AG 53's review turns to how assets behave under stress

For eighteen months, the AG 53 project answered a narrow test—can insurers report private and structured assets consistently?—and that test is now being retired as the framework's center of gravity moves toward the question Insurance AUM Journal's new retrospective foregrounds: how an asset behaves when conditions turn. Supervisors built the reporting machinery first, and the review, published this week, finds they are now using it to decide which reported numbers deserve a second look—the ones a general account's risk profile actually rests on.

When the journal first assessed AG 53 in April 2025, it framed success as three things: consistency, materiality, and coordination. Its eighteen-month answer is that the industry is getting there, and that the discussion has grown more sophisticated—a fair verdict and a reminder of how quickly a framework can move from forms to judgment. The evidence sits with the Valuation Analysis Working Group, whose reviews have spread beyond unusually high net-yield assumptions into cliff risk, illiquidity risk, and Level 3 valuation, territory that concerns the investment committee as much as the examiner.

A net-yield assumption is a figure an actuary can test, whereas cliff risk and illiquidity are behaviors, and probing them means asking how a position unwinds once its assumptions stop holding. Regulators have followed that logic into downside scenarios, running higher-yielding and illiquid holdings through stress cases even when the outcome leaves reserves adequate—a practice the review calls constructive—because an investment that clears an expected-conditions test can still sour once liquidity thins or cash flows arrive late, and a scenario's value is that it surfaces the weakness in disclosure rather than in a downgrade. The same reasoning applies with more force as insurers lean into private credit and structured paper, where risk often sits outside both a credit rating and an asset-class label.

None of this is an argument against complexity, and the review is careful to say so, because private and less liquid assets do real work in a general account, diversifying the portfolio and letting insurers hold exposures that suit long-dated liabilities. The failure mode is a complex asset whose economics hold only while the right assumptions survive stress. Telling those two apart is the supervisory task, and the review frames it as a shared one: regulators must draw the distinction, and insurers need the in-house expertise to reach the same verdict rather than wait for a regulator to reach it first.

The NAIC's year-end 2025 AG 53 guidance is where the shift becomes operational, asking for more consistent reporting and more detail on Schedule BA holdings—feeder funds, collateral loans, structured notes—and it adds a requirement that deserves more weight than its billing suggests: insurers must explain how their investment departments and their asset managers interact.

That line turns alignment from a slogan into something examinable, because asset adequacy testing cannot live entirely inside the actuarial function when the questions that decide it are about how investments behave economically; the actuary holds the liabilities, the investment desk holds the assets, and the asset manager holds the structures, with AG 53 increasingly the place all three meet. An insurer that outsources a general account and cannot describe how the manager's judgment reaches its own investment committee has an answer the guidance now obliges it to give, and the answer is a test of operating model as much as of any single holding.

AG 53 also belongs to a longer sequence, and the review is explicit that it should not be read in isolation. At its 2026 Summer National Meeting, the NAIC adopted changes to the C-1 risk-based capital framework, carrying the same scrutiny from the valuation question to the capital charge that follows it. How far those changes reach into private credit is the open end of the story; the review's account stops at the adoption.

The direction fits what this publication has argued: the NAIC is writing the capital charge that will end the private-credit rating arbitrage, and rating committees already are pricing it. The year-end guidance is the softer half of the same move, and the sequencing is deliberate, because before a charge can be calibrated, someone has to establish what the assets are and who is attesting that they are what they claim to be. The manager-interaction disclosure is where the NAIC gathers that evidence, and it is also where the asset-manager annexation of insurance meets its test: a manager can hold the structures, but the insurer has to be able to explain the economics back to its own committee.

For a general account, the practical read is that the burden of proof moved inward: an insurer able to show how its investment team, its actuary, and its outside manager reach the same view of a structured asset under stress is already where examinations are heading, while one that answers the new requirement with an organizational chart has a harder story to tell. Watch how far the C-1 changes reach into the private-credit book, because the disclosure AG 53 now demands is the evidence that charge will be built on.