If you're interested in reading some detailed analysis of what went wrong in the financial world, the New York Times carried two lengthy articles today that get into it:
First, there's a two-parter, starting with The End Of The Financial World As We Know It. With a title like that, I was hoping that part two would be called "And I Feel Fine", but it's really How To Repair A Broken Financial World. These articles are written by Michael Lewis and David Einhorn, and you might recognize the former as the author of another very interesting article that came out several weeks ago called The End. The new series begins with the story of Harry Markopolis, who had been waving red flags in front of the SEC regarding Bernie Madoff for years, and the authors use this story to illiustrate their point: "What’s interesting about the Madoff scandal, in retrospect, is how little interest anyone inside the financial system had in exposing it." and they then make the same point about the inherent problems throughout the financial system: that it wasn't necessarily that nobody saw warning signs. It's that nobody had incentive to do anything about the warning signs. Not regulators, ratings agencies, traders, executives, boards, stockholders, or investors -- not to mention elected officials.
The other article is just called Risk Mismanagement and it is about the use (or mis-use, or abuse... take your pick) of models that put a dollar figure, known as "VaR", on financial risk. It seems that this number started out as a useful 40,000-foot management tool with known limitations, but it took on a life of its own. It was a number, therefore it must be All-Important, right? Wrong! It was a number and a probability -- and the probability was 0.99, so if the number was X that meant that the model predicted that 99 times out of 100 the risk of loss from a firm's transactions was X or less. The fact that it was expected to be wrong 1% of the time was ignored, and there was never a second number for the multiple of X that might be lost in the 1 in 100 case. By putting too much faith in VaR, risk managers systematically ignored the worst risks, and then the Rand cum Greenspan philosophy that financial institutions are inherently better than regulators at understanding their own risk kicked in: VaR became the accepted formula for determining reserve requirements. That's despite the fact that VaR gave no information at all about the outlier scenarios in which reserves are critical to a financial institution! And to top it all off, the risks that the model considered didn't include the risk of a liquiditiy crisis so even if outliers were considered, the worst-case outlier would still not be accounted for.
Very interesting stuff.
1. ninest12304/23/2016 04:47:25 AM