Issue 56, November 2016 Insurance IFRS Newsletter “The decisions made this month Setting the stage for a should ease some of the operational burden and costs final standard in key areas when This month, the IASB discussed its fndings from the external implementing and testing of a draft of the forthcoming insurance contracts standard applying IFRS 17.” (draft IFRS 17) and addressed some of the issues raised. – Joachim Kölschbach, Level of aggregation KPMG’s global IFRS A portfolio of contracts would be divided at least between those contracts that are insurance leader onerous on initial recognition, contracts that have no signifcant risk of becoming onerous after initial recognition and other contracts. However, entities would be prohibited from grouping contracts issued more than one year apart. Recognition of changes in estimates The Board agreed that when an experience adjustment directly causes a change in Contents the estimate of the present value of future cash fows, the combined effect would be recognised in proft or loss. Summary of feedback received 2 Derivatives used to mitigate fnancial risks Level of aggregation 3 The Board agreed that if an entity uses a derivative to mitigate fnancial risks arising from an insurance contract, subject to the variable fee approach, then the entity Recognition of changes in would be permitted to exclude the effect of those changes in the fnancial risk from estimates 7 the contractual service margin (CSM) when specifc criteria are met. Derivatives used to mitigate fnancial risk 9 Transition The Board made changes to some transition requirements and confrmed that an entity Transition 11 would apply the forthcoming insurance contracts standard (IFRS 17) retrospectively, Other sweep issues 15 unless this is impracticable. If it is impracticable, then an entity would be permitted to choose between a modifed retrospective approach and the fair value approach. Appendix: Summary of IASB’s redeliberations 19 Effective date Project milestones and The Board agreed that an entity would apply IFRS 17 for annual periods beginning timeline 32 on or after 1 January 2021, assuming that it is published in H1 2017. Entities would be able to apply it earlier if they also apply IFRS 9 Financial Instruments and IFRS 15 KPMG contacts 33 Revenue from Contracts with Customers. Keeping in touch 34 Next steps Acknowledgements 36 The staff are continuing the drafting process and expect to issue IFRS 17 in H1 2017. © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. 1 Summary of feedback received Respondents provided As part of the balloting process, which began in February 2016, the staff undertook a topic-based external testing of the draft IFRS 17. The topics included, among generally consistent others, level of aggregation, recognition of changes in estimates, derivatives used feedback on key to mitigate fnancial risk, and transition. areas such as level of The objective was to provide the Board with evidence about how entities interpret aggregation, transition specifc requirements and what operational diffculties, if any, they may encounter in applying the requirements. and variable fee approach scope. Over the past two months, the staff analysed the feedback and this month they presented that feedback to the Board. The key feedback from respondents is included throughout this newsletter in the ‘What feedback was received?’ sections, and the staff’s responses to this feedback and actions taken by the Board based on this feedback are discussed in the ‘What did the staff recommend?’ and ‘What did the IASB decide?’ sections. 2 © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. Level of aggregation A portfolio of contracts What were the proposed requirements? would be divided An entity ow uld have aggregated contracts into groups to determine hw ether at least between to recognise a loss of r a group o f onerous contracts and to measure the MSC atf er initial recognition. hT ose groups ow uld have comprised contracts that on those contracts in itial recognition have uf ture cash o� sw that the entity epx ects iw ll respond that are onerous on similarly in terms o f amount and timing to changes in ek y assumptions and similar initial recognition, epx ectedp� rot� ability 1. contracts that have no signifcant risk of becoming onerous What feedback was received? after initial recognition Generally, external reviewers said that the proposed requirements would result in a very high number of groups of contracts and low numbers of contracts in and other contracts. each group, which would require excessively granular, costly and burdensome calculations that may not necessarily equate with better quality information. Most believed that this level of granularity resulted from the ‘similar proftability’ proposed requirement. In addition, they believed that how to interpret this term was unclear. Most external reviewers noted that the application of these proposed requirements would generally be different from the way they currently manage their business to assess proftability and track contracts’ performance. Further, the number of groups would be much higher than the groups that they currently use for management reporting, and accounting for fnancial and regulatory reporting. eN arly all respondents agreed iw th the principle that losses on onerous contracts should be accounted of r in the statement o f prot� or loss. oH ew ver, most ef lt that the obej ctive o f grouping is not clearly articulated. What did the staff recommend? Grouping requirements The Board’s conclusion is about limiting when contracts that are onerous should be grouped with contracts that are proftable after inception, as follows: − groups of contracts that have a greater risk of being onerous should not be grouped with those that have a lower risk of being onerous; and − the CSM should be allocated to periods in a way that refects the service provided under the contracts. The staff believed that some aspects of the Board’s intention could be met even if entities were permitted to aggregate contracts into bigger groups in some cases, and such a permission would have a signifcant beneft in terms of operational costs for preparers. They also noted that it is generally desirable to align accounting requirements with information that is consistent with internal reporting and risk management. 1. See ‘Topic 1 – Aggregation of contracts’ in the IAs’BS August 2016 testing uq estionnaire for the exact requirements. © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. 3 Accordingly, the staff proposed that operational relief could be provided to entities by: − being clear that determining whether contracts ‘respond similarly to changes in key assumptions’ is expected to be assessed at the level of monitoring and management of key assumptions by management (i.e. contracts within each product line would be expected to have similar risks); and − using a top-down approach, focused on separating groups of contracts rather than grouping individual contracts (e.g. taking a portfolio of insurance contracts and determining whether further disaggregation is necessary based on certain requirements). The staff proposed disaggregating portfolios into at least three groups within each portfolio: − those that are onerous at inception; − those that have no signifcant risk of becoming onerous (i.e. more resilient in their ability to retain proftability as a result of adjustments to the CSM); and − other contracts that are not onerous at inception (i.e. less resilient in their ability to retain proftability as a result of adjustments to the CSM). The staff believed that such a requirement would enable the Board to achieve its intention and reduce the operational burden that may have resulted from the proposed requirement for entities to group contracts according to ‘similar expected proftability’. The staff also proposed prohibiting entities from grouping contracts issued more than one year apart. They noted that such a requirement is operational and retains the desired outcome for the allocation of the CSM. This recommendation would also apply to mutualised contracts where the losses of one policyholder are offset by gains from another within an insurance contract portfolio (although the contracts may be in different annual cohorts). The staff noted that the effects of mutualisation should be taken into account within the fulflment cash fows. Accordingly, losses that arise in a group would not be regarded as onerous if the combined mutualised portfolio is proftable. Discount rates used for determining the interest accretion on the CSM External reviewers also expressed a concern that a group may contain contracts that accrete interest on the CSM at different rates. This would mean that contracts would need to be accreted at different rates, and would result in entities needing to track contracts with those different rates separately within a group. hT e sta f f acnk olw edged this concern and proposed that an entity may use a ew ighteda- verage discount rate of r the interest accretion on the MSC , iw th an averaging period o f up to one year. 4 © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. What did the IASB discuss? Various Board members agreed with the staff’s recommendations, noting that they ‘strike the right balance’ between operationality, cost concerns and achieving the Board’s objectives. Some who supported the recommendations expressed a preference for the previous proposals (i.e. a more principles-based approach), but acknowledged that the external reviewers had interpreted the previous proposals in a way that resulted in a signifcantly larger amount of groups than the Board intended. Another Board member said that it should be clear in the standard what the objectives are and what is included as a way to achieve those objectives. hT e sta f f noted that they iw ll add more guidance on ho w to apply IFR S 1 7 to contracts iw th mutualisation and clarie� d that entities ow uld be prohibited rf om grouping contracts issued more than one year apart. oH ew ver, they may group on a more rf euq ent basis. hT e sta f f also clarie� d that entities may measure a set o f contracts together at inception i f they can determine that those contracts can be grouped iw th others based on available inof rmation at inception. What did the IASB decide? The Board agreed with the staff’s recommendations, as follows. − Retain the defnition of ‘portfolio’ – i.e. a portfolio is a group of contracts subject to similar risks and managed together as a single pool. IFRS 17 would provide guidance that contracts within each product line (e.g. annuities or whole-life contracts) would be expected to have similar risks, and therefore contracts from different product lines would not be expected to be in the same portfolio. − Entities would be required to identify onerous contracts at inception and group them separately from contracts that are not onerous at inception. In addition, they could measure contracts together if they can determine that those contracts can be grouped with others based on available information at inception. − Entities would be required to measure insurance contracts that are not onerous at inception by dividing portfolios into two groups – a group of contracts that have no signifcant risk of becoming onerous and a group of other proftable contracts. IFRS 17 would provide guidance that: - an entity would assess the risk of the contracts in the group becoming onerous in a manner consistent with the entity’s internal reporting about changes in estimates; - an entity would assess the risk of the contracts in the group becoming onerous based on the sensitivity of the fulflment cash fows to changes in estimates which, if they occurred, would result in the contracts becoming onerous; and - an entity would be permitted to divide portfolios further. For example, if the entity’s internal reporting provides information that distinguishes the different risks of contracts becoming onerous. − Entities would be prohibited from grouping contracts issued more than one year apart. − Entities would be permitted to use a weighted-average discount rate for the accretion of interest on the CSM, with an averaging period of up to one year. © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. 5 − The IASB also confrmed its previous proposal that entities would allocate the CSM for a group of contracts on the basis of the passage of time. Therefore, the CSM would be allocated over the current and expected remaining coverage period and the allocation would be based on coverage units, refecting the expected duration and size of the contracts in the group. KPMG insight The revised proposals are expected to result in fewer groups – i.e. units of account – than was previously expected, and therefore may alleviate a signifcant amount of the burden and reduce associated implementation costs. However, entities would still have to identify groups of insurance contracts at a lower level than the portfolio of insurance contracts. Entities would be able to take a ‘top-down’ approach to their level of aggregation assessment – i.e. may be able to begin at the portfolio level and determine whether further disaggregation is necessary. For portfolios of insurance contracts that do not include onerous contracts at inception, and: − where all of the contracts have no signifcant risk of becoming onerous; or − where all of the contracts have a signifcant risk of becoming onerous, the level of aggregation may be consistent with the portfolio level – but only for contracts issued within the same year. This may be more consistent with the level at which the business is managed and currently accounted for than the IASB’s previous proposals. The proposed requirements do not seem to alleviate the potential inconsistency that may result between the IFRS 17 reported results and those that are used by management to manage the business for portfolios where losses at inception on some contracts are offset by gains on others such that the overall portfolio is proftable. 6 © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. Recognition of changes in estimates The combined What were the proposed requirements? effect of experience An entity w ould have regarded epx erience aduj stments as relating to cur rent or adjustments that past services to be recognised in pro� t or loss, and changes in estimates o f uf t ure cash � osw as relating to uf t ure services to be recognised as an aduj stment to the directly cause a change MSC . oH w ever, circumstances in hw ic h the f ormer w ould not apply w ould have in the estimate of the included the ef f ect o f e vents that result in an epx erience aduj stment that causes a present value of future change in the estimate o f uf t ure cash � osw . T he combined ef f ect w ould have been regarded as relating to uf t ure service, giving rise to an aduj stment to the� MSC . 2 cash fows would be recognised in proft or loss. What feedback was received? Many external reviewers had diffculty in interpreting and applying the proposed requirements that deal with the combined effects of experience adjustments and associated changes in estimates of future cash fows. They believed that it was not clear how to distinguish when a change in estimate was caused by an experience adjustment and when it was not. oS me respondents e px ressed that it w ould be operationally less complicated i f the combined ef f ect w ere accounted f or in the statement o f pro� t or loss and not in the MSC . oH w ever, others noted that this may result in accounting volatility in the statement o f pro� t or loss. What did the staff recommend? The staff acknowledged that entities may fnd it diffcult to understand what types of changes in estimates should be considered because past and current experience is sometimes used to determine changes in future assumptions. For example, entities may periodically review their recent experience coupled with that from the past (e.g. mortality rates) via experience studies. These studies are used to determine prospective changes to future assumptions when determining the estimates of future cash fows. The staff did not intend for such a change to be regarded as being caused by experience adjustments that had occurred in the current or past periods. The staff believed that it was possible to make a distinction between: − changes in estimates of cash fows that should be combined with experience adjustments (e.g. an unexpected increase in lapses in the current period that directly causes a reduction in expected coverage units and the related fulflment cash fows); and − other changes in estimates of cash fows (e.g. prospective changes in future lapse assumptions as a result of an annual experience study). The staff proposed that the changes in estimates of cash fows that should be combined with experience adjustments would be those directly caused by an experience adjustment because the experience adjustment changes the future rights and obligations for the group of contracts (i.e. the number of coverage units), and not just the measurement of those rights and obligations. The staff also noted that adjusting the CSM for the combined effect of an experience adjustment and a resulting change in the estimate of the present value 2. See paragraph B93 of the draft IFRS 17, as presented ‘Topic 5 – Recognition of changes in estimates’ of the IAs’BS August 2016 testing uq estionnaire , for the exact requirements. © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. 7 of future cash fows could result in increased implementation costs. However, they believed that these costs could be reduced if this adjustment is recognised in the statement of proft or loss instead. The staff also analysed experience adjustments for contracts measured under the variable fee approach that do not affect the underlying items and that arise from non-fnancial risk. The Board had previously decided that these items would adjust the CSM. However, this would be inconsistent with the treatment under the general model for: − experience adjustments that do not directly cause changes in the estimates of the present value of future cash fows; and − the revised treatment (as proposed by the staff this month) for the combined effects of experience adjustments. hT e sta f f did not believe that there is a good reason of r a dief f rence betew en the general model and the variable ef e approach in the treatment o f these items. hT ereof re, they proposed that these items be recognised in prot� or loss. What did the IASB decide? The Board agreed with the staff’s recommendations that: − when an experience adjustment directly causes a change in the estimate of the present value of future cash fows, the combined effect of the experience adjustment and the change in the estimate of the present value of future cash fows would not adjust the CSM, but would be recognised in proft or loss; − for contracts measured under the variable fee approach, experience adjustments arising from non-fnancial risk that do not affect the underlying items, and any directly caused changes in the estimates of the present value of future cash fows, would not adjust the CSM but would be recognised in proft or loss; and − an experience adjustment directly causes a change in the estimate of the present value of future cash fows only when it causes a change in the future rights and obligations for the group of contracts (i.e. the number of coverage units). A change in the measurement only of existing rights and obligations would not be directly caused by an experience adjustment. KPMG insight Usually, insurance contracts are impacted by experience adjustments in each reporting period. If the combined effect of an experience adjustment and a resulting change in the estimate of the present value of future cash fows were to adjust the CSM, then there would be additional measurement complexity and operational considerations that entities would need to consider when calculating the CSM. The Board’s decision this month is expected to reduce the number of adjustments that would be made against the CSM and should make the subsequent measurement of the CSM less complex. 8 © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. Derivatives used to mitigate financial risk The Board decided What were the proposed requirements? to extend a previous T he B oard had previously agreed that i f an entit y uses a derivative measured at decision to more f air value through pro� t or loss F( PTV L ) to mitigate the n� ancial mark et ris k rf om a n� ancial option or a guarantee embedded in an insurance contract subej ct to the closely align the variable f ee approach, then the entity w ould be permitted to recognise in pro� t general model and or loss the changes in the value o f the embedded n� ancial option or guarantee variable fee approach. determined using uf ll� ment cash � osw , pro vided that certain criteria w ere met.3 What feedback was received? Most external reviewers for whom these requirements were relevant supported the proposals, as they would reduce accounting mismatches. However, the reviewers believed that these proposed requirements should also be available for contracts that are accounted for using the general model. Some external reviewers believed that the scope of the proposed requirements should be extended to include other changes in fnancial risk that are mitigated through hedging with derivatives (e.g. changes in the entity’s share in the underlying items), as well as changes to non-fnancial risks (e.g. mortality or longevity) when they are hedged. Some also stated that applying these proposed requirements on a prospective basis could lead to signifcant accounting mismatches on transition and, as a result, misstate shareholders’ equity at the date of transition and future profts after transition. nO e etx ernal revieew r stated that there ow uld be a signic� ant accounting mismatch due to dief f rences betew en the discount rate used to value the insurance contract cash o� sw and that used in the valuation o f the derivatives. What did the staff recommend? The staff acknowledged that under the variable fee approach, an entity’s use of a derivative measured at FVTPL (which is not considered an underlying item of the contract) to mitigate: − the fnancial market risks from the entity’s share in the underlying items; or − a fnancial option or guarantee embedded in an insurance contract, could result in similar accounting mismatches. The IASB’s previous decision to reduce these accounting mismatches was only permitted for fnancial options or guarantees embedded in insurance contracts. Given the similarities, the staff proposed extending this permission to include other changes in fnancial risk – namely, those arising from an entity’s share of underlying assets. 3. See paragraph B104 of the draft IFRS 17, as presented ‘Topic 3 – Derivatives used to mitigate fnancial market risk’ in the IAs’BS August 2016 testing uq estionnaire , for the exact requirement. © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved. 9 hT e sta f f also considered the other ef edbac k provided by etx ernal revieew rs but did not propose any other changes to the oB ards’ previous decision, primarily because the obej ctive o f this reuq irement is to align the general measurement model and variable ef e approach hw ere necessary. It is not the stas’f f intention to address general concerns related to accounting of r ris k mitigation activities. Further, some o f the ef edbac k had already been considered by the oB ard. What did the IASB decide? hT e oB ard agreed iw th the stas’f f recommendation that i f an entity uses a derivative to mitigate n� ancial rissk arising rf om an insurance contract, subej ct to the variable ef e approach, then the entity ow uld be permitted to ecx lude the eef f ct o f those changes in the n� ancial ris k rf om the MSC hw en specic� criteria are met. hT is etx ends the oB ards’ previous decision to all n� ancial rissk ree� cted in the insurance contract to hw ich the variable ef e approach is applied. 10 © 2016 KPMG IFRG Limited, a UK company, limited by guarantee. All rights reserved.