Insurance Finance & Solvency Models

Chapter 3
Insurance Liabilities, Reserves and Actuarial Assumptions

Post Graduate Program in Insurance Management
Insurance Finance & Solvency Models   •   Post Graduate

Insurance Liabilities, Reserves and Actuarial Assumptions

Introduction and Context

An insurer’s financial strength depends less on the cash it has collected today than on whether it can meet obligations arising from contracts written in the past. Those obligations include future claim payments, maturity benefits, death benefits, surrender values, expenses, commissions, and claim settlement costs. They are measured through liabilities and reserves.

Insurance liabilities are difficult to estimate because payment often occurs long after premium collection, information is incomplete, policyholder behaviour is uncertain, and economic conditions change. A motor insurer may know that an accident occurred but not the final repair cost. A life insurer may know the policy terms but not when death, surrender, or maturity will occur. Consequently, reserve measurement combines contractual analysis, actuarial models, statistical estimation, financial economics, accounting rules, and professional judgement.

This chapter examines premium, claim, and benefit liabilities; life policy reserves; non-life loss development and IBNR; best-estimate assumptions; margins; discounting; reserve uncertainty; and governance implications. The emphasis is on how a technically calculated reserve becomes a managed financial estimate subject to controls, regulation, and review.

Learning Objectives

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

  1. Distinguish premium liabilities from claim and benefit liabilities.
  2. Explain the structure and purpose of life insurance reserves.
  3. Apply basic loss-development and chain-ladder calculations.
  4. Explain IBNR and alternative claims-reserving approaches.
  5. Construct best-estimate assumptions and distinguish them from margins.
  6. Apply discounted cash-flow principles to insurance liabilities.
  7. Evaluate uncertainty, model risk, governance, and managerial consequences.

1. The Architecture of Insurance Liabilities

1.1 Contractual cash-flow perspective

An insurance contract creates expected future cash inflows and outflows. Premiums are generally received before the insurer provides complete coverage, whereas claims and benefits may be paid much later. The liability is therefore not simply the amount of premium collected; it is the present or expected value of future obligations after allowing for future premiums and related expenses, according to the applicable accounting and regulatory framework.

Important liability categories include:

CategoryEconomic meaningTypical examples
Unearned premium liabilityCoverage already sold but not yet earnedAnnual motor policy with six months remaining
Premium liabilityFuture claims and expenses relating to unexpired coverageExpected losses during the remaining policy period
Reported claim liabilityClaims notified but not finally settledA reported bodily-injury claim
IBNR liabilityEvents or claim developments not yet fully reported or knownLate-reported liability claims
Life policy reserveValue of future benefits and expenses less relevant future premiumsWhole-life or endowment policy
Claim settlement expense liabilityCosts of adjusting and settling claimsSurveyor, legal, medical, and administrative costs
Reinsurance recoverableAmount expected from reinsurersRecoveries under proportional or excess-of-loss treaties

The exact labels differ across jurisdictions and reporting bases. A finance professional must therefore identify whether a figure is intended for financial reporting, statutory solvency, pricing, internal economic capital, or management planning.

1.2 Premium and claim liabilities

For a short-duration policy, the unearned premium reserve may be a useful starting point. Under a simple pro-rata assumption, if an annual premium of 1,200 covers twelve months and six months remain, the unearned component is:

UPR = P × (Remaining coverage period ÷ Total coverage period)

where UPR is the unearned premium reserve and P is written premium. This method assumes that risk is approximately uniform through time. It is inappropriate when exposure is seasonal, claim frequency is strongly time-dependent, or policy terms create non-uniform risk.

A premium liability should also reflect expected future claims, expenses, and sometimes profit or risk adjustments associated with unexpired coverage. If expected future claims and expenses exceed the related unearned premium, an additional liability may be required under the relevant framework. This is often called an unexpired-risk, premium-deficiency, or onerous-contract assessment, although terminology varies.

Claim liabilities relate to insured events that have already occurred. They normally include:

  • amounts already paid;
  • case reserves for reported but unsettled claims;
  • IBNR for claims incurred but not reported; and
  • incurred-but-not-enough-reported amounts for claims whose ultimate cost will develop beyond the current case estimate.

The distinction between “reported” and “incurred” is crucial. A claim may be incurred on the accident date, reported later, and settled still later. Reserving must therefore reconcile accident, report, development, and payment dates.

2. Life Insurance Reserves

2.1 Why life reserves are needed

Life insurance contracts can extend for decades. A level premium may exceed the expected cost of mortality in early years and fall below it in later years. The excess collected in early years is not immediately profit: it helps finance later benefits. The reserve represents the insurer’s accumulated obligation under the contract, subject to the valuation basis.

A reserve is not necessarily the amount the insurer would pay if the policy were surrendered. It is a valuation measure based on specified benefits, premiums, mortality, lapse, expenses, interest, and applicable contract rules.

2.2 Net premium prospective reserve

For a traditional participating or non-participating life contract, a simplified net premium prospective reserve at time t can be written as:

Vt = PVt(Future benefits) − PVt(Future net premiums)

where Vt is the reserve at time t, and PVt denotes present value at that valuation date. The net premium basis generally excludes some expenses and uses a defined mortality and interest basis. A gross premium reserve may additionally include future expenses, commissions, premium loadings, lapses, and other contractual cash flows.

For a benefit payable at the end of year k, discounted at annual effective interest rate i, the present value at time zero is:

PV = Benefit × vk, where v = 1 ÷ (1 + i)

If payments are contingent on survival or death, the benefit is multiplied by the relevant probability. For example, a death benefit of B payable at the end of year k has expected present value:

EPV = B × Pr(death in year k) × vk

This is a simplified discrete model. Actual actuarial valuations may use fractional ages, select mortality, multiple decrements, payment timing conventions, expenses, bonuses, options, and stochastic techniques.

2.3 Example: simplified endowment reserve

Assume a policy pays 100,000 at the end of year 10 if the insured survives, or at the end of the year of death if death occurs earlier. Suppose, for illustration, that the expected benefit cash flows have already been calculated as 7,500 at year 3, 8,000 at year 6, and 70,000 at year 10. At an annual effective discount rate of 4%, the present value is:

PV = 7,500(1.04)^−3 + 8,000(1.04)^−6 + 70,000(1.04)^−10

Using approximate discount factors 0.8890, 0.7903, and 0.6756:

PV ≈ 6,667.5 + 6,322.4 + 47,292.0 = 60,281.9

If the present value of future net premiums is 52,000, the simplified reserve is:

V ≈ 60,281.9 − 52,000 = 8,281.9

The example is illustrative rather than a complete statutory valuation. It assumes that the cash flows and probabilities embedded in the expected benefits are appropriate, that the discount rate is applicable, and that no additional expense or option adjustment is required.

2.4 Gross premium and prospective reserves

A gross premium valuation uses realistic or prescribed assumptions for:

  • mortality and morbidity;
  • persistency, lapse, and surrender;
  • renewal and expense inflation;
  • maintenance and claim expenses;
  • investment returns or discount rates;
  • commissions and acquisition costs;
  • policyholder options and guarantees; and
  • future bonuses or management actions.

For a generic liability cash-flow model:

Lt = ∑k=1n [E(CFt+k) ÷ (1 + rk)k] − PVt(Future contractual inflows)

Here Lt is the liability at time t, E(CFt+k) is the expected net cash outflow at future time t+k, rk is the appropriate annual effective discount rate for that maturity, and n is the projection horizon. If cash flows are monthly or continuous, the summation and discounting convention must be changed consistently.

3. Non-Life Claims Reserving and Loss Development

3.1 Claims triangles

A claims triangle organizes cumulative or incremental claims by origin period and development period. The origin period may be an accident year, underwriting year, or report year. Development year measures the time elapsed since origin.

A cumulative paid triangle might appear as follows:

Accident yearDevelopment 1Development 2Development 3Development 4
2021400650760800
2022450720850—
2023500780——
2024560———

The missing cells are future development. The actuary estimates them using historical patterns, exposure information, claim-level data, or alternative models.

3.2 Chain-ladder method

The volume-weighted age-to-age factor from development period j to j+1 is:

fj = [∑i Ci,j+1] ÷ [∑i Ci,j]

where Ci,j is cumulative claim amount for origin period i at development age j. The cumulative development factor to ultimate is:

CDFj = ∏k=jm−1 fk

where m is the final selected development age. The projected ultimate claim amount for origin period i is:

Ultimatei = Ci,j × CDFj

The reserve is ultimate less cumulative claims already paid or incurred, depending on the triangle basis.

3.3 Worked chain-ladder example

Using the triangle above, the age-to-age factors are:

f1 = (650 + 720 + 780) ÷ (400 + 450 + 500) = 2,150 ÷ 1,350 ≈ 1.5926
f2 = (760 + 850) ÷ (650 + 720) = 1,610 ÷ 1,370 ≈ 1.1752
f3 = 800 ÷ 760 ≈ 1.0526

Assume the selected ultimate development factor from development 2 is 1.1752 × 1.0526 ≈ 1.237. For 2022, the development-2 cumulative amount is 720, so:

Ultimate2022 ≈ 720 × 1.237 = 890.64

The indicated reserve is approximately:

Reserve2022 ≈ 890.64 − 720 = 170.64

This method is mechanically transparent but not automatically reliable. It assumes that historical development patterns remain informative for the selected origin period. Inflation changes, claims-management initiatives, legal reforms, catastrophe events, changes in policy mix, and inadequate historical data can invalidate that assumption.

3.4 Bornhuetter–Ferguson and expected-loss approaches

The chain-ladder method relies heavily on observed claims. For immature accident years, a priori expected loss information can be more stable. The Bornhuetter–Ferguson method combines an expected ultimate loss with the proportion of losses expected to have emerged.

Let EL be the a priori expected ultimate loss, and let pj be the expected proportion reported or paid by development age j. Then the expected unreported reserve is:

ReserveBF = EL × (1 − pj)

The indicated ultimate is observed cumulative claims plus this reserve:

UltimateBF = Observed cumulative claims + EL × (1 − pj)

This method is less sensitive than chain-ladder to random early experience, but it can be biased if the expected loss ratio, exposure, or emergence pattern is poor. Professional practice often compares several methods rather than selecting one mechanically.

4. IBNR and Claim Uncertainty

IBNR is commonly used as an umbrella term for obligations not fully represented in reported case reserves. It can include:

  1. claims incurred but not yet reported;
  2. claims reported late;
  3. claims reported but not yet entered into the database; and
  4. future development on known claims.

A narrow definition of pure IBNR may exclude development on reported claims, while management reports may use “IBNR” broadly. Definitions must be documented because inconsistent terminology can lead to double counting or omission.

Claim reserves can be estimated using:

  • chain-ladder and other development methods;
  • Bornhuetter–Ferguson;
  • expected-loss-ratio methods;
  • frequency–severity models;
  • claim-level case reserve models;
  • generalized linear models;
  • Bayesian and stochastic reserving models; and
  • scenario or stress testing.

The frequency–severity approach decomposes expected aggregate loss. If N is claim count and X is claim severity, then under a basic independence assumption:

E(S) = E(N) × E(X)

where S is aggregate loss. Independence is not essential for all models, but if frequency and severity are dependent, the simple product may be inadequate. Catastrophe claims, large losses, social inflation, and policy limits often require separate treatment.

5. Best-Estimate Assumptions and Margins

5.1 Best estimate

A best estimate is an unbiased or central estimate of the expected present value of future cash flows, based on information available at the valuation date. “Best” does not mean optimistic, most favourable to management, or equal to the most likely single outcome. For skewed distributions, the mean may differ materially from the mode or median.

Assumptions should be:

  • evidence-based and relevant to the portfolio;
  • internally consistent;
  • consistent with contract boundaries and data definitions;
  • documented with rationale and approval;
  • regularly monitored against actual experience; and
  • adjusted when credible evidence indicates a change.

Common assumptions include mortality, morbidity, lapse, claim frequency, claim severity, reporting delay, settlement speed, expense inflation, wage inflation, repair-cost inflation, discount rates, recovery rates, and reinsurance collectability.

5.2 Margins and risk adjustment

A best estimate alone does not necessarily compensate for uncertainty. A margin or risk adjustment may be required to reflect the compensation or capital cost associated with non-financial uncertainty. The treatment depends on the accounting or solvency regime.

A generic liability decomposition is:

Insurance liability = Best estimate of future cash flows + Risk adjustment or margin

The margin should not be confused with an arbitrary prudence loading. It should have a defined purpose, measurable methodology, and governance. Possible approaches include:

  • percentile or confidence-level methods;
  • cost-of-capital methods;
  • conditional tail expectation;
  • scenario-based margins; and
  • explicit parameter and model-risk allowances.

If a liability distribution is represented by random variable L, a percentile margin at confidence level q may be expressed as:

Marginq = VaRq(L) − E(L)

where VaRq(L) is the q-quantile of the liability distribution. This is not universally suitable: it may be unstable, may fail to reward diversification appropriately, and does not describe the entire tail. A cost-of-capital method instead discounts projected capital requirements and multiplies them by a prescribed or selected cost-of-capital rate.

5.3 Assumption setting and experience studies

An experience study compares actual outcomes with assumptions. For example, an observed lapse rate can be calculated as:

Observed lapse rate = Number of policies lapsed during period ÷ Policies exposed to lapse risk

The denominator must be defined carefully. Beginning-of-period policies, average in-force policies, policy-months, and exposure-weighted measures produce different results. Credibility increases with volume and homogeneity, but a large dataset can still be unrepresentative after a product redesign or distribution change.

Assumption setting should distinguish:

  • historical experience from future expectation;
  • random fluctuation from structural change;
  • portfolio mix effects from true behavioural change; and
  • data error from economic or demographic deterioration.

6. Discounting Insurance Liabilities

6.1 Principle

Discounting recognizes that a payment received later has a different economic value from an immediate payment. For a deterministic cash flow CFk payable at time k, discounted at annual effective rate rk, the present value is:

PV = ∑k=1n CFk ÷ (1 + rk)k

With a term structure, each cash flow should generally use a maturity-consistent rate. If rates are continuously compounded, the equivalent expression is:

PV = ∑k=1n CFk × e^(−yk k)

where yk is the continuously compounded zero rate. The compounding convention must match the quoted rate.

Discounting may be inappropriate or modified where liabilities are very short-term, where prescribed regulation uses undiscounted amounts, or where the framework requires specific treatment of illiquidity, matching adjustment, or risk-free curves. The analyst must not mix nominal cash flows with real discount rates. If cash flows include inflation, the discount rate must be nominal; if cash flows are expressed in real terms, a real rate is required.

6.2 Duration and interest-rate sensitivity

A liability’s sensitivity to rates can be approximated using duration. Macaulay duration is:

DM = [∑ k × PV(CFk)] ÷ [∑ PV(CFk)]

Modified duration under annual effective rates is:

Dmod = DM ÷ (1 + r)

For a small parallel change in yield ∆r, the approximate percentage price change is:

∆PV ÷ PV ≈ −Dmod × ∆r

The approximation assumes a small, parallel rate movement and unchanged cash flows. Insurance liabilities often have embedded options, changing cash flows, and non-parallel curve exposure, so duration alone is insufficient for full asset-liability management.

7. Reserve Uncertainty and Model Risk

A reserve estimate is uncertain for at least four reasons:

  1. Process risk: future claims naturally fluctuate even if the model is correct.
  2. Parameter risk: selected frequencies, severities, development factors, or mortality rates are estimated rather than known.
  3. Model risk: the model structure may omit relevant relationships or use unsuitable assumptions.
  4. Operational and data risk: coding errors, missing claims, inconsistent dates, and system migrations can distort results.

A useful governance practice is to report a central estimate together with a range, sensitivity analysis, and major sources of movement. Examples include:

  • a 5% increase in claim severity;
  • a slower settlement pattern;
  • higher medical inflation;
  • a reduction in lapse rates;
  • a widening of discount spreads; or
  • reduced reinsurance collectability.

Stochastic reserving generates a distribution rather than one number. However, simulation does not eliminate judgement. A sophisticated model with poor data or inappropriate dependence assumptions may be less reliable than a transparent deterministic model.

8. Illustrative Case: Meridian General Insurance

Background

Meridian General Insurance writes commercial liability and motor business. Its year-end claims triangle shows rapidly increasing bodily-injury costs. Management has also changed its claims-handling system, and a legal reform is expected to increase compensation awards.

Problem

The reserving team’s chain-ladder estimate is 48 million currency units. A Bornhuetter–Ferguson estimate is 53 million because the a priori loss ratio reflects the legal reform. Case reserves are incomplete for several large claims, and the finance director asks whether the lower chain-ladder estimate can be used to improve reported profit.

Analysis

The historical triangle is not fully comparable with the current portfolio. The claims system change affects reporting and development patterns, while legal reform affects severity. These changes undermine the assumption that historical age-to-age factors alone represent future development. The team therefore separates large claims, reviews claim-level case estimates, updates the expected loss ratio, and examines reporting delays before selecting a central estimate.

It identifies the following considerations:

  • ordinary claims remain reasonably represented by historical development;
  • large bodily-injury claims have a heavier and changing severity distribution;
  • the new system temporarily altered reporting lags;
  • reinsurance recoveries depend on attachment points and claim classification; and
  • settlement inflation is higher than the historical average.

Approach and solution

Meridian uses a segmented approach: a development method for stable motor property claims, claim-level analysis for large bodily-injury claims, and a Bornhuetter–Ferguson method for immature liability years. It selects a central estimate of 52 million after reconciling segments, adds an explicitly governed risk adjustment where required, and documents the reasons for rejecting an uncritical chain-ladder selection.

The actuary performs sensitivity tests for severity inflation and delayed settlement. The audit committee receives a bridge from the prior reserve, actual-versus-expected analysis, model changes, assumption changes, and uncertainty ranges.

Outcome and learning points

The selected reserve is higher than the initial chain-ladder indication, but it is better aligned with current exposures and known structural changes. The case demonstrates that reserving is not a contest between methods. It is a process of matching methods to data quality, portfolio characteristics, and causal conditions. It also illustrates why profit management through optimistic assumptions can threaten solvency and financial reporting credibility.

9. Governance, Regulation, and Professional Ethics

Reserve governance should establish clear responsibilities among underwriting, claims, actuarial, finance, risk, internal audit, and the board. A sound framework includes:

  • documented valuation policies and materiality thresholds;
  • controlled data lineage from source systems to reports;
  • independent review of material models;
  • assumption approval and change logs;
  • back-testing and actual-versus-expected analysis;
  • reconciliations between actuarial, finance, and claims records;
  • management of expert judgement; and
  • escalation of adverse development and reserve deficiencies.

Regulatory regimes may prescribe valuation bases, discount curves, contract boundaries, risk margins, admissibility rules, capital requirements, and public disclosures. Financial reporting standards may use different concepts from local statutory accounts. Therefore, one reserve cannot automatically be copied into every reporting purpose.

Ethical issues arise when management pressures actuaries to select assumptions that increase current earnings, when uncertainty is hidden in a single-point estimate, or when model limitations are not disclosed. Professional independence requires the actuary to communicate a technically defensible conclusion, including material uncertainty and alternative outcomes.

10. Technology and Contemporary Developments

Modern insurers increasingly use granular policy and claim data, cloud platforms, automated data checks, machine learning, and real-time dashboards. These tools can improve segmentation, fraud detection, claim triage, and assumption monitoring. Machine learning may identify nonlinear patterns in claim severity or settlement duration, but predictive accuracy does not by itself establish an appropriate liability estimate.

Key controls for AI-supported reserving include:

  • explainability appropriate to the decision;
  • stability under changing claims conditions;
  • monitoring for drift and biased outcomes;
  • clear separation between prediction and valuation judgement;
  • reproducible versioning of data and models; and
  • human review of exceptional or high-impact claims.

Climate change, cyber risk, medical inflation, social inflation, pandemic effects, and geopolitical disruption challenge historical-data methods. Emerging risks may require scenario analysis, expert judgement, external data, and explicit limitations rather than false statistical precision.

Best-Practice Framework

A practical reserve process can follow this sequence:

  1. Define the valuation purpose, reporting basis, contract boundary, and unit of account.
  2. Reconcile exposure, premium, claims, payment, and reinsurance data.
  3. Segment portfolios by genuinely different risk and development characteristics.
  4. Select methods appropriate to maturity, data quality, and structural change.
  5. Set and document best-estimate assumptions.
  6. Discount consistently where required, using appropriate curves and conventions.
  7. Quantify uncertainty through sensitivity, scenarios, or stochastic methods.
  8. Apply any required margin or risk adjustment without double counting.
  9. Perform independent review and management challenge.
  10. Monitor actual emergence and update assumptions when evidence warrants it.

Chapter Summary

Insurance liabilities represent future obligations created by past and current contracts. Premium liabilities concern unexpired coverage, while claim liabilities concern insured events that have occurred, including reported claims, development on known claims, and IBNR. Life policy reserves are commonly valued as the present value of future benefits and expenses less the present value of future premiums, subject to the selected actuarial basis.

Non-life reserving uses claims triangles, development factors, expected-loss methods, claim-level analysis, and statistical models. Chain-ladder is transparent but depends on stable historical development. Bornhuetter–Ferguson introduces a priori expected losses and can be more robust for immature periods. Best-estimate assumptions should be evidence-based, internally consistent, documented, and regularly tested against experience.

Discounting converts future cash flows into present values, but its use requires consistent treatment of inflation, rate conventions, timing, and contract features. A reserve is uncertain because of process, parameter, model, data, and operational risks. Margins or risk adjustments should have a defined purpose and avoid arbitrary or duplicated prudence. Effective governance combines actuarial expertise, financial control, claims knowledge, independent review, ethical judgement, and transparent communication of uncertainty.

Key Terms

Actuarial assumption: A demographic, economic, behavioural, or claim-related estimate used in valuation.

Best estimate: A central estimate of expected future cash flows using information available at the valuation date.

Case reserve: An amount set aside for a reported claim based on information about that claim.

Chain-ladder: A claims-development method that projects ultimate claims using historical age-to-age factors.

Claim liability: The obligation relating to insured events that have already occurred.

Contract boundary: The point beyond which future cash flows are not treated as arising from an existing insurance contract under a specified framework.

Development factor: A ratio describing change in cumulative claims between development periods.

Discounting: Converting future cash flows to present value using an appropriate interest or discount rate.

Gross premium reserve: A reserve based on gross contractual premiums, benefits, expenses, and other relevant cash flows.

IBNR: Incurred but not reported, often used broadly for claims not fully captured by reported case reserves.

Loss development triangle: A table organizing claims by origin period and development age.

Margin: An amount added to a best estimate to reflect uncertainty or required compensation for risk.

Macaulay duration: The present-value-weighted average timing of cash flows.

Premium liability: The expected obligation associated with future coverage under existing contracts.

Prospective reserve: A reserve calculated from the present value of future outgo less future inflows.

Risk adjustment: An amount reflecting non-financial uncertainty, according to the applicable framework.

Ultimate claim: The estimated total cost of a claim cohort after all future development.

Unearned premium reserve: Premium allocated to coverage that has not yet expired.

Valuation basis: The specified rules, assumptions, methods, and conventions used to measure liabilities.

Review Questions

Short Answer Questions

  1. Distinguish a premium liability from a claim liability.
  2. What is the purpose of a life insurance reserve?
  3. Define IBNR and explain why its definition should be documented.
  4. State two limitations of the chain-ladder method.
  5. Why must inflation and discount-rate conventions be consistent?

Long Answer Questions

  1. Explain the prospective method for valuing a life policy reserve and discuss the assumptions that may enter a gross premium valuation.
  2. Describe the chain-ladder method, including development factors, cumulative development factors, and ultimate claims.
  3. Compare chain-ladder, Bornhuetter–Ferguson, and claim-level reserving approaches.
  4. Discuss process risk, parameter risk, model risk, and data risk in insurance reserving.
  5. Explain the governance and ethical responsibilities involved in setting insurance liabilities.

Analytical / Application Questions

  1. An accident year has cumulative paid claims of 900 at development age two. The selected development factor to ultimate is 1.18. Calculate the indicated ultimate claim and reserve, and identify the assumptions behind the calculation.
  2. A life insurer’s lapse experience has fallen materially after a product redesign. Design an experience-study and assumption-governance process for deciding whether the best-estimate lapse assumption should change.
  3. A liability portfolio has a chain-ladder estimate of 40 million and a Bornhuetter–Ferguson estimate of 46 million after a legal reform. Explain what additional evidence you would obtain before selecting a reserve and reporting a margin or uncertainty range.

Multiple Choice Questions with Answers

Question 1

Which item most directly represents a claim liability?

A. Premium for coverage beginning next year

B. Expected cost of an accident that has already occurred

C. Future marketing expenditure unrelated to contracts

D. Shareholder dividend declared after year-end

Correct Answer: B. Expected cost of an accident that has already occurred

Question 2

Under a simplified prospective reserve approach, the reserve equals:

A. Future premiums minus future benefits only

B. Past premiums plus investment income

C. Present value of future benefits minus present value of future net premiums

D. Written premium divided by policy duration

Correct Answer: C. Present value of future benefits minus present value of future net premiums

Question 3

Using cumulative claims of 900 and a selected ultimate development factor of 1.18, what is the indicated reserve?

A. 162

B. 180

C. 1,062

D. 762

Correct Answer: A. 162

Question 4

What is the principal purpose of a claims-development triangle?

A. To calculate shareholder dividends

B. To organize claims by origin period and development age

C. To determine an insurer’s tax rate

D. To replace all claim-level information

Correct Answer: B. To organize claims by origin period and development age

Question 5

The Bornhuetter–Ferguson method combines observed claims with:

A. A priori expected ultimate losses and an emergence pattern

B. Only the oldest accident year

C. The insurer’s market share

D. A zero-reserve assumption

Correct Answer: A. A priori expected ultimate losses and an emergence pattern

Question 6

Which statement best describes IBNR?

A. Only claims that have already been paid

B. Premiums that have not yet been collected

C. Obligations not fully represented in reported case reserves

D. Investment income earned on reserves

Correct Answer: C. Obligations not fully represented in reported case reserves

Question 7

If future cash flows include inflation, the discounting analysis should generally use:

A. A real discount rate with nominal cash flows

B. A nominal discount rate consistent with nominal cash flows

C. No time periods

D. Only the shortest available rate

Correct Answer: B. A nominal discount rate consistent with nominal cash flows

Question 8

Which is an example of parameter risk?

A. A random fluctuation in one claim portfolio

B. Uncertainty in the selected claim severity assumption

C. A database password failure

D. A legal requirement to publish accounts

Correct Answer: B. Uncertainty in the selected claim severity assumption

Question 9

A risk adjustment or margin should ideally be:

A. Arbitrary and undisclosed

B. Designed to conceal poor data

C. Defined, governed, and measured consistently with its purpose

D. Added twice under different names

Correct Answer: C. Defined, governed, and measured consistently with its purpose

Question 10

Which factor most directly weakens the use of historical chain-ladder factors?

A. Stable exposure and unchanged claims practice

B. A structural legal reform that changes claim severity

C. Complete and comparable historical data

D. Consistent development patterns across years

Correct Answer: B. A structural legal reform that changes claim severity

Insurance Finance & Solvency Models
Chapter 3 of 14