LDTI's First Full Year for Non-Public Life Insurers in 2026: What Mutuals, Fraternals, and Smaller Stock Companies Should Be Disclosing

The public-filer cohort has now produced three years of LDTI disclosures. Those filings are a roadmap. Every earnings surprise the SEC filers absorbed between Q1 2023 and Q4 2025, from discount-rate unlock swings to MRB fair-value whiplash to the asymmetric drag of the LFPB zero floor, is waiting in the non-public cohort's first full year of annual statements. The difference is audience. Mutuals and fraternals do not report to analysts on quarterly calls. They report to policyholder advisory committees, state regulators, rating agencies, and boards of directors who are seeing LDTI-framed numbers for the first time.

This article is written for the actuaries, controllers, and disclosure leads inside those non-public insurers who are drafting their first full-year LDTI footnotes right now. It is also for the practicing consultants who are helping them translate implementation work from 2024 and 2025 into defensible, comparable, decision-useful 2026 disclosures. The focus is narrow on purpose: what the standard requires, what the public-filer experience has taught us about where the numbers swing, and how to build management discussion and footnote narrative that holds up to board questions, actuarial audit, and peer benchmarking.

What the Non-Public Cohort Is Actually Filing

ASU 2018-12 was effective for non-public entities for fiscal years beginning after December 15, 2024, with interim reporting requirements extending to fiscal years beginning after December 15, 2025. For a calendar-year insurer, that means the 2025 annual financial statements are the first full set of disclosures prepared under LDTI, and 2026 quarterly reporting introduces interim LDTI disclosure for the first time. Non-public entity in this context follows the FASB definition: essentially any entity that is not an SEC filer and not a conduit bond obligor with publicly traded debt. The population includes mutual life insurers, fraternal benefit societies, smaller stock companies whose debt is not publicly traded, and the life insurance subsidiaries of privately held holding companies.

The core disclosure package under LDTI for these filers consists of four building blocks. The liability for future policy benefits (LFPB) for traditional and limited-payment long-duration contracts, measured at the net premium ratio with assumptions updated annually and discount rates flowing through other comprehensive income. The liability for policyholder account balances for universal life, deferred annuities, and investment contracts, measured at the balance credited to the policyholder with enhanced disclosure of interest crediting and surrender activity. The market risk benefit (MRB) liability or asset for variable annuity guarantees, fixed indexed annuity guarantees, and certain universal life features that protect policyholders from capital market risk and expose the insurer to other-than-nominal capital market risk. And deferred acquisition costs (DAC) amortized on a constant-level basis over the expected term of the related contracts without linkage to gross profits or margins.

Each of the four balances requires a disaggregated rollforward. The rollforward standard is the same one that has produced most of the pain for public filers: present value of expected future benefits, present value of expected future premiums, the net premium ratio, actual versus expected experience, assumption changes, discount rate effect separated from cash flow effect, and the opening-to-closing tie. The disaggregation has to be at a level that is meaningful for users of the financial statements, which the standard does not define precisely. For a mutual with one block of participating whole life, two blocks of universal life by issue era, and a runoff block of variable annuity with GMDBs, the practical disaggregation is usually four to six cohort groups. For a fraternal with a single dominant whole life product and a small annuity book, two or three is defensible.

Where the Public-Filer Experience Points the Non-Public Cohort

From tracking SOA updates and public-filer earnings disclosures through 2023, 2024, and 2025, the volatility drivers have clustered into three categories. The first two are familiar from the implementation literature. The third has been the surprise.

MRB fair value changes. Market risk benefits are measured at fair value using risk-neutral stochastic scenarios. When equity markets fall, the liability for GMDBs, GMWBs, and GMIBs generally rises, and the fair value change flows through net income. When interest rates fall, the present value of projected guarantee payments rises, also hitting net income. The mechanics produce earnings swings that are decoupled from the underlying claim experience, and public filers have consistently cited MRB fair value changes as the largest single driver of quarter-to-quarter GAAP earnings volatility. A non-public insurer with a legacy variable annuity block from the 2005 to 2012 era will see this pattern even if the book is closed to new business, because the in-force guarantees continue to be measured at fair value every quarter. The SOA Financial Reporting Section's comparison of IFRS 17 and LDTI has documented this as the single largest accounting difference between the two regimes: IFRS 17's contractual service margin (CSM) smooths recognition of guaranteed benefit reserves over the coverage period, while LDTI recognizes MRB fair value changes immediately.

Discount-rate unlock. The LFPB discount rate under LDTI is an upper-medium-grade (A-rated) fixed-income instrument yield, updated at each reporting date. The effect of the discount rate change on the LFPB flows through accumulated other comprehensive income (AOCI), not net income. That sounds benign. In practice, for mutuals with large blocks of participating whole life and limited-payment business, the AOCI swing can be enormous relative to the insurer's GAAP equity. A 50 basis point decline in the A-rated yield at year-end can increase the LFPB by a multiple of the insurer's annual net income, with the offset landing entirely in AOCI. The dislocation is not an earnings problem, but it is a governance and communication problem: boards need to understand why policyholder surplus fell sharply on the GAAP balance sheet while statutory surplus moved modestly, if at all.

Assumption review cadence. LDTI requires annual assumption updates at the same time each year. For most insurers, the chosen date is the fourth quarter, which means the Q4 close concentrates mortality, lapse, expense, and utilization assumption changes into a single reporting period. The resulting Q4 net income number is therefore structurally noisier than Q1, Q2, or Q3. Public filers have responded with expanded MD&A narrative separating "baseline" earnings from the impact of the annual assumption review. Non-public insurers do not file MD&A, but rating agency analysts, audit committees, and policyholder advisory committees will ask the same questions. Preparing a parallel narrative for those audiences is one of the most important soft deliverables of a first-year close.

The LFPB Zero Floor and Why It Creates Asymmetric Volatility

One feature of the LFPB measurement model that has produced more board-level confusion than any other is the zero floor. Under LDTI, the LFPB is calculated as the present value of future benefits and expenses less the present value of future gross premiums, using a net premium ratio that is capped at 100 percent. If the net premium ratio calculated at a cohort level exceeds 100 percent, the standard caps it at 100 percent, and the resulting LFPB becomes a liability equal to the present value of future benefits less the present value of future gross premiums (with the net premium ratio applied at the cap). Once a cohort breaches the cap, it is effectively "onerous" in LDTI terms, although the standard does not use that word.

The asymmetry is important. When assumptions deteriorate enough to breach the cap, the LFPB increases sharply and the excess is recognized immediately in net income. If assumptions subsequently improve, the LFPB decreases, but recovery is bounded: once the cap is hit and an immediate loss is booked, subsequent favorable movement cannot produce an immediate gain of the same magnitude without affecting future premium recognition patterns. The net premium ratio can only decline back from 100 percent as experience improves and is revalidated in subsequent annual reviews. In practice, breaching the cap creates a ratchet: losses emerge immediately, gains emerge gradually.

Compare this to IFRS 17. Under IFRS 17's General Measurement Model, the CSM absorbs favorable assumption changes over the remaining coverage period and unfavorable changes are offset against the existing CSM balance until that balance is exhausted. Only when the CSM is fully absorbed does the contract become onerous and is a loss recognized immediately. LDTI has no equivalent smoothing mechanism. Milliman's LDTI-versus-IFRS-17 comparison and the SOA Financial Reporting Section's bridging analysis both flag this as one of the primary economic reasons the two standards produce different earnings patterns even on identical portfolios.

Disaggregated Rollforwards: Where Auditors Spend Their Time

The rollforward is the disclosure item that has generated the most auditor friction during 2025 adoption work. The required rollforward of the LFPB under ASU 2018-12 separates several moving parts: the present value of expected net premiums at the beginning of the period, adjustments for actual experience, assumption changes recognized retrospectively, discount rate effects, interest accretion, benefit payments, and the present value of expected net premiums at the end of the period. Each component has to reconcile exactly, and each has to be disclosed at a level of disaggregation that allows a financial statement user to understand the drivers of the change.

Three practical issues recur across first-year non-public close projects. First, cohorting choices made during implementation often need to be revisited once the rollforward numbers emerge. A cohorting scheme that looked defensible on a static balance-sheet view can produce rollforward components that tell a confused story, with large offsetting cash flow and assumption effects inside a single cohort. The remedy is usually finer disaggregation, which auditors tend to push toward regardless of the preparer's initial preference. Second, actual-versus-expected disclosure is required for mortality, morbidity, and lapse, and the "expected" baseline has to be the assumption used to calculate the opening LFPB, not the revised assumption adopted in the current period. Getting the timing right on this disclosure is a common first-year error. Third, the disclosure of significant inputs, judgments, and sensitivity analyses has to go beyond boilerplate. Public-filer 2024 and 2025 footnotes moved steadily toward more quantitative sensitivity tables as peers set the benchmark, and non-public filers will face the same expectation.

Patterns we have seen in recent public-filer disclosure reviews suggest that the "expected term of liabilities" disclosure is where peer benchmarking first becomes visible. Companies with heavier single-premium immediate annuity blocks disclose longer weighted-average durations; companies with shorter-duration term life disclose shorter durations. For a mutual disclosing this metric for the first time, it is worth comparing to at least three peers of similar product mix to ensure that the calculation method and disclosure framing are aligned with practice.

Practical Challenges for Mutuals and Fraternals

The non-public cohort faces implementation constraints that public filers largely did not. Three are structural.

Limited historical cohort data. LDTI requires MRB measurement at fair value on a full retrospective basis regardless of the overall transition method elected. For a fraternal that has held a small variable annuity block since the early 2000s, reconstructing the inputs needed for retrospective MRB fair-value calculation back to inception can require going to paper records, legacy policy administration systems that have been migrated multiple times, and historical market data that the insurer did not previously maintain at the granularity LDTI requires. The KPMG benchmarking work found that 50 percent of 2025 adopters reported significant use of external resources during implementation; the data reconstruction work is a leading reason.

Cohorting choices in a small-block environment. Public filers have enough product diversity to absorb cohorting errors. A mid-sized mutual with one dominant product line does not. Cohorting has to satisfy ASU 2018-12's requirement that contracts be grouped by issue year and by a basis that is not inconsistent with underwriting or pricing, but it also has to avoid concentrating adverse experience. A pragmatic first-year approach is to cohort at the narrowest level the actuarial system supports and then test aggregated disclosure at progressively broader levels to see where the rollforward story becomes most interpretable. The auditor and the preparer rarely land on the same choice on the first pass, and the conversation is usually most productive when framed around rollforward clarity rather than abstract cohorting principles.

Peer benchmarking gap. Mutuals and fraternals traditionally benchmark primarily against statutory filings, where comparability across peers is high. Under LDTI, GAAP comparability across non-public peers is limited by the fact that most non-public insurers do not publish their financial statements in a form that supports easy peer access. Rating agencies and the NAIC's Financial Analysis Handbook provide some cross-peer data, but a non-public insurer drafting its first LDTI narrative generally has to anchor the discussion to the public-filer cohort, even though product mix, investment strategy, and accounting policy elections may differ meaningfully.

Intersection With PBR, CECL, and Statutory Asset Adequacy

LDTI does not replace statutory reporting, and a non-public mutual or fraternal has to operate three reserve frameworks in parallel. Statutory reserves under the Valuation Manual, including principle-based reserves (PBR) under VM-20 for life and VM-21 for variable annuities. GAAP reserves under LDTI. And CECL estimates for reinsurance recoverables, investment portfolios, and certain receivables under ASC 326. Each framework has different measurement objectives, different assumption sets, and different disclosure requirements.

Deloitte's LDTI regulatory intersection analysis and RSM's U.S. GAAP long-duration improvements practice guide both flag the same practical points. The LDTI assumption set and the PBR deterministic reserve assumption set are not the same: LDTI uses best estimate assumptions with margins embedded only in the discount rate convention, while VM-20 uses prudent estimate assumptions with explicit margins. Cash flows modeled for LDTI cannot be reused without adjustment for PBR, and vice versa. Insurers that tried to unify the two assumption sets during 2025 implementation have generally found it was faster to maintain parallel assumption inventories with documented mapping than to force a single assumption set through both frameworks.

Statutory Asset Adequacy Testing (AAT), conducted under Actuarial Guideline 51 and related VM-30 requirements, adds a third layer. The Appointed Actuary's opinion on asset adequacy relies on cash-flow testing using assumptions that may differ again from both PBR and LDTI. For offshore and affiliated reinsurance arrangements, Actuarial Guideline 55 adds a new layer of asset adequacy disclosure that is due April 1 each year for year-end testing; for mutuals and fraternals with affiliated reinsurance cessions to captive or offshore entities, the AG 55 deliverable sits alongside the LDTI disclosure as part of the first-quarter work calendar.

The intersection is not theoretical. When a board asks why the GAAP LFPB moved sharply but the statutory reserve barely changed, the answer is a direct consequence of the three-framework structure. GAAP moved because the A-rated discount rate moved and the assumption review cycle hit. Statutory did not move because VM-20's discount rate framework is different and the statutory assumption set was not touched. The Appointed Actuary's opinion on adequacy remains supported even as the GAAP equity number swings.

Translating MRB-Driven Swings Into Economic Reality

The Footnotes Analyst has argued in recent commentary that LDTI's MRB fair value recognition produces accounting volatility that does not correspond to economic volatility on a hedged or partially hedged variable annuity block. For an insurer that has hedged its MRB exposure using dynamic futures, options, or swap positions, the hedge gains and losses flow through net income in a pattern that approximately offsets the MRB liability movement. For an insurer that has not hedged, or is partially hedged, the net income swing captures both the economic reality of changed guarantee exposure and the absence of an offset.

The communication challenge for a non-public filer is that the board or policyholder committee is seeing the net number, not the underlying attribution. Three practices from the public-filer cohort have worked well.

First, separate MRB fair value changes from underlying earnings in board reporting, with explicit attribution of the portion offset by hedge gains and losses and the portion that remained unhedged. Some public filers have introduced non-GAAP "core earnings" or "operating earnings" measures that exclude MRB fair value changes; non-public insurers can adopt similar internal conventions without issuing external press releases, simply by using them in board decks and audit committee packages.

Second, present a multi-quarter trailing view of MRB fair value changes. Single-quarter swings look alarming. Four-quarter trailing sums reveal whether the swings are mean-reverting (which they usually are for a hedged block) or trending (which would require investigation). Boards digesting LDTI numbers for the first time benefit substantially from this framing.

Third, cross-walk MRB fair value changes to the economic exposure being measured. The GMDB fair value liability at reporting date is a function of in-the-moneyness, projected guarantee claim frequency, and the risk-neutral discount rate applied to projected guarantee payments. When the fair value swings, one of those three inputs moved. Identifying which, and whether the movement is likely to persist or reverse, is the core of the economic narrative that translates accounting volatility into actionable insight.

Building the First-Year Disclosure Narrative

The LDTI footnote is formally prescribed. The narrative framing is not. Based on reviewing public-filer 2023 and 2024 footnotes against their 2025 versions, several narrative choices have emerged as practitioner norms.

The introduction to the long-duration contract footnote typically acknowledges that the insurer adopted ASU 2018-12 effective January 1, 2025 (for non-public filers), briefly describes the transition method elected, and notes any modified retrospective elections. A sentence acknowledging that prior periods have been restated under the elected transition method is standard. A sentence explaining that MRBs have been measured at fair value on a full retrospective basis regardless of the overall transition method is also standard and worth including because it signals to auditors that the preparer understands the distinct retrospective treatment for MRBs.

The disaggregated rollforward narrative is the most important section. Each cohort's rollforward should be accompanied by a short paragraph identifying the major drivers of the change: actual versus expected experience, assumption updates, discount rate effects, and significant cash flow or benefit events. The narrative does not need to quantify every driver in running text (the rollforward table does that), but it should identify which driver was largest and why. For a mutual with a large participating whole life block, this will typically be the discount rate effect running through AOCI. For a fraternal with a concentrated mortality block, it will typically be mortality assumption updates. For a closed block of variable annuities, it will typically be MRB fair value changes.

The significant inputs and judgments disclosure is where auditor expectations have risen between 2023 and 2025. Quantitative sensitivity to mortality assumption changes, lapse assumption changes, and discount rate changes is now the expected floor rather than an aspirational addition. Non-public preparers who plan to disclose only qualitative narrative should expect audit pushback and may lose the argument even if their draft is technically compliant with the literal text of the standard.

Why This Matters

LDTI's first full year for the non-public cohort is not primarily a technical accounting exercise. The technical work was largely completed during 2025 implementation. What 2026 reveals is whether the cohorting choices, assumption governance, and disclosure framing put in place during implementation hold up under the weight of actual reporting, actual audit review, and actual stakeholder communication.

For mutuals and fraternals, the stakes are governance credibility. Policyholder advisory committees and boards that have spent decades interpreting statutory surplus movements will now see GAAP equity swings that statutory reporting does not produce. The actuary's job is to build the interpretive bridge, not just to produce the numbers that cross it. For smaller stock insurers with non-public debt, the stakes include rating agency dialogue and creditor relationships where LDTI-framed numbers are now a primary data source.

For the profession, the non-public cohort's 2026 disclosures are the second leg of a natural experiment in long-duration contract accounting. The first leg, the 2023 to 2025 public-filer experience, showed that LDTI produces the intended transparency at the cost of earnings volatility that is novel to U.S. GAAP insurance reporting. The second leg will show whether that volatility is manageable in a reporting environment without quarterly analyst calls, and whether the American Academy of Actuaries' May 2025 agenda request for targeted refinements gains the traction that public-filer experience alone did not produce.

---

---

Sources

1. FASB, "ASU 2018-12, Financial Services - Insurance (Topic 944): Targeted Improvements to the Accounting for Long-Duration Contracts," August 2018 - fasb.org
2. Cherry Bekaert, "How LDTI Affects Insurance Accounting Standards," August 2025 - cbh.com
3. SOA Financial Reporting Section, "Bridging the GAAP: IFRS 17 and LDTI Differences Explored," July 2022 - soa.org
4. Milliman, "IFRS 17 vs. US GAAP LDTI: Different Animals?," December 2019 - milliman.com
5. Oracle, "LDTI vs IFRS 17: A Comparison of Long-Duration Insurance Contract Accounting Standards" (white paper) - oracle.com
6. Deloitte, "LDTI: Targeted Improvements to the Accounting for Long-Duration Contracts," 2025 update - deloitte.com
7. FASB Accounting Standards Codification Topic 944, Financial Services - Insurance, as amended by ASU 2018-12, ASU 2019-09, ASU 2020-11, and ASU 2022-05 - fasb.org
8. Footnotes Analyst, "Insurance accounting: Economic versus accounting volatility under IFRS 17 and LDTI" - footnotesanalyst.com
9. KPMG, "Benchmarking LDTI Implementation," December 2024 - kpmg.com
10. American Academy of Actuaries, "Application of ASU 2018-12 to the Accounting for Long-Duration Contracts under U.S. GAAP" (Practice Note), December 2023 - actuary.org
11. American Academy of Actuaries, Life GAAP Reporting Committee Agenda Request to FASB on LDTI, May 2025 - actuary.org
12. RSM, "U.S. GAAP Long-Duration Targeted Improvements: Implications for Insurance Companies," 2024 - rsmus.com
13. SOA, "The New Face of LDTI Under US GAAP" (e-Newsletter), October 2025 - soa.org
14. NAIC Valuation Manual, Sections VM-20 (Life PBR), VM-21 (Variable Annuities), and VM-30 (Actuarial Opinion and Memorandum) - naic.org
15. NAIC Actuarial Guideline 55, Application of Asset Adequacy Testing to Certain Reinsurance Arrangements - naic.org

---

Further Reading on actuary.info

• LDTI / ASU 2018-12 in 2026: The Full Implementation Guide
• AG 55 First Filing Hits: What Life Actuaries Learned
• IFRS 17 Implementation in 2026: The Global Insurance Accounting Transformation
• Life Insurance & Annuity Market Trends 2026
• NAIC Regulatory Developments 2026
• Long-Term Care Insurance Crisis 2026