Recalculate the quarter from the latest numbers
September’s employment report estimated a 29,000 increase in nonfarm payroll jobs, but the more useful comparison includes revisions to the two preceding months. The Bureau of Labor Statistics lowered July from a 21,000 increase to a 10,000 decline and August from a 162,000 increase to 133,000.
Using those updated estimates, July through September totals 152,000 added payroll jobs: minus 10,000, plus 133,000, plus 29,000. Dividing by three produces about 50,700 per month. That calculation describes the latest reported quarter; it is not a forecast or a count of newly hired individual people.
Do not subtract revisions from the wrong month
The combined downward revision for July and August is 60,000. It would be incorrect to subtract that amount from September’s 29,000 and label the result September job losses. The adjustment changes earlier months’ estimates. September’s own estimate remains positive in the October 2 release, which BLS characterizes as little changed.
The agency reports a 4.2 percent unemployment rate for September. The next Employment Situation release is scheduled for November 6. These dates matter because the current figures may be revised again; they are survey estimates, not a final census of every employer.
Jobs and people answer different questions
BLS’ methodology comparison explains that the payroll survey counts nonfarm wage-and-salary jobs. Someone with two covered payroll jobs can appear twice. The household survey instead counts employed people once and includes categories the payroll survey excludes, such as the unincorporated self-employed and agricultural workers.
The surveys also use different reference periods: generally a week containing the twelfth for households, versus an employer pay period containing the twelfth for payrolls. Their monthly changes need not match. For readers weighing claims about the economy, first identify whether the number describes jobs, people, or the share of the labor force unemployed. The three-month calculation here provides a consistent payroll comparison without claiming it captures every form of work.
Sources and further reading
BLS: September 2026 Employment Situation, released October 2 ↗
BLS: Comparing the household and payroll surveys ↗