There are moments when the bond market stops waiting for the Fed and starts doing the job itself. Wednesday was one of those moments.
The September FOMC minutes landed at 2:00 p.m. ET and confirmed what yields had already priced: the committee that voted unanimously to hike in September sees more work ahead. "Many participants" assessed that higher rates would likely be necessary if inflation did not fall. That language is not a hedge. It is a forecast dressed in committee grammar.
Meanwhile, the 10-year Treasury yield touched 5.34% on the session, its highest since May 2002. The S&P 500, which had closed at a record on Tuesday, gave back ground. The Dow fell 341 points. Nvidia slipped. Meta dropped 2.4%. The whole growth complex flinched.
But here is the part the market has not fully absorbed. The September hike was unanimous, 12 to 0. The July hold was 9 to 3 against. The committee did not come around slowly. It snapped. That is a very different animal from a tentative, one-and-done insurance hike. The October 28 meeting is now three weeks away. And this morning's jobless claims report at 8:30 drops right into the middle of that conversation.
Let's get into the signals.
The minutes confirmed what the yield curve already knew
Let's be precise about what the Fed told us Wednesday and what markets chose to hear.
The minutes of the September 15-16 FOMC meeting showed the committee voted unanimously to raise the federal funds rate by 25 basis points to 3.75%-4.00%. That was the first hike since July 2023. The September dot plot moved the median year-end 2026 projection to 4.1%, which implies at least one more quarter-point increase before December.
The key phrase: "many participants assessed that a further firming of the stance of monetary policy would likely be appropriate if incoming information was not consistent with inflation moving back toward the Committee's objective." In FOMC language, "many" means more than half but fewer than all. That is not a minority view. It is a governing consensus.
What the minutes do not capture is everything that happened after the meeting. September payrolls came in at just 29,000, well below the roughly 90,000 expected. August PCE showed core inflation beginning to stabilize. Both of those data points have pulled market-implied odds of an October hike down to roughly 20%. BofA sees it differently, and we will get to that.
The market signal?
The minutes read hawkish, but the post-meeting data reads soft. That gap is why October 28 is the most consequential FOMC meeting of the year. September CPI on October 14 will decide whether the doves or the hawks walk in with the stronger hand.
Elon Musk's $2.1 Trillion Hit List
The SpaceX IPO made Elon Musk the wealthiest man who has ever lived.
Now he's spending it to complete his dynasty...
He's already acquired a $40 billion AI firm and a $1 billion turbine company.
But a handful of companies still stand between him and total control of the space economy.
Click here to see why early investors could walk away with a fortune.
The 10-year is writing policy now
The 10-year Treasury yield has now spent over a week above 5.30%, territory last occupied in May 2002. The 30-year is hovering near 5.70%. To put that in human terms: the 30-year fixed mortgage rate is at 7.3%, its highest since November 2023.
Wednesday's session gave the bulls a momentary reprieve. A solid $39 billion 10-year auction steadied nerves, and yields pulled back slightly from intraday highs. But the broader picture has not changed. The curve is steepening from the wrong end. The 2-year sits near 4.19%, the 30-year near 5.70%. That is over 150 basis points of slope, and nearly all of this month's widening came from the long end rising, not the short end falling.
This is not a rate-cut trade. It is a term-premium trade. When gold jumps, bitcoin rallies, and the long bond sells off simultaneously, the story they are telling is debasement risk, not growth. Gold closed above $4,700 this week. Bitcoin cracked $77,000.
The S&P 500 (SPX) closed Wednesday at 7,801.77, down 0.22%. The Dow (DJI) lost 341 points. The Nasdaq (COMP) slipped 0.22%. The pullback was orderly. But the Shiller CAPE ratio at 42.78, its second-highest reading in history, makes every session at these levels feel borrowed.
The market signal?
As long as the 10-year holds above 5.25%, every equity multiple is being priced against a discount rate the Fed does not control. A sustained break below 5% changes the conversation. Nothing in the next 48 hours makes that likely.
BofA's Cabana is daring the Fed to blink
Bank of America's Mark Cabana has been the loudest voice on rates for months, and on this he has been consistent. His message: the Fed's choice is "hike or bond spike."
BofA's house call remains three quarter-point hikes in 2026: September (done), October, and December, taking the fed funds rate to 4.25%-4.50% by year-end. That view puts BofA well ahead of markets, which are pricing roughly 20% odds of an October move and about 80% odds of at least one more hike by December.
Cabana's argument is structural. Nominal consumer spending is running over 6% year-on-year. That is well above the level historically tied to hot core inflation. Tightening cycles are front-loaded. The Fed almost never stops after one hike. And Chair Kevin Warsh's track record suggests he will tolerate labor-market weakness before he tolerates inflation persistence.
The counterargument is the data. September payrolls at 29,000 were genuinely weak. If Thursday's jobless claims print above 210,000 or Friday's revisions show further softening, the October hike case crumbles. BofA has flagged this risk itself.
The market signal?
If Cabana is right, the 10-year has room to push toward 5.50% and equities re-rate lower on forward earnings. If he is wrong and the Fed pauses in October, yields likely reverse sharply and the next leg of the equity rally begins from a lower base. The binary is uncomfortably clean.
A "bloodbath" Is Coming
Silicon Valley billionaires are hoarding guns, gold, and even military grade gas masks... or fleeing the country altogether. What do they know that you don't? This AI insider is stepping forward with this time-critical message: Move your money before October 31st, ahead of the dangerous next phase of the AI market. Here's the one urgent move you need to make with your money immediately.
This ad is sent on behalf of TradeSmith at 1125 N. Charles Street, Baltimore, Maryland 21201. If you're not interested in this opportunity, pleaseclick here.
Marvell just told the market what 2031 looks like
On Monday, Marvell Technology (MRVL) held its investor day and dropped a number that made the room go quiet: $70 billion to $90 billion in revenue by fiscal 2031.
That is from a company that generated $8.2 billion in fiscal 2026. CEO Matt Murphy is projecting something close to a tenfold expansion in five years. The growth engine: custom AI chips designed for hyperscaler customers (including a $120 billion, six-year deal with Alphabet signed earlier this year) and high-speed optical interconnects that move data inside AI data centers.
Recall Jensen Huang's endorsement at Computex: "This is the next trillion-dollar company." Nvidia (NVDA) invested $2 billion in Marvell in March. The two companies are co-developing silicon photonics. Marvell is not competing with Nvidia. It is building the connective tissue that makes Nvidia's GPUs function as a system.
The investor-day presentation also reframed the competitive landscape. Murphy argued that connectivity, not compute, is about to become the primary bottleneck in AI infrastructure. If he is right, Marvell's optical DSPs and custom ASICs are positioned exactly where the supply constraint is tightest.
The market signal?
At current prices, Marvell would need a price-to-earnings ratio near 50 and approximately $60 billion in revenue to justify a $1 trillion valuation. That could plausibly arrive by 2030 if the growth rate holds. The investor day was designed to make that math visible. Whether the bond market lets investors pay those multiples is the question Marvell cannot answer.
The world is burning through its insurance policy on oil
Last Friday, G7 leaders agreed to release 100 million barrels of crude oil and diesel from emergency reserves over four months, including a frontloaded diesel release within the first 20 days. It was the second massive coordinated drawdown of 2026. The March release, triggered when the Strait of Hormuz effectively closed, moved 400 million barrels.
Add the numbers. That is half a billion barrels of emergency stocks released in seven months. The IEA's members hold just over one billion barrels in total. At this pace, the strategic insurance that took decades to build is being spent in under two years.
US diesel prices hit a record $6.52 per gallon on September 22 before pulling back to $6.37 after the G7 announcement. Brent crude is still above $101. The emergency releases are keeping a lid on the worst-case scenario. But they are not fixing the underlying problem: Hormuz remains constrained, Houthi attacks on Saudi oil infrastructure continue, and the ceasefire framework that was "almost there" in May has not produced a signature.
Strategic reserves are a fire extinguisher, not a fire department. Every barrel released now is a barrel unavailable later. If a true supply shock arrives in 2027, the world's insurance policy will be materially smaller than it was eighteen months ago.
The market signal?
Watch the pace of reserve depletion alongside crude inventories. The G7 releases are buying time for diplomacy. If diplomacy fails, the repricing of energy risk will be violent and the Fed's rate path will tighten further, not less.
Buffett, Gates and Bezos Quietly Dumping Stocks—Here's Why
The world's wealthiest individuals are making huge moves with their money.
Warren Buffett just liquidated billions of shares. Bill Gates sold 500,000 shares of Microsoft. Jeff Bezos filed to sell Amazon shares worth $4.8 billion.
What is going on? One multi-millionaire believes they are preparing for a catastrophic event. But not a crash, bank run, or recession. It’s something we haven’t seen in America for more than a century.
- Weekly jobless claims land at 8:30 a.m. ET this morning. Recent readings have held broadly below 200,000, signaling a labor market that refuses to crack. A print above 210,000 would be the first soft signal ahead of the October 28 FOMC meeting, potentially pulling back October hike odds further. A steady number keeps the hawks on the prowl. Full calendar.
- September CPI drops next Wednesday, October 14, and it is the single most consequential data point before the Fed meets. Kalshi prediction markets are pricing annual CPI somewhere between 3.5% and 3.7%. A print at the higher end would likely push October hike odds back above 30% and send the 10-year toward 5.50%. A cooling print would give equities room to breathe.
- The market signal across both? Thursday and next Wednesday are the two data gates that determine whether the Fed hikes or holds on October 28. Everything else is noise until those numbers print.
That's it for today's Slate. Claims at 8:30 set the tone. Stay close to the feed.
Today's reply prompt: The Fed hiked unanimously in September. Is the October meeting a hold or a hike, and what's your one-sentence case?
3 Stocks at a Major Turning Point
Something unusual is happening beneath the surface of three widely followed stocks.
In each case, the fundamentals are saying one thing... while institutional activity, management signals or the options market are saying something else.
That kind of disagreement can matter.
Because when the evidence stops lining up, the next move in the story often comes down to a handful of signals most investors never think to watch.
Our new FREE Market Tell Special Report breaks down three of these situations, and shows you exactly what we’re watching next.
Tickers: MS FOMC ET PCE CPI SPX DJI COMP CAPE MRVL NVDA IEA POLL