〈BofA Reinforces Three-Hike 2026 Call as Warsh Hawkish Testimony Offsets CPI Disinflation; Leveraged ETF Deleveraging and Hedge Fund Treasury Concentration Highlight Fragility; Oil Surges 16% to $88〉

0. Weekly Arc

The week opened with the benign June CPI/PPI disinflation (core CPI flat m/m, PPI -0.3% m/m) collapsing July hike probability to ~14%. Chair Warsh’s hawkish congressional testimony (July 14-15) — “zero tolerance” for persistent inflation, downplaying the data as one print — partially reversed the dovish repricing. BofA concurrently issued its three-hike 2026 call, citing sticky inflation, strong growth tracking (GDP upgraded to 1.7% for Q2, 2.5% for H2), and a strategic motive to rebuild credibility ^[1][2][3][4]. By week’s end, the September hike probability settled at ~53% ^[5][6], the curve steepened on expectations of rate normalization ^[7][8], and the oil surge to $88/bbl (Brent +15.91% weekly) reintroduced energy-inflation risk ^[6]. The arc ends with a standoff: the data flow argues for a hold, but the committee and sell-side are pricing tightening.

1. Policy Narrative & Expectations

The net change over the past ~24h is a reinforced hawkish institutional stance from BofA’s three-hike call and Warsh’s testimony, partially offset by the market’s CPI-driven repricing of September probability to ~53%. BofA Economics now explicitly forecasts 75bp of hikes in September, October, and December 2026, arguing the Fed’s new leadership under Warsh has a “strategic motive to rebuild credibility” rather than use the task forces to delay action ^[1][2][3]. Warsh’s testimony established that the committee has “no tolerance for persistently elevated inflation” and reframed the Fed’s mission as preventing supply shocks from “spreading out” into general inflation ^[9][6]. The market OIS prices only ~1.5 hikes over the next three quarters, which BofA labels a “low-probability intermediate zone” in a bimodal distribution: either no hikes or a full 3-hike cycle ^[1].

1.1 FOMC Officials’ Remarks

No public FOMC remarks in the past 24h. (The articles contain remarks from Warsh’s July 14-15 testimony and July 13-16 speeches from Waller, Logan, Jefferson, and Schmid — all already covered in prior briefings.)

1.2 Policy Signals & Institutional Communication

• [NEW] BofA three-hike 2026 forecast: BofA Economics expects 25bp hikes in September, October, and December, totaling 75bp, to reverse 2025’s 75bp of easing and tighten financial conditions ^[1][2][3]. Goldman raised its Q2 GDP tracking to 1.7% and H2 average to 2.5% ^[2]. Warsh has a strategic motive to “build credibility” through action rather than delay via task forces ^[1][2].
• [NEW] Core PCE “true” rate estimated at 2.5%: BofA estimates that after removing one-time factors, the underlying core PCE inflation rate is approximately 2.5% — still above the 2% target ^[1][2].
• [ONGOING] CME FedWatch — September hike at ~53%, July hold at ~86%: As of July 19, the market prices July unchanged at 85.6%, September 25bp hike at 53.5%, and the cumulative 75bp hike path at ~8% ^[5]. The July 18 data from CME Fed Watch shows a 51.2% September hike probability ^[6].
• [NEW] Fed blackout period began July 18: The FOMC entered its pre-meeting quiet period on July 18 ^[2].
• [NEW] BofA Bull & Bear Indicator at 9.6 (extreme bullishness): The indicator rose to 9.6, signaling extreme market optimism. Equity inflows were $55.8bn, bond inflows $20bn, and cash outflows $119.6bn for the week ^[10]. Tech funds saw a record $48.8bn inflow over three consecutive weeks ^[10]. 83% of fund managers expect no rate hike before the November midterms ^[10].
• [NEW] Warsh’s reduced communication era analyzed: FT details Warsh’s shift away from forward guidance — he intends to be “very curt” on rate plans ^[9]. Waller argues the chair should say more about scenarios; Brainard noted the biggest two-year yield jump since tariff day occurred after Warsh’s June press conference ^[9]. Lael Brainard and Janet Yellen both warned that reduced guidance creates pricing uncertainty ^[9]. Apollo’s Torsten Sløk warned that “leverage unwind” scenarios could happen suddenly with less forward guidance ^[9].
• [NEW] BIS warns on hedge fund Treasury concentration: Hedge funds’ holdings of US Treasuries doubled between 2023 and 2025 to over $2.5tn ^[9]. The Bank for International Settlements warned that hedge funds may need to rapidly reduce their positions, describing this as “one of the most troubling financial stability risks” ^[9].
• [NEW] AI hyperscalers’ debt issuance boom: The Bank of England reported that AI hyperscalers borrowed more in H1 2026 than in all of 2025, accumulating as much new debt as the UK government ^[9]. US federal debt held by the public is set to exceed pre-WWII highs ^[9].

2. Key Data & Market Read

• [ONGOING] June CPI disinflation: Headline CPI fell 0.4% m/m (first monthly decline since 2020), core CPI flat m/m — both below consensus ^{[2][4][11][6]}. BofA estimates the CPI trend points to 3.9% by year-end, still above target ^[10]. Core PCE after removing one-time factors is ~2.5% ^[1][2]. Narrative impact: Provides the Fed with a window to hold near-term but does not change the medium-term tightening path per BofA and Warsh.
• [ONGOING] June PPI miss: PPI fell 0.3% m/m (vs expected unchanged), energy costs down 6.4% m/m, year-over-year grew 5.5% (below 6.2% consensus) ^[4][11]. Narrative impact: Reinforced the factory-level disinflation narrative but is largely energy-driven.
• [NEW] June retail sales in line, control group strong: Headline retail sales rose 0.2% m/m in line, but the control group (ex-auto, ex-gas) rose 0.5% with upward revisions to April-May data ^[4][6]. Narrative impact: Consumer spending remains resilient, supporting the growth leg and BofA’s upgraded GDP tracking.
• [NEW] NFIB small business optimism and Michigan consumer sentiment above expectations: June NFIB rose to 97.4 (vs 95.5 expected); July Michigan consumer sentiment rose to 54.4 (five-month high) ^[4]. Narrative impact: Both data points show business and consumer confidence recovering — inconsistent with a recession narrative.
• [ONGOING] Oil surge to $88/bbl: Brent crude weekly gain of 15.91% to $88.10/bbl ^[6]. Narrative impact: Reintroduces energy-driven inflation risk that the June CPI/PPI data (from May-June, pre-surge) do not capture.

3. Financial-Conditions Signals

• [NEW] Dollar & rates — curve steepening narrative: Deer Point Macro argues that normalization of the policy rate toward the effective fed funds rate supports a steeper Treasury curve, particularly in the 7-year sector, with recent flattening improving entry points ^[7][8]. BofA maintains a short 2-year position targeting 4.40% from 4.14% ^[3].
• [NEW] Dollar & rates — MOVE index rose to 70.88, 10-year yield ranged 4.55%-4.62%: The MOVE volatility index rose, and the 10-year yield touched 4.62% before settling at 4.55% ^[6]. The dollar weakened slightly, DXY at 100.76, down 0.20% on the week ^[6].
• [NEW] Liquidity — leveraged ETF AUM down 13% from peak: J.P. Morgan’s Flows & Liquidity report shows leveraged equity ETF AUM declined about 13% from the June peak, with storage chip single-stock leveraged ETFs down ~34% ^[12]. Risk parity implied leverage has largely normalized, suggesting future pressure will come from leveraged ETFs, options, and margin accounts ^[12]. Retail call option buying on June 5 approached 14 million contracts, near historical highs that preceded months of tech corrections ^[12].
• [NEW] Liquidity — hedge fund Treasury holdings at $2.5tn (systemic risk): Large hedge funds’ Treasury holdings doubled to over $2.5tn, faster than overall market growth ^[9]. BIS warns of a potential rapid deleveraging in the government bond market as a key stability risk ^[9].
• [NEW] Liquidity — cash outflows extreme: Global money market funds had a record $98.46bn outflow in one week, with the outflow quantile dropping to 1.2% (extremely low) ^[13]. US stock and bond funds saw continued inflows but at reduced pace from the prior week ^[13].
• [NEW] Credit — AI hyperscalers as the new sovereign borrowers: The BOE noted that AI companies borrowed more in H1 2026 than in all of 2025, accumulating as much new debt as the UK government ^[9].
• [ONGOING] Credit — US financial conditions tightened: The Bloomberg US Financial Conditions Index fell from 1.241 to 1.060 over the week, signaling tighter conditions ^[4]. US-EU yield spreads narrowed, US-Japan widened ^[4].

4. Global Central-Bank Linkages

• [NEW] Bank of Korea hiked 25bp to 2.75%: The BOK raised rates to 2.75%, tightening financial conditions and exposing high leverage — the KOSPI fell 8.77% on the week, entering a technical bear market, led by chip stocks (SK Hynix -11% on one day) ^[10][11][6].
• [NEW] ECB — expected to hold next week, September hike possible: BofA expects the ECB to keep rates unchanged at the July meeting but sees a final 25bp hike in September given energy price rebounding ^[1][3]. ECB President Lagarde recently called for a “back-to-basics approach” to communication ^[9].
• [NEW] BOE — two-sided inflation risk: BofA flags that the Bank of England faces a secondary inflation risk from energy prices, with September or November as potential hike windows. However, BofA maintains a steepening view on the Gilt curve, expecting inflation below BOE projections ^[1][3]. Governor Bailey echoed Warsh’s view that too much guidance limits policy flexibility ^[9].
• [NEW] RBA — August hike risk: BofA maintains long Australian rates vs US, noting that a strong labor market or core inflation surprise could raise the risk of an August Reserve Bank of Australia hike ^[3].
• [NEW] Sweden’s Riksbank — Fed meetings have minimal rate impact: Riksbank research indicates that since the 2022 inflation surge, Fed meetings have had little impact on subsequent yield rises ^[9].

5. Asset Implications

This section is inference — no [N]. Anchored to the facts above.

Quadrant	Current probability tilt	Key asset implication	Anchoring narrative
Growth↑ + Inflation↑	Rising	BofA’s three-hike forecast, oil surge to $88, AI hyperscaler debt boom, and the BIS hedge fund Treasury warning create a stagflationary pressure: commodities↑ (oil, gold) TIPS↑ nominal long bonds↓; BofA short 2s; curve steepeners (Deer Point Macro) capture front-end hawkish repricing; tech and semiconductors vulnerable (semiconductor ETF SMH +9.5% in June but mega-cap cloud -14.5%, rotation — but storage chip leverage density 3x avg) → risk of VaR shock	§1.2 (BofA three-hike, core PCE 2.5%); §2 (oil $88, CPI trend 3.9%); §3 (curve steepener call, BIS hedge fund warning, AI debt boom); §19 (Brent +15.91%, KOSPI -8.77%); §3 (leveraged ETF AUM -13%, storage chip leverage 3x)
Growth↑ + Inflation↓	Falling	The “Goldilocks” window narrows: BofA’s GDP upgrade (Q2 1.7%, H2 2.5%) and strong control-group retail sales (+0.5%) support growth, but BofA explicitly links that growth to a need for hikes, not a hold; the Michigan sentiment and NFIB data show confidence, not caution; the Warsh hawkish testimony and 83% of fund managers expecting no pre-midterm hike [10] is fragile — if BofA is right, the binary tail is violent; SMH up 9.5% in June while mega-cap cloud fell 14.5%, inconsistent with pure Goldilocks	§1.2 (BofA GDP upgrade, NFIB/Michigan beats); §2 (control-group retail +0.5%); §8 (83% expect no hike before November — fragility); §3 (SMH +9.5% vs mega-cap cloud -14.5% — rotation, not pure risk-on)
Growth↓ + Inflation↑	Rising	Stagflation tail is the base case for risk parity: oil spike + BofA hike + KOSPI crash (−8.77%, tech bear) + leveraged ETF deleveraging (−13% AUM) + BIS Treasury liquidation risk = stocks and bonds both vulnerable. The BIS hedge fund warning ($2.5tn, doubled, “most troubling stability risk”) is a systemic level threat that, if realized, would force simultaneous liquidation across rates/credit/equities — the worst risk-parity scenario. Gold struggled (weekly −2.53%, below $4000) because rising real yields and dollar resilience offset inflation hedging, but BofA still projects $4250 Q4 2026 and $5000 by mid-2027	§1.2 (BIS hedge fund warning, BofA three-hike); §2 (oil $88, KOSPI −8.77%); §3 (leveraged ETF AUM −13%, storage chip leverage 3x; retail options near historical highs); §4 (BOK hike triggers KOSPI crash); §19 (gold −2.53%, Brent +15.91%)
Growth↓ + Inflation↓	Falling	Long-duration bonds would benefit from a recession+disinflation outcome, but the opposite is priced: BofA shorts 2s (target 4.40%, current 4.14%), curve steepeners are crowded, cash is flowing out aggressively ($119.6bn cash outflow, MMF quantile at 1.2% — the lowest in history, implying extreme bearish cash → bullish risk), and the BIS hedge fund risk means any bond rally could be cut short by forced hedge fund liquidation. The Warsh testimony rules out a dovish pivot; the 53.5% September hike probability means a hold surprise would be a rally, but BofA’s baseline is for the hike to happen. The deer point macro steepening trade (7-year sector carry) is consistent with a “growth resilient, inflation normalizing” but not “growth↓+inflation↓” path	§1.2 (BofA short 2s, three-hike); §3 (cash outflows extreme, retail options at historical highs; curve steepener call; MOVE 70.88); §19 (10-year ranged 4.55-4.62%, gold −2.53%)

Stock-bond correlation call: The regime is leaning toward positive correlation (inflation-driven). Three structural forces argue for stocks and bonds selling together: (1) BofA’s three-hike call and Warsh’s zero-tolerance stance create a repricing of the entire rate path that hits equity durations; (2) the oil surge (+16% weekly) reintroduces an energy-driven stagflation channel that the benign June CPI data does not capture; (3) the BIS hedge fund warning — $2.5tn in Treasury positions that “might have to be rapidly reduced” — is a systemic risk that simultaneously threatens both rates and credit spreads. The leveraged ETF deleveraging (−13% AUM, storage chip −34%) is already happening, and the retail options activity (14 million contracts on June 5, preceding months of tech corrections) suggests the equity side of the positive-correlation cascade is underway. Negative correlation (growth-driven) would require a decisive growth shock — the Michigan/NFIB data rules that out near-term — or a collapse in oil that takes the energy tail off the table. The deer point macro steepening thesis (normalize policy rate toward EFFR → steeper curve) is implicitly a positive-correlation trade: it positions for a hawkish Fed that drives front-end rates up while fiscal/AI supply keeps long-end elevated.

Risk-budget implication:

• Underweight nominal duration (2-5 year) entirely until the September hike uncertainty resolves. BofA’s short 2-year (target 4.40%, current 4.14%) has favorable asymmetry: the 53.5% September hike probability means 25bp of tightening is more than half-priced, but BofA expects three. Shorting 2s yields simple carry + roll if BofA is right; a September hold surprises would hurt. Consider 2-year receiver swaptions collared against a short 2-year futures to cap the downside.
• Overweight the 7-year sector steepener — Deer Point Macro’s recommendation of carry-rich steepeners in the belly ^[7][8] aligns with the policy-rate normalization narrative. The 7-year sector offers positive carry/roll and benefits from Fed front-end repricing. The recent flattening improves the entry point.
• Underweight high-beta tech and semiconductors — J.P. Morgan’s leveraged ETF deleveraging mechanism (−13% AUM, storage chip −34%, retail options at historical highs) is a clear VaR-shock risk ^[12]. The SMH’s 9.5% June gain vs mega-cap cloud −14.5% ^[12] is a rotation into more vulnerable segments. Storage chip leverage density is ~3x average ^[12], meaning higher volatility amplification. The BOK rate hike and KOSPI −8.77% are a real-time EM tech deleveraging event that may spill over through the leveraged ETF channel.
• Overweight gold with a tactical caveat — BofA’s $4,250 Q4 2026 and $5,000 mid-2027 targets ^[1] provide a medium-term anchor, but gold’s near-term weakness (−2.53% weekly, below $4,000) ^[6] reflects rising real yields and the dollar’s resilience. Gold will only break out if the oil surge (which supports stagflation) combines with a Fed credibility failure (e.g., a September hold that the market reads as reactive, which would weaken the dollar). Use cheap call options on gold with strikes at $4,200-$4,500 expiring Q4 2026.
• Underweight the USD from a structural perspective — BofA is bullish USD (USD-JPY target 152 by year-end from 162) ^[2], but the BIS hedge fund Treasury liquidation risk, the BOK hike (which reduces the US rate differential premium), and the extreme cash outflow ($119.6bn in one week) are all dollar-negative in a risk-off scenario. A short USD position should be paired with long EUR or JPY, not a clean short. The ECB/BOE are tightening into a Fed that is at least debating a hold — that rate-differential compression is medium-term dollar-negative.

6. Contrarian & Tail Risks

• Consensus fragility — BofA’s three-hike call is a bimodal bet against the market’s ~1.5 hike pricing: BofA explicitly states the market is in a “low-probability intermediate zone” — either no hikes or 3+ hikes ^[1]. The gap between BofA’s 75bp and the market’s ~38bp is a binary risk. If BofA is right and the market has to reprice up (2-year yields from 4.14% toward 4.40-4.50%), both bonds and equities will sell off. If BofA is wrong and the September hike does not materialize, the 2-year would rally back toward 4.00%, equities would get a relief rally, and gold would rally. The asymmetry in positioning (cash at 3.6%, cash outflow extreme, Bull & Bear at 9.6) favors the hawkish surprise — the market is already fully invested.

• Consensus fragility — leveraged ETF deleveraging as a self-correcting mechanism: J.P. Morgan’s report warns that leveraged equity ETFs have an inherent “self-correcting mechanism” and that the de-leveraging may take about three months ^[12]. With AUM down 13% already and storage chips down 34%, this is not a one-time event — it could continue into October. Historical parallels (retail call option activity at 14 million on June 5, near 2021 and 2025 highs that preceded months of tech corrections) suggest the semiconductor selloff has further to run. Given the BOK rate hike that triggered the KOSPI −8.77% and the concentration of Korean tech firms in the global memory supply chain, the spillover to US-listed storage chip ETFs is the key vector.

• Consensus fragility — the BIS hedge fund Treasury position warning: The BIS flagged hedge funds’ $2.5tn Treasury position doubling as “one of the most troubling financial stability risks in the world today” ^[9]. A sudden unwind in a rising-rate environment (BofA short 2s, curve steepening — which makes the carry in long positions negative) could cause a “sudden” leverage-unwind scenario per Torsten Sløk ^[9]. This is a systemic risk for all bonds and would push yields sharply higher, not lower — the opposite of a flight-to-safety.

• Consensus fragility — lag structure: oil surge to $88 is not in June CPI: The June CPI captured gasoline prices during a May-June window before the Iran escalation pushed Brent from ~$76 to $88. July CPI data (due August 13) will show the energy passthrough. With energy costs up ~16% from mid-May, July headline CPI could re-accelerate, breaking the disinflation narrative entirely. Warsh’s “zero tolerance” and BofA’s pre-positioned three-hike call would gain immediate validation.

• Consensus fragility — 83% of fund managers expect no pre-midterm hike: BofA’s fund manager survey shows 83% expect no rate hike before the November midterms ^[10]. If the September hike occurs, this is a violent positioning unwind: cash at 3.6% (BofA sell signal), Bull & Bear at 9.6, tech fund inflows record $48.8bn over three weeks. The consensus is fragile and directional.

• Second-order — AI hyperscalers’ debt as a mirror of UK gilt risk: The BOE’s observation that AI companies “borrowed more in H1 2026 than in all of 2025” and accumulated “as much new debt as the UK government” ^[9] should be read through the 2022 LDI crisis lens. A 50bp+ rise in 10-year yields would trigger margin calls on AI-company hedges the way US pension swaps triggered the 2022 gilt crisis. The BIS hedge fund Treasury warning is the same mechanism: levered bond positions that break when yields rise.

• Second-order — KOSPI crash as a canary for global tech leverage: The KOSPI’s −8.77% weekly decline, led by SK Hynix (−11% in a single day) ^[11][6], combined with the BOK rate hike and the JP Morgan leveraged ETF report, point to a global technology sector deleveraging chain. Storage chip leverage density is ~3x average — any VaR shock from a large Korean institutional investor forced to liquidate will propagate to US-listed semiconductor ETFs. The SMH’s June 9.5% rise vs mega-cap cloud −14.5% suggests the rotation into higher-beta semiconductors is self-correcting through JPMorgan’s “correcting mechanism.”

• Source quality control: The BofA reports ^{[10][1][2][3]} are primary institutional research (BofA Merrill Lynch). The J.P. Morgan flows & liquidity report ^[12] is secondary (via 硅谷宇宙) but well-attributed and contains detailed numbers (AUM −13%, retail options 14 million, storage chip leverage 3x). The Financial Times article ^[9] is a primary wire and the most authoritative source on the BIS hedge fund warning, AI debt, and Warsh communication analysis. The 华创证券 data ^[4][13] is primary Chinese sell-side research. The CME FedWatch data ^[5] is official. The @deerpointmacro tweets ^[7][8] are single-source and unverified — treat as analytical notes, not facts. The Christophe Barraud post ^[14] is a social/unverified single source.

Appendix: Additional Sources

• ^[15] 金十数据 — Warsh reduced communication; Wall Street uses AI to analyze thousands of documents
• ^[16] 华尔街见闻 — 2-year yield up 75bp since February; Sherman says market already tightened; Chi Chen says market pricing more hawkish than expected
• ^[17] Mohamed El-Erian — Weekly note on divergence, oil, central banks
• ^[18] 华创证券 — China high-frequency data (no Fed relevance)
• ^[14] Christophe Barraud — Week ahead: PMI data, ECB decision

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