IFRS 9 – Impairment of Financial Assets
May 1, 2016 - Monocle Research Department
IFRS 9 is the IASB’s envisaged answer to the criticisms cast at its current financial instruments standard, IAS 39. As part of the three-phase project to bring it about, the IASB has been cooperating with the US FASB on key aspects so as to bring about the convergence and greater consistency the G20 so stresses. In terms of IFRS 9, it is impairment methodology which remains at the heart of the boards’ ongoing deliberations, and it is the focus of our discussion in this paper.
The foundations of the new impairment standard have been laid and spell the end of incurred loss provisioning. In its place comes the notion of expected loss provisioning, which seeks to estimate losses on a probabilistic, ‘expected value’ basis and recognise them in a timely manner. In doing so, the aim is to reflect the deterioration in the credit quality of financial assets.
Delays in finalising this project belie the fervour with which convergence has been pursued. Indeed, broad consultation with the IASB’s Expert Advisory Panel and supervisors and stakeholders in countries the world over has uncovered how divergent current practices are. At the heart of this endeavour, the urgency behind replacing divergent practices which have hurt entities during the financial crisis vies with the necessity for operational, informative, comparable, and consistently applied standards to supersede them.
Expected loss provisioning seeks to estimate losses on a probabilistic ‘expected value’ basis.
The IASB and the FASB have been working toward aligning the IFRS and US GAAP standards they disseminate for close to ten years now. The Group of Twenty (G20) talks at the 2009 Pittsburgh Summit, however, urged the two boards to re-double their efforts in finalising a common set of high quality accounting standards.
Valuation practices for financial instruments played an important role in the run up to the financial crisis. The G20 believes that inadequate and divergent accounting standards served to mask impending trouble and exacerbate it once it arrived. Accounting standards relating to the classification and measurement of financial instruments have come under intense scrutiny as a result.
The G20’s charge to standard-setters in this regard was to:
- Simplify the accounting standards for financial instruments; and
- Place more emphasis on asset impairment measurement by integrating more credit-related information.
Currently in IASB jurisdictions IAS 39 prescribes treatment of financial assets and liabilities in terms of recognition, measurement, impairment, and hedging. It is the intention of the IASB to entirely replace the latter with a new standard, IFRS 9. Development has been divided into three phases:
- Classification and Measurement;
- Impairment methodology (working jointly with the FASB); and
- Hedge Accounting.
Despite completion of the project having been initially envisaged for the second half of 2011, only the first phase remains complete. Consequently, in December 2011 the IASB issued a proposal to defer the mandatory date of IFRS 9 application to 1 January 2015, from 1 January 2013. However, early adoption is permitted and comparative-period financial statements need not be restated.
2. Current definition of Impairment
Under IAS 39 all financial assets must be measured at fair value at initial recognition. For subsequent measurement, assets are classified into four categories: financial assets at fair value through profit or loss; available for-sale financial assets; loans and receivables; and held-to-maturity investments. The latter two categories of assets are subsequently measured at amortised cost, and need to be tested for impairment at the end of each reporting period so that any gains or losses in value can be ascertained. For assets measured at fair value, gains and losses are naturally determined in updating their fair value.
Impairment can be generically defined as follows:
Impairment is the difference between the initial recorded value of an asset (carrying value) and the value of economic benefits (e.g. resale value, or contractual cash flows) that will conceivably accrue to the entity from that asset.
IAS 39 uses an incurred loss model to determine impairment of financial assets measured at amortised cost. The model is so called since it requires that there be objective evidence that an impairment loss has been incurred. The amount of the loss is measured as the difference between the asset’s carrying amount and the present value of estimated future cash flows (excluding future credit losses that have not been incurred) discounted at the financial asset’s original effective interest rate; i.e. the effective interest rate computed at initial recognition.
3. The problem with IAS 39
IAS 39 is a backward-looking impairment model based on financial assets’ actual behaviour, reliant on detection of loss events that have occurred. It effectively assumes that past behaviour is indicative of the future as it takes no account of perceived future losses and economic conditions. As a result allowances for credit losses have typically been found to be too low as the economy slows and greater losses are experienced.
That non-incurred future losses are ignored in determining the present value of estimated future cash flows means that, depending on current economic conditions and how much history is factored in, impairment allowances are too low. Inasmuch as incurred losses act as the triggers for impairment, allowance for loss is made too late.
The incurred loss framework leads to an inconsistency between the initial measurement of financial assets and their subsequent (amortised cost) measurement and impairment. The initial fair value of the asset takes into account its credit risk (the premium over the basic interest rate for credit risk), thus implicitly accounting for expected future credit losses. However, expected losses are expressly ignored when determining the effective interest rate with which to subsequently value the asset at amortised cost. The inflated effective interest rate means interest revenue is overstated in the periods before a loss event occurs and the resulting impairment losses are therefore partly adjustments of that inappropriate revenue recognition.
4. The changes IFRS 9 will bring
IFRS 9 (Phase 1) narrowed the four subsequent asset measurement categories of IAS 39 to two: assets measured at fair value; and those measured at amortised cost. Although the finer details of the new impairment model (Phase 2) have not been finalised, the major concepts have been laid down and agreed on by the two boards. The move is one away from incurred loss provisioning to expected loss provisioning. In this way, a statistical approach is implicitly incorporated in that forward-looking estimates of future cash flows are sought. The guiding principle in terms of impairment recognition is to reflect the deterioration in the credit quality of the financial assets.
To that end the boards have proposed a three-bucket structure into which assets will be placed depending on the deterioration of their credit quality since initial recognition. Bucket 1 is where financial assets would be placed at initial recognition. Here only a portion of lifetime expected losses is recognised as an impairment allowance. The transfer to Buckets 2 and 3 will depend on the ensuing deterioration in the credit quality of the assets. In these two buckets, the full amount of lifetime expected losses would be recognised as an allowance.
Bucket 1 impairment measurement and recognition
The period and recognition of expected lifetime losses is under review. For Bucket 1 assets, losses expected to materialise over either an emergence period or a set number of months will be recognised as a loss allowance.
When to recognise lifetime expected losses
This concept is tantamount to deciding when to transfer an asset to Bucket 2 or 3 – see “Differentiating between Bucket 2 and Bucket 3” below – since lifetime losses are then recognised. This decision will depend on the definition or guidance provided relating to ‘meaningful’ deterioration of credit quality.
Grouping of Assets
Assets are to be grouped into categories by which to evaluate whether transfer out of Bucket 1 is appropriate. The principles behind this categorisation are:
- Assets are to be grouped on the basis of ‘shared risk characteristics’;
- Assets may not be grouped at a more aggregated level if there are shared risk characteristics for a sub-group that would indicate that recognition of lifetime losses is appropriate;
- If assets cannot be appropriately grouped, or are individually significant, then those assets are to be evaluated individually;
- An entity may evaluate assets within a group of similar assets with shared risk characteristics, or individually.
Differentiating between Bucket 2 and Bucket 3
One approach suggests there is a simply a “unit of evaluation” difference between Bucket 2 and Bucket 3, in that Bucket 2 only includes financial assets evaluated collectively and Bucket 3 only includes financial assets evaluated individually. Alternatively, the transfer point from Bucket 2 to
Bucket 3 would be based on deterioration to a particular level of credit risk. In this case, that level would need to be defined.
Application of the credit deterioration model to publicly traded debt instruments and loans
Credit impairment is typically evaluated for debt securities (typically carried at amortised cost) on an instrument-by-instrument basis as in most cases the security represents a unique instrument. The boards do not expect the three-bucket approach to change this practice, but the “Grouping of Assets” guidance above still allows for any collective assessment. Although explored, the boards decided against the use of any explicit rules to trigger recognition of lifetime expected credit losses, such as if the fair value of the security (which should be available since the securities are publicly traded) is less than a certain percentage of its amortised cost over a specified time period.
Similarly, for commercial and consumer loans, the boards also rejected including a presumption of when recognition of lifetime losses is appropriate, such as reaching a particular delinquency status.
Estimating expected losses as an ‘expected value’
In March 2011, the boards suggested that expected losses should be estimated under the framework of an “expected value”. Methods by which to implement this framework include:
- Estimating expected losses based on the probability-weighted mean of possible outcomes, starting with appropriate loss rates, say;
- Estimating the amount of cash flows expected not to be recovered using supportable historical, current, and forecasted information;
- Using probabilities of default (PD), loss given default (LGD), and exposure at default (EAD) estimates as in the Basel II regulatory
In future meetings, it might be decided that further application guidance is necessary as to other appropriate methods that could be used. On the other hand, the boards may conclude that certain methods actually are not justifiable means of achieving the expected value objective.
The boards plan to consider the principle behind recognition of lifetime losses being applied to assets that improve in credit quality so as to warrant a transfer from Bucket 2 to Bucket 1. Practical applications of the expected value objective will be further explored.
A simplified, more principles-based approach to asset impairment has been sought. Such a framework is more flexible and customisable, which promotes more meaningful results for an organisation. At the same time, robust disclosure requirements and appropriate implementation advice will need to be put in place in order to protect the interests of comparability and consistency of application.
It is within the framework of expected loss measurement that decisive guidance is essential as it forms the basis of impairment measurement. A Basel II expected loss approach, as described above, would be favourable in that impairment and credit risk management would bear closer ties and existing systems and data collections could be utilised. Regardless of the method, expected loss estimation and the need to revise these at each reporting period could prove operationally burdensome to implement.
Monocle Solutions has extensive international experience in data management, risk modelling and implementation, and business strategy. Monocle is able to identify challenges and pitfalls and how practicably to overcome them. Our know-how and unique approach, along with keeping abreast of the latest developments in risk and other arenas, ensures that we consistently deliver relevant, valuable solutions.