Maybe the first thing to say about Jamie Dimon's pay raise, from $11.5 million in 2012 to $20 million in 2013, is "accrual accounting." Oh sure, JPMorgan paid "approximately $20 billion in penalties" since his last annual performance review. But most of those fines were incurred in earlier years -- naughty mortgage selling (2005-2007), mainly, with a side salad of London Whale (2012ish) and a bowl of Madoff enabling (1990s-2008) for dessert -- so it's no fair to dock Jamie Dimon's 2013 pay because of them. So far, 2013 seems to have been a year relatively free of fines.
Of course, Dimon got paid $27 million in 2006 and $30 million in 2007, when the biggest fines were incurred. So you can't push that theory too far. It would imply that the only time Dimon can be penalized for legal scandals is if they occur and are resolved within a single calendar year. (Or if the board clawed back compensation from previous years, which, hahahaha. ) On the other hand, he got no bonus and a paltry $1 million total compensation in 2008, when he bought Bear Stearns and Washington Mutual, which brought with them their own exciting collection of (eventual) fines. And last year his pay was cut in half to $11.5 million for the Whale thing. Just as JPMorgan was mostly reserved for its 2013 fines, so, in his way, was Dimon.
There are two big questions in deciding how much to pay Jamie Dimon after a big year of fines. The first is, even if Jamie Dimon is the bestest he can be at making money -- even if you're sure that earnings under Jamie Dimon, net of fines, are billions of dollars more than they'd be under anyone else -- are the fines themselves disqualifying? Is paying a dollar, or $20 billion, in fines to settle government allegations of wrongdoing qualitatively worse than losing a dollar, or $20 billion, on bad loans?
That's such a hard question that some JPMorgan directors were driven to actual physical activity in their efforts to answer it:
The debate pitted a vocal minority of directors who wanted to keep his compensation largely flat, citing the approximately $20 billion in penalties JPMorgan has paid in the last year to federal authorities, against directors who argued that Mr. Dimon should be rewarded for his stewardship of the bank during such a difficult period. During the meetings, some board members left the conference room to pace up and down the 50th-floor corridor.
Obviously, if you believe that "JPMorgan has been unfairly persecuted and that Dimon, as its figurehead, has been unreasonably vilified," then that would be an argument against taking the fines as disqualifying. If you think overseeing massive sort-of-admitted violations of law is a minor personal peccadillo, akin to smoking cigarettes or wearing a bad toupée, then, again, the fines wouldn't bother you much as fines. But Alex Pareene would disagree:
I think that any time you’re looking at the greatest fine in the history of Wall Street regulation, it’s really worth asking should this guy stay in his job. In any other industry — I can’t think of another industry. If you managed a restaurant, and it got the biggest health department fine in the history of restaurants, no one would say “Yeah, but the restaurant’s making a lot of money. There’s only a little bit of poison in the food.”
My guess is that the board thinks of the fines as a negotiated outcome to a business problem like any other: You'd prefer to pay less rather than more, obviously, but your goal is to maximize the expected value of the franchise, not to minimize fines on their own. It sounds terrible to say, well, massive fraud settlements are a cost of doing business as a global bank, but at least some nonzero probability of massive fraud settlements really is a cost of doing business as a global bank. Look at all the mortgage settlements, or all the Libor nitwits. There's a lot of people, and some of them will mess up, and their sheer lack of coordination can itself incur liability. And sometimes you will make the choice to do things that wander into legal gray areas, or that increase the odds of some of your people doing illegal things ("Let's hire a bunch of people to make subprime mortgage loans and pay them for production, what could possibly go wrong?"), because that's where the money is.
If you want to just avoid fines, a good way to do it is to not run a global bank. Nobody is forcing you to run a global bank! You can stay home and not sell anyone any kind of mortgages. But once you decide to get into the business of hiring thousands of people to do risky stuff not under your constant personal supervision, the risk of fines becomes nonzero. Managing that risk is an important function, but it is not the only or most important function.
Presumably the most important function is the obvious one, "make money." And Jamie Dimon is ... I guess the hypothesis is that he's good at that? Even after the fines? I dunno, the question of Jamie Dimon's value-add is pretty profound. JPMorgan makes a lot of money, after fines, though you could quibble over measures like return on equity. But JPMorgan is a big global bank that employs lots of people, many of whom are pretty good at making money, and some nonzero percentage of whom are very good at losing money. Certainly Jamie Dimon plays a significant role -- hiring more people who are good at making money, firing the people whose skills run toward losing it, acquiring whole companies whose money-making records were let's say mixed, setting strategic priorities for where the most money is to be made, and bopping around motivating everyone to make more money -- but the dollar value of what he does in that role is less certain.
CEO value in general is hard to measure, and you can't be sure that Dimon is adding more value in that role than, I don't know, Michael Cavanagh would be, or me, or Alex Pareene. There is a substantial margin of error. His most-likely value over replacement banker might be, I don't know, five billion dollars, but with a standard deviation of three billion dollars. There's a non-trivial chance that his value is in the negative billions.
We have industries in America where an individual worker's added value can be confidently quantified. Those industries tend to be those that generate lots of statistics from repeatable experiments. Finance ... I don't know, is sort of not unlike that? Some consultants have called for banks to "better identify value created by incremental profits generated, rather than the ‘value of the seat’ of the underlying franchise, with pay-out ratios recalibrated accordingly," and you could imagine how that would work for the average trader or relationship banker. The idea does not seem to have gotten much traction among those traders and bankers.
It's harder for CEOs, of course, since they tend not to generate a lot of individual statistics. And as is so often the case in finance, persuasion trumps objective measurement, and the lack of clear indicators of fundamental value leads you to fall back on market comps. For all the differences in banks' performance, and all the metrics referred to in their comp discussions, these days big bank CEOs get pretty similar paychecks. The going rate for the top firms seems to be $10 to $30 million, a pretty narrow range considering that it applies to CEOs with a pretty wide (multiple billions of dollars wide) range of performance and, presumably, an even wider range of added value. Dimon got paid toward the bottom end of that range last year and so I guess he was due for a raise to something more in the middle. Fines or no fines, the reasoning might be as simple as that.
That's mostly a joke. We won't know for sure what bad stuff JPMorgan was getting up to in 2013 until, like, 15 minutes before the statute of limitations expires. Let's see how this China hiring stuff plays out, for instance, though I guess that was mostly pre-2013? Everyone meet back here in December 2018 and we'll talk about what we've learned.
Those numbers are from the narrative compensation discussions, not the summary compensation table, which gets different numbers based on various arbitrary means of measuring. I just used JPMorgan's arbitrary means of measuring, which is how the board talks about it, so seems the most relevant.
You could have a model where a dollar of fines is "worth" $2 or whatever: Like, a CEO who makes $10 billion of pre-fine profits and pays $3 billion of fines gets "credit" for $4 billion (vs. $7 billion of GAAP earnings), while one who makes $7 billion of pre-fraud earnings and pays $1 billion of fines gets $5 billion of credit (vs. $6 billion of earnings). I suppose that's a plausible mental model for saying, "we want you to make money, but fines are bad. Very bad. But not, you know, that bad." But the outlines of the optimization problem remain the same.
Bloomberg (RV) tells me that JPMorgan's ROE was 8.36 percent for 2013, versus 13.99 percent for Wells Fargo, 11.98 percent for Goldman, 11.02 percent for Bank of America, 7.14 percent for Citi, and 5.04 percent for Morgan Stanley. Not bad, not best-in-class. The fines didn't help.
Here is a roundup, and a paper, focused on compensation. These do not strike me as quite addressing the "value over replacement banker" question. Here is an entirely plausible price-based argument that Steve Ballmer's value to Microsoft, at the end of his reign, was negative $18 billion. JPMorgan outperformed the S&P by almost 2 percent on the day after Jamie Dimon got to keep his chairmanship; he had made vague threats of quitting as CEO if he'd lost that vote. That suggests a minimum Jamie Dimon market value of $4 billion (2 percent of a $200+ billion market cap), and probably much more depending how you adjust for the (pretty low, right?) ex ante probability that (1) he'd lose the vote and (2) he'd then quit.
I have criticized that chart/ranking in the past. Bank pay and performance are weird, is the takeaway. Also that chart found that Jamie Dimon's pay was just right, last year, so I guess that says something about whether he should get a raise?
To contact the author on this story:
Matt Levine at email@example.com
To contact the editor on this story:
Tobin Harshaw at firstname.lastname@example.org