RESEARCH DOSSIER
Solidarity and Redistribution in Insurance
Abstract. Insurance redistributes resources across realized states of the world by collecting contributions before losses and paying contingent benefits after losses. The distribution of expected net benefits across people, income groups, risk classes, cohorts, and institutions is a further question. This monograph examines the relation between risk pooling, interpersonal redistribution, intertemporal finance, and solidarity, treating their accounting identities, behavioral mechanisms, legal constraints, and ethical justifications separately. It specifies an ex ante and ex post incidence framework, derives transfers under risk-rated and community-rated premiums, examines subsidy finance and insurer-level risk equalization, and distinguishes within-cohort risk sharing from transfers between birth cohorts. Comparative institutional analyses draw on international social-security principles and health-financing guidance; United States health-insurance rating and risk adjustment; European Union equal-treatment law; and Dutch and Australian community-rating arrangements. The manuscript evaluates selection, moral hazard, affordability, legal equality, vertical and horizontal equity, welfare aggregation, and the distribution of financial protection. Its central analytic conclusion is that the amount and direction of solidarity depend on the valuation date, counterfactual premium, financing boundary, distributional unit, and benefit definition. Uniform premiums may create expected cross-subsidies, actuarially differentiated premiums may still deliver considerable state-contingent insurance, and redistribution can be displaced across taxes, employers, risk-adjustment transfers, and time. A credible distributional assessment therefore reconciles actuarial incidence with tax incidence, utilization, access, and the present value of lifetime burdens. The arguments synthesize established scholarship and institutional evidence; the formal numerical exercises are transparent illustrations rather than empirical estimates or proposed new theoretical doctrines.
Keywords: Insurance solidarity; redistribution; risk pooling; cross-subsidy; community rating; actuarial fairness; social insurance; mutual insurance; risk adjustment; intergenerational transfers; welfare economics; distributional incidence; financial protection; insurance equity.
Introduction
An insurance arrangement begins with the asymmetric timing of payment and loss. Premiums or contributions are collected from many participants, while indemnities, services, or stipulated benefits are delivered to those who satisfy contractual or statutory contingencies. The promise transfers purchasing power between possible future states and, after those states are realized, between actual participants. Questions of solidarity arise because the allocation of the financing burden, the probability of receiving benefits, the ability to enter or remain in the scheme, and the governance of the shared fund can all be designed differently.
An analytic vocabulary is necessary. Risk pooling denotes the aggregation and financing of uncertain individual losses. Ex post redistribution denotes transfers observed after the realization of contingencies. Ex ante redistribution denotes systematic differences between groups' expected benefits and their expected financing contributions, evaluated on a common basis. Income redistribution refers to changes in the distribution of resources across income or wealth strata. Intergenerational redistribution concerns net transfers between birth cohorts; intertemporal smoothing concerns reallocations of resources across dates or ages, which may occur entirely within one person's lifetime. Solidarity denotes an institutional commitment to shared protection or burden sharing under an articulated collective rule. The last concept has a legal, ethical, and political meaning that must be established from institutions and their objectives, rather than inferred from the existence of a large insurer.
Kenneth Arrow's analysis of uncertainty in medical care (1963) identifies incomplete risk-bearing markets and nonmarket institutions as essential to welfare analysis. Mossin's rational insurance purchasing analysis (1968) examines purchasing behavior under risk and loading. Rothschild and Stiglitz's competitive insurance equilibrium study (1976) shows why unequal information may destabilize ostensibly competitive risk classification and pricing. These contributions establish the economic setting, while the ILO's account of Convention No. 102 (2024) and the World Health Organization's health-financing pooling guidance show that collective financing is also a policy choice concerning entitlements and distribution.
The object studied here is the distributional design of insurance pools, including related public financing and risk-equalization mechanisms. The analysis includes contractual insurance, social insurance, mutual and cooperative structures, and hybrid regulated markets, without treating all institutional types as legally interchangeable. The focus is on transfer incidence and its justification. Detailed insurance pricing, general health-system architecture, and welfare-state administration are used as necessary context, while their comprehensive treatment belongs to their own disciplines and Science areas.
Methodologically, the dossier combines a budget-consistent accounting analysis, conditional expected-value comparisons, economic interpretation of selection and insurance demand, source-critical legal comparison, and normative evaluation. Mathematical exercises specify populations, losses, premiums, and timing. Legal claims are restricted to named systems and verified provisions. Institutional guidance is distinguished from binding law, and explanatory examples are distinguished from observational estimates. The research cutoff is November 14, 2026. The references include original scholarly publications, primary legislative and judicial documents, and institutional source pages; access limitations are recorded in the evidence ledger.
Risk Sharing and Redistribution
Contingent protection and the valuation date
Consider a set of individuals indexed by i, a one-period horizon, and an insured indemnity X_i that depends on a future loss state. Let P_i denote that individual's contribution, excluding explicitly identified administration, capital costs, and external subsidies for the moment. Let q_i = E[X_i] be the expected claim cost under the specified information set. The realized net insurance receipt is X_i - P_i. The corresponding expected net receipt is q_i - P_i. For a self-financing pool with no loading and no external funds, the contributions sum to expected claims, so the sum of expected net receipts is zero. The zero-sum budget identity concerns expected monetary payments; it is compatible with a positive aggregate welfare gain from reducing undesirable consumption volatility.
This distinction has practical consequences. A person with a low probability of loss may obtain substantial insurance value even if that person's expected claims exactly equal the premium: a fixed premium replaces a volatile contingent loss. A person who suffers a large loss may receive a large realized transfer without having been a structurally subsidized risk class before the event. The insured's realized net receipt and expected net receipt answer different questions. An ex post recipient of a claim is not, on that fact alone, evidence of an ex ante cross-subsidy between risk classes.
Let an individual's uninsured wealth after loss be W_i - L_i and insured wealth be W_i - P_i - L_i + X_i. With concave von Neumann-Morgenstern utility u_i, the insurance arrangement may increase E[u_i] even when P_i = E[X_i], subject to benefit adequacy and moral-hazard effects. This is the conventional welfare basis of risk transfer, not a proof that the arrangement makes the distribution of income more equal. The Arrow (1963) and Mossin (1968) contributions support this separation: the first analyzes the welfare consequences of missing risk-bearing markets; the second investigates insurance demand when contracts and premiums change the allocation of uncertainty.
The valuation date also controls interpretation. Before uncertainty resolves, two people may have the same expected loss and face the same premium. After uncertainty resolves, their claim receipts diverge. The insurance payout is a redistribution across realized states, and it is also a payment against a preexisting entitlement. Moral or legal evaluation must recognize the latter: claimants receive what the contract or statute promises, rather than gratuitous transfers from remaining contributors. The distinction makes it possible to affirm the social function of risk sharing without redescribing every valid indemnity as a discretionary subsidy.
The accounting perimeter of a pool
A closed pool's simplified period budget is
where K_0 is opening capital, r is its realized net return, G is external public or donor support, A represents administrative and operating expenses, and K_1 is closing capital. For stochastic expenses and claims, this equality holds for realized cashflows with consistent period definitions; actuarial expected-value identities use expectations for both sides. Insurance profits, reinsurance recoveries, capital injections, taxes, and changes in technical provisions can be added to obtain a full institutional ledger. These additions alter incidence, so they cannot be silently treated as transfers between the policyholders alone.
For a risk group g containing N_g members, define the average expected underwriting transfer as
where P_g^{risk} is the group's assigned contribution toward expected claim costs. Under a closed balanced risk fund, (\sum_g N_g T_g = 0). When a tax subsidy or external contribution is present, the balance becomes conditional on including that financier in the accounting perimeter. This is an operational definition for the manuscript, a bookkeeping specification rather than a new concept or normative principle.
A uniform cost loading creates another possible misunderstanding. If every member pays E[X_i] plus an equal administrative fee, naive comparison of gross premiums with claims may label everyone a net contributor. The difference finances services and capital, and the resulting premium-claim difference is not fully a distributive transfer between insured persons. Similarly, a reported net insurance payout should not ignore deductibles, copayments, excluded risks, claims administration, capital charges, or delayed payment.
Risk pooling, diversification, and correlation
The law of large numbers can reduce the relative variation of average losses under appropriate independence or weak-dependence assumptions, but a larger pool need not eliminate catastrophic accumulation or correlated exposures. If independent losses have common variance (\sigma^2), the variance of the sample mean is (\sigma^2/N). If individual losses have a common pairwise correlation (\rho) and equal variance, the variance becomes (\sigma^2[1+(N-1)\rho]/N). The first expression tends to zero as N increases; the second tends to (\rho\sigma^2) for positive common correlation. These elementary results identify a solvency constraint on pooling. A solidarity rule promises transfers only if financing, capital, reinsurance, or governmental fiscal capacity can sustain highly correlated claims.
Risk pooling is therefore both an allocative and a distributional arrangement. The same insurance contract redistributes across states, yet the pooling organization's premium and benefit rules determine whether different people have different expected net positions. WHO's assessment of pool fragmentation explicitly connects fragmentation with limited redistribution capacity, while also recommending larger and more diverse pools and compulsory participation for progress toward universal health coverage. That institutional recommendation addresses health financing; it is not a universal theorem that compulsory enrollment is optimal for every form of private insurance.
Four separate notions of benefit
A complete assessment should distinguish contractual coverage, expected actuarial benefit, realized payment, and welfare-equivalent value. Contractual coverage determines which contingencies qualify. Expected actuarial benefit averages across the probability and severity of qualifying events. Realized payment reflects actual claims and settlement. Welfare-equivalent value includes the value of reduced uncertainty, liquidity, access to timely services, and behavioral effects. These variables can move in different directions. An expensive service may raise claims without proportionately improving welfare; stringent exclusions can make apparently generous nominal limits difficult to use. A low-risk member may obtain high insurance value from protection against a severe tail loss despite a low expected indemnity.
The first conclusion is precise: insurance pooling supplies contingent protection. Redistribution between persons or groups is established only after the observer specifies the benefit measure, contribution measure, reference population, timeframe, and alternative pricing or financing rule.
Intergenerational and Intertemporal Transfers
Time, age, and cohort as different units
Insurance redistributes over time whenever participants prepay uncertain future claims, reserves accumulate before payment, or one generation finances another. These flows are analytically distinct. A young adult's contribution may help finance an older person's current medical claims. The same young adult may later receive similar protection from another set of contributors. A lifetime analysis could show substantial within-person smoothing alongside a small lifetime redistribution, or it could show durable cohort differences if demography, eligibility, coverage, productivity, or tax rules change. A single cross-sectional premium-benefit table cannot determine the long-run balance.
Let a birth cohort be c and let t index calendar periods. Define the present value of the cohort's net financed benefits at a common reference date t0 as
where B is the covered benefit, C is the combined contribution or economic tax burden allocated to that cohort, and d is the stated discount rate. A lifetime study should extend the accounting to the cohort's entire life, including earlier periods, or explicitly identify left truncation. Changing d, life expectancy, employment trajectories, or the allocation of taxes on public debt changes the present-value result. This formula is an elementary accounting expression, not a claim that one discount rate is normatively privileged.
Demographic composition is a first-order determinant of a community-rated system's observed annual transfers. Consider 100 younger members with expected annual insured costs of 2,400 each. The balanced common premium, excluding expenses and capital, is 60,000 more than their expected claims in aggregate, while older members receive $60,000 more in expected claims than they contribute. Nothing in these numbers demonstrates that the younger cohort will necessarily be a lifetime net payer. A lifetime conclusion requires records of later claims, contributions, policy changes, survival, and the financing of future benefits.
A fairness argument that privileges equal contributions at each date differs from an argument that privileges lifetime equality for people with comparable lifetime resources. Both differ from a principle of immediate protection according to present need. Such disputes cannot be resolved by expected-loss estimates alone because they concern the choice of the relevant unit and social objective.
Funded insurance and pay-as-you-go arrangements
A prefunded arrangement builds assets today to finance obligations attributable to current participants. In a simplified funded cohort, the present value of contributions, investment income, benefits, and expenses must reconcile under the valuation assumptions, including capital support or a residual surplus or deficit. In pay-as-you-go financing, current revenue supports contemporaneous benefits, with later cohorts expected to finance later beneficiaries. These are financing architectures rather than exhaustive descriptions of any country's pension, disability, or health system. Real programs can combine reserves, payroll contributions, earmarked taxes, transfers, and borrowing.
Intergenerational redistribution is particularly transparent when transition rules grant benefits to an initial generation before it contributes the actuarial value of those benefits. Later cohorts may finance that initial commitment. It also arises when a system maintains a particular benefit formula while declining contributor-to-beneficiary ratios or increasing longevity alter the total burden. Redistribution can be reversed by growth, tax reform, benefit adjustment, migration, or changes in enrollment; the sign must be calculated for each cohort under explicit assumptions.
The ILO's Convention No. 102 financing principles articulate collective finance and legal entitlements across social risks. Its institutional principles should not be treated as a guarantee that particular countries have a financially balanced lifetime cohort arrangement. The treaty sets minimum standards for ratifying states under the relevant accepted branches; operational sustainability depends on demographic and fiscal conditions and on actual national implementation.
Inflation, investment, and contingent fiscal support
Insurance promises often mature long after contributions are collected. Inflation changes nominal and real burdens, investment performance affects reserves, and changes in discount rates change the recorded present value of liabilities. If statutory benefits are indexed to wages or prices while contributions depend on a different base, the distribution among age groups and successive entrants changes over time. A government's guarantee can redistribute tail losses from a failing insurer or depleted social pool to general taxpayers, potentially including taxpayers not enrolled when the insured exposures arose.
A robust incidence exercise allocates contingent public guarantees under scenarios rather than assigning zero cost merely because no guarantee payment occurred in the observation year. It likewise avoids counting the entire face value of a guarantee as a current cash transfer. The appropriate measure depends on the analytic question: expected fiscal cost, worst-case contingent exposure, realized outlay, or welfare value of the guarantee to beneficiaries.
Retirement, annuities, and longevity pooling
Life annuities redistribute realized lifetime resources from annuitants who die earlier to annuitants who survive longer than anticipated, as part of the contract's longevity insurance. Their expected group transfers may also differ if mortality rates or prices are heterogeneous, and selection can influence who purchases annuities. This is a clear illustration of the two axes: ex post transfers among mortality outcomes and ex ante differences among risk classes. Retirement systems that combine survivor contingencies, minimum pensions, progressive benefits, and general-revenue subsidies contain several overlapping channels of redistribution. Each should be identified in separate line items before their effects are netted.
Disability insurance involves similar separation. Claimants with qualifying disabilities receive larger realized benefits, while the expected net position of groups depends on disability incidence, benefit duration, contribution history, eligibility definitions, and administrative access. A change in evaluation rules might increase spending on one group and alter labor participation or claim duration for another. Scientific inference therefore requires institutional details and behavioral analysis in addition to aggregate premiums and expenditure.
Demographic stress tests and policy transitions
A transition from uniform to experience-rated premiums may make current low-risk members better off immediately but may impose abrupt financing burdens on older or ill members whose earlier contributions helped support the common pool. Conversely, imposing new universal contributions may benefit future entrants differently from existing participants. Phase-in periods, grandfathering, public subsidies, and reserve transfers allocate these transition costs. The ethical problem is not restricted to whether the new steady-state rate is actuarially aligned. Legitimate expectations, earned entitlements, and the distribution of transitional liabilities also matter.
A practical sensitivity analysis should calculate net present values for cohorts under alternative discount rates, wage paths, mortality and morbidity assumptions, migration scenarios, uptake rates, public-finance rules, and catastrophic shocks. It should report the position of each cohort separately from a population-weighted average, since an apparently neutral aggregate result may conceal substantial differences between overlapping generations. The chapter thus places intertemporal solidarity within a real financing constraint: protection promised today requires a specified path of resources and credible governance in future periods.
Solidarity in Social and Mutual Insurance
Collective finance and institutional meaning
The institutional character of solidarity depends on the legal and governance framework. Social insurance frequently rests on legally defined participation, contribution, and benefit rules, with government responsibility or oversight. Mutual insurance typically rests on member relationships, risk sharing, and governance arrangements defined by the mutual's legal form and applicable law. Commercial insurance with community-rating regulation can contain explicit solidarity mechanisms even when the insurer's corporate form is shareholder-owned. Cooperative identity, nonprofit status, statutory compulsion, and solidarity should therefore be described as separate features before their interaction is evaluated.
The ILO minimum social-security standards emphasize state responsibility, collective financing, legal entitlement, sustainability, participatory management, and transparency. The nine branches of social security covered in the framework reflect multiple contingencies rather than a single homogeneous insurance product. Its guidance also distinguishes collective risk finance from placing investment and other risks principally on individuals. The relevance to insurance science is a publicly specified distribution of burdens and entitlements, often extending beyond those who could purchase equivalent voluntary market coverage.
The World Health Organization's pooling policy assessment explains that fragmenting prepaid finance across narrow funds limits redistribution and leaves financial exposure unevenly distributed. Its revenue-raising guidance emphasizes compulsory prepaid collective resources and the distributional importance of the revenue source. A system financed mainly by proportional payroll contributions, one financed by progressive income taxes, and one financed by consumption taxes may offer identical nominal benefits while imposing different income-incidence profiles. The purchasing and benefit allocation stages can also produce divergent actual outcomes.
Netherlands: solidarity within regulated competition
The Dutch government describes a mandatory standard health-insurance package and an acceptance obligation under which insurers must accept applicants for that standard package. Its standard insurance explanation states that the standard package is government-defined and that insurers charge the same premium regardless of age or health status for the standard insurance. It explicitly presents solidarity as a system characteristic. The appropriate interpretation concerns the nationally regulated basic package; supplemental policies and other contribution channels require their own rules.
The Dutch example separates insurer competition over service and organization from an entitlement that limits risk-selective access to basic coverage. The household's full financing burden cannot be inferred from the nominal policy premium alone: income-linked contributions, public support, and the institutional handling of expected risk costs matter. A rigorous household distributional study would therefore combine the premium, applicable fiscal contribution, public allowance, benefit use, and risk-equalization arrangements on a common basis. The government page verifies the basic legal architecture but is insufficient alone to quantify the net redistribution among income deciles or age cohorts.
This example also illustrates that uniformity of a particular premium does not imply uniformity of the total financing burden. Income-linked collection and tax-funded allowances can support a redistributive result outside the premium receipt itself. The level of insurer choice and the burden of navigating it can also influence actual uptake of entitlements; formal access and effective use are different observations.
Australia: community rating and access rules
Australian government guidance states that regulated private health insurance is community rated: insurers must generally accept eligible applicants, apply the same price to the same policy, and recognize renewal rights. These core statements appear in the Department of Health's private insurer operating rules. Product categories and lawful qualifications mean that same-price comparisons must be made for genuinely equivalent policies and relevant regulatory adjustments rather than across every available coverage package.
Community rating changes the relationship between individual expected loss and premium; it does not eliminate cost variation among products, funds, ages at entry under applicable regulatory mechanisms, or eligibility conditions. The analytic question is how the package of rating rules, demand incentives, public rebates, and risk equalization affects realized coverage and cost distribution. A one-variable comparison of headline premiums is therefore insufficient for either a favorable or unfavorable verdict on solidarity.
United States: bounded community rating and transfers among insurers
In the U.S. individual and small-group markets governed by the cited federal rules, premium differences are limited to specified factors rather than set solely by each applicant's expected health claim cost. The federal statutory rating provision allows specified variation by family composition, geography, adult age within a statutory ratio, and tobacco use within a statutory ratio, subject to the relevant statutory and regulatory framework. State law and plan-year changes must be assessed separately. This is a bounded form of community rating, not a universal flat price for every buyer.
Regulating enrollee premiums creates an incentive to attract favorable risks when some risks are more costly than rating rules allow insurers to recognize. The U.S. Department of Health and Human Services describes the risk adjustment transfer program as moving funds from issuers with lower-than-average enrolled risk toward issuers with higher-than-average enrolled risk. Transfers occur among issuers within the relevant risk-adjustment architecture, whereas premium subsidies, if applicable, act through a different financing channel. An insurer-level transfer should not be added to insured benefits as if it were a second medical claim payment; it is primarily an adjustment to the insurer's financing position.
Risk adjustment can improve the compatibility of regulated premiums with insurer solvency and competition, but it requires predictive models, reliable data, safeguards against coding manipulation, and scrutiny of residual incentives. Even an accurate average transfer cannot guarantee that every individual high-cost enrollee is equally attractive to every insurer. The fairness implications depend on which risk factors the model recognizes, how much predictable cost remains uncompensated, and the quality of care incentives.
European Union: equality constraints on classifications
The EU's Directive 2004/113/EC applied the principle of equal treatment between women and men to access to and supply of goods and services, including insurance-related rules on actuarial factors. The Court of Justice held in Case C-236/09, Test-Achats that the indefinite derogation in Article 5(2) was invalid from December 21, 2012. This ruling is significant because a risk predictor can have statistical relevance while its use in contract prices conflicts with the applicable legal norm of equal treatment.
The decision illustrates a normative choice within insurance design: a legal system can prioritize equal treatment along a protected dimension even if expected insured losses differ between classified groups. The resulting expected transfers require measurement under the actual distribution of covered persons and benefits; the judgment itself does not quantify them. It is also essential to distinguish obligations of private insurance providers under the directive from the broad institutional arrangements of statutory social security, which are governed by other legal rules and competence structures.
Mutual associations, reciprocity, and governance
The historical and contemporary idea of mutuality places policyholders or members in a common financial undertaking, potentially combining protection with member governance. Its distributional implications depend on eligibility, contribution design, surplus allocation, assessments after adverse losses, and withdrawal rules. A mutual that prices each class exactly at expected cost may provide strong state-contingent risk sharing with limited ex ante cross-subsidy. A mutual that accepts uniform dues across heterogeneous costs may implement substantial ex ante transfers. A member-owned organization may nevertheless exclude outsiders or distribute benefits regressively among those who participate.
Governance legitimacy matters because solidarity entails commitments about who shares risk with whom and on what terms. Transparent member voting, public statutory authorization, or accountable fiduciary administration provide different institutional channels through which collective obligations can be justified. They do not substitute for the verification of actual coverage and financial incidence. A fund can formally embrace solidarity while imposing barriers that prevent the most exposed from obtaining benefits. Comparative research therefore requires member-level coverage and claims data, not just the insurer's charter or public mission statement.
Lessons from the comparison
Across the institutions examined, solidarity may be implemented by common premiums, income-sensitive revenue collection, open enrollment, mandatory membership, insurer-level equalization, benefit guarantees, and public subsidies. The same policy label can denote different instruments, and identical instruments can generate different outcomes under different risk distributions and tax systems. Institutional evidence permits specific conclusions about obligations and design; quantitative claims about who gains require distributional data and incidence assumptions that go beyond the public source pages reviewed here.
Distributional Incidence
Unit of observation and the question of who pays
Incidence analysis identifies the economic agents who ultimately bear a financing burden and receive an economic benefit. Legal payment responsibility does not always determine economic incidence. An employer may remit a group premium, yet its cost may affect compensation, prices, or profits over time. A government may subsidize an insurer, yet the financing may fall on current taxpayers, purchasers of government debt, or future taxpayers. A mutual may levy assessments that reflect past claims. The incidence question requires an explicit accounting of those channels and appropriate empirical assumptions.
The relevant units vary with the research problem: individuals, households, age groups, morbidity groups, income deciles, insured and uninsured persons, insurers, firms, regions, and birth cohorts. A household may contain several people with different risk types; an employer may insure workers while benefiting from reduced turnover; a regional program may transfer resources through national taxation. Aggregation can alter the apparent pattern of redistribution. Moving from individual-level claims to household welfare, or from current-year net receipts to lifetime net present values, changes the substantive question.
Let expected insured benefit for person i be B_i. Let direct gross premium be P_i, person-level subsidy received for that premium be S_i, taxes economically borne to finance the arrangement be (\tau_i), and other obligatory contributions attributable to the person be (C_i). One useful partial net-transfer measure is
This is defined for a stated accounting perimeter. It does not include the utility value of financial risk protection, and it requires care to prevent counting an employer contribution, government premium grant, insurer transfer, or benefit twice. If P_i already represents the net premium actually paid after subsidy, S_i must not be added again. If taxes reflect all public expenditure, the share financing the specific insurance arrangement must be isolated before it is assigned to that program. If E[B_i] measures provider payments, medical services consumed or privately paid copayments should be assessed separately rather than summed as if they were identical cash receipts.
The WHO revenue-raising analysis highlights the equity effects of different revenue sources. The analysis is therefore incomplete when a policy is described as redistributive only because it visibly subsidizes the premium. Subsidies can be financed with taxes that themselves have different incidence; their net welfare and distributional effects need both sides of the ledger.
Ex ante, ex post, and lifecycle accounting
At least three tables should be produced in an empirical project. The first records expected benefits and expected contributions before the insurance contingency, stratified by observed risk and socioeconomic status. The second records realized claims and contributions over the observation period. The third discounts benefits and financing over complete lifecycles or well-defined policy horizons. Each table should retain the same membership and exposure definitions wherever comparisons are intended. Divergence across tables is informative: ex post claims concentration can reflect the operation of insurance rather than systematic pricing inequity, while a neutral current-year average can conceal a large lifetime transfer.
Survivor and attrition biases are especially important in longitudinal analysis. Older cohorts observed today contain survivors; people who died before the sample date may have contributed premiums or received benefits that are absent from the recorded period. Enrollment can also be endogenous to anticipated costs, subsidy eligibility, and economic conditions. Without these corrections, estimates can confuse changes in who remains observed with changes in the allocation rule.
Decomposition of the insurance transfer
A disciplined calculation proceeds in five steps. First, estimate expected covered loss conditional on the information set and benefit design. Second, separate the risk premium from operating expense, financing costs, taxes, and profit or surplus requirements. Third, identify cross-subsidies within the premium schedule relative to the declared reference risk classes. Fourth, add and allocate external subsidies, income-based taxes, and institutional transfers. Fifth, compare effective access, realized claims, and welfare-equivalent protection across income, risk, age, and other legitimate comparison groups.
The decomposition also prevents analytical double counting. Suppose an insurer collects a $1,250 community-rated premium from each member of the simple two-risk example and later receives a risk-adjustment payment for high-risk enrollment. The payment may change the allocation of revenues among insurers, but it does not automatically change the insured person's expected claims if the benefit contract is unchanged. If the equalization mechanism is budget neutral across insurers, total pool resources remain subject to the same collective constraint. When a government contributes new money, the relevant perimeter expands to the tax-financing system; equalization and subsidy must be recorded in distinct entries.
Distributional metrics and normative benchmarks
A program can be compared by the mean, median, and distribution of net transfers within risk and income classes; the concentration of publicly financed benefits; catastrophic out-of-pocket expenditure; access to essential services; and risk-adjusted consumer welfare. Each measure answers a distinct question. A reduction in variance of household consumption need not lower the Gini coefficient of disposable income. A lower inequality index can coexist with worse protection for a very small and severely exposed group. A higher average benefit may reflect higher input prices rather than greater service value.
Atkinson's foundational inequality analysis (1970) demonstrates that aggregation into a scalar inequality measure depends on distributive assumptions. Consequently, a claim that one insurance scheme is more equitable than another requires a defined welfare or inequality criterion, not merely an unspecified ranking. The distinction between horizontal equity (similarly situated people treated similarly under a chosen criterion) and vertical equity (appropriately different treatment across relevant circumstances) is particularly useful. The operative words are 'similarly situated' and 'relevant': identical premiums across people with different incomes may satisfy one equal-treatment norm yet produce markedly different burdens relative to resources.
A robust report should display at least the following alongside any synthetic inequality score: participation, benefit generosity, out-of-pocket exposure, uninsured risk, socioeconomic distribution of the financing burden, and uncertainty around the estimates. A single percentile comparison may conceal variation among high-risk groups, especially when disease burden, wealth, access to treatment, and age are correlated.
Identification challenges in observed insurance data
Selection and treatment effects arise because insurance participation is rarely randomized and coverage may alter behavior. Individuals may buy more insurance because they anticipate higher losses, or because they are highly risk averse and invest in prevention. Finkelstein and McGarry (2006) show that these informational dimensions can operate in opposite directions. Einav and Finkelstein (2011) review the resulting empirical challenges. Simply comparing claims of insured and uninsured persons therefore cannot establish the causal effect of insurance on health, loss prevention, household consumption, or redistribution.
Descriptive incidence studies can measure fiscal and actuarial flows without making causal claims. When causal inference is required, researchers may use appropriately justified policy discontinuities, natural experiments, randomized allocation where ethically and legally permissible, instrumental variables with defensible exclusion restrictions, or panel models with transparent assumptions. None of these techniques mechanically eliminates endogeneity. Design choice follows the specific policy variation, available data, and target estimand. The present monograph performs comparative document analysis and illustrative accounting; it does not estimate causal effects from an original individual-level dataset.
Missing transfers and incomplete records
Claims and premium datasets often omit employer-borne costs, taxes used for subsidies, uncompensated care, reinsurance and insurer-capital costs, enrollment lapses, claims rejected for documentation reasons, and household financial consequences of delayed treatment. A distributional result that ignores such variables can systematically misclassify contributors and beneficiaries. Public health expenditure includes government-financed services that may be delivered outside the insurance contract; those services must be allocated consistently with the study's definition of benefits.
Data governance adds further limits. Risk classification may use health diagnoses or other sensitive attributes under legal safeguards; small-group estimates may risk reidentification. An ethically and technically sound analysis uses appropriate aggregation, validation, disclosure controls, and secure treatment of records. Claims reported by calendar year can differ from incurred-year liabilities because settlement occurs later, while regulatory reporting can treat transfers on a benefit-year basis. The unit, date, measurement basis, and revision status should accompany every published statistic.
Sensitivity and transparency
A credible incidence finding should survive the explicit display of alternative risk baselines, eligibility assumptions, discount rates, tax-incidence allocations, counterfactual premium schedules, and cost-sharing definitions. If the sign of a transfer depends heavily on one weak assumption, the report should state this dependence. Where official documentation establishes a legal rule but does not quantify its effects, the result is a verified institutional description and an analytically specified question for future measurement, not an unsupported numerical conclusion.
The practical synthesis is to preserve two ledgers and a bridge. One ledger reports money and expected claims; another records welfare-relevant protection, access, and outcomes. The bridge documents the assumptions connecting financial transfers to welfare. The ledgers are ordinary research practices applied together here, not new theoretical systems. They prevent a change in institutional funding from being mistaken for a change in real household protection, and they prevent a valuable insurance promise from being judged solely by the last year's claims-to-premiums ratio.
Equity and Policy Design
Distributional objectives and their conflicts
Insurance solidarity can serve multiple purposes. It can insure uncertain outcomes, maintain continuous coverage after the discovery of a health condition, spread high-cost exposure across a broad membership, guarantee a minimum level of protection, stabilize household consumption, or express a civic judgment about shared vulnerability. Public authorities may additionally seek poverty reduction, labor-market participation, disaster resilience, or equal treatment of protected groups. An institutional design that advances one objective may require compromises in another; these trade-offs should be expressed as explicit choices and evaluated against evidence.
Five questions organize normative assessment. Who is included in the protected community? Which contingencies qualify for benefits? How are contributions determined and collected? Who has legal rights to protection and a fair claims process? How is solvency sustained under adverse future states? These questions turn an abstract appeal to solidarity into verifiable features of policy design. They also reveal the distributive effects of excluding applicants, limiting benefits, permitting unpriced risky activity, or imposing complex claims requirements.
Equal actuarial cost, equal treatment, and ability to pay
Actuarial equity can mean aligning expected claim cost and risk premium within an accepted classification. Equal-treatment rules can prohibit a particular classification even when it has predictive relevance. Ability-to-pay principles allocate contributions according to income or resources. Needs-based principles allocate benefits according to exposure, disability, illness, dependency, or other recognized need. These principles have different informational requirements and can be combined within one program. They cannot all be reduced to an identical price for everyone.
The European Union's Test-Achats judgment establishes one clear case in which legal equality restricts the use of a characteristic in premiums and benefits. The U.S. health-premium rules illustrate a different legally bounded set of permitted pricing dimensions. Neither example supports transplanting one jurisdiction's full statutory approach to another. A comparative recommendation should identify the applicable legal scope, market type, consumer rights, and regulatory competence.
Access and financial protection
Solidarity should be judged by effective protection. Uniform premiums can remain unaffordable for low-income households. A low contribution can coexist with a high deductible, restricted provider network, long waiting period, or narrow covered-benefit package. In such cases the intended transfer may fail to reach the person exposed to severe loss. WHO's health financing overview describes the connection between financing arrangements and access without financial hardship; its pooling guidance emphasizes risk diversification and the structural effects of fragmentation. The institutional outcome to evaluate is the distribution of protection, including the point at which a covered contingency actually becomes payable.
Fiscal subsidies may improve access but must be judged with their financing and eligibility conditions. A subsidy may be concentrated on one income band, have abrupt withdrawal thresholds, or exclude persons who cannot satisfy documentary requirements. A premium assistance program can increase enrollment yet leave remaining out-of-pocket burden high. Careful research therefore analyzes take-up, continuity, underinsurance, and household liquidity as well as the announced subsidy amount.
Risk adjustment and the integrity of collective rules
When multiple insurers compete under constrained premium differentiation, equalization mechanisms can support a broad solidarity commitment without requiring all costs to be borne by insurers with the sickest members. The CMS 2026 risk-adjustment description supplies a concrete institutional example of transfers between issuers. Performance evaluation should examine whether these payments sufficiently compensate predictable differences in enrollee costs while preserving incentives for cost control, prevention, claims accuracy, and service quality.
There are limits to compensation through predicted risk. Incomplete models leave residual cost differences; data manipulation can distort transfers; very small risk groups produce unstable estimates; and overly retrospective reimbursement may weaken cost discipline. Policymakers need to distinguish correction for exogenous health-risk composition from indemnification of avoidable overspending. Monitoring should include residual profitability by risk group, coding changes, market exit, benefit-design differentiation, and access to specialist care. The goal is an intelligible financing rule consistent with effective coverage and sustainable competition.
Incentives, prevention, and reciprocal duties
Public and mutual pools sometimes invoke reciprocal duties: contributors finance others' claims in exchange for future protection and reasonable conduct by co-members. The economic concern is that full coverage can change the marginal cost of prevention or service use. The Pauly (1968) moral-hazard analysis provides a starting point, while Arrow (1963) highlights uncertainty and institutional complexity. A policy response should distinguish preventable behavior, unavoidable health needs, provider-induced expenditure, and measurement error. Crude use of personal responsibility can produce high burdens for persons whose risk is only weakly controllable or for whom prevention is expensive.
The design of deductibles, exclusions, and participation incentives should therefore be informed by causal evidence about actual behavior and service value. Solidarity does not require ignoring incentive effects; neither does the existence of incentives establish that a restrictive benefit condition is equitable. Predictability, proportionality, accessible appeals, and procedural transparency are part of the design of a legitimate collective insurance promise.
Solvency, disaster losses, and the limits of the pool
A redistributive premium rule is sustainable only if the aggregate financing path supports liabilities. The risk-bearing pool must confront correlated catastrophes, inflation, unfavorable investment returns, adverse enrollment shifts, and concentrated high-cost claims. Reinsurance, capital requirements, reserves, and contingent public finance may protect collective commitments, but those protections have their own distributional incidence. A public backstop can reduce the immediate burden on members while shifting contingent obligations to the fiscal community.
A policy evaluation should stress-test both the financial survival of the pool and the distribution of losses after a shock. Cutting benefits to restore balance may place burdens on households least able to absorb them. Raising uniform premiums can disproportionately affect low-income members. A progressive financing supplement or targeted subsidy can alter this incidence, subject to fiscal cost and administrative feasibility. The optimal choice follows the stated welfare objective and reliable estimates; no universal financing mix is established by the comparative evidence reviewed here.
Governance, disclosure, and review
Legitimate solidarity depends on durable, intelligible obligations. Effective governance requires clarity about entry and renewal, premium-setting authority, surplus treatment, benefit adjudication, dispute resolution, fiscal guarantees, and the information participants may inspect. The ILO's collective financing and governance principles furnish an important public-law benchmark in the covered social-security context. Contractual mutual and regulated commercial contexts must be evaluated under their respective legal regimes.
Transparency includes publishing assumptions and limitations. If a ministry describes a system as solidaristic, that description can be reported as an institutional commitment; determining its actual distributional progressivity requires independent incidence work. If a premium change is justified by improved actuarial accuracy, an analysis should show whether it increases risk-based differences, changes access, creates exclusion, or transfers liabilities to public finance. These are measurable consequences that make policy rhetoric accountable.
Comparative conclusions and research priorities
The comparative cases support a restrained but substantive synthesis. Community rating changes the allocation of expected costs between risk classes; open-enrollment obligations can preserve access; income-related financing determines a separate vertical redistribution; insurer risk adjustment can support competition under constrained rating; statutory equality can prohibit particular predictive classifications; and mutual membership gives one organizational route for collective decision-making. The combined distribution can be established only through consistent observation of coverage, premium burden, tax incidence, expected losses, realized benefits, and lifetime outcomes.
Future empirical priorities follow directly. Researchers should produce harmonized country-level incidence studies with explicit accounting boundaries; distinguish short-run transfer incidence from lifetime cohort incidence; examine effective protection for excluded or underinsured persons; test the stability of risk-adjustment outcomes under data and coding changes; and document the fiscal distribution of catastrophic backstops. These are investigative tasks derived from existing methods and evidence. They do not assert a newly created theoretical model, doctrinal principle, or scientific discovery.
Conclusion
Insurance creates enforceable transfers across uncertain states of the world. Its ex ante distributional structure depends on the relation among expected claims, premiums, contributions, subsidies, taxes, and institutional financing. Solidarity is expressed through the rules that determine collective membership and the shared allocation of costs and protection. These propositions establish a coherent interpretation of insurance economics, public finance, and the institutional evidence evaluated in this monograph.
The analysis yields six substantive conclusions. First, pooling and redistribution overlap in realized payments while remaining distinguishable in expected group-level incidence. Second, community rating creates a measurable expected transfer when heterogeneous risk classes face a contribution schedule that departs from their expected covered losses, provided the accounting baseline and pool composition are specified. Third, actuarially risk-rated arrangements may provide valuable protection even with no systematic expected net transfer between risk groups. Fourth, intergenerational redistribution must be measured across complete cohort horizons; an annual cross-sectional difference between younger and older insured persons does not establish lifetime subsidy. Fifth, regulated market solidarity may be implemented through a combination of acceptance requirements, constrained rating, fiscal revenue, and transfers among insurers, each with its own distributional channel. Sixth, equitable performance depends on actual benefit access, financial protection, sustainable funding, and procedural rights as well as nominal premiums.
The institutional comparisons identify distinctive legal and economic arrangements. International social-security guidance foregrounds collective financing and legal entitlements. WHO guidance connects pooling design with financial protection and the limits of fragmented funds. U.S. law illustrates restricted rating factors and an insurer risk-adjustment mechanism; EU law demonstrates the force of equal-treatment constraints; Dutch and Australian guidance document community-rating commitments within their own regulatory systems. These examples support comparative interpretation, not universal numerical estimates of redistribution.
The dossier's original authorial contribution is a source-grounded integration of contingent insurance value, expected and realized incidence, lifetime transfers, legal equality, and the financing perimeter into one continuous analytical account. The illustrative pool examples and accounting identities clarify how to interpret transfers without presenting a new doctrinal system. Conclusions about actual redistributive magnitudes in individual jurisdictions require underlying premium, claims, income, tax, and lifecycle data. A scientific judgment about solidarity is strongest when its normative objective is explicit and every claimed transfer can be traced through a balanced financial record to the people whose protection and resources are affected.
References
Arrow, Kenneth J. (1963). "Uncertainty and the Welfare Economics of Medical Care." American Economic Review 53(5): 941-973. Original article: https://assets.aeaweb.org/asset-server/files/9442.pdf
Atkinson, Anthony B. (1970). "On the Measurement of Inequality." Journal of Economic Theory 2(3): 244-263. https://www.sciencedirect.com/science/article/abs/pii/0022053170900396
Australian Government, Department of Health, Disability and Ageing. (2025, page consulted November 14, 2026). "Private Health Insurance Policies and Inclusions You Can Offer." https://www.health.gov.au/topics/private-health-insurance/operating-rules-for-private-health-insurers-and-providers/private-health-insurance-policies-and-inclusions-you-can-offer?language=en
Centers for Medicare & Medicaid Services (CMS). (2025). "HHS Notice of Benefit and Payment Parameters for 2026 Final Rule." https://www.cms.gov/newsroom/fact-sheets/hhs-notice-benefit-and-payment-parameters-2026-final-rule
Court of Justice of the European Union. (2011). Association belge des Consommateurs Test-Achats ASBL and Others v Conseil des ministres, Case C-236/09, judgment of March 1, 2011, ECLI:EU:C:2011:100. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A62009CJ0236
Council of the European Union. (2004). Council Directive 2004/113/EC of December 13, 2004, implementing the principle of equal treatment between men and women in the access to and supply of goods and services. Official Journal of the European Union L 373. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32004L0113
Einav, Liran, and Amy Finkelstein. (2011). "Selection in Insurance Markets: Theory and Empirics in Pictures." Journal of Economic Perspectives 25(1): 115-138. https://doi.org/10.1257/jep.25.1.115
Finkelstein, Amy, and Kathleen McGarry. (2006). "Multiple Dimensions of Private Information: Evidence from the Long-Term Care Insurance Market." American Economic Review 96(4): 938-958. https://doi.org/10.1257/aer.96.4.938
Government of the Netherlands. (undated official guidance; consulted November 14, 2026). "Standard Health Insurance." https://www.government.nl/themes/family-health-and-care/health-insurance/standard-health-insurance
International Labour Organization (ILO). (2024). "Frequently Asked Questions - Social Security (Minimum Standards) Convention, 1952 (No. 102)." August 6. https://www.ilo.org/frequently-asked-questions-social-security-minimum-standards-convention
Mossin, Jan. (1968). "Aspects of Rational Insurance Purchasing." Journal of Political Economy 76(4), Part 1: 553-568. https://doi.org/10.1086/259427
Pauly, Mark V. (1968). "The Economics of Moral Hazard: Comment." American Economic Review 58(3): 531-537. https://www.jstor.org/stable/1813785
Rothschild, Michael, and Joseph E. Stiglitz. (1976). "Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information." Quarterly Journal of Economics 90(4): 629-649. https://doi.org/10.2307/1885326
United States Congress. (codified federal law, text consulted November 14, 2026). 42 U.S.C. section 300gg, "Fair Health Insurance Premiums." https://uscode.house.gov/us/usc/119-124/t42/s300gg/
United States Department of Health and Human Services. (current regulatory text consulted November 14, 2026). 45 C.F.R. section 147.102, "Fair Health Insurance Premiums." https://www.law.cornell.edu/cfr/text/45/147.102
World Health Organization (WHO). (2010; online bibliographic record consulted November 14, 2026). The World Health Report 2010: Health Systems Financing - The Path to Universal Coverage. Geneva: WHO. ISBN 978-92-4-156402-1. https://www.who.int/publications/i/item/9789241564021
World Health Organization (WHO). (undated official guidance; consulted November 14, 2026). "Pooling Revenues and Reducing Fragmentation." https://www.who.int/activities/pooling
World Health Organization (WHO). (undated official guidance; consulted November 14, 2026). "Health Financing Policy: Revenue Raising." https://www.who.int/teams/health-financing-and-economics/health-financing/health-financing-policy/revenue-raising