Gold fell.
That's the sentence that should give every buyer of Thursday's record high pause. The S&P 500 climbed 0.85% intraday to 7,816.7, a fresh all-time high, per TradingKey, tagged within minutes of a producer price index that came in softer than expected. Risk assets did exactly what the doves scripted. Gold did the opposite.
Softer-than-expected wholesale inflation should be rocket fuel for gold, the asset that thrives when real rates fall and the Fed leans easy. Instead it struggled to extend its rebound. OCBC's read is that the next leg higher "might not be straightforward." Translation: the market that supposedly just got a green light for easing declined to buy the one instrument that pays off if easing is real.
The market that supposedly just got a green light for easing declined to buy the one instrument that pays off if easing is real.
Hold that thought. It's the tell for everything else.
The headline story is clean and seductive. Two straight days of soft inflation data—Wednesday's CPI, Thursday's PPI—lifted the index to a record. Producer inflation was flat month-over-month against expectations for a modest rise. Year-over-year, PPI slowed to 4.7% from 5.5% in June. Core PPI rose 0.2%, under the 0.3% forecast. Cooling, cooling, cooling.
And traders responded by pricing out the hike. According to a social-media commenter, September odds of a 25bp move went from 55% to 40% in a week.
Here's the problem: the hike odds are moving faster than the prints themselves. A fifteen-point repricing on two data points is not a measured response to incoming information. It's positioning racing ahead of the evidence, momentum feeding on momentum.
Because look at what actually drove the disinflation. Final demand goods prices fell 0.7% month-over-month—the second straight monthly drop—and energy did the heavy lifting, down 3.1%. Strip food and energy from goods and prices rose 0.1%. The cool-down wasn't broad. It was concentrated in exactly the components most prone to reversal.
Food helped too, down 0.9%. But the marquee move there was lettuce posting its largest month-over-month price decline on record—a one-off driven by a cyclospora outbreak that spooked consumers. That is not disinflation. That is a produce scare masquerading as a policy signal.
Now the part the tape buried. Energy and airline prices were still feeling upward momentum from supply shocks tied to the Iran war. Ship traffic in the Strait of Hormuz has fallen to near three-month lows. And crude rose over the same week traders were busy pricing out the hike, which is why the CNBC read explicitly flags upside risk in next month's report.
So the disinflation you're celebrating is energy-led, sitting directly on top of a live oil-supply shock that is pushing energy the other way. The single biggest driver of July's soft print is, arguably, among the most likely things to reverse in August.
This is where the professionals stop nodding along.
"There is an inflation problem, but this Fed right now is probably going to be on hold," said BNY Wealth's Alicia Levine. Read that carefully. It is not a dovish statement. It says the underlying problem persists and the Fed's response is paralysis, not accommodation. On hold turns every CPI print into drama—which is exactly the volatility a record high papers over.
Verdence's Megan Horneman went further: "I think the market is getting too optimistic about the Fed and what they're going to do in the coming months." Her precise stance on a hike is unclear—one social-media account claims she wants one—but she's calling the euphoria overdone.
Then there's former Kansas City Fed President Esther George, whose read matters because it removes the usual excuse for easing. "The labor market has changed, but I think overall that aspect of the Fed's mandate has remained relatively stable in the face of a growing economy." If the labor side of the dual mandate is stable and the economy is growing, the Fed has no employment reason to cut and, per Levine, an inflation reason not to. That's the definition of on hold.
Inside the Fed the disagreement is sharper than the rally implies. Barkin says whether inflation is even on a downward trajectory "remains inconclusive." Hammack is arguing for hikes, calling post-shock inflation pressures broad-based. This is not a committee sliding toward relief. It's a committee split, staring at a 4.7% wholesale print.
Which brings us back to gold.
If the market truly believed the data vindicated the doves—that real rates were heading lower and easing was coming—gold should have caught a bid on Thursday's soft PPI. It fell. The asset with the cleanest, most direct claim on a genuine easing cycle voted no.
What rallied tells you what this actually was. Memory-chip names led the tape—SanDisk, Micron, SK Hynix, the Philadelphia Semiconductor Index up 2.07%—because falling bond yields improve the valuation math for long-duration growth. That's the rate-sensitivity mechanism transmitting the macro repricing straight into the most rate-levered corner of the market. The equity-specific verdicts belong elsewhere; the point here is directional. Capital rotated into what benefits most from lower rates, not into what benefits most from actual disinflation.
That is a positioning trade, not a conviction trade. Rotation into rate-sensitive growth on a yield move is not the same as the market believing inflation is beaten. Gold is the control group, and the control group didn't move the way a real easing signal demands.
So price the record honestly. An all-time high notched the instant a friendly PPI hit the tape is a trophy photo, not a floor. The number is real; the interpretation is fragile. Strip out energy and a lettuce scare and the disinflation thins to almost nothing, while the supply shock underneath it points the other way for next month.
The setup is a melt-up built on a rate-path repricing that outran its own evidence—reportedly 55% to 40% on two soft, narrow prints, with the Fed's own former officials pointing at an inflation problem and a stable labor market that argue for exactly nothing.
Levine gave you the frame. On hold turns every print into drama. That works beautifully on the way up, when soft data buys you a record. It works the other way just as fast when crude carries into next month's report and the 4.7% that everyone waved through becomes the number that matters.
Headlines don't get a vote on the stop. Gold already cast its.