BLS finds March jobs estimate was 79,000 too high
The Bureau of Labor Statistics said Friday, August 28, that its preliminary annual benchmark revision put March 2026 total nonfarm employment 79,000 jobs below the earlier estimate. The adjustment equals 0.1% of the total. The preliminary revision for total private employment was larger: 178,000 jobs, also 0.1%.
Those figures do not mean 79,000 jobs suddenly disappeared in March. They show that a more comprehensive employment count came in below the earlier estimate produced from the government’s monthly payroll survey. The figures remain preliminary, and BLS has not changed the currently published official payroll tables.
What the preliminary revision changed
BLS’s Current Employment Statistics, or CES, program produces the monthly payroll report from a survey of employers. Each year, the agency benchmarks those estimates against more comprehensive employment counts from the Quarterly Census of Employment and Wages, or QCEW.
QCEW counts are derived primarily from state unemployment-insurance tax records that nearly all employers are required to file with state workforce agencies. BLS also uses other sources for employment not covered by QCEW, including Census Bureau data, state employment information and Railroad Retirement Board records.
For March 2026, the preliminary comparison produced a downward adjustment of 79,000 jobs for total nonfarm employment. The private-sector estimate moved down by 178,000. Government employment was revised 99,000 higher, which partly offset the larger decline in private employment and left the total revision smaller.
BLS describes the preliminary benchmark as the difference between two independently derived employment counts, each with its own potential sources of error. The result is therefore a preview of the likely benchmark adjustment, not a final employment count.
Why some industries moved more than others
The industry results were uneven. BLS revised employment lower in manufacturing, wholesale trade, retail trade, professional and business services, private education and health services, leisure and hospitality, other services, and mining and logging.
Employment was revised higher in construction, transportation and warehousing, information, financial activities, utilities and government. The largest listed declines included retail trade, down 154,600 jobs, and wholesale trade, down 86,200. Transportation and warehousing was revised 135,100 higher, while government was revised 99,000 higher.
That industry breakdown should not be read as a real-time tally of layoffs. It is a measurement adjustment to the estimated employment level through March, based on broader employer records that became available after the monthly survey estimates were produced.
What it says about job growth
The preliminary result suggests that employment growth through March 2026 was somewhat weaker than the monthly payroll reports indicated. Reuters reported that the private-employment adjustment implies average private-sector job growth of about 24,000 a month over the year through March, compared with 38,000 a month under the previously published data.
Even so, the national adjustment is relatively modest by recent benchmark standards. BLS said annual benchmark revisions over the last 10 years had an absolute average of 0.2% of total nonfarm employment. The March 2026 preliminary revision was 0.1%.
The geographic numbers show why local figures should be interpreted cautiously. Across the 50 states and the District of Columbia, preliminary revisions ranged from a 1.2% decline in North Dakota to a 1.0% increase in Iowa and the District of Columbia. Among the 56 metropolitan areas with populations of at least 1 million, the range was from a 1.9% decline in Tucson, Arizona, to a 4.3% increase in Grand Rapids-Wyoming-Kentwood, Michigan.
July data are separate
BLS’s July Employment Situation report remains the latest regular monthly payroll release. It showed total nonfarm payroll employment changed little in July, with a decline of 23,000, while the unemployment rate was 4.1%. BLS also revised May and June payroll changes lower by a combined 103,000 jobs.
Those monthly revisions are separate from the August 28 preliminary benchmark. The benchmark release did not update the currently published CES estimates, and it does not create a new monthly jobs figure for July.
When the official series will change
BLS said the final national benchmark revision is scheduled to be incorporated with the January 2027 Employment Situation release, which is expected in February 2027. Until then, the official national payroll series remains as previously published.
Final benchmark revisions for state and metropolitan-area series are scheduled for March 2027 with the January 2027 State Employment and Unemployment release.
For workers, employers, policymakers and anyone tracking the economy, the practical takeaway is straightforward: monthly payroll figures are the best early employment indicators available, but they can be revised when more complete employer records arrive. The August 28 release points to slightly weaker employment growth through March, not to a newly measured wave of job losses today.
Sources
Look for updates to this story
Discover more from Interactive News
Subscribe to get the latest posts sent to your email.