Michael Lewis has a great piece on sub prime. This leaps out at me.
In retrospect, pretty much all of the riskiest subprime-backed bonds were worth betting against; they would all one day be worth zero. But at the time Eisman began to do it, in the fall of 2006, that wasn’t clear. He and his team set out to find the smelliest pile of loans they could so that they could make side bets against them with Goldman Sachs or Deutsche Bank. What they were doing, oddly enough, was the analysis of subprime lending that should have been done before the loans were made: Which poor Americans were likely to jump which way with their finances? How much did home prices need to fall for these loans to blow up? (It turned out they didn’t have to fall; they merely needed to stay flat.)
If someone, or many people, screw up, then in a market there will be someone who uncovers that error. And in doing so, make a very large amount of money. Providing that incentive to uncover the screw up.