International

​This Comment Letter was sent by BDO Global Coordination B.V. on behalf of BDO International, to the International Accounting Standards Board on 14 September 2009:

Dear Sir

Exposure Draft ED/2009/07 Financial Instruments: Classification and Measurement

We are pleased to comment on the above exposure draft (the ED) issued by the International Accounting Standards Board (IASB), on behalf of BDO International¹. Our detailed comments are set out in the attached Appendix.

We strongly support the IASB in its project to replace IAS 39, and agree that an appropriate mechanism for the classification and measurement of financial instruments is a combination of an entity’s business model and the characteristics of the financial instruments themselves (with certain characteristics of an instrument being capable of overriding the classification that would be derived solely from a business model approach). However, we believe that it would be helpful to clarify further the approach to be taken, and to make amendments to the draft guidance in a number of areas, to assist in determining whether financial instruments have basic loan features and whether they are managed on a contractual yield basis. We have suggested a number of changes in response to the IASB’s detailed questions.

We also believe that it is important that the IASB, in each phase of its project to replace IAS 39, should seek to minimise the risk that new requirements that are implemented might need to be revisited in the relatively short term. Revisiting might be needed either because there are other projects where the outcome will be relevant or linked to the financial instruments issue(s), or because additional considerations would be identified through a fuller debate. We have set out below approaches to certain aspects of the classification and measurement phase, and to the accounting for investments in equity instruments, that we believe the IASB should consider adopting.

Separate consideration of financial assets and financial liabilities

The ED proposes to align the approach for the classification and measurement of both financial assets and financial liabilities. While we can see the attraction of a symmetrical approach for both sides of the balance sheet, we note that this does not currently arise under the existing guidance in IAS 39. In our view, it is not essential to have this symmetry in the immediate term, and we do not consider that a replacement of the current IAS 39 classification and measurement guidance for both financial assets and financial liabilities is critical for December 2009 financial year end financial statements.

The IASB’s project to replace IAS 39 has been accelerated significantly by the global financial crisis. However, the focus of criticism of financial instrument accounting has been mainly on financial assets, with less emphasis on financial liabilities. We note that the IASB has a number of concurrent projects, including the effect of an entity’s own credit rating on the fair value measurement of its own debt. We also note that financial liabilities are often different in nature to an entity’s financial assets, in that an entity might often have long term structured debt, with financial assets being managed on a shorter term basis. It may be that additional time to consider all aspects of accounting for financial liabilities would be helpful before concluding on the appropriate classification and measurement boundaries for those instruments.

We therefore suggest that the IASB’s phased approach to the replacement of IAS 39 is split further, with the classification and measurement phase being divided into one for financial assets and another for financial liabilities. This would involve retaining the existing guidance for embedded derivatives for financial liabilities pending the completion of that phase of the project.

This would enable the IASB to complete the phase dealing with financial assets in time for adoption in December 2009 financial statements (and therefore meet the calls from a number of constituents for certain key aspects of financial instrument accounting to be dealt with by the end of this year), while allowing full and appropriate consideration to be given to the accounting for financial liabilities. A number of entities have long term structured debt with one or more significant embedded derivatives where the embedded feature(s) are currently accounted for at fair value and the host contract at amortised cost. These instruments would often, under the proposals as drafted, need to be measured at fair value in their entirety. While this may be an appropriate approach in some cases, it does again link to the question of the effect of an entity’s own credit rating on the fair value measurement of its debt. We believe that it would be helpful for fuller consideration and debate to be had on this point in particular before concluding on the appropriate approach.

Accounting for investments in equity instruments

The IASB has proposed an option for entities to account for investments in equity investments as at Fair Value Through Other Comprehensive Income (FVTOCI), with all dividend income also being recognised in OCI and no recycling being permitted. In our detailed response to question 10 we have suggested that, should the IASB proceed with this proposal, it would be appropriate for dividend income to be recognised in Comprehensive Income.

However, we believe that a number of associated issues arise with both the IASB’s proposals and the recognition of dividend income in Comprehensive Income. In particular, we believe that the IASB’s proposals pre-empt the outcome of the ongoing performance reporting project, which we believe would be unsatisfactory. There is also the question of determining whether a dividend payment received represents, in reality, a capital repayment which should properly be recorded within OCI. It is possible that the resolution of this issue would necessarily result in the implementation of a somewhat arbitrary and rules based approach, which we consider would be unsatisfactory.

We also note that phase 2 of the IASB’s insurance project is likely to bring changes to insurance accounting that are likely to link to the appropriate accounting approach for investments in equity instruments.

In consequence, we suggest that the IASB should retain the existing Available for Sale category for investments in equity investments, together with the associated impairment guidance, pending wider consideration of accounting for investments in equity instruments in the context of the IASB’s other related projects.

Special Purpose Entities and waterfall arrangements

We understand that the IASB does not intend that structured arrangements should be ‘looked through’ in determining the appropriate accounting approach to be followed. As noted in our detailed comments to question 4(b), we disagree with the IASB’s view that the most senior tranche of structured debt issued by an SPE should be the only tranche that might qualify as having basic loan features, with all subordinated tranches being accounted for at fair value, and believe that there are structuring opportunities under the approach proposed in the ED which could give rise to inappropriate measurement (whether this is at amortised cost or at fair value).

Although it would inevitably bring some additional complexity, we believe that it is essential that a ‘look through’ approach is followed in determining the appropriate classification and measurement by investors in tranches of loan notes issued by an SPE. We believe that there are some structures where even the most senior tranche should fail the basic loan features test, while in others at least some of the junior tranches should meet the basic loan features test and therefore be capable of being measured at amortised cost.

We hope that our comments and suggestions are helpful; as noted above, we are highly supportive of the IASB in this project and would be pleased to provide further input. Should you wish to discuss any of the points we have raised, please contact either Tracey-Lee Massey at +32 2 778 0130 or Andrew Buchanan, our Global Head of IFRS, at +44 (0)20 7893 3300.


Yours faithfully

BDO Global Coordination B.V.


Appendix

Question 1
Does amortised cost provide decision-useful information for a financial asset or financial liability that has basic loan features and is managed on a contractual yield basis? If not, why?

Yes. If a financial instrument is held primarily for the purpose of generating or incurring interest based income or expense, based on the entity’s business model, and not to generate gains or losses in the short term, then amortised cost measurement is appropriate. In this context, interest has the meaning as set out in the second sentence of paragraph B1.

However, as noted in our response to question 2 below, we believe that certain financial instruments which would, under the proposals, be required to be measured at fair value, should instead be measured at amortised cost. In particular, we disagree with the proposals for assets whose cash flows are subject to waterfall arrangements (paragraphs B7 and B8), and for assets purchased at a discount to their principal amount (paragraph B13), which would result in almost all of these assets being measured at fair value.

Question 2
Do you believe that the exposure draft proposes sufficient, operational guidance on the application of whether an instrument has ‘basic loan features’ and ‘is managed on a contractual yield basis’? If not, why? What additional guidance would you propose and why?

While we support the approach derived from the two principles that are proposed, we believe that additional guidance is needed to make them operational, and to assist in consistency of approach.

Basic loan features

We note that the proposed standard does not itself contain a principle or definition for basic loan features. However, we agree with the guidance and examples set out in paragraphs B1 to B3 and believe that it would be appropriate for certain of this guidance, in particular that set out in the first two sentences of paragraph B1, to be moved into the body of the standard itself. We also note that in its basis for conclusions (paragraph BC21), the Board has set out useful guidance and again we believe that this should be included in the standard.

In order to assist in making the proposed standard operational, we believe that the Board should provide additional guidance in the following areas:

• Linked to our comment above that the standard needs to include a clear principle or definition of basic loan features, it would be appropriate for the standard to state explicitly that instruments with basic loan features are restricted to those with market based compensation for the time value of money, and credit risk that is not leveraged. The associated question of what ‘leveraged’ means could be addressed through guidance which expands on the narrative currently in paragraph BC21. This might note that any feature which could cause interest cash flows to vary disproportionately in comparison to changes in an interest rate as set out in what is currently paragraph B3(a) would be regarded as being a leveraged feature, and hence preclude the instrument as being regarded as having basic loan features.
• Certain instruments should be identified as having features which do not qualify as basic loan features (for example, inverse rate instruments, such as one where the interest rate is 10% less LIBOR, and instruments where the return is linked to a commodity, such as the price of gold).
• Inflation linked instruments should not be precluded from being regarded as having basic loan features. However, the guidance should make it clear that basic loan features are restricted to only those inflation linked loans which result in the instrument having what is effectively a market based floating rate. Therefore, an instrument with an inflation linked rate could qualify as having basic loan features, but if the inflation linked feature gave rise to a leveraged return, the instrument would be required to be measured at fair value.
• It is common for interest only strips to be traded. Where these are at a fixed rate, then the instrument might qualify as having basic loan features (subject to prepayment risk as, if this was significant, the test for repayments of principal and interest would be failed in the event of prepayment of the principal element of the loan to which the interest strip relates). It would appear that variable rate interest only strips would be precluded from being regarded as having basic loan features as they are effectively derivative instruments, and the guidance should state this explicitly.
• Convertible debt, because of the conversion feature, will not qualify to be measured at amortised cost by the holder. However, from the issuer’s perspective it is not clear from the proposals whether the debt element would qualify as having basic loan features. Assuming the debt holder can choose either a cash or share settlement on maturity, we assume that the principal amount outstanding will be the amortising liability during the term of the instrument on which the effective interest rate is based (as the liability will, in addition to attracting cash interest payments, need to be accreted up to the cash settlement amount which may be payable at maturity). We assume that it is intended that this accretion would not preclude the liability component of the instrument from being regarded as having basic loan features. It is less clear whether the ability for the holder to require settlement in shares would violate the requirement currently in paragraph B1 that the instrument should give rise on specified dates to cash flows which are payments of principal and interest on the principal amount outstanding.
• It is not clear how a perpetual instrument would be classified. Because the instrument has no stated maturity date, it could be argued that there is no payment of a principal amount and therefore the instrument cannot be regarded as having basic loan features. We believe that this would be an inappropriate analysis because in substance the interest payments on a perpetual instrument represent a combination of interest and principal, and consider that the Board should clarify that where the interest payment on a perpetual instrument represents compensation for the time value of money and is not leveraged, the instrument can be regarded as having basic loan features.
• Zero coupon bonds are noted as having basic loan features. This should be extended explicitly to cover trade receivables and payables, as the pricing of those instruments compensates the vendor for the time value of money for the permitted credit period.
• Some loans have contractual cash flows which are secured on a particular asset. For example, in the investment property sector, a common structure is ‘single property SPEs’ where loan finance taken out to acquire a property is ringfenced (that is, in the event of default, there is no recourse to other assets of the group). In consequence, although the loan might have stated principal and interest payments, the recovery of these amounts is linked specifically to the rental income and fair value of the investment property. It is not clear whether the loan would be regarded as having basic loan features or instead be viewed as being leveraged due to the potential variation in return. This is a common issue at present, as the loan balance outstanding in a number of these structures currently exceeds the fair value of the investment property meaning that linkage of the recovery of the loan to the fair value of the property could be relevant at the date of initial adoption of the new standard.

Managed on a contractual yield basis (including financial assets acquired at a discount)

A significant difficulty with making the proposals operational is in determining quite what is meant by ‘managed on a contractual yield basis’ and, consequently, where the boundary between amortised cost and fair value measurement lies. For example, the proposals could be taken to mean that financial instruments would need to be managed for their entire term on a contractual yield basis (meaning that the test would not be far away from that to be followed in determining whether a financial asset can be designated as Held to Maturity under the existing IAS 39). Alternatively, the analysis might instead be that provided financial instruments are managed at least to some extent on a contractual yield basis then they could be regarded as being managed on a contractual yield basis. The latter interpretation, which we favour and believe the Board intended, results in there being a need for additional guidance in determining which instruments do and do not qualify as having met the test.

We agree that the manner in which an entity asserts that financial instruments are ‘managed on a contractual yield basis’ needs to be consistent with the way in which those instruments are managed and their performance evaluated by key management, as set out in paragraph B9. We also agree with the conclusion (and the reasons), in paragraph BC35 that there should be no requirement for a documented risk management or investment strategy. However, we consider that there should be a requirement to disclose, in the financial statements, the strategy and policies followed in determining whether financial instruments are managed on a contractual yield basis. This requirement would assist users of financial statements in their understanding of how the reporting entity manages its financial instruments, and would also link to the requirement included in the proposals to disclose the gains and losses generated from the sale or disposal of financial instruments measured at amortised cost before their contractual maturity. For the same reason, any changes to an entity’s strategy and policies should also need to be disclosed.

The disclosed strategy and policies should also include the circumstances in which financial assets might be sold, and financial liabilities settled, where those financial assets and liabilities are measured at amortised cost. This would link to the fact that many entities hold a ‘liquidity reserve’ where surplus funds are invested in debt instruments, which may often be realised before their contractual maturities to meet liquidity requirements. In our view, provided such investments are held to generate an investment return, if they need (or are even planned) to be converted at short notice into cash (the amount of which can be reliably predicted), this should not violate the assertion that they are held on a contractual yield basis because the focus is on the generation of cash flows and not the realisation of holding gains. However, where the purpose of holding a portfolio of such investments is to buy and sell them in response to changes in value in order to crystallise holding gains, or there is a pattern of doing so in practice, these should not be viewed as being managed on a contractual yield basis (this would again link to the stated strategy and policies for holding the instruments).

The Board might consider different terminology for this test, which might more clearly permit the sale/settlement of financial instruments before their contractual maturity. This would also recognise that, for example, trade receivables are not generally managed on a contractual yield basis, with the focus instead being more on cash collection. We suggest the following guidance might be used as a substitute

Managed for the collection of contractual cash flows

Definition
A financial instrument that is held, in accordance with an entity’s strategy and policies which are evaluated by the entity’s key management personnel (as defined in IAS 24 Related Party Disclosures), for the purpose of the collection of contractual cash flows that are generated when the financial instrument is held or issued. These contractual cash flows include any adjustment for accrued or prepaid interest on initial recognition, and any similar adjustment on the settlement/disposal of the financial instrument. A financial instrument that is managed for the collection of contractual cash flows may, but is not required, to be held to its contractual maturity.

Application guidance might include the following:

A financial instrument is regarded as being managed for the collection of contractual cash flows, even if it is settled or disposed of prior to its contractual maturity, if the primary purpose of that settlement or disposal is to realise the principal amount for the purposes of cash flow management.

Financial instruments are not regarded as being managed for the collection of contractual cash flows if an entity has a history of short term purchases and sales of financial instruments, where the proceeds of sale are immediately reinvested in different financial instruments, as this indicates that the entity is making short term investments with the intention of generating short term gains.

An entity which settles or disposes of a financial instrument prior to its contractual maturity in response to unexpected changes in market rates of interest, exchange rates, or other similar factors is not precluded from including that financial instrument in a portfolio which it regards as being managed for the collection of contractual cash flows. In such cases, judgement is required in determining whether it is appropriate for financial instruments to be regarded as being managed for the collection of contractual cash flows. This will depend on considerations which include, but are not limited to, the frequency of such ‘opportunistic’ settlements or disposals and the purpose of the early settlement or disposal, including the use to which the proceeds are put.
Financial assets acquired at a discount

We disagree with the proposal in paragraph B13(b) that a financial asset that is acquired at a discount that reflects incurred credit losses cannot be regarded as being managed on a contractual yield basis. Where a financial asset is acquired at a discount, the implied effective interest rate simply reflects the higher yield that a lender would require at that point if funds were to be lent to the borrower.

It is possible that a purchaser of a portfolio of financial assets could structure arrangements in order to achieve amortised cost measurement by, instead of acquiring the existing financial assets, arranging for contractual settlement of the existing amount owed by each of the debtors, with new loan balances being originated at that point. The newly originated loans to the third parties might then, depending on their terms, qualify for amortised cost measurement.

We also believe that the proposal would not be operational in practice. For example, in the acquisition of a bank it would be expected that the carrying value of certain of the loan asset balances acquired as part of the business combination would reflect incurred credit losses. The proposals would appear to require the portfolio of loan balances to be split between those where the carrying value reflect, and do not reflect, incurred credit losses. The former might qualify for amortised cost measurement, whereas the latter would be required to be measured at fair value, despite the potential for the business model for both portfolios to be the collection of contractual cash flows. The same considerations would apply to any business combination involving the acquisition of an entity which had trade receivables, where the carrying value of some of these balances had been reduced for incurred credit losses.

Question 3
Do you believe that other conditions would be more appropriate to identify which financial assets or financial liabilities should be measured at amortised cost? If so,

(a) what alternative conditions would you propose? Why are those conditions more appropriate?

b) if additional financial assets or financial liabilities would be measured at amortised cost using those conditions, what are those additional financial assets or financial liabilities? Why does measurement at amortised cost result in information that is more decision-useful than measurement at fair value?

(c) if financial assets or financial liabilities that the exposure draft would measure at amortised cost do not meet your proposed conditions, do you think that those financial assets or financial liabilities should be measured at fair value? If not, what measurement attribute is appropriate and why?

Other than as outlined in our responses to questions 1 and 2 above, we do not believe that other conditions would be more appropriate.

Question 4

(a) Do you agree that the embedded derivative requirements for a hybrid contract with a financial host should be eliminated? If not, please describe any alternative proposal and explain how it simplifies the accounting requirements and how it would improve the decision-usefulness of information about hybrid contracts.

We agree that the embedded derivative requirements set out in IAS 39 should, ultimately, be eliminated. However, as noted in our covering letter, we believe that it would be appropriate to retain the guidance for embedded derivatives for financial liabilities, pending the outcome of certain of the Board’s other projects, in particular in respect of the fair value measurement of an entity’s own debt.

We note that the existing guidance for embedded derivatives would continue to apply to contracts with non-financial host contracts, and with financial host contracts which are outside the scope of IAS 39. Consequently, the current requirements would continue to be applicable to leasing contracts. Given the inconsistency that currently exists between IAS 17 and the embedded derivative guidance in IAS 39, where contingent rentals are recognised in the period in which they are incurred rather than being treated, where appropriate, as a separable embedded derivative, we suggest that the Board considers eliminating the applicability of the embedded derivative guidance to lease contracts as part of its project on leases. We also note that in some cases there is little substantive difference between a finance lease asset or liability and a loan receivable/payable, and that the elimination of embedded derivative guidance for loans but not lease contracts gives rise to inconsistency in approach.

We also note that the guidance for embedded derivatives would continue to apply to insurance contracts, and suggest that this is eliminated as part of the proposals for phase 2 of the Board’s insurance project.

(b) Do you agree with the proposed application of the proposed classification approach to contractually subordinated interests (ie tranches)? If not, what approach would you propose for such contractually subordinated interests? How is that approach consistent with the proposed classification approach? How would that approach simplify the accounting requirements and improve the decision-usefulness of information about contractually subordinated interests?

We do not agree with the proposals for contractually subordinated interests in paragraphs B7 and B8, and consider that the proposals are inconsistent with the guidance in paragraph B6. We note that an arrangement involving a series of subordinated tranches (a ‘waterfall’ arrangement) is regarded as giving rise to a series of subordinated tranches that contain credit protection, and that this feature results in the subordinated tranches failing the basic loan features test. However, paragraph B6 notes that although an instrument might be subordinated to another, that instrument may still have basic loan features.

It is our understanding that the Board does not intend any ‘look through’ of structured arrangements to be required. We believe that, while it might add a degree of complexity, ‘look through’ is essential in determining whether amortised cost or fair value is the appropriate measurement attribute for the various ‘tranches’. An alternative approach, which would require ‘look through’ would in our view be operational and would avoid the potential for structuring of arrangements to achieve a desired accounting result.

In our view, in a waterfall arrangement, each of the bond holders who have lent funds to the investing vehicle is being compensated for the time value of money and the credit risk associated with the principal amount outstanding. The credit risk increases according to the level of subordination, and therefore it would be expected that the more a bond is subordinated, the higher the (market) yield it will attract. In our view, when considering whether a loan tranche qualifies as having basic loan features, the more important question is whether the assets in which the funds are ultimately invested would themselves qualify as having basic loan features. If they do not, then we consider that the proposed approach, where the most senior tranche of notes issued by an SPE might qualify to be measured at amortised cost, could often be inappropriate.

As an example, assume SPE 1 issues tranched notes to investors for $10m. SPE 1 then purchases a $10m bond (which has basic loan features) issued by SPE 2 at an interest rate that will, if all principal and interest payments are received, result in the noteholders in SPE 1 being repaid in full. If SPE 2 invests in instruments that themselves qualify as having basic loan features, then in our view all of the tranched notes issued by SPE 1 should be analysed further to determine whether they also qualify as having basic loan features. Alternatively, if SPE 2 invests all of the funds it receives from SPE 1 in complex derivative instruments, all of the notes issued by SPE 1 (including the most senior tranche) should automatically fail the basic loan features test.

We therefore consider that the guidance should be changed to require that, in any waterfall structure, all tranches of notes issued by an SPE or other entity should be required to be measured at fair value unless it is possible to look through to the ultimate investment that is being made and to determine that all or part of the ultimate investment qualifies as having basic loan features. If this first test is passed and the ultimate investment does qualify as having basic loan features, then all tranches of notes issued should then be analysed further to determine whether they, individually, have basic loan features. This test might be linked back to include the guidance we have suggested above in our response to question 2 to assist in determining whether a financial instrument has a return which is leveraged, and consideration of the certainty of recovery of the contracted amounts of principal and interest of each tranche.

As noted above, only part of the ultimate investment might be in instruments that have basic loan features. Where the ultimate investment is a combination of instruments that do and do not have basic loan features then, to the extent that each tranche of notes is fully covered by instruments with basic loan features then those tranches should be analysed further to determine whether they have basic loan features.

For a structure comprised of instruments with and without basic loan features, in its simplest form the ultimate investment might be comprised of a floating rate loan and a floating to fixed interest rate swap with terms that perfectly match those of the loan. The existence of the derivative instrument should not automatically preclude one or more of the ultimate investment tranches from being regarded as having basic loan features. We consider that, if the ultimate investment is capable of being designated in an effective hedging relationship (as set out in IAS 39), then the combined effect of the instruments comprising the ultimate investment should be permitted to be considered to the extent that the hedging relationship is demonstrated to be effective. This approach should apply whether or not the ultimate investment is actually designated in an effective hedging relationship.

However, if it is not possible (or an entity does not carry out the necessary analysis) to determine which tranches are and are not covered by the ultimate instruments that have basic loan features (or can be designated in an effective hedging relationship), then all tranches should be required to be measured at fair value.

A similar approach should apply where the ultimate investments are comprised of, for example, 40% in value with basic loan features and 60% in value which do not have basic loan features. Where, on initial recognition, the returns received by all of the ultimate noteholders will be derived from a mixture of ultimate instruments which have, and do not have, basic loan features, all of the investment tranches should be required to be measured at fair value (subject to whether it is demonstrated that certain of the ultimate investment instruments could be designated in an effective hedging relationship). Alternatively, if one or more of the investment tranches are directly linked (and ringfenced) to instruments with basic loan features, then these tranches should qualify for further analysis to determine whether they themselves have basic loan features.

Question 5
Do you agree that entities should continue to be permitted to designate any financial asset or financial liability at fair value through profit or loss if such designation eliminates or significantly reduces an accounting mismatch? If not, why?

Yes. We suggest that, in the context of the fair value option being used (for example) as a means to account for an economic hedge of an entity’s own debt, the Board considers whether it would be appropriate to exclude the effect of changes in an entity’s own credit rating from the remeasurement of the debt (or, alternatively, to require separate presentation of those remeasurements in order to separate them from the economic hedge).

We also suggest that the Board considers defining, or providing guidance to clarify, the meaning of ‘significantly’ in order to indicate clearly how restrictive this criterion is intended to be. For example, does ‘significant’ mean ‘more than insignificant’ or is the threshold intended to be higher?

Question 6
Should the fair value option be allowed under any other circumstances? If so, under what other circumstances should it be allowed and why?

While we do not believe that it is necessary for the fair value option to be extended, it may be appropriate in the context of the potential structuring that could otherwise be undertaken in order to achieve a desired measurement approach. Under the proposals, an entity might add a marginally substantive embedded feature to a financial instrument so that the instrument fails to have basic loan features and is therefore required to be measured at fair value. Without some form of restriction on this type of structuring, the fair value option would appear to be capable of being open to any financial instrument.

We believe that it would be difficult to make operational any guidance that would be aimed at dealing with the structuring outlined above without that guidance being rules based, which we believe would be likely to set ‘bright line’ boundaries on structuring rather than eliminating it. If the Board is not opposed to an unrestricted fair value option, then we suggest that this would avoid the issue. We acknowledge that a significant reason for the fair value option being restricted was the effect of its use on the measurement of an entity’s own debt, where a deterioration in the entity’s own credit rating would give rise to a gain in the income statement. We agree that this is inappropriate. However, should the Board as part of its associated project on the measurement of an entity’s own debt consider that fair value measurement of an entity’s own debt should not include the effect of changes in its credit rating, then an unrestricted fair value option would appear feasible and operational.

Question 7
Do you agree that reclassification should be prohibited? If not, in what circumstances do you believe reclassification is appropriate and why do such reclassifications provide understandable and useful information to users of financial statements? How would you account for such reclassifications, and why?

No. We agree that an unrestricted reclassification option, between amortised cost and fair value measurement, would be inappropriate. However, linked to our support for the measurement of financial instruments with basic loan features being linked to an entity’s business model, in circumstances where it is clear that an entity’s business model has changed then we believe reclassification should be required.

We acknowledge that in some cases, changes in the measurement attribute would properly be applied only on a prospective basis. In others, existing financial instruments would be affected and the final standard should acknowledge these circumstances. For example, an entity might hold a portfolio of 30 year maturity bonds which are measured at amortised cost and, at the beginning of year 2, undergo an internal restructuring that changes the way in which is manages those bonds such that, if they were acquired after the restructuring, they would be required to be measured at fair value. Under the proposals, the entity would be precluded from remeasuring the bonds to their fair value.

We suggest that the Board permits reconsideration of the classification of financial instruments, but restricts this to circumstances in which an entity’s business model has changed. Any such change would need to be sufficient to require an associated significant change to the strategy and policies disclosed in the financial statements (see our response to question 1 above).

The Board might also consider whether, in the event that a designation at fair value eliminates or significantly reduces a measurement mismatch and the matching asset or liability is subsequently disposed of, the instrument designated at fair value should then be reclassified if an analysis shows that it would, if acquired at the point of the disposal, be required to be measured at amortised cost.

The issue of reclassification links to the Board’s derecognition project, where a new standard is expected to be issued in the first half of 2010. The potential for a financial instrument to be reclassified is relevant in the context of, for example, whether an instrument is derecognised and a new one recognised as a result of the renegotiation of terms.

Question 8
Do you believe that more decision-useful information about investments in equity instruments (and derivatives on those equity instruments) results if all such investments are measured at fair value? If not, why?

While we agree that fair value is a superior measurement attribute, we consider that this does not necessarily apply where a reliable fair value cannot be obtained. However, we consider that it is extremely rare for it to be impossible to obtain a reliable fair value.

Consequently, we suggest that fair value measurement should be mandated, subject to an impracticability exemption (as defined in IAS 8). If it is impracticable to obtain a reliable fair value, cost measurement (subject to impairment) should be required. IAS 8 sets a high threshold for impracticability, meaning that this approach should bring sufficient rigour to the analysis. Where an approach of ‘cost less provision for impairment’ measurement is followed, additional disclosures should be required to explain why it has been impracticable to obtain a reliable fair value, including the approaches which have been undertaken in order to attempt to derive a fair value.

We disagree with the implications of the Board’s analysis in paragraph BC64, that an impairment calculation is no more reliable than measuring the equity investment at fair value. This paragraph ignores the fact that the application of an impairment test would require an impairment trigger and that this would therefore not necessarily be a test that was carried out on an annual basis, unlike fair value measurement.

Question 9
Are there circumstances in which the benefits of improved decision-usefulness do not outweigh the costs of providing this information? What are those circumstances and why? In such circumstances, what impairment test would you require and why?

We are not convinced that the costs involved in complying with the requirement to measure the very small number of equity instruments affected at an unreliable fair value are outweighed by the benefits obtained, or that this would necessarily enhance the quality of information presented in financial statements.

For the extremely limited number of unlisted equity investments that would qualify for measurement at cost under our suggested approach (see our response to question 8 above), we believe that the existing impairment requirements should continue to be applied.

Question 10
Do you believe that presenting fair value changes (and dividends) for particular investments in equity instruments in other comprehensive income would improve financial reporting? If not, why?

While it is arguable whether the proposed approach would improve financial reporting, we acknowledge that this is aimed at providing a solution to those entities which do hold investments in equity instruments for long term strategic purposes.

We note that the ability in the existing IAS 39 to defer reporting unrealised gains and losses of financial assets designated as Available for Sale in the income statement brought associated complexity, in particular in determining when losses should be regarded as representing impairment and be recycled to the income statement. We agree that recycling should ultimately be eliminated, as in principle we consider that gains and losses should be recognised in the performance statements only once.

However, we are concerned that the proposed changes to accounting for investments in equity instruments pre-empt the outcome of certain of the IASB’s other projects, in particular the project covering financial statement presentation. In consequence, our preferred approach at this stage would be to retain the Available for Sale category for investments in equity instruments only, together with the associated impairment guidance. This should then be revisited in the context of the outcome of the IASB’s other projects.

In addition, we note that one of the difficulties in practice in dealing with the impairment of investments in equity instruments is that there is a prohibition on reversal of impairments through profit or loss. This prohibition has made some entities more reluctant to record impairments, as the consequence is the recording of a charge in profit or loss with no prospect of a corresponding credit if the value of the investment recovers. We do not consider that there is a strong conceptual basis for this prohibition, and believe that the Board should, in addition to retaining the Available for Sale category for investments in equity instruments, permit the reversal of impairment through profit or loss.

If the Board considers that it remains appropriate to provide an option for amounts associated with certain equity investments to be reported within other comprehensive income (OCI) as set out in the exposure draft, we suggest the following changes:

• If the option is followed, the entity should be required to present one single statement of comprehensive income (meaning that the option in IAS 1.81(b) to present separate statements of comprehensive income and of other comprehensive income would not be available);
• Consideration should be given as to whether it is appropriate for dividend income to be presented in comprehensive income rather than other comprehensive income. In some industry sectors (such as the insurance industry), entities may borrow funds in order to purchase long term investments in equity securities. Unless dividend income were permitted to be included within the non OCI section of the statement of comprehensive income, a reporting mismatch would be created as total comprehensive income would include finance expense in respect of the loan with dividend income being included within OCI.

Should dividend income be recorded in comprehensive income, it would be necessary to include guidance to be followed in determining whether a dividend represented an amount which should properly be recorded in the income statement or, alternatively, represented a substantive return of capital which should properly be recorded within other comprehensive income. In the absence of such guidance, we believe that it would be possible to structure arrangements in order to achieve a desired accounting result (for example, to arrange for what would otherwise be dealt with as a return of capital, and therefore recorded in OCI, to be legally in the form of a dividend in order that the credit is recorded in the income statement).

It is also not clear whether ‘equity instruments’ are intended to be as defined in IAS 32. Paragraph B26 would imply that this is the case, as it refers to non monetary items. Assuming this is the case, we suggest that the Board reconsiders the use of the definitions in IAS 32 when determining whether an entity holds an equity investment. We note that the use of the IAS 32 definition could give rise to reclassifications when the revised guidance for equity classification is issued and, given that it may be more restrictive that the current IAS 32, entities might be faced with reclassifying out of the FVTOCI category.

Should the Board include an option to take amounts to OCI, we believe that this should be linked to the ongoing project for Income Taxes, as backward tracing of deferred tax balances would be needed in order that there is an appropriate classification of changes in those tax balances.

Question 11
Do you agree that an entity should be permitted to present in other comprehensive income changes in the fair value (and dividends) of any investment in equity instruments (other than those that are held for trading), only if it elects to do so at initial recognition? If not,

(a) how do you propose to identify those investments for which presentation in other comprehensive income is appropriate? Why?

(b) should entities present changes in fair value in other comprehensive income only in the periods in which the investments in equity instruments meet the proposed identification principle in (a)? Why?

If the option is to be included in the final standard, we agree that there should be an irrevocable election on initial recognition.

Question 12
Do you agree with the additional disclosure requirements proposed for entities that apply the proposed IFRS before its mandated effective date? If not, what would you propose instead and why?

Yes, we agree with the proposed disclosure requirements.

Question 13
Do you agree with applying the proposals retrospectively and the related proposed transition guidance? If not, why? What transition guidance would you propose instead and why?

While ideally a new accounting standard should be applied on a fully retrospective basis, we do not consider this to be an absolute requirement and believe that certain reliefs should be considered in order to make implementation for December 2009 year ends operational. Many entities that will be required to adopt the new requirements (and might adopt them early) provide, in addition to the information in the primary financial statements, tables showing a five year trading history and it is not clear whether they would be required to restate all five years.

We note that there is some precedent, on the adoption of significant change, for there to be significant relief from full retrospective restatement. In 2005, when many groups in the European region and elsewhere adopted IFRS for the first time, IFRS 1 provided a number of reliefs which restricted the extent to which entities needed fully to restate transactions.

We suggest that, as a means of making the proposals less complex to implement, in the event that an entity chooses to adopt the new standard for its December 2009 year no restatement should be required prior to 1 January 2009, with an adjustment being made to opening reserves for the effect of the application of the new standard. For entities adopting for the first time in their December 2010 year ends or later, the restatement requirement could be limited to the start of the earliest comparative period presented (subject to an earliest date of 1 January 2009, to accommodate those entities which might present more than one year of comparative information).

The transitional guidance at paragraphs 24 to 33 indicates that it would be possible for an entity to adopt the new standard at any point during its financial year. Assuming this is intended, we disagree with the proposal. If a free choice existed, an entity might choose its date of adoption of the new standard depending on whether it wished to achieve a particular accounting result, such as maximising or minimising accounting mismatches. We suggest that the date on which balances are restated and classifications determined is set as being the later of the date of issue of the final standard and the start of the current reporting period. Combined with our suggested approach above, this would mean that an entity which adopted the new standard in its financial year ending 31 December 2010 would have a date of initial application of 1 January 2010, with the restatements that the transitional guidance requires for the comparative period being limited to no earlier than 1 January 2009.

Question 14
Do you believe that this alternative approach provides more decision-useful information than measuring those financial assets at amortised cost, specifically:

(a) in the statement of financial position?

(b) in the statement of comprehensive income? If so, why?

No. The requirement to measure all financial assets that are traded in an active market at fair value might appear attractive (and we assume that this is the intention – if it is, then the reference should not be to failing the loans and receivables definition, but to being traded in an active market). However, we believe that the same information could be provided through note disclosure and that the additional complexity of reporting would be likely to outweigh the benefits.

The suggested approach is also inconsistent with the proposals, which we support, whereby financial instruments with basic loan features which are managed on a contractual yield basis are measured at amortised cost (subject to our comments in response to questions 1 and 2 above).

Question 15
Do you believe that either of the possible variants of the alternative approach provides more decision-useful information than the alternative approach and the approach proposed in the exposure draft? If so, which variant and why?

No. As noted above, we support the measurement of certain financial assets and financial liabilities at amortised cost, and basing the qualifying criteria on both the characteristics of the instruments and the manner in which they are managed. The proposed approach would be inconsistent with that view, and would add additional and unnecessary complexity.


¹ BDO International is a world wide network of public accounting firms, called BDO Member Firms, serving international clients. Each BDO Member Firm is an independent legal entity in its own country. The network is coordinated by BDO Global Coordination B.V., incorporated in the Netherlands, with an office in Brussels, Belgium, where the International Executive Office is located.


Exposure Draft ED/2009/7 of July 2009 - Financial Instruments: Classification and Measurement, Comments due by 14 September 2009:

ED-2009-7 Financial Instruments-Classification and Measurement.pdf

ED-2009-7 Financial Instruments-Classification and Measurement Basis of Conclusions.pdf

ED-2009-7 Financial Instruments-Classification and Measurement Guidance.pdf