Yet for all the risks they're taking on, banks insist they're safer than ever. They've hired many of the greatest mathematical minds in the world to create impossibly complex risk models... "Right now everything on my screen is flashing red," said Michael Alix, chief risk officer at Bear, Stearns & Co., on May 11, the day after Federal Reserve Chairman Ben S. Bernanke raised interest rates, sending the market gauges he was looking at tumbling. But "that doesn't make me nervous," says Alix. The bank has built such powerful computing systems that Alix can reevaluate every day the risks of thousands of positions across the firm's trading businesses under various stressful scenarios to be sure the firm doesn't hold too much of any risky investment at any one time. That type of analysis used to take a week to complete. "The machine works," he says.This quote is cited in a great post by curious capitalist Justin Fox. People are rightly skeptical of risk management models, but I think their criticisms miss the mark. From what I can tell the models seemed to work just fine, but the problem was human error caused when people like Alix and his superiors dismissed the risk. This could be legitimate, since there is a tradeoff between risk and return, but as we've seen the potential spillover effects were never included in their analysis. Or they were simply ignored.
We had a fascinating conversation about this in class last night and Phil made the excellent point that a firm can only lose up to the value of the firm, even though they were leveraged to many times that value. Taxpayers were left holding the bag in order to limit systemic risk. I don't know what the answer is to this problem, but I do think we should be focusing on the incentive problems and not simply blaming technology.
Posts
No comments:
Post a Comment