Insurance by BogdanovaThe Encyclopedia of Insurance and Risk Science

RESEARCH DOSSIER

Insurance Accounting and Performance Measurement

Abstract. Insurance accounting translates uncertain future obligations and the provision of protection into financial statements. The interpretation of those statements requires careful separation of contractual cash flows, current estimates of obligations, compensation for nonfinancial uncertainty, deferred unearned profit, insurance service, financial investment effects, and capital resources. This monograph examines the conceptual structure and practical consequences of insurance accounting under IFRS 17, the interaction with IFRS 9, contrasts with U.S. GAAP and U.S. statutory accounting, and the relationship between accounting profits, underwriting ratios, solvency measures, and economic performance. Its method combines primary-standard analysis, explicit accounting reconciliations, controlled numerical illustrations, comparative institutional examination, and critical appraisal of published implementation evidence. The central conclusion is that insurer performance is interpretable through a reconciled set of measurements with different purposes, bases, reporting periods, and exposure denominators. The contractual service margin identifies a stock of unearned future service profit rather than immediately distributable earnings; claims estimates and risk adjustment represent different elements of fulfillment; and insurance service result, investment results, cash generation, and regulatory capital provide complementary and sometimes divergent signals. Observed international differences therefore require measurement-basis, business-mix, transition, and reporting-date controls before substantive comparisons.

Keywords: Insurance accounting; IFRS 17; IFRS 9; insurance contracts; insurance revenue; contractual service margin; fulfillment cash flows; risk adjustment; premium allocation approach; reinsurance; underwriting; combined ratio; earnings quality; US GAAP; statutory accounting; solvency; performance measurement; insurance financial reporting.

Introduction

Insurance promises are exchanged in the present while claims, costs, and services are distributed across contingent future states. A premium receipt provides cash today, yet the insurer remains obligated to stand ready, settle claims, and administer coverage. Financial accounting must distinguish money already collected from protection already supplied, and management forecasts from obligations measured at the reporting date. Insurers also invest cash flows on behalf of an intertemporal business, hold capital against adverse experience, and often buy reinsurance. The annual accounting result is consequently a structured summary of several linked economic processes.

The International Accounting Standards Board's IFRS 17 Insurance Contracts established a coherent approach to recognition, measurement, presentation, and disclosure. It applies for annual reporting periods beginning on or after January 1, 2023, subject to jurisdictional adoption. The framework replaces the broad diversity of practice previously tolerated by IFRS 4, updates future cash-flow estimates, and recognizes expected profit as insurance services are provided. Its separation of insurance service result and insurance finance income or expenses improves the visibility of business functions while leaving extensive analytical work to users.

The evidence about actual reporting is more qualified than any promise of immediate worldwide comparability. In their review of 24 major European insurers' 2023 financial statements, ter Hoeven, Borelli, and van Vlijmen (2024) documented significant transition effects, heterogeneity in chosen options, and limitations in reconciliation and sensitivity disclosures; their first-year reporting study therefore provides a necessary empirical counterweight to the normative objectives of the standard. Comparison with the United States adds further complexity because general-purpose U.S. GAAP, insurer statutory returns, and risk-based capital each serve partly different reporting objectives.

This study answers three questions. How do financial statements measure insurance promises at different points in their contractual life? What makes insurance service performance different from reported profit or cash receipts? Which adjustments are needed before ratios from distinct insurers, years, or jurisdictions become economically meaningful? Its geographical scope is international, with the IFRS and U.S. systems examined as consequential examples rather than a universal national-law survey. The research date is November 14, 2026; dynamic regulatory and market figures are explicitly dated to their reporting period.

Insurance Accounting Objects and Research Method

Contract, service, obligation, and financial position

A useful starting point is to identify the economic object before choosing the accounting measure. The contract defines enforceable mutual rights and obligations. The insurance coverage provides a continuing service of standing ready to compensate for an insured event. Claims already incurred create settlement obligations, whereas future coverage creates a remaining-service obligation. Cash flows include premiums, benefit and claims payments, directly attributable expenses, and acquisition costs. Financial statements represent recognized assets and liabilities under a specified measurement framework, not the entire economic value of an insurer's customer relationships or business franchise.

The IFRS 17 key terms distinguish the liability for remaining coverage from the liability for incurred claims. This division is foundational for interpreting earnings: the first concerns future protection and associated contractual profit, the second concerns claims arising from past service. In the standard's general measurement model, cash-flow estimates are combined with a nonfinancial risk adjustment and the contractual service margin. For incurred claims, the accounting emphasizes settlement measurement and changes in estimates rather than recognition of a new contractual margin on already supplied coverage.

Five further accounting objects must remain distinct: insurance revenue as consideration attributed to services; insurance service expenses as incurred service costs; insurance finance income or expenses as effects of financial variables and time value on insurance contracts; investment results on financial assets; and equity as the residual recognized interest in a balance sheet prepared under particular rules. Regulatory own funds and statutory surplus are additional measurement objects governed by different purposes.

Method and inferential discipline

The research method has four layers. First, primary text analysis extracts recognition and measurement requirements from the IASB, FASB, NAIC, IAIS, and European supervisory materials, identifying the exact document and relevant scope. Second, stock–flow reconciliations follow opening carrying amounts through cash movements, assumption changes, service release, and closing carrying amounts. Third, controlled illustrations hold specified claims, premiums, and discount assumptions constant to isolate changes in reported result attributable to timing and classification. Fourth, the analysis evaluates published cross-company research and recent market aggregates according to their samples and denominators.

Accounting identities are treated as identities. Numerical examples are constructions, expressly distinguished from empirical estimates. Cross-jurisdictional comparisons are interpretive and conditional on reporting basis. Market evidence is not extrapolated beyond its source population. The distinction between the 2026 OECD market survey, which aggregates national supervisory submissions for 2025, and entity financial statements under a particular GAAP is maintained throughout.

Performance has several measurement horizons

Financial performance can mean whether this year's insurance service generated margin; whether historical policy issuance will produce future service profit; whether investment holdings earned acceptable returns; whether claim reserves are adequate; or whether capital can withstand adverse shocks. These questions employ different horizons. Premiums written refer to contractual volume, premiums earned or insurance revenue refer to service periods under their respective definitions, investment return concerns asset holdings, and solvency concerns the ability to absorb risk. A single bottom-line profit figure combines these elements without identifying all their causal origins.

Recognition, Boundaries, Aggregation, and Contract Components

Identifying insurance contracts

IFRS 17 applies when an issuer accepts significant insurance risk from a policyholder by agreeing to compensate the policyholder if an uncertain insured event adversely affects that policyholder. Financial risk and insurance risk must be distinguished according to the standard's definitions. The same broad commercial product may contain both protection and financial intermediation. The exact IFRS accounting scope turns on contractual substance, including designated investment contracts with discretionary participation features issued by entities that also issue insurance contracts.

The identification of significant insurance risk precedes revenue measurement. Payments that are economically deposits or investment components cannot automatically be counted as sales of insurance protection. IFRS 17 requires certain embedded derivatives, distinct investment components, and distinct noninsurance service obligations to be separated and accounted for under the applicable other IFRS standards. The standard's separation rules connect the perimeter of an insurance contract to later choices about what counts as revenue.

Unit of account, groups, and annual cohorts

Insurance risk is frequently managed in large portfolios, but recognition is not simply a total of all policies sold by a company. IFRS 17 first identifies portfolios of contracts with similar risks that are managed together. Groups are then separated by expected profitability, including onerous contracts and contracts with no significant possibility of subsequently becoming onerous, and ordinarily contain contracts issued no more than a year apart. This discipline limits the transfer of profit across contracts written in different periods or with materially different profitability.

Aggregation determines when losses become visible and where margin is released. A policy group expected to generate losses cannot ordinarily conceal those losses within a larger pool of profitable, differently grouped contracts. Nevertheless, the measurement is still at group level: it does not create a continuously observable profit figure for each individual policy. The historical data system must preserve issue dates, original group assignment, acquisition cash flows, assumption versions, and the bridge between claims systems and general ledger.

Contract boundaries and recognition dates

The contract boundary determines which expected future premiums and claims belong to a recognized group. Under IFRS 17 this depends on the substantive rights and obligations and the insurer's practical ability to reassess risks and reprice or terminate coverage within the conditions specified by the standard. Projection should not automatically include all potential future renewals, nor may it omit existing enforceable future obligations because management expects renegotiation.

Recognition timing also matters. Under the general model, groups are recognized at the earliest of the beginning of their coverage period, the date the first payment from a policyholder is due, or when facts and circumstances indicate the group is onerous; special rules apply for reinsurance held. The reporting period therefore contains not only cash receipts but new groups, claims incurred, adjustments for future service, financial effects, and service supplied by in-force groups.

Fulfillment Cash Flows and Risk Adjustment

Present value as a reporting-date measure

The general measurement model begins with the present value of expected future cash flows, including probability weighting and discounting. In a compact illustrative notation with net outflows positive,

PVFCF = Σ[t=1..T] E(Outflows_t − Inflows_t) × D_t,

where D_t is the relevant discount factor and the expectations incorporate assumptions current at the reporting date. Cash flows within the contract boundary are evaluated consistently with the characteristics of the insurance liabilities. The actual model is richer than the displayed single factor: currency, liquidity characteristics, stochastic dependence, financial options, and the relevant discount-rate methodology can matter. The IASB measurement explanation explicitly places probability-weighted discounted flows at the center of the liability calculation.

For a two-year illustrative obligation to pay 100 units after one year and 110 units after two years, with a flat illustrative annual discount rate of 4%, the present value is 100/1.04 + 110/1.04² = 197.855 units (rounded to three decimals). At 5%, holding the payments fixed, it is 100/1.05 + 110/1.05² = 195.011. The approximately 2.844-unit decrease arises from the changed discount assumption alone. Such a shift can change reported insurance finance results even while underwriting operations and claim-payment forecasts remain unchanged. The illustration deliberately excludes yield-curve shape, reinvestment, credit, and option effects.

Risk adjustment for nonfinancial risk

The risk adjustment for nonfinancial risk reflects the compensation the reporting entity requires for bearing the uncertainty about the amount and timing of cash flows arising from nonfinancial risk. It is an explicit adjustment in IFRS 17 and relates to the characteristics of the risk borne. It is not an additional expected claim and not the same thing as a prudential risk margin prescribed under Solvency II. The IFRS terminology guidance defines this risk adjustment in terms of the compensation required for bearing insurance risk.

Suppose a simplified group has an expected net present outflow of 900 and a risk adjustment of 60. The fulfillment cash flows measured as a liability are 960 units. The 60 units do not indicate a further expected indemnity payment. They describe compensation for uncertainty in excess of expected values. The model must therefore document the separation of best estimates and uncertainty compensation, the sensitivity to volatility or tail assumptions, and how diversification is reflected at the level permitted by the accounting approach. IFRS 17 does not prescribe a single universal risk-adjustment technique, which can yield comparability differences even where disclosures comply formally.

A change in the risk adjustment associated with future service generally participates in the adjustment of the contractual service margin in the general measurement model; release attributable to risk expired in the current period affects the insurance service result. Both timing and classification must be recorded. A favorable year of claims is not equivalent to the entire release of a risk adjustment: favorable claims experience, revised future uncertainty, and the passage of the service period are separate events.

Estimation, data, and auditability

Insurance cash flows rely on claims histories, morbidity and mortality, lapse behavior, policyholder options, inflation assumptions, expenses, and statistical tail models. Long-duration contracts amplify sensitivity to discounting and behavior. A measurement is trustworthy when assumptions, data lineage, controls, and change explanations can be traced from actuarial projections to the accounts, especially when estimates are revised after an event. The research logic here is consistent with the IAIS supervisory valuation guidance, which calls for reliable, decision-useful, and transparent valuation while recognizing that expert judgment cannot simply be eliminated.

Contractual Service Margin and Profit Recognition

The meaning of unearned profit

The contractual service margin, or CSM, represents the unearned profit component of the carrying amount of a group of insurance contracts. Its financial interpretation is a deferred service margin attached to identified insurance coverage, subject to the requirements for future-service adjustments and release. Under an initially profitable group, measurement is arranged to avoid recognizing a day-one gain solely from undertaking future insurance service obligations. The IASB definition of CSM identifies it as unearned profit recognized as services are provided.

In an intentionally simplified inception example, expected premiums to be received have a present value of 1,300 units, expected claims and attributable expenses have present value of 1,090, and nonfinancial risk adjustment equals 70. Thus the fulfillment cash flows are 1,090 − 1,300 + 70 = −140; the positive CSM of 140 offsets the favorable net fulfillment position, producing a zero net group carrying amount at inception where no other initial cash flows or recognized acquisition assets enter the calculation. Negative fulfillment flows in this sign convention represent a net inflow. This is an algebraic illustration of the mechanics of deferring profit, not a universal requirement that every recognized group has zero liability or asset.

Subsequent measurement and coverage units

After initial recognition the CSM is adjusted for relevant new contracts added to the group, interest accretion where applicable, changes in estimates that relate to future service, currency effects, and the amount recognized in profit or loss as coverage is provided. The allocation uses coverage units expressing the quantity of benefits provided and expected duration. The result is a pattern of service recognition reflecting the services actually provided, rather than merely the dates of premium collection.

For an illustrative group with a 140-unit CSM that has no interest accretion, estimate revisions, new contracts, or currency changes, suppose 40% of the total remaining weighted coverage units are supplied in the current period. A 56-unit CSM release enters insurance revenue through the service accounting mechanism, leaving 84 units for later periods. If instead current coverage is 25% of total remaining units, release is 35 units. Both arithmetic examples concern the same group with different hypothetical service patterns; they show why duration and benefit weighting matter to annual profits.

Future-service changes and the boundary of loss absorption

Favorable or adverse changes in projected future service can adjust the CSM subject to the standard's requirements. A large adverse revision can exhaust the existing CSM. The remaining amount, if it makes the group onerous, is then recognized as a loss rather than carried indefinitely as negative expected profit. Consequently the CSM has an asymmetric role in timing: it absorbs eligible changes within a remaining positive margin, while incurred losses and other specified changes may affect current results immediately. The CSM is not simply a reserve of discretionary capital and its balance cannot be read as a forecast of future dividends or aggregate corporate value.

The official IFRS 17 reinsurance guide sets out comparable logic for margins on reinsurance held, with important differences in the initial recognition and treatment of net gains or costs. The exact issue date and coverage of the reinsurance contract are relevant. Reinsurance cannot be measured by transferring the insurer's original group CSM mechanically to its reinsurer.

Premium Allocation, Onerous Contracts, and Reinsurance

Short-duration coverage and the premium allocation approach

The premium allocation approach (PAA) is a permitted simplification for qualifying groups. At inception it is available when an entity reasonably expects a measurement of the liability for remaining coverage not materially different from the general model, or when each contract's coverage period, including services from premiums within its contract boundary, is one year or less. The condition in IFRS 17 paragraph 53 is an eligibility test, not a rule that every short-duration contract must use PAA. The simplification changes the measurement mechanics for remaining coverage; incurred claims are still recognized and measured under the applicable claims requirements, including discounting and risk adjustment where required.

A simple premium-allocation intuition is that a 1,200-unit amount allocated to twelve months of uniformly supplied coverage produces 100 units of revenue each month when the applicable conditions and assumptions support that pattern. Actual IFRS 17 insurance revenue recognition depends on service and the permitted treatment of acquisition cash flows, financing components, and expected pattern of incurred insurance service expenses. Premiums collected or written, cash-flow timing, and insurance revenue are separate ledger items. The PAA does not justify measuring complex multiyear or materially nonuniform risk coverage as a straight-line subscription without checking the standard's criteria.

Loss components and timely recognition of onerous contracts

The distinction between a profitable group and an onerous group is fundamental to profit reporting. When a group's fulfillment obligations exceed the benefits expected from remaining coverage, or the group otherwise meets the standard's onerous test, a loss is recognized immediately under the applicable accounting mechanics. The liability for remaining coverage is tracked with a loss component to allocate future changes appropriately. The accounting effect is to separate the expected loss from the recognition of earned profit on profitable service periods.

Consider a stylized group with remaining expected cash inflows worth 500, claims and directly attributable costs worth 550, and risk adjustment of 30. The fulfillment cash flows represent an 80-unit net obligation. Assuming no offsetting margin or other measurement changes, the 80-unit expected loss is a current-period loss rather than a 20-unit loss spread over four years. Subsequent changes in the loss component must be allocated under IFRS 17's detailed rules. This example demonstrates the direction of recognition; it does not reproduce the full set of loss-component movements or discounting options.

Ceded protection as a separately measured contract

Reinsurance changes the issuer's net risk position, but the original obligation to policyholders remains a gross obligation unless another legal/accounting event changes it. IFRS 17 separately recognizes and measures reinsurance contracts held, usually as a distinct group with its own recognition date, cash flows, risk adjustment reflecting risk transferred, and margin or net cost. The IASB guide to reinsurance held explains its differences from underlying issued insurance groups, including adjustments related to recovery of losses on onerous underlying business when specified conditions apply.

Two reporting perspectives have to be reconciled. The insurer's gross insurance service result reflects protection issued. The effects of reinsurance held reflect risk ceded and the price paid for that protection. A combined ratio or margin calculated net of reinsurance can change through a cession strategy even when the gross incidence of claims is unchanged. Thus changes in net underwriting ratios should be decomposed into gross claims experience, reinsurance cost, recoveries, commission effects, and retention.

Contractual options and participating arrangements

Insurance products may contain surrender options, guaranteed returns, investment-linked features, and discretionary or direct participation. These features alter the projected cash flows and the treatment of financial-variable changes. Under IFRS 17, some contracts with direct participation features are accounted for using the variable fee approach, in which elements of the insurer's share of underlying items affect the CSM under specified conditions. A general discussion that treats all long-duration insurance as a fixed-fee obligation misses this variation. The method applied, its eligibility, and the financial-variable effects should be disclosed in a form that allows investors to interpret the reporting result.

Insurance Revenue, Expenses, and Financial Results

Revenue as compensation for service

Under IFRS 17, insurance revenue reflects the consideration for insurance contract services in the reporting period and excludes investment components. It is not equivalent to total billed premiums, written premiums, or cash received. Where a contract contains material amounts unconditionally returnable to policyholders, counting those amounts as revenue would gross up commercial turnover without representing service supplied. The distinction is explicit in the IFRS 17 presentation requirements.

Insurance revenue under the general model includes amounts corresponding to expected insurance service expenses released from the liability for remaining coverage, release of the risk adjustment related to current service, and the portion of CSM recognized for service, subject to the standard's prescribed mechanics. Insurance acquisition cash-flow recovery is addressed according to the relevant allocation rules. The components are expressed here for economic interpretation; an audit-ready revenue reconciliation follows the exact IFRS disclosure tables rather than a free-form reconstruction.

A 300-unit cash premium may fund three years of protection, with some amount designated as an investment component. Cash collection affects cash and the contract carrying amount. The portion that represents investment components does not become insurance revenue; the protection component is recognized over the service pattern. Consequently the ratio of revenue to premiums received may differ sharply across a savings-oriented life insurer and a property insurer even when both are performing as contracted.

Service expenses and the release of incurred claims

Insurance service expenses generally include incurred claims and other incurred insurance service expenses, changes relating to past service in the liability for incurred claims, and specified losses and reversals for onerous groups. They exclude repayments of investment components. Presentation distinguishes this from insurance finance income or expenses, enabling users to identify the insurance-service portion of results.

This distinction does not remove actuarial estimation from earnings. A favorable development on previous accident years may improve a current-period service result. Such an improvement records a revised estimate of the cost of previously supplied service rather than performance from newly written policies. Analysts should therefore separate current accident-year operating results, prior-year development, current investment market effects, and release of contractual margins when data allow.

Financial results and other comprehensive income

Time value of money and financial-risk changes influence the carrying value of insurance obligations. IFRS 17 classifies these as insurance finance income or expenses, distinct from the insurance service result. Depending on the applicable measurement model and elected accounting policy, some insurance finance effects may be disaggregated between profit or loss and other comprehensive income (OCI). The corresponding asset-side presentation is driven separately by financial instrument classification and measurement. The result is that two insurers with comparable economic asset–liability exposures can report different immediate profit or loss depending on elections, portfolio characteristics, and asset accounting.

The IASB's IFRS 17 effects analysis describes the expected changes in statements of financial performance and key financial metrics while recognizing product-dependent implementation effects. For performance analysis, a gain recorded in OCI must be considered when evaluating total economic value changes, even though OCI is distinguished from profit or loss under the reporting standards. The relevant comprehensive-income reconciliation and changes in equity are necessary to avoid treating a presentation choice as a change in real claims cost.

Earnings quality and the meaning of recurring profit

A recurring profit measure requires documented exclusions. Removing large catastrophic losses may produce a more stable indicator of attritional underwriting, but catastrophe exposure is part of the insured risk that policyholders pay the company to bear. Removing reserve strengthening may clarify short-term operational performance, yet adequate reserving is integral to responsible insurance operations. Similarly, CSM release under a mature book and positive investment carry can be sustainable under some conditions but still depend on future business, lapse behavior, asset returns, and capital requirements.

An adjusted operating profit should therefore be reconciled to audited statutory or general-purpose figures and accompanied by stable definitions across time. There is no single universal adjusted earnings metric that supersedes the primary accounting statements.

Assets, Investments, and Asset–Liability Effects

Financial assets and contractual liability matching

Insurance financing is a joint asset–liability problem. Premiums are invested in bonds, equities, loans, real estate, and other instruments, while expected claim payments evolve with underwriting exposures, interest rates, inflation, and contractual options. Insurers manage duration mismatch, credit risk, liquidity risk, and reinvestment opportunities. A measure of service profitability that excludes investment return tells only part of the insurer's economic performance; a measure of investment yield that ignores the liabilities those assets fund is similarly incomplete.

The IFRS 9 classification framework categorizes financial assets by business model and contractual cash-flow characteristics and prescribes measurement and impairment treatment. The interaction of IFRS 9 with IFRS 17 matters because changes in asset prices and insurance liabilities may be recognized in different places or periods. Asset and liability changes may economically hedge one another while their separately required presentation creates temporary accounting volatility.

A complete analysis starts with investment income and realized/unrealized gains, then considers credit deterioration, expected credit losses as applicable, cash-flow matching, insurance finance income/expenses, OCI movements, and effects on capital. Claims inflation and interest rates can simultaneously change assets, expected indemnities, expense projections, and solvency requirements. No single investment return percentage captures all these consequences.

Interest rates and the insurer's economic duration

Suppose an insurer holds 1,000 units of fixed-income investments with a modified duration of five years and has discounted insurance obligations of 900 units with a duration of seven years. A small parallel rise of one percentage point produces approximate first-order fair-value movements of −0.05 × 1,000 = −50 in the investments and −0.07 × 900 = −63 in the obligations. A decline of 63 in the liability is economically favorable to the insurer, while the 50 decline in assets is unfavorable, yielding an illustrative positive 13-unit first-order net-value change before convexity, cash flows, spread changes, guarantees, and taxes. This approximation uses the familiar modified-duration sensitivity, assumes matching sign conventions and comparable valuation, and says nothing by itself about reported profit or loss because asset/liability classifications and OCI treatments must be examined.

Duration matching reduces certain rate exposures but can leave cash-flow timing, lapse, credit, liquidity, and basis risks. Further, an insurance obligation's duration may change endogenously when policyholders alter surrender behavior or when participation features respond to market conditions. Therefore a year of unusually high investment income can coexist with deteriorating balance-sheet resilience.

Reinvestment and investment income persistence

Current yield, accounting investment income, and expected total return differ. A bond purchased several years ago may generate a contractual coupon that does not track current market yields. Realized gains may bring forward value that would otherwise emerge later. Higher reinvestment yields can improve future earnings even as a portfolio's fair value falls today; lower rates can produce the reverse. In longer-duration business, the simultaneous effects on liability cash flows, discount curves, and options are essential to any performance interpretation.

Underwriting and Investment Performance Indicators

The combined ratio and its denominator

For many nonlife insurance markets, the combined ratio is an accessible measure of underwriting profitability. A conventional net earned-premium formulation is

Combined ratio = (Net incurred claims + loss adjustment expenses + underwriting expenses) / Net earned premiums × 100%.

The loss ratio measures relevant claims and loss-adjustment costs against earned premiums; the expense ratio captures the specified underwriting expense measure against its denominator. Not all statistical systems use precisely the same expense denominator or treatment of acquisition expenses, and different published ratios may include dividends to policyholders or other specific components. The NAIC insurance glossary identifies the combined ratio through its loss and expense components. The analyst must obtain the actual calculation definition before comparing ratios.

The OECD's 2025 combined-ratio comparison reports values below 100% in 34 of 39 jurisdictions with available data and provides a documented warning against interpreting the market as homogeneous. Its data refer to national reporting arrangements and product mixes. It cannot be mechanically equated with an IFRS 17 insurance service margin calculated from a multinational company's consolidated income statement.

For a purely hypothetical insurer with net earned premiums of 500 units, incurred claims and loss adjustment expenses of 310, and underwriting expenses of 145, the combined ratio is (310 + 145)/500 = 91%. The corresponding underwriting surplus is 45 units before investment, tax, other expenses, and any reconciliations between the chosen ratio convention and GAAP reporting. This figure cannot establish whether reserves are adequate: if previously unreported expected claims of 40 are recognized, the revised ratio is (350 + 145)/500 = 99%, all else equal.

Loss ratios, development, and denominator effects

An improved loss ratio can arise from lower claim frequency, smaller claims, stronger pricing, better risk selection, favorable reserve development, or changes in the mix and timing of exposures. It may also result from increased earned premiums when claims have yet to emerge or be fully reported. Catastrophe claims, long-tail liability development, social inflation, claim settlement costs, and changes in reinsurance retentions require explicit assessment.

Expense ratios likewise depend on reporting practice and scale. Acquisition commissions, deferred costs, premium taxes, technology investment, policy administration, and catastrophe-management spending may be allocated differently. A falling expense ratio can reflect genuine efficiency, growth in the denominator, or timing changes. Where acquisition costs are expensed earlier or deferred under another reporting basis, the apparent efficiency ranking may reverse. Interpretations should be supported by cash expense trends and stable definitions of current and prior periods.

Investment return, return on equity, and capital efficiency

Investment return can be expressed as an income yield on a specified investment base or a total return including valuation changes. The two may diverge materially in changing interest-rate environments. Return on equity divides an earnings measure by an equity measure, usually using average equity for a period; it is sensitive to accounting standards, leverage, valuation changes, capital distributions, and business mix. A higher return on equity can reflect effective underwriting and investment operations, but also thinner equity and higher financial risk. Therefore profitability, capital adequacy, and risk-adjusted resilience belong in one analysis.

The OECD's 2026 insurer profitability chapter reports average 2025 return on equity of 16.7% for life insurers and 15.7% for nonlife insurers among the reporting jurisdictions covered, with positive returns in most jurisdictions. Those aggregates are evidence of observed sector outcomes, not a common company-level accounting benchmark. Company size, statutory and accounting standards, cross-country sampling, and the definition of equity all remain material.

Underwriting, liquidity, and growth

Premium growth can be beneficial when pricing adequately compensates expected losses, operating costs, acquisition, and risk capital. It can be adverse when risk selection deteriorates or long-tail reserves are understated. Cash inflows from newly sold policies may temporarily exceed cash payments even where projected ultimate profitability is negative. Conversely, an insurer can have a profitable current-service result and face near-term liquidity stress from catastrophe settlements or collateral requirements. A responsible performance report therefore reconciles insurance service earnings, cash inflows/outflows, claims settlement timing, investment liquidity, and available capital, with time-horizon differences clearly labeled.

U.S. GAAP, Statutory Accounting, and Prudential Valuation

The scope and purpose of U.S. financial reporting

The United States maintains accounting regimes with substantially different purposes. U.S. GAAP financial statements communicate to investors under the FASB Accounting Standards Codification. State-regulated insurer statutory statements support solvency supervision, policyholder protection, and evaluation of assets available to meet liabilities. Regulatory capital regimes provide intervention thresholds. These should be understood as related but differently constructed systems rather than variant names for the same profit measure.

FASB's ASU 2018-12 on long-duration contracts introduced targeted improvements to Topic 944, including updated assumptions for the liability for future policy benefits, discounting changes, market risk benefit measurement, deferred acquisition cost treatment, and enhanced disclosures. These changes matter for life and long-duration insurance reporting. The targeted amendments are different in design and scope from the comprehensive IFRS 17 general measurement model, including its CSM. For an issuer comparing IFRS and U.S. GAAP results, a reconciliation must examine specific contracts and reporting periods; it cannot assume that both sets of figures become comparable by replacing a single liability label.

U.S. GAAP should also be distinguished from U.S. statutory accounting. The NAIC's statutory accounting principles overview explains the influence of state prescribed and permitted practices and the balance-sheet orientation of solvency reporting. Some assets recognized under general-purpose GAAP may be nonadmitted for statutory purposes, affecting policyholders' surplus. The ability of a statutory report to protect policyholders depends on asset admissibility, reserves, capital adequacy, and the application of jurisdictional rules, rather than GAAP earnings alone.

Capital ratios and supervisory thresholds

Risk-based capital is a supervisory device aimed at detecting potentially weak financial condition. The NAIC's risk-based capital explanation describes required capital in relation to the insurer's size and risks, specific regulatory action levels, and the limitations of RBC ratios as a general ranking tool. A strong RBC measure is relevant to financial condition, but it does not provide an audit of claims reserves, liquidity at all dates, governance, or policyholder outcomes. Nor does a high reported return on equity establish an adequate RBC position.

Solvency II offers a further distinction. Under the Solvency II technical provisions rule, technical provisions generally equal a best estimate and a risk margin, with specified exceptions where replicating instruments provide a reliable market value. The best estimate reflects probability-weighted discounted future cash flows. Its risk margin is calibrated within the solvency framework and should not be substituted directly for the entity-specific IFRS 17 risk adjustment. The IAIS's Insurance Core Principles and ComFrame provide the international supervisory context, including valuation and capital principles, while national implementation determines binding legal obligations.

A reconciliation rather than a conversion factor

Suppose general-purpose statements show net assets of 600 units, while a regulator excludes 45 of assets from admissible surplus, measures technical obligations 25 units higher, and recognizes a 10-unit offsetting eligible adjustment under its own rules. The resulting illustrative adjusted residual is 600 − 45 − 25 + 10 = 540. This is a demonstration of reconciliation logic, not a Solvency II or statutory calculation. Real accounting-to-prudential bridges require itemized differences in asset classifications, technical provisions, tax, reinsurance recoverables, intangibles, capital instruments, and eligibility filters. A single net-asset comparison conceals the actual solvency drivers.

The most important accounting comparison is therefore not whether one framework reports a higher or lower net income. It is which set of economic claims, estimates, recognition rules, capital resources, and timing effects explains the difference. General-purpose accounting provides earnings information, statutory accounting evaluates regulator-defined resources, and economic analysis asks what risks and resources remain after measurement choices have been made.

Comparative Evidence and Interpretive Reliability

First-year European implementation

ter Hoeven, Borelli, and van Vlijmen (2024) examined first-year IFRS 17 disclosure across 24 large European insurers using their 2023 annual reports. They found that many companies reported lower opening equity at transition, while the explanation of changes was hard to compare because preceding IFRS 4 treatments and transition approaches varied. Their assessment also pointed to insufficient disaggregation and sensitivities and significant variation in adjusted operating profit definitions. The European first-year evidence supports a restrained interpretation: greater formal harmonization of standard requirements is a precondition for better comparison, but user-level comparability also depends on elections, detailed disclosures, and the ability to reconcile opening balances.

Puławska and Strzelczyk (2025), using reported assessments from 68 market participants, identified annual cohorts, onerous contracts, discount-rate approaches, and PAA matters among important implementation challenges. Their market participants' study reports professional differences in perceived difficulty. It supplies evidence about implementation experience, not a representative global estimate of accounting error frequency or the financial cost of compliance. The distinction matters because views of practitioners and financial-statement outcomes are different observations.

A Korean examination by Byun and Han (2025) compared the association of reported accounting metrics with stock prices before and after the transition. Its value-relevance study found no significant improvement in the explanatory power of earnings per share and book value per share in its sampled market and period, and questioned the information implications of insurance service profit and CSM. The finding is informative about Korean listed insurers and the researchers' model; it does not contradict the possibility of better economic disclosure in other jurisdictions or over longer post-transition periods. Value relevance is an empirical association influenced by market structure, expectations, control variables, and data quality, not a universal test of the truth of a standard.

The studies collectively establish three points: adoption can improve the architecture of reporting while implementation remains heterogeneous; accounting measures can become more structured without every metric immediately gaining market informativeness; and the evidence is conditional on sample, measurement period, and empirical method. No new universal measurement theory is inferred from these studies.

Market-level evidence and the denominator problem

OECD (2026) reports 2025 insurance-market outcomes based on national authorities' submissions across more than 50 OECD and non-OECD jurisdictions. The report's market survey methodology and scope make it highly useful for observing market developments, but the resulting aggregates cannot be interpreted as a harmonized IFRS 17 financial-statement panel. Reported premium growth, combined ratios, investment returns, and equity measures may follow different supervisory or statistical definitions.

The 2025 OECD evidence reports that the combined ratio remained below 100% in 34 of 39 jurisdictions with available information. A favorable national ratio can coexist with stressed firms, weak lines of business, or a severe deterioration in certain hazards. Conversely, a market with a ratio above 100% may still report positive net income because investment returns and other items contribute to profit. This is why market-level underwriting and entity-level shareholder measures require different denominators and interpretations.

The NAIC's 2025 U.S. P&C industry analysis illustrates the divergence directly: the industry combined ratio improved to 92.9% from 96.9% in 2024, while reported net income declined from approximately 167.2 billion to 150.6 billion U.S. dollars as net investment gains fell. The report's figures describe a national statutory-industry aggregate, not an IFRS group and not a causal estimate. The coexistence of a better underwriting ratio and lower net income is sufficient to demonstrate the need to assess underwriting and investment contributions separately.

Sources of comparability differences

Comparability requires examination of at least the following observed dimensions within every source: scope of consolidating entities, valuation jurisdiction and standard, business mix, accounting periods, reporting currencies, inflation and discount curves, gross-versus-net reinsurance convention, treatment of acquisition costs, prior-year reserve development, investment classification, policyholder participation, transitional relief, and accounting-policy elections. Some differences are disclosed; others require supplementary filings or careful exclusion of incommensurable entities.

Reported growth has additional base effects. An insurer adding an acquired book will exhibit larger premiums and liabilities that may have little relationship to organic profitability improvement. A company switching from a premium measure to IFRS 17 service revenue at transition cannot meaningfully report year-over-year service growth without a comparable restated base. Translation into reporting currency and changes in reporting scope further affect apparent growth.

Numerical Reconciliations and Sensitivity Analysis

A complete simplified profit bridge

Consider a purely hypothetical nonlife insurance operation during a single year. Its net earned premiums under a specified conventional insurance-statistics basis are 1,000 units. Net incurred claims are 640, underwriting expenses are 270, and the net combined ratio is (640 + 270)/1,000 = 91%. Its technical underwriting surplus is therefore 90. Suppose investment income attributable to the operation is 70, a recognized net investment valuation loss is 25, other corporate costs not included in underwriting expenses are 20, and tax is 15. Under these expressly simplified classification assumptions:

Underwriting surplus = 1,000 − 640 − 270 = 90.

Investment contribution = 70 − 25 = 45.

Pretax illustrative result = 90 + 45 − 20 = 115.

After-tax illustrative result = 115 − 15 = 100.

No IFRS 17 CSM is imputed from this conventional ratio example. The illustrative accounts are defined for arithmetic clarity; detailed IFRS 17 reconciliation would require insurance-service revenue and expenses, insurance finance effects, reinsurance-held presentation, and any OCI items under their applicable rules. This is the exact interpretive danger the example exposes: a 91% combined ratio is useful as a nonlife underwriting statistic, but treating its implied 90-unit surplus as audited IFRS service profit would confuse measures.

The effect of reserve development

Assume that subsequent claims investigation increases expected ultimate losses on the prior period by 60 units, charged to current-period underwriting results under the illustrative accounting convention. With current-period net earned premiums unchanged at 1,000, the ratio becomes (700 + 270)/1,000 = 97%, the underwriting surplus falls to 30, and pretax result falls to 55 if all other items remain unchanged. The comparison shows how reported performance can change through measurement of already incurred obligations without any change in premiums issued in the current period. Reserve adequacy is therefore both a measurement and a governance issue.

Separating the assumptions

A disciplined sensitivity analysis changes one assumption at a time where feasible and then models plausible joint variation. For example: an upward revision to expected nonfinancial claim severity raises projected outflows; a decrease in discount rates tends to raise the present value of longer-dated net outflows; a rise in risk uncertainty can increase the risk adjustment; a longer expected coverage duration can alter the annual pattern of CSM release; and a lower retention ratio changes the gross-to-net claim relationship. The direction of reported profit or loss depends on whether the affected cash flow belongs to current, past, or future service and whether an eligible adjustment is absorbed in CSM or recognized immediately.

When input assumptions are correlated, one-at-a-time tests can underestimate aggregate exposure. Higher claim inflation can coincide with increases in interest rates; catastrophic losses can coincide with market stress, reinsurance counterparty deterioration, liquidity outflows, and changes in exposure. A meaningful scenario analysis must preserve the accounting classification in each scenario while also showing the underlying economic total. Hypothetical stress percentages should never be presented as actuarial forecasts absent calibrated data.

An interpretive checklist for reported ratios

For each candidate metric the analyst should be able to answer: What is the numerator? What is the denominator? Are amounts gross or net of reinsurance? Which period supplies exposure? Which claims cohorts contribute? What accounting measurement basis produced the figure? Does it include changes in prior estimates, investment gains, OCI, or taxes? Is the figure reported, derived from published line items, or adjusted by management? How does the metric reconcile to the relevant audited or supervisory statement? An unanswered question is a limitation on the interpretation, not a permissible invitation to infer precision.

Disclosure, Governance, and Research Implications

Reconciliation from statements to actuarial assumptions

Financial-statement disclosure converts a valuation into an explanation of how an insurer's obligation and result have changed. For IFRS 17, important disclosures include roll-forwards of insurance contract balances, cash-flow estimates, risk adjustment, CSM movements, significant judgments and discount assumptions, sensitivity to relevant risks, and information enabling assessment of the nature and extent of insurance risk. The standard's disclosure provisions provide the normative basis. The specific balance sheets and notes, not a formula in isolation, determine whether a reader can actually reconcile a company's results.

Evidence of governance lies in reproducible data lineage: policy data and contract groups; actuarial projection assumptions; independent claims and reserving reviews; discount curve documentation; reinsurance asset estimates and credit adjustments; GL reconciliations; management overlays; and explanation of changes between periods. An insurer reporting a materially favorable CSM release while simultaneously lowering future service forecasts should provide a coherent bridge. It is the causal explanation that permits interpretation of headline earnings.

Performance metrics and incentives

Governance becomes especially important when remuneration or capital-allocation decisions depend on chosen performance measures. A target defined solely as premium growth can encourage volume without adequate pricing; a target based only on short-term underwriting margin can encourage claim-reserve optimism; a target based only on investment yield can induce credit or duration risk; and a target based only on return on equity can favor thin capitalization. A balanced set of measures should communicate current service profitability, reserve development, investment risk, policyholder obligations, and capital sufficiency, accompanied by consistent computation rules.

Such a set is an analytical practice derived from existing accounting and supervisory objectives, not a newly established doctrinal performance system. Metrics retain their formal institutional definitions. Their purpose is to support well-grounded judgment about whether recognized profit is repeatable and compatible with the insurer's long-term capacity to pay claims.

Research limits and further empirical inquiry

The present work synthesizes standards and studies and supplies checkable numerical illustrations. It does not claim to have audited insurance entities, recalculated company-specific actuarial reserves, estimated new causal treatment effects, or measured the entire worldwide IFRS 17 adoption population. The FASB and IFRS standards govern different reporting perimeters; differences are analyzed conceptually, not converted into one invented uniform global accounting basis. The OECD national-market data and NAIC industry numbers serve as dated institutional evidence.

Useful future research can construct a panel of audited insurers with consistent policy-group and line-of-business categories, reconcile GAAP service metrics with statutory underwriting metrics, disclose transition choices, and quantify the influence of risk-adjustment and discount methods. Robust tests would compare within comparable product lines, control for market conditions and mergers, and treat accounting-regime adoption as a potentially nonrandom event. Sensitivity and market outcomes should be examined jointly, with research data and methods available for replication where lawful.

The scientific contribution of this monograph is its explicit synthesis across contractual recognition, actuarial measurement, earnings presentation, empirical comparison, and prudential capital interpretation. It uses established theories, standards, and accounting identities and advances carefully bounded arguments about interpretive practice. No claim of a newly invented accounting theory, doctrine, or universal comparative index is made.

Conclusion

Insurance accounting measures a continuing promise through reporting periods in which coverage, uncertain claims, cash movements, financial markets, and capital requirements follow different clocks. IFRS 17 gives this promise a structured accounting representation by identifying insurance contracts, current fulfillment cash flows, risk adjustment for nonfinancial uncertainty, deferred contractual service profit, the liability for incurred claims, and distinct service and finance results. Its contractual service margin is a measure of unearned service profit attached to an identified group; it is neither immediate operating cash flow nor free capital. Its revenue measure records insurance service rather than simply treating premium receipts or investment deposits as sales.

The analysis establishes that insurance profitability must be read through reconciled perspectives. Underwriting ratios are informative about claim and expense experience relative to defined premium denominators. Insurance service result reflects the recognition of contractual coverage and its measurement adjustments. Investment performance reflects assets, discounting, financial risk, and accounting presentation. Net income and comprehensive income combine these results under reporting conventions. Statutory surplus and risk-based capital evaluate different aspects of policyholder protection and regulatory intervention. A number from one perspective cannot substitute for the other without an exact reconciliation.

Research on early IFRS 17 implementation confirms meaningful improvement in reporting structure alongside continued heterogeneity in transition choices, actuarial judgments, presentation, and disclosure. Market data confirm that favorable underwriting trends can coexist with weaker total net income. Consequently the valid unit of international comparison is a clearly defined metric situated within its contract type, recognition method, reporting perimeter, reinsurance treatment, accounting basis, and historical year.

The methodological conclusion is practical. Every material performance claim should identify its source statement, exposure denominator, accounting or supervisory basis, period, and relationship to other major profit and risk measures. When this is done, reported financial performance becomes a coherent explanation of how insurance promises are priced, fulfilled, financed, and supported across time.

References