E-mini S&P futures settled the Labor Day session at 7,708.75. Convert that at Friday’s basis and the futures are pricing the S&P to reopen near 7,705 — barely above a put wall at 7,700 and thirty-four points above the gamma flip, the thinnest cushion this letter has logged. Crude went the other way again: WTI settled $92.70, Brent $97.00, with Hormuz transits at their lowest since May. The Fed is silent until the 17th, the strip prices a hike at 60.4%, and the option book’s own centre of gravity walks down seventy points between this morning and the decision.
| Instrument | Last | Change | Note |
|---|---|---|---|
| ES Sep Mon settle | 7,708.75 | −0.17% | Range 7,703.50–7,728.50 on 137,293 lots. The low stopped a tenth of a point above the put wall in futures terms |
| NQ Sep Mon settle | 29,603.50 | +0.13% | Green for a second session and the only US index contract that was |
| YM Sep Mon settle | 53,111 | −0.62% | A 329-point hole against a Nasdaq that closed up — same direction it took Friday |
| RTY Sep Mon settle | 2,968.10 | −0.29% | Gave back Friday’s small-cap lead; the front end did not move because it was shut |
| S&P 500 cash, Friday close — still the last print | 7,718.60 | −0.38% | Payrolls day; range 7,706.12–7,750.19; 1.25% under the 13 August record of 7,816.70 |
| Nasdaq / Dow / Russell cash, Friday close | 26,506.99 / 53,414.25 / 2,975.65 | −0.29% / −0.51% / +0.25% | The Russell was the only one green — the tell that shows up when the front end moves |
| WTI Oct / Brent Nov Mon settle | 92.70 / 97.00 | +1.33% / +0.75% | Highs of 93.29 and 98.06, the best since 24 July. Brent traded 200,962 lots — a heavier book than the equity contract |
| Gold Dec / Silver Dec Mon settle | 4,452.0 / 66.815 | −0.55% / +0.10% | Gold’s fourth straight lower session with the Gulf escalating — trading its real-yield beta, not its geopolitics beta |
| Nat gas Oct / Copper Dec Mon settle | 2.966 / 6.7040 | −0.30% / +0.32% | Gas is the leg Goldman prefers to crude for this risk and it did not participate |
| ZN Dec / ZB Dec Mon settle | 107-11½ / 108-16 | −0.10% / −0.17% | Cash Treasuries shut. The note future is sitting on the 107.25 52-week low it printed Friday |
| US 10Y / 2Y / 30Y Friday close | 4.784% / 4.374% / 5.244% | +2.2 / +4.0 / +0.1 bp | Two- and five-year both at 52-week highs; curve +41.0 bp. $119B of coupon supply lands today through Thursday |
| DXY / EUR / USD-JPY live | 98.917 / 1.1622 / 154.35 | −0.26% / +0.08% / −1.21% | The yen is the strongest major for a second session, low 154.07, on BoJ repricing into 18 September |
| Bitcoin / Ether live | 79,189 / 2,498 | −0.79% / −0.7% | BTC back through $80,000 after failing at $82,000; still +25% on the month |
| VIX Sep future Mon settle · expires Sep 16 | 16.30 | +0.20% | Spot VIX did not disseminate Monday. Friday’s Cboe close was 14.53 — a 1.77 vol discount to the contract that settles the morning after the decision |
| Nikkei / Kospi / Hang Seng Monday close | 66,399.84 / 6,995.39 / 25,413 | +2.12% / +4.61% / −0.85% | Korea is the largest print anywhere this week; memory and foundry led. CSI 300 +0.59% |
| Stoxx 600 / DAX / FTSE Monday close | −0.1% / −0.38% / +0.18% | mixed | Thin with New York shut. Oil & gas +1.14% best, healthcare −1% worst on the Lp(a) trial failure |
| Gauge | Reading | Read |
|---|---|---|
| Dealer regime Friday close, all expiries | Long gamma | Net dealer gamma +$30.1bn per 1% move at the cash close, against +$79.9bn two sessions earlier — the dampener is 62% smaller than it was Thursday |
| CNN Fear & Greed Friday close | 41.9 · Fear | Up from 35.2 the prior session; 52.3 a week ago, 60.0 a month ago. It cannot move until cash reopens |
| AAII bull / bear w/e Sep 2 | 39.7% / 37.6% | Spread +2.1. The story is neutral at 22.7%, eight points below its long-run 31.0 — respondents left the fence and split to both poles |
| Cboe put/call Friday | 0.58 equity · 0.89 index | Single-stock calls bought while index puts are held. Complacent at the name level, hedged at the index level |
| VIX complex Friday closes | 14.53 · 9D 11.97 · 3M 17.61 | Deep contango, day 105. VVIX 84.42; SKEW 151.58 — tails bid while the body is sold |
| CME FedWatch Sep 15–16, live | Hike 60.4% | Hold 39.6%, ease 0.0%. Prediction markets still sit near even — Section 09 |
The flow read is a market short the index it cannot move and long the one that is moving. Commitments of Traders as of 1 September has non-commercials net short 75,941 E-mini S&P, another 12,911 shorts on the week, with the micro net short 134,302 after 33,791 were laid on in five sessions — while they covered 15,049 Nasdaq-100 shorts and hold net long 25,890 there. Two positions look exposed: gold longs of 228,124 were trimmed only 16,674 into a fourth down session, and Brent shorts were increased by 8,382 in the week before the US disabled three tankers.
| When | Event | Reference | Why it matters |
|---|---|---|---|
| Tue 06:00 | NFIB small business (Aug) | 99.7 e / 99.3 p | The small-business read on the same wage question the committee is arguing about |
| Tue 11:00 | NY Fed inflation expectations (Aug) | 3.6% e / 3.6% p | With petrol at $4.14 and diesel at a record, an upside print is the anchoring argument the hawks want |
| Tue 13:00 | 3-year note auction | $58B · 4.291% prior | First of $119B across three days, into a two-year at a 52-week high and a session absorbing whatever the weekend did |
| Wed 13:00 | 10-year reopening · Apple event | $39B · 4.683% prior | Treasury’s long-end buybacks double to at least $4B per operation from Wednesday. First Apple launch under John Ternus |
| Thu 08:30 | August PPI · claims · 30-year reopening · ECB | 5.3% y/y e · 4.7% p | A 60 basis point jump in the headline, one day before CPI. The less-watched print, and the one that sees diesel first. ECB expected to hike to 2.50% |
| Thu AMC | Oracle · Adobe | ADBE $6.08 e | Oracle’s $638B backlog against negative free cash flow is the AI-credit read-through |
| Fri 08:30 | August CPI · U. Michigan 10:00 | 0.4% / 3.4% · core 0.2% / 2.4% | The print the meeting turns on. The Cleveland Fed nowcast has core at 2.38% — below consensus. Section 09 |
| Sep 15–18 | FOMC (16) → BoJ and quarterly expiry (18) | — | $6.2 trillion of notional expires on the 18th, the same morning the BoJ meets with a hike 63% priced |
The settled map has not moved since Friday, and it could not: there was no cash session on Monday to reprice open interest against. What moved is the futures contract, and that is the story. ES settled 13.25 lower while the index it tracks sat frozen, so the cushion between spot and the flip has been narrowed by the futures rather than by the index. Convert Monday’s settle at Friday’s +3.40 basis — the last clean pairing available, since cash was shut — and the futures are marking the S&P near 7,705. Against a flip of 7,671.32 that is a cushion of about 34 points, or 0.44%. It was 47 points yesterday and 103 the session before. The regime is still positive. It is running out of room.
| Level | SPX | ES Sep · +3.40 | Role in today’s tape |
|---|---|---|---|
| Call wall all expiries | 7,800 | 7,803.40 | Fourth consecutive session at this strike and the one level every book agrees on. 79,416 calls sit there in aggregate, carrying the largest strike gamma on the board at roughly +$5.9bn per 1% move. It is 1.2% above the futures mark |
| Max pain Tue Sep 8 expiry — 0 DTE today | 7,720 | 7,723.40 | Fifteen points above where the futures are marking cash. Yesterday this was a Tuesday problem; this morning it expires, and the pull dies with it at the close |
| Put wall all expiries | 7,700 | 7,703.40 | Third session unchanged and 64,589 puts deep. Monday’s holiday low was 7,703.50 — a tenth of a point above it. Tested and held twice now without cash participating |
| Gamma flip regime line | 7,671.32 | 7,674.72 | Unchanged because the map is. The one-day implied move on this morning’s expiry is ±39 points, putting the downside boundary at 7,669 in futures — five points through the flip. An ordinary day now reaches the regime line |
Underneath the aggregate, the two books that actually trade this week disagree about support by a hundred points, and that disagreement is the most useful thing on this page. The September monthly, ten days out and carrying the quarterly, puts its heaviest put open interest below spot at 7,600 with roughly 57,200 contracts, then 7,550 with 47,500 — a shelf running 7,550 to 7,600 that sits a full percentage point beneath the aggregate wall. The contract expiring this afternoon defends somewhere else again: its largest put position below spot is 7,660, about 3,720 contracts, with a second block of 2,519 at 7,680. Three books, three floors: 7,700, 7,660, 7,600. The ceiling, by contrast, is unanimous — the monthly’s heaviest call strike above spot is 7,800 with roughly 47,500 contracts, and this morning’s expiring book peaks at exactly the same strike.
Read that as a market that has stopped arguing about upside and started disagreeing about how far down the first air pocket goes. The day book defends 7,660 until four o’clock and then stops existing. The monthly’s 7,600 shelf does not expire until the 18th. Between them sits the aggregate wall at 7,700, tested twice this week and held both times — on a holiday, with no cash market behind it.
The other thing worth knowing about the cushion is when it expires. Of roughly $25.5bn of net dealer gamma spread across the four nearby contracts, about $17.6bn — near enough seventy per cent — sits in the September monthly alone. The dampening everyone is relying on is not a property of the market. It is a property of one expiry, and that expiry is the 18th.
Cannon’s daily levels and Cannon Edge snapshot for September 8th had not published at press time — this letter was closed early because of the holiday. The pivot table returns tomorrow.
The largest single revision on this board, and it went almost unremarked because it landed on a Saturday. Citi has pushed its first Fed cut from October–December 2026 out to June 2027, and now forecasts reductions in June, September and December of next year. The trigger was Friday: 162,000 jobs against roughly 56,000 expected. The detail that matters is that Citi is not forecasting a hot print — it looks for core CPI at +0.18% m/m, a rounding error under the two-tenths line three separate desks have named as the trigger. So the house that just deleted a year of easing is carrying a CPI call that argues against a hike this month. That is not a contradiction; it is the shape of the whole debate. The question stopped being about this print and became a question about where the funds rate settles.
Friday’s note read September as “a roughly 50-50 chance of a rate hike” — materially below the strip, which tells you the house is not chasing the pricing. But the number to take away is a European one: diesel is up 140% since the first of January, with global energy flows “severely restricted.” That is a bigger move than crude, in the product sitting inside every freight rate and therefore inside core goods, published by a fixed income desk rather than a commodities one. Set against it, his warning on volatility: realised vol has been historically low since early June and rate and currency implieds are “unusually low” into an “unusually wide range of outcomes.” Cheap options into a wide distribution is the same observation index skew has been making all month from the other direction.
The sharpest published dissent from the hike consensus, unhedged: hike odds are “on way to zero, in our view.” The reasoning is that Waller’s lean to hold is the tell, that “disinflation forces are underway, so there did not seem to be a rationale to hike,” and that a hawkish Chair “can later point to softer CPI as reasons to hold.” He keeps 8,000 and now sees 7,900–8,000 as the near-term path. Buried in the note is the most interesting inflation claim of the week from anyone: he puts flash memory at 33% of the excess inflation in core PCE. If that is right, a third of the overshoot the committee is preparing to tighten against is a semiconductor shortage — the same shortage that took the Kospi up 4.6% on Monday. Monetary policy does not build fabs.
Sunday’s work is the historical analogue nobody had put numbers on: “The Fed hasn’t hiked in more than three years and they could in two weeks. That first hike isn’t always a bad thing, but size appears to matter.” Across thirty-six years, after a first hike of 25 basis points the S&P was lower a month later 100% of the time and lower three months later 80% of the time, averaging −2.7% — but twelve months later it was higher 100% of the time, averaging +12.5%. The single instance since 1990 of a 50 basis point first move produced double-digit declines at three, six and twelve months. Two things follow: the near-term record argues equities have not finished pricing a hike they are 60% sure is coming, and the twelve-month record is why nobody with a year-end target has cut one.
| Desk | From → To | When | What did it |
|---|---|---|---|
| Citi Research Hollenhorst, Clark | First cut Oct–Dec 2026 → June 2027 | Sat Sep 5 | Friday’s 162k. A full year of easing deleted. Equity target 8,100 unchanged |
| Barclays house call | No change through 2026 → two hikes, Sep and Dec | Mon Aug 31 | Warsh’s Jackson Hole speech, read as notably hawkish. Takes fed funds to 4.00–4.25% — still the most hawkish major house on record |
| Apollo Torsten Slok | “Leaning toward hiking” → house call for September | Sat Sep 5 | Payrolls plus ISM services prices paid at 2021–22 levels. Still the only explicit house hike call from a US asset manager here |
| Morgan Stanley house Fed view | Hold, unchanged — reads the meeting near 50-50 | Sat Sep 5 | Warsh “preserved optionality, not a signal.” Core PCE around 3.1% after revisions. Below market pricing and comfortable there |
| JPMorgan Andrew Tyler, Market Intelligence | Bullish since June → tactically cautious | Mon Aug 31 | Crowded positioning, seasonality, momentum unwind, energy pressure. Note the split inside the building: JPM research still tells clients to buy dips |
| Wells Fargo Ohsung Kwon | Constructive → tactically cautious into September | Tue Sep 1 | AI-rally catalyst exhaustion and seasonality; longer-term view intact. No refreshed index target published |
| Goldman Sachs Jan Hatzius | Hold base case, unchanged | Fri Sep 4 | Friday “does remove an obstacle to a hike” but “doesn’t provide an affirmative case” |
| Fundstrat Tom Lee | 8,000 reaffirmed; near-term 7,900–8,000 | Fri Sep 4 | Hike odds “on way to zero.” The most constructive voice here and the most exposed to Friday |
| BofA Savita Subramanian | Street-low 7,100, unchanged | — | Nothing carried over the holiday. The September Sell Side Indicator has still not surfaced |
| Yardeni Research Ed Yardeni | 8,400, unchanged | — | His own stated wildcard is now the live event: oil that pushes inflation up and forces central banks to resume hiking |
The column that matters is the fourth, and what it does not contain. Citi moved a full year of Fed policy and left 8,100 alone. Barclays wrote in two hikes and nobody there touched 7,800. Fundstrat kept 8,000, Yardeni 8,400, BofA 7,100. Nine days, four genuine changes of view on the path of policy, zero changes to where these desks think the index finishes the year. Either the Street believes a quarter point does not matter to 2026 earnings, or nobody wants to be first to move a target three weeks before a meeting they cannot handicap. Both readings argue the same thing about the next fortnight: risk is being expressed in duration of conviction, not in levels.
| Pressure | Level | Direction of travel |
|---|---|---|
| Hormuz supply risk the dominant driver | Brent 97.00 | Escalating, and now measurable in hulls rather than headlines. The ten-day moving average of transits fell to 10 ships a day on Sunday, the lowest since May, against a pre-war 20 million barrels a day of flow. Two vessels transited Saturday, six Sunday; no VLCC has exited since Wednesday. Twenty-seven projectile-strike incidents since 6 July, and cumulative war-risk claims now exceed $2 billion. Deutsche Bank calls it “a tit-for-tat escalation targeting commercial shipping” — a risk object with no natural ceiling and no negotiating table |
| Iran’s response the mechanism, not the rhetoric | — | Read the detail: the exclusion zone running from the US blockade line through the Strait into the Gulf is enforced by a sanctions list, not by force — any identified ship transiting “will be placed on our sanctions list.” That is an insurance and chartering event, not a closure. It raises the cost of every voyage without giving anyone a casus belli. Tehran doubled non-subsidised petrol to 10,000 tomans a litre from today |
| Distillate the transmission channel | ULSD 4.66, +2.70% | Tightening faster than crude. Heating oil is up 102% year on year; the diesel crack set a record $108.02 last week; East Coast distillate stocks hit a record-low 19.3 million barrels; national stocks sit 14% under the five-year average and refineries have run above 95% utilisation for twelve straight weeks, the longest since 2000. Retail diesel is $5.82 a gallon |
| OPEC+ supply response Sunday’s meeting | 31.01 mb/d | Neutral, and that is the point. The eight-country voluntary group held October production at September levels — Saudi 10.478, Russia 9.949. No new barrels are coming from the cartel side and the next meeting is 4 October. The US strategic reserve is below 290 million barrels, the lowest since 1982 |
| The long end supply week | 30Y 5.244% | $119 billion of coupon supply lands today through Thursday into a two-year and five-year both at 52-week highs. Treasury’s long-end buyback operations double to at least $4 billion each from Wednesday — the offset that matters more than the auction sizes |
| Bank of Japan Sep 18 | JGB 10Y 2.92% | Tightening, and the yen is doing the talking. Dollar-yen fell another 1.2% Monday to 154.35, a second consecutive session as the strongest major, with roughly 63% priced for a hike on the 18th — the same morning as the quarterly expiry |
| AI capex funding watched, not stressed | HY OAS 2.65% | Investment grade 0.81%, both marginally off the tights. Oracle reports Thursday with a $638 billion backlog against negative free cash flow and net debt at 4.5 times EBITDA, with a further $40 billion of debt and equity planned. That is where this stops being a story about multiples |
The de-escalation option keeps getting cheaper and nobody is buying it. Citi’s base case is still that the Strait reopens in the fourth quarter through renewed dealmaking, and it puts a number on what that would mean: a surplus of three to four million barrels a day, against 2.3 to 2.4 billion barrels of Gulf supply removed from 2026 so far. Prediction markets are nowhere near agreeing. Polymarket has traffic normal by year-end at 26%, “no return in 2026” at 77%, and Iran charging transit fees by December at 28%. The largest book of the lot, $73 million, is on whether Kharg Island leaves Iranian control — and it trades at 7%. The market is pricing a long war fought at sea: not an invasion, not a settlement. That is the pricing to argue with if you want an edge, because it is the one everyone has quietly agreed on.
The cleanest gap at 9:30 is a biotech one, and it printed where Americans could not trade it. Novartis’s pelacarsen missed its primary endpoint in the Lp(a)HORIZON trial: it lowered lipoprotein(a) substantially but produced no reduction in major adverse cardiac events against placebo, across more than eight thousand patients. Novartis fell as much as 3.6% in Europe. The read-across did the real damage — Ionis fell 10% and Amgen 5% in European trading, and William Blair put it plainly: the result erases a peak US sales forecast of roughly $6 billion and “suggests meaningful risk in other ongoing trials.” Eli Lilly’s muvalaplin and Silence Therapeutics sit in the same read-across. Those gaps are live and unhedged into the reopening, and healthcare was the worst sector in Europe on Monday.
The memory trade stopped being about Micron some time last week. Korean DRAM export prices went from $16.76 in May to $22.90 in July, up 37%, while export volume fell 13% and export value rose 19%. Falling units with rising value is not the top of a cycle; it is a shortage deepening. HBM3E trades on the spot market at roughly $2,100 a chip against contract prices near $370 to $510 — four to five times. That is the fuel behind Monday’s Asian session: SK hynix +8.26%, Samsung +5.68%, Kioxia +7.93%, SoftBank +6.82%, and a Kospi at its highest since 23 July. Goldman’s Timothy Moe carries a 12,000 Kospi target on the argument that the market underestimates how long the earnings cycle runs.
Which brings the squeeze to Wednesday afternoon. TrendForce has the bill of materials on a 256GB iPhone 18 Pro up roughly 38% year on year, driven primarily by memory, with the Pro line facing 10 to 20% price-hike pressure and the foldable expected to start between $2,099 and $2,299. Apple’s first launch under John Ternus is therefore also the first read on whether the most price-insensitive consumer franchise in the world can pass through a component shock.
Refining is where the physical and equity markets meet. The diesel crack hit a record $108.02 intraday last week with East Coast distillate at a record low, and the four large US refiners cleared roughly $12.6 billion of combined profit in the second quarter. The mirror trade is freight: carriers who cannot recover fuel cost inside contracted rates. The tanker complex is the third leg — a sanctions-list regime in the Strait is a war-risk and protection-and-indemnity event before it is a rate event, with Gulf-to-China crude freight already near four times its five-year average.
Carrying in from Friday, unresolved by a holiday: Lululemon fell about 18%, Guidewire 22%, Fair Isaac 17.5%; Astera Labs added 10% and SanDisk 7%. Four sessions of new lows beating new highs on both exchanges, 46.9% of the S&P above its fifty-day average against an index sitting 1.8% above its own, and a fifty-day breadth reading that fell 7.8% on Friday alone. Equal weight is still ahead of cap weight year to date, +15.3% against +13.5% — the healthiest number on this page and the reason the deterioration still reads as rotation rather than distribution. One mechanical fact to diarise: the buyback blackout begins to accelerate around 12 September, and two-thirds of this year’s largest authorisations sit outside technology, so the bid that thins is the one under the industrials and financials, not the one under the Nasdaq.
The strip is more precise than the commentary around it. CME FedWatch prices a 25 basis point hike on 16 September at 60.4%, a hold at 39.6% and an ease at zero, off a September fed funds contract with 241,034 in open interest. That is up from 59.4% Friday, down from 65.4% a week ago and up from 44.4% a month ago — a market that spiked on Warsh, unwound on Waller, repriced on payrolls and has since gone quiet. There is no cut anywhere in the distribution.
What is unusual is the spread between venues. While the futures strip says 60%, Polymarket has “no change” at 52% against a hike at 49% on $100.7 million of volume, and Kalshi has hold at 48.5%. A ten-point gap between the rate market and the prediction markets on a single binary, two weeks out, is not noise. It is either a hedging premium embedded in the futures or a real difference of opinion between people who trade rates and people who trade outcomes. Whichever it is, one of them pays.
The Fed went dark on Saturday and cannot speak again until 17 September, the day after the decision. The last words on the record are a hawkish Chair — Warsh’s bar of confidence that inflation moves to target “clearly and at sufficient speed” — against Governor Waller saying on 3 September he would be “inclined to support” holding if the inflation improvement continues, though he left the door open if that improvement “has been fleeting.” John Williams: “I think we have to wait and see.” Three officials dissented for a hike in July; nine voted to hold.
Line the houses up and the disagreement is not about the data, it is about what the data obliges. Barclays wants two hikes this year. Apollo wants one this month. Morgan Stanley reads it 50-50 and forecasts a hold. Goldman holds. Citi just deleted a year of cuts and still expects core CPI at +0.18%. Fundstrat says the odds go to zero. Six desks, one data set, a range running from “zero” to “two before Christmas.” Two thresholds carry into Friday: the Cleveland Fed nowcast has August core CPI at +0.20% and 2.38% year on year, below consensus and exactly on the two-tenths line three desks independently named as the trigger; and Thursday’s PPI consensus carries a headline jump from 4.7% to 5.3%. The producer print lands first, is less watched, and sees freight and distillate before the consumer basket does. If the week breaks hawkish, it probably breaks on Thursday.
Max pain is treated as a single number for a single day. It is a curve, and this one has a slope. The strike that maximises option-holder losses is 7,720 for this morning’s expiry, 7,700 Wednesday, 7,725 Thursday and Friday, 7,700 on the 14th, 7,675 on the 15th — and 7,650 on the 16th, the day of the decision. Then it snaps back to 7,705 for the quarterly on the 18th. A seventy-point downhill walk from here to the FOMC, and an immediate recovery afterwards.
Nobody reads it that way because max pain is reported as a daily curiosity rather than a term structure. The shape says something specific: contracts written for the decision date were sold around a lower strike than contracts written for either side of it. Set that against Section 06, where four desks changed their view of Fed policy in nine days and not one moved a year-end target. The people writing options for the sixteenth marked their book down. The people writing research marked nothing. One of those groups has money at risk on the specific day.
“Long gamma, dealers dampen” has been the reassuring sentence in every letter including this one for six of the last seven sessions. It is true. What is never said is where the gamma lives. Of roughly $25.5 billion of net dealer gamma across the four nearby contracts, about $17.6 billion — close to seventy per cent — sits in the September monthly alone. The two front weeklies contribute a few billion each. October is a rounding error.
So the dampening is not a property of the market’s structure. It is a property of one expiry, which dies on the morning of 18 September alongside $6.2 trillion of notional, on the same day the Bank of Japan meets with a hike 63% priced, two days after the FOMC. The market spends eight sessions inside a cushion that vanishes on the ninth. Every scenario map being drawn for this meeting quietly assumes the volatility-suppressing mechanism survives it. It survives it by two days.
Fundstrat put a number on it that nobody has picked up: flash memory accounts for roughly 33% of the excess inflation in core PCE. Set that beside the physical market. Korean DRAM export prices are up 37% since May while volumes fell 13%. HBM3E trades at four to five times its contract price. The bill of materials on the iPhone launching tomorrow is up 38% year on year, driven primarily by memory. And on Monday the Kospi rose 4.61% while Hong Kong fell 0.85% — same session, same news, opposite directions, because Korea was trading the shortage and Hong Kong was trading the Fed.
If a meaningful slice of the core overshoot is a supply shock in one component, a policy rate does nothing to it except squeeze the demand that is not causing it. That is the strongest version of the hold case and almost nobody is making it — the doves are arguing about labour slack and tariff pass-through instead. It also implies something uncomfortable for the other side: if the committee hikes into a memory shortage and the shortage resolves on its own timetable, the tightening will look, in hindsight, like it worked. Both camps have an incentive not to examine this, which is usually the sign that it is worth examining.
This publication is provided by Cannon Trading Company for informational and educational purposes only. Content may include market commentary, technical observations, analyst opinions, and aggregated material derived from publicly available sources. While such information is believed to be reliable, Cannon Trading Company does not author, independently verify, endorse, or guarantee the accuracy, completeness, or timeliness of any third-party information referenced or summarized herein.
The information, opinions, market data, and commentary contained in this publication are subject to change at any time without notice and do not constitute investment advice, a solicitation, or a recommendation to buy or sell any security, futures contract, option on futures, foreign currency transaction, or any other financial instrument.
Past performance is not indicative of future results.
Trading Futures, Options on Futures, retail off-exchange foreign currency transactions, and other derivatives involves substantial risk of loss and is not suitable for all investors. You may lose all or more than your initial investment. Carefully consider whether trading is appropriate for you in light of your experience, objectives, financial resources, and other relevant circumstances.
Cannon Trading Company does not guarantee any profits and makes no representation that the strategies, ideas, analyses, or information presented will result in profitable trades or avoid losses. Any market views, analyst calls, forecasts, or third-party commentary referenced reflect the opinions of their respective authors and may or may not align with the views of Cannon Trading Company.
HYPOTHETICAL PERFORMANCE RESULTS HAVE MANY INHERENT LIMITATIONS, SOME OF WHICH ARE DESCRIBED BELOW. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFITS OR LOSSES SIMILAR TO THOSE SHOWN. IN FACT, THERE ARE FREQUENTLY SHARP DIFFERENCES BETWEEN HYPOTHETICAL PERFORMANCE RESULTS AND THE ACTUAL RESULTS SUBSEQUENTLY ACHIEVED BY ANY PARTICULAR TRADING PROGRAM.
ONE OF THE LIMITATIONS OF HYPOTHETICAL PERFORMANCE RESULTS IS THAT THEY ARE GENERALLY PREPARED WITH THE BENEFIT OF HINDSIGHT. IN ADDITION, HYPOTHETICAL TRADING DOES NO INVOLVE FINACIAL RISK, AND NO HYPOTHETICAL TRADING RECORD CAN COMPLETLEY ACCOUNT FOR THE IMPACT OF FINANCIAL RISK IN ACTUAL TRADING. FOR EXAMPLE, THE ABILITY TO WITHSTAND LOSSES OR TO ADHERE TO A PARTICULAR TRADING PROGRAM IN SPITE OF TRADING LOSSES ARE MATERIAL POINTS WHICH CAN ALSO ADVERSELY AFFECT ACTUAL TRADING RESULTS. THERE ARE NUMEROUS OTHER FACTORS RELATED TO THE MARKETS IN GENERAL OR TO THE IMPLEMENTATION OF ANY SPECIFIC TRADING PROGRAM WHICH CANNOT BE FULLY ACCOUNTED FOR IN THE PREPARATION OF HYPOTHETICAL PERFORMANCE RESULTS AND ALL OF WHICH CAN ADVERSELY AFFECT ACTUAL TRADING RESULTS.
Cannon Trading Company is registered solely as a commodities broker. Nothing contained herein constitutes the provision of investment advisory services.
© 2026 Cannon Trading Company, Inc. All rights reserved.
cannontrading.com • 1-800-454-9572 • (310) 859-9572