This Comment Letter was sent by Brussels Worldwide Services BVBA to the International Accounting Standards Board on 30 June 2010, on behalf of BDO International.
Dear Sir
IASB Exposure Draft ED/2009/12: Financial Instruments: Amortised Cost and Impairment
We are pleased to comment, on the above exposure draft (the ED) issued by the International Accounting Standards Board (IASB), on behalf of BDO International.
We are supportive of the IASB’s project to replace IAS 39 Financial Instruments: Recognition and Measurement with an accounting standard that enhances the quality of information included in financial statements and reduces the complexity of financial reporting for financial instruments. As part of this, we support the development of new guidance for amortised cost and impairment, provided that this guidance sets out clear measurement objectives that will lead to improvements in the consistency of financial reporting, and is operational for all entities to apply.
However, having considered the proposals set out in the ED, we believe that they should not be taken forward in their current form. We have set out a number of principal reasons below, with our responses to specific questions being included in the attached Appendix.
The proposed guidance appears to have been drafted with larger financial institutions in mind, rather than the full range of entities that report using IFRS. While the global financial crisis has, of course, focussed attention on larger financial institutions and improvements that might be made to their financial reporting, it is also right that guidance proposed for all IFRS consituents should take into account the circumstances of that wide range of entities.
It would appear that, while the proposals might be capable of being applied to a large group of homogenous items by a sophisticated preparer of financial statements, for many entities, and for individual balances, the proposals are not operational. We note that the IASB formed an Expert Advisory Panel (EAP) as part of its outreach activities in relation to the ED, and that the EAP raised concerns at its April 2010 meeting about whether a simpler model that might be capable of being applied by smaller banks would even be a close approximation of, and consistent with, the underlying principles of the IASB’s proposals.
While in principle an expected loss model would appear to have certain superior attributes in comparison to the present incurred loss model in IAS 39 (such as the apparent reduction in the initial profit margin on loan receivables during the period up to the point at which an incurred loss event takes place), we are not convinced that an expected loss model would have done anything to identify the onset of the global financial crisis at an earlier stage, or reduce the extent of the impairment losses that were reported after the onset of the global financial crisis. In fact, if anything it would appear that an expcted loss model would have resulted in an increased level of provisioning in comparison to the current incurred loss model, due to the sharp downward revisions of expectations at the point at which the crisis took effect. For example, in the first half of 2007 when loans were being advanced at what can be seen with the benefit of hindsight to have been uneconomic rates, it is doubtful that very much would have been provided for expected losses due to the overly optimistic views taken at that time. Consequently, unless the expected loss model can clearly be demonstrated to bring a significant enhancement in financial reporting which would have assisted in identifying the onset of the global financial crisis at an earlier stage, we do not see the rationale for its introduction.
We note, and agree with, the alternative views set out at the end of the Basis for Conclusions. In particular, we agree that there is a significant risk of earnings management under an expected loss model, that this model would be extremely difficult to audit satisfactorily, and that it may, as part of the overall project dealing with impairment, be appropriate for the Board to revisit the incurred loss model and consider whether this can be refined such that more realistic provisions than seem to have been made in the past would be required.
We also note the acknowledgement in the ED, and comments made by other parties, that the proposed changes would take a very significant period to implement, with three years often being mentioned. Again, and linked to the paragraphs above, in the absence of clear evidence that the new model would have been significantly better in identifying the timing of the onset of the global financial crisis and that the new model can be made operational for the wide range of entities that report under IFRS, it is difficult to see why this model should be adopted.
On a less significant point, we note that the definition of ‘non-performing’ includes a financial asset that is more that 90 days past due. While this does not, on the face of it, seem unreasonable it is also the case that in some jurisdictions it is common for balances to be received very late, meaning that a 90 day threshold may not be meaningful.
We acknowledge and understand that the Board has come under significant pressure from a wide range of constituents and other organisations to revisit the current impairment model in IAS 39 for financial assets measured at amortised cost, and that the incurred loss model has been criticised for having delayed the recognition of losses in comparison to other approaches. We accept that it may be appropriate for the Board to acknowledge the work that is being carried out by the Basel Committee, although it is also the case that not all entities that report in accordance with IFRS are subject to Basel requirements (including some entities in the financial sector). However, it is important that any changes introduced are demonstrated to be superior to the existing requirements, particularly if (as would be the case with an expected loss model) those changes would result in significant work being required to implement them and require systems changes.
Finally, we should note that we are wholly opposed to the introduction into IFRS of any impairment model that bases its provisioning on what is often termed a ‘dynamic’ or ‘through the cycle’ approach with (for example) losses being recorded for loans which have not yet been advanced. While such an approach may be appropriate for prudential regulators, it is not appropriate for financial statements that are intended for investors and lenders.
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
Brussels Worldwide Services BVBA
Appendix
Question 1
Is the description of the objective of amortised cost measurement in the exposure draft clear? If not, how would you describe the objective and why?
We agree that the description of the objective of amortised cost measurement is clear.
We note that paragraphs 1 and 2 of the ED should include a reference to financial assets within the scope of IFRS 9 Financial Instruments.
Question 2
Do you believe that the objective of amortised cost set out in the exposure draft is appropriate for the measurement category? If not, why? What objective would you propose and why?
We agree that, as set out in the ED, the objective of amortised cost is appropriate for the measurement category.
However, while we agree that the concept of ‘effective return’ embodied in the objectives is relevant for financial institutions, it may be less so for non-financial institutions. Providing information about the effective return of a financial asset assumes that the main purpose for holding the financial asset is to earn a return from it. While this may be true for financial institutions, the same cannot be said for non-financial institutions. For many entities, the most significant financial assets will be trade receivables which usually result from the process of extending short-term credit in order to sell goods or services. Trade receivables are not held to generate interest revenue and the impairment costs associated with such receivables are seen as a business expense.
We suggest that the IASB considers redrafting the objectives in paragraphs 3 to 5 of the ED to relate the objective of amortised cost to providing information about the effect return where such information is ‘relevant’.
Question 3
Do you agree with the way that the exposure draft is drafted, which emphasises measurement principles accompanied by application guidance but which does not include implementation guidance or illustrative examples? If not, why? How would you prefer the standard to be drafted instead and why?
We do not agree with the exclusion of implementation guidance or illustrative examples from the proposed standard. While we acknowledge that a common criticism of IAS 39 is that it is difficult to understand, apply and interpret, we do not believe this criticism is due to the implementation guidance or illustrative examples included in IAS 39. This criticism is instead due to complexity and inconsistencies that arise from certain aspects of IAS 39, such the number of categories of financial assets and financial liabilities, the derecognition guidance on financial assets, and hedge accounting. Therefore, we are not convinced that the IASB has sufficiently documented in its basis for conclusions its reasons for excluding implementation guidance or illustrative examples from the proposed standard.
In addition, as noted in paragraphs IN7, and BC28 - BC29 in the basis of conclusions, many respondents to the ‘Request for Information’ expressed concern over the operational challenges in moving to an expected loss model. Although the IASB has published some IASB staff examples on the application of the expected loss model, preparers of IFRS financial statements would benefit from the inclusion of these and/or other numerical examples in any final standard as illustrative examples.
Therefore we recommend that the IASB reconsider this issue and include implementation guidance and illustrative examples in the final standard.
Question 4
(a) Do you agree with the measurement principles set out in the exposure draft? If not, which of measurement principles do you disagree with and why?
(b) Are there any other measurement principles that should be added? If so, what are they and why should they be added?
We generally agree with the measurement principles for amortised cost set out in the exposure draft, except for the approach proposed for the incorporation of expected credit losses for the reasons set out in our covering letter.
We note that the exclusion of an entity’s own non-performance risk in estimating expected cash flows of financial liabilities is not mentioned until paragraph B3 of Appendix B. As this is fundamental to the amortised cost measurement of financial liabilities, we suggest greater emphasis of this point is included in the measurement principles section of any final IFRS. Alternatively, specific reference could be made from the measurements principles to application guidance in Appendix B.
We do not consider that other measurement principles should be added.
Question 5
(a) Is the description of the objective of presentation and disclosure in relation to financial instruments measured at amortised cost in the exposure draft clear? If not, how would you describe the objective and why?
(b) Do you believe that the objective of presentation and disclosure in relation to financial instruments measured at amortised cost set out in the exposure draft is appropriate? In not, why? What objective would you propose and why?
The objective of presentation and disclosure is not entirely clear because the objective is not clearly linked to the concept of ‘effective return’ in the measurement objective. We also have concerns about the extent to which the presentation and disclosure objective is appropriate for trade receivables held by non-financial institutions as the purpose for holding such financial assets is not to earn interest revenue (as noted in our response to question 2 above).
We suggest that paragraph 11 of the ED is redrafted to incorporate the effective return concept set out in the amortised cost measurement objective and to clarify how the objective is relevant to short-term trade receivables held by non-financial institutions.
Question 6
Do you agree with the proposed presentation requirements? If not, why? What presentation would you prefer instead and why?
We generally agree with the presentation requirements proposed in paragraph 13 of the ED for financial institutions. However, we are not convinced that they are wholly appropriate for entities in other industry sectors.
We suggest that paragraph 13 is redrafted to clarify that this presentation is required for financial institutions and for any financial instruments where such disclosures would provide useful and relevant information to the users of financial statements. In addition, simplified presentation requirements (or practical expedients) should be added for short-term trade receivables and for non-financial entities.
Question 7
(a) Do you agree with the proposed disclosure requirements? If not, what disclosure requirements do you disagree with and why?
(b) What other disclosure would you prefer (whether in addition to or instead of the proposed disclosures) and why?
We are concerned that the disclosure requirements appear to have been drafted in the context of financial institutions. The nature and level of certain of the proposed disclosures may not provide useful and relevant information to users of financial statements of entities that are not financial institutions and whose core business is not the provision of financial services. In fact, such disclosure may distract the attention of users from the core business of the reporting entity.
We recommend that the Board considers reducing the level of disclosure for non-financial institutions by redrafting the disclosures into two sections, being one which includes disclosures for all entities and another setting out additional disclosures required for financial institutions and entities whose core business is the provision of financial services.
We also disagree with the proposed requirement for stress testing disclosures set out in paragraph 20. It is possible that some financial institutions would not wish to disclose this type of information, which might then lead to the risk that a requirement would discourage formal stress testing as the disclosure would then be avoided. We note that there can also be practical issues, and that the FASB removed their original draft requirement for such disclosures. We suggest that the IASB might instead encourage (rather than require) stress testing disclosures and provide examples to indicate how it is anticipated this disclosure might be made.
Similarly, it would be helpful for examples to be included of the type of disclosure that is envisaged by paragraphs 19 and 21 b).
We note from page 16 of the ED that the Board might treat the presentation and disclosure requirements as amendments to IAS 1 Presentation of Financial Statements and IFRS 7 Financial Instruments: Disclosures respectively. We believe that this approach should be followed, and that it would be inappropriate for the presentation and disclosure requirements to be set out as part of the new IFRS.
Question 8
Would a mandatory effective date of about three years after the date of issue of the IFRS allow sufficient lead-time for implementing the proposed requirements? If not, what would be an appropriate lead-time and why?
We generally agree that an effective date of about three years after the date of issuance is sufficient lead-time for implementing the requirements, and consider that it is likely to be appropriate for the effective date to be consistent with the effective date of IFRS 9 Financial Instruments. If the Board considers that a different effective date is appropriate, with phased implementation being required or permitted, this should be covered in the basis for conclusions in any final IFRS based on the ED.
Question 9
(a) Do you agree with the proposed transition requirements? If not, why? What transition approach would you propose instead and why?
(b) Would you prefer the alternative approach (described above in the summary of the transition requirements)? If so, why?
(C) Do you agree that comparative information should be restated to reflect the proposed measurement requirements? If not, what would you prefer instead and why? If you believe that the requirements to restate comparative information would affect the lead-time (see Question 8) please describe why and to what extent.
We do not agree with the proposed transition requirements.
We generally support full retrospective application where possible as it provides the most useful information to users of the financial statements. We note from paragraph BC71 (a) in the basis of the conclusions that the Board considered an approach that would have allowed an entity to choose retrospective application if the required information was available without the use of hindsight. We believe that this option should be included in the final standard.
We also have some concern with the use of ‘all available historical data’ in the effective interest rate adjustment set out in paragraphs 25 and 26 of the ED. We are not convinced that it is possible to gather and reconstruct ‘all available historical data’ without applying hindsight given the application guidance in paragraph B30 which suggests approaches that could be taken to include the use of hindsight. Furthermore the cost to gather such information could prove to be significant for smaller entities without sophisticated accounting and data collection and retention systems.
Therefore we suggest the Board considers revising paragraph 26 to change ‘all available historical data’ to ‘all reasonable and relevant historical data that can be obtained without undue cost and effort’. We also suggest paragraphs 26 and B30 be revised to clarify that the use of hindsight is not permitted.
It is not clear from the ED whether the ‘alternative approach’ described in the summary is the ‘customised transition approach’ discussed in paragraph BC71 of the basis of conclusions. We are not convinced that the Board has provided sufficient information in the basis for conclusions to explain its deliberations over the ‘alternative approach’ and to why this approach was considered inferior to the proposed transitional approach in the ED.
We suggest the Board considers including numerical examples of the alternative approach and the proposed transitional approach in the basis of conclusions to help preparers better understand the computations involved and the differences that result.
We agree that comparative information should be restated to reflect the proposed measurement requirements.
Question 10
Do you agree with the proposed disclosure requirements in relation to transition? If not, what would you propose instead and why?
We agree with the proposed disclosure requirements in relation to transition in the ED.
Question 11
Do you agree with the proposed guidance on practical expedients is appropriate? If not, why? What would you propose instead and why?
We do not agree that the proposed guidance on the practical expedients is appropriate. In the summary on pages 12 and 13 of the ED, it appears the Board’s intention for including practical expedients is to facilitate cost-effect ways of determining amortised cost in certain situations. However, other than a brief explanation in paragraph BC 38(d), there is no significant discussion in the basis for conclusions to support how the Board came to this conclusion.
Consequently we suggest that the Board includes some additional commentary in the basis for conclusions to better explain its reasons for the inclusion of practical expedients in the final standard.
Also we are not convinced that the practical expedients will achieve the Board’s intended purpose of being cost-effective. Paragraph B15 of the ED indicates that these expedients may only be used if the overall effect is immaterial. Thus it would seem that for a reporting entity to substantiate that the application of a practical expedient is immaterial, it would need to calculate amortised cost according to the measurement principles and compare the results using the practical expedient. In addition, paragraph B17 provides an example whereby an entity might determine amortised cost using two separate present value calculations. The information necessary to perform such calculations could prove to be costly to obtain for smaller and/or less sophisticated entities.
Therefore, we suggest the Board revises the guidance in paragraphs B15 to B17 for the use of practical expedients in order that its intended purpose of cost-effectiveness can be achieved.
Question 12
Do you believe additional guidance on practical expedients should be provided? If so, what guidance would you propose and why? How closely do you think any additional practical expedients would approximate the outcome that would result from the proposed requirements, and what is the basis for your assessment?
Yes, we believe that additional guidance should be included.
As noted in our responses to certain of the questions above, one of our main concerns with the proposals in the ED is that many of requirements appear to have been drafted in the context of large financial institutions. The nature and level of many of the proposals may not be wholly relevant to entities that are not financial institutions and whose core business is not the provision of financial services.
Therefore we recommend the Board considers simplifying the presentation and disclosure requirements for non-financial institutions in order that these provide relevant information to the users of the financial statements of such entities.
ED-2009-12 Financial Instruments Amortised Cost and Impairment.pdf
BDO Comment Letter 2010 04.pdf