# FED Warsh Overhaul

- status: dormant  |  conviction: high  |  first detected: 2026-06-03  |  last update: 2026-08-17
- canonical page: https://themic.dev/themes/fed-warsh-overhaul
- exposed instruments: ZN, SR3, ES, GC, 6E, DXY

> DORMANT — no substantive updates since 2026-08-17; archived ledger, not a current view.

## Thesis

Warsh's hawkish credibility + a 4-sigma-hot US labour market + forward-guidance removal = bear-flattening/hike-repricing bias (ZN bear, SR3 front-end bear). First full 25bp hike now priced by year-end vs cuts previously; BNP models 75bp through Mar 2027. Tension with the equity bull narrative that runs on falling yields means a faster-than-priced Warsh tightening compresses multiples (ES). Breadth fragility (Big 3 sectors = all job growth since Dec 2022) and subdued wages are the dovish counterweight; institutional-independence tail persists.

## Watching

CPI May (~Jun 12) as the near-term extension catalyst; $58B/$39B/$22B 3Y/10Y/30Y auction wall requiring concession; Warsh first FOMC (~Jun 17-18) guidance mechanics; whether breadth/wage softness caps the hike path; r-star debate repricing.

## Development timeline (dated, source-cited)

- 2026-08-17: Aug 17: dovish tail edged further past the Aug-15 ~30%-Sept-odds / RenMac-contrarian entry — Westpac's Friday recap marks ~20bps of additional 2026 tightening priced (down from 28bps a week ago) on softer retail sales/sentiment, corroborating the ING/Wells Fargo/UBS dovish lean; no fresh hawkish voice today (RenMac/Nordea counter unchanged). FOMC July minutes (Wed Aug 19) is this week's key catalyst per FT and Westpac, ahead of Jackson Hole. Sources: Westpac, FT. [FT, independent channels]
- 2026-08-17: Aug 17 intraday: dovish repricing ACCELERATED past this morning's ~20bp-priced entry — Sept hike now just ~7bp priced (ING) / ~30% (Handelsbanken, from 50% a week ago); BI says markets see only ONE hike by year-end, down from two in July. Lloyds is the lone pushback, framing Friday's retail-sales miss as 'noise' given resilient top-income spending. ING flags the long end vulnerable regardless (fiscal supply/real yields/yen spillover). FOMC minutes Wed the catalyst. Sources: ING, Handelsbanken, Business Insider, Lloyds. [Handelsbanken, Business Insider, Lloyds, independent channels]
- 2026-08-15: Aug 15: past the Aug-14 retail-sales-soft / RMP-to-zero / benign-PCE-preview entries, the split hardened with a NAMED lone contrarian — RenMac ('Total Eclipse of the Hike') argues the bond market is UNDERPRICING Sept hike risk given core PCE annualising ~2.5% and a soft-CPI/firmer-PPI internal mix it reads as more hawkish than consensus positioning. Sept hike odds now ~30% (WSJ/CME, from 56% no-change a week ago), Lloyds ~30%, Handelsbanken ~35%, MS <40%/BMO ~33%. PPI core y/y 4.2% lowest since March. Convergence exceptional across ING/Lloyds/Handelsbanken/WSJ/BI/Bloomberg/Bond Beat/Westpac/Capital Wars on the Fed-fading-vs-long-end-cheapening split. Sources: WSJ, Lloyds, Handelsbanken, The Bond Beat, independent channels. [WSJ, Lloyds, Handelsbanken, independent channels]
- 2026-08-15: Aug 15 (thin weekend window): color-only past the AM RenMac-lone-contrarian / ~30% Sept-odds entry — independent research (robinjbrooks) reiterates a ~10% USD-bear call for the year but frames it explicitly as CYCLICAL (Fed staying on hold), NOT reserve-currency erosion (reserve managers' USD allocations unchanged since pre-Trump); reads yesterday's weak retail sales as 'totally fine' (a fall after a 'massive rise'), consistent with — not incremental to — the fading-hike-odds move already logged. Sources: independent channels. [independent channels]
- 2026-08-14: Aug 14 intraday: Sept-hike odds pared FURTHER — CME FedWatch ~70% no-change (WSJ), Lloyds ~30% hike (from ~70% at month-start); Goolsbee (non-voter) said inflation 'getting a little bit better'. 30y auction detail firmed: cleared ~5.22%, highest 30y clearing yield since 2001 (Lloyds); 10y 4.66%. NEW counter-thread on the AI-disinflation thesis: BofA/Business Insider flag AI buildout as actively INFLATIONARY now (July CPI computer prices +3.5% m/m, software +40% y/y), pushing back on Warsh; Sevens Report reads it as favoring a 'one-and-done' Sept hike rather than none. Sources: WSJ, Lloyds, Business Insider, BofA. [WSJ, Lloyds, Business Insider, BofA]
- 2026-08-14: Aug 14 intraday (2nd window): fresh macro DATA reinforces the disinflation/soft-consumer leg past this morning's ~70% no-change / one-and-done AI-inflation entry — July retail sales -0.6% m/m vs +0.1% exp, weakest headline in >1yr; control group -0.4% vs +0.3% exp (autos -2.4%, internet -2.2% Prime Day hangover), June revised to +0.4%. ING flags 2.7% savings rate (vs 6% long-run) as a genuine soft-consumer signal despite still expecting a Q3 GDP rebound. Fed cut Reserve Management Purchases to ZERO starting today (WolfStreet) — a further incremental balance-sheet-tightening step, sheet at $6.76tn. Post-PPI PCE previews converged benign: Barclays core +0.20%, MS +0.23%, UBS +0.25% m/m; Wells Fargo keeps Fed on hold through 2027, raised 10y forecasts to 4.45%/4.40% (2026/27). Thought piece (Ethan Harris, ex-BofA via The Bond Beat): Warsh's silence on reaction function has returned markets to pre-1994 Greenspan-decoder-ring guesswork. Sources: ING, WolfStreet, The Bond Beat, Wells Fargo. [independent channels]
- 2026-08-13: Aug 13: past the CPI-in / three-FOMC-dissenter tally, the print resolved into a NEW long-end regime read — Sept-hike odds pared to ~40% (Westpac, from 50% pre-print; CME a Sept coin-flip, ~12% tail for 50bp), but 10y held 4.65-4.75% and did NOT rally. ING (Rates Spark) explicitly reframes: REAL yields and the fiscal deficit, not inflation, are now the swing factor on the long end. Bloomberg's Bloomberg ('Chronicle of a Hike Foretold') calls it the first genuinely unsurprising CPI since the pandemic yet argues Warsh still has little room to avoid eventually hiking toward a true 2% target. Sources: ING, Bloomberg, Westpac, The Bond Beat. [Bloomberg, independent channels]
- 2026-08-13: Aug 13 intraday: the CPI reaction split into a genuine THREE-WAY debate past the AM 'real-yields/deficit-not-inflation' entry — ING quantifies actual repricing: Sept hike odds fell to just 9bp priced (from ~14bp pre-print), Dec -10bp, but a FULL 25bp hike still embedded in the year-end curve (hawkish tail survived); Business Insider frames it dovishly (hold + one hike by year-end); Lloyds Bank pushes back hawkishly ('CPI doesn't capture risks' — wealth effects, AI-capex, falling participation = little slack); robinjbrooks' 'core of core' tracker (30% of core, ex-OER/health/transport) flat in July after negative June, arguing MORE underlying disinflation than headline and markets pricing hikes only 'reluctantly.' First real disagreement since the print. Sources: ING, Business Insider, Lloyds, independent channels. [Business Insider, Lloyds, independent channels]
- 2026-08-13: Aug 13 intraday (2nd window): July PPI decelerated MORE than forecast (energy/food fell), a fresh softer print after CPI giving the Fed 'more time' (Bloomberg); Sept-hike odds slid further to ~35-36% (CME/Polymarket, from ~75% a month ago per macro commentators), Fed-funds curve now pricing just ONE hike in December vs 2+ months ago. Sell-side split hardens: Barclays 2.7% Q4 core, Fed on hold rest of year; BMO core 2.5% y/y (lowest since Feb, 'return to pre-war levels'), Sept pause contingent on Aug core ≤0.3%. Bespoke flags the hawkish-dissent puzzle: three sitting voters (Hammack, Kashkari, Logan) voted to hike in July vs unanimous cuts a year ago at HIGHER core CPI. New: Warsh Jackson Hole speech now dated Fri Aug 28. Sources: Bloomberg, The Bond Beat (Barclays/BMO/UBS/MS/Yardeni/Bespoke), independent channels. [Bloomberg, independent channels]
- 2026-08-12: Aug 12 (CPI-day setup): the split-camp debate crystallised into three named thought pieces past yesterday's Hammack-hawkish / Dudley-critique entries — Bloomberg Opinion (Dudley) hawkish credibility critique, independent research (robinjbrooks) defending Warsh as 'politicized' panic (Powell got a bigger pass for a larger 2025 guidance shift), and independent channels (macro commentators) framing the whole thing as a DEBT-SUPPLY story dressed up as a Fed story. CPI-day pricing: CME FedWatch ~52% hike vs Polymarket ~56% hold for Sep 16 — genuinely balanced. NEW: ING's Garvey notes the recent 10y rise is REAL-rate driven (breakevens ~2.25%, below core est), so a 'good' CPI may not rally ZN if real yields are the binding constraint. Sources: Bloomberg, independent channels, ING. [Bloomberg, independent channels]
- 2026-08-12: Aug 12 intraday (CPI-day): a SECOND hawkish-leaning Fed voice past Monday's Hammack / this morning's three-thought-piece entry — Boston Fed's Collins (non-voter, FT interview) called policy only 'mildly restrictive' and flagged 'the possibility that economic conditions in the coming months will require tighter policy.' Two live/near-live FOMC voices now leaning hawkish into today's CPI, against ~50% Sept-hike pricing (Handelsbanken: a soft print tilts toward hold and could pressure the dollar). Sources: FT, Handelsbanken. [FT, Handelsbanken]
- 2026-08-12: Aug 12 intraday (CPI IN): July CPI landed EXACTLY on consensus past the AM Collins-hawkish entry — headline +0.1% m/m / 3.4% y/y (from 3.5%), core +0.2% m/m / 2.5% y/y (from 2.6%), 3m annualised core down to just 1.6% (gasoline -2.9%, shelter subdued +0.1%; airline fares the +2.2%/+25.5% outlier, jet-fuel driven). ING (Knightley) now calls for a Fed HOLD well into 2027 on four disinflation channels (energy pass-through, shelter cooling to 3.2% w/ rents falling outright, ECI private wages +3.1% consistent with 2% CPI, fading tariff pass-through — June IEEPA refunds already +$25.5bn over new tariff revenue), and dismisses 'chipflation' (computer/comms just 0.7pp of basket vs 35% housing). But three FOMC officials now on record dissenting FOR higher rates (theBondBeat), and pre-CPI CME had Dec at 45% one hike / 28% two / ~5% three — so the hawkish-dissent tally hardens even as the print cools. Sources: ING, theBondBeat. [independent channels]

---
Themic macro theme ledger · Not investment advice. Themic synthesises curated third-party research into a dated, source-attributed ledger of market narratives; the tracking, structure and scenarios are its own editorial work. It holds no directional view — a market call appears only where it is attributed to a named source. Source claims are summarised rather than reproduced, and may be incomplete, superseded or wrong. Nothing here is an offer or solicitation to trade.