1A September cut is priced at one chance in two hundred
The Federal Reserve decides on 16 September. Polymarket's contract on that meeting carries $5.76m of liquidity and traded $2.9m in the last 24 hours, so its prices are worth reading. A 25bp increase sits at 51.5%. No change sits at 47.5%. Every cut outcome combined sits at 0.6%.
That last number is the story. On Friday the President told the Fed to lower rates or he would stop trading with the countries America runs a goods deficit with. The market has responded by pricing his stated demand at one chance in two hundred.
Ross Givens, on his own channel, argues the demand is self-defeating on its own terms. Since the first cut of this cycle in September 2024 the Fed has cut 175bp, and over exactly that window the 30-year mortgage went from 6.09% to 6.71% — Freddie Mac's number for the week of 3 September, 62bp higher, not lower. His mechanism is that a mortgage is priced off the 10-year Treasury plus a lender spread of roughly 1.9 points, and the Fed sets neither. What moves the 10-year is expected inflation and trust in whoever sets policy. Pressure applied to the short end therefore leaks into the long end as a risk premium.
Ben Hunt, in a clip played on Excess Returns, puts a measurement behind the same idea. His firm tracks how densely a narrative appears across news, and he says the story "the Fed has lost credibility" flipped from contested to confirmed after Kevin Warsh talked hawkish in late July and then did not move. Hunt's framing is that credibility is a teacup: glue it and it still holds tea, but it is never the same again. Matt Zeigler, on the same show, ties the series to the gold price directly — gold rose during the earlier public fight over the Fed, went sideways once Warsh arrived and credibility recovered, and started moving again after Jackson Hole.
A thinner contract corroborates the fracture. Polymarket's market on how many officials dissent on 16 September has only $19.1k of liquidity, so treat it as weak, but it prices four or more dissents at 30.5%, up 9.5pp on the week, and a unanimous decision at just 10.5%.
The consequence is that the 16 September number is the least informative thing the meeting will produce. Whether the long end and the 2s30s gap tighten or widen on the day is what says whether the market thinks the decision was made independently.
2Oil, not payrolls, is driving the hike bid
The August employment report landed on 4 September and beat expectations. The unemployment rate held at 4.1% and payrolls added 162,000 jobs to 159.08 million. That was the week's dominant story, and over the past week the hike leg of the September contract actually fell 5pp while the hold leg gained 5pp.
The last 24 hours reversed part of that: the hike leg gained 3pp and hold lost 4pp. Payrolls did not change over that window. Oil did. Crude reached six-week highs on worsening Middle East conflict, with Reuters reporting Hormuz traffic at its lowest since May after strikes on shipping, and pre-market commentary putting crude near $97. Goldman Sachs is reported to see $120/bbl risk if attacks on vessels intensify. Copper hit an all-time high on tariff disruption.
The transmission is not the headline oil price on its own. Reuters' Morning Bid names diesel as "the real spoiler for US inflation doves", and the mechanism is a refinery product-mix problem: refiners strained by the war are favouring other products, so a ship-fuel shortage is building. Distillate feeds freight, and freight feeds core goods — which is how an energy shock stops being a headline-only shock.
The inflation contracts show that migration happening. On the August core CPI print due 11 September, Polymarket's month-on-month contract has 0.2% at 54.5%, up 6pp on the week, and 0.3% at 31.5%, up 7.5pp, while the 0.1% outcome collapsed 12.5pp. Both are thin — $2.8k of liquidity — so they are directional colour, not proof. The distribution is moving up, not down, in the week before the print.
For context on what the Fed is working against: headline CPI ran 3.30% year-on-year in July and core PCE 3.34%, against a fed funds rate of 3.63%. Real policy is barely positive. An energy shock into that starting point is a different problem than the same shock into a 5% policy rate.
The consequence is a two-meeting question rather than a one-meeting question. October's contract, on $1.4m of liquidity, now prices a 25bp increase at 30.5%, up 3pp in a day and 4pp on the week.
3The war is being priced to end while its damage is still being priced in
The news flow is uniformly escalatory. A Saudi-led coalition reported 73 injured in Houthi attacks on the kingdom. Iran threatened US energy assets in the Gulf and warned of "economic warfare". Oman evacuated the crew of a Saudi vessel attacked by Iran.
The prediction markets moved the other way, and hard. Polymarket's contract on a US–Iran effective ceasefire, on $144.5k of liquidity and $90.8k of 24-hour volume, prices a ceasefire by 11 September at 70.5% — up 12pp in a day. By 18 September it is 75.0% and by 30 September 80.0%. A separate contract on an Iran–Oman Hormuz management agreement, on $143k of liquidity, prices a deal by 31 October at 49.0%, up 9.5pp in a day and 16.5pp on the week.
There is a source for the divergence rather than a mystery. The Wall Street Journal reported that Iran's top politicians are urging an end to the war — a story the President posted himself. The markets are pricing the political end-state; crude is pricing the physical disruption that has already happened.
Both can be right, and that is the useful part. A ceasefire ends the escalation premium; it does not immediately restore Hormuz traffic, refloat the ships, or unwind the refinery product-mix problem that Reuters identifies as the inflation channel. So the two prices are answering different questions on different clocks.
What would settle it for the 16 September meeting is narrow: whether crude holds near current levels through the 11 September CPI print. If the ceasefire contracts are right and crude gives the move back before then, the last 24 hours of hike-bid was noise. If crude holds, the diesel channel gets a second week to work.
4US equities have not priced any of it yet
The US market was shut on Monday for Labor Day, so the last American close was Friday 4 September. Everything above — the six-week-high crude, the Houthi strikes on Saudi Arabia, the record copper price, the yen at a seven-month high — happened while US equities were closed.
Friday's tape shows the gap plainly. Energy did not rally into the weekend: fell 0.87% and closed unchanged. fell 0.84%, which is the strong-payrolls-means-hike-bets read, not the oil read. fell 0.38% to 770.18.
The single-stock moves that week were idiosyncratic rather than macro. rose 5.91% to $1,015 and 4.52% on memory and semiconductor-specific news, while fell 5.97% and dropped 19% on its own results. None of those are read-throughs to the book; the macro repricing had not started.
The consequence is that Tuesday's open is the first opportunity for US equities to mark any of this. The 11 September CPI print then arrives four days later, and the Fed five days after that, in a week the calendar otherwise leaves empty — on 10 September is the only large-cap earnings event of the week.
What the sources recommended
Four videos produced one single-name recommendation between them, and it was a recommendation against something rather than for it. Three of the four were macro or educational pieces that named no tradeable position at all, which is itself the shape of a week where the argument is about rates rather than about companies.
The one call: Glen Hare, a financial adviser appearing on Equity Mates, was asked whether he would debt-recycle into a geared diversified ETF and answered flatly that he would not. His objection to was double gearing — borrowing to buy a fund that is itself geared — against a recommended holding period of eight to ten years for members in their twenties and thirties whose circumstances change repeatedly over that span. He said he had shut the idea down when members raised it.
Two mentions came with explicit refusals to make them calls. Felix Prehn, on Felix & Friends, cited and as examples of businesses with pricing power that can raise prices without losing customers, and said in the same breath: "not telling you to buy those." He also said he had bought "some railway stock" recently without naming it. Neither is a recommendation and neither is treated as one here.
Dan Niles, in a clip on Excess Returns, described as one of his own disasters this year — a former category leader hurt by years of mismanagement and new competition — and used as a second example of a name people were told to put away for grandchildren. Both were arguments against buy-and-hold as a doctrine rather than sell recommendations, and his framing was that conviction should be strong but loosely held. His broader point rested on : its own investor-relations archive records bookings going from +70% year-on-year to −30% while it was the most valuable company in the world.
The next four days test all of it
The August PPI lands on 10 September and August CPI on 11 September. Those are the direct tests of section 2 — whether the energy shock has reached core. The Fed follows on 16 September, and section 1's argument says to watch the 10-year and the 2s30s gap on the day rather than the headline decision.
On the war, the near-dated ceasefire contract resolves on 11 September, the same day as CPI. That is an unusually clean read: within four days both the inflation question and the de-escalation question get marked.
Earnings are almost absent. on 10 September is the week's only significant report, alongside , and on 9 September.
