Proposed Accounting Standards Update, Financial Instruments Credit Losses (Subtopic )

Tel +44 (0)20 7694 8871 8 Salisbury Square Fax +44 (0)20 7694 8429 London EC4Y 8BB mark.vaessen@kpmgifrg.com United Kingdom Mr Hans Hoogervorst International Accounting Standards Board 1 st Floor 30 Cannon Street London EC4M 6XH Technical Director File Reference No. 2012-260 Financial Accounting Standards Board 401 Merritt 7, PO Box 5116 Norwalk, Connecticut 06856-5116 Our ref Contact MV/288 Mark Vaessen Dear Mr Hoogervorst and Technical Director Comment letter on: ED/2013/3 Financial Instruments: Expected Credit Losses; and Proposed Accounting Standards Update, Financial Instruments Credit Losses (Subtopic 825-15) We appreciate the opportunity to comment on the International Accounting Standards Board s (IASB) Exposure Draft ED/2013/3 Financial Instruments: Expected Credit Losses (the ED) and the Financial Accounting Standards Board s (FASB) Proposed Accounting Standards Update, Financial Instruments Credit Losses (Subtopic 825-15) (the PASU). We have consulted within the KPMG network in respect of this letter, which represents the views of the international network of KPMG member firms, including KPMG LLP (U.S.). This letter is being submitted to both the IASB and the FASB (jointly the Boards). We are submitting one letter in response to both exposure drafts in order to facilitate joint deliberation of the comments by the Boards and to assist the Boards in arriving at a single solution. We continue to believe that it is very important for the Boards to work together in order to achieve a high-quality converged solution for recognition of impairment of financial instruments. It would be disappointing if, in the context of the G20 call for convergence, there were to be divergence on an issue which is so fundamental to the financial reporting of financial and insurance companies and which has been at the root of many commentators concerns arising from the financial crisis. The lack of comparability resulting from the divergence would create difficulties for users trying to understand financial statements. In addition, members of the Boards Expert Advisory Panel have noted that implementation of different IASB and FASB, a UK company limited by guarantee, is a member of KPMG International Cooperative, a Swiss entity. Registered in England No 5253019 Registered office: 8 Salisbury Square, London, EC4Y 8BB

proposals by entities that have operations that apply both U.S. GAAP and IFRS would be a significant operational challenge. Consistent with our 2010 comment letter to the FASB on the Proposed Accounting Standards Update, Accounting for Financial Instruments and Revisions to the Accounting for Derivative Instruments and Hedging Activities issued in May 2010, we support the proposal to establish a single impairment model that would be applied to all financial assets. We continue to agree with the objective of reducing delays in recognition of credit losses and, thus, support an expected credit loss model. However, we continue to believe that it is inappropriate to immediately recognise in profit or loss lifetime 1 expected credit losses on financial assets within the scope of the proposals at the time of their origination or acquisition. Immediate recognition of lifetime expected credit losses ignores the business practice of pricing credit risk into the terms of financial instruments and is not consistent with the economic reality that such losses do not occur immediately and are compensated wholly or partially by interest margin recognised over time. Further, recognition of lifetime expected credit losses on day one contradicts the measurement of financial assets at initial recognition at either fair value (because the effects of uncertainty about future cash flows are included in the fair value measurement) or transaction price (because the transaction price contemplates credit risk). Overall, we support the IASB s proposed model for impairment of financial instruments as we believe that it is a reasonable compromise between reflecting the underlying economics of a lending transaction and aiming to ease the operational complexities that would result from spreading the initial estimate of lifetime expected credit losses as part of the effective interest rate. Although the model proposed in the ED would result in the recognition of some expected credit losses at initial recognition of a financial asset, this is preferable to the recognition of lifetime expected credit losses at that time. However, if as a result of the feedback received and further deliberations, the FASB s model is favoured because of its conceptual simplicity, we would support a model that recognises lifetime expected credit losses in the interest of having one converged standard, provided it does not result in such losses being recognised in profit or loss on initial recognition of a financial instrument. Full recognition of lifetime expected credit losses in profit or loss could be avoided by deferring, in the statement of financial position, a debit balance equal to the initial estimate of lifetime expected credit losses and recognising it in profit or loss over future periods. If this approach were chosen, further work would be required to develop the method of amortisation of the deferred amount to profit or loss in light of the operational complexities involved. 1 We note that the FASB decided not to use the term lifetime expected credit losses in the PASU; instead, expected credit losses is used to convey the same meaning. In this letter, we use lifetime expected credit losses. MV/288 2

We would also like to make the following observations regarding other key aspects of the proposed models: Use of the effective interest rate (EIR) We support the FASB proposals that would require the use of the EIR of a financial asset as the discount rate in measuring the expected credit losses. The Boards have noted that the EIR is conceptually appropriate for calculations of amortised cost. We agree with this assessment. We also agree that calculation of a precise EIR, particularly where an impairment loss is measured on a portfolio basis, may be operationally challenging. Accordingly, we suggest that the Boards clarify that it would be acceptable to use an approximate EIR when measuring expected credit losses on a portfolio basis. Purchased credit-impaired financial assets We support the FASB proposals to set up a credit loss allowance on initial recognition of purchased credit-impaired financial assets and to present the allowance as a gross-up on the statement of financial position. Also, we believe application of this approach is appropriate for all financial assets that are creditimpaired at initial recognition, irrespective of whether they are purchased or originated. We favour this approach because: purchased and originated credit-impaired financial assets would attract a credit loss allowance in a similar way to other financial assets within the scope of the ED; the resulting presentation would be more transparent and informative for users; and this approach would not result in a negative credit loss allowance, which could be the case under the ED. Recognition of interest We support the IASB proposals for recognition of interest on financial assets for which there is objective evidence of impairment rather than the FASB non-accrual approach for financial assets when it is not probable that substantially all principal or substantially all interest will be received. We favour this approach because: the IASB proposals are more consistent with the amortised cost principle; and the FASB non-accrual approach relies on an arbitrary determination as to whether the expected cash flows from an impaired financial asset represent principal or interest; once this determination is made, income recognition depends on the actual cash received. Assets measured at fair value through other comprehensive income (FVOCI) We support the FASB practical expedient permitting an entity not to recognise a credit loss allowance on assets measured at FVOCI if certain conditions are met. We favour this approach because we believe that a practical expedient would reduce the operational burden on preparers and because the statement of financial position would already reflect the fair value of such financial assets. Further, we believe that the approach would not adversely MV/288 3

impact the quality of the information if the practical expedient would be available only when expected credit losses are insignificant. We comment in more detail in Appendix II (our responses to Questions 7 and 14) on the conditions for the application of the practical expedient. Modifications We do not believe that continuing recognition of an asset is appropriate in all cases when a modification meets the definition of troubled debt restructuring (TDR). Also, we believe that the guidance in the final standard on expected credit losses should be limited to modifications of credit-impaired financial assets rather than extended to all modifications of financial assets. We comment in more detail in Appendix I (our response to Question 8) on the IASB proposals relating to modifications of financial assets and in Appendix II (our response to Question 16) on the removal of the TDR distinction. Collateralised financial instruments We support the IASB proposals for collateralised financial instruments, which require that the estimate of expected cash flows reflect the amount and timing of cash flows that are expected from foreclosure less costs for obtaining and selling the collateral. This approach is consistent with the principles of the Boards models under which the impairment allowance reflects supportable forecasts about future conditions as well as current conditions and an unbiased, probability-weighted amount that is determined by evaluating a range of possible outcomes. In contrast, under the FASB s proposed practical expedient for collateral-dependent financial assets, expected credit losses would reflect only the current fair value of the collateral, which ignores, for instance, expected losses that may arise from adverse changes. Disclosure requirements The ED and PASU propose extensive disclosures. Although we agree that adequate disclosures are essential to explain how entities use their judgement to implement the proposals, it is difficult to see how certain proposed disclosures provide information that is decision-useful. We believe that some of the detailed disclosure requirements appear excessive and we question whether they are consistent with the Boards respective initiatives to identify ways of improving and simplifying disclosures, and whether they achieve the right cost/benefit balance. We comment in more detail on the proposed disclosure requirements in Appendix I (our response to Question 7) and Appendix II (our response to Question 18). We attach the following appendices to this letter: Appendix I that contains our responses to the specific questions posed by the IASB; Appendix II that contains our responses to the specific questions posed by the FASB; and Appendix III that contains our additional observations on the IASB proposals. MV/288 4

If you have any questions about our comments or wish to discuss any of these matters further, please contact Mark Vaessen at +44 (0)20 7694 8871, Andrew Vials at +44 (0)20 7694 8473, or Enrique Tejerina at +1 (212) 909-5530. Yours sincerely MV/288 5

Appendix I KPMG s responses to specific questions posed by the IASB Question 1: (a) Do you agree that an approach that recognises a loss allowance (or provision) at an amount equal to a portion of expected credit losses initially, and lifetime expected credit losses only after significant deterioration in credit quality, will reflect: (i) the economic link between the pricing of financial instruments and the credit quality at initial recognition; and (ii) the effects of changes in the credit quality subsequent to initial recognition? If not, why not and how do you believe the proposed model should be revised? As noted in our covering letter, overall, we support the proposed general approach in the ED to recognise a credit loss allowance (or provision) at an amount equal to 12-month expected credit losses initially and lifetime expected credit losses only after significant deterioration in credit quality has occurred since the initial recognition of an instrument. We acknowledge that the proposed model will not faithfully reflect the economic link between the pricing of financial instruments and the credit quality at initial recognition. The model proposed in the Exposure Draft Financial Instruments: Amortised Cost and Impairment published in November 2009 (the 2009 IASB ED) better reflected this economic link. However, the 2009 IASB ED presented significant implementation challenges for preparers and we believe that the current proposals are a reasonable compromise between reflecting the underlying economics of a lending transaction and aiming to ease operational complexities. We believe that the proposed model will reflect the effects of changes in credit quality subsequent to initial recognition since it would require that changes in expected credit losses (either in 12-month expected credit losses or lifetime expected credit losses, as applicable) subsequent to initial recognition are recognised immediately in profit or loss. (b) Do you agree that recognising a loss allowance or provision from initial recognition at an amount equal to lifetime expected credit losses, discounted using the original effective interest rate, does not faithfully represent the underlying economics of financial instruments? If not, why not? We agree that recognising a credit loss allowance or provision from initial recognition at an amount equal to lifetime expected credit losses, discounted using the original effective interest rate, does not faithfully represent the underlying economics of financial instruments. As noted in our covering letter, we believe that immediate recognition of lifetime expected credit losses in profit or loss ignores the business practice of pricing credit risk into the terms of the financial MV/288 6

instruments and is not consistent with the economic reality that such losses do not occur immediately and are compensated wholly or partially by interest margin recognised over time. Further, recognition of lifetime expected credit losses on day one contradicts the measurement of financial assets at initial recognition at fair value because the effects of uncertainty about future cash flows are included in the fair value measurement. Question 2: (a) Do you agree that recognising a loss allowance (or provision) at an amount equal to 12- month expected credit losses and at an amount equal to lifetime expected credit losses after significant deterioration in credit quality achieves an appropriate balance between the faithful representation of the underlying economics and the costs of implementation? If not, why not? What alternative would you prefer and why? We agree that recognising a credit loss allowance (or provision) at an amount equal to 12-month expected credit losses and at an amount equal to lifetime expected credit losses after significant deterioration in credit quality achieves an appropriate balance between the faithful representation of the underlying economics and the costs of implementation. Please refer to our response to Question 1(a). (b) Do you agree that the approach for accounting for expected credit losses proposed in this Exposure Draft achieves a better balance between the faithful representation of the underlying economics and the cost of implementation than the approaches in the 2009 ED and the SD (without the foreseeable future floor)? We believe that the IASB should seek feedback from preparers on the relative costs of implementing the above proposals. As commented in our response to the 2009 IASB ED, we believe that the principal weakness of that model was its substantial operational complexity. The proposals in this ED ease that operational complexity. We believe that the approach in the Supplement to ED/2009/12 Financial Instruments: Amortised Cost and Impairment (without the foreseeable future floor) (the SD) would have represented credit losses more faithfully than the ED as it would have required that lifetime expected credit losses in the good book be time-apportioned rather than recognised immediately. However, in our comment letter on the SD we raised a concern that: the minimum criteria for when a financial asset could no longer be included in the good book may not result in transfers to the bad book at a sufficiently early stage and so may result in delayed recognition of full allowance for credit losses; and MV/288 7

the transfer criteria in the SD were based on management practice rather than on the more objective test of when the credit risk has increased significantly that is included in the ED. The proposals in the SD may have been more challenging to implement than the ED as the SD would have required lifetime expected credit losses to be estimated for all financial assets and a time-proportional amount to have been calculated. We suggest that, unless there is a perception by respondents that the approach in the SD would achieve a significantly better balance between the faithful representation of the underlying economics and the cost of implementation than the approach in the ED, the IASB does not revert to the proposals in the SD at this stage. (c) Do you think that recognising a loss allowance at an amount equal to the lifetime expected credit losses from initial recognition, discounted using the original effective interest rate, achieves a better balance between the faithful representation of the underlying economics and the cost of implementation than this Exposure Draft? We do not believe that recognising a credit loss allowance at an amount equal to the lifetime expected credit losses from initial recognition, discounted using the original effective interest rate, achieves a better balance between the faithful representation of the underlying economics and the cost of implementation than this ED. As noted in our answer to Question 1, we believe that immediate recognition of lifetime expected credit losses in profit or loss ignores the business practice of pricing credit risk into the terms of financial instruments and is not consistent with the economic reality that such losses do not occur immediately and are compensated wholly or partially by interest margin recognised over time. Question 3: (a) Do you agree with the proposed scope of this Exposure Draft? If not, why not? We agree with the proposed scope of this ED. In particular, we believe that the inclusion of certain financial guarantee contracts and loan commitments within the scope reduces complexity and is better aligned with the way that many entities, especially in the financial sector, manage credit risk. Most banks consider the amount receivable and the undrawn amount of the commitment together for risk management purposes, while currently losses related to financial guarantees and loan commitments are generally accounted for under IAS 37 Provisions, Contingent Liabilities and Contingent Assets rather than under IAS 39 Financial Instruments: Recognition and Measurement. However, please refer to our response to Question 9 which discusses our concerns relating to the application of the general impairment model to loan commitments and financial guarantee contracts. MV/288 8

(b) Do you agree that, for financial assets that are mandatorily measured at FVOCI in accordance with the Classification and Measurement ED, the accounting for expected credit losses should be as proposed in this Exposure Draft? Why or why not? We agree in principle that, for financial assets that would be mandatorily measured at FVOCI in accordance with the ED/2012/4 Classification and Measurement: Limited Amendments to IFRS 9 (Classification and Measurement ED), the accounting for expected credit losses should be as proposed in the ED. We believe that recognition and measurement of expected credit losses should be the same, irrespective of whether a financial instrument is accounted for at amortised cost or at FVOCI. Also, the current model in IAS 39 for available-for-sale assets has been heavily criticised as it is based on fair value fluctuations that may be driven by changes in liquidity and interest rates and therefore obscures impairment losses. Using a single model will increase comparability among financial assets and also reduce complexity. However, as noted in our covering letter, we support the practical expedient proposed by the FASB, permitting an entity not to recognise a credit loss allowance on assets measured at FVOCI if certain conditions are met i.e. the fair value of the financial asset is greater than or equal to the amortised cost and the expected credit losses are insignificant. We favour this approach because we believe that a practical expedient would reduce the operational burden on preparers and because the statement of financial position would already reflect the fair value of such financial assets. Further, we believe that the approach would not adversely impact the quality of the information if the practical expedient would be available only when expected credit losses are insignificant. We make further suggestions regarding the practical expedient proposed by the FASB in our responses to Questions 7 and 14 of Appendix II. Question 4: Is measuring the loss allowance (or a provision) at an amount equal to 12-month expected credit losses operational? If not, why not and how do you believe the portion recognised from initial recognition should be determined? We believe that the IASB should seek feedback from preparers on whether measuring the credit loss allowance (or a provision) at an amount equal to 12-month expected credit losses is operational. However, we see no reason why measuring the credit loss allowance (or a provision) at an amount equal to 12-month expected credit losses should not be operational. We believe that it is reasonable to expect that entities have the ability to estimate the likelihood of defaults occurring within a 12-month horizon and the effects of those defaults. Many entities in the financial sector already perform similar calculations for regulatory purposes, although those entities would have to identify and quantify the effect of differences between the regulatory requirements and the ED. MV/288 9

Question 5: (a) Do you agree with the proposed requirement to recognise a loss allowance (or a provision) at an amount equal to lifetime expected credit losses on the basis of a significant increase in credit risk since initial recognition? If not, why not and what alternative would you prefer? We agree with the proposed requirement to recognise a credit loss allowance (or a provision) at an amount equal to lifetime expected credit losses on the basis of a significant increase in credit risk since initial recognition. We believe that such an approach is a reasonable way to segregate financial instruments for which a 12-month expected credit loss allowance is recognised from those for which a lifetime expected credit loss is recognised. As noted in our covering letter, we believe that it is inappropriate to immediately recognise in profit or loss lifetime expected credit losses on financial assets at the time of origination or acquisition. Immediate recognition of lifetime expected credit losses in profit and loss ignores the business practice of pricing credit risk into the terms of financial instruments and is not consistent with the economic reality that such losses do not occur immediately and are compensated wholly or partially by interest margin recognised over time. Further, recognition of lifetime expected credit losses on day one contradicts the measurement of financial assets at initial recognition at fair value because the effects of uncertainty about future cash flows are included in the fair value measurement. We agree that recognition of credit losses should be based on more forward-looking information and that recognition of such losses should not be delayed until a narrowly-defined loss event has occurred. We believe that recognising lifetime expected credit losses on the basis of a significant increase in credit risk achieves an appropriate balance. (b) Do the proposals provide sufficient guidance on when to recognise lifetime expected credit losses? If not, what additional guidance would you suggest? We note that the concept of significant increase in credit risk is not defined in the ED and so would require the application of judgement. However, we believe that accounting for expected credit losses should reflect the way in which credit risk is managed, and so enabling entities to define what is significant in the context of their financial instruments would lead to improved financial reporting. Accordingly, we believe that the proposals provide sufficient guidance on when to recognise lifetime expected credit losses. However, we have concerns relating to: the assessment of the significance of increases in credit risk for revolving facilities such as credit cards; MV/288 10

the assessment of significant increase in credit risk for instruments with credit spreads that reset when the credit rating changes; the explanation of the concept of significant increase in credit risk in certain illustrative examples; and the description of when credit risk should be regarded as low. Credit risk for revolving facilities such as credit cards Certain credit facilities such as credit cards can have a life of many years with balances being drawn daily and repaid (fully or partially) at monthly or other intervals. Such facilities could be viewed as a combination of two financial instruments: a loan commitment; and a loan balance. For such financial instruments, it is not clear what would be the starting point for assessing the significance of increases in credit risk. For example, the assessment could be based on a comparison to the credit risk when a contract is signed, or to the credit risk when each drawdown takes place, although the latter approach would be very complex as draw-downs could take place daily. While for drawn balances, entities may be able to use the 30-day rebuttable presumption to determine when recognition of lifetime expected credit losses is appropriate, this would not be appropriate when more forward-looking information on the borrower is available. Additionally, the issue will not be relevant for facilities that can be cancelled without notice as there would be no present obligation to extend credit. Please also refer to our response to Question 9. We suggest that the IASB consider developing guidance in this area. Instruments with credit spreads that reset when the credit rating changes Certain debt instruments include features under which the credit spread resets when their credit rating changes. It is not clear whether for such instruments the period over which the significance of credit risk deterioration is measured would be affected by the periodic credit spread resets. Also, it is not clear whether such resets should be considered akin to a floating rate of interest please refer to item 3 in Appendix III for our comment that discusses the definition of a floating rate instrument in more detail. We suggest that the IASB consider developing guidance for instruments with similar features. Illustrative examples of the concept of significant increase in credit risk We find that the usefulness of examples 3, 4 and 5 is limited. The fact patterns described as the basis for making individual credit assessments for the example loans are brief and in real life a lender would have access to more relevant information. It is difficult to see how a conclusion MV/288 11

could be reached on such limited fact patterns and so it would be difficult to apply the guidance in the examples in practice. Accordingly, we suggest that these examples either be deleted from the final standard or enhanced to clarify the interpretive point being made. Description of when credit risk should be regarded as low The guidance in paragraph 6 of the ED for determining whether the credit risk of a financial instrument is low appears internally inconsistent. Under paragraph 6, a financial instrument has low credit risk if a default is not imminent and any adverse economic conditions or changing circumstances may lead to, at most, a weakened capacity of the borrower to meet its contractual cash flow obligations on the financial instrument. The first part of the description indicates that credit risk is low unless default is imminent, which suggests a relatively high threshold before recognition of lifetime expected credit losses. However, the second part of the description would require entities to be satisfied that any adverse economic conditions or change in circumstances would lead to at most, a weakened capacity of the borrower to meet its contractual obligations. We believe the second part of the description would suggest an extremely low threshold for recognition of lifetime expected credit losses, because extreme scenarios could always be envisaged where even the best quality borrower would fail to meet its obligations. This would be the case even for investment-grade assets. One way to resolve this inconsistency would be to use the wording similar to that used by rating agencies to describe investment-grade assets, as the ED states that an investment grade asset would be considered to have a low credit risk. For example, Fitch defines investment-grade obligations as follows: BBB' ratings indicate that expectations of credit risk are currently low. The capacity for payment of financial commitments is considered adequate but adverse business or economic conditions are more likely to impair this capacity. [source: FitchRatings- Definition of Ratings and Other Forms of Opinion, May 2013] We also note that the description in paragraph 6 of the ED is inconsistent with the explanation in paragraph BC76 of the ED, which states that the tracking of credit quality and the assessment of deterioration in credit quality is limited to financial instruments whose credit risk is low enough that adverse economic conditions or changes in business or financial circumstances could, at most, lead to the inability to fully recover cash flows in the medium or short term. Such credit risk is typically equivalent to the investment grade market convention. It seems that there may be a typographical error in the first sentence in paragraph BC76 as it appears to indicate that the assessment of deterioration in credit quality would be limited to financial instruments with low credit risk. We believe that the opposite would be true: the assessment of significance of deterioration in credit risk would have to be made for all assets other than those whose credit risk is low. (c) Do you agree that the assessment of when to recognise lifetime expected credit losses should consider only changes in the probability of a default occurring, rather than changes in expected credit losses (or credit loss given default ( LGD ))? If not, why not and what would you prefer? MV/288 12

We agree with the proposal in the ED that the assessment of when to recognise lifetime expected credit losses should consider only changes in the probability of a default (PD) occurring, rather than changes in the loss given default (LGD) or in the amount of expected credit losses. Although in our view, the overall credit quality of an asset is affected by both PD and LGD, monitoring of both indicators for all assets would add complexity. Accordingly, we believe that focusing solely on changes in PD is a reasonable simplification that will ease the operational burden on preparers. (d) Do you agree with the proposed operational simplifications, and do they contribute to an appropriate balance between faithful representation and the cost of implementation? We agree with the following proposed operational simplifications: to recognise 12-month expected credit losses for instruments whose credit risk is low; and the rebuttable presumption that lifetime expected credit losses are recognised when contractual payments are more than 30 days past due. We believe these operational simplifications contribute to an appropriate balance between faithful representation and the cost of implementing the proposals. We support the proposal that if the presumption that lifetime expected credit losses are recognised when contractual payments are more than 30 days past due is not appropriate in particular circumstances then it may be rebutted. However, an illustrative example of circumstances where the presumption could be rebutted would be useful. In addition, we have a concern about the description of when credit risk of an asset is low please refer to our response to Question 5(b). (e) Do you agree with the proposal that the model shall allow the re-establishment of a loss allowance (or a provision) at an amount equal to 12-month expected credit losses if the criteria for the recognition of lifetime expected credit losses are no longer met? If not, why not, and what would you prefer? We agree with the proposal that the model shall allow the re-establishment of a credit loss allowance (or a provision) at an amount equal to 12-month expected credit losses if the criteria for the recognition of lifetime expected credit losses are no longer met. This approach would allow the credit loss allowances for financial instruments with similar cumulative increases in credit risk since initial recognition to be assessed in a similar way and therefore would better reflect the economic position at the reporting date. MV/288 13

Question 6: (a) Do you agree that there are circumstances when interest revenue calculated on a net carrying amount (amortised cost) rather than on a gross carrying amount can provide more useful information? If not, why not, and what would you prefer? We agree that there are circumstances when interest revenue calculated on a net carrying amount (amortised cost) rather than on a gross carrying amount can provide more useful information. We believe that there comes a point at which the probability that the full contractual cash flows will not be collected has increased such that continuing to recognise interest on a contractual basis would result in overstatement of interest income and would no longer faithfully represent the economic substance. (b) Do you agree with the proposal to change how interest revenue is calculated for assets that have objective evidence of impairment subsequent to initial recognition? Why or why not? If not, for what population of assets should the interest revenue calculation change? We agree with the proposal to change how interest revenue is calculated for assets that have objective evidence of impairment subsequent to initial recognition. We believe that when there is objective evidence of impairment, continuing to recognise full interest in accordance with the contractual terms would not reflect the economic substance as it is no longer probable that the contractual cash flows will be collected in full. (c) Do you agree with the proposal that the interest revenue approach shall be symmetrical (ie that the calculation can revert back to a calculation on the gross carrying amount)? Why or why not? If not, what approach would you prefer? We agree with the proposal that the interest revenue approach shall be symmetrical. In circumstances where objective evidence of impairment no longer exists, recognising full contractual interest revenue separately from expected credit losses should be consistent with other assets that do not have objective evidence of impairment. However, we suggest that the IASB consider providing guidance for the accounting for when the calculation reverts from one based on the net carrying amount to gross carrying amount. Cash received in respect of a credit-impaired asset will reduce its amortised cost i.e. the net carrying amount but it is unclear whether it should reduce the gross carrying amount or be treated as an adjustment to the credit loss allowance. For example, full contractual interest payments may be received that exceed interest income recognised by applying the EIR to the net carrying amount. If the excess cash received reduces the gross carrying amount, then the gross carrying amount would no longer equate to the net present value of contractual cash flows discounted using the original EIR. Because the loss allowance is adjusted at the reporting date to an amount reflecting expected credit losses, any of these methods should have the same effect on net profit or loss, but would result in different presentation in: MV/288 14

the statement of profit or loss and other comprehensive income (OCI) i.e. they impact interest income and impairment losses differently; and the statement of financial position i.e. they impact the gross amount and credit loss allowance differently. Question 7: (a) Do you agree with the proposed disclosure requirements? Why or why not? If not, what changes do you recommend and why? As noted in our covering letter, although we agree that adequate disclosures are essential to explain how entities use their judgement to implement the proposals, it is difficult to see how certain proposed disclosures provide information that is decision-useful. We believe that some of the detailed requirements appear excessive and we question whether the proposed requirements are consistent with the IASB s initiative to identify ways of improving and simplifying disclosures and to achieve an appropriate cost/benefit balance. We suggest that in defining mandatory disclosures, the IASB place more emphasis on requiring preparers to apply judgement in determining which detailed requirements are relevant to their particular facts and circumstances. The Enhanced Disclosure Task Force in their report Enhancing the Risk Disclosures of Banks issued in October 2012 (the EDTF report), noted that: The bank should provide disclosures only if they are material and reflect its activities and risks [Principle 3: Disclosures should present relevant information]. The following are examples of proposed disclosures that appear not to provide information that is decision-useful or that is excessive: The requirements in paragraphs 35 and 36 to provide a reconciliation of the gross carrying amounts for each class of assets. The requirements in paragraph 38(a) for modified financial instruments. In addition, we believe that the disclosure for modified financial assets should be limited to modifications of credit-impaired financial assets (please refer to our response to Question 8). The requirements in paragraph 39(b) to provide an explanation of changes in estimates of expected credit losses and the causes of those changes. As banks may typically have hundreds of different portfolios of financial instruments that may experience different changes in the credit loss allowance for different reasons, this disclosure may potentially result in voluminous information that may be costly to prepare and difficult for users to understand. The requirements in paragraph 40(a) to explain any changes in the quality of collateral as a result of deterioration; it is usual for the value of collateral to fluctuate and such fluctuations MV/288 15

may be directionally different for different portfolios and different types of collateral. It appears that the proposals may require entities to capture and aggregate in a meaningful way all instances where the collateral has deteriorated. It is unclear why such granular information on collateral is considered to be useful in all cases. Collateral may be one of many aspects of credit quality of a financial instrument and we believe that it would be more helpful to focus on the key aspects of the credit quality of a particular financial instrument, or a portfolio of financial instruments, rather than applying prescriptive requirements specifically for collateral irrespective of its significance. The EDTF report contains a helpful recommendation in this area (recommendation No 30): Provide qualitative information on credit risk mitigation, including collateral held for all sources of credit risk and quantitative information where meaningful. Collateral disclosures should be sufficiently detailed to allow an assessment of the quality of collateral. The requirements in paragraph 42(b) to explain the changes in the estimates of the credit risk and the cause of those changes relating to determining: where credit risk has increased significantly since initial recognition; and if there is objective evidence of impairment. The changes in the estimates of credit risk and the cause of those changes may be specific to individual financial instruments or portfolios and therefore difficult to aggregate in a meaningful way. Accordingly, the requirement may potentially result in voluminous information that may be costly to prepare and difficult to understand. In addition, we believe that the following disclosure requirements are not clear: Whether the requirements in paragraph 35(d) to disclose the total amount of undiscounted expected credit losses at initial recognition for purchased or originated credit-impaired (POCI) financial assets apply to: only assets acquired during the reporting period (similar to the requirements under IFRS 3 Business Combinations for assets with incurred losses acquired in a business combination); or all assets recognised in the statement of financial position at the end of the reporting period, irrespective of when they were acquired. If the disclosure requirements apply to the latter, then calculating the relevant amounts would be much more challenging for preparers. We do not believe they would provide useful information as the amounts might change significantly after initial recognition and may no longer be relevant. MV/288 16

Whether the requirement in paragraph 40(a) to disclose any changes in the quality of collateral as a result of deterioration is required for changes: since the beginning of the reporting period; or since the initial recognition of the collateralised financial instrument. (b) Do you foresee any specific operational challenges when implementing the proposed disclosure requirements? If so, please explain. Yes, we do foresee operational challenges when implementing the proposed disclosure requirements. Please refer to our response to Question 7(a). We believe that the IASB should seek feedback from preparers regarding specific operational challenges when implementing the proposed disclosure requirements. (c) What other disclosures do you believe would provide useful information (whether in addition to, or instead of, the proposed disclosures) and why? We do not have examples of any additional disclosures that would provide useful information. Question 8: Do you agree with the proposed treatment of financial assets on which contractual cash flows are modified, and do you believe that it provides useful information? If not, why not and what alternative would you prefer? We do not fully agree with the proposed treatment of financial assets whose contractual cash flows are modified. We believe that the guidance in the final standard on expected credit losses should be limited to modifications of credit-impaired financial assets rather than extended to all modifications of financial assets. If the scope of the guidance is limited to modifications of credit-impaired financial assets, then we agree with the proposed accounting treatment. In other cases, the proposed treatment may not always be appropriate. For example, a bank may extend the maturity of a loan of a good customer to meet his business needs and increase the interest rate to reflect market rates for the extended maturity. We believe that in such a scenario, if the modification does not result in derecognition, it would be inappropriate for a bank to recognise the increase in the margin immediately in profit or loss as a gain on modification; rather the EIR should be amended prospectively. Furthermore, there is no guidance in the ED regarding the line item in the statement of profit or loss and OCI in which modification gains or losses should be presented. We suggest that the final standard clarify that modification gains or losses on credit-impaired financial assets are presented in profit or loss in an impairment line, together with other impairment gains or losses. MV/288 17

The interrelation between gains and losses on modification of credit-impaired financial assets and changes in the allowance for credit losses should be reflected in the same line item. In addition, the ED refers to IFRS 9 Financial Instruments for determining when a change in the terms of a financial asset results in derecognition. Under the derecognition provisions of IFRS 9, which remain unchanged from IAS 39, a financial asset is derecognised when the contractual rights to the cash flows expire. However, there is no specific guidance in IFRS 9 or the ED on how this criterion should be applied to modifications of a financial asset to determine whether it should be derecognised as a result of the modification. A submission on this matter was made to IFRS Interpretations Committee (IFRIC) in May 2012 but the IFRIC decided not to add it to its agenda and so the issue remains unclear. We recommend that guidance be provided in this area at it would increase consistency in the application of the derecognition requirements for financial assets. Question 9: (a) Do you agree with the proposals on the application of the general model to loan commitment and financial guarantee contracts? Why or why not? If not, what approach would you prefer? We agree with the proposal to apply the general impairment model to loan commitments and financial guarantee contracts. Please refer to our response to Question 3(a). However, we have concerns relating to: the inclusion of loan commitments only where there is a present contractual obligation to extend credit; the discount rate used to measure expected credit losses for loan commitments; and the measurement of financial guarantee contracts and loan commitments granted at belowmarket interest rates. Inclusion of loan commitments only where there is a present contractual obligation to extend credit In some jurisdictions, certain loan commitments granted to retail customers (such as credit cards or overdrafts) can be withdrawn on demand or with very little notice. This means that, for these products, expected credit losses estimated in accordance with the requirements of the ED would be very small or zero. This result appears counterintuitive as these products often attract a high level of defaults. Although, contractually, the lender may withdraw the credit line on demand or demand an immediate repayment of the drawn balance, it would not normally have information on when it would be advisable to do so. Often the key monitoring tool for such financial instruments is an overdue status and by the time the loan is overdue impairment losses may MV/288 18

have already occurred. The same issue arises in respect of drawn balances for these products. We recommend that the IASB give more consideration to this issue to ensure that expected credit losses on such financial instruments are not understated. Discount rate used to measure expected credit losses for loan commitments Paragraph B29(b) requires that for undrawn loan commitments an entity uses a discount rate that reflects the current market assessment of the time value of money, risk adjusted where appropriate. It is not clear why this requirement is different from the general requirement in paragraph B29(a) allowing use of any reasonable rate that is between (and including) the riskfree rate and the EIR. Measurement of financial guarantee contracts and loan commitments granted at below-market interest rates The ED does not change the general requirements under IAS 39/IFRS 9 to measure liabilities that result from financial guarantee contracts and loan commitments granted at below-market interest rates at the higher of: the amount of the credit provision determined in accordance with the ED; and the amount initially recognised, less cumulative amortisation in accordance with IAS 18 Revenue. This would mean that generally no expected credit loss would be recognised at initial recognition of a financial guarantee, or loan commitments granted at below-market interest rates, as the amount initially recognised i.e. the fair value of the financial instrument is likely to be higher than any expected credit losses. As expected credit losses on a good quality instrument would usually decrease over time, it could be that credit provisions calculated under the ED would generally be less than the amounts initially recognised, less cumulative amortisation. This means that no credit loss allowance would be recognised for many financial guarantees or loan commitments granted at below-market interest rates. Also, inconsistency may arise in situations when the issuer of the guarantee is paid its fee or premium in installments over the life of the guarantee rather than in full at inception. The ED does not discuss whether or how future premium receipts are to be included in the measurement of expected credit losses. For example, if no premium is received at inception but is instead all to be received in future installments, the fair value of the financial guarantee is likely to be zero at inception. In this case, assuming future premium receipts are not included in calculating expected credit losses, it would appear that the amount of credit provision determined in accordance with the ED at inception would always be higher than the fair value at initial recognition. This would mean that whether an allowance for expected credit loss is recognised would depend on the manner in which the guarantee premium is collected. We do not believe MV/288 19