Paper 5A – Assessing when to recognise lifetime expected credit losses
The general impairment model proposed in the March 2013 Exposure Draft Financial Instruments: Expected Credit Losses (ED) uses different measurement objectives depending on whether the credit risk of financial instruments has increased significantly since initial recognition. A key objective of the ED was to recognise lifetime expected credit losses (ECL) on all financial instruments for which there have been significant increases in credit risk since initial recognition.
The IASB addressed concerns relating to the timing and recognition of ECL under 5 headings:
- Issue A: whether to recognise lifetime ECL when the credit risk of a financial instrument has increased significantly since initial recognition.
- Issue B: whether the proposals provide sufficient guidance on when to recognise lifetime expected credit losses.
- Issue C: whether the assessment of when to recognise lifetime ECL should consider only changes in the probability of a default occurring, rather than changes in ECL (or credit loss given default (LGD)) and whether and when that assessment should be based on the lifetime probability of default or the 12-month probability of default.
- Issue D: whether an entity shall re-establish the loss allowance (or provision) at an amount equal to 12-month ECL when the criteria for the recognition of lifetime ECL are no longer met
Issue A: whether to recognise lifetime ECL when the credit risk of a financial instrument has increased significantly since initial recognition.
The ED proposes the following for the recognition of lifetime expected credit losses.
[Par. 5] At the reporting date, the entity shall measure the expected credit losses for a financial instrument at an amount equal to the lifetime expected credit losses if the credit risk on that financial instrument has increased significantly since initial recognition.
The vast majority of respondents, including non-US users of financial statements, agreed with the proposal to recognise lifetime ECL when the credit risk of a financial instrument increased significantly, because it considers the underlying economics of a transaction while easing operational complexities.
Although many preparers commented that they could build upon their internal credit risk management practices to identify significant increases in credit risk, some of them made suggestions to align the proposals more closely with their current credit risk management systems. To this effect they proposed some alternatives.
Many of these alternative suggestions were already considered by the IASB in developing the proposals in the ED. These were:
- Absolute level of credit risk
Using this approach, an entity would recognise lifetime ECL on all financial instruments for which the credit risk is at or above a specified level at the reporting date, including on originated or purchased financial instruments at market terms that are at or above the specified level of credit risk on initial recognition.
However, this was rejected by the IASB because it does not reflect a credit deterioration model (see paragraph BC67 in the Basis for Conclusions of the ED).
- Change in the credit risk management objective
The staff think that this approach is somewhat similar to the approach proposed in the Supplementary Document Financial Instruments: Impairment (SD) which was published in January 2011. The SD required recognition of lifetime ECL if the collectability of a financial asset, or group of financial assets, becomes so uncertain that the entity’s credit risk management objective for that asset, or group, changes from receiving the regular payments from the debtor to recovery of all or a portion of the financial asset. While recognising lifetime ECL when the credit risk management objective changes would be operationally simpler (ie financial instruments that are being managed differently would be identified rather than assessing a change in credit risk since initial recognition), the approach would be similar to the incurred loss model in IAS 39. The management of a financial instrument may only change relatively late compared with when significant increases in credit risk occur, so this may be a less timely approach to recognising lifetime ECL.
There were also some suggested alternative approaches not previously considered by the IASB. These were:
- Credit underwriting policies
Using this approach an entity would recognise lifetime ECLs on a financial instrument if the financial instrument’s credit risk at the reporting date is higher than the credit risk at which the entity would originate new loans for that particular class of financial instruments. In other words, if the level of credit risk of a financial instrument at the reporting date exceeded the credit underwriting limit for this particular class of financial instruments, lifetime ECL would need to be recognised for the financial instrument. However, this approach has similar advantages and disadvantages to an approach based on the absolute level of credit risk or an approach based on the change in the credit risk management objective.
Using this approach an entity would recognise lifetime ECL on all financial instruments it holds with the same borrower if the credit risk of the borrower has reached a specified level at the reporting date (including on newly originated or purchased financial instruments at market terms). This approach is really more akin to an absolute approach, in that once the credit risk of a borrower reaches a particular level (or exceeds it) lifetime ECL would be recognised on all exposures to that borrower. However, the staff did note that the proposals in the ED could support such an approach if applied appropriately (see staff recommendations below).
The staff recommendations for the IASB were:
- to confirm that lifetime ECL shall be recognised when there is a significant increase in credit risk since initial recognition;
- to clarify that the assessment of significant increases in credit risk could be implemented more simply by establishing the initial maximum credit risk for a particular portfolio (by product type and/or region) (the ‘origination’ credit risk) and then comparing the credit risk of financial instruments in that portfolio at the reporting date with that origination credit risk. This would be possible for portfolios of financial instruments with similar credit risk on initial recognition;
- to emphasise that the assessment of significant increases in credit risk can be implemented through a counterparty assessment as long as such assessment achieves the objectives of the proposed model and the outcome will not be different to what it would have been if financial instruments have been individually assessed. This will be when the counterparty assessment neither delays the recognition of lifetime ECL for those financial instruments for which credit risk has increased significantly nor recognises lifetime ECL for financial instruments for which credit risk has not increased significantly.
The staff also rejected the notion that the proposals would be unduly costly to implement. They stated that the costs that would be incurred to implement the changes to processes would be no more than those needed to manage the entity’s business effectively.
One Board member raised concerns with regard to the use of a maximum credit risk approach to assess whether credit risk has significantly increased. He was concerned that such a “fixed point” of comparison would not necessarily take changes of circumstances into account.
Another Board member raised a concern that any clarifications or examples should be written in such a way that it does not appear as though the IASB is prescribing a specific method to achieve the objectives set out in the standard.
However, all 14 board members present voted in favour of the staff recommendation.
Issue B: whether the proposals provide sufficient guidance on when to recognise lifetime expected credit losses.
The proposals in the ED require an entity to consider the following when assessing whether there has been a significant increase in the credit risk of a financial instrument:
- the credit risk of the financial instrument at initial recognition;
- the remaining maturity of the financial instrument; and
- the best information available that might affect credit risk.
The ED includes application guidance in paragraph B20 (see Appendix A) that sets out the types of information that may be relevant when determining whether the recognition of lifetime ECL is required.
Many respondents supported the principle-based approach taken in the ED of providing indicators of a significant increase in credit risk rather than prescriptive rules and ‘bright lines’ about what constitutes a significant increase in credit risk.
However, respondents noted some potential inconsistencies in the proposals of the ED and asked for clarification. In particular, although the proposals set out a range of information that could be considered, some respondents thought that the proposals as currently drafted could be interpreted to explicitly require the use of a probability of default (PD) approach when assessing significant increase in credit risk. Those respondents are concerned that this would require the explicit calculation and storage of the lifetime PD curve for a financial instrument to compare the expected remaining lifetime PD at inception with the remaining lifetime PD at the reporting date.
In contrast to the general level of agreement, some respondents, most notably some regulators and standard-setters, are concerned that allowing entities to use their internal risk management processes may lead to an assessment of whether there has been a significant increase in credit risk that is too judgemental and open to manipulation by the reporting entity. They are concerned that this would harm comparability and would lead to diversity in practice.
The staff are of the view that respondents’ concerns that the assessment would be based on judgement and would be open to manipulation could be addressed in drafting. The final Standard:
- could require that the indicators that are relevant for the particular financial instrument being assessed must be considered when assessing a significant increase in credit risk;
- could place more emphasis on the fact that a significant increase in credit risk occurs earlier than non-performance or default;
- could state the principle underlying the focus on the significant increase in credit risk more prominently in the application guidance—ie that a significant increase in credit risk may have occurred when rates or terms of existing financial instruments would be different because of changes in credit risk if the financial instrument were to be newly originated or had been issued at the reporting date; and
- will provide examples on how non-borrower-specific information (ie macroeconomic information) would be factored into the assessment of a significant increase in credit risk, as tentatively decided by the IASB in its September Board meeting.
The staff recommended to confirm the guidance provided in the ED on how to assess changes in credit risk and provide clarifications to address the concerns on the operation of the model.
All 14 Board members agreed with the staff recommendation.
Issue C: whether the assessment of when to recognise lifetime ECL should consider only changes in the probability of a default occurring, rather than changes in ECL (or credit loss given default (LGD)) and whether and when that assessment should be based on the lifetime probability of default or the 12-month probability of default.
Most respondents agree that an assessment of when to recognise lifetime ECL should only take into consideration the changes in credit risk (ie the risk of a default occurring) rather than changes in ECL (ie the severity of the loss). Although some stated that other factors (such as LGD) do affect the assessment of increases in credit risk, they supported the proposal in the ED because:
- they consider the probability (ie risk) of a default as the most relevant factor in assessing credit risk; and
- tracking only the probability of a default occurring makes the model more operational and less costly to apply because an assessment based on the probability generally aligns with their credit risk management.
However, a few respondents disagreed with the proposal in the ED. They stated that in an expected loss model, a movement to lifetime ECL should not be based solely on the risk of a default occurring but should also be based on changes in the ECL (ie including LGD). However, these are not new arguments and during the development of the proposals, the IASB had already considered recognising lifetime expected losses based on changes in the ECL instead of the risk of a default occurring. The IASB rejected it for the reasons stated in the Basis for Conclusions (see paragraph BC69 and BC 73).
The staff recommendations for the IASB are as follows:
- to confirm that the assessment of when to recognise lifetime ECL should consider only changes in the probability or risk of a default occurring, rather than changes in the amount of ECLs (or the credit loss given default (LGD); and
- to clarify that an assessment based on the change in the risk of a default occurring in the next 12 months is permitted unless circumstances indicate that a lifetime assessment is necessary and provide examples of when a 12-month assessment would not be appropriate and a lifetime assessment would be necessary.
All 14 Board members agreed with the staff recommendation. However, one Board member expressed concern that any guidance or clarification should be provided as a means of achieving the principles and objectives of the standard, but it should avoid any wording that might suggest that the IASB prescribes certain approaches.
Issue D: whether an entity shall re-establish the loss allowance (or provision) at an amount equal to 12-month ECL when the criteria for the recognition of lifetime ECL are no longer met.
The staff recommended to confirm the proposal that a loss allowance measured at an amount equal to 12-month ECL shall be re-established for financial instruments for which the criteria for the recognition of lifetime ECL are no longer met.
All 14 Board members agreed with the staff recommendation.
Paper 5B – Operational simplification – 30 day past due and low credit risk
In the ED there are two operational simplifications:
- there is a rebuttable presumption that credit risk has significantly increased on a financial instrument, over the remaining life, when contractual payments are more than 30 days past due
- the entity need not assess for significant increase in credit risk on instruments that are low credit risk.
30 Days past due
The majority of respondents supported the inclusion of the ‘more than 30 days past due’ rebuttable presumption. They argued that it was in line with credit risk management, and because delinquency is a lagging indicator, a longer period would not be appropriate. In contrast, those that opposed the proposals argued that it was a lagging indicator and indicated concerns that preparers may only rely on the ‘more than 30 days past due’ presumption in isolation, whereas others argued that the line should be further away than 30 days. This is because instruments may move back and forth too readily between stages 1 and 2. There were also mixed views on whether the presumption should be rebuttable. This view was also shared by one board member. However, the staff and other Board members did not think it would be appropriate to require all instruments to be at lifetime expected credit losses when the contractual payments are more than 30 days past due. This is because it may not always be reflective of significant increases in credit risk.
The staff recommended that the Board:
- Keep the rebuttable presumption that there was a significant increase in credit risk when the contractual payments are more than 30 days past due.
- Clarify that the objective of the rebuttable presumption is to serve as backstop identifying those instruments that have experienced a significant increase in credit risk.
- Clarify that the presumption is meant to result in recognition of lifetime expected credit losses before there is objective evidence of impairment.
- Clarify that an entity can rebut this presumption where it has reasonable and supportable information that more than 30 days past due is not the point at which a significant increase in credit risk arises.
The Board voted 12 in favour of the staff recommendations, 2 members were not present.
Low credit risk
The ED proposed an exemption from assessing significant increases in credit risk for financial assets that are low credit risk at the reporting date. Accordingly the entity would continue to recognise the 12 month expected credit loss allowance on these assets. The proposals further included a description of low credit risk and illustrated this description by using an example of ‘investment grade’.
Respondents were mixed in their feedback on the proposals. In addition respondents asked for a number of clarifications. The respondents that supported the proposals argued that:
- It reduces the operational burden and
- It would not be appropriate to recognise lifetime expected credit losses for high quality instruments.
In contrast those that oppose the proposals stated that:
- There is no clear notion of low credit risk
- That downgrades within investment grade can reflect significant increases in credit risk
- In some cases the exemption results in operational burden due to additional layer of assessment
The staff recommended to describe low credit risk as:
- Having a low risk of default
- In the near term the borrower has a strong capacity to meet its obligations
- In the longer term, the borrower’s capacity may reduce, but will not necessarily be reduced by adverse business and economic conditions
The staff also recommended to clarify in an example that the standard does not require that instruments have external ratings to be able to qualify as low credit risk. The staff also recommended that global rating scales should be used rather than national ratings scales.
One board member struggled to accept that if a credit rating drops below investment grade that it would not automatically result in life time expected credit losses. This board member wanted it to be an absolute point at which an entity is required to measure at lifetime expected losses.
When put to the vote, the Board agreed with the staff recommendation with the proviso of making sure that when the staff draft the paper they make the intentions very clear of what low credit risk is intended to achieve.
Paper 5C – Measurement of expected credit losses
The ED proposed that in applying the general model, an entity recognises:
- Lifetime expected credit losses for instruments whose credit risk significantly increased
- 12-month expected credit losses for all other instruments
In measuring the expected credit losses, the ED proposed that the measurement reflects:
- An unbiased probability-weighted amount of expected credit losses
- Time value of money
With regard to the discount rate, the ED proposed to permit, for instruments in Stage 1 and Stage 2, to use any ‘reasonable rate’ between and including the risk free rate and effective interest rate. The Board has consistently said that the effective interest rate is the conceptually correct answer. However, to address the operability concerns with tracking the effective interest rate which was raised in the 2009 proposals, the Board believed that from a cost/benefit analysis permitting a range of discount rates would make the model more operable.
Many respondents did not agree with the proposals and indicated that EIR would be the more conceptual answer.
One board member had particular issue with the fact that an entity could use the risk free rate. He did not believe that the risk free rate was an applicable rate to use and therefore it should not be a possible point in the ‘relevant range’.
The board did not agree on the specific guidance that they would like to include. But rather, the board asked the staff to develop guidance that would prioritise EIR to the extent that it was possible to use it. However the board acknowledge that a range might still be appropriate due to operability issues.
With regard to the other measurement guidance in the exposure draft, the board agreed with a staff recommendation to retain the guidance and add clarifications where required to improve the understanding of the proposals.
Paper 5D – Modifications
The ED proposed that if contractual cash flows of a financial asset are renegotiated or otherwise modified, and modification does not result in derecognition, the entity should adjust the gross carrying amount of the asset to reflect the revised contractual cash flows. The gross carrying amount should be the present value of the estimated future cash flows discounted at the asset’s original EIR.
For the purpose of determining whether a significant increase in credit risk has occurred, the entity shall compare the credit risk at the reporting date based on the modified contractual terms to the credit risk at the initial recognition based on the original, unmodified contractual terms. If the renegotiation or modification results in the derecognition of that financial asset, the date of the modification shall be treated as the date of the initial recognition of the modified asset for the purposes of tracking increases in credit risk.
The staff recommended that the board confirm the proposals in the ED. The Board agreed unanimously. The Board also agreed to add some clarification where it is necessary to make sure the intention of the proposals is clear.