1.1 Brief history and background of the impairment project

During the 2007/8 global financial crisis, the delayed recognition of credit losses that are associated with loans and other financial instruments was identified as a weakness in existing accounting standards. This is primarily due to the fact that the previous impairment requirements under IAS 39 – Financial Instruments: Recognition and Measurement – were based on an incurred loss model, i.e. credit losses were not recognised until a credit loss event had occurred. Since losses are rarely incurred evenly over the lives of loans, there was a mismatch in the timing of the recognition of the credit spread inherent in the interest charged on the loans over their lives and any impairment losses that was only recognised at a later date. A further identified weakness was the complexity of different entities using different approaches to calculate impairment.

As part of the joint approach by the IASB and the FASB to deal with the financial reporting issues arising from the financial crisis, the boards set up the Financial Crisis Advisory Group (FCAG) in October 2008 to consider how improvements in financial reporting could help to enhance investor confidence in financial markets. Not long after, the leaders of the Group of 20 (also known as the G20) published a report Declaration on Strengthening the Financial System in April 2009 that called on the accounting standard setters to reduce the complexity of accounting standards for financial instruments and to strengthen accounting recognition of loan-loss provisions by incorporating a broader range of credit information.¹

In July 2009, the FCAG presented its report to the IASB and the FASB about the standard-setting implications of the global financial crisis. Consistent with the G20's recommendations, the FCAG also recommended both the IASB and the FASB to explore alternatives to the incurred loss model for loan loss provisioning that used more forward-looking information.²

In June 2009, the IASB published a request for information – Impairment of Financial Assets: Expected Cash Flow Approach – on the feasibility of an expected loss model for the impairment of financial assets. Following this, in November 2009, the IASB issued an Exposure Draft – Financial Instruments: Amortised Cost and Impairment (the 2009 ED) that proposed an impairment model based on expected losses rather than on incurred losses, for all financial assets recorded at amortised cost. In this approach, the initial ECLs were to be recognised over the life of a financial asset, by including them in the computation of the effective interest rate (EIR) when the asset was first recognised. This would build an allowance for credit losses over the life of a financial asset and so match the recognition of credit losses with that of the credit spread implicit in the interest charged. Subsequent changes in credit loss expectations would be reflected in catch-up adjustments to profit or loss based on the original EIR. Because the proposals were much more closely linked to credit risk management concepts, the IASB acknowledged that this would represent a fundamental change from how entities currently operate (i.e. typically, entities operate their accounting and credit risk management systems separately). Consequently, the IASB established a panel of credit risk experts, the Expert Advisory Panel (EAP), to provide input to the project. [IFRS 9.BC5.87].

Comments received on the 2009 ED and during the IASB's outreach activities indicated that constituents were generally supportive of a model that distinguished between the effect of initial estimates of ECLs and subsequent changes in those estimates. However, they were also concerned about the operational difficulties in implementing the model proposed. These included: [IFRS 9.BC5.89]

• estimating the full expected cash flows for all financial instruments;
• applying a credit-adjusted EIR to those cash flow estimates; and
• maintaining information about the initial estimate of ECLs.

Also, the proposals would not have been easy to apply to portfolios of loans managed on a collective basis, in particular, open portfolios to which new financial instruments are added over time, and concerns were expressed about the volatility of reported profit or loss arising from the catch-up adjustments.

To address these operational challenges and as suggested by the EAP, the IASB decided to decouple the measurement and allocation of initial ECLs from the determination of the EIR (except for purchased or originated credit-impaired financial assets). Therefore, the financial asset and the loss allowance would be measured separately, using an original EIR that is not adjusted for initial ECLs. Such an approach would help address the operational challenges raised and allow entities to leverage their existing accounting and credit risk management systems and so reduce the extent of the necessary integration between these systems. [IFRS 9.BC5.92].

By decoupling ECLs from the EIR, an entity must measure the present value of ECLs using the original EIR. This presents a dilemma, because measuring ECLs using such a rate double-counts the ECLs that were priced into the financial asset at initial recognition. This is because the fair value of the loan at original recognition already reflects the ECLs, so to provide for the ECLs as an additional allowance would be to double count these losses. Hence, the IASB concluded that it was not appropriate to recognise lifetime ECLs on initial recognition. In order to address the operational challenges while trying to reduce the effect of double-counting, as well as to replicate (very approximately) the outcome of the 2009 ED, the IASB decided to pursue a dual-measurement model that would require an entity to recognise: [IFRS 9.BC5.93]

• a portion of the lifetime ECLs from initial recognition as a proxy for recognising the initial ECLs over the life of the financial asset; and
• the lifetime ECLs when credit risk had increased since initial recognition (i.e. when the recognition of only a portion of the lifetime ECLs would no longer be appropriate because the entity has suffered a significant economic loss).

It is worth noting that any approach that seeks to approximate the outcomes of the model in the 2009 ED, without the associated operational challenges, will include a recognition threshold for lifetime ECLs. This gives rise to what has been referred to as ‘a cliff effect’, i.e. the significant increase in allowance that represents the difference between the portion that was recognised previously and the lifetime ECLs. [IFRS 9.BC5.95].

Subsequently, the IASB and FASB spent a considerable amount of time and effort developing a converged impairment model. In January 2011, the IASB issued with the FASB a Supplementary Document – Financial Instruments: Impairment – reflecting a joint approach that proposed a two-tier loss allowance: [IFRS 9.BC5.96]

• for the good book, an entity would recognise the higher of a time-proportionate allowance (i.e. the lifetime ECLs over the weighted average life of the portfolio of assets) or ECLs for the ‘foreseeable future’; and
• for the bad book, an entity would recognise lifetime ECLs on those financial assets when the collectability of contractual cash flows had become so uncertain that the entity's credit risk management objective had changed from receiving the regular payments to recovery of all, or a portion of, the asset.

However, this approach received only limited support, because respondents were concerned about the operational difficulties in performing the dual calculation for the good book, that it also lacked conceptual merit and, potentially, would provide confusing information to users of financial statements. Moreover, concerns were also raised as to how ‘foreseeable future’ should be interpreted and applied.

Many constituents emphasised the importance of achieving convergence and this encouraged the IASB and FASB to attempt to develop another joint alternative approach. In May 2011, the boards decided to develop jointly an expected credit loss model that would reflect the general pattern of increases in the credit risk of financial instruments, the so-called three-bucket model. [IFRS 9.BC5.111].

However, due to concerns raised by the FASB's constituents about the model's complexity, the FASB decided to develop an alternative expected credit loss model. [IFRS 9.BC5.112]. In December 2012, the FASB issued a proposed accounting standard update, Financial Instruments Credit Losses (Subtopic 825‑15), that would require an entity to recognise a loss allowance from initial recognition at an amount equal to lifetime ECLs (see 1.4 below).

In March 2013, the IASB published a new Exposure Draft – Financial Instruments: Expected Credit Losses (the 2013 ED), based on proposals that grew out of the joint project with the FASB. The 2013 ED proposed that entities should recognise a loss allowance or provision at an amount equal to 12‑month credit losses for those financial instruments that had not yet seen a significant increase in credit risk since initial recognition, and lifetime ECLs once there had been a significant increase in credit risk. This new model was designed to:

• ensure a more timely recognition of ECLs than the existing incurred loss model;
• distinguish between financial instruments that have significantly deteriorated in credit quality and those that have not; and
• better approximate the economic ECLs.³

This two-step model was designed to approximate the build-up of the allowance as proposed in the 2009 ED, but involving less operational complexity. Figure 51.1 below illustrates the stepped profile of the new model, shown by the solid line, compared to the steady increase shown by the black dotted line proposed in the 2009 ED (based on the original ECL assumptions and assuming no subsequent revisions of this estimate). It shows that the two step model first overstates the allowance (compared to the method set out in the 2009 ED), then understates it as the credit quality deteriorates, and then overstates it once again, as soon as the deterioration is significant.

Feedback received on the IASB's 2013 ED and the FASB's 2012 Proposed Update was considered at the joint board meetings. In general, non-US constituents preferred the IASB's proposals whilst the US constituents preferred the FASB's proposals. These differences in views arose in large part because of differences in the starting point of how preparers apply US GAAP for loss allowances from that for most IFRS preparers, while the interaction between the role of prudential regulators and calculation of loss allowances is historically stronger in the US. [IFRS 9.BC5.116].

Many respondents urged the IASB to finalise the proposals in the 2013 ED as soon as possible, even if convergence could not be achieved, in order to improve the accounting for the impairment of financial assets in IFRSs. [IFRS 9.BC5.114]. The IASB re-deliberated particular aspects of the 2013 ED proposals, with the aim of providing further clarifications and additional guidance to help entities implement the proposed requirements. The IASB finalised the impairment requirements and issued them in July 2014, as part of the final version of IFRS 9.

1.2 Overview of the IFRS 9 impairment requirements

The new impairment requirements in IFRS 9 are based on an ECL model and replace the IAS 39 incurred loss model. The ECL model applies to debt instruments (such as bank deposits, loans, debt securities and trade receivables) recorded at amortised cost or at fair value through other comprehensive income, plus lease receivables and contract assets. Loan commitments and financial guarantee contracts that are not measured at fair value through profit or loss are also included in the scope of the new ECL model.

Conceptually, an impairment model is a necessary complement to an accounting model for financial assets that is based on the concept of realisation, i.e. when cash flows are received and paid. For financial assets in the scope of the impairment model, the recognition of revenue follows the effective interest method, and gains and losses relating to changes in their fair value are generally only recognised in profit or loss when the financial asset is derecognised, when that gain or loss is realised. In order to avoid delaying the recognition of impairment losses, those accounting models are complemented by an impairment model that anticipates impairment losses by recognising them before they are realised. The new impairment model of IFRS 9 does this on the basis of ECLs, which means impairment losses are anticipated earlier than under the incurred loss impairment model of IAS 39. In contrast, the accounting models for financial assets in IFRS 9 that are not based on the concept of realisation, i.e. those measured at fair value through profit or loss or at fair value through other comprehensive income without recycling (under the presentation choice for fair value changes of some investments in equity instruments, see Chapter 48 at 8 and Chapter 50 at 2.5) do not involve any separate impairment accounting. Those measurement categories implicitly include impairment losses in the fair value changes that are recognised immediately (and without any later reclassification entries between other comprehensive income and profit or loss when they are realised).

The guiding principle of the ECL model is to reflect the general pattern of deterioration, or improvement, in the credit quality of financial instruments. The ECL approach has been commonly referred to as the three-bucket approach, although IFRS 9 does not use this term. Figure 51.2 below summarises the general approach in recognising either 12‑month or lifetime ECLs.

The amount of ECLs recognised as a loss allowance or provision depends on the extent of credit deterioration since initial recognition. Under the general approach (see 3.1 below), there are two measurement bases:

• 12‑month ECLs (stage 1), which apply to all items as long as there is no significant deterioration in credit risk; and
• lifetime ECLs (stages 2 and 3), which apply when a significant increase in credit risk has occurred on an individual or collective basis.

When assessing significant increases in credit risk, there are a number of operational simplifications available, such as the low credit risk simplification (see 6.4.1 below).

Stages 2 and 3 differ in how interest revenue is recognised. Under stage 2 (as under stage 1), there is a full decoupling between interest recognition and impairment, and interest revenue is calculated on the gross carrying amount. Under stage 3, when a credit event has occurred, interest revenue is calculated on the amortised cost (i.e. the gross carrying amount adjusted for the impairment allowance).

The following example illustrates how the ECL allowance changes when a loan moves from stage 1 to stage 3.

The ECL allowance in each stage is shown below and the detailed calculation is illustrated in Example 51.3 at 5.4.1 below.

There are two alternatives to the general approach:

• the simplified approach, that is either required or available as a policy choice for trade receivables, contract assets and lease receivables (see 3.2 below); and
• the credit-adjusted effective interest rate approach, for purchased or originated credit-impaired financial assets (see 3.3 below).

ECLs are an estimate of credit losses over the next 12 months or the life of a financial instrument and, when measuring ECLs (see 5 below), an entity needs to take into account:

• the probability-weighted outcome (see 5.6 below), as ECLs should not simply be either a best or a worst-case scenario, but should, instead, reflect the possibility that a credit loss occurs and the possibility that no credit loss occurs. Following discussion at the ITG, this is understood to include a need to consider multiple economic scenarios (see 5.6 below);
• the time value of money (see 5.7 below); and
• reasonable and supportable information that is available without undue cost or effort at the reporting date about past events, current conditions and forecasts of future economic conditions (see 5.9 below).

1.3 Key changes from the IAS 39 impairment requirements and the implications

The new IFRS 9 impairment requirements eliminate the IAS 39 threshold for the recognition of credit losses, i.e. it is no longer necessary for a credit event to have occurred before credit losses are recognised. Instead, an entity always accounts for ECLs, and updates the loss allowance for changes in these ECLs at each reporting date to reflect changes in credit risk since initial recognition. Consequently, the holder of the financial asset needs to take into account more timely and forward-looking information.

The main implications for both financial and non-financial entities are as follows:

• The scope of the impairment requirements is now much broader. Previously, under IAS 39, there were different impairment models for financial assets measured at amortised cost and available-for-sale financial assets. Under IFRS 9, there is a single impairment model for all debt instruments measured at amortised cost and at fair value through other comprehensive income. Furthermore, loan commitments and financial guarantee contracts that were previously in the scope of IAS 37 – Provisions, Contingent Liabilities and Contingent Assets – are now in the scope of the IFRS 9 impairment requirements (see 11 below).
• Previously, under IAS 39, loss allowances were only recorded for impaired exposures. The new impairment requirements result in earlier recognition of credit losses, by necessitating a 12‑month ECL allowance for all credit exposures not measured at fair value through profit or loss. In addition, there is a larger allowance for all credit exposures that have significantly deteriorated (as compared to the recognition of incurred losses under IAS 39). While credit exposures in stage 3, as illustrated in Figure 51.2 above, are similar to those deemed by IAS 39 to have suffered individual incurred losses, credit exposure in stages 1 and 2 essentially replace those exposures measured under IAS 39's collective approach.
• The ECL model is more forward-looking than the IAS 39 impairment model. Holders of financial assets are not only required to consider historical information that is adjusted to reflect the effects of current conditions and information that provides objective evidence that financial assets are impaired in relation to incurred losses. They are also required to consider reasonable and supportable information that includes forecasts of future economic conditions including, where relevant, multiple scenarios, when calculating ECLs, on an individual and collective basis.
• The application of the new IFRS 9 impairment requirements has led to an increase in credit loss allowances (with a corresponding reduction in equity on first-time adoption) of many entities, particularly banks and similar financial institutions. However, the increase in the loss allowance has varied by entity, depending on its portfolio and previous practices. Entities with shorter term and higher quality financial instruments are less significantly affected. Similarly, financial institutions with unsecured retail loans are generally affected to a greater extent than those with collateralised loans such as mortgages.
• Moreover, the focus on expected losses will possibly result in higher volatility in the ECL amounts charged to profit or loss, especially for financial institutions. The level of loss allowances will increase as economic conditions are forecast to deteriorate and will decrease as economic conditions are forecast to become more favourable. This may be further compounded by the significant increase in the loss allowance when financial instruments move between 12‑month and lifetime ECLs and vice versa.
• The need to incorporate forward-looking information, including establishing multiple macroeconomic scenarios, determining the probability of their occurrence and assessing how changes in macroeconomic factors will affect ECLs, means that the application of the standard requires considerable judgement. Also, the increased level of judgement required in making the ECL calculation and assessing when significant deterioration has occurred may mean that it will be difficult to compare the reported results of different entities. However, the more detailed disclosure requirements, that require entities to explain their inputs, assumptions and techniques used in estimating ECLs requirements, should provide greater transparency over entities' credit risk and provisioning processes. The Enhanced Disclosures Task Force (EDTF), established in 2012 by the Financial Stability Board to recommend best practice market risk disclosures, has also published guidance to promote greater transparency and comparability about the application of the ECL model (see Chapter 54 at 9.2).
• In financial institutions, finance and credit risk management systems and processes now have to be better connected, because of the necessary alignment between risk and accounting in the new model. Risk models and data are used more extensively to make the assessments and calculations required for accounting purposes, which are both a major change from IAS 39 and a key challenge.
• In addition, financial institutions need to fully understand the complex interactions between the IFRS 9 and regulatory capital requirements in relation to credit losses. The Basel Committee on Banking Supervision has now finalised what it calls an ‘interim’ approach and transitional arrangements, providing national jurisdictions with a framework for any arrangement. This is contained in the BCBS document Standards – Regulatory treatment of accounting provisions – interim approach and transitional arrangements. However, the long-term regulatory treatment of ECL provisions remains to be determined. In most cases, the new IFRS 9 ECL requirements have resulted in a reduction in the regulatory capital of financial institutions.
• The IFRS impairment requirements have had a less significant impact on corporates than on banks and similar financial institutions. To reflect this, we have set out a summary of the main implications for corporates in a separate section (see 4 below). Given that IFRS 16 – Leases – has retained the distinction for accounting by lessors between finance leases and operating leases, the effect of IFRS 9 is largely limited to finance leases. For these, there is a significant increase in allowances, particularly if the lessor has opted to apply the simplified approach and record lifetime allowances (see 3.2 below).

1.4 Key differences from the FASB's standard

On 16 June 2016, the FASB issued an Accounting Standard Update (ASU), Financial Instruments – Credit Losses (Topic 326), that aims to address the same fundamental issue that the IASB's ECL model (in IFRS 9) addresses, namely the delayed recognition of credit losses resulting from the incurred credit loss model. It is therefore also an ECL model but it is not the same as the model in IFRS 9. The most significant differences between the FASB's and the IASB's ECL models are, as follows:

• The FASB's ECL model (known as the Current ECL or CECL model) will not be applied to debt securities measured at fair value through other comprehensive income (i.e. available for sale securities under US GAAP). Rather, for available for sale securities, an allowance is recognised to reflect estimated credit losses and is adjusted over time to reflect both positive and adverse changes in expected cash flows.
• The FASB's ECLs will be calculated based on the losses expected over the remaining contractual life of an asset, considering the effect of prepayments and reasonably expected troubled debt restructurings. An allowance for lifetime ECLs will be required when the loan is initially recognised instead of 12‑month ECLs. As a result, the FASB's model does not require an entity to assess whether there has been a significant deterioration in credit quality, in contrast to the assessment required by IFRS 9. This is similar to the IFRS 9 simplified approach (see 3.2 below).
• The FASB's standard does not require, nor does it prohibit, an entity to use probability weighted outcomes. However, entities are required to incorporate the risk of loss, even if such a risk is remote, into their estimate. It is expected that an allowance of zero will only be appropriate in very limited circumstances.
• For purchased credit-impaired assets defined as ‘acquired individual financial assets (or acquired groups of financial assets with shared risk characteristics) that, as of the date of acquisition, have experienced a more-than-insignificant deterioration in credit quality since origination, as determined by an acquirer's assessment’ and that are measured at amortised cost, the FASB's model will require an entity to increase the purchase price by the allowance for ECLs upon acquisition. In doing so, the FASB model will gross up the asset's amortised cost upon acquisition by the difference between the calculated ECLs and the purchase discount (known as the non-credit discount), by establishing a liability which is subsequently amortised to income over the asset's life (see 3.3 below for the accounting treatment of credit-impaired assets under IFRS 9).
• Unlike IFRS 9, there is no exception for revolving credit facilities (e.g. commitments connected with overdrafts and credit cards) under the FASB's model (see 12 below for the IFRS 9 treatment) and therefore, no impairment allowance is required if the commitment is unconditionally cancellable or legally revocable without any conditions.

The FASB standard has tiered effective dates, starting in 2020 for calendar-year reporting public business entities that meet the definition of a U.S. Securities and Exchange Commission (SEC) filer. Early adoption is permitted for all entities but this cannot be before 2019 for calendar-year entities.

1.5 The IFRS Transition Resource Group for Impairment of Financial Instruments (ITG) and IASB webcasts

In 2014, the IASB set up an ITG that aimed to:

• provide a public discussion forum to support stakeholders on implementation issues arising from the new impairment requirements that could create diversity in practice; and
• inform the IASB about the implementation issues, to help the IASB determine what action, if any, was needed to address them.⁵

However, the ITG did not issue any guidance.

Members of the ITG included financial statement preparers and auditors from various geographical locations with expertise, skills or practical knowledge on credit risk management and accounting for impairment. Board members and observers from the Basel Committee on Banking Supervision and the International Organisation of Securities Commissions also attended the meetings.

The ITG agenda papers were prepared by the IASB staff and were made public before the meetings. The staff also provided ITG meeting summaries which are not authoritative. Both the staff papers and the meeting summaries represent educational reading on the issues submitted.

Following its inaugural meeting in December 2014 to discuss its operating procedures, the ITG met three times, on 22 April 2015, on 16 September 2015, and on 11 December 2015. IFRS 9 is now in effect and therefore the IASB has brought the work of the ITG to a conclusion. Any future support provided by the IASB for the implementation and application of IFRS 9 will be from the IFRIC, the Board and through the publication of educational materials.

On 22 April 2015, the ITG discussed eight implementation issues raised by stakeholders. These included:⁶

• when applying the impairment requirements at the reporting date, whether and how to incorporate events and forecasts that occur after economic forecasts have been made, but before the reporting date, and between the reporting period end and the date of signing the financial statements (see 5.9.4 below);
• whether the impairment requirements in IFRS 9 must also be applied to other commitments to extend credit, in particular, a commitment (on inception of a finance lease) to commence a finance lease at a date in the future and a commitment by a retailer through the issue of a store account to provide a customer with credit when the customer buys goods or services from the retailer in the future (see 11 below);
• whether there is a requirement to measure ECLs at dates other than the reporting date, namely the date of derecognition and the date of initial recognition (see 7.3.1 below);
• whether an entity should consider the ability to recover cash flows through an integral financial guarantee contract when assessing whether there has been a significant increase in the credit risk of the guaranteed debt instrument since initial recognition (see 6.1.1 below);
• the maximum period to consider when measuring ECLs on a portfolio of mortgage loans that have a stated maturity of 6 months, but contain a contractual feature whereby the term is automatically extended every 6 months subject to the lender's non-objection (see 5.5 below);
• the maximum period to consider when measuring ECLs for revolving credit facilities and the determination of the date of initial recognition of the revolving facilities for the purposes of assessing them for significant increases in credit risk (see 12.2.2 and 12.2.4 below);
• whether the measurement of ECLs for financial guarantee contracts issued should consider future premium receipts due from the holder and, if so, how (see 11 below); and
• the measurement of ECLs in respect of a modified financial asset, the calculation of the modification gain or loss and subsequent requirement to measure ECLs on the modified financial asset as well as the appropriate presentation and disclosure (see 8.2 below).

On 16 September 2015, the ITG held its third meeting to discuss six implementation issues raised by stakeholders. These included:⁷

• how to identify a significant increase in credit risk for a portfolio of retail loans when identical pricing and contractual terms are applied to customers across broad credit quality bands (see 6.2.1 below);
• the possibility of using behavioural indicators of credit risk for the purpose of the assessment of significant increases in credit risk since initial recognition (see 6.2.4 below);
• when assessing significant increases in credit risk, whether an entity would be required to perform an annual review to determine whether circumstances still support the use of the 12‑month risk of a default occurring as an approximation of changes in the lifetime risk of a default occurring (see 6.4.3 below);
• when measuring ECLs for revolving credit facilities, how an entity should estimate future drawdowns on undrawn lines of credit when an entity has a history of allowing customers to exceed their contractually set credit limits on overdrafts and other revolving credit facilities (see 12.3 below);
• at what level should forward-looking information be incorporated – at the level of the entity or on a portfolio-by-portfolio basis (see 5.9.5 below); and
• how to determine what is reasonable and supportable forward-looking information and how to treat shock events with material, but uncertain, economic consequences (see 5.9.5 below).

On 11 December 2015, the ITG held its fourth meeting to discuss eleven implementation issues raised by stakeholders. These included:⁸

• what was meant by the ‘current EIR’ when an entity recognises interest revenue in each period based on the actual floating-rate applicable to that period (see 5.7 below);
• what was meant by ‘part of the contractual terms’, specifically whether a credit enhancement must be an explicit term of the related asset's contract in order for it to be taken into account in the measurement of ECLs, or whether other credit enhancements that are not recognised separately can also be taken into account (see 5.8.1.A below);
• whether cash flows that are expected to be recovered from the sale on default of a loan could be included in the measurement of ECLs (see 5.8.2 below);
• application of the revolving credit facilities exception set out in paragraph 5.5.20 of IFRS 9 to multi-purpose facilities (see 12.1 below);
• how future drawdowns should be estimated for charge cards when measuring ECLs if there is no specified credit limit in the contract (see 12.3 below);
• how an entity should determine the starting-point and the ending-point of the maximum period to consider when measuring ECLs for revolving credit facilities (see 12.2 below);
• when measuring ECLs, whether an entity can use a single forward-looking economic scenario or whether an entity needs to incorporate multiple forward-looking scenarios, and if so how (see 5.6 below);
• when assessing significant increases in credit risk, whether an entity can use a single forward-looking economic scenario or whether the entity needs to incorporate multiple forward-looking scenarios, and if so how (see 6.7 below);
• whether there is a requirement to assess significant increases in credit risk for financial assets with a maturity of 12 months or less (see 6 below);
• how to measure the gross carrying amount and loss allowance for credit-impaired financial assets that are not purchased or originated credit-impaired and that are measured at amortised cost (see 14.1.1 below); and
• whether an entity is required to present the loss allowance for financial assets measured at amortised cost (or trade receivables, contract assets or lease receivables) separately in the statement of financial position (see 14.1 below).

The FASB (see 1.4 above) has also set up its own Transition Resource Group (TRG) for credit losses and its discussions may prove relevant to the application of IFRS 9 in areas where the two ECL models are similar.

The IFRIC discussed two issues in relation to the IFRS 9 ECL model in November 2018:

• Whether the cash flows expected from a financial guarantee contract or any other credit enhancement can be included in the measurement of ECL if the credit enhancement is required to be recognised separately applying IFRS Standards (see 5.8.1 below).
• How an entity should present amounts recognised in the statement of profit or loss when a credit-impaired financial asset is subsequently cured, i.e. paid in full or no longer credit-impaired (see 14.1.2 below).

In addition, as part of its activities to support implementation, the IASB has published three educational webcasts since IFRS 9 was published.⁹

• The first, on forward-looking information and multiple scenarios was released on 25 July 2016. It discussed when multiple scenarios need to be considered and the concept of non-linearity, consistency of scenarios, probability-weighted assessment of significant increase in credit risk, and approaches to incorporating forward-looking scenarios (see 5.6 below).
• The second, on the expected life of revolving facilities was released on 16 May 2017. It focused on how credit risk management actions would affect the expected life of revolving facilities for the purpose of measuring ECLs (see 12.2 below).
• The third, on curing of a credit-impaired financial asset was released on 22 July 2019. It summarised the IFRIC's discussion and explained the conclusion it reached on how a company should present amounts in its statement of profit or loss if a credit-impaired financial asset is subsequently paid in full or no longer credit-impaired (cured) (see 14.1.2 below).

1.6 Other guidance on expected credit losses

In November 2015, the Enhanced Disclosure Task Force (EDTF) published its report, Impact of Expected Credit Loss Approaches on Bank Risk Disclosures, in which it recommended disclosures for banks to provide with the implementation of the ECL requirements of IFRS 9 and US GAAP (see 15 below).

In December 2015, the Basel Committee on Banking Supervision issued its Guidance on accounting for expected credit losses, aimed primarily at internationally active banks, which sets out supervisory expectations regarding sound credit risk practices associated with implementing and applying an ECL accounting framework (see 7.1 below).

On 17 June 2016, the Global Public Policy Committee of representatives of the six largest accounting networks (the GPPC) published The implementation of IFRS 9 impairment by banks – Considerations for those charged with governance of systemically important banks (the GPPC guidance) to promote a high standard in the implementation of accounting for ECLs. It aims to help those charged with governance to evaluate management's progress during the implementation and transition phase. A year later, on 28 July 2017, the GPPC issued a paper titled The Auditor's Response to the Risks of Material Misstatement Posed by Estimates of Expected Credit Losses under IFRS 9 to assist audit committees oversee the audit of ECLs (see 7.2 below).

3.1 General approach

Under the general approach, at each reporting date, an entity recognises a loss allowance based on either 12‑month ECLs or lifetime ECLs, depending on whether there has been a significant increase in credit risk on the financial instrument since initial recognition. [IFRS 9.5.5.3, 5.5.5]. The changes in the loss allowance balance are recognised in profit or loss as an impairment gain or loss. [IFRS 9.5.5.8, Appendix A].

Essentially, an entity must make the following assessment at each reporting date:

• for credit exposures where there have not been significant increases in credit risk since initial recognition, an entity is required to provide for 12‑month ECLs, i.e. the portion of lifetime ECLs that represent the ECLs that result from default events that are possible within the 12‑months after the reporting date (stage 1 in Figure 51.2 at 1.2 above); [IFRS 9.5.5.5, Appendix A]
• for credit exposures where there have been significant increases in credit risk since initial recognition on an individual or collective basis, a loss allowance is required for lifetime ECLs, i.e. ECLs that result from all possible default events over the expected life of a financial instrument (stages 2 and 3 in Figure 51.2 at 1.2 above); [IFRS 9.5.5.4, 5.5.3, Appendix A] or
• in subsequent reporting periods, if the credit quality of the financial instrument improves such that there is no longer a significant increase in credit risk since initial recognition, then the entity reverts to recognising a loss allowance based on 12‑month ECLs (i.e. the approach is symmetrical). [IFRS 9.5.5.7].

It may not be practical to determine for every individual financial instrument whether there has been a significant increase in credit risk, because they may be small and many in number and/or because there may not be the evidence available to do so. [IFRS 9.B5.5.1]. Consequently, it may be necessary to measure ECLs on a collective basis, to approximate the result of using comprehensive credit risk information that incorporates forward-looking information at an individual instrument level (see 6.5 below). [IFRS 9.BC5.141].

The standard includes practical expedients, in particular the use of a simplified approach (see 3.2 below) and a provision matrix (see 10.1 below) which should help in measuring the loss allowance for trade receivables.

To help enable an entity's assessment of significant increases in credit risk, IFRS 9 also provides the following operational simplifications:

• a low credit risk threshold equivalent to investment grade (see 6.4.1 below), below which no assessment of significant increases in credit risk is required;
• the ability to rely on past due information if reasonable and supportable forward-looking information is not available without undue cost or effort (see 6.4.2 below). This is subject to the rebuttable presumption that there has been a significant increase in credit risk if the loan is 30 days past due (see 6.2.2 below); and
• use of a change in the 12‑month risk of a default as an approximation for change in lifetime risk (see 6.4.3 below).

The IFRS 9 illustrative examples also provide the following suggestions on how to implement the assessment of significant increases in credit risk:

• assessment at the counterparty level (see 6.4.4 below); and
• a set transfer threshold by determining maximum initial credit risk for a portfolio (see 6.4.5 below).

In stages 1 and 2, there is a complete decoupling between interest recognition and impairment. Therefore, interest revenue is calculated on the gross carrying amount (without deducting the loss allowance). If a financial asset subsequently becomes credit‑impaired (stage 3 in Figure 51.2 at 1.2 above), an entity is required to calculate the interest revenue by applying the EIR in subsequent reporting periods to the amortised cost of the financial asset (i.e. the gross carrying amount net of loss allowance) rather than the gross carrying amount. [IFRS 9.5.4.1, Appendix A]. Financial assets are assessed as credit-impaired using substantially the same criteria as for the impairment assessment of an individual asset under IAS 39. [IFRS 9 Appendix A].

A financial asset is credit-impaired when one or more events that have a detrimental impact on the estimated future cash flows of that financial asset have occurred. Evidence that a financial asset is impaired includes observable data about such events. IFRS 9 provides a list of events that are substantially the same as the IAS 39 loss events for an individual asset assessment: [IFRS 9 Appendix A]

• significant financial difficulty of the issuer or the borrower;
• a breach of contract, such as a default or past due event;
• the lender(s) of the borrower, for economic or contractual reasons relating to the borrower's financial difficulty, having granted to the borrower a concession(s) that the lender(s) would not otherwise consider;
• it is becoming probable that the borrower will enter bankruptcy or other financial reorganisation;
• the disappearance of an active market for that financial asset because of financial difficulties; or
• the purchase or origination of a financial asset at a deep discount that reflects the incurred credit losses.

It may not be possible for an entity to identify a single discrete event. Instead, the combined effect of several events may have caused the financial asset to become credit-impaired. [IFRS 9 Appendix A].

There has also been some debate as to whether financial assets that are considered to be in default (e.g. because payments are more than 90 days past due) but that are fully collateralised (so that there is no ECL) would qualify as credit-impaired and therefore have to be transferred to stage 3. Although the definition of credit-impaired refers to ‘a detrimental impact on the estimated future cash flows’, it is not clear whether this should be read to include any recoveries from the realisation of collateral and IFRS 9 has no explicit requirements to consider collateral when assessing credit-impaired financial assets.

There are some strong arguments in favour of aligning the criteria for transferring an asset to stage 3 with those for assessing whether it is in default. First, IFRS 9 bases significant deterioration on risk of a default occurring and it would therefore seem inconsistent (and potentially confusing for users) if the value of collateral is considered for stage 3 allocation. Also, if collateral value were to influence the stage 3 allocation, this could result in some instability between stage 2 and 3, as exposures would potentially go back and forth depending on the collateral value.

Aligning stage 3 with the default status affects the scope of instruments to which the purchased or originated credit-impaired approach must be applied. However, for any exposure which is fully collateralised and where the expected loss is zero, classification as a purchased or originated credit-impaired financial asset, or classification between stage 1, 2 or 3 does not affect the accounting. If the expected loss is zero, it will not affect the EIR calculation.

Also, IFRS 7 requires a quantitative disclosure about the collateral held as security and other credit enhancements for financial assets that are credit-impaired at the reporting date (e.g. quantification of the extent to which collateral and other credit enhancements mitigate credit risk). [IFRS 7.35K(c)].

In subsequent reporting periods, if the credit quality of the financial asset improves so that the financial asset is no longer credit-impaired and the improvement can be related objectively to the occurrence of an event (such as an improvement in the borrower's credit rating), then the entity should once again calculate the interest revenue by applying the EIR to the gross carrying amount of the financial asset. [IFRS 9.5.4.2].

When the entity has no reasonable expectations of recovering a financial asset, in its entirety or a portion thereof, then the gross carrying amount of the financial asset should be directly reduced in its entirety. A write-off constitutes a derecognition event (see 14.1.1 below). [IFRS 9.5.4.4].

3.2 Simplified approach

The simplified approach does not require an entity to track the changes in credit risk, but instead requires the entity to recognise a loss allowance based on lifetime ECLs at each reporting date. [IFRS 9.5.5.15].

An entity is required to apply the simplified approach for trade receivables or contract assets that result from transactions within the scope of IFRS 15 and that do not contain a significant financing component, or when the entity applies the practical expedient for contracts that have a maturity of one year or less, in accordance with IFRS 15 (see Chapter 28). [IFRS 9.5.5.15(a)(i)]. Paragraphs 60‑65 of IFRS 15 provide the requirements for determining the existence of a significant financing component in the contract, including the use of the practical expedient for contracts that are one year or less.

A contract asset is defined as an entity's right to consideration in exchange for goods or services that the entity has transferred to a customer when that right is conditioned on something other than the passage of time (for example, the entity's future performance). [IFRS 15 Appendix A]. IFRS 15 describes contracts with a significant financing component as those for which the agreed timing of payment provides the customer or the entity with a significant benefit of financing on the transfer of goods or services to the customer. Hence, in determining the transaction price, an entity is required to adjust the promised amount of consideration for the effects of the time value of money. [IFRS 15.60]. However, if the entity expects, at contract inception, that the period between when the entity transfers a promised good or service to a customer and when the customer pays for that good or service will be one year or less, as a practical expedient, an entity need not adjust the promised amount of consideration for the effects of a significant financing component. [IFRS 15.63].

Application of the simplified approach to trade receivables and contract assets that do not contain a significant financing component intuitively makes sense. In particular, for trade receivables and contract assets that are due in 12 months or less, the 12‑month ECLs are the same as the lifetime ECLs.

However, an entity has a policy choice to apply either the simplified approach or the general approach for the following: [IFRS 9.5.5.16]

• all trade receivables or contract assets that result from transactions within the scope of IFRS 15, and that contain a significant financing component in accordance with IFRS 15. The policy choice may be applied separately to trade receivables and contract assets (see 10.1 below and Chapter 28); [IFRS 9.5.5.15(a)(ii)] and
• all lease receivables that result from transactions that are within the scope of IFRS 16. The policy choice may be applied separately to finance and operating lease receivables (see 10.2 below, Chapter 23). [IFRS 9.5.5.15(b)].

The IASB noted that offering this policy choice would reduce comparability. However, the IASB believes it would alleviate some of the practical concerns of tracking changes in credit risk for entities that do not have sophisticated credit risk management systems. [IFRS 9.BC5.225].

In practice, trade receivables may be sold to a factoring bank, whereby all risks and rewards are transferred to the bank. Consequently, the trade receivables are derecognised by the transferring entity and recognised by the factoring bank which obtains the right to receive the payments made by the debtor for the invoiced amount. In such a case, we believe that the ‘factored’ trade receivables are outside the scope of the simplified approach for the purpose of the factoring bank when applying the IFRS 9 ECL model. This is because the simplified approach is limited to trade receivables that result from transactions within the scope of IFRS 15, i.e. based on a contract to obtain goods or services. This is not the case for the factoring bank since it has acquired the trade receivables through a factoring agreement. Moreover, the simplified approach was introduced to assist entities with less sophisticated credit risk management systems. [IFRS 9.BC5.104]. Factoring banks are likely to have more sophisticated credit risk management systems in place.

3.3 Purchased or originated credit-impaired financial assets

On initial recognition of a financial asset, an entity is required to determine whether the asset is credit-impaired. The criteria are set out at 3.1 above. [IFRS 9.5.5.3, 5.5.5, 5.5.13].

A financial asset may be purchased credit-impaired because it has already met the criteria. Such an asset is likely to be acquired at a deep discount. However, this does not mean that an entity is required to apply the credit-adjusted EIR to a financial asset solely because the financial asset has a high credit risk at initial recognition, if it has not yet met those criteria. [IFRS 9.B5.4.7].

It may be also possible that an entity originates a credit-impaired financial asset, for example following a substantial modification of a distressed financial asset that resulted in the derecognition of the original financial asset (see 8 below). [IFRS 9.B5.5.26].

Again, this does not mean that the asset should be considered credit-impaired just because it is high risk. Consider an example of a bank originating a loan of €100,000 with interest of 30% per annum charged over the term of the loan, payable in monthly amortising instalments. The bank's customer has a high credit risk on origination and the bank expects a large portion of this type of customer to pay late or fail to pay some or all of their instalment payments. Although the loan is of high credit risk (which is supported by the high interest rate), none of the loss events listed above have occurred and the loan was not the result of a substantial modification and derecognition of a distressed debt, hence, the bank should assess the loan not to be credit-impaired on origination.

For financial assets that are considered to be credit-impaired on purchase or origination, the EIR (see Chapter 50 at 3) is calculated taking into account the initial lifetime ECLs in the estimated cash flows. [IFRS 9.B5.4.7, Appendix A, BC5.214, BC5.217]. This accounting treatment is the same as that under IAS 39 for similar assets. [IAS 39.AG5]. It is also consistent with the original method for measuring impairment proposed in the 2009 Exposure Draft.

Consequently, no allowance is recorded for 12‑month ECLs for financial assets that are credit-impaired on initial recognition. The rationale for not recording a 12‑month ECL allowance for these assets is that the losses are already reflected in the fair values at which they are initially recognised. The same logic could be applied to all the other financial assets which are not credit-impaired, arguing that they, too, are initially recognised at a fair value that reflects expectations of future losses. However, the distinction is made because the double-counting of 12‑month ECLs on initial recognition would be too large for assets with such a high credit risk since default has already occurred and the 12‑month ECLs are already reflected in the initial fair value. The exclusion of initial ECLs from the computation of the EIR would lead to a distortion that would be too significant to be acceptable.

For financial assets that were credit-impaired on purchase or origination, the credit-adjusted EIR is also used subsequently to discount the ECLs. In subsequent reporting periods an entity is required to recognise:

• in the statement of financial position, the cumulative changes in lifetime ECLs since initial recognition, discounted at the credit-impaired EIR (see 5.7 below), as a loss allowance; [IFRS 9.5.5.13, B5.5.45] and
• in profit or loss, the amount of any change in lifetime ECLs as an impairment gain or loss. An impairment gain is recognised if favourable changes result in the lifetime ECLs estimate becoming lower than the original estimate that was incorporated in the estimated cash flows on initial recognition when calculating the credit-adjusted EIR. [IFRS 9.5.5.14].

For favourable changes that result in a lower lifetime ECLs than the original estimate on initial recognition, IFRS 9 does not provide guidance on where in the statement of financial position the debit entry should be booked. In our view, the impairment gain should be recognised as a direct adjustment to the gross carrying amount. This is supported by the application guidance in IFRS 9 on revision of estimates which requires adjusting the gross carrying amount of the financial asset. [IFRS 9.B5.4.6]. For purchased or originated credit-impaired financial assets, since the ECLs are included in the estimated cash flows when calculating the credit-adjusted EIR, it would be consistent to follow the same principle and adjust the gross carrying amount when revising the original estimates of ECLs. An alternative treatment would be to recognise a negative loss allowance which would reflect the favourable changes in lifetime ECLs.

Along with the other credit risk disclosures requirements (see 15 below and Chapter 54 at 5.3), the holder is required to explain how it has determined that assets are credit-impaired (including the inputs, assumptions and estimation techniques used). It is also required to disclose the total amount of undiscounted ECLs at initial recognition for financial assets initially recognised during the reporting period that were purchased or originated credit-impaired. [IFRS 7.35H(c)].

Once a financial asset is classified as purchased or originated credit-impaired, it retains that classification until it is derecognised, i.e. when fully paid or written-off. The financial asset is never subject to a staging assessment and is required to be disclosed separately from financial assets that are subject to the general approach staging model. Therefore, a purchased or originated credit-impaired financial asset can never return to stage 1, even if the credit risk on the loan significantly improves.

The accounting treatment for a purchased credit-impaired financial asset is illustrated in the following example.

5.1 Definition of default

Default is not defined for the purposes of determining the risk of a default occurring. Because it is defined differently by different institutions (for instance, 30, 90 or 180 days past due), the IASB was concerned that defining default could result in a definition that is inconsistent with that applied internally for credit risk management. In particular, since default is the anchor point used to measure probabilities of default and losses given default in Basel modelling, requiring a different definition would require building a different set of models for accounting purposes. Therefore, the standard requires an entity to apply a definition of default that is consistent with how it is defined for normal credit risk management practices, consistently from one period to another. It follows that an entity might have to use different default definitions for different types of financial instruments. However, the standard stresses that an entity needs to consider qualitative indicators of default when appropriate in addition to days past due, such as breaches of covenant. [IFRS 9.B5.5.37].

The IASB did not originally expect ECL calculations to vary as a result of differences in the definition of default, because of the counterbalancing interaction between the way an entity defines default and the credit losses that arise as a result of that definition of default. [IFRS 9.BC5.248]. (For instance, if an entity uses a shorter delinquency period of 30 days past due instead of 60 days past due, the associated loss given default (LGD) will be correspondingly smaller as it is to be expected that more debtors that are 30 days past due will in due course recover). However, the notion of default is fundamental to the application of the model, particularly because it affects the subset of the population that is subject to the 12‑month ECL measure. [IFRS 9.BC5.249].

The standard restricts diversity resulting from this effect by establishing a rebuttable presumption that default does not occur later than when a financial asset is 90 days past due. This presumption may be rebutted only if an entity has reasonable and supportable information to support an alternative default criterion. [IFRS 9.B5.5.37, BC5.252].

A 90-day default definition would also be consistent with that used by banks for the advanced Basel II regulatory capital calculations (with a few exceptions). We observe that most banks have aligned their regulatory and accounting definitions of default. This generally means using 90 days past due under IFRS 9, with some exceptions for certain portfolios such as mortgages for which the regulatory definition may allow longer delinquency periods. Most banks also aligned the accounting definition of credit-impaired for transfer to stage 3 with the definition of default.

5.2 Lifetime expected credit losses

IFRS 9 defines lifetime ECLs as the ECLs that result from all possible default events over the expected life of a financial instrument (i.e. an entity needs to estimate the risk of a default occurring on the financial instrument during its expected life). [IFRS 9 Appendix A, B5.5.43]. The expected life considered for the measurement of lifetime ECLs cannot be longer than the maximum contractual period (including extension options at the discretion of the borrower) over which the entity is exposed to credit risk. However, there is an exception for revolving facilities (see 12 below).

ECLs should be estimated based on the present value of all cash shortfalls over the remaining expected life of the financial asset, i.e. the difference between: [IFRS 9.B5.5.29]

• the contractual cash flows that are due to an entity under the contract; and
• the cash flows that the holder expects to receive.

As ECLs take into account both the amount and the timing of payments, a credit loss arises even if the holder expects to receive all the contractual payments due, but at a later date. [IFRS 9.B5.5.28].

When estimating lifetime ECLs for undrawn loan commitments (see 11 below), the provider of the commitment needs to:

• estimate the expected portion of the loan commitment that will be drawn down over the expected life of the loan commitment. Except for revolving facilities (see 12 below), the expected life will be capped at the maximum contractual period, including extension options at the discretion of the borrower, over which the entity is exposed to credit risk (see 5.3 below for 12‑month ECLs); [IFRS 9.B5.5.31] and
• calculate the present value of cash shortfalls between the contractual cash flows that are due to the entity if the holder of the loan commitment draws down that expected portion of the loan and the cash flows that the entity expects to receive if that expected portion of the loan is drawn down. [IFRS 9.B5.5.30].

For a financial guarantee contract (see 11 below), the guarantor is required to make payments only in the event of a default by the debtor in accordance with the terms of the instrument that is guaranteed. Accordingly, the estimate of lifetime ECLs would be based on the present value of the expected payments to reimburse the holder for a credit loss that it incurs, less any amounts that the guarantor expects to receive from the holder, the debtor or any other party. If an asset is fully guaranteed, the ECL estimate for the financial guarantee contract would be consistent with the estimated cash shortfall estimate for the asset subject to the guarantee. [IFRS 9.B5.5.32].

5.3 12‑month expected credit losses

The 12‑month ECLs is defined as a portion of the lifetime ECLs that represent the ECLs that result from default events on a financial instrument that are possible within the 12 months after the reporting date. [IFRS 9 Appendix A]. The standard explains further that the 12‑month ECLs are a portion of the lifetime ECLs that will result if a default occurs in the 12 months after the reporting date (or a shorter period if the expected life of a financial instrument is less than 12 months), weighted by the probability of that default occurring. [IFRS 9.B5.5.43].

Because the calculation is based on the probability of default (PD), the standard emphasises that the 12‑month ECL is not the lifetime ECL that an entity will incur on assets that it predicts will default in the next 12 months (i.e. for which the PD over the next 12 months is greater than 50%). For instance, the PD might be only 5%, in which case this should be used to calculate 12‑month ECLs, even though it is not probable that the asset will default. Also, the 12‑month ECLs are not the cash shortfalls that are predicted over only the next 12 months. For an asset defaulting in the next 12 months, the lifetime ECLs that need to be included in the calculation will normally be significantly greater than just the cash flows that were contractually due in the next 12 months.

If the financial instrument has a maturity of less than 12 months, then the 12‑month ECLs are the credit losses expected over the period to maturity.

For undrawn loan commitments (see 11 below), an entity's estimate of 12‑month ECLs should be based on its expectations of the portion of the loan commitment that will be drawn down within 12 months of the reporting date. [IFRS 9.B5.5.31].

As already mentioned at 1.2 above, the IASB believes that the 12‑month ECLs serve as a proxy for the recognition of initial ECLs over time, as proposed in the 2009 Exposure Draft, and they mitigate the systematic overstatement of interest revenue that was recognised under IAS 39. [IFRS 9.BC5.135]. This practical approximation was necessary as a result of the decision to decouple the measurement and allocation of initial ECLs from the determination of the EIR following the re-deliberations of the 2009 Exposure Draft. [IFRS 9.BC5.199].

The stage 1, 12‑month allowance overstates the necessary allowance for each financial instrument after initial recognition. However, this is offset by the fact that the allowance is not further increased (except for changes in the 12‑month ECLs) until the instrument's credit risk has significantly increased and it is transferred to stage 2. For a portfolio of instruments, with various origination dates, the overall provision may (very approximately) be a similar size as might be achieved using a more conceptually robust approach. Although there is no conceptual justification for an allowance based on 12‑month ECLs, it was designed to be a pragmatic solution to achieve an appropriate balance between faithfully representing the underlying economics of a transaction and the cost of implementation.

How accurate a proxy the 12‑month and lifetime ECL model is for a more conceptually pure approach will depend on the nature of the portfolio. Also, the effect of recording a 12‑month ECL in the first reporting period that a financial instrument is recognised will not have a significant effect on reported income if the portfolio is stable in size from one period to the next. The 12‑month ECL allowance may, however, significantly reduce the reported income for entities which are expanding the size of their portfolio.

Although the choice of 12 months is arbitrary, it is the same time horizon as used for the more advanced bank regulatory capital calculation under the Basel framework.¹⁰ The definition of 12‑month ECLs is similar to the Basel Committee's definition of ECL, although the modelling requirements differ significantly.¹¹ The 12‑month requirement under IFRS 9 will always differ from that computed for regulatory capital purposes, as the IFRS 9 measure is a point-in-time estimate, reflecting currently forecast economic conditions (see 5.9.3 below), while the Basel regulatory figure is based on through-the-cycle assumptions of default and conservative estimates of losses given default. However, banks that use an advanced approach to calculate their capital requirements should be able to use their existing systems and methodologies as a starting point and make the necessary adjustments to flex the calculation to comply with IFRS 9.

As mentioned above, the 12‑month ECLs is defined as a portion of the lifetime ECLs that represent the ECLs that result from default events on a financial instrument that are possible within the 12‑months after the reporting date. [IFRS 9 Appendix A]. When measuring 12‑month ECLs, one question is whether the cash shortfalls used should take into account only default events within the next 12 months or subsequent default events as well. The issue arises for instruments that are expected to default and cure (i.e. to restore to performing) and then default again after curing.

IFRS 9 does not explicitly mention the treatment of cures and subsequent defaults when calculating ECLs. However, the ITG briefly talked about this in their discussion about the life of revolving credit card portfolios in April 2015: ‘As regards assets in Stage 2, it was acknowledged that the probability of assets defaulting and curing would have to be taken into account and that it would be necessary to build this into any models dealing with expected credit loss calculations. However, it was noted that materiality would need to be considered.’¹²

We conclude from the ITG discussion that cure events should only be reflected in the calculation of the LGD to the extent that they are expected to be effective. Consequently, if it is predicted that the asset will re-default in subsequent years, this need not be included in the calculation of 12‑month expected losses if the defaults are expected to be unrelated to the first default. In practice, however, IFRS 9 acknowledges that a variety of techniques can be used to meet the objective of ECL and that the definition of default may vary, by product, across a bank and between banks. [IFRS 9.B5.5.37, BC5.252, BC5.265].

When measuring ECLs, the treatment of re-defaults affects both the PD and the LGD. Therefore, the same treatment should be applied consistently to determine both the PD and the LGD.

5.4 Probability of default (PD) and loss rate approaches

As mentioned above, the standard does not prescribe specific approaches to estimate ECLs. Some of the common approaches include the PD and loss rate approaches (see 5.4.1 and 5.4.2 below, respectively).

5.4.1 Probability of default (PD) approach

Following from Example 51.1 at 1.2 above, calculations of the 12‑month and lifetime ECLs are illustrated below.

Probable scenarios	Probability	EAD	Discount rate	Expected net future cash flows	Expected recovery time	ECL of each scenario	Weighted ECL
Scenario 1: Cure	20%	1,030,000	3%	900,000	0.0	$130,000	$26,000
Scenario 2: Restructure	40%	1,030,000	3%	800,000	0.5	$241,737	$96,695
Scenario 3: Liquidation	40%	1,030,000	3%	700,000	1.0	$350,388	$140,155
Weighted average ECL						Lifetime ECL	$262,850
						% of EAD:	26%

Most sophisticated banks have developed their IFRS 9 solutions by adjusting and extending their Basel models. This is true for all types of model component: PD, LGD and EAD. This is perhaps unsurprising given the historical investment large banks have made in their Basel models, and the fact that IFRS 9 shares fundamental similarities in expected loss modelling. But, for many banks, creating lifetime estimates and altering models to satisfy the complex and detailed IFRS 9 requirements necessitates significant work.

However, many non-financial entities do not have models and systems in place that capture such information. One possibility is to make use of historical default rates as collected by credit rating agencies such as Standard and Poor's. This is illustrated in the example below.

5.5 Expected life versus contractual period

Lifetime ECLs are defined as the ECLs that result from all possible default events over the expected life of a financial instrument. [IFRS 9 Appendix A]. This is consistent with the requirement that an entity should assess whether the credit risk on a financial instrument has increased significantly since initial recognition by using the change in the risk of a default occurring over the expected life of the financial instrument. [IFRS 9.5.5.9].

An entity must therefore estimate cash flows and the instrument's life by considering all contractual terms of the financial instrument (for example, prepayment, extension, call and similar options). There is a presumption that the expected life of a financial instrument can be estimated reliably. In those rare cases when it is not possible to reliably estimate the expected life of a financial instrument, the entity shall use the remaining contractual term of the financial instrument. [IFRS 9 Appendix A, B5.5.51].

However, the maximum period to consider when measuring ECLs should be the maximum contractual period (including extension options) over which the entity is exposed to credit risk and not a longer period, even if that longer period is consistent with business practice. [IFRS 9.5.5.19]. Although an exception to this principle has been added for revolving facilities (see 12 below), the IASB remains of the view that the contractual period over which an entity is committed to provide credit (or a shorter period considering prepayments) is the correct conceptual outcome. The IASB noted that most loan commitments will expire at a specified date, and if an entity decides to renew or extend its commitment to extend credit, it will be a new instrument for which the entity has the opportunity to revise the terms and conditions. [IFRS 9.BC5.260].

This means that extension options should only be reflected in the measurement of ECLs as long as this does not extend the horizon beyond the maximum contractual period over which the entity is exposed to credit risk. Extension options at the discretion of the lender should therefore be excluded from the measurement of ECLs. Similarly, a lender's ability to require prepayment limits the horizon over which it is exposed to credit risk. The first prepayment date at the discretion of the lender should therefore represent the maximum period to be reflected in the expected loss calculation.

When assessing the impact of extension options at the discretion of the borrower, an entity should estimate both the probability of exercise of the extension option as well as the portion of the loan that will be extended (if the extension option can be exercised for a portion of the loan only). This is consistent with how lifetime expected losses must be assessed for loan commitments where an entity's estimate of ECLs must be consistent with its expectations of drawdowns on that loan commitment. Although the standard is not explicit on this point, the effect of extension options is best modelled not by estimating an average life of the facility but by estimating the EAD each year over the maximum lifetime. This is because use of an average life would not reflect losses expected to occur beyond the average life. [IFRS 9.B.5.5.31].

Expected prepayments at the discretion of borrowers should also be reflected in the measurement of ECLs. As with extension options, an entity must estimate both the probability of exercise of the prepayment option as well as the portion of the loan that will be prepaid (if the prepayment option can be exercised for a portion of the loan only). As with extension options, the standard does not specify whether prepayment patterns should be reflected through an amortising EAD over the maximum contractual period of the financial instruments or, rather, by shortening the horizon over which to measure ECLs to the average life of the financial instruments. Similar to the treatment of extension options, described above, in our view it is more appropriate to adjust the EAD for the facility each year over the maximum lifetime. We consider this a more transparent way of incorporating product features and potential impacts of different macroeconomic scenarios that can, for example, affect pre-payment patterns and customers' ability to refinance.

Further complexity in assessing expected prepayments and extensions arises if one considers that the behaviour of borrowers is affected by their creditworthiness. This means that prepayment and extension patterns should probably be estimated separately for stage 1 and stage 2 assets. This may represent a significant challenge, as making such estimates would require distinct historical observations for each of the stage 1 and 2 populations, which are unlikely to be available given that these populations were never identified in the past. Prepayment assumptions for stage 2 assets would need to factor in the probabilities that some may subsequently default and some may cure. A further complication is that expected prepayment and extension behaviour may vary with changes in the macroeconomic outlook.

The standard is clear that, for loan commitments and financial guarantee contracts, the time horizon to measure ECLs is the maximum contractual period over which an entity has a present contractual obligation to extend credit. [IFRS 9.B5.5.38]. However, for revolving credit facilities (e.g. credit cards and overdrafts), as an exception to the normal rule, this period is extended beyond the maximum contractual period and includes the period over which the entity is exposed to credit risk and ECLs would not be mitigated by credit risk management actions (see 12 below). This exception is limited to facilities that include both a loan and an undrawn commitment component, that do not have a fixed term or repayment structure and usually have a short contractual cancellation period (for example, one day). [IFRS 9.5.5.20, B5.5.39, B5.5.40].

At its April 2015 meeting, the ITG discussed how to determine the maximum period for measuring ECLs, by reference to the following example.¹⁴

The ITG noted that:

• IFRS 9 is clear that the maximum period to consider when measuring ECLs in this example would be restricted to 6 months, because this is the maximum contractual period over which the lender is exposed to credit risk, i.e. the period until the lender can next object to an extension. [IFRS 9.5.5.19].
• The standard requires that extension options must be considered when determining the maximum contractual period, but does not specify whether these are lender or borrower extension options. However, if the extension option is within the control of the lender, the lender cannot be forced to continue extending credit and therefore such an option cannot be considered as lengthening the maximum period of exposure to credit risk. Conversely, if a borrower holds an extension option that could force the lender to continue extending credit, this would have the effect of lengthening that maximum contractual period of credit exposure.
• The maximum contractual period over which the entity is exposed to credit risk should be determined in accordance with the substantive contractual terms of the financial instrument. To further illustrate this point, a situation in which a lender is legally prevented from exercising a contractual right should be seen as distinct from a situation in which a lender chooses not to exercise a contractual right for practical or operational reasons.
• In the example presented, the facility is not of a revolving nature and the borrower does not have any such flexibility regarding drawdowns. Consequently, it would not be appropriate to analogise the 6-month mortgage loan to a revolving credit facility that has been fully drawn at the reporting date. Hence, the example falls outside the narrow scope exception for revolving credit facilities (e.g. credit cards and overdraft facilities) in which the maximum period to consider when measuring ECLs is over the period that the entity is exposed to credit risk and ECLs would not be mitigated by credit risk management actions, even if that period extends beyond the maximum contractual period (see 12 below). [IFRS 9.5.5.20].
• Consequently, it was acknowledged that there may be a disconnect between the accounting and credit risk management view in some situations (e.g. an entity may choose to continue extending credit to a long-standing customer despite being in a position to reduce or remove the exposure). See further discussion on the application of the revolving credit facilities exception to multi-purpose facilities (see 12 below).

For demand deposits that have no fixed maturity and can be withdrawn by the holder on very short notice (e.g. one day) (assuming there is no contractual or legal constraint that could prevent the holder from withdrawing its cash at any time), the period used by the holder of such demand deposits to estimate ECLs would be limited to the contractual notice period, i.e. one day. This is the maximum contractual period over which the holder is exposed to credit risk. In accordance with paragraph 5.5.19 of IFRS 9, extension periods at the option of the holder are excluded in estimating the maximum contractual period because the holder can unilaterally choose not to extend credit and thus can limit the period over which it is exposed to credit risk. Furthermore, demand deposits do not fall under the revolving credit facility exception (see 12 below) as they do not comprise an undrawn element. [IFRS 9.5.5.20].

5.6 Probability-weighted outcome and multiple scenarios

ECLs must reflect an unbiased and probability-weighted estimate of credit losses over the expected life of the financial instrument (i.e. the weighted average of credit losses with the respective risks of a default occurring as the weights). [IFRS 9.5.5.17(a), Appendix A, B5.5.28].

The standard makes it clear that when measuring ECLs, in order to derive an unbiased and probability-weighted amount, an entity needs to evaluate a range of possible outcomes. [IFRS 9.5.5.17(a)]. This involves identifying possible scenarios that specify:

1. the amount and timing of the cash flows for particular outcomes; and
2. the estimated probability of these outcomes.

Although an entity does not need to identify every possible scenario, it will need to take into account the possibility that a credit loss occurs, no matter how low that probability is. [IFRS 9.5.5.18]. This is not the same as a single estimate of the worst-case or best-case scenario, or the most likely outcome (i.e. when there is a low risk or probability of a default (PD) with high loss outcomes, the most likely outcome could be no credit loss even though an allowance would be required based on probability-weighted cash flows). [IFRS 9.B5.5.41]. It is worthwhile noting that it is implicit that the sum of the weighted probabilities will be equal to one. A simple example of application of a probability-weighted calculation is shown in Example 51.7.

Without taking into account multiple economic scenarios (see below) calculating a probability-weighted amount may not require a complex analysis or a detailed simulation of a large number of scenarios and the standard suggests that relatively simple modelling may be sufficient. For instance, the average credit losses of a large group of financial instruments with shared risk characteristics may be a reasonable estimate of the probability-weighted amount. In other situations, the identification of scenarios that specify the amount and timing of the cash flows for particular outcomes and the estimated probability of those outcomes will probably be needed. In those situations, the ECLs shall reflect at least two outcomes in accordance with paragraph 5.5.18 of IFRS 9. [IFRS 9.B5.5.42].

At the December 2015 ITG meeting, the question was asked as to whether the use of multiple scenarios referred to in the standard relates only to what might happen to particular assets given a single forward-looking economic scenario (i.e. default or no default), or whether application of the standard requires an entity to use multiple forward-looking economic scenarios, and if so how.

The ITG members noted that the measurement of ECLs is required to reflect an unbiased and probability-weighted amount that is determined by evaluating a range of possible outcomes. Consequently, when there is a non-linear relationship between the different forward-looking scenarios and their associated credit losses, using a single forward-looking economic scenario would not meet this objective. In such cases, more than one forward-looking economic scenario would need to be used in the measurement of ECLs.¹⁵ For each scenario the associated ECLs would need to be multiplied by the weighting allocated to that scenario.

The ITG also discussed the use of multiple economic scenarios to assess whether exposures should be measured using lifetime economic losses (see 6.7 below). It was noted by the ITG that if the same variable is relevant for determining significant increase in credit risk and for measuring ECLs, the same forward-looking scenarios should be used for both.

The ITG discussed a particular example in which there are considered to be three possible economic scenarios.¹⁶ This is illustrated in the example below.

The ITG was concerned about the distribution of possible losses often being ‘non-linear’, in that the increase in losses associated with those economic scenarios that are worse than the central forecast will be greater than the reduction in losses associated with those scenarios that are more benign. To use statistical terminology, the distribution is skewed. Depending on how it is calculated, a single scenario gives the mode of this distribution (i.e. the most likely outcome) or the median (the central forecast). In contrast, the standard requires the use of the mean (i.e. a probability-weighted estimation). A possible distribution of the losses in the portfolio consistent with the above example is shown in Figure 51.4 below.

At the ITG meeting, it was noted that there are a number of possible approaches that might be used to incorporate multiple economic approaches. IFRS 9 does not prescribe any particular method of measuring ECLs and the measurement should reflect an entity's own view. What the standard does require is that the expected losses must reflect:

1. an unbiased and probability-weighted amount that is determined using a range of possible outcomes; and
2. reasonable and supportable information that is available without undue cost or effort at the reporting date.

With respect to reasonable and supportable information, ITG members made the following observations:

1. although IFRS 9 does not specifically require an entity to consider external information, an entity should consider information from a variety of sources in order to ensure that the information used is reasonable and supportable;
2. the information considered could vary depending on the facts and circumstances including the level of sophistication of the entity, geographical region and the particular features of the portfolio; and
3. while entities are not expected to consider every possible scenario, the scenarios considered should reflect a representative sample of possible outcomes.¹⁷

ITG members recognised that materiality considerations would need to be taken into account.

In an IASB webcast on 25 July 2016, it was noted that, having considered:

1. whether the effect of non-linearity is material;
2. whether the entity has a reasonable and supportable basis for this multiple scenario analysis; and
3. whether the application is possible without undue cost or effort;

A conclusion may sometimes be reached that it is not necessary to actually use multiple scenarios to apply the impairment requirements in IFRS 9. However, multiple scenarios must always be considered.

At the December 2015 ITG meeting, the ITG also noted that consideration should be given to the consistency of forward-looking information used for the measurement of ECLs and for other purposes within the organisation, such as budgeting and forecasting. ITG members acknowledged that there might be differences, but observed that these should be understood and explainable.

ITG members also observed that the incorporation of forward-looking scenarios will require judgement. Consequently, they emphasised the importance of the IFRS 7 disclosure requirements relating to how forward-looking information has been incorporated into the determination of ECLs (see Chapter 54 at 5.3).¹⁸

Since December 2015, banks have given significant attention to how multiple economic scenarios can be incorporated into ECL calculations. We have seen three main approaches, as follows:

1. Probability weighted scenarios – this is similar to the method discussed at the ITG meeting in December 2015 and illustrated in Example 51.7 above. It involves establishing a number of scenarios (typically three scenarios but we have seen varying numbers, generally between two and four), estimating the losses that would arise in those scenarios and allocating a weighting to each scenario. Unlike Example 51.7 above, these do not normally model economic variables such as unemployment rates in isolation – to do so, would also require complex modelling of the correlations between those variables. Instead, each scenario is normally a coherent combination of economic variables. For example, a scenario relevant to mortgage loans might include assumptions about unemployment, interest rates and house prices. This approach is transparent, but it may be difficult to assign the weightings to each scenario, requiring judgement as well as experience of the past. While selecting scenarios and respective weights, we expect banks to take into consideration the entire distribution of macroeconomic scenarios and select points (i.e. scenarios) from that distribution, with their respective weights representing the portion of the distribution represented by the scenario. We would also expect that the mean of the selected scenarios and weights is similar to that of the entire distribution.
2. The second approach is to calculate ECLs based on a central forward-looking scenario and to adjust the outcome where necessary by a factor to reflect the non-linearity of the loss distribution. In practice it may be that a method similar to (a) above will need to be used in order to calculate this factor – so that it is not a very different approach. However, some banks view the merits of this approach as being less mechanistic and allowing more room for judgement.
3. Monte Carlo simulation – this method seeks to calculate the expected losses associated with the entire distribution of possible scenarios around the bank's central economic forecast. It has the advantage that it does not require the bank to formulate specific scenarios or assign weightings to them, but the simulation is dependent on assumptions that may not be transparent to either users or preparers, so that this solution can seem a ‘black box’. It is also very demanding as to the volume of data that has to be manipulated and it is not how most banks manage credit risk today. This method is quite rarely applied in practice.

The effect of multiple scenarios will affect not just the probability of default but also the losses given default. For instance, for property-based lending it will be necessary to forecast the value of collateral associated with each economic scenario that is modelled. A consequence of this is that there may be a need to record an ECL allowance for an asset that, based on the central forecast of future collateral values, is fully collateralised. (Also, as a result, the loss allowance for a stage 3 asset may be higher than for an impaired asset under IAS 39).

The use of multiple scenarios may also have an effect on the estimated EAD.

A number of other observations can be made about the use of multiple scenarios:

1. Whichever approach is used to calculate the effect of non-linearity, it will be necessary for banks to communicate the result of the calculation in a manner which can be understood by readers of the financial statements. One possible approach would be for banks to report the losses associated with the central forecast and then, separately, the effect of the consideration of other scenarios. This would allow banks to communicate the amounts they expect to lose and would permit comparison between banks of the effect of the adjustment for non-linearity, even if the banks use different methods to make the calculation.
2. It would seem that the effects of non-linearity depend on the countries in which banks operate and the economic characteristics of those countries. For instance, the effect of alternative scenarios of interest rates and unemployment may be greater in countries where there is more of a ‘boom and bust’ economic cycle. The size of the effect is also dependent on origination practices and the particular lending products – variable rate loans being more sensitive to interest rates than fixed-rate ones, while defaults on credit cards are more affected by unemployment rates. In some cases the issue is seen as most relevant for exposures to a particular economic variable, a topical example being lending to companies involved in the oil industry. In this example, banks might model a number of scenarios as to how oil prices could evolve. A similar approach may be relevant for non-banks with similar exposures through long term construction contracts or leasing activities. There is also more likely to be non-linearity in the calculation of ECLs when exposures are collateralised by assets whose values also change in response to the economic conditions that drive the probability of default. An example is residential mortgage loans.
3. It should be stressed that the ITG discussion highlighted the importance of calculating the effect of non-linearity using only reasonable and supportable information, implying that if the information is not available then there is a limit to what can be done. However, banks will also need to take into account their regulators' expectations (see 1.6 above and 7.1 below for Basel Committee guidance).

The process of forecasting future economic conditions is discussed further in 5.9.3 below.

5.7 Time value of money

An entity needs to consider the time value of money when measuring ECLs, by discounting the estimated losses to the reporting date using a rate that approximates the EIR of the asset. [IFRS 9.5.5.17, B5.5.44]. This has two components:

• discounting recoveries to the date of default, hence ‘a credit loss arises even if the entity expects to be paid in full but later than when contractually due’; [IFRS 9.B5.5.28] and
• discounting losses from the date of default to the reporting date. This is needed as the gross amortised cost of the asset is based on the contractual cash flows discounted at the EIR, and therefore not discounting cash flows that are now not expected to be received would overstate the loss.

It is rare that customers just fail to pay amounts when due. In most cases, default also involves payments being paid late, while default can lead to the acceleration of payment of amounts that are not contractually due until a later date. Therefore, modelling losses involves modelling the timing of payments when default occurs and different patterns of timing of recoverable cash flows, such as the time it takes to foreclose on and sell collateral and complete bankruptcy proceedings, before the ECLs can be discounted back to the reporting date.

Of these two components, the first is typically included by banks in their calculation of the LGD (although not necessarily using the EIR). However the second will also need to be calculated to comply with the standard.

The standard and its illustrative examples are silent on how the calculation should be made. In Illustrative Example 9 the present value of the observed loss is assumed and in Illustrative Example 8, a footnote states that, ‘because the LGD represents a percentage of the present value of the gross carrying amount, this example does not illustrate the time value of money’.

One approach would be to model various scenarios as to how cash is collected once the loan has defaulted, and probability-weight the discounted cash flows of these various scenarios.

The discount rate is calculated as follows:

• for a fixed-rate financial asset, entities are required to determine or approximate the EIR on the initial recognition of the financial asset, while for a floating-rate financial asset, entities are required to use the current EIR; [IFRS 9.B5.5.44]
• for a purchased or originated credit-impaired financial asset (see 3.3 above), entities are required to discount ECLs using the credit-adjusted EIR determined on the initial recognition of the financial asset; [IFRS 9.B5.5.45]
• for a loan commitment (see 11 below), entities are required to use the EIR of the asset that will result once the commitment is drawn down. This would give rise to a consistent rate for a credit facility that includes both a loan (i.e. a financial asset) and an undrawn commitment (i.e. a loan commitment). If the EIR of the resulting asset is not determinable, then entities are required to use the current risk-free rate (i.e. the discount rate that reflects the current market assessment of the time value of money). This should be adjusted for risks specific to the cash flows, but only if the cash flows have not already been adjusted for these risks, in order to avoid double counting; [IFRS 9.B5.5.47, B5.5.48]
• for financial guarantee contracts (see 11 below) entities are required to use the current risk-free rate adjusted for risks specific to the cash flows, again to the extent that those cash flows have not already been adjusted for the risks; [IFRS 9.B5.5.48] and
• for lease receivables (see 10.2 below), entities are required to discount the ECLs using the same discount rate used in the measurement of the lease receivables in accordance with IFRS 16. [IFRS 9.B5.5.46].

LGD data available from Basel models should include a discounting factor and sometimes this may be different from the rate required by IFRS 9. Furthermore, the discount rate used in Basel models only covers the period between default and subsequent recoveries. Therefore, entities will have to find ways to adjust their LGDs to reflect the discounting effect required by the standard (i.e. based on a rate that approximates the EIR and over the entire period from recoveries back to the reporting date). Given the requirement to use an approximation to the EIR, entities will need to work out how to determine a rate that is sufficiently accurate. One of the challenges is to interpret how much flexibility is afforded by the term ‘approximation’.

At its meeting in December 2015, the ITG also discussed what was meant by the current EIR when an entity recognises interest revenue in each period based on the actual floating-rate applicable to that period. The ITG first noted that the definition of the EIR in IFRS 9 was carried forward essentially unchanged from the definition within IAS 39. Consequently, similarly to IAS 39, IFRS 9 does not specify whether an entity should use the current interest rate at the reporting date or the projected interest rates derived from the current yield curve as at the reporting date. There should be consistency between the rate used to recognise interest revenue, the rate used to project future cash flows (including cash shortfalls) and the rate used to discount those cash flows (see 5.7 above).

In relation to the guidance in paragraphs B5.5.47 and B5.5.48 on loan commitments when the EIR on the resulting asset is not determinable and for financial guarantee contracts, we make the following observations:

• Although it is not clear in the standard, any adjustment for the risks specific to the cash flows would be a reduction of the risk free rate, not an increase. This would be consistent with the approach applied to provisions in IAS 37 and as was made clear in the staff paper presented to the Board when this treatment was discussed in December 2013. For financial guarantee contracts, the reduction in the risk-free discount rate will increase the present value of the obligation to pay claims to the guarantee holder. This reflects the additional compensation that would be demanded to take on this risky obligation, in particular to bear the risk that claims payments will be higher than the probability-weighted expected amount.
• For loan commitments when the EIR on the resulting asset is not determinable, this approach provides a prudent calculation of ECLs, given that it is likely that the entity which enters into the commitment will receive a credit spread on the loan if it is drawn down. It is in a much better position than the issuer of a financial guarantee contract, who will receive no credit spread should it be required to pay out on the guarantee.
• The idea that the rate should be adjusted only if the cash flows have not already been adjusted for the risks may not be easy to apply in practice. This is because the cash flows should have already been estimated with regard to any non-linearities in the distribution of losses (see 5.6 above) and so will already have been partly adjusted for risk. It may not be easy to calculate the necessary adjustment to reflect a market assessment of the remaining risks.

5.8 Losses expected in the event of default

This section discusses the measurement of ECLs taking into account credit enhancements such as collateral and financial guarantees, cash flows from the sale of a defaulted loan and collection costs paid to an external debt collection agency.

5.9 Reasonable and supportable information

IFRS 9 requires an entity to consider reasonable and supportable information that is available without undue cost or effort at the reporting date about past events, current conditions and forecasts of future economic conditions and that is relevant to the estimate of ECLs, including the effect of expected prepayments. [IFRS 9.5.5.17(c), B5.5.51].

6.1 Change in the risk of a default occurring

In order to make the assessment of whether there has been significant credit deterioration, an entity should consider reasonable and supportable information that is available without undue cost or effort and compare: [IFRS 9.5.5.9]

• the risk of a default occurring on the financial instrument over its life as at the reporting date; and
• the risk of a default occurring on the financial instrument over its life as at the date of initial recognition.

For loan commitments, an entity should consider changes in the risk of a default occurring on the potential loan to which a loan commitment relates.

For financial guarantee contracts, an entity should consider the changes in the risk that the specified debtor will default. [IFRS 9.B5.5.8].

An entity is required to assess significant increases in credit risk based on the change in the risk of a default occurring over the expected life of the financial instrument rather than the change in the amount of ECLs. [IFRS 9.5.5.9]. It should be noted that in a departure from the Basel regulatory wording and to avoid suggesting that statistical models are required (including the PD approach), the IASB changed the terminology from ‘probability of a default occurring’ to ‘risk of a default occurring’. [IFRS 9.BC5.157].

In order to make the IFRS 9 impairment model operational, the IASB considered a number of alternative methods for determining significant increases in credit risk, but these were rejected for the following reasons:

• Absolute level of credit risk: The IASB considered whether an entity should be required to recognise lifetime ECLs on all financial instruments at, or above, a particular credit risk at the reporting date. Although this approach is operationally simpler to apply (because an entity is not required to track changes in credit risk), such an approach would provide very different information. It would not approximate the economic effect of changes in credit loss expectations subsequent to initial recognition. In addition, it may also result in overstatement or understatement of ECLs, depending on the threshold set for recognising lifetime ECLs. [IFRS 9.BC5.160]. However, the IASB noted that an absolute approach could be used for portfolios of financial instruments with similar credit risk at initial recognition, by determining the maximum initial credit risk accepted and then comparing the maximum initial credit risk to the credit risk at the reporting date (see 6.4.5 below). [IFRS 9.BC5.161].
• Change in the credit risk management objective: The IASB also considered whether the assessment of significant deterioration should be based on whether an entity's credit risk management objective changes (e.g. monitoring of financial assets on an individual basis, or a change from collecting past due amounts to the recovery of these amounts). This approach is operationally relatively easy to apply. However, it is likely to have a similar effect to the IAS 39 incurred loss model and, hence, may result in a delayed recognition of ECLs. [IFRS 9.BC5.162].
• Credit underwriting policies: The IASB further considered whether the change in the entity's credit underwriting limit for a particular class of financial instrument at the reporting date (i.e. an entity would not originate new loans on the same terms) should form the basis of assessing significant increase in credit risk. The IASB noted that this approach is similar to the absolute approach above. Moreover, the change in an entity's credit underwriting limits may be driven by other factors that are not related to a change in the credit risk of its borrowers (e.g. the entity may incorporate favourable terms to maintain a good business relationship or to increase lending), or are dependent on circumstances existing at the reporting date that are not relevant to the particular vintages of financial instruments. [IFRS 9.BC5.163, BC5.164, BC5.165].

Similar to measuring ECLs, an entity may use different approaches when assessing significant increases in credit risk for different financial instruments. An approach that does not include PD as an explicit input can be consistent with the impairment requirements as long as the entity is able to separate the changes in the risk of a default occurring from changes in other drivers of ECLs (e.g. collateral) and considers the following when making the assessment: [IFRS 9.B5.5.12]

• the change in the risk of a default occurring since initial recognition;
• the expected life of the financial instrument; and
• reasonable and supportable information that is available, without undue cost or effort, that may affect credit risk.

In addition, because of the relationship between the expected life and the risk of default occurring, the change in credit risk cannot be assessed simply by comparing the change in the absolute risk of default over time, because the risk of default usually decreases as time passes if the credit risk is unchanged. [IFRS 9.B5.5.11].

Entities that do not use probability of loss as an explicit input will have to use other criteria to identify a change in the risk of default occurring. These might include deterioration in a behavioural score, or other indicators, of a heightened risk of default. A collective approach may also be an appropriate supplement or substitute for an assessment at the individual instrument level (see 6.5 below).

A number of operational simplifications and presumptions are available to help entities make this assessment (as described further below).

6.2 Factors or indicators of changes in credit risk

Similar to measuring ECLs (see 5 above), when assessing significant increases in credit risk, an entity should consider all reasonable and supportable information that is available without undue cost or effort (see 5.9.1 above) and that is relevant for an individual financial instrument, a portfolio, portions of a portfolio, and groups of portfolios. [IFRS 9.B5.5.15, B5.5.16].

The IASB notes that it did not intend to prescribe a specific or mechanistic approach to assess changes in credit risk and that the appropriate approach will vary for different levels of sophistication of entities, the financial instrument and the availability of data. [IFRS 9.BC5.157]. It is important to stress that the assessment of significant increases in credit risk often involves a multifactor and holistic analysis. The importance and relevance of each specific factor will depend on the type of product, characteristics of the financial instruments and the borrower as well as the geographical region. [IFRS 9.B5.5.16]. The guidance in the standard is clear that in certain circumstances, qualitative and non-statistical quantitative information may be sufficient to determine that a financial instrument has met the criterion for the recognition of lifetime ECLs. That is, the information does not need to flow through a statistical model or credit ratings process in order to determine whether there has been a significant increase in the credit risk of the financial instrument. In other cases, the assessment may be based on quantitative information or a mixture of quantitative and qualitative information. [IFRS 9.B5.5.18].

6.3 What is significant?

The assessment of whether credit risk has significantly increased depends, critically, on an interpretation of the word ‘significant’. Some constituents who commented on the 2013 Exposure Draft requested the IASB to quantify the term significant, however, the IASB decided not to do so, for the following reasons: [IFRS 9.BC5.171, BC5.172]

• specifying a fixed percentage change in the risk of default would require all entities to use the risk of default approach. As not all entities (apart from regulated financial institutions) use PDs as an explicit input, this would have increased the costs and effort for those entities that do not use such an approach; and
• defining the amount of change in the risk of a default occurring would be arbitrary and this would depend on the type of products, maturities and initial credit risk.

The standard emphasises that the determination of the significance of the change in the risk of a default occurring depends on:

• the original credit risk at initial recognition: the same absolute change in PD for a financial instrument with a lower initial credit risk will be more significant than those with a higher initial credit risk (see 6.1 above and Example 51.15 below); [IFRS 9.B5.5.9] and
• the expected life or term structure: the risk of a default occurring for financial instruments with similar credit risk increases the longer the expected life of the financial instruments. Due to the relationship between the expected life and the risk of a default occurring, an entity cannot simply compare the absolute risk of a default occurring over time. For example, if the risk of a default occurring for a financial instrument with an expected life of 10 years at initial recognition is the same after five years, then this indicates that the credit risk has increased. The standard also states that, for financial instruments that have significant payment obligations close to the maturity of the financial instrument (e.g. those where the principal is only repaid at maturity), the risk of a default occurring may not necessarily decrease as time passes. In such cases, an entity needs to consider other qualitative factors. We note, however, that while the risk of default may decrease less quickly for an instrument with payment obligations throughout its contractual life, normally, the risk of default will still decrease as maturity approaches. [IFRS 9.B5.5.10, B5.5.11].

Some of these challenges are illustrated by examining the historical levels of default associated with the credit ratings of agencies, such as Standard & Poor's.

• It is apparent that the PDs increase at a geometrical, rather than an arithmetic, rate as the credit ratings decline. Hence, the absolute increase in the PD between two relatively low risk credit ratings is considerably less than between two relatively higher risk ratings.
• The relative increase in PD between each of these ratings might be considered significant, since most involve a doubling or trebling of the PD. In contrast, because credit rating is an art rather than a science, the smaller changes in credit risk associated with the plus or minus notches in the grading system are less likely to be viewed as significant.
• In addition, as the time horizon increases, the PDs also increase across all credit ratings (i.e. the PD increases with a longer maturity).

The majority of credit exposures that are assessed for significant credit deterioration will not have been rated by a credit rating agency. However, the same logic will apply when entities have developed their own PD models and are able to classify their exposure by PD levels.

The determination of what is significant will, for the larger banks, be influenced by the guidance issued by banking regulators (see 7.1 below).

Given the exponential shape of the PD curve relative to ratings, some banks are considering that a bigger downgrade, as measured by the number of grades, would be significant for a higher quality loan than for one with a lower quality. The extent to which this is appropriate will depend on how the different grades map to PDs. Also, the calibration of a significant deterioration has to take into account the fact that PD multiples for very good ratings only represent very small movements in absolute risk, whereas the same multiple applied to bad ratings can represent a significant change in the absolute amount of PD.

Banks have varying views on how much of an increase in PD is significant. Also, while they may view a quantitative threshold, such as a doubling of PD, to be significant, many also require a minimum absolute PD increase, such as 50 basis points so as to avoid very high quality assets moving to stage 2 as a result of a very small change and to filter out ‘noise’.

Banks have also introduced various metrics to assess the effect of different approaches to assess significant increase in credit risk and for management information. Examples include the volume of stage 2 assets compared to the total portfolio and compared to 12‑months of lifetime expected losses, the volume of movement (back and forth) between stages 1 and 2, the amount of assets that jump directly from stage 1 to stage 3, the proportion of assets in stage 3 which went via stage 2, and how long assets were in stage 2 before moving to stage 3.

6.4 Operational simplifications

When assessing significant increases in credit risk, there are a number of operational simplifications available. These are discussed below.

6.5 Collective assessment

Banks may have hundreds of thousands, or even millions, of small exposures to retail customers and small businesses. Much of the information available to monitor them is based on whether payments are past due and behavioural information that is mostly historical rather than forward-looking. As a result such exposures tend to be managed on an aggregated basis, combining past due and behavioural data with historical statistical experience and sometimes macroeconomic indicators, such as interest rates and unemployment levels, that tend to correlate with future defaults. Also, even when exposures are managed on an individual basis, as is the case for most commercial loans, the information used to manage them may not be sufficiently forward-looking to comply with the standard.

To address these concerns, the standard introduces the idea of making a collective assessment for financial assets, to determine if there has been a significant increase in credit risk, if an entity cannot make the assessment adequately on an individual instrument level. This exercise must consider comprehensive information that incorporates not only past due data but other relevant credit information, such as forward-looking macroeconomic information. The objective is to approximate the result of using comprehensive credit information that incorporates forward-looking information at an individual instrument level. [IFRS 9.B5.5.4]. Hence, even if a financial asset is normally managed on an individual basis, it should also be assessed collectively (i.e. based on macroeconomic indicators), if the entity does not have sufficient forward-looking information at the individual level to make the determination. The way that this might work is not very different from the IAS 39 requirement to assess an asset collectively for impairment if it has already been assessed individually and found not to be impaired.

Some kind of collective adjustment or overlay will be needed for many retail lending portfolios, given that most customer-specific information will not be forward-looking. In contrast, for commercial loans, the lender will typically have access to much more information and a forward-looking approach may already have been built into loan grading systems. Nevertheless, we are aware of some banks who consider that they need to introduce an additional overlay for commercial loans so as to be more responsive to emerging macroeconomic and other risk developments. Other banks using their existing watch list approaches to supplement their credit grading system when assessing whether there has been a significant increase in credit risk. This is because watch list systems can be more reactive to changing circumstances than formal credit gradings. Any one bank is likely to employ a variety of methods, depending on its products, systems and data.

It is worth noting that the language on when a collective approach is required is not entirely consistent within the standard. Paragraph B5.5.1 states that ‘it may be necessary to perform the assessment’ on a collective basis, which is consistent with the requirement in paragraph 5.5.11, that ‘an entity cannot rely solely on past due information if reasonable and supportable forward-looking information is available without undue cost or effort’. However, paragraph B5.5.4 states that if ‘an entity does not have reasonable and supportable information that is available without undue cost or effort to measure lifetime expected credit losses on an individual instrument basis … lifetime expected credit losses shall be recognised on a collective basis’ (emphasis added for each quotation). Banking regulators will probably ensure that this ‘shall be’ wording will be applied, at least for more sophisticated banks (see 1.6 above and 7.1 below). This raises a second concern: once significant deterioration has been identified for a portfolio, whether the entire portfolio would have to be measured using lifetime ECLs. This outcome would result in sudden, massive increases in provisions as soon as conditions begin to decline. Consequently the Board, in finalising the standard, set out examples of possible methods, using which only a segment or portion of the portfolio would be changed to lifetime ECLs.

Illustrative Example 5 in the Implementation Guidance for the standard illustrates how an entity may assess whether its individual assessment should be complemented with a collective one whenever the information at individual level is not sufficiently comprehensive and up-to-date. The following examples have been adapted from that guidance.

6.6 Determining the credit risk at initial recognition of an identical group of financial assets

In practice, entities may hold a portfolio of debt securities that are identical and cannot be distinguished individually (e.g. all securities have the same international securities identification number (ISIN)) and over the lifetime of the portfolio, entities may acquire additional securities or sell some of those previously acquired. In such instances, entities have to determine the credit risk at initial recognition of those securities that remain in the homogeneous portfolio at the reporting date.

IFRS 9 contains no specific guidance on how to calculate the cost of financial assets for derecognition purposes when they are part of a homogenous portfolio. Under IAS 39, which was also silent on this topic, entities used to choose between three cost allocation methods for available-for-sale securities: the average cost method, the first-in-first-out (FIFO) method or the specific identification method. Specific identification can be applied if the entity is able to identify the specific items sold and their costs. For example, a specific security may be identified as sold by linking the date, amount and cost of securities bought with the sale transaction, provided that there is no other evidence suggesting that the actual security sold was not the one identified under this method.

For IFRS 9, the question arises whether entities can continue to apply one of the above methods for debt instruments, not only for determining the cost of the security at derecognition but also for determining their initial credit risk. We believe that:

• the method used for recognising and measuring impairment losses should normally be the same as that used for determining the cost allocation method on derecognition;
• a FIFO approach or a specific identification method as described above constitute acceptable accounting policy choices to be applied consistently; however
• it would not normally be appropriate to use the weighted-average method to determine the credit risk at initial recognition, as averaging the different levels of initial credit risk of debt securities purchased at different dates would result in an identical initial credit risk for each item. It therefore, would create bias when assessing whether the credit risk of debt securities has increased significantly.

6.7 Multiple scenarios for the assessment of significant increases in credit risk

At its December 2015 meeting the Impairment Transition resource Group (ITG) discussed not only the need to consider multiple scenarios for measurement of ECLs (see 5.6 above), but also for the purposes of assessing whether exposures should be measured on a lifetime loss basis.

Similar to the measurement of ECLs, the ITG members noted that where there is a non-linear relationship between the different forward-looking scenarios and the associated risks of default, using a single scenario would not meet the objectives of the standard. Consequently, in such cases, an entity would need to consider more than one forward-looking scenario. Further, there should be consistency, to the extent relevant, between the information used to measure ECLs and that used to assess significant increases in credit risk. An example of when the information might not be relevant is the value of collateral. It may be necessary to calculate the effect of multiple scenarios to value collateral to measure ECLs, but this information may not be relevant to assessing significant changes in credit risk unless the value has an effect on the probability of default occurring.²¹

As with the measurement of ECLs, the ITG members noted that IFRS 9 does not prescribe particular methods of assessing for significant increases in credit risk. Consequently, various methods could be applied, depending on facts and circumstances and these may include both quantitative and qualitative approaches. An entity should not restrict itself by considering only quantitative approaches when considering how to incorporate multiple forward-looking scenarios. Whichever approach is taken, it should be consistent with IFRS 9, considering reasonable and supportable information that is available without undue cost and effort. Once again, this is an area of judgement and so appropriate disclosures would need to be provided to comply with the requirements of IFRS 7 (see 15 below and Chapter 54 at 5.3).

A further issue was raised at the ITG meeting, which was not referred to in the minutes but was addressed in the 25 July 2016 IASB webcast. If a number of scenarios are applied to an individual asset, in some of which there is no significant increase in credit risk and in others there is, is it possible that it could be measured partly based on 12‑month losses and partly on lifetime losses? It was not the intention of the IASB that an asset should be regarded as being in more than one stage at the same time. For staging as well as for measurement, IFRS 9 applies to the unit of account which is the individual financial instrument. The financial asset cannot be considered to have partly significantly deteriorated and partly not. Hence, for instance, if the staging assessment is based on a mechanistic approach which considers the change in the lifetime probability of default, the entity should use the multiple scenario probability-weighted lifetime probability of default to assess whether there has been a significant increase in credit risk. The asset should then be measured using the weighted 12‑month probability of default if it is considered to be in stage 1, or the weighted lifetime probability of default if it is considered to be in stage 2.

However, as described in 6.5.3 above, the webcast also noted that, for a collectively assessed portfolio of assets, only a proportion of the portfolio may be deemed to have significantly deteriorated while the rest of the portfolio has not, due to differences in sensitivities of credit risk to a change in a particular parameter.

The IASB also illustrated how multiple scenarios can be reflected in a non-PD-based approach, using the example of a scorecard system. If the entity determines that there is non-linearity in the effect of the scenarios on the credit risk of the customers, one possibility is to look at the scorecard inputs and to determine which of these inputs have a non-linear relationship with the macroeconomic parameters. The entity then adjusts the scorecard, for example, using a scaling factor to reflect the impact of non-linearity, assesses whether there has been a significant increase in credit risk and measures ECL on the basis of the adjusted scorecard.

The approach set out in this discussion is broadly the same as ‘the top down’ approach to collective assessments illustrated by Example 51.18 in 6.5 above.

It is important to note that the ITG did not state that it is always necessary to use multiple scenarios and probability-weighted lifetime probabilities of default to assess significant increases in credit risk.

What it did state is that:

1. it is necessary to consider more than one scenario if there is non-linearity in the possible distribution of losses;
2. qualitative approaches may be included as well as quantitative ones, so that, for instance, it might be possible to take account of non-linearities by scaling the output from score cards; and
3. the assessment should be based on reasonable and supportable information that is available without undue cost or effort (see 5.9.1 above).

Nevertheless, the ITG did state that there should be consistency, to the extent relevant, between the forward-looking information used for measurement and for the assessment of significant increases in credit risk. There would not always be a direct mapping of the relevant information, because in some cases information might have an impact on the measurement of ECLs but not on the assessment of significant increases in credit risk (and vice versa). Also, various methods of assessing for significant increases in credit risk could be applied, depending on the particular facts and circumstances, and an entity should not restrict itself by considering only quantitative approaches when considering how to incorporate multiple forward-looking scenarios.

In the July 2016 webcast, the IASB also stressed the importance of adequate disclosures. Because there is no one right approach and because this area involves a high level of judgement, disclosures are very important to enable users of financial statements to understand how entities' credit risk is affected by forward-looking scenarios and how they have affected the application of the ECL model. It would also be useful to disclose if relevant forward-looking information has not been reflected in the assessment of significant deterioration on the basis that it is not reasonable and supportable.

In practice, many banks that use multiple scenarios of lifetime probabilities of default to measure assets in stage 2, use them also for assessing if there has been a significant increase in credit risk. Moreover, as with measurement, banks will need to consider regulators' expectations (see 7.1 below).

7.1 Basel guidance on accounting for expected credit losses

The Basel Committee published the final version of its Guidance on Credit Risk and Accounting for Expected Credit Losses (sometimes referred to as ‘G-CRAECL’, but in this publication, as ‘the Basel guidance’ or just ‘the guidance’) in December 2015 (see 1.6 above). The guidance deals with lending exposures, and not debt securities, and does not address the consequent capital requirements.

The guidance was originally drafted for internationally active banks and more sophisticated banks in the business of lending. The final version does not limit its scope but allows less complex banks to apply, ‘a proportionate approach’ that is commensurate with the size, nature and complexity of their lending exposures. It also extends this notion to individual portfolios of more complex banks. It follows that determining what is proportionate will be a key judgement to be made, which is likely to be guided in some jurisdictions by banking regulators. The guidance issued in June 2016 by the Global Public Policy Committee (GPPC) (see 7.2 below) will also be relevant in making this determination. The final version of the guidance acknowledges that due consideration may also be given to materiality.

The main section of the Basel Committee's guidance is intended to be applicable in all jurisdictions (i.e. for banks reporting under US GAAP as well as for banks reporting under IFRS) and contains 11 supervisory principles. The guidance is supplemented by an appendix that outlines additional supervisory requirements specific to jurisdictions applying the IFRS 9 ECL model.

It is important to stress that the guidance is not intended to conflict with IFRS 9 (and, indeed, this has been confirmed by the IASB), but it goes further than IFRS 9 and, in particular, removes some of the simplifications that are available in the standard. It also insists that any approximation to what would be regarded as an ‘ideal’ implementation of ECL accounting should be designed and implemented so as to avoid ‘bias’. The term ‘avoidance of bias’ is used several times in the guidance and we understand it to have its normal accounting meaning of neutrality. Hence, for instance, if a bank were ever dependent on past-due information to assess whether an exposure should be measured on a lifetime ECL basis, it is guided to ‘pay particular attention to their measurement of the 12‑month allowance to ensure that ECLs are appropriately captured in accordance with the measurement objective of IFRS 9.’²²

Perhaps one of the most significant pieces of guidance provided by the Basel Committee relates to the important requirement in IFRS 9 that ECLs should be measured using ‘reasonable and supportable information’. The Committee accepts that in certain circumstances, information relevant to the assessment and measurement of credit risk may not be reasonable and supportable and should therefore be excluded from the ECL assessment and measurement process. But, given that credit risk management is a core competence of banks, ‘these circumstances would be exceptional in nature’.²³ This attitude pervades the guidance. It also states that management is expected ‘to apply its credit judgement to consider future scenarios’ and ‘[t]he Committee does not view the unbiased consideration of forward-looking information as speculative’.²⁴ The guidance, therefore, establishes a high hurdle for when it is not possible for an internationally active bank to estimate the effects of forward-looking information. It is possible that banking regulators would expect banks to make an estimate of the effects of events with an uncertain binary outcome that is highly significant, such as the result of a referendum as discussed by the ITG in September 2015 (see 5.9.5 above).

A connected piece of the guidance relates to another important principle in IFRS 9, that reasonable and supportable information should be available ‘without undue cost or effort’. The guidance states that banks are not expected to read this ‘restrictively’. It goes on to say that, ‘Since the objective of the IFRS 9 model is to deliver fundamental improvements in the measurement of credit losses … this will potentially require costly upfront investment in new systems and processes’. Such costs ‘should therefore not be considered undue’.²⁵

Much of the guidance relates to systems and controls and so is outside the scope of this publication. The requirements of the main section that relate to accounting include:

1. There should be commonality in the processes, systems, tools and data used to assess credit risk and to measure ECLs for accounting and for regulatory capital purposes.²⁶
2. When a bank's individual assessment of exposures does not adequately consider forward-looking information, it is appropriate to group lending exposures with shared credit risk characteristics to estimate the impact of forward-looking information, including macroeconomic factors (see 6.5 above).²⁷ The grouping of lending exposures into portfolios with shared credit risk characteristics must be re-evaluated regularly (including re-segmentation in light of relevant new information or changes in the bank's expectations). Groupings must be granular enough to assess changes in credit risk and changes in a part of the portfolio must not be masked by the performance of the portfolio as a whole.²⁸
3. ‘Adjustments’ may be used to address events, circumstances or risk factors that are not fully considered in credit rating and modelling processes. But the Committee expects that such adjustments will be temporary. If the reason for an adjustment is not expected to be temporary then the processes should be updated to incorporate that risk driver. The guidance goes on to say that adjustments require judgement and create the potential for bias. Therefore, they should be subject to appropriate governance processes.²⁹
4. The ‘consideration of forward-looking information and macroeconomic factors is considered essential to the proper implementation of an ECL model. It cannot be avoided on the basis that the banks consider the costs to be excessive or unnecessary or because there is uncertainty in formulating forward-looking scenarios’. However, the Committee recognises that an ECL is ‘an estimate and thus may not perfectly predict actual outcomes. Accordingly, the need to incorporate such information is likely to increase the inherent degree of subjectivity in ECL estimates, compared with impairment measured using incurred loss approaches’. Also, the Basel Committee recognises that it may not always be possible to demonstrate a strong link in formal statistical terms between certain types of information and the credit risk drivers. Consequently, a bank's experienced credit judgement will be crucial in establishing the appropriate level for the individual or collective allowance.³⁰
5. Although the final version of the guidance says less about disclosures than the draft version, given the publication of the Enhanced Disclosure Task Force (EDTF) recommendations, disclosure remains one of the key principles (see 15 below or Chapter 54 at 9.2).

The guidance is supplemented by an appendix that outlines additional supervisory requirements specific to jurisdictions applying the IFRS 9 ECL model. The key requirements are outlined below:

1. A bank's definition of default adopted for accounting purposes should be guided by the definition used for regulatory purposes, which includes both a qualitative ‘unlikeliness to pay’ criterion and an objective 90-days-past-due criterion, described by the Committee as a ‘backstop’.
2. The IFRS 9 requirement to assess whether exposures have significantly increased in credit risk ‘is demanding in its requirements for data, analysis and use of experienced credit judgement’. The determination should be made ‘on a timely and holistic basis’, considering a wide range of current information. It is critical that banks have processes in place to ensure that financial instruments, whether assessed individually or collectively, are moved from the 12‑month to the lifetime ECL measurement as soon as credit risk has increased significantly. Credit losses very often begin to deteriorate a considerable period of time before an actual delinquency occurs and delinquency data are generally backward-looking. Therefore, ‘the Committee believes that they will seldom on their own be appropriate in the implementation of an ECL approach by a bank.’ Instead, banks need to consider the linkages between macroeconomic factors and borrower attributes, using historical information to identify the main risk drivers, and current and forecast conditions and experienced credit judgement to determine loss expectations. This will apply not only to collective assessments of portfolios but also for assessments of individual loans. The guidance gives the example of a commercial property loan, for which the bank should assess the sensitivity of the property market to the macroeconomic environment and use information such as interest rates or vacancy rates to make the assessment.³¹
3. In assessing whether there has been a significant increase in credit risk, banks should not rely solely on quantitative analysis. The guidance draws banks' attention to the list of qualitative indicators set out in paragraph B5.5.17 of the standard. Particular consideration should be given to a list of conditions, including an increased credit spread for a particular loan, a decision to strengthen collateral and/or covenant requirements, a downgrade by a credit rating agency or within the bank's internal credit rating system, a deterioration in future cash flows, or an expectation of forbearance or restructuring. Also, the guidance stresses that the sensitivity of the risk of a default occurring to rating downgrades increases strongly as rating quality declines. Therefore, the widths of credit risk grades need to be set appropriately, so that significant increases in credit risk are not masked. Further, ‘if a decision is made to intensify the monitoring of a borrower or class of borrowers, it is unlikely that such action would have been taken … had the increase in credit risk not been perceived as significant.’³²
4. Exposures that are transferred to stage 2 and that are subsequently renegotiated or modified, but not derecognised, should not be moved back to stage 1 until there is sufficient evidence that the credit risk over the remaining life is no longer significantly higher than on initial recognition. ‘Typically, a customer would need to demonstrate consistently good payment behaviour over a period of time before the credit risk is considered to have decreased.’³³
5. IFRS 9 includes a number of practical expedients (see 6.4 above). However, as banks are in the business of lending and it is unlikely that obtaining relevant information will involve undue cost or effort, the Basel Committee expects their limited use by internationally active banks. For instance:
   1. The long-term benefit of a high-quality implementation of an ECL model that takes into account all reasonable and supportable information far outweighs the associated costs.
   2. The use of the low credit risk simplification is considered a low-quality implementation of the ECL model and its use should be limited (except for holdings in debt securities, which are out of scope of the guidance). Also, the reference to an investment grade rating in the standard is only given as an example of a low credit risk exposure. An investment grade rating given by a rating agenda cannot automatically be considered low credit risk because banks are expected to rely primarily on their own credit assessments.
   3. Delinquency is a lagging indicator. Therefore, the Committee expects banks not to use the more-than-30-days-past-due rebuttable presumption as a primary indicator of a significant increase in credit risk. Banks may only use the rebuttable presumption as a backstop measure, alongside other, earlier indicators, while any rebuttal of the presumption would have to be accompanied by a thorough analysis to show that 30 days past due is not correlated with a significant increase in credit risk.³⁴

7.2 Global Public Policy Committee (GPPC) guidance

On 17 June 2016, the GPPC published The implementation of IFRS 9 impairment by banks – Considerations for those charged with governance of systemically important banks (‘the GPPC guidance’). The GPPC is the Global Public Policy Committee of representatives of the six largest accounting networks. This publication was issued to promote high-quality implementation of the accounting for ECLs in accordance with IFRS and to help those charged with governance to identify the elements of a high-quality implementation. It was designed to complement other guidance such as that issued by the Basel Committee (see 7.1 above) and the EDTF (see Chapter 54 at 9.2). It does not purport to amend or interpret the requirements of IFRS 9 in any way. The first half of the GPCC guidance sets out key areas of focus for those charged with governance. This includes governance and controls, transition issues and ten questions that those charged with governance might wish to discuss. The second half of the guidance sets out a sophisticated approach to implementing each aspect of the requirements of IFRS 9, along with considerations for a simpler approach and what is not compliant. Where relevant to understanding the accounting requirements of IFRS 9, this guidance is reflected in this chapter.

The GPPC guidance regards determination of the level of sophistication of the approach to be used as one of the key areas of focus for those charged with governance. Consequently, it provides guidance on how to make this determination for particular portfolios. It sets out factors to consider at the level of the entity, such as the extent of systemic risk that the bank poses, whether it is listed or a public interest entity, the size of its balance sheet and off balance sheet credit exposures, and the level and volatility of historical credit losses. Portfolio-level factors include its size relative to that of the total balance sheet and its complexity, the sophistication of other lending-related modelling methodologies, the extent of available data, the level of historical losses and the level and volatility of losses expected in the future. The document stresses that a simpler approach is not necessarily a lower quality approach if it is applied to an appropriate portfolio.

Also, on 28 July 2017, the GPPC issued its second paper titled The Auditor's Response to the Risks of Material Misstatement Posed by Estimates of Expected Credit Losses under IFRS 9. This second paper was written in an effort to assist audit committees in their oversight of the bank's auditors with regard to auditing ECLs. It is addressed primarily to the audit committees of systemically-important banks (SIBs) because of the relative importance of SIBs to capital markets and global financial stability but is relevant for other banks as well. It should be read in conjunction with the initial guidance published in 2016.

7.3 Measurement dates of expected credit losses

7.4 Interaction between the initial measurement of debt instruments acquired in a business combination and the impairment model of IFRS 9

Consistent with IFRS 9 and IFRS 13 – Fair Value Measurement, IFRS 3 requires financial assets acquired in a business combination to be measured by the acquirer on initial recognition at their fair value (see Chapter 49 at 3.3.4 and Chapter 9 at 5.5.5). [IFRS 3.18, IFRS 3.36]. IFRS 3 contains application guidance explaining that an acquirer should not recognise a separate valuation allowance (i.e. loss allowance for ECLs) in respect of loans and receivables acquired in a business combination for contractual cash flows that are deemed to be uncollectible at the acquisition date. This is because the effects of uncertainty about future cash flows are included in the fair value measure. [IFRS 3.B41].

Consequently, the accounting for impairment of debt instruments measured at amortised cost or fair value through other comprehensive income under IFRS 9 does not affect the accounting for the business combination. At the acquisition date, the acquired debt instruments are measured at their acquisition-date fair value in accordance with IFRS 3. No loss allowance is recognised as part of the initial measurement of debt instruments that are acquired in a business combination.

In contrast, after their original recognition, the subsequent accounting for debt instruments acquired in a business combination is in the scope of IFRS 9. [IFRS 9.5.5, 9.5.2.1, 9.5.2.2]. At the first reporting date after the business combination, following the guidance in IFRS 9, a loss allowance is recognised. [IFRS 9.5.5.3, 9.5.5.5]. This will result in an impairment loss that is recognised in profit or loss (rather than an adjustment to goodwill), just as would be the case if the entity were to originate those assets or acquire them as a portfolio, rather than acquire them through a business combination. [IFRS 9.5.5.8]. The assets will all be measured on the basis of 12‑month ECLs unless the assets have increased significantly in credit risk between the date of the business combination and the first reporting date.

Despite the colloquial reference to a ‘day one’ loss that results from the ECL impairment model in IFRS 9, it is important to understand that the recognition of a loss allowance for newly acquired (whether purchased or originated) debt instruments that are in the scope of the impairment requirements of IFRS 9 is a matter of subsequent measurement of those financial instruments. This means that the acquirer recognises the loss allowance for all debt instruments acquired in a business combination (that are subject to impairment accounting) in the reporting period that includes the business combination but not as part of that business combination, and with a corresponding impairment loss in profit or loss. As a result, these acquired assets are carried at a value below their fair value by what is often referred to as ‘day two’, when the 12‑month ECL is booked. This may seem counter-intuitive to many preparers and users of financial statements, but is clearly what is intended by IFRS 9.

The only exception to the need to record ECLs on day two is the specific accounting for purchased or originated credit-impaired financial assets (see 3.3 above). This applies to financial assets which are already credit-impaired at the acquisition date. A financial asset is credit-impaired when one or more events that have a detrimental impact on the estimated future cash flows of that financial asset have occurred (see 3.1 above). For such assets, no allowance is initially made for ECLs, as the ECL is part of the ‘credit-adjusted effective interest rate’. This differs from financial assets that are not purchased or originated credit-impaired, where an allowance must be made for expected credit losses (see 3.1 and 3.3 above).

It follows that, on a business combination, the acquirer needs to classify the acquired debt instruments that will be recorded at amortised cost or at fair value through OCI, according to whether they are purchased credit-impaired or not. If not, they will be regarded as stage 1 assets and be subject to a 12‑month ECL. If they are purchased credit-impaired, then there will be no need for an additional ECL unless there is a subsequent change in estimated lifetime ECL (see 3.3 above). Furthermore, these purchased credit-impaired loans will always be treated as such and are never subsequently subject to the staging requirements. Also, none of the acquired assets, either performing or purchased credit-impaired, will be classified as stage 2 or stage 3 at the date of initial recognition.

7.5 Interaction between expected credit losses calculations and fair value hedge accounting

Previously, the implementation guidance of IAS 39 made it clear that a fair value hedge adjustment would be included in the carrying amount of a financial asset that is subject to the impairment requirements. Otherwise, a part of its carrying amount would not have a loss allowance or the loss allowance would be overstated (in case of a negative fair value hedge adjustment). This guidance stated that the effect of fair value hedge accounting is to adjust the EIR, which affects the rate used to discount expected future cash flows. [IAS 39.E.4.4]. The rationale given in the example is that the original interest rate before the hedge becomes irrelevant once the carrying amount of the loan is adjusted for any changes in its fair value attributable to interest rate movements.

Similarly, for a financial asset that becomes credit-impaired, IFRS 9 requires impairment to be measured by reference to the gross carrying amount of the asset, which would include the fair value hedge adjustment. Therefore, for a credit-impaired financial asset in stage 3, the EIR would be adjusted to reflect any fair value hedge adjustment. [IFRS 9.B5.5.33].

However, whereas under IAS 39, most assets that are impaired would not generally be those for which fair value hedge accounting has been undertaken, under the new ECL impairment model an allowance is required for assets in stages 1 and 2, in addition to assets in stage 3. Hence, if the discount rate were to be adjusted whenever fair value hedge is applied, then all fair value hedge adjustments would need to be taken into account in calculating ECLs. This would give rise to significant operational challenges.

IFRS 9 is not explicit on this matter, but two points in the standard would seem to be relevant. First, unlike IAS 39, except for credit-impaired assets, the ECL requirements are not based on an asset's ‘carrying amount’ but on the contractual cash flows that are expected to be lost. Second, implementation guidance E4.4 in IAS 39, which stated that a fair value hedge adjusts the EIR, was not carried forward into the new standard. We understand that removing this guidance was not intended to change the accounting treatment in this respect. However another requirement of IAS 39, carried forward into IFRS 9, is that a fair value hedge adjustment is only required to be amortised when the hedged item ceases to be adjusted for changes in fair value attributable to the risk being hedged, which can be read to imply that until then there is no need to adjust the EIR, and hence the rate used to discount ECLs. [IFRS 9.6.5.10].

We believe the requirement is not clear and, so at least until it is clarified, there is an accounting policy choice on the matter. One approach would be to adjust the EIR whenever a fair value hedge adjustment is made and hence change the interest rate used to discount expected losses. The other would not take into account the fair value hedge adjustment until the EIR is adjusted to amortise the fair value hedge adjustment. [IFRS 9.6.5.8]. Such an adjustment to the EIR is permitted to commence at any time but would, at the latest, be required when hedge accounting ceases or when the financial asset becomes credit impaired, i.e. moved to stage 3.

7.6 Changes in ECL methodologies – errors, changes in estimates or changes in accounting policies

Many entities, having now had some experience of applying IFRS 9, may be considering changing various aspects of their ECL methodologies. The potential changes could include the correction of errors found in the model, updating PD, LGD and EAD for recent experience, making changes to how significant increases in credit risk are assessed, and amending the definition of default and write-off policies. The question that these potential amendments pose is whether they should be accounted for as prior year errors or changes in accounting policies, both of which (if material) would be adjusted for retrospectively, or changes in estimates which would be accounted for as they arise.

The majority of such changes are likely to be changes in estimates. The only changes that will likely be treated as errors are those that are ‘omissions from, and misstatements in, the entity's financial statements for one or more prior periods arising from a failure to use, or misuse of, reliable information that:

1. was available when financial statements for those periods were authorised for issue; and
2. could reasonably be expected to have been obtained and taken into account in the preparation and presentation of those financial statements’. [IAS 8.5].

Examples of errors would include failure to calculate ECLs on all relevant exposures, errors in input data and incorrect spreadsheet formulae. Most amendments to ECL calculation methods, including use of more sophisticated or more up to date data, revisions to formulae or measurement on a more granular basis, are likely to be changes in estimates. It is unlikely that amendments to the calculation approach would be regarded as a change in accounting policy.

If changes are made to how credit risk is assessed for the staging assessment, care must be taken to distinguish between changes in how the risk is perceived and how it is measured, and changes in what is regarded as ‘significant’. In general, it is not appropriate to reassess the credit risk of a financial asset on initial recognition. If the initial credit risk turns out to be riskier than expected, then the initial credit risk is not amended; it is possible that the asset will be transferred to stage 2 and the profit or loss impact would be accounted for as a change in estimate. On the other hand, if all that is amended is the way that the credit risk is measured, such as expressing the risk in terms of 12‑month PDs as opposed to lifetime PDs, it may be necessary to recalibrate the initial credit risk using the new measurement basis, so as to measure initial and current credit risk on a consistent basis. Meanwhile, the effect on ECLs of a change in the level of risk that is deemed to be ‘significant’ would normally be treated as a change in estimate.

8.1 Accounting treatment if modified financial assets are derecognised

In some circumstances, the renegotiation or modification of the contractual cash flows of a financial asset can lead to the derecognition of the existing financial asset and subsequently, the recognition of a new financial asset. [IFRS 9.B5.5.25]. This means that the entity is starting fresh and the date of the modification will also be the date of initial recognition of the new financial asset at its fair value. Typically, the entity will recognise a loss allowance based on 12‑month ECLs at each reporting date until the requirements for the recognition of lifetime ECLs are met. However, in what the standard describes as ‘some unusual circumstances’ following a modification that results in derecognition of the original financial asset, there may be evidence that the new financial asset is credit-impaired on initial recognition (see 3.3 above). Thus, the financial asset should be recognised as an originated credit-impaired financial asset. [IFRS 9.B5.5.26].

8.2 Accounting treatment if modified financial assets are not derecognised

In other circumstances, the renegotiation or modification of the contractual cash flows of a financial asset does not lead to the derecognition of the existing financial asset under IFRS 9. In such situations, the entity will:

• continue with its current accounting treatment for the existing asset that has been modified;
• recognise a modification gain or loss in profit or loss by recalculating the gross carrying amount of the financial asset as the present value of the renegotiated or modified contractual cash flows, discounted at the financial asset's original EIR (or the credit-adjusted EIR for purchased or originated credit-impaired financial assets). Any costs or fees incurred adjust the carrying amount of the modified financial asset and are amortised over the remaining term of the modified financial asset (see 3.1 above); [IFRS 9.5.4.3, Appendix A, IAS 1.82(a)]
• assess whether there has been a significant increase in the credit risk of the financial instrument, by comparing the risk of a default occurring at the reporting date (based on the modified contractual terms) and the risk of a default occurring at initial recognition (based on the original, unmodified contractual terms). A financial asset that has been renegotiated or modified is not automatically considered to have lower credit risk. The assessment should consider the credit risk over the expected life of the asset based on historical and forward-looking information, including information about the circumstances that led to the modification. Evidence that the criteria for the recognition of lifetime ECLs are subsequently no longer met may include a history of up-to-date and timely payment in subsequent periods. This means a minimum period of observation will often be necessary before a financial asset may qualify to return to stage 1; [IFRS 9.5.5.12, B5.5.27] and
• make the appropriate quantitative and qualitative disclosures required for renegotiated or modified assets to enable users of financial statements to understand the nature and effect of such modifications (including the effect on the measurement of ECLs) and how the entity monitors its assets that have been modified (see 15 below and Chapter 54 at 5.3). [IFRS 7.35F(f), B8B, 35J].

The following example has been adapted from Example 11 of the Implementation Guidance in the standard to illustrate the accounting treatment of a loan that is modified. It should be noted that it does not consider whether any of the contractual cash flows should be written off, as a partial derecognition. [IFRS 9 Example 11 IE66-IE73].

At its meeting on 22 April 2015, the ITG (see 1.5 above) discussed the measurement of ECLs in respect of a modified financial asset where the modification does not result in derecognition, but the cash flows have been renegotiated to be consistent with those previously expected to be paid.³⁵

The ITG noted that IFRS 9 is clear that an entity is required to calculate a new gross carrying amount and the gain or loss on modification taken to profit or loss should be based on the renegotiated or modified contractual cash flows and excludes ECLs unless it is a purchased or originated credit-impaired financial asset. [IFRS 9.5.4.3, Appendix A]. Consequently, an entity must calculate the gain or loss on modification as a first step before going on to consider the revised ECL allowance required on the modified financial asset. Thereafter, the entity is required to continue to apply the impairment requirements to the modified financial asset in the same way as it would for other unmodified financial instruments, taking into account the revised contractual terms. [IFRS 9.5.5.12]. The revised ECL cannot be assumed to be nil as, in accordance with paragraph 5.5.18 of IFRS 9, an entity is required to consider the possibility that a credit loss occurs, even if the likelihood of that credit loss occurring is very low. [IFRS 9.5.5.18].

The ITG also discussed the appropriate presentation and disclosure requirements pertaining to modifications. These are discussed further in Chapter 54 at 7.1.1.

We note that if an entity has no reasonable expectations of recovering a portion of the financial asset, which is subsequently forgiven, then this amount should arguably be written off, as a partial derecognition. The gross carrying amount would be reduced directly before a modification gain or loss is calculated. [IFRS 9.5.4.4, B5.4.9]. This would mean that the loss will be recorded as an impairment loss, rather than as a loss on modification, and presented differently in the profit or loss account. In practice, it will often be difficult to disentangle the effects of modification and write off, as some forgone cash flows may be compensated for by a higher interest rate applied to the remaining contractual amounts due.

9.1 Accounting treatment for debt instruments measured at fair value through other comprehensive income

The accounting treatment and journal entries for debt instruments measured at fair value through other comprehensive income are illustrated in the following example, based on Illustrative Example 13 in the Implementation Guidance for the standard. [IFRS 9 Example 13 IE78-IE81].

This means that in contrast to financial assets measured at amortised cost, there is no separate allowance but, instead, impairment gains or losses are accounted for as an adjustment of the revaluation reserve accumulated in other comprehensive income, with a corresponding charge to profit or loss (which is then reflected in retained earnings). The tax implications of debt instruments measured at FVOCI is discussed further in Chapter 33 at 10.4.1 and 10.4.2.

As explained in 7.3.1 above IFRS 9 requires a derecognition gain or loss to be measured relative to the carrying amount at the date of derecognition. This necessitates an assessment and measurement of ECLs for that particular asset as at the date of derecognition.

9.2 Interaction between foreign currency translation, fair value hedge accounting and impairment

The above example is relatively straightforward. A more complicated one, based on a foreign currency denominated financial asset which is also the subject of an interest rate hedge, is provided below. It is based on Illustrative Example 14 in the Implementation Guidance for the standard but has been adjusted so as to include the effect of discounting in the measurement of ECLs (see 5.7 above). [IFRS 9.IE82‑IE102]. Note that we do not address the additional complexities that will arise from the consideration of taxation, including deferred tax. The tax accounting implications of debt instruments measured at FVOCI is discussed further in Chapter 33 at 10.4.1 and 10.4.2.

10.1 Trade receivables and contract assets

It is a requirement for entities to apply the simplified approach for trade receivables or contract assets that do not contain a significant financing component. However, entities have a policy choice to apply either the general approach (see 3.1 above) or the simplified approach separately to trade receivables and contract assets that do contain a significant financing component (see 3.2 above). [IFRS 9.5.5.15(a)].

Also, entities are allowed to use practical expedients when measuring ECLs, as long as the approach reflects a probability-weighted outcome, the time value of money and reasonable and supportable information that is available without undue cost or effort at the reporting date about past events, current conditions and forecasts of future economic conditions. [IFRS 9.5.5.17, B5.5.35].

One of the approaches suggested in the standard is the use of a provision matrix as a practical expedient for measuring ECLs on trade receivables. For instance, the provision rates might be based on days past due (e.g. 1 per cent if not past due, 2 per cent if less than 30 days past due, etc.) for groupings of various customer segments that have similar loss patterns. The grouping may be based on geographical region, product type, customer rating, the type of collateral or whether covered by trade credit insurance, and the type of customer (such as wholesale or retail). To calibrate the matrix, the entity would adjust its historical credit loss experience with forward-looking information. [IFRS 9.B5.5.35].

In practice, many corporates use a provision matrix to calculate their current impairment allowances. However, in order to comply with the IFRS 9 requirements, corporates have needed to consider how current and forward-looking information might affect their customers' historical default rates and, consequently, how the information would affect their current expectations and estimates of ECLs. The use of the provision matrix is illustrated in the following example. [IFRS 9 Example 12 IE74-IE77].

It should be noted that this example, like many in the standard, ignores the need to consider explicitly the time value of money, presumably in this case because the effect is considered immaterial.

10.2 Lease receivables

For lease receivables, entities have a policy choice to apply either the general approach (see 3.1 above) or the simplified approach (see 3.2 above) separately to finance and operating lease receivables (see Chapter 23). [IFRS 9.5.5.15(b)].

When measuring ECLs for lease receivables, an entity should:

• use the cash flows that are used in measuring the lease receivables in accordance with IFRS 16; [IFRS 9.B5.5.34] and
• discount the ECLs using the same discount rate used in the measurement of the lease receivables in accordance with IFRS 16. [IFRS 9.B5.5.46, IAS 17.4].

There has been some discussion on whether the unguaranteed residual value (URV) of the asset subject to a finance lease should be included in the calculation of ECLs under IFRS 9. The URV is part of the gross investment in the finance lease, together with the minimum lease payments receivable by the lessor. Changes to URV arise from fluctuations in the price that could be received for the leased asset at the end of the lease term. Paragraph 2.1(b) of IFRS 9 scopes out rights and obligations under leases to which IFRS 16 applies, except for the impairment of finance lease receivables (i.e. net investments in finance leases) and operating lease receivables recognised by a lessor (see 2 above). IFRS 16 provides guidance on measurement of the URV, which means that such measurement is within the scope of IFRS 16 rather than the impairment requirements of IFRS 9. [IFRS 16.77].

The URV of the asset underlying a finance lease should be excluded from the calculation of ECLs under IFRS 9. This means that the collateral that is taken into account in measuring ECLs should exclude any amounts attributed to URV and recorded on the lessor's statement of financial position.

In other words, any collateral taken into account in the calculation of ECLs should be restricted to the fair value of the right of use of the asset and not that of the underlying asset itself.

11.1 Definitions and scope

The description of ‘loan commitment’ and the definition of ‘financial guarantee contract’ remain unchanged from IAS 39. Loan commitments (see Chapter 45 at 3.5) are described in IFRS 9 as ‘firm commitments to provide credit under pre-specified terms and conditions’, while a financial guarantee contract (see Chapter 45 at 3.4) is defined as ‘a contract that requires the issuer to make specified payments to reimburse the holder for a loss it incurs because a specified debtor fails to make payment when due in accordance with the original or modified terms of a debt instrument’. [IFRS 9.BCZ2.2, Appendix A, IAS 39.9, BC15].

The IFRS 9 impairment requirements apply to loan commitments and financial guarantee contracts that are not measured at fair value through profit or loss under IFRS 9, with some exceptions (see 2 above).

The ITG (see 1.5 above) discussed in April 2015 whether the impairment requirements in IFRS 9 must also be applied to other commitments to extend credit such as:

• a commitment (on inception of a finance lease) to commence a finance lease at a date in the future (i.e. a commitment to transfer the right to use an asset at the lease commencement date in return for a payment or series of payments in the future); and
• a commitment by a retailer through the issue of a store account to provide a customer with credit when the customer buys goods or services from the retailer in the future.

The ITG appeared to agree with the IASB's staff analysis that the impairment requirements of IFRS 9 apply to an agreement that contains a commitment to extend credit by virtue of paragraph 2.1(g) if:

• the agreement meets the description of a loan commitment; [IFRS 9.BCZ2.2]
• the agreement meets the definition of a financial instrument; [IAS 32.11] and
• none of the specific exemptions from the requirements of IFRS 9 apply. [IFRS 9.2.1].

The IASB staff paper stated that some contracts, such as irrevocable finance lease agreements, might clearly contain a firm commitment at inception to provide credit under pre-specified terms and conditions. However, other cases might not be so clear cut, depending upon the specific terms of the agreement and other facts and circumstances (e.g. if the issuer of a store account has the discretion to refuse to sell products or services to a customer with a store card and hence can avoid extending credit).³⁶

In the examples discussed above, the finance lease and store account do not meet the definition of a financial instrument until the contractual right to receive cash is established, that is likely to be at the commencement of the lease term or when goods or services are sold. [IAS 32.11, AG20]. Only lease receivables are scoped into the IFRS 9 impairment requirements (see 10.2 above). [IFRS 9.2.1(b)]. Consequently, there is no need to make provision for ECLs, in accordance with IFRS 9, until a financial lease receivable or a financial asset within the scope of IFRS 9 is recognised.

11.2 The calculation of ECLs

The application of the model to financial guarantees and loan commitments warrants some further specification regarding some of the key elements, such as the determination of the credit quality on initial recognition, cash shortfalls and the EIR to be used in the ECL calculations. These specifications are summarised in Figure 51.7 below, which also highlights the differences in recognising and measuring ECLs for financial assets measured at amortised cost or at fair value through other comprehensive income, loan commitments and financial guarantee contracts.

Further discussion on the discount rate to be used for calculating ECLs on loan commitments and financial guarantee contracts is provided at 5.7 above.

11.3 Financial guarantee contracts that require the holder to pay further premiums in the future

At its meeting in April 2015, the ITG (see 1.5 above) also discussed the measurement of ECLs for an issued financial guarantee contract that requires the holder to pay further premiums in the future. Some members of the ITG agreed with the staff's analysis that the issuer of a financial guarantee contract should exclude future premium receipts due from the holder when measuring ECLs in respect of the expected cash outflows payable under the guarantee.³⁷ When estimating the cash shortfalls, the amounts that the entity expects to receive from the holder should relate only to recoveries or reimbursements of claims for losses and would not include receipts of premiums. [IFRS 9.B5.5.32]. Moreover, the expected cash outflows under the guarantee depend upon the risk of default of the guaranteed asset, while the expected future premiums receipts are subject to the risk of default by the holder of the guarantee. Hence, these risks of default should be considered separately. In other words, the ECL measurement should be carried out gross of any premiums receivable in the future.

In addition, an ITG member noted that the terms of a financial guarantee contract may affect the period of exposure to credit risk on the guarantee, for example if the guarantee were contingent or cancellable. This should be taken into consideration when measuring the ECLs of the guarantee.

IFRS 9 requires that financial guarantees and off-market loan commitments should be measured at the ‘higher of’ the amount initially recognised less cumulative amortisation, and the ECL. [IFRS 9.4.2.1(c), 4.2.1(d)]. For a financial guarantee contract issued to an unrelated party in a stand-alone arm's length transaction, premiums that are received in full at inception will likely be the same as the fair value of the guarantee at initial recognition (see Chapter 49 at 3.3.3). In such circumstances, it is likely that no ECLs will need to be recognised immediately after initial recognition, as the initial fair value will normally exceed the lifetime ECLs. However, a financial guarantee contract for which premiums are receivable over the life of the guarantee will have a nil fair value at initial recognition. In such circumstances, the subsequent measurement of the financial guarantee contract is likely to be based on the ECL allowance. This is illustrated in the example below.

Normal loan commitments (i.e. those issued at market interest rates and so are excluded from the scope of IFRS 9 except for impairment and derecognition) [IFRS 9.2.1(g), 2.3] are not subject to the ‘higher of’ test for subsequent measurement. [IFRS 9.2.3(c), 4.2.1(d)]. The consequence is that an ECL is required for all normal loan commitments, whether or not any fees are paid upfront. This is consistent with the general requirement to provide for 12‑month ECLs for any new loans that have not experienced significant increases in credit risk since initial recognition.

11.4 Cross-company guarantees

For the individual entities within a group, there may be instances where cross-company guarantee arrangements will meet the definition of financial guarantee contracts and hence will fall within the scope of the IFRS 9 ECL requirements. [IFRS 9.4.2.1 (d)].

When estimating the ECLs on cross-company guarantees relating to loans to a group entity made by parties external to the group, each entity that is a party to the cross-company guarantee arrangement needs, for its standalone financial statements, to factor in the likelihood of it being called upon to make payments under the arrangement. A practical approach would be to determine the total ECLs and then allocate them to group entities, based on an assessment of which entities will most likely settle the guarantee if payment is required. In doing so, all reasonable and supportable information that is available, including the respective entities' relative financial strength in relation to other group entities will need to be taken into account. If one entity in the group is expected to reimburse the lender, then this will reduce the potential cash shortfall estimated by other entities within the group. For example, if the parent is in the strongest financial position and is expected to fully reimburse the lender for a loan issued to one of its subsidiaries, then the parent's estimate of cash shortfalls would be consistent with the cash shortfalls that the lender could suffer in the absence of the guarantee. However, once the expected payments from the parent are taken into account, this would significantly reduce the expected cash shortfalls estimated by other group entities. This will result in each entity recognising the appropriate probability-weighted estimate of credit losses. Requiring all entities to recognise an ECL amount which does not take into account the likelihood of being called upon would not represent a probability weighted estimate of credit losses and would not provide useful information to users of the financial statements.

Additional disclosures would be required to explain the arrangement, including the maximum amount to which the entity is exposed and how the ECL amount has been calculated. [IFRS 7.36(a), 35G]. Also, IAS 24 – Related Party Disclosures – requires details of any guarantees given to, or received from, related parties to be disclosed. [IAS 24.18, 21].

11.5 Purchased or originated credit impaired loan commitments and financial guarantee contracts

Another question that arises in practice is whether loan commitments and financial guarantee contracts can ever be accounted for as purchased or originated credit-impaired. The definition of ‘purchased or originated credit-impaired’ in IFRS 9 refers only to financial assets, not financial instruments (consistent with the definition of credit-impaired) but loan commitments and financial guarantees are not financial assets. So, if such an instrument is entered into when default is highly likely or has already occurred, and the potential loss is reflected in the price, how should the ECLs be measured, so as to avoid double counting the loss?

This issue could be particularly relevant in the context of business combinations, where an entity may acquire loan commitments or financial guarantee contracts that are already credit-impaired. For financial guarantees, the ‘higher of’ test avoids the double-counting, as the fair value of the guarantee recognised as a liability on initial recognition will be higher than lifetime expected losses. For loan commitments, one view could be to consider they are at below-market interest rates on initial recognition (as the terms were fixed at a time where the loan commitment was not credit-impaired) and apply the higher of test; alternatively, one may consider that the guidance for financial assets may be applied by analogy to loans commitments. This would make sense as the standard treats loans that are drawn from a loan commitment as a continuation of the same financial instrument. For disclosure purposes, we believe such loan commitments and financial guarantees should also be reported as credit-impaired.

12.1 Scope of the exception

The guidance relates to financial instruments that ‘include both a loan and an undrawn commitment component and for which the entity's contractual ability to demand repayment and cancel the commitment does not limit the entity's exposure to credit losses to the contractual notice period’. [IFRS 9.5.5.20]. Despite the use of the word ‘both’, the ITG agreed, in April 2015, that this guidance applies even if the facility has yet to be drawn down. It also applies if the facility has been completely drawn down, as it is the nature of revolving facilities that the drawn down component is periodically paid off before further amounts will be drawn down again in future.

The standard also describes three characteristics generally associated with such instruments: [IFRS 9.B5.5.39]

• they usually have no fixed term or repayment structure and usually have a short contractual cancellation period;
• the contractual ability to cancel the contract is not enforced in day-to-day management, but only when the lender is aware of an increase in credit risk at the facility level; and
• they are managed on a collective basis.

Products that are generally agreed to be in the scope of the exception include most credit card facilities and most retail overdrafts. However, even with these some caution needs to be applied, since we understand that there are credit card facilities which do not enable the issuer to demand repayment and cancel the facility, and which would therefore be out of scope.

What is less clear is the treatment of corporate overdrafts and similar facilities. It is relevant that all the ITG discussions as well as the webcast referred to credit cards and retail customers and not corporate exposures. The problem is partly that the guidance to the standard describes management on a collective basis as a characteristic that revolving facilities in the scope of the exception ‘generally have’, rather than a required feature as listed in paragraph 5.5.20 of IFRS 9. [IFRS 9.B5.5.39]. Some banks consider this is still a determining feature and that many of their corporate facilities are outside the scope of the exception because they are managed on an individual basis. Banks normally have a closer business relationship with their larger corporate customers than with most retail customers, and more data to manage the credit risk, such as access to regular management information. Other banks consider that facilities that are individually managed are still in the scope of the exception, notably because individual credit reviews are generally performed only on an annual basis (unless a significant event occurs). In addition, it is unclear exactly what is meant by ‘managed on a collective basis’ and where to draw the line between large corporates and smaller entities. It should be noted that if a corporate facility is not deemed to be a revolving facility, but can be cancelled at short notice, the ECLs will be limited to those that arise over the notice period.

At its December 2015 meeting, the ITG discussed whether:

• multi-purpose credit facilities, which have the ability to be drawn down in a number of different ways (e.g. as a revolving overdraft, a variable or fixed-rate loan (with or without a fixed term) or an amortising loan such as a mortgage) would fall within the scope exception;
• the general characteristics identified in paragraph B5.5.39 of IFRS 9 should be considered to be required characteristics, or merely examples of typical characteristics; and
• the existence of a fixed term of the loan once drawn down would prevent a facility from falling within the scope exception.

The ITG commented that:

• The supporting application guidance in paragraph B5.5.39 of IFRS 9 reinforces the features described in paragraph 5.5.20 of IFRS 9 by setting out general characteristics which, while not determinative, are consistent with those features.
• The Basis for Conclusions of IFRS 9 provides further context around the type of financial instruments that the Board envisaged would fall within the scope exception. In particular, the exception was intended to be limited in nature and that it was introduced in order to address specific concerns raised by respondents in relation to revolving credit facilities that were managed on a collective basis. Also, it was understood that these types of financial instruments included both a loan and an undrawn commitment component and that they were managed, and ECLs were estimated, on a facility level; i.e. the drawn and undrawn exposure were viewed as one single cash flow from the borrower. [IFRS 9.BC5.254-BC5.261].
• Consequently, both the drawn and undrawn components of these facilities were understood to have similar short contractual maturities, i.e. the lender had both the ability to withdraw the undrawn commitment component and demand repayment of the drawn component at short notice.
• An immediately revocable facility which has a fixed maturity (e.g. 5 years) would be consistent with the type of facility within the scope exception because the fixed term feature does not negate the lender's contractual right to cancel the undrawn component at any time. In contrast, an immediately revocable facility that has no fixed maturity but when drawn, can take the form of a loan with a fixed maturity (i.e. once it has been drawn, the lender no longer has the right to demand immediate repayment at its discretion) would not be consistent with the type of facility envisaged to be within the scope exception. This is because the fixed term feature does negate the lender's contractual right to demand repayment of the undrawn component. However, regarding this characteristic, the ITG members also highlighted the following:
  1. an entity would first need to establish the unit of account to which the requirements of IFRS 9 should be applied. In this regard, they noted that even if there was only one legal contract supporting a particular multi-purpose credit facility, there might be more than one unit of account to consider; and
  2. if the fixed-term feature was for a shorter period, judgement would be required in order to determine whether such a fixed-term feature would prevent a particular financial instrument from falling within the scope exception (e.g. whether the borrower could consider the exposure on the drawn and undrawn components to be one single cash flow).
• IFRS 7 requires an entity to explain, among other things, the assumptions used to measure ECLs. Within the context of multi-purpose credit facilities, such disclosures are likely to be important in order to meet the disclosure objectives (see 15 below and Chapter 54 at 5.3).

While, according to the ITG, the drawn and undrawn exposures are viewed as ‘one single cash flow from the borrower’, the standard's Basis for Conclusions is slightly clearer. [IFRS 9.BC5.259]. It states that the loan and undrawn commitment ‘are managed, and ECLs are estimated on a facility level. In other words, there is only one set of cash flows from the borrower that relates to both components’. Hence, the drawn and undrawn elements of a revolving facility within the scope of the exception would normally be viewed as only one unit of account. The ITG discussion seems to suggest that a new unit of account would be recognised if a borrower chose to draw down on a multi-purpose facility in the form of a term loan, because this is the point where this specific drawn portion ceases to share the key characteristic of a revolving facility, i.e. the entity's contractual ability to demand repayment and cancel the commitment.

At its December 2015 meeting, the ITG discussed charge cards and how ECLs on future drawdowns should be measured if there is no specified credit limit in the contract. The ITG members considered a specific fact pattern where the bank has the ability to approve each transaction at the time of sale based on the customer's perceived spending capacity using statistical models and inputs such as spending history and known income.

The ITG members noted that because the bank has the right to refuse each transaction at its discretion, and on the assumption that the bank actually exercises that right in practice, then:

• the contractual credit limit should be considered to be zero and consequently future drawdowns would not be taken into account; and furthermore,
• the facility described would not fall within the scope exception because there would be no undrawn commitment component (i.e. there is no firm commitment to extend credit).

However, the ITG members noted that their discussions focussed on the very specific fact pattern presented and observed that the conclusion could differ in other situations.

12.2 The period over which to measure ECLs

12.3 Exposure at default (EAD)

To measure ECLs on revolving facilities, such as credit cards, it will be necessary to estimate several components that make up the EAD:

• the credit conversion factor, to determine the portion of the facility that is drawn down in any period (limited, for facilities in stage 1 to the next twelve months);
• the speed at which drawn down facilities are paid off; and
• the level of interest expected to be charged in the future on those facilities that are drawn down.

These components will all need to be estimated based on past experience and future expectations, for sections of the portfolio that are segmented so that they have similar credit characteristics (see 6.5.2 above). The estimation of interest is addressed further in 12.4 below.

At its meeting on 16 September 2015, the ITG (see 1.5 above) discussed how an entity should estimate future drawdowns on undrawn lines of credit when an entity has a history of allowing customers to exceed their contractually set credit limits on overdrafts and other revolving credit facilities.

The ITG members noted that:

• The exception for some types of revolving credit facilities set out in paragraph 5.5.20 of IFRS 9 relates to the contractual commitment period and does not address the contractual credit limit. The standard was clear in this regard and consequently, it would not be appropriate to analogise this specific exception to the contractual credit limit.
• Some members of the ITG pointed out that, in practice, the tenor and amount of revolving credit facilities are inextricably linked, because banks not only extend credit for a period in excess of their maximum contractual commitment period but also allow customers to make drawdowns in excess of the maximum contractually agreed credit limit as notified to the customer. Consequently, if amounts in excess of the maximum contractually agreed credit limits are not taken into account, there would be a potential disconnect between the accounting and credit risk management view.
• However, it was concluded that IFRS 9 limits the estimation of future drawdowns to the contractually agreed credit limit.

12.4 Time value of money

The time value of money is important in measuring ECLs for revolving facilities since interest rates (when interest is charged) are high. Hence it is important that any interest that is expected to be charged on drawn balances is included in the EAD and that an appropriate rate is used to discount ECLs. An additional complexity is introduced by credit cards, because they typically have a grace period in which no interest is charged as long as the amount drawn down is repaid within a specified period of time.

The standard is silent on this topic, however, the ITG discussion on the use of floating-rates of interest to measure ECLs in December 2015 (see 5.7 above) established a useful principle, that there should be consistency between the rate used to recognise interest revenue, the rate used to project future cash flows (including shortfalls) and the rate used to discount those cash flows. While the high rates charged by a credit card issuer are sometimes fixed in the contract, the fact that the rate charged (nil or the high rate) depends on how quickly the customer repays the amount drawn, means that the rate can be thought of as ‘floating’, even if it does not vary with a benchmark rate of interest. This is important since otherwise it would be necessary to assess the EIR on original recognition and keep this fixed unless the facility is derecognised, ignoring any changes in customer behaviour.

Applying this principle, for a credit card customer that is a ‘transactor’, that is, one who repays any amount drawn down within the specified short period and so is charged no interest, it would not be appropriate to discount expected losses. On the other hand, for a credit card customer who is a ‘revolver’ and who only pays off the minimum amounts permitted by the issuer (in effect, using the card to borrow money), the high rate of interest should be included in the forecast cash flows and in the discount rate.

However, any transactor who goes on to default is likely to begin paying off less than the full amount for a period of time before they default. To estimate the expected losses for this scenario, it will be necessary to include any interest that will be charged in this period. A consistent discount rate will then be a blended rate, of nil for the period over which the customer is expected to pay no interest and the high rate over the period in which they will pay.

According to the guidance for ‘normal’ loan commitments, the expected credit losses on a loan commitment must be discounted using the effective interest rate, or an approximation thereof, that will be applied when recognising the financial asset resulting from the loan commitment. [IFRS 9.B5.5.47]. Applying this approach, the losses on the currently undrawn portion of a revolving facility should be discounted based on the rate that is likely to be charged if it is drawn down. If it is expected that interest will be charged at the high rate – which is likely for most facilities that are already ‘revolvers’ – then the discount rate is likely to be the high rate. This approach is consistent with a view expressed at the ITG meeting that the drawn and undrawn balances should be viewed as one unit of account and so discounted at the same rate. If it is projected that a transactor will at some stage become a revolver before it defaults, then it may be appropriate to calculate a blended discount rate.

Because the choice of interest rate used to project cash flows and to discount losses will depend on expectations of the borrower's behaviour, it will need to be made separately for segments of the portfolio with similar credit and payment characteristics.

In practice, it is likely that credit card issuers will often adopt procedures to discount their ECLs that may be less sophisticated than set out above, due to operational constraints and because the objective of the standard is to discount ECLs at an approximation of the EIR. [IFRS 9.B5.5.44]. However, it is necessary to understand what is theoretically required by IFRS 9 in order to be able to assess whether a pragmatic approach is a reasonable approximation.

12.5 Determining significant increase in credit risk

As already mentioned at 12.2 above, at its April 2015 meeting, the ITG discussed the starting reference date when assessing significant increases in credit risk for a portfolio of revolving credit facilities. There will typically be a diverse customer base, ranging from long-standing customers who have been with the bank for many years, to new customers who have only recently opened an account. The standard's general rule is that the starting reference date is the date of original recognition. Consequently, the date of initial recognition for this purpose is the date the facility was issued and this should only be changed if there has been a derecognition of the original facility. As discussed at 12.2 above, it is not altogether clear what would qualify as a derecognition within the context of the revolving facility exception. If the lender derecognises a facility at the end of its contractual term and recognises a new one when it decides to renew or extend credit, consistent with the Basis of Conclusions, it would be consistent to assess if there has been a significant increase in credit risk from when the current facility was first recognised. [IFRS 9.BC5.260]. Similarly, it may make sense to use the date that the limit was increased if a facility is now far larger than would have been granted on original recognition. There is also a view that the credit risk on the date that the facility was last increased may be a useful proxy for the credit risk on the date of original recognition. There is a particular challenge on transition to IFRS 9, since entities may have limited data on the credit risk at the date of original recognition. (See 5.9 above).

However, as discussed at 12.2 above, another view is that only a reduction or cancellation of the facility would lead to the revolving facility being derecognised. In some circumstances, issuing a new card may be indicative that the original facility has been derecognised (e.g. replacement of a student credit card with a new credit card upon graduation), but in other cases, this may be a purely operational process and thus would not indicate that a new facility has been issued.

The ITG did not conclude further on this issue and it was not discussed in the IASB's May 2017 webcast. Consequently, at the date of writing this issue had not been resolved.

14.1 Allowance for financial assets measured at amortised cost, contract assets and lease receivables

ECLs on financial assets measured at amortised cost, lease receivables (see Chapter 23) and contract assets (see Chapter 28) are presented as an allowance, i.e. as an integral part of the measurement of those assets in the statement of financial position.

Unlike the requirement to show impairment losses as a separate line item in the statement of profit or loss, there is no similar consequential amendment to IAS 1 to present the loss allowance as a separate line item in the statement of financial position. [IAS 1.82(ba)].

It is clear from the standard that the definition of amortised cost of a financial asset is after adjusting for any loss allowance and hence, the loss allowance would reduce the gross carrying amount in the statement of financial position (which is why an allowance is sometimes referred to as a contra asset account). [IFRS 9 Appendix A]. Accordingly, financial assets measured at amortised cost, contract assets and lease receivables should be presented net of the loss allowance at their amortised cost in the statement of financial position.

This was confirmed at the ITG meeting in December 2015, when the ITG discussed whether an entity is required to present the loss allowance for financial assets measured at amortised cost (or trade receivables, contract assets or lease receivables) separately in the statement of financial position. The ITG members first noted that irrespective of how the loss allowance is presented or how it is included in the measurement of the financial instrument, IFRS 7 contains disclosure requirements pertaining to the loss allowance for all financial instruments within the scope of the IFRS 9 impairment requirements. The ITG members also noted that, in contrast to the case of financial assets measured at fair value through other comprehensive income, neither IFRS 9 nor IFRS 7 contains any specific requirements regarding the presentation of the loss allowance for financial assets measured at amortised cost (or trade receivables, contract assets or lease receivables) on the face of the statement of financial position. In accordance with the general requirements of IAS 1, the financial statements should fairly present the financial position of an entity. However, the ITG members noted that paragraph 54 of IAS 1 does not list the loss allowance as an amount that is required to be separately presented on the face of the statement of financial position.

14.2 Provisions for loan commitments and financial guarantee contracts

In contrast to the presentation of impairment of assets, ECLs on loan commitments and financial guarantee contracts are presented as a provision in the statement of financial position, i.e. as a liability. [IFRS 9 Appendix A].

For financial institutions that offer credit facilities, commitments may often be partially drawn down, i.e. an entity may have a facility that includes both a loan (a financial asset) and an undrawn commitment (a loan commitment). If the entity cannot separately identify the ECLs attributable to the drawn amount and the undrawn commitment, IFRS 7 requires an entity to present the provision for ECLs on the loan commitment together with the allowance for the financial asset. IFRS 7 states, further, that if the combined ECLs exceed the gross carrying amount of the financial asset, then the ECLs should be recognised as a provision. [IFRS 7.B8E].

14.3 Accumulated impairment amount for debt instruments measured at fair value through other comprehensive income

Rather than presenting ECLs on financial assets measured at fair value through other comprehensive income as an allowance, this amount is presented as the ‘accumulated impairment amount’ in other comprehensive income. This is because financial assets measured at fair value through other comprehensive income are measured at fair value in the statement of financial position and the accumulated impairment amount cannot reduce the carrying amount of these assets (see 9 above for further details). [IFRS 9.4.1.2A, 5.5.2, Appendix A].