The August employment report landed Friday morning at 162,000 jobs, against a Dow Jones consensus of roughly 53,000. That is not a beat. That is a different economy from the one forecasters had modelled three weeks ago, and it arrived after a July print that showed the country losing 23,000 jobs.
The unemployment rate held at 4.1%, exactly as expected. Average hourly earnings rose 10 cents, or 0.3%, to $37.75, up 3.1% over the year. And the revisions cut the same way as the headline: June was revised up 11,000 and July up 44,000, putting the two months a combined 55,000 higher than previously reported. August was the strongest month since March.
The bond market did the arithmetic immediately. Treasuries sold off across maturities, led by an eight-basis-point climb in the two-year — the tenor that trades the Fed and nothing else. Traders moved September hike odds to roughly 60% on CME data, with some venues nearer 53%. Equities, which had rallied hard on Thursday as Governor Christopher Waller urged patience, gave ground on Friday morning.
Our take: The story here is the revisions, not the headline. A single hot print inside a noisy series is a data point; a hot print plus 55,000 added to the two months behind it is a re-levelling of the entire trend, and it retroactively weakens the case that the labour market was cracking in midsummer. That is what moved the two-year eight basis points. It also means the July collapse everyone traded off was, in part, a measurement artefact — a reminder that the first estimate of a jobs number is a draft, and the Fed knows it even when the market forgets.
The setup into September 15–16
Thursday’s close was a broad rally — the Dow up 1.18% to 53,686.11, the S&P 500 up 1.06% to 7,747.71, the Nasdaq up 1.40% to 26,584.06 — built substantially on the idea that the Fed would sit still. Friday’s data does not settle the question so much as reopen it. Inflation is not where the committee wants it, the labour market has just refused to roll over on schedule, and the meeting is eleven days out.
Note the internal tension in the week’s own numbers: ADP put private payrolls at 38,000 for August, with the largest employers carrying nearly all of it. The official private figure came in far above that. When the two series diverge this widely, the honest read is that hiring is concentrated rather than broad, and concentration is exactly the condition that produces volatile monthly prints in both directions.
What to watch
- The two-year, not the ten. If the front end holds its move into next week, the market has genuinely repriced the September meeting rather than reflexed on a headline.
- Household survey detail. A steady 4.1% unemployment rate alongside a 162,000 payroll gain implies participation absorbed the strength. Whether that continues determines how much wage pressure follows.
- Whether the ADP gap closes. Two series, one month, wildly different answers. The revision cycle will pick a side.
- Fed speakers before the blackout. Waller argued for patience on Thursday. Friday’s number is the first test of whether that argument survives contact with data.