Gold Is Right About This Jobs Report And The Nasdaq Is Not

On Friday the United States reported that it had lost jobs, and two very different assets rallied hard on the same headline. The Nasdaq Composite closed up 1.3% at 26,690.62, capping a 5.2% week. Gold settled at $4,350.37 an ounce, up around 6.6% over the week and its strongest five days since January. The S&P 500 finished at a record 7,757.64, up 3.6% on the week. All of it was attributed to one number: a July payrolls print of minus 23,000, against a consensus looking for a gain of about 83,000.

Those trades are not compatible with each other. The equity market rallied on the view that a softer labor market removes the Federal Reserve's reason to tighten, and that this is a benign outcome. Gold rallied on the view that the Fed now will not tighten into inflation running at 3.5%, and that this is a dangerous one. Only one of them is reading the report correctly. My view is that gold has it right, and the equity market has misread what actually happened inside the data.

The unemployment rate fell for the worst available reason

The headline that reassured people was the unemployment rate, which slipped to 4.1% in July 2026. On its own that looks like a labor market that is cooling gently rather than cracking. It is not what happened.

The rate fell because the civilian labor force shrank by 264,000 people over the month. The participation rate dropped to 61.4%, its lowest in more than five years. An unemployment rate is a ratio, and this one improved because the denominator got smaller, not because the numerator did.

This distinction matters more than it sounds. Payroll weakness caused by falling demand for workers creates slack, and slack is disinflationary. Payroll weakness that arrives alongside a shrinking workforce creates far less of it, because the economy's breakeven rate of job creation, the monthly gain required to hold unemployment steady, falls at the same time. Minus 23,000 jobs against a labor force that is contracting is simply not the same quantity of spare capacity that minus 23,000 would have represented two years ago.

Where the losses landed matters too. Government payrolls fell by more than 50,000 in July, and leisure and hospitality shed around 40,000. The first is a policy decision rather than a read on private demand. The second is the most cyclically exposed corner of the labor market. Neither is the profile of a broad-based collapse in hiring appetite across the economy.

The revisions reinforce the point rather than soften it. May was marked down by 66,000 to a gain of 63,000, and June by 37,000 to a gain of 20,000, taking the two months together 103,000 lower. The twelve-month average is now around 34,000 a month. That is a genuinely weak hiring economy. It is also, on these participation numbers, an economy where fewer people are available to hire.

This inflation was never a wage story, so weak jobs do not fix it

Here is the part the Friday rally skipped over. Average hourly earnings in July 2026 came in at $37.62, up 3.2% over the year, below the 3.5% economists had expected. Consumer price inflation was running at 3.5% year over year as of the June 2026 print.

Read those two figures next to each other. Wages are growing more slowly than prices. Real earnings are going backwards. Whatever is keeping American inflation above target, it is not a wage-price spiral, and it has not been for some time. The pressure has come from the supply side, principally energy, with Middle East disruption keeping crude elevated for much of the year.

That is why the market's reflex is misapplied. The logic that weak payrolls mean lower inflation runs entirely through the labor-cost channel. If labor costs were not the source of the problem, then cooling them further tells the Fed almost nothing about the problem. Friday's report contained a great deal of information about American growth and essentially none about American inflation. Yet the market repriced September Fed policy on it, cutting implied hike odds from roughly 55% to somewhere around 40%.

The Fed being priced is not the Fed that exists

There is a second assumption embedded in the equity rally, which is that this Fed reacts to labor weakness the way its predecessors did. That is a large bet on an institution that has just changed hands.

Kevin Warsh became Fed chair on 13 May 2026. At the meeting of 28 and 29 July the Committee held the target range at 3.50% to 3.75% for a fifth consecutive time, and issued a notably shortened statement built around a single commitment: that it will deliver price stability. Warsh used his press conference to say plainly that inflation above 2% is unacceptable, and the Committee signaled a hike was live for September.

A chair three months into the job, whose entire public identity is built on inflation credibility, and who has deliberately stripped the policy statement back to that one promise, is not the profile of someone who abandons a signaled hike because the labor force shrank. He may still not hike. But if he holds, it will be because inflation data cooperated, not because payrolls disappointed, and the market has been pricing the wrong causal chain.

The long end of the curve refused to play along

The cleanest evidence that Friday's equity move was a misread is that the bond market barely participated in it.

The 10-year Treasury yield eased to 4.65% on 7 August 2026. In the week ending 3 August it had touched roughly 4.74%, its highest since January 2025. So a negative payroll print, with three months of downward revisions attached, bought the long end about nine basis points off a multi-year high. That is not the reaction of a bond market that believes growth is rolling over into a comfortable disinflation.

The curve tells the same story. The 10-year to 2-year spread widened out to around 99 basis points from 81 basis points during the week of 27 July, a bear steepening that occurred without any change in Fed policy. Front-end yields can fall on weak growth data. Long-end yields, which carry the inflation risk premium, have refused to follow. When the short end rallies and the long end does not, the market is telling you it expects easier policy and worse inflation outcomes at the same time. That is precisely what gold at $4,350 an ounce, and silver up 11.6% on the week through $64, are also saying.

The strongest case against me

I should put the other side properly, because it is not weak. Two arguments deserve respect.

The first is valuation. The S&P 500 was trading at a forward price-to-earnings multiple of about 20.4 times on 7 August 2026, against a ten-year median near 19.95 times. That is not a mania. If you think American equities are priced for perfection, the multiple does not obviously agree with you, and a market at fair value can absorb a 25 basis point hike without much drama.

The second is energy. Brent crude was trading in the low $80s per barrel in early August 2026, well off the levels above $100 seen earlier in the year, and the Energy Information Administration cut its third-quarter 2026 Brent forecast to an average of $74 a barrel, some $27 lower than its previous outlook, with 2027 seen averaging $65. If crude behaves, headline inflation falls mechanically over the next two quarters and the Fed's problem solves itself.

I accept both. But note what the second one concedes. If disinflation arrives, it arrives from the oil market, not from the labor market. The equity rally on Friday was priced off a payrolls number, and it is taking credit for a disinflation that crude, not employment, would deliver. Being right for the wrong reason is not a durable position, because it means you have no idea which data actually matters to your position.

What would change my mind

This claim is falsifiable, so let me say how. I am wrong if core inflation, stripped of energy, decelerates toward 2.5% over the autumn prints while the labor force stabilizes and participation stops falling. That would mean genuine slack was building after all and the Fed's pause is well founded.

I am also wrong if the 10-year yield breaks decisively below about 4.4% while growth data continue to deteriorate. That would be the bond market ratifying the equity view, and it would tell me the inflation risk premium I think is embedded in the long end has evaporated.

What I do not expect is the arrangement currently being priced: a Fed that stops tightening, an inflation rate that falls anyway, and a labor market that is contracting on the supply side without consequence for costs. Gold going up 6.6% in a week is the market quietly admitting it does not believe that combination either. The Nasdaq has simply not noticed yet.