Recent proposed modifications to Valuation Manual 22 (VM-22) represent a significant step forward in the ongoing evolution of statutory reserve requirements for life and annuity products in the United States. Its introduction is part of a broader shift to move away from prescriptive, one-size-fits-all formulas toward a more dynamic framework that integrates actuarial judgment and asset liability management (ALM) modeling with enhanced transparency and governance. VM-22 principle-based reserve (PBR) establishes a standardized framework for determining reserves for non-variable annuities.
VM-22 PBR is expected to take effect January 1, 2026, and be applicable to products in scope issued on or after that date.
Non-variable, non-exempt, annuity contracts specified in VM Section II, Subsection 2 “Annuity Products,” Paragraph D and applicable contracts in VM Section II, Subsection 3 “Deposit-Type Contracts” are subject to VM-22 requirements. The products in scope fall under two reserving categories: the “Accumulation Reserving Category” and the “Payout Annuity Reserving Category” as shown in the table below. There is also a “Longevity Reinsurance Reserving Category,” which encompasses reinsurance agreements where an insurance company assumes the longevity risk with periodic payments under immediate or deferred payout annuity contracts.
| Accumulation reserving category[1] | Payout annuity reserving category |
|---|---|
| Fixed indexed annuities (FIAs) | Single premium immediate annuities (SPIAs) |
| Single and flexible premium fixed deferred annuities (FDAs) | Pension risk transfer (PRT) |
| Multi-year guarantee annuities (MYGAs) | Deferred income annuities (DIAs) |
| Structured settlement contracts (SSCs) |
Preneed Annuities, Guaranteed Investment Contracts, Synthetic Guaranteed Investment Contracts, Funding Agreements, and other Stable Value Contracts are not in scope.
For products in scope, under the new framework, companies must first determine whether to elect the Single Scenario Test (SST). If a company meets all the conditions of the SST and passes a specific portion of the Stochastic Exclusion Test (SET), it may satisfy VM-22 requirements using the Deterministic Reserve (DR), thereby avoiding the need for complex stochastic modeling. The DR is a reserve based on a single prescribed scenario using company-specific assumptions, offering a simpler compliance path when applicable.
If a company does not elect or qualify for the SST, it may still meet certain SET criteria to continue using pre-VM-22 reserving methodologies. Otherwise, it must comply with VM-22 through the Stochastic Reserve (SR), which is calculated using a range of economic scenarios to capture tail risk and reflect policyholder behavior variability.
In both the DR and SR approaches, companies are also required to calculate the Standard Projection Amount (SPA), which is based on fully prescribed assumptions. However, the SPA result is a disclosure-only item, rather than a binding floor.
At its core, VM-22 is intended to replace older, formulaic approaches — most notably current Commissioners Annuity Reserve Valuation Method (CARVM) Actuarial Guideline XXXIII and XXXV (AG 33 and AG35) — with a methodology that more accurately reflects the underlying economics and risk profile of the business.
| Feature | Current CARVM | VM-22 PBR |
|---|---|---|
| Method | Prescribed | Principle-based |
| Assumptions | Prescribed, limited flexibility | Prudent estimate with company-specific experience, when possible |
| Risk sensitivity | Lower | Higher |
| Use of Scenarios | None | Deterministic and stochastic |
| Reflection of general account and hedge assets’ | No | Yes |
| Product scope | Immediate and fixed deferred annuities and SSCs | Non-variable annuities |
| Governance and documentation | Minimal | Detailed documentation and governance required |
| Modeling requirements | Lower | Higher |
| Aggregation benefits | No | Yes, but with some restrictions for aggregating across reserving categories |
| Standardized floor | N/A | N/A[2] |
Compared with the current valuation methods, VM-22 provides greater flexibility in modeling assumptions and incorporates both deterministic and stochastic scenario approaches, allowing for a broader view of underlying risks and claim paying capabilities. It emphasizes the use of company-specific assumptions where appropriate, while still requiring prudent margins to ensure conservatism and comparability across companies. For actuaries, the regulation brings both new challenges and opportunities, as it raises the bar for assumption setting, model capabilities, governance and cross-functional collaboration in terms of product design and risk management.
The impact on reserves due to the implementation of VM-22 is still being quantified and proves difficult to assess in general terms due to the company- and product-specific nature of the framework. Unfortunately, industry field testing results have shown mixed results, with reserve impacts varying widely across participants. Within the field testing, results even varied in terms of if the SPA or the SR was greater. These findings (or lack thereof) naturally raise a key question: Why do some insurers see increases in reserves while others see decreases within the same product type, when compared with CARVM?
One likely explanation lies in differences in product design and the nature of embedded guarantees. Even among relatively simple products such as fixed deferred annuities (FDAs), variation in key features can lead to significantly different outcomes. For instance, an FDA with a guaranteed minimum interest rate (GMIR) of 1% will behave differently under VM-22 compared with one with a GMIR of 4%, all else being equal. A similar principle applies to more complex products. A fixed indexed annuity (FIA) with a guaranteed living withdrawal benefit rider, as another example, will show more favorable reserve outcomes (i.e., decreases in reserve levels) when prudent estimated policyholder behavior assumptions deviate from fully optimal utilization used in CARVM.
Another explanation can be found in the limitations and caveats reported in the field testing results, such as suboptimal asset selection or assumption choices. A change in the reserving framework will likely lead to different asset selection criteria, which wasn’t reflected in field testing, as insurers adjust and optimize their asset and liability management process under the new regime. This leads to another important insight: The impacts of VM-22 will not be driven by reserve methodology alone. Secondary effects such as changes to asset allocation as well as VM-22-driven enhancements to assumptions will play a significant role as well.
A related yet critical aspect of the new framework is how reserves will behave over time. VM-22 introduces a structure that is inherently more sensitive to changes in economic conditions — particularly interest rates, credit spreads and market volatility. VM-22’s use of deterministic and stochastic scenarios means that reserve calculations will now dynamically respond to fluctuations in the economic environment, aligning statutory valuation more closely with economic value.
The introduction of VM-22 will have ripple effects beyond valuation; it will directly influence product design, pricing strategy, capital management, and reinsurance appetite and structures. As reserve requirements become more reflective of product risk, actuaries will need to recalibrate how new products are designed and priced and how existing blocks are managed.
Product design and pricing strategy. As insurers gain more insight into how VM-22 affects reserving levels and volatility for their products, many may begin to adjust product features to optimize reserve results or limit volatility due to economic sensitivity.
Capital considerations. Because statutory reserves influence statutory capital, VM-22 will make capital levels more product-specific, assumption-sensitive and economically responsive. The result may be weakened or volatile risk-based capital ratios, even if the company’s actual risk hasn’t changed.
Reinsurance. In cases where VM-22 results in lower reserves, insurers may see a reduced need for reserve financing arrangements such as captives or letters of credit. However, insurers may turn to reinsurance solutions, particularly for blocks with significant optionality or tail risk, as a way to stabilize reserves and capital.
With VM-22 set to take effect on January 1, 2026, insurers have a limited window to prepare for a regulation that will significantly impact valuation, pricing, governance and risk management for non-variable annuity products. Getting ready involves more than just model updates; it requires a proactive transition strategy with the following key steps:
For insurers, VM-22 introduces new complexity: Reserve levels may increase or decrease as markets shift, even if the underlying policies remain unchanged. This has important implications for capital planning, earnings volatility and asset-liability management. Actuarial teams will need to consider not just the reserve level at issue but also how reserves might evolve under varying scenarios — and how that volatility could impact the company’s financial position and strategic decisions.
While this article focuses on the foundational elements of VM-22, future pieces will delve deeper into how specific product features within each annuity product type influence VM-22 outcomes by isolating individual product design levers within our actuarial modeling platform.
