Models Don't Kill Banks, Bankers Kill Banks

A post on quantitative risk analysis and its role in the current crisis at The Big Money:

Wild-eyed number crunchers, the theory goes, have unleashed monsters that they are unable to control. Buffett himself has branded derivatives as "financial weapons of mass destruction." George Soros—cited with equal frequency in the emerging genre of crisis-lit—has added that he has long avoided the use of derivatives on the grounds that "we don't really understand how they work."

Well, with all due respect to the billionaire financial geniuses of the world, I just don't buy it. Modern financial models are highly imperfect, to be sure, and we the modelers are insufferably arrogant. But models don't kill banks; bankers kill banks. We geeks may grunt a lot (we want the world to know how laborious our calculations are), but the truth is that our models aren't that hard to build, and they aren't that hard to take apart. When bankers and their advisers fail to question the premises of these models, it's usually because they find those premises quite congenial. The models merely provide an excuse to exercise a faculty for which human beings have always shown a special talent, namely, wishful thinking. What's worse, they also provide the financial community with a rhetorical device for persuading government officials that banks are, by virtue of their purported technical expertise, capable of achieving that audaciously oxymoronic state of being known as "self-regulation."