Last week, the Obama administration’s “pay czar,” Kenneth Feinberg, announced that the government will impose caps on compensation for the 25 highest-paid executives at seven companies that received “exceptional assistance” through the Troubled Asset Relief Program — including American International Group (AIG), Bank of America, Citigroup, Chrysler, Chrysler Financial, General Motors and GMAC. Under the new regulations, salaries will be reduced by an average of 90%, and total compensation (including bonuses and stock options) will be lowered by 50%. Knowledge@Wharton spoke with Wharton accounting professor Wayne R. Guay and then with finance professor Alex Edmans about what these changes could mean for Wall Street, company shareholders and taxpayers.
The following is an edited transcript of the interviews.
Interview with Wayne Guay:
Knowledge@Wharton: In the overall context of reforming regulations after the financial crisis, how important is the executive pay issue?
Wayne R. Guay: We need to think about how executive pay is related to the financial crisis. We want to keep our eyes on the ball … with respect to the risk issues…. Much of the effort that we have heard about in the last week or so has been related to the level of pay that executives might be taking in. I don’t think that was much of a factor in the financial crisis.
The large financial institutions didn’t run into trouble because they were overpaying their executives. The amount of compensation these individuals were taking home is not what caused this problem. The problem was these institutions had a lot of similar risk structures that got hit with some of the same underlying fundamentals in the economy, which caused all of them to experience substantial losses at a similar point in time.
Then the question is: How does executive compensation factor into risk-taking with respect to these financial institutions? There, I would say we want to focus on two angles. One is how executive compensation affects the risk-taking of a single bank or financial institution. And then, how does executive compensation factor into the coordinated or systemic risk that these financial institutions might all be taking at the same point in time. It might be possible to do something about the first one with respect to how executive compensation affects the risk-taking of an individual bank. It is not clear to me how you would structure an executive compensation plan so that an executive of Bank A would consider the risks that Bank B, Bank C, Bank D and Bank E are taking, [even though] those are all coordinated and correlated. The solution to that problem is probably outside of executive compensation.
Knowledge@Wharton: Even if an executive compensation package is tied to short-term gains, do you think it makes a difference in the equation?
Guay: We want to step back. There has been a lot of rhetoric with respect to these executives having a very short-term focus. Some ways that the regulators [plan] to focus more on the longer term is to have short-term bonuses or salaries paid in stock and require executives to hold the stock for two, three, four or five years before they can sell it and have these sort of longer-term performance objectives. We want to … recognize that for most of these top executives, the vast majority of their compensation already is tied to long-term stock price performance. Starting in the early to mid 1990s and through to recent years, the majority of a CEO’s pay — not only at a financial institution but at non-financial institutions in corporate America in general — comes from grants of stock and options.
Those securities typically have a three-to-five year vesting period, and an executive needs to hold them for a substantial period of time. We are focusing on the salary that might be now paid in stock or bonuses. We want to recognize that it is a small component of the typical CEO’s pay anyway. If CEOs are already getting 80% of their pay in stock or options and we make it 85% or 88%, is that really going to make a big difference? My feeling is probably not.
Knowledge@Wharton: Of the groups that have skin in this issue, how are they being treated in terms of fairness and equity? You have the executives. You have taxpayers who supported the rescue of these companies. You have company shareholders, employees and other stakeholders. Are executives being given a fair shake in this process?
Guay: I certainly feel [without] regulatory pressures or actions, executives can look out for themselves. I am not worried so much about the executives, except to the extent that regulators or someone else comes in and forces the issue. Executives can look out for themselves. But if a regulator comes in and says, “You are only allowed to make half what you are currently making,” then you may need to step back and say, “Is the executive getting a fair shake here?” We have heard a lot of discussions about the labor market for executives and this has all been through the financial press.
If you constrain the pay that bank executives are receiving, they can go elsewhere and make money. They can go to hedge funds. They can go to private equity funds. They can go abroad. They can go to private organizations. We have heard all those arguments. But the focus has been on the level of pay, which I don’t think has anything really to do with the financial crisis we are experiencing.
If we now move to the shareholders, in many ways [they] can also look after themselves. At the same time, there is a role for the regulators to coordinate efforts [concerning] corporate governance. You could imagine a setting where 10,000 publicly traded firms in the U.S. all would like to make some basic changes or important changes to their corporate governance. They can each try to make the changes by themselves or they can ask a regulator to step in and coordinate that effort on behalf of all corporations. There is an efficiency argument that comes from regulators stepping in and having one body making changes.
Some of the changes … [include] giving shareholders a bit more access to proxy materials so that when you have directors who should be removed, it [will be made] easier for shareholders to do so. And just in the last few days, there have been some new proposals on ways to do that. It is important for regulators to look out for the shareholders, not so much because they can’t look out for themselves in many cases, but because they can coordinate the corporate governance shifts that most corporations might be looking for at various points in time.
Knowledge@Wharton: And the taxpayers?
Guay: Regulators do need to look out for the taxpayers. A lot of the press in the last week or so has been [about] Mr. Feinberg’s efforts to regulate or alter compensation at seven of the large institutions — financial and non-financial — that took a particularly large amount of support from the government [and] taxpayers. Now taxpayers have a vested interest in those firms, even [more so] than they would normally have. They actually own part of the organizations, and so looking out for their interests is important.
…Regulators would want to think about taxpayers’ interests and stockholders’ interests in a similar way. But if taxpayers have debt or preferred stock in the company, the interest of debt holders might be different than the interest of stockholders. Regulators would want to make sure that the debt securities the taxpayers hold are safe and are going to be paid back. What else we are seeing here is that by squeezing executive compensation a little bit at these organizations, [the companies] will put more effort into paying back taxpayers and getting out from under the thumb of the regulator.
One way for institutions [to not have] their executive pay regulated to such an extent would be to pay back the assistance they received from taxpayers and then have their executive compensation be similar to other organizations.
Knowledge@Wharton: Do you think that excessive risk-taking and executive compensation are closely tied?
Guay: Risk-taking incentives are kind of tricky. [Executives] have an incentive to take a risk when they are protected on the downside and have unlimited or substantial gains from the upside. Stock doesn’t really have that feature. Tying an executive’s compensation to stock doesn’t really provide a lot of risk-taking incentives. There aren’t a lot of substantial risk-taking incentives in these compensation packages now, at least for top executives.
… Some of the issues of compensation revolve around traders in banks and financial institutions, some of which have tremendous upside and fairly limited downside. They have possibly been taking some risks that they should not have been taking. But if you ensure that you have a good, solid board of directors, and the board of directors is looking after the CEO and top executive team and making sure that [their] incentive structures are aligned, then the executives should be able to look out for the traders and other people in the organization.
It is a chain of command you are worried about. Shareholders choose the directors to look after [their] interest. The directors then select an executive management team to look out for the shareholders’ interest. Then the executive management team looks after the other employees, [who should] look after the shareholders. If we think there is a problem with an executive, a trader or someone else’s compensation, the fault at the end of the day has to lie with the board of directors. If you think the board of directors is not a good board, come up with a way to fix the board as opposed to fixing all the bad actions the board is taking.
We have seen a lot of effort over the last 10 years to have more independent boards of directors. People sometimes forget that the compensation committee is made up completely of independent directors. The audit committee is made up of independent directors. Most boards have a vast majority of independent directors. We have made a tremendous amount of [progress in] structuring the board of directors to be independent of the CEO and to look after shareholders’ interests. If we don’t think we have gone far enough, maybe we should think about going a little further in terms of selecting a board. But once we have a board looking out for the shareholders, it is dangerous to second-guess what that board is doing. That’s what I think is going on now.
Knowledge@Wharton: But there have been a lot of suggestions that shareholders and boards have short memories and time horizons. If a financial institution is making huge profits right before heading to a waterfall, they are not highly incentivized to change the board — at least the shareholders aren’t.
Guay: There is a fallacy there. Again, most top executives get the majority of their pay in the form of stock and options. Look at the loss and wealth of the CEOs [and top executives] of — just pick at random — any large financial institution in the last year and a half. They lost billions and billions of dollars. They certainly would have been much better off had this crisis not happened. There is this perception out there that the top executives walked away with a tremendous amount of money and left shareholders holding the bag.
They did walk away with some bonuses and pay, but they were left holding the bag with the shareholders at that point in time. Shareholders as a group — all the shareholders holding the stock of those financial institutions — suffered tremendously. If everyone knew that something was amiss, why didn’t the shareholders holding the shares take action to make a change? Why didn’t the directors? Most of the directors now have a big black mark on their reputations as well — as do the CEOs. The incentives were very tightly aligned with maintaining the value of these institutions. Nobody benefited from having this financial crisis — well, I shouldn’t say nobody. But most people did not benefit from the financial crisis. It is easy to look back in the rearview mirror and say everybody should have seen this coming. But three years from now, my guess is we will come to realize that everybody made a similar mistake and we have all paid a price for it.
Interview with Alex Edmans:
Knowledge@Wharton: Alex, in the overall context of reforming regulations after the financial crisis, how important is the executive pay issue?
Alex Edmans: It is very important politically. A lot of people lost a lot of money in the financial crisis, and it is very natural to want some people to pay for it — specifically those people who were in charge of the companies that ended up going bust or losing a lot of money. Often as economists, we like to belittle the political reasons, but they are very important given that this has affected a huge number of people.
However, economically the importance of executive pay is a bit less clear. This depends on what you think caused the financial crisis. There are three main schools of thought.
One of them is that we just had bad models. Even though we had the most sophisticated and smartest people on Wall Street, the models just weren’t up-to-date for the current financial system. This is similar to [talking about] somebody who gives up the game-winning home run in a baseball game: The incentives are correct but you just make mistakes from time to time. If that is the case, changing pay is not going to affect this because it wasn’t about bad incentives to begin with.
The second school of thought is one known as the behavioral school, where people just made psychological mistakes because, as human beings, we interpret data in an incorrect manner. We tend to over-extrapolate from small trends and so forth. Much like the bad models idea, this is not caused by incentives. Again, under that school, executive pay won’t be so important.
The third school is that because of certain pay systems, executives took too many risks. They took options that were too short-term. But again, it is not so clear that the changes we have seen are tackling the component of executive pay that was the problem. What we have seen is limits to the level of pay.
Many people believe it is not so much the level of pay but the short-term nature of it or there’s an asymmetric risk-reward profile — that if you do well, you get a high bonus; if you don’t do well, you get a large severance package. Looking at just the level of pay may miss the major part of it, which is the incentive structure.
Knowledge@Wharton: There were some components of the proposals from the compensation czar that [would mean] we have a longer-term view of how compensation is structured. Are [the reformers] looking long-term enough?
Edmans: They are. This is one of the most important things that came out of the new proposals. In addition to the reduction in pay levels, which got a lot of press, I was particularly pleased to see Ken Feinberg’s quote that he is requiring [executives] to buy stock that they must hold for a number of years before they can cash out. This is getting more to the heart of the system. While many people will focus on pay levels and people getting paid too much in terms of raw numbers, the incentive structures seem to matter more. This is going to have more bite, even though the media typically focuses on the $1 million versus $500,000 or whatever the amount of salary is.
Knowledge@Wharton: To what extent did some of the popular reaction to this exist before the financial crisis? Executive compensation has been an issue for a long time and a lot of working people have questioned the very high level of salaries and stock options. Does that have any role in the political component right now?
Edmans: It is true that a lot of the debate pre-dated the financial crisis. A big debate among academics and practitioners existed before the start of this crisis. What we saw was the level of U.S. pay going up sixfold since the 1980s, and this was way [higher] than the increase in the average [U.S.] worker’s income and incomes overseas. The view was that executives were setting their own pay. We like to think there is a board of directors acting on behalf of shareholders, which sets pay to maximize shareholder value. But the alternative view is that the CEO has his or her buddies on the board, which enables them to [approve] pay schemes that are good for the CEO but not the shareholders.
However, again, because it is a debate, there were two sides to the story. A prominent view — mainly proposed by Xavier Gabaix and Augustin Landier of NYU Stern School of Business — argued that [the pay level] is justified by competitive forces. Take two baseball players — Babe Ruth and Derek Jeter. Babe Ruth [was] extremely talented and Derek Jeter is extremely talented. But Derek Jeter gets paid far more than Babe Ruth did even though they have the same level of talent. This is because the stakes in the baseball industry have become much higher….
Applying this to a corporate setting, we now have companies that are far bigger than they were 20 years ago. If you have a CEO who is only slightly more talented than the next best guy [and] so adds only 1% more to the firm’s value — and given that [revenue for the firm is], say, $10 billion — 1% of that is $100 million. So it is worth paying top dollar to get the best person.
This is not a widely palatable view because people are willing to accept that Derek Jeter is talented and they can see him hitting homeruns, but executives? Well, they are not really talented. They sit in their office, they play golf a few times a week and this is a job that anybody can do.
But look at some of the very good CEOs. They can create extremely good products, which can change the way people live their lives. Companies that have made transformational innovations, such as Google, are run by talented CEOs.
Knowledge@Wharton: In addition to executives, other groups have skin in the game. Those would be the shareholders and taxpayers, who are now part owners of some of these companies — of the seven that are affected by these rules. Are they getting a fair deal?
Edmans: I will lump shareholders and taxpayers together because, as you said, both can be considered investors in the banks. Again, [the situation] is not clear. If you just looked at pay going down, this has a direct immediate benefit to shareholders and taxpayers because a dollar that you are not paying to the CEO is a dollar that you can return to the government or shareholders.
But if you want to look at the broader implications of changing pay, one of the concerns is that if you reduce the amount of pay, you might not be recruiting the best managers to do the job. Or maybe you are not incentivizing them enough. And if you were to just look at basic numbers, we get pretty outraged if a CEO gets overpaid by $10 million; $10 million is a huge amount to any normal person, but in the grand scheme of things, $10 million is only 0.1% of a $10 billion firm. Whereas if executives are not given correct incentive structures, they [might not be] adding 1% or 2% to the firm’s value. This is several orders of magnitude higher.
One of the concerns that some shareholders and maybe taxpayers may have is that we are throwing the baby out with the bath water. On the one hand, we have got a direct saving in the amount of salary. But on the other, what we are losing in terms of potentially better decisions may vastly outweigh this.
Politically it is tricky. If you don’t give CEOs the correct incentives and they don’t invent a new product, we will never see [this] as outsiders. We can only see bad actions, such as being overpaid, and punish those [responsible for those actions]. But we can never see the [result] of not taking good actions, even though they may be much more important for a firm’s value. Because we never see them, they never get as much attention in this debate.
Knowledge@Wharton: In a paper you wrote this year with some of your colleagues, you suggested a system that escrows compensation for a set period of years stretching into an executive’s retirement. The restrictions proposed by the government on the seven companies don’t go nearly that far. Do you think longer timeframes for measuring executive performance will ever gain traction?
Edmans: Sure. This was a paper I co-authored with Xavier Gabaix and Tomasz Sadzik of NYU Stern and Yuliy Sannikov of Princeton University. You are right. One of the main hallmarks of this was long-run incentives, so that CEOs would not take short-term “mis-actions”. What you have seen are regulations focused more on the level of pay rather than the horizon. But there was that phrase I mentioned earlier about trying to increase the horizon of incentives. Some of this is indeed getting traction.
But more importantly, this doesn’t require any regulation to be implemented. If we are correct and this is sensible, companies will adopt it voluntarily. What we saw in the wake of the financial crisis is a number of investment banks reforming their compensation schemes. UBS has a new compensation model where it links bonuses to long-term performance. Morgan Stanley, Goldman Sachs and some of the other investment banks followed suit by trying to lengthen the horizon of incentives. This doesn’t require government intervention. If a pay package is good, shareholders have the incentives to adopt it themselves.
Knowledge@Wharton: What do you think about the argument put forward by many proponents of pay limits that executive compensation and excessive risk-taking are closely tied?
Edmans: I definitely agree that executive compensation and risk-taking are closely tied. Because if you have an asymmetric risk-reward structure … this will clearly encourage you to take risks. Also, if you are given short-term incentives where you are allowed to cash out early, you have incentives to take risky decisions and cash out before bad consequences are played out.
However, this is not the same as saying pay limits and risk-taking are closely tied. While I agree that executive compensation is a big factor, this is [about] the structure of incentives rather than the actual level of pay. If we see the asymmetric risk-reward structure or the short-term nature of incentives being the driver of risk-taking, then changing the level of pay misses most of the action. This may not get to the heart of the problem. Politically this is something that is very palatable, but economically it is not clear whether this is tackling the main problem.