Blackout begins with the December consensus intact — and the Treasury buyer base, not the Fed, is now the marginal price-setter

0. Weekly Arc

The week ran from framework speech into institutional silence. Hammack’s “smaller, more frequent” argument and Waller’s willingness to let the long end do the tightening framed Oct 13; Collins, Barkin and Williams then hardened the year-end case on Oct 14; a hot CPI reopened the October question on Oct 15 and a four-year-high PPI followed on Oct 16; Chair Warsh’s refusal to pre-commit collapsed October odds below 15% on Oct 17 and, more durably, converted an inflation shock into a real-yield shock — hot prints left the long end lower on Oct 18–19. Today the Committee goes quiet and the sell side consolidates on December.

1. Policy Narrative & Expectations

The net change over the past ~24h is a market that has stopped arguing about October and started arguing about what the Fed’s reaction function does to the term structure. CME pricing puts October at an 82.8% hold against a 17.2% cumulative 25bp, with December split 6.4% hold / 77.8% cumulative 25bp / 15.8% cumulative 50bp ^[1], and a parallel read has traders at an 82% probability of no change ahead of next week’s decision ^[2]; a weekend snapshot had the October hike probability as low as 15% ^[3]. Against that, the published sell side is splitting on level rather than direction: Barclays now expects 25bp hikes in December 2026 and January 2027 with a 4.25%–4.50% terminal range, explicitly more hawkish than its prior one-hike-then-cut path, and has raised its Q4 2026 core PCE forecast to 3.2% ^[4]; JPMorgan argues the opposite on magnitude, seeing only 25bp cumulative to 4.25% by end-2027 against market pricing of 79bp to 4.79%, and calls the market’s tightening expectations excessive ^[5]. The most useful new structural fact is the arithmetic behind the regime: between the March and September SEP, the median end-2026 policy-rate projection rose 75bp while the median inflation projection rose a full percentage point ^[6]. And the mechanism that reconciles hot inflation with flat breakevens is now published — a hawkish Fed lifted real yields while the energy shock lifted nominal yields, so breakevens barely moved ^[6].

1.1 FOMC Officials’ Remarks

• [NEW] No public FOMC remarks in the past 24h: the Committee is in its pre-meeting blackout until next week’s decision, and there is a published consensus that the next rate rise will not come until December ^[7].

1.2 Policy Signals & Institutional Communication

• [NEW] Donald Kohn turns his critique from substance to communication: the former Vice Chair said Chair Warsh has “retreated too far” in his policy communication, per The Paper, and that a central bank must lay out a narrative framework — where the economy is heading and what it is doing — because that is also what lets the public monitor how decisions are made ^[8]. He added that when rates are far from the zero lower bound, specific rate forward guidance is not very necessary or useful and can lock a central bank into a path it ultimately does not want ^[8]. His substantive read is unchanged from prior briefings: he is not sure current policy is restrictive, measured by broad financial conditions, with credit spreads low and equity indices high ^[8].
• [NEW] Kohn on the independence channel, per Yicai: pressure on Fed independence in the US is already very large, and it coincides with fiscal pressure and repeated supply shocks; the best safeguard of independence is delivering price stability, which recurring oil and tariff shocks have made harder; he does not think the US is heading toward fiscal dominance, and his advice is to stay nonpartisan and focus on price and financial stability ^[9]. His sharpest institutional point is that the Fed communicates well with markets and academia and poorly with the public and the legislature, where it falls into technocratic discussion ^[9].
• [NEW] The academic case for and against the Chair’s anti-guidance doctrine: CEPR/VoxEU work by Lustenberger and Rossi finds that during forward-guidance episodes central-bank information is less precise than private information yet professional forecasters over-weight it; in US data the ratio of average forecast-error variance to forecaster disagreement on three-month-ahead T-bill forecasts falls to 0.67 during forward-guidance periods against 3.32 outside them ^[10]. The warning is the relevant part: “During this interim phase, forecast errors and disagreement may intensify alongside a rise in market volatility” ^[10].
• [ESCALATED] Barclays moves to a two-hike path and prices a higher inflation floor: December 2026 and January 2027 hikes to a 4.25%–4.50% terminal range, then unchanged through 2027, with Q4 2026 core PCE revised up to 3.2% and Q4 2027 to 2.6% ^[4]. Its macro weekly frames the same conclusion differently — the UK Budget needs a credible fiscal stance, US inflation above expectations drove the hawkish revision, and the ECB and BOJ are also expected to tighten further ^[4].
• [NEW] JPMorgan pushes back on the market’s hawkishness: “magical thinking” about rapid rate cuts must end because global growth is running above potential and core inflation is above target, but the bank still forecasts only 25bp of cumulative Fed tightening to 4.25% by end-2027 against 79bp priced, and warns that even a mild US recession would historically require very large rate cuts that sticky inflation makes hard to deliver ^[5].
• [NEW] Haitong frames October as a deliberate hawkish pause: it expects the Fed to hold to give elevated Treasury yields more time to tighten financial conditions, preserving an observation window and pushing the real decision to December, and argues that even if inflation trends lower that does not mean rates cannot rise — inflation is not currently a hard constraint on hiking ^[11].
• [NEW] Shanxi Securities bounds the remaining room: it prices total remaining hiking space at roughly 60–75bp, taking the lower of the Taylor-rule implied rate and the terminal-rate credibility cap, with December (~65% priced) as the baseline hike and October a hold; it notes the 10-year sits well above the policy-rate upper bound with the nominal rise borne by real yields while breakevens decline, implying the policy path has not been re-priced ^[12].
• [NEW] A 1997 analogue for the “one more and done” instinct: Greenspan said in March 1997 that he thought the odds were better than 50/50 that the tightening move under consideration would not be the last one — cited by Sinolink as the case against treating a single hike as terminal ^[13] (dated framework piece, 09/13).
• [ONGOING] Reserve-management plumbing: the Fed is holding reserve-management purchases paused under its October–November schedule, with Barclays expecting a resumption in mid-November at a modest monthly pace as bill supply rises and reserve balances fall ^[14].
• [ONGOING] Pricing structure: October 82.8% hold / 17.2% hike; December 6.4% / 77.8% / 15.8% ^[1][2].

2. Key Data & Market Read

• [NEW] Goldman’s activity update keeps the growth leg intact but cooling: Q3 GDP tracking a firm annualized pace, the economic surprise index turning marginally positive, the preliminary October activity indicator easing from September, and — the important half — a core inflation tracker showing no re-acceleration with the wage tracker steady ^[15].
• [NEW] Housing is the cleanest transmission casualty: NAHB housebuilding sentiment fell this month to its lowest in fourteen and a half years, which the reporting attributes directly to the mortgage squeeze ^[7].
• [ONGOING] September CPI, PPI and retail sales: headline and core CPI above expectations with energy the headline driver but housing, transport services and goods also firmer; PPI posting its largest monthly gain since 2022 with transport and warehousing costs rising; and a retail headline beat that was gasoline-driven, with the ex-gasoline and control-group measures clearly weaker than prior and expected ^[16].
• [ONGOING] The composition argument is unchanged: core goods prices have turned positive year-on-year while rent and supercore services remain the sticky components, and the labour market is tight rather than loosening ^[5].
• [NEW] A dissenting read on how much inflation is actually left: Sinolink argues core CPI has quietly fallen to its lowest since 2021 and close to the pre-pandemic average, and that the five most volatile components contributed a large absolute but very small net amount to headline CPI over January–August — i.e. they generate monthly noise rather than a persistent price level ^[13] (dated framework piece, 09/13).
• [NEW] Calendar: September housing starts and pending home sales, weekly ADP figures, Thursday’s jobless claims, Friday’s S&P Global composite PMI flash and the PCE gauge are all due before the decision ^[2].
• [ONGOING] Narrative impact: the growth leg is still being marked up and the inflation leg is being contested from both directions, which is exactly the configuration in which the Committee prefers to say nothing — and it now can, for a week.

3. Financial-Conditions Signals

• [ESCALATED] Credit: dispersion, not direction, is the signal: the JULI index widened 3bp to 96bp while its yield actually fell 12bp on the week, hyperscaler spreads widened 12bp to 148bp near the July wides, and J.P. Morgan reads that move as an overreaction unsupported by fundamentals because new issuance was concentrated among higher-rated borrowers and the cohort’s average rating improved ^[3]. The dispersion measures are the tell: JULI sector-weighted dispersion at its widest since May 2025, the ratio of 7–10 year dispersion to spread at its highest since January 2024, and the BBB-versus-A non-financial spread at its widest since March 2026 ^[3]. CDX.IG and CDX.HY both widened on the day ^[3], and Citi’s checklist notes high-yield spreads have already widened — historically a coincident signal with a yield peak ^[17].
• [ESCALATED] Liquidity: the marginal buyer has become price-sensitive: weekly inflows into US bond funds collapsed to roughly the third percentile of the past six months, driven mainly by long-end government funds flipping to net outflow at the third percentile, while intermediate government funds stayed strong and municipal funds rebounded sharply ^[14]. The official-sector side keeps weakening — Fed custody holdings for foreign official accounts fell another $14.6bn in the week to Oct 14 and Japanese investors sold foreign bills and bonds again ^[14] — money-market fund assets fell about $40bn with the Treasury General Account at roughly $935bn, and heavy bill and coupon settlement is absorbing the cash ^[14].
• [NEW] The buyer-base composition is now the structural story: the foreign official share of Treasury holdings fell to 11.8% in August from 40% during the financial crisis, while the hedge-fund share rose to 33.5% from around 10% — an increasingly leveraged, price-sensitive buyer absorbing ever more supply, which the reporting links directly to higher long-run yields and higher borrowing costs for mortgages and business ^[18].
• [ESCALATED] Financial conditions tightened on every published gauge, from a loose base: the Chicago Fed index hit its tightest since April and Goldman’s equivalent its tightest since March, Morgan Stanley’s rates-dollar-energy composite its tightest since 2022 — yet the Chicago Fed measure remains well below its 20-year average ^[7]. Goldman’s own index tightened modestly on the week, dragged by equities and partly offset by lower 10-year yields ^[15]. Over the two months since Jackson Hole the 10-year is up almost 50bp, thirty-year mortgage rates up 70bp, and junk spreads out almost 100bp ^[7].
• [NEW] Cross-asset conditions are doing the tightening the Fed is not: the S&P 500 and Nasdaq are basically unchanged since Jackson Hole, so equities have stopped offsetting; the VIX hit a four-month high on Friday and MOVE has surged back to March levels; crude is up more than 20% over two months and the dollar index almost 3%; economically sensitive stocks have lost nearly 10% since August ^[7]. Gold is being capped by elevated real yields even as allocation money adds counter-trend ^[19], and French bond stress is spilling over into US rates ^[20][21].
• [NEW] Driver narrative — an inflation shock transmitted as a real-yield shock: breakevens sit around 2.4%, roughly where they started the year, while the rise in yields has been in real terms. FT Alphaville, drawing on Morgan Stanley, states the conclusion plainly: “the core reason why bond yields are higher is still inflation stemming from the Iran war-triggered energy shock, despite break-even rates staying quiescent” ^[6]. The mechanism is that investors trust the new Fed leadership to lean against inflation, so a hawkish Fed lifts real yields and energy lifts nominal yields, and the two cancel in the breakeven ^[6]. Citi’s cross-asset checklist refuses to confirm a top: oil has not clearly fallen back, front-end breakevens are still strengthening, and the 5s30s curve has reversed over the past 25 sessions — so the signals conflict ^[17].
• [NEW] Risk parity is named explicitly in the flow research: with the stock-bond correlation sustained at a high level, the equity allocation increase is capped; the bond allocation sits below median because its hedging value has fallen; and the commodity allocation is near a ten-year high and rises further at rebalancing precisely because of its negative correlation ^[22]. Systematic positioning is the near-term headwind — vol-control equity exposure has risen to about 95% of its yearly high with little room to add, CTA models imply selling in US large cap, small cap and tech, CTAs remain short Treasuries with the deepest shorts in the long end, and CTAs are deeply short gold with the short now crowded ^[22].

4. Global Central-Bank Linkages

• [NEW] PBoC — a framework change, not a cut: Governor Pan Gongsheng said at the Financial Street Forum on 10/19 that the interbank overnight repo rate has in practice become the operating target, and that the central bank will study the timing to switch the main open-market operation tenor from 7 days to overnight, making the overnight repo rate the main policy rate ^[23][24]. He also proposed moving loan pricing from a single LPR reference to more diversified benchmarks, with stronger firms able to choose DR001 or LPR for floating-rate loans ^[24]; announced a liquidity support tool for large redemptions at non-bank institutions, to be activated on group-wide liquidity stress and a redemption-selloff loop ^[25][26]; extended the standing lending facility to domestically incorporated foreign bank entities ^[25][26]; and moved to retire the deposit and lending benchmark rates ^[24][26]. Analysts across houses stress this is not a rate cut — Goldman notes that if the 1.25% overnight rate were adopted it would imply a 15bp reduction versus the current 7-day rate, but keeps its baseline of no policy-rate or RRR cuts this year given stronger-than-expected Q3 GDP, with part of 2027 easing delivered through the benchmark transition ^[26]; Société Générale and Tianfeng expect a multi-month transition requiring more frequent operations to test market reception ^[27].
• [NEW] BOJ: Reuters reports, citing sources, that the Bank of Japan may raise its inflation forecast while standing pat on rates ^[28] (single source / unverified). Barclays has brought its next BOJ hike forward to December 2026 with the terminal rate raised to 2.0% ^[4], and JPMorgan expects 100bp of BOJ tightening to 2.25% by end-2027, above the 83bp priced ^[5].
• [NEW] ECB: JPMorgan expects 50bp to 3.00% by end-2027, below the 77bp priced ^[5]; Barclays expects a final 25bp in December taking the deposit rate to 2.75% ^[4]. The Bund Summit panel supplied the institutional frame: Frenkel argued independence means fulfilling the statutory mandate with price stability first and that data-dependent central banks are lagged by construction; Knot said the point of independence is that it sharply lowers the cost of inflation, citing euro-area expectations peaking at only 2.6% during a double-digit inflation episode; Lane noted the ECB operates inside a legislative framework in supervision and on the digital euro; and Shirakawa warned that independence is being challenged because central banks are asked to do too much and that price stability alone does not capture financial imbalances ^[9].
• [NEW] Bank of Canada: after the central bank’s latest surveys showed inflation expectations rising, more economists and traders expect a hike next week, with Bank of Montreal, RBC Dominion Securities and Desjardins flipping their calls on Monday ^[29].
• [NEW] BOE and the synchronised cycle: markets expect the Bank of England to raise rates soon ^[9], and JPMorgan sees the euro area and the UK each adding 50bp by end-2027 as part of a 36bp average rise in developed-economy policy rates ^[5]. September 2026 marked the Fed, ECB and BOJ all hiking — the largest wave of net tightening since 2023 — with the rise in risk-free rates concentrated at the long end ^[9][12].

5. Asset Implications

Quadrant	Current probability tilt	Key asset implication	Anchoring narrative
Growth↑ + Inflation↑	Unchanged at elevated levels	Commodities↑ TIPS↑ nominal long bonds↓ — Barclays marking Q4 2026 core PCE up to 3.2%, JPMorgan’s “core stuck near 3% and magical thinking must end”, Kohn’s 2.5%–3% landing point, and a SEP in which a full point of inflation revision bought only 75bp of rate revision	§1.2 / §2
Growth↑ + Inflation↓	Rising at the margin	Stocks↑ long bonds↑ gold↓ — Goldman’s contained core and wage trackers with a firm activity read, Sinolink’s argument that core CPI is at its lowest since 2021 and that volatile components net out, and the FT/Morgan Stanley finding that breakevens are anchored at roughly their start-of-year level	§2 / §3
Growth↓ + Inflation↑	Rising	Commodities↑ gold↑ stocks↓ — housing sentiment at a fourteen-and-a-half-year low with mortgage rates up 70bp since Jackson Hole, junk spreads out almost 100bp, the VIX at a four-month high, economically sensitive stocks down nearly 10%, and Citi’s regime model shifting to late cycle with equities and credit underweight	§3 / §2
Growth↓ + Inflation↓	Falling to a low probability	Long bonds↑↑ stocks↓ credit spreads↑ — this quadrant has lost its buyer: weekly bond-fund inflows at the third percentile with long-end government funds in net outflow, foreign official holdings at 11.8% of the stack, and custody holdings still falling, so any duration rally lacks a marginal bid	§3

Stock-bond correlation call: the positive, inflation-and-supply-driven regime remains the base case, but today’s evidence says the transmission channel has migrated from breakevens to real yields. That distinction matters for a risk-parity book. If yields were rising because inflation compensation was rising, TIPS would be the natural hedge and the loss would be concentrated in nominals. Instead breakevens sit near their start-of-year level while real yields do the work ^[6], which means TIPS and nominals lose together, equities lose on the discount rate, and the only asset that reliably gains is the one whose supply is disrupted. A credible hawkish Fed is therefore a positive-correlation machine: it is precisely the credibility that keeps breakevens anchored and pushes the entire adjustment into the real-yield leg, which is the leg that repriced both sleeves simultaneously. Three corroborations: gross issuance keeps rising against a buyer base that is now a third leveraged hedge funds and only about a twelfth foreign official ^[18]; rate volatility has returned to March levels while equity volatility sits at a four-month high, so the bond sleeve is no longer a low-volatility diversifier ^[7]; and the published flow research now states mechanically that a high stock-bond correlation caps the equity increase, pushes the bond allocation below median on lost hedging value, and pushes commodities to a ten-year high ^[22]. The counterweight is Citi’s refusal to confirm a yield top combined with its late-cycle regime switch — underweight equities and credit, overweight non-US bonds and commodities ^[17] — which is the same conclusion reached from the growth side rather than the inflation side.

Risk-budget implication: under this correlation structure the bond sleeve should carry less risk budget than its volatility suggests, and the reduction should be taken in the long end specifically rather than in duration generally. The reason is not valuation but market structure: the marginal long-end buyer has become price-sensitive and is now absent at the third percentile of inflows, so the same duration position carries more price risk per unit of yield than it did when official accounts absorbed supply ^[14][18]. Duration is better expressed in the intermediate and front end, where fund demand is still in the upper percentiles, and in the belly rather than as a level bet — with curve structures preferred to outright longs, given that CTA shorts are deepest at the long end and would cover first there ^[22]. The commodity sleeve earns an above-neutral weight precisely because it is the only leg with a reliably negative correlation to both others in this regime, and because the energy shock is simultaneously the inflation story, the term-premium story and the reason December stays priced ^[22][7]. Within credit, hold front-end financials as the defensive expression and treat the hyperscaler widening as dispersion rather than deterioration, since the cohort’s average rating improved even as spreads widened ^[3]. In equities, size to a discount-rate scenario rather than an earnings scenario: vol-control exposure is already near its yearly high, systematic flows imply selling across paths, and the crowded long-gold short means the metal is a reserve-diversification sleeve rather than a rates hedge at this point in the positioning cycle ^[22].

6. Contrarian & Tail Risks

• Consensus fragility: the baseline is now “October is off, December is the meeting, and the sell side has converged.” Four assumptions are falsifiable. First, the sell side has not converged at all — Barclays moved more hawkish to December-plus-January with a 4.25%–4.50% terminal, while JPMorgan calls the market’s 79bp of pricing excessive and forecasts only 25bp, a 54bp disagreement about the same cycle ^[4][5]. Second, the pricing itself is three different statistics sampled at different moments — an 82.8% hold probability, an 82% hold probability, a 15% hike probability and a 17.2% hike probability all describe the same meeting ^[1][2][3]. Third, the inflation data are being read in opposite directions from comparable inputs: core CPI is variously reported as having fallen to its lowest since 2021 and as sticky near 3% with core goods turning positive, and Barclays marked core PCE up to 3.2% while Goldman’s tracker shows no re-acceleration ^{[13][5][4][15]}. Fourth, and most structural, the December consensus rests on a buyer base that has already walked: if long-end demand is genuinely price-sensitive and currently absent, then the tightening the Fed is counting on to substitute for a hike may reverse faster than the Fed expects, which cuts both ways.
• Second-order transmission: four chains are not in the price. The market-structure chain is the most important and is now quantified: foreign official holdings at 11.8% from 40%, hedge funds at 33.5% from around 10%, and gross issuance rising — a buyer base that must be compensated with volatility, which means higher long-run yields irrespective of the policy path ^[18]. The AI-financing chain remains live on both sides: CoreWeave’s technical selling is judged a funding issue rather than a fundamental one, but hyperscaler spreads near the July wides show the market charging the cohort that also supplies index earnings ^[4][3]. The real-yield chain is the third: because the adjustment is running through real yields rather than breakevens, the same Fed credibility that anchors expectations also guarantees that any growth scare hits TIPS and nominals together, removing the inflation hedge from the portfolio at exactly the moment it is needed ^[6]. The communication-vacuum chain is the fourth and is dated to this week: with the Committee in blackout, no official can correct a mispriced path for seven days, and the published research warns that exiting forward guidance carries transitional costs in the form of larger forecast errors, greater disagreement and higher market volatility ^[10][7].
• Source quality control: volume is not confirmation. Several items are single-source relays and are carried as colour: the BOJ report that it may raise its inflation forecast while holding rates is a Reuters source story relayed through a social post ^[28]; the French-to-US rates spillover is a single social post ^[20]; the gold-and-equities item is a headline with no body ^[19]; the WSJ Market Talk item is truncated and yields only the proposition that Treasury demand has changed over the decade ^[30]; and one contributed item carries no extractable Fed or US macro fact at all and is excluded ^[31]. One large block is a dated framework piece rather than today’s news — Sinolink is dated 09/13 and is used for its structural arguments, not as a fresh signal ^[13]. Direct conflicts deserve side-by-side reading rather than splicing: core CPI appears at its lowest since 2021 in one house and near 3% in another, which is either a different series or a different vintage and must not be averaged ^[13][5]; the October hike probability appears at 15%, 17.2% and an implied 18% across three readings ^[3][1][2]; Barclays appears three times in this batch with related but distinct framings of the same revision and should be read as one call, not three ^[4][32][14]; and the “Fed may quietly welcome an equity selloff” line is one outlet’s inference, not reporting ^[19]. Finally, several items are model- or framework-derived — the FCI levels and percentile rankings, Citi’s 21-episode base rate, the SEP decomposition, the Taylor-rule bounded hiking space, and the systematic-positioning estimates — and frame a distribution rather than a price.

---

Appendix: Additional Sources

• ^[33] Wallstreetcn — China’s monetary framework shifting from quantity anchors to a price anchor; not a cut, a hike or a policy turn
• ^[27] Bloomberg — Société Générale and Tianfeng: a multi-month transition to an overnight policy rate will require more frequent operations to test market reception
• ^[25] Huachuang Securities — Financial Street Forum read-through including the non-bank redemption liquidity tool and the SLF expansion
• ^[23] Yicai — Pan Gongsheng: DR001 already the operating target; the 7-day-to-overnight switch should not be equated with a rate cut
• ^[34] 郁言债市 — six signals from the Financial Street Forum, including LPR diversification and the CSRC’s market-resilience measures
• ^[24] CICC — the framework upgrade is about perfecting a price-based regime and financial stability, not signalling aggregate easing
• ^[26] Goldman Sachs — the overnight switch implies a 15bp reduction if the 1.25% rate is adopted, but the no-cut baseline for this year is maintained
• ^[12] Shanxi Securities — remaining hiking space bounded by the Taylor-rule rate and the terminal-rate cap; December the baseline
• ^[16] 铜冠金源期货 — September CPI, PPI and retail detail; second-round pass-through and consumer momentum are the watch items
• ^[35] Soochow Securities — geopolitical risk plus a cooling AI trade drove long-end yields lower and flattened the curve

---

This report is a macro-mechanism analysis, not investment advice.