RESEARCH DOSSIER
Life Insurance Affordability and Coverage Persistence
Abstract. Life insurance affordability is a dynamic relation between a household's capacity to sustain premiums, the adequacy and stability of insured benefits, and the consequences of changing or terminating coverage. Coverage persistence is the continuation of an enforceable protection arrangement over time, subject to contract status, premium requirements, benefit adjustments, and the evolving need for protection. This monograph investigates how income, liquid resources, premium schedules, health and family shocks, contractual options, sales incentives, and consumer-protection rules jointly shape continuation, reduction, surrender, and lapse of life insurance. It draws on the economics of contingent claims and dynamic contracts, U.S. life policy experience studies, household survey research, a structural investigation of lapse decisions, recent British supervisory evidence, Indian product regulation, European consumer research, and international market statistics. The analysis establishes that a low initial premium, an observed in-force contract, and an administratively classified lapse are each incomplete indicators of affordable and durable survivor protection. A policy may become expensive relative to available resources while remaining economically valuable; it may stay in force with materially reduced coverage; and its termination may follow financial hardship, beneficial replacement, diminished need, contract design, or sales incentives. Original analytical synthesis combines existing methods to distinguish household budget affordability, actuarial continuation, and the retention of adequate death benefits, while specifying observational limits and policy-state transitions. Stylized arithmetic demonstrates the compounding consequences of lapse hazards and heterogeneous benefit preservation. The empirical literature supports substantial heterogeneity across contract forms and populations, but it does not establish one universal affordability threshold or a single causal explanation for lapse. The dossier proposes no new theory, doctrine, independently estimated causal effect, or proprietary dataset.
Keywords: Life insurance; affordability; coverage persistence; life insurance lapse; surrender; nonforfeiture; premium burden; disposable income; household liquidity; death-benefit adequacy; term insurance; whole life; universal life; premium persistency; policyholder behavior; dynamic contracts; bequest motives; distribution incentives; consumer protection; survival analysis; competing risks; causal identification; market conduct; comparative insurance regulation.
Introduction
Life insurance transforms an uncertain death-related financial loss into a contractual promise whose continued existence depends on both insurer performance and policyholder conduct. A household purchases protection at one date but may need it years later, when income, debt, caregiving responsibilities, mortality risk, available savings, health, and the premiums required to retain the contract have all changed. Affordability is therefore experienced repeatedly. The original purchase establishes the possibility of protection; continued eligibility, payment, and benefit preservation determine its realized durability.
This longitudinal perspective changes how the principal research question should be formulated. The relevant object is not simply whether an individual can buy a policy today. It is whether the contract's projected premium and benefit schedule can coexist with household resources throughout the period in which a death benefit remains valuable, and how contractual and institutional arrangements respond when those resources change. Term life policies, savings-oriented whole life contracts, and flexible-premium universal life policies express this problem through different legal and actuarial mechanisms. Their persistence measures should not be merged without contract-specific definitions.
Yaari's uncertain-lifetime consumption analysis situates life insurance in intertemporal decision-making under mortality uncertainty. Hendel and Lizzeri's dynamic-contract evidence shows that the timing of life premiums and the limits of consumer commitment affect contract design and lapse. Fang and Kung's structural study of lapse decisions demonstrates heterogeneous responses to income, health, bequest motives, and other shocks. These foundations establish the need to analyze affordability and continuation jointly, rather than treating the insurance decision as a static choice at issue.
The empirical material is deliberately heterogeneous. Insurer experience reports provide substantial policy exposure and termination information but limited direct household budget observations. Consumer surveys measure perceptions, ownership, and self-reported barriers but rarely observe the full contractual histories needed for lapse attribution. Regulators examine conduct and product features within legal regimes, while international statistics aggregate businesses with diverse insurance and savings functions. The contribution of this dossier is a disciplined synthesis of these evidentiary layers and transparent deductions from their differences.
The geographic scope is international. Specific legal provisions are attributed to their jurisdictions and dates. U.S. experience studies, British supervisory material, Indian surrender regulation, European consumer research, and OECD market statistics are examined as distinct observations. No national provision is presented as a universal insurance rule, and no population figure is treated as a verified global estimate of household affordability.
Definitions, Historical Context, and Source-Critical Foundations
Scientific Object and Three Separate Outcomes
Life insurance affordability denotes the feasibility, over relevant future dates, of paying the amounts necessary to maintain a contract and its desired benefits while meeting essential household obligations and preserving an acceptable level of financial resilience. This is an operational economic definition for the present analysis. Its meaning depends on premiums, fees, expected future price increases, disposable resources, liquidity reserves, the timing of income, and the household's alternative financial obligations. No common numerical ratio is declared a universal affordability standard.
Policy persistence refers to survival of an insurance contract or administratively identified policy state through a measurement period. Benefit persistence concerns the continuation of a specified amount and scope of effective death coverage. Adequate-protection persistence adds a needs-based threshold: protection remains sufficient for the surviving household, given the structure of dependents, debts, replacement income, available assets, and other coverage. A policy can satisfy the first condition and fail the second or third.
The distinction has actuarial and consumer importance. A permanent life contract may retain reduced paid-up insurance even after scheduled premiums stop. A universal life contract can remain in force temporarily although premiums received fall below the originally planned amount, depending on the account value, charges, and guarantees. A term policy may lapse after premiums cease, while a replacement policy may maintain some household protection with a different premium and benefit structure. Contract status, premium receipts, and effective sum insured must therefore be recorded independently.
For this monograph, affordability is not equated with willingness to pay, survey-reported discomfort, or low premium relative to national income. A person can afford a premium but rationally decline coverage after dependents become financially independent. Another may willingly pay a premium at the expense of basic necessities. Neither observation, alone, establishes affordability or the adequacy of coverage.
Contract Types and Policy States
Term life generally provides a specified death benefit over a defined insurance period, with conditions governing renewal, conversion, and continuation. At the end of the initial level-premium period, renewal prices can rise materially if permitted by the contract. Whole life combines death protection with contractual values and, where applicable, nonforfeiture alternatives. Universal life usually permits flexible premium payments, but the maintenance of death protection depends on account deductions and any secondary guarantees. The NAIC life insurance guide explains the core consumer distinction between term and permanent protection.
A minimum taxonomy of contract states is essential. Active at full benefit is different from active at reduced benefit. Active in a grace period is different from terminated without value. Surrender with payment of value is different from forfeiture after default. Contract maturity, insured death, expiration of term, replacement, and cancellation at the policyholder's election must be separately coded. A single label, “lapse,” is often used more broadly in experience studies than in a jurisdiction's legal definitions; the measurement convention must be stated.
A further distinction concerns premium-payment behavior. For flexible-premium products, paid premiums may be less than the planned premium without immediate termination. Premium persistency can be measured by premiums collected relative to an original or current plan, yet a resulting percentage cannot be interpreted as the fraction of households whose death benefits remain enforceable. The SOA/LIMRA universal life experience study explicitly treats premium persistency and policy lapse through different submitted data and metrics.
Economic and Historical Foundations
The economic reason to preserve protection is connected to the distribution of resources after the policyholder's death. Under uncertain lifetime, a death benefit can support dependent consumption and debt repayment, and premium payments compete with living consumption and precautionary savings. The Yaari (1965) formulation provides a foundational framework for this intertemporal insurance decision, while its formal assumptions cannot establish the observed behavior of a particular contemporary population.
Long-duration pricing introduces an additional complication. Premiums are paid in one sequence, while the costs of mortality protection, acquisition, maintenance, and insurer capital change over policy duration. Frontloaded premiums can partly finance later protection and affect the cost of exiting. Hendel and Lizzeri (2003), through their analysis of contractual commitment, find patterns consistent with a role for premium prepayment and partial consumer lock-in in observed life contracts. This is evidence about the relation between contract design and lapse, not a general instruction to increase initial premium burdens.
The legal history of cash values and nonforfeiture expresses another aspect of continuity: a consumer who has already financed part of a contract's value may be entitled to a different insurance benefit or surrender payment if premium payments cease. Those rights vary by jurisdiction and product. Inference about persistence requires the original policy form and the rules applicable when the relevant premium was due.
Sources, Populations, and Interpretation
The evidentiary hierarchy used here puts enacted legal instruments and regulator-issued documents first for jurisdiction-specific obligations; insurer experience data first for explicitly defined experience populations; primary scholarly articles first for the studies' empirical methods and findings; and consumer surveys first for responses within their sampling frames. Market-wide aggregates are used for context and never substituted for individual policy histories.
The SOA Research Institute and LIMRA's 2015–2022 term and whole life study describes policy exposure from 28 contributing U.S. companies. The experience is restricted to specified fully underwritten, single-life policies and is not a random sample of all countries, policy designs, or distribution arrangements. The 2025 Insurance Barometer measures U.S. consumer perceptions and ownership, a different evidentiary object. The FCA's pure protection lapse analysis examines British distribution and contract outcomes; legal coverage and product composition differ from the SOA experience files.
Each use of evidence in the following chapters identifies the unit observed, the endpoint measured, and the interpretation justified. A pooled study's volume of exposure strengthens descriptive experience measurement for its contributors; it cannot alone reveal which individual contracts were financially unaffordable. A consumer's stated cost barrier identifies a perceived impediment; it cannot establish the premium at which the same respondent would actually maintain a given policy in the future. A causal finding from a structural model depends on its data, identification assumptions, and counterfactual construction.
Coverage Persistence, Lapse, and Actuarial Consequences
Policy-State Transitions and Competing Outcomes
Contract continuation is a multi-state process. Let A represent active full-benefit coverage, G a grace-period state, U active reduced or otherwise modified coverage, S a surrender with value paid, L termination without a death benefit, D an insured death claim, and X expiration or maturity under contractual terms. These labels are descriptive, not a newly proposed actuarial model. They apply established multi-state reasoning to the specific research problem. From A, a contract may remain A, enter G or U, be surrendered, terminate, pay a claim, or expire. Rules governing possible transitions differ across product forms and jurisdictions.
It follows that “still in force” is an incomplete protection outcome. Where a reduced paid-up policy preserves a smaller death benefit, A-to-U is a material reduction in household protection despite contract continuity. Where a contract is replaced, a termination in the original insurer's file may coincide with continued, increased, or reduced aggregate household coverage. Administrative records restricted to a single issuing company cannot always distinguish the competing explanations.
For a defined original issuance cohort, one may measure full-benefit persistence at duration t as the number of policies still delivering the original specified death benefit divided by the appropriately defined starting or exposed cohort, with consistent exclusions for insured death and scheduled expiration. Contract persistence uses a broader numerator. Premium persistency uses premium cash flows and planned amounts, and may fall without contract termination. These measures should be displayed together rather than combined into one misleading indicator.
Measuring Persistence Correctly
A conventional lapse rate approximates terminations of a designated type during an exposure period divided by policies or policy-years at risk, subject to experience-study timing conventions. The precise exposure basis, treatment of mid-period events, and inclusion of surrenders must follow the reporting source. A policy-count-weighted rate describes the experience of contracts; a face-amount-weighted rate describes the distribution of insured amounts; a premium-weighted rate describes the funds associated with continuing or ending contracts. Their denominators differ, and their rankings across products may change.
The SOA/LIMRA term and whole life experience reports approximately 135.9 million policy exposures and 5.4 million combined surrenders and lapses, with roughly $30.6 trillion face amount exposed and $1.4 trillion face amount in combined terminations. These are aggregated study exposure and event totals across contributed company-years, not counts of 135.9 million distinct living policyholders. Dividing the event total by aggregate policy exposure produces a rough exposure-weighted descriptive ratio around 4.0%; it is not a measured causal response to affordability, nor a universal life-insurance lapse probability. Published results separate durations and other covariates, and the detailed experience coverage is constrained by product eligibility and company participation.
For flexible-premium universal life, the SOA/LIMRA 2015–2021 experience summary identifies a premium-persistency contribution from 14 companies with 11.9 million policies exposed and approximately 417,000 lapse/surrender terminations, while a separate lapse/surrender contribution from 24 companies covers approximately 33.5 million policy-years and 1.3 million terminations. The studies have different reporting populations and different definitions. Treating their event counts as interchangeable, or using premium collections as a proxy for a contract survival rate, would be invalid. A discrepancy between the landing-page and report text for one cumulative collected-premium total is excluded from this monograph's quantitative conclusions.
Evidence from Term, Whole, and Universal Life
Term products are exposed to premium-schedule discontinuities, especially after an initial guaranteed level period. In those settings the observed end of coverage may partly reflect an anticipated end-of-term replacement decision, a price jump, declining need for protection, or health-based constraints on a new purchase. Without observing renewal offers and the existence of replacement coverage, an elevated rate after the level period does not uniquely measure financial distress. The SOA policyholder behavior catalog identifies separate studies of post-level term lapse and mortality, emphasizing that these behaviors warrant product-specific analysis.
Whole life introduces cash-value and surrender considerations. Surrender can deliver money to a household facing a short-term shock but can also extinguish death protection and cause a loss relative to the household's long-run needs. The net value of surrender must be compared with alternatives such as permitted policy loans, reduced paid-up status, available nonforfeiture elections, and substitute protection, using the actual contract. A gross surrender amount is neither the contract's economic profit nor proof that cancellation was advantageous.
Universal life adds the distinction between contributions and deductions. When a policyholder reduces premium payments, account value may temporarily fund charges; later depletion may lead to termination unless a guarantee applies. Consequently, an early decrease in premium persistency can precede loss of the death benefit by several years. In this product family, the timing of lapse is shaped by the evolving account and charges as well as by current household capacity, and the analysis should measure coverage and premium outcomes separately.
Illustrative Continuation Mathematics
Suppose an otherwise homogeneous hypothetical cohort faces a constant annual probability q of termination from an explicitly defined state, with no competing exit and no changes in policy status. Then the fraction of original contracts expected to survive ten independent annual intervals is (1 − q)^10. At q = 0.06, the hypothetical continuation fraction is approximately 0.539; at q = 0.10 it is approximately 0.349. The difference, about 19 percentage points, is a consequence of the specified assumptions, not an empirical estimate or a forecast. If actual lapse risk varies by duration, health, product, income, or replacement, the calculation requires corresponding conditional hazards and an appropriate treatment of competing deaths and expirations.
For a duration-dependent conditional lapse probability q_t, pure policy continuation through T years can be represented as the product across dates of (1 − q_t), if the probabilities are defined among those still exposed and the relevant competing events are handled coherently. The insurance benefit retained at T has an additional dimension: even surviving contracts may have reduced face amounts. Thus an expected full-benefit indicator and an expected insured-benefit amount need not move together. A duration-independent survival calculation is useful only as a sensitivity illustration.
If a cohort contains two equally sized groups with the same original benefit but hypothetical termination probabilities of 2% and 12% per year, the ten-year unweighted continuation proportion equals one-half times 0.98^10 plus one-half times 0.88^10, approximately 0.548. It is not equivalent to treating everyone as facing the averaged 7% annual hazard, which yields approximately 0.484. Heterogeneity matters because survival conditioning changes the mixture of remaining policyholders over time. This simple arithmetic reveals why aggregate lapse percentages cannot always be extrapolated to future coverage without stratifying the underlying risks.
Selection and Insurer Exposure
Persistence changes insurer liabilities, acquisition-cost recovery, required liquidity, and the composition of remaining mortality risks. High termination rates may reduce some future death-benefit obligations but can also eliminate future premium flows and accelerate surrender cash outflows. Which direction dominates depends on reserve structure, contract guarantees, investment conditions, and the characteristics of policyholders who leave. A lapse assumption cannot be translated into a simple profit effect independent of product design.
Health-sensitive lapse can generate selection. If less healthy individuals recognize the increased value of already-underwritten protection, they may remain insured while healthier individuals lapse or replace coverage. Conversely, liquidity shocks can force a person with elevated mortality risk to surrender valuable insurance, potentially creating a serious household welfare loss. These mechanisms imply heterogeneous relationships among lapse, future mortality, and financial hardship; they are propositions for empirical investigation, not quantified findings of this dossier.
Mass lapse is a separate portfolio stress problem. Biagini, Huber, Jaspersen, and Mazzon (2021) investigate extreme cancellation rates with simulation and company-level panel information, emphasizing the challenge of estimating rare joint exits. Their results caution against universal mass-lapse stress calibration. Individual household hardship and a market-wide correlated surrender episode can produce the same administrative event type but require different explanations and risk management.
Consumer Protection, Product Design, and Distribution
Notices, Grace Periods, and Nonforfeiture
Consumer-protection rules influence the interval between an initial payment difficulty and the irreversible loss of a life benefit. Notices can inform policyholders of upcoming premiums or threatened termination. Grace periods can preserve coverage for a specified period and create time to restore payments. Nonforfeiture provisions can preserve some accumulated contractual value through cash surrender, reduced paid-up insurance, or other choices, where the policy form and legal regime provide them. These mechanisms differ in both legal effect and the protection they preserve.
A jurisdiction-specific example is New York Insurance Law §3211. It sets detailed rules about premium notices and lapse for identified life and disability policies, including timing for scheduled-premium notices and separate treatment of flexible premiums. Its provisions must be read in their complete statutory setting, with the applicable contract and administrative practice; it does not establish a global notice period or a universally transferable grace rule.
The existence of a notice obligation is not equivalent to reliable receipt or understanding. Research should examine recorded address, consent to electronic delivery, policyholder language and accessibility needs, delivery time, the timing of bank collection, and whether a notice describes the exact benefit at risk. A formally compliant notice may still be poorly understood; a direct test would connect notice characteristics with later payment recovery and preserved coverage, controlling for initial financial position.
Intermediary Incentives and Quality of Advice
Distribution incentives affect the policy a person enters and the reasons the policy is later discontinued. An intermediary compensated heavily at sale can have incentives that diverge from the consumer's long-run ability to maintain the recommended premium. Commission clawback arrangements may change the timing of these incentives. Effective governance requires information on sale suitability, expected premium paths, early-policy behavior, replacement, and complaints, rather than relying only on a commission percentage.
The UK FCA's 2026 pure protection analysis expressly discusses that policy lapse can follow unsuitable or unaffordable sales, changing circumstances, or beneficial switching. In its insurer sample, 12.4 million policies were in force at the end of 2024; about 53% had been in force for at least five years and 24.3% for at least ten years. Those are tenure distributions among contracts remaining in force, and therefore are not cohort probabilities of survival from origination. The FCA reports average lapse-rate ranges of 1%–7% for the broader in-force material and 2%–7% for categories of intermediated pure protection. These estimates should not be merged with U.S. universal-life or whole-life data because products, periods, sampling, definitions, and regulation differ.
A strong advice-quality analysis follows the insured household after any lapse or replacement. If the former contract ends but a suitable lower-cost policy starts with continuous effective protection, consumer welfare may improve. If a replacement destroys advantageous underwriting terms or creates an unrecognized gap, the same administrative lapse can signal harm. The distinction can be resolved only with outcome information about both contracts, the client's circumstances, and continuity of benefits.
Comparative Legal Instruments
India provides a different institutional example centered on life products with surrender and nonforfeiture rights. The Insurance Regulatory and Development Authority of India's Insurance Products Regulations 2024 govern specified product designs, surrender values, and related obligations. Their treatment depends on product class, premium-payment history, and the applicable schedule. A guaranteed or special surrender value under a qualifying savings product must not be imputed to a pure risk term contract without examining whether the relevant rule applies.
IRDAI's 2024 life-products master circular additionally directs insurers to examine persistency across distribution channels and address misleading sales practices associated with surrender or lapse. These are supervisory expectations directed at market conduct and product value. They support the analytical proposition that persistence is relevant to the consumer experience, while the actual effect of the rule on future surrender rates requires a distinct empirical evaluation. A general statement about “Indian life insurance” without distinguishing linked, nonlinked, protection, savings, and limited-premium classes would obscure material legal differences.
European supervisory analysis emphasizes the quality and value of insurance outcomes. EIOPA's Consumer Trends Report 2024, published in January 2025, examines insurance and pension consumers' experience, digital distribution, and perceived value. It does not provide a universally harmonized life-insurance lapse panel. Its contribution here concerns the institutional importance of understandability and value for money rather than a direct causal estimate for life-policy persistence.
Information, Switching, and Complaint Evidence
A meaningful disclosure should answer what death benefit exists today, what the household must pay to keep it, under which conditions the premium can change, what happens if payment is missed, and which rights survive surrender or reduction. For products combining protection and saving, it should also distinguish guaranteed values, projected values, and charges. The way those facts are displayed can influence decisions, but any asserted quantitative behavioral effect requires a study of actual presentation and subsequent action.
Consumer-reported “cost” is a broad survey category that may mix inability to pay, perceived low value, uncertainty about future prices, mistrust, and allocation of funds to more immediate needs. The LIMRA/Life Happens insurance ownership research is useful for describing the stated barriers within a U.S. survey population. It cannot, by itself, allocate observed insurer lapse totals among these mechanisms. Such an allocation would require merging permitted survey responses with payment events and policy histories while protecting confidentiality.
Complaints, cancellation calls, repeated late-payment attempts, policy illustrations, and replacement records provide complementary evidence. A high number of early lapses concentrated in one distribution channel may be a useful audit trigger, particularly if accompanied by affordability complaints or poor understanding of future price changes. A low lapse rate can arise from valuable long-duration protection or from exit penalties and limited awareness of alternatives. Persistence must therefore be evaluated with protection adequacy, consumer understanding, and costs of exit.
Comparative Empirical Evaluation and Causal Interpretation
Household Surveys and Insurer Administrative Data
Household surveys and policy administration systems have different strengths. Household data can observe employment, income, wealth, family changes, and stated needs. Administrative records can observe premium due dates, premium receipt, policy duration, face amounts, surrender payments, loans, and contract status with greater operational precision. Each side alone leaves crucial questions unresolved. Surveys may not report exact policy terms or dates, while insurer files rarely record all household assets and new insurance purchased from competitors.
The SOA/LIMRA 2015–2022 study provides valuable descriptive experience by product, duration, and other dimensions, within explicitly limited participating business. The U.S. insurance barometer addresses consumer attitudes and insurance ownership in an entirely different design. A research conclusion that blends these sources must respect their separate denominators. A stated ownership gap cannot be converted into an observed annual lapse rate; the number of policy exposures cannot be converted into an estimate of unique unprotected families.
An ideal longitudinal dataset would track the same policyholder over multiple years with consented observations of income and essential expenditure; premium schedules, account value, death benefits, and riders; claims, surrenders, reductions, and reinstatements; advice and distribution; and replacement coverage with other insurers. Death, term expiration, and emigration from the dataset must be coded appropriately as competing exits or censoring events. Missingness and attrition should be separately assessed, especially when financial distress makes respondents harder to follow.
Longitudinal Inference and Confounding
A simple comparison of lapse rates between low- and high-income buyers is generally confounded. Income covaries with age, occupation, education, health, underwriting, contract type, number of dependents, financial assets, and marketing access. The observed policyholder population itself is selected by the decision to buy coverage. A product with more stringent underwriting or a different channel can show higher persistence even if the causal effect of its premium is smaller than that of another product.
The Fang–Kung dynamic choice study is instructive because it explicitly incorporates income, health, bequest-related incentives, and unobserved shocks. Its structural counterfactuals rely on a specified behavioral model and assumptions concerning states and unobservables. Their existence does not justify presenting the resulting mechanisms as experimentally identified in all markets. A replication should describe the HRS observations used, the model's available choice set, identification restrictions, parameter uncertainty, and how changes in measured coverage are distinguished from switching.
A policy premium change can sometimes supply useful variation, but its causality is not automatic. Renewal increases are correlated with duration, age, product guarantees, and possibly interest rates. A difference-in-differences approach would need comparable treatment and control groups, a credible parallel-trends argument, complete reporting of anticipated premium changes, and checks for simultaneous changes in terms. A discontinuity design at a predetermined contract anniversary must address endogenous contract selection and the possibility that policyholders anticipate the adjustment. No causal coefficient is estimated in this monograph.
International Comparability
The OECD's Global Insurance Market Trends 2026 reports average nominal life premium growth of 8.7% in 2025 across reporting jurisdictions. The report also describes moderation of surrender-related pressures: Portugal's surrender payments declined by around 19%; Italy's nominal life claims including surrenders fell by around 7%; and lower surrender activity contributed to changes in French life claims. Interest-rate movements and savings-linked product conditions help interpret these developments. They cannot be used as evidence that household term-life insurance became more affordable by a corresponding amount.
Comparability is restricted by the institutional meaning of “life insurance.” Some countries' life premium statistics include long-duration savings and investment-linked contracts, and others are dominated by protection insurance or pension-linked products. Interest-rate-driven transfers away from savings policies may dominate an aggregate series without corresponding changes in survivor-protection needs. An international comparison must report product composition, exchange-rate and inflation treatments, contract termination categories, and the distinction between paid claims and paid surrender values.
The OECD's market-wide insurance summary contrasts total insurance penetration in high-income and lower-middle-income jurisdictions. Penetration is premium volume as a share of gross domestic product, not a household budget ratio, insurance coverage rate, or continuation measure. A low national penetration value may coexist with strong insurance protection among some households and severe gaps among others. Country comparisons must be linked to household or policy data before they support an affordability conclusion.
The FCA's British pure protection evidence and the SOA's U.S. individual fully underwritten experience similarly cannot be ranked directly as if a single globally consistent lapse definition and cohort frame existed. Measurement differences may be as large as genuine behavioral differences. All cross-border estimates in this dossier are therefore descriptive, with source-specific terminology retained.
Stress Tests and Sensitivity Analysis
The first useful stress dimension is a household liquidity shock. A projected premium budget can be tested under temporary unemployment, a reduction in work hours, unexpected essential expenditure, and delayed income recovery. A second dimension is premium-schedule stress: an announced or permitted renewal increase, lower interest credits in an account-value policy, or changes in premiums for supplementary benefits. A third dimension is protection-need stress: new dependents, increased debt, or the exhaustion of alternative savings.
Stress analysis should also include access to contractual alternatives. In one policy, a missed payment may be cured within a grace period. In another, existing value can fund deductions briefly. In a third, a qualifying paid-up alternative may preserve a smaller death benefit. Each scenario produces a separate trajectory for protection and cash flow. A single “lapse severity” parameter cannot describe these institutional differences without losing important welfare and liability information.
The numerical continuation illustrations above demonstrate another sensitivity: a seemingly modest change in annual conditional termination rates compounds over long protection periods. In real data, a credible stress test would use duration-specific transitions and evaluate higher lapse probabilities jointly with changes in mortality, interest rates, and liquidity needs. A portfolio stress scenario should be reconciled with observed experience and transparent assumptions, following the methodological concerns in Biagini and colleagues (2021).
Evidence Hierarchy and Boundaries
The evidence can be ordered by what it establishes. Contract and legal documents establish the available rights and payment obligations for specified products. Experience studies establish events recorded under a documented exposure definition. Survey responses establish perceptions and reported circumstances in the sample. Dynamic modeling can estimate structural relationships conditionally on assumptions. Comparative international tables establish aggregates according to national reporting definitions. Every stronger inference requires additional linkage and assumptions.
There are several predictable errors. First, the combination of lapse and surrender can mask important differences in the cash received by the policyholder. Second, a consumer's preference not to renew can be misclassified as financial exclusion. Third, a contract held in force at reduced coverage can be counted as persistent even after a material loss of protection. Fourth, tenure among contracts currently in force can be misread as the share of all original contracts surviving to that duration. Fifth, a premium-weighted measure can be mistaken for a count-weighted lapse rate. Sixth, monetary trends can be compared without separating nominal and real changes.
All six errors alter how affordability and coverage persistence are interpreted. A high-quality study records the original risk set, each relevant policy state, the contractual terms, and the population underlying each published figure. When that information is absent, the appropriate outcome is a bounded descriptive finding rather than an invented causal narrative.
Integrated Scientific Findings and Research Implications
Protection Continuity as a Joint Outcome
The first integrated finding is that affordable and durable life insurance is a joint result of household liquidity, the future cost of maintaining specified benefits, and the availability of contractual pathways that preserve protection during temporary financial difficulty. This conclusion follows from the interaction of dynamic premium obligations with the observed institutional distinction between premium receipts and contract status. It is an analytical synthesis of established economics and actuarial state distinctions, not a new insurance theory.
The central unit is the household's exposure to an adverse financial outcome upon an insured death, conditional on the actual benefits enforceable at that date. A continuous administrative policy identifier is valuable evidence, but the benefit amount and relevant household obligations determine whether continuity has material meaning. For a household with several sources of life coverage, this assessment should be aggregated across contracts without double-counting insured benefits.
The second finding is that affordability is intertemporal and state-dependent. Initial price comparisons are incomplete when premium paths, future household income, and the ability to preserve benefits differ. The need for continuing protection can also change. A policy may become unaffordable exactly when replacing its underwriting terms is difficult; conversely, a policy can become unnecessary when financial dependents are no longer exposed to the contemplated loss. A scientifically adequate outcome measure must identify both continuation and current protection need.
Interpretation of Lapse and Surrender
The third finding is that administrative lapse records support multiple explanations. The FCA's consumer-outcomes investigation explicitly identifies affordability, unsuitable sales, changing circumstances, and beneficial switching as possible pathways. The Fang–Kung policyholder decision analysis adds evidence of age-dependent income, health, bequest, and idiosyncratic factors. These bodies of evidence support a multi-causal interpretation with source-specific limits.
A lapse is therefore an event to classify, not a complete welfare verdict. Surrender can release valuable cash to an overburdened household while reducing death protection. Retaining an expensive policy can sustain essential coverage or cause avoidable household hardship. Replacing a policy can create continuity or a dangerous protection gap. The relevant comparison examines cash flows, actual benefits, replacement terms, and household needs around the event.
The fourth finding is that empirical persistence measures require an explicit denominator and event definition. The SOA/LIMRA life policy experience studies provide substantial data on specified contracts, but their aggregate terminations do not identify a universal affordability response. Their primary contribution is the analysis of reported policyholder behavior within well-defined product cohorts. Extrapolation beyond those populations requires evidence that the relevant contractual, economic, and selection mechanisms transport.
Evaluation for Households and Insurance Institutions
For household welfare analysis, a defensible evaluation has at least four elements: the premium and expense burden across dates; the value of preserved death benefits relative to dependents' needs; the financial consequences of surrender or reduction; and feasible alternative protection arrangements. The analyst should report which elements are directly observed and which are estimated or assumed. This approach allows contract comparisons without treating cheapness and durable adequacy as synonyms.
For insurer management, the corresponding evidence is a policy-state experience analysis that distinguishes premium deviations, full terminations, benefit reductions, and replacement when observed. The pricing and liability implications depend on the pattern of cash inflows, death-benefit liabilities, reserves, charges, and surrender values. A short-duration deterioration in affordability can increase payment delays without immediate lapse in one product, while producing rapid terminations in another. Aggregating these paths risks inaccurate reserve, liquidity, and conduct assessments.
For supervisors, persistence is useful when triangulated with early-policy customer outcomes, premium shocks, distribution channels, complaint signals, and evidence of switching. The Indian 2024 product regulation and the British 2026 conduct analysis illustrate distinct institutional responses to product value and policyholder outcomes. Their legal effects remain nationally bounded, while the underlying empirical question—whether household protection continues on fair and comprehensible terms—has broad relevance.
Research Design, Data Requirements, and Limits
A rigorous empirical program should predefine its population at policy issue and record product family, underwriting, distribution channel, age, face amount, scheduled premiums, annual changes, nonforfeiture options, and intended protection term. It should follow premium receipt, benefit amount, account values, policy-state transitions, and death or scheduled expiration. Household income, liquid assets, dependents, debt, and other insurance should be linked only where lawful, consented, and scientifically defensible.
For causal research, the question should specify the treatment—an income shock, premium jump, notice intervention, payment flexibility change, or a distribution reform—and the outcome—payment continuity, full-benefit retention, any contract survival, or welfare-sensitive adequacy. Appropriate identification may include longitudinal fixed effects, natural experiments with carefully defended assumptions, event studies with pre-trend analysis, structural estimation, or randomized interventions when ethically and legally feasible. Each method answers different questions; no single estimator provides a universal explanation.
A full quantitative comparison would also report confidence intervals, model uncertainty, policyholder selection, missing replacement coverage, differential survey attrition, censoring, competing risks, and the treatment of inflation. For international samples it should document currency conversion, purchasing-power assumptions, product classification, and differences in statutory lapse and surrender regimes. These requirements are derived from the shortcomings of treating heterogeneous sources as one universal dataset.
This manuscript has not obtained confidential insurer microdata, matched household-policy records, experimental evidence, or a harmonized cross-national lapse dataset. It therefore offers a documentary and analytical synthesis with transparent illustrative calculations, not a new numerical estimate of the causal effect of affordability on persistence. Its original contribution consists of the explicit three-outcome distinction; the crosswalking of economic, actuarial, and legal meanings; error identification across evidence sources; and an integrated interpretation of continuation and benefit preservation using established methods.
Conclusion
Life insurance affordability and coverage persistence are best studied together over the full period of intended protection. The cost of continuing a contract is shaped by premium schedules and household cash flow, while the value of continuation depends on the protection still needed and the benefit actually preserved. A policy's formal survival, its premium receipts, and its effective death benefit describe related but distinct outcomes.
The source record supports several substantive conclusions. Contractual design influences the timing and cost of continued protection. Household income, health, bequest motives, and idiosyncratic circumstances contribute differently to lapse decisions across ages and populations. Administrative lapse rates contain terminations arising from financial hardship, beneficial switching, changes in need, and conduct concerns. The interpretation of persistence therefore depends on the population studied, the policy form, and the observation of household protection after a change in contract status.
Contemporary sources make the measurement limits concrete. SOA/LIMRA experience studies provide large U.S. policy exposure samples under defined product restrictions. British supervisory work investigates lapse as a possible consumer-outcome signal while acknowledging its several explanations. Indian rules show the importance of product-specific surrender and nonforfeiture protections. OECD international statistics contextualize savings-related surrender pressures but cannot measure individual household payment capacity. Survey information describes barriers and perceptions without revealing all policy histories.
The resulting scientific conclusion is precise: an assessment of durable life protection requires evidence of continuing payment capacity, actual contract and benefit states, and the household's evolving exposure to financial loss on death. Research that integrates those observations can identify genuine affordability failures and distinguish them from efficient changes of coverage. The current literature supplies credible components of that analysis; the absence of harmonized household-policy data limits the scope of causal and international numerical conclusions. The monograph's contribution is original analytical integration within established economic and actuarial approaches. It establishes no new theoretical system or unsupported claim of scientific priority.
References
Biagini, Francesca, Tobias Huber, Johannes G. Jaspersen, and Andrea Mazzon. 2021. “Estimating Extreme Cancellation Rates in Life Insurance.” Journal of Risk and Insurance 88 (4): 971–1000. Publisher article and DOI.
Board of Governors of the Federal Reserve System. 2023. “Expenses.” In Report on the Economic Well-Being of U.S. Households in 2022. Washington, DC: Federal Reserve Board. Official household survey report.
European Insurance and Occupational Pensions Authority (EIOPA). 2025. Consumer Trends Report 2024. Published January 15, 2025. Frankfurt am Main: EIOPA. Official consumer trends report.
Fang, Hanming, and Edward Kung. 2021. “Why Do Life Insurance Policyholders Lapse? The Roles of Income, Health, and Bequest Motive Shocks.” Journal of Risk and Insurance 88 (4): 937–970. Journal article and DOI.
Financial Conduct Authority (FCA). 2026. Pure Protection Market Study: Annex 1, Intermediated Distribution, Remuneration and Consumer Outcomes. London: Financial Conduct Authority. Official market study annex.
Hendel, Igal, and Alessandro Lizzeri. 2003. “The Role of Commitment in Dynamic Contracts: Evidence from Life Insurance.” Quarterly Journal of Economics 118 (1): 299–328. Journal article and DOI.
Insurance Regulatory and Development Authority of India (IRDAI). 2024. Master Circular on Life Insurance Products, IRDAI/ACTL/MSTCIR/MISC/89/6/2024, June 12, 2024. New Delhi: IRDAI. Official life-products circular.
Insurance Regulatory and Development Authority of India (IRDAI). 2024. Insurance Regulatory and Development Authority of India (Insurance Products) Regulations, 2024. New Delhi: IRDAI. Copy hosted by Institute of Actuaries of India. Insurance products regulations.
LIMRA and Life Happens. 2025. 2025 Insurance Barometer Study. Research abstract dated June 25, 2025. Windsor, CT: LIMRA. Official study overview.
National Association of Insurance Commissioners (NAIC). n.d. “Life Insurance.” Consumer Insurance Search and Guide. Kansas City, MO: NAIC. Accessed November 14, 2026. Official life insurance guide.
New York State Legislature. Insurance Law §3211. “Notice of Premium Due under Life or Disability Insurance Policy; Notice to Assignees of Non-Payment of Premium.” Current online statutory text consulted November 14, 2026. New York statutory text.
Organisation for Economic Co-operation and Development (OECD). 2026. Global Insurance Market Trends 2026. Paris: OECD Publishing. Life insurance market chapter.
Organisation for Economic Co-operation and Development (OECD). 2026. Global Insurance Market Trends 2026. Executive Summary. Paris: OECD Publishing. Official report executive summary.
Society of Actuaries (SOA) Research Institute. n.d. “Individual Life — Policyholder Behavior Experience Studies.” Schaumburg, IL: Society of Actuaries. Accessed November 14, 2026. Official experience study index.
Society of Actuaries (SOA) Research Institute and LIMRA. 2024. 2015–2022 Term and Whole Life Insurance Policy Surrender/Lapse Experience Study. December 2024. Schaumburg, IL: SOA Research Institute and LIMRA. Official report and study summary.
Society of Actuaries (SOA) Research Institute and LIMRA. 2024. 2015–2021 Universal Life Premium Persistency and Lapse/Surrender Experience Study. July 2024, report revised December 2024. Schaumburg, IL: SOA Research Institute and LIMRA. Official experience study summary.
Yaari, Menahem E. 1965. “Uncertain Lifetime, Life Insurance, and the Theory of the Consumer.” Review of Economic Studies 32 (2): 137–150. Journal article and DOI.