The Standard-Examiner carries a compact Washington Post editorial this morning, briefly discussing the series of events (over two decades) which eventually led to the 2008 downfall of insurance mega-giant AIG, and the resultant daisy chain collapse of the whole world financial sector. It's a good "nutshell" piece, boiling down the root causes of the current global financial crisis in a tidy 428 words:
• A cautionary tale of downfallThe editorial also refers to a three-part investigative article series running last week in the Washington Post, under the bylines of Robert O'Harrow Jr., Brady Dennis and Bob Woodward. We'd followed this series ourselves, and had intended to feature it in an upcoming WCF article. Thanks to the Standard-Examiner's introductory lead, we're delighted to provide this article series right on cue.
For readers interested in a more detailed explanation of how Wall Street innovation outpaced Washington regulation and brought the world economy to its knees, the following three WaPo articles are a definite must read:
1. Greed on Wall Street and blindness in Washington certainly helped cause the financial system's crash. But a deeper explanation begins 20 years ago with a bold experiment to master the variable that has defeated so many visionaries: Risk:
• Part 1: The Beautiful Machine2. By 1998, AIG Financial Products had made hundreds of millions of dollars and had captured Wall Street's attention with its precise, finely balanced system for managing risk. Then it subtly turned in a dangerous direction:
• Part 2: A Crack in The System3. How could a single unit of AIG cause the giant company's near-ruin and become a fulcrum of the global financial crisis? By straying from its own rules for managing risk and then failing to anticipate the consequences:
• Part 3: Downgrades And DownfallDon't forget to bookmark this, Weber County Forum economics wonks. This is by far the most detailed and sophisticated post-crash post-mortem reporting and analysis that we've come across so far -- Pulitzer Prize quality material -- in our never-humble estimation.
Comments are invited as always.
10 comments:
Once again we witness the folly of computer modelling. Garbage in... garbage out.
And as a good counter-measure to those still-blinkered alleged "conservatives" who think FDR and the New Deal made the Depression worse, and who are determined to tip us into a new one, don't miss Paul Krugman's
recent column in the NY Times. The inability of Conservatives, who allegedly revere tradition and the lessons of the past, to learn from history is truly breathtaking.
"Greed and blindness." That's actually more my business (theology), than economics. I have been wondering whether a new business school course called "Human Nature and the Nature of Risk" should be required. The artciles, however, point out the flaw in this approach. Those who made the models that led to this debacle thought they were "the smartest guys in the room." In the end it is not even greed or blindness, but hubris, the sin of pride, that is at the root.
And in conducting the various "why the Titanic sank" investigations that are already underway, let us not forget the role of the so-called "independent" ratings companies, which wrote some truly creative fiction, certifying the bonds and health of essentially bankrupt businesses as "AAA rated." Good article on the hows and whys can be found here.
To believe that the sophisticated financiers running AIG believed that there was no risk in writing billions of dollars of insurance based on the assumption that housing prices could and would forever continue to increase at a rate significantly higher that the rate at which wages were increasing is to believe in the Easter Bunny. These people knew full well what they were doing and they new full well what the risks were. They just didn't care because they pocketing huge sums of money. The sad thing is that they'll pay no penalty for the damage they've wrought.
In defense of computer models.
Now everybody wants to blame the over-reliance on complex computer models for the failures. Computer models are as good as the assumptions that are entered as inputs. It looks like they never ran a model with the assumption that the AAA rating may be lost. And they never ran a model with the assumption that housing prices may fall. Or maybe they ran these models, did not like the results and declared the assumptions unlikely. The problem is not the use of models, the problem is that flawed humans manipulate these models.
So it really was a version of 1929, when the death knell sounded: "Margin; you'll have to put up more margin." The greedy never learn.
So, tell me again how smart are these guys? Are they smarter or dumber than Mayor Godfrey?
I want to know how really stupid Matt Godfrey actually is.
Deidre
Actually the Lil' Lord is quite smart. His problem, which becomes ours of course, is that he is just plum loaded with what "Theologian" above identifies - Hubris.
The one problem I had with the editorial is where they say that this derivative bull shit some how makes for a more liquid and smooth running market and there fore is good in principle. I would love to have some one explain how that is true. As far as I can see it is just one more overcomplicated game devised by greedy operators to put another profit layer into the market that allows them to put billions in their own pockets without contributing anything to the market or the economy. The concept seems to only truly represent greed and the above mentioned hubris.
That small risk is based on computer modeling. Reality is not. The article also does not state the time frame for that risk factor. Is it .15% risk over a day, week, year, or decade? Risk evaluation on a highly complex, interactive, human-controlled system is inherently risky in itself. A single missed or inaccurate factor can easily throw off the calculation. Leverage, total investment involved, and number of different entities using the system will increase the risk. Once the program propagated over several different companies with various units within these companies, with various managers and analysts involved, and trillions of dollars invested the risk factor multiplied to 99.85% chance of catastrophe on a market downturn.
To err is human, to totally screw things up takes a computer.
A computer can take a single human error and execute it a billion times in a blink of an eye.
Post a Comment