Does Warsh's higher-for-longer turn remove the rate-cut NII overhang priced into cash-sweep-heavy brokers (SCHW, IBKR) and force an upward re-rate?

A net-new chain surfaced in the DAILY step-2 financials bucket (2026-06-18). The market spent early 2026 pricing a Fed easing cycle and sell-side baked rate-cut haircuts into brokerage net-interest income (NII). Warsh's hawkish FOMC debut (2026-06-17) reverses that consensus — so the embedded NII-decline assumption in the cash-sweep-geared brokers is now wrong-footed.

The chain

1. Forcing function: Warsh's FOMC (2026-06-17) strips the easing bias, 9 of 18 participants project a 2026 hike, framed as "higher-for-longer regime change." (From 2026-06-18-autoresearch-bucket-warsh-higher-for-longer-brokerage-nii-insurer-float) — confirmed-grade (multi-source: FXStreet, TheStreet, CNN, beincrypto).
2. → Consensus had priced 2026 cuts; sell-side baked rate-cut NII haircuts into the brokers. A 25bp cut reduces IBKR NII by ~$108M, and BofA flagged future cuts as the key near-term risk. (From 2026-06-18-autoresearch-bucket-warsh-higher-for-longer-brokerage-nii-insurer-float)
3. → Removing/reversing the cuts removes the haircut: IBKR Q1-2026 NII $904M (+17% YoY); SCHW sweep deposit base (sweep APY 3.21% as of Jun 1) is the most rate-geared NII engine among brokers. (From 2026-06-18-autoresearch-bucket-warsh-higher-for-longer-brokerage-nii-insurer-float)
4. → SCHW / IBKR re-rate as the embedded NII-decline reverses to flat/up (⚠ unverified — the gap to research: that consensus still carries a 2026-cut NII haircut, and that the re-rate hasn't already happened in the YTD run).

Why it matters

Tradeable, US-listed, liquid: long SCHW (sweep-deposit NII gearing; already on EARNINGS.md), long IBKR (cleanest, quantified $108M-per-25bp sensitivity, growing accounts). The asymmetry: a consensus built on cuts that Warsh just cancelled. Distinct cluster from the AI-infra book (financials/rate-regime — a thin vertical per the breadth report).

• michael-howell in 2026-08-15-macro-voices-macrovoices-545-michael-howell-warsh-vs-the: "This everything bubble will inevitably come to an end and you're starting to see signs that liquidity is rolling over." Warsh-vs-markets framing: a cut-eager White House against a cycle that has already peaked. Fork noted, not adjudicated; still status: hypothesis.

• darius-dale in 2026-08-15-forward-guidance-the-growth-strategy-trapping-the-fed: run-the-economy-hot (Paradigm C) as the administration's choice since Apr-2025; success = stay in the growth phase without hitting the print phase. Tightrope for Treasury Secretary and Fed Chair. Fork with Warsh-hawkish talk noted, not adjudicated.

• darius-dale in 2026-08-20-podcast-macro-voices-macrovoices-546-darius-dale-darius-dale-for-potus: they are "already starting to tiptoe our way into paradigm D" (print / YCC). Same-week tension with the hike-path: Warsh talks hawkish, Treasury bills-for-duration is easy. Fork still not adjudicated; still status: hypothesis. See post-midterm-brake-on-debasement.

Why it may not work

• Two-way rate bet (weakest link): higher-for-longer helps the rate but cash-sorting (clients moving sweep cash to higher-yield products) shrinks the balance — most acute for SCHW. IBKR (less sweep-reliant, more margin-lending) is cleaner.
• Already priced: the brokers ran hard in early 2026 (IBKR +18% YTD by late Jan). The chain pays only if consensus still embeds a 2026-cut NII haircut for the surprise to remove.
• Dovish pivot: a growth scare that forces cuts after all removes the forcing function.

Update (2026-06-19) — the "peak-hawkishness" counter-read sharpens the weakest link

2026-06-19-podcast-forward-guidance-a-new-era-is-beginning-in-markets-weekly-roundup (Forward Guidance hosts Jack Farley / Quinn / Tyler, the week of Warsh's debut) reads the same FOMC the opposite way at the margin — "peak hawkishness may already be behind us." Their case: the hawkish 2026-hike dots came from (mostly non-voting) presidents and are reactive — oil is down ~30% from the prior dot plot, tariffs are now "net-zero" (refunding), break-evens and 1yr inflation swaps are "back to where the war started," and "headline inflation is going to be in the gutter next month." Quinn: "the market right now is pricing in two hikes by mid-2027… it's strange to me"; Farley: "there's no way in hell that we're hiking at the end of this year." They expect less-hawkish policy eventually.

This directly sharpens this hypothesis's weakest link (the "dovish pivot" falsifier + "already priced"): if the hawkish signal is a reactive head-fake and the Fed eases into disinflation, the NII-cut overhang this re-rate is supposed to remove comes back — defanging the trade. The chain is now explicitly two-sided: yesterday's joseph-wang read (regime change, higher-for-longer durable) vs today's Forward-Guidance read (peak hawkishness, disinflation incoming). The conversion trigger tightens to: does higher-for-longer hold past the July CPI print, or does oil-led disinflation force the cuts back in? Held status: hypothesis, priority toward low-medium — the forcing function is now contested, not confirmed.

One corroboration that survives both reads (relevant to the sibling private-credit chain): Tyler — "high-yield credit spreads barely budged this whole thing. The CapEx cycle is completely intact" — credit spreads at lows still funding the AI buildout, consistent with ai-capex-derate-to-private-credit-contagion's mark-lag (credit stress not in spreads yet) and the SemiAnalysis capex-intact read.

Update (2026-06-29) — two-sided again: no-cut framing supports the NII thesis, but the hawkish-hike trigger is removed; the regime is data-contingent, not structural

Two new reads, pulling in opposite directions — net keeps this a low-medium, explicitly two-sided hypothesis.

• Supports the no-cut / higher-for-longer side (the NII-overhang stays removed). From 2026-06-26-podcast-the-compound-and-friends-too-early-to-get-off-the-wave-with-ryan-detrick: "count the votes, not the dots … I don't think they're going to hike … they're basically going to run it hot." I.e. no cuts (which is what the brokerage-NII re-rate needs — the embedded cut-haircut stays cancelled) — but also no hike (the hawkish-surprise upside fades). For the NII thesis specifically, "run it hot / no cuts" is the supportive read: the rate stays elevated, so the cut-haircut the re-rate is supposed to remove doesn't come back.
• But the hawkish hike trigger is removed (the oil-driven urgency is gone). Same source: "Crude oil just fell from 120 to 70. So if that was the urgent reason to hike, you don't have to worry about that anymore." The 2026-hike dots that powered the aggressive version of this thesis lose their forcing function. lyn-alden in 2026-06-25-podcast-macro-voices-macrovoices-538-lyn-alden-is-the-war-really-over reinforces that the hawkishness is data-contingent, not structural: "if they start to show that inflation's rolling over, we might start to see more patient language."

Net (consistent with the 06-19 two-sidedness): the no-cut outcome supports the brokerage-NII re-rate (the overhang stays removed), but the framing is now explicitly data-contingent, not a structural regime change — oil-led disinflation could pull the cuts back in (the original "dovish pivot" falsifier + the Forward-Guidance "peak hawkishness" read). The hawkish-hike upside is largely gone. Held status: hypothesis; priority medium → low-medium; the load-bearing fork remains "does higher-for-longer survive the July CPI print, or does oil-led disinflation force the cuts back in."

Update (2026-07-14) — a same-regime first-party NII datapoint (bank, not broker)

jeremy-barnum (CFO, JPMorgan) in 2026-07-14-earnings-jpm-q2-fy2026 raised JPMorgan's FY2026 NII guide to ~$105.5B total / ~$96.5B ex-Markets — the largest US bank lifting, not cutting, its rate-sensitive NII outlook. JPM is a deposit bank (not a cash-sweep broker), so this corroborates the regime (higher-for-longer NII holding/being revised up), not the SCHW/IBKR cash-sweep mechanics specifically. It supports the "no-cut / overhang-stays-removed" side of this two-sided hypothesis; it says nothing about whether the broker consensus still carries a 2026-cut haircut left to remove.

Update (2026-07-20) — IBKR's per-25bp sensitivity corroborated; SCHW sweep mechanics + a second earnings lever; TWO new beneficiary legs (custodians, exchanges)

An autoresearch pass on brokerage/bank NII under higher-for-longer (2026-07-20-autoresearch-brokerage-bank-nii-rate-regime) sharpens this chain on three fronts and widens the beneficiary set beyond SCHW/IBKR.

• IBKR — Step 2's number tightens and gets management-attributed. The chain previously carried "a 25bp cut reduces IBKR NII by ~$108M." The source restates it as management's own estimate: a 25bps Fed cut reduces annual NII by $80M ($117M including other benchmark rates). Q1 2026 NII +17% YoY to $904M, margin loans and idle-cash credits each ~+35% YoY, accounts +31%. This corroborates the load-bearing "no cuts preserves the run-rate" assumption first-party-ish (management-quantified). See interactive-brokers.
• SCHW — the sweep mechanics, plus a second earnings lever the chain didn't name. Schwab pays just ~0.19% on transactional sweep cash (~$461.5B, ~9.6% of client assets, +14% YoY on the deposit line) — and, critically, elevated rates let it pay down high-cost bank supplemental borrowings, an earnings lever beyond the raw sweep spread. Bear case is unchanged (cash-sorting / de-sweeping, with a jpmorgan AI cash-management tool as the competitive threat). ⚠ The SCHW figures here are snippet-sourced (primary returned 403) — treat as needing corroboration against the Q2 print. See schwab.
• Two NEW beneficiary legs this question didn't cover. The same forcing function benefits custodians (state-street, bny-mellon NII +18% to $1.4B, northern-trust) that reinvest client cash, and exchanges/clearinghouses (cme-group, intercontinental-exchange) that earn near-costless interest on mandatory margin/clearing collateral. These are filed as a sibling concept — spread-on-other-peoples-cash — and mechanism — client-float-interest-to-custodian-exchange-nii. The two-sided qualifier: custodian deposit betas are reported >100% (sophisticated institutional clients), so the custodian NII benefit is muted; the exchange-collateral leg (near-zero beta, mandatory) is the cleanest expression. regions-financial (retail-funded, mid-30s beta) sits at the opposite, high-retention end and quantifies the regime (FY NII +2.5–4%) — also snippet-sourced (SEC EDGAR 403), treat as needing corroboration.
• Contradiction flagged (not merged): the source's JPM Q2 figures (net income $21.2B, EPS $6.14 on $58.02B rev, "highest quarterly profit ever") are snippet-sourced (techtimes) and disagree with the first-party jpmorgan entity built from 2026-07-14-earnings-jpm-q2-fy2026 (net income $16.9B, EPS $7.70 on $57.35B rev). Trust the earnings call; the snippet number is not folded into the JPM entity.
• Direct near-term test: SCHW and IBKR both report in the 2026-07-21 window — watch SCHW sweep-balance QoQ, the rate paid, the pace of bank-supplemental-borrowing paydown, and whether the IBKR/SCHW guides still carry a 2026-cut haircut left to remove.

Update (2026-07-21) — SCHW Q2 print day: the NIM story is deleveraging, not rate-beta (actuals pending)

SCHW reported Q2 2026 pre-market 07-21 (consensus EPS $1.55 / rev $6.89B / NIR ~$3.28B on record $13.14T client assets). The chain-relevant structural fact: Schwab's NIM was already expanding — 2.88% in Q1 2026 vs 2.53% YoY — driven by wholesale-funding paydown + lower funding-cost rates, not by the rate level. That reframes the long-SCHW case as a self-help/deleveraging NIM tailwind that survives even without cuts (and is arguably strengthened by a higher-for-longer Warsh Fed keeping reinvestment yields high while funding costs are managed down). Deposits/NII have stabilized since the 2023 episode. Caveat: the actual Q2 print (sweep balance QoQ, cash-sorting commentary) could not be fetched results-day morning (IR PDF 403, wires unindexed) — confirm via earnings-ingest next run. From 2026-07-21-schw-q2-2026-brokerage-custodian-nii-chain.

What to watch (evidence to convert to active)

• A sell-side NII model that still embeds 2026 cuts (the haircut to be removed) — confirm the overhang exists before claiming it's removed.
• SCHW deposit-balance trend (cash-sorting): are sweep balances stabilizing or still bleeding? Stabilizing = the long-SCHW case; bleeding = prefer IBKR.
• Next prints: SCHW / IBKR Q2 2026 NII guidance vs the Warsh-reset rate path.
• Does higher-for-longer survive the July CPI print, or does oil-led disinflation force the cuts back in (the peak-hawkishness counter-read)? Now the load-bearing fork.

Update (2026-07-21) — SCHW Q2 print resolves the mechanism: the re-rate driver is lending-mix self-help, not rate-beta — which makes it MORE durable, not less

The SCHW Q2 2026 call (2026-07-21-earnings-schw-q2-fy2026, CFO mike-verdeschi) materially strengthens the durability of this thesis while re-attributing its engine. The confirmed print: revenue $7.1B (+21%), adj EPS $1.62 (+42%), NII +19% YoY. Crucially, Verdeschi attributes the YoY NIM expansion "primarily [to] lending activity" — PAL originations +59% YoY, bank loans $67B (+33%) at >100bps incremental spread — not to rate-beta. FY guide: NIM 3.0–3.10% (Q4 exit 3.25–3.30%) on an assumption of only one December hike (no 2026 P&L impact), and "if rates resume a hiking pattern you'll see even more expansion."

This partly dissolves the two-sided fork that had held this at low-medium: the re-rate no longer depends on higher-for-longer holding, because a self-help / deleveraging + lending-penetration tailwind expands NIM even in the peak-hawkishness-is-behind-us / no-hike world. Higher-for-longer becomes upside optionality on top, not the load-bearing driver. Weakest link (step 4) is now substantially evidenced on the SCHW leg: the NII expansion is real, printed, and structurally sourced. The cash-sorting risk is also muted in the print — sweep cash grew (+$24.2B, partly long-short-strategy driven). Net: the SCHW leg graduates toward active; IBKR leg still open. Priority nudged low-medium → medium.

Update (2026-08-14) — Dillian's intentional-steepener read: Warsh holds funds, long end does the tightening

jared-dillian in 2026-08-05-podcast-forward-guidance-the-portfolio-built-to-survive-every-crash-jared inverts the "Warsh made a mistake / lost credibility" tape: "I think this was completely intentional, 100% intentional. I think … Warsh knew the curve would steepen a lot if he kept rates the same." Mechanism: "The long end got obliterated. And that actually is the monetary policy he was looking for" — a hike would have flattened (stimulative via mortgages); holding funds "had an immediate tightening effect" (10Y and mortgage rates up). Path: "the curve's going to steepen for the next six to 12 months" with "Fed funds come down to three" and "the long end stays pretty high." He does "not think Warsh is going to hike, period."

This is a new path on the same forcing function, not a new mechanism: short rates eventually down (NII haircut returns for cash-sweep brokers) while the long end stays high (tightening via mortgages / bank NIM via steepener). It re-opens the two-sided fork the SCHW Q2 self-help print had partly dissolved — if Dillian is right, the SCHW lending-mix tailwind still holds, but the "no-cut overhang stays removed" IBKR rate-beta leg is not the 6–12 month path. Held status: hypothesis. Interested (Dillian is talking a steepener). Portfolio-construction / gold-oil-defensive advice from the same interview is out of scope.

Sources

• 2026-09-07-autoresearch-jobs-print-162k-cpi-week-ferc-wuxi-still-no — BLS August NFP +162k vs ~53k; FedWatch ~58.4% hike at Sept 15–16. Still hypothesis. Do not graduate SCHW.
• 2026-09-08-research-reopen-week-gates — PPI Thu / CPI Fri confirmed; FedWatch 58.3% hike as of Sept 7. Still hypothesis. Do not graduate SCHW.
• 2026-09-09-research-event-day-gates — FedWatch 60.4% hike / 39.6% hold as of Sept 9. PPI tomorrow, same morning as the first doubled buyback. Still hypothesis. Do not graduate SCHW.
• 2026-09-10-research-event-day-gates — PPI not yet printed. FedWatch ~60.2/39.8. Still hypothesis. Do not graduate SCHW.
• 2026-08-31-autoresearch-warsh-jackson-hole-keynote — delivered keynote (CNBC PDF of prepared remarks + CNBC recap). "Work to do" / no reaction function. Do not graduate.
• 2026-08-31-podcast-odd-lots-richmond-fed-s-tom-barkin-on-the-surprisingly — tom-barkin at the Lodge after the speech. Color, not a conversion.
• 2026-08-28-autoresearch-warsh-jackson-hole-morning-preview — Reuters preview of Warsh's first Jackson Hole keynote (10:00 a.m. EDT today). Funds held 3.50–3.75% since December; July PCE 3.7%. Schmid/Hammack hawkish color is not a Warsh reaction function. Keep hypothesis; do not graduate.
• 2026-08-27-podcast-forward-guidance-druck-calls-out-bessent-will-jackson-hole-derail — FG speaker B: Warsh will not hike the front end; cannot attack Bessent on the long end at Jackson Hole without "throwing gasoline." Fork, not a close.
• 2026-08-25-autoresearch-warsh-jackson-hole-real-yield-decomposition — FRED DFII10 2.40% (08-21); Cleveland Fed real +53 bp vs inflation +16 bp YTD; Warsh first JH keynote Fri 08-28. Keep hypothesis; do not graduate.
• 2026-08-24-podcast-the-compound-and-friends-why-every-trader-on-earth-is-watching-the-10-year — nick-colas: 30-year decomposed into flat inflation expectations vs rising real residual.
• 2026-08-20-autoresearch-warsh-hold-through-2026-brokerage-nii — UBS hold-through-2026 + hike-pricing-too-aggressive; Jackson Hole 08-29 mute (calendar superseded 08-25 by Fri 08-28 keynote); core CPI 2.5%. Keep hypothesis.
• 2026-07-21-earnings-schw-q2-fy2026 — SCHW Q2 first-party call: NIM expansion attributed to lending-mix (PAL +59%, bank loans +33%) not rate-beta; FY NIM 3.0–3.10% on a one-December-hike assumption; "if rates resume a hiking pattern you'll see even more expansion." The self-help reframe that de-risks this thesis's rate-dependence.
• 2026-06-19-podcast-forward-guidance-a-new-era-is-beginning-in-markets-weekly-roundup — Forward Guidance hosts' "peak hawkishness behind us" counter-read (oil −30%, tariffs net-zero, disinflation incoming) + "credit spreads barely budged, capex cycle intact."
• 2026-06-26-podcast-the-compound-and-friends-too-early-to-get-off-the-wave-with-ryan-detrick — Detrick: "count the votes, not the dots … they're basically going to run it hot" (no cut, no hike) + "crude fell 120→70, so [the] urgent reason to hike … you don't have to worry about that anymore" (removes the hawkish-hike trigger).
• 2026-06-25-podcast-macro-voices-macrovoices-538-lyn-alden-is-the-war-really-over — Alden: hawkishness is data-contingent — "if … inflation's rolling over, we might start to see more patient language."
• 2026-06-18-autoresearch-bucket-warsh-higher-for-longer-brokerage-nii-insurer-float
• 2026-07-21-schw-q2-2026-brokerage-custodian-nii-chain — SCHW Q2 print day: NIM 2.88% (from 2.53%) on wholesale-funding paydown (deleveraging tailwind, not rate-beta); deposits stabilized since 2023; actual Q2 print pending (IR PDF 403).
• 2026-07-20-autoresearch-brokerage-bank-nii-rate-regime — IBKR per-25bp NII sensitivity (~$80–117M, management-quantified); SCHW sweep mechanics (~0.19% paid) + bank-supplemental-borrowing paydown lever; new custodian (STT/BK/NTRS) and exchange (CME/ICE) beneficiary legs; custodian deposit betas >100%. SCHW/RF figures snippet-sourced (403s) — flagged for corroboration.
• 2026-06-17-podcast-forward-guidance-the-warsh-fed-will-look-nothing-like-before — joseph-wang (former NY Fed trader) primary read of the forcing function: "the committee will deliver price stability, period… a mic-drop moment… the market is now pricing in a little bit more than one hike this year… the entire curve basically shifted upward." Confirms higher-for-longer is the regime, not just a forecast.
• 2026-08-05-podcast-forward-guidance-the-portfolio-built-to-survive-every-crash-jared — jared-dillian: Warsh hold = intentional steepener (long-end tightening); funds to ~3%, long end stays high; will not hike.

Related

• p-and-c-insurer-float-income-rate-regime — sibling rate-regime beneficiary (float vs sweep NII); both keyed to Warsh higher-for-longer.
• spread-on-other-peoples-cash — the unifying beneficiary-basket concept (sweeps + custody deposits + exchange collateral), with the deposit-beta two-sidedness.
• client-float-interest-to-custodian-exchange-nii — the custodian/exchange leg as a filed mechanism (the new beneficiaries this question doesn't cover).
• ai-capex-derate-to-private-credit-contagion — the risk side of the same forcing function (higher-for-longer stresses cash-flow-optimism credit).
• dealer-balance-sheet-as-market-constraint — adjacent rate-regime market-structure chain.
• jared-dillian
• global-liquidity-rollover-to-2027-rebound

Update (2026-08-15) — Treasury-dovish layer under Warsh-hawkish talk

jack-farley / tyler-neville in 2026-08-15-forward-guidance-washington-is-suppressing-volatility argue vol control has moved to Treasury (ESF/FIMA yen intervention; QRA language flip to "changes" = decrease-tail). Warsh still talks hawkish; Bessent acts dovish on the long end. This is the same two-sided fork Dillian re-opened 2026-08-14 — not adjudicated. Held status: hypothesis. See treasury-vol-suppression-to-ai-capex-statecraft.

Update (2026-08-20) — UBS: hold through YE2026; hike pricing "too aggressive"; July CPI is in

From 2026-08-20-autoresearch-warsh-hold-through-2026-brokerage-nii (UBS this week): July FOMC left funds at 3.50–3.75% with three hawkish dissents; Warsh "watchful thinking." July core CPI 2.5% (from 2.6%). UBS base case: no move the rest of 2026, 2027 easing pivot; market hike pricing over the coming year is "somewhat too aggressive." Forward guidance already stripped in June; Warsh has not submitted dots. Jackson Hole 2026-08-29 — AGBI: do not expect reaction-function hints. 10y ~4.70%; 30y >5.2%.

Two-sided, still hypothesis. The 06-18 forcing function (no 2026 cuts) still holds. The 06-19 "peak hawkishness / hike pricing is the overshoot" counter just got named-house corroboration. The conversion bar moves from July CPI (now printed: hold, not hike) to Jackson Hole 08-29 / September dissent count. Do not graduate. Do not re-rate SCHW/IBKR. Private-credit $1.4T+$1.4T sizing attaches to fomc-private-credit-outflows-alt-managers / bdc-redemption-spiral-to-private-credit-repricing — not a new chain.

Update (2026-08-25) — real yields did the work; Warsh's first Jackson Hole keynote is Friday 08-28. Still hypothesis.

From 2026-08-25-autoresearch-warsh-jackson-hole-real-yield-decomposition (FRED DFII10 + Cleveland Fed via Piedmont Crescent, 2026-08-23):

• 10-year TIPS real yield 2.40% on 2026-08-21 (FRED DFII10, updated Aug 24). Cleveland Fed YTD: expected-inflation +16 bp vs real rate +53 bp; market 10-year TIPS +44 bp while the 10-year breakeven fell 6 bp. FOMC June/July minutes already said higher nominal yields "reflected higher real rates." 30-year closed ≥5% for 34 consecutive sessions and touched 5.34%.
• This independently corroborates nick-colas on The Compound (2026-08-24): the 30-year breakout is a real-yield residual with inflation expectations "dead flat" for ~15 years. Colas is speaker color on the same spine, not a second chain.
• Dated catalyst is Friday 2026-08-28 — Chair Warsh's first Jackson Hole keynote (KC Fed symposium Aug 27–29, theme "Financial Innovation"). Slot reported ~10:00 a.m. ET; confirm on the KC Fed agenda Aug 26. Nineteen days before the Sept 16 FOMC. Score the speech on MOVE >70 through Sept 16 and forecast dispersion, not on the 30-year print the day of. Prior "08-29 mute" calendar from the 08-20 UBS pass is superseded by the Friday keynote date.
• ⚠ Unpriced collision, not a graduation: Treasury doubled long-end buybacks (Aug 19; Bessent) while a Warsh task force is examining a shorter SOMA — Operation Twist run backwards against the buyback window (Piedmont). Attaches to treasury-buyback-twist-to-hard-asset-debasement Step 1 (already confirmed on size/calendar). Do not emit GLD/IBIT/XLE off this preview.

Still hypothesis. Do not graduate SCHW/IBKR. The forcing function (higher real rates / no-cut) is stronger on this tape. The conversion tests are unchanged and still unmet: a sell-side NII model that still embeds 2026 cuts was not fetched this pass; SCHW sweep vs IBKR 25bp last resolved at Q2 as lending-mix self-help. Jackson Hole can move the term premium without resolving that arithmetic. SCHW marked −0.8% off its 52w high on 08-24 closes — no valuation gap to chase.

Update (2026-08-28) — the keynote is this morning; this run is still a preview. Stay hypothesis.

From 2026-08-28-autoresearch-warsh-jackson-hole-morning-preview (Reuters, 2026-08-28) plus 2026-08-27-podcast-forward-guidance-druck-calls-out-bessent-will-jackson-hole-derail:

• Chair Warsh's first Jackson Hole keynote is today at 10:00 a.m. EDT. This 05:00 run is before the speech. Next FOMC is September 15–16. Jackson Hole is not a rate meeting.
• Funds have been held in 3.50–3.75% since December. July PCE 3.7%, "steady but well above target." Kansas City Fed President Jeffrey Schmid (host): policy is not restricting spending/investment; inflation "still stubborn." Cleveland Fed President Beth Hammack: "now is the time to act." Boston Fed President Susan Collins more mixed. That is regional-president color, not a Warsh reaction function.
• FG unnamed co-host (speaker B): Warsh will not hike the front end; the hawkish avenue is a smaller, less-duration SOMA, but he cannot attack Bessent on the long end at Jackson Hole without "throwing gasoline." jack-farley independently: the US long-end move has been term premium, not rate expectations — a confidence/fiscal read, not a hike-path.
• Conversion tests unchanged and still unmet. Do not graduate. Do not re-rate SCHW/IBKR off a preview. Score the speech after 10 a.m. on MOVE and on whether he describes a reaction function.

Update (2026-08-31) — the speech was delivered. He refused a reaction function. Stay hypothesis.

From 2026-08-31-autoresearch-warsh-jackson-hole-keynote (CNBC-hosted prepared remarks, 2026-08-28; CNBC recap):

• Title "In Our Time," 100th day as Chair. He told the audience they could call it an outline or a trail map, "just don't call it forward guidance." He quoted the request for an explicit reaction function / Taylor-style rule back and declined: knowledge "just doesn't extend that far." Closer: "I stand here today committed to a discipline, not to a decision."
• Inflation: twelve-month PCE 3.7 percent; six-month 4.1 percent. Summer prints "better than expected" but "do not tell me that underlying trends have meaningfully improved." Standard: "We must be confident that underlying inflation is moving to our objective, clearly and at sufficient speed. Otherwise, we have work to do." Two percent PCE is a "firm, fixed target" and "not self-executing."
• Activity: equipment and intangibles ~9 percent; "more than half of the cap-ex growth this year can likely be ascribed to the buildout related to AI." Unemployment 4.1 percent. "I would be hard pressed to describe broad financial conditions as restrictive." He did not mention Bessent buybacks.
• Market: two-year +8 bp to 4.31 percent; CME FedWatch September hike odds 55.7 percent (~+20 ppt). Street color split (hike door by October/December vs wait for another print).
• Conversion bar was a reaction function. He named it and declined. Dated catalyst is now FOMC September 15–16. Sell-side NII models that still embed 2026 cuts were not re-fetched. Do not graduate. Do not re-rate SCHW/IBKR.

tom-barkin in 2026-08-31-podcast-odd-lots-richmond-fed-s-tom-barkin-on-the-surprisingly (recorded at the Lodge): "If you choose not to do forward guidance, you don't do forward guidance." On the 65-month inflation clock: a "47 month view followed by an 18 month view" is "a perfectly defensible" split (shocks after March 2025), vs "don't give me your excuses. It's been 65 months… And maybe rates aren't that restrictive." Color on the same fork, not a conversion. joe-weisenthal: Warsh "described policy as not restrictive" so the tape fills in hikes — "But he didn't quite say that."

Update (2026-09-02) — FOMC is still the conversion test. Stay hypothesis.

From 2026-09-02-autoresearch-wuxi-dod-answer-extended-sept-18 (also-checked, financials bucket): no new Warsh reaction function. Next dated test remains FOMC September 15–16. Do not graduate SCHW.

Update (2026-09-03) — Haber: maybe one Q4 hike as hawkish theater. Still hypothesis.

bob-haber in 2026-09-03-forward-guidance-fiscal-dominance-is-breaking-the-60-40-portfolio: "We may in this, let's call it fourth quarter, get the Fed to raise once as kind of a … attempt to show that we are … hawkish." He thinks Bessent and Warsh "believe that we're going to AI our way out of this problem" via productivity growth, would be "really shocked if it comes before the election," and would treat a hike as a hard-asset buying opportunity. matt-hougan same episode: the Fed is "not the big dog"; this is a fiscally driven environment; agrees a hike would be an opportunity. Practitioner color on the existing two-sided fork (hawkish theater vs AI-grow-out), not a reaction function and not a conversion. Next dated test remains FOMC September 15–16. Do not graduate SCHW. From 2026-09-03-autoresearch-apple-sept-9-and-treasury-buyback (also-checked): no new Chair print.

Update (2026-09-04) — August jobs print is this morning. Still hypothesis.

From 2026-09-04-autoresearch-jobs-print-morning-ferc-wuxi-18a-still-no: BLS August employment is today at 8:30 a.m. ET. This 05:00 run is before the print. ING named this print and September 11 CPI as the hike/hold gates into FOMC September 15–16. A calendar is not a reaction function. Do not graduate SCHW. Do not re-rate off a print that has not happened.

Update (2026-09-07) — the print is in: +162k vs ~53k. Still hypothesis.

From 2026-09-07-autoresearch-jobs-print-162k-cpi-week-ferc-wuxi-still-no: BLS August nonfarm payrolls +162,000 vs ~53k consensus; unemployment 4.1%; participation +0.2 pp; June+July revised +55k; food services +59k; local-gov education +42k; information −23k (CNBC jobs; BLS HTML 403 from this env). Monday CME FedWatch ~58.4% for a 25 bp hike at September 15–16 (CNBC gold). Direction is tighter. Conversion is still the FOMC statement, not a single payrolls print. ING's second gate — CPI — is this week (PPI Thursday Sep 10 / CPI Friday Sep 11). Do not graduate SCHW.

Update (2026-09-08) — first session after Labor Day. Odds still a coin-flip. Still hypothesis.

From 2026-09-08-research-reopen-week-gates: BLS calendar confirms PPI Thursday Sep 10 and CPI Friday Sep 11 (both 8:30 a.m. ET). CME FedWatch via Gate 58.3% hike / 41.7% hold as of Sept 7. Schwab's last disclosed scenario still assumed a December hike, not September. Futures odds are not a reaction function. Conversion remains the FOMC statement. Do not graduate SCHW.

Update (2026-09-09) — odds ticked to 60.4%. Still a coin-flip. Still hypothesis.

From 2026-09-09-research-event-day-gates: CME FedWatch via BlockBeats 60.4% hike / 39.6% hold as of Sept 9 (a small tick from yesterday's 58.3%). PPI is tomorrow (same morning as the first doubled Treasury buyback operation). CPI Friday. Conversion remains the FOMC statement September 15–16. Do not graduate SCHW.

Update (2026-09-10) — PPI has not printed. Odds still ~60%. Still hypothesis.

From 2026-09-10-research-event-day-gates: this pass is before the 8:30 a.m. ET August PPI print. PANews citing CME FedWatch: 60.2% hike / 39.8% hold. FXStreet (06:40 GMT) had 61.2% ahead of the print; consensus 5.3% YoY. Schwab's last disclosed scenario still assumed a December hike. Conversion remains the FOMC statement September 15–16. Do not graduate SCHW.

Update (2026-09-11) — PPI printed in-line. CPI has not. Still hypothesis.

From 2026-09-11-research-event-day-gates: August PPI +0.4% m/m / +5.4% y/y (matched the monthly consensus); core ex-food-and-energy +0.2% m/m (a tick under some 0.3% prints); energy +4.2% m/m on renewed US–Iran hostilities (Haver; Reuters / Journal Record). FedWatch ~62% hike ahead of that print. CPI for August prints today, 8:30 a.m. ET — this pass is before it. Conversion remains the FOMC statement September 15–16. An in-line PPI is not a reaction function. Do not graduate SCHW.

Update (2026-09-14) — August CPI printed. Core a tick hot. Still hypothesis.

From 2026-09-14-event-day-gates-august-cpi-printed-fomc-tomorrow-siri-beta (BLS CPI, September 11; Reuters / Mutikani): headline +0.4% m/m / +3.4% y/y (index 334.980), matching estimates; core +0.3% m/m (above a 0.2% consensus) / 2.4% y/y. Gasoline +3.9% m/m (more than a third of the monthly rise); energy +2.1% m/m / +16.3% y/y; shelter +0.3%. FedWatch jumped from ~70% Thursday to ~87% (some prints 91% then settled) for a 25 bp hike at September 15–16. Funds still 3.50–3.75%. Next CPI: October 14.

jack-farley in 2026-09-14-forward-guidance-the-bond-market-is-trapping-the-fed-weekly (recorded Thursday, pre-CPI): PPI core came in 0.2% vs 0.3% consensus while headline was hot; FedWatch ~70% hike before CPI; the 10-year vs NGDP regression he walks is a fair-value sketch around 5.8%. Pre-print color, not a conversion.

jens-nordvig in 2026-09-14-the-compound-and-friends-the-most-interesting-macro-moment: long-end already at multi-decade highs even if funds are not; hyperscaler capex still accelerating so issuance beta rises; Bessent intervening in both yen and the long-end. Independent practitioner color on the same two-sided fork — not a reaction function.

Futures odds are not a reaction function. Schwab's last disclosed scenario still assumed a December hike. Conversion remains the FOMC statement, September 15–16 (tomorrow). Do not graduate SCHW. Do not re-rate SCHW/IBKR off the print.

Update (2026-09-15) — FOMC day 1. Statement is tomorrow. Still hypothesis.

From 2026-09-15-event-day-gates-fomc-day-1-statement-tomorrow: the meeting started today. TradingKey and C-SPAN put the statement, SEP, and Warsh press conference on Wednesday September 16 (2:00 / 2:30 p.m. ET). Futures still ~87–93% for a 25 bp hike to 3.75–4.00%. CNBC, September 14: Jackson Hole "work to do." Boston Globe / Hassett: White House prefers no hike into the midterms.

A meeting's first day is not a reaction function. Conversion remains Wednesday's statement. Do not graduate SCHW. Do not re-date a SCHW signal that is not live.

Update (2026-09-16) — FOMC day 2. Statement is this afternoon. Still hypothesis.

From 2026-09-16-event-day-gates-fomc-statement-this-afternoon (Newsquawk; Kiplinger live; CryptoTimes): statement, SEP, and Warsh press conference today 2:00 / 2:30 p.m. ET — not yet printed. FedWatch ~92.7% for a 25 bp hike to 3.75–4.00% (from 93.5% Monday). Funds still 3.50–3.75%. Warsh expected to skip his own SEP/dots. Minutes October 7.

A preview is not a reaction function. Conversion remains this afternoon's statement. Do not graduate SCHW. Do not re-date a SCHW signal that is not live.

Update (2026-09-17) — the hike printed. Conversion bar is still a sell-side NII haircut. Stay hypothesis.

From 2026-09-17-event-day-gates-fomc-hiked-warsh-words (primary: FOMC statement; SEP tables; Warsh statement PDF): 12–0 hike 25 bp to 3.75–4.00%. IORB 3.65%, ON RRP 3.50%, effective 18 September. Median SEP funds 4.1% at year-end 2026 and year-end 2027. PCE 3.7% (2026) → 2.3% (2027). Chair Warsh: "I would be hard-pressed to describe current financial conditions as restrictive." And: "The Committee's 2 percent inflation objective is a commitment, not a forecast. That standard has not been satisfied."

The forcing function printed (higher-for-longer, not a one-and-done). The conversion bar did not: no new sell-side NII haircut on SCHW/IBKR consensus, and Schwab's last disclosed scenario still assumed a December hike. A 12–0 vote is a reaction function for funds; it is not a brokerage re-rate. Do not graduate SCHW. Do not re-date a SCHW signal that is not live.

dean-curnutt in 2026-09-17-podcast-forward-guidance-the-market-is-mispricing-a-correlation-shock-dean (recorded Monday September 14, pre-statement): optionality "out of step too low relative to the uncertainties"; the 10-year "is the risk asset." Pre-print color. Does not change the conversion bar.

Update (2026-09-18) — Bassman: the hike is credibility, not economics. Stay hypothesis.

harley-bassman in 2026-09-18-macro-voices-macrovoices-550-harley-bassman-in-fed-we-trust (Macro Voices #550, recorded Wednesday afternoon after the hike): the 25 bp print plus a second hike before Christmas is "not a good result"; it should have been "0 or 50." The load-bearing claim is that the hike is about trust, not the funds rate: "we have a lack of trust in the government in a lot of our institutions, and maybe about the Fed" and "the Fed needs to go and regain the trust of the market." Rates have risen 150 bps while 10-year TIPS breakevens sit at 2.34–2.35 — "inflation. No one cares about inflation." Hyperscaler bonds are "probably money good" (Meta, Google, Amazon, Microsoft, Oracle can cover coupons); "the equity I can't comment on." Non-bank private credit "reduces the ability of the Fed to transmit their policy."

This is independent practitioner color on the same forcing function that printed yesterday. It does not supply a sell-side NII haircut on SCHW/IBKR. Conversion bar unchanged. Stay hypothesis. Do not graduate SCHW.

Update (2026-09-21) — still no post-hike NII model. Stay hypothesis.

From 2026-09-21-autoresearch-financials-schwab-nii-still-hypothesis: nothing on Schwab IR or EDGAR this morning isolates post-hike net interest income. The Q2 call (July 21) is still the last primary: CFO Mike Verdeschi's scenario assumed one 25 bp hike in December, "minimal impact on 2026," expansion in 2027. The September 16 hike to 3.75–4.00% with median SEP 4.1% / 4.1% is earlier than that scenario. It is not a revised company or sell-side NII haircut. Conversion bar unchanged. Stay hypothesis. Do not graduate SCHW. Do not re-date a SCHW signal that is not live.

Update (2026-09-24) — Tuesday −6% tape is restated Q2 guide, not a post-hike NII model. Stay hypothesis.

From 2026-09-24-autoresearch-financials-consumer-transport-buckets (TIKR): Schwab closed Tuesday $100.35 (−6.11%) and Wednesday $99.50 (−0.85%). The note attributes the drop to AI-disruption fear, not a NIM miss, and restates CFO Mike Verdeschi's Q2 guide: Q4 exit NIM 3.25%–3.30%, ~$250–$300 million annualized NIR per 25 bp. That guide was already on the Q2 transcript. A restatement of a July guide is not the post-hike sell-side NII model this hypothesis asked for. Conversion bar unchanged. Stay hypothesis. Do not graduate SCHW.