On FAS 157 and Measurement Fallacy

Posted on by

Roger Ehrenberg has a great post today, “Straight-talk on FAS 157: Blackstone and their Banker Buddies Have it Wrong,” which I suggest you all read.

Although I was taken with the entire discussion, the last paragraph caught my attention:

So why do risk managers and bank managements’ so consistently make bad decisions? Probably because there is an over-reliance on measures that are seemingly quantifiable. They can measure delta. They can measure vega. They can measure theta. They can measure gamma (or at least they think they can). They can estimate credit loss ratios. But what about liquidity? When you are quantifying factor exposures, how exactly do you model liquidity as other risk factors change? It is a very, very hard question. Sometimes risk management requires judgment beyond computers, which is hopefully one of the biggest take-aways from the current credit melt-down. My sense is that there is currently a fear to manage without a machine telling you what to do. It is kind of like the drunk looking for his lost keys by the streetlight, simply because this is where he can see. But the likelihood of his keys being within the illumination of the streetlight is very, very low. Some of the best risk managers, guys like Gus Levy of Goldman Sachs and Ace Greenberg of Bear Stearns, didn’t rely on computers but relied on instinct, savvy and experience. We need more of this. It’s called leadership. Let’s not cloud the issue. It’s not about FAS 157 or any other accounting rule. It has been and always will be about management.

If that viewpoint resonates with you, you might enjoy a longer-form treatment in Across the Board (Conference Board) article, “Management’s Great Addiction.”

Print Friendly, PDF & Email


  1. JKH

    The entire Ehrenberg post is a true gem.

    Gap management was the original form of bank balance sheet risk management. It’s still important in commercial banks, although securitization deflected focus to some degree. And Investment banks in particular developed a very lax attitude to the importance of the idea in whatever risk monitoring they did, (although they should have learned from the Salomon Brothers event.)

    The importance of FAS 157 on liability valuation is also critical, but gets little attention.

    As far as the final paragraph is concerned, it’s gold.

    But it’s also the case that the best financial services CEOs, who understand risk management does not mean quantitative slavery, also tend to see FAS 157 as accounting and capital slavery. They argue that FAS 157 content should be applied rigidly to disclosure, rather than inflexibly to accounting.

    The CEOs of the 3 best run financial institutions in Canada (Manulife Financial, Royal Bank, Bank of Nova Scotia) have recently and separately made quite public statements that are remarkably similar connecting these two points on risk management and FAS 157. (And none of these institutions has any major problems with sub-prime or securitization exposures).

  2. Doc Holiday

    Re: “It is kind of like the drunk looking for his lost keys by the streetlight, simply because this is where he can see. But the likelihood of his keys being within the illumination of the streetlight is very, very low. “

    That is funny– and the reality often results in the fact that the keys were in the pant pockets from the start — and the street light is just a beacon that distracts a person from something obvious.

    Computer models depend on people to fit the data to a problem and then try to fit solutions into equations that fit into a finite loop that makes everyone happy — for a short time, then the model needs adjusted, fine tuned and as more time passes, the model becomes more and more meaningless. This is very much like a drunk looking around for the obvious and then relying on the convenient illumination of something easy to gravitate towards, versus having self control and not getting so drunk that you can’t ponder the obvious!

  3. Anonymous

    One thing I think people misunderstand is that mark to market does not mean an inflexible, calculation based approach to valuation. If you actually look at the IPEV standard (which is the gold standard for VC/PE mark to market methodology) it allows for enormous application of judgement and discretion.

    Good post on this subject at Carriedinterest.com

    Link is: http://www.carriedinterest.com/2009/01/the-valuation.html

Comments are closed.