RESEARCH DOSSIER
Participating Life Insurance Dividend and Surplus Mechanisms
Abstract. Participating life insurance integrates guaranteed mortality-contingent protection and long-term accumulation with an institutionally governed sharing of favorable experience. Its research problem is the conversion of aggregate insurer experience into policyholder dividends, bonus additions, premium offsets, and other benefits under long-duration financial constraints. This monograph examines the contractual basis of participation, the actuarial determination of divisible surplus, the contribution principle, dividend-factor classes, asset-share and smoothing practices, guaranteed and discretionary benefits, interest-rate and surrender options, consumer disclosures, and policyholder–shareholder conflicts. It compares U.S. actuarial standards and state law, U.K. with-profits regulation, European Solvency II and IFRS 17 measurement, and the documented Hong Kong supervisory setting. An original analytical synthesis connects favorable experience, authorized distributable surplus, cohort allocation, declaration, and policyholder realization. Carefully labeled hypothetical numerical reconciliations illustrate the difference between arithmetic consistency, distributive fairness, and legal entitlement. Peer-reviewed contingent-claims and optimal-stopping studies establish why guarantee costs and surrender rights alter the economics of distributions. The documentary and conceptual findings are supported by primary sources and original scholarly publications; the dossier presents no proprietary new theory or empirical estimates from insurer microdata.
Keywords: Participating life insurance; policyholder dividends; divisible surplus; contribution principle; with-profits; asset share; smoothing; guarantee; discretionary benefits; reversionary bonus; terminal bonus; paid-up additions; insurer capital; policyholder rights; surrender option; actuarial governance; IFRS 17; intergenerational equity.
Introduction
Participating life insurance is a distinctive institutional form of long-duration risk sharing. The insurer undertakes specific contractual obligations while retaining a process through which part of favorable investment, mortality, expense, or other experience can return to eligible policyholders. A participating policy therefore has several economically distinct benefit states: guaranteed benefits specified in the contract; dividends or bonuses already declared and credited; future discretionary distributions; and projected values contingent on assumptions that may never be realized. Contract interpretation, actuarial surplus analysis, financial-option valuation, accounting measurement, and consumer communications each describe part of the same arrangement.
The central questions of this dossier are: how the participation entitlement is defined; how a participating block's experience is translated into divisible surplus; how dividend-factor and smoothing practices affect fairness among policies and generations; what guarantees and surrender options do to liability value; and what an empirical study would have to observe to establish actual policyholder outcomes. These issues arise across jurisdictions, but the obligations and terminology are locally specific. In the United States, actuarial contribution and state participating-policy statutes are central. In the United Kingdom, with-profits fund principles, documented discretion, and policyholder governance matter. European prudential accounting separately identifies future discretionary benefits and surplus funds. Hong Kong supervision adds a contemporaneous fund-governance setting.
The research method combines a source-specific legal and regulatory comparison with critical analysis of actuarial standards and peer-reviewed financial economics. It reconstructs the economic sequence from premiums and claims through declared benefits, presents transparent hypothetical arithmetic, and specifies empirical variables and identification challenges. Primary statutes are the authorities for statutory claims; actuarial standards are authorities for recommended professional practice; regulatory guidance governs only its operative jurisdiction; financial reporting defines an accounting basis; and academic models establish results only within their assumptions. This hierarchy prevents a modeled policy value, an illustrated scale, or an accounting surplus balance from being misrepresented as a particular policyholder's enforceable right.
Participation in dividends is distinct from the related research dossier devoted to accounting and allocation inside participating life funds. This monograph emphasizes the contract, the flow of surplus into policyholder benefits, and the implications for households and risk-bearing institutions. General mortality forecasting, reinsurance, reserve financing, and product pricing are brought in only where they determine the participating mechanism. The full Science on life insurance supplies the larger disciplinary setting.
All numerical examples are hypothetical. No insurer-specific contract, confidential experience dataset, or external peer-reviewed original study was supplied for empirical estimation. Conclusions concern verified legal and actuarial structures, the logical properties of documented financial mechanisms, and the transferability of published mathematical analyses. Claims about future dividends are necessarily conditional on contract rules, experience, and financial conditions.
Definitions, Historical and Documentary Foundations
Participating contracts and the meaning of surplus
Participating life insurance is a policy form under which a holder can be eligible to receive favorable-experience distributions in addition to specified guarantees. These distributions may be described as policy dividends, regular or final bonuses, credited additions, or shareable fund benefits. The terminology by itself does not determine the holder's rights. Contract clauses, applicable law, insurer practice, and the declaration decision establish when an amount becomes a liability and whether it can later be revised. The generic insurer promise, a dividend scale, and the contract's enforceable coverage schedule must be interpreted separately.
The term surplus has multiple technical meanings. An insurance company's accounting surplus or own funds is a residual subject to a valuation basis and regulatory rules. A participating block can produce favorable experience relative to an actuarial assumption without the entire amount being immediately distributable. Divisible surplus is the amount made available for distribution under a particular company's dividend framework. An asset-share calculation may provide a reference value for policyholder allocation but does not automatically equal cash surrender value. A reported future discretionary benefit is a valued insurance liability rather than a dividend check. The definitions in ASOP No. 15 distinguish dividend factors, divisible surplus, policies, experience, and contribution. Keeping these meanings separate is essential to every comparison.
The United States retains a documentary framework of annual dividends and policyholder elections. Massachusetts General Laws § 140 specifies participating-policy surplus distribution after provisions for liabilities and reserves and describes cash, premium-reduction, paid-up additions, and accumulation options, subject to its detailed conditions. New York Insurance Law § 4231 supplies another state-specific framework for participation in surplus. These are separate state enactments; neither is a general statement of every U.S. contract.
The participating arrangement is historically associated with mutual life insurance, but the distribution mechanism and the ownership of the insurer must be analyzed independently. A stock insurer can issue participating business and a mutual can offer nonparticipating coverage. In both, senior contractual liabilities, costs of capital, retained financial buffers, and the rights of distinct policy classes remain relevant. A claim that mutual ownership removes all distribution conflicts fails when current policyholders, future policyholders, surrendering holders, and managers have different interests.
The documentary architecture of with-profits insurance
The U.K. uses the regulatory category with-profits for relevant participating insurance business. The FCA's COBS 20 rules include fund identification, the fair treatment of policyholders, Principles and Practices of Financial Management (PPFM), communications, and governance arrangements. The rules require applicable firms to document their principles and practices and address competing policyholder and shareholder interests. Some sections were amended on June 26, 2026; a study of earlier historical decisions must use the rules operative at the earlier date rather than treating the current consolidated chapter as timeless.
The PRA's SS14/15 with-profits guidance sets out prudential supervisory expectations. The current listed version was published November 15, 2024 and took effect December 31, 2024. The FCA and PRA have complementary institutional roles: fairness and conduct expectations under one framework, prudential resilience and the treatment of funds under another. Neither approach implies that an individual policyholder owns a fixed fraction of every accounting gain when that gain arises.
Hong Kong's regulator maintains information about participating policies, and its official guideline directory lists a revised GL34, Governance and Management of Fund(s) of Participating Business, effective March 31, 2026. Direct automated retrieval of that guideline's new PDF and consumer information page was restricted during this review. The verified finding here is the existence and effective date of the revised governance instrument. No specific paragraph of the inaccessible PDF is represented as examined.
What each source establishes
The regulatory source hierarchy matters because identical terms can refer to different acts. A statutory requirement to apportion divisible surplus is a legal condition. A professional actuarial principle provides the methodology by which classes can be treated fairly. A regulator's illustrated-scale rule addresses representations to a purchaser. Financial accounting determines when and how an insurer measures insurance liabilities and profit. Research on the valuation of guarantees characterizes economic exposures under specified option, interest-rate, and surrender assumptions.
For example, ASOP No. 15 on participating dividends concerns professional determination of dividends and the allocation of divisible surplus; the standard does not require an insurer to distribute every estimated favorable difference in a year. By contrast, NAIC Model Regulation No. 582 concerns the contents and presentation of life insurance illustrations. Its requirements apply through applicable legal authority, subject to model scope and state adoption. Treating one document as proof of obligations governed by another creates false precision.
The original research literature adds a third type of evidence. Grosen and Jørgensen (2000) study participating contract liabilities, guarantees, bonuses, and surrender rights through contingent-claims analysis. Bauer, Kiesel, Kling, and Ruß (2006) use risk-neutral valuation in a German-oriented cliquet-style setting. Those papers establish that dividend and guarantee rules can have economic option values; they do not establish the legal terms of an unnamed Massachusetts or Hong Kong policy.
Participation and the historical allocation problem
Historically diverse participating systems converge around the problem of how to share collective gains over time when claims and market outcomes are uncertain. A participating life policy is issued at one time, its reserves and assets evolve over many years, and some policyholders terminate while others persist to later durations or death. Thus a dividend rule allocates resources both among contemporaneous classes and among successive generations. The same positive investment experience can support distributions, reserves, or retained capacity; the outcome depends on a choice whose beneficiaries are not necessarily present in the same year.
This intertemporal structure explains why divisible surplus is never adequately understood from a one-year cash balance alone. The emergence of favorable experience may have required historical capital provided by another block, while future capital requirements may be borne by continuing policyholders. The source-based research task is to identify the cohort, experience period, valuation date, governing rule and benefit state, then trace the movement of resources across them. Without a traceable distinction between current gain and available distributable resources, a claim of equitable sharing remains unauditable.
Participation Entitlements and Contractual Architecture
The guaranteed core and contingent participation
A participating contract combines several layers of entitlement. The first comprises explicitly guaranteed death benefits, premiums, cash or nonforfeiture values, and any contractual crediting floors. The second comprises eligibility for participation under a dividend or bonus framework. The third comprises specific benefits after a declaration, credit, or election has occurred. The legal quality of each layer depends on the document. An amount entered on a sales illustration as expected future accumulation remains conditional unless the contract itself guarantees that amount.
The NAIC's consumer illustration overview, updated January 8, 2026, distinguishes guaranteed values determined at issue from non-guaranteed values affected by current assumptions. The NAIC illustration model requires guaranteed and non-guaranteed elements to be identified separately where the model applies through operative law. This means a currently payable scale, disciplined illustrative scale, and historical dividends can inform the policyholder without creating identical obligations. An amount already credited under a contract can have a different legal status from a future non-guaranteed projection.
The status of a reversionary bonus or dividend-funded addition needs separate examination. A declared reversionary bonus may be incorporated into a policy's guaranteed death or maturity benefit under its terms. A terminal or final bonus may be evaluated when a qualifying claim occurs and can depend on the fund and event. A cash dividend can be withdrawn, retained, or applied to additional insurance. The names are informative, but the relevant legal question remains whether and when a particular amount has vested, may be reduced, or becomes payable.
Divisible-surplus participation as an eligibility and allocation process
Participation entitlements are not necessarily fixed percentages of every dollar earned by an insurer. The insurer determines an aggregate pool under the applicable law and business framework, then uses actuarial allocation factors to distinguish policies that contributed differently to the pool. The contribution principle in ASOP No. 15 calls for such an approach unless an actuary judges a different basis preferable, reasonable and appropriate. It expressly recognizes annual or extended-period application and practical approximations.
Dividend-factor classes may reflect policy type, mortality and underwriting, issue cohort, guarantees, expense structure, investment experience, duration, loans, taxes, and reinsurance. Credible experience supports meaningful differentiation; sparse classes can require grouping or external data. An insurer that allocates a favorable investment gain entirely in proportion to premium size may be fair in a simplified illustration but unfair where one cohort has a materially stronger credited-rate guarantee or different invested asset share. Allocation therefore requires relevant exposure and liability bases, not simply premium receipts.
A transparent simplification sets a declared pool D and positive dividend factors w for classes c. The hypothetical proportional identity D_c = D × w_c / Σw_j illustrates allocation and ensures that the distributed components sum to D. It is not a general industry dividend formula. Actual methods may involve mortality dividends, expense dividends, interest differences, termination dividends, reserved pools, shareholder transfers, and adjusted contribution periods. The formula has instructional value only when the researcher identifies what each class weight measures and whether it is economically appropriate.
Four uses of a declared policy dividend
A declaration changes available policyholder choices but does not predetermine how the holder realizes value. Depending on the applicable policy, a dividend can be paid in cash, applied to a premium, used to purchase paid-up additional insurance, or left to accumulate with the insurer. The alternatives are expressly visible in Massachusetts's participating-policy statute. Their exact availability, automatic default, and loan and surrender treatment are contract-specific.
For an invented $600 annual declared dividend, cash election supplies $600 of immediate liquidity. Using $600 to offset a $2,400 required premium reduces current out-of-pocket expenditure to $1,800 where permitted. If a hypothetical net single premium for paid-up additional insurance were $30 per $100 of added face amount, the same $600 would purchase $2,000 of additional death benefit. The $30 figure is an arbitrary teaching assumption, not an insurance quotation, and does not incorporate a policy's actual age, fees or underwriting. Retention on deposit leaves the $600 with the insurer under a separate accumulation and withdrawal framework whose future credited interest may be conditional.
These four uses change the timing and character of benefits. Cash raises near-term consumption capacity; premium reduction preserves near-term liquidity without changing a stated premium obligation elsewhere; paid-up additions alter future insurance coverage and possibly its later participating base; accumulation leaves credit and counterparty exposure with the insurer. A comparison that focuses only on the $600 initial nominal dividend would miss the resulting differences in mortality-dependent payoff, future surrender values, tax treatment and reinvestment risk.
Guarantees, vesting, and consumer expectations
The holder's present enforceable claim and expected economic benefit are separate values. A guaranteed paid-up addition can create an enforceable incremental death benefit, while expectations of further future dividends remain sensitive to the insurer's board and experience. A bonus declared only at maturity may have different significance for a holder who surrenders earlier. Policy loans may reduce death and surrender proceeds despite favorable dividend history. These effects require the policy terms, crediting option and event-date values to be read together.
One problem is the mistaken inference that a dividend-financed future premium is no longer due. The New York DFS illustration guidance discusses restrictions on misleading “vanishing premium” representations and on portraying non-guaranteed elements as assured. The 2024 ASOP No. 24 illustration standard supplies contemporary professional guidance for actuarial services supporting illustration certification. If actual dividends later decline, the policyholder may still be required to fund scheduled premiums to preserve guarantees.
A serious contract audit therefore retrieves the original issued terms, endorsements, dividend option election, historical benefit additions, in-force illustration and outstanding loan information. These sources identify which numbers are legal obligations, which are currently declared, and which remain projections. An annual notice showing a dividend for one period is evidence about that period, not independent evidence that the same amount is due in every subsequent year.
Surplus Emergence, Allocation, and Distribution
The economic sources of participation
Surplus emerges from the interaction of priced or reserved expectations and realized outcomes. Mortality experience can become favorable when claims are below an appropriate insured-cohort base, but its financial effect depends on insured amounts, durations and reserve movements. Investment experience depends on asset income, realized and unrealized values, reinvestment and matching of long-duration liabilities. Expenses can be favorable because retention, distribution and administration costs differ from assumptions. Persistency, tax, reinsurance, and policy loan movements can also affect performance. The actuarial dividend factors addressed by ASOP No. 15 are designed to reflect the relevant financial and policy characteristics.
The transition from favorable experience to a dividend pool requires reserves, guarantee obligations, available capital and the governing decision on retention to be reconciled. A strong investment year is insufficient to prove that equally strong dividends can be declared. Existing liabilities may become more expensive because interest rates move, mortality assumptions change, or the cost of options rises. Conversely, a temporarily weaker year can coincide with a deliberately maintained dividend scale supported by prior reserves or smoothing resources. Analysts must state the starting valuation basis and the decision date before identifying distributable gains.
Original illustrative surplus reconciliation
The following worked calculation is constructed solely for this monograph. It represents a hypothetical participating block for one review period. Amounts are in millions of U.S. dollars, and the assumed comparison basis is held fixed throughout the exercise.
| Hypothetical experience or decision | USD millions |
|---|---|
| Investment experience above starting basis | +5.20 |
| Favorable insured mortality experience | +0.80 |
| Excess operating expenses | −0.35 |
| Additional cost of financial guarantees | −1.10 |
| Additional reserve strengthening | −0.60 |
| Attributable taxes and other adjustments | −0.40 |
| Net favorable experience after adjustments | +3.55 |
| Illustrative retained buffer for resilience | −0.95 |
| Illustrative divisible amount | +2.60 |
The arithmetic is 5.20 + 0.80 − 0.35 − 1.10 − 0.60 − 0.40 = 3.55; then 3.55 − 0.95 = 2.60. This is a transparent demonstration of how a favorable result might be separated from an amount chosen for distribution. It is not a statutory balance sheet, a model solvency assessment, an observed insurance company's accounts, or a recommendation that 26.8% of gains always be retained.
If three actuarially supported dividend classes have hypothetical weights 1.30, 0.80 and 0.50, the $2.60 million distribution assigns $1.30 million, $0.80 million and $0.50 million to the respective classes. The amounts sum to the declared pool. However, arithmetic consistency establishes neither actuarial contribution nor legal fairness. To judge those outcomes one must validate the class definition, policy factors, data credibility, guarantee costs and the reasons for retaining the $0.95 million buffer.
The example has an important sensitivity. If a scenario changes the cost of guarantees from $1.10 million to $2.10 million while the other items and retained buffer remain fixed, the illustrative amount available for distribution falls from $2.60 million to $1.60 million. A future higher-rate guarantee, stronger persistency under adverse yields or alternative reserve valuation could therefore change the distribution sharply even while investment income appears favorable. This sensitivity is arithmetic, not an estimated loss distribution.
The contribution principle and class equity
The contribution principle relates dividend allocation to the contribution of participating policy classes to divisible surplus. The professional guidance of ASOP No. 15 allows applications over annual or longer periods and admits reasonable adjustments. The time period matters because experience in a young cohort may differ systematically from later years. A large acquisition expense can be borne early, while subsequent investment returns and mortality outcomes emerge over decades. A simplistic one-year allocation might penalize new business despite defensible long-term contribution estimates, or over-reward mature contracts if past capital support is ignored.
Experience credibility also matters. Small mortality classes may appear exceptional because of random claims. Expense differences may be caused by reclassification. A new guarantee may generate an option cost that is invisible in simple realized income. A fair allocation method requires explicit assumptions about exposures, cohort comparability, expected-versus-actual experience and how uncertainty is shared. When an actuary uses broader experience or a simplified factor, the justification should be documented. This preserves fairness as an evidence-based process rather than an unexplained assertion.
Cross-subsidies can occur within and between cohorts. Some are part of the underlying insurance risk pool, in which uncertainty is deliberately shared; others arise from dividend formula choices, asset allocation, management expenses or retained resources. The evaluator must distinguish contractual risk pooling from a transfer that undermines the promised allocation rules. Equal dividends per dollar of premium, equal returns on estimated asset shares and equal lifetime expected contribution are different criteria. A single numerical fairness test cannot be selected without the contract and governance framework.
Smoothing the timing of benefits
Smoothing allocates investment or other experience over a longer horizon rather than allowing the full annual volatility to flow directly into bonuses. It can protect the usability of long-term savings contracts and reduce policyholder exposure to temporary price fluctuations. It also requires resources, changes the timing of recognition, and can create different outcomes for policyholders who enter, die, surrender or mature at different dates. A valid smoothing evaluation must include these distinct cohorts and the opportunity cost of retaining resources.
Suppose a wholly hypothetical annual dividend budget is $4 million, while adverse experience reduces what the current period alone supports by $1 million. A smoothing rule might use a preexisting permissible buffer to declare an intermediate amount and replenish the buffer in later favorable periods. The economic cost of that decision depends on the liquidity and solvency of the fund and the timing of exits. A smoothing practice that persistently favors maturing policies at the expense of remaining policyholders may be inequitable even when reported bonus rates look stable.
In Grosen and Jørgensen's participating-liability study, bonus policy and surrender rights can materially change modeled contract value. In Ruß and Schelling's client-side smoothing analysis, preferences and long-term return smoothing are evaluated in a defined analytical setting. The research implication is that smoothing must be assessed jointly as a policyholder benefit and as a cost or risk allocation for the insurer. Neither paper demonstrates that every existing smoothing scale maximizes welfare.
Asset shares, inherited estates, and discretion
A with-profits asset share can estimate the accumulation attributable to a policy or cohort after premiums, investment returns, expenses, risk claims and other relevant movements. It may guide bonus-setting and surrender or maturity calculations. Its exact status depends on the insurer's PPFM and the contract; it is not invariably identical to an individual policyholder's legal ownership claim or immediate cash value. Current surrender payments may be affected by smoothing, market-value reductions, terminal bonuses and fund arrangements.
Inherited estate resources create a difficult intergenerational issue. Some funds contain accumulated past gains and capital support whose source and legal treatment are different from one year's divisible result. A proposal to release these assets to current holders needs to account for the interests of future claimants, existing guarantees, shareholder rights and policyholder classes. The current FCA with-profits rules require the firm to address competing interests and explain how it handles distribution and retained resources. That process is more informative than a simple historical dividend-rate comparison.
Governance of declaration decisions
The final declared amount is a business and legal decision informed by actuarial evidence. A robust record should identify the experience period, methods, divisor classes, retained capital rationale, board approval, date of effect, and communication sent to policyholders. The fairness of a dividend is judged against both individual entitlements and the continued ability of the insurer to meet its guarantees. For a stock insurer, shareholder transfers and cost attribution also require review; a mutual similarly needs transparency over management incentives and cross-generation financial support.
The FCA's with-profits committee requirements are an institutional example of scrutiny over discretion, bonus rates, smoothing and competing rights. The general scientific lesson is that actuarial mathematical consistency and effective governance are complementary evidence. An elegant calculation can support an inequitable decision when its classes or starting assumptions are wrong. Conversely, a board's appeal to fairness cannot replace the need to show where the distributed resources originated and how they were assigned.
Guaranteed and Discretionary Benefits Under Stress
The contingent cost of guarantees
A policy may guarantee a minimum rate or accumulation value while granting upside dividends when experience is favorable. Its guarantee becomes more expensive under adverse interest-rate and asset-return paths. The resulting liability is non-linear because the insurer can be required to honor the floor while sharing some favorable performance. An embedded surrender option adds the possibility that a policyholder will exercise a claim at a financially unfavorable time for the insurer. The risk-neutral valuation by Bauer and colleagues (2006) examines such participating contracts under specified German cliquet-style guarantees.
This matters for surplus allocation. An insurer that distributes all current favorable income without funding the cost of guarantees could impair future solvency. In a falling-rate path, the guaranteed liability may become more expensive despite a positive accounting investment yield. In a volatile market, the option to take guaranteed value at surrender can increase losses when the insurer has to liquidate assets. The insurer's capacity to pay a dividend therefore depends on contingent future obligations, not solely realized portfolio income.
The Grosen and Jørgensen valuation decomposes participating contract exposure into bond-like, bonus and surrender features and examines their interaction. The Chiarolla, De Angelis and Stabile (2022) study develops a theoretical optimal-stopping analysis for a specific policy with a minimum guarantee, surrender right, bonus and solvency-related termination. The latter source's exercise boundary results depend on its stated contract and stochastic dynamics. They cannot be assigned without qualification to all whole-life policies, all markets or actual human surrender behavior.
Four adverse scenarios
An investment-rate shock is the first scenario. Bond returns on reinvestment can fall below assumptions while guaranteed credits continue. The present value of some long-duration obligations rises. Management may need to reduce the prospective dividend scale, strengthen liabilities or retain experience rather than distribute it. The shock can be invisible in a short-run dividend history, particularly under smoothing.
A second scenario is market-value loss combined with increased surrender demand. A surrendering holder may choose a contractual value that exceeds the economic liquidation value of assets supporting it. If many holders exercise together, liquidity requirements can reinforce asset sales and the cost of maintaining guarantees. A terminal bonus can be lower than previously illustrated while a contractual surrender guarantee is still binding. The researcher must model all these effects without counting the same asset loss twice.
A third scenario concerns mortality. An unexpected rise in claims can reduce the experience available for distribution. A cohort with large death benefits can have a different financial experience from one with similar premium volume but lower protection. Comparing dividends only against investment yield would erase the core insurance risk transfer. Mortality assumptions should therefore reflect insured lives and actual benefit exposures rather than general population death rates.
A fourth scenario is adverse expenses, lapse and persistent inflation. Higher servicing costs can offset investment experience, while policyholders facing higher living costs may surrender or alter premium payments. The option to reduce premiums using dividends becomes valuable to households yet may be less effective if the insurer reduces its current scale. The ability to maintain guaranteed benefits should be considered under both insurer-wide and household-level shocks.
Illustration discipline and consumer effects
A life policy illustration is a projection at specified assumptions. The NAIC illustration model requires non-guaranteed elements to be identified and warns that results may be more or less favorable than shown. The current NAIC overview distinguishes basic, supplemental and in-force illustrations, along with guaranteed and non-guaranteed benefit representations. The illustrative projection is part of a consumer's decision context, but it is not equivalent to an enforceable future benefit schedule.
Consider an invented year-20 example in which a contract guarantees $80,000 of cash value and shows a projected $114,000 on the current dividend scale. The additional $34,000 is a conditional projection and can change. A projected future reduction in premium outlays financed by dividends is similarly conditional; the nominal contractual premium may continue to be required. The New York illustration guidance targets misleading representations of this kind and has differences from the NAIC model.
An adequate consumer appraisal also explains policy loans, cash available upon surrender, premium timing, and the status of declared paid-up additions. Guaranteed death benefits, projected account accumulation and net surrender payments may have different amounts and sensitivities. A statistically expected dividend can also differ from an individual's realized value because the holder may die, lapse, surrender or select an option before the illustration horizon. Policyholder welfare must be assessed as contingent cash flows, not as the highest figure on one projection column.
Solvency II: discretionary liabilities and surplus funds
The EU prudential framework distinguishes the valuation of future discretionary benefits from the classification of accumulated surplus funds. Article 24 of Delegated Regulation (EU) 2015/35 specifies how the best estimate should reflect asset-dependent discretionary benefits and assumptions consistent with the relevant risk-free interest-rate term structure subject to the stipulated adjustments. Article 25 requires separate determination of future discretionary benefits within technical provisions. That measured amount is not the same as the amount currently declared as a dividend.
Under Solvency II Directive Article 91, accumulated profits not made available for distribution have a surplus-fund classification, and qualifying treatment outside liabilities is subject to national law and capital criteria. This is a particularly important restriction on language. A balance called “surplus” can represent financial capacity or accumulated profits, while a modeled future discretionary benefit represents an expected liability. They cannot be added as if they were independent cash reserves belonging to policyholders.
IFRS 17: a separate accounting question
IFRS 17 Insurance Contracts governs the measurement and presentation of insurance contracts for applicable financial reporting periods beginning on or after January 1, 2023. Some contracts with direct participation features may meet the criteria for the variable-fee approach, depending on contractual identification of underlying items and the stipulated characteristics of the share and variability of benefits. It would be unsound to assert that every traditional participating contract automatically qualifies for the same accounting model.
The IASB's 2017 Effects Analysis explains aspects of the measurement rationale but contains an original projected 2021 effective date that was later superseded. The current IFRS Foundation standard page establishes the applicable 2023 date. Accounting recognition of profit and contract liabilities is separate from both statutory solvency assessment and the insurer's annual board decision on divisible surplus. An author comparing balance sheets and dividends must disclose the standard and valuation date and avoid using reported earnings as an unconditional dividend capacity figure.
Empirical Research and International Comparison
Comparison across four settings
| Setting | Primary participating object | Governing evidence used in this dossier | Evaluation emphasis |
|---|---|---|---|
| U.S. states | Divisible surplus and policy dividend factors | ASOP No. 15, NAIC Model No. 582, Massachusetts § 140, New York § 4231 | Contribution, declaration, legal options, illustration honesty |
| United Kingdom | With-profits fund, PPFM, asset shares and bonuses | FCA COBS 20 and PRA SS14/15 | Discretion, smoothing, intergenerational fairness, disclosure |
| European Union | Future discretionary benefits and surplus funds | Solvency II Articles 24, 25 and 91 | Prudential valuation and eligible own-funds distinction |
| Hong Kong | Participating-business fund governance and consumer benefit information | Insurance Authority directory and consumer materials | Governance and stated non-guaranteed benefit characteristics |
The comparison uses function rather than identical labels as its common denominator. A U.S. annual dividend, a U.K. final bonus, a Solvency II future-discretionary-benefit liability and a participating fund's accounting surplus are distinct objects. Their nominal comparison without contract and valuation adjustments would be invalid. Even within the U.S., different New York statutory provisions and Massachusetts dividend options demand state-specific verification. The FCA's PPFM and governance framework addresses a different legal architecture.
The Hong Kong Insurance Authority's guideline listing confirms the March 31, 2026 GL34 transition. Its participating-policy consumer page and fulfillment-ratio information identify consumer-facing issues but were accessible through indexed official results rather than unrestricted direct automated page readback. Accordingly, the present study neither quotes detailed new GL34 clauses nor asserts the numerical ratios of particular Hong Kong insurers. These access limits are a documented research boundary.
Reconstructing an individual policy path
For an empirical study, an anonymized policy record should include issue date, version of contractual terms, insured age and risk characteristics, policy benefits, premium pattern, chosen dividend option, guaranteed cash values, loans, surrender behavior and death or maturity event. The participating fund record should include relevant asset returns, mortality, expenses, reinsurance, reserve movements, dividend-factor class, board declaration and allocation basis. Without a join between these datasets, a researcher cannot demonstrate how aggregate favorable experience became the value realized by one policyholder.
Time needs to be recorded in four distinct dimensions. First is the period when experience emerged. Second is the period in which the actuary determined factors and the board declared the scale. Third is the point at which the benefit became contractually vested or credited. Fourth is the payment or realization date, which may be years later. A study that merges all four into one reporting year can overstate the relation between current returns and the cash ultimately paid.
The measurement basis also matters. An asset-share figure can use a different allocation and valuation method from financial statement fair value. The insurer's booked investment income can diverge from market total return. Mortality gains should use exposure-weighted insured populations and benefit sizes. A guaranteed addition purchased with a dividend should be recognized once as the dividend's chosen use and again only when later contingent benefit payments actually arise, with appropriate present-value treatment; summing the purchase cost and the full eventual face amount as contemporaneous gains would double count.
Empirical identification strategies and uncertainty
One descriptive question is whether dividend scales respond to actual favorable experience with a documented lag. A panel of participating classes can estimate associations between investment income, mortality gains, expense deviations and next-year dividends. But a coefficient on investment income does not establish causation. The insurer may smooth outcomes, retain capital or change guarantee provisions while contemporaneous economic conditions affect lapses and acquisitions. Causal claims require a credible source of exogenous variation or a structural model that explicitly accounts for management discretion and policyholder selection.
A second question concerns fairness across generations. One could estimate cumulative dividend payments relative to documented class contribution and pre-specified risk and capital adjustments, following issue cohorts through maturity, death or termination. However, there is no neutral single fairness metric. Equal dividends per premium, equal expected lifetime contribution and equal distribution relative to asset share are different norms. The ASOP No. 15 dividend-factor framework encourages a considered allocation and permits practical adjustments. A study must declare its fairness norm, sensitivity to alternatives, and legal scope before reporting a result.
A third question is whether smoothing improves consumer outcomes. Observed low annual dividend volatility may reflect deliberate reserves, a stable underlying asset portfolio, or delayed recognition of bad experience. A proper analysis would compare realized present-value outcomes for equivalent cohorts and assess the liquidity, risk and lost returns of smoothing, including terminating policyholders. The Ruß and Schelling study offers a client-side valuation perspective but should not be interpreted as experimental evidence that every individual prefers the same policy.
A fourth question concerns disclosure. Consumers may misread a current illustrative scale as a promised lifetime benefit. Disclosure experiments could compare interpretations under guaranteed-only and combined guaranteed/non-guaranteed projections, with random assignment of presentation formats. The outcome would be comprehension and downstream decisions, not merely stated preference. Transferability would depend on the governing contract, participant characteristics and distribution channel. No such experiment has been performed in this dossier; the design is a transparent agenda derived from the documentary problem.
Financial valuation research and external validity
The examined valuation literature demonstrates the limitations of a deterministic dividend forecast. Grosen and Jørgensen (2000) show how guarantees, bonus policies and surrender options affect modeled liability values. Bauer and colleagues (2006) develop a risk-neutral analysis in a specific participating-contract environment. Chiarolla, De Angelis and Stabile (2022) analyze optimal surrender for a particular theoretical contract. The models are not substitutes for actual insurer dividend data or the text of an issued policy; they identify the economic dependence of contract value on options and management rules.
A central limitation is the tendency of empirical summaries to select surviving contracts. Policies that lapse or surrender early disappear from in-force samples, often before projected dividends or terminal bonuses are realized. A historical fulfillment ratio for persistent policies can therefore overstate the typical realized long-term outcome of the original issue cohort unless departures and benefits are recorded. Similarly, high past dividends may coincide with favorable interest-rate conditions that cannot be extrapolated into the future. The regulator's non-guaranteed-element disclosures are directly relevant to avoiding this inferential mistake.
The evidence also supports skeptical treatment of three simplistic propositions. Dividend stability does not by itself establish solvency or equitable allocation; stability may be financed by assets that will be needed later. Exact individual contribution cannot always be observed and may require defensible cohort pooling. Retention of favorable experience is not automatically an unfair denial of distribution, because financial guarantees and future contingencies can justify buffers. In every case, a scientific conclusion depends on the contract, fund resources, historical cohorts and management decisions, not a slogan about profit sharing.
Data and evidence required for stronger conclusions
Future work should collect actual contracts and revisions, confidential or anonymized insurer experience triangles, dividend-factor scales, fund-level asset and liability accounts, board decision records and longitudinal policyholder outcomes. A full policy cohort should be followed from issue to death, surrender, maturity or administrative censoring. Each figure needs clear units, valuation basis, reference date, jurisdiction and document version. Validation should address missing data, selection, recoding, rate changes, unknown past options and transferred business.
Comparative datasets require careful harmonization. An American dividend used to buy paid-up additions is a different realization mechanism from a British terminal bonus. A Solvency II future-discretionary-benefit liability is a regulatory valuation, not the sum of cash currently declared. An IFRS 17 measurement is a financial reporting amount, not an individual dividend allocation. Using one common category called “policyholder profit” without a traceable benefit map would generate apparently precise but invalid statistics.
Integrated Scientific Synthesis
The full mechanism from experience to benefit
The verified sources support a coherent institutional sequence. Premiums and assets finance promised protection and savings elements. Claims, expenses, investment results and policyholder behavior generate favorable or adverse experience relative to specified assumptions. The insurer's legal obligations, reserves and capital policy determine which financial resources may support distribution. The approved dividend framework allocates the declared amount among classes and contracts. Declaration and dividend elections then change the individual policy's benefit state. Death, survival, surrender and later crediting determine when and whether those values are realized.
This sequence explains several otherwise confusing observations. Favorable asset returns need not raise dividends proportionately because guarantees or reserve requirements can absorb the gain. More aggregate divisible surplus need not generate equal percentage increases for every cohort because dividend factors reflect different contributions. A higher declared dividend does not necessarily increase cash paid today when the holder elects added insurance. A high projected future dividend does not create a current guaranteed claim. These are explanatory deductions from the documented contract and allocation architecture; they constitute an original analytical synthesis, not a newly named theory or proprietary system.
Equity across policyholders and time
Participating life insurance pools mortality and financial uncertainty over decades. The distribution system therefore has an intergenerational dimension. Contracts issued at different times can carry different guarantees, expense loads, investment exposures and pricing assumptions. Policies surrendered early can receive different value from policies retained to maturity or death. Current dividends can consume assets that otherwise support future guarantees, while excessive retention can defer benefits beyond the reasonable life of a cohort. Fair allocation must combine contribution and a documented view of the timing and beneficiaries of financial obligations.
The ASOP No. 15 contribution principle and the FCA's governance of competing interests address different sides of this problem. One establishes actuarial discipline around attribution; the other requires an institution to address fairness across policyholder classes and shareholders. The synthesis is that a reasonable distribution policy should be testable both quantitatively, through credible class experience and actuarial factors, and institutionally, through a transparent decision record explaining exceptions, retention and smoothing.
Reconciling legal, economic and reporting categories
A policy guarantee is a contractual obligation; a projected dividend scale is conditional; a declared addition may become vested under the policy; a future discretionary benefit can be included in prudential technical provisions; accumulated undistributed profits may have surplus-fund treatment subject to national law; IFRS 17 financial reporting measures an insurer's contractual obligations and services under its own rules. These values may derive from overlapping assets and profits, so additive comparisons can be economically wrong.
The Solvency II surplus-funds provision and separate valuation of future discretionary benefits provide a particularly clear example of distinct categories. The current IFRS 17 reporting framework adds another perspective. Analysts should state precisely whether a figure is a projection, a regulatory liability estimate, a legally distributable amount, an actual declared benefit or a realized cash flow.
Original contribution and unresolved empirical questions
The author's original contribution is the integrated source-grounded reconstruction of participating benefits across contract rights, actuarial allocation, management discretion, capital support and household outcomes. The two hypothetical numerical examples expose differences between a financial gain, a justified distribution pool and the alternative economic uses of one declared dividend. The cross-jurisdiction synthesis shows why superficially similar terms require distinct legal and valuation interpretations. The critical literature review clarifies which results follow from published contingent-claims theory and which would need additional empirical evidence.
The most valuable future tests concern actual class contributions relative to paid dividends; the distributional effect of smoothing across entering, surrendering and persisting generations; the interaction of guarantees, insurer capital and management discretion during stress; and the ability of consumers to distinguish contractually guaranteed benefits from future non-guaranteed illustrations. These tests demand original insurer or policyholder data that have not been collected here. The documentary conclusions and specified numerical identities remain valid within their stated assumptions, while causal generalizations about observed markets remain open.
Conclusion
Participating life insurance creates a governed bridge between the collective financial experience of a life insurer and the individual protection and savings outcomes of its policyholders. The bridge is constructed through legal participation rights, actuarial identification of divisible surplus, class-based allocation, a documented declaration decision and a policyholder's choice of how declared value is applied. Economic experience and realized policyholder benefit therefore differ in amount, timing and legal certainty.
The American sources establish contribution-oriented dividend determination and state-specific policy options and disclosures. The British rules make with-profits governance, smoothing and conflicting policyholder interests explicit. European prudential provisions separate valued future discretionary benefits from surplus funds, and IFRS 17 provides a distinct reporting perspective. Scholarly valuation research demonstrates the economic importance of interest-rate guarantees, bonus rules and surrender options. Hong Kong's documented 2026 GL34 revision establishes a contemporary supervisory point of comparison, with direct-source access limitations expressly preserved.
Fairness requires credible class experience, explicit time horizons, reserves adequate for guarantees, transparent management of retained capital and communication that accurately distinguishes guaranteed, declared and projected amounts. The illustrative $3.55 million favorable experience, $0.95 million retention and $2.60 million distribution reconciliation demonstrates that a positive insurer result is only one stage of the dividend process. The hypothetical $600 policy dividend demonstrates that cash, premium reduction, paid-up additions and accumulation produce different household outcomes. Scientific analysis is complete at the level of its documentary and conceptual evidence; insurer-specific empirical claims require datasets and contracts beyond this investigation.
References
Actuarial Standards Board (2011 revision). Actuarial Standard of Practice No. 15: Dividends for Individual Participating Life Insurance, Annuities, and Disability Insurance. Revised May 2011; originally effective August 1, 2006.
Actuarial Standards Board (2024). Actuarial Standard of Practice No. 24: NAIC Life Insurance Illustrations Model Regulation. Revised September 2024; effective December 1, 2024.
Bauer, Daniel, Rüdiger Kiesel, Alexander Kling, and Jochen Ruß (2006). Risk-neutral valuation of participating life insurance contracts. Insurance: Mathematics and Economics 39(2):171–183. doi:10.1016/j.insmatheco.2006.02.003.
Chiarolla, Maria B., Tiziano De Angelis, and Gabriele Stabile (2022). An analytical study of participating policies with minimum rate guarantee and surrender option. Finance and Stochastics 26:173–216. doi:10.1007/s00780-022-00471-0.
Commonwealth of Massachusetts (current codification). Massachusetts General Laws, Chapter 175, Section 140. Participating life and endowment policy surplus and dividend provisions.
European Union (2009). Directive 2009/138/EC, Article 91: Surplus Funds. Solvency II Directive; EIOPA Single Rulebook.
European Union (2015). Commission Delegated Regulation (EU) 2015/35, Article 24: Future Discretionary Benefits. EIOPA Solvency II Single Rulebook.
European Union (2015). Commission Delegated Regulation (EU) 2015/35, Article 25: Separate Calculation of Future Discretionary Benefits. EIOPA Solvency II Single Rulebook.
Financial Conduct Authority (2026 consolidated). COBS 20: With-profits. FCA Handbook; consolidated content last updated June 26, 2026.
Grosen, Anders, and Peter Løchte Jørgensen (2000). Fair valuation of life insurance liabilities: The impact of interest rate guarantees, surrender options, and bonus policies. Insurance: Mathematics and Economics 26(1):37–57. doi:10.1016/S0167-6687(99)00041-4.
Hong Kong Insurance Authority (n.d.). Understanding a Participating Policy. Official consumer information; direct automated page retrieval restricted.
Hong Kong Insurance Authority (n.d.). Fulfillment Ratio. Official consumer information; direct automated page retrieval restricted.
Hong Kong Insurance Authority (2026). Guidelines: GL34, Governance and Management of Fund(s) of Participating Business. Official guideline directory; new GL34 effective March 31, 2026. Full guideline PDF access restricted.
International Accounting Standards Board (2017). IFRS 17 Insurance Contracts: Effects Analysis. IFRS Foundation, May 2017; historical explanatory report, not operative text.
International Accounting Standards Board (current standard). IFRS 17: Insurance Contracts. IFRS Foundation, effective for annual periods beginning on or after January 1, 2023.
National Association of Insurance Commissioners (2001). Life Insurance Illustrations Model Regulation (Model No. 582). NAIC Model Laws, Regulations, Guidelines and Other Resources, April 2001.
National Association of Insurance Commissioners (2026). Life Insurance Illustrations. Official background page, last updated January 8, 2026.
New York State Department of Financial Services (2008). Guidance for Life Insurance Policy Illustrations. Official New York Regulation 74 interpretation and NAIC comparison.
New York State Legislature (current codification). New York Insurance Law § 4231: Participation in Surplus. Official New York statutory text.
Prudential Regulation Authority (2024). Supervisory Statement SS14/15: With-profits. Current version published November 15, 2024; effective December 31, 2024.
Ruß, Jochen, and Stefan Schelling (2021). Return smoothing in life insurance from a client perspective. Insurance: Mathematics and Economics 101(A):91–106. doi:10.1016/j.insmatheco.2021.03.012.