The ratio of the pay of corporate chief executive officers (CEO) to ordinary workers has exploded in the last 40 years. It was a bit over 20 to 1 at the start of the 1970s, now it is well over 200 to 1, and in good years for CEOs, it can be more than 300 to 1. Steven Clifford’s book, The CEO Pay Machine (Blue Rider Press) is an effort to explain how this happened and what we can do about it.
Clifford speaks from firsthand experience, having spent 13 years as a CEO of major corporations and having subsequently sat on more than a dozen corporate boards. Clifford makes it clear at the onset that he deplores the run-up in CEO pay, but he is trying to explain why it happened.
Clifford’s basic story is that the process that determines pay has become hopelessly corrupted. At the most basic level, the corporate boards that are supposed to represent shareholders and put a check on CEO pay have little interest in doing so. Clifford describes a process of whereby boards are captured by CEOs and other top management. Being a board member is a cushy job, typically paying well over $100,000 a year for around 150 hours of work a year, by Clifford’s calculation. With many boards paying $300,000 or $400,000 a year, the pay can be in the range of $3,000 an hour.
And, as Clifford notes, it is almost impossible to get fired from a board by shareholders. More than 99 percent of the directors who are nominated by the board for reappointment win their election. Furthermore, the boards are typically used to working with the CEOs. The CEOs and their staff are the ones who provide them with information. Often the CEO himself is a board member, usually the chair.
In this context, board members have little reason or incentive to ever challenge CEO pay. After all, the CEO is their friend, why would anyone object to giving their friend more money at the expense of a diverse group of shareholders, the vast majority of whom the director does not know. Furthermore, what difference would a few extra dollars make to the typical shareholder, if this is what it takes to add a few million to the CEOs pay?
Clifford goes through all the rationales for CEOs getting paid salaries in the tens or even hundreds of millions of dollars. Part of the story is compensation consultants, who always seem to want to raise pay, in part because they are often doing other business for the CEO. Part of the story is the belief that their CEO is always above average and the company would suffer if they lost their CEO to a competitor.
Clifford dismisses the idea that CEOs are worth anything comparable to their current pay. He argues that the inequality created by bloated CEO pay is destructive both to the companies themselves and to democracy. I would add to Clifford’s list of complaints that excessive CEO pay contributes to a distorted pay structure throughout the economy. After all, if a moderately competent CEO of a mid-sized company can pocket $5 million a year, then certainly the president of a major university or non-profit foundation is worth at least $1 or $2 million. And more pay for these people means less for everyone else. That is how arithmetic works.
Clifford’s solution includes some reforms of corporate boards, but most importantly a 100 percent luxury tax (taken from team salary caps in professional sports) on CEO pay in excess of $6 million. In effect, this would mean that the company must pay a dollar to the government for every dollar it pays CEOs in excess of this $6 million ceiling.
While most people would likely agree that CEOs are seriously overpaid, it is worth backing up a bit in this story in order to assess Clifford’s proposed solution. At the most basic level there is the question of whether CEOs are worth their pay in the sense of producing returns to shareholders that warrant their seven, eight, and even nine figure salaries. We’ll ignore for purposes of this discussion that companies may have other goals than returns to shareholders. The question is simply whether CEOs produce returns to shareholders.
Clifford cites several major studies indicating that pay is not closely correlated with returns. He notes that pay is often associated with a large element of luck, such as the CEO of an oil company getting a huge payout because the share price of the company surged along with world oil prices. Clifford also points out that CEOs are generally not especially mobile. They have firm-specific skills, so even a hugely talented CEO may not have a good second option if her current company won’t agree to a big raise.
This point is actually quite important and does not get nearly enough attention in the debate. If even an unusually good CEO has nowhere else to go then the company does not have to pay an exorbitant salary to keep her. A CEO who isn’t getting paid an amount equal to their value to the company could decide to simply quit and drop out the CEO business, but this works the other way as well. A CEO who has accumulated hundreds of millions of dollars may opt to retire at an early age, since they would have little need of more money.
Anyhow, if we accept that CEOs are not worth their pay to shareholders, then the implication is that shareholders are being ripped off. If a CEO is getting paid $20 million or perhaps even $200 million, in a context where a $5 million CEO could have provided the same return, then this is the same as if a contractor overcharged the company by that amount. The shareholders should be upset and want to do something about it.
Clifford argues cogently that the corporate governance structure largely blocks the ability of shareholders to challenge CEO pay, but this raises an interesting issue with his proposed luxury tax, the big bazooka in his anti-CEO pay arsenal. If the corporate governance structure is so corrupt as to hand out tens of millions of dollars of shareholders’ money in excess pay to CEOs, is there any reason to believe that we will stop the rip-off with Clifford’s luxury tax? There are good reasons for skepticism.
First, the tax will of course require a law passed by Congress. It is very easy to put clauses in a CEO luxury tax law that will make it worthless for all practical purposes.
If that sounds like an outlandish claim, we need look no further than Clifford’s book to find a precedent. In 1993, President Clinton followed through on a campaign pledge to end the tax deductibility of any CEO compensation in excess of $1 million a year. This passed Congress as an amendment to a huge budget bill with the small provision that pay related to performance was exempted.
This exemption was big enough to drive a truck through. In subsequent years the vast majority of CEO pay took the form of stock options or grants of restricted stock, both of which were deemed as being related to performance and therefore exempted from the cap. If anything, the change in the law accelerated the rise in CEO pay, since in the 1990s the value of options did not even have to be included as an expense on corporate balance sheets. Companies were effectively able to hand millions to their top executives at no cost on their books. (Accounting rules were changed in the last decade so that the value of options is now listed as an expense.)
If we could somehow muster the political power to impose a luxury tax along the lines recommended by Clifford, is there any reason to believe that we will more successful in making it a real tax than in 1993? Certainly the folks who installed the performance pay loophole understood what they were doing. But very few of the people concerned about excessive CEO pay realized that this exercise was a complete waste of time. Do we need Round II of this charade?
But let’s suppose that we get a serious luxury tax, without the loopholes, but do nothing about the corruption of corporate governance. Won’t directors still want to help their friends in the CEO suite? There are many ways they could evade this cap, most obviously by providing a range of luxury services (think of yachts, corporate jets, vacation homes, etc.) which could go far towards making up for the pay cut.
Providing these items for personal use may be a violation of the law, but if the people on the inside want to do it, how successful will the I.R.S. or other enforcement agencies be in cracking down? It will all officially be listed as being for business purposes, so it would take some serious sleuthing to determine that this was not the case.
If we don’t address the governance problem, it’s not clear that the luxury tax would lead to much of a clamp down on pay, even if we got it without loopholes. After all, if a board is willing to pay $20 million to a CEO who is worth $3 million, maybe they would pay $34 million for the CEO as well (the pay, plus a $14 million tax). To the directors, this is still just other people’s money.
If the basic problem is a corrupt corporate governance structure, then the solution must involve reforming the governance structure. This means changing the incentives for directors and restructuring the voting process so that insiders do not tightly control it.
In the former category, directors need some real incentive to crackdown on CEO pay. One mechanism that I have proposed elsewhere is putting some teeth in the now advisory Say on Pay votes. Suppose that directors lost their annual stipend if a Say on Pay vote when down to defeat. This would create some real downside risk to overpaying a CEO.
Ideally this could be an amendment to the law requiring Say on Pay (amendment provision in Dodd-Frank), but this is also the sort of measure that activists could pressure companies to put in place voluntarily in their corporate charters. After all, less than 3 percent of compensation packages are defeated. How many companies would like to say that they are in the bottom 3 percent of corporate governance in terms of their ability to rein in CEO pay?
Also, as a political matter, giving more voice to shareholders in setting CEO pay seems a considerably stronger position than arguing for a stiff tax on compensation the government has determined to be excessive. The former is arguably making the market work better. Why would anyone be opposed to giving shareholders more control over what their companies do? The latter is clearly the government interfering with a contract between a company and its CEO. There are other routes for giving the directors a serious downside risk for excessive CEO pay, but this seems an essential governance reform in order to ensure that directors have the right incentive to put downward pressure on CEO pay.
The other issue is changing who gets to vote in shareholder elections. As it stands now, the fund managers that act as trustees for shares held in mutual funds often cast the majority of votes in shareholder elections. These managers have no direct interest in holding down the pay of CEOs. In many cases they have ongoing business relationships with CEOs who they count on to keep them informed of developments with the company.
It seems reasonable to strip these managers of their votes unless the actual owners of the shares explicitly instruct them in casting their votes. Here again, the direction of the change is to give the ostensible owners of the company more control over the company. Can a believer in free market capitalism oppose this?
In short, Clifford’s book is an entertaining insider’s take on the problem of excessive CEO pay and the corruption of corporate governance. But when it comes to solutions, it falls somewhat short.