RESEARCH DOSSIER
Insurer Investment Portfolio Construction
Abstract. Insurer investment portfolio construction is the selection, sizing, financing, and governance of financial assets within an enterprise whose cash obligations arise from insurance contracts. It is a joint problem of intertemporal asset allocation, contingent liability valuation, liquidity provision, regulatory capital, and institutional control. This monograph establishes an integrated analytical account of that problem by relating modern portfolio selection, actuarial immunization, stochastic liability cash flows, economic surplus, solvency requirements, and multi-horizon liquidity constraints. The analysis distinguishes the investment decision for life, annuity, property and casualty, and health insurance general accounts, while accounting for the materially different economics of policyholder-directed investment products. It compares the international supervisory principles of the IAIS with relevant European Union prudential requirements, U.S. state-based risk-based capital oversight, and the separate measurement discipline of IFRS 9 and IFRS 17. Recent official evidence from the NAIC, EIOPA, IAIS, and Federal Reserve demonstrates both the continued importance of fixed income and the growing analytical significance of privately originated credit, indirect exposures, collateral calls, and nontraditional funding. Mathematical sections derive a liability-relative allocation objective, duration sensitivity of surplus, scenario-constrained liquidity coverage, and a transparent illustrative asset mix. The original synthesis identifies portfolio feasibility as the intersection of contingent claims, time-specific cash availability, capital resilience, and governable exposures. The contribution is analytical and editorial: it introduces no new named theory, regulatory measure, or claim of empirical estimation.
Keywords: Insurance investments; insurer asset allocation; portfolio construction; asset-liability management; economic surplus; prudent person principle; insurance capital; duration and convexity; credit spread risk; private credit; liquidity stress; insurer governance; IFRS 9; IFRS 17; risk-based capital; solvency; investment mandates.
Introduction
The financial assets of an insurer are simultaneously invested capital and a means of discharging contractual obligations. Portfolio construction therefore starts with the promises issued to policyholders. Insurance liabilities determine the currencies, potential payment dates, embedded financial guarantees, optional withdrawals, and contingencies that an asset portfolio must support. Capital requirements and supervisory standards restrict the losses that the insurer may absorb; operational capabilities determine the exposures that it can evaluate and control. The resulting investment objective is contingent on the insurer's liability structure and its entire balance sheet.
The distinction can be seen directly in the historical foundations. Markowitz's portfolio selection analysis established the relevance of the covariance of returns for diversification. Redington's life-office valuation analysis focused on the sensitivity of assets relative to insurance obligations, including the circumstances under which a fund could be locally immunized against interest-rate changes. Together they supply two essential dimensions of insurer investment choice: diversification among assets and consistency between asset and liability values. Neither dimension can be replaced by the other.
A further distinction concerns time. A portfolio that appears liquid and profitable at a year-end valuation may fail to supply cash during a wave of surrenders, a catastrophe, or a derivatives margin shock. A portfolio that maximizes accounting income over one reporting period may create unfavorable economic exposure to rate movements or future reinvestment conditions. A high-yield asset may appear attractive within the investment account while worsening the insurer's solvency position when its default risk, liquidity cost, or interaction with insurance obligations is considered. These are separate, jointly binding questions.
The institutional importance is observable in official investment records. The NAIC's year-end 2025 asset-mix report measures approximately $9.58 trillion in U.S. insurer cash and invested assets on a book/adjusted carrying-value basis, with bonds representing approximately 59.8% of the total. Its report is an aggregate U.S. statutory-filing description that includes affiliated and unaffiliated investments; the numbers are not a universal international portfolio target. The EIOPA October 2026 insurance risk dashboard describes a different EU reporting population and different exposure definitions. Its Q2 2026 reported median bond exposure of 49.7% of total assets cannot be subtracted directly from the NAIC's year-end industry aggregate investment-share measure to infer a cross-regional portfolio preference.
The scope of this monograph is the construction of insurers' strategic portfolios, including asset selection, allocation, liability-sensitive risk budgets, executable liquidity, and governance. Detailed calibration of credit-spread derivatives, the internal mechanics of every private-credit structure, or the specialized rules of individual insurance products belong to adjacent research dossiers. They enter here insofar as they alter portfolio feasibility, concentration, liquidity, accounting, or solvency. The method combines primary regulatory reading, analysis of foundational financial literature, reproducible stylized calculations, and comparison of supervisory evidence. Monetary examples and parameters described as hypothetical are demonstrations, not empirical portfolio estimates or investment recommendations.
Insurance Investment Universe — Concepts, History, and Scientific Object
The insurer's investment balance sheet
An insurer receives premium and other financing inflows, invests available resources, and owes benefits and claims whose timing and amount depend on contractual terms and uncertain events. At date zero, a simplified economic balance sheet has market-valued assets A, economically valued liabilities L, and economic surplus S = A - L. This identity is useful for exposition but does not substitute for either regulatory own-funds definitions or financial-reporting equity. In particular, liabilities can contain options and nonfinancial contingencies whose economic valuation responds differently to financial scenarios than the accounting carrying amount.
Investment assets may be grouped by contractual cash-flow structure and liquidity characteristics: sovereign and agency debt; public corporate bonds; cash and deposits; mortgage and structured credit; common and preferred equity; property and infrastructure; private placements and private loans; investment funds; and derivatives used for hedging, efficient management, or other purposes. These categories overlap under look-through analysis. For example, a fund unit may represent loans or equities held indirectly, and a structured note may embed multiple market and credit exposures. Portfolio reporting must identify both the legal security held and its relevant underlying economic risks.
The insurer's economic interest in fixed income is explained by the relationship between bond cash flows and liabilities payable over time. A coupon-paying bond contributes scheduled receipts that may fund claims and benefits. Its market value nevertheless varies with discount rates and credit spreads; principal repayment creates reinvestment exposure; call and prepayment terms change the timing of receipts. A long-maturity bond may improve the duration match for an annuity portfolio while making the insurer vulnerable to credit deterioration or a near-term liquidity shock. Risk is defined relative to the complete asset-liability and funding position, not solely the label of an asset class.
The investment functions of equities and real assets differ. They can supply participation in economic growth, real income, or diversification across return drivers, but their payments and market values are generally less predictable than those of high-quality fixed obligations. Their inclusion has to be assessed against the insurer's risk appetite, policyholder guarantees, underwriting volatility, capital consumption, and capacity to realize value at the dates when money is needed. Private securities additionally require rigorous assessment of valuation reliability, recovery assumptions, document rights, exit options, and concentration in borrowers, sponsors, servicers, managers, and underlying collateral.
Origins of the analytical discipline
The history of insurer portfolio construction links actuarial valuation with financial economics. Redington (1952) studied a life office as a matched structure of inflows and obligations. The first-order sensitivity condition subsequently associated with immunization reflects how the combined fund responds locally to yield changes; its usefulness depends on valuation assumptions and an appropriate second-order condition. Fisher and Weil's interest-rate risk study developed bond-holding strategies that address fluctuations in the term structure and provide a further basis for immunization and horizon analysis. These results established the importance of liability payment dates and the dynamics of reinvestment, beyond treating a bond as a single-yield instrument.
Markowitz (1952) formalized asset diversification through the relation between expected return and portfolio variance. The contribution survives in insurer portfolios because positions in different securities, issuers, sectors, and currencies may be exposed to common and distinct shocks. The limitations arise when asset-only variance obscures liabilities, capital rules, liquidity constraints, and tail dependence. Campbell and Viceira's long-horizon portfolio analysis demonstrates why the investment horizon and changes in the investment opportunity set matter. Long-dated income obligations may make instruments that are volatile over a short mark-to-market horizon appropriate hedges over a long liability horizon.
The historical progression is therefore additive. Insurance valuation contributes the structure of obligations; modern finance contributes diversification and intertemporal choice; institutional regulation and governance determine which combinations can be carried by a licensed insurer. The relevant theoretical object is the feasible asset strategy conditional on its obligations and institutional constraints.
Distinctions among insurance business models
Traditional life and annuity insurers can carry decades-long benefits, surrender rights, crediting promises, and embedded options. Their allocation often centers on long-duration fixed income and other cash-generating assets, while requiring explicit allowance for policyholder actions and liquidity provision. For guaranteed products, an increase in rates may reduce economic liability values while increasing the attractiveness of surrender or new product alternatives. These effects have different timing: a valuation gain from liability discounting does not itself provide cash for withdrawals.
Property and casualty insurers face claim obligations linked to underwriting events and settlement practices. Their liabilities may be shorter or more volatile, with catastrophe and litigation exposures capable of producing concentrated settlement demands. An insurer writing long-tail liability insurance also faces material inflation and claims-development risk. Health insurance portfolios may have short cash settlement cycles and high operational claims throughput, though product and jurisdictional design vary considerably. A universal strategic asset allocation across these lines would ignore the basic mechanism by which their obligations differ.
Unit-linked and similar investment-linked arrangements require particular care. Where the policyholder bears the investment risk and benefits track designated underlying assets, matching may mean holding the linked units or corresponding assets rather than seeking an independently maximized spread. EIOPA's Article 132 prudent-person rule makes this distinction explicit for the relevant life insurance contracts. The general account and policyholder-directed investment account should therefore be analyzed separately even when both are legally held by an insurance entity.
Asset labels versus underlying risks
Portfolio classification can conceal systematic exposure. A public investment-grade bond and a privately placed loan may share borrower, industry, collateral, and refinancing risks. A securitized position may contain layers of credit exposure and structural leverage. Affiliated asset-management platforms can produce repeated economic exposures across bonds, fund units, warehouse facilities, and related-party transactions. In a Federal Reserve analysis of risky credit intermediation, Carlino, Foley-Fisher, Heinrich, and Verani (2025) document the role of life insurers in direct and indirect corporate credit investment and the increasing complexity of affiliated channels.
The NAIC's private-credit category analysis states that its approximately $1.21 trillion estimate for 2025 is based on a broad definition, with approximately $544 billion concentrated in categories of particular supervisory interest such as privately rated bonds, business development companies, and private-credit funds. Those categories are not economically homogeneous. A sound investment inventory assigns each position more than one tag: legal form, asset class, ultimate borrower or risk factor, investment liquidity, valuation basis, counterparty or sponsor, encumbrance, currency, and capital treatment. Aggregation by a single reporting label is insufficient for a material exposure assessment.
The portfolio as an institutional commitment
The strategic allocation is enacted through investment policy, asset-manager mandates, capital plans, risk models, internal limits, custodian records, and contingent funding procedures. The portfolio is also exposed to changes in premium production, claims, reinsurance recoveries, and shareholder actions. Portfolio construction must therefore be repeated as a controlled decision process rather than regarded as a static vector of percentages. The allocation is admissible only if the insurer can identify the assets, validate valuations, forecast relevant liability cash requirements, monitor concentration, and fund foreseeable stressed obligations.
Liability-Informed Portfolio Objectives — Financial and Actuarial Mechanisms
Economic surplus and state-dependent obligations
Let A_t denote the market or economic value of investment assets at time t and L_t the economic value of liabilities under a stated valuation convention. The surplus is S_t = A_t - L_t. Both values depend on the state of the world: interest rates, inflation, credit conditions, exchange rates, equity markets, insurance experience, policyholder behavior, and management actions. The surplus sensitivity to a factor x is given formally by dS/dx = dA/dx - dL/dx. This derivative expresses a balance-sheet condition. Its empirical magnitude must be estimated with assets and liabilities evaluated consistently; otherwise it mixes incompatible definitions of value.
A purely asset-based optimizer might maximize E[R_A] - (gamma/2) Var(R_A), where R_A is asset return. A liability-aware formulation evaluates the distribution of the change in S, possibly after operating cash flows, distributions, financing, and management interventions. One simple economic objective over horizon T is to maximize E[S_T] minus a penalty for downside surplus outcomes, subject to hard bounds on capital and timely cash. The objective must specify the treatment of shareholder value, policyholder protection, and the capital required to absorb uncertainty. A stochastic enterprise decision can use multiple constraints without implying that social, regulatory, and contractual objectives can all be represented by one scalar risk-aversion coefficient.
Present value and duration
For deterministic liability cash flows C_t discounted at effective yields y_t, a stylized liability value is L = sum_t C_t (1 + y_t)^(-t). For continuous compounding it can be written L = sum_t C_t exp(-r_t t). Actual insurance liability valuation is more complex: stochastic cash flows, mortality, lapses, embedded options, nonfinancial risk adjustment, and the applicable reporting or supervisory discount methodology must be incorporated. The formula isolates the interest-rate mechanism.
Define modified duration D as minus the proportional first derivative of value with respect to a parallel yield change: D = -(1/V)(dV/dy). For small changes in a common yield factor, delta A is approximately -D_A A delta y and delta L approximately -D_L L delta y. Hence delta S is approximately -(D_A A - D_L L) delta y. A first-order parallel-rate neutral surplus requires D_A A = D_L L. The frequent shortcut D_A = D_L works only when the asset and liability present values are equal or the surplus holding is modeled separately.
As a deliberately hypothetical calculation, let A = 100, L = 92, D_A = 5, and D_L = 8, all on a consistent valuation basis. Then the asset dollar-duration is 500 and liability dollar-duration is 736. For a parallel 100-basis-point yield rise, the first-order asset change is -5, the liability change is -7.36, and surplus increases by approximately 2.36 units, from 8 to 10.36. A corresponding fall produces an approximate 2.36-unit decrease in surplus. This is a local symmetric approximation. Convexity, key-rate mismatches, credit spreads, optional withdrawals, and changing liabilities can invalidate symmetry and magnitude, especially under large moves.
Convexity, curve shape, and optionality
The next Taylor term adds one-half of convexity times value times squared rate change. For the surplus, the contribution is approximately 0.5 (C_A A - C_L L) (delta y)^2. Redington's immunization logic requires both local sensitivity control and attention to the curvature of the fund's excess value. Even those conditions apply to specified yield transformations, asset cash flows, and valuation conventions. Real yield curves move by level, slope, and curvature, so a single duration match may leave significant key-rate exposure. An insurer should compare asset and liability sensitivities at meaningful maturity nodes, including the long end relevant to annuity liabilities.
Embedded insurance options amplify the problem. Guaranteed minimum crediting, annuitization, death benefits, surrender rights, and other features alter obligations when policyholders respond to rates, markets, and personal circumstances. The insurer may hedge a financial option using liquid derivatives while retaining uncertainty in lapse and exercise behavior. Model validation must distinguish observable market inputs from assumptions about human behavior; the most precise bond-pricing model cannot remove uncertainty in the contingent contract cash flows it is intended to support.
Liability segmentation and investment horizons
The aggregate balance sheet can conceal the differences among cohorts and legal funds. Useful segments include currency, contract guarantee, product type, surrender pattern, expected claim settlement, and jurisdictionally restricted assets. Segmentation supports an attribution of assets to liability demands, but group-wide treasury needs remain relevant. Ring-fenced portfolios and linked product assets may be unavailable for another business segment. Thus the same instrument may be liquid to one legal entity and unavailable to another, even under a consolidated liquidity metric.
Long-horizon analysis changes what counts as safe. Campbell and Viceira (2002) show that long-term bonds can hedge long-horizon needs, while short-term holdings expose the investor to changing future reinvestment opportunities. For an insurer, the appropriate long-term instrument also depends on the nominal or real denomination of liabilities, counterparty credit, default risk, and the capacity to realize cash before scheduled maturity. It is analytically incomplete to identify either cash or long-duration government bonds as universally risk free.
Profitability, surplus, and cash are separate outputs
A well-performing insurer investment mandate should report at least four outcomes: asset income or total return; economic surplus and its scenario sensitivities; capital requirement or solvency resilience; and cash availability over operational horizons. Accounting earnings supply an additional, necessary measurement. The outputs may diverge. An unrealized gain may improve economic surplus without generating settled cash; a reported interest margin can coexist with increased spread or default exposure; regulatory capital can rise during a period of positive asset returns; and asset sales to meet claims can alter the remaining portfolio's future earnings and duration.
The distinction allows clearer decisions about the fundamental objective. A proposed allocation should be accepted only on evidence that the gain in its chosen economic objective compensates for incremental capital, liquidity and governance commitments. This is a testable decision logic rather than a new formal theory. It follows directly from treating the insurer as an issuer of contingent liabilities with a finite capacity to absorb loss and fund payments.
Risk and Return Allocation — Quantitative Methods and Assumptions
From exposures to a constrained decision
Let w_i be the fraction of available economic investment assets assigned to asset category i, with sum_i w_i = 1. For a long-only strategic portfolio, w_i is nonnegative; derivatives and financing may produce economically negative or leveraged exposures and require a different representation. With expected total-return vector mu and covariance matrix Sigma, an asset-only mean-variance objective is w' mu - (gamma/2) w' Sigma w. This familiar representation captures only part of the insurer problem. A complete allocation also requires a map from asset positions to liability changes, capital consumption, liquidity sources and uses, regulatory eligibility, and the organization's capacity to govern the exposures.
Suppose Z denotes stochastic external factors and L_T(Z) the value or cash settlement of liabilities at horizon T. The insurer's asset payoff is A_T(w,Z), and a simplified terminal surplus is S_T(w,Z) = A_T(w,Z) - L_T(Z). A risk-sensitive choice may maximize E[S_T] - lambda ES_alpha(-S_T), where ES is expected shortfall of a suitably defined loss at confidence level alpha. Capital, liquidity, and portfolio-composition conditions act as additional constraints. This is a schematic objective; an applied implementation must establish the accounting numeraire, whether capital is infused or distributed during the horizon, the governance-approved tail measure, the liability valuation model, and the treatment of cash flows. A mathematically solvable program is not automatically an admissible insurer decision.
Rockafellar and Uryasev's conditional value-at-risk optimization shows how an expected-tail-loss objective can be handled in a tractable scenario-based optimization. It is useful for distributions where extreme losses drive solvency concerns. Yet empirical tail modeling is especially demanding for insurers: historical covariance matrices may not capture correlated underwriting catastrophes, simultaneous market shocks and surrender behavior, or forced selling. Tail estimates should accompany nonprobabilistic stresses and explicitly specified severe scenarios rather than being treated as complete statements of enterprise risk.
Risks that require separate measurement
Interest-rate risk includes level, shape, and basis moves in discount curves. Credit risk includes issuer default, downgrade, migration, spread revaluation, collateral deterioration and recovery uncertainty. Equity and real-asset risks include price fluctuations, liquidity discounts, and exposure to economic conditions. Currency risk arises when the denomination of assets, liabilities, and settlement needs diverges. Inflation exposure depends on which claims and expenses are nominal, contract-indexed, or subject to claims-cost inflation. Counterparty risk includes derivatives and reinsurance receivables, custodial and settlement exposures, and concentrations in affiliated financial entities. Operational and legal risks include investment authorization, custody, collateral enforceability, valuation conflicts, and outsourced manager failure.
Covariance-based risk measures describe dependence under a chosen probability distribution and horizon. Structural concentration describes economic exposures that a short history may miss. For example, holdings may be diversified across legal bond issuers but tied to the same commercial property market, asset manager, reinsurance transaction, or refinancing channel. The portfolio control system must capture issuer, ultimate parent, sector, geography, risk driver, collateral, originator, manager, counterparty, and currency concentrations, with look-through wherever the necessary data are available. An absent look-through record is an identifiable information limitation and can itself justify an investment restriction.
A transparent illustrative allocation
Consider a hypothetical insurer general account with 100 units of investable economic assets and four broad asset sleeves. The values below are scenario parameters selected for analytical demonstration. They are not market estimates, regulatory factors, or allocations recommended for any insurer.
| Asset sleeve | Weight | Illustrative gross annual expected return | Illustrative capital proxy per unit |
|---|---|---|---|
| High-quality fixed income | 65% | 3.8% | 3% |
| Public equities | 10% | 7.0% | 25% |
| Cash and short-term liquidity | 10% | 2.0% | 0% |
| Illiquid credit and other alternatives | 15% | 6.0% | 20% |
| Total | 100% | 4.27% weighted average | 7.45 units weighted proxy |
The expected return arithmetic is 0.65(3.8%) + 0.10(7.0%) + 0.10(2.0%) + 0.15(6.0%) = 4.27%. The illustrative linear capital proxy is 100[0.65(0.03) + 0.10(0.25) + 0.10(0) + 0.15(0.20)] = 7.45 units. This latter number is explicitly not a Solvency II Solvency Capital Requirement, U.S. RBC result, or IAIS Insurance Capital Standard result. Real capital measures use regime-specific risk calibrations, correlations, look-through, offsets, tax and loss-absorbing effects, entity boundaries, and valuation rules. The proxy exists solely to make the trade-off arithmetically transparent.
Now shift ten percentage points from high-quality fixed income into the illiquid alternative sleeve, holding all other assumptions constant. The gross expected return rises by 0.10(6.0% - 3.8%) = 0.22 percentage points, from 4.27% to 4.49%. The simple capital proxy increases by 100(0.10)(0.20 - 0.03) = 1.70 units, from 7.45 to 9.15. The shift may also worsen liability cash-flow matching and near-term available liquidity. Even in this intentionally optimistic deterministic illustration, the apparently favorable expected-return increment is bought with a considerably larger hypothetical capital allocation. A decision cannot be based on the return increment alone.
Sensitivity and model risk
Portfolio rankings are fragile when expected returns differ by small amounts. If the illiquid sleeve's assumed excess gross return over high-quality fixed income is 2.2 percentage points, a 10-point allocation shift appears to add 22 basis points to total portfolio return. A 2.2-point reduction in that assumed excess return removes the entire modeled advantage. A transaction cost, credit impairment, or forced-sale discount can eliminate it even when the expected-return assumptions remain unchanged. Since such inputs are uncertain, the investment committee should examine break-even assumptions rather than presenting an optimizer's point estimate as a decision.
Covariance estimation is similarly sensitive. Private assets may be valued infrequently, creating smoothed return series with understated short-horizon volatility and correlation. Models can mistake the absence of quoted price changes for diversification. A defensible analysis should compare recorded returns with cash-flow-based scenarios, public market proxies, independent valuation marks, credit and collateral migration, and realized recovery assumptions. It should document which risk factors cannot be measured reliably from available data and what additional capital or investment limits are imposed to reflect that uncertainty.
Immunization and multi-factor hedging
Duration matching of liabilities can be implemented through cash bonds, interest-rate derivatives, or combinations of both. The measurement should be made in comparable valuation units. A portfolio with assets of 100 and liabilities of 92 can achieve zero first-order parallel-rate surplus sensitivity by setting its asset modified duration to 7.36 when liability duration is 8, since 100(7.36) = 92(8). This does not guarantee cash matching, credit safety, or neutrality to curve twists. A hedge using swaps or futures may reduce interest-rate exposure while generating daily variation-margin needs. The liquidity consequence must be included at inception, not merely after rates have moved.
Fisher and Weil (1971) provide a basis for evaluating rate shifts over an investment horizon, while the investment literature emphasizes exposure to a changing set of opportunities. The choice between a physical long-duration asset and a derivative overlay depends on credit risk, counterparty arrangements, funding, collateral terms, capital treatment, and governance capacity. An allocation process should identify the instrument that best supplies the required balance-sheet effect after these constraints are applied; nominally equal durations may have unequal economic reliability.
Scenario vectors and aggregation
Each material scenario should define a consistent vector of changes: rates by tenor, spread movements by credit and sector, defaults, equity and property prices, inflation, currency, liability cash flows, surrender rates, reinsurance payments, derivatives margin, asset sale haircuts, and financing capacity. The model then maps the scenario into A, L, solvency measures, and cumulative cash at several horizons. It should report the effect of the scenario on surplus and liquidity separately, since the sign and timing can differ. Scenario aggregation can reveal when an individually modest exposure creates an enterprise vulnerability through shared factors.
Scenario design requires evidence as well as judgement. EIOPA's 2024 insurance stress test evaluated combined adverse financial and insurance shocks and compared fixed-balance-sheet with management-action assumptions. Its published findings describe sensitivity to simultaneously rising rates and risk premia while reporting that the tested participants held enough liquid assets to meet the simulated needs. That result pertains to the exercise, participating insurers, defined stresses, and associated assumptions. It does not establish universal protection against every liquidity shock or certify a particular investment allocation.
Solvency Constraints — Insurance Application and Comparative Evidence
International supervisory baseline
The IAIS's Insurance Core Principles and ComFrame, December 2024 edition, establish an international supervisory reference framework rather than a directly operative law in every jurisdiction. ICP 15 addresses investment and recognizes the importance of asset quality, liability dependence, diversification, and controls. ICP 16 calls for an integrated enterprise risk management framework encompassing asset-liability management, investment and liquidity risk, capital adequacy and the own risk and solvency assessment. The text makes clear that quantitative restrictions need to be complemented by qualitative management of investment risk. This supports an insurer-specific mandate in which the ability to understand and control an investment is relevant alongside its modeled profitability.
At the level of internationally active insurance groups, the IAIS Insurance Capital Standard was adopted in December 2024 as a global, risk-based group capital measure. Its scope and implementation architecture must be distinguished from entity-level domestic laws. The ICS is an important international reference for group capital discussions, but a manuscript cannot substitute an ICS figure for a legal requirement applicable to every insurance company or use one capital ratio as if its definitions were identical across regimes.
A useful supervisory distinction is between the permission to hold an asset and the economic ability to carry its associated risks. Legal eligibility, portfolio concentration limits, asset-valuation rules, capital requirements, and prudent-management duties may operate simultaneously. Compliance with one rule does not determine the portfolio's complete solvency quality. The investment process must identify every applicable requirement and document which constraints bind within each insurer and subsidiary.
European Union: prudent investment and capital sensitivity
Under Solvency II, Article 132 requires the prudent person principle across the whole portfolio. The undertaking must invest in assets whose risks it can properly identify, measure, monitor, manage, control, and report. It must also consider security, quality, liquidity, and profitability of the portfolio as a whole and the nature and duration of the obligations covered. These requirements are relevant to investment policy, model validation, asset eligibility, and the degree of complexity that the insurer can reasonably supervise.
The Solvency II Article 105 market-risk module includes interest-rate, equity, property, spread, currency, and market concentration components. Crucially, its description connects market-price shocks with the values of both assets and liabilities and specifically recognizes their structural mismatch, including duration. An EU insurer's allocation should therefore be examined as a balance-sheet risk position. The relative regulatory consequences of two portfolios cannot be inferred from their return volatility or bond ratings alone.
The regulatory basis requires temporal precision. EIOPA's Solvency II framework overview identifies Directive (EU) 2025/2 as an amending instrument whose new rules take effect on January 30, 2027. This manuscript is assessed as of November 15, 2026; it treats the future-effective amendments as prospective and refrains from presenting their January 2027 requirements as the current law governing a 2026 allocation. Applied work must also verify the transposition and applicable delegated or implementing texts for the precise valuation date.
The prudent-person principle is not an argument for a single universal safe asset mix. Its whole-portfolio test asks whether an asset's risk is comprehensible and manageable in relation to the insurer's obligations and risk capacity. A diversified portfolio can breach its own liquidity or concentration constraints; a concentrated duration hedge can lower a particular liability-relative risk while increasing credit or execution risk. The practical construction objective is to solve these interacting constraints together and state any trade-offs explicitly.
United States: state oversight and risk-based capital
The U.S. insurance supervisory system is state-based and uses statutory accounting, investment regulation, risk-based capital standards, and other supervisory tools. The NAIC's risk-based capital explanation, updated June 30, 2026, states that RBC requirements reflect insurer size and inherent financial and operating risk, and are intended as regulatory intervention thresholds. The NAIC cautions that RBC is not necessarily the total capital an insurer needs to meet its objectives, and is not an appropriate general ranking of insurers. It follows that treating the minimum RBC threshold as a comprehensive investment risk budget would misinterpret its purpose.
State investments laws vary, and the NAIC maintains both defined-limits and defined-standards variants among its investment model laws. A model law does not automatically operate identically in each state. An actual U.S. portfolio optimization must take as inputs the specific domicile's adopted statutes, statutory statement requirements, asset admissibility, group transactions, concentration limits, and insurer-type RBC instructions. A generic linear "capital charge per asset" table has pedagogical value but cannot replace these computations.
The NAIC 2025 asset-mix study demonstrates the diversity of assets in actual U.S. insurance accounts: at year-end 2025, bonds accounted for 59.8% of industry cash and invested assets, common stock 13.6%, mortgage loans 9.1%, Schedule BA and other long-term investments 6.7%, and cash and short-term investments 6.2%. These are book/adjusted carrying-value classifications and percentages of an aggregate; they are neither risk weights nor a prescription for an individual insurer. Differences among life and property-and-casualty business mix, affiliated equity holdings, and liability character matter to the interpretation. In particular, the report describes substantial common-stock holdings within property-and-casualty insurers, including affiliated positions, while life insurers have a far smaller direct common-stock share.
Private credit and capital-quality questions
The NAIC's private credit report for 2025 estimates approximately $1.21 trillion of U.S. insurer exposure under a broad definition, about 13% of cash and invested assets, and identifies approximately $544 billion in categories warranting especially close attention for complexity and opacity. The report stresses that definitions overlap and available statutory reporting data impose limitations. Portfolio researchers must avoid adding overlapping labels as if they were disjoint assets, treating the broad-definition total as equivalent to direct lending, or inferring that every dollar has an identical risk profile.
A risk-based reading combines the actual underlying borrower and collateral with structural seniority, loss waterfalls, valuation evidence, leverage, liquidity, and the extent of independent credit analysis. A privately rated instrument requires attention to the reliability and comparability of credit assessment. A fund position requires look-through to commitments, uncalled capital, leverage, and asset-level exposures, with documentation of opacity where transparency is incomplete. The Federal Reserve's 2025 analysis of life insurers' credit channels provides independent evidence of growing interaction among insurers, affiliated asset managers, and direct and indirect lending structures. The existence of these channels makes a governance-quality assessment necessary, in addition to a numerical spread comparison.
Accounting measurement is distinct from regulatory solvency
Financial reporting changes how investment performance and insurance finance effects appear in accounting statements. The IFRS Foundation's IFRS 9 Financial Instruments guidance describes classification based on the business model and the contractual cash-flow characteristics of financial assets. In the relevant cases it distinguishes amortized cost, fair value through other comprehensive income, and fair value through profit or loss. A sale, hedging strategy, or change in investment-management objective can therefore have accounting implications that are analytically separate from the asset's economic risk.
The IFRS Foundation's IFRS 17 Insurance Contracts framework measures insurance contract groups using current fulfilment cash flows and a contractual service margin where applicable, distinguishes insurance service results from insurance finance effects, and permits specified presentation choices for finance income or expenses. IFRS 17 applies for annual periods beginning on or after January 1, 2023, subject to its transition provisions and adoption by reporting jurisdictions. Asset and liability measurement and presentation must be considered jointly when interpreting reported volatility; an accounting mismatch does not necessarily identify an economic mismatch, and the absence of accounting volatility cannot establish economic safety.
Three reporting views are therefore required in a rigorous assessment: economic or market-consistent balance-sheet effects, applicable supervisory capital and admissibility, and the accounting classification and recognition of income and value changes. They should be reconciled with a documented bridge rather than collapsed into a single figure labeled risk or profitability. The bridge is particularly important for longer-dated liabilities, alternative assets valued with significant assumptions, and hedging strategies whose cash and valuation consequences occur at different times.
Limits of international comparison
The EIOPA October 2026 dashboard reports Q2 2026 median exposures of roughly 49.7% of total assets to bonds and 6.0% to equities and participations, on the definitions specified for its indicators. It also reports median cash holdings of approximately 0.7% of total assets and a median liquid-asset share of 46%. These quantities describe different categories and denominators from the NAIC's aggregate book/adjusted-carrying-value portfolio data. They are not direct cross-jurisdictional matched samples. Their proper use is to illuminate institutionally meaningful exposures while prompting further reconciliation of denominator, legal entity, coverage, treatment of funds, valuation basis, and reporting date.
The IAIS's Global Insurance Market Report overview identifies private credit, geoeconomic fragmentation, and other emerging exposures as areas of supervisory attention. The report's framing is global, but capital adequacy and portfolio law remain jurisdiction-sensitive. A robust research conclusion is that common risk factors can be studied internationally while legal feasibility must be established locally. This distinction is essential to the international scientific character of insurer portfolio research.
Liquidity and Costs — Stress Testing, Model Limits, and Verification
Liquidity is a dated cash requirement
The relevant liquidity question is whether the insurer can deliver the correct currency and amount of cash at the required date, while preserving a viable remaining portfolio. Total reported assets and even total holdings of marketable securities do not answer that question. Available cash depends on settlement periods, custody restrictions, collateral commitments, sales capacity, price haircuts, market depth, encumbrance, and legal-entity transfer restrictions. A long-term solvent insurer can face short-term stress if claim payments or withdrawals arrive before cash can be realized from assets without unacceptable losses.
A practical cash-flow projection separates contractual and stochastic inflows and outflows by day, week, month, quarter, and longer horizons. Outflows include paid claims and benefits, policy surrenders, operating expenses, reinsurance settlements, collateral and margin requirements, debt maturities, tax obligations, committed investments, and capital transactions. Inflows include premiums, coupon and maturity payments, recoveries, asset sales, repos and other financing if legally and operationally available. Each contingent inflow should be assessed for collectibility and timing under the very scenario that generates the outflow. For example, a reinsurance recovery received months after a catastrophe may improve ultimate profitability but not fund tomorrow's catastrophe claims.
The calculation must distinguish the cash that is legally owned by an insurer from funds restricted by policyholder investment arrangements, local regulations, collateral obligations, or ring-fenced portfolios. A liquid bond cannot be sold and simultaneously counted at full value as pledged collateral. Similarly, a liquidity line whose draw conditions fail in systemic stress should not be assigned the same credibility as cash already available at the operating bank. These are accounting, operational, and legal constraints as well as economic ones.
A dated stress-coverage test
For a chosen scenario s and time horizon h, define cumulative stressed cash demand N_s(h) as cumulative outflows minus reliable contractual inflows. Define E_s(h) as immediately available unencumbered cash plus realizable net sales proceeds and committed funding that can actually settle by h after scenario-specific haircuts and costs. A necessary liquidity condition is E_s(h) >= max(0,N_s(h)) for each relevant h and s. This is a feasibility test, not a new regulatory liquidity ratio. It must be evaluated along the time path: resources available at month-end do not cure a payment shortfall on an earlier day.
Consider a stylized insurer facing 12 units of stressed payments over five working days. It has 4 units of immediately available cash and 10 units of unencumbered government bonds that can settle within the horizon. If the bonds can be sold with a 5% all-in stressed haircut, realizable sale proceeds are 9.5 and eligible sources total 13.5, creating 1.5 units of headroom. If only 7 units of the bonds can be settled within five days, realizable proceeds are 6.65, total sources are 10.65, and the five-day deficit is 1.35. If 4 units of the 10 are encumbered, the available portion similarly falls to 6 units before applying a haircut. The illustration shows that a reported holding value does not determine dated liquidity. These values are hypothetical and should not be read as observed insurer liquidity conditions.
The scenario should also analyze how sales affect future duration, credit concentration, liquidity buffers, and investment income. Selling the most liquid bonds first can leave the firm holding a more illiquid residual portfolio while its balance-sheet risk worsens. This adverse selection matters in a prolonged stress: an insurer may meet the first wave of outflows and become less resilient to a second wave. A liquidity plan therefore needs a post-sale portfolio test, not only a single minimum-cash constraint.
Policyholder behavior, catastrophe payments, and collateral calls
Life insurance cash flows can be rate-sensitive because new investments or alternative savings products become more attractive as interest rates rise. Mass surrender or lapse assumptions are scenario-specific and should be grounded in contract design and the policyholder population. A catastrophe can create rapid and geographically concentrated claims payments that need a different liquidity profile. The timing of reinsurance recoveries may diverge from the timing of primary claims. Hedging programs can generate margin calls following rate or market changes even when the hedge improves long-run economic surplus. A portfolio optimization that models only asset market value will miss these cash demands.
The Financial Stability Board's margin and collateral preparedness recommendations apply to a range of nonbank market participants including insurers. They emphasize liquidity risk governance, stress testing, and collateral management for market-wide increases in margin and collateral calls. Their relevance is not that every insurer engages in identical derivatives activity, but that a liquid hedge portfolio must include operational access to eligible collateral under stressed settlement conditions.
EIOPA's 2024 insurer stress-test results provide an institutional example of the difference between cash holdings and a wider liquid-asset buffer. The participating insurers in that exercise experienced modeled liquidity pressure under a severe joint financial and insurance scenario and were assessed using both fixed-balance-sheet and constrained-balance-sheet assumptions. Participants had a sufficient broader stock of liquid assets in the exercise even though cash alone was insufficient to cover all stressed outflows. The result depends on the tested population, scope, allowed actions, and assumed ability to liquidate instruments; it should be interpreted as evidence for robust stress modeling rather than a universal assurance about insurer liquidity.
Financing structures and the erosion of apparent liquidity
Repo transactions, securities lending, funding agreements, collateralized borrowing, and derivatives change the time structure of the insurer's obligations and the availability of assets. A financing instrument may support an otherwise attractive longer-term investment strategy while imposing repayment or collateral needs on a shorter clock. A Federal Reserve discussion of insurer funding reports continuing use of nontraditional life insurer liabilities, including funding-agreement-backed securities, Federal Home Loan Bank advances, and cash associated with securities lending and repos. The Federal Reserve identifies associated maturity and liquidity questions. These exposures should enter the financing cash-flow schedule and collateral register rather than being treated as costless additions to investable capital.
The economically relevant liquidity buffer has a dynamic value. It can reduce expected investment income compared with a higher-yielding asset but lowers the probability that unfavorable market conditions require forced liquidation. Evaluating it solely as a return drag disregards its option-like economic service: readiness to meet contingent claims without disrupting the remaining asset strategy. Conversely, an excessively large cash balance can create reinvestment risk and reduce long-horizon liability matching. The correct buffer follows the profile of liabilities, funding, and scenario-dependent transaction capacity.
Transaction, valuation, and management costs
A gross yield comparison omits transaction spreads, bid-ask losses, custody, management fees, derivatives margin finance, taxes, legal structuring, credit surveillance, valuation services, capital opportunity costs, and losses arising from a forced exit. Alternative investment fees may be paid at fund, vehicle, and underlying-asset levels. Credit enhancement and securitization expenses may alter the investor's realized return. Execution and operational demands matter particularly when investment portfolios are managed across several legal entities or outsourced platforms.
The net economic contribution of an asset is its expected scenario-dependent benefit after relevant expenses, credit and market losses, funding, and the burden it places on capital and liquidity. A higher contractual coupon can be associated with lower realizable net benefit when those costs are counted. A rigorous research record should distinguish quoted yield, accounting interest recognition, modeled expected total return, and realized cash yield. These quantities have different definitions and should never be substituted for one another without reconciliation.
Reliability of data and models
Investment data quality determines what the insurer can credibly optimize. Public fixed-income holdings may have observable traded prices but require careful modeling of term structure, spread, default and optionality. Private loans and real assets often rely on appraisals, manager valuations, comparable instruments, or discounted cash-flow assumptions. Valuation lags can understate measured volatility. Structured investments need cash-flow waterfalls and borrower-level data. Derivative valuation requires collateral agreements, basis risks, and closeout assumptions. Liability data require contract-level exposures, policy options and actuarially credible stochastic experience assumptions.
The model inventory should record data source, observation date, valuation hierarchy, currency, legal ownership, model owner, independent challenge, back-testing results, and material changes. For tail risk, the model should distinguish a lack of observed extreme events from affirmative evidence that extreme loss is unlikely. Under correlated stress, a model calibrated to benign historical observations can misstate both the loss distribution and the depth of secondary markets. An investment committee's uncertainty budget should be expressed in explicit sensitivity limits and investment constraints where robust probabilities are unavailable.
Methodological verification requirements
Reproducible research requires that empirical numbers identify the source population, reporting date, accounting or regulatory valuation basis, denominator, and any overlap across asset categories. The NAIC's U.S. year-end 2025 report uses a book/adjusted carrying-value basis, while EIOPA's Q2 2026 indicators use the Solvency II reporting population and specified exposure indicators. These figures are informative but are not interchangeable observations in a harmonized panel. A valid international empirical comparison would need definitions aligned on holdings, consolidation, derivatives, funds, valuation, time and insurance line. Until such reconciliation is possible, this monograph uses the datasets descriptively.
The numerical demonstrations in this work can be replicated with arithmetic and the stated assumptions. The four-sleeve example sums to 100%, has gross return 4.27%, and yields a linear hypothetical capital proxy of 7.45 units. The alternative-allocation sensitivity adds 0.22 percentage points of expected return and 1.70 proxy units of capital. The duration example uses first-order price changes. The five-day liquidity example uses 12 units of required payments, 4 units of cash, and specified bond sales after a 5% haircut. None estimates a real insurer's capital requirement, expected return, or liquidity coverage, and no inference of statistical significance is made.
Optimization and Governance — Integrated Interpretation and Implications
The feasible investment mandate
An investment mandate converts a balance-sheet strategy into implementable controls. It specifies permitted instruments, issuer and ultimate-risk limits, target and tolerance weights, currency boundaries, duration and key-rate ranges, credit-quality standards, maximum illiquidity, derivatives permissions, collateral rules, monitoring frequency, escalation triggers, authorized counterparties, and valuation procedures. The mandate is coherent only when its limits are consistent with liability forecasts, capital policy, and the actual organization of investment decisions. A mandate that calls for both low risk and high yield without specifying the relevant risk metric cannot be consistently optimized or governed.
The prudent-person approach in Solvency II Article 132 is a useful institutional expression of this requirement: asset risks must be identifiable and manageable, and the whole portfolio must meet security, quality, liquidity and profitability considerations relative to obligations. An insurer may choose a more conservative internal policy than a regulatory maximum. Those internal limits are legitimate responses to uncertainty in liabilities, capital access, or valuation rather than evidence of defective mathematical optimization.
The decision sequence
A defensible sequence begins by inventorying insurance obligations and forecasting liability cash flows with stochastic policyholder and claims behavior where material. The enterprise then determines the economic surplus sensitivities, funding requirements, regulatory restrictions, and target capital buffers. It establishes an investable universe whose risks can be understood, valued, and monitored. Only after these steps does it optimize weights or individual positions under return assumptions, scenario constraints, transaction costs and execution limits. The proposed allocation is reviewed against adverse scenarios, model uncertainty, and operational capacity, and is formally approved within the insurer's governance framework.
This order has practical consequences. If an optimizer first selects instruments from a broad universe and later adds rough capital or liquidity adjustments, the solution may depend on assets that should never have been eligible. Pre-trade constraints on data transparency, legal rights, and settlement feasibility can be more reliable than imposing a large generic risk penalty in the objective. Some assets may be attractive in isolation but unsuitable because an insurer cannot price their optionality, trace underlying credit, or monetize them within its required liquidity horizon. Eligibility is a prior research conclusion that should be supported by specific evidence.
Dynamic control, rebalancing, and management actions
A strategic portfolio is monitored as liabilities run off, new business arrives, asset cash flows mature, economic conditions change and capital becomes more or less abundant. Rebalancing policies should specify the circumstances under which exposures are restored, hedges modified, or asset sales delayed. The insurer should evaluate the transaction costs and procyclicality of rebalancing under stress. A mechanical target-weight rule can force sales after market declines or require expensive purchases when market liquidity is scarce. A liability-relative strategy can instead specify permitted bands for surplus duration, cash coverage, credit exposures, and capital headroom.
Management actions must be legally and operationally feasible in the scenarios where they are credited. Selling assets, changing credited rates, drawing a funding line, adjusting hedges, using reinsurance, or reducing dividends can have material effects on the enterprise. Each action requires a documented trigger, authority, time to execute, market assumptions and second-order balance-sheet effects. EIOPA's stress-testing approach illustrates why comparing fixed-balance-sheet and management-action variants helps reveal what portion of projected resilience depends on intervention rather than the starting portfolio.
Governance of outsourcing and affiliated management
External asset management can broaden investment expertise while creating information and principal-agent problems. The insurer still has responsibility for understanding the risks it carries. Mandates should specify reporting of ultimate exposures, pricing and valuation methods, conflicts of interest, dealing with affiliated entities, capital and liquidity implications, collateral practices, and the circumstances under which the insurer may suspend new investment or redeem an allocation. The insurer must be able to assess whether investment fees, origination incentives or reported performance create pressure to increase risk beyond the approved liability-relative mandate.
The documented expansion of affiliated private-credit investment channels in the Federal Reserve study by Carlino and colleagues reinforces the importance of examining exposures through the economic chain. The portfolio should not be deemed diversified merely because assets sit in several legal investment vehicles managed by related parties. Ultimate borrower, collateral, asset manager, underwriting policy, leverage and funding channels should be consolidated for appropriate risk analysis, with clearly recorded gaps in look-through information.
Monitoring, escalation, and independent review
The insurer needs an investment-risk dashboard that reports asset and liability market-value sensitivities, surplus movements, capital headroom, dated liquidity resources, collateral availability, issuer and factor concentrations, valuation uncertainty, realized losses, and deviations from the mandate. The monitoring schedule should reflect risk speed. Derivatives margin and liquid asset settlement can require daily or intraday attention; private asset valuations and underwriting sensitivities may be updated less frequently but should be challenged with event-driven indicators. Material model changes should be independently reviewed, and stress-test assumptions should be approved and preserved for audit.
An effective internal review is structured around reproducible questions. Which contractual obligations require funding under the stated scenario? Which holdings produce cash, and when? What loss and capital effect follows the same scenario? Which risks are common across apparently distinct positions? What data are direct observations, and which are model-generated? Which transactions can be executed at the assumed time and price? What remains after an intervention? The answers convert an investment policy from a statement of general objectives into an auditable economic and operational design.
Comparative interpretation of available evidence
Official data suggest a persistent role for fixed income alongside greater scrutiny of complex credit and liquidity. The NAIC's 2025 aggregate investment composition documents bonds as the largest asset class in U.S. insurers' investment accounts. Its separate private credit analysis illustrates the need to distinguish long-standing private placements from opaque or more complex holdings. EIOPA's October 2026 risk dashboard indicates heightened market-risk attention and continuing supervisory monitoring of private credit quality within the European reporting population. The FSB's liquidity recommendations show why margins and collateral remain part of investment resilience even when an insurer's central objective is long-term investment.
These observations support a limited but substantive conclusion: asset composition can remain relatively stable at the aggregate level while the complexity, financing structure and concentration of its underlying exposures change. A stability claim based solely on category weights is therefore weaker than a stability claim supported by look-through, cash-flow and capital-risk evidence. The conclusion follows from the interaction of the source definitions and balance-sheet mechanics; it is an analytical interpretation, not a claim that the cited studies establish a universal causal law.
Boundaries of the synthesis
This monograph offers an integrated analytical method and documented comparison of primary sources. It does not claim to estimate a globally representative efficient frontier, calibrate insurers' true risk tolerance, identify an optimal percentage allocation for a real institution, or prove the predictive superiority of a particular portfolio model. The official data cover different jurisdictions, populations, accounting bases, and dates. Illustrative scenarios do not replicate supervisory capital engines or proprietary portfolio systems. The validity of an applied allocation requires its own claims, liabilities, asset-level data, governance evidence and jurisdiction-specific legal review. These restrictions define the scope of the result and the work required for empirical application.
Conclusion
Insurer investment portfolio construction is the selection and control of asset exposures within an institution that has issued contingent promises and is subject to dated cash requirements, capital standards, accounting measurement, and supervisory duties. The portfolio problem is resolved by combining the financial logic of diversification with actuarial matching and enterprise-wide constraints. Expected return and covariance explain some of the economics of assets; surplus sensitivities, optionality, credit risk, cash-flow availability, and legal and regulatory feasibility determine whether an allocation can support insurance obligations.
The first central finding is that portfolio quality is liability-relative. A short-duration portfolio may have low stand-alone price volatility and substantial mismatch with long-dated guarantees. An apparently risky long-duration asset can supply a meaningful hedge while introducing credit or funding exposure. Duration and convexity should therefore be expressed in comparable asset and liability values, extended to key-rate and option sensitivities, and tested against cash requirements.
The second finding is that profitability, solvency and liquidity are distinct dimensions. A return-enhancing asset can consume disproportionate capital, become difficult to sell, or increase losses under a common stress. The hypothetical examples make these interactions auditable: a 0.22 percentage-point modeled return improvement can coincide with 1.70 additional units of illustrative capital proxy; a substantial bond holding can fail to cover a five-day cash need when settlement or encumbrance is considered. Their importance lies in the exposed mechanisms, rather than in the selected illustrative parameters.
The third finding concerns supervision and governance. The IAIS establishes a global conceptual baseline for investment and enterprise risk management; Solvency II, U.S. state law and RBC, and applicable group standards impose different legal architectures. IFRS 9 and IFRS 17 introduce a separate financial-reporting perspective. These systems must be reconciled by jurisdiction, measurement and date. The November 2026 analysis preserves the prospective status of the January 2027 EU amendments and distinguishes international standards from directly applicable local law.
The fourth finding follows from contemporary supervisory evidence. Fixed income remains central to insurer investing, while private-credit complexity, related-party investment channels, collateral needs, and nontraditional financing complicate the interpretation of apparently familiar asset classes. Portfolio reporting that stops at broad asset categories cannot fully describe these exposures. Look-through information, independent valuation, scenario-based liquidity analysis, and documented investment mandates are therefore necessary elements of sound construction.
The resulting scientific conclusion is that an insurer's investable portfolio is the set of strategies that can be supported simultaneously by its liability cash flows, capital capacity, operational liquidity and governance. A portfolio that maximizes one dimension while breaching another does not solve the institutional investment problem. The analytical synthesis presented here organizes existing financial, actuarial and supervisory knowledge into one decision structure. It creates no separate named theory or new regulatory doctrine.
References
Board of Governors of the Federal Reserve System. 2026. "Funding Risks." Financial Stability Report, May 2026. Federal Reserve report.
Campbell, John Y., and Luis M. Viceira. 2002. Strategic Asset Allocation: Portfolio Choice for Long-Term Investors. Oxford: Oxford University Press. Publisher and DOI.
Campbell, John Y., and Luis M. Viceira. 2002. "Who Should Buy Long-Term Bonds?" In Strategic Asset Allocation: Portfolio Choice for Long-Term Investors, 48-87. Oxford University Press. Chapter DOI.
Carlino, Sydney, Nathan Foley-Fisher, Nathan Heinrich, and Stephane Verani. 2025. "Life Insurers' Role in the Intermediation Chain of Public and Private Credit to Risky Firms." FEDS Notes, March 21. Board of Governors of the Federal Reserve System. Federal Reserve research.
European Insurance and Occupational Pensions Authority (EIOPA). 2024. "Insurance Stress Test 2024." European Insurance and Occupational Pensions Authority. Official study and findings.
European Insurance and Occupational Pensions Authority (EIOPA). 2026. "Insurance Risk Dashboard: October 2026." EIOPA-BoS-26-495. European Insurance and Occupational Pensions Authority. Official dashboard.
European Insurance and Occupational Pensions Authority (EIOPA). 2026. "Solvency II." Regulatory overview, including temporal status of Directive (EU) 2025/2. Official framework overview.
European Union. 2009, as amended. "Directive 2009/138/EC: Article 105, Calculation of the Basic Solvency Capital Requirement." Solvency II Single Rulebook hosted by EIOPA. Article 105 text.
European Union. 2009, as amended. "Directive 2009/138/EC: Article 132, Prudent Person Principle." Solvency II Single Rulebook hosted by EIOPA. Article 132 text.
Financial Stability Board (FSB). 2024. "Liquidity Preparedness for Margin and Collateral Calls." Policy recommendations, December 10. FSB publication.
Fisher, Lawrence, and Roman L. Weil. 1971. "Coping with the Risk of Interest-Rate Fluctuations: Returns to Bondholders from Naive and Optimal Strategies." Journal of Business 44 (4): 408-431. DOI.
IFRS Foundation. 2026 access. "IFRS 17 Insurance Contracts." Issued Standards. International Accounting Standards Board. Standard overview.
IFRS Foundation. 2026 access. "IFRS 9 Financial Instruments." Issued Standards. International Accounting Standards Board. Standard overview.
International Association of Insurance Supervisors (IAIS). 2024. "Insurance Capital Standard." Adopted in 2024. Official ICS overview.
International Association of Insurance Supervisors (IAIS). 2024. Insurance Core Principles and Common Framework for the Supervision of Internationally Active Insurance Groups. December 2024, especially ICP 15 and ICP 16. Basel: IAIS. Official IAIS standard.
International Association of Insurance Supervisors (IAIS). 2025-2026 access. "Global Insurance Market Report (GIMAR)." Financial Stability publication series. Official GIMAR catalog.
Markowitz, Harry. 1952. "Portfolio Selection." Journal of Finance 7 (1): 77-91. DOI.
National Association of Insurance Commissioners (NAIC). 2026 access. "Model Laws." Including Investments of Insurers Model Acts 280 and 283. Official model-law catalog.
National Association of Insurance Commissioners (NAIC). 2026. "Private Credit and Insurance: A Data-Driven View from State Regulators." NAIC Capital Markets Bureau, analysis of year-end 2025 data. NAIC report landing page.
National Association of Insurance Commissioners (NAIC). 2026. "Risk-Based Capital." Insurance Topics, updated June 30. NAIC supervisory explanation.
Redington, Frank M. 1952. "Review of the Principles of Life-office Valuations." Journal of the Institute of Actuaries 78 (3): 286-340. DOI.
Rockafellar, R. Tyrrell, and Stanislav Uryasev. 2000. "Optimization of Conditional Value-at-Risk." Journal of Risk 2 (3): 21-41. DOI.
Wong, Michele, and Jennifer Vento. 2026. "U.S. Insurance Industry's Cash and Invested Assets Grow to $9.6 Trillion as of Year-End 2025." NAIC Capital Markets Bureau Special Report, May. Official asset-mix report.