Inside the Numbers: How CSM Run-off, Risk Adjustment, and Onerous-Contract Testing Are Quietly Reshaping Earnings

 Three IFRS 17 mechanics that most Caribbean boards do not yet fully understand — and which determine reported profits more than any underwriting decision

 

Three IFRS 17 mechanics — the Contractual Service Margin run-off pattern, the risk adjustment calibration, and the onerous contract test — together drive the majority of reported insurance earnings under the new standard. Each is governed by methodology choices made at transition and rarely revisited since. Each can move reported profits in any given period by amounts that would, in any other industry, be considered material. And each is now attracting scrutiny from auditors, regulators, and rating agencies in ways that boards should be asking different questions about than they were two years ago.

 

Why these three mechanics matter more than the rest

IFRS 17 contains hundreds of pages of guidance, dozens of measurement choices, and an extensive disclosure architecture. Most of it is procedural — important to get right, but rarely consequential to reported earnings in any single period. Three mechanics are different. They share two characteristics: each translates assumption choices directly into reported profit, and each operates with degrees of management discretion that the standard explicitly acknowledges. Together, they account for most of what moves the profit and loss account between periods under IFRS 17.

Article 03 in this series argued that the strategic transition to IFRS 17 in the Caribbean is unfinished. Article 04 goes inside the three mechanics where that unfinished work is most visible — and where the next 24 months of audit, regulatory, and rating-agency attention will be most concentrated. This is the technical article in the IFRS 17 pair, written for chief actuaries, audit committees, and external auditors. The exposition is plain English but does not avoid the substance.

Three mechanics translate assumption choices directly into reported profit — and each operates with degrees of management discretion that the standard explicitly acknowledges.

 

Mechanic 1 — Contractual Service Margin run-off

What the CSM actually does

The Contractual Service Margin is the unearned profit that an insurer expects to earn over the remaining life of a group of contracts. At inception, the CSM is calculated such that no profit is recognised on day one for a profitable group of contracts — the profit instead sits as a liability that is released to the profit and loss account over the period during which services are provided. The CSM is, in this sense, the most distinctive liability in insurance accounting: it is genuinely an unearned-profit balance, sitting on the balance sheet, waiting to flow through the income statement.

The mechanics of CSM release are governed by two choices that, in combination, shape almost every IFRS 17 income statement in the Caribbean. The first is the coverage unit pattern — the basis on which the carrier measures the quantity of services provided in each period. The second is the treatment of assumption changes — specifically, which assumption updates adjust the CSM (and therefore future profit) versus which flow directly through the profit and loss account in the current period.

Where the choices bite

Coverage unit selection is the choice that most boards have not appreciated as a profit-shaping decision. For a group of long-duration life insurance contracts, coverage units can be measured on the basis of expected face amount in force, number of policies in force, expected premium in force, or several other defensible bases. Each gives a different pattern of CSM release. A face-amount basis releases CSM more quickly in the early years of a level-premium product; a policy-count basis tends to release more slowly. Across a typical Caribbean life book, the choice can move reported insurance service result by amounts that materially affect period-on-period earnings comparison.

Assumption-change treatment is the second consequential choice. Changes in non-financial assumptions — mortality, lapse, expense — adjust the CSM. Changes in financial assumptions, depending on the carrier’s election under IFRS 17, may either adjust the CSM at a locked-in rate or flow through profit and loss at current rates. The locked-in versus current-rate distinction sounds technical. It is in fact one of the most consequential choices a carrier makes at transition, because it determines how much of the income statement is exposed to interest-rate movements quarter on quarter. Caribbean carriers, broadly, have made the right structural elections. Few have explained those elections to their boards in terms boards can use to interpret quarterly results.

What good practice looks like

A carrier with strong CSM discipline reviews the coverage unit pattern at least annually against actual experience, refreshes assumption-setting governance such that assumption changes are attributed cleanly between CSM-adjusting and current-period items, and produces an earnings walk for the board that shows how much of any quarter’s reported insurance service result is driven by CSM release versus current-period items versus assumption changes. None of this is required by the standard, but all of it transforms the CSM from a passive balance into a managed liability.

THE BOARD-LEVEL QUESTION If the audit committee cannot answer the question ‘how much of the change in CSM this quarter was driven by experience emergence, how much by assumption updates, and how much by the underlying release pattern?’, then the CSM is not yet being managed. It is being calculated. Calculation is mandatory; management is what creates the analytical value the standard was designed to provide.

Mechanic 2 — The risk adjustment

What the risk adjustment actually does

The risk adjustment is the compensation an insurer requires for bearing non-financial risk — mortality, morbidity, lapse, and expense uncertainty. It sits within the measurement of insurance liabilities and is released to profit and loss as the underlying risks expire. The standard deliberately leaves the methodology open: carriers may use a confidence-level calibration (the most common approach), a cost-of-capital approach, or any other technique provided it produces a defensible risk-adjustment outcome and supports the required disclosures. The standard does not specify a method; it specifies a principle.

The risk adjustment matters for three reasons. It moves with each period’s actual versus expected experience. It is one of the most disclosed numbers in IFRS 17 financial statements. And, more than any other mechanic, it is the number that signals to capital markets and regulators how a carrier understands its own risk profile. A risk adjustment that does not move when material assumption changes occur, or that bears no relationship to the carrier’s own risk appetite framework, signals that the calibration is mechanical rather than analytical.

Where Caribbean carriers stand

Most Caribbean carriers selected a confidence-level approach at transition — typically calibrating the risk adjustment to a target percentile (often somewhere between the 65th and 80th percentile of the distribution of insurance liability outcomes). The percentile choice was usually documented, defensible, and consistent with the practice of regional peers. Three years on, two problems are emerging. First, the percentile was set and has not been challenged. Second, the resulting risk adjustment bears limited demonstrable relationship to the carrier’s actual capital costs, pricing margins, or reinsurance economics — raising the question of whether the calibration genuinely reflects the carrier’s view of risk, or simply produces a defensible disclosure number.

What good practice looks like

Strong risk-adjustment practice connects the calibration to the carrier’s own risk-appetite framework. Whatever percentile or cost-of-capital level is chosen, that level should be re-tested annually against the carrier’s actual risk metrics: solvency margin movements, reinsurance retention decisions, pricing target loss ratios, and capital management actions. The risk adjustment becomes credible when the audit committee can ask why the percentile is what it is, and the answer references the carrier’s own risk-appetite statements rather than industry custom. The carriers that achieve this connection will pass audit, regulator, and rating-agency scrutiny in a single conversation. Those that have not will face that scrutiny in three.

THE BOARD-LEVEL QUESTION If the chief actuary is asked why the risk adjustment is calibrated at the level it is, and the answer is ‘because that is what we selected at transition and the auditor accepted it’, then the risk adjustment is not yet connected to the business. The correct answer references the carrier’s own risk appetite, the cost of capital actually paid by the carrier, the pricing margins targeted on new business, or the reinsurance retentions actually held. Any of these is defensible. The first answer is not.

Mechanic 3 — Onerous contract testing

What the onerous contract test actually does

IFRS 17 requires that groups of contracts expected to be unprofitable at inception — generating losses over their lifetime, taking into account the time value of money and the risk adjustment — be identified as onerous and recognised immediately as losses, rather than spread over the term of the contracts. The mechanic forces transparency about pricing inadequacy at the point of issue. Done well, it is a powerful pricing-discipline signal. Done as a compliance exercise, it produces a balance-sheet entry without changing the underlying behaviour.

The test runs at the level of groups of contracts — contracts that share similar risk profiles and are managed together, with the standard limiting how broadly contracts may be aggregated for this purpose. The aggregation rules are designed specifically to prevent profitable contracts from masking onerous ones; the granularity is therefore meaningful and the testing can surface pockets of unprofitability that would not be visible in aggregate.

Where the gap is

Across the Caribbean, onerous testing has largely become a retrospective compliance exercise. Groups are issued, the test is run after the fact, an onerous loss is recognised where indicated, and the carrier moves on. The pricing function and the IFRS 17 measurement function operate in parallel. Onerous results are reported to the audit committee as part of the financial statement close package; they are rarely fed back into the pricing committee in time to change new-business pricing or in-force renewal strategy. The signal that the standard architects intended to produce is being produced; it is simply not being received.

What good practice looks like

Strong onerous-testing practice runs the test pre-issue on representative new-business cohorts — stress-testing the pricing assumptions against the IFRS 17 measurement basis before the cover is bound. It runs on a rolling basis on in-force groups, with deteriorating groups flagged in time for renewal-pricing or product-redesign intervention. It feeds the pricing committee, not just the audit committee. And it treats onerous emergence as a leading indicator of pricing discipline rather than a lagging indicator of past mistakes. None of this is required by the standard. All of it changes onerous testing from accounting output to pricing discipline — which is what the standard architects intended.

THE BOARD-LEVEL QUESTION If the pricing committee and the audit committee are looking at different views of contract profitability, the carrier is leaving signal on the table. The most useful single integration a Caribbean insurer can make in the next 24 months is to have onerous-test results in front of the pricing committee at the same time they are in front of the audit committee, with the same granularity and the same forward-looking implications. The technical change required is minimal. The behavioural change required is substantial.

How the three mechanics interact

The three mechanics are not independent. A change in mortality assumption, for example, can simultaneously adjust the CSM (releasing a different amount of future profit), shift the risk adjustment (changing the current-period release), and tip a borderline group into onerous classification (triggering immediate loss recognition). The interactions produce earnings movements that, on the surface, can look counter-intuitive. A carrier reporting strong underwriting experience may see reported insurance service result decline because the experience caused an assumption update that released CSM more conservatively. A carrier reporting weak underwriting experience may see reported insurance service result hold up because the risk adjustment release offset the experience loss.

These interactions are not a flaw in the standard. They are a faithful representation of the economics of long-duration insurance contracts — in which the measurement of profit in any single period is genuinely complex, and in which the carrier’s choices about how to attribute movements between mechanics affect what reported profit looks like even when the underlying economics are unchanged. The carriers that explain these interactions to their boards, their auditors, and their analysts in a structured way — typically through a quarterly earnings walk that decomposes the period movement — are positioned favourably. The carriers that report the mechanics in isolation are leaving the reader to do the decomposition unaided, which is a poor user experience and a missed signalling opportunity.

THE EARNINGS WALK The single most useful disclosure innovation an IFRS 17 reporter can produce is a quarterly earnings walk that decomposes the period movement in insurance service result into its CSM, risk adjustment, onerous contract, experience variance, and assumption change components. It is not required by the standard. It is increasingly expected by analysts and rating agencies. And it is, in practice, the cleanest single signal a board can give that it understands its own income statement.

Where scrutiny is heading on each mechanic

Three years into IFRS 17, the technical-correctness phase of audit and regulatory attention is closing. The methodology phase is opening. On each of the three mechanics, the questions being asked by auditors, supervisors, and rating agencies have shifted in distinctive directions.

On CSM

Audit attention is increasingly focused on whether the coverage unit pattern remains appropriate as portfolios mature, whether assumption-change attribution is being applied consistently, and whether the locked-in versus current-rate elections continue to produce sensible outputs. Regulator attention is on whether the CSM movements over time tell a coherent story about the carrier’s emerging profitability. Rating-agency attention is on whether the CSM release pattern looks sustainable or whether it has been front-loaded in ways that flatter near-term earnings.

On risk adjustment

Audit attention is on whether the calibration has been genuinely re-tested rather than mechanically rolled forward. Regulator attention is on whether the risk-adjustment level is consistent with the carrier’s stated risk appetite and capital management framework. Rating-agency attention is on whether the risk adjustment moves credibly in response to changes in the portfolio mix or reinsurance arrangements — a non-moving risk adjustment in a changing portfolio is a flag.

On onerous contract testing

Audit attention is on the rigour of the aggregation: whether groups are being defined narrowly enough to satisfy the standard. Regulator attention is on whether onerous emergence is being addressed in pricing and product design or simply absorbed quietly through additional capital. Rating-agency attention is on whether the carrier’s onerous-loss recognition pattern indicates a pricing problem the carrier has not yet acted on.

In each case, the direction of travel is the same: from ‘is the calculation right?’ to ‘does the calculation reflect the underlying business?’ Carriers that have built the analytical infrastructure to answer the second question will face shorter and easier conversations. Carriers that have not will face longer ones.

The direction of travel is from ‘is the calculation right?’ to ‘does the calculation reflect the underlying business?’

Why this matters now

Article 03 in this series argued that the IFRS 17 transition produced machinery that is now operational but has not yet been connected to the business decision-making the standard was designed to enhance. Article 04 has gone inside three of the mechanics where that disconnection is most visible — and most expensive to leave unresolved. CSM, risk adjustment, and onerous testing are the three places where reasonable methodology choices, applied with discipline, produce measurably better-informed boards and measurably easier conversations with auditors, regulators, and rating agencies. Each is buildable in the next 24 months. Each requires modest investment in actuarial and finance capacity rather than enterprise-scale capital commitment. And each, once built, becomes a permanent capability that compounds with each reporting cycle.

The carriers that take these three mechanics seriously over the next two years will reach a point at which the IFRS 17 income statement tells them something they did not previously know about their own business. The carriers that do not will continue to produce the disclosures the standard requires without extracting the analytical value the standard was designed to provide. Both outcomes are within reach. The choice is, again, made one mechanic at a time.

ABOUT THE SERIES The Caribbean Actuarial Imperative is a 16-article series from Dawgen Global’s Actuarial & Insurance Regulatory Advisory Division. The series examines the structural shifts reshaping Caribbean insurance — pricing, reserving, reinsurance, enterprise risk, regulation, experience data, modelling technology, and transactions — and what insurance boards, executives, and regulators should be doing about them. The Actuarial & Insurance Regulatory Advisory Division is Fellowship-led, independent of any global broker or reinsurance group, and integrated with Dawgen Global’s broader Risk Advisory, Audit & Assurance, Tax Advisory, M&A, IT, and Cybersecurity practices. Enquiries: [email protected] Please reference ‘Actuarial Division’ in your subject line.

 

PREVIOUSLY IN THE SERIES Article 03 IFRS 17, Three Years On: What Caribbean Insurers Got Right, What They Got Wrong, and What Comes Next

NEXT IN THE SERIES Article 05 The Reinsurance Trap: Why Caribbean Cedants Pay Too Much and Capture Too Little

◆ ◆ ◆

Big Firm Capabilities. Caribbean Understanding.

About Dawgen Global