Federal Deposit
Insurance Corporation

June 29, 2005

Mr. Robert E. Feldman
Executive Secretary
Federal Deposit Insurance Corporation
550 17th Street, NW
Washington, DC 20429

Dear Mr. Feldman:

Re: Interagency Proposal on the Classification of Commercial Credit Exposures

Thank you for the opportunity to comment on the proposed “Classification of Commercial Credit Exposures”.

Overall, the proposal is a step in the right direction. The current framework for the classification of commercial credits essentially addresses only Probability of Default (PD) and does not consider loss mitigants. This one dimensional classification system does not give due consideration to loss severity. Under the current classification system, it is possible to have two different loans with significantly different levels of expected loss receive the same rating and therefore the same required capital allocation. The proposed framework addresses this limitation with the addition of a facility rating that takes into effect the loss potential in the event of default.

While the proposal appears reasonable, comments on some of the specific areas within the framework are as follows:

1) “Commercial Credit Exposure”: More clarification is needed on what will be considered “commercial”. Our business portfolio is separated into those loans considered to be “Small Business” and those considered to be “Commercial”. Will the examiners use each bank’s definition of “Commercial” or some standard definition of “Commercial”? If the new framework only includes “Commercial” loans, what if any changes will be made for Small Business and Retail loans? These questions need to be addressed in the proposal.

2) “Default” borrower rating: The classification “Default” could potentially lead to confusion between legal default (e.g. Loan covenant violation) and the risk rating “Default”. A suggestion would be to call this rating “Impaired” instead of “Default”.

3) Loss severity: The proposal states that an institution may estimate loss severity by “mirroring the institution’s allowance for loan and lease losses (ALLL) methodologies.” The proposal does not say that the methodologies should be consistent with FASB 114. In order to eliminate any potential confusion about the methodology, the proposal should state that methodologies for estimating loss severity should be consistent with FASB 114.

FASB 114 works well for large loans that are individually analyzed but does not work for small loans that are not individually analyzed. The proposal does allow an institution to use an alternative rating approach for small loans. The flexibility allowed is appreciated; however, the challenge is having enough data to establish Loss Given Default (LGD), or loss severity, for all collateral types. Our institution is capturing loss data but does not have sufficient data to establish LGD for all collateral types.

The proposal refers to “loss severity” and not specifically to “Loss Given Default” (LGD). Loss severity could mean “Loss Given Loss” which is a different concept. The proposal should clearly explain loss severity, especially for smaller loans that may be aggregated into pools.

Further, as a practical matter, statistical validation of the difference between 0% loss severity and loss severity of 5% or less for pooled loans will be difficult. Several years of data will be necessary to statistically distinguish loss severity to this degree of precision.

4) Remote Risk of Loss: The proposed framework specifically mentions cash, marketable securities, commodities, livestock, agricultural crops, and asset-based facilities as possibilities of collateral that may result in a “remote risk of loss”.

The use of these examples in the body of the proposal causes confusion and could lead to interpretation that is too prescriptive and rules driven. The proposal should allow an institution to demonstrate that there is a remote risk of loss regardless of the type of collateral used to secure the loan in question.

5) Split Facility Rating: This idea is presented in the proposal but is not fully explained. The concept should be more fully developed in the framework.

6) Effective Date: There was no effective date proposed. To implement the proposal, significant data systems programming changes will be required. Institutions should be given sufficient time to transition to the new framework.

Again, thank you for the opportunity to comment on this proposal. I would encourage continued dialogue on this proposal with the industry.

Sincerely,

Kenneth L. Daniels
Executive Vice President

Copy to:
W. Kendall Chalk, BB&T
Mike Morgan, FDIC