This case does not name where the next systemic crisis starts. It scoreboards the question. Private credit's default rate reads anywhere from 2.73% to 6.0% depending on methodology, both watched, both disputed.[1] Office CMBS delinquency sits at a record ~10-11%, concentrated in regional banks holding two-thirds of bank CRE debt.[2] California's FAIR Plan required its first assessment since 1994 after the January 2025 LA fires.[3] Non-bank leverage through the Treasury basis trade runs an estimated ~$1 trillion notional, a plumbing risk that had a real stress rehearsal in April 2025.[4] Official watchlists — the IMF, the Fed, the BIS, the Bank of England — rank stretched valuations, NBFI opacity, CRE concentration, and sovereign fragility as the top systemic risks in that order, in 2025-26.[5] The honest complication, and it's the whole point of this case: 2008 started in subprime mortgages and shadow banking, which regulators weren't watching closely; SVB failed from interest-rate and duration risk, not the credit risk everyone was scrutinizing after 2008. The risk that gets a name, a watchlist entry, and a stress test is — by that same logic — the one market participants have already started pricing and de-risking against. This case names four falsifiable triggers and sets a review for January 31, 2027, after the Q4 data lands. Until then, or until two or more triggers cross at once, the honest answer is the scoreboard, not a guess.
The temptation, at the end of a cluster tracing how systemic risk migrated after 2023, is to name the place it will resurface — to say private credit, or CRE, or insurance is where the next crisis starts. This case refuses that temptation on purpose. The honest position in mid-2026 is that all three are genuinely watched, genuinely stressed, and genuinely unresolved, and the discipline is to state what would settle the question rather than to guess which piece breaks first.
The official watchlists agree on the shortlist, if not the order. The IMF's GFSR, the Fed's Financial Stability Report, BIS bulletins, and the Bank of England's Financial Policy Committee rank stretched asset valuations colliding with leverage first, non-bank financial intermediation (NBFI) opacity second, CRE concentration in regional banks third, and sovereign fiscal fragility fourth — with a newer, less-examined line item emerging in 2025-26: private-equity-owned life insurers ceding risk to offshore reinsurers in Bermuda and the Cayman Islands, a channel with limited transparency that none of this cluster's five cases have yet examined directly.[5]
The base rate argues for humility about all of it. Paul Samuelson's line that the stock market predicted nine of the last five recessions has aged into a durable truth about early-warning indicators generally: most flagged risks de-escalate or grind on for years without producing a discrete crisis.[6] Reinhart and Rogoff documented that crises recur but are systematically mistimed and mislabeled in advance — the confident “this time is different” framing is itself usually the tell. And there's a sharper, structural version of the same point specific to this cluster: 2008 started in subprime mortgages and off-balance-sheet shadow banking, areas regulators were not scrutinizing closely at the time; SVB in 2023 failed from interest-rate and duration risk in its held-to-maturity bond book, not the credit risk everyone had spent fifteen years stress-testing since 2008.[7] By that logic, the risk that gets a name, a watchlist ranking, and an annual stress test is partially inoculated by the attention itself — capital gets raised against it, exposures get trimmed, buyers price it in. The dangerous exposure, historically, is the one nobody is measuring yet.
So this case sets four falsifiable, dated triggers instead of a prediction. Trigger one: private credit's default rate sustains above 5% with rising PIK income, currently reading 2.73-6.0% depending on index.[1] Trigger two: office CMBS delinquency crosses 15%, or a named regional bank fails and cites CRE — currently ~10-11%, no failure confirmed since NYCB.[2] Trigger three: a second insurer-of-last-resort assessment lands within twelve months of the first — California's FAIR Plan levied its first since 1994 in February 2025.[3] Trigger four: the Treasury basis trade or bank-to-nonbank lending channel dislocates disorderly enough to force Fed intervention — a real rehearsal already happened in April 2025.[4] None of the four has fully fired as of July 2026. The rule going forward: this case stays HELD OPEN unless two or more cross simultaneously, consistent with the base rate that single flags usually fizzle. Review: January 31, 2027, after Q4 2026 data — the FDIC's Quarterly Banking Profile, Trepp's December CMBS print, and the FSB's NBFI Global Monitoring Report — are all in hand.
The four signals a forward-looking analyst is watching — and their status as of July 2026.
The disputed range currently reads 2.73% (Proskauer, Q1 2026) to 6.0% (Fitch, record, 12 months to April 2026) depending on methodology. Neither index has sustained clearly above the 5% threshold with confirmed rising PIK share yet.[1]
Not FiredOffice CMBS delinquency sits at a record ~10-11% — elevated, but below the 15% threshold. No confirmed new CRE-driven regional bank failure has occurred since NYCB in January 2024.[2]
Not FiredCalifornia's FAIR Plan levied its first assessment since 1994 in February 2025, following the LA fires. The trigger is a second one within 12 months of the first — as of July 2026, that clock hasn't produced a second event.[3]
Not FiredThe basis trade runs an estimated ~$1T notional. April 2025 tariff-driven Treasury volatility was a real stress test of this plumbing — but it did not force a disorderly unwind or emergency Fed intervention. A rehearsal, not the event.[4]
Rehearsed, Not FiredThe FDIC's Q4 2026 Quarterly Banking Profile, Trepp's December CMBS print, and the FSB's December NBFI Global Monitoring Report will all be available. Review then: does any trigger cross, and does the October 2026 IMF GFSR shift the ranked watchlist. Until then, the verdict stays open.
ReviewThe stock market has predicted nine of the last five recessions. — Paul Samuelson, economist, 1966 — the durable warning about watchlists and false positives
| Dimension | Evidence |
|---|---|
| Revenue (D2) Origin · 78 | The unresolved question across all three sectors is a capital one: whether a return, a recovery, or a repricing arrives before a loss is forced into the open.[1][2][3] D2 is the origin because none of the individual triggers is really about technology or operations in the abstract — each measures whether the revenue/recovery curve catches the loss curve before a forcing event does, the same variable UC-255 tracked for AI capex, generalized here across three financial sectors.Return Before Recognition |
| Operational (D6) L1 · 80 | The operational tell is the four triggers directly: default rates, delinquency rates, assessment counts, basis-trade notional.[1][2][3][4] D6 amplifies from D2 because these are the most concrete, dated, and falsifiable expressions of the capital question — an analyst can argue about forward returns, but a default-rate index or a delinquency print is a fact on a filing.The Triggers Themselves |
| Customer (D1) L1 · 70 | The demand/confidence side: fund-investor redemption pressure (UC-256), depositor/investor confidence at NYCB (UC-257), and policyholder exit patterns in California and Florida (UC-258) are the visible, human-facing signals underneath the institutional triggers.[1][2][3] D1 amplifies alongside D6 as the observable-to-ordinary-people layer of the same underlying question. |
| Quality (D5) L2 · 68 | Whether structural containment holds is the quality dimension — the UC-259 counterexample this capstone must weigh in full. If illiquidity, capital buffers, and slow recognition genuinely absorb these stresses the way UC-259 argues, no trigger crosses no matter how elevated the underlying numbers get.[6] D5 is where the bull case for containment has its strongest structural argument. |
| Employee (D3) L2 · 56 | The workforce and accountability thread already visible across the cluster — Tricolor's DOJ-charged executives (UC-256), NYCB's leadership changes (UC-257) — is a smaller but real dimension of how these risks surface: through individual accountability actions before, sometimes, the systemic numbers move. |
| Regulatory (D4) 74 | D4 is the longest-lag, highest-stakes dimension and the explicit subject of this capstone: the ranked IMF/Fed/BIS/BoE watchlists, the base-rate literature on false positives, and the 2008/SVB historical precedent for where risk actually surfaces versus where it's watched.[5][6][7] This is where an individual sector's stress becomes — or doesn't — a financial-stability event, and it is the slowest dimension to move and the most consequential if it does.The Dimension This Capstone Is Built Around |
The cascade originates in D2 — Revenue — because the unresolved question across all three sectors is fundamentally a capital one: whether the return or the recovery arrives before the loss gets forced into the open. From D2 it runs to D6 (the operational triggers themselves — default rates, delinquency, exposure levels) and D1 (the demand-side and depositor/policyholder-facing signals), then D5 (whether structural containment, argued in UC-259, actually holds) and D3 (the workforce and executive-accountability threads already visible in UC-256's fraud prosecutions). D4 (the systemic/regulatory dimension — where the BIS, IMF, and Fed watchlists sit) is the longest-lag and highest-stakes dimension, and the one this capstone is explicitly built around. This is the cluster capstone: it synthesizes [UC-256] (private credit), [UC-257] (CRE), and [UC-258] (insurance) into one forward scoreboard, and it must weigh [UC-259] — the illiquidity-shield counterexample — as the case most capable of keeping any single trigger from mattering as much as it looks like it should. Confidence is deliberately low (0.40): a capstone that faked certainty about which sector breaks first would betray the discipline the whole cluster is built to demonstrate.
-- UC-260: Where Does the Next One Start?: 6D Prognostic Capstone
-- Systemic risk scoreboard (synthesizes UC-256/257/258; weighs counter UC-259)
FORAGE next_one_starts
WHERE verdict_held_open = true
AND four_triggers_observable_and_dated = true
AND base_rate_argues_humility = true
ACROSS D2, D6, D1, D5, D3, D4
DEPTH 3
SURFACE next_one_starts
WATCH private_credit_default WHEN default_rate_sustains_above_5pct_with_rising_pik = true
WATCH cre_delinquency WHEN office_cmbs_delinquency_exceeds_15pct = true
WATCH cre_bank_failure WHEN named_regional_bank_fails_citing_cre = true
WATCH insurer_last_resort WHEN second_assessment_within_12_months_of_first = true
WATCH nonbank_leverage_unwind WHEN treasury_basis_trade_dislocates_disorderly = true
DRIFT next_one_starts
METHODOLOGY 88
PERFORMANCE 40
FETCH next_one_starts
THRESHOLD 1000
ON WATCH CHIRP medium 'Where does the next systemic crisis start - private credit, CRE, or insurance? Four dated triggers: private-credit default rate above 5pct (currently 2.73-6.0pct), office CMBS delinquency above 15pct or a named regional bank failure (currently ~10-11pct, none confirmed), a second insurer-of-last-resort assessment within 12 months (first was Feb 2025), or a disorderly non-bank leverage unwind. None fully fired as of July 2026. 2008 started where regulators weren't looking; SVB failed from a risk nobody was watching. The watched risk may not be the real one'
SURFACE review ON '2027-01-31'
SURFACE analysis AS jsonRuntime: @stratiqx/cal-runtime · Spec: cal.semanticintent.dev · DOI: 10.5281/zenodo.18905193
All four watched signals sit below their threshold as of July 2026: default rates elevated but under 5%, office delinquency elevated but under 15%, one insurer assessment but not a second, real Treasury volatility but not a disorderly unwind. A genuinely open scoreboard, not a foregone conclusion dressed up as one.[1][2][3][4]
Proskauer's 2.73% and Fitch's 6.0% aren't a data error — they're two legitimate methodologies disagreeing about how much stress is already in the system, which is itself the most honest available answer to “how bad is it right now.”[1]
2008 and SVB both broke where attention wasn't. If that pattern holds, the sectors with an IMF chapter, a BIS bulletin, and a Fed stress test devoted to them — private credit, CRE — may be exactly the ones least likely to be the actual trigger. The offshore-reinsurer channel this case flags but doesn't examine is the kind of place that pattern would predict.[7]
A prognostic capstone is not judged on naming the right sector early. It is judged on stating four triggers precise enough to be proven wrong, and a review date — January 31, 2027 — precise enough to hold itself to it. Confidence is 0.40 on purpose.
Seven sources, held two-sided by design: the four trigger-specific data sources (default indices, CMBS delinquency, FAIR Plan exposure, the Treasury basis-trade estimate), the ranked official watchlists, the base-rate literature on false-positive warnings, and the historical 2008/SVB precedent for where risk actually surfaced versus where it was watched.
Watch the four triggers. When two fire together, the question closes. Until then, anyone naming the next crisis with confidence is guessing, not scoring.