Regime Check — July 11, 2026: The Dot Plot Flips, Oil Round-Trips, and the Last Reservoir Narrows

Builds-on: regime-check-june-10-2026 Related: the-eccles-inversion-and-the-may-13-collision, energy-and-stagflation-forecast-2026-2031, how-inflation-dies-the-empty-reservoir, failure-cascade-index, two-economy-gauge, the-involution-import-open-weight-deflation-and-frontier-pricing-power, cyclical-20-and-the-ai-capex-mask, japan-debt-trap-thesis-audit

One-month catch-up (June 10 → July 11), run deliberately before the July 14 – Aug 7 print corridor (CPI, PPI, FOMC, all four hyperscalers, GDP advance, PCE/ECI, payrolls) so the standing state is on record going in. The short version: the Fed flipped hawkish and capitulated on its inflation forecast in the same SEP, Iran signed a memorandum and then started charging tolls in the strait, oil gave back the entire war premium ($88-90 → ~$71), core PCE broke the "core is contained" comfort at 3.4%, the labor freeze thinned further (+57K), and the AI tape not only recovered the June rout but re-concentrated to the point where one stock drove ~70% of a day's index gain. Meanwhile this week's research found stage 4 of the demand-destruction sequence already firing at the small-business tier and the hyperscaler funding rotation (cash flow → debt) firing on schedule. The system is more resolved and more concentrated than a month ago — in both directions at once.

1. The Fed: hawkish flip + forecast capitulation in one meeting

June 17 resolved exactly as priced (hold, unanimous, 3.50-3.75%) but the details were the story (CNBC, Chase takeaways):

• The dot plot flipped to hikes. Median year-end 2026 moved 3.4% → 3.8% — from one cut to net hikes. Nine of 18 project at least one hike; six project two. The political-cut vector isn't just dead, it's inverted at the institutional-median level.
• The SEP capitulated to the vault's inflation world. Year-end PCE projection moved 2.7% → 3.6%. The Fed's own central forecast now sits on the energy-and-stagflation-forecast-2026-2031 "modal ~3.5%, not 2%" path. When the referee moves his forecast onto your number, the disagreement that remains is about the response, not the diagnosis.
• Warsh's fingerprint is institutional, not directional. A 130-word statement (from 341), the easing bias stripped exactly as Waller flagged — and Warsh declined to submit a dot, openly questioning the dot plot as an institution. Read with the Eccles-inversion frame: he's dismantling communication apparatus before touching the balance sheet. No accord document, no composition operations yet — the the-eccles-inversion-and-the-may-13-collision composition trap is intact and untriggered.

2. Iran: the memorandum-with-tolls — a scenario the taxonomy didn't have

June 10's reweight put 45% on re-escalation. What actually happened is stranger and needs its own name (Britannica summary, Wikipedia: 2026 Strait of Hormuz crisis):

• June 17: the Islamabad Memorandum — a signed US-Iran MoU formalizing the war-end process with a 60-day negotiation window (expires ~August 16) and a Hormuz reopening framework. Signed the same day as the FOMC.
• Then Iran started charging rent on the strait. Repeated threats and attacks on shipping to coerce vessels into Iranian-set routes and protocols — a toll regime, not a blockade and not free transit. Three ships attacked July 6-7 → US strikes → Trump declared the truce "over" July 7, walked it back July 8, talks still on. The US Navy's JMIC opened a widened Oman-side route June 27 to contest the toll regime directly.
• Oil round-tripped the entire war premium anyway. WTI ~$71 (from $88-90 on June 10; the April wartime peak was WTI ~$106-113 intraday / Brent briefly $126 — earlier "$130-140" figures circulating in synced notes were wrong, corrected 2026-07-15); national average gasoline $3.88, down from the $4.56 May 21 peak (though it ticked up a nickel on the July 6-7 flare). The market has priced the memorandum as a done deal and the toll regime as noise.

This is the June-10 "Wright fade" warning at 10x scale: the tape is fully pricing the 60-day negotiation succeeding while the mechanism that would break it (toll-regime enforcement meeting US freedom-of-navigation strikes) fires weekly. The asymmetry has inverted since May: then, insurance was expensive and the premium was in the price; now the war premium is gone while the war is technically unresolved. If August 16 fails, the repricing starts from $71, not $90. (Update 2026-07-19: the "oil faded anyway" anomaly is now substantially resolved by china-oil-buffer-thesis-audit — China's buying strike (~11M→7.8M bpd, no stockpile draw) accounted for ~74% of the global crude-trade decline. The fade was physical, not just psychological — and it has an expiry: JPM expects Chinese imports to recover from August 2026.)

Scenario reweight against June 10's (re-escalation 45 / protracted 30 / theater 15 / deal 10): armistice-with-tolls (the new modal state) ~45%, durable deal ~20%, re-escalation ~25%, full war resumption ~10%. The energy doc's structural supply thesis (Permian peak, OPEC cohesion breaking) is unchanged underneath — this is a demand-side-of-fear repricing, not a supply fix.

3. Inflation: the legs swapped

June 10's configuration was "energy screaming, core contained." Both halves flipped:

• The energy leg is mechanically unwinding. Oil -20% in a month; gasoline -68c from peak. July CPI (prints mid-August) will carry the fade; June CPI (July 14) is the transition month. The energy>60%-of-the-print driver is going away — headline should roll toward core over the next two prints.
• Core broke the containment story. May core PCE printed 3.4% YoY (June 25) — highest since October 2023, headline PCE 4.1%. Core CPI 2.9% had been the "supply-shock-not-demand-pull" comfort; core PCE at 3.4% with tariff pass-through peaking through Q2 says the shock has been broadening into the sticky categories while everyone watched gasoline.
• Net: the stagflation mix is rotating from energy-headline to tariff-core. That's worse for the Fed's reaction function (core is what they can't look through) and consistent with the SEP capitulation. The vault's probability-weighted path (~3.7% 2026), which was "running low" against 4.2% headline in June, is back on track as oil fades — the modal-3.5% world, arriving via mix shift.

4. Labor: frozen, and thinning faster

June payrolls +57K; May was revised down 172K → 129K — reversing the June-10 "priors revised UP" pattern and partially rehabilitating the demand-side-audit-may-2026 revision thesis that regime-check-june-10 had corrected. U-3 fell to 4.2%, but via participation dropping to 61.5% (lowest since March 2021) — the unemployment rate is improving for the wrong reason, which keeps Sahm mechanically suppressed (~0.1, not triggered). Wages +3.5% and decelerating; quits still frozen. Gains averaging ~40K/month since January 2025.

Stage-5 tripwires (U-3 4.6-4.8%, Sahm 0.30, claims 1.95M): not fired. But the freeze is thinning — each month of 40-57K prints with negative revisions moves the labor market closer to where a single shock (the stage-4 failure wave broadening, an AI-complex event) converts freeze into synchronized firing. The failure-cascade-index fuse gauges say the pressure for that conversion is building: absorption wedge open, NFIB desperation signature active, SubV filings +50% H1.

5. Credit: the two-tier market confirms the tiering principle

Aggregate credit got more benign — HY OAS 267bp (July 7), tighter than June 10's 309bp. But underneath: the CCC-to-single-B differential now exceeds 600bp, having widened ~200bp YTD, CCC defaults are running above long-term averages, and private credit is showing rising PIK-toggle usage with late-2025 leveraged-loan defaults described as the toughest environment since 2008. This is the credit-market version of the K-shape: the aggregate spread reads "no recession" while the weakest tier reprices toward one — the same masking structure as Case-Shiller, payrolls, and GDP-with-AI-capex. The failure cascade is visible in credit exactly where the tiering principle said to look, and invisible exactly where most people look.

6. The AI complex: recovered, re-concentrated, and now formally the last reservoir

• The June 3-5 rout fully round-tripped. SPX 7,575 (July 10), above the May 27 high; Nasdaq 26,282. NVDA drove ~70% of the S&P's Friday gain while small caps fell — breadth narrower than before the rout.
• SK Hynix's $26.5B US listing — the largest ever by a foreign company — popped 13% on debut (July 10). An HBM pure-play doing a record foreign listing into peak AI concentration is blow-off texture — the 2026 analog of 1999-2000's record IPO windows.
• This week's research (see the-involution-import-open-weight-deflation-and-frontier-pricing-power) formalized the two-sided state: revenue delivery genuinely accelerating (Anthropic $47B RR, Jevons consumption) while the funding structure degrades (capex crosses operating cash flow ~Q3 2026, $175B debt issuance, Amazon TTM FCF $26B→$1.2B) and open-weight competition caps the price ceiling (US OpenRouter share 70%→30%, GLM 5.2 within ~1% of frontier at one-sixth cost, OpenAI weighing price cuts). Blow-off configuration: fundamentals improving and structure weakening, simultaneously, at record concentration.
• Breakevens are quietly voting for the bear half. 10y breakeven 2.24% (July 6), compressed from 2.45% on May 27 — while headline CPI printed 4.2% and core PCE 3.4%. The bond market is pricing the oil fade and the demand story, not the current prints. 10y nominal ~4.54% → real yields ~2.3%, still restrictive.

Portfolio mapping (positions unchanged; two gauges now near trigger)

• TIPS (SCHP/FIPDX/STIP): underperformed nominals in the window as breakevens compressed 21bp. This is the insurance bleeding as designed, same as XLE in the May peace rally. The uncomfortable note from how-inflation-dies-the-empty-reservoir stands: the book is positioned for A/D, and the week's research grew B's plausibility. BE 2.24% is now sitting almost exactly at the ~2.2% rotation threshold — but the rule requires the stage-5 tripwire to fire while BE holds above it, and stage 5 has not fired. No action; watch both dials weekly through the corridor.
• XLE: oil -20% = the drag is back. Standing guidance (don't chase, don't cut — it's insurance against the August 16 failure mode, which starts from $71 now) holds.
• IAU: no fresh read in-window; check on the next pass.
• The one new input for the duration question: if July CPI (mid-Aug) rolls hard on energy while core stays 3.4%+, TIPS accrual stays paid while the disinflation narrative builds — the awkward middle where both TIPS and nominals argue. The corridor prints decide it; nothing to pre-empt.

Scenario weights (empty-reservoir frame, formally updated)

Scorecard

Right: fade-the-announcement (oil gave back the entire premium — the vault's most durable call this year); XLE don't-chase discipline; the stage-2→3 consumer read (staples pass-through failure now consensus in earnings language); "frozen labor" over "collapsing labor"; the AI crack-then-recover pattern (why-the-market-refuses-to-crash logic — good news stopped working in June, then the structural bid reasserted).

Wrong / corrected: "Core is contained" — core PCE 3.4% broke it; the June-10 correction of the demand-side audit's revision thesis was itself half-wrong (May's +172K became +129K — the QCEW-style downward-revision pattern is back); CPI modal path was "running low" in June and is now roughly on track — the lesson is the path error was composition (energy vs core), not level.

New this cycle (from the July 11 research day): stage 4 confirmed firing at the small-business tier (SubV +50-67%) — a finding the aggregate data actively hides; the funding-rotation tripwire fired on the hyperscalers' own guidance; the last-reservoir thesis formalized with the asset-side audit; three standing gauges built (failure-cascade-index, two-economy-gauge, normalization-vs-destruction-test).

What to watch (the corridor, in order)

1. Jul 14 CPI / Jul 15 PPI — the wedge update: does headline start rolling toward core; does PPI confirm the pipeline cooling. First test of the C-scenario mechanism.
2. Jul 22 Alphabet, Jul 29 Microsoft + Meta, Jul 30 Amazon + Q2 GDP advance — FCF prints and any first 2027 capex language. An Amazon guide-down is the efficiency-cliff catalyst; four clean guides-up extends the blow-off.
3. Jul 28-29 FOMC — no SEP this meeting; the statement language and any composition-operations hint are the Warsh fingerprints to read.
4. ~Aug 16: Islamabad Memorandum 60-day expiry — the single largest binary in the corridor. Repricing starts from $71.
5. Aug 7 payrolls — stage-5 tripwires against a thinning freeze; watch the May/June revisions as much as the July print.
6. Breakevens vs the 2.2% threshold weekly — the rotation rule's arming condition.
7. CCC-B differential >600bp and widening — the tier that breaks first, already breaking; a move through ~750-800bp with HY aggregate still <300bp would be maximal masking, maximal fragility.
8. Yen 160-162 band — the record ¥11.73T ($73B) intervention failed within six weeks; yen at 161 with JGB 10y at 2.64%. Channel 1 of regime-cascade-architecture is loading quietly under everything else.
9. OpenAI pricing decision + Beijing export-restriction decision — the two involution-import wires; either one moves the frontier-sticker-discipline gauge.
10. Early-Aug Epiq July filings — SubV: base effect or acceleration.

Sources

• Fed decision June 17 (CNBC) · Dot plot flips to hike (StockTitan) · Warsh takeaways (Chase) · Bloomberg dot plot chart
• 2026 Iran war (Britannica) · Strait of Hormuz crisis (Wikipedia) · Ceasefire reactions June 18 (Al Jazeera) · July 10 liveblog (Al Jazeera)
• May PCE 4.1%/3.4% core (CNBC) · dshort/Advisor Perspectives · AAA gas prices
• June jobs +57K (CNBC) · Marketplace wage read
• HY OAS 267bp (Convex/FRED) · State Street Q3 credit outlook (CCC-B >600bp) · Private credit outlook (With Intelligence)
• July 10 market close (Yahoo) · NVDA 70% of gain (ts2) · SK Hynix debut (Motley Fool)
• 10y breakeven 2.24% (Convex/FRED) · WTI ~$71 (TradingEconomics)
• Yen intervention failure (CNBC) · Yen past 161 (CNBC)