I. Summary
At its Summer 2026 National Meeting on August 12, 2026, the NAIC's Statutory Accounting Principles (E) Working Group ("SAPWG") exposed a new agenda item, Ref #2026-10 (SSAP No. 26, Embedded ALM Risk) (the “ALM Proposal”), addressing a fast-growing category of insurer investment that regulators have termed "multi-collateral structured credit investments.” The item proposes targeted revisions to the principles-based bond definition ("PBBD"), which took effect on January 1, 2025, in SSAP No. 26 that would preclude bond treatment for structured securities found to carry "significant embedded asset-liability management (ALM) risk."
The proposal certainly highlights some items that genuinely warrant consideration. But as H.L. Mencken observed: "For every complex problem there is an answer that is clear, simple, and wrong." Based upon our work in the structured credit markets since their inception, we would respectfully submit that the relevant consideration is not how complex or how long-dated a product is, but rather whether the underlying cash flow is durable and whether the structural framework is adequately protective.
This OnPoint summarizes NAIC proposal and the reasoning behind it, and offers our observations on the concerns the NAIC has identified.
II. What Is the NAIC?
The National Association of Insurance Commissioners ("NAIC") is the U.S. standard-setting and regulatory support organization created and governed by the chief insurance regulators of the 50 states, the District of Columbia, and five U.S. territories. The NAIC itself has no direct regulatory authority over insurers; rather, its model laws, accounting guidance, and risk-based capital frameworks are developed collaboratively by state regulators and, once adopted, are implemented by individual states through legislation, regulation, or administrative practice. As a practical matter, however, state insurance departments look to NAIC guidance closely, and changes to the NAIC's statutory accounting principles, such as those proposed here, tend to work their way into state reporting, capital, and disclosure requirements over time.
III. Background: A New Asset Class Draws Scrutiny
The ALM Proposal responds to NAIC's identification of an emerging asset class distinguished from traditional structured credit by the diversity of its underlying collateral. Whereas conventional structured securities are typically backed by a single, homogenous collateral type, RMBS by residential mortgages, CLOs by corporate loans, student loan ABS by student loans, the agenda item observes that multi-collateral structured credit investments (the “Instruments”) can be supported by effectively any asset class, and that the structures regulators have reviewed generally include a broad mix of investment-grade and non-investment-grade bonds, direct loans, smaller allocations to unrated debt and mortgage loans, equity investments, and in some cases tranches of other securitizations. These structures generally satisfy the PBBD today because the underlying fixed income investments generate cash flows to service and repay the debt and a residual (typically equity) tranche provides the credit enhancement the PBBD requires.
The agenda item flags that a subset of these structures also exhibits a duration mismatch between the debt and the underlying collateral: regulators point to observed cases in which the stated maturity of a debt tranche runs as long as 40 years while the collateral backing it carries a weighted average duration in the neighborhood of 8 years or less. Regulators view this as materially different in degree from the reinvestment periods common to structured credit generally, which the agenda item describes as typically 5 years or less and subject to early-termination triggers.
It is worth pausing on that framing. First, stated maturity is not the critical measure of duration for these Instruments. The expected weighted average life ("WAL") of a debt tranche issued in a multi-collateral structured credit investment, which reflects prepayment assumptions, reinvestment dynamics, structural amortization mechanics, and market convention, is typically considerably shorter than the stated legal final maturity, which functions as an outer limit rather than an expected payoff date. Senior investment-grade tranches in securitizations often have WALs that are half of the length of their legal final maturity. Focusing regulatory concern on a stated maturity that functions as the latest possible date for a debt tranche to be repaid rather than the expected WAL that actually takes into account projected cash-flows, reinvestment and other structural considerations will systematically overstate the duration exposure these instruments actually present.
Second, a longer expected WAL is not inherently a source of increased risk. It is, for a significant portion of the insurance market, precisely the feature that makes these Instruments useful parts of their portfolio of investments. Life insurers and annuity writers carry liabilities with durations measured in decades. The longstanding challenge for such institutions is not that long-dated assets are hard to manage, but that long-dated, high-quality assets in sufficient supply are genuinely hard to find. Corporate bond markets do not reliably supply 30-to-40-year investment-grade paper at adequate spread. Government bonds supply duration but not credit premium. The Instruments fill precisely this gap, offering duration matching for long-liability portfolios alongside spread premium and structural credit enhancement that has historically delivered performance superior to equivalently rated corporate bonds. Ironically, the NAIC steering insurance companies towards shorter-dated ABS paper could cause the same mismatch on insurance companies’ investments when compared to their longer-dated liabilities that the ALM Proposal is intended to address.
Third, the risk of the inability to reinvest over a longer reinvestment period in multi-collateral structured credit investments is mitigated by diversity of the collateral base. The ALM Proposed identified the risk that the longer dated reinvestment period of multi-collateral structured credit investments could result in the “[in]ability to reinvest at sufficient yields over several decades.” However, this risk should theoretically be lower in a multi-collateral structured credit investment when compared to a single asset class securitization of the same maturity because there is a more diverse collateral base from which to acquire new assets. For instance, if the commercial real estate market suffered a market disruption that resulted in the inability to acquire new CRE collateral, a CRE CLO with a long reinvestment period may be subject to some of the reinvestment risk that the ALM proposal is concerned about; but a multi-collateral structured credit investment with a diverse collateral base would have the flexibility to manage around any asset class specific risk during the reinvestment period.
The historical record across comparable long-dated asset classes bears out that point. U.S. franchise whole business securitizations covering brands such as Domino's, Dunkin', and Sonic have demonstrated that even operationally-generated revenue streams can support long-dated investment-grade debt when the underlying brand economics are durable and structural protections are robust. The same principle held for data center ABS, backed by long-term, non-cancellable lease payments from investment-grade hyperscale tenants, which has similarly supported strong credit performance, even though the sector remains young and has not yet been tested through a full credit cycle. In each case, the determining factor in performance was not duration. It was the durability of the underlying cash flow source.
IV. Regulatory Concerns
The agenda item states that regulators have raised three distinct concerns:
Complexity, Transparency, and Interconnectedness. The first item relates to complexity, transparency, and interconnectedness: the diversity of collateral within these structures, in the regulators' view, creates greater potential for circular or overlapping exposure across insurer portfolios than would arise from a single-collateral-type structure, and makes it more difficult to track an insurer's true aggregate economic exposure where the same underlying asset may be held both directly and indirectly. The agenda item notes, however, that the credit quality and diversity of the collateral, together with the vehicle's capital structure, often supports a view that the credit risk is comparable to, or even lower than, equivalently rated single-collateral structures. The concern raised is about transparency and aggregation, not necessarily about the credit risk of any individual structure being higher.
That distinction matters. A structure diversified across hundreds of obligors and multiple asset classes is inherently more complex to describe than a single bilateral loan or a single-exposure structured note, but that added complexity is also what allows idiosyncratic underperformance to be absorbed across a pool and by subordination, rather than concentrated entirely in one credit. Complexity and credit risk are analytically distinct, and the agenda item's reasoning does not fully separate them. As Andrew Haldane, then the Bank of England's Executive Director for Financial Stability, observed in a widely cited 2012 paper examining the perils of complex financial regulation, "[b]ecause complexity generates uncertainty, not risk, [regulation] requires a regulatory response grounded in simplicity, not complexity." Treating a structure's complexity as itself a proxy for elevated credit risk risks exactly that conflation. In addition, any concerns about the complexity of these transactions can be mitigated by enhanced transparency on par with, or greater than, that which is provided in the structured credit markets more generally. Such transparency should mitigate concerns around the interconnectedness of the portfolios since the aggregate exposure can be reported and monitored on a portfolio-wide basis.
Asset-Liability Duration Risk. The second item is embedded ALM risk. Where reinvestment periods extend cash flow uncertainty over long horizons, the ability to pay contractual amounts as scheduled can depend on the ability to reinvest at sufficient yields over an extended period. The agenda item observes that mitigants exist in the market, including hedging instruments, accelerated amortization triggers, and subordinated equity tranches that absorb reinvestment-yield compression before it reaches the debt tranches, while also cautioning that unmitigated ALM risk may remain material to debt holders notwithstanding those mitigants. The agenda item's central point is that the PBBD, as originally developed, did not contemplate embedded ALM risk: the substantive credit enhancement concept was designed around subordination that absorbs credit losses, not subordination used to absorb reinvestment or duration risk.
This concern is somewhat difficult to square with the historical record given that outside the world of static securitizations many products have at least similar flavors of reinvestment risk. What the historical record of managed reinvestment structures shows, however, is that structurally bounded reinvestment undertaken by qualified investment managers can be a way to minimize losses, not create them. CLOs, the most rigorously tested managed reinvestment vehicle in global credit markets, have reinvested continuously into leveraged loans across multiple credit cycles, including the deepest credit stress in a generation in 2008 to 2009 and the sharp dislocation of March 2020, and AAA and AA CLO tranches have suffered no material principal losses across more than two decades of market cycles. That performance reflects the benefits of structured reinvestment and the value added by having that reinvestment done by some of the most skilled investment managers in the world and, as discussed below, it is precisely this track record that the proposed ALM risk test would put at risk by sweeping in CLOs, CRE CLOs and asset-based loans without a clear basis for doing so.
Whether Investors Understand the ALM Risk These Structures Contain. The third concern mentioned in the agenda item is whether the institutions investing in these Instruments fully understand and can monitor the ALM risk embedded within them. No specific shortcoming is identified as evidence for this. In fact, it is perhaps more plausible that U.S. insurance companies invest in these Instruments not because they do not understand them, but precisely because they do. The appropriate regulatory response to this concern would appear to be to require that investors demonstrate adequate analytical process before investing in this product.
V. The Proposed Revisions
NAIC staff recommend exposing revisions to SSAP No. 26 as a scope-limited change to statutory accounting principles, rather than a full reconsideration of the bond definition, given the narrow focus of the proposal. As drafted, the revisions would work in two steps.
First, if the financial assets backing a structured security do not self-liquidate, meaning they are not structured to convert to cash on their own over a defined period under their contract terms, the transaction would have to clear the rebuttable presumption that already applies under the PBBD to debt instruments backed by equity-like interests, rather than qualifying through the ordinary ABS pathway.
Second, a new requirement applies where the underlying assets are self-liquidating but the transaction documents permit reinvestment of collected cash flows instead of using them immediately to pay down the debt. In that scenario, the reporting entity must determine whether the instrument carries what the proposal calls "significant embedded ALM risk," meaning exposure to the possibility that changes in reinvestment rates or spreads could interfere with timely payment of the contractual cash flows. If that risk is found to be significant, the instrument loses bond eligibility under the PBBD irrespective of any other favorable characteristics it may have.
We note that separately managed accounts (through which insurers may hold the same underlying credit exposure) may present a materially different analysis than a rated note structure. Because a separately managed account is analyzed asset by asset rather than as a rated note, a managed account holding directly originated, self-liquidating loans may avoid entanglement in the proposed revisions. It is possible that this may make SMAs comparatively more attractive for some insurer investors.
The agenda item notes this proposal is intentionally narrow, and that a separate, broader follow-on project is anticipated to address reporting for multi-collateral structured credit investments, feeder funds, and other fund finance more generally.
VI. OnPoint Observations
We believe a few points are worth flagging as this proposal moves through the comment process.
Other products may get hung up on the self-liquidating test. The ALM Proposal is framed around concerns about multi-collateral structured credit, but the non-self-liquidating test in the proposal is not limited to those transactions. As discussed above, the proposal would require any ABS whose financial assets do not "convert themselves to cash over a defined period in accordance with their contract terms" to overcome the rebuttable presumption against bond treatment rather than qualify through the ordinary ABS pathway. Several transaction types absent from the SAPWG's discussion including data center securitizations, collateralized fund obligations and rated note feeders backed by private equity or infrastructure funds, whole business securitizations, and IP/royalty and fiber securitizations might fail that threshold. Each of these structures shares the same underlying feature: repayment depends on distributions or operating cash flows that are not fixed by contract, rather than on a defined, self-liquidating repayment schedule, so the presumption against bond treatment attaches from the outset. The ramification is that each transaction type would need to be documented, transaction-specific support to overcome the equity-backed presumption, a heavier diligence and disclosure burden with direct consequences for bond versus non-bond classification and risk-based capital treatment. Because the exposure draft does not name or discuss these transaction types, sponsors and insurance company investors in these markets have a clear interest in raising them during the comment period.
The proposal is broader than it purports. The agenda item's background discussion repeatedly frames extreme duration mismatch, debt stated maturities of up to 40 years against materially shorter underlying asset WALs, as the animating concern, and expressly distinguishes it from the shorter, trigger-limited reinvestment periods (5 years or less) that the agenda item treats as an ordinary, unremarkable feature of structured credit generally. The new embedded ALM risk test, however, is not written as a quantitative or duration-based standard. It applies to any self-liquidating asset-backed security with a reinvestment feature. On its face, it is not confined to the extreme scenario described in the background discussion, and could sweep in any structured product with a reinvestment period (including CLOs and commercial real estate (CRE) CLOs) that meets the dangerously low bar discussed below. The NAIC should consider a brightline test that more narrowly addresses the concerns and facts that motivated it.
The "significant embedded ALM risk" definition sets a dangerously low bar. As drafted, embedded ALM risk is significant if cash flows could be impacted by changes in reinvestment rates or spreads. There is no materiality qualifier. There is no probability or likelihood standard. There is no safe harbor or specific thresholds. Under that framing, virtually any structured instrument with a reinvestment feature qualifies because the cash flow profile of any reinvesting structure is in some theoretical sense sensitive to reinvestment rates and spreads. That cannot be what the NAIC intends and would unintentionally jeopardize the treatment of trillions of dollars in structured products. Meaningful calibration, a materiality threshold, a likelihood qualifier, or a safe harbor for transactions where reinvestment risk has been independently identified, stress-tested, and structurally mitigated, is essential if the test is to function as intended.
Stated maturity is the wrong metric. There is a pervasive flaw in the NAIC’s observations about these Instruments in support of the proposals. To the extent duration matters to the analysis, the relevant measure is expected WAL, not stated legal final maturity. Expected WAL reflects actual amortization dynamics and structural mechanics, and is typically considerably shorter than the legal final for these Instruments. A rule focused on stated maturity will systematically overstate duration exposure and penalize instruments whose actual cash flow profiles are considerably more conservative than their outer limit maturities suggest.
At this stage, the proposals are premature. As Sherlock Holmes once observed, "It is a capital mistake to theorize before one has data.” Multi-collateral structured credit investments are a relatively new addition to the structured credit landscape, and, as with any new asset class, there is not yet a meaningful body of data documenting defaults, losses, or reinvestment-related underperformance in these structures. Nor does the agenda item point to any instances of such impairments, including any nascent or incipient ones. In the absence of such evidence, a proposal that would strip bond eligibility from a class of instruments before there is data establishing that the risk it targets has actually materialized is, at best, premature, and the better approach would seem to be for the NAIC to continue to monitor performance as the asset class matures before deciding how to address its concerns.
The interconnectedness concern calls for a different tool. To the extent that this is concern about the collateral overlap across successive transactions, a security-level bond-classification test is not in our view the best way to address this concern. The NAIC has approached comparable transparency questions elsewhere through disclosure-based tools, investment characteristic codes and look-through reporting, as in its separate work on rated notes and rated feeder funds, rather than reclassification. That approach is better suited to this concern, and more consistent with the analytical capacity of the sophisticated institutions the NAIC regulates.
The agenda item is silent on effective date and transition. The exposure draft does not include a proposed effective date or transition guidance. It is accordingly unclear whether any final revisions would apply only prospectively to new issuances or would instead require insurers to re-evaluate the classification of securities they already hold, as occurred when the PBBD itself was first adopted. That distinction has significant practical consequences for insurers currently holding affected paper. This is not a case of European regulators, who have at times shown a greater institutional willingness to adopt measures with retroactive application; U.S. insurance regulators do not operate within that tradition, which makes it all the more important that the NAIC state plainly, and soon, that any final rule will apply only prospectively and will grandfather existing holdings.
The consequences of a reclassification could be severe. Losing bond status typically means the instrument drops out of Schedule D-1 reporting and instead lands on Schedule BA, or in some cases is treated as an equity interest altogether. Because bonds carrying strong ratings are ordinarily assigned very modest capital charges, a shift of this kind could push the applicable risk-based capital charge up toward the 30 percent range, a change that could occur even though the underlying instrument's credit rating has not moved at all.
The interaction with SSAP No. 43R warrants clarification. Because the proposal operates through the existing rebuttable presumption framework, commenters should seek clarity on how the new test is intended to interact with SSAP No. 43R and the rebuttable presumption more generally, so that the two frameworks are not applied inconsistently to the same instrument.
VII. Next Steps
The proposal is now on SAPWG's active agenda for comment, with comments due by October 2, 2026. Insurers holding or considering multi-collateral structured credit investments, particularly those with reinvestment features, longer-dated debt tranches, or exposure across multiple deals from the same sponsor platform, should assess how the proposed revisions would apply to their current and prospective holdings, and should consider whether comments would be useful to submit during the comment process. Sponsors and insurer clients, directly or through trade associations, may wish to use the comment period to address the underlying necessity and breadth of the proposal; the meaning of "significant" embedded ALM risk; whether structural mitigants and subordination should be given express credit; the interaction with SSAP No. 43R and the existing rebuttable presumption framework; and the need for prospective application and grandfathering of existing holdings. We are continuing to monitor this item and will provide updates as it develops. Please contact us if you would like to discuss how this proposal may affect existing or contemplated transactions.
