U.S. private sector added only 38 thousand jobs in August — below forecasts

Edited by: Tatyana Hurynovich

The U.S. private sector added just 38 thousand jobs in August 2026. This is the weakest result since January and noticeably below analyst expectations.

According to the ADP National Employment Report, published on September 2, the figure came in below the consensus forecast of 47 thousand. The July result was revised upward to 46 thousand.

The largest contributions came from education and health care — plus 45 thousand positions. Construction added 12 thousand, and leisure and hospitality 16 thousand. Financial activities gained 6 thousand.

At the same time, manufacturing cut 17 thousand jobs. Professional and business services lost 16 thousand, and trade, transportation, and utilities 5 thousand. The information sector and natural resources with mining also showed declines.

Large companies with more than 500 employees accounted for nearly all of the gain — 34 thousand positions. Small firms with fewer than 50 workers added only 3 thousand, while medium-sized ones remained flat.

ADP chief economist Nela Richardson noted that "paychecks tell a lot about the current hiring instability." In her words, one can understand the dynamics only by deeply analyzing where pay growth is accelerating or slowing and for whom exactly.

The ADP report traditionally serves as an important guide ahead of the official statistics from the U.S. Department of Labor, which will be released in the coming days. How will this picture affect the overall assessment of the labor market?

ADP data are based on anonymized payroll information from more than 26 million private-sector employees in collaboration with the Stanford Digital Economy Lab. They show a slowdown in hiring amid mixed industry trends.

35 Views

Sources

  • Private-sector hiring slowed in August

Comments

Read more articles on this topic:

Korea Aerospace Industries says its AI module on the 12U BEE-1012 CubeSat, launched on Nuri Oct 7, passed an orbital pilot Oct 20. Given a simulated fault, it diagnosed cause and scope and sent back a fix plan. A main demo starts in December. Image Credit: Yonsei University

Image
Reply

Great paper from Anthropic on AI agents doing ML-for-science research. (bookmark it) If you build AI-scientist systems, this one is worth your time. They propose Agent4Sci, which gives frontier coding agents 20 scientific ML benchmarks across 11 domains and asks them to beat

Image
Reply
Did you find an error or inaccuracy?We will consider your comments as soon as possible.