US Jobs Report, August 2026 — Payrolls Up 162,000 as July's Decline Is Revised Away

The August report, released on 4 September 2026, showed nonfarm payrolls up 162,000 with the unemployment rate unchanged at 4.1 percent. That was far above the pre-release consensus, which clustered around 53,000 to 58,000.^[1]

The more consequential change is to July. The same release revised July from −23,000 to +21,000 and June from +20,000 to +31,000, leaving the two months 55,000 higher than previously reported. In other words, the outright job loss that defined the July report no longer appears in the data; it now reads as a modest gain.^[1]

This is the revision mechanism described below acting on this very page. The figure quoted for July throughout the rest of this document was the first print, and it moved by 44,000 — a reminder that a single month's first estimate is a provisional reading rather than a settled fact.^[1]

Gains were concentrated in a few areas: food services and drinking places added 59,000 and local government education added 42,000, the latter largely offsetting a drop the month before. The information sector lost 23,000 jobs.^[1]

Wages cooled further. Average hourly earnings rose 10 cents, or 0.3 percent, to $37.75, and were up 3.1 percent over the year — below the 3.2 percent pace recorded in July. The labour force participation rate edged up to 61.6 percent, though it remains 0.5 percentage point lower than in January.^[1]

That participation detail matters for reading the unemployment rate. Forecasters had expected the rate to hold at 4.1 percent because the labour force was shrinking. It did hold — but with participation ticking up rather than down, so the flat rate this month is not the shrinking-denominator story that had been anticipated.^[1]

July's report was the clearest sign yet of a stalling job market: payrolls contracted by 23,000 against expectations of a modest gain, dragged down by a 53,000 drop in government employment and softness in retail and leisure. Healthcare, long the reliable engine of job growth, added jobs at a slower-than-usual pace.^[1][2]

Revisions compounded the weakness — May and June were marked down by a combined 103,000 jobs. Paired with 3.2% wage growth, the lowest since May 2021, the report strengthened market expectations that the Fed would prioritize the employment side of its mandate heading into the Jackson Hole symposium.^[1][2][3]

The most common confusion in this release is that payrolls and the unemployment rate come from two different surveys with different denominators. Payrolls count jobs reported by employers. The unemployment rate comes from a household survey and is the share of the labour force — people working or actively looking — who are without work. In July 2026 payrolls fell by 23,000 while the unemployment rate edged down to 4.1 percent, which reads as a contradiction only if both are assumed to measure the same thing.^[1][2]

A rate can fall because more people found work, or because people stopped looking and left the labour force, which shrinks the denominator. Those two causes point in opposite directions for the economy but produce the same movement in the headline rate. This is why the participation rate is read alongside it rather than after it.^[1][2]

Each month's payroll figure is revised twice, in the two releases that follow, as more employer responses arrive. A month that reads as a small gain on first publication can become a decline two months later, and the reverse also happens. When revisions are large, the direction of the trend can change even though no single month's headline was wrong at the time it was published.^[1][2]

The next report, covering August 2026, is scheduled for Friday, September 4, 2026 at 8:30 a.m. Eastern Time. It will carry revisions to July alongside the new month, so the July figure quoted here is subject to change on that date.^[1][4]

Before the August 2026 report was published, searching for it returned the August 2025 release instead. That trap has not fully closed: the two years are still easy to confuse because the framing is similar. The 2025 report showed payrolls of +22,000 with unemployment rising from 4.2 to 4.3 percent. Those are not 2026 figures.^[4]

Two checks separate them quickly. The unemployment path has to be continuous: July 2026 was 4.1 percent, so a report describing a move from 4.2 to 4.3 percent belongs to a different year. And monthly revisions run in the tens of thousands, so a figure implying a swing of about a hundred thousand for the same month is a sign that two different years have been mixed.^[4]

The August 2026 report was due at 8:30 a.m. Eastern on Friday, 4 September 2026. Going in, economists expected payrolls to rise by about 58,000 with the unemployment rate holding at 4.1%, though the consensus figure varied by survey — some outlets cited a number closer to 53,000. The actual figure was 162,000, roughly three times the consensus, which is a useful reminder of how wide the error band on these forecasts can be.^[1][5]

One leading indicator was already out. The ADP National Employment Report showed private payrolls up 38,000 in August, below both July's 46,000 and the 47,000 economists expected. Forecasts spanned a wide range as a result: Fifth Third Commercial Bank projected a 25,000 decline, citing the effect of TPS cancellation on work authorisation.^[5]

The unemployment rate is the number most likely to mislead. Forecasters expected it to hold at 4.1% because the labour force is shrinking, not because hiring is strong — the labour force fell by about 1.3 million workers in the twelve months through July. A flat rate produced by a shrinking denominator is not the same signal as a flat rate produced by steady hiring.^[5]

The year mix-up described above is not only about the numbers. The August 2025 coverage was written around an expected rate cut. The 2026 debate runs the other way: commentary ahead of this release framed unexpectedly strong hiring as something that could push the Fed toward considering rate hikes if inflation failed to cooperate, with the central expectation being that policy stays on hold.^[5]

So a summary that pairs a jobs figure with an imminent rate cut is describing 2025, not 2026. Checking the policy framing is a faster year-check than comparing the payroll numbers themselves.^[5]