Jan

29

the banker from monopoly[Editor's Note: TBTF = Too Big To Fail]. My friend Rocky Humbert posed the question: ‘If 1000 mini-AIG's and mini-Fannies are all imploding, why is that less of a catastrophic event than a single mega-AIG? Arguably, it's a more serious systemic risk…as the possible chain-reaction will be like the whack-a-mole game at the video arcade.’

His statement presupposes that 1000 separate actuarial teams would ignore or miscalculate risks inherent to derivatives. The assortment of risk management methods and individual company’s investigations initiated because of pricing disparities among competitors will illuminate many risk pricing issues that went unnoticed in the past. Notwithstanding past widespread risk pricing blunders, it is likely that risk management performed by 1000 separate competitive entities will lower overall systemic risk.

Rocky Humbert begs to differ:

I respectfully disagree with your premise for these reasons:

Thousands of supposedly independent and uncorrelated investors BOUGHT the housing related derivatives over the past several years; and all their models presupposed the same generalized assumption - that housing prices wouldn't decline nationally. This demonstrates beyond any doubt that a systemic event is systemic exactly because of a widely held belief. Any time large groups of people share the same belief, it becomes a systemic risk. How can you ignore the exception that proves the rule?

The wave of bank failures in the early 1930's was spread across the country in small and medium sized banks too. Similarly, the RTC which spent billions cleaning up the S&L failures of the late 1980's. Who was the TBTF institution in the 1930's ?

The equity quant debacle in the summer of 2007 demonstrated that most of these independent minds were using similar models. This was not a systemic risk event — but it demonstrates the illusion of independence among participants…both hedge funds and broker-dealers.

There are many many illustrations of similar phenomenon in the natural world — feedback loops, harmonic amplifiers, etc etc.

George Parkanyi adds:

Banks should be allowed to grow as big as they want, but not allowed to be counter-parties to each other where their own capital is involved. And certainly no borrowing from each other or insuring each other. (They are supposed to be competitors after all). That way, they would be transacting with each other only on clients` behalf (e.g. letters of credit, wire transfers, cheque clearing etc.). They should be able to take deposits, borrow from the Treasury with the transparency associated with that, and they should be allowed to trade their own capital on 3rd-party exchanges (again, not directly with each other). And of course they could form syndicates to spread risk when financing 3rd parties. This way their business, and sphere of exposure would be limited to the business they do with their clients and their proprietary trading through exchanges that have well-established margin rules and centralized, neutral clearing mechanisms. There would be relatively little linkage to allow a chain reaction to occur. (There would be some through lending to the same clients. One guy calling in a loan at a bad time could cause problems for the others. But this would just have to be risk-managed - you can`t take the risk out of everything).

Something like this would compartmentalize risk without having to treat institutions differently simply because of their size. Thoughts?


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search