Federal Deposit
Insurance Corporation

From: Greg Brown 
Sent: Monday, May 23, 2005 3:59 PM
To: Comments
Subject: FIL-22-2005/Classification of Commercial Credit Exposures

We are a $220mm+ community bank.

We are opposed to the proposed changes to current Loan Classification 
methodology.

1.  The current methodology has served the banking system and the 
insurance fund very well, even through the turmoil of the '80s.  It is 
well understood by bankers and examiners alike, and every bank's 
internal and/or external review system inevitably ties back to the 
existing classifications.

2.  The existing methodology has been in place for over 65 years, 
during which time revisions have been necessary on only four occasions, 
one of which was the interagency adoption of a common definition of the 
"Special Mention" rating.  No showing has been made of the necessity of 
now junking a proven system, in favor of an unproven one.

3.  The Federal Register notice, in the Uniform Agreement, states that, 
"Institutions may incorporate this framework into their internal risk 
rating systems or, alternatively, they may map their internal rating 
system into the supervisory framework."

  a. This unnecessarily increases the burden on our loan operations 
  departments, in that they will have to continue their current work of 
  assessing lending risks, and then adapt those findings to the new 
  methodology.

  b. Apparently no consideration has been given to state-chartered 
  institutions, and the effect a Federal methodology which differs from 
  their state's methodology will have on their overhead and examination 
  processes.

4.  The proposal seems to be aimed, at least partially, at providing 
for a formulaic procedure which presumably would be applied to the same 
extent at all examined institutions, hopefully eliminating " . . . 
differing applications of the current classification system by 
institutions and the agencies."  The examination process, at least in 
the lending area of the bank, is by its very nature a subjective one.  
It will continue to be, at least as long as we have humans involved in 
the process.  The proposed methodology will not eliminate that 
subjectivity.

5.  Finally, there is the Law of Unintended Consequences, and we wonder 
what might unexpectedly occur either during the transition process, or 
after full implementation, that none of us can currently predict.  
Given the long range success of the existing systems, is this a risk we 
are willing to take?

Greg B. Brown,
Chairman and CEO
Union Bank
Benton, Arkansas