This comment letter, which reflects the views of the international BDO network of independent member firms, was sent by BDO IFR Advisory Limited to the International Accounting Standards Board on 3 April 2011.
A copy of the exposure draft is attached at the foot of this page.International Accounting Standards Board30 Cannon Street
London
EC4M 6XH
1 April 2011
Dear Sir
IASB/FASB Joint supplementary document Financial Instruments: Impairment
We are pleased to comment on the above supplementary document (the SD). Following consultation, this letter summarises the views of the BDO network. We note that substantially the same letter is being submitted under separate cover to the FASB by our US member firm, BDO USA, LLP.
We appreciate the efforts to pursue improvements to the separate impairment models initially proposed by the Boards and to find a converged solution. Consistent with our comments on other recent proposals, we believe that a key objective of any proposal issued at this time should be the development of converged high quality accounting standards, in particular to avoid the risk of accounting arbitrage as entities look to the (differing) guidance issued by each of the Boards. We are also encouraged that the IASB has favourably considered the requests made by many constituents for the decoupling of interest income and expected credit losses.
However, we have significant reservations about the proposals set out in the SD and the extent to which they are supported by clear principles. In certain respects, we believe the proposals would be likely to lead to excessive inconsistency in practice, particularly the requirement to look forward for the ‘foreseeable future’.
Key concerns
The proposals set out in the SD appear to reflect more of a compromise between the two Boards than convergence of views, with the result that it is difficult to determine the extent to which core principles support the proposals, and how they are to be applied. In that context, if the proposed model is to be taken forward, additional clarification and appropriate implementation guidance will be necessary. However, even with additional guidance we are concerned that inconsistencies in application are likely to arise both among different entities and different jurisdictions.
In consequence, based on the proposals as drafted we believe that the potential benefits of the SD are outweighed by our concerns and, therefore, are not able to support them. We believe that, if the Boards continue with the proposals as drafted, substantial field testing will be needed before the new standard is finalised to assist in determining whether its requirements can be implemented satisfactorily in practice. We also consider that outreach activities should extend to include regulators of financial institutions.
Assuming that the Boards take the proposals forward, we believe that several key concepts will need substantial additional clarification and illustrative examples, such as the definitions of new terms, the point at which a loss event occurs, more specific guidance and illustrations of when loans should be transferred from the ‘good book’ to the ‘bad book’ and whether such a reclassification is meaningful.
Our detailed responses set out in Appendix A elaborate on those concerns. Without enhancement, we do not believe that the proposals would be operational for preparers, or informative to users, whether in the financial services sector or otherwise.
Timing of the project
We note that many of the Boards’ constituents have been focussed on year-end closing procedures and audit requirements, and therefore may not have been able to devote an ideal amount of time to reviewing the SD. In that context, and in view of the significant reservations that we have about the proposals as drafted, it is likely to be appropriate for the Boards to redeliberate the impairment model, expand it to cover closed portfolios and individual asset balances, and re-expose comprehensive proposals for comment later in 2011. While we fully appreciate the pressures that the Boards are under, and the desire to complete a number of projects in the short term, we would be supportive of the Boards taking additional time to develop a high quality accounting standard, even if this means that Board decisions are required in the second half of 2011.
Consequential amendments
We note that the Boards have not exposed for comment detailed amendments to current IFRSs or US GAAP, including the US Accounting Standards Codification. We assume that the Boards do not believe these are necessary for an understanding of the proposals. However, given the significance of the financial instruments and other MoU projects, it seems likely that any final amendments will require a separate exposure period to mitigate unintended consequences that otherwise would only be identified through post implementation reviews or additional standard setting.
Our responses to the specific questions included in the ED are set out in the attached Appendices.
We hope that our comments and suggestions are helpful. If you would like to discuss any of them, please contact Andrew Buchanan at +44 (0)20 7893 3300.
Yours faithfully
BDO IFR Advisory Limited
Appendix A
As indicated in our covering letter, we have significant reservations about the operationality of the proposals in the SD. However, the following responses are provided in the context of the potential that the Boards will continue with their proposals.
Question 1
Do you believe the approach for recognition of impairment described in this supplementary document deals with this weakness (ie delayed recognition of expected credit losses)? If not, how do you believe the proposed model should be revised and why?
It is difficult to conclude whether the model in the supplementary document (SD) will satisfactorily resolve the delayed recognition of credit losses since it does not address measurement, nor does it provide recognition guidance for individual loans or debt securities, as discussed in paragraph IN20. However, transitioning from an incurred loss model to a forward-looking approach suggests some degree of credit losses will be recognised more quickly than they are under current standards.
While we have significant reservations with the proposals set out in the SD, we believe it is to an extent an improvement over both Boards’ recent proposals. First, the SD captures near-term foreseeable losses, considering all available information including future conditions based on reasonable and supportable information. As such, not all losses would be recognised on Day 1, nor would all expected losses be “smoothed” over the life of the instrument. While imperfect, in particular as it appears that an adjustment might be made for certain losses on initial recognition, this indicates the timing of loss recognition may more closely approximate the actual loss events compared to recognising all expected losses at Day 1, a concept we do not support.
Second, we agree with the decision to decouple interest revenues from credit losses. In our view, the SD approach is preferable to a model under which interest income is recognised on assets net of the allowance for credit losses. We note many users were not supportive of the ‘net’ approach (although some were). In contrast, they expressed a preference for yields calculated on the basis of the instrument’s contractual terms. Further, calculating interest income ‘gross’ instead of ‘net’ will also be easier for preparers to implement since many creditors use separate information systems to track interest income and manage credit risk.
However, we have significant concerns around the operationality of the proposals as drafted, and believe that significantly more guidance would be needed to make them operational on a consistent basis, in particular how the ‘foreseeable future’ is to be determined.
Question 2
Is the impairment model proposed in the supplementary document at least as operational for closed portfolios and other instruments as it is for open portfolios? Why or why not?
Although the supplementary document seeks views on whether the proposed approach is suitable for open portfolios, the boards welcome any comments on its suitability for single assets and closed portfolios and also comments on how important it is to have a single impairment approach for all relevant financial assets
Generally, the concepts for recognising losses on an open (dynamic) portfolio should be at least as easy to apply to a closed (static) portfolio. However, we suggest that the Boards discuss and determine whether the model for closed portfolios could be made more comparable and precise. For instance, if discounting is applied, the Boards could consider specifying the effective interest rate as the discount rate for closed portfolios. Consistent use of the effective interest rate would improve comparability and be conceptually more appropriate than applying the risk-free rate or any other discount rate. In contrast to open portfolios, where an option to choose the discount rate is proposed, it may be far less complex to determine the actual effective interest rate for closed portfolios and individual items.
As it relates to individual loans, we are unclear whether the SD would result in an allowance at Day 1 in all cases. We disagree with the recording of losses on initial recognition because we believe it is inconsistent with the amortised cost measurement attribute and suggest the Boards provide clarification along these lines in the final standard.
“Open portfolio” and “closed portfolio” appear to be new terms under IFRS and US GAAP. To avoid confusion, we recommend these terms be defined, as were “portfolio segment” and “class” in Accounting Standards Update 2010-20. FASB Accounting Standards Topic 310: Disclosures about the Credit Quality of Financing Receivables and the Allowance for Credit Losses We also recommend providing implementation guidance to illustrate both concepts.
Question 3
Do you agree that for financial assets in the ‘good book’ it is appropriate to recognise the impairment allowance using the approach described above? Why or why not?
We understand factors that motivated the Boards to reach the ‘higher of’ compromise reflected in the SD. Paragraph 2 indicates “At each reporting date, an entity shall recognise an impairment allowance that is the total of: (a) for assets for which it is appropriate to recognise expected credit losses over a time period, the higher of: (i) the time-proportional expected credit losses; and (ii) the credit losses expected to occur within the foreseeable future (which shall be no less than twelve months after an entity’s reporting date); and (b) for all other assets, the entire amount of expected credit losses.” However, they have resulted in a model that doesn’t readily reflect an underlying principle. If the Boards ultimately affirm the ‘higher of’ model as a practical resolution to the ‘too little, too late’ problem cited by constituents, users will ultimately decide whether the resulting credit allowances provide decision-useful information. If they do not find the information improves their ability to make investment decisions, the model’s benefits would be unclear.
Without a single principle such as incurred loss, expected loss or another alternative, we agree the notions of ‘time proportion’, a ‘foreseeable future’ floor (i.e., 12 months or more) and the entire amount of credit losses for ‘bad book’ assets are necessary components for making the model operational. Otherwise, practitioners would be likely to reach widely different conclusions about the amount and timing of losses to record, reducing comparability for users. However, as drafted, we believe that the SD permits a significantly greater degree of flexibility than is desirable.
Question 4
Would the proposed approach to determining the impairment allowance on a time-proportional basis be operational? Why or why not?
The mechanics in the illustrative examples beginning in paragraph IE1 appear operational, once the actual impairment amounts have been determined. The challenges in practice will be determining the amount of credit losses to record, the estimated life over which they should be recorded, and the discrete periods to which they should be allocated. Further, we note the ‘higher of’ approach effectively requires practitioners to apply two different models in order to determine the amount of credit losses to recognise. Applying two models will, unavoidably, be more complex and costly than applying one.
The time-proportional approach requires a determination of the weighted average age and weighted average life of the portfolio (or components of the portfolio). However, no guidance is provided on the computations and the SD appears to assume that these are commonly understood concepts in practice. Determining these dynamic weighted averages based on the due date, maturity and renewal characteristics for different types of loans—such as demand loans, lines of credit, term loans that are expected to renew, and loans with no maturities like credit card receivables—could be challenging. Field testing should be conducted if the Boards intend to move forward with this guidance. Otherwise, post implementation reviews or additional standard setting may be necessary.
While the ability of all entities to reasonably predict lifetime losses diminishes as the length of time increases, the cost/benefit assessment will differ between the largest financial institutions and all other creditors. This is particularly true of non-financial entities. As such, some might question the extent to which estimating lifetime losses for long term open pools will result in meaningful information in some situations, although we note that these types of pools are likely to arise principally at financial institutions.
The variables in estimating time proportional losses will be equally challenging to audit. We recommend disclosure of the factors comprising the total losses, key assumptions made in determining the amount recognised to date, and a discussion of significant changes from the prior period, including reasons for the change. These disclosures should enhance consistent estimates each period and serve as a deterrent for potentially abusive changes in estimates.
Lastly, we note one of the objectives of the financial instruments project in general, as well as the credit impairment model in particular, is to reduce complexity. We are not convinced that the proposals set out in the SD achieve that objective. As noted elsewhere in this letter:
the ‘higher of’ approach is effectively two models in one (time proportional and a floor);
it is not clear how the ‘foreseeable future’ would be determined consistently among different entities and jurisdictions;
the Boards have yet to decide whether it will apply to all debt instruments;
it is not clear when a loan should be reclassified between the good and bad book;
the Boards still need to agree on measurement guidance; and
additional definitions and examples will be needed to promote consistency.
As such, it is unclear whether existing complexity will be meaningfully reduced, or whether the SD contains complexities of its own.Question 5Would the proposed approach provide information that is useful for decision-making? If not, how would you modify the proposal?See our responses to questions 3 and 4. Question 6Is the requirement to differentiate between the two groups (i.e. ‘good book’ and ‘bad book’) for the purpose of determining the impairment allowance clearly described? If not, how could it be described more clearly?As described in paragraph 3, the SD distinguishes between the two groups on the basis of when management’s ‘objective’ changes from receiving payments to cost recovery. This language appears similar to the concept of ‘management’s intent’ in Topic 320 Investments—Debt and Equity Securities and IAS 39 Financial Instruments: Recognition and Measurement for distinguishing between trading, available-for-sale and held-to-maturity securities. In its recent financial instruments exposure draft, the FASB concluded it was more appropriate to assess the entity’s business strategy rather than providing detailed guidelines about assertions of intent, holding periods, and so forth. We believe the same rationale suggests distinguishing between the good book and bad book on the basis of observable behaviour (a change in business activities), rather than the more abstract notion of a change in concept.However, while business activities may be a helpful guide for the distinction between a good and bad book, we caution against the use of this as the only driver. We believe that, in addition to a requirement to look at how an entity manages its assets, guidance should be provided setting out circumstances in which there is a requirement for a reclassification from good to bad book. Linked to this, we consider that the illustrative examples towards the end of paragraph B3, setting out actions that an entity may take that illustrate that ‘bad book’ classification is appropriate, are too vague and need to be tightened. For example, the act of making contact with a debtor by phone is not always in itself a definitive indicator that a loan should be moved to the ‘bad book’. Moreover, as stated in our response to question 7, we recommend that the Boards provide additional guidance to illustrate the intended application of this guidance. In particular, it would be helpful for the Boards to more clearly articulate whether a reclassification from one category to another is expected to result in a significant charge (or credit) to earnings. We do not think the guidance in BC54 is sufficiently clear on this point. We assume that it is anticipated that, on transfer from the good book to the bad book, there would be a two step approach with the existing provision being transferred and an additional provision then being recorded. However, it might reasonably be concluded from the proposals that the amount of credit losses expected to occur within the foreseeable future should converge with the entire amount of expected credit losses as a debt instrument “floats to the bottom” of the good book. If those two amounts are expected to be the same, the incremental relevance of a separate “bad book” category is not entirely clear. In other words, even if management’s business activities with respect to a problem loan differ between the good and bad book, but they have no discernible impact on the amount of the allowance to record, additional guidance from the Boards would be helpful in understanding the reason for the distinction and how it will differ from past practice.Question 7Is the requirement to differentiate between the two groups (i.e. ‘good book’ and ‘bad book’) for the purpose of determining the impairment allowance operational and/or auditable? If not, how could it be made more operational and/or auditable?We believe additional clarity is needed. As indicated in question 6, we believe the language for distinguishing between the good and bad book should more objectively describe risk management business activities with clearer guidance setting out triggers that would typically lead to reclassification. In addition, examples should be provided for different sizes and types of entities. To improve comparability, the Boards’ intent and meaning must be clear. We are not certain the Boards’ conception of good and bad books sufficiently encompasses the various distinctions used by creditors in practice today. For example, paragraph B3 of the SD states that migration from the good book to the bad book occurs when the collectability becomes so uncertain that the entity’s credit risk management objective changes from receiving the regular payments to recovery of all or a portion of the financial asset. In this light, creditors may classify similar loans differently, even though they are performing the same activities because they have different views about when the objective for holding the asset changes. We further understand some banks differentiate between good loans and bad loans on the basis of whether they are assessed for impairment in a homogenous pool or individually under US GAAP. These factors result in migration analyses that are typically based on portfolio types and internal ratings since standards do not exist for determining migrations between ‘good’ and ‘bad’ categories. Although the Boards contemplated diverse risk management policies in paragraph B3 of the SD, additional examples for different types and sizes of entities would promote greater comparability. Otherwise, the generality of the principle may be so vague as to be nonsubstantive. As indicated previously, additional field testing would be appropriate. Therefore, we request clarity on what the Boards intend to be the migration threshold, including financial assets that are not currently held for recovery, but are on a ‘watch list’ ie in an intermediate state of being doubtful as to recovery but not yet held for recovery. Without additional guidance, it may also be difficult to determine the appropriate grouping of portfolios, including any interaction between the SD and loans that have been modified in a troubled debt restructuring (TDR). Question 8Do you agree with the proposed requirement to differentiate between the two groups (ie ‘good book’ and ‘bad book’) for the purpose of determining the impairment allowance? If not, what requirement would you propose and why?See our responses to questions 6 and 7. We agree with the notion of two categories, although we believe additional guidance from the Boards is necessary to clarify the definitions and accounting implications associated with each category.Question 9The boards are seeking comment with respect to the minimum allowance amount (floor) that would be required under this model. Specifically, on the following issues:(a) Do you agree with the proposal to require a floor for the impairment allowance related to the ‘good book’? Why or why not?Although we do not believe it would be appropriate to avoid recognising an incurred loss, we do not believe that it would be appropriate to recognise an impairment loss on the initial recognition of a financial asset. While it is not entirely clear, the proposals suggest that this could arise in practice.Consequently, we request the Boards to clarify their intention as to when losses would be regarded as ‘occurring’, as contemplated in paragraph 2(a)(ii) of the SD. In addition to the question of the approach at initial recognition, there is often an extended period of time between when a loan is deemed nonperforming, doubtful or otherwise problematic, to the time of ultimate recovery, which could easily be 12 months or more. It is not clear in the SD whether the loss estimate for the foreseeable future should be determined by reference to the time when it first becomes problematic or when the actual loss is expected to be realised, for instance after a period of 12 months or later.
The concept of a foreseeable future period may also lead to unintended consequences. For instance, more sophisticated entities are likely to have models with greater predictive reliability. Therefore, do the Boards intend for these entities to record greater losses than less sophisticated entities simply because they can better predict the future? Similarly, some entities might choose a 12 month ‘look forward’ period if the Boards adopt it as the required minimum. Is it intended that an entity that could look forward for an additional period, but chooses not to, should be capable of recording a smaller impairment loss than an identical entity that decides to build a longer period into its calculation?
It is also possible, and probably likely, that the foreseeable future will differ by type of portfolio, for instance, a credit card portfolio with ample historical metrics compared to a pool of less homogenous commercial loans. These factors may contribute to complexity and affect comparability.
(b) Alternatively, do you believe that an entity should be required to invoke a floor for the impairment allowance related to the ‘good book’ only in circumstances in which there is evidence of an early loss pattern?
Within the parameters of the Boards’ ‘higher of’ model, in the event that the proposals in the SD are taken forward we believe the floor should be applied consistently, regardless of whether there is evidence of an early loss pattern.
(c) If you agree with a proposed minimum allowance amount, do you further agree that it should be determined on the basis of losses expected to occur within the foreseeable future (and no less than twelve months)? Why or why not? If you disagree, how would you prefer the minimum allowance to be determined and why?
As stated in our response to question 9(a), we recommend that the Boards provide additional guidance to illustrate the intended application of this guidance. It would be helpful for that guidance to contemplate swings in the credit cycles and dynamic economic conditions, such as the recent credit crisis, as it relates to gauging the ‘foreseeable future’. Specifically, volatile market conditions can render historic data less relevant for purposes of future projections. This could lead to the counterintuitive result that, in deteriorating conditions, the foreseeable future might be a shorter period of time than in normal conditions. In turn, this may potentially reduce the overall impairment loss allowance that is recorded for the good book.
We do accept that, under the proposals as drafted, a specified minimum period (whether 12 months or some other period of time) is necessary. However, we find it difficult to articulate the choice of a period of 12 months in the context of a clear principle. As such, it might be helpful for the Boards to describe the minimum time period in practical terms or perhaps as an abuse-prevention mechanism in the final basis for conclusions.
(d) For the foreseeable future, would the period considered in developing the expected loss estimate change on the basis of changes in economic conditions?
We believe it could change. See our response to question 9(c).
(e) Do you believe that the foreseeable future period (for purposes of a credit impairment model) is typically a period greater than twelve months? Why or why not? Please provide data to support your response, including details of particular portfolios for which you believe this will be the case.
We recommend that the Boards carry out field-testing in this regard.
(f) If you agree that the foreseeable future is typically a period greater than twelve months, in order to facilitate comparability, do you believe that a ‘ceiling’ should be established for determining the amount of credit impairment to be recognised under the ‘floor’ requirement (for example, no more than three years after an entity’s reporting date)? If so, please provide data and/or reasons to support your response.
We believe a ‘ceiling’ should be defined only in terms of the reliability of the information available to develop the estimate. The Boards may wish to consider providing indications of when such information is not considered ‘reasonable and supportable’ as described in paragraph B5.
Question 10
Do you believe that the floor will typically be equal to or higher than the amount calculated in accordance with paragraph 2(a)(i)? Please provide data and/or reasons to support your response, including details of particular portfolios for which you believe this will be the case.
We recommend that the Boards carry out field-testing in this regard.
Question 11
The boards are seeking comment with respect to the flexibility related to using discounted amounts. Specifically, on the following issues:
(a) Do you agree with the flexibility permitted to use either a discounted or undiscounted estimate when applying the approach described in paragraph B8(a)? Why or why not?
Without measurement guidance, we are unable to form a definitive view on this question. However, from a conceptual standpoint, most loans represent the present value of the future cash flows to be collected (principal and interest). This suggests the use of discounting is appropriate, unless undiscounted amounts are materially the same.
(b) Do you agree with permitting flexibility in the selection of a discount rate when using a discounted expected loss amount? Why or why not?
We do not believe discounting should be elective, although we agree flexibility is needed to make the SD operational. Therefore, we would suggest the following hierarchy
I. Initially, the effective interest rate should be used if it is available without undue cost and effort. The Boards may also wish to consider an “impracticable” threshold as that term is used in ASC 825-10-50-17 or as defined in IAS 8 Accounting Policies, Changes in Accounting Estimates and Errors.
II. Otherwise, the risk free rate should be used based on the closest approximation of the portfolio’s maturity profile.
III. Lastly, straight-line recognition could be used if not materially different than the effective or risk free rates.
We believe this hierarchy would promote greater comparability at an acceptable cost.
Question 12
Would you prefer the IASB approach for open portfolios of financial assets measured at amortised cost to the common proposal in this document? Why or why not? If you would not prefer this specific IASB approach, do you prefer the general concept of the IASB approach (ie to recognise expected credit losses over the life of the assets)? Why or why not?
Please see our covering letter.
Question 13
Would you prefer the FASB approach for assets in the scope of this document to the common proposal in this document? Why or why not? If you would not prefer this specific FASB approach, do you prefer the general concept of this FASB approach (i.e. to recognise currently credit losses expected to occur in the foreseeable future)? Why or why not?
Please see our covering letter.
Appendix B – IASB questions
Question 14Z
Do you agree that the determination of the effective interest rate should be separate from the consideration of expected losses, as opposed to the original IASB proposal, which incorporated expected credit losses in the calculation of the effective interest rate? Why or why not?
From a conceptual perspective, we believe that the determination of the effective interest rate should be combined with the consideration of expected losses. However, we acknowledge the concerns raised by constituents about the operationality of the proposal to combine these features and, in consequence, from a practical perspective agree that it is appropriate to decouple the calculation of the effective interest rate and the adjustment for expected losses.
Question 15Z
Should all loan commitments that are not accounted for at fair value through profit or loss (whether within the scope of IAS 39 and IFRS 9 or IAS 37) be subject to the impairment requirements proposed in the supplementary document? Why or why not?
We are not wholly convinced that this would be an appropriate approach. We note that it is acknowledged that loan commitments and loans are often managed using the same business model and information systems. A natural consequence of this is that it would appear appropriate to follow a consistent approach to impairment.
However, there is also the question of what this would be intended to achieve, in particular for those entities outside the financial sector. It is possible that any adverse change in market conditions (from the lender’s perspective) between the date of issue of the loan commitment and the issue of the loan itself would result in the recording of an impairment (with the amount of the impairment representing the fair value of the difference between the committed interest rate and the market interest rate). This would appear to result in remeasurement being required for all loan commitments, regardless of whether they are required to be measured at fair value through profit or loss. This is in contrast to current practice, which would typically be to record an adjustment only when the loan commitment became onerous.
Question 16Z
Would the proposed requirements be operational if applied to loan commitments and financial guarantee contracts? Why or why not?
We see no reason why the proposed requirements would not be operational, as it would appear possible for the necessary calculations to be carried out. However, as noted above, we would question the rationale for the expansion in scope for the remeasurement of loan commitments and financial guarantee contracts.
Question 17Z
Do you agree with the proposed presentation requirements? If not, what presentation would you prefer and why?
We agree with the proposals.
Question 18Z
a) Do you agree with the proposed disclosure requirements? If not, which disclosure requirements do you disagree with and why?
b) What other disclosures would you prefer (whether in addition to or instead of the proposed disclosures) for the proposed impairment model and why?
We agree in principle with the proposed disclosure requirements. However, in order to avoid excessive disclosures, it may be appropriate to require the tabular disclosures illustrated by paragraph BZ22 for the classes of assets in their entirety, with individual tables being required only for individually significant classes of assets.
We also suggest that any additional proposed disclosures are considered in the light of those already required by IFRS 7, in order to ensure that any new requirements are substantively different, and to eliminate any existing disclosure requirements that are no longer necessary.
Question 19Z
Do you agree with the proposal to transfer an amount of the related allowance reflecting the age of the financial asset when transferring financial assets between the two groups? Why or why not? If not, would you instead prefer to transfer all or none of the expected credit loss of the financial asset?
We agree with the proposal, which would apply to those assets that were being transferred from the ‘good book’ to the ‘bad book’. We assume that the transfer would involve a two step approach, under which the age proportionate allowance would first be transferred with the related asset and then the allowance would be increased to the entire amount of expected credit losses.
ED Financial Instruments Impairment.pdf
BDO Comment Letter 2011 04.pdf