The S&P 500 closed Friday at 7,757.64, up 0.62% and its highest close on record. The Nasdaq Composite rose 1.3% to 26,690.62. The Dow Jones Industrial Average added 151.83 points, or 0.28%, to 54,036.93. Across the five sessions the S&P gained about 3.6% and the Nasdaq roughly 5% — the biggest weekly advance for the major U.S. indexes since April.
The catalyst was bad news. Nonfarm payrolls fell 23,000 in July against a consensus near +83,000, and May and June were revised down by a combined 103,000. The labour market did not soften; it contracted. Stocks rose anyway, because in this cycle a weak payroll print is not read as a warning about earnings. It is read as an argument against a rate rise.
That is the entire trade. Interest rate futures now put the odds of a September hike at 43.9%, down from 57% before the report, with the probability of a hold climbing to 60.4% from 43.2%. On the prediction market Kalshi, traders see roughly a 65% chance the Fed stands pat. Ten days ago, in the hours after the July meeting, CME FedWatch had September hike odds at 59%. One release moved the market’s base case from tightening to waiting.
The Fed being priced has three dissenters
On July 29 the FOMC left the federal funds target at 3.50%–3.75%, where it has sat since December. The vote was not unanimous. Cleveland’s Beth Hammack, Minneapolis’s Neel Kashkari and Dallas’s Lorie Logan each preferred a quarter-point increase; nine officials, Chair Kevin Warsh among them, voted to hold. “There is no soft inflation target,” Warsh said afterwards. “There’s only a target, and it’s 2%.”
Warsh has also deliberately narrowed the Fed’s forward guidance, giving markets fewer signals about where policy is heading — and has said that reticence was itself a factor in bond yields rising as investors were left to weigh the data themselves. So equities are now pricing a specific September outcome out of a committee that has stopped telling anyone what it intends to do, and in which a quarter of the voters wanted to hike two weeks ago on numbers that did not yet include a payroll contraction.
Our take: This is not a rally in earnings. It is a rally in the discount rate — and a discount-rate rally reprices in both directions with equal speed. The index is at an all-time high on the strength of a single labour print, read through a Fed that has withdrawn its own guidance. That is a fine position to be in while the inflation data cooperates. Wednesday is when we find out whether it does.
Wednesday is the test
July CPI lands at 8:30 a.m. ET on Wednesday, August 12, with producer prices the following day. It is the last major inflation reading before the September meeting, and the one number that can reverse Friday’s arithmetic outright. A hot print does not merely trim the odds of a hold — it restores the exact case the three dissenters were making, and it does so against an index that has already spent the rally.
There is a second fragility underneath the first. The S&P 500’s blended Q2 earnings growth rate of 50.4% is heavily flattered by unrealised gains on private AI stakes at two companies. Strip those out and growth is 32%. The market is leaning on a rate call for its multiple and on non-operating gains for part of its earnings.
What to watch
- Core CPI, Wednesday 8:30 a.m. ET. The hold-versus-hike split inside the FOMC turns on whether price growth is still easing. This is the number that settles it.
- Whether hold odds stay above 60%. Futures moved 13 points on one report; they can move back on the next one.
- The size of the next revisions. Another month of large back-revisions makes July’s contraction much harder to dismiss as noise — and turns a rate story into a growth story.
- Whether the rates trade stays broad. Gold had its best week since January on the same logic. If it and equities start disagreeing, one of them is wrong about September.
The week was a good one. It was also a single bet, placed on a committee that has made a point of not showing its hand.