Federal Deposit
Insurance Corporation

In reviewing the proposal my initial assessment is that it is definitely a step in the right direction.  As we all know there are two components to loss - default probability and loss in the event of default.  The current inter-agency classification guidelines address the former to some extent but do a poor job of assessing the latter.  In discussions with examiners the issue of a well-defined weakness is normally not an issue of contention but the probability of loss typically is.  This proposal establishes a good framework for determining a bank's potential loss exposure and is preferable to the current system that assigns common reserve adequacy factors without considering the underlying collateral and/or other credit enhancements. 

While the framework appears reasonable establishing loss factors could be an issue.  Under default the proposal defines a low loss severity of less then or equal to 5%.  Intuitively how do you classify a loan as default with that type of loss factor?  It may be a moot point but if a loan is classified default, moderate or high loss severity appear to be the appropriate and common facility ratings.   

1. Our institution is about $1.3 billion in total assets and I do not envision a problem in implementing this change.  Currently, we assess each criticized and classified loan over $50M and assign it a specific reserve factor.  Loans under $50M are pooled and assigned a risk factor based on the typical collateral that supports said pool.  We would continue to utilize our current risk grading system for pass loans.

2. Implementation expenses should be negligible given our current level of criticized loans.  We would make manual changes to our loan subsystem to reflect the proposed changes.

3. Training expenses would be expected to be moderate as we would likely send out a circular to all lenders and follow up with an in-house training session explaining the changes.  Credit Administration currently monitors all criticized loans and would continue to do so.  I would not expect these changes to impact programming as we would simply develop a different user code to identify the different classification categories.

4. The proposal examples provide a clear thought process and were definitely beneficial. The resultant ratings were reasonably thought out and determined.

5. The proposed treatment of guarantors is reasonable in that the inherent value of a guaranty structure is the guarantor's capacity and willingness to support the underlying debt.  If the guarantor does not evidence performance then the default risk definitely increases and the bank's classified segment should accordingly reflect this.

Thank you for the opportunity to comment and I support the proposed changes.

Ed Gomez, VP
Exchange Bank
Santa Rosa, CA