Footnotes

 

 

 

 

  1. New York Times; 2/3/09.

 

  1. Berlin, Isaiah; The Crooked Timber of Humanity (ed.); Princeton University Press (Princeton, 1990); p. 19.

 

  1. Gorton, Gary; “The Panic of 2007”; Yale School of Management and NBER; p. 12. According to the Fed, the national average for 30 yr. fixed rate mortgages was 6 ¼%.

 

  1. Merrill Lynch 2006 Annual Report; p. 50.

 

5. The 2/22/09 London Sunday Times provides a description of what it was like in the trenches, by a risk manager for a major financial company. He describes mania at full throttle:

 

 

“I once tried to block a large derivative deal because the maths (a British term) didn’t make sense. It wasn’t going to make enough profit to justify the 30 years of credit risk it was going to leave on the balance sheet. But the head salesman, who was behind the trade, would be paid his bonus in one year, not 30. Needless to say, he disagreed with my analysis.

 

My boss and I tried to face him down. He went straight for the jugular and threatened to take the matter to the head of Fixed Income, accusing us of curtailing good client business and short-changing the shareholders. I tried to make my case but my boss’s appreciation of derivatives was weak, to say the least. I could see he had lost my point in the first sentence of my argument….

 

Many senior managers have been as clueless as the outside world as how to value these (derivatives) trades. They are simply sitting at the top of the pile, praying they have timed it well to enjoy a couple of good years and allow themselves the opportunity to cream off the mother of all bonuses.

 

Job protection is a career in itself. All these complex products are designed and controlled by complicated computer programs. It has been known for the analysts who build the valuation systems to add pages and pages of unnecessary computer code so as to conceal what is really happening.

 

The managers are unable to check their work….”

 

 

We do not know if this manager worked for RBS, the Royal Bank of Scotland. The 2/24/09 WSJ reports that RBS would like to government insure  £325 billion in toxic assets. Compare this with RBS’ average 2007 shareholder equity of  £43 billion, levered at 27:1. We suspect this problem is fairly typical.

 

Level 3 Assets are assets held on the books of financial institutions priced using unobservable inputs and management judgment. They are probably to be viewed with some skepticism.

 

 

6.      Niebuhr, Reinhold; The Essential Reinhold Niebuhr (ed.); Yale University Press (New Haven, 1986); p. 172.

 

  1. Madison, James; The Federalist No. 51; Independent Journal (1788).

 

  1. Greenspan, Alan; FT.Com; 10/23/08.

 

According to David Frum, on the 3/15/09 Meet the Press, the economic news during the George W. Bush administration was not good. The only good news was the increase in house prices, and “no one wanted to get in the way of the story.” Mr. Frum was the former president’s speechwriter. Major events are often caused by perfect storms, at the intersection of several historic factors that this essay discusses. 

 

  1. In a 2/9/09 address to a Conference on Stress Testing, Andrew Haldane of the Bank of England furthermore said, “It is perhaps no coincidence that the last three truly systematic crises – October 1987, August 1998, and the credit crunch which commenced in 2007 – were roughly separated by a decade. Perhaps ten years is the threshold heuristic for risk managers.” Note that ten year average figure.

 

The problem with all the value at risk statistical models (even assuming the statistical models to be valid) was that they were calibrated in the “Golden Decade” of 1997-2007, and the risk analyses were “precisely calibrated in-sample.” Thus, a comparison of the statistics of that decade with the long-run data of English GDP since 1857 (they do have an historical perspective) reveals that the period of moderation was most atypical:

 

 

   

                                  Distribution of UK GDP Time Series

           

                                                 Golden Decade     Long-run

 

Mean                                              2.9                       2.0

Standard Deviation                          .6                       2.7 (note especially)

Skew *                                            0.2                     -0.8

Kurtosis *                                      -0.8                     -2.2

 
 
 


           

 

 

 

 

 

 

 

 

 

 

* These are more general measures of statistical distribution. Skew measures lopsidedness and kurtosis measures whether a distribution is tall and skinny or short and squat, relative to a normal distribution with the same standard deviation. We think that the use of quantitative measures is ultimately a qualitative decision. Because, as Warren Buffet says, “It is better to be approximately right rather than exactly wrong.”

 

 

 

RETURN TO PREVIOUS PAGE