RESEARCH DOSSIER
Comparative COSO and ISO 31000 Frameworks for Insurers
Abstract. Enterprise risk management in insurance is an institutional system for deciding which contingent obligations an insurer will accept, how the insurer will finance and diversify them, which risk limits govern strategic action, and how it will demonstrate its continuing capacity to meet policyholder obligations. The 2017 COSO enterprise risk management framework and ISO 31000:2018 both make risk management integral to organizational decisions. They differ in their analytical emphasis, architecture, terminology, and the kinds of practical evidence their application makes readily visible. This monograph undertakes a source-critical international comparison grounded in the original framework publishers, the International Association of Insurance Supervisors (IAIS), European Solvency II provisions, United States National Association of Insurance Commissioners (NAIC) ORSA materials, Australian Prudential Regulation Authority requirements, and peer-reviewed insurer ERM research. The central finding is that COSO's strategy–performance–governance articulation and ISO 31000's principles–framework–process articulation can provide complementary evidence of an insurer's risk decisions, but neither generic framework replaces a jurisdiction's prudential duties, quantitative capital tests, ORSA, control-function independence, or legally accountable governing body. An insurer's actual supervisory fit is determined by demonstrable decisions and controls at the appropriate legal-entity and group levels. The comparison explains COSO's 2004-to-2017 transition, ISO's context-based treatment process, the separate status of international supervisory principles and binding domestic law, challenges of risk appetite and aggregation, differences among insurer business models, and limits of observational evidence on ERM value. A detailed qualitative comparison and worked hypothetical exercises establish transparent decision tests without claiming a new theory, named model, or causal estimate. Empirical literature on ERM adoption cannot establish that one of these two frameworks outperforms the other. The research contribution is an original, traceable synthesis of institutional responsibilities, decision evidence, methodological fit, and scope conditions for using the frameworks in insurers.
Keywords: COSO; ISO 31000; enterprise risk management; insurer risk governance; risk appetite; insurance solvency; risk management framework; strategic performance; own risk and solvency assessment; ORSA; IAIS Insurance Core Principles; ICP 8; ICP 16; Solvency II; NAIC Model Act 505; APRA CPS 220; internal controls; risk aggregation; insurance group; underwriting; reinsurance; asset–liability management; governance evidence; organizational risk culture; comparative institutional analysis.
Introduction
Insurance contracts convert uncertain future events into present contractual obligations. The enterprise receives premiums and invests assets while maintaining reserves, capital, operating resources, legal rights against reinsurers, and access to liquidity. Its exposures interact. Catastrophe claims may coincide with asset-price stress; inflation may increase claims severity while changing interest-rate assumptions; an emerging liability risk may affect underwriting, reserving, litigation, market conduct, and reputation at once. The risk system must therefore preserve the connection between an accepted insurance obligation and the entity that ultimately bears its financial consequences.
Organizational risk guidance became influential partly because functional divisions often manage their own exposures with different measures and horizons. Underwriters examine expected loss and pricing; actuaries examine ultimate liabilities; asset managers examine returns, duration, and liquidity; legal and compliance functions evaluate obligations; finance evaluates capital and reported earnings. These are legitimate specialized perspectives. The enterprise needs a mechanism by which a local decision can be evaluated under a shared objective, risk tolerance, relevant uncertainty, and accounting and solvency constraints.
The COSO ERM guidance and ISO 31000:2018 guidelines address that organizational problem through different presentations of risk management. COSO foregrounds mission, strategy, performance, governance, and reporting. ISO foregrounds principles, the organizational framework, and a repeatable process for establishing context, assessing and treating uncertainty, communicating, monitoring, and recording results. The comparison concerns the evidentiary consequences of those presentations for insurers: what decisions they induce management to document, which questions supervisors can answer, and where additional insurance-specific legal and actuarial requirements must be supplied.
The study uses a documentary and analytical method. Its primary comparator set comprises official COSO and ISO pages, an official December 2024 version of IAIS Insurance Core Principles, the published EU Solvency II Articles 44 and 45, NAIC ORSA guidance and official explanatory materials, and APRA CPS 220. Empirical propositions are drawn from identified peer-reviewed studies. Assertions about a document are confined to the edition or publisher material actually reviewed; publisher summaries of proprietary standards receive narrower evidentiary treatment than complete accessible legislation. Illustrative scenarios are labeled and are never presented as observed firm outcomes.
Three questions organize the inquiry. First, how do the concepts and historical architecture of COSO and ISO differ, and which similarities withstand close comparison? Second, through which operational decisions could an insurer translate either framework into strategy, risk acceptance, risk response, capital, and governance? Third, what can regulatory requirements and empirical evidence establish about the validity, adoption, effectiveness, and limits of those translations?
The subject is deliberately placed beside the published Enterprise Risk Management in Insurance, Insurance Governance and Market Conduct, and Insurance Regulation Science areas. This dossier examines the framework comparison itself, rather than restating a general treatise on insurance capital, actuarial validation, reinsurance optimization, or regulatory implementation. Cross-cutting examples serve to test the frameworks' coverage and the traceability of organizational decisions.
Units of analysis and evidentiary standard
The analytical unit can be a regulated insurer, a regulated insurance group, a wider conglomerate, an underwriting program, or an individual strategic decision. These units must be distinguished. A group can possess an excellent enterprise dashboard while a single subsidiary lacks sufficient capital or faces a local liquidity constraint. A common risk appetite statement can help group-wide communication without extinguishing a subsidiary board's legal obligations. The research treats regulatory scope as an attribute of the applicable law, not as a universal property of the framework.
A source-based comparison has four evidence levels. The first is the text of a law, official prudential standard, or an authoritative organizational framework within its accessible scope. The second is the issuing institution's explanation of implementation or version history. The third is empirical research reporting methods and findings for defined samples. The fourth is an explicitly reasoned analytical inference about possible organizational mechanisms. These levels are connected by argument, while remaining separate in attribution and certainty. In particular, regulatory conformity and market valuation are different outcomes with different evidence requirements.
Definitions, History, and Documentary Foundation
Enterprise risk management as a decision institution
For the purposes of this study, enterprise risk management means an institutionally organized set of responsibilities, processes, information flows, decision criteria, and corrective mechanisms through which an insurer evaluates and manages material uncertainty relative to its obligations and objectives. This is a working analytical definition, not a newly proposed theory or a claim to amend COSO, ISO, or IAIS terminology. It emphasizes the features that can be observed: who decides, what information is used, how alternatives are assessed, which limits bind, and how an adverse result is escalated.
A prudent comparison must distinguish risk from its numerical representations. A capital figure, loss scenario, risk register entry, projected combined ratio, or qualitative heat-map rating represents only part of the underlying uncertainty. It may omit dependencies, longer horizons, actions taken by counterparties, or severe but poorly observed events. Risk governance determines how those partial representations enter an actual decision. Insurer-specific economics makes the connection especially important: Froot (2007) links risk allocation, product and capital market frictions, underwriting, reinsurance, and capital policy. The insurer's choice to accept a contract has consequences for surplus, external risk transfer, cash flows, and potential financial distress.
COSO's organizational history includes the 2004 Enterprise Risk Management—Integrated Framework and the 2017 Enterprise Risk Management—Integrating with Strategy and Performance update. The official COSO publication history records this revision as a response to a more complex risk environment, growing board oversight expectations, and the demand to integrate risk management with strategic decisions and reporting. The conceptual significance of the change is that risk is examined when an enterprise chooses objectives and strategies, rather than considered only after activities have already been authorized.
The ISO sequence includes ISO 31000:2009 followed by ISO 31000:2018, the second edition issued in February 2018. ISO reports that this edition was subsequently reviewed and confirmed in 2023 and that revision work is underway; a prospective revision is not a substitute for the presently identified edition. ISO describes a general-purpose guideline usable across sectors and organizational sizes. Its official page makes a particularly important institutional distinction: the standard itself does not provide a certification scheme. An insurer's claim to hold an “ISO 31000 certificate” requires scrutiny of exactly what a private assessor certified, because the standard is guidance rather than a certifiable management-system requirement.
The two objects of comparison
The COSO and ISO documents have a large common field: integration with objectives, leadership responsibility, continuous adaptation, communication, and organization-wide coordination. There are also substantive differences in emphasis. COSO presents an articulated relationship among governance and culture, strategy and objective setting, performance, review and revision, and information, communication, and reporting. ISO presents principles, a management framework, and a process intended to be embedded, tailored, iterated, and communicated throughout activities. The difference is especially visible when management explains why a strategic risk was accepted: COSO draws attention to the strategy–performance rationale, whereas ISO demands clarity about context, criteria, assessment, treatment, and monitoring.
Neither document is insurance legislation. Neither automatically establishes compliance with a local solvency calculation, technical provision regime, actuarial function mandate, ORSA filing, or insurer board's fiduciary and statutory responsibilities. Their relevance to an insurer is mediated by the documented organization and the jurisdiction in which the insurer operates. The IAIS ICPs and ComFrame constitute an international supervisory benchmark; national or regional instruments translate supervisory expectations into their legally applicable forms. A governance method may be excellent and still require specific operational enhancements to meet that translation.
Historical asymmetries and source editions
The founding date of a framework, its current edition, and the effective date of a supervisory requirement answer different historical questions. The COSO 2004 text and COSO 2017 revision are not interchangeable descriptions of one timeless framework. The ISO 2009 and 2018 editions similarly reflect a documented revision. The supervisory context also changes: IAIS provides a December 2024 revised ICP package; EU Solvency II is read through identifiable directive provisions and amendment history; US ORSA requirements depend on adopted state versions of Model Act 505; APRA identifies July 1, 2019 as the in-force date of CPS 220. A comparative claim that simply says “under current ERM requirements” loses this legally material chronology.
The IAIS online supervisory tool identifies the December 2024 changes to ICPs and ComFrame. The study uses that edition for its international benchmark. The ISO 2018 edition remains identifiable as the published standard despite prospective revision. NAIC materials also require date discipline: a publicly indexed 2026 ORSA file carries a December 2026 cover date, which is later than this manuscript's November 15 research cutoff. This study does not treat that future-dated edition as already operative. Instead, it anchors implementation details in the dated 2025 guidance and the NAIC's public ORSA topic description, while recognizing that later changes can require fresh application-specific review.
Four forms of adequacy
A comparison is more precise when it asks four separate questions. Conceptual adequacy concerns whether management has a coherent vocabulary and view of risks. Procedural adequacy concerns whether decisions are assessed, treated, reviewed, and recorded. Prudential adequacy concerns whether the insurer satisfies applicable legal, capital, liquidity, governance, and ORSA obligations. Empirical effectiveness concerns whether the process actually improves outcomes under observable conditions. A checklist derived from COSO or ISO may be informative about the first two; it cannot alone prove the latter two. This distinction informs the chapters that follow.
COSO Framework Evolution
The strategic turn in the 2017 framework
COSO's principal contribution to the present comparison is its explicit treatment of strategy and performance as part of ERM. The 2017 COSO framework organizes its account around five interrelated components: governance and culture; strategy and objective setting; performance; review and revision; and information, communication, and reporting. The underlying 2017 publication specifies twenty principles across these components. The architecture is useful for insurers because strategic objectives and risk acceptance are inseparable: an underwriting growth target, a capital return target, or an asset-allocation decision affects the capacity to pay claims under adverse conditions.
The historical comparison with the 2004 framework should be substantive. An insurer might previously have mapped business objectives, events, control activities, and monitoring onto a multi-component enterprise control architecture. In a 2017-aligned analysis, management must show how alternatives to a proposed strategy were considered and whether accepted risk is consistent with enterprise performance objectives. This does not mean that earlier COSO approaches ignored objectives. It means that the newer publication makes the relationship between strategy choice and the distribution of possible outcomes unusually explicit. The design of board packs, executive incentives, and post-decision reviews follows from that emphasis.
Consider an illustrative property insurer proposing rapid expansion into a coastal region. The conventional business case compares anticipated premiums, expenses, expected loss, acquisition costs, and capital consumption. A COSO-oriented ERM case also asks the board to consider alternative growth strategies, catastrophe accumulation, reinsurance terms, correlation with current exposures, constraints on claims-handling capacity, the insurer's financial and cultural incentives, and the effect of a severe event on enterprise objectives. The material improvement comes from requiring evidence that strategic alternatives were evaluated together with their potential downside, rather than appending a generic risk matrix to an approved expansion plan.
Governance and culture
Governance is operationally concrete. The board approves or oversees the relevant appetite and strategic boundaries; executive management allocates authority; specialized risk and control functions provide challenge and independent perspectives; management owns the risk consequences of day-to-day decisions; and information systems make those consequences visible. COSO's governance component is analytically useful when defining the accountability chain from a proposed exposure through underwriting authority to enterprise limits, senior management escalation, and the board.
Culture becomes observable through decisions. A board's statement that it prizes disciplined underwriting acquires evidentiary meaning when bonus structures, exception logs, renewal decisions, product withdrawals, escalation records, and remediation all support that claim. Conversely, a company can have a formally elaborate ERM charter while allowing repeated limit exceptions without documented deliberation. The argument is not that culture is reducible to one metric. It is that a culture claim should generate a pattern of institutional behavior open to review.
The supervisory baseline makes those observations consequential. The IAIS risk-management principles place corporate governance and independent internal-control responsibilities within an insurer's sound functioning. In a COSO implementation, the board's strategic oversight and the legal responsibilities of actuarial, compliance, risk-management, and internal-audit functions must be aligned without conflating management ownership with independent review.
Strategy, objectives, and risk appetite
Risk appetite links willingness to take risk with the pursuit of objectives; risk tolerances and operating limits translate that willingness into constraints for particular risks or portfolios. It is tempting to express appetite only as a single solvency ratio. Yet solvency is one of several dimensions of acceptable exposure. An insurer may maintain a satisfactory aggregate solvency position while accepting a catastrophe concentration, a severe short-term liquidity vulnerability, conduct harm, or operational disruption that violates the board's tolerances.
An analytically sufficient appetite statement therefore connects qualitative and quantitative conditions. It can express a threshold for stressed capital coverage, an exposure concentration bound, a reserve deterioration trigger, a liquidity horizon, outsourcing or cyber-risk tolerances, limits on product conduct risk, and an explicit escalation path. Those variables require consistent measurement dates and confidence or scenario assumptions. The objective is that management can identify an approaching limit before a strategy has irreversibly increased exposure.
COSO's strategic orientation asks whether appetite constrains strategic alternatives and whether strategic choices change appetite's practical meaning. For example, expansion into a long-tail liability line can change the capital horizon and model uncertainty. A fixed numerical capital buffer copied from a prior short-tail portfolio may be an inadequate expression of the revised risk context. The governance decision is whether to accept a different distribution of exposures and establish limits and resources suitable to that acceptance.
Performance and portfolio view
The performance component is valuable where decentralized decisions affect a common risk portfolio. The insurer may calculate exposure severities by product, geography, legal entity, asset class, counterparty, and time horizon. Management then needs an aggregated view and an intelligible explanation of the dependence assumptions. Diversification credits are decision-relevant only insofar as the underlying exposures and extreme-event dependencies justify them. An aggregate capital number alone is an opaque summary when underlying catastrophe, lapse, spread, credit, or liquidity shocks can interact.
Suppose two business units independently propose additions to risk limits. Each proposal may satisfy its unit-level threshold, while both depend on the same regional economy or reinsurer. A COSO-informed portfolio review asks whether the combined increase is consistent with enterprise objectives and who owns the cross-unit constraint. The test is whether shared information changes the approval decision when it should. The existence of a central risk committee, by itself, does not establish this result.
Performance review must also distinguish leading from lagging indicators. Claims emergence, reserve development, exposure accumulations, investment spread exposures, control failures, and consumer complaints can provide early warning before year-end capital or earnings measures register the full consequences. Indicators become useful when trigger levels, data sources, responsible owners, and response deadlines are specified. A monitoring dashboard becomes part of governance when its alerts have actual decision consequences.
Review, revision, and reporting
The premise of review is that organizational decisions are revised when conditions or understanding change. A new legal interpretation, a catastrophe model revision, a treaty renewal, a technology failure, or a shift in inflation expectations can make a prior assumption obsolete. A mature system records the assumption, its owner, the evidence of change, the new assessment, the chosen action, and the reason for accepting any residual exposure. This is a practical audit trail from information to decision.
The information, communication, and reporting component is particularly important in insurance groups because numerous legal entities and functional departments report incompatible metrics. A risk report should disclose its aggregation boundary, measurement assumptions, time horizon, limitations, and connection to board decisions. A consolidated group report may summarize exposure but must preserve material subsidiary-specific constraints. The correct level of aggregation is determined by the decision being made.
COSO's strengths and limitations for insurers
The 2017 articulation is especially strong as a diagnostic for strategic accountability: whether management assessed alternatives, linked risk to objectives, adopted appetite consistently, reviewed real performance, and furnished decision-quality information. This is a comparative interpretation of the framework architecture, not a measured claim that insurers adopting COSO become more profitable. Its limitations arise from the generality that gives it wide use. It does not itself supply insurance-specific reserving methodology, a capital confidence standard, reinsurance credit assessment, ORSA methodology, or legal criteria for independent control functions.
An insurer that uses COSO as the highest-level organizing vocabulary still needs a rigorous process for identifying, analyzing, treating, communicating, and periodically reassessing each material exposure. That need opens the comparison with ISO 31000.
ISO 31000 Risk Process
Principles, framework, and process
The ISO 31000:2018 standard is organized around risk-management principles, an organizational framework, and a risk-management process. ISO's official description stresses organizational integration, tailoring to context, continual improvement, and management of uncertainty across decisions. The value of this architecture to an insurer is its disciplined attention to the context and criteria of a specific decision, from initial identification through action and reassessment.
The three dimensions should be read together. Principles articulate expected properties of sound risk management. The framework concerns leadership, integration, organizational arrangements, evaluation, and improvement. The process concerns communication and consultation, scope/context/criteria, risk assessment, risk treatment, monitoring and review, and recording and reporting. A process performed outside governance can become mechanical; a framework without functioning assessments can become ceremonial. Their interaction is the basis of the ISO comparison.
ISO supplies an introductory risk-management brochure that contextualizes the standard's broad application. The authoritative detailed text is a copyrighted standard sold or previewed through ISO, and this research does not claim licensed full-text access to every clause. Accordingly, detailed clause-by-clause compliance assertions are outside the verified scope. The structural and practical analysis here rests on the publisher-confirmed principles, framework/process account and independently corroborated risk-management practice.
Establishing scope, context, and criteria
An insurance risk assessment starts with a precise question. Examples include whether to launch a cyber product, enter a new jurisdiction, renew a catastrophe treaty, adjust annuity hedging, outsource claims administration, or expand a bond allocation. Each question determines its unit of analysis, timeframe, affected obligations, dependencies, stakeholders, and relevant risk criteria. Without that specification, risk ratings across proposals may appear comparable when they measure different things.
For a property catastrophe underwriting proposal, context includes insured locations, perils, exposure period, vulnerability assumptions, geographic correlations, existing catastrophe concentrations, claims inflation, treaty wording, counterparty credit, legal-entity boundary, market conditions, and the maturity of data. Risk criteria might include expected underwriting return, downside loss distribution, catastrophe exceedance measures, liquidity required immediately after an event, solvency buffer, and authority thresholds. The relevant risk is then understood relative to objectives and to capacities for response.
The treatment of context differs meaningfully between a claims-processing outsourcing decision and an asset portfolio trade. Outsourcing places weight on operational continuity, third-party concentration, information security, customer fairness, and contingency management. Portfolio changes place weight on market, credit, liquidity, valuation, asset–liability matching, and concentration. A universal scorecard that gives every category the same numerical scale obscures these differences. ISO's tailoring orientation encourages management to justify the criteria, not merely to apply them.
Identification, analysis, and evaluation
Risk identification requires recognizing events, sources, drivers, consequences, and possible responses, including exposures created by strategic decisions themselves. For insurers, the relevant data include policy terms, aggregate insured values, claims histories, reserving triangles, modeled catastrophe loss, asset holdings, liabilities, reinsurance placements, vendor dependencies, and regulatory reporting. These data have different limitations and cannot be reduced to one homogeneous observation set.
Analysis then evaluates the nature and potential consequences of the identified uncertainty. Some risks permit probability models with sufficient experience data; others demand stress scenarios, reverse stress tests, expert assessment, qualitative controls review, or structured sensitivity analysis. The IEC 31010 assessment guidance establishes the role of selecting and applying assessment techniques under uncertainty. Its existence reinforces the distinction between an organizational risk guideline and the concrete techniques used in individual assessments.
Evaluation compares results against decision criteria. Expected loss may be relevant for pricing; tail loss and capital consumption may be relevant to solvency; near-term cash demand may be decisive for liquidity; a regulatory prohibition may make acceptance impermissible regardless of expected profit. A credible evaluation thus keeps several outcome measures visible. It avoids transforming all risk criteria into one scalar score whose weights are neither approved nor defensible.
Analysis must also consider epistemic uncertainty. Historical claims cannot always represent an emerging peril; catastrophe models can have structural errors; reinsurer recoverables depend on legal enforceability and counterparty solvency; operational incidents are often sparse and heterogeneous. The institution's risk decision should reflect both modeled uncertainty and limitations in the evidence from which the model is built.
Risk treatment as an insurance decision
Treatments can include limiting an activity, altering contractual terms, repricing, diversification, purchasing reinsurance, hedging financial exposure, strengthening controls, increasing capital or liquidity, or declining a proposed risk. Different treatments transfer, transform, defer, or retain different parts of the exposure. Reinsurance may lower net underwriting loss while creating counterparty and basis risk. An investment hedge may reduce market sensitivity while requiring collateral cash under stress. Outsourcing may alter operational capabilities without removing the insurer's responsibility to policyholders.
A useful ISO-oriented decision record includes the untreated exposure, considered alternatives, assessed treatment effectiveness, residual risk, responsible decision-maker, implementation date, contingency measures, and review trigger. If a treatment is approved, the operational evidence should show that the contract was executed, systems changed, exposure limits updated, or control deployed. An action plan that never reaches implementation is incomplete risk management.
Treatments require evaluation for unintended effects. A lower deductible or broader policy limit may improve sales but increase correlated loss. A cheaper reinsurance treaty may contain narrower event definitions. A capital-protection trade may increase exposure to forced asset sales. A centralized vendor can improve efficiency while creating common-cause outage risk across insurance subsidiaries. This is where ISO's process-oriented analysis connects to COSO's organization-wide strategic and portfolio review.
Communication, consultation, and reassessment
Risk criteria may be developed by risk specialists but require input from people who understand the contracts and operations. Underwriting, claims, actuarial, treasury, legal, compliance, technology, and the business owner may each possess essential information. Consultation is analytically valuable because it can reveal hidden assumptions, conflicts between departments, and unavailable data. It is effective when a decision record identifies who contributed and what unresolved disagreement remained.
Communication changes the usefulness of a calculation. A catastrophe model output should identify the model version, exposure snapshot, peril definition, simulation horizon, financial terms, reinsurance assumptions, and model uncertainty. A statement that one portfolio is “high risk” conveys much less information. An escalation notice should identify the breached criterion and available actions. When data can be revised materially, the record should make the revision and its governance consequence traceable.
Monitoring and review ask whether relevant conditions and treatments remain valid. An insurer may set quarterly reviews for strategic exposures, continuous controls for operational thresholds, and event-driven reviews following severe losses or regulatory change. Recording and reporting connect the risk process to institutional memory and supervisory examination. The decision history is part of the risk-management asset: it reveals whether management repeatedly saw a problem, evaluated alternatives, and acted.
ISO's strengths and limitations for insurers
ISO 31000 is particularly useful for the systematic specification of decision context, criteria, assessment, treatment, and reassessment. This makes it adaptable to diverse insurance products, entity structures, maturity levels, and emerging risks. The same generic quality creates limits. ISO alone does not determine the amount of capital legally required, the applicable insurer actuarial standards, the definition of ORSA, the governance of statutory key functions, or whether a risk appetite statement satisfies a particular supervisor.
The ISO publisher's own description states that ISO 31000 is guidance and cannot itself be used for certification. External independent assessments may be informative, but the standards' legal and evidential status should be described exactly. A practical ISO process can support insurance supervision when its outputs are incorporated into accountable decisions and backed by jurisdiction-specific requirements. The next chapter tests this connection.
Fit with Insurance Supervision
International standard-setting and legal force
The IAIS ICP and ComFrame framework is the global supervisory reference for insurance regulators and internationally active insurance groups. ICP 7 concerns corporate governance, ICP 8 concerns risk management and internal controls, and ICP 16 concerns enterprise risk management for solvency purposes. In the December 2024 ICP text, ERM for solvency includes coordinated identification, measurement, reporting and management of insurer risks, risk appetite, risk limits, capital adequacy, asset–liability and underwriting policies, group risks, ORSA, and recovery planning.
This international framework performs a different function from COSO or ISO. It defines supervisory expectations for insurers and guides regulators toward consistent assessment. It does not automatically enact identical private-law obligations in every jurisdiction. National and regional law determines the enforceable duties of particular legal entities and the procedural obligations of the supervisory authority. A COSO- or ISO-aligned insurer should therefore demonstrate substantive concordance with the applicable ICP-related domestic requirements, rather than treat a generic reference to the ICPs as proof of compliance.
ICP 8 demands an effectively governed risk and internal-control system, including relevant control functions. ICP 16 focuses more explicitly on insurer solvency, material risk and capital relationships. The comparison reveals a division of labor: the generic frameworks help management organize judgment and processes; supervisory provisions specify financial-resilience, legal accountability, and sector-specific outcomes. The same risk event may therefore require strategic assessment under COSO, process assessment under ISO, and solvency evidence under the relevant ICP implementation.
European Union: risk management and ORSA
Solvency II Article 44 requires insurance and reinsurance undertakings to maintain an effective risk-management system that identifies, measures, monitors, manages, and reports risks on a continuous basis at individual and aggregated levels, including interdependencies. It places that system inside the undertaking's organization and decision processes and encompasses both risks included in the Solvency Capital Requirement and risks inadequately captured by its calculation. This is a legal demand for substantive operational integration.
Solvency II Article 45 sets the ORSA obligation as part of the insurer's risk-management system. It addresses overall solvency needs in light of the specific risk profile, approved tolerances, and business strategy; continuing compliance with capital and technical provision requirements; and the degree to which the company's risk profile differs from assumptions underlying the capital calculation. These are distinct inquiries. A company with a compliant regulatory capital ratio can still have material unmodeled risk and elevated own solvency needs.
The EIOPA ORSA guidelines add a dedicated supervisory guidance layer, with an official publication record dating to September 2015 and later materials identifiable on the publication page. The editorial distinction matters: a directive article, regulatory technical implementation, EIOPA guideline, and company's internal COSO or ISO procedure have different legal status and evidentiary function. A valid comparative study must retain that hierarchy.
COSO can strengthen the ORSA's strategic rationale, showing how alternative strategies, appetite, and financial outcomes interact. ISO can strengthen the ORSA's process traceability, specifying scope, assessment methods, residual risk, treatments, monitoring, and changes in assumptions. An effective EU implementation uses both types of evidence while retaining the prescribed ORSA governance, reporting and prudential measurements.
United States: state law, model requirements, and confidential reporting
The NAIC's ORSA topic guidance describes the Risk Management and Own Risk and Solvency Assessment Model Act (Model 505), introduced as part of the US solvency modernization initiative, and explains its adoption through state insurance laws. The NAIC page states general premium thresholds of more than $500 million annually for individual insurers and more than $1 billion for groups, subject to the model's exemptions, regulatory discretion, and the enacted requirements of each state. Covered insurers conduct ORSA at least annually and provide a confidential summary. A model act has legal effect through the jurisdiction's adoption and amendments, not by publication of the model alone.
US ORSA links enterprise risk management with prospective solvency. The 2025 ORSA manual details updated expectations touching the scope of captives, first reporting after thresholds are crossed, internationally active groups, and the debt-service capacity of an insurance group. These issues illustrate how a process designed for one entity must be carefully reconciled with holding-company structure, intercompany exposures, and the actual location of capital and liquidity.
The broader NAIC ERM architecture incorporates ORSA, Form F enterprise risk reporting, corporate governance disclosure, internal audit requirements, and risk-based capital tools. The relationship to COSO and ISO is one of practical compatibility and complementarity, not formal replacement. A report describing COSO governance components or an ISO risk process does not automatically satisfy ORSA filing, ownership, scope, attestation, or confidentiality obligations.
For a US insurer, the relevant question is whether the chosen generic framework creates defensible records of risk culture and governance, identification and prioritization, appetite and limits, management and controls, and communication, and whether those records connect to the ORSA summary and underlying company analysis. The supervisor remains interested in the insurer's actual risk exposure and prospective financial condition.
Australia: explicit board accountability
The Australian CPS 220 standard, identified by APRA as in force from July 1, 2019, applies across covered APRA-regulated institutions including general, life, and private health insurers. It requires a board-approved risk appetite statement, a risk management strategy and framework consistent with business strategy, a board-approved business plan, accountable risk functions, and mechanisms for managing and reporting material risk. It also makes the board ultimately responsible for the framework's appropriateness and requires material framework breaches to be addressed through specified supervisory notification.
The Australian example shows where the difference between organizational guidance and binding supervision is clearest. COSO's governance and culture component can help define board oversight and institutional behavior. ISO's framework and process can help define criteria, treatment, and escalation. CPS 220 independently specifies mandatory structures, risk-appetite approval, and the relationship to the business plan. A board cannot discharge those obligations merely by announcing alignment with either external framework.
The international comparison must be proportional. A small specialist insurer, a large multiline stock company, a mutual insurer, and a globally active group have different exposure complexity and resources. Their regulatory frameworks likewise differ. Proportionality permits appropriately scaled arrangements; it does not eliminate the need to identify material risks or to assign legal responsibility for decisions.
Crosswalk: institutional questions and evidence
| Supervisory question | COSO-emphasized inquiry | ISO 31000-emphasized inquiry | Insurer-specific required evidence |
|---|---|---|---|
| Who owns risk? | Governance, culture, decision authority | Leadership, roles, consultation | Board mandates, control functions, legal-entity duties |
| Why accept risk? | Strategy, objectives, appetite, alternatives | Scope, context, criteria | Underwriting authority, capital and conduct constraints |
| What can go wrong? | Portfolio performance and enterprise view | Identification, analysis, evaluation | Insurance stress scenarios, reserving, dependencies |
| What changes exposure? | Risk responses linked to performance | Treatment choices and residual risk | Reinsurance, limits, hedging, control effectiveness |
| When is response required? | Reporting, review and revision | Monitoring, reporting and reassessment | Limit breaches, ORSA triggers, supervisory notifications |
| How is resilience established? | Strategy and performance consistent with appetite | Evidence of process and treatment | Own solvency needs, SCR/RBC or jurisdictional capital, liquidity |
The table is an interpretive crosswalk constructed from verified primary provisions. It does not purport to reproduce an official COSO–ISO mapping or a supervisory equivalence determination. It supports a testable proposition about organizational evidence: when a material insurance decision is sampled, records should show the governing objective, the relevant risk criteria, the alternatives evaluated, the residual risk, and its legal-entity solvency implications. Those records can be compared against the actual jurisdictional provisions without assuming that one generic framework guarantees compliance.
International Uptake and Limitations
What adoption research actually measures
Empirical work on enterprise risk management can illuminate why insurers adopt more coordinated approaches and whether adoption correlates with firm value. Yet most observational studies do not identify whether a company specifically applied the 2017 COSO framework or ISO 31000:2018. Many study different eras, proxies, ratings, or disclosure-based classifications. Their results cannot be repurposed as direct evidence that one framework produces higher insurance returns.
Hoyt and Liebenberg (2011) study US insurers and report a roughly 20% valuation association for firms classified as practicing ERM in their specification. That is an important signal of potential economic significance. It is not a randomized estimate of the effect of adopting COSO, ISO, or any specified framework edition, and Tobin's Q is a market-based outcome influenced by expectations, organization, portfolio mix, and measurement.
McShane, Nair, and Rustambekov (2011) use Standard & Poor's ERM assessment information for insurers. They report a positive relationship between increasing conventional risk-management capability and firm value but no additional value associated with the higher ERM rating in their sample. The contrast is analytically informative because it separates the consequences of substantive risk-management capability from the assumption that all higher organizational maturity labels must command a premium. The studies employ different exposure measures and designs, so a direct numerical comparison of their estimates would be invalid.
Baxter and colleagues (2013) examine ERM program quality and value relevance in financial services, including crisis context; their research makes quality measurement and heterogeneity central to interpretation. The Gatzert and Martin (2015) review synthesizes empirical ERM adoption and value studies and finds that data and model specification differences, industry and geography limit generalization. These results support a disciplined account of uncertainty rather than an unconditional adoption premium.
Why selection and causality matter
Insurers may choose more comprehensive ERM because they are larger, internationally active, already profitable, under pressure from supervisors, or exposed to complex risks. Those same characteristics influence firm value and operational outcomes. A regression association between disclosed ERM and Tobin's Q therefore combines multiple possible mechanisms: real improvements in risk selection, better governance, investor confidence, superior reporting, existing organizational capability, and selection into the reporting category.
A credible causal study of COSO versus ISO would need verified adoption dates, actual implementation scope, comparable insurers, stable outcomes, information about concurrent regulatory changes, and detailed evidence about how adoption changed decisions. Ideally, it would distinguish adoption from compliance and execution. A public announcement that a company is “COSO-aligned” is not equivalent to independent evidence that a board materially altered risk acceptance or capital policy. A similar issue arises when an ISO-alignment statement describes an administrative assessment without showing how actual controls performed.
Observed outcomes should include more than shareholder valuation. Possible dimensions include regulatory solvency margin and variability; capital requirement breaches; reserve development; catastrophe and counterparty concentrations; claims settlement performance; consumer harm; liquidity resilience; limit breaches and escalation time; and the speed and reliability of remediation. The sign and magnitude of a framework's association with one outcome may differ from another. A framework that reduces an excessive risk appetite could lower short-run growth while improving policyholder security.
Jurisdiction and entity heterogeneity
The same generic framework travels across countries, yet its operational meaning depends on the host jurisdiction. An EU insurer's ORSA has explicit Solvency II Article 45 functions. A US insurer's ORSA report depends on applicable state law and model-act implementation. An Australian institution's risk appetite and board requirements derive from CPS 220. These are different mechanisms of supervision and should remain distinct in empirical classification.
Differences in company structure are equally material. A mutual insurer may emphasize policyholder protection and long-term stability; a listed insurer may balance shareholder returns and contractual security; a reinsurer can be particularly exposed to aggregate catastrophe, retrocession, and extreme-event modeling; a life insurer faces duration, options, lapse and longevity risks; a specialist cyber underwriter may face correlated and poorly observed events. Framework language may remain constant while the evidence needed for a competent decision changes.
A group-level risk management framework can conceal legal-entity frictions. Surplus cannot always be transferred without restriction, reinsurance recoverables may have different legal terms, and a parent company's liquidity cannot be treated as freely available to every subsidiary. Governance analysis should therefore describe internal group boundaries, applicable jurisdictions, ring-fencing, risk transfer, and the location of claims-paying assets.
The limits of documentary conformity
A compliance statement or a framework crosswalk is an observation about documents. It is not direct evidence of risk-management effectiveness. A well-designed risk register may fail to incorporate model updates; a board committee may receive timely reports but take no action; a risk appetite policy may contain limits that are routinely overridden; and a sophisticated ORSA may rely on implausible assumptions. The decisive evidence is the consistency of decisions, limits, controls, and realized responses under conditions of uncertainty.
There are also limits to source access. The complete proprietary text of ISO 31000 is not reproduced here, and the manuscript's ISO structural claims rely on publisher metadata and officially accessible guidance. Peer-reviewed empirical articles are interpreted primarily from verified original publisher abstracts and citation records unless the record supports deeper inspection. Confidential insurer ORSA materials, board deliberations, nonpublic supervisory assessments, and micro-level claims experience are outside this study. These restrictions define the boundary of what can be claimed.
Change across time
Risk governance is dynamic. Insurance contracts, financial markets, catastrophe exposure, climate hazards, inflation, technologies, and legal interpretations evolve. COSO's review and revision and ISO's monitoring and improvement are both responsive to that reality. Their adequacy can only be judged relative to the relevant date, risk context, approved policies, and supervisory edition. Evaluating today's program under tomorrow's draft standard, or judging a historical insurer against requirements introduced later, produces a misleading comparison.
The same temporal discipline applies to a document marked for future issuance. As of this manuscript cutoff, a NAIC file with a December 2026 cover date is future-dated. Its presence on an official website is evidence that the file exists, not evidence that its December edition already governed November filings. The NAIC's established ORSA description and dated 2025 manual supply the historical baseline here.
Comparative Synthesis and Research Conclusions
The frameworks as complementary organizational lenses
The comparative evidence supports a precise synthesis. COSO is particularly articulate about the relationship among strategic alternatives, objectives, appetite, performance, culture, and oversight. ISO 31000 is particularly articulate about establishing risk-management context, choosing criteria, performing assessments, selecting treatment, and returning to the decision as conditions change. Both insist on integration, leadership, communication, and review. Their practical distinction lies in the emphasis that readers are likely to place on the decision record.
This synthesis does not establish a new hybrid doctrine and does not propose that insurers must implement both branded frameworks. A well-executed single framework can incorporate practices described by the other. The research question is whether the insurer can demonstrate both strategic accountability and process discipline, then connect these to binding prudential responsibilities. The frameworks offer different routes to ask those questions; the legal and economic requirements determine how complete the answers must be.
An insurer may implement COSO through a strategic decision template specifying objectives, alternatives, appetite consistency, portfolio effects, accountable owners, and performance review. It may implement ISO through a process template specifying scope, context, criteria, assessment, treatment, residual risk, monitoring and recordkeeping. The two are analytically commensurable where they concern the same decision, while their vocabulary remains distinct. The integration is an ordinary implementation judgment, not a claim that a new standard or certification has been created.
Worked hypothetical: coastal underwriting expansion
Consider a hypothetical multiline property insurer with 1,000 million units of available capital, an existing modeled one-year severe-loss requirement of 650 million, and management's proposed coastal expansion. Assume the illustrative incremental stand-alone risk measure is 180 million, but dependence with current exposures produces an illustrative combined requirement of 780 million. These figures are purely pedagogical and are neither actual insurer data nor a lawful capital calculation. The existing headroom under the stated measure is 350 million; after expansion it is 220 million. These arithmetical relationships establish only the effect under the assumed model.
A COSO-focused evaluation asks whether the business case examined alternative regions, product structures, prices and reinsurance strategies; whether management's performance targets remain coherent with the board's appetite; whether the exposure is consistent with portfolio concentration; and whether information about catastrophe model limitations reaches the appropriate decision-makers. An ISO-focused evaluation establishes the scope and evaluation criteria, identifies catastrophe and other material risks, analyzes model and dependence uncertainty, compares risk-treatment alternatives and residual exposure, and sets monitoring triggers.
The prudential layer then requires a separate step. Under a Solvency II legal entity, management considers Article 44 risk-management expectations and Article 45 ORSA questions, including own solvency needs and divergence from standard formula assumptions. Under an applicable US state ORSA regime, it connects the decision to the company's enterprise risk framework and prospective capital assessment. Under Australian CPS 220, it documents consistency with board-approved appetite and the risk management strategy. None of those steps can be completed merely by showing the two illustrative capital numbers.
If purchasing reinsurance reduces modeled net catastrophe risk but creates collateral or recovery uncertainty, the risk treatment must be reassessed for its financial and operational consequences. An insurer should identify the contract wording, counterparty strength, exhaustion points, reinstatement obligations, event definitions and claims timing; these determine whether the transaction improves the actual ability to meet policyholder obligations. Both frameworks can support the decision record. Insurance-specific analysis supplies the substance.
Worked hypothetical: outsourcing and common-cause technology risk
Consider a second hypothetical insurer that migrates policy servicing and claims administration to a common vendor used by several subsidiaries. Management expects lower unit cost and faster routine claims handling. The risk assessment should include simultaneous outage, data integrity failure, delayed claim payments, cyberattack propagation, recovery time, and inability to reconstruct a claims ledger. The exposure may be difficult to price from historical loss data because the technology stack, contractual dependencies, and cyber threat environment change quickly.
Under COSO, the board examines whether the outsourcing strategy is consistent with customer obligations and operational risk appetite, whether performance incentives obscure control weaknesses, and which cross-subsidiary risks the group risk committee should oversee. Under ISO, the team defines context, identifies failure pathways, evaluates service dependency and control effectiveness, considers alternate vendors or resilience controls, and specifies trigger-based monitoring and contractual treatment.
A jurisdiction-specific supervisory review must then consider the entity's legal responsibility for outsourced processes, continuity of essential services, prudential reporting and third-party risk requirements. That inquiry can involve additional operational-resilience rules not studied as primary sources in this dossier. The two generic frameworks facilitate the analysis but do not exhaust the rules governing outsourcing or cyber incidents.
Criteria for evaluating implementation evidence
For each material strategic or operational decision, the following questions provide a research-derived checklist of observable evidence rather than a new named framework:
- Which legal entity or group bears the ultimate exposure, and which board or management body authorizes it?
- Which organizational objective is being pursued, and which plausible strategic alternatives were considered?
- Which risk definitions, datasets, models, scenarios, time horizons and evaluation criteria informed the decision?
- Which dependencies, concentrations and residual uncertainties were included or consciously documented as limitations?
- Which appetite statements, tolerances, operating limits and statutory requirements constrain acceptance?
- Which treatments were selected, who is accountable for implementation, and how was treatment effectiveness tested?
- Which reporting, escalation, monitoring and revision mechanisms ensure that new information changes action?
- Which documents allow an independent reviewer to trace the link from the original assessment to the final decision and its subsequent performance?
The first and fifth questions expose entity and prudential accountability. The second connects most visibly to COSO's strategy–performance orientation. The third, fourth and sixth connect strongly to ISO's risk process. The seventh and eighth are required by both organizational approaches and supervisory governance. Their value is methodological: they identify missing evidence and guide empirical observation. They do not imply that ticking eight boxes establishes successful risk management.
Research design for future empirical comparison
A direct empirical comparison could construct a longitudinal panel of insurers that disclose verifiable adoption of a specified framework edition and identify comparable risk-management practices independently of promotional statements. Adoption dates would be checked against board-approved policies, reports, and changes in the decision process. Data would require legal-entity financial and solvency information, macroeconomic and catastrophe exposures, reporting rules, ownership forms, and contemporaneous regulatory amendments.
The ideal estimand would distinguish three effects: adopting a framework label, changing actual risk governance practices, and achieving particular outcomes. Without this distinction, a difference between “COSO adopters” and “ISO adopters” could reflect company composition, regulatory jurisdiction, and self-selection. Appropriate research methods could include within-firm comparisons before and after verified procedural changes, well-matched controls, pre-trend tests, and event-based examination of decisions where the framework altered a documented risk response. These are standard methods proposed as an evaluation agenda, not an empirical study completed here.
Outcomes should be aligned with the stated mechanism. If the hypothesis concerns better catastrophe accumulation control, evaluate concentration and loss development, not solely reported profits. If it concerns board oversight, test response to limit breaches and remediation timing. If it concerns policyholder security, evaluate insurer-specific financial resilience and continuity of claims settlement. The insurer's business model, regulatory framework and risk exposure must be controlled explicitly. Reporting an association without these conditions can support only a narrower conclusion.
Main comparative conclusions
The review establishes six connected findings. First, COSO 2017 and ISO 31000:2018 share the objective of embedding risk management into decisions, although they organize the decision record differently. Second, COSO's explicit strategy and performance articulation is particularly informative when evaluating board-level business alternatives and appetite. Third, ISO's process specification is particularly informative when evaluating exposure-level assessment, criteria, treatment and monitoring. Fourth, insurer-specific supervisory requirements are a separate, legally authoritative layer that demands evidence of risk control, financial capacity and governance at the proper entity level. Fifth, adoption of a generic framework cannot by itself certify ORSA adequacy or superior financial performance. Sixth, the existing empirical ERM literature does not offer a defensible direct causal ranking of COSO against ISO for insurers.
These findings are original comparative interpretations anchored in the identified materials. They are formulated as conclusions about governance evidence, not new scientific principles, proprietary methods, or a named hybrid framework. Their limitations are explicit: no confidential supervisory case files, no licensed full-text verification of every ISO clause, no proprietary insurer panel, no econometric replication and no controlled adoption experiment.
Conclusion
Risk management frameworks matter to insurers when they govern actual acceptance, retention, transformation and financing of uncertain obligations. COSO 2017 makes the strategic and organizational conditions of those decisions particularly visible. ISO 31000:2018 makes the disciplined assessment and treatment process particularly visible. Both demand responsible leadership, information and review; both require insurance-specific implementation. The comparative question resolves at the point where organizational method meets legal entity, supervisory duty, and demonstrated financial resilience.
The relevant insurance supervisory instruments establish that an insurer must be able to explain risks and dependencies, maintain appropriate governance and controls, connect risk appetite with strategy, and assess prospective solvency where applicable. IAIS ICPs provide the international benchmark; EU, US, and Australian instruments illustrate distinct forms of legal application. Generic framework adoption can assist these tasks and cannot substitute for them. Nor does an external framework label establish improved economic outcomes. The available peer-reviewed studies of ERM offer heterogeneous observational associations and leave the direct COSO-versus-ISO comparison empirically unresolved.
The lasting research conclusion is operational and verifiable: an insurer's ERM should be evaluated through the quality of its documented decisions, the rigor of its assessment and risk treatment, the alignment of governing authority and risk limits, and the adequacy of its jurisdiction-specific solvency and policyholder-protection arrangements. A comparison that preserves these dimensions is more scientifically informative than an abstract preference for one framework or a generic maturity rating. The present monograph is a completed documentary and analytical synthesis within the stated source-access and empirical boundaries.
References
Australian Prudential Regulation Authority. (2019). Prudential Standard CPS 220: Risk Management (effective July 1, 2019). APRA. https://www.apra.gov.au/standards/cps-220
Baxter, R., Bedard, J. C., Hoitash, R., & Yezegel, A. (2013). Enterprise risk management program quality: Determinants, value relevance, and the financial crisis. Contemporary Accounting Research, 30(4), 1264–1295. https://doi.org/10.1111/j.1911-3846.2012.01194.x
Committee of Sponsoring Organizations of the Treadway Commission. (2017). Enterprise risk management—Integrating with strategy and performance (and official comparison with the 2004 Integrated Framework). COSO. https://www.coso.org/guidance-erm
European Insurance and Occupational Pensions Authority. (2015). Guidelines on own risk and solvency assessment (ORSA) (publication page dated September 14, 2015; subsequent language files and consultation materials identified separately). EIOPA. https://www.eiopa.europa.eu/publications/guidelines-own-risk-solvency-assessment-orsa_en
European Insurance and Occupational Pensions Authority. (n.d.-a). Risk management: Article 44 of Directive 2009/138/EC. Solvency II Single Rulebook. https://www.eiopa.europa.eu/rulebook/solvency-ii-single-rulebook/article-2127_en
European Insurance and Occupational Pensions Authority. (n.d.-b). Own risk and solvency assessment: Article 45 of Directive 2009/138/EC. Solvency II Single Rulebook. https://www.eiopa.europa.eu/rulebook/solvency-ii-single-rulebook/article-2128_en
Froot, K. A. (2007). Risk management, capital budgeting, and capital structure policy for insurers and reinsurers. Journal of Risk and Insurance, 74(2), 273–299. https://doi.org/10.1111/j.1539-6975.2007.00213.x
Gatzert, N., & Martin, M. (2015). Determinants and value of enterprise risk management: Empirical evidence from the literature. Risk Management and Insurance Review, 18(1), 29–53. https://doi.org/10.1111/rmir.12028
Hoyt, R. E., & Liebenberg, A. P. (2011). The value of enterprise risk management. Journal of Risk and Insurance, 78(4), 795–822. https://doi.org/10.1111/j.1539-6975.2011.01413.x
International Association of Insurance Supervisors. (2024a). Insurance Core Principles and ComFrame Online Tool (December 2024 revision notice). IAIS. https://www.iais.org/icp-online-tool/
International Association of Insurance Supervisors. (2024b). Insurance Core Principles and Common Framework for the Supervision of Internationally Active Insurance Groups (December 2024 edition). IAIS. https://www.iais.org/uploads/2024/12/IAIS-ICPs-and-ComFrame-adopted-in-December-2024.pdf
International Electrotechnical Commission. (2019). IEC 31010:2019: Risk management—Risk assessment techniques (2nd ed.). IEC/ISO publication listing. https://www.iso.org/standard/72140.html
International Organization for Standardization. (2018). ISO 31000—Risk management (free introductory brochure). ISO. https://www.iso.org/publication/PUB100426.html
International Organization for Standardization. (2018). ISO 31000:2018: Risk management—Guidelines (2nd ed., February 2018; confirmed in 2023). ISO. https://www.iso.org/standard/31000
McShane, M. K., Nair, A., & Rustambekov, E. (2011). Does enterprise risk management increase firm value? Journal of Accounting, Auditing & Finance, 26(4), 641–658. https://doi.org/10.1177/0148558X11409160
National Association of Insurance Commissioners. (2025). Own risk and solvency assessment (ORSA) guidance manual (revision memorandum dated August 12, 2025). NAIC. https://content.naic.org/sites/default/files/committee_related_documents/naic-orsa-guidance-manual-final_0.pdf
National Association of Insurance Commissioners. (n.d.-a). Enterprise risk management. NAIC Insurance Topics. https://content.naic.org/insurance-topics/enterprise-risk-management
National Association of Insurance Commissioners. (n.d.-b). Own risk and solvency assessment. NAIC Insurance Topics. https://content.naic.org/insurance-topics/own-risk-and-solvency-assessment