GYP - Market Intelligence — Friday, August 7, 2026
The US economy lost 23,000 jobs in July. The market still prices 28bp of hikes by December.
Published for educational and informational purposes only. Nothing here is investment advice or a recommendation to buy or sell any security. Past performance isn’t indicative of future results.
The US economy lost 23,000 jobs in July. The market still prices 28bp of hikes by December.
Hold those two facts together, because everything else follows from the fact that they can both be true.
Payrolls were expected at +83,000. They came in at −23,000 — an outright contraction, and a 106,000 swing from consensus. June was revised down to +20,000 from +57,000. And yet unemployment fell, to 4.1% from 4.2%.
Markets did what you’d expect: gold +3%, the yen +1%, the dollar lower against everything, and the Nasdaq leading a broad rally. September hike odds roughly inverted — Kalshi now prices 62% no change against 34% for a hike, having been the other way round midweek.
But rate futures still carry 28bp of hikes by December, down only from 32bp. The labour market contracted this morning and the market barely blinked on the direction of the next move. That’s what 64 straight months of core PCE above target does to a reaction function.
Volatility
VIX 15.28, down 3.66% from 15.86 and roughly 22% below its long-run average near 19.5. But the shape of the curve is the interesting part:
The front fell four times harder than the back. The curve steepened rather than shifting down in parallel — the market is pricing calm now, not calm in three months. That’s a 1.22 VIX3M/VIX ratio, firm contango.
And bond volatility went the other way. MOVE rose to 76.12 from 73.58, up 3.45%, on the same session equity vol fell 3.7%. Equity vol down and rates vol up, simultaneously, is the tell: the hedging bid has moved from growth risk to inflation and fiscal risk.
Rates & Fed
The number that matters is the one nobody expected to be arguing about: the next Fed move may be a hike.
The Fed held at 3.50%–3.75% on July 28–29 on a 9-3 vote, with all three dissents hawkish — Hammack, Kashkari and Logan, each voting to raise 25bp. That’s the first three-member hawkish dissent since September 2016.
Their stated reason: inflation above 2% for more than five years. Core PCE ran 3.3% in June, the 64th straight month above target.
CME FedWatch had September hike odds at roughly 62% as of midweek, down from about 82% in late July. Not a certainty, but the debate is genuinely about up.
The next FOMC is September 15–16, with a fresh dot plot. July minutes land August 19. Between here and there: this morning’s payrolls, and July CPI on August 12.
The other side of the argument, for balance: Citi’s Veronica Clark still sees three cuts by January 2027 on unemployment rising above 4.5% — the mirror image of the dissenters.
Equities & Breadth
The rotation flipped, and the jobs report amplified it. For four straight sessions, mega-cap technology lagged a rising tape — and on Thursday the Nasdaq was the only major index in the red. It came into Friday leading, and after the print it ran away:
Long duration and high beta led; cyclicals lagged. Nasdaq gained double the S&P, small caps nearly as much, the Dow trailed, and crude extended its decline. That is the textbook signature of a market repricing the Fed lower, not one repricing growth higher.
The dollar confirmed it. Every one of six major crosses rose against the dollar after the print — Australian dollar +0.57%, yen +0.74%, Mexican peso +0.60%, Canadian dollar +0.45%, euro +0.44%, sterling +0.29%. Before 8:30 the dollar had been firmer against nearly all of them. A broad, uniform dollar sell-off like that is a rates repricing, not a currency story.
But look underneath the index and the dispersion is extraordinary. These are year-to-date moves in a single watchlist:
A market where one name is up 430% year to date and still 46% below its own 52-week high is not a market you can describe with an index level. SanDisk fell 6.8% yesterday. Micron fell 1.3%. The year’s biggest winners were the day’s biggest losers.
Amazon is +15.6% on the week, Marvell +14.9%, Nvidia +12.3%, Meta +9.4%. That’s the AI rebound the wires are crediting for the week — and it’s concentrated, not broad.
Cross-Asset
Metals were the story before the number and are still the story after it. Spot silver rose 5% to $64.56/oz overnight. Gold ramped 1.6% pre-print and has since added roughly 3% to $4,365.86 on the jobs data.
That puts the gold/silver ratio near 66, down from about 70 yesterday. Silver leading gold is the higher-beta, more inflation-sensitive version of a precious-metals move — it tends to show up when the market is pricing something more than a simple flight to safety.
Keith McCullough, who spent Thursday telling subscribers to book some gold gains ahead of a potentially hawkish jobs report, reversed this morning: gold is “back to signaling Bullish TRADE and TREND.” Worth noting that he changed his mind inside 24 hours, and before the event he was trimming for.
One detail makes this a cleaner signal than it looks. The dollar is flat — DXY sitting right at 99.95–100.00. Metals aren’t rallying because the dollar is falling. That makes this a real-rate and debasement story rather than a currency story, which is a more durable reason for a metals bid than an FX swing.
Credit is priced for perfection. High-yield spreads are around 284bp against a long-run median near 450bp — the richest decile in history. That’s not a warning by itself, but it is the definition of a market with no compensation left for being wrong.
And rate-sensitives took the hawkish hit this week: Utilities down more than 4%, REITs more than 2%. TLT is within about 60 cents of its 52-week low.
Crude keeps recovering. WTI $77.11, up about 1% over 24 hours and a third straight session of gains off the Hormuz collapse. The disinflation impulse that drove Monday’s tape is steadily fading.
Bitcoin +0.68% pre-market at $65,060, after several sessions of not participating in the equity rally.
Catalysts
July nonfarm payrolls: −23,000. Here is what the market was expecting, against what it got:
Every forecaster was on the wrong side of zero. The closest — Kalshi’s prediction market at +72K — was still 95,000 too high. The most optimistic missed by 138,000. This wasn’t a soft print; the economy shed jobs.
Two revisions matter as much as the headline. June came down to +20,000 from +57,000. That is the pattern Danielle DiMartino Booth has been hammering — 24 of the last 30 months revised negative — confirmed live on the tape.
And unemployment fell anyway, to 4.1%. Read that carefully: the economy lost jobs and the unemployment rate improved. That only happens when people leave the workforce rather than find work in it. It is the same mechanism as household employment being down 833,000 this year. The rate got better for a bad reason.
Going in, July ADP had already come in at +44,000 against +65,000 expected, the weakest since January. The warning was there.
And here is the number that reframes the whole series. Household employment is down 833,000 so far in 2026. The unemployment rate has stayed low not because hiring is strong but because participation is falling — people are leaving the workforce rather than finding work in it. A 4.2% unemployment rate built that way is a very different signal from a 4.2% built on hiring.
Alongside it: Q2 productivity came in at +1.4% and unit labour costs at +1.3%, both below forecast.
July CPI lands Wednesday, August 12 — the last major inflation read before the September FOMC. PPI follows August 13, retail sales and preliminary UMich August 14, and the FOMC minutes August 19.
Sentiment
Greed at the index level, real hedging demand underneath. The CNN Fear & Greed gauge has been reading “Greed” all week, while CBOE equity put/call has been printing above 1.0 — unusual for equity-only flow, and a sign somebody has been buying protection into the melt-up.
The put/call series has been genuinely noisy: 0.55 on Monday, 1.11 on Tuesday, around 1.05 midweek. Treat single prints with caution.
Breaking Headlines
Trump says the Hormuz deal is “moving along.” Iran and Oman are discussing a framework, though Iranian lawmakers are simultaneously considering restrictions on US- and Israeli-linked ships. That single negotiation has been driving oil, gold, rate expectations and sector leadership all week, and it has reversed direction more than once.
The US is reviewing how China may be accessing Nvidia’s advanced AI chips through offshore channels, following reported breakthroughs in Chinese AI capability. That is a live policy risk in a tape already jumpy about semiconductors.
Jane Street is in talks to shift $11bn of debt to investors, per the FT — worth filing alongside the private-credit concerns several strategists have been raising.
Trusted Voices
Charlie Bilello (Aug 5) — a melt-up in price on a genuinely strong fundamental base: the S&P above 7,700 for the first time, Q2 earnings tracking +47% year over year, the best since Q2 2021, and margins at a record 16.7%.
But he flags that leadership broke: SOXX fell 21% in July, DRAM names 32%, while Value is beating Growth by more than 20% through seven months — on pace for the largest value outperformance on record. Against the melt-up he sets the 30-year at 5.27%, the six-year bond bear market, and national debt up $3.6T in 13 months heading toward $40T.
Danielle DiMartino Booth — calls the July vote the most hawkish since September 2016. Her argument is that the bond market has already done the tightening while labour cracks widen: 24 of the last 30 monthly payroll figures revised negative, and teen hiring tracking the lowest since 1948.
Keith McCullough (Aug 5) — has broadened his framework from a single regime to a rolling “Quad 3-1-2” path, which is why he is pushing rotation over index calls. Notably, only 5 of his 37 daily risk-range signals are bearish on trend — a more constructive stance than his spring positioning. The core 2026 trade is unchanged: fade the mega-cap complex into a slowing earnings rate-of-change, rotate into small caps and the commodity/gold/bitcoin complex.
On this morning’s print he is characteristically blunt: “It’s Friday and a good day for government guys to manipulate markets — get your napkins ready.”
Liz Ann Sonders’ most recent dated commentary is from early July, so she is not quoted as current.
Bottom line
The labour market contracted, and the market still won’t price a cut.
That is the whole story, and it is genuinely unusual. Normally a −23,000 payroll print with a downward revision to the prior month sends the front end scrambling to price easing. Instead, hike odds for September fell from roughly two-thirds to about one-third, and December still carries 28bp of hikes.
The reason is sitting in the other half of the mandate: core PCE at 3.3%, above target for 64 consecutive months, with three FOMC members already dissenting to raise. A weak labour market doesn’t buy a cut when inflation has been persistent for five years. It buys a pause.
Three things follow.
Precious metals front-ran it and are still leading. Silver was up 5% before the number, gold has since added roughly 3% to $4,365. The dollar is lower against everything. With DXY flat before the print, this was already a real-rate story rather than a currency one — the jobs data just added fuel.
The index level is still telling you very little. A name up 430% year to date sits 46% below its own high. The week’s biggest winners were yesterday’s biggest losers. Selling premium against an index and selling it against a single name have rarely been more different trades.
And the cheap-volatility setup got more interesting, not less. The VIX curve was pricing calm now and less calm in three months. A contracting labour market with an inflation-constrained Fed is precisely the shape of risk that sits in the back of that curve, not the front.
For a premium seller, an event that has passed is generally a better place to sell than one that hasn’t. But the September meeting is now genuinely two-sided, and August 12’s CPI is the next thing that moves it.
Every trade we take is posted for members. This content is published for educational and informational purposes only and is not investment advice or a recommendation to buy or sell any security. Past performance isn’t indicative of future results. Options involve risk and aren’t suitable for all investors.
Not a member yet? Join at growyourpile.com.
— Tony Rihan & Tony Battista