At this point, Nassim N. Taleb has explained many times (1) what caused the financial crisis and (2) what will cause the next one. This is actually pretty simple stuff and as he's sick of repeating it, we'll summarize:
(1) the Nobel Prize Committee, and
(2) knowing things.
We learned the second part when Taleb dropped by the U.S. Congress last week to warn them away from the dangerous trap of setting up an Office of Financial Research to collect statistical data from banks and try to analyze it to reduce risk. Taleb's prepared testimony included some polite objections to this plan:
Allow me to present my conclusions upfront and in no uncertain terms: this measure, if I read it well, aims at the creation of an omniscient Soviet-style central risk manager. It makes us fall into the naive illusion of risk management that got us here —the same illusion has led in the past to the blind accumulation of Black Swan risks.
It's what Taleb has been saying for years - backwards-looking data can provide no guidance about future risks, any indication that you can beat the market or manage risk is just a temporary anomaly, and the only way to protect yourself is to buy out-of-the-money puts and/or Taleb's books. Brad DeLong summarizes:
His argument is that you should actively seek to be completely ignorant of everything, for if you learn something then you will be overconfident about how much you know and will run unacceptable risks as a result. Better to just shut your eyes and act completely at random.
It sounds weird, we know. But Taleb can prove he's right:
Numerous experiments provide evidence that professionals are significantly influenced by numbers that they know to be irrelevant to their decision, like writing down the last 4 digits of one's social security number before making a numerical estimate of potential market moves. German judges rolling dice before sentencing showed an increase of 50% in the length of the sentence when the dice show a high number, without being conscious of it.
So if you go around measuring VaR or bank capital or the amount of subprime paper outstanding, even if you know that they're bullshit, your subconscious will fool you and you'll end up blurting out stupid things like "mortgage securitization caused the crisis" when it turns out the right answer was "Black Swan AAAAAAHHHHH!!"