Financial instruments – Impairment

Date recorded:

Alternate approach developed by the working group

During the May 2011 Board meeting, the IASB and FASB had asked a small working group of staff and Board members to develop a variation of the proposals in the Supplementary Document (SD) considering the feedback received through outreach activities and comment letters. The IASB staff presented the Boards with a high level summary of the approach developed by the working group.

The model developed by the working group would seek to reflect the general pattern of deterioration of the credit quality of loans. Similar to the proposals in the original exposure drafts and the SD, the impairment model would be based on expected credit losses and would develop those expectations of future credit losses based on historical, current and forward looking information that is reasonable and supportable.

The model would divide assets subject to impairment into three categories based on levels of credit deterioration. The first category would be the general portfolio level of assets that do not meet the criteria for either of the second two categories. For this category the working group has developed three possible alternatives for recognising expected credit losses:

  • Alternative A — recognise an impairment allowance equal to 12 months' worth of expected losses only based on current loss expectations by using an annual loss rate multiplied by the balance of assets in category 1
  • Alternative B — recognise an impairment allowance equal to a time-proportional amount of expected credit losses based on current loss expectations by calculating the lifetime expected credit losses multiplied by the factor of the weighted average age divided by the weighted average life
  • Alternative C — recognise an impairment allowance equal to 12 months' worth of expected losses based on initial expectations plus the full amount of any subsequent changes in expected credit losses; this involves two separate calculations, the first is similar to that required under Alternative A and the second is based on the lifetime effect of changes in expectations of future lifetime losses.

The second category would consist of a group of assets that have been impacted by the occurrence of an observable credit event which indicates a possible future default but the specific assets that will default cannot be specifically identified. For the second category, the full lifetime expected credit losses would be recognised as an allowance, but the allowance would be calculated on a portfolio basis.

The third category would consist of assets where information is available that specifically identifies that credit losses have, or are expected to occur for individual assets. Similar to the second category, the third category would recognise the full lifetime expected credit losses as an allowance, but the allowance would be calculated on an individual asset basis.

The IASB staff provided a simple example to illustrate the application of these three categories. Bank Z's entire portfolio consists of loans in Country X that includes mortgage loans in Town ABC and Town XYZ. The national GDP in Country X decreases such that the general level of credit defaults for Bank Z's entire portfolio increases. This change of circumstances would be considered in the amount of expected losses estimated for the first category but would not result in a transfer of loans to the second category where the entire lifetime expected losses would be recognised. However, if housing prices in Town ABC decrease to an level that defaults are expected to increase, all mortgage loans to borrowers in Town ABC would be transferred from category 1 to category 2 and the entire lifetime expected losses would be recognised immediately. This is because the declining house prices have a direct relationship to potential future defaults but it is not clear yet which specific mortgages to borrowers in Town ABC are likely to default. The working group notes the cliff effect resulting from transferring loans from category 1 to category 2 with the recognition of all lifetime expected losses and therefore the working group believes it appropriate to have a notion of an observable event causing loans to be transferred between categories.

The IASB staff also provided the Board with a table that summarises their considerations for each of the approaches which has been replicated below.

General approach Alternative A Alternative B Alternative C
The extent to which changes in information is captured in Bucket 1 (ie the Alternative used for Bucket 1) affects the importance of timing of move to Bucket 2 Operationally simple More responsive to changes in information compared to Alternative A Easiest to rationalise conceptually because represents original expectation of losses plus full effect of changes in remaining lifetime expectations
How to differentiate between Buckets 1 and 2 – clarify when this happens Only one year's worth of expected loss recognised in Bucket 1 allowance balance May be difficult to rationalise conceptually (why apportion future expectations to time period passed?) Most responsive to changes in information compared to Alternative A and Alternative B
Moving from Bucket 1 to Bucket 2 could have a dramatic effect on allowance balance Less responsive to changes in information compared to Alternative B and C Must calculate weighted average age Less operational in an open portfolio setting. May require much data tracking
Difficult to differentiate between Bucket 1 catch-up (changes in lifetime) amount and Bucket 2 (full lifetime) amount
The staff mentioned they had reached out to some of the expert advisory panel (EAP) members to solicit views on the three alternatives for category 1. For operational reasons they generally supported Alternative A over Alternative B and then Alternative B over Alternative C. Some EAP members also had issues with the staff's description of Alternative C as being conceptually the best alternative.

The staff also mentioned that they would consider incorporating the guidance around incurred but not reported (IBNR) losses that exists in current IFRS and US GAAP to provide additional guidance around transfers from category 1 to category 2.

The Boards tentatively agreed to allow the staff to further develop an impairment model based on three categories using credit risk or credit deterioration as the distinguishing feature between the categories. They also tentatively agreed that for the assets in category 2 and category 3 the allowance would be recognised as the full lifetime expected loss and that for category 1 the objective of the allowance would be to recognise an impairment allowance equal to 12 months' worth of expected losses based on initial expectations plus the full amount of any subsequent changes in expected credit losses. However, the Boards acknowledged this objective was based on it conceptual merits and that there may be operational difficulty in complying with that objective; therefore they requested the staff consider how best to make the objective operational.

The IASB Chair also mentioned the Boards would like to re-expose a revised impairment model in September, therefore several decisions would need to be made during the month of July in order to give the staff drafting time in August in preparation to issue an exposure draft.

Related Topics

Related Meeting Notes


Correction list for hyphenation

These words serve as exceptions. Once entered, they are only hyphenated at the specified hyphenation points. Each word should be on a separate line.