Actuarial Governance, Regulation and Future Practice

Introduction and context

Actuarial mathematics is often introduced through probability, statistics, financial mathematics and risk models. In professional practice, however, a technically correct calculation is only one part of a defensible actuarial decision. The calculation must be based on appropriate data, justified assumptions, a suitable model, documented methods, effective review and compliance with law and professional standards. It must also be communicated clearly to decision-makers who may not understand the mathematics but remain accountable for the consequences.

Actuarial governance is the system through which an insurer ensures that actuarial work is reliable, independent, transparent, challengeable and connected to risk-taking decisions. It includes authority structures, professional standards, assumption-setting, model controls, validation, peer review, documentation, audit trails and escalation procedures.

The strategic importance of governance has increased because insurers now face interconnected risks: longevity and mortality uncertainty, inflation, catastrophe and climate risk, cyber risk, financial-market volatility, operational disruption, data-quality weaknesses and risks arising from artificial intelligence. Solvency regulation and financial reporting frameworks, especially risk-based capital regimes and IFRS 17, require actuaries to explain not only the amount of an estimate but also its uncertainty, movement and sensitivity.

This chapter examines how actuarial work is governed, regulated and likely to develop. It treats governance not as administrative compliance, but as a control system that protects policyholders, supports sound management and preserves public confidence.

Learning objectives

By the end of this chapter, a learner should be able to:

1. Explain the purpose of actuarial governance and distinguish governance, management and assurance.
2. Evaluate professional standards, ethical duties, documentation and actuarial communication.
3. Design an assumption-governance process covering evidence, approval, monitoring and change control.
4. Explain model validation, independent review, audit controls and limitations of actuarial models.
5. Relate actuarial work to solvency regulation, enterprise risk management and IFRS 17.
6. Assess the actuarial implications of climate risk, cyber risk, digital insurance and artificial intelligence.
7. Recommend practical controls for a robust actuarial function.

1. The architecture of actuarial governance

1.1 Meaning and purpose

Actuarial governance is the framework of responsibilities, controls and decision processes used to ensure that actuarial opinions and calculations are fit for purpose. Its objectives are to:

• protect policyholders and beneficiaries;
• support adequate pricing, reserving and capital;
• ensure compliance with legislation and professional standards;
• make assumptions and judgements transparent;
• identify model limitations and emerging risks;
• prevent conflicts of interest and inappropriate influence; and
• provide evidence that decisions were properly made.

Governance differs from actuarial management. Management allocates resources and executes business decisions. Governance determines who has authority, what evidence is required, how challenge occurs and how accountability is assigned. Assurance provides independent confidence that controls and outputs operate as intended.

A typical insurance governance structure includes the board, board risk or audit committee, chief executive, chief risk officer, chief financial officer, appointed or chief actuary, actuarial function, internal audit, compliance, external audit and the supervisor. The precise legal titles vary by jurisdiction, but the underlying principles are widely applicable.

1.2 Three lines of responsibility

A useful control model is:

External audit and the insurance supervisor are not simply a “fourth line”; they provide external assurance or regulatory oversight with distinct mandates. The board remains responsible for effective governance even when specialist work is delegated.

1.3 Proportionality and materiality

Controls should be proportionate to the nature, scale and complexity of the insurer. A small insurer may use a simpler model than a multinational group, but it cannot dispense with documentation, conflict management or validation. Materiality should be assessed in relation to possible effects on policyholders, capital, profit, solvency and management decisions. A seemingly small model error can become material when repeated across millions of policies or combined with correlated assumptions.

2. Professional standards and ethical duties

2.1 Standards as a professional control

Actuarial standards generally address competence, integrity, objectivity, communication, data, assumptions, methods, documentation and review. They may be issued by a national actuarial body, an international body, a regulator or a combination of these institutions. Standards can be principles-based, rules-based or hybrid.

Principles-based standards require professional judgement and explain the objective that should be achieved. Rules-based standards prescribe more specific procedures. Principles-based standards are adaptable but may produce inconsistent interpretations; rules-based standards are more predictable but may become obsolete or encourage mechanical compliance. A mature actuary understands both the mandatory requirements and the reasoning behind them.

A professional standard does not convert an uncertain estimate into a fact. It requires the actuary to identify uncertainty, use appropriate methods and communicate limitations. Compliance should therefore be evidenced through working papers, review records and clear disclosures.

2.2 Core ethical duties

The central ethical duties are:

• Integrity: act honestly and avoid misleading statements or omissions.
• Competence and care: accept work only when the necessary knowledge, experience, time and resources are available.
• Objectivity: resist pressure to select assumptions or results solely because they improve reported performance.
• Confidentiality: protect personal, commercial and policyholder information, subject to legal reporting duties.
• Professional behaviour: comply with law, standards and legitimate supervisory requirements.
• Public interest: recognise that actuarial work affects policyholders, claimants, pensioners, investors and society, not merely the appointing client.

Conflicts may arise when an actuary advises both a policyholder group and an insurer, receives performance incentives linked to a result, or reports to management while expected to challenge management. Controls include disclosure, recusal, independent review, separation of duties and escalation to the board or supervisor.

2.3 Communication and professional judgement

An actuarial report should normally identify its purpose, intended users, period of validity, data sources, methods, assumptions, material limitations, uncertainty, sensitivities and conclusions. It should distinguish facts, assumptions, judgement and recommendations. Technical precision is not sufficient if the communication creates a false impression of certainty.

A useful communication test is whether a reasonably informed decision-maker can answer four questions:

1. What was estimated or concluded?
2. Why was this method and these assumptions used?
3. How sensitive is the result to plausible alternatives?
4. What action or limitation follows from the result?

3. Documentation and auditability

Documentation is the evidence chain linking a decision to its inputs, method, judgement, review and approval. It should enable a qualified independent person to reproduce the result or understand precisely why reproduction is not possible.

A robust actuarial file commonly includes:

• scope, purpose and intended users;
• data extracts, reconciliation and data-quality assessment;
• policy wording and legal interpretation where relevant;
• model version, code, software environment and parameter settings;
• assumptions, evidence and rationale;
• expert judgements and alternatives considered;
• validation and sensitivity tests;
• review comments and their resolution;
• approval records and sign-off;
• limitations, uncertainty and subsequent events; and
• retention, access and change-control records.

Version control is essential. A result should be linked to a specific data date, code version, assumption set and model configuration. “Final.xlsx” is not an adequate control when several materially different files may exist. Access rights should prevent unauthorised changes, while logs should identify who changed what, when and why.

Documentation also supports regulatory review and litigation defence. It does not mean recording every thought; it means recording material decisions sufficiently to demonstrate a rational and controlled process.

4. Assumption governance

4.1 Nature of actuarial assumptions

Assumptions are propositions about uncertain future experience. Examples include mortality, morbidity, lapse, expenses, inflation, claim frequency, claim severity, discount rates, reinsurance recoveries and policyholder behaviour. Some assumptions are observable from market data; others require expert judgement.

For a future cash-flow estimate, a simplified expected-value representation is:

Here, V is the estimated present value, C_t is the cash flow at time t, E[C_t] is its expected value under the selected probability and economic assumptions, v_t is the discount factor, and T is the projection horizon. The formula assumes that the cash flows, probability model and discounting convention are defined consistently. It is not a complete valuation formula for every product: options, guarantees, risk adjustments, taxes and contract boundaries may require additional terms or stochastic methods.

Assumptions should not be chosen merely because they reproduce a target profit. They should reflect evidence, intended use, materiality and the relevant accounting or regulatory basis.

4.2 Assumption lifecycle

A disciplined lifecycle consists of:

1. Identification: determine which assumptions drive the decision.
2. Evidence collection: obtain internal experience, external studies, market data, expert evidence and relevant legal or economic information.
3. Analysis: test data quality, credibility, trends, seasonality, heterogeneity and structural breaks.
4. Selection: choose a central estimate, margins or stress parameters as appropriate.
5. Challenge: obtain independent review and compare alternatives.
6. Approval: assign authority according to materiality and governance policy.
7. Implementation: record the assumption in a controlled model environment.
8. Monitoring: compare actual experience with expected experience and investigate deviations.
9. Change control: document triggers, impact and approval for revisions.

Experience monitoring may use an actual-to-expected ratio:

For claim counts, A/E > 1 indicates more claims than expected; for mortality rates, the interpretation depends on the definition of the numerator and denominator. The ratio is a diagnostic measure, not automatically evidence that an assumption should be changed. Random fluctuation, exposure changes, coding errors and selection effects must be considered.

4.3 Uncertainty and sensitivity

A central estimate should be accompanied by sensitivity analysis. If a result Y depends on assumptions θ₁, θ₂, …, θ_n, a local approximation is:

This approximation is useful when changes are small and the model is sufficiently smooth. It may be unreliable for discontinuous decisions, extreme scenarios or highly nonlinear guarantees. Scenario testing and stochastic simulation may then be more appropriate.

Correlation is especially important. Changing mortality, lapse and interest rates independently may produce unrealistic outcomes if policyholder behaviour responds to economic conditions. Assumption governance must therefore consider dependencies, not just individual parameters.

5. Model governance and validation

5.1 Models as representations, not reality

An actuarial model is a structured representation of an insurance system. It simplifies reality through data definitions, equations, algorithms and assumptions. Model risk is the possibility that the model is unsuitable, incorrectly implemented, misused or misunderstood.

Model governance should cover the entire lifecycle: inventory, classification, development, testing, approval, use, monitoring, change and retirement. Models should be classified by materiality and complexity. A capital model used for strategic decisions requires more extensive controls than a simple management report, although even a spreadsheet can be material.

5.2 Validation dimensions

Validation should be sufficiently independent from development and should address at least:

• Conceptual soundness: Is the model appropriate for the product, risk and intended use?
• Data integrity: Are data complete, accurate, relevant and consistently defined?
• Implementation accuracy: Does the code implement the documented method?
• Calibration: Do parameters reflect credible evidence and the appropriate time horizon?
• Performance: Does the model explain or predict experience adequately, subject to its purpose?
• Sensitivity and stress: Does behaviour under changed assumptions make sense?
• Benchmarking: Do alternative models or external references reveal material differences?
• Limitations: Are weaknesses documented and controlled?

Back-testing compares model expectations with subsequent experience, but it must account for estimation error, changing portfolios and conditions. A model can perform poorly in a short sample and still be appropriate, or perform well by chance while being structurally wrong.

5.3 Independent review and audit controls

Important controls include segregation between model development and approval, controlled code repositories, automated tests, reconciliations to source systems, reasonableness checks, input/output controls, exception reporting and formal sign-off. Internal audit should evaluate whether the control framework operates effectively, not re-perform every actuarial calculation as if it were the actuarial function.

A strong control environment distinguishes errors from model limitations. An arithmetic coding error should be corrected. A limitation caused by sparse catastrophe data should be disclosed, stress-tested and managed; it cannot necessarily be “fixed” by more coding.

6. Actuarial work and solvency regulation

6.1 Risk-based solvency

Modern solvency frameworks generally require an insurer to hold capital related to the risks it bears rather than applying only a fixed premium or reserve percentage. A simplified solvency relationship is:

A solvency ratio may be expressed as:

“Available capital” and “required capital” are regulatory concepts whose definitions differ by jurisdiction. The ratio is meaningful only when the valuation basis, eligibility rules, risk coverage and supervisory calibration are understood. It should not be compared across regimes without adjustment.

Actuaries contribute to technical provisions, risk capital, stress testing, reinsurance analysis, own-risk and solvency assessments, recovery planning and supervisory reporting. The objective is not to maximise the ratio mechanically. Excessively conservative assumptions may reduce reported risk but distort pricing and capital allocation; excessively optimistic assumptions threaten policyholder security.

6.2 Enterprise risk management

Actuarial analysis supports an enterprise risk-management cycle:

1. identify risks and concentrations;
2. measure exposures using appropriate metrics;
3. assess correlations and scenarios;
4. compare risk with appetite and available capital;
5. select controls, mitigation or transfer;
6. monitor indicators and emerging risks; and
7. report and escalate.

Risk appetite should be translated into measurable limits, such as catastrophe exposure, duration mismatch, liquidity needs, reinsurance counterparty exposure or concentration by geography. Models inform decisions but do not replace board judgement.

7. IFRS 17 and actuarial governance

IFRS 17 establishes principles for measuring and presenting insurance contracts. Implementation requires close cooperation among actuaries, accountants, finance, data specialists, risk managers and auditors. The exact calculation depends on contract features and the applicable model, but key concepts include fulfilment cash flows, discounting, risk adjustment for non-financial risk and the contractual service margin.

A simplified conceptual representation is:

The signs and presentation must follow the entity’s accounting policy and IFRS 17 requirements; the expression is a conceptual summary rather than a substitute for the standard. The contractual service margin generally represents unearned profit released as insurance services are provided, subject to the applicable measurement model and subsequent adjustments.

Governance challenges include:

• defining contract boundaries consistently;
• grouping contracts into appropriate portfolios and cohorts;
• selecting discount rates and illiquidity adjustments;
• determining methods for risk adjustment;
• allocating acquisition cash flows;
• maintaining data lineage between actuarial and general-ledger systems;
• explaining movement analyses; and
• controlling judgements that affect profit emergence.

Actuarial estimates under IFRS 17 should be reconciled to source data and financial statements, but reconciliation does not prove correctness. Differences may arise from timing, scope, measurement basis or accounting classification and must be explained.

8. Climate and environmental risk

Climate risk has both physical and transition dimensions. Physical risk includes floods, storms, drought, wildfire, heat and disease patterns. Transition risk arises from changes in technology, regulation, consumer behaviour, carbon pricing and asset values. Liability risk may arise from litigation or changing legal duties.

Actuarial implications include changes in frequency and severity, non-stationary historical data, geographic concentration, protection gaps, asset impairment, reinsurance availability and correlated losses across lines. Historical averages may be poor guides when the underlying hazard is changing.

A practical climate framework should:

• establish materiality and exposure maps;
• identify relevant time horizons;
• combine climate science with insurance and financial data;
• use scenarios rather than a single forecast;
• test underwriting, reserving, investment and liquidity effects;
• consider adaptation and mitigation actions; and
• disclose uncertainty and model limitations.

Climate scenarios are not predictions. They are internally coherent descriptions used to test resilience. Long-horizon scenarios are particularly uncertain because policy, technology and behaviour can change.

9. Cyber risk, digital insurance and artificial intelligence

Cyber risk affects insurers as both risk carriers and institutions. It includes data theft, ransomware, service interruption, fraud, supply-chain compromise and accumulation across many insureds. Actuarial challenges include limited historical data, rapidly changing threats, dependent losses and difficulty defining policy coverage.

Controls include data minimisation, access management, encryption, incident response, vendor oversight, backup testing and scenario analysis. Actuaries should understand whether cyber exclusions, sublimits, waiting periods and aggregation controls are reflected in models and pricing.

Digital insurance uses automated underwriting, telematics, online claims, APIs and real-time data. Benefits include speed, personalisation and potentially better risk management. Risks include biased data, privacy breaches, opaque decisions, unfair discrimination, proxy variables and operational concentration.

Artificial intelligence can assist with anomaly detection, document classification, claims triage and experience analysis. It should not be treated as inherently objective. Governance should address:

• purpose and permitted use;
• data provenance and representativeness;
• explainability appropriate to the decision;
• validation against a suitable benchmark;
• drift monitoring;
• human override and appeal;
• privacy and security;
• fairness and prohibited discrimination; and
• accountability for automated decisions.

A model that predicts claims accurately may still be unacceptable if it uses unlawful or unfair characteristics. Actuarial professionalism requires consideration of social consequences, not only predictive performance.

10. Illustrative case: governance failure in a rapidly growing insurer

Background

The fictional Meridian Digital Insurer sells short-duration motor policies through an online platform. Its pricing model uses telematics, customer characteristics and external data. Premium growth is rapid, and management plans to raise capital. The actuarial team is small and reports to the commercial director.

Problem

The insurer experiences worsening claims but continues to report favourable projected profitability. An investigation identifies four weaknesses: a data extract excludes some late-reported claims; an external vendor changes a rating variable without notification; model documentation does not state the training period; and commercial staff can change selected pricing parameters in production. No independent validation has occurred since launch.

Analysis

The issue is not simply a pricing error. It is a governance failure involving data lineage, model change control, independence, monitoring and accountability. The apparent profit is not reliable because the model and actual experience are not comparable. The omission of late claims creates an artificial improvement, while the unrecorded vendor change creates an uncontrolled basis change.

The actuary should first preserve evidence and quantify the impact using a controlled reconstruction. The insurer should suspend uncontrolled parameter changes, reconcile claims to the source system, assess customer impact and notify the board and supervisor where required. A temporary pricing or underwriting restriction may be necessary if continuing to write business would increase harm.

Approach and solution

The insurer establishes a model-governance committee, appoints an independent validation team and creates a model inventory. It introduces dual approval for production changes, automated data reconciliations, monthly actual-to-expected monitoring and documented override limits. The actuarial function is given direct access to the risk committee. The vendor contract is amended to require change notification, audit rights and data-quality standards.

The model is recalibrated using a controlled data set, but management also recognises that recalibration cannot solve structural uncertainty. Scenario tests are performed for claims inflation, telematics failure, adverse selection and cyber interruption. Customer remediation is considered for policies affected by materially unfair pricing.

Outcome

Reported profitability falls, but the insurer obtains a more credible view of capital needs and improves supervisory confidence. Growth is temporarily slowed. The board concludes that a lower growth rate with reliable governance is strategically superior to rapid expansion based on uncontrolled model risk.

Learning points

1. Data governance is part of actuarial governance.
2. Independent validation must occur before material reliance, not only after a problem.
3. Commercial ownership does not justify uncontrolled model changes.
4. A model result is not credible unless its lineage and version are known.
5. Ethical duties extend to customers affected by automated decisions.

11. Contemporary developments and future practice

Future actuarial practice is likely to combine classical actuarial methods with causal analysis, machine learning, geospatial data, climate science, behavioural economics, cyber security and real-time monitoring. The professional role will shift from producing isolated estimates toward governing decision systems.

Several capabilities will become increasingly important:

• translating technical uncertainty into strategic choices;
• validating automated and externally supplied models;
• understanding data architecture and software controls;
• analysing distributional and fairness effects;
• designing scenario-based rather than purely historical risk assessments;
• integrating finance, risk and actuarial information;
• communicating with boards, regulators and the public; and
• maintaining professional independence under commercial pressure.

There is an important scholarly debate about explainability. A highly interpretable model may be less accurate, while a complex model may perform better but be difficult to challenge. The appropriate choice depends on purpose, materiality, legal requirements and the consequences of error. Explainability should not be reduced to a technical feature; it is also a governance requirement enabling challenge and accountability.

Another debate concerns the boundary between expert judgement and automation. Automation can reduce repetitive error and improve consistency, but it may also reproduce historical discrimination or hide uncertainty behind a numerical output. The future actuary must be able to interrogate both the mathematics and the institutional context in which it is used.

Best-practice framework

An insurer seeking a mature actuarial function should implement the following minimum practices:

No framework eliminates uncertainty. Its purpose is to ensure that uncertainty is recognised, managed and communicated rather than concealed.

Chapter Summary

Actuarial governance connects technical actuarial work with professional accountability, policyholder protection and strategic decision-making. It requires clear responsibilities, proportional controls, ethical conduct, reliable documentation and communication that distinguishes evidence from judgement.

Assumption governance is a lifecycle involving evidence, analysis, challenge, approval, implementation and monitoring. Model governance addresses conceptual suitability, data, implementation, calibration, performance, sensitivity and limitations. Independent validation, segregation of duties, version control and audit trails are central safeguards.

Actuaries contribute directly to solvency regulation through technical provisions, capital assessment, stress testing and enterprise risk management. IFRS 17 increases the importance of controlled cash-flow projections, discounting, risk adjustment, contractual service margin analysis and financial reconciliation. Climate, cyber, digital and artificial-intelligence risks require scenario analysis, data governance and attention to correlation, fairness and uncertainty.

The future actuarial professional will need both quantitative expertise and governance judgement. The most valuable actuarial answer is not merely a number; it is a well-supported, transparent and appropriately qualified basis for responsible action.

Key Terms

Actuarial governance: The framework of responsibilities and controls governing actuarial work.

Assumption governance: The process for selecting, approving, monitoring and changing actuarial assumptions.

Available capital: Regulatory capital eligible to support an insurer’s risks.

Model risk: The risk of unsuitable, incorrectly implemented or misused models.

Materiality: The significance of an error or uncertainty to decisions or users.

Professional standard: An authoritative requirement or principle governing actuarial practice.

Professional judgement: Reasoned expert evaluation where evidence does not determine a unique answer.

Data lineage: The traceable path from source data through transformations to reported output.

Validation: Independent assessment of a model’s suitability, implementation and performance.

Back-testing: Comparing model expectations with subsequent observed experience.

Sensitivity analysis: Analysis of changes in results caused by changes in assumptions or inputs.

Stress testing: Evaluation of outcomes under adverse but plausible conditions.

Solvency ratio: A regulatory comparison of eligible capital with required capital.

Enterprise risk management: Integrated identification, measurement and management of organisational risks.

IFRS 17: The international financial reporting standard for insurance contracts.

Risk adjustment: An amount reflecting uncertainty about non-financial insurance risk.

Contractual service margin: A component representing unearned profit released as insurance services are provided.

Physical climate risk: Loss arising from climate-related physical hazards.

Transition risk: Risk arising from changes toward a lower-carbon economy.

Model drift: Deterioration or change in model performance as data or conditions change.

Review Questions

Short Answer Questions

1. What is the purpose of actuarial governance?
2. Name three core ethical duties of an actuary.
3. Why is version control important in actuarial modelling?
4. What is the difference between physical and transition climate risk?
5. State two dimensions that should be included in model validation.

Long Answer Questions

1. Explain how the three lines of responsibility can operate within an insurer’s actuarial governance framework.
2. Describe the complete lifecycle for governing actuarial assumptions and explain why monitoring is necessary.
3. Discuss the relationship between actuarial work, risk-based solvency regulation and enterprise risk management.
4. Explain the major actuarial governance challenges created by IFRS 17.
5. Evaluate the ethical and control issues involved in using artificial intelligence for insurance underwriting or claims.

Analytical / Application Questions

1. An insurer’s actual-to-expected claims ratio has exceeded 1.10 for three months. Analyse the evidence that should be collected before changing the pricing assumption.
2. A capital model produces a favourable solvency ratio, but its catastrophe data are sparse and its dependencies are undocumented. Recommend a validation and governance response.
3. Design a governance plan for a digital insurer using telematics data, including controls for data quality, fairness, model change and customer appeals.

Multiple Choice Questions with Answers

Question 1

Which statement best describes actuarial governance?

A. A process used only to calculate insurance premiums

B. A framework of responsibilities and controls governing actuarial work

C. A method for eliminating all uncertainty from forecasts

D. A replacement for board accountability

Correct Answer: B. A framework of responsibilities and controls governing actuarial work

Question 2

Which control most directly supports reproducibility of an actuarial result?

A. Increasing the marketing budget

B. Removing independent review

C. Linking the result to controlled data, code and assumption versions

D. Using only historical averages

Correct Answer: C. Linking the result to controlled data, code and assumption versions

Question 3

An actual-to-expected ratio greater than 1 for claim counts generally indicates:

A. Fewer claims than expected

B. More claims than expected

C. No uncertainty in the estimate

D. That the model must immediately be discarded

Correct Answer: B. More claims than expected

Question 4

Which is a conceptual-soundness validation question?

A. Is the model appropriate for the risk and intended use?

B. Has the model been given a shorter name?

C. Does the model produce a favourable result?

D. Was the model created by the largest team?

Correct Answer: A. Is the model appropriate for the risk and intended use?

Question 5

In the simplified solvency ratio, the denominator represents:

A. Eligible assets

B. Required capital

C. Premium income only

D. Claims paid only

Correct Answer: B. Required capital

Question 6

Which item is conceptually included in simplified IFRS 17 fulfilment cash flows?

A. Only historical premiums

B. Present-value cash flows plus a risk adjustment for non-financial risk

C. Marketing expenditure alone

D. The insurer’s share price

Correct Answer: B. Present-value cash flows plus a risk adjustment for non-financial risk

Question 7

Which is an example of transition climate risk?

A. A flood damaging an insured property

B. A storm increasing claims in a coastal region

C. A regulatory change affecting carbon-intensive assets

D. A database password failure

Correct Answer: C. A regulatory change affecting carbon-intensive assets

Question 8

Why should an actuarial model using artificial intelligence be monitored for drift?

A. Because relationships and data conditions may change over time

B. Because monitoring guarantees fairness

C. Because drift means the model has no code

D. Because all automated models are invalid

Correct Answer: A. Because relationships and data conditions may change over time

Question 9

Which arrangement best supports independence in material model validation?

A. The model developer approves their own model without review

B. An independent team validates the model and reports findings through governance channels

C. Commercial staff alter parameters without records

D. Validation is performed only after regulatory failure

Correct Answer: B. An independent team validates the model and reports findings through governance channels

Question 10

A professional actuarial communication should normally include:

A. Only the final number

B. Purpose, methods, assumptions, uncertainty and limitations

C. No information about intended users

D. An assurance that the estimate is certain

Correct Answer: B. Purpose, methods, assumptions, uncertainty and limitations