In March 2013, the IASB issued the Exposure Draft Financial Instruments: Expected Credit Losses, which proposed a new model for the recognition, measurement, presentation and disclosure of expected credit losses. The comment period ended 5 July 2013. 

At this meeting the IASB discussed feedback from

• outreach activities
• field work
• comment letters

Feedback from outreach activities

Outreach activities were undertaken in order to supplement the comment letter process and interact with a broader range of interested parties during the comment period. The participants included preparers, users, auditors, national standard-setters, regional bodies and regulators. 

Generally speaking, the majority of the participants (including users) in the outreach activities supported the proposals in the exposure draft. They did not support a model that recognises lifetime expected losses for all assets at all times. The vast majority of the participants supported a deterioration model that distinguishes between financial instruments that have experienced significant deterioration in credit quality and those that have not, because:

• It achieves an appropriate balance between the economics of lending and operational complexity of an expected credit loss model
• It is closely aligned to credit risk management practices
• It will provide useful, relevant and timely information about expected credit losses
• It measures lifetime expected credit losses on only those items that have deteriorated in credit quality

Outreach from field work

The IASB invited a small number of preparers who represented the major geographical regions across the world and who were at different levels of sophistication to participate in the fieldwork to test and discuss the proposals.  The boards spent considerable time discussing the results from the field work done of both staffs. Some of the findings are summarised below:

• Operational challenges for implementation – Some participants were initially concerned that the assessment of significant deterioration is based on the change in the credit risk of individual instruments and not on the changes in the counterparty credit risk. However, over the course of the fieldwork a number of participants found ways to deal with the difference between the change in the counterparty credit risk and the change in the credit risk of the instrument since origination. Ultimately, they applied the proposed model at an instrument level and no longer stated this as an area of concern. 
• Responsiveness of the proposed model compared to IAS 39 – By asking respondents to estimate anticipated allowance balance under IAS 39 and allowance balance under the exposure draft, they were able to assess the anticipated responsiveness of the model. Participants found that the proposed model was more responsive to changing economic conditions than the current IAS 39 model. Participants were provided a series of economic information so their proxy forecasting was more perfect than it would be in reality. The staff acknowledged that tis assessment has imperfections, however, it did provide an estimate of the responsiveness of the proposed model. 
• Directional impact on allowance balances – Almost all participants observed a noticeable increase in the allowance balance on transition and throughout the entire economic cycle compared to the current IAS 39. 
• Portfolios other than mortgage portfolios – On transition, the allowance measured under the proposed model is between 25% and 60% higher than compared under IAS 39. The allowance measured equal to lifetime expected credit losses on all the financial assets is between 50% and 140% higher than compared to IAS 39. At the point in the cycle where the allowances are the highest, i.e. where the economic forecast is the worst, the allowance measured in accordance with the exposure draft is between 50% and 150% higher than in the same period compared to IAS 39.
• Mortgage portfolios – On transition, the allowance measured in accordance with the exposure draft is between 30% and 250% higher than compared to IAS 39. The allowance measured equal to lifetime expected credit losses on all of the financial assets is between 130% and 730% higher compared to IAS 39. At the point in the cycle where the allowances are the highest, i.e. where the economic forecast is the worst, the allowance measured in accordance with the exposure draft is between 80% and 400% higher than in the same period compared to IAS 39. 

Some of the results are not yet finalised and the staff will provide a full analysis of the results of the fieldwork during the September 2013 meeting. 

Feedback from comment letters

The feedback from the comment letters (which largely overlaps with the comments received from the outreach activities) covered the following areas:

• 12-month expected losses – most accept the 12-month expected credit losses and welcome the ability to use different methods to calculate it. However, some are concerned that a probability of default approach is implicitly required and request clarification that other methods could be used.
• Significant deterioration in credit quality – the vast majority of respondents support this criterion for recognising lifetime expected credit losses on a financial instrument. However, respondents did raise a number of detailed questions and concerns. 
• Expected credit losses on financial assets mandatorily measured at fair value through other comprehensive income (FVTOCI) – irrespective of their views on the proposed introduction of a mandatory FVTOCI measurement category, most support a single impairment model for all financial instruments. 
• Low credit risk simplification – respondents had mixed views on the exception that a financial instrument is not considered to have significant deterioration if it has a low credit risk (eg, it is equivalent to investment grade) at the reporting date. However, many commented that the ‘investment grade’ exception could conflict with the principle of significant deterioration.
• 30 day rebuttable presumption – the majority agree with the rebuttable presumption in the ED that financial instruments have experienced a significant deterioration on credit quality if they are more than 30 days past due. However, some are concerned that it would be applied as a bright line and would in some cases be inconsistent with the principle of significant deterioration.
• Discount rate – although there was no specific question on this in the exposure draft, many commented that the effective interest rate should be used to discount the allowance rather than a rate between the effective interest rate and risk-free rate.
• Interest revenue – the vast majority of respondents agree that conceptually interest revenue should be calculated on the net basis for financial assets in stage 3 and that this would be consistent with IAS 39 Financial Instruments: Recognition and Measurement today. However, most prefer the non-accrual of interest revenue which would be similar to regulatory requirements in some jurisdictions, or the gross presentation of interest revenue in all circumstances. 
• Effective date – a range of estimates are provided, however most respondents indicate they would need a three-year lead time for implementation of the exposure draft’s proposals.
• Transition – the majority of respondents agree with the transition proposals overall, but requested some additional clarifications.
• Disclosures – the majority of respondents are concerned about the proposed disclosure requirements and encourage the IASB to consider making the disclosure more consistent with the credit risk management practices, and more focused on the judgements and assumptions made by management. 

Constituents’ feedback on the FASB impairment proposals

The FASB also provided feedback from constituents at this meeting on its impairment model (current expected credit losses).  The FASB found striking differences between the views of users and preparers.  A large majority of investors (3-1 margin) prefer an approach where all expected credit losses are recognised at initial recognition. On the other hand, preparers prefer a model that recognises only some expected credit losses or maintains a threshold that must be met before all expected credit losses are recognised. Those users that prefer the full recognition of expected credit losses on initial recognition do so on the grounds of delayed loss recognition and adequacy of reserves.

The FASB found that preparers generally found the proposed model operational, noting the significant cost involved in implementing and applying the model.  However, the FASB did have to clarify to preparers that they would not necessarily be expected to forecast economic conditions over the remaining life of the asset.

General observations

At the meeting it was clear that the IASB and the FASB are approaching expected credit losses from different conceptual starting points.  The feedback that the IASB received seemed to favour an approach that provided accurate reflections of performance in the income statement.  Conversely, the FASB seemed to have received feedback from users that favour an approach that reflects total expected losses in the balance sheet.  One of the board members’ concluding remarks were that if the two boards were to attempt to reach a converged solution, they will have to consider the feedback received given the two perspectives to provide information that is suitable to both needs.