CEO Salaries

Rajeet Guha

CEO Salaries: Trends, Key Issues and Way Forward


1. Introduction
Corporate governance has been in the spotlight for several years especially since a wave of accounting scandals swept through corporate America in 2002. Within this discourse the most divisive debates have been around the issue of CEO salaries and emoluments whether in Congress, in courts, in the business schools, on the streets or in living rooms. Almost not a day passes in America without the media passing some jibe or revealing some disgusting detail of unbridled extravagance in granting CEO bonanzas. Even yesterday (30th November, 2007) the on line media carried several such articles with such screaming headlines as: The 5 Richest payoffs for fired CEOs (MSN Money); The CEOs Beer Money (MSN Money) etc. In the last one year alone the Economist carried over 40 articles on this issue: a little less than one article every week. Clearly there is a lot of popular discontent about CEO salaries. In this essay we shall seek to go beyond the popular perceptions and (a) examine current trends regarding CEO compensation in America in comparison with that in Europe and other countries (b) unravel the different ethical, legal and economic issues underlying the discontent, and (c) suggest a way forward.

2. Trends in US
In terms of the reality of the situation one trend is almost universally true: the salaries of top managers in the corporate sector have risen at levels, often exponentially when compared to that of other categories of staff and workers. For almost half a century from the 1940s till 1990 the ratio of salary of top executives to the mean earnings i.e. median earnings remained unchanged at about 40 times. From the 1990s the ratio rose several fold dramatically in 2004. This is demonstrated by the graph below.



As the Economist article says: “At that point the worker on an American shop floor was earning in a year roughly what his boss on the top floor took home each evening.” As Professors Lucian Bebchuk and Jesse Fried point out that in the US between 1991 and 2003 the average large-company CEO's total remuneration increased from 140 times the pay of an average worker to 500 times average pay. At the same time in Australia between 1992 and 2002 CEO remuneration increased from about 22 times average weekly earnings to 74 times average earnings.


The Economist article goes on to say: “Most people think they know what lay behind this. Greedy chief executives, abetted by weak, sycophantic boards, gorged themselves at the expense of savers—more often than not the very pension and mutual-fund investors who, as workers, had seen their salaries and benefit packages fail to grow.” The common perception is also that such high payments routinely reward mediocrity and often even downright incompetence. The MSN Money article captioned, The 5 Richest Payoffs for Fired CEOs, referred to the $161.5 million retirement package former Meryl Lynch CEO Stanley O’Neil was given as he left after presiding over disturbing exposure to the collapsing sub-prime mortgage market. The Chief of Countrywide Financials Angelo Mozilo who perhaps takes a substantial part of the sub prime crisis which has brought America to the edge of a recession, will get more than $73 million. Returning to the Economist article it points out “The Corporate Library, an American corporate-governance consultancy, last year identified 11 large and well known but poorly governed companies, including AT&T, Merck and Time Warner, where the chief executive had been paid at least $15m a year for two successive years even as the company's shares had underperformed.”

3. Comparative Trends: US, Australia, Europe
Executive salary is controversial across geographical boundaries. As the size of the package increases, sharp rise in inequality with median earnings of companies rise exponentially, potential windfalls through share options grow and mismatch between company performance and salaries grow this controversy deepens. As Greg Cornish of the CEO Forum Group says globally today there is a new model and structure of CEO remuneration which comprises the following:
Base Pay
Cash salary
Flexible benefits
Statutory benefits
May include expatriate payments
Annual Incentives
Linked to business plan
Delivered in cash and/or equity
May include deferral
May link to value-added measure
Long-term incentives
3 to 5 year timeframe
Cash LTI
Equity - shares and/or options
Shareholder value focus
Referenced against peer group
The mix (salary vs. short term and long term incentives) of rewards as do the actual size of remuneration varies considerably between US, UK and Australia:
United States (485%)
United Kingdom (382%)
Australia (100%)

.



Clearly the remuneration levels (the last table is in Australian dollars), the incentive packages and the inequalities between remuneration at the median level are highest and sharpest in the US. In 2006 the 20 highest paid European CEOs made an average of $12.5 million, only one third that of what the 20 highest earning US CEOs made. The European CEOs earned less despite often leading larger firms. The 20 European firms with the top 20 CEOs had sales of $65 billion compared to $46 billion for the US counterparts. Part of the reason for this difference lies in social attitude: Europeans have strong cultural objections to paying their CEOs the sort of salaries that American CEOs get. The reason is also political: inequality has risen faster under the Republicans than under the Democrats. The unions are not just weaker than before but almost disenfranchised in the US.

4. Ethical, Economic and Legal Issue and Debates

Rising Inequality:
At a most fundamental level it is a reality that inequalities today are rising faster than ever before. As economists Thomas Picketty and Emmanuel Saez (2003) have shown, the 1 percent of households with the highest incomes now earn about 16 percent of all income (excluding volatile capital gains). This is truly a stunning level of inequality in America which frames the context within which the question of CEO remuneration needs to be seen. According to one view this situation has come about because of a never before seen rise in returns to talent and skills. It is as if a free market in talent has ended up spawning huge inequalities and CEOs sit at the top of this pyramid. According to the 14th Annual CEO Compensation Survey CEOs of large US companies last year made as much money from just one day on the job as average workers made over the entire year. The top 20 private equity and hedge fund mangers made an average of $657 million or 22,255 times the pay of an average US worker. Workers at the bottom rung received just the first minimum wage increase in a decade (after the Democrats won a majority in the house), which even after the revision is in real terms 7% below where it was a decade ago. American Business Executives earn much more than their European counterparts: in 2006 20 highest paid European managers made on an average $12.5 million, only one third as much as 20 highest earning US executives.




The staggering gap between CEO and Worker’s Pay is shown in the graph above. The recent minimum wage increase has made no dent in the almost unethical inequality between the top and the bottom of the American economy. The real value of the minimum wage, with last year’s increase from $5.15 to $5.85, now stands 7 percent below the minimum wage’s value in 1996. Average worker’s pay rise has also lagged behind quite steeply behind CEO compensation rise.

CEO Remuneration: Driven by Market Forces?
Although these numbers are shocking, American society seems to have become blasé about inequalities and some conservative viewpoints explain this as rising returns to talent and skill in a free market of talent. Obviously America, “the country with the world's freest market in talent is seeing a dramatic increase in inequality”. (Ibid: Economist: revenge of the Bell Curve”…) The argument is that in facing global competition, companies continue to keep their wage bills under strict control but at the same time are desperate not to lose their top talent to rival companies and are therefore willing to pay whatever it takes to retain them. In the process they end up keeping their wage bills down even while giving larger and larger shares to the top performers.
According to the conservative view the link between talent and inequality is being strengthened not just in the corporate sector or Hollywood but across society: even Universities are shedding their egalitarian and equity image and are increasingly willing to pay unusually high salaries for a very few academic stars “because they will attract other high-flyers.” (Ibid) The same is true for lawyers, hedge fund managers, athletes etc.
As Steven Kaplan argues, the increase in CEO salaries, particularly from 1994 to 2004 is a part of (not the cause of) the general increase in economic inequality being witnessed over the last several decades. If we are content saying that the remunerations of other groups have been driven by market forces, it is difficult to contend that CEO salaries have not been largely market driven. Kaplan identifies the market forces as changes in technology leading to dramatic productivity increases from the talented, the easy access to information of investors, brokers and traders, the easier and much larger reach of (say) athletes, sportsmen through the media etc. Other market factors include globalization via trade. As the share of trade in the GDP rises and as America goes on exporting goods requiring highly skilled labour and importing goods with low labour skill requirements, the demand for skilled labour in the US rises even as the demand for lower skilled labour decreases: thus increasing salaries of the more talented.
As the Economist article (the Revenge of the Bell Curve” goes on to say “The talent war is producing a global meritocracy—a group of people nicknamed “Davos men” or “cosmocrats” who are reaping handsome rewards from globalization. These people inhabit a socio-cultural bubble full of other super-achievers like themselves. They attend world-class universities and business schools, work for global organizations and speak the global language of business. Countries that still insist on clinging to egalitarianism are paying a heavy price. Sweden, for instance, finds it hard to attract foreign talent. And across Europe, egalitarian universities are losing out to their more elitist American rivals.” (Ibid)
Link between Pay and Performance: Rewards for Non-Performance: the Debate
The assumption in the foregoing view is that it is talent and therefore performance that commands a huge premium representing the scarcity value of that talent. There is however a contrary view that questions this very assumption and adduces strong empirical evidence to show that “in fact US CEO compensation is inefficient pay without performance”.
The evidence for increasing decoupling of CEO pay from performance is out in public for all to see. In 2006, departing CEOs Henry McKinnell of Pfizer and Robert Nardelli of Home Depot both received exit packages of more than $200 million. Both companies underperformed during their tenures, although their excessive pay was an issue in itself. In some cases, CEOs were entitled to receive generous exit packages, despite their involvement in the stock options backdating scandal. Former CEO Bruce Karatz departed because of options backdating at KB Home, but because he retired and was not fired, the terms of his employment agreement entitled him to an exit package worth as much as $175 million.
Excessive CEO pay inevitably leads to the question of governance. The board of directors is supposed to protect shareholder interests and ensure that CEO pay reflects performance. However, at approximately two-thirds of companies , the CEO also chairs the board. When a single person serves as both chair and CEO, it is impossible to objectively monitor and evaluate his or her own performance. CEOs also dominate the election of directors. The vast majority of directors are hand picked by incumbent management. A rent extractive perspective argues that some CEOs may make their own decisions about their compensation packages due to lack of shareholder oversight. In today’s situation, given diffuse ownership of large corporations, there is substantial discretion in the hands of the professional CEO in running the company and often CEOs use such discretion to sanction themselves huge remunerations and not maximize shareholder value. Although top executives generally have some degree of influence over their board, the extent of their influence depends on various features of the firm’s governance structure. Poor governance such as a CEO serving as chair of the board of directors, having few board meetings etc. may make it possible for CEOs to pursue their self interests. Executives who have more power vis-à-vis their boards receive higher pay—or pay that is less sensitive to performance—than their less powerful counterparts. A substantial body of evidence does indeed indicate that pay has been higher, and less sensitive to performance, when executives have more power.

Bebchuk and Fried (Ibid) show that non-equity compensation such as salary, bonus (acquisition and signing), severance payments etc., are not tied to performance (or weakly tied). Further CEOs are often paid for stock price increases that are unrelated to the CEOs own performance but are instead related to general bullish rise in the stock markets. Why should CEOs be rewarded for a bull market since their actions have no direct impact on the market? Similarly executives are often paid bonuses even when their performances are poor. All these are evidence consistent with the view that CEO remuneration is not tied to performance. Executives also get payments such as bonuses for acquisitions although acquisitions frequently result in stock price decreases after the purchase. The authors question why if the acquisitions result in value loss for the shareholders, should CEOs be rewarded. Finally they critique the widespread system of paying severance packages for departing CEOs even when the firms have under performed.

Bebchuk and Fried have also unraveled several aspects of equity based payments which expose the de-coupling of pay and performance. They refer to such practices as use of at-the-money options, option re-pricing, reload options, restricted stock in lieu of options, and executive’s ability to unwind their equity positions which all in some ways allow even non performing CEOs to receive huge payments thus defeating the overt purpose
of equity related payments to link CEO compensation with company performance. Managers according to Bebchuk and Fried have almost unfettered discretion to sell their vested stock and options, thus weakening their incentives to maximize share holder value. Further managers have almost total control when they unload their stock options and this gives them the opportunity even to engage in insider trading through their superior information over others.

This has considerable significance and implications. Today nearly half of Americans own some stock, with many middle-class families relying on capital market investments to finance their retirement security or help their children afford college. Workers with traditional pension plans, including many public employees, also have a stake in the long-term financial performance of companies their pensions are invested in. As a result, the notion that retirement security that middle-class Americans work a lifetime to earn can be thrown into jeopardy by lavish CEO pay and retirement packages that reward executives with hundreds of millions of dollars even if they perform poorly or behave unethically is a very emotive and strongly moral one. Excessive CEO pay not only cuts into the company’s bottom line, but can also provide incentives for CEOs to manipulate earnings or encourage mergers that pump up their compensation packages but are unprofitable for the company as a whole – directly undermining shareholder value. Simply put: when CEOs get sumptuous rewards for inferior or mediocre work, middle-class Americans get short-changed on the retirement benefits they’ve earned.
In the last twenty years there has been considerable development in empirical economic testing to see if performance-based incentives are effective. For many years the conclusion of all these studies was that there was no correlation between CEO compensation and company performance. However, these arguments including those of Bebchuk and Fried have been countered by more recent research by Steven Kaplan (Ibid) and Mathew S. Lilling. Steve Kaplan (Ibid) shows from his research that in fact CEOs who perform better earn more pay, i.e. contrary to popular perceptions, CEOs are paid for performance. According to them, critics sometimes mix up theoretical pay (what the boards give to the CEOs as estimated pay) and actual pay. The former includes salary, bonus, value of restricted stock issued and the estimated/theoretical value of options issued to the CEO that year. But this is not what the CEO actually gets to take home is salary and bonus, while he does not get to cash the stock options. Their research shows that firms with CEOs in the top decile of pay earned stock returns 90% greater than those of other firms in their industries over the previous years. Firms with CEOs in the bottom decile of pay actually underperformed their industries by almost 40% (Steve Kaplan: ibid) during the same period. Mathew Lilling in his research also shows significant positive relationship between market value of a company and CEO remuneration. He therefore recommends that incentive based contracts are effective due to this positive relationship between CEO compensation and market value of a firm.
Steven Kaplan (Ibid) had done considerable research on this and his conclusions quite decisively question the pay-performance de-coupling notion on several fronts. First he argues that the US Economy has done extremely well over the 16 years (till very recently following the sub prime crisis and dollar slide) in absolute terms and in comparison with other countries. The stock market rise and productivity growths have been cited as evidence for this: since 1996 the US Stock market consistently increased by 6% net of inflation. At a broad level therefore it may be reasonable to say that US CEOs have done remarkably: ensuring consistent productivity growth and share holder returns and therefore desire to be rewarded more than others. Secondly he questions whether they really overpaid. He points out the often somewhat facile way CEO compensation is calculated. There is the estimated or theoretical value of CEO pay which includes salary, bonus, the value of restricted stocks issued and the theoretical value of options issued to the CEO during the year. This is not a measure of the CEO’s take home pay! The CEO takes home only his salary and bonus but does not get to cash his options. Thirdly as we have pointed out in the foregoing section, salaries have risen across the board and therefore rise in CEO salary is more an effect of market forces. He also questions the hypothesis about boards being dominated by CEOs and shows the high CEO turnover levels as proxy evidence to the contrary. Since 1998 in any single year, one in six Fortune 500 CEOs lost their jobs compared to one out of ten during the 70s. He also posits that such high CEO turnover in strongly related to poor performance and that this trend is the result of pressure by the boards driven by institutional share holders and hedge funds. .
The underlying issue in this debate is the principal-agent problem. The board member’s goal is to figure out a way to compensate the CEO fairly, while still giving him the incentive to look out for the best interests of the company. The CEO is here the agent while the shareholders are the principals. The CEO is looking at his own interest and wants to paid more and more while the principal (share holder) has a stake in the company and so wants the company to do well above all other considerations. The board member must find a way to compensate the CEO so that the company’s high performance will translate into a considerable salary- One way is to use an incentive based contract wherein the CEO is paid besides a base predictable salary is rewarded with a performance-based bonus which can be in the form of cash, stocks or stock options. The Mercer Issues Annual Study (2007) of CEO Compensation reports that in 2007 there is a definite increase in trends to link long term incentives to performance. The number of CEOs receiving performance shares, including performance shares, performance-contingent restricted stock, jumped from 111 in 2005 to 178 in 2006. The portion of CEO payment that was made up of performance-based shares and units jumped from 2005 to 2006 from 21% of the pay mix to 31%.
In terms of linking pay to incentives, overtly the US does much better than Australia or Europe as the pie chart on page 5 of this essay shows: long term and short term incentives comprise much larger shares of CEO salary in the US than in these two continents. I am using the word “overt” because as the research by Bebchuk and Fried (Ibid) shows the picture is quite murky even when it comes to “incentives” which are generally equity value linked payments. The traction between such payments and performance is often very weak.
5. The Legal Environment
Rules of Disclosure: Are They Adequate?
As public concern in the US over corporate governance and specifically over CEO compensation has mounted, so have attempts to use laws and regulations in particular to get some traction between CEO compensation and CEO performance. There are broadly two legal streams relevant in this issue: one is the Sarbanes Oxley Act which broadly seeks to strengthen CEO and Board accountability and the SEC Disclosure Rules which require companies to be more transparent about CEO compensation: understood to be a first step towards holding boards of directors accountable for excessive CEO compensation. These new disclosure rules were brought into effect by the US Security and Exchange Commission (SEC) in 2006. These new rules go further than ever before in making transparent CEO compensation including information such as pension and estimated severance package totals.
However this advance has also been marked by determined attempts to obfuscate attempts at complete transparency especially regarding stock options. In fact although the new rules represent a substantial improvement in pay disclosure, an 11th-hour change by the SEC resulted in less meaningful disclosure of stock options. An absurd example of how such obfuscation and confusion can happen when the rules are applied in practice as cited in the reference is the case of Ian Cockwell CEO of Brookfield Homes. According to the SEC rules he made -$2,296,918 in 2006. “However, Cockwell did not give money back to the company. The method with which Cockwell’s past equity awards were calculated for financial reporting purposes resulted in a negative figure that was larger than the rest of his 2006 compensation. In actuality, Cockwell’s cash compensation increased from 2005. His total compensation for 2006 is $2,015,499, including the grant date fair value of stock and options awarded in 2006”. (AFL-CIO: New SEC Disclosure Rules: Ibid) The new rules are widely seen as benefiting CEOs and providing escape hatches rather than really strengthening accountability.
In the wake of growing criticism of insufficient accountability to shareholders regarding CEO compensation, Congress has been considering legislative action to cap/regulate CEO salaries. The House Financial Services Committee has earlier this year approved a “say on pay” bill requiring advisory votes by shareholders on CEO compensation. According to this proposed bill shareholders would vote once a year on pay packages. While the votes would be non-binding but they are widely expected to push boards to be more independent of management. On the other hand some critics say such a vote would plunge boards into political discussions and detract them from discussing and deciding on strategic matters regarding the running of the company. Others argue that shareholders have no business dictating CEO salaries. This is a management function and not the shareholders’ concern. Steven Kaplan on his part makes the point that this bill (HR 1257) would not bring additional benefits and instead impose costs. He argues that this would result in companies which are doing badly to have negative votes, which even now most such companies get though it is not a mandatory voting. On the other hand if made mandatory even the companies doing well will have to go through this voting resulting in additional costs.

6. The Way Forward
The truth of the entire matter probably lies in the middle. While it is true that along with CEO remuneration, salaries in different sectors including education have also risen as part of a post globalization rise in economic and social inequalities, the ratio of CEO and worker pay in the US at 260:1 is completely indefensible. There have also been major cases of failure of corporate governance and CEO pay abuses, although sometimes a few bad cases do exaggerate and sensationalize the issue. The delinking of pay and performance is also too frequent a occurrence to be passed off as aberrations. Ultimately, shareholders have to be able to trust their boards of directors to set responsible CEO pay packages. For this reason, CEO pay will be reformed only when corporate boards become more accountable. Until then, CEOs will continue to influence the size and form of their own compensation, and CEO pay will continue to rise.
We also need to see the issue in a larger perspective which not only includes corporate governance in general but also larger issues of social equity and accountability. While there is need for strong market regulatory mechanisms that create more transparencies and redress mechanisms, reform of tax systems would also be important. People of this country (88%) think the top earners are not taxed enough and that there are explicit and implicit tax subsidies for the very rich which need to be plugged. For instance corporations can now deduct as “business expense” the excessive pay packages that they offer to the CEOs by defining such excesses as performance incentive. There are proposals for capping the amount of CEO pay that companies can deduct as business expense to 25 times the pay of the lowest paid worker. They calculate that if this was introduced last year on the 326 Fortune 500 companies surveyed, an additional $1.4 billion would have been raised through taxes in 2006 and this could have gone to education or health reforms.
Other proposals include ending the preferential tax treatment of Private Investment Company Executive Income; imposing a cap on Tax Free “Deferred Pay Accounts” for CEOs which also carry guaranteed interest etc. Another suggestion is to increase marginal tax rates on high incomes. The merit of some of these tax reform proposals is that they would address the larger issue of social inequalities without singling out CEO pay.
We end this essay with a quote from a blog from Robert Reich who is America’s 22nd Labour Secretary and Professor at Berkeley: “The real scandal of CEO pay is that it has become so far removed from the pay of average workers. And average workers aren’t doing so well. Median wages are still below where they were in 2000, adjusted for inflation. Every two weeks, Scott Lee, Jr., the CEO of Wal-Mart, rakes in roughly the same amount of money that his average employee earns in a lifetime.
We should stop worrying about linking CEO pay to company performance, and start worrying about linking CEO pay more closely to the pay of average workers. How? One simple starting place: Restore progressivity to the personal income tax.”