RESEARCH DOSSIER
Cash Flow Matching and Reinvestment Risk
Abstract. Cash-flow matching is a financial and actuarial technique for coordinating the amounts and payment dates of assets with future obligations. Its apparent simplicity conceals a demanding economic problem: insurance liabilities are frequently contingent, investment cash flows can be delayed or accelerated, and the short-term investment rates available after a receipt are unknown when a portfolio is constructed. This monograph examines the conditions under which dedicated fixed-income portfolios can fund long-term insurance obligations and the mechanisms by which coupon reinvestment, principal prepayment, calls, defaults, curve reshaping, policyholder behavior, and adverse liability-timing developments undermine a baseline cash-flow match. The analysis integrates Redington's actuarial immunization argument, Fisher and Weil's bond-return research, liability-driven fixed-income methods, primary insurance supervision, actuarial standards, and transparent numerical calculations. It derives period-by-period funding constraints, distinguishes exact cash-flow dedication from duration immunization, characterizes the irreducible timing choices in investment portfolios, and compares insurance regimes in the United States, European Union, and United Kingdom as of November 15, 2026. Four reproducible numerical investigations demonstrate the value of dated payments, the dependency of coupon terminal wealth on reinvestment, the narrowness of duration matching, and the negative effect of an early call. The central finding is that matching quality must be judged against contingent settlement needs and jointly stressed asset and liability paths, with regulatory eligibility, credit performance, and operational liquidity treated as separate constraints. The contribution is a critical, original synthesis of established financial and actuarial techniques and explicit pedagogical derivations; no new theory or doctrine is claimed.
Keywords: cash-flow matching; cash-flow dedication; reinvestment risk; coupon risk; prepayment risk; call risk; asset-liability management; insurance finance; fixed-income immunization; Redington; Fisher and Weil; duration; convexity; key-rate sensitivity; asset adequacy; contingent claims; stochastic cash-flow testing; annuities; matching adjustment; Solvency II; Solvency UK; actuarial practice.
Introduction
Insurance converts premiums and invested resources into promises of future payments. The funding problem begins when the dates on which money is received differ from the dates on which money must be paid. Fixed bonds can distribute cash through coupons and scheduled principal; life insurers may owe survival-contingent annuity benefits, guaranteed surrender amounts, maturity benefits, and death claims whose dates are influenced by mortality and policyholder behavior. A precisely projected asset receipt therefore has financial meaning only in relation to the amount, timing, contractual status, and availability of the corresponding liability payment.
Cash-flow matching, in its strict deterministic form, is the construction of asset payments that satisfy specified obligations at specified dates without dependence on subsequent sale prices or uncertain reinvestment rates. A weaker strategy can achieve economic present-value or duration matching while leaving funding gaps between dates. A third strategy can hold assets expected to pay the right amounts but containing embedded borrower or issuer options that change the timing of payment when rates change. The distinction determines whether an insurer possesses a genuine dated funding arrangement or a forecast that depends on market conditions.
International supervisory practice recognizes the distinction. The International Association of Insurance Supervisors, in ICP 16.5 on asset-liability management, requires an explicit ALM policy that considers whether cash inflows meet liability outflows and how both sides respond to scenarios. Its guidance treats matching as effective management of mismatches rather than an abstract requirement that every payment be identical. That institutional judgment is important because insurance portfolios must also meet claims promptly, absorb credit losses, maintain regulatory capital, preserve investment diversification, and operate within real trading and accounting systems.
The principal research question is when a portfolio with a satisfactory initial cash-flow schedule continues to fund liabilities after financial markets, counterparties, and insured populations move away from baseline assumptions. Three supporting questions guide the inquiry. How should a genuine dated match be distinguished from duration immunization? When do coupons, calls, amortization, and other early principal receipts expose insurance funding to reinvestment rates? Which decision and supervisory structures translate projected cash flows into credible future settlement capacity?
The dossier addresses insurance assets and liabilities globally, with detailed references to the EU, UK, and US arrangements for which verified source materials are available. It is a theoretical and comparative scientific analysis. Its numerical scenarios are constructed illustrations, not observations about named insurers, market forecasts, portfolio instructions, or personal financial advice.
Research Method, Evidence, and Scope
The study combines textual interpretation of original actuarial research, formal cash-flow mathematics, deterministic counterfactual calculations, and comparative reading of primary institutional rules. Historical claims concerning immunization rely on the original bibliographic record and accessible text of Redington's 1952 valuation paper, together with the indexed metadata for Fisher and Weil (1971). Fisher and Weil's full subscription-restricted text was not represented as fully independently reviewed; its historical significance is also cross-checked against the later fixed-income literature. Technical contemporary definitions are supported by the CFA Institute's bond-return treatment and its liability-driven strategies discussion.
Primary supervisory and professional instruments establish obligations within their stated jurisdictions and effective periods. The IAIS supplies internationally relevant supervisory principles. EIOPA's published Solvency II rulebook supplies EU legal provisions. The UK Prudential Regulation Authority supplies UK supervisory expectations. US actuarial standards govern the professional work to which their scope clauses apply and do not substitute for individual state laws or the NAIC's adopted valuation requirements. IFRS standards govern accounting classification and measurement for entities to which they apply, rather than guaranteeing economic asset-liability equivalence.
The analysis deliberately uses simple annual dates and specified rates for numerical transparency. These assumptions isolate contractual cash-flow logic from day-count conventions, taxation, transaction costs, collateral constraints, bid-ask spreads, stochastic mortality, and multi-curve discounting. The limitations are stated at the point of each calculation. A realistic insurer model must restore those dimensions. All illustrative figures were independently recalculated from the displayed formulas using arithmetic code on November 15, 2026.
The information cutoff is November 15, 2026. Prospective regulations are identified by their future application dates. A direct URL in the bibliography is a record of the exact source checked; a linked phrase in the chapters points to the proposition for which that source provides support. Bibliographic completeness and claim-to-source consistency are kept separate from legal or empirical certainty.
Cash Flow and Maturity Maps
The map of dated obligations
Let the ordered settlement dates be , measured from a valuation instant. Let represent benefits, claims, expenses, and related net liability payments at date under state . The state can encode survival, surrender, claim inflation, litigation timing, interest rates, and reinsurance performance. Let denote cash made available by the assigned investments at the same date. In a deterministic projection the state index is omitted.
For a known net liability schedule, a sufficient exact-flow condition is at every relevant date, provided investments perform and every receipt is legally and operationally usable for settlement. Where surplus receipts can be stored safely, the weaker cumulative condition is also relevant: , where
Here is the cash account after the period's settlement and is the actually realizable return on the preceding surplus cash. If is negative, payment requires another resource: a sale, borrowing, additional cash contribution, derivative settlement, or an explicitly maintained buffer. A nominally positive aggregate present value cannot substitute for this date-specific condition.
The financing test changes when an early surplus is invested in an instrument maturing later than the next obligation, or when the assumed reinvestment rate is itself random. A strategy satisfying the cumulative condition at a forecast rate can still fail under a falling-rate stress. The economic distinction is especially important for insurers because policy benefits generally must be paid on their due dates, while excess asset value may be tied up in securities unavailable without sale or financing.
Present value, currency, and discounting
With deterministic cash flows and a compatible discount function , define
Equality is one scalar equation. Exact matching requires the relevant vector of payment dates and amounts to align. Two positive cash-flow portfolios can share a present value while producing entirely different funding patterns. Currency adds an additional coordinate: a cash flow in one currency is not an unrestricted substitute for a contractual benefit payable in another without a settled foreign-exchange arrangement.
Consider liabilities of 60, 100, and 120 units at years one, two, and three. At a flat annual effective discount rate of 4%, their present value is
An exact zero-coupon ladder paying 60, 100, and 120 at the respective dates matches the assumed nominal dates and avoids interim coupon reinvestment on those matched pieces. An alternative portfolio costing the same 256.827492 may concentrate all cash at year three; it would then require a funding mechanism for the obligations in years one and two. The equality of initial purchase cost is not evidence that early claims can be met.
Instruments and dated payment engineering
A zero-coupon instrument places its principal and investment return at one maturity date. The US Treasury's STRIPS structure separates eligible bond principal and interest into independently tradable zero-coupon pieces. A strip maturing on a specified date can support a dated nominal liability without coupons to reinvest before maturity. This illustrates the financial mechanism, while issuer credit, denomination, taxes, currencies, settlement calendars, market depth, and regulatory treatment remain relevant.
Coupon-bearing sovereign and corporate securities provide multiple payments. Amortizing loans provide declining principal and interest streams. Mortgage-backed securities distribute principal according to borrower repayment and transaction structures. Floating-rate securities reset their interest; inflation-linked securities vary with an index; callable securities grant their issuers specified redemption options. Thus, identifying an asset as a bond says little by itself about whether its payment schedule is fixed under the risk factors that matter to an insurer.
The historical actuarial distinction
In Redington's analysis of life-office valuations, matching is the broader actuarial objective and immunization concerns protection against specified interest-rate changes. The later study by Fellows (1981) on pension matching illustrates how actuarial liability patterns complicate any single duration description. Historical immunization arguments depend on how liabilities are valued; they do not make the projected timing of contingent payments contractually certain.
The foundational insight remains powerful: the investment problem is defined by liabilities. However, a present-value sensitivity result should not be interpreted as a proof that future benefits can be paid in cash. Both the 1952 actuarial framework and subsequent bond research emerged before the current complexity of many embedded-option, collateral, and market-linked insurance structures. Their central derivations can be retained while their simplifying assumptions are made explicit.
Dedicated Portfolios
Defining dedication
A dedicated portfolio is a specified collection of assets assigned to finance a liability schedule. At the strict end, exact dated payments deliver the required money at the same dates and in the same currency. At the flexible end, surplus inflows are held, reinvested, or exchanged to fund later dates. The latter may be efficient, but it embeds additional return, credit, and operational assumptions.
A deterministic cash-flow dedication problem can be expressed with , the quantity of bond ; , its initial purchase price; and , its contractual payment at date . For hard annual matching one solves
This is a linear programming statement only while cash flows and costs are deterministic, divisible, and appropriately specified. In reality purchase lots, issuer limits, asset eligibility, collateral, fees, default, and early redemption can turn the problem into a mixed-integer or nonlinear optimization. A model can minimize initial cost yet leave the institution exposed to credit losses, concentrated issuers, or unexpected early claims if its constraints omit those risks.
An overfunded annual constraint can be economically wasteful when early surpluses could be carried forward. To model reinvestment explicitly, introduce surplus accounts and the balance equation from the preceding chapter. This formulation can lower portfolio cost, but its feasibility now depends on available reinvestment instruments and rates. The distinction between hard dedication and reinvestment-dependent dedication must be visible in investment committee materials.
Why duration is not a dated match
For a fixed cash-flow bond valued at yield with annual compounding, Macaulay duration is
Modified duration is under the same yield convention. It approximates the proportional change in price for a small parallel change in yield. If the present values of assets and liabilities agree and their dollar durations agree, their first-order parallel rate sensitivities coincide locally. This is the familiar valuation basis of immunization; it is an approximation evaluated at a specified market point.
The CFA Institute's liability-driven treatment explicitly treats multiple-liability immunization through value-weighted sensitivity and rebalancing. The result can fail under nonparallel curve shifts, credit spread changes, large moves that activate options, or a shortage of cash at an intermediate date. Duration equality is not a schedule of actual deposits into the claims account.
Reproducible barbell calculation
Assume one unconditional liability of 100 at year three and an annual effective flat discount yield of 4%. The present value is . Construct a portfolio with a year-one zero and a year-five zero, each contributing half this initial present value. The resulting payments are approximately 46.227811 at year one and 54.080000 at year five. The assets have duration three years at the initial yield: . Their initial present value equals the liability's.
Under a uniform 50-basis-point increase to 4.5%, the portfolio value is 87.633692 and the liability value is 87.629660, producing a small positive surplus of 0.004032. With a uniform one-percentage-point fall to 3%, the portfolio value is 91.531253 and the liability is 91.514166, a positive surplus of 0.017087. The difference reflects the greater convexity of this two-zero barbell relative to the single year-three obligation in the assumed frictionless setup. These results are local or scenario-specific; they are not evidence of a strategy that cannot lose money.
In a deliberately nonparallel thought experiment, discount the year-one asset payment at 5%, the year-five payment at 3%, and leave the year-three liability at 4%. The asset present value becomes 90.676369 and the liability remains 88.899636, for a 1.776734 surplus. Reversing the twist would materially alter the outcome. More fundamentally, the asset portfolio pays at years one and five while the liability is due at year three. The first receipt is insufficient to fund the liability unless it can be invested at an exceptionally high rate or additional cash is supplied, and the year-five receipt arrives too late. Thus a seemingly attractive valuation hedge need not provide year-three payment capacity.
This example isolates three different facts: equality of initial value, equality of a duration statistic, and availability of cash at the liability date. Only the third resolves the immediate settlement problem. Reinvestment and interim liquidity assumptions bridge the difference.
Yield-curve shapes and basis exposures
Even noncallable bond cash flows can be exposed to the way the discount curve changes. A parallel shift assumption moves all maturities together; real yield curves can steepen, flatten, or experience localized shocks. Key-rate durations describe sensitivities to changes at designated maturities, while key-rate buckets can obscure shape variation within each bucket. Corporate bonds add credit-spread risk; collateralized assets introduce price and timing sensitivity; foreign-currency bonds add exchange risk.
The Basel Committee's interest-rate risk framework classifies gap, basis, and option risk in banking books. Those banking requirements are not insurance rules, but their analytical distinctions are transferable when an insurer models repricing and options. Such cross-sector borrowing must preserve the institutional difference between bank earnings models and the claims-funding obligations of insurers.
Coupon Reinvestment
Why a promised coupon is not a promised horizon return
A fixed coupon is a contractual amount payable on a specified date, subject to issuer performance. Its compounded contribution to a later funding horizon depends on rates available after receipt. A bond's yield to maturity is therefore a compound-return statement conditional on a reinvestment assumption and holding conditions, as the CFA Institute's interest-rate analysis makes clear. The coupon itself may be contractually fixed even when the return earned between its receipt and the liability date is uncertain.
For coupon receipts that are reinvested at annual rates until horizon , their horizon contribution is
The final coupon has no remaining reinvestment period; an early coupon has more periods at risk. When is stochastic, this sum is a random variable even for a nondefaulting fixed-rate bond. The return distribution cannot be inferred from its opening yield alone.
Five-year coupon case
Consider a five-year, par-100, annual 5% coupon bond without options or default, held until maturity. If every coupon can be reinvested at 5%, the year-five wealth from the 100 principal and five coupons is
If the reinvestment rate is 1%, terminal wealth is
With zero reinvestment return the same contractual bond produces terminal cash wealth of 125.000000. The difference between 127.628156 and 125.505025, equal to 2.123131, is caused solely by the stipulated reinvestment path. It should not be described as a loss of contractual coupons: all five coupons and the 100 principal are still received.
A year-five liability of 127.628156 is exactly funded by the first return assumption, but under the 1% alternative the available terminal wealth falls short by 2.123131. In this isolated example, the shortfall arises without issuer default, a fall in asset coupon, or an increase in the contractual liability. That is the defining financial mechanism of coupon reinvestment risk.
Reinvestment risk and price risk
Market-rate changes affect the resale price of a bond and the investment return on coupons in opposing directions under standard fixed-cash-flow assumptions. When yields rise, the price of an existing fixed bond falls, but coupons can be reinvested at more favorable rates. When yields fall, its sale price tends to rise, but coupons are reinvested at less favorable rates. These effects can offset at an appropriately specified investment horizon. The classical literature associated with Fisher and Weil's 1971 study develops the relevance of term structure and holding-horizon returns to immunization.
The offset requires a common valuation convention, appropriate holding horizon, absence or modeling of options and defaults, and a consistent yield-shock assumption. Insurance liabilities commonly span many horizons. A strategy immunized to one horizon can remain exposed at another. The common statement that price and reinvestment risks balance at Macaulay duration concerns a narrow stylized configuration, not guaranteed equality of future claims coverage under all economic states.
Maturity-wall reinvestment
Reinvestment also arises when an asset's principal matures before a corresponding liability. An insurer funding a year-ten obligation with bonds maturing at year five must replace future investment income or invest the principal for the remaining term. If the expected forward returns are unavailable at year five, a positive opening spread or an attractive asset book yield may prove insufficient to cover the original liability economics.
The reverse mismatch matters as well: bonds maturing after a suddenly accelerated liability payment can be sold, pledged, or hedged, but doing so introduces market value and liquidity risk. Reinvestment risk and funding-liquidity risk are therefore related through the timing map; they are not interchangeable labels. In the first, early cash must earn an unknown return. In the second, required cash may have to be obtained before investments pay.
Credit and inflation interaction
An extra promised yield on a corporate security can represent compensation for default, downgrade, liquidity, and spread exposure rather than a freely available funding surplus. If credit deterioration reduces expected payments, reinvesting coupons more favorably cannot necessarily restore the missing principal. Similarly, a fixed nominal coupon ladder may match nominal liability payments while failing against inflation-indexed benefits or expenses. The asset and liability valuation currencies, indexes, and credit conventions must be consistent before a yield margin is called protective.
Regulatory accounting adds distinct measurement objectives. Under the IFRS Foundation's IFRS 9 classification framework, an asset's business model and contractual cash-flow characteristics affect its accounting category. Amortized cost, fair value through other comprehensive income, and fair value through profit or loss do not imply identical economic risk exposures. The insurer can face reinvestment risk even when a market-value decline does not flow immediately through reported earnings.
Prepayment and Call Risk
State-dependent early principal
An asset scheduled to repay principal in a distant year may repay earlier. The issuer of a callable bond can exercise its contractual call; borrowers can refinance loans; a mortgage pool can return principal faster through refinancing and turnover. Early receipt creates a new dated cash balance that must be invested or held until the corresponding liability is due. The SEC's definition of prepayment risk identifies the mechanism of early repayment followed by reinvestment at lower prevailing rates.
The exercise behavior is often rate-dependent. Falling rates increase the economic incentive to refinance expensive borrowing and can shorten mortgage cash-flow lives, although actual prepayments also depend on housing turnover, costs, underwriting, legal arrangements, and borrower circumstances. The SEC's mortgage-backed securities explanation warns that mortgage prepayments can return principal when reinvestment opportunities are unattractive. Contractual, modeled, and realized payment schedules must therefore be treated as distinct objects.
The interaction is asymmetric. In a low-rate state, the investor often receives principal earlier and reinvests at a lower rate. In a high-rate state, repayment may slow and asset maturity can extend while liability dates remain fixed or change differently. The investor is exposed to both contraction and extension. A baseline expected weighted-average life is not an enforceable maturity date.
Five-year callable-bond case
Consider a hypothetical five-year par-100 bond paying a 6-unit annual coupon, with an issuer option to redeem the principal at 100 immediately after the year-two coupon. Assume no credit loss, no option premium adjustment, and a constant 2% annual reinvestment rate for cash received before the year-five liability horizon. If the bond remains outstanding through year five, year-five accumulated wealth is
If the issuer exercises its call after year two, the investor receives coupons of 6 at years one and two and principal of 100 at year two. The corresponding year-five accumulated wealth is
The early-call scenario generates 12.241600 less terminal wealth than the noncalled scenario. Both computations include all contractually appropriate cash receipts under their respective scenarios and reinvest them under the same assumed rate. The arithmetic does not measure the fair option price at purchase, expected loss, or investment return on the initial transaction. Those quantities would require the call option's ex ante valuation and possibly a different issue price. It demonstrates precisely why contractual principal acceleration can obstruct a later liability-matching objective.
A five-year obligation of 131.224241 would be covered in the noncall scenario and underfunded in the call scenario. The key is not an assumed deterioration in the issuer's willingness to pay; the issuer has exercised a contractually permitted option. A security can be performing exactly as written while defeating a funding plan based on the uncalled cash-flow schedule.
Option-adjusted cash flows
Cash-flow projections should distinguish legal call terms, contractual amortization, prepayment assumptions, and option exercise rules. When embedded options are material, the asset cash flows become functions of an interest-rate path and other relevant state variables. A possible model has , where describes interest-rate history, credit or spread conditions, and borrower or issuer behavior. For complex securities, option-adjusted spread valuation may provide one market-consistent perspective, while realized funding cash flows still require scenario analysis.
Different mortgage tranches redistribute, rather than universally eliminate, the timing exposure. A tranche designed to protect certain payments can depend on collateral structure and the behavior of other tranches. The financial comparison must examine legal waterfall terms and credit enhancement. A nominal yield advantage may be economically offset by exposure to early principal return, extension, or impaired collateral.
Insurance-specific correlation
Insurers may face a harmful combination in which assets return money early just as the cost of funding long-term guarantees rises. The liability side can also carry options: guaranteed minimum crediting rates, partial withdrawals, annuity choices, surrender rights, and dynamic lapses. Asset calls and liability withdrawals can respond to rates with different sensitivities or opposite timings. The correct stress question is how both sides behave in the same state, rather than applying an isolated fixed percentage shock to one side while freezing the other.
Investment committees therefore need to distinguish the contractual upper and lower bounds of an asset's payments from a single expected schedule. Forecast cash-flow maps should display conditional branches at possible call dates, refinancing shocks, and extension states. The willingness to pay a lower yield for truly fixed and suitable dated cash flows can be understood as purchasing reduced timing uncertainty, subject to pricing, credit, and market constraints.
Liability Timing Uncertainty
Survival-contingent payments
An annuity payment at year is due only if the relevant beneficiary remains entitled to it. Its expected value depends on mortality, survival, joint-life status, guarantees, indexation, and optional benefit elections. For a benefit amount payable upon survival, a simplified expected cash flow is , where is the probability of survival to time conditional on age . This expectation is useful for valuation and portfolio planning but is not an exact schedule for individual or aggregate settlements.
In a large portfolio, mortality pooling reduces some sampling volatility; systematic longevity improvement, changing socioeconomic composition, model error, and adverse selection can move many annuity payments together. A zero-coupon ladder matching mean expected annuity benefits will not pay any unexpected longevity surplus automatically. The insurer must assess the distribution of aggregate benefits and the capacity of investments and capital resources to meet adverse outcomes.
A death benefit reverses the contingency: earlier-than-expected mortality can accelerate a payment. Claims from disasters and litigation create their own timing concentrations. The common mathematical point is that expected cash flow can be a stable valuation input while realized cash flow remains stochastic and event-dependent. A portfolio should be tested on joint distributions rather than only matching expected payment dates.
Policyholder options and behavior
Lapse and surrender rates can depend on interest rates, policy guarantees, distribution conditions, taxation, changing household needs, and competition. When market yields rise above an existing contract's crediting rate, some policyholders may prefer alternative investments and surrender where contract terms permit. When guaranteed rates become relatively attractive, persistency may rise, prolonging guarantee exposure. Behavioral responses are neither constant across products nor fully determined by rates.
Cash-flow testing should therefore treat policyholder behavior as a model component. A deterministic assumption such as a constant annual surrender probability can be used as a baseline, but it is inadequate when material option incentives change in stress. Where behavior is uncertain, scenario designs can include rational exercise bounds, empirical estimates, and prudent adverse assumptions, together with sensitivity to model choice.
The US Actuarial Standards Board's current ASOP No. 7, revised December 2025 and effective for covered work on or after June 1, 2026, addresses life or health cash-flow analysis and the treatment of relevant cash-flow risks and options. Its scope is professional actuarial practice, not a universal investment statute. The standard supports the need for reasoned cash-flow assumptions and documentation when such analysis is performed.
Asset adequacy, reserves, and states
The US ASOP No. 22 provides guidance for opinions based on asset adequacy analysis when applicable law calls for them. Its discussion of moderately adverse conditions and methods illustrates why a current balance-sheet value may be insufficient evidence of ongoing reserve support. Testing should assess whether assets and related resources sustain benefits under adverse paths, including assumptions about future returns and behavior.
A state-contingent funding condition can be written
Feasibility under every imaginable state is generally economically impossible. Practical approaches specify scenario sets, probability measures, risk limits, capital buffers, and tolerance for residual risk. Stochastic analysis can estimate a distribution of shortages, yet a modeled tail probability is only as reliable as the data, options, correlation structures, and stress design. Historical correlations may change during crises precisely when an insurer needs cash.
Liquidity before terminal solvency
A balance sheet can show positive discounted surplus while an insurer has an imminent payment deficit. To settle a claim, a company needs cash or an accessible instrument of payment at the right time. The ability to sell a bond is affected by market depth, bid-ask spreads, settlement, collateral and encumbrances. Investment policies should therefore evaluate near-term net cash positions and contingent calls on resources separately from long-horizon expected returns.
An elementary example has liabilities of 40, 80, and 50 at years one through three, and asset receipts of 70, 20, and 80. If the year-one surplus of 30 earns 2% for one year, the available cash for year two is 30.6 plus the year-two asset receipt of 20, or 50.6. The year-two obligation of 80 leaves a funding deficit of 29.4 despite a later year-three asset payment of 80. A high terminal aggregate inflow does not cure a missing year-two receipt without bridging finance. This is why cumulative cash-flow tests matter independently of present value.
Reinsurance recoveries can create an additional settlement lag: the insurer may owe a policyholder before the reinsurer pays, and a failed or disputed recovery can enlarge the gap. Similar timing problems arise with derivatives that require collateral or margin before their hedging benefit is realized. These facts are incorporated here as funding constraints within the narrow cash-flow matching subject; comprehensive enterprise liquidity governance belongs to the wider ALM discipline.
The international supervisory setting
The EU Solvency II prudent person principle, Article 132, requires investment risks to be identifiable, measurable, monitored and managed, and emphasizes security, quality, liquidity, profitability, diversification, and the nature and duration of the insurer's liabilities. Its demands reach beyond a spreadsheet matching score. Asset assignment must be understood inside the overall prudent portfolio and regulatory treatment.
The EU Article 77b matching-adjustment conditions specify circumstances in which supervisory approval permits a matching adjustment for qualifying life obligations, including assigned portfolios and cash-flow replication requirements. Article 77c's calculation rules determine the adjustment and address the fundamental spread related to retained risks. The subordinate Article 53 calculation provision explains which assigned cash flows enter the calculation. An investment technique called cash-flow matching is not itself a legal entitlement to a matching adjustment; eligibility and supervisory authorization are distinct.
The EU review's effective-date boundary matters. EIOPA's July 2026 implementation announcement states that the reviewed framework and associated measures apply from January 30, 2027. As of this manuscript's November 15, 2026 cutoff, those future measures are not treated as already effective. The historically applicable instrument and any announced replacement must be separated in comparative work.
In the United Kingdom, the PRA's SS7/18 version history identifies the October 2025 version effective October 27, 2025, and a July 2026 future version effective December 31, 2026. A matching-adjustment portfolio is governed by UK-specific permission and eligibility rules; a predicted payment schedule cannot alone establish compliance. This jurisdictional account states effective dates without importing an inapplicable future supervisory threshold into the present.
Financial statements have another object. The IFRS Foundation's IFRS 17 overview concerns recognition and measurement of insurance contracts. Its financial reporting measurement is distinct from a settlement cash account or a jurisdiction-specific matching adjustment permission. Reconciling accounting and regulatory values can help explain risk management, but neither ledger substitutes for actual liquidity when a payment falls due.
Optimized Matching
A coherent decision problem
The most useful formulation begins with contractually described liabilities, a modeled distribution of contingent benefit payments, and an explicit universe of legally available assets. An insurer selects assets to meet dated funding requirements while controlling investment cost, residual mismatch, credit concentration, and the losses arising when the baseline ceases to hold. Objective functions may include discounted purchase cost, expected surplus, capital consumption, or a risk measure. These choices express management preferences within established optimization methods; no new scientific framework is needed to describe them.
For scenarios , let be asset 's payment at time , the liability payment, and the available cash balance. A basic scenario-constrained expression is
subject to
with for scenarios designated as hard funding requirements, and additional concentration, eligibility, cash-buffer, and trading constraints. The term is a specified residual risk functional; its definition and calibration must be disclosed in an actual implementation. The parameter expresses a decision trade-off, not an empirically universal optimal coefficient.
Where scenario sets are large, a manager may use chance constraints, expected shortfall, or robust bounds. Each has a limitation. A chance constraint can accept rare catastrophic deficits; expected shortfall depends on a sufficiently credible tail distribution; a robust set can become prohibitively costly if it includes mutually impossible extremes. Optimization should report the risk remaining under scenarios outside the design set, not merely the minimal objective value.
Priorities in portfolio engineering
The first engineering choice is whether dated liability payments are fixed enough to justify close nominal matching with zero-coupon or similarly predictable instruments. For a guaranteed annuity payment, a ladder may offer a transparent starting point. For survival-contingent annuities, matching expected payments leaves systematic longevity residuals. For benefits triggered by policyholder exercise, state-dependent liquidity becomes central. A manager may retain reserves, select short-liquidity assets, buy hedges, reinsure risks, or maintain lines of financing; each introduces legal, capital, price, or counterparty constraints.
A second choice concerns how much reinvestment dependency is acceptable. A low-coupon or zero-coupon security can reduce intermediate coupons but may have greater price sensitivity and different liquidity characteristics. A higher-coupon bond may provide earlier cash for near-term claims but create a larger unknown terminal accumulation for distant claims. The better choice is defined by the liability schedule, credit quality, and complete risk budget rather than the isolated stated yield.
A third choice is treatment of options. A callable or amortizing asset may be attractively priced on an option-adjusted basis but still be unsuitable for a liability that requires a particularly secure future payment at one date. Prepayment models and call-sensitive scenarios should be monitored separately from contractual maturity. Credit deterioration, asset replacement, and issuer concentration can be tested in the same scenario schedule.
Integrated stress design
A robust research-grade study should compare at least the following types of joint changes: parallel rate increases and decreases; curve steepening and flattening; short-rate and long-rate shocks at different dates; deteriorating credit performance and slower cash recovery; faster and slower asset principal return; accelerated surrenders; prolonged annuity survival; adverse claims inflation; and simultaneous liquidity demands from collateral or reinsurance settlement delays. The relevant set depends on insurance products and the assets actually held.
A useful output has a period-by-period table of asset cash inflows, contractual liability payments, expected and stressed contingent payments, surplus cash carried forward, funding deficits, and required intervention. Reports should also include an economic value sensitivity map so that favorable valuations are not mistaken for settlement cash. Rebalancing plans must specify transaction costs, execution capacity, and the circumstances under which permitted asset changes would destroy the original match.
The IAIS ALM supervisory guidance supports coordinating investment strategy, liability strategy, pricing, and capital under scenarios. The practical analytical consequence is that cash-flow matching must remain a monitored decision process. An opening portfolio can be well constructed and later become inappropriate after issuance of new business, changes in liability estimates, asset calls, credit downgrades, or altered constraints. Such developments call for re-estimation and controlled adaptation.
Validation and reproducibility
Validation begins with data lineage: contractual asset terms, call features, expected principal schedules, policy benefit rules, mortality tables, lapse assumptions, discount curves, reinvestment conventions, and collateral arrangements. It then tests the mathematical implementation. For fixed instruments, discounted prices must reconcile with stated cash flows and compounding rules. For loans and mortgage securities, projections should react plausibly to prepayment shocks. For liabilities, survival probabilities and policy options should affect settlement dates in the correct direction.
A disciplined model review separates input verification, formula verification, sensitivity testing, outcome reasonableness, and decision governance. Back-testing can test realized behavior where comparable history exists; stress testing examines plausible conditions that may be absent from observed records. The presence of sophisticated software is not evidence that a missing contract option or correlation has been modeled correctly. In a model of funding capacity, omissions of dated cash flows may be more consequential than refinements in the final decimal place of present value.
The present dossier's numerical examples can be reproduced without proprietary models. The three-liability discount calculation yields 256.827492. The duration-matched year-one and year-five zero payments are 46.227811 and 54.080000; they produce 0.004032 surplus after a 50-basis-point parallel upward yield shock under the stated repricing convention. The coupon case yields terminal values of 127.628156 and 125.505025 under 5% and 1% reinvestment rates. The hypothetical callable bond yields 131.224241 when not called and 118.982641 when called at year two, under 2% reinvestment. Stating these intermediate values makes arithmetic checking possible and discourages unsupported extrapolation from stylized outcomes.
Analytical synthesis and limits
Five conclusions follow jointly from the formal derivations, institutional sources, and scenarios. First, payment-date matching and present-value matching answer different questions. Second, coupon and early-principal flows transfer a portion of the funding plan to future market rates unless an offsetting hedge or dated reinvestment arrangement is secured. Third, option-bearing assets must be evaluated in the same states that change insurance liabilities. Fourth, actuarial expectations and prudential asset eligibility are different layers of the analysis, so a mathematically attractive portfolio can remain institutionally inappropriate. Fifth, the relevant performance measure is the ability to meet claims throughout time, together with a transparent account of residual risks and available capital.
These are integrated deductions from known instruments and methods. They do not constitute a newly invented matching theory, actuarial model, optimization principle, or regulatory doctrine. Important unresolved empirical questions remain outside the scope of this desk-based study: the true joint distribution of interest rates and policyholder exercise for particular products, the reliability of private-asset contractual cash flows, the cost of maintaining contingent liquidity under stress, and the effect of heterogeneous national rules on actual allocation decisions. Reliable empirical estimates require portfolio-level data and identification methods that the present sources do not provide.
Conclusion
Cash-flow matching is strongest when its promised asset receipts occur in the required currency and amount at the dates an insurer must settle obligations, with credit performance and operational availability explicitly recognized. This form of dedication can substantially reduce the need to rely on reinvestment or premature asset sales. The result is conditional on the liabilities actually being due as projected and on the assets paying as scheduled.
The study establishes that reinvestment risk is an independent funding problem even when contractual coupons and principal arrive exactly as written. A bond with predictable coupons may fail to generate the horizon amount implied by its opening yield if later reinvestment rates decline. An issuer call or borrower prepayment can accelerate principal into the very rate environment in which the investor's subsequent returns are least attractive; a rise in rates can instead extend asset lives. A duration-matched portfolio can hedge small parallel rate movements and still leave a large timing mismatch at the benefit date. Mathematical valuation equality therefore needs to be interpreted alongside the dated cash account and state-contingent settlement conditions.
The insurance dimension deepens this result. Longevity, mortality, surrender rights, claim inflation, reinsurance timing, and liquidity constraints affect the liability map. Exact replication of a single expected schedule cannot remove the volatility of contingent insurance promises. Cash-flow testing and stress design should examine joint asset and liability behavior, the effect of contractual options, and the practical interventions available when cash is short.
The comparative supervisory evidence gives the final institutional conclusion. IAIS ALM principles, EU Solvency II investment and matching-adjustment rules, UK matching-adjustment supervision, US actuarial cash-flow and asset-adequacy standards, and IFRS accounting frameworks address related but distinct purposes. Successful economic matching does not automatically confer a legal matching adjustment, a reserve sufficiency opinion, or an accounting classification. A credible funding decision is therefore an explicitly documented conjunction of dated cash flows, contingent liabilities, model quality, asset admissibility, and governance.
This monograph contributes an original synthesis, reproducible quantitative illustrations, and source-specific comparative analysis within the established disciplines of actuarial science and financial economics. Its scientific conclusions are confined to the assumptions disclosed and the authorities cited. The work is designed as a research reference within Insurance by Bogdanova, with first publication and public canonical activation reserved for a separately authorized release.
References
Actuarial Standards Board. (2025). Life or Health Cash Flow Analysis: Actuarial Standard of Practice No. 7 (revised December 2025; effective June 1, 2026). American Academy of Actuaries. https://www.actuarialstandardsboard.org/asops/life-or-health-cash-flow-analysis/
Actuarial Standards Board. (2021). Statements of Actuarial Opinion Based on Asset Adequacy Analysis for Life Insurance, Annuity, or Health Insurance Reserves and Other Liabilities: Actuarial Standard of Practice No. 22. American Academy of Actuaries. https://www.actuarialstandardsboard.org/asops/asop-no-22-statements-of-actuarial-opinion-based-on-asset-adequacy-analysis-for-life-insurance-annuity-or-health-insurance-reserves-and-other-liabilities/
Basel Committee on Banking Supervision. (2024; framework effective January 1, 2026). Basel Framework, SRP31: Interest Rate Risk in the Banking Book. Bank for International Settlements. https://www.bis.org/committees/bcbs/basel-framework/standard/srp/31/inforce/2026-01-01/published/2024-07-16
CFA Institute. (2026). Interest Rate Risk and Return. Professional Learning, Refresher Readings. https://www.cfainstitute.org/insights/professional-learning/refresher-readings/2026/interest-rate-risk-and-return
CFA Institute. (2026). Liability-Driven and Index-Based Strategies. Professional Learning, Refresher Readings. https://www.cfainstitute.org/insights/professional-learning/refresher-readings/2026/liability-driven-index-based-strategies
European Insurance and Occupational Pensions Authority. (2021, maintained). Solvency II Single Rulebook: Directive 2009/138/EC, Article 132, Prudent Person Principle. https://www.eiopa.europa.eu/rulebook/solvency-ii-single-rulebook/article-2219_en
European Insurance and Occupational Pensions Authority. (2021, maintained). Solvency II Single Rulebook: Directive 2009/138/EC, Article 77b, Matching Adjustment to the Relevant Risk-Free Interest Rate Term Structure. https://www.eiopa.europa.eu/rulebook/solvency-ii-single-rulebook/article-2162_en
European Insurance and Occupational Pensions Authority. (2021, maintained). Solvency II Single Rulebook: Directive 2009/138/EC, Article 77c, Calculation of the Matching Adjustment. https://www.eiopa.europa.eu/rulebook/solvency-ii-single-rulebook/article-2163_en
European Insurance and Occupational Pensions Authority. (2021, maintained). Solvency II Single Rulebook: Delegated Regulation (EU) 2015/35, Article 53, Calculation of the Matching Adjustment. https://www.eiopa.europa.eu/rulebook/solvency-ii-single-rulebook/article-2466_en
European Insurance and Occupational Pensions Authority. (2026, July 15). EIOPA Completes Solvency II Review Mandate with Final Guidelines and Draft Technical Standards before Revised Framework Takes Effect Early Next Year. https://www.eiopa.europa.eu/eiopa-completes-solvency-ii-review-mandate-final-guidelines-and-draft-technical-standards-revised-2026-07-15_en
Fellows, D. E. (1981). Pension Fund Liabilities and Asset Matching. Journal of the Institute of Actuaries, 108(2), 211–228. https://www.cambridge.org/core/journals/journal-of-the-institute-of-actuaries/article/abs/pension-fund-liabilities-and-asset-matching/E78DEFDD588ACBEF7413B6C19C13C3D9
Fisher, L., & Weil, R. L. (1971). Coping with the Risk of Interest-Rate Fluctuations: Returns to Bondholders from Naive and Optimal Strategies. The Journal of Business, 44(4), 408–431. https://ideas.repec.org/a/ucp/jnlbus/v44y1971i4p408-31.html (DOI: 10.1086/295402).
IFRS Foundation. (2017, amended). IFRS 17 Insurance Contracts. https://www.ifrs.org/issued-standards/list-of-standards/ifrs-17-insurance-contracts/
IFRS Foundation. (2014, as issued and amended). IFRS 9 Financial Instruments. https://www.ifrs.org/issued-standards/list-of-standards/ifrs-9-financial-instruments/
International Association of Insurance Supervisors. (2024, maintained). Insurance Core Principles and Common Framework, ICP 16.5: Asset-Liability Management. https://www.iais.org/icp-online-tool/
Prudential Regulation Authority, Bank of England. (2018–2026). Supervisory Statement SS7/18: Solvency II: Matching Adjustment (current October 2025 version; July 2026 future version scheduled effective December 31, 2026). https://www.bankofengland.co.uk/prudential-regulation/publication/2018/solvency-2-matching-adjustment-ss
Redington, F. M. (1952). Review of the Principles of Life-Office Valuations. Journal of the Institute of Actuaries, 78(3), 286–340. https://www.cambridge.org/core/journals/journal-of-the-institute-of-actuaries/article/abs/review-of-the-principles-of-lifeoffice-valuations/42DE39108ADCAD8D476B8E633F8E606F (DOI: 10.1017/S0020268100052811).
U.S. Department of the Treasury, Bureau of the Fiscal Service. (n.d.). STRIPS: Separate Trading of Registered Interest and Principal of Securities. TreasuryDirect. https://www.treasurydirect.gov/marketable-securities/strips/
U.S. Securities and Exchange Commission. (n.d.). Mortgage-Backed Securities and Collateralized Mortgage Obligations. Investor.gov. https://www.investor.gov/introduction-investing/investing-basics/glossary/mortgage-backed-securities-and-collateralized
U.S. Securities and Exchange Commission. (n.d.). Prepayment Risk. Investor.gov. https://www.investor.gov/introduction-investing/investing-basics/glossary/prepayment-risk