〔Day Digest〕Three Shocks Collide to Crash 10Y Yield 19.3bp to 4.95% as Fed Hike Pricing Slips to January 2027 — Oil at $114 and French 10Y at 5.04% Frame Warsh's Credibility Test

0. Daily Arc

The bond market reset came in three distinct blows. AI tech earnings disappointed — Tesla cut FY capex from $25B+ to ~$20B; Meta lifted its FY capex floor to $135B but the implied growth still disappointed; Lam Research was strong without exceeding the upper end of expectations ^[1]. French sovereign spreads blew out ^[1][2]. Middle East tensions pushed Brent to a five-month high of $114/bbl ^[3]. Net result: a 19.3bp single-day drop in the 10Y to 4.952% ^[1] — the largest since the March 2023 SVB crisis, with only 15 trading days in the past decade matching that move ^[1]. Nomura cross-asset strategist Charlie McElligott wrote that "bonds are re-hedging" and that markets are repricing "the probability of a US 'growth scare' driven by an AI reset" in real time ^[1]. A note flagged that Thursday's move "looks more like position stampede than pure safe-haven" ^[4].

1. Fed Path and the Warsh Test

• **[ESCALATED] Hike pricing pushed to January 2027:** Fed funds futures now fully price a 25bp January move, slipping from the prior December timing ^[1].
• **[NEW] October hold is near-certain:** Macro daily tracker shows October hike probability at 14.9% (-0.2pp d/d); December at 87.1% (+1.7pp d/d) ^[5].
• **[NEW] Deutsche Bank hawkish lean:** Chief US economist Matthew Luzzetti expects the Fed to hold in October but still deliver 25bp hikes in December and March 2027, cumulative 50bp ^[6]. Deutsche cites September inflation as "fully surprising to the upside" with core PCE tracking 3.1-3.2% y/y ^[6].
• **[NEW] Warsh credibility test:** Deutsche expects statement changes to be minimal beyond "maintain" wording; the press conference is the focus — Warsh must explain the pause logic, reaffirm the price-stability commitment, clarify the relationship between tightening financial conditions and the policy rate, and characterize AI-related volatility ^[6]. His July press conference was read as softening, raising the credibility bar for this meeting ^[6].

2. The Three Shocks

• **[NEW] AI unwind:** Tesla cut FY capex from $25B+ to ~$20B; Meta raised the FY capex floor to $135B but the implied growth still underwhelmed ^[1]. Goldman/JPM/Bloomberg all tag the AI-vs-Fed collision as the dominant tape driver today ^[7][8][9].
• **[ESCALATED] French sovereign crisis:** French 10Y +4bp to 5.039%, a 24-year high; French 5Y CDS breached 100bp; France plans to deploy troops ahead of fresh protests; ECB Executive Board member Panetta said rising bond yields will dampen the economy ^[2]. ECB expected to deliver a "hawkish pause" at its October meeting ^[10].
• **[NEW] Oil shock:** Brent climbed to a five-month high of $114/bbl on Middle East tensions ^[3]. The market read high oil as a growth threat, not an inflation threat — the inverse of the 2022-2023 interpretation ^[1][3]. The macro weekly warned that energy's consumer-crowding-out effect "deserves continued market attention" ^[10].

3. Cross-Asset Read

• **[NEW] Asia caught the selloff:** KOSPI -4% intraday, -21% month-to-date — one of October's worst major markets globally; Nikkei 225 -2%; MSCI Asia Pacific -1% on track for a fourth weekly loss ^[3].
• **[NEW] Gold paradox:** Gold broke below $4000/oz despite safe-haven flows, -0.7% on the day; silver -2% intraday ^[3]. Logic: high oil reinforces expectations of Fed-hold-higher, raising the opportunity cost of holding non-yielding gold ^[11]. Gold set for its biggest weekly drop in over four months ^[12][11].
• **[NEW] FX and rates:** USD/JPY -0.2% to 156.44; USD/CHF -0.1% to 0.8196, near a one-month low ^[3]. Japan 10Y -10bp to 2.78%, 5Y -10bp to 2.19%; Australia 10Y -10bp to 5.164% ^[3]. US yields edged up Friday after Thursday's slide, with the focus on France ^[13].
• **[NEW] EM central bank moves:** Philippines surprised with a 50bp hike to 5.5%; Turkey cut 100bp to 36% as expected; Swiss National Bank said FX intervention necessity has eased; Indonesia flagged readiness to intervene ^[2].

4. The Long-End Supply Counter

• **[ONGOING] US fiscal load:** CBO reports FY2026 federal deficit at $1.99T, +$218B y/y; federal spending $7.4T, +6% y/y ^[14]. Publicly held debt/GDP projected at 107.7% by 2030, above the 1946 record of 106.1% ^[14].
• **[ONGOING] Long-end premium:** FT flags Treasury Secretary Bessent's tilt toward short-term debt; if inflation and debt rise and AI optimism fades, the danger of reducing Treasury maturities grows ^[15].
• **[NEW] DoubleLine underweight:** Stays heavily underweight long-dated Treasuries, sees short-term safe-haven demand unable to reverse the supply pressure ^[2][16]. A separate note warned that the 10Y breaking 5% "does not signal a return of cheap money" ^[17].

5. What Would Falsify the Pivot

• The macro weekly report cautions that the apparent "pause" is really the market pushing tightening expectations further out, not the disappearance of tightening ^[10]. September retail sales beat expectations but the "control group" used in GDP slowed sharply, with gas-station sales masking weakness elsewhere ^[10].
• The US labor market remains split: healthcare +372K, social assistance +149K and construction +109K over the year to September 2026, while federal government shed 232K jobs (-8%), information lost 120K and finance lost 107K ^[18].
• Source quality: the "19.3bp, only 15 days in a decade" framing ^[1] and the McElligott quote ^[1] are single-sourced through one Wall Street outlet. The position-stampede thesis ^[4] and the "breaking 5% ≠ cheap money" warning ^[17] are headlines without full article text. Kansas City Fed's note on stable loan modifications despite rising delinquencies ^[19] is the only direct central-bank piece in the packet. Motley Fool's "terrible news" inflation framing ^[20] and the policy-deadlock diagnosis ^[21] rest on summaries only — quote as views, not as data.