All that said, what I think people really need to do is confront the fact that obscene banker compensation is really a social justice issue rather than a financial regulation issue. Which is to say the sky-high pay seems wrong just as such rather than because there’s a specific bad incentives issue that needs to be corrected.
But Ygelsias's comments miss a key point about the both the study and the structure of investment banks--the problem is not with the CEOs* (which the study measures). Rather, it's the traders swapping complex derivatives, bankers packaging shady mortgages, and salesman foisting these crappy products on customers we have to worry about. These people very clearly have incentives to think about the short-term gains rather than the long-term losses. The pay structure encourages bank employees to gamble with shareholders' money in a way that gives the bankers huge rewards if they win, while sticking them with little of the downside.
True, there probably aren't many, if any, studies that could empirically prove this. But that's because it's impossible to get actually get data on this to measure. CEO pay has to be reported to the SEC, but the numbers for non-top management employees are closely guarded. Banks are reluctant to turn them over to federal law enforcement agents when asked. Do you think they're going to hand them over to academics for research that will mostly be used to criticize their practices?
*Most of the CEOs of these banks were bond traders or salesman in a time when finance was much less complicated. I highly doubt the really understood what was going in many of the more complex areas of the bank, which is why the overly relied on measures like Value at Risk.