The U.S. economy added 29,000 payroll jobs in September, and unemployment edged up to 4.2%. That headline offers a restrained picture of a labor market that is still adding jobs, but at a slower pace than many workers and businesses would prefer. The revisions matter just as much: July payrolls were revised from a gain of 21,000 to a loss of 10,000, and August’s reported gain was cut by 29,000. Together, the two months were 60,000 weaker than previously estimated.

This is not evidence that the country is already in a broad jobs downturn. It is evidence that hiring has been modest and that early monthly estimates can overstate the underlying pace. For workers, the distinction matters: finding a new position may take longer even when aggregate employment is not falling. For employers, slower hiring can signal caution, but the report alone cannot tell whether firms are postponing recruitment or responding to weaker demand.

A small gain in a noisy series

The payroll survey counts jobs reported by employers; the separate household survey estimates how many people are employed or unemployed. In September, payroll employment increased by 29,000, while the unemployment rate was 4.2% and the number unemployed was 7.1 million. The rate has remained between 4.1% and 4.3% since March, according to the Bureau of Labor Statistics (BLS).

The BLS said payrolls averaged 45,000 monthly gains over the previous 12 months. September’s result is below that pace. But monthly changes are estimates, not a precise census, and the report’s own technical note says an increase or decrease of roughly 122,000 jobs is needed for a month-to-month payroll change to be statistically significant. That does not make a 29,000 gain meaningless; it means a single month is weak evidence for declaring a turning point.

Revisions are a regular feature of the survey. Employers provide additional responses after the first estimate, and seasonal factors are recalculated. July’s downward revision is especially consequential because it changes the month from job growth to a small decline. Reading the series as a whole therefore produces a softer picture than the first estimates had suggested.

Hiring is uneven beneath the total

Most major industries changed little in September. Health care added 17,000 jobs, less than its 33,000 average monthly gain over the prior year. Construction added 11,000, close to its recent average. Manufacturing increased by 9,000 and was up 72,000 from its December 2025 low. Financial activities lost 7,000 jobs and remained 129,000 below its May 2025 peak.

These sector details help explain why one national number cannot describe every job seeker’s prospects. A health-care employer’s hiring needs are not the same as a bank’s or a factory’s. The report also showed employment in information and professional and business services declining modestly, while transportation, retail, and leisure and hospitality posted small gains. Such movements can be revised too, so they are useful as clues rather than definitive evidence of a durable industry trend.

The household survey adds another dimension. The number of people unemployed for at least 27 weeks was 1.9 million, or 27.1% of the unemployed. About 4.5 million people were working part time for economic reasons, meaning they would have preferred full-time work but had reduced hours or could not find a full-time job. Those figures suggest the pressure of a slower market may show up in longer job searches or hours, not only in outright layoffs.

What workers and employers can infer

For workers, the practical implication is to avoid treating a stable unemployment rate as proof that job changes remain easy. If a search is taking longer, that experience is consistent with slower hiring, though this report cannot establish the reason in any one occupation or region. People weighing a voluntary job change may want to consider their own field’s demand and the time required to secure another role rather than rely on the national payroll headline.

For businesses, slower employment growth can reflect restrained hiring plans, labor supply constraints, or fewer available candidates; the report does not isolate those causes. Employers should distinguish a genuine drop in demand from a temporary pause in recruitment before changing staffing plans. The mix of sector results also argues against assuming every business faces the same labor conditions.

Pay rose, but not sharply: average hourly earnings increased 0.1% in September and 3.0% over 12 months. The average workweek held at 34.4 hours. That combination is consistent with ongoing income growth, but the aggregate does not reveal how pay is changing for individual workers or whether wage increases are keeping pace with each household’s expenses.

What to watch next

The next employment report will show whether September’s weak payroll gain and the downward revisions persist. Several months of data will be more informative than one result, especially if job growth, hours, unemployment, and revisions move in the same direction. For now, the best reading is neither “the labor market is collapsing” nor “nothing has changed.” Hiring has slowed, the recent record has been revised down, and the evidence is not yet enough to say that employment is on a sustained downward path.