Tuesday, May 29, 2007

Review of Pay without Performance: the Unfulfilled Promise of Executive Compensation

Pay without Performance: The Unfulfilled Promise of Executive Compensation
Lucian Bebchuk and Jesse Fried
Cambridge, Mass.: Harvard University Press, 2004
278 pages

Pay without Performance is a well-written book that provides a post-'90s update to Graef Crystal's 1991 In Search of Excess. Bebchuk and Fried recast the 75-year old debate about executive compensation that began with Berle and Means's 1932 Modern Corporation and Private Property into two alternative models: the arm's length bargaining and the managerial power models. Their assessment, which is similar to Crystal's, is that managerial power has guided executive compensation practices and led to excesses that the authors catalogue, updating Crystal's book. Their book is more theoretical than Crystal's, but it is theoretical in an accessible and elegant way that is a pleasure to read.

Like Crystal, Bebchuk and Fried argue that top managers' extraction of rents has not been and will not be resolved by spontaneous market forces. In other words, top executives have found loopholes in the structure of the American corporation and have learned to use the loopholes to their financial advantage. The amounts of excess compensation that executives extract are probably too small per shareholder to concern the shareholders. This is so even though executives extract large sums.

Bebchuk and Fried do not discuss the broader social implications of the separation of pay from performance, but it is probably serious. Public support for the market system and belief in hard work is likely eroding because of the example that CEOs have set.

Also, Bebchuk and Fried do not discuss the role of both macroeconomic factors and human resource management factors that would potentially address the problem of excessive CEO compensation. First, with respect to macroeconomic factors, low interest rates make this issue more salient than it would be in a free market environment without loose Fed monetary policy.

As Bebchuk and Fried point out, managers have taken advantage of general stock market returns that have been boosted by low interest rates. That's not the sole cause of excessive compensation because managerial compensation is not very well-correlated even with general stock market returns. However, declining interest rates cloak shoddy management by boosting stock prices. We have been living in a twenty-five year old stock market bubble caused by increasing money supply. Any bubble is characterized by excesses in fields like executive compensation. When interest rates rise, top pay will subside. Top manager self-dealing and pontification will always exist, but it will seem less disturbing as the worst-managed firms go under and the amounts shrink.

I don't believe that improving the design of incentives will alter the pattern of self-dealing to the degree that Bebchuk and Fried believe. There is an unresolved paradox in the executive compensation literature. First, Jensen and Fama (1983), for example, claim that the hiring of insiders as directors and the informal hiring processes that characterize selection of board members can be explained by the quest to optimize board expertise. "(I)t is natural that when the internal decision control system works well, the outside members of the board are nominated by internal managers. Internal managers can use their knowledge of the organization to nominate outside board members with relevant complementary knowledge."

In fact, Jensen and Fama's contention diverges from decades of research with respect to employee selection. Formal selection methods such as structured interviews, IQ tests, job knowledge tests and work samples are superior (in psychological parlance more valid) than informal processes such as impression or subjective evaluation. Anyone with real world knowledge of the corporate world knows that executives "manage" impressions about their competence.

Moreover, there are agency problems involving board appointment such as quid pro quo, the hiring of board members who are unlikely to rock the boat and interlocking appointments. That agency theory was oblivious to the distortions of the abstract model of board governance, and the open secret of "ol' boy networks" in board appointments seems hard to believe in the post-Enron world.

The paradox is that at the same time that agency theorists have argued that "ol' boy" network board hiring practices result in more capable boards, they also argue that because of special failings of CEOs, such as risk-aversion, that result from the practices, executives deserve higher pay. As Bebchuck and Fried note, top managers tend to be risk averse and so need to be motivated with stock options. Jensen and Fama, Murphy and others claim that supply of qualified executives is wage inelastic because there are so few qualified executives. This seems contradictory. If current executives have problematic personality traits such as risk aversion, just what about them is so desirable that their supply is wage inelastic? More importantly, no one in the executive compensation literature seems to have been eager to ask these questions: What are the specific competencies, if any, that qualify executives? Where are the studies that validate the supposed competencies? Why aren't corporations falling over themselves to identify this obviously valuable information? Why, instead, do corporations insist on the secrecy of selection methods, competence information and validation of board member and executive selection?

In the book Good to Great, Jim Collins (2001) points out that the best executives focus on teamwork, set up their successors for greater success than they have achieved, display compelling modesty, are self-effacing and understated, focus on sustained results, and attribute success to factors other than themselves but take full responsibility for failure. Collins notes that these traits are not rare among the general population. They are just rare among current top managers! In other words, boards have failed to even begin to apply systematic methods to top manager selection. This would seem to be related to the character weaknesses that Bebchuk and Fried find commonly displayed in top managers' indifference to their fiduciary duties. Collins notes (p.39):

"I believe that (such) leaders exist all around us, if we just know what to look for, and that many people have the potential to evolve (into great leaders)."

If Collins is right, then there are important staffing issues. Why these aren't discussed in the compensation and governance literature is a puzzle. Where are the competency studies to identify traits necessary for directors? For CEOs? Are there systematic approaches to selecting directors, or do CEOs hire to boards guys with whom they play golf or shoot pool? I wonder how many firms use objective selection criteria in choosing a CEO or director before they are interviewed.

In 1990, Lex Donaldson (1990, paid access) criticized organizational economics and agency theory, which holds that interests of principals and agents diverge; that their interests can be aligned through incentives; and that shirking or non-compliance with ethical or contractual obligations needs to be addressed with efficient contracts and rewards. Donaldson argued that agency theory puts too much methodological emphasis on individualism. In the management literature, Donaldson pointed out, agency theory is known as MacGregor's "theory x". MacGregor argued that people who believe in theory x believe that workers cannot be trusted, that they need to be motivated by financial incentives and that they need to be monitored. In contrast, people who believe in "theory y", which MacGregor advocated, believe that people can be trusted, that they are naturally creative and that they aim to achieve.

Donaldson pointed out that most of the time people need to be motivated with a combination of theory y and theory x, but agency theory is too narrowly focused on theory x. "Organizational economics creates a theoretical scenario in which managers act opportunistically."

Donaldson argued that stewardship theory is an alternative model to agency theory. Rather than a conflict of interest between principals and agents, stewardship theory would propose that managers are team players and that "the optimal structure is one that authorizes them to act in the owners' interests". Donaldson argued that many of the governance reforms that Bebchuck and Fried propose, such as having increasingly independent directors and allowing shareholders to select directors independently of the CEO, would contradict the prescriptions of stewardship theory. Donaldson was concerned that both agency theory and stewardship theory might be too broad.

Things seem to have deteriorated since Donaldson's article. In the 1980s, there was talk of what William Ouchi (1981) called Theory Z: high trust human resource management and tight organizational culture. Ouchi argued that such arrangements are optimal because they reduce transactions costs. Similarly, until his death in 1993, W. Edwards Deming (2000) the creator of many of the ideas behind total quality management, argued that long term thinking, driving fear out of the workplace and elimination of performance appraisal were essential to good management.

Current CEO pay practices seem to have thrown the "stewardship" theory out the window. Bebchuk and Fried do not even consider that organizational culture and high trust necessitate current pay practices. If not, then it would seem that the best hopes of management theorists like MacGregor, Ouchi and Deming have been dashed. Rather than behave as part of a coherent, organic organization in which social relationships with employees and with society play a paramount role, today's executives in Bebchuk's and Fried's view function as manipulators who can be influenced by incentives, if only their tendency to outsmart the incentive plan can be overcome. This is very far from traditional "stewardship" management theory.

Instead, Bebchuk and Fried contrast the agency model with the managerial power model. According to the agency model, CEOs may engage in (p.16) "empire building, which can increase their prestige, perks, compensation and other private benefits...They may run companies in ways that are personally satisfying or convenient even if they come at the expense of shareholders. They may be tempted to pursue pet projects, for example, or they may fail to take actions that are personally costly, such as firing mediocre subordinates who also happen to be their friends." According to agency theory, directors (p.17) "take an independent position vis-a-vis executives" and bargain at arm's length." In order to counteract such tendencies, firms need to link pay with performance. But because executives are characteristically risk averse, linking pay to performance means that pay amounts must increase.

Bebchuk and Fried argue against the agency model of arm's length bargaining. Executives do not (p. 23) "instinctively seek to maximize shareholder value" so there is "no reason to expect a priori that directors will act in this way." In particular, directors of the thousand largest firms earn $116,000 just for serving on the board. They would not want to jeopardize their compensation by fighting over the CEO's pay given that the CEO is largely the one who appoints them. Moreover, most boards emphasize collegiality, so even if money is not the issue, the cultures of corporate boards involve norms that inhibit questioning the CEO's pay demands. There are also factors like friendship, reputational risks and, if the directors are themselves highly paid CEOs, cognitive dissonance.

Shareholders have a few ways to intervene, such as through law suits, voting on option plans, or resolutions, but these methods turn out to be ineffective. Likewise, takeover bids are unlikely just because of executive compensation. The 40 percent premium that takeovers require are usually too great for the one to three percent effect on stock prices due to excessive executive pay to justify.

The relatively weak correlation between executive pay and firm performance; the reissuing of stock options at lower strike prices; the use of camouflage whereby true pay amounts are not admitted or are disguised through the use of pension formulas or by failing to disclose the tax effects of deferred compensation all suggest that the exercise of managerial power rather than arm's length bargaining determine executive pay. Bebchuk and Fried do not consider the possibility that these behaviors are part of the adaptive corporation that Alfred Dupont Chandler (1962) described in Strategy and Structure. Such a claim would be far fetched.

Bebchuk and Fried present research findings that suggest that CEO power plays a role in determining pay. CEO compensation increases in tandem with the size of the board. The probability of getting a golden parachute is greater when more board members have been directly appointed by the sitting CEO. Lower institutional ownership leads to higher and less performance sensitive executive compensation. Managerial power seems to be linked to pay.

Bebchuk and Fried do an excellent job of cataloguing the different ways that boards camouflage pay through retirement pensions, deferred compensation, post-retirement perks, consulting fees and loans. They also point out that (p. 121) "managers have used their power to secure pay without performance." Cash compensation has at most weakly correlated with performance. It is, however, associated with cash windfalls that firms receive, for instance when they win a lawsuit or terminate a pension plan with excess assets. In those cases executives get a nice piece of the windfall. Executives of acquiring companies are often paid bonuses, even though acquisitions almost always result in a decline in the acquiring firm's stock price. Only 30 percent of share prices reflect corporate performance with the remaining 70 percent due to market factors like interest rates. Yet, almost all executives are compensated on the raw share price rather than on a sector-adjusted share price.

The authors, like Crystal (1991) argue against restricted stock. I take issue with that particular argument. If the value of the restricted stock is adjusted so that executives are equally well off, I would think that restricted stock is better than options. The reason is that options might encourage excessive risk taking. A large holding of stock (and nothing else) would seem to put executives in the same position as shareholders. Of course, the parenthetical "and nothing else" is the devil in the deep blue sea.

The authors conclude with a convincing set of recommendations that focus on the issue of compensation. As I have already noted, my claim is that the problem may be one of executive and board selection rather than governance and compensation.

Overall, this is a very good book. It is clearly written, practical, theoretically sound and a joy to read.


Collins, J. 2001. Good to Great: Why Some Companies Make the Leap and Others Don't. New York: Harper Business.

Chandler, A.D. 1962. Strategy and Structure. Cambridge, Mass.: MIT Press

Deming, WE. 2000. Out of the Crisis. Cambridge, Mass.: MIT Press.

Donaldson, L. The Ethereal Hand: Organizational Economics and Management Theory. 1990. Academy of Management Review, 15:3. 369-381.

Ouchi, W. G. 1993. Theory Z.

Jensen and Fama. 1983. "The Corporation and Private Property." Journal of Law and Economics. 26:2


Paul said...
This comment has been removed by the author.
Paul said...

I think that in order to be a leading company, it is always necessary to make board appointments every two or three years.