That is the question. The (current) incurred loss methodology strongly encourages pooling loans. Most financial institutions only conduct an individual evaluation on impaired credits, although technically you can individually evaluate an unimpaired credit and determine the incurred loss just for that credit (Look up the FDIC’s slide deck on the allowance if you are wondering about this).
There is a problem with pooling credits for loss characteristics, however. One of the things we find quite often when we perform an ALLL model validation is that there is little support behind the creation of the pools. Many Banks use the Call Report pools as the pools for the historical look back. And while this is not a bad thing, there is a need to support why this is the right pool structure for your Bank. We often note that there are too few loans in a pool or one loan that dominates the historical loss rates for that particular pool
Moving to an expected loss methodology changes the way a financial institution looks at its reserving process by capturing the underwriting characteristics into the loss calculation. With expected loss, you (can) begin with the underwriting and the interest rate you have calculated for the loan. That contains the probability of default and the loss given default (or at least it should). For loans with like characteristics in pools big enough to demonstrate loss behavior (think credit cards and auto loans, for example), the expected loss can be tied in to the expected duration of the loan. Pools work very well for these types of loans.
For commercial loans, however, most community financial institutions do not have a sufficient number of like loans to determine the expected loss. But if you consider the use of underwriting data, or even a dual factor grading approach, you can, at origination, determine the expected loss and then use that to identify the loss given default of the loan which can then be plugged into your reserve model.
If it all sounds theoretical at this stage it probably is. Especially for community banks. This is why you need to sit down and really analyze the loans in your loan portfolio to see how you can best approach expected loss. And don’t forget to document your process, especially your assumptions. I like to say, “we lend for the good times and document for the bad times”.
As a follow up note, we are more than happy to discuss this with you. If you are interested please call or email me.