Impracticability of full retrospective transition approach

Reference to standard: IFRS 17 paras C3, BC 381–BC 382; IAS 8 para 5

Background

According to paragraph C3 of IFRS 17, an entity should apply IFRS 17 retrospectively unless impracticable. Only if it is impracticable for an entity to apply paragraph C3 for a group of insurance contracts can an entity apply alternative approaches.

Retrospective application involves assessing each component – that is, fulfilment cash flows (FCF) and contractual service margin (CSM) – at each point in time from inception (the year of sale for the policy with the longest duration in the group of contracts), using all known factors including in-year actual experience and relevant assumptions.

The CSM would be adjusted at each period for certain assumption changes and the prospective impact of experience variances (where these affect future cash flows). As such, this would require complete policyholder data, a complete set of assumptions (demographic and economic), and pre-coverage cash flows including premiums and acquisition costs. This process would be repeated, adjusting the CSM for each reporting period based on the specific assumptions for each reporting period until the date of transition.

Question

How should an entity assess whether or not a retrospective approach of IFRS 17 to groups of in-force contracts is ‘impracticable’ at the date of transition to IFRS 17?

Answer

IAS 8 states that, “Applying a requirement is impracticable when the entity cannot apply it after making every reasonable effort to do so”. Effort relates to collection of information, but it also relates to the creation of information where an entity did not collect the relevant data in the prior periods. In the process of collecting or creating information, the additional guidance in paragraph 5 of IAS 8 should be considered:

“For a particular prior period, it is impracticable to apply a change in an accounting policy retrospectively or to make a retrospective restatement to correct an error if:

·        the effects of the retrospective application or retrospective restatement are not determinable;

·        the retrospective application or retrospective restatement requires assumptions about what management’s intent would have been in that period; or

·        the retrospective application or retrospective restatement requires significant estimates of amounts and it is impossible to distinguish objectively information about those estimates that:

o   provides evidence of circumstances that existed on the date(s) as at which those amounts are to be recognised, measured or disclosed; and

o   would have been available when the financial statements for that prior period were authorised for issue from other information.”

Effects are ‘not determinable’ if there is not enough data available to be able to determine the impact of IFRS 17 components at inception or, between inception and the transition date, to perform the CSM roll-forward to the transition date. For example, best estimate assumptions might not have been documented or saved. There will be judgement required in determining whether existing data is available, or can be recreated, at an appropriate granularity and at a quality that can be used in the retrospective calculations. These reasons will be less likely to be supportable for groups of contracts issued closer to the transition date.

IAS 8 highlights the need to make estimates in applying accounting policies, and it notes that developing estimates is potentially more difficult in retrospective application, because of the longer time period that might have passed since the affected transaction occurred. Paragraph 53 of IAS 8 states that hindsight should not be used, so there might be areas of the calculations where retrospective application will not be possible. This means that, for retrospective application to be practicable, the measurement of the present value of future cash flows, risk adjustment and CSM, at inception and in subsequent periods, should apply management’s estimates at that point in time, with only the information that would have been available at that point in time.

Examples of amounts needed for retrospective application, but that might often be impracticable to be measured for a group of insurance contracts at the transition date (because they can be determined only using hindsight), are included in paragraph BC 378 of the Basis for Conclusions to IFRS 17:

a. the estimates of cash flows at the date of initial recognition;

b. the risk adjustment for non-financial risk at the date of initial recognition;

c. the changes in estimates that would have been recognised in profit or loss for each accounting period because they did not relate to future service, and the extent to which changes in the fulfilment cash flows would have been allocated to the loss component;

d. the discount rates at the date of initial recognition; and

e. the effect of changes in discount rates on estimates of future cash flows for contracts for which changes in financial assumptions have a substantial effect on the amounts paid to policyholders.

Further, Basis for Conclusions to IFRS 17 indicate that, for contracts in force on transition, performing the following assessments at the initial recognition of the contracts would often be impracticable (that is, they would be impossible without the use of hindsight):

·        eligibility for the variable fee approach;

·        level of aggregation; and

·        effect of discretion on estimated cash flows for contracts subject to the general model.

In addition, the Basis for Conclusions explains that, for the fully retrospective approach, it will be necessary to estimate the effect of contracts derecognised before the transition date on the allocation of the CSM between past and future periods on transition date. This might be particularly challenging for large portfolios of long-term contracts, for which terms and circumstances (for example, size and number of contracts issued in prior reporting periods) often change.

Entities will need to consider, for each component within the measurement model, what data is available per group for contracts, and thus to determine what historical data is available, or can be recreated, and to assess the extent to which the retrospective approach calculation could be performed without the use of hindsight.

Why does IFRS 17 permit three different transition approaches?

Reference to standard: IFRS 17 paras C3, C4

IAS 8 generally requires retrospective application of new accounting policies, unless it is impracticable to do so. This is also the starting point for initial application of IFRS 17. Insurers might issue long-term contracts, especially in the life insurance business. Gathering information for contracts issued many years before transition could be either very costly or not possible. Determining assumptions and estimates for such contracts at the contract inception date might not be possible without the use of hindsight. So the IASB offers insurers additional alternatives for transition where full retrospective application is impracticable.

The IASB wants to achieve a transition amount that is as close to the retrospective application as possible where only some information is missing. In such a situation, insurers could use the modified retrospective approach that provides specified simplifications compared to the full retrospective approach. However, each modification to the fully retrospective approach would mean that the outcome is more different than one that would have been obtained if a fully retrospective approach had been applied.

At the same time, the IASB acknowledged that there might be very little information available on transition. Application of the modified retrospective approach in such cases could be costly. So the IASB permitted a fair value approach as an accounting policy choice to reduce the implementation costs.

Will there be differing results from applying each of the three transition approaches?

Reference to standard: IFRS 17 paras C4, C5

Yes. The result of measurement of the same group of contracts, from applying the three transition approaches, will be different. The measurement using the full retrospective approach and the modified retrospective approach might be relatively close, compared to the fair value approach, depending on the availability of information and simplifications used. The fair value approach might provide a very different measurement compared to the two retrospective approaches. For example, there might be no contractual service margin when applying the full retrospective approach or the modified retrospective approach if the contracts are onerous, while there might be a contractual service margin when the fair value approach is applied, because IFRS 13 indicates that the fair value includes the profit margin that a market participant would require to accept obligations under insurance contracts.

The IASB acknowledged that the optionality available on transition will result in less comparable financial statements until the contracts written before the transition date are derecognised. However, the IASB concluded that the costs associated with the full retrospective approach or modified retrospective approach might exceed the benefits if there is little information available on the transition date. Thus, the fair value approach is permitted as an accounting policy choice in accordance with IAS 8 if a full retrospective approach is impracticable.

How are contracts that are acquired in a business combination measured on transition?

Reference to standard: IFRS 17 para B94

As discussed, the coverage period, contractual service margin (CSM) and locked-in assumptions might be different in the financial statements of an acquired subsidiary and in the consolidated financial statements of the group, because the contract inception date is:

·        the date when the contracts were written, in the financial statements of a subsidiary; and

·        the date when contracts are purchased, in the consolidated financial statements of a group.

Measurement on transition in the financial statements of an acquired subsidiary and in the consolidated financial statements will be different, depending on the transition approaches that are used.

Measurement will be different, both on transition and after transition, if the full retrospective or modified retrospective approaches are used. The dates of initial measurement of contracts and the amount used to calculate the CSM will be different for the acquired subsidiary and the group, resulting in different retrospective measurement. Coverage periods for the acquired subsidiary might be shorter than coverage periods for the acquiring entity, and this will result in different measurement of the CSM after transition.

There will be no differences in measurement of insurance contracts on transition if the fair value approach is used, because the fair value is expected to be the same, whether viewed from a subsidiary or from a group perspective. However, after transition the measurement might be different, because the coverage periods might still be different, resulting in different measurement of the CSM.

What are some of the practical implications of IFRS 17 transition choices?

Reference to standard: IFRS 17 para C2 Reference to standing text: 50A.254 Industry: Insurance

The choice between the modified retrospective approach and the fair value approach on transition will impact shareholders’ equity on transition and the release of profit from the insurance contracts in force after transition. It is also likely to affect operational complexity and the cost of IFRS 17 implementation. In addition, some profits from insurance contracts might not be recognised at all in profit or loss (that is, they would not have been recognised in profit or loss under IFRS 4 and will be recognised as an adjustment to equity on transition to IFRS 17), while other profits might be recognised in profit or loss twice (that is, they would have been recognised in profit or loss under IFRS 4 and will be recognised in the CSM and then in profit or loss after transition to IFRS 17). This is an unavoidable result of differences between measurement approaches used under IFRS 4 and on transition to IFRS 17.

Equity on transition and release of profit from insurance contracts after transition

The CSM on transition represents the profit from insurance contracts in force that insurers will earn after transition. On transition, the higher the CSM is, the lower is the accumulated profit from insurance contracts recognised in shareholders’ equity and the more profit insurers will recognise in future periods until the end of the coverage of insurance contracts in force on transition. This might impact the ability to pay future dividends, solvency capital or taxation, depending on local legal and regulatory requirements. This might also impact the way in which investors assess the performance of the entity on transition and at future dates until the end of the coverage period of the contracts in force on transition. Additional disclosures will be required, as explained.

Operational complexity and cost of IFRS 17 implementation

The complexity of application of the different transition approaches and the cost of transition to IFRS 17 might be different, depending on the availability of information. Generally, for insurance contracts issued a long time before the transition date, the full retrospective approach and the modified retrospective approach will be more expensive to apply. For short-term contracts and contracts issued close to the transition date, more information is likely to be available, and the fair value approach might be more complex and more expensive compared to the alternatives.

Can IFRS 17 accounting policies be adopted prior to the effective date as a change in accounting policies under IFRS 4?

Reference to standard: IFRS 4 paras 22, BC 123; IAS 8 paras 19(b), 27

Yes; however, following IAS 8 guidance on retrospective application will not permit the IFRS 17 specified transitional approaches (modified retrospective application and fair value approach) when entities change their accounting policies under IFRS 4.

Under IFRS 4 an entity can change its existing accounting policies for insurance contracts where the change results in presenting information that, compared to existing policies, is more relevant and no less reliable, or more reliable and no less relevant. IFRS 4 notes that “as the Board’s conclusions for phase II develop, they will give insurers further context for judgements about whether a change in accounting policies meets the criteria”. Therefore, changing to an accounting policy that is an IFRS 17 requirement is considered to meet the criteria for changes. However, a change in accounting policies under IAS 8 requires retrospective application, unless impracticable. The simplified transitional approaches in IFRS 17 (modified retrospective application and fair value approach) are not available for use in such situations, because they are not part of IAS 8 or IFRS 4.

Consequently, if an entity chooses to change its IFRS 4 accounting policies to IFRS 17 policies prior to the effective date, and retrospective application is impracticable, paragraph 27 of IAS 8 notes that “When it is impracticable for an entity to apply a new accounting policy retrospectively, because it cannot determine the cumulative effect of applying the policy to all prior periods, the entity, in accordance with paragraph 25, applies the new policy prospectively from the start of the earliest period practicable …” . As a result, entities that cannot apply the new policy retrospectively, and instead elect to adopt prospectively under IFRS 4, will need to perform another transition for past contracts under IFRS 17 transition rules when adopted.

In addition, the disclosure requirements will continue to follow IFRS 4 until the entity adopts IFRS 17. However, an entity is permitted to include other disclosures, and so it could include the additional disclosures required by IFRS 17 where relevant.

On initial adoption of IFRS 17 when applying IFRS 17 retrospectively, should the impact of reporting at interim periods be considered?

Reference to standard: IFRS 17 para C4; IAS 8 para 22

Question

When an entity applies IFRS 17 for the first time, in calculating the contractual service margin (CSM), is an entity required to perform the calculation as if it were following its actual interim reporting in accordance with IAS 34 in presenting the effect of adoption on all past interim and annual reporting periods?

Answer

Yes, subject to materiality considerations. IAS 8 notes that, if an entity is changing its accounting policies retrospectively, it should adjust equity (and the comparatives) as if the new accounting policy had always been applied.

Some of the adjustments to the CSM might be different in an interim period compared to the annual reporting period. For example, IFRS 17 requires the CSM to be adjusted for changes in cash flows relating to future service, and future service is measured from the end of the current reporting period (including any interim reporting period). Therefore, the calculation of CSM adjustments for an entity subject to interim reporting could differ from an entity that only does annual reporting. Applying IFRS 17 retrospectively means that an entity needs to consider the retrospective impact following its normal interim reporting under IAS 34, calculating the ‘roll forward’ of the CSM from inception, considering each interim reporting period, subject to materiality considerations.

Can contracts issued more than one year apart be grouped together when applying the modified retrospective approach on transition?

Reference to standard: IFRS 17 paras C8, C10

Yes. IFRS 17 require an entity to establish groups that do not include contracts issued more than one year apart, if the entity has reasonable and supportable information to do this, when applying the modified retrospective approach on transition. The IASB acknowledged that it might not always be practicable for entities to group contracts written in the same year retrospectively, and therefore it provided this transitional relief so that entities would not have to divide contracts that are in place on transition into annual cohorts.

How could the fair value of a group of contracts be determined under IFRS 13?

Reference to standard: IFRS 17 para C20

IFRS 13 requires a fair value measurement of a liability to be a value that a market participant would require in return for assuming all of the obligations related to that liability. Entities are already using IFRS 13 for measurement of insurance contracts at fair value where they have business combinations accounted for in accordance with IFRS 3 or portfolio transfers. For business combinations and portfolio transfers, insurers generally have a market price, which is the transaction price. On transition to IFRS 17, there will be few or no observable market prices for many insurance contracts. This will require additional judgement compared to the use of IFRS 13 for business combinations or portfolio transfers today.

Discount rate

Cash flows and discount rates should reflect the assumptions that market participants would use when pricing the asset or liability. To avoid double counting or omitting the effects of risk factors, discount rates should reflect assumptions that are consistent with those inherent in the cash flows.

In contrast to the measurement requirements under IFRS 17: own credit risk related to the liability should be considered in a fair value measurement; and an asset-based discount rate could be used if such approaches are used by market participants in pricing liabilities in actual transactions.

Profit margin

Irrespective of the approach used for measurement of the fair value, entities should compare the results to any actual market transactions, and they should adjust the measurement to consider the actual market transactions or quotes for similar groups of contracts. In particular, an acquirer of a group of insurance contracts is likely to require a profit margin (above the probability-weighted, risk-adjusted discounted cash flows), even if the contract would have no unearned profit if a full retrospective approach had been applied on transition.

Application of the fair value approach might be expected to result in the fair value of a group of insurance contracts being higher than its fulfilment cash flows. This is because IFRS 13 considers the fair value of insurance contracts from the perspective of a market participant, and a market participant would generally require a profit margin on top of the fulfilment cash flows to accept a liability. This will be different from the full retrospective measurement or modified retrospective approach, where it is generally expected that onerous insurance contracts will have no contractual service margin, because losses will be immediately recognised in profit or loss.

Can an alternative existing measure of economic value be used as the fair value of contracts for the purposes of transition to IFRS 17?

Reference to standard: IFRS 17 para C20

Historically, many insurers issuing long-term contracts applied Market Consistent Embedded Value (MCEV) or European Embedded Value (EEV) Principles, issued by the European Insurance CFO Forum, to measure insurance contracts for supplementary reporting purposes. Where still produced, entities will be able to use these measurements or other economic-based regulatory measures (such as Solvency II in Europe) as a starting point for the fair value approach on transition to IFRS 17. However, entities should ensure that such measurement is consistent with the IFRS 13 requirements, and they should adjust it for any differences.

Insurance acquisition cash flows when using the fair value approach on transition

Reference to standard: IFRS 17 para C20

Question

When applying the fair value transition approach, for reporting periods subsequent to the transition date, should insurance acquisition cash flows that occurred prior to the transition date be included in the contractual service margin (CSM), and so be recognised as revenue and expense in the statement of profit or loss after transition?

Answer

No. This question was discussed by the IASB Transition Resource Group for IFRS 17 Insurance Contracts (‘TRG’) in February 2018. It was clarified that, when applying the fair value approach on transition, paragraph C20 of IFRS 17 requires the amount of insurance acquisition cash flows included in the measurement of the CSM to be only the amount occurring after the transition date that is also included in the fulfilment cash flows. Applying the fair value approach, an entity is not permitted to include any insurance acquisition cash flows occurring prior to the date of transition in the measurement of the CSM. Therefore, insurance acquisition cash flows that occurred prior to the date of transition are not included in the measurement of the CSM at the transition date, and so they are not included in the presentation of insurance revenue and expenses after transition.

Are there some other requirements where additional comparative information is mandated?

Reference to standard: IFRS 17 para C25

Some regulatory bodies require the presentation of comparative information for more comparative periods than is required under IFRS.

The IFRS 17 transition date is the beginning of the annual reporting period immediately preceding the date of initial application. This provides some relief from full retrospective application for all periods presented for entities filing with some regulatory bodies, such as the US Securities and Exchange Commission, which require presentation of financial statements and financial information for periods greater than two years. An entity should clearly identify the information that has not been adjusted, disclose that it has been prepared on a different basis, and explain that basis. Alternatively, an entity can present adjusted comparative information applying IFRS 17 for any earlier periods presented, but it is not required to do so.