PwC Comments on FASB Discussion Paper: Preliminary Views on Insurance Contracts

PwC comment letter (FASB) 11/30/2010 by Assurance services

PwC believes the FASB should continue to work with the IASB in evaluating and making changes responsive to comments received, resolving current differences between the Boards, and developing a global insurance standard. PwC expresses concerns about certain aspects of both the IASB and FASB proposals. One key concern is how the proposals would interact with other standards that are relevant to insurers' financial statements. PwC believes the FASB and IASB need to resolve their remaining differences on this project, address the concerns regarding the proposed models raised in PwC's letter to the IASB, and finalize one converged financial instruments standard, in order for a proposed insurance contracts accounting standard to be sufficiently beneficial to warrant its adoption in the United States.

Comment letter

November 30, 2010

Technical Director
Financial Accounting Standards Board
401 Merritt 7
PO Box 5166
Norwalk, CT 06856-5116

File Reference No. 1870-100

PwC appreciates the opportunity to comment on the FASB Discussion Paper, Preliminary Views on Insurance Contracts (the "Discussion Paper"). We support the FASB's objective of working with the IASB to develop converged standards, including a standard on accounting for insurance contracts. We believe that convergence to a single set of high quality accounting standards will benefit the global capital markets and the financial statement users who participate in those markets.

We believe the FASB should continue to work with the IASB in evaluating and making changes responsive to comments received, resolving current differences between the Boards, and developing a global insurance standard. As described in more detail in the attached PwC letter to the IASB, we have concerns about certain aspects of both the IASB and FASB proposals. One key concern is how the proposals would interact with other standards that are relevant to insurers' financial statements. For example, for many insurers, standards dealing with the accounting for financial instruments are as important as standards on accounting for insurance contracts. In our view, to produce relevant, decision useful information, the interaction of other standards with standards on accounting for insurance contracts must be considered. We encourage the Boards to do so. In addition, due to the pervasiveness of the proposed changes, we recommend the Boards work closely with the insurance industry to comprehensively test the proposals using real data before finalizing the standard. This would help to ensure that the model will produce information that is relevant to the decision making needs of users and is cost effective for preparers.

US standards on accounting for insurance contracts have been in use for several decades and are well understood by preparers and users. From our discussions with certain members of each group, they seem comfortable that the existing models produce relevant information for users. We generally concur with that view. Thus, we believe the FASB and IASB need to resolve their remaining differences on this project, address the concerns regarding the proposed models raised in PwC's letter to the IASB, and finalize one converged financial instruments standard, in order for a proposed insurance contracts accounting standard to be sufficiently beneficial to warrant its adoption in the United States.

We have attached to this letter the cover letter from the PwC network of firms (PwC network) to the IASB in response to its Exposure Draft - Insurance Contracts followed by our responses to the specific questions raised in the Discussion Paper. The final attachment is the PwC network's responses to the specific questions raised in the IASB Exposure Draft, for your reference.

If you have any questions, please contact Paul Kepple at 973-236-5293 or Donald Doran at 973-236-5280.

Sincerely,

PricewaterhouseCoopers LLP

Attachment A - PwC Network of Firms Cover Letter to the IASB Exposure Draft Response

Sir David Tweedie
Chairman
International Accounting Standards Board
30 Cannon Street
London
EC4M 6XH

30 November 2010

Dear Sir

Exposure Draft: Insurance Contracts

We are pleased to respond to your Exposure Draft – Insurance Contracts. Following consultation with members of the PwC network of firms, this response summarises the views of the member firms that commented on this Exposure Draft. “PwC” refers to a network of member firms of PricewaterhouseCoopers International Limited, each of which is a separate and independent legal entity.

We recognise the efforts the IASB has made towards developing a comprehensive model of accounting for insurance contracts. The development of a comprehensive standard for insurance contracts is essential because the transitional arrangements established in IFRS 4 do not provide the level of transparency and comparability necessary for the users of financial statements. Furthermore, the current accounting for insurance contracts lacks a consistent measurement approach which users of financial statements demand.

We believe the Board’s proposal is a significant improvement that addresses these concerns. Overall, we support the proposed use of a measurement model for all insurance contracts that portrays a current assessment of the amount, timing and uncertainty of the future cash flows that the insurer expects its existing insurance contracts to generate. We welcome the move away from the current exit value notion. We support the fulfilment objective proposed in the exposure draft which reflects the economics of the insurer’s business and uses management’s estimates of cash flows based on an entity’s own strategy and efficiency. We agree with the use of market observable variables when they are used to determine the expected future cash flows arising from the insurance contract. We believe that if our concerns noted below are addressed the proposed model will provide a reliable source of data and useful information for users of the financial statements.

Detailed measurement approach

Whilst we support the Board’s proposal overall, we do have some concerns with respect to certain aspects of the proposed measurement model which are described further below. We support the use of a current value measurement that assumes performance according to the terms of the contract for measuring insurance liabilities. We believe that users of financial statements would generally prefer to look to a current value measurement to obtain information about the present value of future cash flows.

Discount rate

We agree the discount rate used by the insurer for non-participating contracts should reflect the characteristics of the insurance contract liability. We also agree that the discount rate should not reflect the own credit risk of the insurer. Furthermore, the exclusion from profit and loss of the changes in the measurement of a liability due to changes in an entity’s own credit risk in circumstances where they are not expected to be realised by transfer to a third party is consistent with the conclusions reached by the IASB in the recent project on financial liabilities. However, while we support the use of the risk free rate with an adjustment for illiquidity for the subsequent measurement, we note that in some instances the exclusion of own credit risk from the discount rate may result in the recognition of a day one accounting loss. This could occur, for example, where the insurer priced the insurance contract with an expected rate of return to the policyholder that is greater than the risk free discount rate. We recommend that the Board assess the prevalence and significance of such apparently uneconomic outcomes in conjunction with its field testing to determine whether an exception to the overall model is warranted or whether some additional disclosure should be given in such circumstances.

We also believe that insurers should be allowed an option to apply a locked-in discount rate as established at the inception of the contract. Under IFRS 9, ‘Financial Instruments’ certain assets supporting the insurance contracts would be measured at amortised cost. The use of a current discount rate could introduce an accounting mismatch in the income statement if the insurer in applying IFRS 9 uses amortised cost for some assets backing the insurance liability. An option to use a locked in discount rate at the inception of the contract would be consistent with the fulfilment objective for measuring insurance contracts. An insurer typically has the intention to fulfil its legal obligation to pay contractual amounts in the event of a claim or the maturity of the contract. This option, applied at portfolio level, should be available if it eliminates or reduces an accounting mismatch and is consistent with the insurer’s business model of fulfilling the contracts.

Risk adjustment

We recommend that the Board work closely with the insurance industry to understand the practical implications and operationality of both the explicit risk adjustment and the alternative composite margin model before finalising the proposed standard. We also believe the Board should work with users to ensure that the recognition of an explicit risk adjustment will provide sufficient decision-useful information to make this approach cost beneficial to adopt. A post implementation review of the proposed standard would also enable the Board to assess whether there is comparability between risk adjustments calculated by different companies on different bases.

If the risk adjustment is ultimately included in the measurement model, we do not believe it is appropriate in a principle-based standard to limit the risk adjustment calculation to three prescribed methods. Whilst we believe the guidance provided for the three risk adjustment techniques will be useful for the preparers of the financial statements, limiting the range of permitted risk adjustment techniques would not allow for the use of new improved risk adjustment techniques that may be developed over time. However, we would suggest requiring the use of a consistent methodology for calculating the risk adjustment for similar contracts within the group financial statements.

Residual margin

We also have concerns with the immediate recognition of all changes in estimates in the statement of comprehensive income whilst the residual margin is not remeasured. An approach that could be taken would be to recalibrate the residual margin at each reporting period using updated cash flow estimates with all current information (including the actual cash flow experience to date) discounted at the original discount rate. The residual margin would be recalibrated (remeasured) as at inception of the contract and amortised over the coverage period using updated information and assumptions of future cash flows. This would allow the remaining residual margin and its release to represent the current expected profit. If a contract becomes onerous subsequent to issuance, we believe recognising a loss in the current period performance statement followed by the amortisation of profits from the release of the locked in residual margin in subsequent periods does not provide useful information. It would also prevent overly conservative initial assumptions of cash flows to be reduced subsequent to initial recognition and taken immediately to income. This proposal applies the principle applied at initial recognition on a consistent basis throughout the life of the contract at each reporting period. However, we acknowledge that the subsequent recalibration of the residual margin may be operationally challenging and therefore the Board should work closely with industry to ensure such an approach can be practically applied.

Short-duration contracts

For short-duration contracts, we believe the modified measurement approach should be permitted but not mandated. Some insurers may find it operationally easier to apply the full model to all of their contracts rather than to have two different accounting methodologies, and we would not want to preclude them from doing so if they choose. We also do not believe the criteria for the use of the simplified measurement should be limited to a coverage period of approximately 12 months or less. We would permit its use for longer duration contracts if it is expected to be a reasonable approximation for the measurement under the full building blocks model. For example, this could be the case for a 24 month contract where the distribution of expected cash flows of future claims is not expected to change because new information will not become available or the distribution rarely changes. Although the converse could also be possible where expected cash flows of future claims may change during a 12 month contract, for example contracts that provide hurricane coverage, we would suppor. 12 month contracts being permitted to use the modified measurement approach as a practical expedient.

Unbundling

We believe it is appropriate to unbundle some components that are included in an insurance contract. However, the proposals as drafted are unclear on what is meant by “closely related” and the interplay between the principle and the examples in analysing product features. For example, it is unclear whether an account balance which does not meet any of the specified unbundling criteria but is considered not to be closely related to the host insurance contract, is required to be unbundled. It is also unclear how to apply the guidance to (a) some universal life contracts which have an account balance with an explicit crediting rate but where the rate is not contractually required to pass all investment performance to the policyholder or (b) a mortgage loan which waives some or all contractual cash flows on death or disability.

We would prefer that components be required to be unbundled if the economic characteristics and risks of the component are not significantly interdependent with those of the host insurance contract. This recognises that an insurance contract is difficult or sometimes impossible to analyse and separate into its individual parts without specific allocation rules and therefore many components should not be unbundled. In addition, this would eliminate the potential "bundling" of two non-interdependent contracts with the same customer if they were closely related activities, for example, where claims administration services and stop loss insurance coverage are combined in a single contract. However, we would support the optional unbundling of interdependent account balances in insurance contracts to enable, for example, loans that are waived on death to be unbundled into the underlying loan and the insurance element.

Transition

We support the Board’s proposal that the proposed standard should be applied retrospectively. However, we do not support the Board’s proposals to not recognise any residual margin for contracts in existence on transition to the new standard. Such an approach would distort an insurer’s reported income for years into the future. Therefore, we would support full retrospective application of the proposed model in line with the requirements in IAS 8 to determine the residual margins at transition date as if the new accounting policy had always been applied. This treatment will allow future earnings to properly reflect profit emergence from all contracts (pre- and post-transition) on a consistent basis.

However, the Board should not underestimate the complexity of retrospective application. If it is impracticable to determine the cumulative effect by applying the new accounting policy to all prior periods at the transition date, we believe the insurer should apply the Board’s proposals to measure the insurance contracts at the present value of the fulfilment cash flows. We believe that using previous measurement policies to determine the future profits would distort an insurer's reported profit for years into the future and we do not support such a transition method.

Timetable and field testing

The proposed standard will bring about pervasive changes to the way insurers measure insurance contracts, for example unbundling components of an insurance contract, the use of risk free discount rates with an illiquidity adjustment, the calculation of an explicit risk adjustment and amortisation of the residual or composite margin, as well as changes to some investment contracts with discretionary participating features. The current field testing being undertaken by the Board has a very tight timeline and this could impede the ability of the participants to fully test the proposals on a wide range of products. The European insurance industry has already had experience with such testing in the context of the Quantitative Impact Studies carried out to support the development of the Solvency II regulations. This testing demonstrated that field testing can enhance the understanding of the proposals and identify problems of interpretation and implementation of the measurement and disclosure requirements. In addition, it has also highlighted the amount of time that is required by the industry to implement extensive changes to systems and processes that are likely to be required. We recommend that the Board work closely with the insurance industry to comprehensively test the proposals with real data before finalising the proposed standard to ensure the finalised model will produce information that is relevant to the decision-making needs of users and on balance cost beneficial to produce. The Board should also take the results of the field testing into consideration when setting the effective date for the proposed standard.

We have expanded on the above and responded to the specific questions raised in your Exposure Draft in the appendix to this letter.

If you have any questions in relation to the letter please do not hesitate to contact John Hitchins, PwC Global Accountant (+44 207 8042497) or Gail Tucker (+44 117 9234230).

Yours faithfully

PricewaterhouseCoopers LLP

Attachment B - PwC Response to FASB Discussion Paper Questions

Definition and Scope

1. Are the proposed definitions of insurance contract and insurance risk (including the related guidance) understandable and operational?

We agree with the definition of an insurance contract and insurance risk as amended by the scope exceptions discussed below. The current definition in IFRS 4 is consistent with the proposed definition and has not led to significant divergence in practice for existing IFRS preparers. The current detailed guidance on the definition of an insurance contract in IG Example 1 of the implementation guidance to IFRS 4 should be included in the new standard as this will be useful, especially for US entities that have not already applied IFRS 4.

2. If the scope of the proposed guidance on insurance contracts is based on the definition of an insurance contract rather than on the type of entity issuing the contract, would financial reporting be improved?

We agree that accounting that is activity based rather than entity based is preferable. Such an approach enables transactions with similar economics to be reflected similarly regardless of the nature of the entity involved. However, the proposed definition of insurance contracts appears overly broad and, as a result, we have concerns about the type of activities that might be inadvertently subject to the insurance guidance. See our comments on question 3 below.

3. Do you agree with the proposed scope exclusions? Why or why not?

We agree with the scope exclusions in paragraphs 28(a) – (d), (f) and (g). We also agree with the intention of the Board in paragraph 28 (e) that fixed-fee service contracts issued by entities with non-insurance business should not be in the scope of the proposed insurance contracts standard but should be covered by the revenue recognition standard. However, we believe that the wording of the scope exemption should be clarified. The purpose of the last sentence in 28(e) is unclear, as an insurer is defined as anyone issuing an insurance contract (which would include fixed service contractors). Interpreting the words “primary purpose” will involve subjective judgment and may be difficult to apply consistently. It is also unclear how maintenance contracts differ from extended warranty insurance and whether both types of contracts would be excluded from the scope of the proposed standard. Similarly, there are healthcare provider organizations that may accept a fixed fee from insurance companies to service all members of an insurance policy.

We disagree that contracts currently defined as financial guarantee contracts should be brought into the scope of the proposed insurance standard. Conceptually, we believe the credit risk arising from a failure to pay is a type of financial risk and these contracts would be more appropriately addressed by the financial instruments standard. Therefore, the definition of financial guarantee contracts in ASC 815 should be retained and these contracts should be scoped out of the insurance standard and addressed using an expected loss accounting model as part of the financial assets impairment project. However, we recognize that this solution will be a concern for companies writing credit insurance and suggest that the Board consult further to identify whether these products can be distinguished from those more commonly issued by banks or other entities.

4. Should benefits that an employer provides to its employees that otherwise meet the definition of an insurance contract be within the scope of the proposed guidance? Why or why not?

We do not think that compensation to employees, including benefits that would otherwise meet the definition of an insurance contract, should be accounted for as an insurance contact. Compensation to employees should be considered under a separate model regardless of the form of that compensation. We also believe that the calculation of the composite/residual margin would be overly complicated as it would be based on the fair value of the compensation received.

5. The Board’s preliminary view is that participating investment contracts should not be accounted for within the proposed model for insurance contracts but, rather, should be included in the scope of the proposed model for accounting for financial instruments. Do you agree? Why or why not?

We do not believe that investment contracts with discretionary participating features should be in the scope of the proposed insurance contract standard. These contracts do not contain significant insurance risk and should therefore be treated as financial instruments. However, the Board should note that these contracts comprise a significant proportion of life insurers’ business in some countries. While we believe these should be accounted for as financial instruments, there is significant complexity in applying financial instruments accounting to these contracts. In particular, there are difficulties in applying existing literature to these hybrid contracts to determine the debt/equity components of the contracts (especially for policies with recurring monthly premiums) because of the multiple embedded options/guarantees, as well as the existence of obligations to pay benefits to a group of policyholders although in some cases no contractual requirement exists with respect to a specific policyholder.

We recommend the Board consider the treatment of investment contracts with discretionary participation features within the financial instruments with characteristics of equity project. However, we note the Board’s decision not to issue an exposure draft on financial instruments with characteristics of equity in the near future due to the current work load of the Board. Until the financial instruments with characteristics of equity project addresses these contracts, as an interim measure, we recommend that investment contracts with discretionary participation features be included within the scope of the proposed insurance contracts standard.

6. Do you support the approach for determining when noninsurance components of contracts should be unbundled? Why or why not?

We believe it is appropriate to unbundle some components that are included in an insurance contract. However, the proposals as drafted are unclear on what is meant by “closely related” and on the interplay between the principle and the examples in analyzing product features. For example, it is unclear whether an account balance that does not meet any of the specified unbundling criteria but is considered not to be closely related to the host insurance contract is required to be unbundled. It is also unclear how to apply the guidance to (a) some universal life contracts that have an account balance with an explicit crediting rate but where the rate is not contractually required to pass all investment performance to the policyholder or (b) a mortgage loan that waives some or all contractual cash flows on death or disability.

We would prefer that components be required to be unbundled if the economic characteristics and risks of the component are not significantly interdependent with those of the host insurance contract. This recognizes that an insurance contract is difficult or sometimes impossible to analyze and separate into its individual parts without specific allocation rules and therefore many components should not be unbundled. In addition, this would eliminate the potential "bundling" of two non-interdependent contracts with the same customer if they were closely related activities, for example, where claims administration services and stop loss insurance coverage are combined in a single contract. However, we would support the optional unbundling of interdependent account balances in insurance contracts to enable, for example, loans that are waived on death to be unbundled into the underlying loan and the insurance element.

If the unbundling criteria in the proposed standard are retained, we would exclude the criteria in paragraph 40(a)(2) for account balances. The inclusion of those criteria will prevent a loan bearing a fixed or variable rate of interest or the account balance within a universal life contract from being unbundled.

Paragraph 40 only refers to insurance contracts and insurance coverage and should be amended to also include investment contracts with discretionary participating features if these would be in the scope of the proposed standard. Policyholders often have the choice to invest in more than one investment portfolio (portfolios with discretionary participation features and portfolios without). We see no reason why these contracts should not be subject to the same unbundling principle.

The proposed standard should provide guidance on the allocation of consideration and acquisition costs between the different contract components that are unbundled. This guidance could be based on the principles in the revenue recognition exposure draft.

We understand that the unbundling guidance for embedded derivatives in insurance contracts under IFRS is different from the guidance in US GAAP. The Board should converge to the IASB guidance in this area. However, as it relates to that guidance, BC225 of the IASB exposure draft should clarify why paragraph 8 of IFRS 4 on the bifurcation of surrender options is no longer needed in the proposed standard. If reliance will be placed on AG 33(h), it is unclear why the paragraph was needed in IFRS 4. However, we agree that the surrender options described in paragraph 8 of IFRS 4 should not be required to be unbundled.

Paragraph 11 of the IASB exposure draft states that the term “insurance contract” refers to the components that remain after unbundling components of an insurance contract. The proposed standard should clarify that the classification of a contract as an insurance contract occurs before applying the unbundling requirements.

The IG Example 2 in IFRS 4 illustrating the treatment of embedded derivatives contained in insurance contracts and investment contracts has been useful implementation guidance and should be retained in the proposed insurance standard.

Recognition and Measurement

7. Do you agree with the use of the probability-weighted estimate of net cash flows to measure insurance contracts? Does that approach faithfully represent the economics of insurance contracts? Is it an improvement over existing U.S. GAAP?

Insurers do not typically transfer insurance contracts in the normal course of business and therefore a fulfillment objective better reflects the substance of the contract compared to the exit price approach based on hypothetical market participants. We agree that the measurement of an insurance contract should include the expected present value of the future cash flows that will arise as the insurer fulfills the insurance contract. We support the proposed measurement model as it captures the nature of an insurance contract, which is a bundle of rights and obligations. Estimates of entity specific cash flows for non-market observable variables, such as policy administration or maintenance expenses and mortality, are relevant and faithfully reflect the economics of the contract.

We agree with the use of probability weighted estimates of future cash flows. However, the inclusion of the term “mean” in the body of the proposed standard would clarify the measurement basis. This is currently only implied by paragraph B76 of the IASB exposure draft, which discusses risk adjustment techniques. Explicitly referring to the estimate of mean cash flows in the proposed standard will ensure that the expected cash flows are not confused with the median, mode or a range of cash flows.

8. Do you think that an entity’s estimate of the net cash flows should include a risk adjustment margin?

We conceptually support the inclusion of an explicit risk adjustment in the measurement of an insurance contract. However, we have concerns regarding whether the proposed model will produce comparable results among insurers. Furthermore, we are also concerned with whether the proposed amortization methods for both the residual margin and the composite margin will recognize the income on a contract in an appropriate manner. For example, in a long duration contract, using expected claims as a driver of amortization would defer profit recognition towards the end of the contract term since the probability of death or maturity increases in the later years. In a risk adjustment model, we believe that amortization of the residual margin based on provision of coverage is more appropriate and is consistent with the concept that the uncertainty/probability of cash flows is covered by the risk adjustment. Similarly, the use of claims paid as an amortization driver of the composite margin for contracts where claims are structured to be paid out over many years also seems to result in too much deferral of the margin when the earnings process is essentially complete upon agreement of the structured pay out pattern.

As noted in our cover letter, we recommend that the Board work closely with the insurance industry to understand the practical implications and operationality of both the explicit risk adjustment and the composite margin models before finalizing the proposed standard. We also believe the Board should work with users to ensure that the recognition of an explicit risk adjustment will provide sufficient decision-useful information to make this approach cost beneficial to adopt. A post implementation review of the proposed standard would also enable the Board to assess whether there is comparability between risk adjustments calculated by different companies on different bases.

We also have concerns over the amortization pattern for the residual margin. The pattern for the release of the residual margin will significantly impact the earnings pattern of some insurers. The measurement of the insurance contract liability does not include elements such as non-incremental acquisition costs, general overheads, and profit margins included in pricing the contract. To the extent these items are not included in the fulfillment cash outflows but are priced for in determining the premiums for insurance contracts, they will be recognized as part of the residual margin. We do not agree with the principle in the IASB exposure draft that the residual margin should be recognized in a manner that reflects the probability of incurring a claim, as this is incorporated in the explicit risk adjustment. Without the benefit of field testing, we are unable to propose any specific amortization pattern but would support the residual margin being recognized in the income statement over the coverage period in a pattern that is reflective of the time value of money. Allowing insurers to select an amortization driver would result in a lack of comparability and be arbitrary due to the variety of elements included within the residual margin. It will be important for the Board to consider the results of the field testing and work closely with the insurance industry to make informed decisions on the appropriate amortization of the residual margin.

Additionally, we are concerned about the immediate recognition of all changes in estimates in the statement of comprehensive income while the residual margin is not remeasured. We would support recalibrating the residual margin at each reporting period to update previous cash flow estimates for all current information (including the actual cash flow experience to date) at the original discount rate. The residual margin would be recalibrated (remeasured) as at inception of the contract and amortized over the coverage period using updated information and assumptions of future cash flows. This would allow the remaining residual margin and its release to represent the current expected profit. It would also prevent overly conservative initial assumptions of cash flows to be reduced subsequent to initial recognition and taken immediately to income. This proposal applies the principle used at initial recognition on a consistent basis throughout the life of the contract at each reporting period. However, we acknowledge that the subsequent recalibration of the residual margin may be operationally challenging and therefore the Board should work closely with industry to assess whether such an approach can be implemented in practice.

In the IASB exposure draft, it is unclear how the proposed amortization methods should be applied, in particular to the timing of benefits. “Benefits” should be clearly defined in the proposed standard. If maturity benefits are considered part of policy benefits it could heavily weight profit recognition towards the end of the contract term, which would not be reflective of the exposure from providing insurance coverage. The IASB exposure draft is also unclear whether the recognition pattern should be locked in at inception or whether subsequent changes in expected timing of insurance claims will impact the amortization pattern.

9. Is the objective of the risk adjustment margin understandable? If so, do you think that the techniques for estimating the risk adjustment margin (see paragraph 52(b)), faithfully represent the maximum amount that the fulfillment cash flows exceed those expected?

We agree that the risk adjustment should depict the maximum amount the insurer would rationally pay to be relieved of the risk that the ultimate fulfillment cash flows exceed those expected. We considered other potential objectives but concluded that the use of the word “maximum” was necessary to identify where in the range of management’s possible assessments the measurement of the risk adjustment should be. We also felt this objective would reduce the instances when gains or losses on reinsurance transactions would occur, since this level of measurement essentially equates to the amount the insurer would be willing to pay in a reinsurance transaction.

The proposed standard should provide the principle for the risk adjustment. It is inconsistent for a principles-based standard to limit the number of permitted risk adjustment techniques to the three included in the Discussion Paper. The proposed standard should allow for the use of any risk adjustment technique that complies with the objective and principle established for the risk adjustment technique. This will allow insurers to use more appropriate and reliable methods to calculate the risk adjustment that may be developed over time. However, we believe the proposed standard should clarify that within group financial statements a consistent methodology should be used to calculate the risk adjustment for all similar contracts.

We agree that the risk adjustment should be determined at the portfolio level. Given the wide range of insurance contracts offered by many insurers, the application of the definition of a portfolio can prove problematic and be subject to a diverse interpretation. Experience in many situations indicates that insurers frequently include non-homogeneous risks in portfolios of contracts managed together in order to benefit from diversification. We therefore recommend amending the definition of a portfolio to exclude the word “broadly” to encourage a consistent approach to defining a portfolio. Measurement reflecting diversification between all of an entity's portfolios would in essence result in measurement of a value of the business rather than a value of the contracts themselves.

10. Do you think that the risk adjustment margin would be comparable for entities that are exposed to similar risks?

As discussed in our response to question 8 above, we conceptually support the inclusion of an explicit risk adjustment in the measurement of an insurance contract. However, we have concerns regarding whether the proposed model will produce comparable results among insurers.

11. Do you agree with the description of cash flows that should be included in the measurement of an insurance contract? Is the proposed guidance operational?

We believe that for a principles-based accounting framework, the proposed guidance on estimates of future cash flows appropriately addresses the fulfillment objective of the model. Expanding in any more detail the level of cash flow guidance would result in setting detailed rules rather than providing guidance on the implementation of the measurement principle.

However, the cash flow guidance regarding “future policyholders” in paragraph B61(j) of the IASB exposure draft is unclear. We understand that referring to future policyholders for participating contracts where some of the investment returns earned in one period will be paid to future policyholders in the next period will eliminate an accounting mismatch between different reporting periods. However, referring to future policyholders contradicts the principle of the contract boundary of the existing contracts covered by the proposed standard. There is also currently divergence in the treatment of accumulated estates relating to participating contracts. Without further guidance as to the use of the term “future policyholders,” this diversity may continue and may result in the elimination of retained earnings o. mutual insurers.

12. Do you agree that the carrying amount of all insurance contracts should be discounted if the effect is material? Do you agree with the proposed guidance on the discount rate that should be used to measure the carrying amount of insurance contracts? If not, which discount rate should be used?

We agree with the objective for the discount rate to reflect the characteristics of the insurance liability. We believe that the management decision of which assets to purchase to back an insurance liability should not influence the measurement of the insurance contract liability unless the amount, timing, and uncertainty of the cash flows arising from the insurance contract depend on the performance of specific assets.

However, we would support an option to apply a locked-in discount rate determined at the inception of the contract. This option should only be available at the level of a portfolio if it eliminates or significantly reduces an accounting mismatch and when it is consistent with the business model of the insurer. This option will allow insurers that hold bonds and other debt securities as part of an asset liability matching strategy to measure these assets at amortized cost instead of electing the fair value option in an attempt to mitigate the mismatch. Under IFRS 9, Financial Instruments, and consistent with our comments to the FASB on financial instruments measurement, we believe these assets would typically be held to collect contractual cash flows and should be measured at amortized cost. It is unclear why insurers should be required to measure these assets at fair value with changes in fair value recognized in net income using the fair value option when their business model is to hold these assets to collect the contractual cash flows. Allowing a locked-in discount rate for insurance contracts would eliminate an accounting mismatch that would arise due to interest rate changes arising in the measurement of the insurance liabilities and the assets backing these liabilities if our recommendation to FASB on permitting amortized cost for certain financial instruments measurement were adopted.

Paragraph 32 of the IASB exposure draft has created some confusion, with some readers suggesting that this paragraph permits the use of an asset-backed discount rate for the measurement of the entire contract when only some of the contract’s cash flows depend on the performance of specific assets. The final standard should be clear that embedded investment options and guarantees, and indeed all cash flows that are not dependent on the performance of specific assets, should not be measured using the rate based on the assets held to back the liability. Therefore, an insurance contract may be measured using different discount rates for the different elements of the contract as set out in the recently published IASB staff paper on the discount rate for participating contracts.

We agree that the discount rate should be adjusted to reflect the liquidity characteristics of the insurance contract. In practice, significant judgment will be required to determine the extent of, and when to apply, the illiquidity adjustment to the discount rate. For long-duration contracts, the effect of the illiquidity adjustment can have a significant impact on the measurement and, therefore, it may be appropriate to provide guidance on when an adjustment should be made to ensure the consistent treatment by issuers of insurance contracts. For example, the guidance could address scenarios such as whole-life contracts where the contract provides the policyholder with a liquid cash surrender value option but also an illiquid death benefit and clarify whether in such a situation an illiquidity adjustment should be made for death benefit cash flows but not for surrender cash flows.

We support the use of a current value measurement that assumes performance according to the terms of the contract for measuring insurance liabilities. We believe that users of financial statements would generally prefer to look to a current value measurement to obtain information about the present value of future cash flows. We agree that the discount rate used by the insurer for non-participating contracts should reflect the characteristics of the insurance contract liability. We also agree that the discount rate should not reflect changes in the insurer's own credit risk. The exclusion from profit and loss of the changes in the measurement of a liability due to changes in an entity’s own credit risk in circumstances where they are not expected to be realized by transfer of the obligation to a third party is consistent with the conclusions reached by the IASB in the recent project on financial liabilities. However, while we support the use of the risk free rate with an adjustment for illiquidity for the ongoing measurement, we note that in some instances the exclusion of the entity's own credit risk from the discount rate may result in the recognition of a day one accounting loss. This could occur, for example, where the insurer priced the insurance contract with an expected rate of return to the policyholder that is greater than the risk free discount rate. We recommend that the Board assess the prevalence and significance of such apparently uneconomic outcomes in conjunction with its field testing to determine whether an exception to the overall model is warranted or whether some additional disclosure should be given in such circumstances. An avenue of exploration could include deferring any day one loss resulting solely from the difference in the risk free rate and the rate of return to the policyholder, and amortizing it over the life of the contract.

Relative to the discount rate, we would observe that the Boards have proposed new guidance involving measurement of fulfillment obligations for insurance, leasing, and revenue recognition, each of which requires different discount rates and measurement approaches. In addition, there is a current debate in practice as to the discount rate to be used for provisions. We find these proposals difficult to bridge in a conceptually consistent manner and suggest the Boards reconcile or conform, if at all possible.

13. Do you think that acquisition costs should be included as one of the cash flows relating to the contract? If not, how would you account for acquisition costs?

We agree that incremental acquisition costs should be included as part of the initial measurement of the insurance contract. The principle of incremental costs at the contract level is currently used in other standards and has been reliably implemented. In addition, we believe that incremental acquisition costs should reduce the pre-claims liability under the modified approach as noted in our response to question 19.

14. Do you agree that acquisition costs included in the cash flows used in the measurement of the insurance contract should be limited to those that are incremental at the individual contract level? If not, which acquisition costs, if any, would you include in the measurement of the insurance contract?

Yes, please see our response to question 13.

15. Do you agree with the use of either the composite margin approach or two-margin approach to measure the net insurance contract? Does either approach faithfully represent the economics of insurance contracts? Is either approach an improvement over the measurement used in current U.S. GAAP?

See our response to questions 8 and 9. We conceptually support the inclusion of an explicit risk adjustment in the measurement of an insurance contract. However, we have concerns regarding whether the proposed model will produce comparable results among insurers. Furthermore, we are concerned with whether the proposed amortization methods for both the residual margin and the composite margin in the FASB model will recognize the income on a contract in an appropriate manner.

We believe the FASB and IASB need to resolve their remaining differences on this project, address the concerns regarding the proposed models raised in PwC's letter to the IASB, and finalize one converged financial instruments standard, in order for a proposed insurance contracts accounting standard to be sufficiently beneficial to warrant its adoption in the United States.

16. Do you think that the composite margin should be recognized in earnings in subsequent periods using the ratio described in paragraph 83? If not, how would you recognize the composite margin in earnings?

We have concerns with the amortization of the composite margin similar to those noted above for the residual margin. For example, there is an issue as to how premium should be allocated for purposes of the amortization formula and whether claim and benefit payments are the appropriate drivers in the formula. These could lead to the inappropriate recognition in later periods of the composite margin, especially for life contracts or structured payout settlements of casualty claims. The Board should work with the industry to apply the proposals to real data. However, if the Board were to adopt a composite margin approach, we would recalibrate the composite margin at each reporting period in the same way as the residual margin above, subject to ensuring that such an approach can practically be applied.

See our response to question 19 as it relates to the need for a margin in the post-claims liability under the modified measurement approach.

17. Do you agree that interest should not be accreted on the composite margin? Why or why not?

We agree that interest should not be accreted on the composite margin. We do not believe it is useful to gross up the composite margin with interest by recognizing an interest charge that will be released to profit again in a future period. This has the potential in early periods to increase the composite margin, which we do not believe is useful. We also would not accrete interest on the residual margin.

18. Do you think that all insurance contracts should be recognized and measured using one approach or that some insurance contracts should be recognized and measured using an alternative approach (for example, the modified approach)? Why or why not?

We consider the modified measurement approach for the pre-claims liability of short duration contracts to be a proxy for the full measurement model. We therefore suggest that the Board permit but not require the use of the simplified measurement approach for the pre-claims liabilities of short-duration contracts. This would allow insurers that issue long- and short-duration contracts to measure the results from all their insurance contracts in a consistent manner. Insurers are required to apply the building blocks model in any event as part of the onerous contract test for the modified measurement approach and therefore they should be given the opportunity to use the building blocks approach as their measurement model. Since the modified approach is simply a proxy for the full measurement model, we do not see this as introducing an option of a different measurement model but simply an easier measurement approach for contracts where the expected cash flows and risk adjustment are unlikely to change significantly over the coverage period.

19. If an alternate approach is required for some insurance contracts, what recognition, measurement, and presentation provisions should be applied (including those items noted in paragraph 106)?

We support the principle that the use of the simplified measurement approach should be limited to a portfolio for which the results are not expected to be materially different from the building blocks approach, as further described in our response to question 21 below.

Measurement. Under the risk adjustment approach, as noted in the IASB exposure draft, there should be a risk adjustment for the claims liability. Under the composite approach, we disagree with paragraph 102 of the Discussion Paper, which implies that under the modified approach, the claims liability is measured as the probability-weighted estimate of the net cash flows with no consideration of an additional margin. We recommend that under the modified approach, an equivalent composite margin be determined for the post-claims period and released in a similar manner as it would be in the building block approach. We believe that such an approach would be consistent with the FASB's composite margin approach. We do not believe that the claims liability should be different for an insurance liability estimated using the building blocks approach as opposed to the modified approach; an 18 month policy should not have a different claims liability measurement than a 12 month policy.

The guidance is unclear whether a claim that changes the remaining risk coverage on a contract could allow the remaining pre-claims liability to be amended or released. We believe that if a claim changes the remaining coverage on the contract, the insurer should be allowed to amend or release the corresponding element of the pre-claims liability.

Acquisition costs. As noted in our response to question 13, we believe that incremental acquisition costs should reduce the pre-claims obligation under the modified approach, consistent with the IASB exposure draft proposal for application of the modified approach. However, the IASB exposure draft defines premium taxes as direct contract costs rather than as incremental acquisition costs. While this distinction is not important in the building blocks approach (because both types of costs are included in contract cash flows), it will impact the treatment of such costs in the IASB's modified approach, where only those costs defined as "incremental acquisition costs" will effectively be capitalized. We believe that since the modified approach is a proxy for the building blocks model, premium taxes should be treated consistently. They should therefore be treated as acquisition costs and under the modified approach recognized as a deduction from the pre-claims liability.

Interest accretion. We do not believe that interest should be accreted on the carrying amount of the pre-claims liability. The purpose of the modified approach is to permit a practical expedient, and simplicity should be part of that objective.

Onerous contract test. If the Board decides on an explicit risk adjustment approach, the onerous contract test would follow the guidance in paragraph 60 of the IASB exposure draft. If the composite margin approach were used, the risk adjustment should be excluded. Aside from the exclusion of the explicit risk adjustment in the calculation, the measurement, recognition, and grouping for this calculation would be the same as described in paragraph 60 of the IASB exposure draft.

Presentation. We believe that contracts subject to the modified measurement approach should be presented in the income statement as described in paragraph 75 of the IASB exposure draft and should be presented in the balance sheet as described in paragraphs 57 and 58 of the IASB exposure draft.

20. Do both the building-block approach and the modified approach (with the latter approach applied only to certain short-duration contracts) produce relevant and decision-useful information? Why or why not?

We believe that the proposed measurement model with a fulfillment objective will produce relevant information for users but we propose some changes as noted in our responses throughout this letter.

21. How should the scope of insurance products for each approach be defined (for example, duration of coverage period, duration of claims payment period, or type of insurance)?

We do not believe the use of the simplified measurement approach should be restricted to a coverage period of approximately 12 months or less as described in the IASB exposure draft. We support the principle that the use of the simplified measurement approach should also be permitted where an insurer can demonstrate for a coverage period greater than 12 months that applying the modified measurement approach to a portfolio should result in a similar amount that is not expected to be materially different from the building blocks approach. For example, this could be the case for a 24 month contract where the distribution of expected cash flows of future claims is not expected to change because new information will not become available or the distribution rarely changes. Although the converse could also be possible where expected cash flows of future claims may change during a 12 month contract, for example contracts that provide hurricane coverage, we would support 12 month contracts being permitted to use the modified measurement approach as a practical expedient.

Short duration contracts contain surrender options that can significantly affect the variability of the cash flows under the contract. One of the criteria to be met in order to qualify for the use of the modified measurement approach in the IASB exposure draft is that the contract should not contain any embedded option that significantly affects the variability of cash flows. We recommend that the proposed standard clarify that these surrender options do not prevent the use of the simplified measurement approach.

22. Are there specific types of insurance contracts for which the approaches would not provide decision-useful information?

See our response to question 3.

23. What are the implications of the recent U.S. healthcare reform to the application of the proposed contract boundary principle, including whether health insurance contracts written under the new reforms would meet the conditions in the proposed guidance to be accounted for under the modified approach?

Under the recent US healthcare reform, an insurer may not have the legal right at contract inception or subsequently to underwrite a prospective policyholder based on his/her individual health risks, or to charge a varying premium based on those health risks. Instead, because of government regulation, the insurer will be required to offer coverage to all who request it, at a government approved rate. In such cases, since renewal policyholders are charged the same rate as new policyholders with the same risk characteristics, the insurer is effectively receiving a market premium rate from renewal policyholders rather than a rate that is price constrained.

We recommend the Board clarify how the boundary of the contract principle should be applied when an insurer's ability to underwrite individual policyholders and/or to set premium rates is subject to external regulatory constraints. We suggest that the proposed standard be clarified to note that "A price fully reflects the risk, and therefore the conditions in paragraph 46(b) have been met, if an insurer can exit the existing contract at either an individual or portfolio level, and if a new policyholder and a renewal policyholder with the same risk characteristics would receive the same price. In that case, the insurer is effectively receiving a market premium rate from the renewal policyholder, rather than a rate that is price constrained.”

24. What other changes should be considered to both improve and simplify U.S. GAAP for short- and long-duration insurance contracts?

If a converged standard is adopted we have the following additional comments not addressed in other questions.

Contract modifications. Unlike current US GAAP (ASC 944 on internal replacements of insurance contracts), no guidance is provided on when a modification to an insurance contract should be treated as a modification versus an extinguishment of the existing contract and the issuance of a new contract. Guidance should be provided for when a change to a contract should be treated as a modification with changes captured in the current accounting period and when changes should be recognized as an extinguishment of one contract (with derecognition) and the issuance of a new contract, as this could impact earnings recognition patterns through the elimination of the composite (or residual) margin upon termination of a contract.

Initial recognition. We agree that no gain should be recognized at the initial recognition of an insurance contract. This is consistent with the proposals/principles in the revenue recognition exposure draft as well as the application guidance in IAS 39 for financial instruments that are valued using significant non-market observable inputs. In our survey of insurance analysts, three-quarters of participants were opposed to the recognition of profit at the inception of the contract.

We agree that a day one loss should be recognized immediately in the income statement if a contract is onerous. The proposal in the IASB exposure draft is consistent with existing requirements for recognition of a provision for an onerous contract. We acknowledge the fact that a day one accounting loss, although not an economic loss, could arise due to differences between inputs used to measure a contract and the inputs used to price the contract, for example the discount rate or the extent of diversification benefits allowed for. There are many elements for which the pricing of the contract will differ from the measurement. Although we do not believe that it is appropriate to adjust the initial measurement to avoid a day one loss for an onerous contract, we recommend that the Board assess the prevalence and significance of apparently uneconomic outcomes in conjunction with its field testing to determine whether an exception to the overall model is warranted or whether some additional disclosure should be given in such circumstances.

We agree that at initial recognition, an insurer should estimate the residual margin (or composite margin) at a level that aggregates insurance contracts into a portfolio of insurance contracts and, within a portfolio, by similar date of inception of the contract and by similar coverage period

Unit-linked contracts. We agree with the IASB proposals in paragraph 71 of the IASB exposure draft to present assets and liabilities arising from unit-linked contracts as single line items and would extend this requirement to unit-linked investment contracts measured under IAS 39. However, the proposals in the exposure draft would not require account balances that have been unbundled from unit-linked insurance contracts to be shown as a single line item as they would not be within the scope of the proposed insurance contracts standard. We believe the requirement for unit linked assets and liabilities to be shown separately should be specifically prescribed in financial instruments guidance to capture all unit-linked contracts (including unbundled account balances), not just those meeting the insurance contracts definition.

We agree with the exception to traditional accounting for treasury shares and owner occupied property described in Appendix C of the IASB exposure draft to the extent they relate to the interest of unit-linked contract holders. The exception should also apply to unit-linked investment contracts, not just to unit-linked insurance contracts issued by an insurer. However, we would not extend this to employee compensation schemes or situations other than unit-linked contracts. Consistent with the proposal for treasury shares, we would also apply this principle to debt instruments. We would limit the recognition of treasury shares and holdings of an insurer’s own debt instruments to those traded in an active market and backing policyholder liabilities. These treasury shares should include shares of the parent held by a subsidiary in the group financial statements of the parent.

Transition. We do not agree with the transition requirements proposed in the IASB exposure draft. We support a full retrospective application of the full building blocks model with the ASC 250-10-45-9 impracticability exception. If full retrospective application is impracticable, we agree with excluding any residual margin on contracts in existence on transition to the new standard as proposed in the IASB exposure draft.

We are concerned that the Board’s proposal would result in the elimination of all future earnings, except for the release of the risk adjustment, from the insurance business in force at the transition date. For long-term contracts this could be a significant amount of future profit. This treatment will result in disproportional earnings from existing contracts compared to new business written after the implementation of the new standard and such an approach would distort an insurer’s reported income for years into the future. We understand that the disproportionate earnings between existing contracts and new contracts was one of the reasons why the Board did not support recognizing the difference between the existing measurement basis and the fulfillment cash flows as the residual margin on transition.

We believe insurers should be allowed to reclassify to amortized cost financial assets designated at fair value with changes in fair value recognized in net income if account balances are unbundled from insurance contracts. This would also be necessary if an insurer is able to apply a locked-in discount rate as proposed by us.

The transition arrangements should provide guidance for past business combinations where the present value of fulfillment cash flows exceeds the fair value of the insurance contract liability. We recommend that the positive difference between the fair value and the fulfillment value on past acquisitions be treated as an adjustment to goodwill.

The requirement to present claims development tables for at least 5 years at transition is problematic due to the requirement to reconcile the claims in the development table to the carrying amounts of the insurance liability recognized in the statement of financial position. This will not be possible for the period before adopting the new standard and the transition provisions should provide relief by requiring this information only from the transition date forward.

If the Board were to adopt the composite margin approach, we believe it would not be cost beneficial to require a risk adjustment to be calculated for transition purposes only but not for subsequent measurement. Consistent with our proposal for transition, we would support the full retrospective application of the building blocks model with the ASC 250-10-45-9 impracticability exception. If impracticable, we agree with the proposed composite margin transition model.

We believe it is very important that the effective dates of this proposed standard and any new financial instruments standard are aligned. Both standards will result in a significant change to the recognized amounts in the financial statements and to have an insurer make significant restatements of comparative information in successive years would not be helpful to the users of the accounts and would make comparison of results difficult. Simultaneous implementation of both standards will also eliminate the need for any exceptions for insurers to the transition provisions and eliminate the need to change previous fair value option elections on adoption of any new financial instruments standard.

We support the plan to solicit input on effective date for all Memorandum of Understanding projects. We believe that a coordinated consideration of the most appropriate transition for all of the new accounting standards, including insurance, will be well received by the Board’s constituents.

Insurers will need sufficient time to implement the proposals in the final standard. Insurers will have to change the data they capture to be able to comply with the proposed measurement model. The requirement for insurers to develop systems to model expected future cash flow scenarios should not be underestimated. Insurers should be allowed sufficient time to ensure that their measurement models will provide reliable information.

25. What are the incremental costs of adopting the alternatives described in this Discussion Paper? Please separately describe one-time costs and ongoing costs.

Due to the pervasiveness of the proposed changes, we recommend the Boards work closely with the insurance industry to comprehensively test the proposals using real data before finalizing the standard. This would help to ensure that the model will produce information that is relevant to the decision making needs of users and is cost effective for preparers.

Reinsurance

26. The scope of the proposed guidance includes reinsurance contracts that an insurer issues or acquires. However, insurance contracts held directly by other policyholders would be excluded from the scope of the proposed guidance. Do you agree with this exclusion? Why or why not?

As noted in our response to question 3, we agree with the scope exclusion in paragraph 28g. of the Discussion Paper.

27. Should there be symmetry between the recognition and measurement of reinsurance contracts and the underlying contract ceded?

We agree with the use of an expected loss model for reinsurance assets as this is consistent with the insurance contract expected cash flow model.

We do not support the recognition of an initial gain by a cedant on entering into a reinsurance contract. Consistent with the proposals for underlying insurance contracts and other standards, we would defer the recognition of any day one gain except when the gain offsets a loss recognized at inception of an onerous contract. We are concerned that this could encourage the structuring of reinsurance transactions to allow cedants to recognize immediate gains on entering into a reinsurance contract.

Paragraph 110 of the Discussion Paper requires the use of the building blocks model to measure reinsurance contracts. This would seem to preclude the use of the modified measurement model for reinsurance contracts. The use of the modified model should be permitted when the cedant applies this model to the reinsured insurance contracts.

It is unclear whether a “risk attaching” reinsurance contract that reinsures annual contracts that begin during a 12 month period would meet the criteria to be measured using the simplified measurement for pre-claims liabilities. It is also unclear whether the coverage period for this contract is 12 or 24 months. We believe it would be appropriate to apply the modified measurement model to these contracts, as mentioned on the IASB reinsurance webcast.

It is not clear whether the reinsurance guidance is meant to capture reinsurance contracts where both insurance contracts and investment contracts (which are accounted for as financial instruments) are reinsured through a single reinsurance contract. It is unclear how the building blocks approach should be applied to the reinsured investment contract element when the direct liability will either be measured at amortized cost or fair value under financial instrument accounting rather than using fulfillment cash flows.

Paragraph B28 of the IASB exposure draft (part of the guidance on significant insurance risk) indicates that contracts that are entered into simultaneously with a single counterparty or contracts that are otherwise interdependent form a single contract. There is currently diversity in practice when applying these requirements to fronting arrangements within a consolidated group. This is particularly the case where an operating entity within a consolidated group transfers risk through insurance to an independent insurer and this insurer passes the risk back to a captive insurer in the same consolidated group as the operating entity. The Board should provide further guidance on how interdependence should be interpreted in this situation.

Presentation and Disclosure

28. The margin presentation approach highlights the changes in the insurance liability, rather than the current approach in U.S. GAAP, which presents, among other items, premium revenues, benefits paid, operating costs, and changes in loss estimates. Would this change improve your understanding of the performance of an entity that provides insurance (for some types of insurance or for all)? Please explain.

The summarized margin approach will be useful to users as it is reflective of the measurement model. However, further consideration should be given to the presentation for entities that issue insurance contracts while undertaking significant non-insurance activities. Under the summarized margin model, insurers will not be presenting those elements that are included in the fulfillment cash flows (premiums, incremental acquisition costs or expenses), which will hinder comparability with other unrelated operations.

The proposed standard should also indicate how reinsurance balances should be presented in the statement of comprehensive income to ensure that a consistent approach is adopted between companies.

29. Should insurance contracts measured under the building-block approach be presented using a margin presentation approach or a premium presentation approach that would require a true-up amount as described in paragraph 119 (for example, the written allocation presentation approach or the allocated premium presentation approach)?

As noted in our response to question 28, we support the proposed summarized margin approach. However, volume information on the level of business of the insurer is typically included in the income statement under current presentation models for both life and non-life insurers. We understand that users find volume information useful and the Board should consider requiring disclosure of appropriate volume information that depicts the level of activity of the insurer. Specifying the disclosure of volume measures, such as, for example, annualized premium equivalent or gross written premium, will reduce the need for disclosure of additional information by companies on an ad hoc basis which could lead to inconsistencies among insurers.

30. Should short- and long-duration (or nonlife and life) contracts be presented in a similar manner even if such contracts are measured under different approaches?

No, as noted in our responses to questions 28 and 29, we believe that the summarized margin approach should be used for contracts following the building blocks approach and that the expanded premiums and claims presentation be permitted for contracts following the modified measurement approach.

For short-duration contracts where the underwriting margin is expanded in the statement of comprehensive income, the proposed standard should state whether differences in the initial estimate of claims incurred are presented in the claims incurred line within the underwriting margin or in the experience adjustment line. Given the potential significance of the changes in the initial estimate of claims incurred, we believe a consistent presentation is important.

31. Do you agree with the proposed disclosures in the IASB’s Exposure Draft? Why or why not? If not, what would you recommend and why?

We agree with the disclosure principle. The measurement model is based on future expected cash flows and therefore the disclosure should provide information about these future cash flows. The objective of the disclosure refers to amount, timing, and uncertainty of future cash flows. However, the liquidity risk disclosure focuses only on insurance liabilities and net cash outflows. Cash flow information on insurance contract balances that are assets would also be useful.

We agree with the proposed disclosure requirements subject to the changes outlined below. We are not supportive of the Board’s proposed disclosure requirements for measurement uncertainty as required by paragraph 90(d) of the IASB exposure draft. In estimating the future cash flows in measuring the insurance contracts, alternative probable outcomes are already included in the probability weighted estimated cash flows. While relevant and useful information is key to understanding the financial statements, we believe the proposed change in the disclosure will be difficult and costly to implement, and we have serious reservations as to whether it will provide meaningful information in addition to that which is already provided on the inputs to the building blocks measurements.

We support the requirements in paragraph 90(a) of the IASB exposure draft to require appropriate information about the method and inputs used to develop the measurements. We would however explicitly refer to these amounts separately for the estimated cash flows, the risk adjustment, and the residual margin. We believe the insurer should provide quantitative information about these inputs and paragraph 90(a) should exclude the word “practicable.” There should also be a requirement to disclose any change in the methods used to estimate the measurements.

We do not support the disclosure of confidence level information as a “comparable benchmark” as this could be misleading when the pattern of claims is not a normal distribution. We are also not convinced that the benefit of producing this information would exceed the cost of producing it. If the CTE or cost of capital method is used, a significant amount of work will be required to convert the risk adjustment as determined by these methods to a corresponding confidence level. In addition, providing this information at a portfolio level will be too voluminous, whereas at an aggregated level the disclosure would lack any meaningful information for comparison with other insurers due to the combination of different portfolios with different distributions and would possibly introduce the need to recognize the diversification benefit between portfolios. However, appropriate disclosure about the risk adjustment technique and the objective and inputs into the technique will be important. Insurers should disclose information about the risk adjustment to help users of financial statements understand and make their own judgments about the maximum amount the insurer would be willing to pay to be relieved of the risk. The risk adjustment technique applied should be described with an explanation of why the entity’s risk adjustment meets the objective for a risk adjustment.

Paragraph 92(e)(i) of the IASB exposure draft should not refer to “equity,” as all changes in the insurance liability will be recognized in the statement of comprehensive income under the proposals.

Where elements related to discretionary participating contracts are included in equity, these should be separately disclosed. Paragraph 86(d) of the IASB exposure draft on the reconciliation of reinsurance assets should also refer to reinsurance liabilities.

Additional Question for Respondents

32. After considering your views on the specific issues contained in this Discussion Paper and the IASB's Exposure Draft, what do you think would represent the most appropriate improvement to U.S. GAAP?

a. Pursue an approach based on the IASB’s Exposure Draft?

b. Pursue an approach based on the IASB’s Exposure Draft with some changes? Please explain those changes.

c. Pursue an approach based on the Board’s preliminary views in this Discussion Paper?

d. Pursue an approach based on the Board’s preliminary views in this Discussion Paper with some changes? Please explain those changes.

e. Make targeted changes to address specific concerns about current U.S. GAAP (for example, items included in paragraph 7). Please describe those changes.

We believe the FASB should continue to work with the IASB in evaluating and making changes responsive to comments received, resolving current differences between the Boards, and developing a global insurance standard. As described in more detail in this letter and the attached PwC letter to the IASB, we have concerns about certain aspects of both the IASB and FASB proposals. One key concern is how the proposals would interact with other standards that are relevant to insurers' financial statements. For example, for many insurers, standards dealing with the accounting for financial instruments are as important as standards on accounting for insurance contracts. In our view, to produce relevant, decision useful information, the interaction of other standards with standards on accounting for insurance contracts must be considered. We encourage the Boards to do so. In addition, due to the pervasiveness of the proposed changes, we recommend the Boards work closely with the insurance industry to comprehensively test the proposals using real data before finalizing the standard. This would help to ensure that the model will produce information that is relevant to the decision making needs of users and is cost effective for preparers.

We are also concerned that the areas identified as potential targeted improvements to US GAAP would not individually significantly improve current US GAAP, could not individually be implemented without potentially reconsidering the entire model, and would not move the US any closer to a converged global standard. We believe that it would be more efficient and effective to work with the IASB as it field tests and redeliberates with a longer term goal of convergence for all standards.

Attachment C - PwC Network of Firms Response to IASB Exposure Draft Questions

Question 1

Do you think that the proposed measurement model will produce relevant information that will help users of an insurer’s financial statements to make economic decisions? Why or why not? If not, what changes do you recommend and why?

As noted in our covering letter, the proposed measurement model will be an improvement for the insurance industry which currently uses a diverse range of accounting policies to present the results from issuing insurance contracts. We believe that the proposed measurement model with a fulfilment objective will produce relevant information for users but we propose some changes as noted below.

In particular, we would include an option of a locked-in discount rate at initial recognition for the insurance liability given the ability to carry some assets supporting insurance contracts at amortised cost under IFRS 9, ‘Financial Instruments’. We also encourage the Board to assess during its field testing the prevalence and significance of day one accounting losses, due solely to the use of risk free discount rates, to determine whether an exception to the overall model is warranted. Furthermore, we would also permit rather than mandate the use of the simplified measurement model for the pre-claims liability of short-duration contracts and introduce different criteria for when to apply this model. Refer to our more detailed comments in questions 3 and 8 below.

Question 2

a)Do you agree that the measurement of an insurance contract should include the expected present value of the future cash outflows less future cash inflows that will arise as the insurer fulfils the insurance contract? Why or why not? If not, what do you recommend and why?

b)Is the draft application guidance in Appendix B on estimates of future cash flows at the right level of detail? Do you have any comments on the guidance?

(a. Insurers do not typically transfer insurance contracts in the normal course of business and therefore a fulfilment objective better reflects the substance of the contract compared to the exit price approach based on hypothetical market participants. We agree that the measurement of an insurance contract should include the expected present value of the future cash flows that will arise as the insurer fulfils the insurance contract. We support the proposed measurement model as it captures the nature of an insurance contract which is a bundle of rights and obligations. Estimates of entity specific cash flows for non-market observable variables, such as policy administration or maintenance expenses and mortality, are relevant and faithfully reflect the economics of the contract.

We agree with the use of probability weighted estimates of future cash flows. However, the inclusion of the term “mean” in the body of the proposed standard would clarify the measurement basis. This is currently only implied by paragraph B76 on risk adjustment techniques. Explicitly referring to the estimate of mean cash flows in the proposed standard will ensure that the expected cash flows are not confused with the median, mode or a range of cash flows.

(b. We believe that for a principle based standard, the proposed guidance on estimates of future cash flows appropriately addresses the fulfilment objective of the model. Expanding in any more detail the level of cash flow guidance would result in setting detailed rules rather than providing guidance on the implementation of the measurement principle.

However, the cash flow guidance is unclear regarding “future policyholders” in paragraph B61(j). We understand that referring to future policyholders for participating contracts where some of the investment returns earned in one period will be paid to future policyholders in the next period will eliminate an accounting mismatch between different reporting periods. However, referring to future policyholders contradicts the principle of the contract boundary of the existing contracts contained within the proposed standard. There is also currently divergence in the treatment of accumulated estates relating to participating contracts. Without further guidance as to the use of the term “future policyholders” this diversity may continue and may result in the elimination of retained earnings in mutual insurers.

Question 3

a)Do you agree that the discount rate used by the insurer for non-participating contracts should reflect the characteristics of the insurance contract liability and not those of the assets backing that liability? Why or why not?

b)Do you agree with the proposal to consider the effect of liquidity, and with the guidance on liquidity (see paragraphs 30(a), 31 and 34)? Why or why not?

c)Some have expressed concerns that the proposed discount rate may misrepresent the economic substance of some long-duration insurance contracts. Are those concerns valid? Why or why not? If they are valid, what approach do you suggest and why? For example, should the Board reconsider its conclusion that the present value of the fulfilment cash flows should not reflect the risk of non-performance by the insurer?

a. We agree with the objective for the discount rate to reflect the characteristics of the insurance liability. We believe that the management decision of which assets to purchase to back an insurance liability should not influence the measurement of the insurance contract liability unless the amount, timing and uncertainty of the cash flows arising from the insurance contract depends on the performance of specific assets.

However, as noted in our covering letter we would support an option to apply a locked-in discount rate determined at the inception of the contract. This option should only be available at the level of a portfolio if it eliminates or significantly reduces an accounting mismatch and when it is consistent with the business model of the insurer. This option will allow insurers that hold bonds and other debt securities as part of an asset liability matching strategy, to measure these assets at amortised cost instead of electing the fair value option in an attempt to mitigate the mismatch. Under IFRS 9, ‘Financial Instruments’, these assets would typically be held to collect contractual cash flows and would be measured at amortised cost. It is unclear why insurers should be required to measure these assets at fair value though profit and loss using the fair value option when their business model is to hold these assets to collect the contractual cash flows. Allowing a locked in discount rate for insurance contracts would therefore eliminate an accounting mismatch that would arise due to interest rate changes arising in the measurement of the insurance liabilities and the assets backing these liabilities.

Paragraph 32 has created some confusion with some readers suggesting that this paragraph permits the use of an asset backed discount rate for the measurement of the entire contract when only some of the contract’s cash flows depend on the performance of specific assets. The final standard should be clear that embedded investment options and guarantees, and indeed all cash flows that are not dependent on the performance of specific assets, should not be measured using the rate based on the assets held to back the liability. Therefore an insurance contract may be measured using different discount rates for the different elements of the contract as set out in the recently published IASB staff paper on the discount rate for participating contracts.

b. We agree that the discount rate should be adjusted to reflect the liquidity characteristics of the insurance contract. In practice, significant judgement will be required to determine the extent of, and when to apply, the illiquidity adjustment to the discount rate. For long duration contracts, the effect of the illiquidity adjustment can have a significant impact on the measurement, and therefore, it may be appropriate to provide guidance on when an adjustment should be made to ensure the consistent treatment by issuers of insurance contracts. For example, the guidance could address scenarios such as whole-life contracts where the contract provides the policyholder with a liquid cash surrender value option but also an illiquid death benefit and clarify whether in such a situation an illiquidity adjustment should be made for death benefit cash flows but not for surrender cash flows.

c. We support the use of a current value measurement that assumes performance according to the terms of the contract for measuring insurance liabilities. We believe that users of financial statements would generally prefer to look to a current value measurement to obtain information about the present value of future cash flows. We agree the discount rate used by the insurer for non-participating contracts should reflect the characteristics of the insurance contract liability. We also agree that the discount rate should not reflect the own credit risk of the insurer. Furthermore, the exclusion from profit and loss of the changes in the measurement of a liability due to changes in an entity’s own credit risk in circumstances where they are not expected to be realised by transfer to a third party is consistent with the conclusions reached by the IASB in the recent project on financial liabilities. However, while we support the use of the risk free rate with an adjustment for illiquidity for the ongoing measurement, we note that in some instances the exclusion of own credit risk from the discount rate may result in the recognition of a day one accounting loss. This could occur, for example, where the insurer priced the insurance contract with an expected rate of return to the policyholder that is greater than the risk free discount rate. We recommend that the Board assess the prevalence and significance of such apparently uneconomic outcomes in conjunction with its field testing to determine whether an exception to the overall model is warranted or whether some additional disclosure should be given in such circumstances.

Question 4

Do you support using a risk adjustment and a residual margin (as the IASB proposes), or do you prefer a single composite margin (as the FASB favours)? Please explain the reason(s) for your view.

We conceptually support the inclusion of an explicit risk adjustment in the measurement of an insurance contract. However, we have concerns regarding whether the proposed model will produce comparable results among insurers. Furthermore, we are also concerned with whether the proposed amortisation methods for both the residual margin and the composite margin in the alternative model will recognise the income on a contract in an appropriate manner. For example, in a long duration contract, using expected claims as a driver of amortisation would defer profit recognition towards the end of the contract term since the probability of death or maturity increases in the later years. We believe that amortisation based on provision of coverage i. more appropriate and is consistent with the concept that the uncertainty/probability of cash flows is covered by the risk adjustment. Similarly, the use of claims paid as an amortisation driver of the composite margin for contracts where claims are structured to be paid out over many years also seems to result in too much deferral of the margin when the earnings process is essentially complete upon agreement of the structured pay out pattern.

As noted in our cover letter, we recommend that the Board work closely with the insurance industry to understand the practical implications and operationality of both the explicit risk adjustment and the alternative composite margin model before finalising the proposed standard. We also believe the Board should reach out to users to ensure that the recognition of an explicit risk adjustment will provide sufficient decision-useful information to make this approach cost beneficial to adopt. A post implementation review of the proposed standard would also enable the Board to assess whether there is comparability between risk adjustments calculated by different companies on different bases.

Question 5

a)Do you agree that the risk adjustment should depict the maximum amount the insurer would rationally pay to be relieved of the risk that the ultimate fulfilment cash flows exceed those expected? Why or why not? If not, what alternatives do you suggest and why?

b)Paragraph B73 limits the choice of techniques for estimating risk adjustments to the confidence level, conditional tail expectation (CTE) and cost of capital techniques. Do you agree that these three techniques should be allowed, and no others? Why or why not? If not, what do you suggest and why?

c)Do you agree that if either the CTE or the cost of capital method is used, the insurer should disclose the confidence level to which the risk adjustment corresponds (see paragraph 90(b)(i))? Why or why not?

d)Do you agree that an insurer should measure the risk adjustment at a portfolio level of aggregation (i.e. a group of contracts that are subject to similar risks and managed together as a pool)? Why or why not? If not, what alternative do you recommend and why?

e)Is the application guidance in Appendix B on risk adjustments at the right level of detail? Do you have any comments on the guidance?

(a. We agree that the risk adjustment should depict the maximum amount the insurer would rationally pay to be relieved of the risk that the ultimate fulfilment cash flows exceed those expected. We considered other potential objectives but concluded that the use of the word “maximum” was necessary to identify where in the range of management’s possible assessments the measurement of the risk adjustment should be and also felt this objective would reduce the instances when gains or losses on reinsurance transactions would occur, since this level of measurement essentially equates to the amount the insurer would be willing to pay in a reinsurance transaction.

(b. The proposed standard should provide the principle for the risk adjustment. It is inconsistent for a principles based standard to limit the number of permitted risk adjustment techniques to the three included in the exposure draft. The proposed standard should allow for the use of any risk adjustment technique that complies with the objective and principle established for the risk adjustment technique. This will allow insurers to use more appropriate and reliable methods to calculate the risk adjustment that may be developed over time. However, we believe the proposed standard should clarify that within group financial statements a consistent methodology should be used to calculate the risk adjustment for all similar contracts.

(c. We do not support the disclosure of confidence level information as a “comparable benchmark” as this could be misleading when the pattern of claims is not a normal distribution. We are also not convinced that the benefit of producing this information exceeds the cost of producing the information. If the CTE or cost of capital methods is used, a significant amount of work will be required to convert the risk adjustment as determined by these methods to a corresponding confidence level. In addition, providing this information at a portfolio level will be too voluminous, whereas providing it at an aggregated level the measure would lack any meaningful information for comparison with other insurers due to the combination of different portfolios with different distributions and would possibly introduce the need to recognise the diversification benefit between portfolios.

(d. We agree the risk adjustment should be determined at the portfolio level. Given the wide range of insurance contracts offered by many insurers the application of the definition of a portfolio can prove problematic and subject to a diverse interpretation. Experience in many situations indicate that insurers frequently include non-homogeneous risks in portfolios of contracts managed together in order to benefit from diversification. We therefore recommend amending the definition of a portfolio to exclude the word “broadly” to encourage a consistent approach to defining a portfolio. Measurement reflecting diversification between all of an entity's portfolios would in essence result in measurement of a value of the business rather than a value of the contracts themselves.

(e. The risk adjustment guidance provided will be useful to insurers and is at the right level of detail for a principle based standard. However, we believe the risk adjustment guidance should not be authoritative application guidance but educational implementation guidance, updated to allow a choice of any appropriate methodology that meets the objective for the risk adjustment.

Question 6

a)Do you agree that an insurer should not recognise any gain at initial recognition of an insurance contract (such a gain arises when the expected present value of the future cash outflows plus the risk adjustment is less than the expected present value of the future cash inflows)? Why or why not?

b)Do you agree that the residual margin should not be less than zero, so that a loss at initial recognition of an insurance contract would be recognised immediately in profit or loss (such a loss arises when the expected present value of the future cash outflows plus the risk adjustment is more than the expected present value of future cash inflows)? Why or why not?

c)Do you agree that an insurer should estimate the residual or composite margin at a level that aggregates insurance contracts into a portfolio of insurance contracts and, within a portfolio, by similar date of inception of the contract and by similar coverage period? Why or why not? If not, what do you recommend and why?

d)Do you agree with the proposed method(s) of releasing the residual margin? Why or why not? If not, what do you suggest and why (see paragraphs 50 and BC125–BC129)?

e)Do you agree with the proposed method(s) of releasing the composite margin, if the Board were to adopt the approach that includes such a margin (see the Appendix to the Basis for Conclusions)? Why or why not?

f)Do you agree that interest should be accreted on the residual margin (see paragraphs 51 and BC131–BC133)? Why or why not? Would you reach the same conclusion for the composite margin? Why or why not?

(a. We agree that no gain should be recognised at the initial recognition of an insurance contract. This is consistent with the proposals/principles in the revenue recognition exposure draft as well as the application guidance in IAS 39 for financial instruments that are valued using significant non-market observable inputs. In our survey of insurance analysts, three-quarters of participants were opposed to the recognition of profit at the inception of the contract.

(b. We agree that a day one loss should be recognised immediately in the income statement if a contract is onerous. The proposal in the exposure draft is consistent with existing requirements for recognition of a provision for an onerous contract. We acknowledge the fact that a day one accounting loss, although not an economic loss, could arise due to differences between inputs used to measure a contract and the inputs used to price the contract, for example the discount rate or the extent of diversification benefits allowed for. As noted below, there are many elements for which the pricing of the contract will differ from the measurement. Although we do not believe that it is appropriate to adjust the initial measurement to avoid a day one loss for an onerous contract, we recommend that the Board assess the prevalence and significance of apparently uneconomic outcomes in conjunction with its field testing to determine whether an exception to the overall model is warranted or whether some additional disclosure should be given in such circumstances.

(c. We agree that at initial recognition, an insurer should estimate the residual margin at a level that aggregates insurance contracts into a portfolio of insurance contracts and, within a portfolio, by similar date of inception of the contract and by similar coverage period.

(d. We have concerns over the amortisation pattern for both the residual margin and the composite margin. The pattern for the release of the residual margin will significantly impact the earnings pattern of some insurers. The measurement of the insurance contract liability does not include elements such as non-incremental acquisition costs, general overheads and profit margins included in pricing the contract. To the extent these items are not included in the fulfilment cash outflows but are priced for in determining the premiums for insurance contracts, these items will be recognised as part of the residual margin. We do not agree with the principle that the residual margin should be recognised in a manner that reflects the probability of incurring a claim, as this is incorporated in the explicit risk adjustment. Without the benefit of field testing, we are unable to propose any specific amortisation pattern but would support the residual margin being recognised in the income statement over the coverage period in a pattern that is reflective of the time value of money. Allowing insurers to select an amortisation driver would result in a lack of comparability and be arbitrary due to the variety of elements included within the residual margin. It will be important for the Board to consider the results of the field testing and work closely with the insurance industry to make informed decisions on the appropriate amortisation of the residual margin.

We also have concerns with the immediate recognition of all changes in estimates in the statement of comprehensive income whilst the residual margin is not remeasured. We would support recalibrating the residual margin at each reporting period to update previous cash flow estimates for all current information (including the actual cash flow experience to date) at the original discount rate. The residual margin would be recalibrated (remeasured) as at inception of the contract and amortised over the coverage period using updated information and assumptions of future cash flows. This would allow the remaining residual margin and its release to represent the current expected profit. It would also prevent overly conservative initial assumptions of cash flows to be reduced subsequent to initial recognition and taken immediately to income. This proposal applies the principle applied at initial recognition on a consistent basis throughout the life of the contract at each reporting period. However, we acknowledge that the subsequent recalibration of the residual margin may be operationally challenging and therefore the Board should work closely with industry to assess whether such an approach can be implemented in practice.

If the proposals in the exposure draft are retained, we note that it is unclear how the proposed amortisation method should be applied, in particular to the timing of benefits. “Benefits” should be clearly defined in the proposed standard. If maturity benefits are considered part of policy benefits it could heavily weight profit recognition towards the end of the contract term which would not be reflective of the exposure from providing insurance coverage. The exposure draft is also unclear whether the recognition pattern should be locked in at inception or whether subsequent changes in expected timing of insurance claims will impact on the amortisation pattern.

(e. We have concerns with the amortisation of the composite margin similar to those noted above for the residual margin. For example, there is an issue as to how premium should be allocated for purposes of the amortisation formula and whether claim and benefit payments are the appropriate drivers in the formula. These could lead to the inappropriate recognition in later periods of the composite margin, especially for life contracts or structured payout settlements of casualty claims. The Board should work with the industry to apply the proposals to real data. However, if the Board were to adopt a composite margin approach, we would recalibrate the composite margin at each reporting period in the same way as the residual margin above, subject to ensuring that such an approach can practically be applied.

It is also unclear to us how one would apply a composite margin model in the simplified measurement model for the pre-claims and post claims liability and whether claims arising are recognised at just the estimated probability weighted cash flows. We recommend that under the simplified model, a composite margin should be determined for the post-claim period and released in a similar manner as it would be in the building block approach.

(f. We do not agree that interest should be accreted on the residual margin. As proposed above, the residual margin could be amortised in a pattern that is reflective of the time value of money. We do not believe it is useful to gross up the residual margin with interest by recognising an interest charge which will be released to profit again in a future period. This has the potential in early periods to increase the residual margin which we do not believe is useful. We would also not accrete interest on the composite margin.

Question 7

Do you agree that incremental acquisition costs for contracts issued should be included in the initial measurement of the insurance contract as contract cash outflows and that all other acquisition costs should be recognised as expenses when incurred? Why or why not? If not, what do you recommend and why?

Question 8

a)Should the Board (i) require, (ii) permit but not require, or (iii) not introduce a modified measurement approach for the pre-claims liabilities of some short-duration insurance contracts? Why or why not?

b)Do you agree with the proposed criteria for requiring that approach and with how to apply that approach? Why or why not? If not, what do you suggest and why?

(a. We consider the modified measurement approach for the pre-claims liability of short duration contracts to be a proxy for the full measurement model. We therefore suggest that the Board should permit but not require the use of the simplified measurement approach for the pre-claims liabilities of short-duration contracts. This would allow insurers that issue long and short duration contracts to measure the results from all their insurance contracts in a consistent manner. Insurers are required to apply the building blocks model in any event as part of the onerous contract test for the modified measurement approach and therefore they should be given the opportunity to use the building blocks approach as their measurement model. Since the modified approach is simply a proxy for the full measurement model, we do not see this as introducing an option of a different measurement model but simply an easier measurement approach for contracts where the expected cash flows and risk adjustment are unlikely to change significantly over the coverage period. The purpose of the modified approach is to permit a practical expedient, and simplicity should be part of that objective. We therefore also do not believe that interest should be accreted on the carrying amount of the pre-claims liability.

The proposed standard defines premium taxes as direct contract costs rather than as incremental acquisition costs. While this distinction is not important in the building blocks approach (because both types of costs are included in contract cash flows), it will impact the treatment of such costs in the modified approach, where only those costs defined as "incremental acquisition costs" will effectively be capitalised. We believe that since the modified approach is a proxy for the building blocks model, premium taxes should be treated consistently. They should therefore be treated as acquisition costs and under the modified approach recognised as a deduction from the pre-claims liability.

(b. We do not believe the criteria for the use on the simplified measurement should be restricted to a coverage period of approximately 12 months or less. We support the principle that the use of the simplified measurement approach should also be permitted where an insurer can demonstrate for a coverage period greater than 12 months that applying the modified measurement approach to a portfolio should result in a similar amount that is not expected to be materially different from the building blocks approach. For example, this could be the case for a 24 month contract where the distribution of expected cash flows of future claims is not expected to change because new information will not become available or the distribution rarely changes. Although the converse could also be possible where expected cash flows of future claims may change during a 12 month contract, for example contracts that provide hurricane coverage, we woul. support 12 month contracts being permitted to use the modified measurement approach as a practical expedient.

Short duration contracts contain surrender options that can significantly affect the variability of the cash flows under the contract. One of the criteria to be met in order to qualify for the use of the simplified measurement approach is that the contract should not contain any embedded option that significantly affect the variability of cash flows. We believe that the proposed standard should clarify that these surrender options should not prevent the use of the simplified measurement approach.

The exposure draft is unclear whether a claim that changes the remaining risk coverage on a contract could allow the remaining pre-claims liability to be amended or released. We believe that if a claim changes the remaining coverage on the contract, the insurer should be allowed to amend or release the corresponding element of the pre-claims liability.

Question 9

Do you agree with the proposed boundary principle and do you think insurers would be able to apply it consistently in practice? Why or why not? If not, what would you recommend and why?

We agree with the contract boundary principle. In most circumstances the proposed principle can be consistently applied in practice because it relies on the rights and obligations contained in the contract. It relies on the contractual right of the insurer to set the price that fully reflects the risk of a particular policyholder and this can be determined without too much judgment or consideration of management intention.

However, we believe the Board should clarify how the boundary of the contract principle should be applied when an insurer's ability to underwrite individual policyholders and/or to set premium rates is subject to external regulatory constraints. For example, for certain types of healthcare contracts, an insurer may not have the legal right at contract inception or subsequently to underwrite a prospective policyholder based on his/her individual health risks, or to charge a varying premium based on those health risks. Instead, because of government regulation, the insurer may be required to offer coverage to all who request it, at a government approved rate. In such instances, since renewal policyholders are charged the same rate as new policyholders with the same risk characteristics, the insurer is effectively receiving a market premium rate from renewal policyholders rather than a rate that is price constrained. We suggest that the proposed standard be clarified to note that: "A price fully reflects the risk, and therefore the conditions in paragraph 27(b) have been met, if an insurer can exit the existing contract at either an individual or portfolio level, and if a new policyholder and a renewal policyholder with the same risk characteristics would receive the same price. In that case, the insurer is effectively receiving a market premium rate from the renewal policyholder, rather than a rate that is price constrained.”

Question 10

a)Do you agree that the measurement of insurance contracts should include participating benefits on an expected present value basis? Why or why not? If not, what do you recommend and why?

b)Should financial instruments with discretionary participation features be within the scope of the IFRS on insurance contracts, or within the scope of the IASB’s financial instruments standards? Why?

c)Do you agree with the proposed definition of a discretionary participation feature, including the proposed new condition that the investment contracts must participate with insurance contracts in the same pool of assets, company, fund or other entity? Why or why not? If not, what do you recommend and why?

d)Paragraphs 64 and 65 modify some measurement proposals to make them suitable for financial instruments with discretionary participation features. Do you agree with those modifications? Why or why not? If not, what would you propose and why? Are any other modifications needed for these contracts?

(a. We agree that insurance contracts with discretionary participating features should include participating benefits on an expected cash flow basis consistent with other cash flow estimates in the building blocks approach. Although we support inclusion of cash flows relating to discretionary participating features to be paid to future policyholders, additional guidance is required on the treatment of orphan estates that have been accumulated by insurers and estates in mutual insurers as explained in our response to question 2 above.

(b. We do not believe that investment contracts with discretionary participating features should be in the scope of the proposed insurance contract standard. These contracts do not contain significant insurance risk and should therefore be treated as financial instruments. However, the Board should note that these contracts comprise a significant proportion of life insurers’ business in some countries. Whilst we believe these should be accounted for as financial instruments, there is significant complexity in applying financial instrument accounting to these contracts. In particular, there are difficulties in applying IAS 32 to these hybrid contracts to determine the debt/equity components of the contracts (especially for policies with recurring monthly premiums) because of the multiple embedded options/guarantees, as well as the existence of obligations to pay benefits to a group of policyholders although in some cases no contractual requirement exists for a specific policyholder.

We recommend the Board considers the treatment of investment contracts with discretionary participation features within the financial instruments with characteristics of equity project. However, we note the Board’s decision not to issue an exposure draft on financial instruments with characteristics of equity in the near future due to the current work load of the Board. Until the financial instruments with characteristics of equity project addresses these contracts, as an interim measure, we recommend that all investment contracts with discretionary participating features are included within the scope of the proposed insurance contracts standard.

(c. We do not agree with the proposed definition of a discretionary participation feature within the proposed insurance contracts standard. As noted above, we would treat all investment contracts with discretionary participation features as financial instruments outside the scope of the proposed insurance contracts standard, regardless of whether they participate in the same pool of assets or fund as insurance contracts, although as noted above as an interim measure we would include them within the proposed insurance contracts standard. However, if only some investment contracts with discretionary participation features are included in the scope of the future insurance standard, the definition should clarify that these contracts should only be within the scope of the future standard if the insurance contracts represent a significant proportion of the combined pool or company.

(d. For investment contracts with discretionary features that are included within the proposed insurance contracts standard we support the proposals for the boundary of the investment contracts with discretionary participation features and the basis for the release of the residual margin.

Question 11

a)Do you agree with the definition of an insurance contract and related guidance, including the two changes summarised in paragraph BC191? If not, why not?

b)Do you agree with the scope exclusions in paragraph 4? Why or why not? If not, what do you propose and why?

c)Do you agree that the contracts currently defined in IFRSs as financial guarantee contracts should be brought within the scope of the IFRS on insurance contracts? Why or why not?

(a. We agree with the definition of an insurance contract. The current definition in IFRS 4 is consistent with the proposed definition and has not led to significant divergence in practice for existing IFRS preparers. The current detailed guidance on the definition of an insurance contract in IG Example 1 of the implementation guidance to IFRS 4 should be retained in the new standard as this will be useful guidance, especially for transitioning territories that have not already applied IFRS 4.

(b. We agree with the scope exclusions (a) – (d), (f) and (g). We also agree with the intention of the Board in paragraph 4(e) that fixed-fee service contracts issued by entities with non-insurance business models should not be in the scope of the proposed insurance contracts standard but covered by the revenue recognition standard. However, we believe that the wording of the scope exemption should be clarified. The purpose of the last sentence in 4(e) is unclear, as an insurer is defined as anyone issuing an insurance contract (which would include fixed service contractors). Interpreting the words “primary purpose” will involve subjective judgement and may be difficult to apply consistently. It is also unclear how maintenance contracts differ from extended warranty insurance and whether both type of contracts are excluded from the scope of the proposed standard. Similarly, there are healthcare provider organisations which may accept a fixed fee from insurance companies to service all members of an insurance policy.

(c. We disagree that contracts that are currently defined as financial guarantees contracts should be brought into the scope of the proposed insurance standard. Conceptually, we believe the credit risk arising from a failure to pay is a type of financial risk and these contracts would be more appropriately addressed by the financial instruments standard. Therefore, the definition of financial guarantee contracts should be retained and these contracts should be scoped out of the proposed insurance standard and addressed using an expected loss accounting model as part of the financial assets impairment project. One advantage of including financial guarantees in the financial asset impairment project is that group companies that enter into intra-group financial guarantees will not be required to use the insurance contract measurement model. However, we recognise that this solution will be a concern for companies writing credit insurance and suggest that the Board consult further to identify whether these products can be distinguished from those more commonly issued by banks or other entities.

Question 12

Do you think it is appropriate to unbundle some components of an insurance contract? Do you agree with the proposed criteria for when this is required? Why or why not? If not, what alternative do you recommend and why?

We believe it is appropriate to unbundle some components that are included in an insurance contract. However, the proposals as drafted are unclear on what is meant by “closely related” and the interplay between the principle and the examples in analysing product features For example, it is unclear whether an account balance which does not meet any of the specified unbundling criteria but is considered not to be closely related to the host insurance contract, is required to be unbundled. It is also unclear how to apply the guidance to (a) some universal life contracts which have an account balance with an explicit crediting rate but where the rate is not contractually required to pass all investment performance to the policyholder or (b) a mortgage loan which waives some or all contractual cash flows on death or disability.

If the unbundling criteria in the proposed standard are retained, we would exclude the criteria in paragraph 8(a)(ii) for account balances. The inclusion of this paragraph will prevent a loan bearing a fixed or variable rate of interest or the account balance within a universal life contract from being unbundled.

Paragraph 8 of the exposure draft only refers to insurance contracts and insurance coverage and should be amended to also include investment contracts with discretionary participating features if these are in the scope of the proposed standard. Policyholders often have the choice to invest in more than one investment portfolio (portfolios with discretionary participation features and portfolios without). There is no reason why these contracts should not be subject to the same unbundling principle.

We are unclear in the basis for conclusions in BC225 why paragraph 8 of IFRS 4 on the bifurcation of surrender options is no longer needed to be included in the proposed standard. If reliance will be placed on AG 33(h), it is unclear why the paragraph was needed in IFRS 4. However, we agree that surrender options described in paragraph 8 of IFRS 4 should not be required to be unbundled.

Paragraph 11 states that the term “insurance contract” refers to the components that remain after unbundling components of an insurance contract. The proposed standard should clarify that the classification of a contract as an insurance contract however occurs before applying the unbundling requirements.

Question 13

a)Will the proposed summarised margin presentation be useful to users of financial statements? Why or why not? If not, what would you recommend and why?

b)Do you agree that an insurer should present all income and expense arising from insurance contracts in profit or loss? Why or why not? If not, what do you recommend and why?

(a. The summarised margin approach will be useful to users as it is reflective of the measurement model. However, further consideration should be given to the presentation for entities that issue insurance contracts whilst undertaking significant non-insurance activities. Under the summarised margin model, insurers will not be presenting those elements that are included in the fulfilment cash flows (premiums, incremental acquisition costs or expenses) which will hinder comparability with other unrelated operations.

The proposed standard should also indicate how reinsurance balances should be presented in the statement of comprehensive income to ensure that a consistent approach is adopted between companies.

As noted above we support the proposed summarised margin approach. However, volume information on the level of business of the insurer is typically included in the income statement under current presentation models for both life and non-life insurers. We understand that users find volume information useful and the Board should consider requiring disclosure of appropriate volume information that depicts the level of activity of the insurer. Specifying the disclosure of volume measures such as, for example, annualised premium equivalent or gross written premium, will reduce the need for disclosure of additional information by companies on an inconsistent basis.

(b. We agree that all income and expenses arising from insurance contracts should be presented in profit and loss as described in the exposure draft. However, we continue to believe that the Board should develop a set of consistent principles to govern the use of other comprehensive income (OCI) and we strongly believe that the Board needs to add to its post 2011 agenda, a project to address the purpose of OCI, what types of items should be recognised in OCI and to what extent recycling is appropriate.

Question 14

a)Do you agree with the proposed disclosure principle? Why or why not? If not, what would you recommend, and why?

b)Do you think the proposed disclosure requirements will meet the proposed objective? Why or why not?

c)Are there any disclosures that have not been proposed that would be useful (or some proposed that are not)? If so, please describe those disclosures and explain why they would or would not be useful.

(a. We agree with the disclosure principle. The measurement model is based on future expected cash flows and therefore the disclosure should provide information about these future cash flows. The objective of the disclosure refers to amount, timing and uncertainty of future cash flows. However, the liquidity risk disclosure focuses only on insurance liabilities and net cash outflows. Cash flow information on insurance contract balances that are assets would also be useful.

(b. We agree with the proposed disclosure requirements subject to the changes outlined below. We are not supportive of the Board’s proposed disclosure requirements for measurement uncertainty as required by paragraph 90(d). In estimating the future cash flows in measuring the insurance contracts, alternative probable outcomes are already included in the probability weighted estimated cash flows. While relevant and useful information is key to understanding the financial statements, we believe the proposed change in the disclosure will be difficult and costly to implement, and we have serious reservations as to whether it will provide meaningful additional disclosure in addition to those that are already provided on the inputs to the building blocks measurements.

(c. We support the requirements in paragraph 90 (a) to require appropriate information about the method and inputs used to develop the measurements. We would however explicitly refer to these amounts separately for the estimated cash flows, the risk adjustment and the residual margin. We believe the insurer should provide quantitative information about these inputs and paragraph 90(a) should exclude the word “practicable”. There should also be a requirement to disclose any change in the methods used to estimate the measurements.

As noted in our response to question 5 above, we do not support the disclosure of confidence level information in those scenarios where an insurer determines the risk adjustment using one of the other two risk adjustment techniques as this implies that the disclosures will be comparable which will not be the case if the pattern of claims is not a normal distribution. However, appropriate disclosure about the risk adjustment technique and the objective and inputs into the technique will be important. Insurers should disclose information about the risk adjustment to help users of financial statements understand and make their own judgements about the maximum amount the insurer would be willing to pay to be relieved of the risk. The risk adjustment technique applied should be described with an explanation of why the entity’s risk adjustment meets the objective for a risk adjustment.

Paragraph 92(e)(i) should not refer to “equity” as all changes in the insurance liability will be recognised in the statement of comprehensive income under the proposals of the exposure draft.

Where elements related to discretionary participating contracts are included in equity these should be separately disclosed. Paragraph 86(d) on the reconciliation of reinsurance assets should also refer to reinsurance liabilities.

Question 15

Do you agree with the proposals on unit-linked contracts? Why or why not? If not what do you recommend and why?

We agree with the proposals to present assets and liabilities arising from unit-linked contracts as single line items and would extend this requirement to unit-linked investment contracts measured under IAS 39. However, the proposals in the exposure draft would not require account balances that have been unbundled from unit-linked insurance contracts to be shown as a single line as they would not be within the scope of the proposed insurance contracts standard. We believe the requirement for unit linked assets and liabilities to be shown separately should be included in IAS 1 to capture all unit-linked contracts (including unbundled account balances) and not just those meeting the insurance contracts definition.

We agree with the exception in the proposed amendments to IAS32, IFRS 9 and IAS 16 for treasury shares and owner occupied property to the extent they relate to the interest of unit-linked contract holders. The exception should also apply to unit linked investment contracts, not just to unit linked insurance contracts issued by an insurer. However, we would not extend this to employee compensation schemes or situations other than unit-linked contracts. Consistent with the proposal for treasury shares, we would also apply this principle to debt instruments. We would limit the recognition of treasury shares and holdings of an insurer’s own debt instruments to those traded in an active market and backing policyholder liabilities. These treasury shares should include shares of the parent held by a subsidiary in the group financial statements of the parent.

Question 16

a)Do you support an expected loss model for reinsurance assets? Why or why not? If not, what do you recommend and why?

b)Do you have any other comments on the reinsurance proposals?

a. We agree with the use of an expected loss model for reinsurance assets as this is consistent with the insurance contract expected cash flow model.

b. We do not support the recognition of an initial gain by a cedant on entering into a reinsurance contract. Consistent with the proposals for underlying insurance contracts and other standards, we would defer the recognition of any day one gain except when the gain offsets a loss recognised at inception of an onerous contract. We are concerned that this could encourage the structuring of reinsurance transactions to allow cedants to recognise immediate gains on entering into a reinsurance contract.

Paragraph 43 requires the use of the buildings block model to measure reinsurance contracts. This would seem to preclude the use of the simplified measurement model for reinsurance contracts. The use of the simplified model should be permitted when the cedant applies this model to the reinsured insurance contracts.

It is unclear whether a reinsurer that issues a “risk attaching” reinsurance contract that reinsures annual contracts that incept during a 12 month period would meet the criteria to be measured using the simplified measurement for pre-claims liabilities. It is unclear whether the coverage period for this contract is 12 or 24 months. We believe it would be appropriate to apply the simplified measurement model to these contracts as was mentioned on the IASB reinsurance webcast.

It is not clear whether the reference to “contracts” in paragraph 44 is meant to capture those reinsurance contracts where both insurance contracts and investment contracts (which are in the scope of IFRS 9, ‘Financial Instruments’) are reinsured through a single reinsurance contract. It is unclear how the building blocks approach should be applied to the reinsured investment contract element when the direct liability will either be measured at amortised cost or fair value under IFRS 9 rather than using fulfilment cash flows.

Paragraph B28 of the exposure draft (part of the guidance on significant insurance risk) indicates that contracts that are entered into simultaneously with a single counterparty or contracts that are otherwise interdependent form a single contract. There is currently diversity in practice when applying these requirements to fronting arrangements within consolidated group scenarios. This is particularly the case where an operating entity within a consolidated group transfers risk through insurance to an independent insurer and this insurer passes the risk back to a captive insurer in the same consolidated group as the operating entity. The Board should provide further guidance on how interdependence should be interpreted in this situation.

Question 17

a)Do you agree with the proposed transition requirements? Why or why not? If not, what would you recommend and why?

b)If the Board were to adopt the composite margin approach favoured by the FASB, would you agree with the FASB’s tentative decision on transition (see the appendix to the Basis for Conclusions)?

c)Is it necessary for the effective date of the IFRS on insurance contracts to be aligned with that of IFRS 9? Why or why not?

d)Please provide an estimate of how long insurers would require to adopt the proposed requirements.

(a. We do not agree with the proposed transition requirements. We support a full retrospective application of the full building blocks model with the IAS 8 impracticability exception. If full retrospective application is impracticable, we agree with excluding any residual margin on contracts in existence on transition to the new standard as proposed in the exposure draft.

We are concerned that the Board’s proposal results in the elimination of all future earnings, except for the release of the risk adjustment, from the insurance business in force at the transition date. For long term contracts this could be a significant amount of future profit. This treatment will result in disproportional earnings from existing contracts compared to new business written after the implementation of the new standard and such an approach would distort an insurer’s reported income for years into the future. We understand that the disproportionate earnings between existing contracts and new contracts was one of the reasons why the Board did not support recognising the difference between the existing measurement basis and the fulfilment cash flows as the residual margin on transition.

We believe insurers should be allowed to reclassify financial assets designated at fair value through profit and loss to amortised cost if account balances are unbundled from insurance contracts. This would also be necessary if an insurer is able to apply a locked in discount rate as proposed by us.

The transition arrangements should provide guidance for past business combinations where the present value of fulfilment cash flows exceeds the fair value of the insurance contract liability. We recommend that the positive difference between the fair value and the fulfilment value on past acquisitions should be treated as an adjustment to goodwill.

The requirement to present claims development tables for at least 5 years at transition is problematic due to the requirement to reconcile the claims in the development table to the carrying amounts of the insurance liability recognised in the statement of financial position. This will not be possible for the period before adopting the new standard and the transition provisions should provide relief to only provide this information from the transition date.

(b. If the Board was to adopt the composite margin approach, we believe to require a risk adjustment to be calculated for transition purposes only but not for subsequent measurement would not be cost beneficial. Consistent with our proposal for transition, we would support the full retrospective application of the building blocks model with the IAS 8 impracticability exception. If impracticable, we agree with the proposed composite margin transition model.

(c. We believe it is very important that the effective dates of this proposed standard and IFRS 9 are aligned. Both standards will result in a significant change to the recognised amounts in the financial statements and to have an insurer make significant restatements of comparative information in successive years would not be helpful to the users of the accounts and would make comparison of results difficult. Simultaneous implementation of both standards will also eliminate the need for any exceptions for insurers to the transition provisions and eliminate the need to change previous fair value option elections on adoption of IFRS 9.

(d. As noted in the covering letter, insurers will need sufficient time to implement the proposals in the final standard. Insurers will have to change the data they capture to be able to comply with the proposed measurement model. The requirement for insurers to develop systems to model expected future cash flows scenarios should not be underestimated and sufficient time should be allowed to ensure these measurement models will provide reliable information.

Question 18

Do you have any other comments on the proposals in the exposure draft?

Unlike IAS 39, no guidance is provided on when a modification to an insurance contract should be treated as a modification versus when it should be treated as an extinguishment of the existing contract and the issuance of a new contract. Guidance should be provided for when a change to a contract should be treated a modification with changes captured in the current accounting period and when changes should be recognised as an extinguishment of one contract (with derecognition) and the issuance of a new contract as this could impact on earning recognition patterns through the elimination of the residual margin upon termination of a contract.

The section on business combinations should refer to business combinations that are in the in the scope of IFRS 3, ‘Business Combinations’, thereby excluding common control transactions which are scoped out of IFRS 3.

The definition of reinsurance contracts should refer to “losses on one or moreinsurance contracts issued by the cedant”.

Question 19

Do you agree with the Board’s assessment of the benefits and costs of the proposed accounting for insurance contracts? Why or why not? If feasible, please estimate the benefits and costs associated with the proposals.

As noted in our cover letter, we believe the development of a comprehensive standard for insurance contracts is essential because of the current lack of transparency and comparability of insurer’s financial statements. The current accounting for insurance contracts lacks a consistent measurement approach which users of financial statements demand.

The proposed standard will bring about pervasive changes to the way insurers measure insurance contracts and the current field testing being undertaken by the Board will be paramount to enhance the understanding of the Board’s proposals and to identify solutions to potential problems of interpretation and implementation of the measurement and disclosure requirements. We recommend that the Board work closely with the insurance industry to ensure the finalised model will produce information that is relevant to the decision-making needs of users of insurers’ financial statements.

PwC Comments on FASB Discussion Paper: Preliminary Views on Insurance Contracts

Comment letter

About the author

About the author

Assurance services

What's new on CFOdirect

PwC Comments on FASB Discussion Paper: Preliminary Views on Insurance Contracts

Comment letter

About the author

About the author

Assurance services

Weekly newsletter and alerts

What's new on CFOdirect