Jeff Judy

Jeff's Thoughts - September 14 , 2011

The (Unfortunate) Evolution of Risk Rating Systems

In the years that have passed since risk rating systems became a core component of every bank's credit risk management practices, those systems have evolved. And that is not always a good thing.

We tend to think of evolution in the "survival of the fittest" context. We think that as species evolve, they always become more viable over the long term. Evolution produces creatures that are more finely tuned to their immediate environments, giving them competitive advantages over less adapted species.

It is easy to forget, then, that the vast majority of species that have ever lived on this planet are extinct. The problem is that the more specialized an organism becomes to fit a particular environment, the more vulnerable it is to changes in that environment. A sudden shift in conditions wipes out highly adapted species very quickly.

The evolutionary trend in risk rating has been the compression, if you will, of risk rating bands into just a few categories. These days, many banks lump all credits of acceptable quality into one rating, or maybe two. All the "passing" credits end up together, instead of being spread across several ratings, as in the original design.

And this is not a complete surprise, given some of the "evolutionary pressures" in our industry of late. In particular, banks tend to move in this direction through their increased interactions with regulators, particularly the FDIC.

The FDIC thinks very much in this simple pass/fail way. But that makes sense for them, because the FDIC is basically an insurer. They are interested in whether or not they will have a "claim" to pay.

In terms of credit risk management, however, you are giving up an awful lot of useful information by drifting into a system that only has a few "risk buckets" in which to place credits. How can you track trends effectively if the only place a "good" credit can go is into the "bad" bucket? It is hard to see underlying patterns of different portfolio segments when you can only draw a line between two points on your risk rating scale.

It is rather like trying to monitor the educational results of a school district that only gives out C's and F's. There just isn't enough information about changes over time to be very useful.

Sophisticated analysis tools give banks the power to predict the changing shapes of their credit portfolios more precisely than ever before. Such calculations open the door to more efficient allocation of reserves and capital.

A better backstop at less cost, in other words, leaving more resources to build your business.

But it all starts with a meaningful, information-rich risk rating system. And that means that you have to do a better job of grading credits than just handing out C's and F's.