |
Source: Bureau of Labor Statistics
Note: Data in thousands; NSA = nonseasonally adjusted; SA = seasonally adjusted
By industry, the largest revisions were a reduction of 154,600 jobs in estimated retail employment and an increase of 135,100 jobs in estimated transportation and warehousing employment. Nine sectors saw downward revisions and six saw upward revisions, including an estimated increase of 99,000 jobs in government employment during a period when the Trump administration was actively cutting the federal workforce.
The U.S. job creation rate has been decelerating over the last two years, in part because of slower demand for labor from businesses uncertain about the economic outlook and whether the artificial intelligence boom would allow them to replace workers with tech tools. Also, there has been a reduction in the pool of available workers because of retirements and President Donald Trump's aggressive immigration crackdowns.
By contrast, about a year ago BLS had initially estimated a downward revision to employment of 911,000 jobs, or 0.6%, in the 12 months through March 2025, a figure that was later revised to a reduction of 862,000, or 0.5%. On a seasonally adjusted basis — only provided with the final benchmark adjustment — employment over those 12 months was revised down by 898,000, or 0.6%.
The benchmark revisions have been abnormally large since the pandemic due to measurement model challenges and low survey response rates.
Friday's revision estimate implied that nonseasonally adjusted nonfarm payroll gains averaged about 11,000 per month through March over the preceding 12 months instead of 18,000. On a seasonally adjusted basis, job gains by current estimates averaged about 23,000 over that span.
The revision to estimated U.S. private employment was somewhat larger at negative 178,000, or 0.1%. That data implies a larger reduction to the pace of private-sector job growth to an average of 24,000 a month from 38,000 based on previously published data.
U.S. employment grew slightly less than previously estimated in the 12 months through March, the government said on Friday, suggesting the modest pace of job creation that has prevailed since President Donald Trump returned to the White House was even weaker through at least the first months of 2026.
The preliminary annual benchmark revision estimate to the closely watched payrolls data from the Labor Department's Bureau of Labor Statistics (BLS) was released weeks after it reported a surprise drop in employment in July, but was nothing like the massive revisions that have become the norm since the COVID-19 pandemic. BLS will issue a final benchmark revision for the period in February 2027 with the publication of the employment report for January.
On a nonseasonally adjusted basis, the level of total U.S. employment as of March is now estimated to be 79,000 jobs lower than previously estimated, a reduction of about 0.1%, BLS said.
