ALLL: Incurred Loss Model vs. an Expected Loss Model

6/23/2011

By David J. Bayens, CPA

In the financial and banking press we have all seen articles lately dealing with the accounting profession’s new look at loan impairment. Specifically, the issue under consideration is the migration from an “incurred loss model” to an “expected loss model.”

Under current GAAP, the incurred loss model is used in computing credit loss reserves. Under this model, evidence of a loss (a trigger event) is necessary before assets can be written down. This model was widely criticized during the latest economic downturn. Criticisms included:

• Reserves were not adequate- reserves at the start of the downturn were woefully below actual losses.
• Timing of loss recognition- losses tended to be recognized at the end of the crisis which reduced the public’s confidence in banks’ financial reporting.
• Multiple approaches to impairment recognition- this further reduced public confidence.

The concept of an expected loss model was proposed by FASB in January 2011. This proposal was designed to address a part (credit impairment) of FASB’s and the IASB’s project for converging the accounting for financial instruments. A primary objective was to address the weaknesses in the incurred loss model.

Under this proposal an impairment allowance is determined based on the total of:

• for assets for which it is appropriate to recognize expected credit losses over time period, the higher of:
  • the time-proportional expected credit losses, and
  • the credit losses expected to occur within the foreseeable future (which shall be no less than twelve months after an entity’s reporting date); and
• for all other assets, the entire amount of the expected credit losses.

The proposal introduces new concepts and definitions to the credit estimation process including:

• time-proportional expected credit losses,
• foreseeable future, and
• “good book” vs. “bad book” assets- similar to the FAS 5 and FAS 114 concepts, but not the same.

The result of the application of this model is the acceleration of the recognition of losses.

Numerous banks commented on the proposal along with banking and accounting organizations and bank regulators. Most respondents agreed that a change in the ALLL methodology was necessary. However, many found fault with the proposal. Criticisms included:

• The new concepts may be overly complex.
• The lack of clear key definitions invites confusion and concern that users will interpret a new standard differently.
• New concepts were introduced but old concepts were seemingly abandoned.
• There were significant areas that weren’t addressed including: TDR’s, the definition of write-offs, and purchased impaired loans.

A number of respondents recommended that the incurred loss model not be abandoned but be expanded to take into account estimated losses dependent upon future events. Recognizing that a change in concepts would have dramatic implications, certain respondents recommended that FASB:

• Re-expose the entire impairment proposal for comment and insure that the individual elements fit into a coherent standard,
• Provide for adequate field testing, and
• Consider condensing required financial statement disclosures for smaller banks.

Comments of the proposal were due April 1, 2011. At this point it is not clear if or when a new proposal will be issued.