Regression to the Mean, JPMorgan Edition

I haven’t been writing about the JPMorgan debacle because, well, everyone else is writing about it. One theme that has stuck out for me, however, has been everyone’s reflexive surprise that this could happen at JPMorgan, supposedly the best and most competent of the big banks. For example, Lisa Pollock of Alphaville, who has provided some of the most detailed analyses of what happened, asked, “could this really happen under CEO Jamie Dimon’s watch?” Dawn Kopecki and Max Adelson at Bloomberg referred to “JPMorgan’s cultivated reputation for policing risk.” Articles about Ina Drew’s resignation are sure to point out her relative success at dealing with the financial crisis of 2007–2009.

“Highly intelligent women tend to marry men who are less intelligent than they are.” Why? Is it that intelligent men don’t want to compete with intelligent women?

No. It’s mainly because if you take two draws from a random distribution, and the first is at the high end, the second is almost certain to be lower, even if the two are somewhat correlated. This example comes straight from Thinking, Fast and Slow by Daniel Kahneman, which I’m finally reading (chapter 17).

The performance of anyone doing anything will exhibit regression to the mean. If you do well at something, it’s because of some combination of skill and luck. If JPMorgan came through the financial crisis well, it was some combination of skill and luck. Remember, JPMorgan didn’t have as big a portfolio of toxic assets as its competitors because it was late to the party; only in retrospect do we ascribe this good fortune to the supposed skill of Jamie Dimon. JPMorgan was never as good as people (both supporters and critics) made it out to be, so we shouldn’t be so surprised that it just lost $2 billion (and counting).

The more disturbing thing isn’t that commentators fell for this statistical red herring. It’s that people inside JPMorgan seem to have fallen for it, too. This was Dimon’s response to a question about whether the Chief Investment Office was becoming more aggressive, as reported by Bloomberg:

“I wouldn’t call it ‘more aggressive,’ I would call it ‘better,’” Dimon told analysts yesterday. “We added different types of people, talented people and stuff like that.” Until recently, they were careful and successful, he said.

People don’t suddenly go from being good to bad overnight. What happens is they go from lucky to unlucky. They are the same people doing the same things.

“Inside JPMorgan, leadership is stunned by the situation, according to two senior executives,” also as reported by Bloomberg. If that’s true, that’s bad news for all of us. It’s one thing if, as many of us thought, JPMorgan was consciously trying to take on more risk (as has been amply documented, Dimon pushed the Chief Investment Office into profit-seeking trades) while denying it to regulators and the press. That’s what we expect.

It’s another thing if the bank didn’t realize it was taking on risks of this magnitude. That implies that JPMorgan executives had started believing their own hype—that is, they believed that they really were just good, not lucky. And that should make all of us very worried.