Risk Management: Rebuilding the Model

http://knowledge.wharton.upenn.edu/article.cfm?articleid=2268

As you may recall, we mentioned in our 6/12 blog on "Economic Recovery" that Diebold at Wharton had developed a new statistical tool (the Arouba-Diebold-Scotti Index) which provides a simpler formula to track overall economic performance. In this article from K@W, Diebold is participating as part of a panel at Wharton's 12th Annual Financial Risk Roundtable on what it would take to build a more informed financial risk model.

The basic perspective of the panel is that there are, and were, many things about risk modeling that are not, as yet, quantifiable. And yet, the "risk modelers" became overconfident prior to the ww financial crisis because they had become very good at what they COULD quantify. This is, as Diebold says, like the old story about the person who has lost their car keys at night and is looking for them under a lamp post because that's where the "light" was.

Diebold refers to a project that is ongoing at the Wharton Financial Institutions Center in conjunction with the Sloan Foundation on the known, the unknown, and the unknowable that might be worth looking into. For me, the implications here are that risk modelers didn't know what they needed to know but didn't know it - this is a form of ignorance.

Last, Diebold's perspective is not about how we will get through this latest financial crisis (because we will get thru it one way or the other), but, more importantly, how do we avoid things like this in the future. This suggests the need for ways of dealing with moral hazard balanced with criteria for rescuing financial institutions.

There are a group of very bright people in the Obama administration working on this so, while a solution to the problem of "balance" between latitude to act and regualtions may be a while in coming, the early effort to give more power to the Federal Reserve looks like a good start.