This site uses cookies to provide you with a more responsive and personalised service. By using this site you agree to our use of cookies. Please read our cookie notice for more information on the cookies we use and how to delete or block them.
The full functionality of our site is not supported on your browser version, or you may have 'compatibility mode' selected. Please turn off compatibility mode, upgrade your browser to at least Internet Explorer 9, or try using another browser such as Google Chrome or Mozilla Firefox.

Financial instruments – Impairment

Date recorded:

The FASB and IASB met in Norwalk to continue its previous week's discussion on credit impairment. The discussion comprised:

  • A presentation by Darrin Benhart and Randy Black, Office of the Comptroller of the Currency, on loan loss data
  • A high-level overview of the impairment models discussed previously.

The meeting was for informational purposes only and accordingly, no decisions were made.

Presentation on loan loss data

To aide in its discussion on credit impairments, the Boards invited Mr. Benhart and Mr. Black to provide some specifics regarding loan loss data compiled from portfolios of different financial institutions in the U.S. Highlights of the credit loss data presented to the Boards are as follows:

  • Losses do not occur smoothly throughout the life of a financial asset
  • When losses occur, the losses tend to be significant
  • Financial assets that have a low internal risk rating (e.g., BBB-rated) typically have a quicker default period.

When asked whether this U.S.-based credit loss data could indicative of data on a global level, Mr. Benhart indicated that similarities may exist but without further testing, no definitive conclusions could be drawn.

High-level overview of impairment models

The FASB and IASB staffs outlined the various alternative models when recognising credit impairment losses that were briefly discussed by the Boards in the previous week. The models are as follows:

  • Alternative 1: Immediate recognition of lifetime expected losses. This alternative would have the following features:
    • Amount of credit loss estimate: The amount of the expected credit losses would be the full amount of losses expected by the entity for the portfolio of the loans, regardless of when they occur during the lifetime of the loan portfolio. For open portfolios, the amount of the credit loss would be determined by applying a loss rate to the portfolio balance (which is an undiscounted principal amount) at each reporting date.
    • Timing of recognition: An entity would recognise all expected credit losses (and changes in expected credit losses) in the current period.
  • Alternative 2: Immediate recognition of losses expected to occur in a shorter emergence period. This alternative would have the following features:
    • Amount of credit loss estimate: The amount of the expected credit losses would be the amount of losses expected by the entity for the portfolio of the loans in a shorter emergence period. For example, this shorter emergence period could correspond to the timeframe used for incorporating into the information set expectations about future conditions. If that timeframe is 2-3 years out, then the coverage period could likewise be 2-3 years worth of losses. For open portfolios, the amount of the credit loss would be determined by applying a loss rate to the portfolio balance (which is an undiscounted principal amount) at each reporting date.
    • Timing of recognition: An entity would recognise all expected credit losses (and changes in expected credit losses) as determined above in the current period.
  • Alternative 3: Recognition of losses expected to occur in a shorter emergence period over the emergence period. This alternative would have the following features:
    • Amount of credit loss estimate: The amount under this alternative would be consistent with Alternative 2; that is, the amount of the expected credit losses would be the amount of losses expected by the entity for the portfolio of the loans in a shorter emergence period.
    • Timing of recognition: An entity would recognise expected credit losses for the shorter emergence period over that emergence period.
  • Alternative 4: Recognition of lifetime expected credit losses using a time-proportionate approach. This alternative would have the following features:
    • Amount of credit loss estimate: The credit loss estimate would be the full amount of the losses expected over the life of the portfolio of assets. However, the timing of recognition would depend on whether an asset is in the good book or in the bad book.
    • Timing of recognition of credit losses: The expected loss (EL) estimate is made at the end of each period for the assets in the portfolio at that date. If the assets are in the good book, the EL estimate is then allocated over the weighted average life of the portfolio. In an open pool setting, a constantly updated EL estimate is allocated over the total weighted average life of the portfolio. For the bad book, ELs are fully provided for (i.e., when an asset is moved to the bad book, the lifetime ELs are recognised fully in the allowance account, as are the effects of any subsequent changes in EL estimates on the bad book).
  • Alternative 5: Time-proportionate approach with notional sub-portfolios to accelerate recognition of expected losses. This alternative would have the following features:
    • Amount of credit loss estimate: The amount of the expected credit losses would be the same as in Alternative 4.
    • Timing of recognition of credit losses: For assets in the good book the allocation of the expected credit losses would be driven by notional sub- portfolios with different ages. This would accelerate the recognition of some expected losses in particular circumstances which could align the recognition of credit losses more closely with the expected loss rate for the group of assets.
  • Alternative 6: Recognition of credit losses using a 'middle' book to accelerate recognition of expected losses. This alternative would have the following features:
    • Amount of credit loss estimate: The amount of the expected credit losses would be the same as in Alternative 4.
    • Timing of recognition of credit losses: The recognition of the expected credit losses would be similar to that in Alternative 4; however, an additional credit loss estimate would be recognised immediately for assets in the middle book, to provide earlier provisioning of credit losses.
  • Alternative 7: Steven Cooper's alternative approach. The model is a combination of a simplified IASB ED approach but with a minimum balance overlay. The approach gives the same information as the IASB ED, including a full catch up. The minimum balance adjustment addresses the problem that the loan loss allowance may not cover actual losses and the concern that upcoming or foreseeable losses are not covered. It combines the IASB 'allocation' and the FASB 'cover upcoming/ foreseeable losses' philosophies.

The Boards made no definitive decisions but did provide the staffs with direction by asking the staffs to further develop Alternative Models 2, 4, and 5. In doing so, the Boards asked the staffs to illustrate the alternative models using a similar fact pattern that applies to a single asset and open and closed portfolios. The Boards are hopeful to redeliberate these models at the next joint meeting in December.

Related Topics

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.