Why are U.S. Executive Salaries So High?Published: February 25, 2011 in Knowledge@Emory
At the moment, one of the few indicators in which the U.S. economy still leads is the amount its top executives earn. To Shivaram Rajgopal, a professor of accounting at Emory University’s Goizueta Business School who studies executive pay, this suggests that something is wrong with the compensation-setting system. In particular, he sees two problems: Wall Street’s preference for short-term profit over long-term investment, and cozy corporate boards.
In an interview with Knowledge@Emory, Rajgopal details his research and the problems inherent in America's executive compensation policy.
Knowledge@Emory: Are U.S. executive salaries really excessive?
Rajgopal: As a scientist, I have to say it’s very difficult to show that CEOs are systematically overpaid. Having said that, as a citizen, I don't believe that the executive labor market is efficient because I've seen enough cases where there is abuse. Look at Larry Ellison [CEO of Oracle]. He earns so much money, does he need another billion-dollar option grant to keep him motivated? I find it hard to believe.
Knowledge@Emory: What’s created this dynamic?
Rajgopal: If you're worried, as many CFOs and CEOs are, of being kicked out, you’ll ask for more money upfront. In Japan or Korea, the implicit understanding is, look, even if you screw up for awhile, it's okay, because products take a long time to materialize, and maybe the long-lead projects are the ones where we will make more money in the future.
Knowledge@Emory: Does the insecurity of some corporate executives matter?
Rajgopal: It does seem to lead to riskier behavior. Most executives expect their time with firms to be much shorter than the life of the contracts they signed. If things go bad, it's somebody else's problem.
Knowledge@Emory: Does this kind of short-term thinking have other consequences?
Rajgopal: You can frame the whole financial crisis in terms of corporate short-termism, from top to bottom. CEOs typically have tenures of four or five years. If you ask them how long they expect to stay, they would likely say less than six years. I saw a CFO.com survey which showed that a CFO's expected tenure is four and a half years, which boggles my mind. How can you do anything in four years? And if you look at their compensation contracts, many of their options vest in four years, and they have a big severance package if things go badly, so whatever happens, they are fine.
Knowledge@Emory: Do investors take a longer view than employees?
Rajgopal: Not at all. Most investors now think quarter to quarter.
This has a corrosive effect on some companies. I’ve asked CEOs and CFOs, “why do you really care about short-term stock price changes?” They say, “oh, we get so much flack. In a conference call, analysts are going to beat us up if the price has fallen 5 or 10 percent if you miss a forecast. There's so much negative press, eventually it snowballs and people start calling for our head.”
I have written a couple of papers that found that the risk to bonuses, option grants, and even the probability of being kicked out goes up if you miss quarterly earnings, which is kind of strange because quarterly earnings shouldn't be that important in valuing the company. An IBM or a Microsoft has been around for decades, so why is one quarter important? But it has become important, so it creates incentives to do funny things like earnings management or even taking on projects that look good from an accounting standpoint but don't necessarily add value.
Knowledge@Emory: Does short-termism lead to bad decisions?
Rajgopal: There is some evidence it does. We have a paper where we asked 400 CFOs what they would do if they had to choose between meeting or beating an early forecast or investing in a positive Net Present Value (NPV) project. More than half the CFOs we surveyed said that they would not take the project. This shows you that silly things like quarterly earnings matter a lot and people are giving up positive NPV projects to be able to meet that number.
Knowledge@Emory: Would vesting options and other pay at later dates help build in a longer-term attitude?
Rajgopal: It should. Options are supposed to be long-term compensation. But if the average stock grant vests in four years, how is that long term? There were some proposals that some incentive compensation should be put away in a bank so that if things went wrong you could actually take that money back, but these haven’t caught on.
Knowledge@Emory: But aren’t there some companies that get this right?
Rajgopal: There are pockets in the economy where you do find that excess compensation is not a problem. At Berkshire Hathaway, Warren Buffett's firm, for example, you don't find excess compensation, you don't find earnings management, but you do find his investees enjoy substantially higher rates of return, long periods for which sales growth outperform their peers, and lower volatility.
Knowledge@Emory: Why can’t everyone be like Berkshire Hathaway?
Rajgopal: One reason is that many boards simply don’t know enough about their business as Berkshire Hathaway does. At the time of the crash, a lot of the big investment bank directors were socialites. It is hard enough to figure out what they are doing even if you have a doctorate in finance. Forget about it if you're the head of the Museum of Modern Art or something. There's no chance that you’ll be able to figure anything out.
The bigger reason, however, is that even if Buffett has a couple of bad years or bad quarters, investors will forgive him and not sell their stock. But not everybody has such a reputation. In particular, hedge funds often won't explain what they do. When that’s the case, if I’m an investor, the only thing I can go by is their performance and if they don't deliver on performance, I’ll get nervous and I take my money somewhere else, which in turn creates incentives for the hedge funds to worry about every quarter.
Knowledge@Emory: Couldn’t a company just copy Buffett’s compensation plans, at least?
Rajgopal: Borrowing another company’s compensation plan in totality is not a good idea because context is important. What makes sense for one company and one industry won’t necessarily make sense for another.
The missing ingredient here is monitoring. We need better governance. But there are huge disincentives for managers to make that happen. Why would they want to make their own life harder?