Microsoft Word Remuneration7 4 04 1.doc

Transcript

1 Remuneration: Where we’ve been, how we got to here, what are the problems, and how to fix them Michael C. Jensen Finance Working Paper N°. 44/2004 Harvard Business School, Monitor Group, July 2004 Cambridge, Massachusetts and ECGI Kevin J. Murphy USC Marshall Business School with the assistance of Eric G. Wruck, Econalytics © Michael C. Jensen and Kevin J.Murphy 2004. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit per- mission provided that full credit, including © notice, is given to the source. This paper can be downloaded without charge from: http://ssrn.com/abstract=561305 www.ecgi.org/wp

2 ECGI Working Paper Series in Finance Remuneration: Where we’ve been, how we got to here, what are the problems, and how to fix them* Working Paper N°. 44/2004 July 2004 Michael C. Jensen Kevin J. Murphy with the assistance of Eric G. Wruck * While this is an independent study and the views expressed are solely those of the authors, we wish to thank BP for fi nancial support, Siew Hong Teoh for sharing her accounting research expertise and data, and Joe Fuller, Amy P. Hutton, and Karen H. Wruck for useful comments and suggestions. The authors are solely responsible for all errors of fact or interpretation. © Michael C. Jensen and Kevin J. Murphy 2004. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.

3 Comments Welcomed Negotiation, Organizations and Markets Research Paper Series Harvard Business School NOM Research Paper No. 04-28 Remuneration: Where we’ve been, how we got to here, * what are the problems, and how to fix them by Michael C. Jensen [email protected] Jesse Isidor Straus Professor of Business Administration, Emeritus Harvard Business School; Managing Director, Organizational Strategy Practice, The Monitor Group Kevin J. Murphy [email protected] Vice Dean of Faculty and Academic Affairs, E. Morgan Stanley Chair in Business Administration, USC Marshall School of Business with the assistance of Eric G. Wruck [email protected] Cofounder, Econalytics July 12, 2004  Michael C. Jensen and Kevin J. Murphy 2004. All rights reserved This paper can be downloaded without charge from the Social Science Research Network Electronic Paper Collection: http://ssrn.com/abstract=561305 * While this is an independent study and the views expressed are solely those of the authors, we wish to thank BP for financial support, Siew Hong Teoh for sharing her accounting research expertise and data, and Joe Fuller, Amy P . Hutton, and Karen H. Wruck for useful comments a nd suggestions. The authors are solely responsible for all e rrors of fact or interpretation.

4 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION ABSTRACT Currently, we are in the midst of a reexamination of chief executive officer (CEO) remuneration that has more than the usual amount of energy and substance. While much of the fury over CEO pay has been aimed at executives associated with accounting scandals and collapses in the prices of their company’s shares, the controversies over GE CEO Jack Welch and NYSE CEO Richard Grasso sign al a watershed. In their cases the competence and pe rformance of both men were unquestioned: the issue s eems to be the perception that they r eceived “too much” and that there was inadequate disclosure. zen recommendations for reforming the system s urrounding We provide, history, analysis and over three do executive compensation. Section I introduces a conceptual fr amework for analyzing remuneration and incentives in organizations. We then analyze the agency problems between managers and shareholders and between board members and shareholders, and discuss how well designed pay packages can mitigate the former while well designed corporate governance policies and processes can mitigate the latter. We say “mitigate” b ecause no solutions w ill eliminate these agency our discussion includes problems completely. Since bad governance can eas ily lead to value destroying pay practices analyses of corporate governance as well as pay design. Because optimal remuneration policies cannot be designed and managed without consideration of th elations and interactions between the financial markets and the e powerful r firm, its top-level executives and the board, we devote significant space to these factors. Section II offers a brief history of executive remuneration from 1970 to the present. Section III examines and explains the forces behind the US-led escalation in share options. We argue that boards and managers falsely perceive stock options to be inexpensive because of accounting and cash-flow considerations and, as a result, too many options have been awarded to too many people. Section IV defines and discusses the overvalued equity as the source of recent c orporate agency costs of scandals. Agency problems asso ciated with overvalued equity are aggravated when managers have large holdings of stock or options. Because neither the market for c orporate control or the usual incentive compensation systems can solve the agency problems of overvalued equity, they must be resolved by c orporate governance systems. And few governance systems were strong enough to solve the problems . As the overvalued equity problem illustrates, while remuneration can be a solution to agency problems, it can also be a source of agency problems. Section V discusses several widespread problems with pay processes and practices, and suggests changes in both corporate governance and pay design to mitigate such problems: including problems with the appointment and pay-setting process, problems with equity-based pay plan s, and problems with the de sign of tradition al bonus plans. We show how traditional plans encourage managers to ignore the cost of capital, manage earnings in ways that destroy value, and take actions to deceive investors and capital markets. ii

5 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Section VI defines and analyzes a new concept: what we call the issue. This is Strategic Value Accountability the accountability for making the link between strategy formulation and cho ice and the value consequences of those choices — basically the link between internal managers and external capital markets. The critical importance of this measurement and remuneration have long been accountability, its assignment, and its implications for pe rformance unrecognized and therefore ignored in most organizations. Section VII analyzes the complex relationships between managers, analysts, and the capital market, the incentives firms have to manage earnings to meet or beat analyst forecasts, and shows how managers playing the earnings-management game systematically erode the integrity of their organization and destroy organizati onal value. We highlight the puzzling equilibrium in this market that seems to suggest collusion between analysts and managers at the expense of investors — an area that is ripe for further research. iii

6 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Executive Remuneration T A B L E O F C O N T E N T S Introduction and Summary ... 1 ...3 ... Overview ... 7 List of Recommendations and Guiding Principles I. The Conceptual Foundations of Executive Remuneration...15 The Governing Objective of the Corpor ation ...15 rd Gives No Scor e ...17 The Balanced Scoreca Enlightened Sta Enlightened Value Maximization and Theo ry...17 keholder Firm Value Maximization Does Not Imply t Run Sto ck Pr ice...17 Maximization of Shor Remunera The Economics of tion ...19 1. The expected total benefits associated with the job or position (including the costs ry aspects of and benefits of non-pecunia ob). ...19 the j 2. The composition of the remuneration package. ...19 3. The relation between pay and performance (what for shorthand we call the pay- performance relation). ...19 Agency Problems and Ex n...21 ratio ecutive Remune II. A Brief History of Executive Remuneration ...23 The Worldwide Econ omic Envir onmen t...23 ve Remune ration ...24 Trends in Executi ...26 The 1970s ... The 1980s ... ...27 The 1990s ... ...29 Trends in CEO Demographics ...32 III. The US-led Opti on Explosion...35 The Cost and Value of Options...38 IV. Corporate Scandals and the Agency Costs of Overvalued Equity ...44 Managerial and Organizational Heroin ...45 Failed Governance and Failed Incentives ...47 The Solution?... ...49 V. Executive Remuneration as an Agency Problem...50 Problems with the Appointment and Pay-Settin g Pro cess...50 How pay decisi ons are made...50 Pay negotiations and the market for CE Os ...51 CEO ...53 Judgment calls go to the iii

7 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION ation Consu ltant The Role of Compens s...55 Problems with Equ ity-Based Plans ...57 No skin in the game ... 58 Problems with traditi onal stock opti ons ...60 Cost of Capital Adjusted Options ...61 Benefits of In centive Pa y...65 Risk Aversion and the Cost and Make Unwinding Rights Explicit in Incentive Remune ration plans ...66 Problems with Annual Bonuses ...68 Poorly designed pay-performance relations...70 ..71 Paying people to lie... Poorly designed perf ormance standards...75 ormance me Poorly designed perf asures ...77 VI. Strategic Value Accountability and Remuneration Policy ...81 Tensions Between Outside Markets and Internal Management ...81 rategic Value Account ability ...83 The Critical Importance of St The importance of risk and trust in management an d governance...86 VII. Relations with Capital Markets: The Earnings Management Game...87 d Punish Ma nagers ...87 How Markets Reward an Ethical and Value Consequences of the Earnings Management Game ...89 How Managers Reward an d Punish An alysts ...90 Evidence on the Collusive Nature of Earnings Forecasts an d Realizations...92 VIII. Conclusions ...98 ...99 References iv

8 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R T A B L E O F F I G U R E S Figure 1 S&P 500 Firm s, 1970-2002 ...25 Average Cash and Total Remune ration for CEOs in on for CEOs in S&P 1970-2002 Average Cash Remunerati ...26 Figure 2 500 Firms, Average Remuneration 500 Firms, 1992-2002 ...31 Figure 3 for CEOs in S&P Outside Hires as Percentage of New CEO Appointments in Large US Firms, 1970- Figure 4 32 2000... Dow Jones Industrial Average Cash and Total Remuneration for CEOs in S&P 500 Figure 5 1970-2002 ...36 Firms, Grant-Date Values of Employee Stock Options in the S& P 500, Figure 6 ...37 1992-2002 Figure 7 Grant-Date Number of Employee Stock Options in the S& P 500, 1992-2002 ...38 Firms timed reductions in retiree h Figure 8 accounting ealthcare benefits to boost reported earnings ...41 Figure 9 alue Creation and No Cost-of-Capital Indexed Options Have Higher Payoff for V eat Capital as Cos Incentives for Managers to Tr tless. It pays managers to choose the indexed option plan as long as they b elieve they can create more value than the ...63 breakeven level of $20 in this example. Base Salary and Bonus for a Typical Annu al Bonus Plan ...70 Figure 10 Figure 11 rformance Relations eliminate gaming and budget-related How Linear Pay-Pe incentives to lie...74 Figure 12 Market-Adjusted Returns for Growth & Value Firms in Response to Quarterly Earnings S urprises...88 Figure 13 The Puzzling Systematic Positive Long-Term Bias and Short-Term Neg ative Bias in Analyst Earnings For ecasts ...93 Figure 14 More Evidence on Lying About Earnings: Frequency Distribution of Earnings Per Share For ecast E rror ...95 L I S T O F T A B L E S ...33 Table 1 Summary Statistics for Newl y Appointed CE Os, 1 970-2000 v

9 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Introduction and Summary Few issues in the history of the modern corporation have attracted the international attention garnered by what the largest corporations pay their top executives. Fueled by disclosure criticizing executive remune ration has been a requirements and human envy, analyzing and popular sport among business pundits for decades. Currently, however, we are in the midst of a reexamination of chief executive officer (CEO) rem uneration that has more than the usual amount of substance. In their 1990 study of executive of energy and more than the usual amount compensation Jensen and Murphy pointed out that CEO pay had not risen in real terms from the 1930s: bonuses. Despite the headlines, top executives are not receiving record salaries and Salaries and bonuses have increased over the last 15 years, but CEO pay levels are just now catching up to where they were 50 years ago. During the period 1934 through 1938, for example, the average salary and bonus for CEOs of leading companies o n change was $882,000 (in the New York Stock Ex 1982 1988 dollars). For the period through 1988, the average salary and bonus fo r CEOs of comparable companies was 1 $843,000. itten 14 years As we now know, things have changed dramatically since these w ords were wr ago, and the result has been much controversy. Over the past two years, much of the fury over CEO pay has been aimed at executives associated with accounting scandals and collapses in the prices of their company’s shares. However, two landmark ev ents may prove to be even more important in signaling changing remuneration policies, practices, and processes. First, in September 2002 the elations of lavish reputation of legendary General Electric CEO Jack Welch was shattered by rev personal retirement benefits that were allegedly not disclosed to the GE Board or GE shareholders. Second, in September 2003 Richard Grasso was forced to resign as CEO of the New Y ork Stock Exchange after revelations that he was to r eceive total accrued retirement and savings benefits of nearly $190 million. We believe these events signi fy a watershed because the competence and performance of both men were unquestioned: the issue seems to be the perception that they r eceived “too much” and that there was inadequate dis closure. Undoubtedly the reactions have been affected by the contemporaneous failures in organizations other than GE and the NYSE; including widespread revelation of failed corporate governance systems, corporate misdeeds, manipu lated financial reporting, fraud, bankruptcy and liquidation that has occurred contemporaneously with the loss of trillions of dollars in equity values associated with declines in worldwide stock prices. 1 Jensen and Murphy, "CEO Incentives-It's Not How Much You Pay, But How",. -1-

10 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Our purpose here is to review where we’ve been in the last several decades in executive remuneration, how we’ve gotten to where we are now, and to assess how we might re-think executive remuneration to provide a solid foundation on which to formulate current and future remuneration policy. We focus on top-executive rem uneration which is but a part of the overall orporate labor market. Because similar issues pertain to employees who are not at the top of the c hierarchy, the thinking embodied in this report will be useful in considering remuneration policy for these employees as well. In analyzing trends and practi ces and reaching conclusions, we draw on the extensive and growing academic literature on executive remuneration in accounting, economics, finance, and 2 ce of recent exchange listing guid In addition, we note the existen organizational behavior. elines and reports from industry groups, especially reports from the Conference Board (2002), the 3 These Business Roundtable (2003), and the National Association of Corporate Directors (2003). reports provide thorough analyses of the role of remuneration committees, and also offer thoughtful recommendations on improving practices, most (but not all) of which we endorse and many of which mirror our own reco mmendations discussed below. The existence of these reports has relieved us of the obligation to describe the ma ny roles, functions, processes and obligations of remuneration committees in detail, but instead allows us to focus our effort here pr imarily on rethinking the conceptual foundati ons of executive co mpensati on. While our primary focus is process remuneration, we also discuss where necessary the major forces influencing the pay-setting that are critical to achieving well-designed pay systems, including corporate governance systems, bor market, and the compensation consultants, external financial markets, the managerial la government. We acknowledge that much of our focus is on r emuneration practices in the US. This is due partly to data limitations and disclosure policies in other countries, but also because (for better or worse) the US is the undisputed trendsetter in executive remuneration practices. 2 See the survey article by Murphy, 1999, "Executive Compensation", in ed. Ashenfelter and Ca rd, Handbook of Labor Economics, 3, North Holland for an overview of the academic literature, including cites to nearly 200 academic articles related to execu tive incentives, r emuneration, and t urnover. Reprints of 45 of the most influential academic articles on executive pay ar Hallock and Murphy, 1999, The Economics of Executive e available in Compensation V. I & II, Elgar Reference Collection, International Library of Critical Writings in Economics, Cheltenham, UK: Edward Elgar Publishing. 3 In particular, we refer the reader to the Confere nce Board’s “Commission on Public Trust and Private Ente rprise” published in September 2002; the Business Roundtable’s “Executive Compensation: Principles and Co mmentary” published in November 2003; and the NACD’s “Blu e Ribbon Commission on Executive Compensation and the Role of the Compensation Committee,” published in December 2003. -2-

11 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Interspersed throughout this report are recommendations and guidelines for improving both the governance and design of executive remuner ation policies, processes, and practices. These Introduction, are designa ted as R-1, R-2, etc. Some recommendations, summarized at the end of this of our recommendations are specific prescriptions for designing efficient remuneration plans. Others are better thought of as “guiding principles ” that can be applied broadly across and within organizations. We have not attempted to design an optimal remuneration policy since such a policy must be specific to each organization taking in to account its idiosyncrasies and the sp ecific competitive and organizational strategies, culture and the laws and regulatory conditions it must deal with. We alert the reader to the fact that an appropriate remunerati on policy for a par ticular organization must take account of the tradeoffs that are inevitably involved to achieve balance and fit with their own organization and people. There is no cookbook solution for remuneration in all organizations. And while a well-constructed remunerati on program will require a t horough understanding of the general issues and guidelines offered here, we emphasize up front that with remuneration, it is the details that matter. Well-designed general principles can be thoroughly undone by the details of implementation. Finally, wh ile simplicity in rem uneration programs is important because they must be understandable to the people they cover; simplicity can be a danger because such programs can easily have dramatic unintended consequences as evidenced by the recent experiences of many organizations. Overview Section I introduces a conceptual framework for analyzing remuneration and incentives in organizations. We begin by defining (and justifying) the ob as maximizing l ong- jective of the firm fine the purpose of attracting, retaining, and run total firm value. Next, we de remuneration as then lay out the th ree critical dimensions of motivating executives (and other employees). We emuneration, the composition of the remuner ation remuneration — the expected total level of r package, and the relation between pay and performance. Finally, we analyze the agency problems between managers and shareholders and between board members and share holders, and discuss how well designed pay packages can mitigate the former wh ile w ell-designed c orporate governance policies and processes can mitigate the latter. We say ecause there are no solutions “mitigate” b that will eliminate these agency problems completely. We conclude that corporate governance and remuneration policies are highly inte rrelated: bad governance can easily lead to value-dest roying pay practices, a nd our discussion of remuner ation problems and their solutions will therefore include analyses of corporate governance as well as pay desi gn. S imilarly, b ecause op timal remuneration policies cannot be designed and managed without clear understanding and -3-

12 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION ons between the financial markets and the firm, consideration of the powerful relations and interacti its top-level executives and the board, we devote significant space to these factors. Section II offers a brief history of executive remuneration from 1970 to the present. We show that fundamental changes in the global economy ha ve led to dramatic increases in US CEO pay over the past three decades, driven by an explosion in grants of share options. In add ition, we document that CEO openings in the US are increasingly filled through external hires rather than through internal promotions. Section III examines and explains the forces behind the US-led escalation in share options. The increased emphasis on equity-based incentives may w ell have been initiated by an increased focus on shareholder value creation beginning in the mid-1980s. However, this emphasis does not explain why grant-date values of options have varied systematically with movements in US stock markets, and why option grants have increasingly been extended to lower levels in the corporate hierarchy. We argue that boards and managers falsely per ceive stock options to be inexpensive because of accounting and cash-flow considerations and, as a result, too many options have been awarded to too many people. Section IV defines and discusses the agency costs of overvalued equity as the source of ally realize they ca nnot generate the recent corporate scandals. Managers in overvalued firms eventu performance necessary to support their sky-high stock price. So, they use the firm’s overvalued equity as currency to make acquisitions to satisfy growth expectations. They use access to cheap capital to engage in excessive internal spending. They make increasingly aggressive accounting and operating decisions that shift future revenues to the present and current expenses to the future. Eventually when these fail to resolve the issues, managers — under incredible pressure to preserve high stock prices —turn to further manipulation and even fraud. The result of all these actions is to destroy part of the core value of the firm. The agency problems associated with overvalued equity are aggravated when managers have all to its large holdings of stock or options that will expire worthless if the stock price is allowed to f true value. Since equity-based pay makes the agency problem of overvalued equity worse not b etter, and since the market for corporate control cannot solve it, the agency problems of overvalued equity must be resolved by corporate governance systems. Few governance systems were strong enough to solve the problems, and the results are continuing to show in the worldwide business press and in the courts. As the overvalued equity problem illustrates, wh ile remuner ation can be a solution to agency problems, it can also be a source of agency problems. Section V discusses several widespread -4-

13 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R governance and problems with pay processes and practices, and suggests changes in both corporate pay design to mitigate such problems. In particular, we discuss problems with the appointment and pay-setting process, problems with equity-based pay plans, and problems with the design of traditional bonus plans. We show how traditional plans encourage managers to ignore the cost of capital, manage earnings in ways that destroy value, and take actions to d eceive investors and capital markets. Section VI begins an analysis of a subtle but important issue in governance and str ategy. In it we define and analyze a new concept: what we call the Strategic Value Accountability issue. This is rategy formulation, and choice, and the value the accountability for making the link between st consequences of those choices — basically the link between internal managers and external capital accountability, its assignmen markets. The critical importance of this t, and its implications for performance measurement and remuneration has l ong been unrecognized and therefore ignored in most organizations. ial organization speak very different languages The capital markets and the internal manager and the result is that the two groups virtually ignore each other. The responsibility for managing the capital markets that speak the tension between the internal management of an orga nization and the language of financial results and value creation lies with the CEO and the Board. In every organization it is necessary that someone or some entity take accountability for making the link between strategy formulation and choice and the value consequences of those choices. The individual or entity that accepts this Strategic Value Accountability is vested with the obligation to bridge the gap between strategic choices and the predicted value consequences of those choices. Organizations systematically try to avoid having to deal with the difficult linkages value consequences as reflected in the capital between strategic and operational choices and their market reaction. This is a critical and common mistak e and one that organizations must struggle to correct. The value consequences of organizational choices are difficult to predict, but such predictions must be made. If they are made accurately both the organizati on and society are much better off. Solving this issue also goes a long way to resolving the c onflict between the internal management structure and the capital markets, and with great benefit. LBOs, MBOs, venture capital izations have done an esp ecially good job at resolving these issues and and private equity organ public corporations can learn much from their approach. Management teams receive substantial re wards and pen alties t hrough the cap ital markets as well as through their direct remuneration from the company. In Section VII, we analyze the complex relationships between managers, analysts, and the cap ital market, examine the incentives firms have to manage earnings to meet or beat analyst forecasts, and show how managers playing the earnings- -5-

14 E R EMUNERATION Jensen, Murphy and Wruck XECUTIVE management game systematically erode the integrity of the organization and dest roy organizational value. We highlight the puzzling equilibrium in this market that seems to s uggest collusion between analysts and managers at the expense of investors — an area that is ripe for further research. Finally, Section VIII offers a brief set of conclusions that are intended to comp lement the summary of our recommendations that follows dir ectly below. We resist the temptation in an already long report to summarize or justify each recommendation or guiding principle here. Instead we list them here and provide page references where they appear in the document along with a brief comment where warranted. -6-

15 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION List of Recommendations and Guiding Principles R-1. ace enlightened value maximization/enlightened stakeholder theory Companies should embr in which “creating firm value” is not one of many objectives, but the firm’s sole or scorecard. And this governing ob jective must governing objective: the score on their be complemented by a statement of corporate vision and strategy that guides and motivates the organization in creating value. . ...18 Properly understood enlightened value creation makes use of much of what is generally called stakeholder theory, bu t insists on long-term value creation as the firm’s governing objective. This resolves the indeterminacy of stakeholder theory and its inability to provide any principled basis for decision making or evaluation of success or failure of the firm or management. In this sense it is identical to enlightened stakeholder theory. See Jensen (2001b) R-2. Remuneration committees should develop a “remuneration philosophy” that ref lects and is consistently faithful to the governing objective, and the corporate vision and strategy. ... ...18 R-3. Employment contracts for CEOs and top managers should be discouraged and when they do exist they should not provide for compens ation when a manager is terminated for incompetence or cause. ...29 R-4. The cost to the corporation of granting an option to an employee is the opportunity cost the firm gives up by not selling the option in the market, and that cost should be recognized in the firm’s accoun ting statements as nse. ...38 an expe When a company grants an option to an employee, it bears an economic cost equal to what an outside investor would pay for the option. ppropriate With a downward adjustments for early exercise and forfeiture (and ignoring potentially valuable inside information held by executives), the Black and Scholes (1973) formula yields a reasonable estimate of the company’s cost of granting an option to an employee. R-5. Remuneration committees should carefully re -examine fixed-share and fixed-option grant programs, fully understand the cost and incentive implications of fixed-share/option vs. fixed-value plans, and communicate to share recipients the value (as measured by their opportunity cost to the firm) of the grants they receive and not just the number of options...40 shares or Given the seemingly widespread ignorance of the value of options and their cost to the firm (witness the economically empty, but continuing, claim that there should be no accounting based charge to earnings for such awards) one can see how a compensation committee could be led to a fixed-share scheme rather than fixed-value option award p lan. Suppose we aw ard the CEO 100,000 options this year, the stock price doub les, and we award him 50,000 options the following year to keep the cost the same. How do we deal with the claim that we are penalizing him for success? Indeed, dealing with this conflict is almost impossible if we refuse to calculate th e grant-date dollar value of the option award and continue to argue that there is no cost to the firm of option awards. -7-

16 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION High equity-based compens quires increased monitoring by the R-6. ation for management re committees of re porting policies and the board and remuneration and audit company’s relations with the capital markets in general. ...48 Because incentives are greater in the presence of high equity-based compensation (both to increase value and to avoid destruction of value), boards must understand that additional monitoring is likely to be required. Because of the increased benefits of manipulating financial reports and/or operating decisions to pump up the stock and th erefore generate larger payoffs in the short term, remuner ation and audit committees must increase their monitoring. In addition, they should pay careful attention to ensuring that their managers cannot benefit from short-term increases in stock prices that are achieved at the expense of long-term value destruction. R-7. If our company’s stock pr ice starts to become overvalued we must resist the temptation to enjoy and encourage it. We must make sure that we are communicating to the markets the information regarding the firm’s curre nt and long run health and pros pects . ...49 Management and the board should not be in the business of telling the markets what value is. That is for the markets and the analysts to d etermine. Management must be accountable for infor ming markets on the firm’s strategy and its progress (or lack of it) on executing it. Managers must work to make their organizations far more transparent to investors and to the markets. Companies should state their strategies clearly, identify the relevant value drivers and report auditable metrics on their progress in executing the strategy. This reporting should address that part of the firms share price not directly linked to opportunities observable cash flows through a clear description of the growth they foresee and a willingness to tell the markets when they perceive their stock price is overvalued. Audit co mmittees and boards should establish regular communication with substantial s hort R-8. any’s st ock...49 sellers of the comp ny’s Those who have bet their own money on the future decline of a compa stock are potentially valuable s ources of information regarding poten tial overvaluation of our firm. Therefore the board and particularly the audit committee should be very interested in hearing the logic behind short s ellers actions. There may be good reasons why the board and committee choose to uating it (for example, some short s ellers ignore such information after eval might have an interest in disseminating incorrect information so as to profit in the short run), but it would be foolhardy not to be informed of the views of short sellers of our securities. This will require a substantial change of attitude in most boards and management teams where short sellers are commonly t hought of as the “enemy”. R-9. Business edu cators, while teaching students the desira bility of maximizing value, must also teach them about the dangers of overvalu ation. Max imizing firm value does not mean ...49 maximizing the price of the stock. -8-

17 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Rem take full control of the r emuneration process, policies, R-10. uneration committees must and practices. ...51 should jealously gua In particular remuneration committees initiation rd their rights over executive remuneration. Th ey must abandon the role of simply ratifying management’s remuneration initiatives. Obviously guarding their initiation rights does not mean that committees should make decisions and recommendations to the whole board without discussions with management, but this is quite different de facto seize the from allowing management to remuneration initiation rights. Remuneration committees can ask for data or information from corporate human resource officers, but these officers should report directly to the committee (and not to top management) for committee- related assignments. Similarly, compensation consultants should be hired by, and report directly to, the remuneration committee and not to management. R-11. Firms should resolutely refuse as a matter of policy to pay the fees for the contr acting agents negotiating for the CE O or other t op-manag ers. ...52 Such reimbursements would appear to be a violation of the boa rd’s fiduciary responsibility to the firm, and have clearly undesirable incentive effects on ss and time managers’ decisions to hire such agents and for the aggressivene such agent’s spend in the negotiation process. R-12. ation committees should employ their own professional contracting agents when Remuner hiring new top-level managers...53 It is especially important for the committee to do so when the manager being ent. The conflicts of interests in such recruited has hired his or her own ag negotiations are high with current managers and even current board members (who quite reasonably wish to bring a new person on board in a climate of cooperation and good will). Therefore, bringing in an outsider who answers handle much of the details of the solely to the remuneration committee to negotiation can help put balance back into such negotiations. Moreover, boards should be wary of announcing the new appointment before the terms of engagement have been agreed upon. R-13. Change the structural, social and psychological environment of the board so that the directors (even those who fulfill the requirements of independence) no longer see themselves as effectively the employees of the CEO ...54 We frame the guideline this way not because it is the cause of the problem, but rather because it is a highly productive frame from which to view the symptoms and causes of fundamental problems with governance. Changing this mindset cial, will not be an easy task. It will require major changes in the so psychological, and power structures in boards. And when it is accomplished boards will no longer see their role as one of primarily supporting the CEO rather than monitoring the CEO as is so common in the American model. The support role is clearly important but mu st be strictly s ubordinate to the boa rd’s role as monitor. Consider the following: the CEO does most of the recru iting for the board and extends the offer to join the board. And, except in unusual circumstances, board members serve at the pleasure of the CEO. The CEO generally sets the agenda for the board. Moreover, it is rare that the board meets outside of the CEO’s presence or without his explicit permission. Fin ally, virtually all information board members receive from the company originates from or passes through the CEO, except in highly controlled or unusual circumstances. Changes in these practices will require a major change in the power relationship between the board and the CEO. -9-

18 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R The board should be chaired by a person who is not the CEO, was not the CEO, and will R-14. not be the CEO ...54 un the process that ev The critical job of the Chair is to r aluates, compensates, hires and fires the CEO and top management team. In general this job ca nnot be performed adequately by the CEO, the past CEO, or the future CEO. CEOs sitting on the boa rd ...55 R-15. Limit the number of outside board members for many obvious reasons. Outside CEOs offer advantages as What generally has gone unaddressed are the disadvantages they bring to the board. It is natural for them to subconsciously (if not consciously) view the board through CEO eyes — a lens where the power of the CEO is not seriously t of serious problems such as obvious challenged, except perhaps in the even incompetence or malfeasance. R-16. The CEO s hould be the only member of the management team with board membership. ...55 ... While members of the management team can add value by participating in members. board discussions there is little reason to have them be formal voting When other members of the management team are voting members of the board we increase the likelihood that the board will consider its job to be that of supporting, not monitoring, the CEO. Members of management that can add value to board discussions can and should do so by being at the meetings regularly as ex-officio members. ration committees should seldom, if ever, use compensation consultants for R-17. Remune executive remuneration purposes who are also used by the firm for actuarial or lower- level employee remuneration assignments...56 Conflicts between these dual roles of compensation consultants dramatically disadvantage the remuneration committ ee and the firm and facilitate more- generous executive pay packages. Consider the situation of a consultant who hopes to close a multi-million dollar actuarial or lower-level employee engagement. The same consultant engaged as an advisor on CEO and top- manager remuneration policies (that might amount to only a high five-figure or low six-figure fee) would be put at a significant disadvantage in reco mmending value-creating remuneration policies inconsistent with what the CEO desires. The reasons for avoiding these conflicting roles are essentially the same as the rules that are emerging that limit the use of a firm’s auditor as a consultant. R-18. Managers should be required to have “skin in the game” by purchasing stock or options or by explicitly and deliberately accepting reductions in other forms of compensation. ...59 ... Important advantages to requiring managers to have skin in the game is that it encourages them to recognize the opportunity cost of capital to the company and to reveal the private information and beliefs they have about the value- creation potential in their strategic plan. If managers are not willing to bet their own money on the plan, it is probably not a good bet for the shareholders either. R-19. Executive share option contracts should, whenever possible, adjust the exercise price of the option for any dividends or return of cap ital paid to holders of the sh ares . ...61 This removes any artificial incentives that manager have to withhold dividends when they have options. -10-

19 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Remuneration committees should give careful consideration to issuing executive share R-20. ny’s cost of cap ital (less the options with exercise prices that increase with the compa ...61 dividend yield if the option is not dividend adjusted). Consider an example where the cost of capital is 10% net of the dividend yield, the current stock price is $10 and the exercise price of the option is $10. Such options would pay managers nothing if the stock price failed to rise over any period by an amount greater than the cost of capital less the dividend y ield. This means managers earn nothing on their options unless shareholders do better than breakeven. Since cost of capital indexed options are less valuable firms can award more of them to managers for the same cost to the firm and thereby create more high-powered incentives for the same cost. R-21. Ma nagers should receive annual statements that clearly summarize in one place the changes in their wealth in the prior year from all sources of remuner ation from the firm (including changes in the present value of future retirement and deferred compensation). ...62 Because managers often do not know the sources and amounts of their total compensation we advocate giving them annual s tatements detailing the changes in their wealth in the prior year due to the grant date value of options received during the year, and changes in the present value of their holdings of options, mpany. If the stock shares, other bonuses, and retirement be nefits from the co price in our example rises to $11 over the first year it is exactly equal to the new exercise price, and the exercise tion is still ze ro. Thus, managers value of the op would be taught in a graphic way that the cost of equity capital is not zero. In cases where the cost to the company of emoluments can be calculated these should also be included in the report. Such accounting will be helpful both to the managers and to the remuneration committee that is managing the process. Remuneration committees should give serious consideration to offering executives the R-22. opportunity to bet on their strategy along with shareholders by offering to sell them in-the-money cost-of-capital indexed share options at the nominal price equal to the difference between market and exercise prices at date of issue. ...65 Selling executives cost-of-capital indexed options causes executives to have skin in the game, motivates th em to understand that the cost of equity capital is not zero (or the dividend yield for divide nd-paying stocks), motivates executives to self select in or out of the firm based on their private information and beliefs firm, guarantees to shareholders that about their ability to create value in the managers’ options pay off only when shareholders do better than breakeven at the cost of capital, and solves the option re-pricing problem when options are far out of the money. R-23. Remuner ation committees should give conscious consideration to the tradeoffs associated with allowing an executive to unwind incentives, and the timing and am ounts of such unwinding to be allowed. ...67 Remuneration committees should include explicit unwinding constraints (or required permissions) in executive incentive awards. They should monitor the portfolio holdings of top-level executives and related parties to ensure that they are not inappropriately unwinding the incentives that have been put in place b y the committee and the board and paid for by the company. -11-

20 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Remuneration committees should require pre-trading disclosure of intention to trade for R-24. all insiders as a condition of employment. ...68 ill not Firms should adopt these policies as a way to guarantee that executives w use insider information to disadvantage the shareholders to which they owe a fiduciary duty. And it is important for all effective sales to be treated this way (for example, an executive’s repayment of a company loan by transfer of any options, stock or debt security to the company). R-25. employment, prohibit t op-level Remuneration committees s hould, as a condition of executives from trading in derivatives of any kind, but especially those related to the the fi rm. ...68 securities of These constraints are required to enforce the unwinding constraints established in R-23 and insider trading constraints in R-24. Design bonus plans with “linear” pay-performance relations...75 R-26. are linear over a broad range, with Better-designed pay-performance relations very high (or non-existent) caps, and “ bonus banks” that allow bonuses to be negative as well as positive. Bonus banks can be created in a number of ways unpaid including, for example, paying a bonus out over three years, where the bonus is available to make up some or all of a negative bonus in the c urrent year. See Stewart (1990). ...76 R-27. Avoid intern ally influenced performance standards Internally influenced performance standards are those where the bonus-plan participants can take actions (often value destroying) that increase bonuses b y reducing the standard rather than by improving performance. Do not measure performance anywhere in an organization with ratios. Simply put: If it is R-28. a performance measure and a ratio, it’s wrong ...78 performance can often be made to work by a Ratio measures of ppropriately changing the decision rights of the agent, but this is almost never done. For example, return on assets can be made to work if the agent is given only the right to decide which assets to use, not the quantity of assets. R-29. Use “line-of-sight” performance measures when possible and give each employee the decision rights to do their job...78 This prescription actually involves several dimensions. To provide incentives, employees must be able to affect the performance measure, and also must understand they can aff ect the pe rformance measure,. To be able to affect how the performance measure they must have the appropriate decision rights to do so. R-30. Use pe rformance measures that reduce compensation risk while maintaining incentives. ... ...78 Since employees “charge” to bear compensation risk, performance measures that reduce risk without reducing incentives increase efficiency and company profits. However, when risky compensation is an add-on to c urrent compensation there is no need to further compensate managers for that added risk. -12-

21 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Pay par R-31. ticular attention to the choice of group versus individual performance measures. ... ...79 ncies in productivity between the actions When there are substantial interdepende of two or more people or groups, defi measure ne the extent of the performance to incorporate the interdependencies. Using individual performance measures in situations where cooperation is important will create conflict, lack of cooperation and reduced performance. And the same principle applies to the choice of the time interval over which performance is measured. R-32. Bonus plans should include a subjective co mponent ...79 There are often no low-risk objective measures of the individual’s contribution to firm value. The objective measures that exist often are too risky (i.e., based on factors that are highly variable), provide insufficient direction (the employee might not know how to affect the measure), or provide incentives to do the wrong thing. However, even when no appropriate objective measures are available, an individual’s contribution to value can often be assessed subjectively by supervisors or managers. Subjective assessments can also be used to reduce the noise in good objective measures, to reduce the “distortion” in bad objective measures, and can also adjust bonus payouts for unan ticipated shocks (such as terrorist attacks or shocks to world oil prices). R-33. Every bonus system including option and other equity-based programs should provide for recovery of rewards (clawback) if and when there is future revision of cr itical indicators on which bonus payments were based or received. ...79 This clawback should include, but not be restricted to, amounts due because of formal restatements of accounting numbers such as earnings or revenues. Moreover, provisions should be made whenever possible to recover the am ounts or retirement benefits when it is from bonus banks, deferred payments ectly. In the absen ce of these clawback impossible to recover the amounts dir provisions we are unintentionally rewarding and therefore providing incentives for people to lie and game the system Encourage managers to build and preserve trust. Because precedent matters we must R-34. ...80 beware of too much tinkering with the system. Hold managers accoun table for the long-run eff ects of their performance evaluations. Encourage them to pay particular attention to the destruction of trust, and the perceived insecur ity of contracts, pro mises, and commitments regarding bonuses and performance measurement when the rules of the game are changed too often by "too much tinkering” with the system. R-35. Mana gers should be held accountable for factors that are beyond their control if they can control or affect the impact of those uncontrollable factors on performance. ...81 For example, we would be foolish to ignore the affect of ga soline prices on the performance of a manager of a fleet of vehicles. We want that manager to be cognizant of what will happen to the cost of r unning the f leet if the price of gasoline goes up or down substantially, and there are certainly things such a manager can do to help the firm prepa re for and to adjust efficiently to major changes in gas prices. Holding him accountable for the effects of changes in gas prices will motivate him to be creative in managing the impact of those uncertain changes. -13-

22 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION The remuner seeing that Strategic Value R-36. ation committee should take the lead in hose who have the unique combination of Accountability is clearly assigned to t business judgment, financial knowledge, wisdom, and willingness to take on the critical task of managing the interface between the operating organization and the ...86 capital markets so as to create value. Let us be clear that the assignment of the decision rights for managing relations with the capital markets is much more than simply talking to investors and institutions to assess their interests, opinions, desires and advice. It goes to the core of what it means to direct the organization so that choices are made that will maximize the chance of competitive success and the efficient use of society’s scarce resources (human, capital, technological and material) entrusted to the organization. Remuneration committees must confront these issues. The committee must see to it that this talent and capacity is recruited into the organization and retained. They must see to it that those who have accepted the Strategic Value Accountability task are held to the value consequences even when they turn out poorly. R-37. Firms must restart the c onversation between corporate managers and Wall Street by “just saying no” to the old game of earnings management and earni ngs guida nce. ...97 This will not be easy. However, eliminating or reducing the influence of these corrupting forces on the firm will be an important step in bolstering the integrity of corporations. There is a window of opportunity now that analysts and the financial institutions that employ them have fallen into disrepute. It is and to estimate their implications for the analyst’s job to forecast earnings value. People are highly aware of the malaise that has gripped the business world. Executives are wondering how to invest in the integrity of their companies. Researchers are starting to examine some of the issues. But this window won’t remain open forever and if we don’t seize this moment to learn from it, and change the system we identify the problem, talk about it, and could find ourselves trapped once again in a vicious, destructive cycle. And let’s ement game (as Coca Cola, Gillette and be clear, ending the earnings manag not mean ending communications with USA Neworks, and others have), does analysts and the capital markets. R-38. Senior managers must communicate with the cap ital markets. They must understand what drives value in their organization and align internal goals with those drivers, not with analysts’ expectati ons. ...97 To limit wishful thinking, managers should reconcile their company’s projections to industry and rivals’ projections. When the company’s expectations lie outside what is widely v iewed as the industry’s growth rate, managers should explain how and why they w ill be able to outpe rform their market. Some will a rgue that making th is all clear to the analysts w ill reveal valuable information to their competitors. “ To this, we have a simple response: If your strategy is based on your competitor not knowing what you are doing as opposed to not being able to do what you can do, you cannot be su ccessful in Fuller and Jensen (2002)) the long run no matter who knows what.” ( -14-

23 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION The Conceptual Foundations of Executive Remuneration I. 4 The Governing Objective of the Corporation 5 jective, The corporate objective function, or more simply its governing ob plays a critical role are, and the accountability of managers and in corporate productivity and efficiency, social welf directors. Yet there is much misunderstanding and confusion about whether there s hould be such a single governing objective and if so, wh at it should be. We can provide an immediate answer to the first of these issues: since it is logically impossible to maximize in more than one d imension, purposeful behavior requires a jective. As someone once said, single-dimensional governing ob “multiple objectives is no objective”. A random sampling of annual reports will pred ictably reveal a variety of s tated company objectives, including maximizing shareholder value, increasing customer and employee satisfaction, building the highest-possible quality products, and furthering charitable ties to the local community. However, two-hundred years of research in economics and finance have produced the result that if our objective is to maximize the efficiency ources (that is to avoid with which society utilizes its res waste and to maximize the size of the pie), then the proper and unique ob jective for each company 6 in the society is to maximize the long-run total value of the firm. Firm value will not be maximized, fore, consistent of course, with unhappy customers and employees or with poor products. There 7 with “stakeholder theory” ned about relations with all their value-maximizing firms will be concer constituencies. A firm cannot maximize value if it ignores the interest of its stakeholders. 4 on Jensen, 2001b, "Value Maximization, Stake holder Theory, and the This section draws heavily Corporate Objective Function", European Financial Management Review, V. 7, No. 3: pp. 297-317 (av ailable from the Social Science Research Network eLibrary at: http://papers.ssrn.com/Abstract=220671 ), and the reader will find extensive discussion of these complex issues there. 5 1994, The Value We borrow the “governing objective” language from McTaggart, Kontes and Mankins, Imperative: Managing for Superior Shareholder Returns , New York: The Free Press. 6 This conclusion holds under the assumption that the government effectively blocks private monopolies from exercising their monopoly pricing power, that it sets the rules of the game so as to cause each company and individual to internalize the costs and benefits of the physical (not value) eff actions on others, and that ects of their all goods are priced. See Jensen, "Value Maximization, Stakeholder Theory, and the Corporate Objective Fun ction", for an extensive discussion of these issues and others associated with stakeholder theory and value maximization. 7 Stakeholder theory, argues that managers should make decisions so as to take account of the interests of all stakeholders in a firm (including not only financial claimants, but also employees, customers, communities, governmental officials, and others). -15-

24 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R But unlike “stakeholder theory”, value maximization gives boards and managers a principled way to think about and to make the tradeoffs that must be made between various cons tituencies. Because advocates of stakeholder the ory refuse to specify how to make the necessary tradeoffs among these competing interests, they ory that makes it impossible to leave managers with a the make purposeful decisions. for example, says that customer and employee Value maximization, eased to the point where further increases in satisfaction and product quality should only be incr ple that we want a firm to expend a dollar’s each would reduce firm value. This means for exam worth of resources to generate benefits for any constituency up to the point where that constituency values those benefits at one dollar or more. Beyond that we are wasting both the firm’s and society’s resources. Social welfare maximization thus implies that creating value should not be simply one of many competing corporate objectives, but the preeminent or governing objective. than acceptance of value max imization as the organizational Creating value takes more objective. As a statement of corporate purpose or vision, value maximization is not likely to tap into the energy and enthusiasm of employees and managers to create value. Seen in this light, change in scorecard that managers, directors, and others use to assess long-term market value becomes the success or failure of the organization. Since we never know when something has been max imized it is better to think of value creation rather than value maximization. Choosing value creation as the ategy and tactics that unite corporate scorecard must be comp lemented by a corporate visi on, str 8 participants in the organization in its struggle for dominance in its competitive arena. ement to decide on a clearly One way to see the how important it is for the board and manag rning objective is that without it there can be no principled defined single-dimensional gove evaluation or monitoring of management performance. Ther efore, it is the precursor to most all the critical activities of the remuneration committee. The governing objective for a corporation is like the provide a score “score” for a sporting event. One thing that is critical for a scorecard is that it must so that one can distinguish w ho won. Stakeholder theory provides no score, and therefore there is no way within it to tell whether the firm is better or worse off. And without a score, there is no principled way to hold management accountable for its pe rformance as steward of the firm’s resources. Without a single-dimensional governing objective managers are either left unmonitored ss in which the evaluator can change the weights or are subject to the vagaries of an evaluation proce on a set of agreed upon critical dimensions to arrive at any score he or she pleases. 8 In his recent article, David Kay, 2004, "Forget How the Crow Flies", Financial Times, January 18, pp. W1, W2 provides an excellent discussion of obliquity, the paradoxical phenomenon in which “goals are more likely to be achieved when pursued indirectly” — another reason why the governing objective must not be inte rpreted as an organization’s vision, purpose or strategy. -16-

25 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R The Balanced Scorecard Gives No Score Since the Balanced Scorecard (Kap ial equivalent of lan and Norton (1996)) is the manager stakeholder theory, the same conclusions hold. Balanced Scorecard theory is flawed because it presents managers with a scorecard that gives no score — that is, no single-valued measure of how they have performed. Thus managers evaluated with such a system (which can easily have two dozen measures and provides no information on the tradeoffs between them) have no way to make principled or purposeful decisions, and the evaluators have no way to make principled evalu ations. (single dimensional) score for measuring pe The solution is to define a true for the rformance organization or division (and the organization’s strategy must be consistent with it). Given this we can then encourage managers to use measures of the drivers of performance in the balanced scorecard to understand better how to maximize their score. And as long as their score is defined properly, (and for lower levels in the organization it need not be and will generally not be value) this will enhance their contribution to the firm. Enlightened Value Maximization and Enlightened Stakeholder Theory Because value maximization has gotten a bad name in many circles and because stakeholder theory has suffered similarly we offer a solution. We call the solution enlightened value , and it is identical to what we call maximization . Enlightened value enlightened stakeholder theory maximization utilizes much of the structure of stakeholder theory but accepts maximization of the long run value of the firm as the criterion for making the requisite tradeoffs among its stakeholders. from enlightened Managers, directors, strategists, and management scientists can benefit imization or value stakeholder theory. Enlightened stakeholder theory specifies long-term value max jective and there seeking as the firm’s governing ob fore solves the problems that arise from the multiple objectives that accompany traditional stakeholder theory. Firm Value Maximization Does Not Imply Maximization of Short Run Stock Price Firm value is not technically the s ame as shareholder value, because “firm value” also includes the values to all other fina ncial claimants such as creditors, debt holders and preferred shareholders. Because shareholders are the residual claimants of the firm, we often call shareholders the “owners” and can loosely speak of the objective of the company as creating long- run shareholder value. But it is w ell to be aware that it is possible for manag ement and a board to make decisions that decrease total firm value and total social value while increasing shareholder value. This clearly undesirable result can happen, for example, as long as the changes transfer enough value from creditors and debt holders to shareholders to more than offset the decline in firm value. -17-

26 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Throughout this report, we will assume that the appropriate objective of the firm is to maximize total long-run firm value, and that well -designed compensation plans will encourage actions that destroy value. We will managers to take actions that increase this value while avoiding not assume that “creating long-run shareholder value” is synonymous with “maximizing company 9 share prices in the short run.” If stock markets are efficient in the strong-form sense that all knowable information is immed iately im pounded into share prices, then any change in long-run shareholder value will, indeed, be immediately ref corresponding change in the share lected into a price. However, top executives will routinely and inevitably possess i nformation not av ailable to investors. In these situations, changes in short-run share prices will not imply a similar change in long-run shareholder value. In addition, we do not assume the absence of noise traders (those who trade without information and create noise in the system). Indeed, as discussed at length in Section IV below, we b elieve that many of the corporate scandals over the last two years were driven, in large part, by executives desperately trying to justify or increase short-run stock prices at the expense of long-run value creation. In addition, the f act that non-public information executives and board members w ill inev itably be in possession of valuable means that the board and especially the remuneration committee must be especially sensitive to ensuring that those who possess it do not use it to the disadvantage of the shareholders and debt holders to which they owe fiduciary duty. We discuss these issues in more length below. R-1. Companies should embrace enlightened value maximization/enlightened stakeholder theory in which “creating firm value” is not one of many objectives, but the firm’s sole or governing objective: the score on their scorecard. And a statement of corporate vision and strategy that guides and motivates the organization in creating value must complement this governing objective. Properly understood enlightened value creation makes use of much of what is generally called stakeholder theory, but insists on long-term value creation as the firm’s governing objective. This resolves the indeterminacy of stakeholder theory and its inability to provide any princip led basis for decision making or evaluation of success or failure of the firm or management. In this sense it is identical to enlightened stakeholder theory. See Jensen (2001b) R-2. Remuneration committees should develop a “remuneration philosophy” that reflects and is consistently faithful to the governing objective, and the corporate vision and strategy. 9 Following Jensen, 1969, "Risk, the Pricing of Capital Assets, and the Evaluation of Investment Portfolios", Journal of Business, V. 42, No. 2: pp. 167-247 financial economists categorize different degrees of market eff iciency as “weak form” (stock prices reflect all information in past stock prices), “semi-strong form” (stock prices ref lect all publicly available information), and “strong form” (stock prices reflect all publicly and priv ately available information). -18-

27 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION The Economics of Remuneration A well-designed remuneration package for executives (or for employees at all levels of the organization), will accomplish three thi ngs: attract the right executives at the lowest cost; retain the right executives at the lowest cost (and encourage the right executives to leave the firm at the appropriate time); and motivate executives to take actions that create l ong-run shareholder value and avoid actions that destroy value. There are three critical dimensions in the design of any remuneration policy that is to accomplish these objectives: 1. The expected total benefits associated with the job or position (including the costs and benefits of non-pecuniary aspects of the job). This dimension determines where someone works. Every person will be in the job for which the expected total benefits associated with the job is highest (taking into account the risk of the benefits and the cost of switching employers). We say because expected some of the costs or benefits of the job will be uncertain (from the perspective of the employee) and it is therefore the risk-adjusted expected value that is r elevant. Therefore, it is the expected total benefits that determine whether we attract and retain the right executives and encourage the right ones to leave. 2. The composition of the remuneration package. This dimension i nvolves the determination of the individual elements of the remuneration package (including for ex ample the amount of cas h received as s alary, the amount of risky performance-related cash, stock, restricted stock or options, retirement benefits, non-pecuniary benefits such as prestige, the emolument package, and so on). It is in the interest of both the employer and employee to structure the composition of the package so that it is efficient in the sense that for any given total cost to the company the benefits to the employee are maximized. Or for any given total benefits to the individual the total cost to the company is minimized. Thus the c orrect remuneration package w or optimal composition of the ill be efficient in the sense that no resources are wasted. It’s a little m ore complicated for aspects of the working the productivity of the employee, but the environment or pay package that affect 10 substance does not change. 3. The relation between pay and performance (what for shorthand we call the pay-performance relation). This dimension defines which actions and results are rewarded and which are penalized, and therefore determines what an employee works on, how hard the employee w orks, and the employee’s productivity. Note that our discussion of the pay-performance relation makes no distinction between for what w ill be reward and motivation. If what we reward (ex post) in this period sets a precedent 10 Taking into account the productivity effects of any item personally valued by the employee means we define the cost to the employer of an increment in such a bene fit as the full incremental cost minus the value of any pos itive productivity effects (or plus the value of any re ductions in produc tivity). For example, offering med ical coverage is costly but also provides benefits to the firm (by facilitating h ealthier employees); the value of these benefits to the firm should be subtracted from the cost to derive the n et cost. To the extent that lib eral health benefits attract less healthy employees to the firm those costs would be added to derive the net full cost. -19-

28 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION rewarded in future periods in the mind of the executive, then the reward this period signals something about the pay-performance relation for future periods and therefore affects motiv ation or for what will incentives. If what we reward this period has no relation in the mind of the executive purely random be rewarded in the future, then it can have no affect on incentives or motivation. A reward system would do this. The three dimensions of executive pay all have to be managed by the remuneration co mmittee, and there are significant policy implications of each dimension. For example, it s hould be clear that there is no conflict between the executive and th e company about the co mposition of the package. On the other hand there is a conflict between the executive and the company over the level of compensation — the committee wants to hire and retain the executive at the smallest possible premium over his or her best alternative employment and the executive wants to be paid only a small amount less than his or her maximum value to the company. To the extent that the pay-performance relation encourages the executive to create more value and given that there can be sharing of this higher value, both want to get the pa y-performance relation right. However, elements of c onflict, negotiation, and gaming can enter the discussion over the pay-performance relation for the executive. And if these are not managed properly they can lead to value destru ction and inefficient remuneration packages. actions between the three Well-designed packages will carefully manage the subtle inter dimensions of remuneration. For example, consider two remuneration packag es that offer the same pective executive — one with a high s expected total benefits to the pros alary and no retirement benefits, and a second with a low salary but generous (unvested) retirement benefits. Wh ile the two ill provide packages provide identical “attraction incentives” to the executive, the latter w stronger “retention incentives” once the executive has accepted the offer. Similarly, a package with a low salary but high bonus opportunity w ill provide better motivation than a package with a high s alary and lower bonus opportunity (assumi ng the bonus is earned by activi ties that result in l ong-run firm-value creation). Moreover, the composition of the package can affect the types of executives gh retirement benefits will appeal to potential the company can attract: the package with the hi executives planning on staying at the firm for an extended time, and the package with the high bonus opportunity will appeal to executives who are less risk-averse, more op timistic, and more confident about their ability to create value. Well-designed packages will also carefully manage the riskiness of the pay package. Bonus opportunities and other performance-contingent pay (such as share options or long-term incentive plans) are inherently volatile and impose undiversifiable risk on plan participants. Wh ile company firm-specific risk, executives tend to be risk averse shareholders are well diversified and neutral to -20-

29 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION and inherently undiversified (with their monetary as well as human capital invested disproportionately in their company) and will often prefer a pay package promising a fixed base salary over a risky package with the same expected value. Risk-averse executives w ill “charge” for bearing risk by discounting the value of the risky elements of pay, or demanding higher expected pay levels. This means that the firm faces a tradeoff between having b etter-motivated employees working hard toward the right outcomes and attracting the right quality executives at the lowest possible cost. This also suggests that the common prescription of “putting more pay at risk” can be misguided when taken alone: while a company indeed wants better-motivated executives, and better motivation almost inevitably involves higher risk, the increased risk by itself increases expected remuneration costs and we want to in cur these expected costs only when the expected productivity benefits exceed the costs. Companies imposing risks on executives through their pay packages must ascertain that the associated incentive benefits exceed the increased expected cost of the package. In add ition, there are ways that a company can structure its r emuneration and employment plans that will cause executives who truly believe they can create large value to self select into the firm and those who do not believe they can create value to self select out of the firm. We discuss these issues below in Section V. Agency Problems and Executive Remuneration If the manager of a firm owned 100 percent of the firm’s shares, then (risk aversion and self control problems aside) the decisions made by that manager would be presumed to be those that maximize long-run shareholder value, and there would be no need for add itional incentive plans. However, decisions in companies are made not by owners but rather by managers who hold far less than 100 percent of the company’s stock. These managers, although hired for their expertise and managerial talent, ca nnot be expected to make the same decisions as the owners would have made themselves. This “agency problem” is esp ecially pre valent for decisions that are personally costly for managers (such as decisions to layoff employees and sell divisions or the entire company) for decisions that because the managers bear a disproportionate share of the cost vs. benefits, and benefit managers (such as buying corporate aircraft or remodeling the corporate hea dquarters) because the managers reap a disproportionate share of the benefits vs. cost (Jensen and Meckling (1976)). The inherent conflict between managers and shareholders is iden tical to the conflict arising whenever a principal hires an agent to take actions on the principal’s behalf. And we should note ooperative that agency problems are a part of all situations in which two or more people engage in c -21-

30 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R activities. They always have some incompatibility of interests and therefore agency problems will always be present — in corporations, partnerships, non-profits, governmental agencies, and even in families. However, there is an added agency problem in public corporations: top executives are hired, monitored, and rewarded not directly by owners but rather by boards of directors who are elected by, but are not perfect agents for, the shareholder-owners. Well-designed executive remuneration packages can mitigate the former type of agency problems by aligning interests of -designed corporate governance policies (including managers and shareholders. Similarly, well director remuneration) can mitigate the latter type of agency problems by defining rules, processes, checks, and balances that help ensure boards of directors faithfully fulfill their fiduciary duties to shareholders. Moreover, since well-designed pay policies cannot resolve all conflicts of interest and agency problems between executives and the firm, well-designed corporate governance systems implemented by directors of high integrity must be in place to resolve those conflicts that cannot be handled by remuneration policies alone. Remuneration decisions are not made by owners but rather by boards of directors (upon recommendation from the remuneration committees). In a ddition, as discussed at length in Section V below, remuneration committees routinely lack the information, expertise and negotiating skills necessary for hard-nosed contract negotiations w ith incumbent and in coming executives. As a result, many pay packages and processes are poorly designed, and therefore pay packages can create as well as reduce agen (or at cy problems in organizations, by attracting the w rong managers too high a cost), retaining the wrong managers, and motivating the wrong behavior. B ecause managers are self interested and because remuneration committee members are spending the firm’s resources, not their own, there is major potential for the participants in the system to behave in ways that will exacerbate, not reduce, agency problems. Thus, corporate governance and remuneration policies are highly inte r-related: bad governance can easily lead to value-destroying pay practices, and many notorious excesses in pay 11 can be traced to poor governance. Our discussion of remuneration problems and their solutions in Section V below will therefore include analyses of corporate governance as well as details of pay design. 11 Bertrand and Mullainathan, 2001, "Are CEOs Rewarded for Luck? The Ones Without Principals Are", Quarterly Journal of Economics, V. 116, No. 3: pp. 901-932 find that CEOs in b etter-governed firms (defined as firms with large shareholders, low CEO tenure, small board sizes, and boards composed with a majority of outside dir ectors) are less likely to be rewarded for “luck” (e.g., CEOs in petroleum firms rewarded for increases in world oil prices) and are more likely to be “charged” for option grants (through reductions in other forms of pay). -22-

31 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION II. A Brief History of Executive Remuneration The Worldwide Economic Environment ecades Worldwide economic, regulatory, and technological changes over the past three d fundamentally altered the global economy Jensen (1993) char acterizes as the and constitute what Modern Industrial Revolution. The beginning of this re volution can be traced to the conglomeration era of the 1960s and the oil-price shocks of 1973 and 1977. The next two decades witnessed the far-reaching effects of technology, declines in regu lation, the defeat of rapid improvements in communism and socialism, growing worldwide capitalism, and globalization of trade which brought billions of laborers earning less than $10 per day into comp etition with workers in the west. These tectonic shifts created massive excess capacity in many of the world’s industries, including automotive, retail trade, steel, tires, textiles, computers, and defense. Corporate managers, loathe to shut down capacity and distribute excess cash back to shareholders, responded by wasting huge 12 amounts of free cash flow through unwarranted diversification and investment programs, which in turn planted the seeds for hostile takeovers, leveraged buyouts, and the use of high-yield debt as both a financing instrument and as a means to force managers to disgorge their excess cash. The late 1980s brought sweeping changes in both US financial markets and global geopolitics. Court decisions and legislation in the US brought the hostile takeover market to a virtual halt. The high-yield debt market was cri ppled by the indictment and subsequent guilty pleas of Michael Milken and Drexel Burnham Lambert and by restrictions on high-yield debt holdings imposed on savings institutions, commercial banks, and insurance firms, and by major punitive changes in the US bankruptcy law that made it uneconomic to reorganize troubled firms outside of bankruptcy. The prospects for worldwide capitalism soared by the end of the decade, marked by the zechoslovakia, Bulgaria, collapse of Soviet-backed regimes in Poland, Hungary, East Germany, C and Romania, the fall of the Berlin Wall, a nd (in 1990) the re unification of Germany and the disintegration of the Soviet Union. The 1990s witnessed the emergence of the Internet and, with it, the rise of the “new economy” and so-called “dot.com” firms. The strong stock market in the latter half of the decade 12 By free cash flow we mean cash flow in excess of that which can be reinvested at returns equal to or better than the cost of capital. See Jensen, 1986, "Agency Costs of Free Cash Flow: Corporate Finance and Takeovers", American Economic Review, V. 76, No. 2: pp. 323-329 (av ailable from th e Social Science Research Network ) eLibrary at: http://ssrn.com/Abstract=99580 -23-

32 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION spurred a boom in initial public offerings, spin-offs, “tracking” shares (shares issued on a division of a larger company), and large mergers financed primarily with equity. Dividend yields plummeted 13 By the early as firms paid lower dividends and substituted share repurchases over dividends. many firms (especially new-economy firms) were 2000s it became apparent that the shares of propped up by questi onable or fraudulent accounting, legal, grossly overvalued, and in many cases brokerage, investment banking and other financial practices. Share prices plummeted, and many large US companies—including Enron, Arthur Andersen, KPMG, Lucen t, WorldCom, Tyco, 14 HealthSouth and Xerox — became embroiled in accounting scandals. Trends in Executive Remuneration The fundamental changes in the global economy have led to (and to some extent have been influenced by) pronounced changes in executive remuneration practices. The pay for chief executive officers (CEOs) in large US firms increased dr amatically over the past three decades, driven by an explosion in grants of share options. Figure 1 shows that average to tal remune ration for CEOs in S&P 500 firms (adjusted for inflation using 2002-constant dollars) in creased from about $850,000 in 1970 to over $14 million in 2000, falling to $9.4 million in 2002. 13 2001, "Disappearing Dividends: Changing Firm Char acteristics or Lowe r Propensity to See Fama and French, Journal of Financial Economics, V. 60, No. 1: pp. 3-43 and DeAngelo, DeAngelo and Skinner, 2002, "Are Pay?" USC Center for Law, Dividends Disappearing? Dividend Concentration and the Consolidation of Earnings", Economics and Organization (CLEO) Working Paper No. 02-9, forthcoming in J onomics , ournal of Financial Ec (available from the Social Science Research Network eLibrary at: http://papers.ssrn.com/Abstract=318562 ). 14 Erickson, Hanlon and Maydew, 2002, "How Much Will Firm s Pay for Earnings that Do Not Exist? Evidence of Taxes Paid on Allegedly Fraudulent Earnings", University of Chicago Working Paper, November 1, 2002, ). Chicago, IL (available from the Social Science Electronic eLibrary at: http://papers.ssrn.com/abstract=347420 -24-

33 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Figure 1 Average Cash and Total Remuneration for CEOs in S&P 500 Firms, 1970-2002 $16,000 $14,000 $12,000 Average CEO Total Remuneration including options valued at grant date $10,000 $8,000 $6,000 Average CEO Cash Remuneration $4,000 (2002 dollars, in 000s) Average CEO Remuneration $2,000 $0 1990 1985 1995 1980 1975 1970 2000 orbes and ExecuComp. CEO total pay Note: S ample is based on all CEOs included in the S&P 500, using d ata from F includes cash pay, restricted stock, payouts from long-term pay programs and the value of stock options granted using ExecuComp’s modified Black-Scholes approach. (Total pay prior to 1978 excludes option grants, realized while total pay between 1978 and 1991 is computed using the amounts from exercising stock options during the year, rather than grant-date values.) k-Scholes value of opti ons soared from near Over this time period, the average grant-date Blac $7.0 million in 2000, falling to $4.4 million in 2002. The difference between zero in 1970 to over the $7.0 million option grant value in 2000 and the $14 million total compens ation is made up of cash compensation, restricted stock, retir payouts from a var iety of long-term ement benefits, and incentive plans. Even base salaries and bonuses ( “cash remuneration”) tripled over this time period. As shown in Figure 1 and emphasized in Fi gure 2, inflati on-adjusted cash remuneration increased from about $850,000 in 1970 to over $2.2 million by 2002. -25-

34 E EMUNERATION Jensen, Murphy and Wruck XECUTIVE R Figure 2 Average Cash Remuneration for CEOs in S&P 500 Firms, 1970-2002 $2,500 Average CEO Cash Remuneration $2,000 $1,500 $1,000 (2002 dollars, in 000s) $500 Averag CEO Remuneration $0 1975 1995 2000 1970 1980 1990 1985 ata from Note: S ample is based on all CEOs included in the S&P 500, using cash remuneration (salary and bonus) d Forbes and ExecuComp. In this section, we chronicle the trends in executive remuneration in the US over the past three decades and show how the trends relate to contemporaneous changes in the economic environment. The 1970s Throughout the 1970s, executive remuneration packages consisted almost entirely of base salaries and bonuses tied to annual performance measures. During this decade, almost half of the cross-sectional variation in cash remuneration in by company size (us ually the US was explained measured by firm revenues), and the highest-paid executives routinely were at the helm of the largest conglomerates and largest s teel, automotive, and oil comp anies. These implicit incentives to increase revenue help e ication and investment programs in the xplain the unproductive diversif 1970s, which in turn contributed to increases in excess cap acity that further depressed company share prices. Executives in the 1970s had little incentive to increase company share prices. Executive share options, popular in the 1960s, fell out of favor in the 1970s following a prolonged depression in the US stock market: the nominal value of the bell- weather Dow Jones average was basically flat from the beginning of 1965 through the early 1980s (falling from 903 in January 1965 to 871 in January -26-

35 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION 1982, and only surpassed 1050 on one day over these seventeen years). Over this time period, many firms “repriced” their existing options (by lowering the original exercise pr ice), while others vor of accounting-based “l ong-term incentive plans” entirely abandoned their option program in fa promising more predictable payouts to executives. Companies in industries with excess capacity typically generate free cash flow, that is, cash in excess of the amount th at can be productively invested within the company or the i ndustry. In this situation, value is created by downs izing and by returning cash to shareholders who can invest the cash in companies and industries with more promising opportunities. The trad itional remuneration practices at the time, however, rewarded size and growth and not valu e creation. In addition, non- monetary aspects of remuneration—including powe r, prestige, and community standing—also tend to be positively linked to firm size and survivability and not to value creation. Over all, the f ailure of corporate governance and remuneration policies of the 1970s helped fuel the cr eation of excess capacity, which in turn set the stage for the capital market restructuring revolution of the 1980s. The 1980s Although there were no large changes to compensation plans and compens ation-related incentives in the early 1980s, managers and boards were subject to increasing pressure and incentives from the market for corporate control. ss cash became targets of Companies with exce hostile takeovers from savvy outsiders — referred to inappropriately at the time as corporate raiders — seeking to put the cash to better uses. These companies also became acquirers (reflecting incumbent managers seeking uses for their excess cash). In both cases the largely cash transactions ition, acquirers financed returned cash to shareholders who could invest it more productively. In add transactions by taking on debt (made possible by the emergence of the active market for hi gh-yield debt), which committed the acquirers to return large portions of current and future cash flows to debt holders instead of wasting the cash on unproductive investments. Managers of potential targets, in turn, thwarted takeover attempts by taking on additional debt through leveraged recapitalizations, simultaneously making their firms a less-attractive target while creating value through the commitment to return future free cash flow to the owners of capital. The availability of high-yield debt also f acilitated the eme rgence of leveraged buyouts in which incumbent managers and outside investors would collaborate to take the company private after paying existing shareholders large premiums for their shares. Managers of these new highly levered organizations faced the discipline of debt (which mitigated incentives to make wasteful investments) and had large nontradeable equity stakes that provided incentives to create long-run value. In contrast to traditional corporations, owned by widely dispersed and passive shareholders and governed by large boards of directors composed primarily of insiders, the new buyout firms -27-

36 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION were owned and governed by concentrated active investors (often through “LBO associations” such as Kohlberg-Kravis-Roberts) who created and managed i governance systems that nnovative 15 facilitated even greater attention to value creation. Although ill-advised court decisions and legislation virtu ally shut down the hostile-takeover e late 1980s, the renewed focus on and LBO market in the US by th creating shareholder value endured. It became apparent that trad itional management incentives focused on company size, stability, and accounting profitability destroyed rather than cr eated value. By this time, shareholder activists and academics (including the first two authors of this report) were increasingly demanding that executive pay be tied more closely to company value through increases in share options and other forms of equity-based incentives. As evident from Figure 1, cash remuneration continued to grow in real terms after th e mid-1980s, but became a smaller part of the total compensation package. Another pay-related development in the takeover market of the 1980s was the evolution of “Golden Parachute” agreements that awarded payments to incumbent managers who lost their jobs 16 in connection with a change in control. Although often introduced as a takeover defense (since re a firm), these agr eements f acilitated trans actions by these agreements make it more costly to acqui lessening incumbent management resistance to takeovers. Change-in-control agreements were fairly rare in the US before passage of the Def icit Reduction Act of 1984, when th e US government imposed a special excise tax on payments 17 ation. exceeding three times the executive’s average recent remuner Ironically, although the cap was meant to reduce the generosity of parac hute payments, the government action appeared to hundreds increase them. The new rules were followed by the introduction of golden parachutes in of companies that had no change-in-control agreements. Apparently compensation committees and 15 See Jensen, 1989a, "Active Investors, LB atization of Ba nkruptcy", Statement before the House Os, and the Priv Ways and Means Committee, February 1, 1989.; Journal of Applied Corporate Finance, V. 2, No. 1: pp. 35-44 the Social S (available from http://papers.ssrn.com/Abstract=244152 ); Jensen, cience Research Network eLibrary at: 1989c, "The Effects of LBOs and Corporate Debt on the Economy", in Remarks before the Subcommittee on Telecommunications and Finance, U .S. House of Representa tives Hearings on L everaged Buyouts , Washington, D.C.: Government Accounting Office; Kaplan, "M anagement Buyouts: Evidence on Post-Buyout Operating Journal of Financial Economics; Kaplan, 1988, "Management Changes", Buyouts: Efficiency Gains or Value Transfers", Chicago, IL: University of Chicago, Unpublished manuscript.; Kaplan, 1989, "The Effects of Management Buyouts on Operating Performance and Value", Journal of Financial Economics, V. 24, No. 2: pp. 217-254; Kaplan, 1990, "Sources of Value in Management Buyouts", Journal of Financial Economics , 16 eceive control pa youts even without losing their In a minority of change-in-control agreements, managers can r jobs. 17 Weston, Mitchell and Mulherin, 2004, Takeovers, Restructuri ng, and Corporat e Governance , Upper Saddle River, N.J.: Pearson Prentice Hall, describe the tax regulations related to parachute payments. -28-

37 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R ell as managers took the regulation as effectively endorsing such change-in-control agreements as w the three times average remuneration (which became the standard). throughout corpor Change-in-control agreements are now ate America: 70 commonplace percent of the largest 1000 companies ha d change-in-control agreements in p lace in 2000, up from McGowan (2001)). In add ition, we b elieve 41 percent in 1988 and 57 percent in 1996 (Alpern and their widespread use has contributed to the emergence of comprehensive employment agreements designed to protect executives from termination for reasons other than a control change. Virtually all these agreements now provide compensation for executives termin for reasons other than ated moral turpitude, gross negligence, or felony convictions. Notably compensation cannot be denied for termination due to incompetence, and we are unable to understand how such provisions could be al that these contracts ha ve become so extreme and in the interests of the firm. We believe in gener so abusive that they call into question the in tegrity of important parts of the remuneration process and the fiduciary responsibilities of boards and remuneration committees. An extreme example is the case in which Th under the stewardship of CEO Mich ael Eisner, paid e Walt Disney Company, Michael Ovitz stock options and cash worth over $100 million do llars when he was released after 14 months on the job at Disney (and amid widespread rumors/accusations of his incompetence). R-3. Employment contracts for CEOs and top managers should be discouraged and when they do exist they should not provide for compensation when a manager is terminated for incompetence or cause. The 1990s Although the US business press had followed CEO pay for d ecades, the CEO pay debate achieved international prominence in the early 1990s. The controversy heightened with the November 1991 introduction of Gr aef Crystal’s exposé on CEO pay, In Search of Excess , and 1992, exploded following President George H. W. Bush’s ill-timed pilgrimage to Japan in January accompanied by an entourage of highly paid US executives. What was meant to be a plea for Japanese trade concessions dissolved into accu sations that US co mpetitiveness was hindered by its excessive executive compensation practices as attention focused on the “huge pay dispar ities 18 between top executives in the two countries.” Consistent with Time magazine’s labeling of CEO pay as the “ populist issue that no 19 politician can resist,” CEO pay b ecame a maj or po litical issue in the US. High CEO salaries emerged as a bipartisan campaign issue among the leading candidates in the 1992 presidential 18 "SEC to Push for Data on Pay of Executives", 1992, Wall Street Journal, January 21. 19 , May 5. McCarroll, 1992, "The Shareholders Strike Back: Executive Pay", Time -29-

38 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION 20 Legislation was introduced in the House of Representatives disallowing deductions for election. aid worker, and the “ Corporate Pay Re sponsibility compensation exceeding 25 times the lowest-p Act” was introduced in the Senate to give shareholders’ more rights to propose compens ation- related policies. The Securities and Exchange Commission (SEC) preempted the pending Senate bill in February 1992 by requiring companies to in clude non-binding sharehol der resolutions about 21 CEO pay in company proxy statements, and announced sweeping new rules affecting the disclosure of top executive compensation in the annual proxy s tatement in October 1992. In 1994, the Clinton tax act (the Omnibus Budget Recon ciliation Act of 1993) defined non-performance- related compensation in excess of $1 million as “unreasonable” and therefore not deductible as an ordinary business expense for corporate income tax purposes. Ironically, although the populist objective was to reduce “excessive” CEO pay levels, the ultimate outcome of the controversy (similar to what happened in response to the Golden Parachute tive compensation, driven by an escalation in option restrictions) was a significant increase in execu grants that satisfied the new IRS regulations and allowed pay signif million icantly in excess of $1 million to be tax deductible to the corporation. It appears from the data that once the Act defined $1 22 and compensation as reasonable many companies increased cash compensation to $1 million, then began to add on the performance based pay that satisfied the act. 20 "Politics and Policy—Campaign ’92: From Quayle to Clinton, Politicians are Pouncing on the Hot Issue of Top Executive’s Hefty Salaries", 1992, Wall Street Journal, January 15. 21 "Shareholder Groups Cheer SEC’s Moves on tive Compens ation" , 1992, Wall Street Disclosure of Execu Journal, February 14. 22 Rose and Wolfram, 2002, "Regulating Executive Pay: Using the Tax Code to Influence Chief Executive Off icer Compensation", Journal of Labor Economics, V. 20, No. 2: pp. S138-S175 document a “spike” in base s alaries at $1 million that did not exist before the new tax rules. -30-

39 E R XECUTIVE EMUNERATION Jensen, Murphy and Wruck Figure 3 Average Remuneration for CEOs in S&P 500 Firms, 1992-2002 $14,042 $14,000 $12,940 17% $12,000 15% $10,030 $10,000 $9,389 $9,325 15% 17% 17% 49% $8,000 54% $6,935 17% 47% $6,000 41% $5,309 47% 17% 37% $4,000 $3,644 Average CEO Pay ($000s) 17% $3,257 34% 18% 14% 20% $2,715 15% $2,618 18% 17% 28% 17% Other 16% 21% 28% 22% $2,000 Options 24% 23% 24% 23% 22% 24% Bonus 17% 18% 23% 20% 19% 24% 27% 31% 34% 38% 37% Salary $0 2001 2000 2002 1999 1997 1996 1995 1994 1993 1992 1998 Note: Average pay levels (in 2002-constant dollars) based on ExecuComp data for S&P 500 CEOs. Total remuner ation ed on date of (indicated by bar height) defined as the sum of salaries , bonuses, benefits, stock options (valu grant using ExecuComp’s modified Black-Scholes formula), stock grants, and other compensation. of CEO pay in S&P 500 firms from 1992 to 2002, Figure 3 shows the composition and level reported in inflation-adjusted 2002 dollars. In 1992, base salaries accounted for 38 percent of the $2.7 million average CEO pay package, wh ile share options (valued at grant date using the B lack- Scholes formula) accounted for 24 per cent. By the peak pay year 2000, base salaries accounted for only 17 percent of the average $14 million pay, wh ile options accounted for half of pay. By 2002, average pay fell to $9.4 million, but options still co mprised nearly half of the t ypical CEO’s pay package. -31-

40 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Figure 4 Outside Hires as Percentage of New CEO Appointments in Large US Firms, 1970-2000 s 1980-1989 Avg: 17.2% 1970-1979 Avg: 14.9% 1990-2000 Avg: 26.5% 35% 30% 25% 20% 15% 10% 5% 0% Outside Hires as % of New CEO Appointment 1970 1990 0 1985 1980 1995 200 1975 Note: Figure shows the fraction of newly appointed CEOs hire d from the outside. Executives serving in their firm for less than a year before the CEO appointment are considered external hires, while those employed for more than a year are considered inside hires. Data include all companies appearing in Forbes annual surveys between 1970 1,323 companies. (The full and 2000, and include 2,783 ne wly appointed CEOs from database includes Forbes 4,633 executives and 2,144 firms, but we exclude CEOs appointed prior to the first year the company is included in the surveys, and also exclude CEOs appointed after the last year the company is included i n Forbes Forbes surveys). the Trends in CEO Demographics The increase in US CEO pay over the past thirty years is well documented. An equ ally pronounced, but less analyzed trend in US corporate governance is the increasing prevalence of filling chief executive officer (CEO) openings t hrough external hires rather than through internal promotions. Figure 4 shows the relative frequenc y of external vs. inte rnal CEO re placements for 23 companies in the annual surveys from 1970 through 2000. During the 1970s and 1980s, Forbes outside hires accounted for 15% and 17% of all CEO replacements, respectively. In contrast, during 23 Executives serving in their firm for less than a year be fore the CEO a ppointment are considered external hires, while those employed for more than a year are consid ered inside hires. D ata are based on 2,783 newly a ppointed CEOs from 1,323 companies. -32-

41 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R CEOs was hired from outside the company. the 1990s (and through 2000) more than one in four Boards going outside to replace poorly performing incumbents do not explain the increase in external hiring. Table 1 Summary Statistics for Newly Appointed CEOs, 1970-2000 1990-2000 1970-1979 1980-1989 Panel A 903 888 992 Newly Appointed CEOs 10.2% As % of All CEOs 11.3% 10.0% Panel B Age at CEO Appointment All New Appointments 53.3 yrs 53.9 yrs 53.3 yrs 53.2 yrs 53.5 yrs Internal Promotions 53.5 yrs External Hires 51.9 yrs 52.1 yrs 53.2 yrs Panel C Tenure at CEO Appointment All New Appointments 14.1 yrs 18.2 yrs 17.2 yrs Internal Promotions 21.3 yrs 19.2 yrs 20.7 yrs  1.0 yrs  1.0 yrs  External Hires 1.0 yrs Note: From Murphy and Zabojnik (2003). Executives serving in their firm for less than a year before the CEO appointment are considered external hires, while those employed for more than a year are considered inside hires. Data include all companies appearing in Forbes annual surveys between 1970 and 2000, and include 2,783 newl y appointed CEOs from 1,323 companies. Table 1 reports summary statistics for the 2,783 newly appointed US CEOs depicted in Figure 4. The full sample is drawn from 4,633 executives and 2,144 companies appearing in Forbes annual surveys between 1970 and 2000. As shown in Panel A, new appointments account for a bout 10% of the Forbes CEOs from 1970 to 1989; CEO turnover increased modestly to 11.3% in the 1990s. Panel B shows that the average age of newly appointed CEOs has increased slightly over the past three decades. Executives promoted internally tend to be older than those hired from the outside. Panel C of Table 1 shows that the average job tenure (prior to CEO a ppointment) has declined substantially over the last thirty years, driven in a large part by the increased prev alence of outside hires (who, by construction, have a y ear or less of tenure upon a ppointment) and to a smaller degree by a decline in the average tenure for inside appointments. -33-

42 E R EMUNERATION Jensen, Murphy and Wruck XECUTIVE Murphy and Zabojnik (2003) show that CEOs hired from the outside earn higher levels of remuneration than CEOs promoted internally. In addition, CEOs in industries with a higher prevalence of outside hiring are paid more than CEOs in industries characterized by internal promotions. We inte rpret these facts as ind icating that the managerial la bor market has become relatively more important for top executives in the US, and the result has been an increase in overall pay levels. -34-

43 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R III. The US-led Option Explosion Executive remuneration in the US has skyrocketed over the past thirty years, propelled in large part by increases in the grant-value of option awards. Identifying options as a leading cause of pay escalation does not “explain” the es calation but merely trans forms the question to: Why has 24 the option component of the pay package increased so dramatically? or section, two poten tial explan ations for the option Based on the discussion in the pri escalation are apparent. First, the increase in option-based pay may reflect the increased focus on equity-based compensati time by sharehol der groups and academic on as advocated at the researchers (especially Jensen and Murphy (1990a); Jensen and Murphy (1990b)). Jensen and Murphy showed that CEOs of large companies were paid like bureaucrats in the sense that they were primarily paid for increasing the size of their organizations, received small rewards for superior performance, even smaller penalties for failures, and that the bonus com ponents of the pay packages showed very little variability. changes in disclosure and tax rules that Second, the increase may reflect contemporaneous reinforced stronger linkages between stock performance and executive pay. The new disclosure rules, for example, implicitly encouraged share options by emphasizing shareholder return and requiring companies to report only the number of, rather than the value of, options granted in the Summary Compensation Table (as found in the company’s proxy s tatement filed annually with the SEC). In addition, under the new tax rules, share options are generally considered “performance- based” and therefore exempt from the $1 million cap on deductible remuneration. o potential explanations are, at best, par However, two facts suggest that these tw tial explanations for the escalation in option compensation for US top executives. First, the trends in the grant-date values of options have varied syst ematically with market share-price movements (although the realized values of options will naturally vary with market movements, there is no obvious reasons why the ex ante grant-date value of options s hould be so c orrelated). Second, the escalation in option-based compensation has not been act has extended limited to CEOs but in f down the corporate hierarchy. The efficiency-explanation for increased equity pay (i.e., providing incentives to take actions that increase shareholder value) is most relevant to t op-level executives who can take direct actions to affect share prices, but not for lower-level employees. In addition, the 24 The discussion in this section draws heavily from Hall and Murphy, 2003, "T he Trouble with Stock Options", V. 17, No. 3: p. 49+. Journal of Economic Perspectives, -35-

44 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Figure 5 Dow Jones Industrial Average Cash and Total Remuneration for CEOs in S&P 500 Firms, 1970-2002 12,000 15000 Dow Jones Industrials Average 10,000 12500 Average CEO Total Remuneration (including options valued at grant-date) 8,000 10000 6,000 7500 Average CEO Cash Remuneration 4,000 5000 Average CEO Pay ($000s) 2,000 2500 Dow Jones Industrials Average 0 0 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 S Note: Dow Jones Industrials based on monthly closing averages. ample is based on all CEOs included in the S&P 500, using data from Forbes and ExecuComp. CEO total pay includes cash pay, restricted stock, payouts from long-term pay programs and the value of stock options modified Black-Scholes granted using ExecuComp’s approach. (Total pay prior to 1978 excludes option grant s, while total pay betw een 1978 and 1991 is computed from exercising stock options during the year, rather than grant-date values.) using the amounts realized disclosure and tax explanation is explicitly relevant only for the top five executives, and not for executives below the top. The first fact is illustrated in Figure 5, which repeats Figure 1 (p. 25) but overlays the Dow O cash compensation is weakly correlated with Jones Industrial Average. The figure shows that CE general market movements, but CEO total compensation is strongly correlated with the stock 25 market. The second fact is illustrated in Figure 6 and Figure 7. Figure 6 shows the average awarded by the average firm in the S&P 500 from inflation-adjusted grant-date values of options 1992-2002. Over this decade, the value of options granted increased from an average of $22 million per company to $238 million pe r company by 2000, falling to $141 million per company in 2002. Employees and executives ranked below the top five have received an increasing share of the total option awards: grants to this group have grown from less than 85 percent of the total in the mid- 1990s to over 90 percent by 2002. Figure 7 shows average annual option grants as a fraction of 25 The total compensation data in Fi gure 5 prior to 1992 are based on amounts r ealized from exercising options, but options were relatively unimportant during this period and average amounts r ealized during the year were closely correlated with average amounts granted. From 1992 forward the to tal compen sation d ata are based on grant-date option values using Black-Scholes values, and ther e is no obvious reason why it would be optimal for firms to award more value in options as the market rises. -36-

45 E XECUTIVE R EMUNERATION Jensen, Murphy and Wruck Figure 6 Grant-Date Values of Employee Stock Options in the S&P 500, 1992- 2002 $250 $238 $226 $200 CEO $146 $150 $141 (4.2% in 2002) Other Top 5 (5.3% in 2002) $101 $100 $71 Employees Below Top 5 $48 $50 (90.5% in 2002) $34 Average Option Grant ($ millions) $30 $25 $22 $0 1993 2000 1999 1998 1997 1996 1995 1994 2001 1992 2002 Note: Figure shows the grant-date value of options (in millions of 2002-constant dollars) granted to all employees in an average S&P 500 firm, based on data from S&P’s ExecuComp data. Grants below the Top 5 are estimated based on “Percent of Total Grant” disclosures; companies not granting options to any of their top five executives are excluded. Grant-values are based on ExecuComp’s Black-Scholes calculations. The number in parentheses indicates the fraction of the grant, on average, that is awarded to the indicated employee (or employee group). Fiscal 2002 results are based on the April 2003 “cut” of ExecuComp, which includes only companies with fiscal closings in December 2002 or earlier. total common shares outstanding. In 1992, the average S&P 500 company granted its employees shares. From 1998 to 2002, in spite of the bull options on about 1.4 per cent of its outstanding market that increased share prices (that, in t urn, increased the value of each granted option), the average S&P 500 company granted options on more than two percent of its shares. So, given these facts why has option compensation increased? Why has it increased with the market? And why has it increased t hroughout the hierarchy? We believe the reason is that option- grant decisions are made by board members and executives who believe (incorrectly) that options are a low-cost way to pay people and do not know or care that the value (and cost) of an option rises as the firm’s share price rises. In the next section, we explore this claim beginning with the fundamental distinctions between the company’s cost of granting an opti on, the value a risk averse employee-recipient places on that option, and th e “perceived cost” of the option from the perspective of corporate decision makers. -37-

46 E EMUNERATION Jensen, Murphy and Wruck XECUTIVE R Figure 7 Grant-Date Number of Employee Stock Options in the S&P 500, 1992- 2002 3.0% 2.6% 2.5% 2.5% 2.5% 2.4% CEO 2.1% 2.0% (.17% in 2002) 1.9% 2.0% 1.6% Other Top 5 (.18% in 2002) 1.4% 1.4% 1.5% 1.4% 1.0% Employees Below Top 5 (1.71% in 2002) 0.5% Average Option Grant (% of common) 0.0% 2000 1992 2001 2002 1993 1995 1996 1997 1998 1994 1999 Note: Figure shows the grant-date number of options as a fraction of total common shares outstanding granted to all employees in an average S&P 500 firm, based on data from S&P’s ExecuComp data. Grants below the Top 5 are estimated based on “Percent of Total Grant” disclosures; companies not granting options to any of their top five executives are excluded. The number in parentheses indicates the fraction of the grant, on average, that is awarded to the indicated employee (or employee group). Fiscal 2002 results are based on the April 2003 “cut” of ExecuComp, which includes only companies with fiscal closings in December 2002 or earlier. The Cost and Value of Options R-4. The cost to the corporation of granting an option to an employee is the opportunity cost the firm gives up by not selling the option in the market, and that cost should be recognized in the firm’s accounting statements as an expense. When a company grants an option to an employee, it bears an economic cost equal to what an outside investor would pay for the option. ppropriate With a downward adjustments for early exer cise and forfeiture (and ignoring poten tially valuable inside information held by executives), the Black and Scholes (1973) formula yields a reasonable estimate of the company’s cost of granting an option 26 to an employee. 26 See Bulow and Shoven, 2004, "Accounting for Stock Options", in Stanford Research Paper Series Paper No. 1848. , March, Palo Alto (available from the Social S cience Research Network eLibrary at: http://ssrn.com/abstract=521882 ), Bodie, Kaplan and Merton, 2003, "For the Last Time: Stock Options Are an Expense", Harvard Business Review , March: pp. 63-71; Merton, 2004, "Su mmary of the Oral Testimony of Robert ated Finance, C. Merton, H.R. 3574: Stock Option Accounting Reform Act", March 3, Washington, DC , Integr -38-

47 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION undiversified, and b ecause they are However, because employees are risk averse and hort-selling prohibited from trading the options or taking actions to hedge their risk (such as s 27 company stock), employees will naturally value options less than they cost the company to grant. Therefore, because the company’s cost exceeds the employee’s value, options are an expensive way to convey compensation to risk-averse employees. And just as in the case with risky compensation in general discussed earlier, it is important for the remuneration co mmittee and board to ensure that the productivity benefits the company expects to get from awarding costly options are more than enough to make up for the pay premium they have to offer to employees receiving the options. In our experience, US companies granting options generally do not make a careful comparison of the cost and value of options, but rather treat options as being essentially free to grant. When a US company grants an option to an employee, it bears no accounting charge and incurs no outlay of cash. Moreover, when the option is exercised, the company (usually) issues a new share to the executive, incurs no cash outlay, and receives a cash benefit in the form of a tax deduction for the spread between the stock price and the exercise price. These factors make the “perceived cost” of an option to the company much lower than the ec onomic cost, and often even lower than the value of the option to the employee. As a result, too many options are granted to too many people, and options with favorable accounting treatment will be preferred to better incentive plans with less favorable accounting treatment. We believe the perceived-cost view explains why the value of options has tracked movements in market stock prices, as documented in Figure 5. If remuneration committees think of option awards as low cost or even no cost to the firm and measure their magnitude by the number of options, this correlation between the grant-date values of the option awards and the level of stock the grant-date value of an option is approx prices will be built in to the system. Because imately proportional to the level of the stock price, awarding the same number of options after a doubling of of the option award. In the period stock prices amounts to doubling the value nnual 1992-1998 the a option awards to CEOs rose from 0.17% of the firm to 0.27% in 1998 a nd back to 0.17% in 2002. Since the Dow Jones index almo st tripled during the period 1992-98 this means the annual dollar value of the option awards increased by more than 300% and we can find no reasonable value- called maximizing basis for this dramatic increase in compensation. The widespread use of so- “fixed-share” option award plans (in which roughly the same number of share options are awarded each year) is consistent with this non-rational non-value-creating result. 2004, "Proposal by Integrated Finance Limited for Expensing Employee Compensatory Stock Options for Financial Reporting Purposes", New York . 27 In making this statement, we ignore inside information held by the employee about the prospects of the firm, and also ignore the potential incentive benefits accruing to shareholders when employees hold options. -39-

48 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION R-5. Remuneration committees should carefully re-examine fixed-share and fixed- option grant programs, fully understand the cost and incentive implications of fixed-share/option vs. fixed-value plans, and communicate to share recipients the value (as measured by their opportunity cost to the firm) of the grants they receive and not just the number of shares or options. ead ignorance of the value Given the seemingly widespr of options and their cost to the firm (witness the economically empty, but continuing, claim that there should be no accounting based charge to earnings for such awards) one can see how a compensation committee could be led to a fixed-share scheme rather than fixed-value option award plan. Suppose we award the CEO 100,000 options this year, the stock price doubles, and we award him 50,000 options the following year to keep the cost the same. How do we deal with the claim that we are pen alizing him for success? Indeed, dealing with this conflict is almost impossible if we e of the option award and continue to refuse to calculate the grant-date dollar valu argue that there is no cost to the firm of option awards. The perception that options are nearly free to grant is readily acknowledged by practitioners and compensation consultants, but is usually dismissed by economists because it implies systematic suboptimal decision-making and a fixation on accounti ng numbers that defies economic logic. But managers often respond to accounting concerns in wa ys that seem irrational to economists, and in a way that is consistent with the notion that it is very common for top managers and boards to be ignorant of what truly creates value in the firms they serve. As an instructive case study, consider what happened when the Financial Acc ounting Standard Board (FASB) changed the accounting treatment for anticipated post-retirement h ealthcare 28 liabilities. Historically, the annual costs of retiree medical benefits were reported on a “pay-as- 29 you-go” basis in company financial statements. xposure In February 1989, FASB issued an “e draft” of a proposed rule change that would force co mpanies to record a current accounting charge for anticipated future medical costs (bringing accounting for retiree medical benefits in line with the current treatment for pension benefits). In December 1990, FASB issued a slightly modified Employers’ Accounting for Postretirement Ben efits version of the exposure draft as SFAS 106 ( Other than Pensions ) and required firms to adopt the new standards no later than 1993. 28 The discussion of retiree healthcare costs draws on Amir, 1993, "The Market Valuation of Acc ounting Information: The Case of Post-retirement Benefits other than Pensions", The Accounting Review, V. 68, No. 4: pp. 703-724; Espahbodie, Strock and Tehranian, 1991, "Impact on Equity Prices of Pronouncements Related to Nonpension Postretirement Benefits", Journal of Accounting & Economics, V. 14, No. 4: pp. 323-346 and Mittelstaedt, Nichols and Regier, 1995, "SFAS N o. 106 and Benefit Redu ctions in Employer-S ponsored R etiree Health Care Plans", V. 70, No. 4: pp. 535-556. The Accounting Review, 29 Prior to 1984, the annual costs of r etiree medical benefits were included al ong with other compens ation-related expenses and not separately reported in financial statements. In late 1984 FASB mand ated separ ate disclosure of retiree medical costs (SFAS 81: Disclosure of Postretirement Health Care and L ife Insurance Benefits ), but did not change the underlying accounting treatment. -40-

49 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION for companies with high anticipated Once adopted, SFAS 106 would reduce reported income future retiree health-care costs (relative to the current “pay-as-you-go” costs) and increase reported income for companies with relatively lower future costs. However, SFAS 106 had no direct cash- flow effects: companies could continue funding retiree health-care costs without changing any current or future cash flows. Nonetheless, a significant number of firms rushed to reduce their benefits. Figure 8 makes it clear that the timing of the cuts was no accident. Some 89% of the firms making benefit cuts did so within a year of adopting SFAS 106. The new accounting rule apparently increased the perceived cost of these benefits, putting them more in line with their actual economic cost, and as a result companies reduced benefit levels. Figure 8 Firms timed reductions in retiree healthcare benefits to boost reported accounting earnings 40 Cuts in 1989 35 Cuts in 1990 30 Cuts in 1991 25 Cuts in 1992 20 15 Number of Firms Cutting Benefits 10 5 0 Prior year Same year Next Year >= 2 years earlier Year Relative to SFAS 106 Adoption Source: Mittelstaedt, Nichols, and Regier (1995), Table 2. The disappearance of option repricing also illustrates how companies respond to accounting on, but controversial, practice virtually rules that have no affect on company cash flows. This comm disappeared after December 1998, when FASB imposed an accounting charge for repr iced options (see Murphy (2003), and Carter and Lynch (2003)). Many companies with declining stock pr ices circumvented the accounting charge on repriced options by canceling existing options and re- (2003)). But this issuing an equal number of options after waiting six months or more (Zheng -41-

50 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R replacement is not neutral. It imposes substantial risk on risk-averse employees since the exercise price is not known for six months and can con the original exercise price. In ceivably be above addition, canceling and reissuing stock options in this way provides perverse incentives to keep the stock-price down for six months so that the new options will have a low exercise price. All of this to avoid an accounting charge that has no affect on the firm’s real costs? From the perspective of many boards and top executives who perceive options to be nearly actice is the trouble associated with ob taining costless, the relevant “cost” of options in pr shareholder approval for additional grants coupled with the cost of additional dilution. Advisory firms often base their shareholder voting recommendations primarily on the option “overhang” (that is, the number of options granted plus options remaining to be granted as a percent of total shares outstanding), and not on the ec onomic cost of ition, the number of the proposed plan. In add options granted is included in fully diluted shares outstanding and therefore increased grants will decrease fully diluted earnings per share. These perceived costs vary with the number of options granted, and not with the dollar-value of the grants, and are consistent with the observed excessive focus on the number of options and not their cost to the firm. We believe that the low-perceived-cost view of options explains why options are granted in workers, and also explains why grant-date such large quantities to large numbers of opportunity cost values rose dramatically and subsequently declined with the stock market as shown in Fi gure 5. We speculate that as grants for top executives increased (for the reasons offered a bove), companies faced growing pressure to push grants down throughout the organization (see, for example, Flanigan (1996)). Employees clamored long as other for broad-based grants, as components of their compensation were not lo wered. Boards readily su ccumbed, especially since nts in mid-2003) shareholder approval was required (prior to changes in exchange listing requireme not required for broad-based plans. In for plans concentrated among top executives but was ills that encouraged broad-based stoc addition, several b k option plans were introduced in 30 Congress. As a result of these pressures, the number of options granted (expressed as a fr action of outstanding shares) grew modestly (Figure 7). But, the economic cost of these grants to the firm followed general stock-market mov ements, and theref ore grew dramatically t hrough 2000. We can 30 For example, H.R. 5242, 2002, Workplace Employee St ock Option Act of 2002 (T o amend the Internal Revenue Code of 1986 to enc ourage the granting of employee stock options .), U.S. House of Represen tatives, Boehner; S. 2877, 2002, Rank and File Stock Option Act of 2002 (To amend the Internal Revenue Code of 1986 to ensure that stock options are granted to rank-and-file employees as well as officers and directors), U.S. Senate, (107th Congress, 2nd Session), Boxer and H.R. 2788, 1997, Employee Stock Option Bill of 1997 (To amend the Internal Revenue Code of 1986 to promote the grant of incentive stock options to non-highly compensated employees), U.S. House of Representatives, (105th Congress, 1st Session), Houghton. -42-

51 E R EMUNERATION Jensen, Murphy and Wruck XECUTIVE ation committees and find little or no value-cr eating rationale for these facts and b elieve remuner boards can and should take actions to stop it. -43-

52 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION 31 IV. Corporate Scandals and the Agency Costs of Overvalued Equity The recent wave of corporate scandals has been associated with the ruination of many fine companies, record numbers of senior executives going to jail (See D'Avolio, Gildor, and Shleifer (2002)), and a major decline in the public view of business and corporate executives. More will come before this period is over. The scandals have highlighted the failure of tems governance sys and motivated substantial new regulations such as the Sarbanes-Oxley legislation in the US, new regulations of accounting practices, and new rules by exchanges such as the NYSE and NASD. In 2002, the International Accounting Standards Board (IASB) and the US Financial Acc ounting Standards Board (FASB) held their first joint meeting and have been working together to harmonize accounting regulations. By January 2005, European firms are to have a dopted 32 international accounting standards. Before we can “solve” current problems we mu st be sure we understand their root cause. The root cause was not that many executives suddenly decided to be crooks, but rather lies with the system in which they were working. (Saying the root cause lies with the system does not mean we absolve the executives invol ved from personal accountability for their acti ons.) Para doxically, the problems arose when the equity of many firms b ecame dangerously overvalued and CEOs, CFOs and boards of directors were caught up in a vicious cycle of ever higher stock values that created forces that caused the destruction of part or all of the core values of their firms — what Jensen (2002) characterizes as the “agency costs of overvalued equity.” These costs result from the damaging managerial and organizational incentives created when a firm’s equity becomes substantially overvalued. By substantially we mean not 10 percent but 100 or 1000 percent as was true for many companies in the recent bubble. Equity is overvalued when a firm’s stock price is higher than its underlying value. Note that definition a firm whose stock is substantially overvalued will not be able to deliver the by performance the market expects to justify that valuation. The situation faced by managers and the board of such a company is fraught with confusion and mixed signals that makes it extremely 31 n, 2002, "The Agency Cost of Overvalu ed Equity and the Current State of This section draws heavily on Jense Lecture European Finan cial Mana gement Associati on)", June 2002, L ondon (av ailable Corporate Finance (Keynote from the Social Science Research Network eLibrary at: http://ssrn.com/abstract=590961 ); Jensen, 2004, "Agency Costs of Overvalued Equity", Negotiations, Organ Working Paper No. 04-26, and izations and Markets (NOM) European Corporate Governance Institute (ECGI) Working Paper No. 39/2004, May (available from the Social Science Research Network eLibrary at: http://ssrn.com/Abstract=480421 ). 32 "FASB Backs Project to Study Unification of Accounting Rules", 2002, Wall Street Journal, October 3 and Weil, 2003, "Fixing the Numbers Problems -- Accounting Standards Board Takes on Hot-Button Issues in Timely January 13, p. C1. Manner", Wall Street Journal, -44-

53 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION of the firm — especially when coupled with a difficult to limit the destruction of the core value general lack of understanding of the dangers of overvaluation. In part, the massive overvaluation of equity that occ urred in the late 1990s and early 2000s was an understandable market mistake. Society often seems to overvalue what is new — in this case, high-tech, telecommunications, and internet ventures. But this catastrophic overvalu ation was also the result of misleading data from managers, large numbers of naïve investors, and breakdowns in the agency relationships within companies, in investment and commercial banks, and in audit and law firms many of whom knowingly contributed to the misinformation a nd manipulation that fed the overvaluation. Managerial and Organizational Heroin Like taking heroin, manning the helm of an overvalued company feels great at first. Like heroin for an addict, overvalued equity generates highly misleading signals for an organization and its board and managers. The capital markets (both equity and debt) are wide open to such a firm. Its managers are likely to get much favorable media attention and their equity-based compensation is contributing to greatly increased personal wealth. But as drug users learn, massive pain lies ahead. It becomes ever more clear to the managers of such organizations that it is difficult to generate the performance necessary to support the sky-high stock price. And knowing that the market will hammer the stock price if it becomes clear the expected performance w ealized, managers ill not be r generate the required pe begin to take actions that will at least appear to rformance. They use the firm’s overvalued equity as currency to make ac quisitions to s atisfy growth expectations. They use access to cheap capital to engage in excessive internal spending in risky greenfield inves tments. They make increasingly aggressive accounting and operating decisions that shift future revenues to ally when these f ail to resolve the issues, the present and current expenses to the future. Eventu managers, under incredible pressure, turn to further manipulation and even fraud. None of these actions truly improve performance. In fact when they are taken not to create real value, but to give the impression of value-creating growth, they destroy part or all of the firm’s core value. But such value destruction is not immediately obvious because in situations like these the market can be fooled during the time it takes for the actual results to be fully revealed. But how can a CEO and CFO argue persuasively to their board that they must take action to reduce the price of the stock? That is especially difficult in a world in which managers and boards have not learned effectively that long-run firm value creation does not necessarily mean maximizing the price of the stock. -45-

54 E Jensen, Murphy and Wruck XECUTIVE EMUNERATION R Let’s understand in more detail the situation such managers find themselves in with their $70 b illion, E nron was actually board. Consider Enron. Our guess is that at its peak value of about nnovative company with a good business, but it was not nearly worth about $30 billion. It was an i as good a business as some argued. For example, one analyst at Deutsche Bank Alex, Brown went so far as to predict that Enron “...would dominate the wholesale energy market for electricity, 33 atives, bandwidth, and energy serv ices on three continents.” natural gas, coal, energy deriv Enron’s management could have helped the market reduce its expectations. They could have found the courage to reset the company’s value. Instead, in choosing to defend the $40 b illion overvaluation (which was going to disappear anyway) they destroyed the entire $30 b illion core 34 value of the company. But imagine the typical board’s response if a CEO says, “we have to reduce the market value of our company by $40 billion.” Investors in the company are not going to see this as a value resetting, a mirage, that was going to go away anyway. They will see it as a real value loss. The board meanwhile, has no clear understanding of the overvaluation, and how defending it destroys part or all of the real core value of the firm. Coup le this with the board’s clear understanding of the ’s market value, and it is not surprising that few massive pain associated with resetting the firm boards or managers mustered the courage to resist the pressure to defend the overvalu ation. Indeed, most CEOs making this argument to the board would probably be fired with the mantra that “ i f you can’t do it we’ll get someone who can”. The situation is par ticularly bad in times like we recently went through in the technology, telecommunications and dot-com sectors where so many firms were dramatically overvalued. Such common overvalu ation makes it even more difficult to y to point to competitors and conclude that they distinguish it for a particular firm because it is eas are managing to deliver the value the market expects. And as boards in these industries now have seen, those competitors weren’t succeeding either. Because top managers and board members have not had the language to talk about the dangers of overvalued equity, few have fully understood it. And even those who have sensed the ’ stock price rose dr the game. When eToys amatically on problem have been unable to stop playing its first day of trading on the NYSE in May 1999, CEO Toby Lenk, reportedly said to his CFO, 35 An interesting comment given that the value of “This is bad. We’re going to live to regret this.” Lenk’s stock had just reached $850 million on the opening day. 33 Tirello, 2000, "Enron Corporation: The Industry Standard for Excellence", Analyst Report, Deutsche Banc Alex Brown, September 15, 2000, New York . 34 See Swartz and Watkins, 2003, Power Failure: The Inside Story of the Collapse of Enron , New York: Doubleday 35 June 9. Sokolove, 2002, "How to Lose $850 Million -- And Not Really Care", New York Times Magazine, -46-

55 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION (undoubtedly pushed by Lenk knew something was wrong, but he and his management team acity to support those who had to prolong the stock price fall until they could sell) built the cap $500 million in sales, and advertised similarly. But sales peaked at $200 million and in February 2001, just 21 months after that first day of public trading the company filed for bankruptcy and was 36 eventually liquidated. This did not have to happen. Failed Governance and Failed Incentives The market for corporate control solved many of the problems of undervalued equity in the through hos tile takeovers, leveraged buyouts, a nd management 1970s and 1980s buyouts. It could not, however solve the agency problems of overvalued equity. It is difficult, to say the least, for an outsider to buy an overvalued company, eliminate its overvaluation and make a profit. In addition, equity-based comp ensation t hrough opti ons, restr icted, unrestricted or phantom stock holdings by executives could not solve the problem either. In fact, in the presence of significantly overvalued equity such equity-based incentives are like throwing gasoline on a fire — 37 they make the problem worse, not b etter. This is but one example of problems that cannot be 36 There were many eToys type exper iences. See the story of TheStreet.com, c hronicled in Cr amer, 2002, Confessions of a Street Addict , New York: Simon & Schuster. Originally slated to go public for $9 to $11 per share, shares in TheStreet.com opened at $61 per share and reached $70 early in its first trading day on May 11, 1999. Its stock never rose from its opening day, falling to $20 per share by year-end 1999 and to $3.00 per share by year-end 2000. In an attempt to placate analysts and keep the stock high, CEO Kevin English advocated spending money to buy up other internet companies and began expansion into Europe. English was fired in November 1999 immediately slashed expenses, (walking away with a large grant of fully vested stock). The new CEO came in and fired hundreds of employees, and dismantled the European operation. 37 sstate financial orporate managers mi Consistent with thisEfendi, Srivastava and Swanson, 2004, "Why do c orporate governance, and other factors", unpublished working paper, statements? The role of option compensation, c from the So cial S cience Mays Business School, Texas A&M U., May 17, College Station, Texas (av ailable http://papers.ssrn.com/abstract=547922 ) in their recent study of 100 firms who Research Network eLibrary at: restated their earnings in 2000 and 2001 document that firms with CEO’s who have large amounts of “in-the-money” options are much more likely to be involved in restatements. Indeed, as compared to their control sample of 100 matched firms with no restatements the average value of in-the-money options for CEOs of restating firms is $30.1 million vs. $2.3 million for the no-restatement firms. Ericks on, Hanlon and Maydew, 2003, "Is There a Link Between Executive Compensation and Accounting Fraud?" University of Chicago Working paper, October 3, 2003, Chicago, IL find that the higher the proportion of stock-based compensation to total compensation for the 5 highest paid executives, the greater the likelihood of accounting fraud. (Stock-based compens ation is defined as the sum of the Black-Scholes value of current year stock option grants and the market value of restricted stock grants. Total compensation includes stock-based compensation, salary and bonus.) They find, for example, that stoc k-based pay accounts for an average of 56 percent of total pay for the top five executives in firms accused of fra ud, but only 41 percent in firms not accused of fraud. See also Harris and Bromiley, 2004, "Incentives to Cheat: The Influence of CEO Incentive Compensation and Relative Firm Performance on Financial Misrepresentation", Univ. of Minnesota Carlson School of Management Working Paper, ; Johns on, Ryan Jr. and Tia n, 2004, "Executive Compensation and Corporate Fraud", Ourso College of Business Administration Working Paper, May, Louisiana State University ). (available from the Social Science Research Network eLibrary at: http://ssrn.com/abstract=395960 -47-

56 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION solved by compensation/incentive systems alone. There will always be issues that require good control systems and monitoring by principled people in a well-designed governance system. hould dr amatically reduce or prohibit managers from holding equity; Some have argued we s 38 . We believe it unwise to return to the old days in which managers were paid like bureaucrats see 39 As we’ve said above, all compensation and all the problems associated with that situation. schemes have the potential to both reduce and to increase agency problems. Many, but not all of the problems with equity-based remuneration can be traced to the lack of required long-term horizons that can be resolved through eff ective holding-period constraints and other policies that are discussed below. ective Thus, in the end, the only solution to the agency problems of overvalued equity is an eff corporate governance system. And we witnessed massive failure in which the boards of directors of company after company failed to stop the charades that eventually led to corruption and the associated destruction of organizational v alue. Many scholars have warned for d ecades that corporate governance systems were woefully inadequate. The results of the last few years have substantially buttressed this position and have led to widespread re-examination and calls for reform of governance systems that ba sically leave top management in many organizations effectively unmonitored. R-6. High equity-based compensati on for management requires in creased monitoring by the board and remuneration and audit committees of reporting policies and the company’s relations with the capital market in general. Because incentives are greater in the ation presence of high equity-based compens understand (both to increase value and to avoid destruction of value), boards must that additional monitoring is likely to be required. Because of the increased benefits of manipulating financial reports and/or operating decis ions to pump u p the stock and therefore generate larger payoffs in the short term, remuner ation and audit committees must increase their monitoring. In addition, they should pay managers cannot bene careful attention to ensuring that their fit from short-term increases in stock prices that are ach ieved at the expense of long-term value destruction. Firms that are more likely to require external financing are also more likely to commit accounting fraud. Although not always significant in their statistical models, they also uncover evidence that the weaker the c orporate governance system (as indicated when the same person holds the CEO and chairman titles), the more likely the fraud. 38 Martin, 2003a, "Taking Stock: If you want managers to act in their shareholder's best interests, take away their company stock", Harvard Business Review, V. 81, No. 1: p. 19; Martin, 2003, The Wrong Incentive: Executives Taking Stock Will Behave Like Athletes Placing Bets Barron's Online, (An electron ic version is available at: http://online.wsj.com/barrons/article/0 ,,SB107187920976382100,00.html ) 39 Frey and Osterloh, 2004, "Yes, Managers S hould Be Paid Like Bureaucrats", University of Zurich, March, Zurich take the opposite view. -48-

57 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION The Solution? It might well be impossible to solve the overvalu ation problem once it occurs. The real solution is to stop it from happening in the first place. This means going against the universal term pain for long-term benefits. We must refuse to play the human reluctance to endure short- 40 earnings-management game. our stock to We must refuse to contribute to the problem by hyping analysts. This means that as managers and board members we must stop creating and consuming the overvaluation heroin. R-7. If our company’s stock price starts to become overvalued we must resist the temptation to enjoy and encourage it. We must make sure that we are communicating to the markets the information regarding the firm’s cu rrent and long run health and prospects. Management and the board should not be in the business of telling the markets what value is. That is for the markets and the analysts to determine. Manag ement must be accountable for informing markets on the firm’s strategy and its progress must work to ma ke their organizations (or lack of it) on executing it. Managers 41 far more transparent to investors and to the markets. Companies should state their strategies clearly, identify the relev ant value drivers and report auditable metrics on their progress in executing the strategy. This reporting should address that part of the firms share price not directly linked to observable cash flows through a clear description of the growth opportunities they foresee and a willingness to tell the markets when they perceive their stock price is overvalued. Audit committees and boards should establish regular communication with R-8. substantial short sellers of the company’s stock. Those who have bet their own money on the future decline of a compa ny’s stock are potentially valuable sources of information regarding potential overvaluation of our firm. Therefore the board and pa rticularly the audit committee should be very interested in hearing the logic behind short sellers actions. There may be good reasons why the board and committee choose to ignore such i nformation after evaluating it (for example, some short sellers might have an interest in disseminating incorrect information so as to profit in the short run), but it would be foolhardy not to be informed short sellers of our secur ities. of the views of This will require a substantial change of attitude in most boards and manag ement 42 teams where short sellers are commonly thought of as the “enemy”. R-9. Business educators, while teaching st udents the desira bility of maximizing value, must also teach them about the dangers of overvaluation. Maximizing firm value does not mean maximizing the price of the stock. 40 For a more extensive discussion of these issues and recommendations see Fuller and Jensen, 2002, "Just Say No To Wall Street: Putting A Stop To the Earnings Game", Journal of Applied Corporate Finance, V. 14, No. 4, Winter 2002: pp. 41-46 (available from the Social Science Research Network eLibrary at: http://ssrn.com/Abstract=297156 ). 41 Diller, CEO of USA Networks, provides analysts with actual budgets broken down by business segments. 42 We are indebted to Jeff Skelton for pointing out the potential value of communications with short sellers. -49-

58 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Executive Remuneration as an Agency Problem V. ource of agency While remuneration can be a solution to agency problems, it can also be a s problems. However well intentioned, boards a nd remuneration committees are not spending their own money, so there is an agency problem between boards and the company that they are there to represent. In addition, even the best-designed plans contain exploitable flaws, and because they have a huge information advantage clever executives can inevitably manipu late the remuneration process to benefit themselves at the expense of the co on, we mpany if they choose to do so. In this secti identify several widespread problems with pay processes and practices, and suggest changes in both corporate governance and pay design to mitigate the problems. Problems with the Appointment and Pay-Setting Process A primary role of boards of directors is to hire, fire, and set the remuneration of the CEO and other top executives. Most large public corporations have remuneration or compensation committees charged with evaluating the CEO’s performance and making r ecommendations relating to executive pay. In our experience, remuneration committee members approach their jobs with good intentions, intelligence, and integrity, but are not as diligent as they would be if they were spending their own money. In add ition, remuneration tinely lack the i nformation, committees rou ecessary for hard-nosed contract negotiations with incumbent and expertise and negotiating skills n ecutive contracts are almost inev itably tilted towards the benefit incoming executives. As a result, ex of top executives. In this sub-section, we identify several of the problems and root causes of the appointment and pay-setting process. How pay decisions are made Although all major decisions related to top-level pay are passed through the remuneration committee, the committee rarely conducts market studies of comp etitive pay levels or in itiates or proposes new incentive plans, and only seldom retains its own compensation experts. Rather, in itial recommendations for pay levels a nd new incentive plans typically emanate from the company’s human resource department, often working in conjunction with outside accountants and compensation consultants. These recommendations are usually sent to top managers for a pproval and revision before being delivered to the compensation committee for consider ation. The CEO typically participates in all committee deliberations, except for discussions sp ecifically dealing with the level of the CEO’s pay. The committee either accepts the recommendations or sends them back -50-

59 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R for revision. If accepted, the committee passes its recommendations on for approval of the full board of directors. Remuneration committees, which typically meet only six to eight times a year, lack both the time and expertise to be involved in the minutia of performance evalu ation and pay design. The fact that initial recommendations are made by company management and not by the remuner ation committee may be an efficient outcome given the time and resource constraints faced by the committee, but it calls into question the integrity of the remuneration process. The fact that the committee only sees plans that have already b een “blessed” by top managers creates an mmittee members are environment that invites abuse and bias. Put differently, although individual co generally competent and well motivated, the governance system itself is corrupted and tilted in the direction of management in a way that will almost inevitably lead to excesses in executive pay levels. Remuneration committees must take full control of the remuneration process, R-10. policies, and practices. In particular remuneration committees should jealously guard their in itiation rights over executive remuneration. Th ey must abandon the role of simply ratifying management’s remuneration in itiatives. Obviously guarding their d make decisions and initiation rights does not mean that committees shoul recommendations to the whole board without discussions with management, but this is quite different from allowing management to seize the de facto remuneration initiation rights. Remuneration committees can ask for data or information from corporate human resource officers, but these officers should report directly to the committee (and not to top management) for committee- related assignments. Similarl should be hired by, and y, compensation consultants report directly to, the remuneration committee and not to management. Pay negotiations and the market for CEOs Remuneration committees almost invariably pay “too much” for newly appointed CEOs, especially for those hired from outside the firm. Corporate directors seeking new CEOs from outside typically hire a professional search for the pos ition firm to identify qualified candidates (Khurana (2002a, b)). The pool of qualified candidates is narrowed through extensive research, background and reference checks, and interviews until a single individual is selected for the position. Negotiations over pay typically begin only after the favored candidate is identified and told that he is to be the new CEO. At this point the board is anxious to secure his services, and the combination of these two factors dramatically shifts the bargaining power to the seller (the candidate) rather than the buyer (the firm). This procedure is a reasonable way to identify top candidates when “price” is not an issue, but is clearly a recipe for sys tematically paying too much for managerial talent. -51-

60 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION acerbated by potential The tendency to pay too much and to pay it in the wrong way is ex CEOs who hire skilled contract agents to negotiate on their behalf. The most famous hired who charges close to $1,000 per hour to extr act as much negotiator in the US is Joseph Bachelder, as possible from the company in the form of salaries, target (and often guaranteed) bonuses, option 43 grants, retirement benefits, perquisites and generous severance arrang Mr. Bachelder is ements. especially proficient in extracting high “si gn-on bonuses” to offset unvested options and benefits from the candidate’s former employer. In a fin al bit of irony, Bachelder routinely negotiates for the company to pay his fees, which can run as high ngle contract. It is not hard to as $100,000 for a si see how what a CEO asks for can easily escalate when he or she is not personally doing the asking, and when the agent is urging the candid ate to not ask for less than the maximum the agent knows has been obtained by anyone else remotely comparable in the country. In such an environment it is not surprising that agents such as Mr. Bachelder increasingly represent executives. In contrast, remuneration committees rarely retain their own expert negotiators. The outcome is a clear mismatch: no matter how well intentioned, the typ ical remuneration committee is no match against a professional nego tiator, and gene rous pay packages become ubiquitous. But, sometimes the problem is worse: the incoming CEO (and his professional agent) negotiate not with the remuneration committee but rather with the company’s general counsel port to the CEO when the or head of human resources. These internal managers know they w ill re contracting is complete, providing strong incentives to make their new boss pleased with his financial arrangements and making it very diff tiations icult for them to play hardball in the nego knowing that any residual anger will unlikely disappear once the deal is concluded. For example: . . . Mr. Bachelder takes private delight in spotting the other side’s weaknesses during negotiations. “Joe took me aside after one contract,” says Michael Valentino, an executive who has worked at several drug companies, “and told me: ‘I knew on Day One that we were going to get everything you wanted.’ “When Mr. Valentino asked why, Mr. Bachelder told him that the hiring company had mistakenly put its general you’re going to be that guy’s counsel in charge of the talks. “When this is over, lained. “He knows that. He can’t fight you too hard on boss,” Mr. Bachelder exp anything.” Anders (2003). R-11. Firms should resolutely refuse as a matter of policy to pay the fees for the contracting agents negotiating for the CEO or other top-managers. Such reimbursements would appear to e board’s fiduciary be a violation of th responsibility to the firm, and have clearly undesirable incentive effects on managers’ decisions to hire such agen ts and for the aggressiveness and time such agent’s spend in the negotiation process. 43 Information on Mr. Bachelder is drawn from Anders, 2003, "Upping the Ante: As Some Decry Lavish CEO Pay, Joe Bachelder Makes It Happen", Wall Street Journal, June 25, p. A1; Kampel, 2000, "Five Questions for Joseph E. Bachelder, Engineer of the Executive Pay Express", New York Times, June 11 and Whitford, 1998, , V. 137, No. 11, June 8, p. 281+. "Becoming CEO? Call Him First", Fortune -52-

61 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Remuneration committees should employ their own professional contracting R-12. agents when hiring new top-level managers. for the committee to do so when the manager being It is especially important recruited has hired his or her own agent. The conflicts of interests in such negotiations are high with current man agers and even current board members (who quite reasonably wish to bring a new person on board in a climate of cooperation and good will). Therefore, bri nging in an outsider who answers solely much of the details of the negotiation to the remuneration committee to handle can help put balance back into such negotiations. Moreover, boards should be wary of announcing the new appointment before the terms of engagement have been agreed upon. Judgment calls go to the CEO lt it is to manage the negotiation process involved in Once we fully appreciate how difficu hiring a new and highly desirable CEO from outside the firm, we begin to have an idea how difficult ryone in the firm already answers to him and it is to manage the process for a sitting CEO. Eve process is complete therefore expects to have to continue to work with him after the remuneration each year. And although the committee members do not formally answer to the CEO it is not uncommon for them to behave as if they effectiv ely do in many firms. In addition, neither the board nor the committee wishes to spoil relations with a su ccessful CEO, much less start an internal war over the annual remuneration process. Even after the hiring decision, remunerati on committees inevitably pay too much. There is in the most well-intentioned boa tematically tend little question that judgment calls even rdroom sys a range of market data on comp etitive pay levels, committees tend to to favor the CEO. Faced with err on the high side. Faced with a choice between a sensible compensation plan and a slightly ill defer to manag ement. Similarly, faced with a inferior plan favored by the CEO, the committee w discretionary choice on bonus-pool funding, the co mmittee will tend to over- rather than under- fund. Bebchuk, Fried, and Walker (2002) and Bebchuk and Fried (2003) argue that the r ecent escalation in executive pay reflects the actions of incumbent executives who can raise their own pay by exercising influence over hand-picked directors. While the factors we have outlined are consistent with their arguments that CEOs will use their power to extr act rents from the firm, their assessment is somewhat overstated, because it cannot explain the dramatic increase in stock options (which, in turn, largely explains the pay escalation). In addition, executives hired from outside the company typically earn higher wages than execu tives promoted internally, suggesting that poor negotiating expertise on the part of remuneration committees may explain more of the increase than captive board members catering to entrenched managers. The inherent biases in the pay-s etting process are not easily solved by enhancing board independence (the obvious prescription to solve -53-

62 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION r) but rather requi mmittees to the problem as framed by Bebchuk-Fried-Walke re remuneration co invest in much greater information and negotiation expertise and to change the very structure of the huk, evaluation and pay-setting process. See Murphy (2002) for an extensive discussion of the Bebc Fried and Walker arguments. Much has been made about the power of CEO’s in the boardroom, and much of this discussion focuses on the notion of independe nce of board members from the CEO and from the company. We agree that independent directors are important to a well -functioning governance system, but we are concerned that this discussi on can sometimes miss the point in the larger picture in which the governance function is carried out. In im proving the remuneration process we must issues that have been responsible for the f ailure of concern ourselves with resolving some of the governance systems. We make three general recommendations here that we believe are requisite to a well-functioning board and remuneration system. R-13. Change the structural, social and psychological environment of the board so that the directors (even those who fulf ill the requ irements of independence) no longer see themselves as effectively the employees of the CEO We frame the guideline this way not because it is the cause of the problem, but rather because it is a highly productive frame from which to view the symptoms and causes of fundamental problems with governance. Changing this mindset will not be an easy task. It will require major changes in th e social, psycholog ical, and will no longer see power structures in boards. And when it is accomplished boards their role as one of primarily supporting the CEO rather than monitoring the CEO as is so common in the American model. The support role is clearly important but must be strictly subordinate to the board’s role as monitor. Consider the ecruiting for the board and extends the following: the CEO does most of the r offer to join the board. And, excep t in unusual circumstances, board members serve at the pleasure of the CEO. The CEO generally sets the agenda for the board. Moreover, it is rare that the board meets outside of the CEO’s presence or members without his explicit permission. Finally, virtually all information board or passes through th e CEO, except in receive from the company originates from highly controlled or unusual circumstan ces. Changes in these practices will require a major change in the power relationship between the board and the 44 CEO. R-14. The board should be chaired by a person who is not the CEO, was not the CEO, and will not be the CEO The critical job of the Chair is to run the process that evaluates, compensates, hires and fires the CEO and top management team. The CEO cannot perform that job adequately. 44 For a more complete discussion of these issu es see Jensen and Fuller, 2002, "What's a Dir ector to Do?" Best Practice: Ideas and Insights form the World's Foremost Business Thinkers, Cambridge, MA. and London: Perseus Publishing and Bloomsbury Publishing (available from the Social Science Research Network eLibrary at: ). http://papers.ssrn.com/Abstract=357722 -54-

63 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Limit the number of outside CEOs sitting on the board R-15. board members for many obvious reasons. Outside CEOs offer advantages as What generally have gone unaddressed are the disadva ntages they bring to the board. It is natural for them to subconsciously (if not consciously) view the board through CEO eyes — a lens where the power of the CEO is not seriously challenged, except perhaps in the event of serious problems such as obvious incompetence or malfeasance. The CEO should be the only member of the management team with board R-16. membership. While members of the management team can add value by participating in board discussions there is little reason to ha ve them be formal voting members. When other members of the management team are voting members of the board we increase the likelihood that the board will consider its job to be that of su pporting, not monitoring, the CEO. Members of management that can add value to board discussions can and should do so by being at the meetings regularly as e x-officio members. The Role of Compensation Consultants Most companies rely on compensation c survey information on i ndustry onsultants to provide and market pay practices and to design incentive arrangements. Although these consultants undoubtedly serve a useful purpose, we believe they have also contributed to abuses in executive pay. Compensation consultants are rarely r on committee but are rather etained by the remunerati retained by company management, and work directly for and with the head of human resources, the chief financial officer, or the CEO. Simply put, the client is the CEO, not the compensation committee. This hiring situation creates an obvi ous potential conflict of interest, since the consultants make recommendations on the pay of the individuals who hire them. More importantly, many of the largest integrated human resource consulting firms (such as eceive fees from their actuarial or (lower- Hay, Hewitt, Towers Perrin, Mercer, WatsonWyatt, etc.) r level) employee pay practices that are orders of magnitude larger than the fees charged by their executive pay practices. Decisions to engage the consulting firm in these more lucr ative firm-wide practice areas are often made by the same top executives who are affected by the consultant’s executive pay recommendation. These prospects for cross-selling other services dramatically increase the conflicts of interest faced by the compensation consultants, the top-level executives, and the firm. It is not realistic to expect a management compensation consultant to aggressively a rgue against overpaying a CEO who the consultant knows is going to rule on hiring him to perform a vastly more lucrative actuarial or rank and file consulting contract. -55-

64 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION nsultants R-17. Remuneration committees should seldom, if ever, use compensation co for executive remuneration purposes who are also used by the firm for actuarial or lower-level employee remuneration assignments. Conflicts between these dual roles onsultants dramatically of compensation c disadvantage the remuneration co mmittee and the firm and facilitate more- generous executive pay packages. Consider the situation of a consultant who hopes to close a multi-million dollar actuarial or lower-level employee engagement. The same consultant engaged as an advisor on CEO and top- manager remuneration policies (that might amount to only a high five-figure or low six-figure fee) would be put at a significant disadvantage in reco mmending value-creating remuneration po licies inconsistent with what the CEO desires. The reasons for avoiding these conflicting roles are essentially the same as the rules that are emerging that limit the use of a firm’s auditor as a consultant. Companies retain compensation consultants in large part to get access to survey information used for competitive “benchmarking.” The surveys, which report a var iety of pay percen tiles ( e.g., th th th , 50 25 , 7 5 groupings or through simple ), typically adjust for company s ize either t hrough s ize log-linear regressions of Log(Pay) on Log(Size). Size is traditionally measured using company revenues, although market capitalization is increasingly used (especially in start-ups with low revenues but high capitalization). As suggested by Baker, Jensen, and Murphy (1988), the size adjustments used in the survey instrumen elation ts both formalize and reinforce the observed r between compensation and company size. In other words, what starts out as a simple empir ical correlation between size of firm and size of remuneration for t op-level managers is turned into a ing the size of the firm causal mechanism that rewards managers for increas hough s they lead even t they may destroy value in doing so. We have no doubt that these factors coupled with the increase in power, visibility and other non-pecuniary benefits associated with larger size play a role in reducing managerial willingness to shrink a firm when that is the value-creating action and similarly increase the motivation to grow a firm even when it destroys value. In addition, we believe that misuse of survey information provided by compensation consultants has led to systematic increases in executive pay levels. Language is powerful and especially so if we are unaware of the nuances of labels. Since pay below the 50th percen tile is often labeled “below market” while pay between the 50th and 75th is considered “competitive,” atchet” eff ect in executive pay levels as firms c hoose to target the surveys have contributed to a “r th percentile. The result is the CEO equivalent of the Lake Wobegon effect: all their pay above the 50 45 CEOs are paid above average (or at least try to be). 45 Lake Wobegon is the fictional Minnesota setti ng for radio entertainer Garrison Keillor’s tales and yarns. He concludes each weekly show with, “That’s the ne ws from Lake Wobegon, Minnesota where all the women are strong, all the men are good-looking, and all the children are above average.” See http://www.prairiehome.org. -56-

65 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Problems with Equity-Based Plans The primary forms of equity-based compensation are stock options, restricted stock, and performance shares. Stock options are contracts that give the recipient the right (but not the obligation) to buy a share of stock at a pre-specified price (called the “exercise” or “strike” price) for a pre-specified term. Conceptually, the exercise price can be set above, below, or at the grant- date share price, or can vary over time with inflation or movements in broad stock indices or with the firm’s cost of capital. The standard practice in both the United States and the United Kingdom is to fix exercise prices at the grant-date share price. Options in the US typically are granted with terms of 10 years; in the UK option maturity terms range between 7 and 10 years. Stock options are usually not exercisable immediately at grant, but rather only become exercisable with the passage of time and/or the attainment of performance thresholds. The standard practice in the US is 46 rformance, while for options to “vest” and become exercisable over time independent of pe options in the UK often include performance criteria. Employees leaving the firm are typically allowed to exercise vested options upon their termination, while unvested options are forfeited. ecipients cannot sell or transfer Restricted stock grants are “restricted” in the sense that the r leave the firm prior to vesting. the shares for a period of time, and will forfeit the shares if they delivery of shares, conditional on continued Performance shares are essentially a promise for future employment (like restricted st ock) and on meeting various performance thresholds. Both restricted alence is much lower stock and performance share plans are common in the US, although their prev than stock option plans. The standard practice in the UK is to grant performance shares (often called “Long-Term Incentive Plans” or LTIPs) and not restricted stock; indeed, in recent years, LTIPs have replaced options as the predominate form of equity-based pay among UK CEOs. Closely related to equity-based plans are cash plans that replicate the payouts from stock plans. For example, “stock appreciation rights” give recipients the right to the spread between the ice, and thus replicate the pa youts from stock options. market price of the stock and the exercise pr Phantom stock plans give recipients the value of shares at some future d ate, sub ject to continued employment and/or performance criteria, and thus replicate the payouts from restricted stock and performance share plans. These cash plans are pa ublic and closely held rticularly popular in non-p corporations, where traditional stock-based plans are impossible or unattractive. Equity-based incentive plans seem a natural if not obvious way to align the interests of managers with those of shareholders. However, poorly designed or implemented equity-based plans can yield excessive levels of compensation and provide incentives to destroy rather than cr eate 46 Some US plans accelerate vesting for superior pe rformance, but the options w ill vest independent of performance as long as the recipient remains employed by the firm throughout the option term. -57-

66 E Jensen, Murphy and Wruck XECUTIVE EMUNERATION R organizational value. In the remainder of this section, we discuss several common problems in equity-based plans, drawing in large part on themes developed earlier in this report. No skin in the game Conceptually, companies offering restricted stoc k or options could reduce base salaries (or licitly other components of remuneration) to partially offset their cost, or alternatively could exp 47 In practice, however, require the executive to buy the stock or options using personal funds. equity-based pay is often layered on top of existing competitive pay packages without requiring any meaningful offset (through direct payments or reductions in other remuneration). The temptation to offer options or stock without an offset follows from the observ ation (discussed in Section 3) that 48 ee” to grant. boards perceive these instruments to be effectively “fr The obvious result of this practice is that executives are systematically ee” but overpaid. Once enough companies offer “fr valuable stock and options they become part of the “competitive” pay package that all firms have to offer to attract scarce managerial talent. The e nd result is a ratcheting-up of pay driven by an escalation in equity-based incentives. A less obvious but potentially more important result is that providing stock and options as an “add-on” erodes the incentives asso ciated with equity-based pay. It is human nature to care more about something purchased through sweat or hard-earned cash than something r eceived for free. Moreover, managers who purchase shares will naturally recognize the opportunity cost of the shares and will strive to earn a fair return on their investment. For example, s uppose that a manager t opportunities purchases $100,000 in stock with his own funds, and that his alternativ e investmen for the money (with the same risk character istics as the company stock) offer an exp eturn of ected r recognize that, if the shares appreciate by 10 percent. This manager will be likely to $5,000 in one poor ret urn on his inves tment and has lost money relative to the year, he has earned a opportunity cost of his capital. In contrast , suppose that the company simply gave the manager $100,000 in stock vesting in one year. If the shares a ppreciate by $5,000, the manag er has lost nothing and in fact walks way with $105,000 worth of shares given to him for free. Indeed, even if the shares lose 5 percent in value, the manager will still end the year with a gift worth $95,000. Thus outright stock grants are very expensive and provide no incentives to self select out of the firm when managers believe they cannot create value. 47 In either case, the cost to the firm will not be completely offset by the contri bution from the manager b ecause (as discussed in Section 3) risk-averse executives w ill gener ally value stock and options less than they cost the company to grant. This will not be true when managers have information that leads them to believe that they can create value. 48 Restricted stock is perceived to be more costly to grant than options, because companies granting stock incur an accounting charge equal to the grant-date value of the stock amortized over the ves ting period. However, companies can grant stock without spending any cash, which makes stock seem cheaper than cash-based forms of pay. -58-

67 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Managers should be required to have “skin in the game” by purchasing stock R-18. or options or by explicitly and deliberately accepting reductions in other forms of compensation. Important advantages to requiring managers to have skin in the game is that it encourages them to recognize the opportunity cost of capital to the company and to reveal the private information and belief s they have about the value-creation potential in their strategic plan. If manag ers are not willing to bet their own money on the plan, it is probably not a good bet for the shareholders either. Managers can get skin in the game through outright purchases of stock and options using after-tax dollars. Historicall y, such purchases have often been financed through loans provided by or guaranteed by the company. We see nothing wrong with such loans, even when offered on a non-recourse basis, provided that the executive is required to put up a sizable down payment and that the loan carries an appropriate interest rate (approximating the company’s cost of cap ital). However, as a response to some high-profile abuses i nvolving loans at a ha ndful of companies, the Sarbanes-Oxley Act now mistakenly forbids US companies from providing or guaranteeing loans to its top executives. As an alternative to using after-tax dollars to purchase equity, executives can get skin in the game through exchange programs that offer stock options in lieu of cash compensation. In order to provide incentives to both recognize the cost of capital and reveal confidence in the strategic plan, these exchange programs must be explicit: it must be clear to all parties what the executive is giving up. Although exchange programs in practice take a var forms, most i nvolve exchanging cash iety of 49 bonuses or current or future in icted stock or options. creases in base salaries Executives for restr participating in exchanges typically receive a “risk premium” for accepting equity-based pay rather than cash. Since executives are risk averse, it is natural for them to expect a compensating differential for accepting riskier pay. However, execu tives demanding excessive risk premiums are likely signaling a lack of confidence in their business strategy rather than risk aversion, and remuneration committees will do well to pay particularly careful attention to such information (more on this below). 49 Some US companies recently completing such exch anges include ADC Telecommunications, Arkla, Avon, Baxter, Black & Decker, Cl orox, EKCO, General Mills, Harnischfeger, International Mu ltifoods, Mead, Merck, PacifiCorp, Panhandle Eastern, Santa Fe Pacific, Sun Company, Teledyne, T oro, Triarc, Union Carbide, Un ited Airlines, and USAir. -59-

68 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R 50 Problems with traditional stock options Traditional stock options, granted as an add on with a fixed exercise price equal to the grant- date market price, are especially prone to misleading executives into thinking that the cost of equity traditional options can reward managers even when value is capital is zero. To illustrate how destroyed, we offer an example based on a real boardroom situation that occurred in a Fortune 500 the top-management team of the company a number of years ago. In its board str ategy presentation company informed the board that if it ratified the strategy the stock price of the firm would rise ement from its current price of $57 per share to $100 per share in five years. The board and manag had already agreed that the cost of equity capital for this company was 15 percent and the company percent per year (based on the beginning-of-year regularly paid an annual dividend of about 2.5 stock price). Given these assumptions, the breakeven value of the equity in five years that leaves $102.72 = shareholders whole (just earning their cost of equity capital net of dividends) would be 5 . So if management’s projection of a stoc k price of $100 in five years were true, $57(1.125) shareholders would lose $2.72 per share (measured in dollars five years in the future). Thus, the roy shareholder value because if the market b elieved that the plan, if executed perfectly, would dest plan would be realized and that the management’s $100 forecasted stock price would be the result, 5 the current stock price would fall to $100/(1.125) = $55.49, an $1.51 immediate loss in value of per share. But the manager’s options awarded at the current market price of $57 per share would in five years be worth $43 = $100 - $57 per share on exercise. Thus, shareholders would lose money while managers would make $43 per share. We would be paying managers handsomely to destroy value for shareholders. Another way to see this is that the typical executive stock option program eff ectively communicates to managers that the cost of equity capital to the company net of the dividend yield is ital. Even t hough one board member in the example zero and therefore encourages the waste of cap discussed above pointed out that this was a value-destroying plan even if executed perfectly (and most board members believed the $100 price was far too optimistic), the board approved the plan anyway. It did so because manag ement did not know what else to do and the board was unable to agree on a solution. After several more years of value destruction the board removed the CEO. plans typically teach executives that the Unfortunately the problems created because option cost of capital is zero (or the divide nd yield for a dividend paying compa ny), is co mpounded 50 This material follows that of Jensen, 2001a, "How Stock Options Reward Managers for Destroying Value and What To Do About It", Harvard Business School Negotiations a nd Markets (NOM) W orking Paper, April 17, 2001 ) (available from the Social Science Research Network eLibrary at: http://ssrn.com/Abstract=480401 -60-

69 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R because most executive stock options do not adjust the exercise price for dividends on the stock. Because paying dividends lowers the value of options dollar for dollar non-dividend-adjusted options thus provide managers incentives to avoid paying dividends or to pay only nominal 51 Unless managers use the excess cash to buy back stock (and there is some evidence dividends. they do) the result is they keep excess cap (free cash flow) in the firm rather than paying it out, ital 52 and this destroys more value. Executive share option contracts should, whenever possible, adjust the exercise R-19. price of the option for any dividends or return of capital paid to holders of the shares. This removes any artificial incentives that managers have to withhold dividends when they have options. Cost of Capital Adjusted Options One solution to the problem associated with fixed-exercise-price options is to change the ital is. In his nature of executive stock options so that they teach managers what the true cost of cap article “Remaking the Corporation from Within” Bennett Stewart (1990) recommends awarding managers options whose exercise price rises at the cost of capital net of the dividend yield. These cost-of-capital-indexed options have many advantages over the typical executive stock option. R-20. Remuneration committees should give careful consideration to issuing executive share options with exercise prices that increase with the company’s cost of capital (less the dividend yield if the option is not dividend adjusted). Consider an example where the cost of e dividend yield, capital is 10% net of th the current stock price is $10 and the exercise price of the option is $10. Such options would pay managers nothing if the stock price failed to rise over any period by an amount greater than the cost of capital less the dividend yield. This means managers earn nothing on their options unless shareholders do better than breakeven. Since cost of capital indexed options are less valuable firms can award more of them to managers for the same cost to the firm and thereby create more high-powered incentives for the same cost. 51 Lambert, Lanen and Larcker, 1989, "Executive Stock Option Plans and Corporate Dividend Po licy", Journal of Financial and Quantitative Analysis, V. 24, No. 4: pp. 409-425 find that “expected dividends” decrease following the initial adoption of top-management stock option plan Martin, 1987, "Executive s. Lewellen, Loderer and Compensation and Executive Incentive Problems: An Empirical Analysis", Journal of Accounting & Economics, V. 9, No. 3: pp. 287-310 find that dividend payout ratios are negatively (but not significantly) r elated to CEO stock- based compensation. 52 Free cash flow is all cash in excess of that required to fund all positive net present value investments. By definition such cash must be paid out in order to avoid dest roying value. See Jensen, "Agency Costs of Free Cash Flow: Corporate Finance and Takeovers",. -61-

70 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION R-21. Managers should receive annual statements that clearly summarize in one place the changes in their wealth in the prior year from all sources of remuneration from the firm (including changes in the present value of future retirement and deferred compensation). Because managers often do not know the sources and amounts of their total compensation we advocate giving them annual statements detailing the changes in the grant date valu e of options received their wealth in the prior year due to during the year, and changes in the present value of their holdings of options, shares, other bonuses, and retirement benefits from the company. If the stock price in our example rises to $11 over the first year it is exactly equal to the new exercise price, and the exercise value of the op tion is still ze ro. Thus, managers would be taught in a graphic way that the cost of equity capital is not zero. any of emoluments can be calculated these In cases where the cost to the comp should also be included in the report. Such accounting will be helpful both to the managers and to the remuneration committee that is managing the process. Some might argue that we cannot get managers to accept these indexed options if other firms are awarding the current flawed options. We agree this is an issue, but there are reasons this can turn out to be an advantage. Because these inde xed options are not as valuable as the standard flawed executive stock options we can (and indeed, must) award more of them if we are to keep managers whole. Moreover, depending upon how much firm value managers think they can cr eate they may even prefer these new indexed options because their payoffs can be greater than under the standard options. And in these conditions the indexed options can induce managers to stay who believe their strategy can create value while encouraging others to leave. Figure 9 provides a graphical illustration of the conditions in which a manager would be better off with cost-of-capital indexed options than the standard options. Assume that the cost of market price at issue of option, exercise price capital is 10%, the exercise price equals the grows at 10% per year for the indexed option and zero for the standard option, and the grant-date value of 53 The figure shows the exercise total value the two packages is equal at a two to one ratio. of 1,000 standard options and 2,000 cost-of-capital indexed options for 4.3 years after award when the 4.3  $10. At this point with stock pr ice equal to exercise price of the indexed option is $15 = (1.1) $15, the 1,000 standard options with an exercise price of $10 are worth $5,000 and the 2,000 options with cost-of-capital-indexed exercise price of $15 are worth zero. But since the slope of the ill earn indexed option payoff is twice that of the standard option the manager with such options w more than one holding the standard options if he or she can create any value in excess of the breakeven point of $20. For example, as the figure sho ws the to tal payoff for the 2,000 indexed 53 For example, suppose that the current stock pr ice is $10, risk-free rate is 6%, the stock-price vo latility is 30%, and the dividend yield is zero. Under these assumptions, the value (using the binomial option value formula to allow for early exercise) of a 10-year option with an exercis e price beginning at $10 and increasing 10% annually is one- half the value of a 10-year option with an exercise price fixed at $10. -62-

71 E EMUNERATION Jensen, Murphy and Wruck XECUTIVE R Cost-of-Capital Indexed Options Have Higher Payoff for Value Creation Figure 9 and No Incentives for Managers to Treat Capital as Costless. It pays managers to choose the indexed option plan as long as they believe they can create more value than the breakeven level of $20 in this example. 35,000 Profit at stock price of $35 for: Total Value Std Options = $25,000 of 2,000 28,000 Indexed Indexed Options = $40,000 Options 21,000 $ Value of Total Option Claims Total Value 14,000 of 1,000 Exercise Price of Standard Standard Options Breakeven Options Point 7,000 0 21 28 35 14 7 0 Price of Stock Assumptions: • Cost of equity capital net of dividend yield = 10% • Exercise price of standard option is $10 / share = Market price at issue 4.3 $10  Graph is for 4.3 years after issue when exercise price of indexed option is $15 = (1.1) • • Equal Black-Scholes value of options is assumed to be 2 indexed options = 1 standard option (this will vary depending on firm, variance, and standard option provisions) ile the total options package at a stock price of $35 per share is $40,000 at 4.3 years after award wh payoff for the 1,000 standard options is only $25,000. We can see several things from the analysis: Ex post it costs shareholders nothing to award cost-of-capital-indexed options to their managers if managers are not able to create value in excess of the cost of capital. Managers share in the gains only when the gains to shareholders are positive in a real, not accounting sense. This means shareholders can give managers a larger share of the pie when they succeed in creating value. ital in the firm unless Considering option gains only, managers have no incentive to hold cap they can earn at least the cost of capital on it. And to the extent that they do hold free cash flow in the firm and invest it at below cost of capital they reduce their option gains. -63-

72 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Managers have incentives to choose the indexed options when they are more confident they can create substantial value in the firm. This mean s there is a desirable self-selection property to these plans. If a competitor offers our managers an option package consisting of 1000 standard leave if they think they ca options as in the graph, our managers are more likely to take it and nnot create much value in our firm, and are much more likely to stay if they believe they can create great value in our firm Let’s assume that our manager’s know they cannot create value in a firm competing for their services. In the case portrayed in Figure 9 above our managers would leave for the comp etitor offering the standard package if they do not believe they can create at least the breakeven ret urns of $20 in our firm. And if we can replace them with managers who can at least earn breakeven returns, we want the current managers to leave. Their choices thus reveal valuable information that the remuneration committee can never get by simply asking managers — either current or potential ones. (A complete analysis of these retention issues means we have to evalu ate the entire executive compensation plan incl uding salary, cas h bonuses, r etirement and other benefits as w ell, but we ation (and assuming our managers ignore them here for simplicity.) Thus in a comp etitive situ understand the tradeoffs portrayed in Figure 9), managers will self-select themselves into our firm to the extent they believe they can create more value in our firm than elsewhere. The self-selection principle explained in the discussion above can be extended with great effect we believe by changing share option plans in one additional aspect. But first note that if a share option grant is offered to an executive as a pure add-on with no give up in any form of aith compensation, he or she would be a fool to refuse to accept it — even if he or she had no f whatsoever in the likelihood of successful ex ecution of the strategic plan. There may be considerable upside from random or market factors alone (especially if it is not indexed to the cost of capital) while there is clearly no downside to accepting the add on options. While it is true such a grant will increase incentives, the executive’s d ecision to accept the opti ons reveals no information to the board that the executive b rategic plan just accepted by the board has no serious elieves the st chance of being executed or if ex ecuted of creating any real value for the firm. If these are the executive’s beliefs, they are something the board and the remuneration co mmittee s hould know. There is no way to find out by simply asking the executive. But there is a way to find out by offering them a real choice. Suppose we offer managers a share option grant where the exercise price is indexed to the cost of equity capital less the dividend rate and we set the exercise price 10% below the market price on the grant date. But instead of giving the option to the executives we offer to sell them the options at a nominal price equal to the difference between the grant date market price and the exercise price. -64-

73 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION $100 do ice to $90 exercise pr So if the price of the stock is llars we offer to sell the option with a Black-Scholes pricing theory that the option is worth them for $10 per option. We know from ally on the order of 30% to 50% of the pr ice of the considerably more than the $10 spread (gener share). In fact we are charging them nothing for the value of the option itself. Now the executives’ choice is not so obvious. To get the option they have to put their own skin in the game. In the example we are using it would cost the executive each option, and $10 for now if they have little or no confidence in their or the company’s str ategic plan they will be reluctant to accept the offer. This is important information for the co mmittee that s hould be asking whether it wants to vest responsibility for executing the company’s strategy in a CEO or management team that is unwilling to bet that their strategy is going to create value by earning more than the cost of capital. This sale of options at a bargain price to executives also provides a way to solve the repricing problem — the pressure from executives whose options are far underwater because the firm’s stock price has fallen below the exercise price. Now the committee’s response can be quite simple. If you ategy we will issue new the executive now have confidence that you can execute a value-creating str options to you with the same terms as the last. You keep the old options but can buy new ones with exercise price 10% below the current market price, but you have to pay the 10% in cash from personal funds. Again, refusals to take this offer will reveal much about whether the executive believes in the strategy. Remuneration committees should give serious consideration to offering R-22. executives the opportunity to bet on their strategy along with shareholders by offering to sell them in-the-money cost-of-capital indexed share options at the nominal price equal to the difference between market and exercise prices at date of issue. Selling executives cost-of-capital indexed options causes executives to have skin in the game, motivates them to understand that the cost of equity capital is not zero (or the dividend yield for dividend-paying stocks), motiv ates executives to self select in or out of th vate i nformation and beliefs e firm based on their pri about their ability to create value in the firm, guarantees to shareholders that managers’ options pay off only when shareholders do better than breakeven at the cost of capital, and solv es the option re-pricing prob lem when options are far out of the money. Risk Aversion and the Cost and Benefits of Incentive Pay Finally, there are those who argue (correctly) that because managers are risk averse and unable to perfectly hedge the risks in the option packages we give them, managers w ill value the options at less than the cost to the firm of granting them. Hall and Murphy (2002), Meul broeck ill value a standard option (2000) and others estimate under reasonable conditions that managers w -65-

74 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R package at about 55% of the cost to the firm of providing it. Put another way, the firm could give a manager straight cash compensation equal to 55% of the Black-Scholes value of its options and off — assumi ng, of course, that the manager has no special information or leave them just as w ell beliefs about his ability to create value in the firm. We are, however, willing to pay managers in a less efficient way (and therefore pay them in ways that will cost us more in aggregate) if the incentive effects of paying through options cr eate bureaucratic purely enough value to more than cover the extra compensation cost. If we start at a fixed-cash-compensation (that by definition is independent of value created or destroyed) it is desirable for a firm to substitute risky outcome -based pay for some riskless fixed pay. The value- et the pay-performance relations of the compensation correct) will creation benefits (if we g generally far outweigh the risk-bearing premium we have to pay to compensate the manager. Consider a $1 billion firm whose CEO is being paid a no-risk salary of $7 million and is expected to create value in excess of the cost of capital of 1% or $10 million. Suppose we offer the CEO cost-of-capital indexed options with grant date value of $4 million in exchange for him giving up $2 million of riskless cash compensation. If this exchange is exp ected to increase incentives by an amount sufficient to generate another .2% gain in value the firm will break even. Make Unwinding Rights Explicit in Incentive Remuneration plans licitly Recognizing that we must pay a ri sk premium either explicitly or imp for risky incentive remuneration plans raises an important and often overlooked issue. Boards of directors who are sed scheme must prohibit their managers paying managers through a risky performance-ba from using financial engineering to hedge those risks away in the capital markets. If we fail to do this we incentives with value will have paid a premium to get better alignment of creation and gotten little or no benefits. We cannot allow managers and their financial advisors to eliminate or reduce the risks and along with it the incentives we so carefully designed into the plans (and paid for). In principle this is 54 all transactions in th e company’s secur ities to the easy to do by requiring managers to report board and to prohibit them from trading in contingent claims of any type on their firm. generally considered inappropriate for the remuner ation Unfortunately, in many boards it is committee or the full board to inquire into the equity or portfolio holdings of its t op-level managers. It is our belief that if remuneration policies are to be truly effective this cultural practice must change. 54 To be complete we must require the executives to re port all transactions to the company so that we can d etect the construction of composite securities to accomplish the forbidden hedging. -66-

75 E R XECUTIVE EMUNERATION Jensen, Murphy and Wruck ied for many years as to why mally restrict Indeed, we have been mystif boards do not for managers’ freedom to unwind the incentives the remuneration committee constructs for them. ting restrictions imposed by boards and co mmittees Academic researchers routinely analyze the ves that determine the point at which an executive can leave the firm and take the options or equity with 55 him. And boards regularly pay attention to these vesting provisions. Equally important, but generally unaddressed, however, are constraints on the point in time at which the executive can exercise the option (for example) and sell the shares, thus unwinding the 56 incentives. There are many reasons why we might wish to constrain an executive’s right to unwind the incentives (thereby turning the stock or options into cash) for some period of time beyond the vesting date. We have undoubtedly paid a premium to compensate for the increased risk in order to gain the incentive benefits, so we should be thoughtfully examining when it makes sense strategically to give the executive freedom to cash in and unwind the incentives and how much unwinding he or she can do at each point in time. Indeed, the unwinding terminology should become as common in considerations of executive remuneration as the vesting terminology is now. R-23. Remuneration committees should give conscious consideration to the tradeoffs associated with allowing an executive to unwind incentives, and the timing and amounts of such unwinding to be allowed. Remuneration committees should include explicit unwinding constraints (or required permissions) in executive incentive awards. They should monitor the portfolio holdings of top-level executives and related parties to ensure that they are not inappropriately unwinding the incentives that have been put in place b y the committee and the board and paid for by the company. There is at least one other reason for remuneration committees to monitor and constrain the portfolio policies of its executives. Executives regularly come into possession of insider information regarding the value of the firm’s securities. Indeed, if they are doing a good job they are creating it. While there are laws preventing ex ecutives from trading on inside information those form allow executives considerable freedom in their most restrictive laws are not uniform and even to effectively trade on such information and thereby transfer wealth from the firm’s security holders 55 See, for example, Kole, 1997, "The Complexity of Compensation Contracts", Journal of Financial Economics, V. 43, No. 1: pp. 79-104. 56 The concept of unwinding constraints is discussed in Fried, 1998, "Reducing the Profitability of Corporate Insider Trading Through Pretrading Disclosure", Southern California Law Review, V. 71: 303-392; Ofek and pp. Yermack, 2000, "Taking Stock: Equit y-based Compens ation and the Evolution of Managerial Ownership", Journal ial Power and Rent of Finance, pp. 1367-1384 and Bebc huk, Fried and Walker, 2002, "Manager V. 55, No. 3: Extraction in the Design of Executive Compensation", University of Chicago Law Review, V. 69, June: pp. 751- ). 846 (available from the Social Science Research Network eLibrary at: http://papers.ssrn.com/abstract=316590 -67-

76 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION 57 bove to themselves. However, it is well documented that corporate insiders consistently earn a 58 normal profits on their trading in shares of their firms. timates that Fried (1998) (p. 323) es corporate insiders earn almost $5 billion per year in insider trading profits on common equity alone. Whether or not such trading violates insider-trading laws, it is certainly inconsistent with the fiduciary responsibility that boards and executives have to the holders of their securities. And since these wealth gains come directly from the public holders of the securities it is well within the purview and responsibilities of the remuneration co mmittee to monitor and limit these gains. Fried (1998) provides extensive analysis of these issues and suggests the best solution to limit such inappropriate profit taking while not overly constraining the liquidity of insiders is to require a pre- latory solution, but trade disclosure of the executive’s intention to trade. He argues for a legal /regu we believe wise boards can and will do it for themselves on a voluntary basis. R-24. Remuneration committees should require pre-trading disclosure of intention to trade for all insiders as a condition of employment. Firms should adopt these policies as a way to guarantee that executives w ill not use insider information to disadvantage the shareholders to which they owe a fiduciary duty. And it is important for all effective sales to be treated this way (for example, an executive’s repayment of a company loan by transfer of any options, stock or debt security to the company). R-25. Remuneration committees should, as a condition of employment, prohibit top- level executives from trading in derivati ves of any kind, but especially those related to the securities of the firm. These constraints are required to enforce the unwinding constraints established in R-23 and insider trading constraints in R-24. 59 Problems with Annual Bonuses Annual bonuses offer several advantages over equity-based plans for providing incentives in organizations. First, equity-based plans reward managers for incr eases in stock prices but they do not in and of themselves provide managers with any guidance on how they can increase equity 57 See Fried, "Reducing the Profitability of Corporate Insider Trading Through Pretrading Disclosure", and Bebchuk, Fried and Walker, "Managerial Power and Rent Extraction in the Design of Executive Compensation", (particularly pp. 829ff). 58 Journal of Law & Economics, V. 35, No. See Seyhun, 1992, "The Effectiveness of Insider Trading Sanctions", 1: pp. 149-182. 59 Journal of This section draws extensively from Murphy, 2000, "Performance Standards in Incentive Contracts", Accounting & Economics, V. 30, No. 3, D ecember 2000: pp. 245-278 and Jensen, 2003, "Paying People to Lie: The Truth About the Budgeting Process", European Financial Management, V. 9, No. 3, 2003: pp. 379-406 ). (available from the Social Science Research Network eLibrary at: http://papers.ssrn.com/Abstract=267651 -68-

77 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R as it is in many organizations it can leave managers and values. If this issue is left unaddressed boards ignorant of what actually creates value in their organizations — although it does provide incentives for them to discover how to create v alue. The evidence cited above on the systematic mistakes that boards and managers have made in awarding too many options to too many people, and awarding too many post-retirement benefits prior to FASB 106, are consistent with our observation that managers and boards are of ten unaware of what truly creates or dest roys value in ming top management does know what creates value their organizations. In light of this, and presu provide incentives (and this is a important assumption), annual bonus programs can be structured to focused on specific operational objectives that will lead to value creation. The obvious problem here is that if the board and top management do not know what creates value, moving away from equity- based plans removes an important incentive for them to find out. ating top managers and others (perhaps Second, equity-based incentives are best at motiv employees) who can directly a ffect company stock prices, while some key engineering or technical mized to individuals or divisions throughout the performance measures in bonus plans can be custo organization. Third, the immediacy and tangi bility of cash awa rds can provide st ronger incentives ally, annual bonus plans can than distant and uncertain paper gain s in unvested equity plans. Fin contain subjective elements not easily available in explicit equity-based plans. Although well-designed annual bonus plans can hroughout an provide meaningful incentives t in practice contain important design flaws that limit their organization, most bonus plans effectiveness. They create incentives to destroy rather than increase value through shirking, value- destroying smoothing of results, or depending on the form of the pay-performance relation they can increasing create incentives to destroy value by . Perhaps more importantly the variability of results most bonus plans create incentives for the organization to destroy information critical to the effective coordination of disparate parts of large complex firms, and incentives for participants to lie and engage in other unethical behaviors. Yet, in spite of these costs, virtu ally every for-profit company offers an annual bonus plan covering its top executives that is based on a single-year’s performance. In spite of substantial heterogeneity across companies and industries, executive bonus plans can be categorized in terms of three basic components: performance measures, performance targets or budgets, and the structure of the pay-performance relation. Figure 10 illustrates these ba sic compone nts for a “typical” bonus plan. Under the typical pla n, no bonus is paid until a threshold pe rformance (usually expressed as a percentage of the target performance) is achieved, and a “threshold bonus” (usually expressed as a percentage of the target bonus) is paid at the threshold performance. Target bonuses are paid for achieving the target performance, and there is typically a “cap” on bonuses -69-

78 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Figure 10 Base Salary and Bonus for a Typical Annual Bonus Plan Cash The "Incentive Zone" Remuneration Base Salary + Bonus Cap Base Salary + Pay-Performance Target Bonus Relation Base Salary + Threshold Bonus Base Salary 0 50 Performance Threshold Target Measure Performance Performance paid (again expressed as a percentage or multiple of the target bonus). The range between the threshold and cap is labeled the “incentive zone,” indicating the range of performance where incremental improvement in performance corresponds to incremental improvement in bonuses. Poorly designed pay-performance relations majority of The pay-performance relation illustrated in Fi gure 10 is descriptive of the vast bonus plans in practice, and yet is replete with incentive problems. These plans are frequently integrated with the firm’s annual budget cycle and the targeted level of performance is one of the nvolve the The problems with these systems i major outputs from the budgeting process. counterproductive incentives that are introduced at any point the pay-performance relation is anything other than a straight line. elieve Consider the effect of the kink at the Threshold Performance level. Managers who b (or “save” profits for next year will stop producing they cannot achieve threshold performance this year by delaying revenues or accelerating expenses). The figure shows that there is no bonus -70-

79 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION penalty for missing the Threshold by a lot instead of a little. And if we see that we are not going to make it we are then better etter next period. off to take a bigger hit this period so we can do b till b elieving they can make it have incentives — Managers struggling to make the Threshold but s atever is n ecessary (including destroying value) — provided by the large threshold bonus to do wh to achieve the threshold. Because of these incentives we can predict that a disproportionate number of managers will end up at or slightly above the threshold. At the other end of the “incentive zone”, managers capable of producing well above the “cap” w ill tend to stop producing once they “max out” on their bonuses and will transfer performan ve been realized this ce results that could ha period into the next period (but often at less than a 1 for 1 basis). In most companies, the range between the threshold performance and the cap in Figure 10 is fairly na rrow, typically covering performance from 80 percent to 120 percent of target performance. More generally, “non-linear ities” in the pay-performance r elation cause predictable incentive problems, especially when managers are able to “trade -off” current for future pe rformance. When the pay-performance relation is concave (or bowl shaped) managers can increase their total bonus our discussion payouts by increasing the variability of their performance. We saw this illustrated in above of what happens when a manager is near the threshold where the kink is associated with a hurdle bonus at the threshold. On the other hand, in pay-performan ce relation is situations when the convex (or upside down bowl shaped) in the relevant range, managers have incentives to smooth performance by capping truly superior performance and saving as much of it as possible for a rainy , when managers are either increasing or reducing next period. This is the situation at the cap. Thus tem they are inevitably destroying v the variability of results to game the bonus sys alue for the organization. What started out as a system to motivate increased performance ends up motiv ating 60 counter productive behavior. Paying people to lie There is another large set of problems induced by these non-linear pay-performance relations. We’ve just described the damage caused by non-linear pay-performance relations once the targets are set. Let us move back a step to see what happens in the process that leads to the setting of the targets, threshold and caps. We focus for simp licity on the setting of the target or budget. Almost all firms go through an annual budget cycle in which lower-level managers, depar tments, and 60 And the costs can be high. Examples are legion. In one case managers intent on satisfying a sales target to earn a bonus shipped unassembled parts to a ware house near its customer in a distant c ountry at the end of the year to conclude the sale. They then had to assemble the parts at great cost to the firm in the foreign environment to satisfy the customer. Profits went down, but the managers earned their bonus. See Jensen, "Paying People to Lie: The Truth About the Budgeting Process", for many other examples of value-destroying behavior induced by these budget- linked non-linear pay-performance relations. -71-

80 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R divisions submit budgets for targeted outputs in the following year. After much conflict and negotiation final budget targets are reached and assigned. In this process managers learn that those punished by ge who tell the trut tting more demanding targets. Those h about what they can do get who can successfully low-ball the process get rewarded with less demanding targets. But there is an important casualty in this process. The information that is critical to coordinating the disparate activities of a large co mplex organization gets unnecessarily muddied or destroyed in the process. Since top-level managers know that lower level managers will lie about what they cannot do, top- level managers are led to lie about what their subordinates can do. No one thinks of these games as tell lying, it’s just a negotiation. But think about it, almost no one in the system has the incentives to the truth and reveal the cr itical i nformation that they have (or can discover) about what can and 61 should be done in the next year. itive to the controversy caused by One of the authors is less sens characterizing these bonus/budgeting systems as “paying people to lie”, and believes that eliminating such behavior in organizations can result in productivity improvements in the range of 62 50 to 100%. Although most managers and analysts understand that budget gaming is widespread, few understand the huge costs it imposes on organizations and how to lower them. The key lies not in 63 eliminating budgeting systems, but in changing the way organizations pay people. In particular to stop this highly counter-productive behavior we must stop using budgets or targets in the loyees and managers. compensation formulas and promotion systems for emp This is accomplished 61 There are situations in which it can pay managers to overstate what they can produce next peri od. For example, suppose that if I promise to produce more I can get more res ources (labor, capital, materials), and suppose further shed in the system. In that if I don’t actually produce what I promised I do not get puni this game there will be systematic overstatement of next year’s performance. 62 See Jensen, "Paying People to Lie: The Truth About the Budgeting Process", p. 390. 63 lish budgets entirely. While in some cases this may be desirable we are not Some do argue that we should abo convinced it is necessary or will work in all organizations. See Hope and Fraser, 1997, "Beyond budgeting: Breaking Through the Barrier To 'The Third Wave'", Management Accounting, V. 75, No. 11: pp. 20-23; Hope and Fraser, 1999a, "Beyond Budgeting White Paper, Beyond Budgeting Round Table", May: Fraser Hope Associates, CAM-I 1999b, "Be yond Consortium for Advanced Manufacturing, International; Hope and Fraser, Budgeting: Building a New Management Model for the Information Age", ounting-London, V. 77, No. 1: pp. 16-21; Management Acc Hope and Fraser, 2000, "Be Budgeting", Strategic Finance, V. 82, No. 4: pp. 30-35; Hope and Fraser, 2003, yond Beyond Budgeting: How Managers Can Break Free from the Annual Performance Trap , Boston, Mass.: Harvard Business School Press; Kersnar, 1999, "Re-Inventing the Budget", CFO Asia onday , July/August; Lester, 2000, "M from Management: Managers Count Blessings As Budgets Begin To Lose Currency--Some Firms Long for Freedom the Burden of Budgeting", and Thomas, 2000, "Toss Your Budget Out the Wi ndow", Business Review Irish Times, Weekly , V. 72, Sep tember 8, who summarize the experiences of a number of mostly Scandinavian companies including Svenska Handelsbanken (Sweden’s largest bank which abandoned budgets in 1970), Air Liquide, SKF, Ericsson, Skania, Schlumberger, Skandia, Swedish Post, Tetrapak, Diageo, Borealis, Volvo Cars, IKEA, and F okus Bank which have abandoned budgets or are in the process of doing so. The Consortium for Advanced Manufacturing International (CAM-I) has established a Beyond Budgeting Roundtable to understand and report on these developments. -72-

81 E R EMUNERATION Jensen, Murphy and Wruck XECUTIVE by (1) taking all kinks, discontinuities and non-linearities out of the pay-for-performance profile of each employee and manager, and (2) committing not to change the pay-for-performance profile from year to year based on budgets, prior-year performance, or any other metric influenced by managers in the current or prior years. Such purely linear compensation formulas provide no 11 provides a incentives to lie, or to withhold and distort inform ation, or to game the sy stem. Figure simple proof of why this is true. As long as th relation is linear with no kinks or e pay-performance discontinuities, a manager’s bonus does not depend on what his or her target is. If the actual performance in the figure is the same, the figure shows that the actual salary plus bonus is the same even if two very different targets were set. Therefore, managers have no incentives from the bonus target setting process to lie about what can be acco mplished in the next period, and planning can take place with more accurate information. -73-

82 E EMUNERATION Jensen, Murphy and Wruck XECUTIVE R How Linear Pay-Performance Relations eliminate gaming and budget-related Figure 11 incentives to lie Cash Remuneration Note: The actual bonus is now independent of the target or budget. Thus, budgets can be used for planning and coordination Pay-Performance ActualBonus Relation Base Salary 0 50 Performance Budget Budget Actual Measure Target #2 Performance Target #1 , discontinuities and curv ilinearity from the pay-for- The solution to the budget gaming problem: Remove all kinks performance function. Note that the bonus for actual performance in th e figure (the Actual Bonus) is the same whether the budget target was set at #1 or at the higher #2 level. Therefore, because managers are rewarded for what they produce and n o t for what the target is set at, they have no incentive to lie about what they can or cannot do in the budget setting process. from the manageme Moreover, eliminating budget/target-induced gaming nt system will eliminate one of the major forces leading to the general loss of integrity in organizations. Once taught to lie in these systems people generally cannot help but extend that behavior to all sorts of other relationships in the organization. While remuneration committees and boards are not generally involved in setting the compensation of ev eryone in an organization, si nce they are in the nvolved in resolving these end accountable for the integrity of the organization, they must be i integrity-damaging budget-based problems. -74-

83 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Design bonus plans with “linear” pay-performance relations. R-26. broad range, with Better-designed pay-performance relations are linear over a bonus banks” that allow bonuses to be very high (or non-existent) caps, and “ Bonus banks can be created in a number of ways negative as w ell as positive. over three years, where the unpaid including, for example, paying a bonus out negative bonus in the current year. bonus is available to make up some or all of a See Stewart (1990). Although linear pay-performance relations offer important advantages over non-linear a trade-off between providing linear pa relations, we recognize that there is sometimes yoffs and non-linear payoffs that have higher “leverage”. For example, in Figure 9 the trad itional stock options had a lower exercise price and hence a broader incentive zone (as defined in Figure 10) than the cost-of-capital indexed option. However, because the indexed option was less costly to the company, it could grant approximately twice as many indexed options as traditional options for the same total cost to the company. Similarly, for the same cost to the company, a firm might be able to offer (1) a bonus offering 1 percent of net profits, or (2) a bonus offeri ng 5 percent of net profits bove payable only above $100 million in profits. The latter has more leveraged incentives a $100 million, but str ictly worse in ell below $100 million, and par ticularly bad incentives right centives w around the $100 million mark. Poorly designed performance standards Bonuses are usually not, in practice, based strictly on a performance measure, but rather on performance measured relative to a performance standard. Examples include net income measured relative to budgeted net income (as we discussed extensively above), earnings per share (EPS) vs. last year’s EPS, cash flow vs. a charge for cap ital, pe rformance measured re lative to a pee r-group, or performance measured against fi nancial or non-financial strategic “ milestones.” Pe rformance standards typically correspond to “expected precisely, the level of performance” or, more performance required to attain the executive’s “target bonus.” The efficacy of alternative performance standards depends on the extent to which managers can influence the standard-setting process: managers can increase bonuses either by taking actions that increase the performance measure or by taking actions that decrease the performance standard. For example, when standards are based on prior-year pe rformance, managers will tend to avoid unusually positive performance outcomes, since good c urrent performance is penalized in the next period through an increased standard. Similarly, when standards are based on meeting the pre- determined company budget, managers have incen tives to negotiate easy budgets and to avoid actions this year that might have an undesirable effect on next year’s budget. Standards based on the performance of co-workers create incentives to sabotage co-worker performance or collude and -75-

84 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R provide collectively shirk, while standards based on the performance of an industry peer group incentives to select “weak” industries and peers. In spite of their obvious problems, most companies use standards that are readily influenced by managers. For example, Murphy (2000) finds that 177 large US companies 89% of a sample of base standards on budgets or prior-year performance. These issues are so important that the board etting general policy that limits the counter- nvolved in s and remuneration committee should be i productive effects of these systems. The integrity of the company depends on it. R-27. Avoid internally influenced performance standards Internally influenced performance standards ar e those where the bonus-plan participants can take actions (often value destroying) that increase bonuses by reducing the standard rather than by improving performance. pay-performance relation, some of the problems with budget- In addition to linearizing the the performance standard; that is, by basing based bonuses can be mitigated by “externalizing” standards on objective measures beyond the direct control of managers (although managers will s till have incentives to use value-destroying means to achieve the standard). For example, in LBOs and laced by an ob other highly leveraged organizations, the objective of “making budget” is rep jective of generating sufficient cash flow to service the debt. Combined with the large equity holdings of LBO managers and directors, this yields a pay-performance relation that has only one non-linearity (at the point of default and bankruptcy). An effective way to understand this critical advantage of the LBO/MBO organization is to see the debt negotiation with the outside supplier of credit as “externalizing” the budget process. Similarly, to the extent that budget-based internal control systems play a more im portant role making budget is reduced following spin-offs and in large diversified corporations, the focus on clical divestitures. Relative performance incentive plans, increasingly popular in utilities and cy industries, replace “making budget” with “bea ting the industry.” But when boards use such external standards they must realize that once such beat-the-industry standards are put in pl ace, the board rather than the management team mu st retain the decision rights over what i ndustry the firm is to be in. Not doing this can s e management team doing w ell by staying in a imply result in th flawed industry and destroying less value than its competitors. This can delay the compe titive adjustment required to move resources to more highly valued uses and thereby dest roy so cial value as well. Similarly, in companies or divisi ons measured by EVA, budget-based ob jectives are replaced by generating cash flows in excess of the cost of capital. Incentive plans such as Disney’s well- designed but now defunct plan that paid its CEO 2 percent of income over a fixed threshold (which -76-

85 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION was approximately 11 percent of the book value of net assets), also externalize and makes linear the performance standard. In all these examples, th nue to use budgets for planning e organizations conti and communication purposes, but stop short of usi ng budgets to define performance targets and determine rewards. Poorly designed performance measures Business history is littered with firms that got what they paid for. At the H.J. Heinz Company, for example, division managers r eceived bonuses only if earnings increased from the prior year. The managers delivered consistent earnings growth by manipulating the timing of shipments to customers and by prepaying for serv ices not yet r eceived (Post and Goodpaster (1981)). At Dun & Bradstreet, salespeople earned no commission unless the customer bought a larger subscription to the firm’s credit-report services than in the previous year. In 1989, the company faced millions of dollars in lawsuits following charges that its salespeople d eceived customers into buying larger subscriptions by fraudulently overstating their historical usage (Roberts (1989)). In 1992, Sears abolished the commissi on plan in its auto-repair shops, which airs authorized by customers. Mechanics misled paid mechanics based on the profits from rep customers into authorizing unnecessary repairs, leading California officials to threaten to close 64 Sears’ auto-repair business statewide (Patterson (1992)). In each of these cases, employees took actions to increase their compens ation, but these actions were at the expense of long-run firm value. The problem is illustrated succinctly by the title of Steven Kerr’s famous 1975 article, “On the folly of rewarding A, while hoping for B” (Kerr compensation and incentive systems can be (1975)). Indeed, many examples of dysfunctional traced to inappropriate performance measures. Piece-rate schemes, for example, provide incentives to increase quantity at the expense of quality. Bonus plans based on net income provide incentives to increase accounting profits while ignoring the cost of capital and thereby destroying value. Plans on equity, ret urn on assets, based on “returns” (e.g., return etc.) provide incentives to pursue only the highest return pro urns. Returns are a ject, ignoring profitable projects earning slightly lower ret particularly interesting performance measure because they are an example of the pitfalls associated with ratios as performance measures. Our rule is: 64 See also Baker, Gi bbons and M urphy, 1994, "Sub jective Pe rformance Measures in Op timal Incentive V. 109, No. 4, November 1994: pp. 1125-1156. Contracts", Quarterly Journal of Economics, -77-

86 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Do not measure performance anywhere in an organization with ratios. Simply R-28. put: If it is a performance measure and a ratio, it’s wrong often be made to work by appropriately Ratio measures of performance can but this is almost never done. For changing the decision rights of the agent, example, return on assets can be made to work if the agent is given only the right to decide which assets to use, not the quantity of assets. Although almost never directly measurable, the “perfect” performance measure is the individual’s contribution to firm value. While we cannot offer precise guidelines on how to identify the most appropriate measure in offer some general “guiding principles” to any situation, we can help design better plans. R-29. Use “line-of-sight” performance measures when possible and give each employee the decision rights to do their job. This prescription actually involves several dimensions. To provide incentives, able to affect the pe rformance measure, and also must employees must be they can affect the performance measure. To be able to affect the understand how performance measure they must have the appropriate decision rights to do so. Incentive compensation can’t be effective unless employees are given s ecision ufficient d rights to exercise discretion over the performance measures. Moreover, even seemingly perfect performance measures will fail if employees don’t understand how their actions affect the bout performance measures. (Tying pay to the measure, however, will induce employees to learn a their potential effect.) R-30. Use performance measures that reduce compensation risk while maintaining incentives. Since employees “charge” to bear compensation risk, performance measures that reduce risk without reducing incentives increase efficiency and company profits. However, when risky compensation is an add-on to current compensation there is no need to further compensate managers for that added risk. sts an important trade-off between using narrow and broad This guiding principle sugge performance measures. Narrow measures — such as measures based on individual performance or short-term performance — have low risk but f ail to capture potentially important interdependencies (i.e., how a person’s current actions affect the performance of others in the firm or in future years). Broad measures — such as measures based on team, group, or division performance or long-term performance — capture more interdependencies but are often riskier. -78-

87 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Pay particular attention to the choice of group versus individual performance R-31. measures. When there are substantial interdependencies in productivity between the actions of two or more people or groups, define the extent of the performance measure to incorporate the interdependencies. Using individual performance measures in onflict, lack of c situations where cooperation is important w ill create c ooperation nciple applies to the choice of the and reduced performance. And the same pri time interval over which performance is measured. Bonus plans should include a subjective component. R-32. There are often no low-risk objective measures of the individual’s contribution to firm value. The objective measures that exist often are too risky (i.e., based o n factors that are highly variable), provide insufficient direction (the employee might not know how to affect the measure), or provide incentives to do the w rong thing. However, even when no appropriate ob jective measures are av ailable, supervisors or managers can often assess an individual’s contribution to value subjectively. Subjective assessments can also be used to reduce the noise in good objective measures, to reduce the “distor tion” in bad objective measures, and can also adjust bonus payouts for unanticipat ed shocks (such as terrorist attacks or shocks to world oil prices). Finally, every bonus plan should include a subjective component if for no other reason than to prevent value destruction by those who game the sys rformance measure is subject to tem. Every pe gaming. Every bonus system s for den ial or modification of a bonus that is earned by allow hould inappropriate actions, actions that harm others, or actions that punch a hole below the waterline. R-33. Every bonus system including option and other equity-based programs should provide for recovery of rewards (clawback) if and when there is future revision of critical indicators on which bonus payments were based or received. This clawback should include, but not be restricted to, amounts due b ecause of formal restatements of accounting numbers such as earnings or revenues. Moreover, provisions should be made whenever possible to recover the amounts from bonus banks, deferred possible to recover th e amounts directly. In the payments or retirement benefits when it is im absence of these clawback provisions we are unintentionally rewarding and there fore providing incentives for people to lie and game the system. There are many challenges involved in incorporating subjective assessments in reward systems. First, no one likes receiving unfavorable performance evaluations, and few managers enjoy giving them. And when they do get them, individuals have ways of imposing costs on those who give such negative feedback. But without fee dback on one’s mistakes, no one can effectively learn. Given all this, it is not s urprising that managers dislike giving unfavorable performance appraisals and avoid making distinctions among employees based on performance. (And we have no trouble in predicting that members of remuneration committees are no different in this respect.) In addition, in most organizations managers have few incentives to make t horough onducting careful assessments, since there is often little reward for giving careful appraisals. And c -79-

88 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION activities. (In these performance appraisals takes time away from other more highly rewarded ppropriately situations the performance measures of the manager doing the assessments are not a weighting the performance appraisal function—one of a manager’s most important duties, and yet a common problem in organizations.) Also, employees often don’t trust managers to make unbiased assessments, or don’t believe managers possess sufficient information to make a ppropriate assessments. ooner or later in the r emuneration co mmittee’s assessment of All the above issues come up s the performance of senior executives. And here the stakes and pride are extremely high so it is not surprising that part time people (directors) are often unwilling to devote the time and the personal effort and courage to provide accurate, frank and effective performance a ppraisals of CEOs and other top managers. contract by opportunistically Finally, managers are often tempted to br eak the performance refusing to pay large bonuses to those whose performance warrants it — on the basis that “no one deserves that much”. These problems suggest that employee-employer trust is crucial in ation incorporating subjective performance assessments in bonus plans, and that remuner attention to precedent and the long run effects of their committees and managers must pay careful hout trust we current assessment and reward decisions. One casualty of failure here is trust. And wit will have to pay more for risk. R-34. Encourage managers to build and preserve trust. Because precedent matters we must beware of too much tinkering with the system. Hold managers accoun table for the long-run eff ects of their performance evaluations. Encourage them to pay particular attention to the destruction of trust, and the perceived insecurity of contracts, promises, and commitments regarding bonuses and performance measurement when the rules of the game are changed too often by “too much tinkering” with the system. It is commonly believed that managers should not be held accountable for factors that are beyond their control. This often leads to excuses, irresponsible behavior, and extremely poor incentives. Managers should be held accountable for factors that are out of their control, when the manager can control or affect the impact of those non-controllable factors on performance. -80-

89 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION Managers should be held accountable for factors that are beyond their control R-35. if they can control or affect the impact of those uncontrollable factors on performance. For example, we would be foolish to ignore the affect of gasoline prices on the performance of a manager of a fleet of vehicles. We want that manager to be cognizant of what will happen to the cost of running the fleet if the price of gasoline goes up or down substantially, and there are certainly things such a manager can do to help the firm prepare for and to adjust efficiently to major changes in gas prices. Holding the manager accountable for the effects of changes in gas prices will motiv ate him to be creative in managing the impact of those uncertain changes. 65 Strategic Value Accountability and Remuneration Policy VI. Tensions Between Outside Markets and Internal Management Recent events reveal pressures impinging on top management teams and boards of directors from both the capital markets and the internal manager ial organization. These two groups speak different languages and top management and the board must manage the tensions between the two. The capital markets speak the language of financial results and value creation, paying scant attention to the underlying operational decisions and strategy that drive those results. Managers, particularly ppropriately, on the those below the executive ranks, do not understand that dialect. They focus, a ssues surrounding the management of the enterprise and the day-to-day and week-to-week i execution of its strategy. One “party,” the capital markets, effectively ignores the other “party,” operating management. And the latter finds the former almost unintelligible. Each perceives the firm and its purposes in very different ways and evaluates performance according to their specific worldview. Neither is universally right on these issues. The capital market’s perspective held the upper hand in this tug of war in the last bubble. And in this envi ronment many firms resorted to a damaging practice — using the market’s consensus earnings forecast as the primary input into their budgeting process. We believe it is a propitious leading firms to br eak this cycle of time for dependence to put more balance back into th e relations between intern al management and external market forces. The last several years demonstrated that many of the standard solutions for the agency problem failed. Providing equity-based incentives based on stock options grants failed to align executives’ interests comprehensively with those of their firms in this period of generally 65 This section draws on ongoing work by Joe Fuller and Michael C. Jensen. -81-

90 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R ement overvalued equity. Extending those programs to the second, third and fourth level of manag operating manag ement to act in the principals’ best did not in the end provide proper incentives to interest. rm’s internal organization al structure and policies New thinking is required that integrates a fi 66 (what we call organizational strategy ) with its external competitive strategy, linking them in a mutually supportive way. This will require us ing more specific local measures of performance than changes in share price or firm value in gauging how an individual executive influences outcomes. That suggests identifying the few, specific decisions or actions within any given executives purview that truly influence outcomes for the firm. This means that firms should design incentives in the internal managerial organization based on those four or five classes of decisions taken within each ccess and important level or unit of the organization that make a difference between strategic su failure. Rather than relying on legitimate but insufficiently specific measures like im provement in firm value, a company should design metrics for operating manag ers around successfully executing those high choices in a “high qu ategic outcome ality” fashion (i.e., a manner consistent with a str that supports the creation of value). The pri hould of every organization s ncipal design logic therefore revolve around identifying those decision-making processes and ensuring that the organizational rules of the game support making “high quality” decisions. strategic thinking for companies w ill shift from “top In doing this, we believe the logic of up.” Companies down” (or, perhaps more appropriately from outside analysts in) to “bottom- should create incentives for executive teams to cr eate assets of lasting value — not only productive and financial assets, but also intellectual a nd human capital. In some sense, this s uggests that Total Quality Management type logic be mapped more explicitly into executive decision-making. In that sense, executive compensation should roll up from the bottom not down from the top or in from outside capital markets. General management should be assessed in terms of the outcomes they generate for firm value through their su ccess in overseeing their organization’s eff ective implementation of an associated strategic logic. Viewed in this way, one could portray the role of top management as supporting value creation in the organization they supervise, rather than imposing it on that organization using the cudgel of capital market’s discipline. 66 , Cambridge, MA: Harvard University Press. See Jensen, 1998, Foundations of Organizational Strategy -82-

91 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION The Critical Importance of Strategic Value Accountability Accomplishing the appropriate integration and isolation between the internal manag ement delicate and di fficult matter. We call this task the organization and the external capital markets is a management of Strategic Value (for lack of a better label at this time). Simply put, the cr itical function we are highlighting is not just the formulation and implementation of the firm’s competitive and organizational st rategy. Strategic Value Acc ountability is the accountability for the ultimate value consequences of the following activities and choices: 1) estimating the long-run value consequences of the alternative strategic choices an organization creates, 2) motivating and managing a unique choice among those alternatives, and 3) implementing (and revising) that choice over time. Strategic Value Accountability must be assigned to a person or a small group if an organizations’ governance and control system is to be comp lete. We frequently observe executives and boards who are comfortable with steps 1 through 3, but recoil from taking accountability for the value consequences of those choices and actions. The expressed concern is that no one can tell now when and how the capital market will value the choice and implementation of our strategy, especially if it is likely to be years before the fruits of the effort are revealed. One manifestation of this reluctance is reflected in the common complaint that there is no easy or straightforward way to measure the contribution to firm value of a person, group, unit, or division. The oft-recommended solution then is to avoid the problem by es or performance by something timating the contribution we can observe today, say earnings per share, or cash flow, or sales, or margin. The fact is that these value estimates have to be made and in a well-run organization there will be someone or some group that is fully accountable for managing this task. It seems to us that there are only two candidates, the CEO (perhaps in a ssociating with a small number of top managers) or the Board of Directors. And those organizations th at do not face up to th e necessity for some en tity to be accountable for the value consequences of the choices regarding ategy, organizational str competitive strategy, and implementation will be at a serious disadvantage. A key challenge in implementing Strategic Value Accountability is deciding how to measure and reward the performance of the person or en tity that is guiding the formulation and execution of the firm’s organizational and competitive strategy. That person or entity cannot be measured on the degree to which it meets the strategic benchmarks that it sets without putting us back in the budget/target setting game (discussed above) in which we pay people to lie and destroy value. Indeed, we believe the only way to measure and reward this person or entity is t hrough the r ealized execution. And this means that the uncertainty long-run value consequences of the strategy and its -83-

92 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION ure and how they will do it will be born by this regarding what the markets will reward in the fut person or entity. We believe few boards or management teams consciously address the issue of who w ill be accountable for these value consequences. And often that has meant that no one is accountable. This sadvantage. Ther leads almost surely to a competitive di forms that have e are organizational executed this function better than others. Their growth and value creation over the last three decades reflects their comparative advantages in strategic value accountability. (Although many have suffered poor results in the post bubble years, ther e have been few of the catastrophic failures seen in the publicly held sector.) LBO Associations, Venture Capital, and Private Equity Funds are organizational examples where strategic value accountability is generally nizations th e decision well executed. In these orga rights and accountability for managing relations with the capital markets generally lies with the 67 board. The KKR’s or Kleiner Perkins of the world are experts in valu ation and in dealing with the capital markets. They play the major role in the interface with the players in these markets. While they respect capital markets they do not slavishly follow their every whim. They also have expertise in the formulation a nd execution of organizational and comp etitive str ategies in their operating organizations (obviously all of which must be comp lemented by the ability to choose highly talented managers with deep specific knowledge of their businesses). The partners in these innovative organizations then take the strategic value accountability for creating governance sys tems ers and for integrating and coordinating the that effectively work with the operating manag competitive and organizational strategies of their operating companies with the capital markets. In these organizations long run value (as repre sented by IPO values, or the value str ategic buyers will pay for the entity) is the measure of performance. The rewards for the partnership sponsible for the headquarters re tem and its value results are dir ectly related to the governance sys value created in the portfolio firms. For example, LBO Association partners regularly receive a 20% held by their buyout funds. Thus they r eceive override on the 60% to 80% of the firm’s equity eated in the venture. Venture capital f unds are roughly 12% to 18% of the increased equity value cr similar. 67 For a description of the typical LBO Association as an innovative model of management and governance see Jensen, "Active Investors, LBOs, and the Privatization of Bankruptcy"; Jensen, 1989b, "Eclipse of the Public Harvard Business Review, V. 67, No. 5: pp. 61-74 (Revised 1997) (av ailable from the So cial S cience Corporation", Research Network eLibrary at: http://papers.ssrn.com/Abstract=146149 ). For a description of the workings of the typical venture capital organization see Sah lman, 1990, "The Structure and Governance of Venture-cap ital V. 27, No. 2: pp. 473-521. Organizations", Journal of Financial Economics, -84-

93 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION In the typical publicly held corporation it is often not clear where the Strategic Value Accountability lies or should lie. In many corporations it is not assigned to or accepted by anyone, and this is a recipe for underperformance and value destruction. Sometimes it is best held in the in other cases wher hands of the CEO or a small group of top-level managers, t financial e the relevan and value expertise lies in the board of directors it is best held there. Let’s deal with a simple case to illustrate the thinking involved here. S uppose we have a CEO whom the board believes is deeply knowledgeable about the organizational and strategic imper atives the firm faces. He is then a natural to lead the strategy formulation and execution process. He leads the initiatives that produce relevant strategic milestones and how to motivate the organization to achieve them. But we cannot measure his performance and base his rewards on whether the strategic milestones are achieved. If we do so we are back into the paying-people-to lie syndrome with all the gaming and associated value destruction that entails. In this situation, the solution is to measure and reward the CEO wh o has accepted the Strategic Valu e Accountability with substantial equity or equity-like claims that are sufficiently long term so as to avoid the temptations for short- term manipulations. This means that the unwinding constraints on the CEO’s incentives must be explicitly a part of the remuneration system and they must be rigorously enforced. for the board to manage This solution should work well in normal times, but will be difficult if the firm becomes substantially overvalued for all the reasons we enumerated in the section a bove on the agency costs of overvalued equity. As we said above, in ma ategic value accountability is not formally ny cases this critical str nyone in the manag assigned to, or accepted by, a and governance structure. These situ ations ement tend to be those where the responsibility is de f acto vested in the CFO and/or director of investor relations who usually are in no position to w ield the n ecessary power and influence over the CEO, Board and operating organization to get the job done properly. They simply do not have the O and Board, are not taking on the strategic value decision rights to do so. And those that do, the CE accountability. This situation is classically asso ciated with the state where the management and governance structure attempts in various ways to deny the relevance of th e capital markets to the firm and anyone’s accountability for value creation — often under the mantra that holders of its securities are not long-term investors and therefore should be ignored to one degree or another. The issue is not whether some or all security holders should be ignored or catered to. The issue is whether the governance sys tem of the organization has faced up to the difficult and demanding task of assigning the Strategic Value Accountability to individuals or entities in the system who have the talent, desire, and proper rewards and punishments to get the job done as well as it can be done. -85-

94 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION accountability is not a (or is vaguely Those organizations in which strategic value ssigned assigned) typically have performance measurement and reward systems that are nominally based on value creation (say with equity or options) but are vitiated when the value consequences turn out are but one example of multitudes of ways in badly. Repricing and reissuing underwater options which remuneration committees avoid holding anyone accountable for the value consequences of the organizations actions and experience. R-36. The remuneration committee should take the lead in seeing that Strategic Value Accountability is clearly assigned to those who have the unique combination of business judgment, financial knowledge, wisdom, and willingness to take on the critical task of managing the interface between the operating organization and the capital markets so as to create value. Let us be clear that the assignment of the decision rights for managing relations with the capital markets is much more than simply talking to investors and institutions to assess their interests, opinions, desires and advice. It goes to the core of what it means to direct the organization so that choices are made that will maximize the chance of competitive success and the efficient use of society’s scarce resources (human, capital, technol ogical and material) entrusted to the organization. Remuneration committees must c onfront these issues. The committee must see to it that this talent and capacity is recruited into the organization and retained. They must see to it that those who have accepted the Strategic Value Accountability task are held to the value consequences even when they turn out poorly. The importance of risk and trust in management and governance major If one looks at the root causes of the epic failures in governance that befell a number of global companies recently, risk is a common denominator. Boards failed to comprehend the risks its unbridled acquisition being run by WorldCom in nron in its undisciplined campaign, by E investments in international assets and special purpose entities, by Vivendi in assuming that synergy would arise from a collection of assets in rela ted, but nonetheless, distinct industries. In those instances and a host of less celebrated cases, the perceived incentive and reward sys tems rewarded managers for minimizing or obscuring the operating risks incurred by stretching to reach budget goals. And they often did not hold managers accountable for the value destruction that occurred. Top-down processes tend to motivate lower managers to take risks with the enterprise’s future competitiveness in order to meet the current performance demands (as set by senior management and its dialogue w ith the capital markets). The “cascade” of executive compensation must account for this. Boards must encourage and comp ensate senior manag ement to investigate and take a conscious posture against unwise risk taking, and management must ensure that the traditional budgetary target and bonus process does not destroy value by encouraging the assumption of unmanaged or hidden risks. -86-

95 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Relations with Capital Markets: The Earnings Management Game VII. The publicly held firm and its management team are subject to substantial rewards and penalties from the capital markets. The reactions of securities analysts, investors and finan cial institutions to managements’ policies and announcements play an important role in the determination of stock prices, the ability of the firm to raise debt and equity capital, and therefore on the success or failure of its competitive strategy. Therefore it is important for the remuneration team committee to understand the complex dynamic relationship between the t op-level management and the capital markets in some detail. While the board as a whole must play a role in ratifying the firm’s strategy for dealing with the capital markets, the remuneration committee must take that ation strategy and the forces of the capital market into consideration in formulating its remuner purpose in this s policies for the top management team. Our ection is to review the more important ital markets s nformation flows between the aspects of the relationship with the cap urrounding the i firm and the markets. How Markets Reward and Punish Managers We’ve explained in Section V how manager budget-based remuner ation s in typical annual systems have pay-performance relations that are non-linear, kinked or discontinuous. We also summarized how these systems induce managers into gaming the system in ways that destroy nformation required to coordinate dispar value; both by destroying the cr itical i ate parts of large complicated organizations and by taking actions that destroy value in order to meet their budgeted targets. Interestingly, the relation between a firm’s management team and the capital board and top- quilibrium that replicates many counterproductive as pects of budget or markets has resulted in an e target-based bonus systems. When a firm produces earnings that beat the consensus analyst forecast the stock price rises on average by 5.5%. For negative earnings surprises the stock price 68 falls on average by –5.05%. Interestingly the stock price rises by 1.63% for zero surprises. 68 See Skinner and Sloan (2002), p. 297. Note that the average stock price response is an abnormal quarterly stock return (i.e., adjusted for the performance of a size-mat ched portfolio) for the 1984-1996 pe riod. The figures presented by Skinner and Sloan appear to be larger in magnitude than the findings reported by Bartov, Givoly and Hayn, 2002, "The Rewards to Meeting or Beating Earnings Expectations", Journal of Accounting & Economics, V. 33, No. 2: pp. 173-204 (available from the Social S cience Research N etwork eLibrary at: http://papers.ssrn.com/paper=247435 ), who report that when initial forecasts are pess imistic, beating the final consensus forecast yields a 3.2% higher quarterly abnormal return than for firms that fail to meet the final consensus -87-

96 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Figure 12 Market-Adjusted Returns for Growth & Value Firms in Response to Quarterly Earnings Surprises Note: The graph plots quarterly abnormal returns for growth and value firms as a function of earnings surprise at the end of the quarter. Forecast error is from ata is elative to the quarter-end stock price. D measured as the earnings surprise r I/B/E/S database for the final month of the fiscal quarter for which earnings is being forecast. Sample size i s 103,274 firm-quarter observations in the period 1984-1996. Source: Skinner and Sloan (2002), p. 299. Thus there are substantial rewards for firms that meet or beat the analysts’ consensus forecast and penalties for those that miss it. In addition, as Figure 12 shows, the relation between the magnitude of the quarterly abnormal stoc urprise has the same k return and the quarterly earnings s general ramped shape as bonus pa y-performance r elation ear managerial the typical non-lin illustrated in Figure 10 (p. 70). Figure 12 also shows that growth stocks (i.e., stocks trading at a high market-to-book ratio) react more strongly to earnings surprises of a given magnitude than non-growth stocks (what the profession calls value stocks). Beating the analysts forecast by 1% (measured by the ratio of the bnormal stock price increase for 10% a ter-end stock price) leads to a earnings surprise to the quar 15%. While the growth stocks. But missing the analyst forecast by 1% leads to a pr ice decline of shape of the price reaction for non-growth stocks for any given is similar the magnitude is smaller error. This makes sense given that earnings farther into the future represent a larger fraction of the value of a growth stock than for non-growth stocks. forecast. Their study covers a somewhat different period, 1983-1997, uses a different methodology for computing abnormal returns (a beta-adjusted return), and they present their findings in the form of the expectations-surprise path. -88-

97 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION This simple empirical regularity — a highly non-linear relation between quarterly earnings surprises and abnormal stock return — ement to focus on provides strong incentives for top manag earnings and surprises. Thus stock-price reactions to earnings announcements reward and punish managers in several ways. Favorable stock-price reactions reward firms and managers by making capital available to them at lower cost in both the debt and equity markets. And to the extent managers have equity-based compensation such as options, stock, or restricted stock they are directly rewarded or punished through changes in th eir wealth. Thus, while no one exp licitly designed it this way, the earnings game managers are i nvolved in with analysts cr eates kinked pay- performance relations and therefore creates incentives late earnings to meet for managers to manipu or beat the analysts forecast, to manipulate analysts forecasts, and to take actions to meet those forecasts even when those actions destroy long run value. Ethical and Value Consequences of the Earnings Management Game Indeed, for more than two decades it has generally been understood that part of the job of every top-level manager has been to “manage earnings”. What people have not generally f aced up to are the ethical issues this practice raises and the l ong-run value consequences that managing earnings engenders. actions that are anything other than Let’s be clear: when “managing earnings” means taking those required to maximize the long-term value of the firm, managing earnings amounts to lying. And it amounts to lying to the very stockholders or potential stockholders to which managers have a ecame common in fiduciary responsibility. We use this strong language because these practices b ealization or the culture of top management and boardrooms with almost no discussion, r confrontation of the ethical or value issues i nvolved. Indeed, these practices were already widely ensconced in the worldwide management culture as a result of the almost universal prevalence of the 69 budget-based bonus and promotion systems discussed a Hence, one important potential bove. control mechanism, the integrity of managers and board members on this issue b ecame powe rfully and effectively disabled in a way that was virtually invisible to those who were involved in the system. Simply put, “it was just the way things were done”, and few could imagine doing it any differently. It’s time to change. And the change probably has to start with the remuner ation committee and the board taking action to invest in th e integrity of their orga nization and its relation to the capital markets. 69 I/B/E/S (Institutional Brokers Estimate System) began compiling consensus earnings estimates in the early 1970s (followed by competitors First C all, Zacks, and Nelson’s). And when firms began using those estimates as targets in their budgets and management bonus plans, executives’ motivation to meet or beat the analysts’ forecasts elevated so that top managers and analysts became even more tightly linked. -89-

98 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R How Managers Reward and Punish Analysts The earnings management game, which itself is an agency problem within corporations, as well as in investment banks and the analyst community, extends beyond the manipulation of currently reported earnings to managing the analysts to affect their forecasts. This happens when managers reward analysts who make forecasts they prefer and punish those who issue forecasts they do not like. Such rewards and punishments are illustrated by the fre quent practice of providing ss to information about the firm — a practice that cooperative analysts favorable treatment in acce has since been made illegal by the SEC’s October 2000 Regu lation Fair Disclosure (FD) — as or other services to firm’s whose analysts well as granting lucrative commissions for banking cater to management’s preferences. And when investment banks or other financial institutions cave in to from clien such pressures or when they request such favorable treatment ts or potential clients they are indulging agency problems of their own and damaging their reputations and integrity. Prior to Regulation FD, private communications between research analysts and the companies they were covering were commonplace. In one instance, a Bear Stearns banking client emailed a a downgrade of the client’s stock, “Y Bear Stearns analyst following our earnings estimates are on track, however, given the downgrade, I sure would have liked to see you give us a lower bar on timate, they were defin revenue...[W]hile we affirmed the revenue es itely a stretch. Seems a shame to waste a downgrade by not buying the opportunity for us both to ove r-perform going 70 forward...” Unfortunately, there is substantial evidence th at the collusive cooperation requested and expected in this communication to the Bear Stearns analyst commonly took place. The private exchanges between analysts and company executives, dubbed guidance, were the subject of a survey conducted by the National Investor Relations Ins titute in 2001. Guidance includes, for example, company review and critique of analysts’ spreadsheet models. Making use of the survey results, no-guidance Hutton (2003) studied analyst forecasts for 457 guidance firms and for another 59 firms over the 1998-2000 period. Where firms provided guidance, analysts’ quarterly forecasts were downwardly biased (therefore rewarding ma nagers by giving them the ability to show pos itive earnings surprises) but more accurate (as measured by lower mean-squared error) than firms that provided no guidance. Given that analysts are rewarded in part for accuracy, it is not s urprising that 70 SEC, 2003, "Securities and Exchange Commission v. Bear Stearns & Co., Inc.," U.S. District Court (Southern District of New York) , ¶ 67. -90-

99 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION 71 accuracy. Both the analysts and the manager’s win — analysts would trade off unbiasedness for in the short run — while their ultimate clients, shareholders, lose. It became general practice for investment banking firms to rely on their analysts to help for doing so. For example generate lucrative new banking business and to compensate the analysts 72 in 2000 Credit Suisse First Boston offered extra pay (in addition to salary and a cash performance bonus) to a prosp ective analyst for helping the firm win stock and high-yield debt transactions. Depending on the level of contribution in winning the business, the analyst could earn 1% to 3% of the firm’s net profit per transaction up to a capped amount of $250,000. Some analysts reportedly were rewarded for a half-dozen such transactions annually. drop analyst coverage was another common tactic used by inves tment banks to The threat to strong-arm companies to give them their banking business — a practice they never rev ealed to the investors who were using the bank’s investment analysis and recommendations. In another internal Bear Stearns email, dated April 3, 2000, one inves tment banker wrote to discuss a banking client’s decision to drop Bear Stearns from a follow-on offering. He wrote, “I expressed signif icant disappointment with the fact that they neglected to discuss this issue with us prior to this time and that they left us no choice but to drop research coverage and tradi ng, since they obviously did not 73 value our support to date.” In the wake of US financial scandals, analyst behavior has justifiably come under intense recommendations (that did not scrutiny. It is believed that highly fa vorable analyst coverage and reflect the privately expressed opinions of the analysts) played a role in heightening and prolonging the recent bubble and the enormous losses that followed. A recent investig ation of ten prominent $1.4 b investment banking firms resulted in the so-called illion (in fines and penalties) Global 74 Research Analyst Settlement. In addition, two well-known “star” analysts, Jack Grubman (formerly of Salomon Smith Barney) and Henry M. Blodget (formerly of Merrill Lync h), paid fines of $15 milli ectively, and were permanently barred from the US on and $4 million, resp 71 See Hong and Kubik, 2003, "Analyzing the Analysts: Career Concerns and Biased Earnings Forecasts", Journal of Finance, V. 58, No. 1: pp. 313-351 for add itional information about analyst for ecasts and career rewards and punishments. 72 Credit-Suisse example from Gasparino, 2002, "Analysts' Contracts Link Pay to Deal Work", Wall Street Journal, May 6, p. C1. 73 SEC, "Securities and Exchange Commission v. B ear Stearns & Co., Inc.," ¶ 54. Ultimately, in this case the Bear Stearns analyst, despite pressure from the Head of Research, did not drop coverage of the unnamed company. 74 The ten firms that were investigated and agreed to pay penalties as part of the settlement agreement are Bear, Stearns, Salomon Smith Barney, Credit Suisse First Bos ton Corporation, Goldman, Sachs, J.P. Morgan Securities, Lehman Brothers, Merrill Lynch, Pierce, Morgan Stanley, UBS Warburg, and Bancorp Piper Jaffrey. See http://www.sec.gov/spotlight/glbalsettelement.htm -91-

100 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION securities industry. The investigation focused on the relationship and dealings between the investment bankers and their research analysts . Each of the investment banks was charged with violations of various NASD and NYSE rules. Conflicts of interest between inves tment banking and research, and unwarranted or exaggerated claims were often among the claims leveled by the SEC. Evidence on the Collusive Nature of Earnings Forecasts and Realizations icious set of phenomena that bear The earnings management game has produced a susp further research, investigation and explanation. In 1997, for example, Microsoft had equaled or beat 75 analyst estimates in 41 out of the 42 quarters since going public. If analysts were just as likely to over-estimate as under-estimate a company’s quarterly earnings and there were no collusion or gaming by either Microsoft or the analysts, the likelihood of the company outperforming the 76 Indeed, in 2002 the SEC forecasts in 41 out of 42 quarters would be less than 1 in 100 billion. investigated Microsoft for lying about its earnings (in this case Microsoft used ina ppropriate 77 Microsoft reserves to report earnings that were smaller than would otherwise have been reported). many managers in typical budget-based bonus sy stems appears to have been behaving much like that wish to hide very good results because of a fear that others will simply raise the bar for the future. But the data seem to i yond ndicate that the practice of manipulating earnings extends far be just a few companies. Research shows that analysts tend to issue systematically positively biased forecasts of one- year-ahead future earnings (See Figure 13). These forecasts then are revised down sys tematically over time to more closely approximate the actual realized earnings announcement for the year. Indeed, prior to 1992, by the end of the forecast year, analysts’ earnings forecasts would equal actual earnings and thus show a zero forecast error. This is sensible because if the error did not end up at zero analysts could easily take the bias into account, and thereby eliminate it. 75 port Fox, 1997, "Learn to Play the Earnings Game (and Wall Street will Love You). The Pressure to Re Smooth, Ever Higher Earnings Has Never Been Fierce r. You Don't Want to Miss the Consensus Estimate by a Fortune , V. 135, No. 6, March 31, p. 76+. Penny—And You Don't Have To", 76 underestimate earnings Given the simplistic assumptions in the text above, the probability that analysts would 42  (1/2) . in 41 out of 42 quarters would be 42 77 As documented in the SEC’s cease-and-desist order, Microsoft “...failed to maintain internal controls that were adequate under the federal securities laws. Specific ally, during the relevant period, Microsoft maintained between approximately $200 million a nd $900 million in uns upported and undisclosed reserves, a signif icant portion of which did not comply with GAAP, which resulted in material inaccuracies in filings made by Microsoft with the In the Matter of Micros oft Corporation, Res pondent: Order Instituting Commission.” See SEC, 2002, " Administrative Proceedings Pursuant to Section 21C of th e Securities Exchange Act of 1934, Making Findings and Imposing Cease-and-Desist Order," June 3, 2002, Securities and Ex change Commission (an electronic version is ), p. 2. available at: http://www.sec.gov/litigation/admin/34-46017.htm -92-

101 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R The Puzzling Systematic Positive Long-Term Bias and Short-Term Figure 13 Negative Bias in Analyst Earnings Forecasts -11 -10 -9 -8 -7 -6 -5 -4 -3 Scaled Forecast Error, 1986-1988 Scaled Forecast Error, 1989-1991 -2 Scaled Forecast Error, 1992-1994 -1 Scaled Forecast Error, 1995-1997 Month Relative to Annual Earnings Announcement 0 Scaled Forecast Error, 1998-2001 0 -.015 -.01 -.005 Median Scaled Forecast Error Source: Richardson, Teoh, and Wysocki (2003). Data provided through private conversations with one of the authors. positive at the time of However, since 1992 analysts average forecast errors remain slightly udy concludes: “Or, to put it in more realization of actual earnings and this is pu zzling. As one st concrete terms, how could analysts continue to underestimate Microsoft’s quarterly earnings 41 78 times in a row?” Thus an equilibrium has evolved in which analysts systematically overestimate the longer- e forecasts as the actual term future earnings and then “walk down” thos date of the earnings report underestimate approaches. And in the end they systematically the final earnings. The result gives management the best of both worlds. Copeland, Dolgoff, and Moel (2002) show that farther out earnings forecasts have a larger effect on the stock price than next quarter’s earnings so the pos itive bias for the far out earnings helps justify excessi en when the bias turns vely high stock prices. Th negative at the actual end of quarter announcement, managers and firms benefit from a pos itive earnings surprise that further increases the value of the stock. 78 Bartov, Givoly and Hayn, "The Rewards to Meeting or Beating Earnings Expectations", p 203. -93-

102 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION that investors ar fooled into If this interpretation is true it implies e systematically being consistent with the d ata. As Skinner and Sloan overpaying for these stocks. And this appears to be orta (1996) a nd Dechow and Sloan (1997) shows that “analysts’ (2002) explain, research by La P long run EPS forecasts are systematically overoptimistic for growth firms, and that the magnitude of the over optimism in these forecasts is sys tematically r elated to the inferior stock price performance of growth firms.” (p. 291) atively to the walk-down What is puzzling is why the market does not appear to respond neg 79 in the forecasts and then responds pos itively to the final earnings s urprise. It almost appears that 80 collusion is taking place, but there is no indication as to how this occurs. This is clearly an issue that requires further research and understanding. What matters here for board policy is that once a firm’s managers get into this earnings management game with analysts and the market, th ere is no way for them to win in the long run — except by pure luck. Pushing expenses into the future and bringing revenues from the future to the present to meet analyst forecasts only compounds the problem of meeting the forecasts in the future. And to the extent that doing so actually destroys future value, it is even less likely that management will win the game. In the end, it appears that analysts understand the game because when a firm misses an earnings forecast by even a penny the stock can suffer a large price decline. The sharp decline in stock prices for growth firms in response to small negative errors in Figure 12 is consistent with if management can’t find another pe nny to report they must this observation. The argument is that 81 And at this point the stock price pe be in serious trouble. nalty can be extremely severe and 79 Indeed, Ibid. , document that firms are rewarded with a statistically significant stock price increase for firms with positive earnings surprises even when top management appears to have managed earnings (through the use of accounting accruals) or expectations (where there has been a walk-down resulting in a positive earnings surprise). 80 The sophisticated business press is aware of these peculiarities. For example, Nocera, 1997, "Who Really Loves the Market? Securities Analysts are Wall Street's New Stars", Fortune , V. 136, No. 8, October 27, p. 90+ summarizes analyst coverage of Intel: “The great bul ny’s k of the 67 analysts who track Intel follow the compa guidance slavishly. They put their earnings estimates just low enough to make it possible for the company to ‘surprise’ them quarter after quarter. They spend most of their time assuring clients th at Intel w ill ‘make the quarter,’ rather than searching anything more fundamental to say about the company.” 81 “At least partly by this expectational interplay, the price of missing by a penny has risen sharply . . . In the growth stock fraternity, ‘missing by a penny’ now implies the height of corporate boneheadedness — that is, if you couldn’t find that extra penny to keep Wall Street happy, then your company must really be in trouble, and since missing by a penny is already going to send your stock plummeting, you’re better off missing by a dime or two and saving those earnings for the next quarter.” From Fox, "Learn to Play the Earnings Game (and Wall Street will Love You). The Pressure to Report Smooth, Ever Higher Earnings Has Never Been Fiercer. You Don't Want to Miss the Consensus Estimate by a Penny—And You Don't Have To", . -94-

103 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Figure 14 More Evidence on Lying About Earnings: Frequency Distribution of Earnings Per Share Forecast Error Note: The figure plots the distribution of the forecast error, the company’s EPS less the analysts’ consensus EPS forecast, over the quarters 1974-1996. The black area below the graph represents the density “shortfall” shortfall relative to a bin equidistant from zero on the other side of the hist (0)  ogram. The refers to a test-statistic devised by the authors to assess statistically a discontinuity in the distribution. round for In this case, the statistic rejects the hypothesis that the density is smooth a ecast errors of zero. Source: Degeorge, Patel, and Zeckhauser (1999), Figure 6, p. 20. ital markets for funding and to the w ealth of therefore damaging to the firm’s access to the cap managers with substantial equity-based compensation. “Earnings Manag atel, In their excellent study of ement To Exceed Thresholds,” Degeorge, P and Zeckhauser (1999) examine quarterly earnings data on 5,387 firms with over 100,000 quarterly earnings observations (although some of the samples are considerably smaller than this). Their mmetrically ecast errors is not sy results show dramatically that the statisti cal distribution of for distributed about zero, as one would expect if such errors were random and unbiased. Figure 14 shows the deviations are what we would expect from is being manipulated. an earnings process that forecast e There are far too many zero or slightly positive quarterly earnings rrors (of +1, +2, and +3 cents per share). In addition, there are far too few forecast errors of –1, –2 and –3 cents per share as well as to few of +4 cents or above per share. The bars to the right of zero in Figure 14 represent positive earnings surprises and those to lackened area beneath the graph denotes the “s the left, negative earnings surprises. The b hortfall” -95-

104 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION of observations in that area compared to what would occur if the forecast error distribution were symmetric. Thus, the earnings management process is yielding too few small negative earnings surprises, and too few large positive earnings surprises. This is consistent with manag under- ement reporting earnings that would yield large earnings s urprises (like the Microsoft example described above) and using the “stored” earnings to generate more small pos itive earnings s urprises. This then yields too many large neg ative earnings s urprises consistent with what we would exp ect to happen when companies finally lose the earnings management game. Their results are a strong indictment of the erosion of integrity in the earnings reporting process. The authors quote the conclusion of a study by Bruns a nd Merchant (1990) “we have no doubt that short-term earnings are being manipulated in many, if not all, companies.” Other evidence is consistent with the hypothesis that top management is aware of the walk- down phenomenon and that they often exploit the phenomenon to time the sale of shares. As pointed out by Richardson, Teoh, and Wysock i (2003), due to the 1988 Insider Trading and adopted insider trading blackout periods that Securities Fraud Enforcement Act, most firms have typically cover the two months prior to an earnings announcement. As a result, management is 82 They effectively constrained to transact only in the month following the quarterly announcement. are positive after an earnings announcement the find, for example, that when net insider sales ssociated with a pos itive earnings s urprise is 66% that is frequency that the announcement was a significantly different from the 54% frequency for firms without subsequent net insider s ales. Looking at it somewhat differently, the probability that insiders w ill s ell shares follo wing a positive 83 (including zero) earnings surprise is 70% and only 60% following a negative earnings surprise. 82 In the US, the Insider Trading and Securities Fraud Enforcement Act of 1984 and 1988 limit trading by company insiders in the company’s stock. In response to the 1988 law, firms designed and instituted policies regarding insider trading. Over 80% of the plans bar option exercises and stock sales except after a relatively short wi ndow following earnings announcements. See Bettis, Coles and Lemmon, 2000, "Corporate Policies Restricting Trading by Insiders", Journal of Financial Economics, V. 57, No. 2: pp. 191-220. 83 Private communication to the authors from Siew Hong Teoh, 2003. -96-

105 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION R-37. Firms must restart the conversation between corporate managers and Wall old game of earnings management and Street by “just saying no” to the 84 earnings guidance. This will not be easy. However, eliminati ng or reducing the influence of these corrupting forces on the firm will be an im portant step in bolstering the integrity of corporations. There is a window of opportunity now that analysts and the financial institutions that employ them have fallen into disrepute. It is the analyst’s job to forecast earnings and to estimate their imp lications for value. People are highly aware of the malaise that has gripped the business world. Executives are wondering how to invest in the integrity of their companies. Researchers are starting to examine some of the issues. But this window w on’t remain open forever and if we don’t s eize this moment to identify the problem, talk about it, and learn from it, and ch ange the system we could find ourselves trapped once again in a vicious, destructive cycle. And let’s be clear, ending the earnings management game (as Coca Cola, Gillette and USA Networks, and others have), does not mean ending communications with analysts and the cap ital markets. R-38. Senior managers must communicate with the capital markets. They must understand what drives value in their organization and align internal goals with those drivers, not with analysts’ expectations. To limit wishful thinking, managers should reconcile their compa ny’s projections to industry and rivals’ projections. When the compa ny’s expectations lie outside what is widely viewed as the industry’s growth rate, managers s hould explain how and why they will be able to outperform their market. Some w ill a rgue that making this all clear to the analysts w aluable i nformation to their ill reveal v competitors. “ To this, we have a simple response: If your strategy is based o n your competitor not knowing what you are doing as opposed to not being able t o do what you can do, you cannot be successful in the long run no matter who knows what.” ( Fuller and Jensen (2002)) 84 See Fuller and Jensen, "Just Say No To Wall Street: Putting A Stop To the Earnings Game", for a full discussion. -97-

106 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Conclusions VIII. In their 1990 study of CEO compensation Jensen and Murphy (1990a) had this to say: “Are attract the best and brightest individuals to current levels of CEO compensation high enough to careers in corporate management? The answer is, probably not.” urmised, Jensen and Murphy would not give As the reader of this report has undoubtedly s that answer today. Indeed, we have emphasized here that while executive compensation can be a powerful tool for reducing the agency conflicts between managers and the firm, compensation can if it is not managed properly. And as we’ve also be a substantial source of agency costs summarized, there is substantial evidence that we can do much better in the future. While our ab ize the phenome non underlying recent problems in executive ility to character that the causes are sys remuneration is not perfect, we are confident creation of a new temic. The regime in compensation practice will entail considerable thought. Otherwise, one risks re-cr eating the type of systems failure we have witnessed unfold in many major companies over the last few years. ill not be In addition, the changes required to put balance back in the remuneration sys tem w easy to implement. The issues are complex. There will be conflict at the highest of corporate levels, and there will be mistakes made. But this is a forward-looking managers and time where wise and acing the difficult choices in remuneration, boards can achieve a competitive advantage by f tments in the governance, and relations with the capital markets. It is a time in which proper inves nerate considerable bene integrity of the organization and its systems will ge fits in both the short and long run. Wise CEOs as well as wise board members will encourage these inves tments b ecause they will understand that well-functioning governance and monitoring syst ems will help to ensure not only organizational success, but personal success. The evidence of the damage to personal reputations as well as organizations is in the daily headlines, not only in the US, but also in the rest of the world. And even for some who have su cceeded in preserving w ealth acquired in the face of honor succeeded in avoi ding jail, the question remains, how good is life wit hout scandal, and have and respect? -98-

107 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R References Alpern, Richard L. and Gail McGowan. 2001. Guide to Change of Control: Protecting Companies and Their : Executive Compensation Advisory Services. Executives Amir, Eli. 1993. "The Market Valuation of Accounting Inform -retirement Benefits other ation: The Case of Post than Pensions." The Accounting Review, V. 68, No. 4: pp. 703-724. Anders, George. 2003. "Upping the Ante: As Some Decry Lavish CEO Pay, Joe Bachelder Makes It Happen." Wall Street Journal , June 25, p. A1. urphy. 1994. "Sub jective Pe rformance Measure s in Optimal nd Kevin J. M Baker, George P., Robert Gibbons a Incentive Contracts." Quarterly Journal of Economics, V. 109, No. 4: November 1994, pp. 1125-1156. n J. Murphy. 1988. "C ompensation and Incentives: Pr Baker, George P., Michael C. Jensen and Kevi actice vs. Theory." V. 43, No. 3: pp. 593-616. Available from the Social S cience Research Network Journal of Finance, http://papers.ssrn.com/Abstract=94029 . Reprinted in Michael C. Jensen, Foundations of eLibrary at: Organizational Strategy, Cambridge: Harvard University Press, 1998. ectations." Bartov, Eli, Dan Givoly and Carla Hayn. 2002. "The Rewards to Meeting or Beating Earnings Exp Journal of Accounting & Economics, V. 33, No. 2: pp. 173-204. cience Research Available from the Social S Network eLibrary at: http://papers.ssrn.com/paper=247435 . Journal of Bebchuk, Lucian Arye and Jesse M. Fried. 2003. "Executive Compen sation as an Agency Problem." Economic Perspectives, V. 17, No. 3: p. 71+. 2002. "Manager Bebchuk, Lucian Arye, Jesse M. Fried and David I. Walker. ial Power and Rent Extraction in the Design of Executive Compensation." University of Chicago Law Review, V. 69: June, pp. 751-846. Av ailable http://papers.ssrn.com/abstract=316590 . from the Social Science Research Network eLibrary at: CEOs Rewarded for Luck? The Ones Without Principals Bertrand, Marianne and Sendhil Mullainathan. 2001. "Are Quarterly Journal of Economics, V. 116, No. 3: pp. 901-932. Are." Bettis, J.Carr, Jeffrey L. Coles and Michael L. Lemmon. 2000. "Corporate Policies Restricting Trading by Insiders." Journal of Financial Economics, V. 57, No. 2: pp. 191-220. Black, Fischer and Myron S. Scholes. 1973. "The Pricing of Options and Corporate Liabilities." Journal of Political Economy, V. 81, No. 3: pp. 637-654. Bodie, Zvi, Robert S. Kaplan and Robert C. Merton. 2003. "For the Last Time: Stock Options Are an Expense." Harvard Business Review : March, pp. 63-71. Management Bruns, William J. and Kenneth A. Merchant. 1990. "The Dangerous Morality of Managing Earnings." Accounting, V. 71: August, pp. 22-25. ohn B. Shoven. 2004. "Accounting for Stock Options". In Stanford Research Paper Series Bulow, Jeremy I. and J Paper No. 1848. , March. Palo Alto. Available from the Social Science Research Network eLibrary at: http://ssrn.com/abstract=521882 . Business Roundtable. 2003. "Executive Compensation: Principles and Commentary." November. Carter, Mary Ellen and Luann J. Lynch. 2003. "The Consequences of the FASB's 1998 Proposal on Accounting for Stock Option Repricing." Journal of Accounting & Economics, V. 35, No. 1: pp. 51-72. Conference Board. 2002. "Commission on Public Trust and Private Enterprise." September. Copeland, Tom, Aaron Dolgoff and Alberto Moel. 2002. "The Role of Expectations in Explaining the Cross- Section of Stock Returns." Monitor Group Working paper, November. Cambridge, MA. -99-

108 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Confessions of a Street Addict . New York: Simon & Schuster. Cramer, James J. 2002. D'Avolio, Gene, Efi Gildor and Andrei Shleifer. 2002. "Technology, Information Production, and Market Economic Policy for the Information Economy , August, 2001. Jackson Hole, Wyo.: Federal Efficiency". In Reserve Bank of Kansas City. Available from the Social Science Research Network eLibrary at: http://papers.ssrn.com/abstract=286597 . http://www.kc.frb.org/PUBLICAT/SYMPOS/2001/papers/S02shle.pdf Skinner. 2002. "Are Dividends Disappearing? Dividend DeAngelo, Harry, Linda E. DeAngelo and Douglas J. Concentration and the Consolidation of Earnings." USC Center for Law, Economics and Or ganization (CLEO) Working Paper No. 02-9, forthcoming in Journal of Fi nancial Economics . Available from the Social S cience Research Network eLibrary at: . http://papers.ssrn.com/Abstract=318562 Dechow, Patricia M. and Richard G. Sloan. 1997. "Returns to Contrarian Investment Strategies: Tests of Naive Journal of Financial Economics, Expectations Hypotheses." V. 43, No. 1: pp. 3-27. Degeorge, François, Jayendu Patel and Richard Zeckhauser. 1999. "Earnings Management to Ex ceed Thresholds." Journal of Business, V. 72, No. 1: pp. 1-33. d Edward P. Swa nson. 2004. "W hy do c orporate managers miss Efendi, Jap, Anup Srivastava an cial tate finan statements? The role of option compensation, corporate gove rnance, and other fact ors." unpublis hed working paper, Mays Business School, Texas A&M U., Ma y 17. College Station, Texas . Available from the Social . Science Research Network eLibrary at: http://papers.ssrn.com/abstract=547922 Erickson, Merle, Michelle Hanlon and Edward Maydew. 2002. "How Much Will Firms Pay for Earnings that Do Not Exist? Evidence of Taxes Paid on Allegedly Fraudulent Earnings." University of Chicago Working Paper, November 1, 2002. Chicago, IL. Available from the Social Science Electronic eLibrary at: . http://papers.ssrn.com/abstract=347420 Erickson, Merle, Michelle Hanlon and Edward Maydew. 2003. "Is There a Link Betw een Executive Compensation and Accounting Fraud?." University of Chicago Working paper, October 3, 2003. Chicago, IL. Espahbodie, Hassan, Elizabeth Strock and Hassan Tehranian. 1991. "Impact on Equity Prices of Pr onouncements Journal of Accounting & Economics, V. 14, No. 4: pp. 323- Related to Nonpension Postretirement Benefits." 346. Fama, Eugene F. and Kenneth R. French. 2001. "Disapp earing Dividends: Changing Firm Characteristics or Lower Propensity to Pay?" Journal of Financial Economics, V. 60, No. 1: pp. 3-43. Wall Street Journal , October 3. "FASB Backs Project to Study Unification of Accounting Rules." 2002. Los Angeles Times Flanigan, James. 1996. "It's Time for All Employees to Get Stock Options." , April 21. Fox, Justin. 1997. "Learn to Play the Earnings Game (and Wall Street will Love You). The Pressure to Re port Smooth, Ever Higher Earnings Has Never Been Fiercer. You Don't Want to Miss the Consensus Estimate by a Penny—And You Don't Have To." , V. 135, No. 6 March 31, p. 76+. Fortune Frey, Bruno S. and Margit Osterloh. 2004. "Yes, Managers Should Be Paid Like Bureaucrats." University of Zurich, March. Zurich. ing the Profitability of Corporate Insider Trading Through Pretrading Disclosure." Fried, Jesse M. 1998. "Reduc Southern California Law Review, V. 71: pp. 303-392. Fuller, Joe and Michael C. Jensen. 2002. "Just Say No To Wall Street: Pu tting A Stop To the Earnings G ame." Journal of Applied Corporate Finance, Winter 2002, pp. 41-46. Av ailable from the Social V. 14, No. 4: Science Research Network eLibrary at: . http://ssrn.com/Abstract=297156 , May 6, p. C1. Gasparino, Charles. 2002. "Analysts' Contracts Link Pay to Deal Work." Wall Street Journal -100-

109 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R k Options for Undiversified Executives." Journal of Accounting & Hall, Brian J. and Kevin J. Murphy. 2002. "Stoc 3-42. Available from the So ce Research Network eLibrary at: Economics, cial Scien V. 33, No. 1: pp. http://papers.ssrn.com/abstract_id=252805 . Trouble with St ock Options." Journal of Economic Persp Hall, Brian J. and Kevi n J. Murphy. 2003. "The ectives, V. 17, No. 3: p. 49+. Hallock, Kevin F. and Kevin J. Murphy. 1999. The Economics of Executive Compensation V. I & II. Elgar Reference Collection, International Library of Critical Writings in Economics. Cheltenham, UK: Edward Elgar Publishing. Harris, J and P. Bromiley. 2004. "Incentives to Cheat: The Influence of CEO Incentive Compens ation and Relative Univ. of Minnesota Carlson School of Manag ement Firm Performance on Financial Misrepresentation." Working Paper. Hong, Harrison and Jeffrey D. Kubik. Concerns and Biased Earnings 2003. "Analyzing the Analysts: Career Journal of Finance, V. 58, No. 1: pp. 313-351. Forecasts." Hope, Jeremy and Robin Fraser. 1997. "Beyond budgeting: Breaking Through the Barrier To 'The Third Wave'." V. 75, No. 11: pp. 20-23. Management Accounting, Hope, Jeremy and Robin Fraser. 1999a. "Beyond Budgeting White Paper, Beyond Budgeting Round Table", May: Fraser Hope Associates, CAM-I Consortium for Advanced Manufacturing, International. Hope, Jeremy and Robin Fraser. 1999b. "Beyond Budgeting: Building a New Management Model for the Information Age." Management Accounting-London, V. 77, No. 1: pp. 16-21. Strategic Finance, Hope, Jeremy and Robin Fraser. 2000. "Beyond Budgeting." V. 82, No. 4: pp. 30-35. Hope, Jeremy and Robin Fraser. 2003. Can Break Free from the A nnual Beyond Budgeting: How Managers . Boston, Mass.: Harvard Business School Press. Performance Trap Consequences of Managerial Earnings Guidance Prior to Regu lation Hutton, Amy P. 2003. "The Determinants and Fair Disclosure." Tuck School of Business, Dar tmouth College, April 2003. Ha nover, NH. Available from the Social Science Research Network eLibrary at: http://ssrn.com/abstract=317160 . Integrated Finance, Ltd. 2004. "Proposal by Integr ated Finance Limited for Expensing Employee Compensatory Stock Options for Financial Reporting Purposes." New York. Jensen, Michael C. 1969. "Risk, the Pricing of Capital Assets, and the Evaluation of Investment Portfolios." Journal of Business, V. 42, No. 2: pp. 167-247. ency Costs of Free Cash Flow: Corporate Finance and Takeovers." American Jensen, Michael C. 1986. "Ag Available from the Social S cience Research Network eLibrary V. 76, No. 2: pp. 323-329. Economic Review, http://ssrn.com/Abstract=99580 . at: Os, and the Privatization of Bankruptcy." Statement before the Jensen, Michael C. 1989a. "Active Investors, LB House Ways and Means Comm ittee, February 1, 1989.; Journal of App lied Corporate Fi nance, V. 2, No. 1: pp. 35-44. Available from the Social Science Research Network eLibrary at: http://papers.ssrn.com/Abstract=244152 . Reprinted in Michael C. Jensen, A Theory of the Firm: Governance, , Cambridge: Harvard University Press, 2000. Residual Claims, and Organizational Forms Harvard Business Review, V. 67, No. 5: pp. 61-74 Jensen, Michael C. 1989b. "Eclipse of the Public Corporation." (Revised 1997). Available from the Social Science Research Network eLibrary at: http://papers.ssrn.com/Abstract=146149 . Jensen, Michael C. 1989c. "The Effects of on the Economy". In Remarks before the LBOs and Corporate Debt Subcommittee on Telecommunications and Finance, U .S. House of Representa tives Hearings on L everaged Buyouts . Washington, D.C.: Government Accounting Office. Jensen, Michael C. 1993. "The Modern Industrial Revoluti on, Exit and the Failure of Internal Control Systems." eLibrary Journal of Finance, V. 6, No. 4: pp. 831-880. Av ailable from th e Social S cience Research Network -101-

110 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R . Also published in Journal of Applied Corporate Fi nance 6, No. 4, at: http://papers.ssrn.com/Abstract=93988 A Theory of the Firm: Governance, Residual Claims, and 1994. Reprinted in Michael C. Jensen, Organizational Forms , Cambridge: Harvard University Press, 2000. Jensen, Michael C. 1998. Foundations of Organizational Strategy . Cambridge, MA: Harvard University Press. Jensen, Michael C. 2001a. "How Stock Options Reward Managers for Destroying Value and What To Do About It." Harvard Business School Negotiations and Markets (NOM) Working Paper, April 17, 2001. Available from the Social Science Research Network eLibrary at: . http://ssrn.com/Abstract=480401 Jensen, Michael C. 2001b. "Value Maximization, Stakeholder Theory, and the Corporate Objective Function." V. 7, No. 3: pp. 297-317. Available from the Social S cience European Financial Management Review, http://papers.ssrn.com/Abstract=220671 . An earlier version of this paper appears Research Network eLibrary at: in Breaking the Code of Change eds, Harvard Business School Press, 2000. , Michael Beer and Nithin Nohria, Business Ethics Quarterly , Vol. 12, No.1, Jan 2002, and the Reprinted in Journal of Applied Corporate , Vol. 14, No. 3, Fall 2001. Finance Jensen, Michael C. 2002. "The Agency Cost of Overvalu Corporate Finance State of ed Equity and the Current 2002. L ondon: forthcoming in (Keynote Lecture European Finan cial Management Association)", June European Financial Management, 2004. Available from the Social Science Research Network eLibrary at: http://ssrn.com/abstract=590961 . Jensen, Michael C. 2003. "Paying People to Lie: The Truth About the Budgeting Process." European Financial V. 9, No. 3: 2003, pp. 379-406. cience Research Network eLibrary Management, Available from the Social S at: http://papers.ssrn.com/Abstract=267651 . An executive summary version of this article appears in the Harvard Business Review, November, 2001 under the title "Corporate Budgeting Is Broken: Let's Fix it". A ticle appeared in th short version of this ar e Wall Street Journal, Manager's Journal Column, January 8, 2001 under the title "Why Pay People to Lie?" Jensen, Michael C. 2004. "Agency Costs of Overvalued Equity." Negotiations, Organ izations and Markets ( NOM) 2004, Working Paper No. 04-26, and European Corporate Governance Institute (ECGI) Working Paper No. 39/ May. Available from the Social Science Research Network eLibrary at: http://ssrn.com/Abstract=480421 . Best Practice: Ideas and Insights form Jensen, Michael C. and Joe Fuller. 2002. "What's a Dir ector to Do?," in ed., the World's Foremost Business Thinkers . Cambridge, MA. and London: Perseus Publishing and Bloomsbury Publishing. Available from the Social Science Research Network eLibrary at: . http://papers.ssrn.com/Abstract=357722 the Firm: Managerial Beha vior, Agency Costs, and Jensen, Michael C. and William H. Meckling. 1976. "Theory of Ownership Structure." Journal of Financial Economics, V. 3, No. 4: October, pp. 305-360. Available from the http://papers.ssrn.com/Abstract=94043 . Reprinted in The Modern Social Science Research Network eLibrary at: , Michael C. Jensen and Clifford W. Smith, Jr., Editors, New York: McGraw- Theory of Corporate Finance Hill, Inc., 1984. Jensen, Michael C. and Kevin J. Murphy. 1990a. "CEO Incentives: It's Not How Much You Pay, But How." Harvard Business Review, V. 68, No. 3: pp. 138-153. Available from the Social Science Research Network Foundations of http://papers.ssrn.com/Abstract=146148 eLibrary at: . Reprinted in Michael C. Jensen, Organizational Strategy , Cambridge: Harvard University Press, 1998. Jensen, Michael C. and Kevin J. Murphy. 1990b. "Pe rformance Pay and Top Management Incentives." Journal of Political Economy, V. 98, No. 2: pp. 225-265. Available from the Social S cience Research Network eLibrary Foundations of Organizational at: http://papers.ssrn.com/Abstract=94009 . Reprinted in Michael C. Jensen, Strategy , Cambridge: Harvard University Press, 1998. Johnson, Shane. A, Harley. E. Ryan Jr. and Yisong. S. Tian. 2004. "Executive Compensation and Corporate Fraud." Ourso College of Business Administration Working Paper, May. Louisiana State University. Av ailable . from the Social Science Research Network eLibrary at: http://ssrn.com/abstract=395960 -102-

111 E R Jensen, Murphy and Wruck XECUTIVE EMUNERATION New Kampel, Stewart. 2000. "Five Questions for Joseph E. Bachelder, Engineer of the Executive Pay Express." , June 11. York Times nd David P. Norton. 1996. The Balanced Scorecard . Boston, MA: Harvard Business School Kaplan, Robert S. a Press. Kaplan, Steven N. "Management Buyouts: Evidence on Post-Buyout Operating Changes." Journal of Financial . Economics Kaplan, Steven N. 1988. "Management Buyouts: Efficiency Gains or Value Transfers". Chicago, IL: University of Chicago. Unpublished manuscript. Kaplan, Steven N. 1989. "The Effects of Management Buyouts on Operating Performance and Value." Journal of Financial Economics, V. 24, No. 2: pp. 217-254. Kaplan, Steven N. 1990. "Sources of Value in Management Buyouts." Journal of Financial Economics . Kay, David. 2004. "Forget How the Crow Flies." , January 18, pp. W1, W2. Financial Times Kerr, Steven. 1975. "On the Folly of Rewarding A, While Hoping for B." Academy of Management Journal, V. 18, No. 4: pp. 769-783. Kersnar, Janet. 1999. "Re-Inventing the Budget." CFO Asia , July/August. Khurana, Rakesh. 2002a. "The Curse of the Superstar CEO." Harvard Business Review : September, pp. 3-8. Khurana, Rakesh. 2002b. Charismatic CEOs . Princeton, Searching for a Corporate Savior: The Irrational Quest for N.J.: Princeton University Press. Kole, Stacey. 1997. "The Complexity of Compensation Contracts." Journal of Fi nancial Ec onomics, V. 43, No. 1: pp. 79-104. V. 51, No. 5: La Porta, Rafael. 1996. "Expectations and the Cross-Section of Stock Returns." Journal of Finance, pp. 1715-1742. Lambert, Richard A., W. Lanen and David F. Larcke r. 1989. "Executive Stock Opti on Plans and Corporate Dividend V. 24, No. 4: pp. 409-425. Journal of Financial and Quantitative Analysis, Policy." Lester, Tom. 2000. "Monday Management: Managers Count Blessings As Budgets Begin To Lose Currency--Some Firms Long for Freedom from the Burden of Budgeting." Irish Times . Lewellen, Wilbur G., Claudio Loderer and Kenneth Martin. 1987. "Executive Compensation and Executive Incentive Journal of Accounting & Economics, V. 9, No. 3: pp. 287-310. Problems: An Empirical Analysis." act in their shareholder 's best interests, take away Martin, Roger. 2003a. "Taking Stock: If you want managers to their company stock." Harvard Business Review, V. 81, No. 1: p. 19. The Wrong Incentive: Execu tives Taking Stock W ill Behave Like Athletes Placing Bets . Martin, Roger. 2003. Barron's Online. An electronic version is available at: http://online.wsj.com/barrons/article/0,,SB107187920976382100,00.html . McCarroll, Thomas. 1992. "The Shareholders Strike Back: Executive Pay." Time , May 5. McTaggart, James M., Peter W. Kontes and Mich ael C. Manki ns. 1994. The Value Imperative: Managing for Superior Shareholder Returns . New York: The Free Press. Merton, Robert C. 2004. "Summary of the Oral Testimony of Robert C. Merton, H.R. 3574: Stock Option Accounting Reform Act." March 3. Washington, DC. Meulbroeck, Lisa. 2000. "The Efficiency of Equity-Linked Compensation: Understanding the Full Cost of Awarding Executive Stock Options", p. 62. Boston, MA: Harvard Business School. W orking Pape r No. 00-056. . Available from the Social Science Research Network eLibrary at: http://papers.ssrn.com/abstract_id=215530 -103-

112 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R "SFAS No. 106 and Benefit Reductions in Employer- Mittelstaedt, Fred, William Nichols and Philip Regier. 1995. V. 70, No. 4: pp. 535-556. Sponsored Retiree Health Care Plans." The Accounting Review, Handbook of L rd, abor Murphy, Kevin J. 1999. "Executive Compensation," in ed. Orley Ashenfelter and David Ca Economics , 3: North Holland. Contracts." Journal of Accounting & Economics, V. Murphy, Kevin J. 2000. "Pe rformance Standards in Incentive 30, No. 3: December 2000, pp. 245-278. ial Power vs. the Per ceived Cost of Stock Murphy, Kevin J. 2002. "Explaining Executive Compensation: Manager Options." University of Chicago Law Review, V. 69, No. 3: pp. 847-869. Journal of Accounting & Economics, Murphy, Kevin J. 2003. "Stock-Based Pay in New Economy Firms." V. 34, No. 1-3: pp. 129-147. Murphy, Kevin J. and Jan Zabojnik. 2003. "Managerial Capital and the Market for CEOs." USC Working paper, October 2003. of the NACD Blue Ribbon Commission on Executive National Association of Corporate Directors. 2003. "Report Compensation and the Role of the Compensation Committee."NACD Washington, D.C. Nocera, Joseph. 1997. "Who Really Loves the Market? Securities Analysts are Wall Street's New Stars." Fortune , V. 136, No. 8 October 27, p. 90+. mack. 2000. "Taking Stock: Equit y-based Compens Ofek, Eli and David Yer ation and the Evolution of Managerial Ownership." Journal of Finance, V. 55, No. 3: pp. 1367-1384. Patterson, Gregory. 1992. "Distressed Shoppers, Disaffected Workers Prompt Stores to Alter Sales Commission." Wall Street Journal , June 1. "Politics and Policy—Campaign ’92: From Quayle to Clinton, Politicians ar e Pouncing on the Hot Issue of Top Wall Street Journal Executive’s Hefty Salaries." 1992. , January 15. Post, Richard J. and Kenneth E. Goodpaster. 1981. "H.J. Heinz Company: of Policy". Boston, The Administration MA: Harvard Business School. HBS Case No. 9-382-034. Richardson, Scott, Siew Hong Teoh and Peter Wysocki. 2003. "The ecasts: The Walk-down to Beatable Analyst For Role of Equity Issuance and Insider Trading Incentiv es." Working paper, Whart on School, University of Pennsylvania, October 20, 2003. Philadelphia, PA. Roberts, Johnnie L. 1989. "Credit Squeeze: Dun & Bradstreet Faces Flap Over How It Sells Reports on Businesses." , March 2. Wall Street Journal Rose, Nancy L. and Catherine D. Wolfram. 2002. "Regulating Executive Pay: Using the Tax Code to Influence Journal of Labor Economics, V. 20, No. 2: pp. S138-S175. Chief Executive Officer Compensation." Sahlman, William A. 1990. "The Structure and Governance of Venture-cap izations." Journal of Financial ital Organ Economics, V. 27, No. 2: pp. 473-521. SEC. 2002. "In the Matter of Microsoft Corporation, Res pondent: Order Instituting Admin istrative Pro ceedings Pursuant to Section 21C of the Securities Exchange king Findings and Imposing Cease-and- Act of 1934, Ma Desist Order." June 3, 2002.Securities and Exchange Commission. An electronic version is available at: http://www.sec.gov/litigation/admin/34-46017.htm . SEC. 2003. "Securities and Exchange Commission v. Bear Stearns & Co., Inc.."U.S. Distr ict Court (Southern District of New York). "SEC to Push for Data on Pay of Executives." 1992. Wall Street Journal , January 21. Seyhun, H. Nejat. 1992. "The Effectiveness of Insider Trading Sanctions." Journal of Law & Economics, V. 35, No. 1: pp. 149-182. -104-

113 E XECUTIVE EMUNERATION Jensen, Murphy and Wruck R Wall Street Journal , "Shareholder Groups Cheer SEC’s Moves on Disclosure of Executive Compensation." 1992. February 14. Sloan. 2002. "Earnings Surprises, Growth Expectations, and Stock Returns or Skinner, Douglas J. and Richard G. Don't Let an Earnings Torpedo Sink Your Portfolio." V. 7, No. 2-3: pp. 289- Review of Accounting Studies, 312. New York Times Magazine , June Sokolove, Michael. 2002. "How to Lose $850 Million -- And Not Really Care." 9. Stewart, G. Bennett. 1990. "Remaking the Corporation from Within." V. 68, No. 4: pp. Harvard Business Review, 126-137. Swartz, Mimi and Sherron Watk ins. 2003. . New York: Power Failure: The Inside Story of the Collapse of Enron Doubleday. Thomas, Tony. 2000. "Toss Your Budget Out the Window." Business Review Weekly , V. 72, September 8. port, Deutsche Tirello, Edward J., Jr. 2000. "Enron Corporation: The Industry Standard for Excellence." Analyst Re Banc Alex Brown, September 15, 2000. New York. H.R. 5242. 2002. Workplace Employee Stock Option Act of 2002 (To amend the Internal Revenue Code of 1986 to .). U.S. House of Represe encourage the granting of employee stock options ntatives. Reps Houghton & Boehner. Code of 1986 to ensure S. 2877. 2002. Rank and File Stock Option Act of 2002 (To amend the Internal Revenue rank-and-file employees as well as officers and directors). U.S. Sen ate (107th that stock options are granted to Congress, 2nd Session). Sens Lieberman & Boxer. H.R. 2788. 1997. Employee Stock Option Bill of 1997 (To amend the Internal Revenue Code of 1986 to promote the grant of incentive stock options to non-highly compensated employees). U.S. House of Represen tatives (105th Congress, 1st Session). Rep. Houghton. Weil, Jonathan. 2003. "Fixing the Numbers Problems -- Accounting Standards Board Takes on Hot-Button Issues in Timely Manner." Wall Street Journal , January 13, p. C1. Weston, J. Fred, Mark L. Mitchell and J. Harold Mulherin. 2004. Takeovers, Restructuring, and Corporate Governance . Upper Saddle River, N.J.: Pearson Prentice Hall. Whitford, David. 1998. "Becoming CEO? Call Him First." Fortune , V. 137, No. 11 June 8, p. 281+. Zheng, Liu. 2003. "Six-Month-One-Day Option Exchange: The Impact of the Accounting Rule on Stock Option Repricing." University of Southern California Working Paper. Los Angeles, CA. -105-

114 about ECGI corpo- The European Corporate Governance Institute has been established to improve rate governance through fostering independent scientific research and related activities. The ECGI will produce and disseminate high quality research while remaining close to the concerns and interests of corporate, financial and public policy makers. It will draw on the expertise of scholars from numerous countries and bring together a critical mass of expertise and interest to bear on this important subject. The views expressed in this working paper are those of the authors, not those of the ECGI or its members. www.ecgi.org

115 ECGI Working Paper Series in Finance Editorial Board Paolo Fulghieri Editor , Professor of Finance, University of North Carolina, INSEAD & CEPR Consulting Editors Franklin Allen , Nippon Life Professor of Finance, Professor of Economics, The Wharton School of the University of Pennsylvania Patrick Bolton , John H. Scully ‘66 Professor of Finance and Economics, Princeton University, ECGI & CEPR Marco Pagano , Professor of Economics, Università di Salerno, ECGI & CEPR Luigi Zingales , Robert C. McCormack Professor of Entrepreneurship and Finance, University of Chicago & CEPR Julian Franks , Corporation of London Professor of Finance, London Business School & CEPR Xavier Vives , Professor of Economics and Finance, INSEAD & CEPR Editorial Assistant : Cristina Vespro , ECARES, Université Libre De Bruxelles Financial assistance for the services of the editorial assistant of these series is provided by the European Commission through its RTN Programme on Understanding Financial Architecture: Legal and Political Frameworks and Economic Efficiency (Contract no. HPRN-CT-2000-00064). www.ecgi.org\wp

116 Electronic Access to the Working Paper Series The full set of ECGI working papers can be accessed through the Institute’s Web-site (www.ecgi.org/wp) or SSRN: Finance Paper Series http://www.ssrn.com/link/ECGI-Fin.html Law Paper Series http://www.ssrn.com/link/ECGI-Law.html www.ecgi.org\wp

Related documents

Corporate Surveillance in Everyday Life

Corporate Surveillance in Everyday Life

Wolfie Christl CORPORATE SURVEILLANCE IN EVERYDAY LIFE How Companies Collect, Combine, Analyze, Trade, and Use Personal Data on Billions EPORT BY CRACKED LABS A R Vienna, June 2017 Author: Wolfie Chri...

More info »
Christl Networks  K .indd

Christl Networks K .indd

Wolfie Christl, Sarah Spiekermann Networks of Control

More info »
Access to Capital and Market Liquidity

Access to Capital and Market Liquidity

Report to Congress Access to Capital and Market Liquidity As Directed by the Explanatory Statement to the Consolidated Appropriations Act, 2016 (P.L. 114-113) This is a report by the Staff of the Divi...

More info »
Who Is On the Other Side

Who Is On the Other Side

BLUEMOUNTAIN INVESTMENT RESEARCH W H O I S O N T H E O T H E R S I D E ? optimism or pessimism, leaving mispriced Being successful as an active investment securities in our wake. And then there are ma...

More info »
AI Now 2017 Report

AI Now 2017 Report

AI ​ ​ Now ​ ​ 2017 ​ ​ Report Authors ​ Campolo, ​ ​ New ​ ​ York ​ ​ University Alex ​ ​ ​ ​ New ​ Sanfilippo, York ​ ​ University ​ ​ Madelyn ​ Whittaker, ​ ​ Google ​ ​ Open ​ Meredith ​ Research,...

More info »
Gilson&Kraakman Book final

Gilson&Kraakman Book final

G ILSON RAAKMAN _B OOK (D O N OT D ELETE ) 3/19/2014 9:12 AM &K MARKET EFFICIENCY AFTER THE FINANCIAL CRISIS: IT’S STILL A MATTER OF INFORMATION COSTS ** * and Reinier Kraakman Ronald J. Gilson I NTRO...

More info »
kd0419345enn

kd0419345enn

Competition policy for the digital era A report by Jacques Crémer Yves-Alexandre de Montjoye Heike Schweitzer Competition

More info »
Automation, Big Data, and Politics: A Research Review

Automation, Big Data, and Politics: A Research Review

Internat (20 16 ), 5032 – 5055 1932 – 8036/2016 000 5 10 ional Journal of Communication Automation, Big Data, and Politics: A Research Review 1 SAMANTHA SHOREY University of Washington, USA PHILIP HOW...

More info »
12 Risks with infinite impact

12 Risks with infinite impact

Global System Major Asteroid Global Artificial Future Bad Extreme Collapse Impact Pandemic Intelligence Global Governance Climate Change Global System Major Asteroid Global Artificial Future Bad Extre...

More info »
ssrn id2154873

ssrn id2154873

, “Sell in May and go i ndicator The Halloween Away”: an even bigger puzzle * Ben Jacobsen University of Edinburgh Business School [email protected] Cherry Y. Zhang , China Nottingham University B...

More info »
v 1.0 2012 02 16

v 1.0 2012 02 16

Written Testimony of John W. Van Alst Attorney, National Consumer Law Center and Director of NCLC’s Working Cars for Working Families Project Before the Committee on Financial Services Subcommittee on...

More info »
Media Manipulation and Disinformation Online

Media Manipulation and Disinformation Online

Media Manipulation and Disinformation Online Alice Marwick and Rebecca Lewis

More info »
MGI big data full report

MGI big data full report

McKinsey Global Institute June 2011 Big data: The next frontier for innovation, competition, and productivity

More info »
Microsoft Word   Buccafusco Online

Microsoft Word Buccafusco Online

C OPYRIGHT 2016, V IRGINIA L AW R EVIEW A SSOCIATION © A THEORY OF COPYRIGHT AUTHORSHIP * Christopher Buccafusco I ... NTRODUCTION ... 1229 L AW AND I. C RITINGS OF A UTHORS ... 1234 OPYRIGHT W Feist ...

More info »
10749

10749

Cash transfers: what does the evidence say? A rigorous review of programme impact and of the role of design and implementation features Francesca Bastagli, Jessica Hagen-Zanker, Luke Harman, Valentina...

More info »
Entering the Law Teaching Market

Entering the Law Teaching Market

Entering the Law Teaching Market Entering the Law Teaching Market • 2015-2016 YALE LAW SCHOOL • CAREER DEVELOPMENT OFFICE • NEW HAVEN, CT 2018-2019 DEVELOPMENT SCHOOL • YALE LAW OFFICE CAREER www.law....

More info »
A Financial System That Creates Economic Opportunities: Capital Markets

A Financial System That Creates Economic Opportunities: Capital Markets

U.S. DEPARTMENT OF THE TREASURY A Financial System That Creates Economic Opportunities Capital Markets OCTOBER 2017

More info »
Establishing cryptocurrency equilibria through game theory

Establishing cryptocurrency equilibria through game theory

AIMS Mathematics, 4(3): 420–436. DOI:10.3934 / math.2019.3.420 Received: 02 March 2019 Accepted: 25 March 2019 Published: 07 May 2019 http: journal / Math // / www.aimspress.com Research article Estab...

More info »
What Do We Know About Base Erosion and Profit Shifting? A Review of the Empirical Literature

What Do We Know About Base Erosion and Profit Shifting? A Review of the Empirical Literature

Univ ersit hic ago L aw Sc hool y of C ago U d un Chic nbo Coase-Sandor Working Paper Series in Law and Coase-Sandor Institute for Law and Economics Economics 2014 What Do We Kno w About B ase Erosion...

More info »
WorkingPaper 2

WorkingPaper 2

Has the Fed Been a Failure? George Selgin Department of Economics Terry College of Business University of Georgia William D. Lastrapes Department of Economics Terry College of Business University of G...

More info »