Signal Scanner · FINANCIAL SERVICES & FUTURE OF MONEY · 14 September 2026

US Insurance Regulators Take the Capital Charge Back from the Rating Agencies

The capital a US insurer must hold against a private asset is being re-derived by the regulator rather than read off a credit rating, with three NAIC instruments biting on the 31 December 2026 balance sheet. Exposed: life insurers, private-credit managers, CLO arrangers and rating agencies.

The consensus argument about insurers and private credit is an argument about credit quality. Life insurers' holdings have more than doubled in a decade, the assets are illiquid and hard to price, and the fear is a real default cycle. Underneath that, something narrower has moved. Three separate NAIC instruments now sit between a credit rating and the capital charge it produces, and all three land on the 31 December 2026 balance sheet. The rating still exists; it has stopped being the answer. Required capital on a private asset can rise without a downgrade, a default, or any change in the asset itself, and most capital plans are built as though it cannot.

Signal Identification

A regulatory pivot with a governance problem inside it. The instrument is not a rule about what insurers may buy but a rule about who converts what they own into a number. Assurance moves from a market participant paid by the issuer to a standard-setter with no market discipline. The exposure is capital planning, asset selection and legal challenge at once.

Time horizon: 1-4 years (SVO discretion in force 1 January 2026; rating-provider due diligence proposal re-exposed 23 September 2026; CLO C-1 factors at the 31 December 2026 filing; collateral-loan look-through 31 December 2027)
regrading binds202620272028to 2030
Plausibility band: Medium–High
LowMediumHigh
Geographic / Jurisdictional Scope: Primary: the United States, where the NAIC sets designations and domiciliary regulators give them effect. Spillover: Bermuda and the Cayman Islands, where ceded annuity blocks sit, and the UK and EU-27, where supervisors run parallel work on private-credit opacity.
PrimaryUnited StatesAgreement States
SpilloverBermudaCayman IslandsUKEU-27
Sectors exposed:
US life insurance and annuitiesCredit rating agenciesCLO managers and arrangersPrivate-credit and alternative asset managersRated feeder fundsData-centre and infrastructure debtInsurance equity and credit investorsActuarial and capital management functions

What's Changing

Three instruments, adopted separately, converge. Since January the NAIC has held the power to challenge and override a filing-exempt rating that differs from its own analysis by more than three notches, across public and private ratings (Insurance Business, 12/06/2026). On 23 June its risk-based capital group adopted new C-1 factors for CLOs effective for reporting as of 31 December 2026, with tranche thickness “measured on a current basis, using the most recent trustee report” rather than at origination (Sidley Austin, 09/07/2026). And the due diligence rules governing which rating providers may be relied on at all return for exposure on 23 September (NAIC, accessed 14/09/2026).

The direction is not uniformly upward. Senior investment-grade tranches generally see lower charges than the corporate-bond treatment they replace, particularly AAA through A, while an 11.77% pretax surcharge lands on below-investment-grade broadly syndicated CLO tranches of 4% thickness or less, and the 45% charge on residual tranches stays (Sidley Austin, 09/07/2026). What changed is the logic, not the level: capital now turns on where a tranche sits and how thick it is, so “similar ratings would not necessarily produce similar RBC outcomes” (DLA Piper, 30/06/2026).

The book being regraded is large, and what insurers report about it has just been shown to be unreliable. Senator Elizabeth Warren's 11 September letter to the NAIC cites its own data putting life insurers' private credit at $849 billion in 2024 against $386 billion in 2014, and follows Delaware Life's restatement of related-party holdings from about $1.4 billion, roughly 3% of investments, to more than $17 billion, upwards of 39% of invested assets (Insurance Business, 11/09/2026). CLOs alone accounted for $276.8 billion at the end of 2024, about 5.1% of insurer bonds (NAIC, 01/04/2026).

Where the regulator has cut into the chain from asset to capital charge

HOW A PRIVATE ASSET BECOMES A CAPITAL CHARGE Private asset CRP rating NAIC designation RBC C-1 factor Required capital 1 2 3 1 Rating-provider due diligence rules, revised proposal re-exposed 23 Sep 2026 2 SVO discretion to challenge a rating three or more notches out in force 1 Jan 2026 3 CLO C-1 factors keyed to rating and tranche thickness 31 Dec 2026 filing US LIFE INSURER PRIVATE CREDIT, NAIC DATA, $BN 0 300 600 900 Private credit, 2014 386 Private credit, 2024 849 of which CLOs, 2024 276.8

Sources: NAIC data cited in Senator Warren's letter of 11 September 2026, reported by Insurance Business; the NAIC issue brief of April 2026 for the CLO figure; Sidley Austin and the NAIC Credit Rating Provider Working Group for the gates and dates.

Disruption Pathway

Stage one runs to the end of 2026 and is mechanical. Insurers file 31 December risk-based capital on the new CLO factors, thickness read off current trustee reports, while the due diligence rules for rating providers close their exposure round. Stage two runs through 2027 and 2028, where discretion becomes visible: the first challenges to filing-exempt ratings, the first contested designations, the collateral-loan look-through at the 31 December 2027 filing, and the thickness logic migrating to rated feeders and other asset-backed structures, which DLA Piper expects (DLA Piper, 30/06/2026). Stage three is the legal test, when a domiciliary regulator first overrides a rating an insurer priced an acquisition on.

Stress concentrates in three places. Valuation is the first: bonds carrying private placement numbers reached 23.4% of insurers' admitted bonds in 2025 against 18.3% in 2021, and positions with the least market evidence are hardest to defend when a regulator asks (Insurance Business, 12/06/2026). Concentration is the second, with five asset managers accounting for a third of global private-credit loan commitments, so one re-grading travels at once (Chatham House, 15/07/2026). Liability is the third, because a designation that can be contested is a judgement someone owns. Two adaptations follow. Origination moves upstream, with tranche sizing negotiated for insurance capital treatment at issue rather than at purchase. And insurers build internal shadow-designation capability, because whoever models the charge before the regulator does can price the asset.

Why This Matters Now

This lands on life insurance boards and chief investment officers, on the alternative managers whose insurance balance sheets fund their origination, on rating agencies whose regulatory franchise is being made conditional, and on the analysts who model insurer capital. The decision architecture that needs revising treats the credit rating as the capital input and the capital charge as a stable function of it. Neither holds from the 31 December filing onward. Insurers should model required capital under adverse designation outcomes as well as adverse credit outcomes, and should know today which positions sit three or more notches from a plausible regulatory view. Taken together, the sources suggest capital plans resting on ratings alone will break first, and quietly, in a filing rather than in a market.

Decision-action posture for this signal: Prepare — the discretion power is in force and the CLO factors are adopted, but the systems to operate a challenge are still being built and the due diligence rules re-enter exposure on 23 September, so positions should be sized against the first contested designations rather than committed against mechanics that are still moving.

Counter-Argument

The strongest objection is that this is calibration dressed as a change of regime, and that the evidence never supported it. Powell and Morris find that life insurers with larger private-debt allocations show stronger financial profiles rather than weaker, conclude that “the available evidence therefore does not support the view that private credit creates risks existing insurance regulation fails to capture”, and note that the CLO recalibration “was not motivated by poor historical performance” (International Center for Law & Economics, 05/06/2026). The NAIC's own brief records about 80% of insurer CLO holdings as investment grade or better (NAIC, 01/04/2026).

That may be right about the motive and still leave the consequence standing. A power used rarely still changes how every position is underwritten, because a three-notch challenge must be priced whether or not it is exercised. And the CLO factors involve no discretion at all: they apply to the whole $276.8 billion book at the 31 December filing, on mechanics no rating agency controls.

Implications

This is durable rather than cyclical, because it relocates an authority rather than adjusting a number. Once a designation is a regulatory judgement that can be contested, it does not revert to being a purchased opinion, and the NAIC states plainly that it “has adopted a discretion process to challenge ratings for regulatory use when they do not reasonably reflect investment risk” (NAIC, 01/04/2026). The inflection window runs from the 31 December 2026 filing to the first contested designation. Insurers with modelling capability, thick senior positions and clean valuation evidence gain. Holders of thin subordinated tranches, affiliated paper and assets with little market evidence lose, and they lose before any borrower misses a payment.

Early Indicators to Monitor

Disconfirming Signals

Strategic Questions

Keywords

NAIC designation; filing exempt; SVO discretion amendment; credit rating provider due diligence; risk-based capital; CLO C-1 factors; tranche thickness; private letter ratings; life insurer private credit; rated feeder funds; regulatory capital arbitrage; insurer solvency

Bibliography

Source tiers: Tier 1, governments, regulators and intergovernmental bodies. Tier 2, think-tanks, academic institutes, major consultancies and quality data providers. Tier 3, quality journalism and specialist trade press. Tier 4, vendor, company and practitioner sources, used only as directional corroboration.


Prepared by Shaping Tomorrow: 14 September 2026