U.S. Stock, Fixed Income & Cross-Asset Closing Daily Wednesday, September 9, 2026 · Full Market Close Report Levels are the 9 September session unless stated. Companion file: US_CrossAsset_Daily_2026-09-09_DataNotes.txt |
|
| The tape in one paragraph. Two things the market had been waiting on arrived on Wednesday and both landed badly, and neither of them was a data release. No release rated "Very high" landed in the past twelve hours — Wednesday carried no scheduled U.S. macroeconomic release at all — and the next twenty-four hours carry none either, with Thursday's PPI and claims at 08:30 rated High and CPI on 11 September the first Very-high print. What the calendar did carry was Treasury's expanded buyback, and Treasury announced that Thursday's operation would be up to $6bn concentrated in the 10-to-20-year sector, not the 20-to-30-year bucket the long end had been positioned for. Ninety minutes earlier the 10-year auction had stopped at 4.834% against 4.683% at the prior auction. The curve cheapened at every coupon point: the 3-year rose 5 bp to 4.49%, the 5-year 4 bp to 4.61% and the 10-year 3 bp to 4.83%, with the intraday high at 4.8528%, the cheapest since November 2023. The equity damage was breadth rather than level. The S&P 500 fell only 0.48% to 7,636.49, but 397 of 494 captured components closed lower against 97 higher — a 4.09-to-1 ratio, the widest of the reporting window and up from 2.63-to-1 on Tuesday — because Meta rose 6.55% to $653.69 on the launch of Muse, its autonomous artificial-intelligence agent, and carried roughly a fifth of a point of index on its own. Alphabet went the other way at -2.28%. The Dow fell 0.77% to 52,380.66 and the Russell 2000 1.28%, so small caps were again the worst of the majors. The second event was the barrel. Brent settled $101.75, up 3.91% and above $100 for the first time since May, and WTI $97.15, up 4.43%, after CENTCOM destroyed five Iranian crude carriers and Iran struck a U.S. base in Jordan and Saudi energy infrastructure. Energy was the only Finviz group to close green, at +0.95%. Three tells worth carrying. Apple closed -0.28% at $315.34 after its first product event under new chief executive John Ternus, having traded $319.15 to $309.90 — the most expensive iPhone line ever announced, and a 2.98% intraday range that resolved into nothing. Comcast fell 6.65% and Charter 8.13% on a chief-financial-officer comment about persistent broadband losses and "irrational" fibre pricing, which is a bigger single-day cable repricing than any earnings report this year produced. And the distillate crack widened $5.73 to $104.55 while the gasoline crack fell $5.34 to $38.24 on the same 4.43% crude rally — the two legs moved eleven dollars apart in one session. |
|
| Index / Instrument | Close | Chg | % | Note | | S&P 500 | 7,636.49 | -37.03 | -0.48% | Breadth 4.09-to-1 negative | | Dow Jones Industrial Average | 52,380.66 | -405.41 | -0.77% | Range 52,314.61-52,707.90 | | Nasdaq Composite | 26,253.34 | -168.07 | -0.64% | Range 26,184.21-26,366.56 | | Nasdaq 100 | 29,421.55 | -86.15 | -0.29% | Second consecutive down session | | Russell 2000 | 2,922.35 | -37.85 | -1.28% | Worst major for a second day | | SOX (Philadelphia Semiconductor) | 11,931.3 | +43.5 | +0.37% | Third straight lead; high 12,016.1 | | VIX | 16.46 | +0.74 | +4.71% | Range 15.57-16.68 | | UST 2-year | 4.43% | +4 bp | — | Front end led the cheapening | | UST 3-year | 4.49% | +5 bp | — | Largest coupon move; Tuesday's hump gone | | UST 5-year | 4.61% | +4 bp | — | Belly cheapest point on the week | | UST 10-year | 4.83% | +3 bp | — | Intraday 4.8528%, cheapest since Nov 2023 | | UST 30-year | 5.28% | +3 bp | — | 20s30s back to flat at 0 bp | | UST 3-month bill | 3.95% | +1 bp | — | Bills lagged the coupons | | Brent (Nov, ICE) | $101.75 | +$3.83 | +3.91% | Above $100 first time since May | | WTI (Oct, NYMEX) | $97.15 | +$4.12 | +4.43% | Brent-WTI $4.60 | | Heating oil (Oct) | $4.8023 | +$0.2345 | +5.13% | Best move on the board | | Gold (Comex Dec) | $4,447.35 | +$8.35 | +0.19% | Range $4,384.86-$4,478.70 | | Silver (Comex Dec) | $67.925 | +$0.925 | +1.38% | Gold-silver ratio 65.47 | | DXY | 98.762 | — | -0.06% | Week -0.83%; +0.45% year to date |
|
| 2 · Market Hot Spots (ranked by tradability) |
|
| 1. | Treasury told the long end where the bid would be, and it was not where the long end was standing. Treasury announced that Thursday's expanded liquidity-support operation will buy up to $6bn in the 10-to-20-year sector. The doubling itself was confirmed a week ago — from a $2bn maximum to at least $4bn per operation through the 4 November refunding — but the bucket was not, and the market had positioned for support at the very long end. Yields rose across every coupon point on the announcement, with the 10-year touching 4.8528%, its cheapest since November 2023, and the 30-year cheapening 3 bp to 5.28%. The mechanism is worth naming precisely: this was not a repricing of the policy path, because the September meeting probability moved +0.8 of a point and the 2-year cheapened in line with everything else. It was a supply-and-sponsorship event. | | 2. | The 10-year auction did the damage first. The 13:00 ET auction stopped at 4.834% against 4.683% at the prior auction — 15.1 basis points of concession in a month — and the secondary market took the print as the tell. The sequence matters for anyone marking the day: the auction result preceded the buyback announcement, so the afternoon's cheapening had two independent causes stacked in ninety minutes, and neither of them was macroeconomic. | | 3. | Meta bought the index a fifth of a point and hid the worst breadth of the window. META +6.55% to $653.69 on 34.51m shares, on the launch of Muse, an autonomous agent that shops, books travel and manages calendars on tiered subscriptions. Morgan Stanley framed the addressable opportunity in the trillions and the market read it as the first monetisation path for Meta's capital expenditure. The second-order tell is the peer set: Alphabet fell 2.28% and Snap 2.21% on the same news, so this was share-shift pricing rather than a sector re-rating. Strip Meta out and the S&P 500's 0.48% decline becomes roughly 0.66%, against 397 decliners and 97 advancers. | | 4. | Broadband repriced on a sentence, not a filing. Comcast's chief financial officer told an investor conference that broadband subscriber losses will persist and that fibre competitors' pricing is "irrational". Comcast fell 6.65% to $24.58 on 41.01m shares and Charter 8.13% to $133.89, the worst S&P 500 performer. T-Mobile fell about 3% in sympathy on the other side of the same trade. The reason to rank this high is that no number changed: the second-quarter figures — Comcast domestic broadband revenue down 5.5% year on year, Charter internet revenue down 3.2% — were already public. What changed is the admission that the trajectory is structural, and Paramount Skydance -5.37% and Omnicom -3.49% show the read-across running into the wider media complex. | | 5. | Brent through $100 with the equity refusing to keep up. Brent settled $101.75, +3.91%, its first close above $100 since May, and WTI $97.15, +4.43%, after CENTCOM destroyed five Iranian crude carriers on 8 September in reply to a ballistic-missile attack on a U.S. warship, with Washington signalling one tanker destroyed per subsequent Iranian strike. Iran hit a U.S. base in Jordan and Saudi energy infrastructure. Yet energy equities rose only 0.95% against a 4%-plus barrel, and Exxon +2.22% and Chevron +1.91% were the megacap contributors. A 4.43% crude session that buys 0.95% of sector is the fade to watch: either the equity does not believe the level holds, or the market is pricing the demand destruction alongside the supply shock. | | 6. | The distillate crack is now the cleanest expression of the war. On an October basis against $97.15 WTI, heating oil rose 5.13% to $4.8023 while gasoline fell 0.89% to $3.2236. The distillate crack widened $5.73 to $104.55 and the gasoline crack fell $5.34 to $38.24, taking the differential $11.07 wider to $66.31 in a single session. Physical corroboration is in the retail numbers: diesel at $5.94 a gallon and gasoline at $4.22, both the highest September prints on record. This reverses the two-session collapse this report recorded a week ago and does it with force. | | 7. | Europe sold its own bond market harder than America sold its. Bunds cheapened 8.2 bp to 3.438%, OATs 10.8 bp to 4.332%, BTPs 10.3 bp to 4.286% and gilts 9.0 bp to 5.268%, against a 4.0 bp move in the U.S. 10-year on the same vendor board. Equities followed: CAC 40 -1.94%, DAX -1.66%, Euro Stoxx 50 -1.58%, IBEX -1.51%, SMI -1.80%. The proximate cause is Thursday's European Central Bank decision, where the market prices a move to a 2.65% refinancing rate from 2.40% and a deposit rate of 2.50% from 2.25%. A continent repricing its own central bank harder than the United States is repricing the Fed is not the configuration of the past six months. | | 8. | Small caps broke, and this time rates explain it. The Russell 2000 fell 1.28% to 2,922.35, its second consecutive session as the worst major and a 0.80-point underperformance of the S&P 500. Unlike Tuesday's healthcare-driven decline, Wednesday's has a clean transmission: floating-rate borrowers repricing against a curve that cheapened 3 to 5 bp at every coupon point, with industrials -1.84% and real estate -1.06% carrying the same signature. | | 9. | Apple's new chief executive launched the most expensive iPhone ever and the stock did nothing. John Ternus, in his first product event since succeeding Tim Cook, unveiled the foldable iPhone Duo at $1,999. AAPL closed -0.28% at $315.34 on 63.37m shares after trading as high as $319.15 and as low as $309.90 — a 2.98% intraday range that ended 1.20% below the high and 1.76% above the low. A new chief executive, a new form factor and a price increase produced twenty-eight basis points of decline. That is a market that has already paid for the cycle. | | 10. | The insurance and data cohort kept selling, and the credit bureaus split. Fair Isaac rose 5.35% to $983.19 — its first real bounce since the 16.68% bi-merge rout — while Equifax fell a further 2.06% to $168.09, a third consecutive decline. Aon -3.51%, Willis Towers Watson -3.57%, Fidelity National Information -3.42% and Fiserv -4.35% extended Tuesday's rout in the same names. The split inside the bureaus is the tradable part and it runs directly against the position in Section 12 item 6. |
|
| 3 · Sector Performance — September 9, 2026 |
|
| Sector | 1-Day | 1-Week | YTD | | Energy | +0.95% | +0.87% | +41.89% | | Technology | -0.05% | +2.50% | +25.60% | | Basic Materials | -0.19% | +1.45% | +19.82% | | Communication Services | -0.45% | +0.82% | -1.76% | | Healthcare | -0.51% | -2.87% | +6.87% | | Financial | -0.74% | +0.27% | +7.69% | | Consumer Defensive | -0.97% | -2.07% | +4.65% | | Real Estate | -1.06% | -1.48% | +6.84% | | Utilities | -1.25% | +0.76% | -0.44% | | Consumer Cyclical | -1.50% | -1.56% | -6.66% | | Industrials | -1.84% | +0.64% | +9.66% |
|
| Source: Finviz group screener, Performance table view (g=sector&v=140&o=name), read after the close. 1-Day is the Change % column, 1-Week Perf Week, YTD Perf YTD. Finviz classification, not GICS. |
| One group green out of eleven, and the leadership is the barrel. Energy's +0.95% is the only positive print and it extends the year-to-date lead to +41.89%, more than sixteen percentage points clear of technology. Underneath, the ranking is a rates ranking: industrials -1.84%, consumer cyclical -1.50%, utilities -1.25% and real estate -1.06% are the four worst, and they are the four most duration-sensitive groups on the board. Healthcare at -0.51% was the fifth-best group on the day after being the worst by four times on Tuesday, so the Lp(a) liquidation exhausted itself in a single session — though the -2.87% weekly figure is still the worst on the board. |
| The YTD reconciliation, and two groups have drifted. Compounding each group's 8 September YTD by Wednesday's one-day move reproduces the published YTD to within 0.03 percentage points for nine of the eleven groups. Worked examples: energy 1.4058 × 1.0095 = 1.41916, or +41.92% against a published +41.89%, deviation 0.03; healthcare 1.0741 × 0.9949 = 1.06862 → +6.86% against +6.87%, deviation 0.01; utilities 1.0082 × 0.9875 = 0.99560 → -0.44% against -0.44%, deviation zero. Two groups fail that test and they fail in opposite directions. Technology implies 1.2613 × 0.9995 = 1.26067, or +26.07%, against a published +25.60% — a 0.47 pp shortfall. Financial implies 1.0794 × 0.9926 = 1.07141, or +7.14%, against a published +7.69% — a 0.55 pp excess. Two roughly half-point deviations of opposite sign on the two largest groups, on a day when neither moved more than 0.74%, is not price drift; it is a constituent change inside the vendor's own group definitions. Both are flagged and carried forward for confirmation rather than smoothed. |
| The composition traps, and one of them inverts a three-day pattern. Technology at -0.05% overstates the weakness in semiconductors for a fourth consecutive session — SOX rose 0.37% with AMD +3.04%, Teradyne +3.20%, Micron +2.75% and Intel +1.69%, while the hardware and software legs split: Hewlett Packard Enterprise +5.12%, HP Inc +3.91%, IBM +3.38%, Akamai +4.87% and F5 +3.85% were among the day's best, against ServiceNow -2.31%, Salesforce -1.99% and Super Micro -3.30%. Communication services at -0.45% conceals the widest intra-group dispersion of the year: Meta +6.55% against Charter -8.13%, Comcast -6.65%, Paramount Skydance -5.37% and Alphabet -2.28%, a 14.68-point spread between the group's best and worst megacaps. Consumer cyclical at -1.50% is where the travel and housing complex sits: Booking -3.81%, Norwegian Cruise -3.51%, Tractor Supply -3.66%, Dollar Tree -4.56% and Ford -3.71% on 59.46m shares. |
| 4 · Movers & Single-Name Catalysts |
|
| Higher — one megacap, the barrel, and the hardware complex. |
- Meta Platforms (META) +6.55% to $653.69 on 34.51m shares, the best S&P 500 performer, on the launch of Muse, an autonomous artificial-intelligence agent sold on tiered subscriptions with a free entry point. Morgan Stanley's note framed the agent market as the first credible monetisation of Meta's capital expenditure. Range $638.56 to $657.86; the close was 0.64% below the high.
- Fair Isaac (FICO) +5.35% to $983.19, Hewlett Packard Enterprise +5.12% to $58.90, HCA +4.93%, Akamai +4.87%, HP Inc +3.91%, F5 Networks +3.85%, IBM +3.38% to $239.94, Teradyne +3.20%, AMD +3.04% to $521.10.
- APA Corp +2.96%, CF Industries +2.81%, Micron +2.75% to $1,027.77 on 23.12m shares, Archer-Daniels-Midland +2.41%, Exxon Mobil +2.22% to $164.23, Sysco +2.17%, Wells Fargo +1.94%, Chevron +1.91% to $213.81, Intel +1.69% to $106.24 on 94.07m shares — the heaviest volume on the board for a second session.
- Universal Health Services +2.26%, VeriSign +1.67%, DaVita +1.64%, Verisk +1.42% reversing part of Tuesday's 5.54% rout, Corning +1.51%, Qualcomm +1.33%, Newmont +1.27%, ConocoPhillips +1.16%, Texas Instruments +1.03%, Western Digital +1.04%.
- Intuitive Surgical +0.91%, AbbVie +0.86%, Marriott +0.85%, Becton Dickinson +0.39%, Zimmer Biomet +0.41% — the medical-device complex stabilised after Tuesday's liquidation, with only Baxter (-1.89%) still falling hard.
|
| Lower — cable, media, housing and everything with duration. |
- Charter Communications (CHTR) -8.13% to $133.89 on 4.02m shares, the worst S&P 500 performer, and Comcast (CMCSA) -6.65% to $24.58 on 41.01m shares. Comcast's chief financial officer told an investor conference that broadband subscriber losses will persist and described competitors' fibre pricing as "irrational". Second-quarter context already public: Comcast domestic broadband revenue -5.5% year on year, Charter internet revenue -3.2%, Comcast broadband average revenue per user -3.8% in the quarter.
- Cooper Companies (COO) -6.22% to $63.48 on 7.84m shares into its own after-close results — the only S&P 500 reporter of the session, and the market de-risked it in the cash session rather than waiting.
- Shopify -5.45%, Enphase -5.43%, Paramount Skydance -5.37%, International Paper -5.07%, First Solar -4.76%, Dollar Tree -4.56%, Fiserv -4.35%, PG&E -4.25% on 50.22m shares, Domino's Pizza -4.16%, Amentum -4.12%, Kimberly-Clark -4.10%, FMC -4.10%.
- Booking -3.81% to $173.43, Ford -3.71% to $13.48 on 59.46m shares, Blackstone -3.66%, Tractor Supply -3.66%, NRG -3.57%, Willis Towers Watson -3.57%, Aon -3.51%, Norwegian Cruise -3.51%, Omnicom -3.49%, Simon Property -3.47%, Fidelity National Information -3.42%, Super Micro -3.30%, KLA -3.21%.
- Alphabet A -2.28% to $330.65 and Alphabet C -2.09% on the Meta read-across, ServiceNow -2.31%, GE Vernova -2.09%, Seagate -2.04%, Salesforce -1.99%, UnitedHealth -1.94% on a $406.53-to-$378.08 range, Baxter -1.89%, Amazon -1.78%, Constellation Energy -1.72%, General Mills -1.59%, Lam Research -1.43%.
- Nvidia (NVDA) -0.91% to $223.67 on 75.45m shares, Broadcom -1.13%, Netflix -0.96%, Microsoft -0.47%, Apple -0.28% to $315.34, Tesla -0.10% — six of the seven largest weights fell, and only Meta did not.
- Equifax -2.06% to $168.09, a third consecutive decline, against Fair Isaac's +5.35%; Kroger -1.29% into Friday's results; Oracle -0.55% to $161.63 and Adobe -0.93% to $254.86 into Thursday evening's prints; Copart -1.75%.
|
| Analyst and corporate actions, 9 September. |
- Morgan Stanley's Muse note is the day's most consequential sell-side call by market impact — it framed the autonomous-agent opportunity as a multi-trillion-dollar market and Meta added 6.55%, roughly $40 a share.
- Comcast's chief-financial-officer remarks at an investor conference functioned as a guidance revision without being one, and the market treated the "irrational" pricing comment as the operative disclosure.
- Goldman Sachs raised its December 2026 Brent forecast to $85 from $80 and its 2027 forecast to $80 from $75, sketching above $120 if Gulf output stays materially below pre-war levels and the $60s if it recovers. The forecast is $16.75 below Wednesday's $101.75 settle, which is the single most striking number in the energy complex.
- UBS told clients to stay long oil pending evidence of diplomatic progress, with the note headline that the U.S.-Iran "off-ramp remains elusive"; Bannockburn Capital Markets described the outlook as a "strong and steady grind higher as the war enters seven months."
- Enbridge agreed to buy Tallgrass's crude oil business for $2.55bn; Dell Technologies is seeking a $4bn bond sale for refinancing, one of the few large investment-grade prints of a thin September calendar.
- Percentage upside is asserted only where a same-session closing price was independently captured. See Data Notes.
|
| 5 · S&P 500 Earnings Calendar — Current & Next Week (S&P 500 components only) |
|
| Sourcing, disclosed. The Earnings Whispers day pages remain behind a cookie-and-usage-agreement consent banner, which this unattended session did not accept. The rosters below are captured from the Nasdaq earnings calendar API for each date and screened name by name against an S&P 500 constituent list. Nasdaq publishes a before-open or after-close bucket rather than a clock time, so no clock times are asserted this session; confirm every time against company investor relations before trading a date. |
| Current week (Sep 7 - Sep 11) — remaining sessions |
|
| Thu 9/10. AMC: Oracle (ORCL), Adobe (ADBE), Copart (CPRT). |
| Fri 9/11. BMO: Kroger (KR). |
| Next week (Sep 14 - Sep 18) |
|
| Mon 9/14. No S&P 500 reporter on either bucket. |
| Tue 9/15. No S&P 500 reporter on either bucket. |
| Wed 9/16. AMC: Lennar (LEN). |
| Thu 9/17. No S&P 500 reporter on either bucket. |
| Fri 9/18. No S&P 500 reporter on either bucket. |
| Changes vs. the prior calendar (9/8 report): |
- No additions, no removals and no re-datings across either week. Every name and bucket in the 9/8 capture repeats exactly: Oracle, Adobe and Copart on 9/10 after the close, Kroger on 9/11 before the open, Lennar on 9/16 after the close. This is the first fully unchanged capture of the reporting window.
- Cooper Companies (COO) has dropped out of the forward calendar because it reported, after the close on 9/9. Under the forward-only rule the day is deleted rather than marked; the cash-session reaction that preceded the print is in Section 4.
- Oracle's 9/10 slot is now confirmed by a third consecutive agreeing capture, so the 9/8 date this report carried for four captures earlier in the window is fully retired.
- LEN.B appears on 9/16 alongside LEN and is deduped as a dual listing, unchanged from the prior capture.
- Four of next week's five sessions carry no S&P 500 reporter at all, unchanged, and it remains the emptiest forward week of the window.
- Non-members on the covered dates, listed so nobody mistakes their absence for an omission: Descartes (DSGX), Macy's (M), RH, National Beverage (FIZZ), Hub Group (HUBG), Zumiez (ZUMZ), Designer Brands (DBI), Lovesac (LOVE), 1-800-Flowers (FLWS), Tenneco (TEN) and IBEX on 9/10; KT Corp, Hooker Furnishings (HOFT), Rent the Runway (RENT), Children's Place (PLCE) and MoneyHero (MNY) on 9/11; Grifols (GRFS), Apartment Investment (AIV), Dave & Buster's (PLAY), Hain Celestial (HAIN) and Komatsu (KMTS) on 9/14; Trip.com (TCOM) and Vera Bradley (VRA) on 9/15; Manchester United (MANU) and Cracker Barrel (CBRL) on 9/16; American Battery (ABAT) and Innate Pharma (IPHA) on 9/17. Borderline membership cases are listed in Data Notes and conservatively excluded.
- What the forward calendar hands the desk. The shape is unchanged and the pressure on it is not. Thursday after the close still carries Oracle, Adobe and Copart into a single overnight window, and it now lands into a bond market that cheapened 3 to 5 bp at every coupon point on a supply event, into a European Central Bank decision fifteen minutes before the U.S. open, and into a barrel that has moved 5.00% across two sessions. The artificial-intelligence capital-expenditure disclosure that Oracle carries is the one number on the calendar that speaks directly to the day's biggest equity move, because Meta's 6.55% was a monetisation story rather than a spending story and Oracle reports the other side of that ledger. Kroger on Friday morning is the only consumer read and it now arrives with diesel at $5.94 and gasoline at $4.22 a gallon in its own cost base. After that the desk gets four blank sessions out of five, with Lennar on 9/16 the single reporter and the first day of the FOMC meeting. The reaction function to carry is the one Section 4 has been recording all week: guidance was punished 17.39% one session, bought back 2.56% the next, and de-risked 6.22% ahead of the print on this one.
|
| 6 · U.S. Treasury Yields — Official Par Curve |
|
| Source: U.S. Department of the Treasury Daily Treasury Par Yield Curve Rates, Text View for September 2026, read after publication on the evening of 9 September. Changes are versus the 8 September official row (1-day) and the 2 September official row (1-week). |
| Tenor | 9 Sep | 1-Day | 1-Week | | 1 Mo | 3.81% | 0 bp | -2 bp | | 3 Mo | 3.95% | +1 bp | +3 bp | | 1 Yr | 4.17% | +2 bp | +1 bp | | 2 Yr | 4.43% | +4 bp | +4 bp | | 3 Yr | 4.49% | +5 bp | +4 bp | | 5 Yr | 4.61% | +4 bp | +7 bp | | 7 Yr | 4.71% | +3 bp | +5 bp | | 10 Yr | 4.83% | +3 bp | +4 bp | | 20 Yr | 5.28% | +2 bp | +1 bp | | 30 Yr | 5.28% | +3 bp | +1 bp |
|
| Spread | 9 Sep | 1-Day | 1-Week | | 2s10s | +40 bp | -1 bp | 0 bp | | 3M10Y | +88 bp | +2 bp | +1 bp | | 2s30s | +85 bp | -1 bp | -3 bp | | 20s30s | 0 bp | +1 bp | 0 bp |
|
| The read: a front-led bear flattener driven by supply, and Tuesday's hump in the belly is gone exactly as this report said it would be. The 8 September row marked the three-year a basis point richer at 4.44% while WSJ's 5:00 p.m. board marked it a basis point cheaper at 4.463%; this report published the official row, named the conflict and wrote that if the vendor was right "the hump will be gone from Wednesday's official row." The three-year cheapened 5 bp to 4.49%, the largest move on the coupon curve, which restores the monotonic shape and confirms the 3:30 p.m. snapshot rather than the level. That is a resolved conflict, recorded rather than quietly dropped. |
| Every coupon point cheapened and the front cheapened most. The 2-year rose 4 bp to 4.43% and the 3-year 5 bp, against 3 bp at the 7-year, 10-year and 30-year and 2 bp at the 20-year. That is a bear flattener with the pivot at the belly, and the diagnostic is not policy: the September meeting probability rose 0.8 of a point and the 2027 strip cheapened by more than the front. What moved was paper. The 13:00 ET 10-year auction stopped at 4.834% against 4.683% at the prior auction, and Treasury then said Thursday's expanded buyback would take up to $6bn in the 10-to-20-year sector rather than the very long end. The intraday 10-year high of 4.8528% is the cheapest print since November 2023. |
| 2s10s flattened 1 bp to +40 bp, 2s30s flattened 1 bp to +85 bp and is 3 bp flatter on the week, and 3M10Y steepened 2 bp to +88 bp because the bill leg lagged. On a one-week view the shape is a clean belly-led cheapening: the 5-year is 7 bp cheaper and the 7-year 5 bp, against 1 bp at the 20-year and 30-year and 2 bp richer at the 1-month. A week in which the belly cheapens seven basis points while the long end cheapens one and the bills richen two is a week of supply and financing, not a week of term-premium repricing. |
| The 20s30s inversion is gone. Both the 20-year and the 30-year print 5.28%, so the spread is exactly 0 bp after two sessions at -1 bp and seven at zero before that. The 20-year cheapened 2 bp and the 30-year 3 bp, which is the 20-year outperforming into a buyback announcement that named the 10-to-20-year bucket. For the position in Section 12 item 2 that is a 1 bp gain and the first move in its favour; the invalidation at -3 bp is now three basis points away rather than two. |
| The vendor cross-check reconciles on direction at every point for the first time this window. WSJ's 5:00 p.m. bond board marks the 10-year at 4.844% (+5.0 bp), the 30-year at 5.293% (+4.5 bp), the 7-year at 4.731% (+5.8 bp), the 5-year at 4.624% (+5.4 bp), the 3-year at 4.527% (+4.3 bp), the 2-year at 4.446% (+4.2 bp) and the 1-month bill at 3.721% (-0.5 bp). Every sign agrees with the official row. The levels sit 1.3 to 3.7 basis points cheaper, widest at the three-year, which is the same ninety-minute gap between Treasury's 3:30 p.m. bid-side snapshot and the 5:00 p.m. quotes — and on a session that cheapened into the close, a cheaper 5:00 p.m. mark is what that gap should produce. Bloomberg marks the 10-year at 4.84%. |
| The bill curve lagged the coupons, which is the tell that this was not a policy session. The 1-month was unchanged at 3.81% and the 3-month rose only 1 bp to 3.95%, against 4 to 5 bp at the 2-year and 3-year, taking the 3-month-minus-1-month spread to 14 bp from 13 bp. Off the published table, the 4-month rose 4 bp to 4.06% — the largest bill move — the 2-month rose 2 bp to 3.93%, the 6-month rose 1 bp to 4.01% and the 1.5-month was unchanged at 3.88%. Two consequences. The four-month-above-six-month inversion widened to 5 bp from 2 bp, its widest of the window. And the 1.5-month-to-2-month gap re-widened to 5 bp from 3 bp after collapsing 9, 7, 3 across the previous three observations, so the anomaly this report called closing is not closed. All four tenors are off-table under the trimmed set and are cited here because they carry the financing story; the funding read sits with the money-market data in Section 9 block b. |
| 7 · U.S. Macroeconomic Calendar |
|
| Source: Federal Reserve Bank of New York Economic Indicators Calendar for September 2026 (all times Eastern), cross-checked against the Investing.com economic calendar read this session and the Wall Street Journal market-data calendar. Consensus figures for 10 September were independently re-captured this session; the 11 September figures are carried from the prior edition's Wall Street Journal capture and are flagged as such in Data Notes. |
| Current week (Sep 7 - Sep 11) — still to come |
| Date | Time ET | Release | Period | Consensus | Sensitivity | | Thu 9/10 | 08:30 | Producer Price Index (PPI) | Aug | +0.4% m/m, prior 0.0% | High | | Thu 9/10 | 08:30 | Core PPI | Aug | +0.3% m/m, prior +0.2% | High | | Thu 9/10 | 08:30 | Core PPI (year on year) | Aug | 4.6%, prior 4.2% | High | | Thu 9/10 | 08:30 | Initial Jobless Claims | wk ended 9/5 | 205K, prior 206K | High | | Thu 9/10 | 08:30 | Continuing Claims | wk ended 8/29 | 1,780K, prior 1,779K | Medium | | Thu 9/10 | 10:00 | NAR Existing Home Sales | Aug | 3.97m | Medium | | Thu 9/10 | 10:30 | EIA Natural Gas Storage | wk ended 9/4 | — | Low | | Thu 9/10 | 12:00 | EIA Petroleum Status Report | wk ended 9/4 | — | Medium | | Thu 9/10 | — | Treasury buyback, 10-to-20-year sector, up to $6bn | — | — | High | | Fri 9/11 | 08:30 | Consumer Price Index | Aug | +0.4% m/m | Very high | | Fri 9/11 | 10:00 | Michigan Consumer Survey (Preliminary) | Sep | 51.4 | Medium |
|
| Next week (Sep 14 - Sep 18) |
| Date | Time ET | Release | Period | Consensus | Sensitivity | | Mon 9/14 | — | No release on the calendar | — | — | — | | Tue 9/15 | 08:30 | Empire State Manufacturing Survey | Sep | — | Medium | | Wed 9/16 | 08:30 | Advance Retail Sales | Aug | No verified consensus published | High | | Wed 9/16 | 08:30 | Import & Export Prices | Aug | — | Medium | | Wed 9/16 | 10:00 | Business Inventories | Jul | — | Low | | Wed 9/16 | — | FOMC meeting begins (two days) | — | — | Very high | | Thu 9/17 | 08:30 | Initial Jobless Claims | wk ended 9/12 | — | High | | Thu 9/17 | 08:30 | Housing Starts & Building Permits | Aug | — | Medium | | Thu 9/17 | 10:00 | Philadelphia Fed Business Outlook | Sep | — | Medium | | Thu 9/17 | 14:00 | FOMC decision, projections and press conference | — | — | Very high | | Fri 9/18 | 09:15 | Industrial Production & Capacity Utilisation | Aug | — | Medium |
|
| The look-ahead: the energy shock has arrived in front of the inflation data rather than behind it, and that changes what a hot print means. The barrel has risen 5.00% across the two sessions since the holiday and 4.43% on Wednesday alone, with Brent above $100 for the first time since May and retail diesel at $5.94 and gasoline at $4.22 a gallon. PPI on 10 September at 08:30, consensus +0.4% month on month against a 0.0% prior, therefore measures August, before this week's move — which makes it the cleaner read on underlying pipeline pressure and the noisier guide to what September will look like. The core year-on-year consensus of 4.6% against a 4.2% prior is the number to watch, because a four-tenths acceleration in core would be difficult to attribute to energy at all. Initial claims the same morning at 205K against a four-week average of 207.25K remains the cleanest high-frequency check on the labour side; a print near consensus keeps the labour market out of the argument entirely. Fifteen minutes earlier, at 08:15 ET, the European Central Bank is priced to raise its refinancing rate to 2.65% from 2.40% — the first time this window a foreign central bank sits directly in front of an American inflation release, and European curves have already cheapened 8 to 11 basis points into it. CPI on 11 September at 08:30, consensus +0.4%, is the only Very-high release before the meeting and the print Governor Waller made the deciding input. Then retail sales on 16 September, the morning the FOMC convenes, and the decision at 14:00 on the 17th. The asymmetry has changed shape rather than size. A hike is the base case at 60.2% and the distribution barely moved on a session that cheapened the whole coupon curve, which says the market is treating this week's supply and energy news as real but not policy-relevant. That is a defensible view for one session and an expensive one for three: if CPI comes in at or above +0.4% with the barrel now above $97 and clear of the August survey period, the argument that the energy pass-through is still ahead of the data stops being a forecast and starts being a fact, and the front of the strip is the cheapest place that has not priced it. |
|
| 8 · Fed Funds Futures & Rate Path |
|
| Current target range: 3.50%-3.75%. A tailing auction, a disappointing buyback bucket and Brent through $100 moved the September meeting eight-tenths of a point. |
| CME FedWatch headline — 16 September 2026 meeting. Data as of 9 Sep 2026, 05:02:15 p.m. CT (6:02 p.m. ET), read from the FedWatch probability table. Contract ZQU6. |
| Target rate (bps) | NOW | 1 DAY (8 SEP 2026) | 1 WEEK (2 SEP 2026) | 1 MONTH (7 AUG 2026) | | 350-375 (current) | 39.8% | 40.6% | 36.8% | 55.6% | | 375-400 | 60.2% | 59.4% | 63.2% | 44.4% |
|
| Provenance of every column, stated — and for the first time this window the prior edition needs no correction. The footer timestamp reads 05:02:15 CT with no meridian; the page's own "Last Updated" line reads 09 Sep 2026 05:14:05 PM CT, twelve minutes later, so it resolves as p.m. and is an indicative read taken about an hour after the 4:00 p.m. CT ZQ session close rather than a settlement snapshot. The 1 DAY column carries the legend date 8 September and prints 59.4%, which is exactly what this report published from CME's live column on Tuesday evening — a zero-point correction, against +0.8 pp on each of the two preceding sessions. That matters beyond bookkeeping: the live-read bias this report has been carrying was consistent in sign and size for two sessions and has now vanished on the session when the live read happened to be clean, so it is a sampling artefact rather than a systematic bias, and it should not be adjusted for. 1 WEEK (2 September) at 63.2% and 1 MONTH (7 August) at 44.4% carry genuine reference dates and are used below. The Investing.com matrix underneath is timestamped 9 Sep 2026, 05:45 p.m. EDT and is the primary source for parts (a), (b) and (c). |
| The CME-versus-Investing.com gap, quantified, and it collapsed to the tightest of the window. CME puts the September hike at 60.2% at 6:02 p.m. ET; Investing.com at 60.1% at 5:45 p.m. ET — a 0.1 percentage-point difference across seventeen minutes, against 1.0 point on Tuesday and 0.2 on each of the two before that. Investing.com publishes the September future at 96.300 against 96.303 on Tuesday. Because the 16 September meeting sits mid-month, only about 47% of the contract's averaging period is affected, so one basis point of ZQ price is worth roughly ten percentage points of hike probability — which means the 0.3 of a basis point the front contract moved is the whole of the 1.7-point probability change, and the vendor gap is now smaller than a rounding step. |
| One-day, one-week and multi-day momentum, and the front finally moved. The September hike rose 0.8 of a point on CME's own columns, 60.2% from 59.4%, and 1.7 points on Investing.com's own columns, 60.1% from 58.4%. ZQU6 cheapened 0.3 bp to 96.300 and ZQZ6 1.5 bp to 96.045 — the first move in the front of the strip since 3 September, after three calendar days in which September and December 2026 were literally unchanged to the third decimal. The multi-day read: the meeting sat at 63.2% a week ago and 44.4% a month ago, so it is 3.0 points less hawkish than a week ago and 15.8 points more hawkish than a month ago, and the week's path is 63.2 down to 59.4 and back up to 60.2. Further out on the vendor's own columns, October's cumulative-above rose to 72.1% from 69.9% and December's to 87.6% from 86.2%. The probability of a cut at any 2026 meeting remains 0.0%. |
| (a) Current-year meeting distributions |
| Investing.com Fed Rate Monitor, updated 9 Sep 2026 05:45 p.m. EDT. Format: current [prior day] [prior week]. Modal range in bold. |
| Meeting | 3.50-3.75 (hold) | 3.75-4.00 (+25) | 4.00-4.25 (+50) | 4.25-4.50 (+75) | Cumulative above | Cumulative below | | Sep 16 | 39.9% [41.6] [39.6] | 60.1% [58.4] [60.4] | 0.0% | 0.0% | 60.1% | 0.0% | | Oct 28 | 28.0% [30.1] [28.7] | 54.1% [53.8] [54.7] | 18.0% [16.1] [16.7] | 0.0% | 72.1% | 0.0% | | Dec 9 | 12.4% [13.9] [11.5] | 39.5% [41.1] [39.1] | 38.1% [36.4] [39.4] | 10.0% [8.7] [10.0] | 87.6% | 0.0% |
|
| Sums are 100.0% at September and December and 100.1% at October. Three observations. First, the vendor's prior-day column reproduces the prior edition exactly for a third consecutive session — 41.6% and 58.4% at September — which is the cleanest run of the reporting window and removes the fixed-snapshot caveat that has been standing since late August. Second, the December mode is now within 1.4 points of losing: one hike leads at 39.5% against 38.1% for two hikes, where the gap was 4.7 points on Tuesday, 5.6 on Friday and 9.9 on Thursday, so the compression is unbroken across four sessions and has accelerated. Third, the tails were bought for a third session and harder — October's +50 bucket to 18.0% from 16.1% and December's +75 to 10.0% from 8.7%, together 3.2 points of tail mass added on top of Tuesday's 1.3 and Friday's 3.4. A distribution that adds three points of upper tail while the modal meeting moves eight-tenths of a point is a market that has stopped arguing about September and started hedging what comes after it. |
| (b) Next-year meeting path |
| Modal range, its probability, and the cumulative probability above and below the current 3.50%-3.75% range, with the contract price that draws it. |
| Meeting | Future price | Modal range | Prob. | Cumulative above | Cumulative below | | Jan 27, 2027 | 95.995 | 4.00-4.25 | 38.5% | 90.9% | 0.0% | | Mar 17, 2027 | 95.890 | 4.00-4.25 | 36.0% | 94.6% | 0.0% | | Apr 28, 2027 | 95.835 | 4.00-4.25 | 33.7% | 95.5% | 0.0% | | Jun 9, 2027 | 95.755 | 4.00-4.25 | 30.5% | 96.4% | 0.0% | | Jul 28, 2027 | 95.735 | 4.00-4.25 | 29.5% | 96.9% | 0.0% | | Sep 15, 2027 | 95.710 | 4.25-4.50 | 29.2% | 96.9% | 0.0% | | Oct 27, 2027 | 95.705 | 4.25-4.50 | 29.2% | 96.9% | 0.0% | | Dec 8, 2027 | 95.715 | 4.00-4.25 | 28.7% | 95.7% | 0.2% |
|
| The modal range moved up a bucket for the first time in this reporting window. At the September and October 2027 meetings the mode is now 4.25%-4.50% at 29.2%, edging 4.00%-4.25% at 28.7%; every prior edition of this table had all eight 2027 meetings modal at 4.00%-4.25%. The margin is four-tenths of a point and it will flip back on any soft print, so it is a threshold crossing rather than a regime change — but it is the third consecutive session of back-end cheapening and the largest. The eight contracts print 95.995, 95.890, 95.835, 95.755, 95.735, 95.710, 95.705, 95.715 against Tuesday's 96.010, 95.910, 95.855, 95.780, 95.760, 95.745, 95.740, 95.755 — 1.5 to 4.0 basis points cheaper, with the cheapening rising monotonically along the curve and roughly double Tuesday's amplitude. Set that against the front, where September 2026 cheapened 0.3 bp and December 2026 1.5 bp. So the market added to the destination roughly three times as fast as it added to the route, on a session whose news was a Treasury auction and a buyback bucket. Cumulative-above at December 2027 rose to 95.7% from 94.5% and the first non-trivial cut probability fell to 0.2% at 3.25-3.50% from the 0.4% published on Tuesday. |
| (c) Year-end probability ladders |
| Year-end 2026 — the 9 December meeting. |
| Outcome | Range | Probability | | -75 bp | 2.75-3.00 | 0.0% | | -50 bp | 3.00-3.25 | 0.0% | | -25 bp | 3.25-3.50 | 0.0% | | Hold | 3.50-3.75 | 12.4% | | +25 bp | 3.75-4.00 | 39.5% | | +50 bp | 4.00-4.25 | 38.1% | | +75 bp | 4.25-4.50 | 10.0% | | +100 bp and beyond | 4.50 and higher | 0.0% |
|
| Cumulative above the current range: 87.6%. Cumulative below: 0.0%. Sum: 100.0%. |
| Year-end 2027 — the 8 December meeting. |
| Outcome | Range | Probability | | -50 bp | 3.00-3.25 | 0.0% | | -25 bp | 3.25-3.50 | 0.2% | | Hold | 3.50-3.75 | 4.0% | | +25 bp | 3.75-4.00 | 15.8% | | +50 bp | 4.00-4.25 | 28.7% | | +75 bp | 4.25-4.50 | 28.2% | | +100 bp | 4.50-4.75 | 16.2% | | +125 bp | 4.75-5.00 | 5.6% | | +150 bp | 5.00-5.25 | 1.1% | | +175 bp | 5.25-5.50 | 0.1% | | +200 bp and beyond | 5.50 and higher | 0.0% |
|
| Cumulative above the current range: 95.7%. Cumulative below: 0.2%. Sum: 99.9%. |
| Rounding, transparently. Every figure is reproduced at the vendor's own one-decimal precision. Column sums of 99.9% or 100.1% are rounding artefacts of that precision, not missing probability mass; no cell has been rescaled, and cells shown as 0.0% are ranges the vendor's own card either omits entirely or publishes as zero, which under CME methodology means a probability below the rounding floor. The 2027 year-end ladder sums to 99.9% and the October 2026 row to 100.1% for exactly this reason; the September and December 2026 rows sum to exactly 100.0%. |
|
| (a) IG and HY credit spreads |
| ICE BofA option-adjusted spreads via FRED. FRED publishes with a one-business-day lag: the levels below carry the 8 September 2026 as-of date, not the 9 September close. Same-day direction is cross-checked against the cash-market proxies underneath and against Bloomberg and WSJ credit coverage. 1-Week is versus the 1 September row. |
| Series | FRED code | 8 Sep | 1-Day | 1-Week | YTD (from 2 Jan 2026) | | IG credit spread (ICE BofA US Corporate OAS) | BAMLC0A0CM | 81 bp | 0 bp | 0 bp | +2 bp (from 79) | | HY credit spread (ICE BofA US High Yield OAS) | BAMLH0A0HYM2 | 267 bp | -1 bp | +2 bp | -16 bp (from 283) | | CCC & lower credit spread | BAMLH0A3HYC | 1,056 bp | +1 bp | +7 bp | +168 bp (from 888) | | CDX IG 5y | — | Not retrievable this session | — | — | — | | CDX HY 5y | — | Not retrievable this session | — | — | — |
|
| CDX — the six-step ladder was worked again in the local Chrome browser, and four of the six steps could be executed. (1) Bloomberg in Chrome: /markets and /markets/rates-bonds both rendered fully, and a full-text scan of the rates-and-bonds page returns zero occurrences of the string and zero of "credit default"; the fixed-income tables carry Bloomberg's own aggregate indices, not credit-default-swap levels. (2) WSJ Market Data bonds page in Chrome: rendered with its Treasury, consumer-rate, government-bond and calendar tables populated, and a full-text scan returns zero occurrences. (3) ICE / S&P Global / Cbonds index pages: the dedicated CDX.NA.IG 5Y page that carried a subscription-masked quote a week ago still returns "no such page," and the index search sits behind the same wall. (4) FT Markets Data and (5) Barchart are both refused by the Chrome extension's domain policy ("Navigation to this domain is not allowed"), so those two steps could not be executed at all — that is a tooling failure rather than an absence of data, and it is recorded as such rather than reported as a clean failure. TradingView symbol searches for CDX continue to resolve to unrelated instruments and CME's credit-index product pages carry specifications rather than levels. (6) Cash-market proxies, labelled as proxies: HYG closed $78.98, -0.18%, and LQD $105.31, -0.16%. No CDX level is published here. |
| The cash market broke its silence and it broke it on yield, not spread. After four sessions in which HYG and LQD moved a cumulative 0.12% and 0.26%, LQD fell 0.16% to $105.31 and set a 52-week low at $105.08 intraday — the lowest the investment-grade cash proxy has traded in a year, on a session when the ICE BofA index spread was unchanged at 81 bp. That is the whole configuration in one observation: the spread series is flat and the price is at a one-year low, because what is repricing investment-grade credit is the 4.83% ten-year underneath it and not the compensation over it. HYG fell 0.18% to $78.98, a larger move than any of the previous four sessions, against a 52-week low of $78.57 that is still 0.52% away. On the index series, HY tightened 1 bp to 267 while CCC widened 1 bp to 1,056 on the 8 September stamp, taking the CCC-minus-HY differential to 789 bp from 787 — a fifth consecutive update in which the tail has done all of the work, and the differential is now 20 bp wider across six updates. IG unchanged at 81 bp for a third consecutive update. |
| (b) Money-market & funding plumbing |
| New York Fed reference rates, published at approximately 8:00 a.m. ET for the prior business day. The 4 September 2026 row published this session, so the table below advances one business day from the prior edition's 3 September basis; no 8 September row had published to the reference-rates endpoint at capture. The operations table beneath it is dated 9 September. Rate up = red. |
| Rate | 4 Sep | 1st pct | 25th pct | 75th pct | 99th pct | Volume | | SOFR | 3.65% | 3.60% | 3.64% | 3.70% | 3.73% | $2,888bn | | EFFR | 3.63% | 3.60% | 3.62% | 3.63% | 3.64% | $103bn | | OBFR | 3.63% | 3.50% | 3.62% | 3.63% | 3.68% | $225bn | | TGCR | 3.64% | 3.53% | 3.63% | 3.64% | 3.67% | $1,173bn | | BGCR | 3.64% | 3.53% | 3.63% | 3.64% | 3.68% | $1,197bn |
|
| Facility / balance | Latest (9 Sep) | Prior (8 Sep) | Note | | SOFR - IORB | 0 bp | +1 bp | IORB 3.65%; SOFR back to administered, 4 Sep basis | | Overnight reverse repo take-up | $432m | $626m | 9 Sep operation, 6 counterparties accepted | | Standing repo facility | $5m | $21m | 9 Sep operation; third consecutive non-zero | | Reserve balances (WRESBAL) | $2.8945tn | $2.8945tn | Week ended 2 Sep; no new print until 10 Sep |
|
| The plumbing eased at the rate and tightened in the facility, which is an unusual combination. SOFR fell a basis point to 3.65% on the 4 September row, exactly level with the 3.65% IORB, so the one-basis-point premium to the administered rate that has stood through the window has gone, and the 99th percentile fell to 3.73% from 3.74% on $2,888bn of volume, $61bn lighter. Against that, the standing repo facility took $5m on 9 September — smaller than Tuesday's $21m but non-zero for a third consecutive operation, which is the longest run of the window — and reverse repo take-up fell to $432m from $626m while the counterparty count rose to six from three. More counterparties taking less cash is a widening of participation at a shrinking size, which is what the front end looks like when the marginal lender is becoming the marginal borrower. No new reserve print landed; the H.4.1 for the week ended 9 September publishes on 10 September, and the standing figure is still $2.8945tn after a $30.4bn weekly drain, with quarter-end now fifteen days away. |
| The off-table bill tenors belong here, and this session they lagged rather than led. The 1-month was unchanged at 3.81% and the 1.5-month unchanged at 3.88%, while the 2-month rose 2 bp to 3.93%, the 3-month 1 bp to 3.95%, the 4-month 4 bp to 4.06% and the 6-month 1 bp to 4.01%. Two consequences, and both cut against Tuesday's reading. The four-month-above-six-month inversion widened to 5 bp from 2 bp, its widest of the window, which localises the cheapening in a single bill maturing just after the December meeting. And the 1.5-month-to-2-month gap re-widened to 5 bp from 3 bp after three observations of 9, 7 and 3 — so the anomaly this report called closing on Tuesday is not closing, and the correct read is that it oscillates with bill supply rather than trending. A bill curve that moves one basis point while the 2-year and 3-year move four and five is a curve saying the pressure is in coupon supply, not in funding. |
| (c) Rates volatility & swap spreads |
| Measure | Level | Change | Note | | MOVE index | 76.14 | Withheld | Vintage 8 September; no 9 September value published | | VIX | 16.46 | +4.71% | Range 15.57-16.68 | | MOVE / VIX | 4.63 | — | On a one-day-stale MOVE numerator; indicative |
|
| The MOVE vintage advanced and the level moved materially. The Investing.com series now carries an 08/09 date stamp at 76.14, with a day range of 73.10 to 76.14 — so the 8 September value that had not published when this report went out on Tuesday has since appeared, and it is 3.04 points, or 4.16%, above the 73.10 that carried a 4 September vintage. Rate volatility therefore rose 4.16% across the two sessions before Wednesday's supply event, which is the first genuine directional information this series has given in nine sessions and it corroborates the curve rather than the equity surface. What has not appeared is any 9 September value: the card's change field reads 0.00 and it fails its own internal consistency check for a tenth consecutive session, with a "previous close" of 95.74 sitting outside its own 73.10-76.14 day range. The level is published with its vintage; the change is withheld, and the MOVE-versus-VIX ratio is computed on a one-day-stale numerator and flagged. Alongside it, VIX rose 4.71% to 16.46 with a range of 15.57 to 16.68, on a session the index fell 0.48% and the coupon curve cheapened 3 to 5 bp — so for the first time in the window equity volatility and rate volatility are moving in the same direction, even if the rate leg is measured a day late. Swap spreads at the 2-year, 10-year and 30-year were not obtainable from a primary source this session and are not asserted; the substitute evidence is the 20s30s returning to exactly 0 bp in Section 6 on the session Treasury named the 10-to-20-year bucket. |
| (d) Issuance, leveraged loans & private credit |
| The primary market is still shut and the level just got worse. The constraint this report has named for a fortnight is the yield rather than the spread, and Wednesday moved the yield: the 10-year closed at 4.83% and traded 4.8528% intraday, the cheapest since November 2023, while LQD set a 52-week low. Post-Labor-Day investment-grade issuance is running at its weakest since 2020, after an August that set a record near $130-145bn against a post-2019 August average of roughly $95bn, and year-to-date supply through August above $1.68tn, up 27% on 2025, against a full-year projection of $2tn or more that the current pace makes a stretch. The concrete prints of the session were Dell Technologies seeking $4bn for refinancing and Enbridge agreeing to buy Tallgrass's crude oil business for $2.55bn, an acquisition that has to be funded somewhere. JPMorgan's 24 August syndicate view that the market can absorb the September stampede remains untested because the stampede has not started, and the window before the FOMC is now two data mornings and a Friday. On the loan side no updated Morningstar LSTA print was obtainable this session and none is asserted. The named private-credit watch item is unchanged: Broadcom's contingent residual-value guarantees to two artificial-intelligence laboratories are vendor financing that no published spread series captures, and Broadcom closed -1.13%. |
| The credit take. The divergence this section has been flagging finally produced a price, and it produced it in the wrong instrument. IG credit spreads were unchanged at 81 bp for a third consecutive update while LQD set a 52-week low at $105.08 — an investment-grade cash market at a one-year low with a spread series that has not moved a basis point in three sessions. That is not calm; it is a market in which the entire repricing is happening in the risk-free leg, and it is exactly what a 4.83% ten-year does to a 5%-yielding asset class. The tail continues its own separate story: CCC at 1,056 bp is 7 bp wider on the week and 168 bp wider on the year, and the CCC-minus-HY differential at 789 bp has widened on five of six updates. What would break the pattern is not a spread move but a deal: the deferred September calendar has to meet a 4.83% ten-year in two data mornings and a Friday, and the concession it clears at will be the first honest mark on where the bid actually is. The nearer risk is still the plumbing, and it got quieter at the rate and busier at the facility — SOFR back level with IORB, but the standing repo facility used on three consecutive operations and reverse repo participation broadening to six counterparties at a third less cash, with reserves down $30.4bn on the last print and quarter-end fifteen days out. Watch the CCC-minus-HY differential at 789 bp; a move through 800 while IG sits at 81 would say the tail has decoupled rather than lagged, and it is now eleven basis points away. |
|
|
| Levels from the TradingEconomics currency board read after the U.S. close. Nine major rows carry a live 06:15 clock stamp rather than a Sep/09 date stamp — the dollar index, the euro, the Australian and New Zealand dollars, the Swiss franc, the yuan, the peso, the krone and the rand — so for those the vendor's %Chg column measures from the 00:00 UTC boundary to the capture, not a settled session. Both bases are published: the Session column is the vendor's own change and the vs 8 Sep column is this report's own 24-hour calculation against the prior edition's published levels for the same vendor. Worked examples are in Data Notes. Corroborated against the Investing.com majors board. Quote basis: USD per unit for EUR, GBP, AUD and NZD; units per USD for JPY, CHF, CAD, KRW, TWD, CNY, MXN, SGD and INR. |
| Pair | Level | Session | vs 8 Sep | Week | YTD | Context | | DXY | 98.762 | -0.03%\* | -0.06% | -0.83% | +0.45% | Flat on a 3-to-5 bp curve cheapening | | USD/JPY | 153.507 | -0.30% | -0.14% | -3.28% | -2.07% | Yen held its seven-month high | | USD/CAD | 1.38027 | +0.14% | +0.14% | -0.29% | +0.60% | Weakened against a 4.43% barrel | | USD/CHF | 0.80983 | +0.05%\* | +0.06% | -0.38% | +2.13% | Franc offered on the day Swiss equities fell 1.80% | | EUR/USD | 1.16361 | +0.10%\* | +0.08% | +0.41% | -0.89% | Unmoved by an 8 bp Bund selloff | | GBP/USD | 1.35489 | +0.06% | +0.06% | +0.46% | +0.66% | Gilts 9 bp cheaper, sterling six basis points firmer | | AUD/USD | 0.72182 | +0.02%\* | 0.00% | +0.69% | +8.17% | Unchanged to five decimals for a second session | | NZD/USD | 0.58398 | -0.24%\* | -0.26% | -0.19% | +1.44% | The weakest major for a second session | | USD/KRW | 1339.49 | -0.05% | -0.07% | -1.39% | -7.02% | Won firm on the Kospi's 1.40% | | USD/TWD | 31.4820 | -0.12% | -0.12% | -0.86% | +0.43% | Third consecutive session of quiet strength | | USD/CNY | 6.70518 | -0.02%\* | -0.02% | -0.19% | -3.89% | The most inert cross on the board again | | USD/MXN | 16.8920 | -0.12%\* | -0.12% | -0.47% | -6.25% | Peso firm on the barrel | | USD/SGD | 1.26431 | -0.06% | -0.03% | -0.54% | -1.70% | Tracking the yuan | | USD/INR | 95.1110 | +0.28% | +0.32% | +0.69% | +5.83% | The only Asian currency to weaken; Nifty -0.86% |
|
| \*Rows marked with an asterisk carry the vendor's live 06:15 clock stamp rather than a Sep/09 date stamp. |
| The take: the dollar did not move on a session that repriced the whole American coupon curve, and that is the anomaly of the day. DXY at 98.762 is 0.06% below Tuesday's 98.822 and 0.83% lower on the week, on a session when the 2-year cheapened 4 bp, the 3-year 5 bp and the 10-year 3 bp. A four-to-five basis-point front-end move that produces six basis points of dollar index is a currency market that has decided the American rate move is about supply rather than policy — which is the same conclusion Section 6 reached from the curve shape and Section 8 reached from a distribution that added tail mass without moving its mode. Three markets reading one event the same way is unusual enough to be worth naming. |
| The euro is the sharpest version of it. EUR/USD rose 0.10% to 1.16361 on a session when Bunds cheapened 8.2 bp against the U.S. 10-year's 4.0 bp on the same vendor board — a four-basis-point relative move in the euro's favour, into a European Central Bank decision priced to deliver a 25 bp hike the following morning, and the currency produced ten basis points. Sterling was worse still: gilts cheapened 9.0 bp and GBP/USD rose 0.06%. The read is that European rate markets are pricing the same thing European currency markets are: a central bank raising into a weakening tape, which is a growth negative that offsets the carry positive almost exactly. |
| Asia bought its own currencies and India sold its. USD/KRW fell to 1339.49 on a session the Kospi rose 1.40%, so for once the won and the equity market agreed — the opposite of the configuration this report flagged in August, when the won weakened on the Kospi's best day. USD/TWD fell 0.12% to 31.4820 for a third consecutive session of quiet strength and USD/CNY was inert again at 6.70518, the most immobile cross on the board for a second edition. Against that, USD/INR rose 0.32% to 95.111, the only Asian currency to lose ground, on the day the Nifty 50 fell 0.86% and the Sensex 1.08% — India is now +5.83% year to date against the dollar and remains the weakest major Asian currency by a wide margin, which on a session driven by a 4.43% crude rally is exactly what the largest crude importer in the region should look like. |
|
| Settlement basis, stated, and reconciled to the prior edition. The Investing.com per-contract historical board remains the settle series of record, unchanged from the prior edition. The prior edition's 8 September figures were captured while those rows were still forming and have since finalised materially lower or higher; the finalised values are used as the base for every daily change below, and the full restatement is in Data Notes. Today's rows were likewise captured at approximately 18:35 ET and are corroborated against TradingEconomics spot and against Bloomberg's and WSJ's boards. WTI, RBOB, heating oil and natural gas are on the October contract; gold, silver and copper on December; Brent on the November contract, front since the 23 August rollover. No front-month roll occurred this session. Weekly and year-to-date columns are TradingEconomics spot returns on the vendor's Sep/09 stamp. |
| Contract | Settle | Chg | %Chg | Week | YTD | Driver | | Heating oil (Oct) | $4.8023 | +$0.2345 | +5.13% | +2.71% | +126.68%\* | Best move on the board; diesel at a record $5.94 retail | | WTI (Oct, NYMEX) | $97.15 | +$4.12 | +4.43% | +6.83% | +69.33%\* | Five Iranian tankers destroyed; +5.00% in two sessions | | Brent (Nov, ICE) | $101.75 | +$3.83 | +3.91% | +6.76% | +67.77%\* | First close above $100 since May | | Silver (Comex Dec) | $67.925 | +$0.925 | +1.38% | +3.00% | -5.59%\* | Outperformed gold for a fifth session | | Copper (Comex Dec) | $6.8583 | +$0.0348 | +0.51% | +4.20% | +19.22%\* | Third consecutive gain | | Gold (Comex Dec) | $4,447.35 | +$8.35 | +0.19% | +0.34% | +1.92%\* | Range $4,384.86-$4,478.70 | | Gasoline RBOB (Oct) | $3.2236 | -$0.0289 | -0.89% | +3.85% | +88.40%\* | Fell on a 4.43% crude session | | Natural gas (Oct) | $2.805 | -$0.111 | -3.81% | -5.14% | -23.92%\* | Worst move on the board for a second session |
|
| \*YTD figures marked with an asterisk are TradingEconomics spot year-to-date returns, not futures returns on the contracts quoted above. Weekly columns are TradingEconomics one-week changes on the same caveat. Mixing the two would be a basis error; they are presented in separate columns for exactly that reason. |
| The crack spreads did something they have not done in this window: they moved eleven dollars apart in one session. On a consistent October basis against $97.15 WTI, and with 8 September restated onto the finalised historical-board values: |
- Distillate crack: $4.8023 × 42 - $97.15 = $104.55, up $5.73 from a restated $98.82.
- Gasoline crack: $3.2236 × 42 - $97.15 = $38.24, down $5.34 from a restated $43.58.
- The differential widened $11.07 to $66.31 from $55.24.
|
| That is the largest single-session move in the differential this report has recorded, and it is the mirror image of the $9.59 two-session collapse of a week ago. The mechanism is clean and it is not refinery margin arithmetic: a supply shock that removes crude carriers removes middle distillate first, because diesel and jet are what the affected barrels are cracked into and because the physical market was already tight. Retail diesel at $5.94 a gallon and gasoline at $4.22 are both the highest September prints on record, and heating oil is +126.68% year to date on the spot basis against gasoline's +88.40% — a 38-point structural gap that has now widened again after narrowing for a fortnight. A gasoline crack that falls 89 basis points on a day crude rises 443 is the demand side of the same shock: end-of-driving-season gasoline cannot pass the barrel through, and distillate can. |
| Energy is where the whole session's conviction sat, and the equity did not follow it. Brent settled $101.75 and WTI $97.15, taking the Brent-WTI differential to $4.60 from $4.89 — a narrowing on a session when the marginal supply risk is seaborne, which is the wrong direction and worth watching. Two sessions have now added 5.00% to WTI and taken the weekly spot return to +6.83%, and energy equities have added 2.14% across the same two sessions. Against that, Goldman Sachs raised its December 2026 Brent forecast only to $85 from $80 and its 2027 forecast to $80 from $75 — $16.75 below Wednesday's settle — with an above-$120 scenario if Gulf output stays materially below pre-war levels and a $60s scenario if it recovers. A sell-side base case sixteen dollars under spot is the most explicit statement available that the street treats this level as a war premium rather than a price. |
| The metals refused to participate, and gold's range is the tell. Gold added $8.35 to $4,447.35 but traded $4,384.86 to $4,478.70 — a $93.84 range, 2.1% of the contract, to produce nineteen basis points of settle. On a session with a 4.43% crude rally, a 5 bp front-end cheapening and a flat dollar, gold ran to within $10 of Tuesday's high, failed, sold $94 and closed almost unchanged. That is the second consecutive session of failure at roughly $4,480, and it is now a level rather than a coincidence. Silver rose 1.38% to $67.925, outperforming for a fifth consecutive session, so the gold-silver ratio fell to 65.47 from 66.25, its lowest of the window. Copper added 0.51% to $6.8583 for a third straight gain. Silver and copper rising while gold stalls is an industrial bid inside a monetary complex that has stopped working, and it is the cleanest evidence available that the metals are not trading the same shock the barrel is. Natural gas fell 3.81% to $2.805, the worst move on the board for a second consecutive session and -23.92% on the year, still the only major commodity deeply negative — and it did that on the day crude rose 4.43%, which is a two-and-a-half-percentage-point-per-session divergence inside the same complex. |
|
| Desk-style ideas for institutional investors. Each carries an explicit expression, catalyst and invalidation. These are not personalized investment advice; verify independently and size to your own mandate before acting. |
| 1. The rates trade — long ZQZ6 against short ZQZ7 paid a fourth time and paid biggest; hold the quarter |
| Mark first. Long ZQZ6 (December 2026) against short ZQZ7 (December 2027), DV01-matched one-for-one at $41.67 per basis point per contract, entered on 3 September at 96.085 / 95.820 for a spread of 26.5 bp, quarter size. Wednesday's mark: ZQZ6 96.045, ZQZ7 95.715 — a spread of 33.0 bp. That is +2.5 bp on the session, worth +$104.17 per contract pair before costs on a quarter, and the position is +6.5 bp from entry, or +$270.83 per pair. |
| Why it paid, and the mechanism finally changed. For three sessions the spread widened purely because the back leg cheapened while the front sat frozen. On Wednesday both legs moved and the back moved 1.5 bp more: ZQZ6 cheapened 1.5 bp and ZQZ7 4.0 bp. That is a healthier version of the same trade, because a spread that only widens when one leg moves is a one-sided position dressed as a spread. It is also the first session since entry in which the front carried information at all — ZQU6 and ZQZ6 had been unchanged to the third decimal for three calendar days. |
| The modal path, base case and tails. Modal path: a 25 bp hike on 16 September at 60.2% on CME and 60.1% on Investing.com, with ease at 0.0%; one hike is modal at October (54.1%) and, by 1.4 points, at December (39.5% against 38.1%); and the 2027 strip has moved its mode up a bucket at the September and October 2027 meetings, to 4.25%-4.50%. Base case: one hike in September, a second priced by December at 87.6% cumulative-above, and a terminal rate that has now risen on three consecutive sessions while the schedule has moved eight-tenths of a point. Tail one, and it is still the whole trade: CPI on 11 September at 08:30, consensus +0.4%, now arriving with Brent above $100. A hot print takes September through 70% and cheapens the front leg, which costs this spread; the offset is that it should cheapen the back harder, and Wednesday is the first evidence that it does. Tail two: a soft CPI that reinstates a parallel richening, which is the configuration that narrows this spread and the only one that has hurt it. Practical implication: at roughly ten percentage points of probability per basis point of ZQ price for a mid-month meeting, and with the front now moving again, the position has stopped being a pure 2027 view and has become a genuine calendar spread — which argues for holding the quarter rather than adding, because both legs can now hurt. Catalyst: PPI and claims 9/10; the ECB at 08:15 ET on 9/10; CPI 9/11; the 16-17 September FOMC. Invalidation, unchanged: the spread through 22.0 bp; or the September cumulative hike back below 45% on either vendor; or the December 2026 mode reverting to two hikes — which is now 1.4 points away, from 4.7 on Tuesday, and is much the closest invalidation to firing. Sizing: a quarter, at $41.67 per basis point per pair. |
| 2. Long the 20-year against the 30-year — the buyback landed in the wrong bucket and the trade made money anyway; hold the half |
| Expression: long the 20-year bond against short the 30-year, DV01-neutral, half size. Mark: 20s30s at exactly 0 bp, with both the 20-year and 30-year at 5.28% — the 20-year 2 bp cheaper and the 30-year 3 bp cheaper, so +1 bp on the session and flat against entry after two sessions at -1 bp. The honest reading, because the thesis was half wrong. The entry note said the operation "stops being a thesis and becomes a printed result." It did, and the result was not the one the trade wanted: Treasury announced Thursday's operation at up to $6bn in the 10-to-20-year sector, not the 20-to-30-year bucket. The trade was a bet that the bid would spread across both buckets so the 20-year's greater cheapness would attract the marginal flow. The bid did not spread — it concentrated in the bucket the 20-year sits at the far end of, which is the more favourable of the two ways the trade could have been wrong, and the 20-year outperformed by a basis point on a session everything cheapened. That is a lucky right answer to a wrong question and it is recorded as such. Action: hold the half; do not add on the result, because a single announcement is not a run rate and the operation itself prints on Thursday. Catalyst: the 10 September operation and its actual accepted maturities; Treasury's 4 November quarterly refunding and the buyback guidance that comes with it. Invalidation, unchanged: 20s30s through -3 bp, now three basis points away rather than two, or an explicit Treasury statement ruling out changes to long-end auction sizes. Sizing: a half, unchanged. |
| 3. Protection on the CCC cohort funded in IG — paid a second consecutive time; hold the half |
| Expression: long CCC-exposed credit protection (or short a levered-loan / CCC-heavy vehicle) against long IG cash. Mark: CCC 1,056 bp, +1 bp; HY 267 bp, -1 bp; IG 81 bp, unchanged on the 8 September FRED update, taking the CCC-minus-HY differential to 789 bp from 787 — a 2 bp gain, the largest single-update gain of the position, for a cumulative +23 bp across six updates. The honest reading, and one leg of it now needs a caveat. The pair has been paid on five of six updates. But the funding leg is no longer inert: LQD set a 52-week low at $105.08 intraday and closed $105.31, -0.16%, on a session the IG spread series did not move. A long-IG-cash funding leg that loses money on duration while its spread is unchanged is a leg that is no longer free, and that is a genuine deterioration in the structure rather than in the thesis. The thesis itself is intact and sharper: the aggregates have been flat for three updates while the tail has widened 7 bp on the week. Action: hold the half; do not add, and consider expressing the funding leg in a duration-hedged form if the ten-year keeps cheapening. Catalyst: the September IG calendar, still at its weakest post-Labor-Day pace since 2020, which has to clear into a 4.83% ten-year; PPI 9/10 and CPI 9/11; quarter-end funding from the 15th. Invalidation, unchanged: the differential back through 750 bp, or IG widening beyond 90 bp. Sizing: a half, unchanged. |
| 4. Short the packaged-food cohort against long consumer defensive — closed on schedule at a small loss |
| Final mark.* The nine-name basket closed Campbell's +0.55%, Bunge -0.31%, McCormick -0.81%, Conagra -0.98%, Tyson -1.03%, JM Smucker -1.08%, Kraft Heinz -1.14%, General Mills -1.59%, Lamb Weston -2.57% for an average of -1.00%, against consumer defensive at -0.97% — a 0.03-point gain on the pair. The close. The three-session window written into the idea has expired and the position is marked out here rather than extended. Cumulative across the window: +0.57, -0.90, +0.03, for -0.30 points. The invalidation — the basket recovering more than half of the 3.77-point dispersion within three sessions — never fired: total recovery was 0.30 points against a 1.885-point trigger. So the trade was neither invalidated nor right; it simply did not work, and it cost thirty basis points of pair performance on a half. What it taught: the dispersion event that opened it was a single-session guidance shock and the cohort mean-reverted to its own group inside three days, which argues for shorter windows on dispersion trades built off one print. **Closed. No further marks. |
| 5. Long volatility on the semiconductor complex — the clock still cannot advance; keep the quarter |
| Mark. October volatility on the semiconductor proxy, a quarter, expressed in premium rather than delta. The clock, precisely. The invalidation is "SOX outperforming the Nasdaq 100 on two consecutive up sessions." Wednesday was not one, because the Nasdaq 100 fell 0.29% — SOX outperformed by 0.66 points but not on an up session, so for the second consecutive edition the counter cannot advance. This is now the mechanical problem with the invalidation rather than with the trade: the Nasdaq 100 has not had an up session since 4 September, so a condition that requires two consecutive up sessions has been unable to evaluate for three sessions running. That is stated plainly rather than treated as evidence for the position. The internals. SOX +0.37% with AMD +3.04%, Teradyne +3.20%, Micron +2.75% and Intel +1.69%, against Nvidia -0.91%, Broadcom -1.13%, KLA -3.21% and Super Micro -3.30%. Three consecutive sessions in which the complex has risen while its largest member fell, and the dispersion inside the index is now wider than the index's own move by a factor of three. Action: keep the quarter, do not add. Mark it out on the next genuine up session for the Nasdaq 100; if the condition cannot evaluate for two more sessions, close it on time decay rather than waiting for a test that the tape will not deliver. Catalyst: Oracle, Adobe and Copart on 10 September after the close (Section 5); CPI 9/11. Invalidation: as written, with the time-decay override above added explicitly. Sizing: a quarter. |
| 6. Short the credit-bureau complex against long the S&P 500 financials — the worst session of the position; halve it |
| Mark. Fair Isaac +5.35% to $983.19 and Equifax -2.06% to $168.09; TransUnion again did not appear in the 494-line component capture and no independently sourced close is asserted, so the basket is marked on two of three names at +1.65%, against S&P 500 financials at -0.74% — a 2.39-point loss on the pair, and cumulatively -2.13 points across two sessions. The honest reading, and it goes against the thesis. The argument was that the bi-merge risk sits with the bureaus that could lose the tie-break rather than with the scoring monopoly. Equifax kept falling, which supports it. Fair Isaac rose 5.35%, which does not — and FICO is the larger, more liquid leg. The stock has now recovered 5.46% of its 16.68% decline in two of the five sessions the invalidation allows; the trigger is half, or 8.34%, and it is 2.88 points away. That is close enough to act on before it fires. Action: halve the position to a quarter now rather than waiting for a mechanical invalidation on a basket that can only be marked on two of three legs. The residual quarter keeps the Equifax leg, which is doing the work. Catalyst: any FHFA follow-through on bi-merge; sell-side mortgage-channel revisions; the bureaus' own investor communications. Invalidation, unchanged for the residual: the basket recovering more than half of the 4 September decline within five sessions — three sessions remain — or an explicit FHFA statement retreating from bi-merge. Sizing: cut from a half to a quarter, dollar-neutral. The risk named honestly: this remains a crowded short after a 17% day, borrow is expensive, and the mark rests on two of three legs. |
| 7. New — long the October distillate crack against the gasoline crack, quarter size |
| Expression: long the October heating-oil crack against short the October gasoline crack, barrel-for-barrel, quarter size, at Wednesday's $104.55 / $38.24 for a differential of $66.31. The thesis. A supply shock that destroys crude carriers removes middle distillate first, and the physical market corroborates it: retail diesel at $5.94 a gallon and gasoline at $4.22, both record September prints, with heating oil +126.68% year to date on the spot basis against gasoline's +88.40%. Wednesday delivered the mechanism in a single session — heating oil +5.13% and gasoline -0.89% on a 4.43% crude rally — and the seasonal turn out of driving season is behind it rather than in front of it. The honest caveat, and it is why this is a quarter. This report closed exactly this pair on 3 September at a $61.48 differential and then watched it collapse and rebuild; re-entering $4.83 wider than the level it was closed at is buying back a trade at a worse price, and that is stated rather than presented as a fresh idea. The justification for re-entering is that the driver has changed from margin seasonality to physical supply. Catalyst: the EIA petroleum status report at 12:00 ET on 10 September, which will show whether distillate inventories confirm the paper move; any de-escalation headline out of the Gulf. Invalidation: the differential back through $55, which would undo the whole of Wednesday's move; or a crude session above 2% in which heating oil underperforms gasoline, which would break the mechanism. Sizing: a quarter, barrel-for-barrel. |
| Prior closes, marked forward. The AI-halo basket against long Nvidia, closed on 1 September, would have lost again: Palo Alto -0.56%, ServiceNow -2.31%, Fortinet -0.16% and Adobe -0.93% for a four-name average of -0.99% against Nvidia -0.91%, a 0.08-point loss on the pair, a second consecutive losing session for a closed idea. The short-debasement basket against long dollar, closed on 3 September, would have lost on both legs for the first time: gold rose 0.19% and DXY fell 0.06%. |
| The vol note. VIX closed 16.46, up 0.74 points or 4.71%, with a session range of 15.57 to 16.68, on a day the index fell 0.48% — a 9.8-to-1 ratio of volatility gain to index decline, against 14-to-1 on Tuesday and 3.9-to-1 on Friday. The five-observation path is 14.32 → 14.53 → 15.72 → 16.46, so the surface has added 2.14 points, or 14.9%, in three sessions while the S&P 500 has fallen a cumulative 1.06% across the last two. A 16.46 handle asks for roughly a 1.04% daily move, against a market that has just delivered 0.38%, 0.58% and 0.48% at the index level while producing breadth of 4.09-to-1 and a 14.68-point spread inside communication services. The argument for dispersion over index vol is stronger than it was on Tuesday, not weaker, and it now has a second leg: MOVE published an 8 September value of 76.14, up 4.16% from its 4 September vintage, so rate volatility is confirmed rising for the first time in nine sessions (Section 9 block c) — which means the two surfaces are moving together and the index-level realised move is the thing that is not. The instruction is unchanged: own the 10-11 September inflation block outright, finance it in the 17-18 September meeting expiry where a 60.2% meeting still carries the richest premium on the curve, and prefer dispersion to index vol. |
|
| Crowded consensuses to stress-test, with the numbers. |
| 1. | "The buyback is a bid for the long end." It is a bid for the 10-to-20-year sector, and Treasury said so on Wednesday: up to $6bn on 10 September in that bucket, from a programme doubled to at least $4bn per operation through 4 November. Stress test: the long end sold anyway — the 30-year cheapened 3 bp to 5.28% and the 10-year traded 4.8528%, its cheapest since November 2023 — and it did so on the same afternoon the 10-year auction stopped at 4.834% against 4.683% at the prior auction. A programme designed to support liquidity in older, less liquid paper is not a programme designed to cap yields, and the market spent one session confusing the two. The next test is the operation itself on Thursday and the actual maturities accepted. | | 2. | "The terminal rate is anchored." It has now moved on three consecutive sessions in the same direction and the third was the largest. Every 2027 contract cheapened 1.5 to 4.0 bp with the move rising monotonically along the curve, cumulative-above at December 2027 rose to 95.7% from 94.5%, the first non-trivial cut probability fell to 0.2%, and the modal range moved up a bucket to 4.25%-4.50% at the September and October 2027 meetings for the first time in this reporting window. Stress test: the margin is four-tenths of a point and reverses on any soft print, so this is a threshold crossing rather than a regime change — but three consecutive sessions of back-end-only cheapening with a nearly frozen front cannot be explained by the reaction function, because there has been nothing to react to. Supply and destination are still indistinguishable until the primary market opens. | | 3. | "Energy is a supply shock the equity market can look through." It looked through it again: WTI +4.43% and Brent +3.91% to $101.75 bought 0.95% of energy sector, and two sessions of +5.00% in the barrel have bought 2.14%. Stress test: the pass-through is not in the equity, it is in the crack. The distillate crack widened $5.73 to $104.55 while the gasoline crack fell $5.34 to $38.24, an $11.07 move in the differential in one session, with retail diesel at $5.94 and gasoline at $4.22, both record September prints. And the street does not believe the level: Goldman's December 2026 Brent forecast is $85, sixteen dollars and seventy-five cents below Wednesday's settle, with a $120 upside case and a $60s downside case. A market whose base case is sixteen dollars under spot is a market pricing a war premium it expects to expire, on the day the war entered its seventh month. | | 4. | "Credit is fine because credit spreads are fine." IG at 81 bp is unchanged for a third consecutive update and HY tightened 1 bp to 267, while LQD set a 52-week low at $105.08 and closed -0.16%. Stress test: the investment-grade cash market is at a one-year low with a spread series that has not moved a basis point, which means the entire repricing is happening in the risk-free leg and the spread series is describing a market with post-Labor-Day issuance at its weakest since 2020 and therefore no new paper in it. Meanwhile CCC widened 1 bp to 1,056, 7 bp on the week and 168 bp on the year, and the CCC-minus-HY differential at 789 bp has widened on five of six updates. The plumbing risk is quieter at the rate — SOFR back level with IORB at 3.65% — and busier at the facility: the standing repo facility has now been used on three consecutive operations and reverse repo participation broadened to six counterparties at $432m from three at $626m, with reserves down $30.4bn on the last print and quarter-end fifteen days out. | | 5. | "Volatility has caught up." It has added 2.14 points in three sessions to 16.46, and it is still describing the wrong risk. Stress test: a 16.46 handle asks for roughly a 1.04% daily move, and the index delivered 0.48% on a session with 397 decliners against 97 advancers — a 4.09-to-1 ratio, the widest of the window — a 6.55% gain in one megacap against 8.13% and 6.65% declines in two others, and a 14.68-point spread inside a single sector. That is realised dispersion far above what a 16 handle implies at the index level. The one thing that has improved is the cross-check: MOVE finally published, at 76.14 on an 8 September stamp, up 4.16% from its 4 September vintage, so rate volatility is confirmed rising — but the series has still failed its own internal consistency check for ten consecutive sessions and has published no 9 September value. |
|
| The two-sided geopolitical tape. Escalation is now kinetic and maritime, and the barrel answered on the day. CENTCOM destroyed five Iranian crude carriers on 8 September in reply to a ballistic-missile attack on a U.S. Navy warship, with Washington signalling one tanker destroyed for each subsequent Iranian strike; Iran attacked a U.S. base in Jordan, intensified strikes on Saudi energy infrastructure and rejected the latest diplomatic overtures. Brent closed above $100 for the first time since May. UBS told clients the "off-ramp remains elusive" and to stay long; Bannockburn described a "strong and steady grind higher as the war enters seven months." Against that sit the deflators: Goldman's $85 December base case, the $60s scenario if Gulf output recovers, elevated inventories that absorbed the initial crisis, and the standing observation that Iraq's August exports ran at 2.34m barrels a day against 1.35m in July. Canada's retaliation on $20bn of U.S. goods remains in effect with 50% tariffs live and potential U.S. measures dated 29 September, and it is now showing up in the consumer-cyclical tape rather than the headlines. |
| Structural watch items. Treasury's expanded buyback landing in the 10-to-20-year bucket rather than the long end, with the operation printing on 10 September; a 10-year auction stop of 4.834% against 4.683% a month earlier; reserve balances at $2.8945tn with no new print until Thursday and a standing repo facility used on three consecutive operations; reverse repo at $432m across six counterparties, more participants at less size; a four-month bill 4 bp cheaper at 4.06%, widening the four-month-above-six-month inversion to 5 bp, and a 1.5-month-to-2-month gap that re-widened to 5 bp from 3 after three sessions of narrowing; post-Labor-Day IG issuance at a six-year low into a 4.83% ten-year and an LQD 52-week low; Dell's $4bn refinancing and Enbridge's $2.55bn Tallgrass acquisition as the visible primary calendar; the European Central Bank at 08:15 ET on 10 September priced to 2.65% with European curves 8 to 11 bp cheaper into it; Japan's 10-year 2.4 bp richer at 2.885% with the yen holding a seven-month high and a Bank of Japan decision on 18 September; the broadband repricing that took Comcast down 6.65% and Charter 8.13% on a chief-financial-officer remark; and Apple's first product cycle under John Ternus, launched at a $1,999 entry price into a 0.28% decline. |
| What VIX is and is not pricing. At 16.46 — up 0.74 points and 4.71% — VIX is pricing roughly a 1.04% daily move into PPI and claims on 10 September, the European Central Bank fifteen minutes earlier, CPI on 11 September at a +0.4% consensus, and a meeting priced 60.2% to hike. It is not pricing the breadth. A session in which the index moved 0.48% contained 397 decliners against 97 advancers, Meta +6.55%, Charter -8.13%, Comcast -6.65%, Cooper -6.22% and Alphabet -2.28% — and it is not pricing what happens when the one stock holding the index up is the one that reported nothing. It is not pricing the supply calendar: a 10-year auction that tailed to 4.834%, a buyback in the wrong bucket, and an investment-grade primary market that has to clear into a 4.83% ten-year in two data mornings and a Friday. It is not pricing a barrel through $100 whose sell-side base case is sixteen dollars lower, or a distillate crack that moved eleven dollars away from gasoline in a session. It is not pricing a funding system with the standing repo facility used three operations running fifteen days from quarter-end. And it is now, for the first time in nine sessions, not un-corroborated: MOVE published at 76.14, up 4.16% on an 8 September stamp, so both volatility surfaces are rising while realised index moves stay under sixty basis points. Two surfaces bidding while the index refuses to move is either an unusually good hedge or an unusually expensive one, and the way to find out is the 08:30 print on Thursday. |
|
| Sourcesindex levels and single-name closes from the Investing.com major-indices, Nasdaq 100, Philadelphia Semiconductor and S&P 500 component boards, read after the 16:00 ET close in the local Chrome browser; sector performance from the Finviz group screener in Performance table view; the official par yield curve from the U.S. Treasury Text View; rate-path data from CME FedWatch and the Investing.com Fed Rate Monitor; credit spreads from FRED (ICE BofA series); money-market rates and operations from the Federal Reserve Bank of New York; FX and commodity spot returns from TradingEconomics; futures settles from the Investing.com per-contract historical boards; cash credit proxies from StockAnalysis; earnings dates from the Nasdaq earnings calendar API; macroeconomic consensus from the Investing.com economic calendar, with the 11 September figures carried from the prior edition's Wall Street Journal capture; and narrative cross-checks from Bloomberg, the Wall Street Journal, Reuters, CNBC and The Motley Fool. |
| Companion file: US_CrossAsset_Daily_2026-09-09_DataNotes.txt — full Data Notes & Conflicts, the Overnight / Asia & Europe read-through and the categorized source links. |
| |
| Prepared for institutional readers after the U.S. cash close and the 5:00 p.m. ET bond close. Nothing here is personalized investment advice; verify independently and size to your own mandate before acting. |