Citation
Style vs. substance

Material Information

Title:
Style vs. substance the nature of CEO evaluation criteria in management controlled and owner controlled firms
Creator:
Silva, Paula
Publication Date:
Language:
English
Physical Description:
vi, 102 leaves : ; 29 cm.

Subjects

Subjects / Keywords:
Boards of directors ( jstor )
Business structures ( jstor )
Chief executive officers ( jstor )
Economic indicators ( jstor )
Musical agency ( jstor )
Outside directors ( jstor )
Performance metrics ( jstor )
Personnel evaluation ( jstor )
Regression coefficients ( jstor )
Shareholders ( jstor )
Dissertations, Academic -- Management -- UF ( lcsh )
Management thesis, Ph.D ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2000.
Bibliography:
Includes bibliographical references (leaves 79-88).
Additional Physical Form:
Also available online.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Paula Silva.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
023009463 ( ALEPH )
45080385 ( OCLC )

Downloads

This item has the following downloads:


Full Text












STYLE VS. SUBSTANCE: THE NATURE OF CEO EVALUATION CRITERIA
IN MANAGEMENT CONTROLLED AND OWNER CONTROLLED FIRMS
















By

PAULA SILVA
















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF THE DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA

2000



















This research is dedicated to my mother and father,

Mary and Alberto Silva. I will always be grateful for their support and encouragement. Thank you for always believing in me without hesitation, and showing me that no dream is unattainable.















ACKNOWLEDGEMENTS

I am forever indebted to my advisor, Henry Tosi, whose guidance, direction, support, and inspiration were instrumental to this achievement. I am grateful to him for encouraging me in all my endeavors at the University of Florida, and helping me reach my fullest potential as an academic.

I would also like to thank the members of my

dissertation committee, Rich Lutz, Heather Elms, Steve Motowidlo and Judy Scully, for their insight and words of encouragement during the dissertation process. I am also grateful to Virginia Maurer, chair of the department, for always being thoughtful and supportive of my efforts at the University of Florida.

I am also thankful to have had supportive family

members and good friends that made the Ph.D. road a little smoother.













iii
















TABLE OF CONTENTS


page
ACKNOWLEDGEMENTS . . iii

ABSTRACT . . vi

CHAPTERS

1 INTRODUCTION AND LITERATURE REVIEW . 1

The Board of Directors . 3
Theory and Hypotheses . .8
Conclusion. . . 26

2 METHODS . . 27

Data Collection Method and Sample . 27 Measures . . ... 28
Independent Variables . 28
Dependent Variables .. . ...32
Control Variables . . 43
Summary .. . .44

3 RESULTS . . 46

Analysis of the Data . ...46
Regressions of Independent Variables by Dependent
Variable ..... ... .................... 49
Report of the Tests of the Hypotheses 55 Summary . . 64

4 DISCUSSION . . ... 65

Theoretical Implications . ...67
Limitations . . 74
Conclusion . . ...75

REFERENCES . . .79







iv










APPENDICES

A MAIL SURVEY MATERIALS . 89

B TABLES OF FACTOR ANALYSIS FOR OUTCOME BASED MEASURE AND BEHAVIOR BASED MEASURES 97 BIOGRAPHICAL SKETCH ............. 102





















































V
















Abstract of Dissertation Presented to the Graduate School of
the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

STYLE VS. SUBSTANCE: THE NATURE OF CEO EVALUATION CRITERIA
IN MANAGEMENT CONTROLLED AND OWNER CONTROLLED FIRMS By

Paula Silva

May 2000

Chairman: Henry L. Tosi, Jr. Major Department: Management

This dissertation examines the CEO performance

evaluation process and makes predictions concerning the relationship between the type of criteria used in the evaluation and the ownership structure of the firm, CEO influence, and board membership. The results suggest that the criteria used by boards in the evaluation of CEO performance are different for management and ownercontrolled firms. CEOs who are allowed considerable discretion in a firm are evaluated by indicators that are subjective and reflective of CEO style, while CEOs with minimum influence are evaluated by financial performance measures.





vi















CHAPTER 1
INTRODUCTION AND LITERATURE REVIEW

In 1992, the Securities and Exchange Commission (SEC) mandated that firms report in their proxy statements the salaries of their top paid executives and justify these compensation levels by measures of corporate performance (Tobin, 1994; Leonard, 1994). This ruling was not surprising given the amount of criticism in the media regarding excessive Chief Executive Officer (CEO) salaries. For example, it is estimated that CEO pay is 157 times higher than that of the average factory worker (Nett, 1993). The justifications for this disparity vary substantially. Some argue that CEOs are worth the compensation they receive (Marino, 1998; McNatt & Light, 1998), while others argue that CEOs are more than adequately paid (Kristie, 1998; Sridharan, 1996).

This dissertation addresses this general issue by examining the nature of the criteria that boards of directors use in determining the compensation of chief executive officers and how these criteria are justified in large firms in the United States. While numerous studies examine performance appraisals of those at lower organizational levels (Kirchner & Reisberg, 1962; DeNisi & Stevens, 1981; Landy & Farr, 1983), only a few studies have

1






2


investigated how boards of directors evaluate chief executive officers (Bushman, Injejikian & Smith, 1996; Antle & Smith, 1986; Gibbons & Murphy, 1990).

Research on CEO evaluation is a problem because

systematic data about CEO appraisals is difficult to get since the information is considered to be very confidential and boards of directors are reluctant to reveal specific information about such evaluations. However, this very paucity of research makes CEO performance appraisals a fruitful area for research since systematic examination of the CEO appraisal process may result in a better understanding of how CEO compensation packages are determined.

What is known empirically about the bases for CEO

compensation is derived from research that investigates the economic and social correlates of CEO compensation (Hallock, 1998; Hall & Liebman, 1998; Schaefer, 1998; Yermack, 1998; Gaver & Gaver, 1998; Westphal & Zajac, 1997; Wang, 1997; Zajac & Westphal, 1996b; Westphal & Zajac, 1995; Baker, Jensen & Murphy, 1988; Wade, O'Reilly & Chandratat, 1990). Specifically, the research on CEO performance evaluation has focused on two issues. First, researchers have investigated the utility of evaluating CEOs relative to the performance of CEOs in comparable industries (Antle & Smith, 1986; Janakiraman, Lambert & Larcker, 1992; Gibbons & Murphy, 1990). Second, a limited amount of research has shown nonfinancial, or subjective, criteria to be important in the






3


determination of bonuses for the CEO (Ittner, Larcker & Rajan, 1997; Bushman, Injejikian & Smith, 1996). Bushman, Injejikian and Smith (1996) suggest that CEOs with more power are able to influence the board to assess them by more subjective criteria that can be more prone to manipulation. However, this conclusion about the behavior, or actions, of boards of directors is inferred from the nature of the archival financial data used in the research rather than a more direct assessment of what boards themselves actually do.

This dissertation addresses the question of CEO

evaluation by the board by seeking to understand the process more directly, using survey report data from an informed observer within the firm to determine correlates of the use of financial and nonfinancial criteria by boards of directors in the evaluation process. This chapter reviews the role and responsibility of the board of directors, examines performance criteria used in the evaluation, and concludes with a discussion of theories that address the nature of the control relationships between CEOs and boards and their implications for the CEO evaluation process. Chapter 2 sets forth the methodology employed in the research; the results are reported in Chapter 3 and the conclusion and implications are set out in Chapter 4.

The Board of Directors

Boards of directors are legally charged with "the power to hire, fire, and compensate the top-level decision






4


managers and to ratify and monitor important decisions" (Fama and Jensen, 1983b: 311). The board of directors, in essence, protects the shareholders' interests by limiting opportunistic behavior by the CEO (Fama, 1980). This is achieved through control of the decision making of the CEO, and by ensuring that the CEO will not be solely responsible for significant wealth decisions of the firm (Fama and Jensen, 1983b).

The board of directors consists of inside and outside directors. Inside directors are top managers who work for the firm. Outside directors have no other link to the organization except maybe in the form of equity holdings. It is suggested that outside directors who hold equity in a firm might be better at protecting shareholder interests than inside directors (Hoskisson, Johnson, and Moesel, 1994; Morck, Shleifer & Vishny, 1988). This is because outside directors, in general, are assumed to be "independent" and more likely to act in the interests of shareholders, since inside directors are, by definition, bound to management in some manner (compensation, family ties, etc.) (Vance, 1990).

Research on this question has been inconsistent. It has been shown that as the percentage of outside directors increases, so does the proportion of equity-based CEO compensation that better aligns the interests of the CEO with the shareholder (Mehran, 1995). On the other hand, it has been demonstrated that CEO compensation is higher in firms consisting of boards with greater proportions of






5


outsiders (Westphal & Zajac, 1995; O'Reilly, Main & Crystal, 1988). This is attributed to the outsiders' increased reliance on the CEO for firm performance information, which can be subject to manipulation (Baysinger & Hoskisson, 1990), and also because outsiders may be more sympathetic to the CEO's purposes. Outsiders, some argue, may be more sympathetic in serving management interests rather than shareholders (Zajac & Westphal, 1996a; Westphal & Zajac, 1995). Zajac & Westphal (1996a) show that outside board members are appointed to boards according to their prior experience in strong or weak boards. Outsiders who participate in structural changes to the board (i.e., decreasing ratio of outsiders to insiders) which favor powerful CEOs are subsequently nominated to boards at firms where there is low board control. Likewise, board members who have made changes to the board that exemplify passivity towards management are nominated to boards with high board control (Zajac & Westphal, 1996a).

Boards that seek to increase their independence by making structural changes face CEOs determined to uphold their power. Westphal (1998) shows a relationship between a firm's increase in board independence and high level of ingratiating and persuasive tactics by the CEO that result in increases in CEO compensation. In firms where structural changes of the board increase board independence (i.e., increases in the number of outsiders, splitting the CEO/Chairman role), CEO interpersonal influence tactics






6


increase, decreasing the power of the board (Westphal, 1998). This finding suggests that even independent boards cannot isolate themselves from an influential CEO and that structural changes to the board of directors (i.e., increasing number of outsiders) to presumably increase board independence is not an effective method in reducing CEO power over the board.

The task of deriving the CEO compensation package is most often left up to the compensation committee of the board of directors (Tosi & Gomez-Mejia, 1989). Lear (1998) maintains that a strong, independent compensation committee is essential to controlling CEO compensation levels. The importance of the compensation committee in the determination of CEO pay has become increasingly important over time; in 1972, only 69% of boards of directors had compensation committees compared to 91% in 1992 (Townley, 1993). However, there is some evidence that membership on the compensation committee is not desired by directors (Crystal, 1991). This might explain the relative "passivity" the compensation committee shows in attending to compensation packages (Crystal, 1991).

The compensation committee decides what performance criteria should be linked to CEO pay. If the purpose of criteria is to align the interest of the CEO with the owners of the firm, then pay should be linked to objective measures of a firm's financial performance, as these measures are easily quantifiable and presumably reflect positive or






7


negative organizational performance from the perspective of equity holders. Among these economic criteria are shortterm and long-term performance goals using standard accounting measures (i.e., profit margins) and market measures (i.e., increasing price of the stock) (National Association of Corporate Directors, 1994;Verespej, 1994; Goldstein 1985; Nadler, 1986; Callen & Jet, 1993).

CEOs can also be evaluated by more subjective criteria, or nonfinancial measures, such as their leadership and management skills (i.e., planning, organizing, and decisionmaking), and behaviors, like establishing vision and direction, succession planning, customer satisfaction, and innovation (Ittner, Larcker & Rajan, 1997; Bushman, Indjejikian & Smith, 1996; Truskie, 1995; Kaplan & Norton, 1992) that contribute to the effective functioning of the organization. Truskie (1995) reports that leadership skills and behaviors are the common qualitative indicators assessed by the board of directors in CEO performance evaluations. Bushman, Indjejikian and Smith (1996) assert that financial measures are insufficient in capturing all the dimensions of CEO performance, and that nonfinancial measures can add important performance information beyond these financial measures. Further, the use of nonfinancial measures is especially important for firms where high information asymmetry exists between the CEO and equity holders and where there is substantial noise in the financial measures






8


(Ittner, Larcker, & Rajan, 1997; Bushman, Injejikian & Smith, 1996).

Theory and Hypotheses

The primary research issue that this dissertation addresses is the understanding of how powerful CEOs can exercise considerable control over the type of criteria used in their performance evaluation that is derived by the board of directors. As stated in the previous section, boards of directors use a combination of nonfinancial measures in conjunction with financial performance measures in the evaluation of CEO performance. Nonfinancial measures are inherently more subjective than financial performance measures and subject to manipulation and bias (Bushman, Injejikian & Smith, 1996). Further, Tosi and Gomez-Mejia (1989) argue that CEO pay is more highly correlated with firm economic performance when CEOs are not influential in the pay process, suggesting that financial performance measures are more important when boards are stronger. Thus, while CEOs are evaluated by both financial measures of firm performance as well as nonfinancial measures by the board of directors (Truskie, 1995), a CEO's relative influence over the board of directors is likely to influence the types of criteria he/she is evaluated against. Powerful CEOs may be more likely to obtain evaluations comprised of nonfinancial measures, as they are easily manipulated and transfer less compensation risk to the CEO.






9


Three theories that contribute to understanding these issues and which are useful for the development of specific hypotheses about the relationship between the CEO and board and the nature of criteria used in the CEO evaluation are agency theory, the theory of managerial capitalism, and attribution theory.

Agency Theory

In agency theory, the firm is defined as a set of explicit and implicit contracts that specify the responsibilities and reward structure of individuals within the organization (Jensen & Meckling, 1976; Ross, 1973). These contracts define the relationship between the owner (principal) and the agent (manager) (Fama and Jensen, 1983b), where the agent agrees to abide by the terms of the principal, and in return, is rewarded for such compliance.

Agency theory is based on the premise that a divergence of interest exists between those who own the firm and those who manage the firm (Jensen & Meckling, 1976), and that conflicts arise between management (agents) and owners (principals). Most typically, research on agency theory refers to the relationship between the board of directors (owner) and the CEO (manager) (Jensen & Meckling, 1976). Both the agent (the CEO) and the principal (the owners) are assumed to be rational and self-serving in the interests they pursue (Hunt & Hogler, 1990; Baiman, 1990). Therefore, logically, the CEO's goal is to maximize personal wealth and






10


will do so at a cost to owners, whose goal is to maximize shareholder wealth.

The crux of the agency problem is in determining and

enforcing the contract. These problems manifest themselves in "agency costs" to the owner that include monitoring the agent's behavior, and losses when the agent's behavior is not in alignment with the owner's interest (Fama & Jensen, 1983b). Agents can behave in a self-serving manner due to their relative proximity to organizational activities and the fact that agents possess information that could be difficult to get by the principal (Hunt & Hogler, 1990; Baiman, 1990). This information asymmetry between the principal and agent (Fama & Jensen, 1983a) maximizes chances for opportunistic behavior. The owner's goal is to reduce the opportunity for self-serving behavior by the agent (agency costs). To curb self-serving behavior by the agent, the principal utilizes two control mechanisms: monitoring and incentive alignment. Monitoring involves direct or indirect observation of the agent's activities by the owner (Jensen & Meckling, 1976).

Monitoring can be achieved through the board of directors, who are the appointed "stewards" of the shareholders. The board of directors is responsible for protecting the interests of the shareholders through their governance of the compensation contract (Jensen & Meckling, 1976), curbing the opportunistic behavior of the agent, and in general, reducing the agency costs (Fama & Jensen,









1983b). Monitoring the agent's (CEO) behavior includes designing contracts that focus on behavior-oriented criteria or outcome-based evaluation criteria that are related to improving the welfare of owners or increasing shareholder wealth. Behavior-based contracts provide for evaluating and rewarding the performance of the agent by their behaviors or actions. Outcome-based contracts reflect the evaluation and reward of the agent's performance by changes in firm performance measures (i.e., increases in the price of the stock) (Ouchi, 1979).

Incentive alignment is a mechanism through which the

interests of the managers are aligned with the owner through compensation packages that tie the agent's pay to firm performance (Eisenhardt, 1989). Tying pay to measurable results (those determined to be in the owner's best interests) increases the risk shared by the owner and manager. Therefore, when pay is tied to firm performance, agents are more likely to make decisions that are in the interests of the owner (i.e., increasing shareholder wealth) (Meredith, 1992). Firms that tie pay to firm performance are using outcome-based contracts that evaluate the agent's performance according to the firm's financial performance (Ouchi, 1979). These outcome-based contracts are the best method to reduce opportunistic behavior by the agent by transferring some risk to the manager and better aligning the interests of the agent with the owner (Jensen & Meckling, 1976).






12


Monitoring and incentive alignment may be

interchangeable control mechanisms. For example, in situations where the agent's behavior is not easily monitored, incentive alignment is expected to be the best control mechanism to reduce agency costs. When monitoring is costly or difficult, the contract is expected to be outcome-based to minimize self-serving behavior by the agent (Demski & Feltham, 1978) in that outcome-based contracts rely on incentive alignment to ensure the agent's behavior is aligned with shareholders. When the agent's behavior is easily monitored, a behavior-based contract is most efficient to minimize the risk to the agent (Demski & Feltham, 1978).

Managerial Capitalism

The basic assumption in agency theory is that the owner is able to structure controls in an efficient way that is in the best interests of equity holders. However, if the CEO has considerable influence over the board of directors, the nature of the relationship between the principal and agent changes and the agent, in essence, controls the firm.

The theory of managerial capitalism is an approach

based on the idea that the managers who are responsible for running the day-to-day activities of the firm have control of the firm when stock ownership is widely dispersed (Berle & Means, 1932). Therefore, owners are not able to oversee these activities and are not able to discipline these managers (Berle & Means, 1932). Firms in which ownership






13


holdings are widely dispersed so that owners cannot effectively exert decision controls are called managementcontrolled firms. Firms in which there are equity holders with large enough equity holdings to exert decision controls are called owner-controlled firms (Berle & Means, 1932).

Since, all other things being equal, managers prefer a contract that involves less monitoring and a reward system where they will bear minimal risk (Harris & Raviv, 1979), a manager in a management-controlled firm will design a contract that maximizes his/her self-interest through influence over the board. This often is reflected in strategies to grow the firm (a focus on increasing firm size) rather than making decisions that would maximize shareholder wealth (Berle & Means, 1932; Larner, 1966; Marris, 1964). The reason is that compensation risk is lower when linking pay to firm size, larger size accords more power and prestige to those managing the firm, and larger firms have a greater likelihood of survival (Baumol, 1967; Marris, 1964).

There is substantial evidence supporting managerial capitalism (Tosi & Gomez-Mejia, 1989; Singh & Harianto, 1989; Tosi & Gomez-Mejia, 1994; Salancik & Pfeffer, 1980; Amihud & Lev, 1981; McEachern, 1975; Palmer, 1973). The evidence suggests that 1) there is less pay risk in management-controlled firms compared to owner-controlled firms (Baumol, 1967; Hambrick & Finkelstein, 1995; GomezMejia, Tosi, & Hinkin, 1987; McEachern, 1975) and 2)






14


monitoring activities are greater in owner-controlled firms compared to management-controlled firms (Tosi & Gomez-Mejia, 1989; Tosi & Gomez-Mejia, 1994). For example, studies show that in management-controlled firms, changes in CEO pay are positively associated with change in sales, while in ownercontrolled firms, changes in pay are positively associated with changes in firm performance (Baumol, 1967; Hambrick & Finkelstein, 1995; Gomez-Mejia, Tosi, & Hinkin, 1987; McEachern, 1975). Further, in owner-controlled firms, CEOs may be penalized for decreases in profit (Hambrick & Finkelstein, 1995).

In a management-controlled firm, the manager is likely to control the nature of his compensation package through a contract that is behavior-based. The reason is that a behavior-based contract minimizes performance risk in that CEO pay is less dependent on financial output measures. Instead, a behavior-based contract relies on information systems that evaluate performance by making judgments about CEO behavior (Demski & Feltham, 1978). Thus, in managementcontrolled firms, because there is no large single shareholder to influence firm strategies and practices, the evaluation of the CEO is more likely to be based on nonfinancial measures that evaluate CEO behavior (i.e., management or leadership skills of the CEO), since these indicators involve no pay risk. Therefore: Hypothesis 1: The use of the behavior-based measures of CEO performance will be greater in management-controlled firms than in owner-controlled firms.






15


On the other hand, though pay/performance sensitivity is not as strong in management-controlled firms (GomezMejia, Tosi & Hinkin, 1987), when financial performance is strong, the pay risk is diminished due to higher firm performance. In this case, the powerful CEO is likely to look toward the use of these traditional economic indicators and base a substantial part of the compensation on these output indicators (Gomez-Mejia, Tosi & Hinkin, 1987). Thus: Hypothesis 2: The importance of outcome-based performance measures should be significantly more important in CEO evaluation in high performing management-controlled firms than in low performing management controlled firms.

In owner-controlled firms the shareholders are likely to take an active role in monitoring the performance of the CEO. This should result in the evaluation and rewarding of CEO performance using an outcome-based contract. An outcome-based contract aligns the interests of the shareholder and manager (CEO) in tying compensation to financial performance measures of the firm. Therefore: Hypothesis 3: The use of outcome-based performance measures will be greater in owner-controlled firms than in management-controlled firms.

From the arguments and hypotheses above, it follows that the relative importance of the behavior-based and outcome-based criteria should also be different for management and owner-controlled firms. The financial performance measures should be given more importance in owner-controlled firms versus management-controlled firms (Tosi & Gomez-Mejia, 1994). Therefore:






16


Hypothesis 4: The use of outcome-based performance measures will be significantly more important in owner-controlled firms than in management-controlled firms while behaviorbased measures will be more important in managementcontrolled firms.

Other Bases of CEO Power

While equity concentration is a useful indicator of the power of the CEO relative to the board of directors, there are other proxies for CEO influence. This influence would be reflected in the relative length of time spent as CEO (or tenure), holding the chairman of the board position, and control over board meetings. These influence indicators are examined in the hypotheses that follow. CEO tenure

Generally, CEOs are able to exert influence over the

board of directors when they have held their position as CEO for a relatively long period of time. As CEO tenure increases, so does the relationship between pay and firm size, where there is a negative relationship between pay and stock returns (Hill and Phan, 1991). Further, there is a positive relationship between the use of nonfinancial criteria and CEO tenure (Bushman, Injejikian & Smith, 1996). Salancik and Pfeffer (1980) and Finkelstein and Hambrick (1989) report that CEO tenure is higher in managementcontrolled compared to owner-controlled firms, while controlling for firm performance. This suggests that in owner-controlled firms, boards of directors terminate CEOs for performance deficiencies more quickly than in






17


management-controlled firms. In management-controlled firms, the CEO is afforded relative discretion in decisionmaking, and opposition by any owner or group of owners of the firm is unlikely given the wide dispersion of ownership (Salancik & Pfeffer, 1980).

This leads to the following hypotheses:

Hypothesis 5: The use of behavior-based measures will be greater in firms where CEO tenure is high versus firms where tenure is low.

Hypothesis 6: The use of outcome-based peformance measures will be greater in firms where CEO tenure is low versus firms where tenure is high.

CEO duality

CEO duality refers to whether the CEO holds, at the same time, the position of board chair. It is estimated that over 70% of CEOs also hold the position of chairman of the board (Fierman, 1990). Permitting the CEO to hold this dual role has been cited to be one of the causes of failure of the board to govern properly (Fierman, 1990; Ettorre, 1992). The chairman of the board is responsible to the shareholders and is supposed to be relatively detached from the organization. However, the dual role not only obscures the separation of the ownership and management interests, but also leaves the CEO/chairman with seemingly no one to answer to (Zajac & Westphal, 1996a; Ettorre, 1992). Research shows that CEOs have higher compensation in firms where they serve as chairman of the board compared to firms in which these two roles are held by different persons









(Fosberg, 1999; Sridharan, 1996). Thus, the CEO who has the additional power of the chair of the board of directors should be in a position to affect favorably the design of his/her compensation contract. Therefore: Hypothesis 7: The use of behavior-based measures will be greater in firms where the CEO is chairman of the board compared to firms in which the CEO is not chairman. Hypothesis 8: The use of outcome-based performance measures will be greater in firms where the CEO is not chairman of the board versus firms where the CEO is chairman.

Board meeting control

In a comprehensive study of boards of directors, Lorsch and MacIver (1989) suggest that the CEO's relative power base stemmed from control of board meetings by possessing superior knowledge of the company's activities, controlling the agenda, monitoring the types of information received by the board, and the leading of the boardroom's discussions. Boards do not seem to view their position as one of "power" to wield over the CEO, but rather an advisory and consultative position (Chitayat, 1985). Lorsch and MacIver (1989) found that directors' respect for the CEO was very important in board meetings and expressing opposition to the CEO was not appropriate. Tosi and Gomez-Mejia (1989) have shown that the CEO in the management-controlled firm is the most important actor in the CEO compensation setting process. Therefore, it is expected that when the CEO does have substantial control of board meetings, regardless of






19


the nature of the ownership structure, he or she is able to reduce compensation risk. Thus: Hypothesis 9: The use of behavior-based measures will be greater in firms where CEO influence in board of directors meetings is high than in firms where CEO influence at board meetings is low.

Hypothesis 10: The use of outcome-based performance measures will be greater in firms where CEO influence in board of directors meetings is low versus firms where CEO influence at board meetings is high.

Attribution Theory

The evidence from managerial capitalism theory suggests that boards of directors are not strong in protecting shareholders' interests when the manager controls the firm (Tosi & Gomez-Mejia, 1989; Singh & Harianto, 1989; Amihud & Lev, 1981). This would imply, as hypothesized above, that boards are using different types of contracts in evaluating CEO performance. When the owner controls the firm, there is more pay risk (Hambrick & Finkelstein, 1995; Gomez-Mejia, Tosi & Hinkin, 1987), suggesting the use of outcome-based contracts that rely on financial performance measures to evaluate CEO performance. In management-controlled firms, on the other hand, there is less pay risk and monitoring of the CEOs behavior (Tosi & Gomez-Mejia, 1989; Hambrick & Finkelstein, 1995), suggesting the use of behavior-based contracts. If this is so, that boards are using different criteria to evaluate performance, and with the increased attention to CEO pay in the media and the SEC rulings, it is important that directors are able to justify the criteria






20


they use, whether behavior-based or outcome-based. The main point in this study is that since outcome-based criteria are more likely to align management and owner interests, boards should tend to make more use of them and discount, as a general rule, behavior-based indicators. Since the evidence suggests that they do not do this in the case of powerful CEOs, there is bias in the CEO evaluation process.

This bias can be understood in terms of attribution theory (Weiner, 1971). Attribution theory centers on the explanation of how people make attributions about "causeeffect" relationships. For purposes here, these causeeffect relationships can be seen in terms of the types of CEO evaluation criteria: outcome-based measures can be thought of as causes of firm performance, behavior-based criteria can be seen as the effects of the CEO behavior. The question becomes one of whether the focus of the evaluation is on the cause (behavior-based criteria) or on the effect (outcome-based criteria).

Among the set of factors that can explain biased

evaluation is affect, or likeability, towards an individual being rated. For example, "liking" can bias a rating in that raters respond more negatively and intensely when rating poor-performing-but-disliked individuals. However, when rating poor-performing-but-liked individuals, raters attribute poor performance to external forces or factors that are beyond control by the ratee (Dobbins & Russell, 1986; Cardy & Dobbin, 1986). These occurrences of biased






21


ratings have been demonstrated in research on performance ratings, where, for example, it has been shown that raters attribute female success at work to luck rather than ability (Feldman-Summers & Kiesler, 1974).

Such results suggest that boards that have positive

affect for their CEOs are more likely to respond positively and intensely when making performance evaluations of the CEOs simply because they like the CEO. CEOs are likely to select board members who share the same ideals and beliefs about the strategic direction of the firm. This shared vision can result in biased performance evaluations in that poor firm performance will be attributed to uncontrollable forces, and good performance to the CEO's leadership (Kerr & Kren, 1992). Further, CEOs can use interpersonal influence tactics to increase positive affect for them by board members (Westphal, 1998). Westphal (1998) shows that CEOs increased the use of persuasion and ingratiation tactics when boards of directors attempted to increase independence through structural changes. These interpersonal influence tactics, used as an alternative source of power, resulted in increases in CEO compensation (Westphal, 1998).

In addition, board members may be selected based on

their similarities to the CEO. What has been termed social comparison theory (Festinger, 1954) is the process individuals go through in the selection of others who will evaluate them. In the firm's selection of outside board members, they will look for other CEOs who are perceived as






22


equal to or better than the focal CEO, and this benefits the CEO (O'Reilly, Main & Crystal, 1988). For example, O'Reilly, Main and Crystal (1988) show that boards remunerate the CEO similarly to their salaries in their own firm. These authors postulate that board members are selected on the basis of their salary, increasing the likelihood that they will compensate the CEO similarly. Further, Westphal and Zajac (1995) found that increased demographic similarity (i.e., similar educational and functional background) between the CEO and board members is significantly related to increases in compensation levels of CEOs.

Placing these findings in an attribution theory framework leads to a set of hypotheses based on the inference that positive affect by the board will be higher when board members are similar to the CEO and when they they are obligated by the firm or the CEO in other ways. This should lead the design of behavior-based contracts with lower pay risk for the CEO. Thus, CEOs can wield considerable discretion in firms where they are able to influence the composition and operation of the board. The board characteristics hypothesized to be important indicators of CEO influence are the proportion of inside directors of a firm, the proportion of outside directors who are past or present CEOs of other firms, and board pay (Westphal & Zajac, 1995; O'Reilly, Main & Crystal, 1988; Sridharan, 1996; Boyd, 1994).






23


Board Characteristics Hypotheses Inside/outside directors

There are conflicting theories and evidence about the role of inside and outside directors as board members. Agency theory argues that larger proportions of outside directors should restrain the discretion of CEOs (Fama & Jensen, 1983b). They are presumably better "stewards" of the company's resources. In management-controlled firms where dispersion of ownership is greater, monitoring in the form of greater proportions of outsiders on the board of directors is predicted to align the interests of management and shareholders (Li, 1994). However, contrary evidence suggests that the role of the outsider might not be so (i.e., Mallette & Fowler, 1992). Research shows that CEO compensation is higher in firms with greater proportions of outside directors controlling for firm size and performance (Westphal & Zajac, 1995; O'Reilly, Main & Crystal, 1988).

Inside directors, who essentially work for the

organization, are not expected to be good monitors of CEO behavior (Fredrickson, Hambrick & Baumrin, 1988). Research has shown a positive relationship between both a relatively high proportion of inside directors and CEO board control in the nomination of new directors (Westphal & Zajac, 1995), and CEO compensation (Sridharan, 1996; Williamson, 1963).

Another theoretical perspective, tournament theory,

suggests that inside directors should be better monitors of CEO performance, as they have more to gain from removing a






24


CEO from office since they are typically senior executives in the firm who might be vying for the CEO's position (Lazear & Rosen, 1981). On the other hand, insiders may have a lot more to lose than gain in firms where they actively seek to jeopardize the CEO's position.

The hypotheses that follow are based on agency theory rather than tounament theory because it seems to be logical to postulate that a board with a high ratio of insiders to outsiders is more likely to focus on outcome-based measures of firm performance, while a board dominated by insiders is more likely to use behavior-based measures in the evaluation of CEO performance. Since the insider works with the CEO on a daily basis, he/she is likely to not only share the same strategic vision for the firm, but also have a collegial and friendly relationship with the CEO. Therefore: Hypothesis 11: The larger the proportion of inside directors, the greater the use of behavior-based measures in the CEO performance evaluation.

The conflicting evidence about the role of the outsider suggests the existence of mediating factors in the relationship between CEO influence and proportion of outsiders. Since there is research that shows that boards tend to nominate outside members who are either current or past CEOs of other companies (Aigner, 1992), this suggests that one of these mediating factors might be the proportion of outsiders on a board who are presently or formerly CEOs of other firms. The rationale for this is that CEOs are likely to ascribe to the philosophy "do unto others as you






25


would have done to yourself" (Aigner, 1992 ; Hiltzik, 1996). Therefore it is proposed that CEOs of other firms who are on boards are likely to look after each other and recommend higher cash compensation packages, leading to the following hypothesis.

Hypothesis 12: The larger the proportion of outside directors who are present or past CEOs, the greater the use of behavior-based measures in the CEO performance evaluation.

Board compensation

There is a positive relationship between the

compensation that board members receive for their services and CEO compensation (Boyd, 1994). The implication of this finding is that board members with high compensation contracts may not be as vigilant as board members who make less money, as the board would have a similar-to-me attitude in CEO compensation decisions. Further evidence suggests that outside board members who are top executives at other firms remunerate CEOs similarly to the compensation they receive at their respective firms (O'Reilly, Main & Crystal, 1988). O'Reilly, Main and Crystal (1988) found a positive relationship between outside directors' compensation and CEO compensation. One would expect this similar-to-me attitude to extend to the remuneration of board members for board service, where the more you pay board members, the greater the CEO compensation. Hypothesis 13: The higher the remuneration of the board of directors, the greater the use of behavior-based measures in CEO performance evaluation.






26


Conclusion

The performance evaluation of the CEO is an important factor for CEO compensation packages. It can be used as a determinant of the package or as a justification for it. While there are a multitude of studies examining CEO pay and its determinants, what has been overlooked is the performance appraisal process in determining CEO pay. By understanding the performance evaluation process, it may be possible to discern how CEO pay decisions are made and maybe provide some answers to the public outcries for justification of CEO salaries.















CHAPTER 2
METHODS


This chapter discusses the research methodology

utilized in the testing of the hypotheses discussed in Chapter One. It is organized into two sections: data collection method and sample and the measures.

Data Collection Method and Sample

A questionnaire was sent to 1,000 American Compensation Association (ACA) members who were top executive compensation officers of their respective firms. They were selected based on the premise that they would be knowledgeable informants about the nature of CEO compensation in the firm and the criteria used in the evaluation for that compensation. The ACA members received a cover letter, the survey, a postage-paid envelope for returning the completed survey, and a postage-paid postcard that could be returned separately if they wanted to receive a copy of the results of the study. The cover letter explained the nature of the study in furthering understanding of the area of CEO compensation and ensured all participants strict confidentiality and anonymity (see Appendix A for copies of the mail survey materials). Approximately three weeks after the initial mailing, 27






28


reminder postcards were sent to all respondents in order to increase the response rate. Phone calls were made after a period of several weeks after the reminder postcards to solicit additional responses. Of the 1,000 surveys that were sent out, 201 were returned for a response rate of 20.1%.

Measures

The questionnaire consisted of a set of items

describing firm characteristics and demographics, board characteristics, items used to assess the importance of behavior-based criteria and outcome-based CEO evaluation criteria, and items determined to be important to the testing of the hypotheses. These specific measures are listed below.

Independent Variables

Ownership Structure

Ownership structure is typically determined by the

dispersion of shares owned by individuals or institutions. A firm is called owner-controlled if an individual or institution holds more than 5% of that firm's shares. A firm is called management-controlled if no one individual or institution holds greater than 5% of that organization's shares (McEachern 1975; Gomez-Mejia, Tosi & Hinkin, 1987; Tosi & Gomez-Mejia, 1989; Hambrick & Finkelstein, 1995). It is called owner-managed when an internal manager holds 5% or more of the stock.






29


Firms were classified into these three ownership categories. The ownership structure of the firm was determined by two items used in prior studies (i.e., Tosi & Gomez-Mejia, 1989). The first item asked respondents whether there was "any single individual or institution outside the firm that owns 5 percent or more of the company stock." Firms were classified as owner-controlled if the answer was "yes." The total number of owner-controlled firms in this study was ninety-eight (n=98). Firms were coded as management-controlled if the answer was "no." The total number of management-controlled firms in this study was ninety (n=90). Ownership structure was coded as a dummy variable where the answers from owner-controlled firms were coded as "1," and management-controlled firms "0."

The second item assessed whether a firm was ownermanaged and asked the respondent whether there was "any individual manager within the firm (e.g., the CEO) who owned 5 percent or more of the company stock." An answer of "yes" to this question indicated the firm was owner-managed. Since the number of owner-managed firms was only four, owner-managed firms were removed from the analysis. Board Characteristics

The study assessed three characteristics of the board. These were: proportion of insiders relative to outsiders, whether or not any of the board members had been past or present CEOs of other firms, and board pay.






30


Proportion of insiders

This variable was assessed by the subject's response to the following items: "how many directors are also executives in the firm" and "how many directors are from outside the firm." These two items were summed together and divided by the number of inside directors. Directors past/present CEOs

This variable was assessed by the subject's response to the following item: "how many of the outside directors are present or past CEOs of other firms?" Board pay

Board compensation was determined by the subject's

response to the following item: "what is the average cash compensation for board membership per year. CEO Influence and Power

There were three different measures of CEO influence: CEO tenure, CEO duality, and board meeting control. CEO tenure

This variable was assessed by the subject's response to the following item: "how long has the CEO held this position with your firm."

CEO duality

CEO duality was assessed by the subject's response to

the following item: "is the current CEO also chairman of the board." Duality was coded as a dummy variable. If the CEO was also chairman, the answer was coded as "1," and if not, 11 0 .






31


Board meeting control

The extent to which the CEO has influence in running the meetings of the board was assessed by a set of four items describing CEO board meeting control on a seven-point Likert-type scale ranging from "none"(1) to "very high"(7). The items were subjected to a principal axis factor analysis with varimax rotation which revealed that all items loaded on a common factor accounting for 85% of the variance. The scale had a reliability of alpha = .94. The following four items (shown with factor loadings) were used to assess CEO board meeting influence.

1. How much influence does the CEO have in reviewing

and approving the types of information received by

board members? (.66)

2. How much influence does the CEO have in setting the

agenda at board meetings? (.84)

3. How much influence does the CEO have in the running

of board meetings? (.80)

4. How much influence does the CEO have in important

board decisions ? (.72)

Firm Performance

Since more conventional archival measures of firm

performance were not available because of the anonymity of the subjects, overall firm performance was assessed by the subject's response to an item that asked him or her to indicate the firm's financial performance relative to other firms in its industry. This provided a subjective estimate






32


of relative performance in the industry and, as such, was a control for industry differences.

Such a measure of performance has been shown to correlate highly with conventional archival economic indicators of performance (Tosi & Gomez-Mejia, 1994) in that measures reported by respondents and those collected through archival means are shown to correlate, suggesting that respondents are able to accurately categorize performance, and that this method of categorization correctly represents a firm's industry performance. Subjects responded to the following item: "please rate your firm's financial performance relative to others in your industry." The scale measuring firm performance consisted of a five-point scale (1= top performers to 5= lowest performers) dividing relative firm performance into percentages ranging from top 20% to lowest 20%.

Dependent Variables

There are two types of dependent variables in this

research. One type are outcome-based performance measures that are typically objective measures of a firm's financial performance. The second are behavior-based measures of CEO performance. These are more subjective indicators that measure different facets of the CEO's style of managing. Both were assessed with Likert-type scales on which respondents indicated the relative importance of each in CEO evaluation on a seven-point scale ranging from "not at all important"(1) to "very important"(7).






33


Outcome-Based Performance Measures of CEO Performance

There are two outcome-based performance measures of CEO performance that may be used in the evaluation of CEO performance by boards of directors: accounting measures of firm performance and\or market measures of firm performance. Accounting measures assess a firm's current profitability by analyzing financial information from operating statements and balance sheets of the firm. Accounting measures are important in top management compensation decisions and also in decisions related to the viability of a business.

Subjects were asked to indicate the extent to which the following six outcome-based measures of firm performance were used in the evaluation of the CEO.

1. Return on assets (ROA). Return on assets (ROA) is a

measure of a firm's effectiveness in deploying its

assets. ROA is the ratio of net income relative to

total assets. This measure is used in investment and

operating decisions (Logue & Steward, 1997).

2. Return on equity (ROE). Return on equity (ROE) is a

measure of return on invested capital. ROE is based on

net income relative to owner's equity. This measure

tells shareholders how effectively their money is being

used, and provides a basis for comparisons with

competitors (Logue & Steward, 1997).

3. Return on investment (ROI). Return on investment (ROI)

expresses profit in direct relation to investment. ROI

is a ratio of net income to total assets. It is used






34


for comparing firms or divisions in terms of management

efficiency and as the basis for whether the firm is

maximizing the interests of the owners (Logue &

Steward, 1997)

4. Sales growth. Sales growth reflects increases or

decreases in sales in a firm. It is usually measured as the change of sales from previous fiscal years (the

ratio of total sales this year to total sales last

year). Sales growth is used to determine how well a

firm's product or service is doing over time, and also

used to compare product lines.

5. Stock price increase. Changes in the price of the

stock of a firm is an indicator of the capacity of the firm to create shareholder value as it signals equity

holders how much their shares are worth relative to

earlier points in time.

6. Economic value added (EVA). Economic value added

reflects the economic returns generated to shareholders (Logue & Steward, 1997). It is a measure of the "gain

(or loss) that remains after levying a charge against

after-tax operating profits for the opportunity cost of

all capital equity as well as debt used to produce

those profits" (Ehrbar, 1999: 20). EVA is essentially

a measure of operating performance that tells the

shareholder how valuable the business is, and is therefore a good indicator of shareholder value.

In the typical research on CEO compensation that






35


examines whether it correlates with firm performance, there are two types of outcome-based measures used: accounting measures of firm performance and/or market measures of firm performance. Accounting measures assess a firm's current profitability by analyzing financial information from the operating statements and balance sheets of the firm. Accounting measures are important in top management compensation decisions and also in decisions related to the viability of a business. Accounting measures are thought to be less valid indicators of firm performance than market indicators because they reflect the accounting performance rather than the cash flow performance of a firm (Logue & Steward, 1997). Additionally, accounting measures are susceptible to manipulation in firms, as they can be overstated, and also distort the assessment of firm performance (Saloman & Smith, 1979). Thus, some believe that accounting measures do not represent the true value of the firm (Logue & Steward, 1997).

Market measures of firm performance reflect stock

market indicators that signal the worth of the equity of a firm. These measures represent the shareholders' interest as they are indicators of the value of the firm. They are preferred by some (Logue & Steward, 1997) because they are not as easy to manipulate by managers of a firm as are accounting measures (Rosen, 1990). However, they are subject to shifts in value that have no relationship to financial performance changes (Rosen, 1990). For example,






36


stock prices commonly fluctuate without any changes in the firm's value. Additionally, shifts in the market discount rate can also distort the interpretation of market measures (Rosen, 1990). However, there is evidence that neither is a superior measure to the other, so both may be useful measures of firm performance for research purposes (Rosen, 1990).

For this study, a research choice was not made to try

to assess these two different classes of indicators based on the assumption that the informants would not be able to discriminate usefully between them. Therefore, the set of financial criteria listed above was chosen. A factor analysis of all the outcome-based performance measures was executed to assess their relationship to each other for this research (see Appendix B). Three factors emerged, but the items that loaded within each factor had no apparent logic or theoretical basis of the type that one might expect. Because of the exploratory nature of this research and the lack of more objective archival measures, it was decided to use a pooled outcome-based performance measure. This was done by simply summing the subject' s responses to each of the six items above to attain a value used in the analysis. Pooled or combined measures of firm performance have been used elsewhere in studies that have used both archival financial data as well as data reported by knowledgeable informants (Tosi & Gomez-Mejia, 1994; Gomez-Mejia, Tosi &






37


Hinkin, 1987). The reliability for this pooled financial performance measure is .60.

Behavior-Based Measures of CEO Performance

The popular business press is replete with reports of what distinguishes effective CEOs from ineffective CEOs. There is, however, no well-developed theoretical or empirical literature on this point. It is true that there is an abundance of research on effective leadership in organizations (Bass, 1985; Bass, 1990; Fiedler, 1967; House, 1971) but it is fair to say that this research was done on effective managers at lower to middle organizational levels. Since this is the case, this research draws on theoretical literature that seems to be relevant to those characteristics described in the popular literature.

Three aspects of CEO behavior are examined as

indicators of behavior-based measures that boards of directors might use in the evaluation of CEO performance. These are: the behavior of the CEO in the managerial role (Mintzberg, 1973), the CEO as a leader (Bass, 1985), and the organizational citizenship of the CEO (Organ, 1988). These aspects of managerial behavior are conceptually quite different. The fundamental distinction is that CEO leadership behavior involves the direction of members to organizational goals, where managerial role performance focuses on the specific activities in different roles that leaders perform. Citizenship behavior reflects activities that go beyond the manager's role prescriptions. These






38


behaviors are expected to contribute to organizational functioning in that they facilitate the accomplishment of organizational goals.

Managerial role performance

Mintzberg (1973) developed a taxonomy of managerial roles from his study of CEOs. He contends there are 10 managerial role behaviors that fall into three general categories of managerial role performance:

1. The interpersonal role describes activities involving

the manager's relationships with others and consists of

three roles; figurehead, liaison, and leader. All of these roles involve behavior that is social in nature

and encompasses the manager's relationships with

employees, customers, the public, and stockholders.

2. The informational role includes those roles that deal

primarily with information, and consists of three

roles: monitor, disseminator, and spokesman. These

roles involve the receiving, transmitting, and general

handling of information by the manager.

3. The decisional role reflects the organizational

decisions that managers are responsible for. It

includes four roles: entrepreneur, disturbance handler,

resource allocator, and negotiator. These roles

reflect the manager's involvement in strategic

planning, scheduling work and budgeting resources, and

negotiating with outside organizations.

The subjects responded to a shortened version (29






39


items) of a managerial role behavior scale developed by Lau

and Pavett (1980). In order to determine whether these

items could be grouped into the same categories as Mintzberg

developed, they were subjected to a factor analysis using

principal axis with varimax rotation (see Appendix B for

factor analysis). A priori, 10 factors were expected to

emerge consistent with Minztberg's (1973) categorization,

but the analysis resulted in four factors. The factors that

did emerge were uninterpretable and inconsistent with prior

research. Given the exploratory nature of this study, it

was decided to keep the scale as originally devised to

assess one factor, managerial role performance. The

reliability for this measure is .89.

The items used to assess managerial role performance are:

1. Taps into the grapevine to keep abreast of what is
going on in the company
2. Is available to outsiders (e.g., consumers, suppliers,
the public) who want to go to the "person in charge"
3. Keeps up with market changes and trends that might have
an impact on the organization
4. Surveys the organization to identify ways to improve
performance or cut costs
5. Integrates subordinates' goals (e.g., career goals,
work preferences) with the organization's work
requirements
6. Keeps professional colleagues informed about the
organization
7. Entertains clients on a regular basis
8. Negotiates with groups outside the organization for
necessary materials, support, commitment, etc.
9. Attends to staffing requirements in different units,
such as hiring, firing, promoting, and recruiting
10. Provides guidance and direction to company employees 11. Keeps members of the organization informed of relevant
information through meetings, conversations, and
written information
12. Develops new contacts by personally answering requests
for information
13. Takes immediate action in response to a crisis or "fire
drill"






40


14. Stays tuned to what is going on in competitor
organizations
15. Passes along unofficial information 16. Joins boards, organizations, clubs, or does public
service work that might provide useful, work-related
contracts
17. Passes on the desires and preferences of top management
to others in the firm
18. Keeps executive members, consumers, or other important
groups informed about the organization's activities and
capabilities
19. Evaluates the outcomes of internal improvement projects 20. Plans and develops improvements in work flow or work
methods
21. Programs work for different units within the
organizations (what is to be done, when and how)
22. Resolves disputes between subordinates or work groups 23. Allocates resources (e.g., personnel, money, material)
among units within the organization
24. Determines the long-range plans and priorities of the
organization
25. Is willing to be the official spokesperson of the
company at gatherings outside of the organization 26. Resolves work flow problems between units 27. Personally answers letters or inquiries about the
organization
28. Attends outside conferences or meetings 29. Personally handles complaints from customers or
subordinates

CEO leadership behavior

The transformational leadership behavior of the CEO was

assessed by the informant in each firm, drawing on the

conventional wisdom of the business press that effective

CEOs are good leaders. For example, an effective

transformational leader like Jan Carlzon, CEO of SAS

Airlines, encourages a participative environment where the

key stakeholders of the firm are involved with the

formulation of the organizational vision. Carlzon believes

that a leader must inspire others to pursue the vision, but

the vision must be inspiring in order to be followed

(Ackoff, 1999).






41


There is evidence that certain leadership styles or behaviors are associated with improved subordinate performance (Avolio & Bass, 1988; Bass & Avolio, 1990; Seltzer & Bass, 1990). For example, Seltzer and Bass (1990) found that transformational leadership is positively related to beneficial organizational outcomes though these outcomes were not outcome-based performance measures of the type discussed earlier. Other research has shown the effects that transformational leaders can have on organizations (Waldman, Bass & Einstein, 1987). For example, Waldman, Bass and Einstein, (1987) show that transformational leadership is positively correlated with employee satisfaction with performance evaluation processes.

CEO leadership behavior was measured with a seven-item scale that was a modification of a measurement instrument developed by Podsakoff et al. (1990). Subjects' responses were factor analyzed using principal axis with varimax rotation, and the resulting scales emerged (see Appendix B for factor analysis). Two factors emerged that were inconsistent with prior findings (Podsakoff et al., 1990), and not interpretable. It was therefore decided to use the entire scale in the analysis to assess CEO leader behavior. The reliability for this measure is .77. The items used to assess CEO leadership behavior are:

1. Provides employees with new ways of looking at things
2. Leads by doing rather than simply by telling
3. Insists on only the best performance
4. Seeks new opportunities for the organization
5. Inspires others with his/her plans for the organization
6. Encourages employees to be "team players"






42


7. Shows respect for individuals' personal feelings Citizenship behavior

Effective CEOs are also often seen contributing in other ways than filling their leadership and management roles. Organizational citizenship behavior is a concept that captures this aspect of CEO behavior (Organ, 1997). Organizational citizenship behavior is defined as "performance that supports the social and psychological environment in which task performance takes place" (Organ, 1997: 95). There are five dimensions of citizenship behavior. These are altruism, conscientiousness, sportsmanship, courtesy and civic virtue. An abbreviated scale of fourteen items to assess these five dimensions was taken from earlier existing scales (MacKenzie, Podsakoff & Fetter, 1991; Bateman & Organ, 1983; Organ, 1988; Podsakoff & MacKenzie, 1994; Smith, Organ & Near, 1983).

Responses were factor analyzed using principal axis

with varimax rotation (see Appendix B for factor analysis). The factors that emerged were inconsistent with prior research (Bateman & Organ, 1983; Organ, 1988) and not clearly interpretable, therefore it was decided to combine all the items to create a single factor that assesses citizenship behavior. The reliability for this measure is .58. The items used to assess citizenship behavior are:

1. Acts cheerful
2. Trains or helps others to perform their jobs better
3. Keeps up with developments in the company
4. Expects themselves and others to conscientiously follow
company regulations and procedures






43


5. Exhibits punctuality in arriving at meetings
6. Attends functions that are not required, but that help
the company image
7. Willingly risks disapproval in order to express beliefs
about what's best for the company
8. Takes steps to try to prevent problems with other
executive members, board members and/or other employees
in the company
9. Cooperates well with others in the organization 10. Makes innovative suggestions to improve the overall
quality of the company
11. Focuses on the positive side of a situation rather than
the negative side of it
12. Willingly gives of his/her time to help others 13. Returns phone calls and responds to other messsages and
requests for information promptly
14. "Touches base" with others before initiating actions
that might affect them

The Relative Importance of Outcome-Based and Behavior-Based Measures of CEO Performance

One item was developed to understand the overall

relative importance of behavior-based and outcome-based

criteria in the CEO evaluation by the board of directors.

The following item was used to assess the relative

importance of the different types of criteria: "Rate the

current relative importance of quantitative criteria or

measures as compared to qualitative criteria or measures in

the determination of the CEO compensation package." A

seven-point scale was used in this assessment with the

following ranges:

1 "Qualitative most important"

4 "About equal"

7 "Quantitative most important"

Control Variables

The following control variables were used in the analysis.






44


Industry

In order to control for industry effects, two items

were used. One asked the respondents to report the firm's SIC classification, with the intention to control for a specific industry in the analysis of the data. There was a total of eight SIC categories. Because a majority of respondents in this study were from manufacturing firms (n=95), this variable was coded as a dummy variable with "0," assigned to manufacturing firms and all other firms coded as "1".

The second item assessed whether the firm's main products were durable or nondurable goods, since it is possible that the type of product a firm produces can have an effect on different firm performance measures (Tosi & Gomez-Mejia, 1994). This measure was used as a dummy measure in the analysis with the coding of "0," if the product was nondurable, and "1" if the product was durable. Firm Size

Firm size was also used as a control variable

(McEachern, 1975). Firm size was measured in this study by the subject's response to the following item: "how many people are employed by your firm."

Summary

This chapter describes the survey methodology utilized in the testing of the hypotheses. The specific measures used were items that had been developed in prior studies and some that were specifically developed for this study. As a






45


result of inconsistent findings from the factor analyses of the behavior-based and outcome-based performance measures, the decision was made to use single scales instead of subscales for each of the measures. The next chapter will describe the data analysis used to test each of the hypotheses and report results.














CHAPTER 3
RESULTS


The data were gathered from 201 firms representing 7 industries. The sample used in the analysis contained 188 cases due to incomplete data and data that came from firms that were not comparable to the rest of the sample.

Analysis of the Data

The primary objective of this study is to examine the effects of the seven independent variables on the two dependent variables as specified in the hypotheses. The means, standard deviation, and bivariate correlations for all variables are reported in Tables 3.1 and 3.2.

The secondary objective is to examine the relationships between CEO duality, tenure, influence at board meetings, and board characteristic variables on the outcome and behavior-based variables by ownership structure.

Regression models were run to test these relationships between the seven independent variables and each of the four dependent variables as outlined above. Linear regression was used to estimate the coefficients of the model.

This chapter is divided into two main sections because several different independent variables in various regression models were used to test the hypotheses. The first part of this chapter reports regressions for each 46







47



Table 3.1
Means and Standard Deviations


Variables Mean Standard Deviation
1. Size(# of emp) 32,405 83,558 2. Performance (quartile ranking) 2.23 1.12 3. SIC .49 .50 4. Sector .33 .47 5. Ownership Structure .52 .50 6. Duality .65 .47 7. Tenure (years) 6.46 4.52 8. CEO Bd. Mtg. Influence 6.37 1.13 9. % Dir Past/Present CEOs .41 .19 10. % Insiders .21 .98 11. Board Pay $30,000 $18,000 12. Citizenship behavior 3.58 .51 13. CEO leadership behavior 4.88 .86 14. Managerial role performance 4.35 .72 20. Outcome-based performance measure 4.14 .97 21. Importance outcome vs behavior-based 5.56 1.19

N = 188


















10o




I n

NA
H
C4'm




10H



100N
41 'C



H C3


mmoN ILn w 0 (
4' (
14 H IN H%



IHMHlNN
0 imNMOH


r-l 0* m
5000><>0
0 I-A 0 .0 10 H


(


SH m O m w v

SCHNmOO H

HI OD -r-W HI I



w -W ~Ln 1-4 00 0 N r-l
4' 4' 4K
0~ ~~~~ '.0 m7 m CD' NNNoL r z




'T 0 H (NHNMNM

H I
4'14
(1' H m~CD ( 0 mIN m M H 0 7 H m Ocoomoooooos
In H, (N m ( C' C' '3' I






(dk HO Nmm ammNH
Nq m oooooooa O ** C -r
0 savoo mars wea O
N OOOHH HOOOO HOH

UNumN~~ommem~mo
I I I w m N N 0 m) nHLm
0 NlrA0 0 IN 0 0.)lr-










-H
(1)


-ro
Q). a)


U

4 0 U
H0 (10 (N .D m ~ N E-44 m WH 0oHO H 0a 44 4' U H







r-A
O

44 U 1 N a H Q(NN m




OMO~





ra W Q) 0 4 *
(N I 4 MOOHHHOO NO )HO 0 H C)-H H H .) 0 0 0 0 4 Hr >0 r E4





N ~ &4 C000 0 r (
H 1 OQU 0
-OO






CD *4 vi >I 00 m HHO 90 OummbdomOU
C 4 >E mu -MIC H H 0 0CD



-OHm me O



CD CDO 000000 4dC DN OOMMC: ( dOQ~HO-Hampy( O
D m o u VI a')CD0~C:CD D CD .5a. *CJ0 CCDCS4CD--4 4C4Jo H NH010 CDm ~ ~~~~~~ 0 ~ -4-U~4H CU f






49


dependent variable with all the hypothesized independent variables and control variables. It also reports regressions of the relationship between the dependent variable on independent variables by ownership structure to ascertain differences in management and owner-controlled firms (Tables 3.3- 3.6). The second part reports the specific regression coefficients that are the tests of each of the hypotheses (Tables 3.7-3.18). Note, that these coefficients are drawn from Tables 3.3-3.6.

Regressions of Independent Variables by Dependent Variable

Outcome-Based Performance Measure

Table 3.3 shows the regressions for the effects of

ownership structure, board characteristics and CEO influence on the use of the outcome-based performance measure. These relationships will be examined in greater detail in the reporting of the hypotheses section.

Table 3.3 also displays regression coefficients for the effects of the same independent variables and dependent variables for owner-controlled and management-controlled firms.

In owner-controlled firms, the use of outcome-based

performance measures is negatively related to board pay (b=.16, p .05). Board members who are compensated less tend to use criteria that reflect the financial performance of a firm.






50



In both management-controlled (b=.53, p .01) and ownercontrolled firms (b=.34, p .01), larger proportions of


outsiders who are past or present CEOs of other firms are

positively related to the use of outcome-based performance

criteria.

In owner-controlled firms, firm size is positively

related to outcome-based performance criteria (b=.30,

p .01). Thus, the larger the firm the greater the use of


outcome-based performance measures.

Table 3.3
Regression Model of the Effects of the Independent
Variables on the Use of the Outcome-Based Performance Measure

Variables Total Sample Manager Control Owner Control
Std. B Std. Er. Std. B Std. Er. Std. B Std. Er. Ownership .29** .15
Duality -.06 .18 -.12 .30 .06 .29 Tenure -.04 .16 -.01 .24 -.05 .22 CEO Bd Mtg .02 .16 .05 .21 -.08 .30 Influence
% Dir. CEOs .40** .27 .53** .44 .34** .35 % Insiders -.01 .71 .10 1.01 -.06 .98 Board Pay -.03 .00 .13 .00 -.16* .00 Performance -.01 .14 .13 .18 -.12 .20 Size .20** .14 .03 .21 .30** .18 SIC -.01 .13 -.01 .18 -.05 .20 Sector .07 .15 .05 .22 .07 .21 R sq. .30 .36 .26 Adj. R sq. .26 .28 .17 F 6.91 4.55 3.03 P-value .000 .000 .002 N 90 98 Chow Test(F) 1.75

*p.05, **p.01






51


Behavior-Based Measures

Managerial role performance

Table 3.4 shows the regressions for the effects of

ownership structure; CEO duality, tenure and influence at

board meetings; and board characteristics on the use of the

behavior-based measure, managerial role performance. These

relationships are examined in the next section on hypotheses

testing.

Table 3.4
Regression Model of the Effects of the Independent Variables
on the Use of the Behavior-Based Measure,
Managerial Role Performance


Variables Total Sample Manager Control Owner Control Std. B Std. Er. Std. B Std. Er. Std. B Std Er. Ownership -.15* .11 Duality .07 .13 .17 .22 .04 .22 Tenure .02 .12 .06 .18 -.07 .17 CEO Bd Mtg Influence .07 .12 .07 .16 .10 .23 % Dir. CEOs .02 .20 .07 .34 -.03 .27 % Insiders -.01 .53 -.24* .80 .15 .75 Board Pay -.35** .00 -.33** .00 -.39** .00 Performance .14* .10 .18* .14 .14 .15 Size .03 .10 -.12 .16 .09 .14 SIC -.19** .10 -.18* .14 -.19* .15 Sector .02 .11 .04 .17 .08 .16 R sq. .21 .21 .25 Adj. R sq. .16 .16 .17 F 4.30 4.30 3.00 P-value .00 .00 .00 Chow Test (F) 1.43

*p.05, **p.01

Table 3.4 also reports regressions of the effect of the

independent variables on managerial role performance by

ownership structure. In management-controlled firms, the

proportion of insiders measure is related to the use of the

behavior-based measure, managerial role performance (b=-.24,






52


p .05). The smaller the proportion of insiders in management-controlled firms, the greater the use of this behavior-based criteria.

Board pay is negatively related to the use of this behavior-based measure in management (b=-.33, p .01) and owner-controlled firms (b=-.39, p .01). Board members who are compensated less in firms tend to use behavior-based measures, like managerial role performance. CEO leadership behavior

Table 3.5 shows the regressions for the effects of

ownership structure; CEO duality, tenure and influence at board meetings; and board characteristics on the use of the behavior-based measure, CEO leadership behavior. Specific relationships are discussed in the hypotheses section.

Table 3.5 also reports regressions of the relationship between the independent variables on the use of the measure, CEO leadership behavior by ownership structure. In management-controlled firms, CEO duality is significantly related to the use of this behavior-based indicator (b=.30, p<.01). When CEOs also serve as chairman of the board, the greater the use of the measure, CEO leadership behavior in management-controlled firms.

In management-controlled firms, the proportion of insiders measure is negatively related the use of this behavior-based measure (b=-.23; p .05). Board of directors with low proportion of insiders in management-controlled






53


firms are less likely to use this indicator in CEO

evaluation.

Table 3.5
Regression Model of the Effects of the Independent Variables
on the Use of the Behavior-Based Measure, CEO Leadership Behavior


Variables Total Sample Manager Control Owner Control Std. B Std. Er. Std. B Std. Er. Std. B Std. Er. Ownership .00 .14 Duality .25** .16 .30** .29 .15 .27 Tenure -.08 .14 -.14 .23 -.03 .20 CEO Bd Mtg Influence .04 .15 -.01 .20 .15 .27 % Dir. CEOs .15* .24 .15 .43 .17* .32 % Insiders -.12* .65 -.23* 1.02 -.05 .91 Board Pay -.32** .00 -.33** .00 -.26** .00 Performance .06 .12 .07 .18 .06 .18 Size .05 .12 -.09 .21 .14 .17 SIC -.04 .12 -.12 .18 .07 .18 Sector .06 .13 .08 .21 .14 .19 R sq. .19 .18 .25 Adj. R sq. .14 .08 .17 F 3.81 1.80 2.99 P-value .00 .03 .00 Chow Test (F) .72
*p.05, **p.01



In both management (b=-.33, p .01) and owner-controlled


(b=-.26, p .01) firms, board pay is negatively related to

the use of the behavior-based measure, CEO leadership

behavior. Low board compensation is related to the use of

this measure. In owner-controlled firms, the proportion of

outside directors who are CEOs of other firms is positively

related to the use of the measure, CEO leadership behavior

(b=.17, p .05). The greater the proportion of these

directors in owner-controlled firms, the greater the use of

this indicator.






54


Citizenship behavior

Table 3.6 shows the regressions for the effects of

ownership structure; CEO duality, tenure and influence at

board meetings; and board characteristics on the use of the

behavior-based measure, citizenship behavior. Again,

relationships between these variables are examined in the

hypotheses section.

Table 3.6
Regression Model of the Effects of the Independent Variables
on the Use of the Behavior-Based Measure, Citizenship Behavior


Variables Total Sample Manager Control Owner Control Std. B Std. Er Std. B Std. Er. Std. B Std. Er Ownership -.20** .08 Duality .06 .09 .04 .18 .18 .15 Tenure .03 .08 .01 .14 .07 .11 CEO Bd Mtg Influence -.09 .09 -.01 .13 -.19 .15 % Dir. CEOs -.04 .14 .03 .27 -.06 .18 % Insiders .05 .38 .06 .65 .10 .50 Board Pay -.30** .00 -.31** .00 -.29** .00 Performance .09 .07 .18* .11 .00 .10 Size .02 .07 -.11 .13 .16 .09 SIC -.16* .07 -.18* .11 -.15 .10 Sector -.03 .08 -.12 .13 .04 .10 R sq. .18 .21 .15 Adj. R sq. .13 .11 .06 F 3.66 2.09 1.61 P-value .00 .03 .06 Chow Test (F) .73
*ps.05, **p .01


Table 3.6 also reports the regressions testing the

effect of the independent variables on the behavior-based

measure, citizenship behavior by ownership structure. In

management (b=-.31, p .01) and owner-controlled (b=-.29,


p .01) firms, board pay is negatively related to the use of

this measure. As with the other behavior-based measures,






55


the less board members receive for board service, the greater the use of the measure, citizenship behavior in CEO evaluation.

Report of the Tests of the Hypotheses Hypothesis 1

Hypothesis 1 states that the use of behavior-based

measures of CEO performance will be greater in managementcontrolled firms than in owner-controlled firms. The regression coefficients between ownership structure and the three behavior-based measures are shown in Table 3.7. Ownership structure is negatively related to the use of the behavior-based measure, managerial role performance (b=-.15; p .05) and citizenship behavior(b=-.20; p .01). Only CEO leadership behavior was not found to be significant. These results generally support Hypothesis 1.

Table 3.7
Regression Coefficients Testing the Effects of Ownership
Structure on the Use of the Behavior-Based Measures Dependent Variable Ownership Structure Managerial Role Performance
(tCoefficient taken from Table 3.4) -.15*t CEO Leadership Behavior
(tCoefficient taken from Table 3.5) .00t Citizenship Behavior
(tCoefficient taken from Table 3.6) -.20**t

*ps.05, **p .01

Hypothesis 2

Hypothesis 2 states that the importance of outcomebased performance measures should be significantly more important in CEO evaluation in high performing management-







56



controlled firms than in low performing management

controlled firms. The results of the regressions testing

Hypothesis 2 are reported in Table 3.8. In managementcontrolled firms, firm performance is significantly related

to the rated relative importance of outcome-based


performance criteria (b=.41, p .01). That is, these


outcome-based performance measures are used to a greater

extent in high performing management-controlled firms.

Table 3.8
Regression Model of the Effects of Ownership on the Rated Relative Importance of Outcome-Based Performance Measure
Compared to Behavior-Based Measures

Variables Total Sample Manager Control Owner Control Std. B Std. Er. Std. B Std. Er. Std. B Std. Er. Ownership .42** .17
Duality -.18** .20 -.18 .36 -.06 .29 Tenure -.01 .18 -.04 .29 .04 .22 CEO Bd Mtg -.01 .18 -.02 .26 -.05 .29 Influence
% Dir. CEOs -.12* .30 .03 .55 -.25** .35 % Insiders .00 .79 .05 1.30 .04 .97 Board Pay .17** .00 .14 .00 .26** .00 Performance .17** .15 .41** .23 -.07 .20 Size -.05 .15 -.10 .27 -.01 .18 SIC .05 .15 .09 .23 -.01 .19 Sector -.08 .16 -.09 .27 -.16 .21 R sq. .37 .25 .20 Adj. R sq. .33 .15 .11 F 9.40 2.64 2.19 P-value .000 .01 .02 N 90 98 Chow Test(F) 1.84*

*p .05, **p .01



Hypothesis 3

Hypothesis 3 states that the use of outcome-based

performance measures will be greater in owner-controlled

firms than in management-controlled firms. The results of

regressions that test Hypotheses 3 are reported in Table






57


3.9. Hypothesis 3, that ownership structure will be positively associated with the use of the outcome-based performance measure, is supported (b=.29, p 5.01). Ownercontrolled firms use outcome-based performance criteria in the evaluation of CEO performance to a greater extent than management-controlled firms.

Table 3.9
Regression Coefficients Testing the Effects of Ownership
Structure on the Use of the Outcome-Based Performance Measure

Dependent Variable Ownership Structure Outcome-Based Performance Measure .29** (see Table 3.3)
*p.05, **p.01

Hypothesis 4

Hypothesis 4 states that the use of outcome-based

performance measures will be significantly more important in owner-controlled firms, while behavior-based measures will be more important in management-controlled firms. The results of the regressions testing Hypothesis 4 are reported in Table 3.8. The regression results show that ownership structure is related to the relative importance of outcomebased performance measures compared to behavior-based measures (b=.42, p:.01). This suggests that outcome-based performance criteria are more important in owner-controlled firms, and behavior-based criteria are more important in management-controlled firms.






58


Hypothesis 5

Hypothesis 5 states that the use of behavior-based

measures will be greater in firms where CEO tenure is high versus firms where CEO tenure is low. The regression coefficients between CEO tenure and the three behavior-based measures are shown in Table 3.10. CEO tenure is not related to the use of the behavior-based measures. CEO tenure is not significantly related to managerial role performance, CEO leadership behavior or citizenship behavior.

Table 3.10
Regression Coefficients Testing the Effects of CEO
Tenure on the Use of Behavior-Based Measures Variables CEO Tenure Managerial Role Performance .02 (see Table 3.4)
CEO Leadership Behavior -.08 (see Table 3.5)
Citizenship Behavior .03 (see Table 3.6)
*p.05, **p.01

Hypothesis 6

Hypothesis 6 states that the use of outcome-based performance measures will be greater in firms where CEO tenure is low versus firms where CEO tenure is high. The regression coefficients between CEO tenure and the outcomebased measure are shown in Table 3.11. Hypothesis 6 is not supported in that low CEO tenure is not related to the use of outcome-based performance criteria in CEO evaluation. Hypothesis 7

Hypothesis 7 states that the use of behavior-based

measures will be greater in firms where the CEO is chairman






59


Table 3.11
Regression Coefficients Testing the Effects of CEO Tenure on the Use of the Outcome-Based Performance Measure Variable CEO Tenure Outcome-Based Performance Measure -.04 (see Table 3.3)
*p.05, **p.01

of the board versus firms in which the CEO is not chairman. The regression coefficients between CEO duality and the use of behavior-based measures is shown in Table 3.12.

Table 3.12
Regression Coefficients Testing the Effects of CEO
Duality on the Use of Behavior-Based Measures Variable CEO Duality Managerial Role Performance .07 (see Table 3.4)
CEO Leadership Behavior .25** (see Table 3.5)
Citizenship Behavior .06 (see Table 3.6)
*p .05, **p .01

CEO duality is positively related to the use of the behavior-based measure, CEO leadership behavior (b=.25; p .01). That is, when CEOs are chairman of the board, the greater the use of the measure, CEO leadership behavior in CEO evaluation. CEO duality is not significantly related to managerial role performance or citizenship behavior. Hypothesis 8

Hypothesis 8 states that the use of outcome-based

performance measures will be greater in firms where the CEO is not chairman of the board versus firms in which the CEO is chairman. The regression coefficients between CEO duality and the outcome-based performance measure are






60


displayed in Table 3.13. CEO duality is not significantly related to the greater use of outcome-based performance measures. That is, when CEOs are not chairman of the board, there is not a greater incidence of outcome-based performance criteria in CEO evaluation.

Table 3.13
Regression Coefficients Testing the Effects of CEO Duality on the Use of the Outcome-Based Performance Measure Variable CEO Duality Outcome-Based Performance Measure -.06 (see Table 3.3)
*p .05, **p.01

Hypothesis 9

Hypothesis 9, states that the use of behavior-based measures will be greater in firms with high CEO board meeting influence versus firms with low CEO board meeting influence. The regression coefficients between CEO board meeting influence and the behavior-based measures are displayed in Table 3.14. CEO board meeting influence is not significantly related to managerial role performance, CEO leadership behavior or citizenship behavior. That is, CEOs with high board meeting influence are not evaluated to a greater extent by behavior-based measures. Hypothesis 10

Hypothesis 10 states that the use of outcome-based

performance measures will be greater in firms with low CEO board meeting influence versus firms with high CEO board meeting influence. The regression coefficients between CEO board meeting influence and the outcome-based performance






61


Table 3.14
Regression Coefficients Testing the Effects of CEO Board Meeting Influence on the Use of the Behavior-Based Measures Variables CEO Board Meeting Influence Managerial Role Performance .07 (see Table 3.4)
CEO Leadership Behavior .07 (see Table 3.5)
Citizenship Behavior -.09 (see Table 3.6)
*p.05, **p.01

measures are displayed in Table 3.15. CEO board meeting influence is not significantly related to the use of outcome-based performance measures. That is, boards of directors in firms where the CEO has little influence over board meetings, are not more likely to use outcome-based performance criteria in the CEO evaluation.

Table 3.15
Regression Coefficients testing the Effects of CEO Board Meeting Influence and the Outcome-Based Performance Measure Variable CEO Board Meeting Influence Outcome-Based Performance Measure .02 (see Table 3.3)
*p .05, **p .01

Hypothesis 11

Hypothesis 11, states that the larger the proportion of inside board members the greater the use of the behaviorbased measures. The regression coefficients between the proportion of insiders and the behavior-based measures are displayed in Table 3.16. The proportion of insiders measure is significantly related to CEO leadership behavior (b=-.12,

*p .05). That is, the smaller the proportion of insiders, the greater the use of this indicator. Managerial role






62


performance and citizenship behavior are not significantly related to the measure, proportion of insiders.

Table 3.16
Regression Coefficients Testing the Effects of the
Proportion of Insiders on the Use of the Behavior-Based Measures

Variables Proportion of Insiders Managerial Role Performance -.01 (see Table 3.4)
CEO Leadership Behavior -.12* (see Table 3.5)
Citizenship Behavior .05 (see Table 3.6)
*p5.05, **p.01

Hypothesis 12

Hypothesis 12 states that the larger the proportion of board members who are presently or formerly CEOs of other firms, the greater use of behavior-based measures. The regression coefficients between the proportion of outsiders who are past or present CEOs and the behavior-based measures are displayed in Table 3.17. The proportion of outside directors who are past or present CEOs of others firms is significantly related to CEO leadership behavior (b=.15, p .05). That is, the higher the proportion of board members who are presently or formerly CEOs of other firms, the greater use of this behavior-based measure. Neither managerial role performance or citizenship behavior are significantly related to larger proportions of board members who are presently or formerly CEOs of other firms.






63


Table 3.17
Regression Coefficients Testing the Effects of the
Proportion of Outsiders Who Are Past/Present CEOs on the Use
of the Behavior-Based Measures

Variables %Outsiders Past/Present CEOs Managerial Role Performance .02 (see Table 3.4)
CEO Leadership Behavior .15* (see Table 3.5)
Citizenship Behavior -.04 (see Table 3.6)
*p .05, **p .01

Hypothesis 13

Hypothesis 13 states that board pay will be positively

related to the use of the behavior-based measures. The

regression coefficients between board pay and the behaviorbased measures are displayed in Table 3.18.

Table 3.18
Regression Coefficients Testing the Effects of Board Pay on
the Use of the Behavior-Based Measures

Variables Board Pay Managerial Role Performance -.35** (see Table 3.4)
CEO Leadership Behavior -.32** (see Table 3.5)
Citizenship Behavior -.30** (see Table 3.6)
*p .05, **p .01

Board pay is negatively related to the use of the

behavior-based measures. That is, board members with lower

pay are more likely to evaluate the CEO with the behaviorbased measures; managerial role performance (b=-.35, p:.01),

CEO leadership behavior (b=-.32, p5.01), and citizenship

behavior(b=-.30, p!.01).






64


Summary

The results in this chapter show evidence that there is generally, a greater use of behavior-based measures when the CEO is influential. In management-controlled firms the use of behavior-based measures are dominant in the CEO evaluation. Specifically, isolating factors representing the three dimensions (managerial role performance, CEO leadership behavior, and citizenship behavior), there are relationships when the CEO holds the chairman position. In owner-controlled firms, the use of outcome-based performance indicators is used to a greater extent in CEO evaluation.

Board pay seems to be an important predictor of the

type of criteria used in CEO evaluation. When board members receive low compensation for their services, the greater the use of all measures representing behavior-based measures (managerial role performance, CEO leadership behavior, and citizenship behavior) in the CEO evaluation. There is also evidence that in owner-controlled firms, when board members receive low compensation for their services the greater the use of outcome-based performance criteria. These results, along with other findings will be discussed in the next chapter.















CHAPTER 4
DISCUSSION

The nature of CEO compensation and how boards determine CEO pay packages has received increased attention over the last decade. The questions surround the seemingly excessive salaries of CEO's at their respective firms. The primary interest of this study is to determine how boards of directors justify CEO compensation decisions by examining the performance appraisal of the CEO by the board. This study assesses the nature of the criteria used in the board of director' evaluation of CEO performance, specifically, the use of outcome-based and behavior-based measures in CEO evaluation.

It is clear that the use of behavior and outcome-based measures are different in management-controlled versus owner-controlled firms. As speculated in Chapter One, based on the theory of managerial capitalism and agency theory, the nature of the criteria revolves around an issue of control over the board of directors or the firm. The CEO is afforded considerable discretion over the evaluation process when they are in control.

For example, the results show that there is a greater use of outcome-based measures (essentially measures of firm financial performance) in owner-controlled firms than in

65






66


management-controlled firms. They are also more frequently used in large owner-controlled firms and when there is a larger proportion of outside directors who are past or present CEOs of other firms. There are some instances when behavior-based measures are used in owner-controlled firms. One instance is when there is a larger proportion of outside directors who are past or present CEOs of other firms. The other is when the members of the board are less well paid.

In management-controlled firms, however, the tendency is to use behavior-based measures such as leadership, managerial role performance and CEO citizenship behavior in CEO evaluation. A major exception, for example, when outcome-based measures are rated as more important than behavior-based measures in management-controlled firms, is when firm financial performance is high.

There also are some facets of corporate governance mechanisms that are related to the use of behavior-based criteria for CEO evaluation. This occurs when board members are less well paid, when there is a larger proportion of outside directors, when the CEO is chairman of the board, and when more members of the board are currently or have been CEOs of other firms.

Some other conventional measures of CEO power were not related to the use of either behavior-based or outcomebased CEO evaluation criteria. Specifically, CEO tenure and CEO influence over board meetings or meeting agenda were found to have no effects.






67


Theoretical Implications

The agency theory model is based on the premise that, in the absence of an optimal sharing of risk between the agent and the principal, the principal can design a contract to control the agent by aligning the incentives of both parties or by indirectly or directly monitoring the agent's actions or behavior (Jensen & Meckling, 1976). Monitoring and incentive alignment are necessary because both the principal and agent, being rational actors, will seek to maximize their own utility. This can be a problem because of information asymmetry, moral hazard and adverse selection issues, which put the manager in a position to act opportunistically and reduce his or her risk at the expense of the equity holder. Thus, agency theory assumes that the CEO will be evaluated on criteria based on the objectives of the equity holders and that the relationship of the CEO and owner will be based on a contract that provides the appropriate level of monitoring and incentive alignment (Jensen & Meckling, 1976).

These criteria, given the difficulties of creating

optimal risk sharing contracts, would most likely be in the form of outcome-based measures that tie compensation to the performance of the firm and are easily observable. The use of outcome-based measures in CEO evaluation reduces the conflict of interests that exists between the owner and the CEO (Ouchi, 1979). On the other hand, behavior-based measures do not align the pay of the CEO with firm






68


performance. According to agency theorists, these criteria would only be optimal in situations in which the behavior of the CEO is easily monitored and is known to lead to outcomes desired by the equity holders (Demski & Feltham, 1978).

While agency theory is built on the idea that contracts can be designed to minimize managerial opportunism, managerial capitalism theory posits that because the equity holdings of large firms, particularly, are widely dispersed, stockholders are not able to exert enough control over managers, putting managers in a position to control the firm (Berle & Means, 1932). Therefore, when managers control the firm, they will be able to design compensation contracts that minimize their compensation risk and maximize their self-interest (Harris & Raviv, 1979).

There is ample evidence to support this idea. For example, managers in management controlled firms may (1) choose accounting methods that state results more favorable to them, (2) make investment decisions that are less optimal for owners but which minimize their downside risk, (3) undertake acquisitions and mergers that transfer higher agency costs to owners, (4) use internal political strategies to block control mechanisms intended to check their opportunism, and (5) use organizational resources to block the disciplining mechanism of the market (Tosi, et al., 1999). However, this evidence is plagued by an important limitation. It treats the firm as a "black box" in that it is based on archival sources of financial and






69


demographic data from which behavioral and process inferences are drawn. As Jensen and Murphy (1990) suggested, it does not actually focus on the internal processes and political forces that are at play in the monitoring process.

The research reported in this study is part of a more recent stream of research that seeks to understand the political forces at play in the implementation of the agency contract (Tosi & Gomez-Mejia, 1989; 1994; Tosi, Katz & Gomez-Mejia, 1997; Antle & Smith, 1986). The general conclusion of this research is that in management-controlled firms, the CEO is an important actor in the actual process of determining his or her compensation and other actors are less influential, while the opposite is the case in ownercontrolled firms.

This study extends that line of research by showing how CEOs can minimize compensation risk by managing the nature of the criteria used in evaluating their performance. Effective contracts, from the principal's point of view, can use behavior-based criteria or outcome-based criteria. However, in both instances it is necessary that the behavior and the outcomes be clearly and precisely known, defined and observable.

This can be a problem for at least two reasons. First, the CEO's task includes a wide variety of components, ranging from making choices that would be reflected in the firm's financial performance to taking a leadership role in influencing and motivating others in the firm. Second, the






70


precise nature of the behavior-outcome relationships of these different tasks, individually and in combination, is difficult to exactly specify. Given that these relationships are not clear, it is likely that the use of outcome-based measures transfer more performance risk to a CEO than behavior-based criteria.

The complexity of the managerial job and the lack of

precise relationships between behaviors and outcomes leaves room for discretion and interpretation, particularly with respect to whether behavior-based or outcome-based criteria will be used in the evaluation process. The way that this discretion and interpretation of specific CEO evaluation criteria is shown in the results discussed earlier. The CEO evaluation in the owner-controlled firm follows the prediction of agency theory in that owners and managers can design a contract that tends to align the interests of both. The evaluation of CEOs in owner-controlled firms is more likely to be based on outcome-based criteria; those derived from indicators of the firm's economic performance.

There are different results for the managementcontrolled firms. In these firms, it is already known that the CEO is the dominant actor in the process of setting his or her pay (Tosi & Gomez-Mejia, 1989). Here it is shown that one manifestation of this dominance is the use of less risky behavior-based criteria. The use of these behavior criteria in management controlled firms is also more likely as the proportion of outsider directors with experience as






71


CEOs increase. Since outside board members are not able to observe the CEO in action, it is more difficult for them to estimate the role of CEO behavior relative to firm performance. They could, then, simply base their evaluation on their own biases and experiences in the top managerial position in a firm. This result is consistent with the social comparison arguments advanced by O'Reilly and his colleagues (O'Reilly, Main & Crystal, 1988).

There are some cases, however, in which outcome-based measures are more likely in management-controlled firms. When firm performance is high, the manager can minimize risk by creating a compensation contract based on outcome-based criteria. Outcome-based criteria tie compensation to firm performance. CEOs in high performing management-controlled firms appear to prefer outcome-based criteria in that these would generate a higher compensation package. In this instance, the CEO of the high performing managementcontrolled firm uses more traditional economically justified measures of firm performance and avoids potential criticism for the use of more subjective behavior based indicators.

Research shows that the level of monitoring is higher in owner-controlled firms than in management-controlled firms, and compensation tends to be in the form of incentive based and more closely tied to maximizing shareholder value (Tosi & Gomez-Mejia, 1989). The results discussed above support this research in that governance activities in management-controlled firms are weaker than in owner-






72


controlled firms as reflected by the type of criteria used in the CEO evaluation. In owner-controlled firms the use of outcome-based measures is the dominant criteria used in the CEO evaluation. Outcome-based measures tie compensation to firm performance and align the interests of the CEO and shareholder.

Overall, the results show that the type of criteria

used in the CEO evaluation are significantly different for management-controlled versus owner-controlled firms. It also provides some evidence of how these criteria are related to certain elements of the governance process.

The results provide some insight into some of the

mechanisms of corporate governance that affect the choice of CEO criteria. For example, powerful CEOs are allowed considerable discretion in the selection of boards of directors and operation of the board. This might lead them to select board members who are prior or present CEOs of other firms. The results here show that this type of board member is more likely to emphasize the use of behavior-based measures. They would be more likely to share common interests with the CEO they are evaluating and this may bias the evaluation process.

Powerful CEOs can also influence the operation of the

board through the control of board pay. While board pay has been shown to be positively related to CEO pay (Boyd, 1994; O'Reilly, Main & Crystal, 1988), in this study it was found to have a negative relationship to the use of behavioral-






73


based measures, i.e., the lower the board pay, the greater the use of behavioral-based CEO criteria. One explanation for this could be that these lower paid boards are less vigilant in their evaluations of CEO performance. Board members may not feel committed to representing the owners best interest (i.e. monitoring the CEO) when board membership is not ostensibly valued, as exemplified by the remuneration of board members. There is a substantial amount of evidence to support the position that individuals who feel inequitably paid, are less motivated to complete a task (i.e. Pritchard, Dunnette & Gorgenson, 1972). This issue needs further exploration.

A powerful CEO can also control the composition of the board. In this study, the proportion of insiders on the board of directors is hypothesized to have a positive relationship with the behavior-based measures. This hypothesis is not supported in that the proportion of insiders' measure is negatively related to the use of the behavior-based measure, CEO leadership behavior. Other research shows a positive relationship between the proportion of insiders and CEO compensation (Sridharan, 1996; Williamson, 1963).

These findings imply a positive relationship with the proportion of outsiders and behavior-based measures. That is, the greater the proportion of outsiders on a board, the greater the use of behavior-based measures. So, while these results are inconsistent with the research on the behavior






74


of insiders, they are nonetheless consistent with prior work on the role of the outsider (Westphal & Zajac, 1995; O'Reilly, Main & Crystal, 1988). Outside members are often nominated by a CEO with whom they share some affiliation outside the firm. Therefore, the decisions these board members make may be in the best interests of their respective CEO, rather than the shareholder. Summary

This study draws on two theories to make suppositions about the CEO performance appraisal. These theories have been previously applied to the study of activities within the firm, and specifically the study of CEO compensation. The findings of this study extend the applications of managerial capitalism and agency theory. Additionally, the results are consistent with these theories prior findings: 1) Managers will behave opportunistically when they control the firm, and 2)the structure of the board of directors prohibits the autonomous role of the board.

Limitations

While the results of this research are consistent with theory and previous research, there are some necessary caveats which should be stated. First, this study is exploratory in being the first empirical examination of outcome-based and behavior-based criteria used by the board in the CEO evaluation process. Secondly, the use of a single source, i.e., the chief compensation officer in the






75


firms, to collect the data limits its interpretation, as common method variance could be a problem.

Third, other factors should be investigated for their

effects on CEO evaluation criteria. The dependent variables used are not an exhaustive list of the measures that could have been collected. The outcome-based measures used were traditional and conventional indicators used in previous studies. The results here were also consistent with results that have used similar outcome-based measures as a dependent variable. However, the behavior-based measures were not drawn from prior empirical work, but chosen based on the implications of the theory. Since some of the measures had little predictive power with the independent variables this suggests the need for more work on these measures. Additionally, there are other individual differences that might be more important in the CEO evaluation than the ones that were assessed in this study (i.e. personality variables).

Conclusion

This dissertation explores the performance evaluation process of the CEO. It contributes to this area in that it illuminates the types of criteria used in CEO evaluation. It also adds to the understanding of the board of director's role in the evaluation process.

Boards of directors have come under increased scrutiny by the public and media due to the excessive salaries awarded CEOs in many of America's leading corporations. The






76


questions have been, "why so much and how does the board make these decisions?" This study answers how the board makes these decisions by understanding the type of measures board members use in the evaluation of CEO performance. It is clear that boards of directors use different criteria across firms in the CEO evaluation.

The board must justify compensation decisions to not

only the public (namely the SEC), but also to shareholders. The CEO performance evaluation is part of the justification used in the derivation of CEO compensation packages. Boards must decide whether to use outcome or behavior-based criteria in the CEO evaluation. The logical criteria to use are those that align the interests of the CEO and owner. These outcome-based measures are easily verifiable and justifiable by the board. Using behavior-based measures are problematic in that they are abstract and not easily verifiable. The dominant use of these measures in firms where the CEO is influential suggests that board members are not fulfilling their role as "stewards" of the firm.

Board of directors cannot fulfill their role when there is a lack of separation of their role and the CEO. It is especially problematic to separate the role of the board and CEO when the CEO is often the chairman of the board. Powerful CEOs can dictate the type of criteria used in the evaluation of their performance. Behavior-based measures are preferred over outcome-based as they minimize the compensation risk to the CEO.






77


Separation of the role of the board from the CEO is crucial for a board fulfilling its role as an effective governance device for its shareholders. However, with current board membership of many firms consisting of present or former CEOs of other firms and also board members who are nominated and selected by the CEO, it is uncertain whether a board can ever be truly autonomous.

Moving beyond board membership issues, there is also a problem with the decision making process of the board. In many of the firms studied for this dissertation, the CEO had significant influence over the agenda and running of board meetings, as well as influence in important board decisions. In order for the board to fulfill their role as guardians of the shareholders interests, they need to truly govern themselves without the intervention of the CEO. Therefore, boards of directors should minimize involvement of the CEO in board decisions, and limit the CEO role to one of purely informational.

Future studies should examine the role of the board

more closely, in that we do not know whether board members feel that their role is being undermined by the CEO. Understanding the motivation of board members in the undertaking of their duties can also reveal their preferences. An examination of board compensation could provide insights into whether board pay plays a role in board member's value of their job. Further, understanding the types of decisions board members make when their






78


compensation consists of salary versus stock options or both could contribute to an understanding of the board decisions. In conclusion, more empirical work is needed in order to understand why the gap between the board and CEO seems to be so small.
















REFERENCES

Ackoff, R.L. (1999). Transformational leadership. Strategy
and Leadership, 27, 20-25.

Aigner, D. (1992). Fixing executive compensation. Journal
of Commerce and Commercial, 394, 6A.

Amihud, Y., and Lev, B. (1981). Risk reduction as
managerial motive for conglomerate mergers. The Bell
Journal of Economics, 12, 605-617.

Antle, R., and Smith, A. (1986). An empirical investigation
of the relative performance evaluation of corporate
executives. Journal of Accounting Research, 24, 1-39.

Avolio, B.J. and Bass, B.M. (1988). Transformational
leadership, charisma, and beyond. Pp. 29-49 in J.G.
Hunt, B.R. Baliga, H.P. Dachler and C.A. Schriesheim
(Eds.), Emerging leadership vistas. Lexington, MA:
Lexington Books.

Baimon, S. (1990). Agency research in managerial
accounting: A second look. Accounting, Organizations
and Society, 15, 341-371.

Baker, G.P., Jensen, M.C., and Murphy (1988). Compensation
and incentives: Practice vs. theory. Journal of
Finance, 43, 593-616.

Bass, B.M. (1985). Leadership and performance beyond
expectations. New York: Free Press.

Bass, B.M.(1990). Bass & Stoqdill's handbook of leadership:
Theory, research and managerial applications (3rd ed.).
New York: Free Press.

Bass, B.M. and Avolio, B.J. (1990). Developing
transformational leadership: 1992 and beyond. Journal
of European Industrial Training, 14, 21-28.

Bateman, T.S., and Organ, D.W. (1983). Job satisfaction and
the good soldier: The relationship between affect and employee "citizenship." Academy of Management Journal,
26, 587-595.
79






80





Baumol, W. (1967). Business, behavior, value and growth.
New York: Harcourt, Brace and World.

Baysinger, B., and Hoskisson, R.E. (1990). The composition
of the board of directors and strategic control:
Effects of corporate strategy. Academy of Management
Review, 15, 72-87.

Berle, A.A. and Means, G.C. (1932). The modern corporation
and private property. New York: MacMillan.

Boyd, B.K. (1994). Board control and CEO compensation.
Strategic Management Journal, 15, 335-344.

Bushman, R.M., Indjejikian, R.J., and Smith, A. (1996). CEO
compensation: The role of individual performance
evaluation. The Journal of Accounting and Economics,
21, 161-93.

Callen, J.H., Jr., and Jett, C.C. (1993, April). Critical
skills and the CEO. Chief Executive, pp. 34-38.

Cardy, R.L., and Dobbins, G.H. (1986). Affect and appraisal
accuracy: liking as an integral dimension in
evaluating performance. Journal of Applied Psychology,
71, 672-678.

Chitayat, G. (1985). Working relationships between the
chairman of the boards of directors and the CEO.
Management International Review, 25, 65-70.

Crystal, G.S. (1991). Why CEO compensation is so high.
California Management Review, 34, 9-29.

Demski, J., and Feltham, G. (1978). Economic incentives in
budgetary control systems. Accounting Review, 53, 336359.

DeNisi, A.S., and Stevens, G.E. (1981). Profiles of
performance, performance evaluations, and personnel
decisions. Academy of Management Journal, 24, 592-602.

Dobbins, G.H., and Russell, J.M. (1986). The biasing
effects of subordinate likeableness on leaders'
responses to poor performers: A laboratory and field
study. Personnel Psychology, 39, 759-77.

Ehrbar, A. (1999). Using EVA to measure performance and
assess strategy. Strategy and Leadership, 27, 20-24.






81




Eisenhardt, K. (1989). Agency theory: An assessment and
review. Academy of Management Review, 14, 57-74.

Ettorre, B. (1992). Evolution inside the boardrooms.
Management Review, 81, 10-16.

Fama, E.F. (1980). Agency problems and the theory of the
firm. Journal of Political Economy, 88, 288-307.

Fama, E.F., and Jensen, M.L. (1983a). Agency problems and
residual claims. Journal of Law and Economics, 26,
327-349.

Fama, E.F., and Jensen, M.L. (1983b). Separation of
ownership and control. Journal of Law and Economics,
26, 301-325.

Feldman-Summers, S., and Kiesler, S.B. (1974). Those who
are number two try harder: The effect of sex on the
attribution of causality. Journal of Personality and
Social Psychology, 30, 846-855.

Festinger, L. (1954). A theory of social comparison
processes. Human Relations, 7, 117-140.

Fiedler, F.E. (1967). A theory of leadership effectiveness.
New York: McGraw-Hill.

Fierman, J. (1990, March). The people who set the CEO's
pay. Fortune, pp. 58-62.

Finkelstein, S., and Hambrick, D.C. (1989). Chief executive
compensation: A study of the intersection of markets
and political processes. Strategic Management Journal,
10, 121-134.

Fosberg, R.H. (1999). Leadership structure and CEO
compensation. American Business Review, 17, 50-56.

Fredrickson, J.C., Hambrick, D.C., and Baumrin, S. (1988).
A model of CEO dismissal. Academy of Management
Review, 13, 255-270.

Gaver, J.J., and Gaver, K.M. (1998). The relation between
nonrecurring accounting transactions and CEO cash
compensation. Accounting Review, 73, 235-253.

Gibbons, R., and Murphy, K.M. (1990). Relative performance
evaluations for chief executive officers. Industrial
and Labor Relations Review, 43 (special issue), 30s51s.






82






Goldstein, M.L. (1985, January). Grading the CEO. Industry
Week, 224, 48-49.

Gomez-Mejia, L.R., Tosi, H.L., and Hinkin, T. (1987).
Managerial control, performance, and executive
compensation. Academy of Management Journal, 30, 5170.

Hall, B.J., and Liebman, J.B. (1998). Are CEOs really paid
like bureaucrats? Quarterly Journal of Economics, 113,
653-691.

Hallock, K.F. (1998). Layoffs, top executive pay, and firm
performance. American Economic Review, 88, 711-723.

Harris, M., and Raviv, A. (1979). Some results on incentive
contracts with imperfect information. Journal of
Economic Theory, 20, 231-259.

Hambrick, D.C., and Finkelstein, S. (1995). The effects of
ownership structure on conditions at the top: The case
of CEO pay raises. Strategic Management Journal, 16,
175-193.

Hill, C.W.L., and Phan, P. (1991). CEO tenure as a
determinant of CEO pay. Academy of Management Journal,
34, 707-807.

Hiltzik, M.A. (1996, May 10). Taking stock of CEO pay. Los
Angeles Times, p. Al.

Hoskisson, R.E., Johnson, R.E., and Moesel, D.D. (1994).
Corporate divestiture intensity in restructuring firms:
Effects of governance, strategy, and performance.
Academy of Management Journal, 37, 1207-1252.

House, R.J. (1971). A path goal theory of leader
effectiveness. Administrative Science Quarterly, 16,
321-328.

Hunt, H.G., III, and Hogler, R.L. (1990). Agency theory as
ideology: A comparative analysis based on critical
legal theory and radical accounting. Accounting,
Organizations and Society, 15, 437-455.

Ittner, C.D., Larcker, D.F., and Rajan, M.V. (1997). The
choice of performance measures in annual bonus contracts. The Accounting Review, 72, 231-255.






83






Janakiraman, S.N., Lambert, R.A., and Larcker, D.F. (1992).
An empirical investigation of the relative performance evaluation hypothesis. Journal of Accounting Research,
30, 53-70.

Jensen, M., and Meckling, W.H. (1976). Theory of the firm:
Managerial behavior, agency costs and ownership
structure. Journal of Financial Economics, 3, 305-360.

Kaplan, R.S., and Norton, D.P. (1992). The balanced
scorecard Measures that drive performance. Harvard
Business Review, 70, 71-80.

Kerr, J.L., and Kren, L. (1992). Effect of relative
decision monitoring on chief executive officer
compensation. Academy of Management Journal, 35, 370397.

Kirchner, W.K., and Reisberg, D.J. (1962). Differences
between better and less effective supervisors in
appraisal of subordinates. Personnel Psychologq, 15,
295-302.

Kristie, J. (1998). Is there no decency? Directors and
Boards, 23, 6.

Landy, F.J., and Farr, J.L. (1983). The measurement of work
performance. New York: Academic Press.

Larner, R. (1966). Ownership and control in the 200 largest
nonfinancial corporations, 1929 and 1963. The American
Economic Review, 56, 778-787.

Lau, A.W., and Pavett, C.M. (1980). The nature of
managerial work: A comparison of public and private sector managers. Group and Organization Studies, 5,
453-466.

Lazear, E., and Rosen, S. (1981). Rank-order tournaments as
optimum labor contracts. Journal of Political Economy,
89, 841-864.

Lear, R.W. (1998, September). Rethinking the compensation
committee. Chief Executive, p. 16.

Leonard, B. (1994, April). CEO compensation packages tied
to performance. HRMagazine, pp. 51-53.






84





Li, J. (1994). Ownership structure and board composition:
A multi-country test of agency theory predictions.
Managerial and Decision Economics, 15, 359-368.

Logue, D.E., and Steward, J.K. (1997). Handbook of modern
finance. Boston: WG&L/RIA Group.

Lorsch, J.W., and MacIver, E. (1989). Pawns or potentates:
The reality of America's corporate boards. Boston:
Harvard Business School Press.

MacKenzie, S.B., Podsakoff, P.M., and Fetter, R. (1991).
Organizational citizenship behavior and objective
productivity as determinants of managerial evaluations of salespersons' performance. Organizational Behavior
and Human Decision Processes, 50, 123-150.

Mallette, P., and Fowler, K.L. (1992). Effects of board
composition and stock ownership on the adoption of
poison pills. Academy of Management Journal, 35, 10101035.

Marino, S.F. (1998, April). Chief executives are underpaid.
Industry Week, 247, p. 22.

Marris, R. (1964). The economic theory of managerial
capitalism. New York: Free Press.

McEachern, W.A. (1975). Managerial control and performance.
New York: D.C. Health.

McNatt, R., and Light, L. (1998, March). Why the CEO may be
worth $100 million. Business Week, p. 6.

Mehran, H. (1995). Executive compensation structure,
ownership, and firm performance. Journal of Financial
Economics, 38, 163-184.

Meredith, D.R. (1992, Jan-Feb). Bold thinking. Chief
Executive, pp. 46-49.

Mintzberg, H. (1973). The nature of managerial work.
Englewood Cliffs, NJ: Prentice-Hall, Inc.

Morck, R., Shleifer, A., and Vishny, R.W. (1988).
Management ownership and market evaluation. Journal of
Financial Economics, 20, 293-315.






85




National Association of Corporate Directors (1994). NACD
Blue Ribbon commission report on performance evaluation
of CEOs, boards, and directors Washington, D.C.:
NACD.

Nadler, P.S. (1986, December). What makes a CEO effective.
Bankers Monthly, pp. 4-5.

Nett, R. (1993, April). What do Japanese CEOs really make?
Business Week, pp. 60-61.

O'Reilly, C.A., III, Main, B.C., and Crystal, G.S. (1988).
CEO compensation as tournament and social comparison:
A tale of two theories. Administrative Science
Quarterly, 33, 257-275.

Organ, D.W. (1997). Organzational citizenship behavior:
Its construct clean-up time. Human Performance, 10,
85-97.

Organ, D.W. (1988). Organizational citizenship behavior:
The good soldier syndrome. Lexington, MA: Lexington.

Ouchi, W. (1979). A conceptual framework for the design of
organizational control mechanisms. Management Science,
25, 833-848.

Palmer, J. (1973). The profit-performance effects of the
separation of ownership from control in large U.S.
corporations. Bell Journal of Economics and Management
Science, 4, 293-303.

Podsakoff, P.M., and MacKenzie, S.B. (1994). Organizational
citizenship behavior and sales unit effectiveness.
Journal of Marketing Research, 31, 351-363.

Podsakoff, P.M., MacKenzie, S.B., Moorman, R.H., and Fetter,
R. (1990). Transformational leader behaviors and their
effects on followers' trust in leader, satisfaction,
and organizational citizenship behaviors. Leadership
Quarterly, 1, 107-142.

Pritchard, R.D., Dunnette, M.D., and Gorgenson, D.O. (1972).
Effects of perception of equity and inequity on worker
performance and satisfaction. Journal of Applied
Psychology, 56, 75-94.

Rosen, S. (1990). Contracts and the market for executives.
Cambridge, MA: Working paper for the National Bureau
of Economic Research.






86





Ross, S. (1973). The economic theory of agency: The
principal's problem. American Economic Review, 63,
134-139.

Salomon, G.L., and Smith, E.D. (1979). Corporate control
and managerial misrepresentation of firm performance.
Bell Journal of Economics, 10, 319-328.

Salancik, G.R., and Pfeffer, J. (1980). Effects of
ownership and performance on executive tenure in U.S.
corporations. Academy of Management Journal, 23, 653664.

Schaefer, S. (1998). The dependence of pay-performance
sensitivity on the size of the firm. Review of
Economics and Statistics, 80, 436-443.

Seltzer, J., and Bass, B.M. (1990). Transformational
leadership: Beyond initiation and consideration.
Journal of Management, 16, 693-703.

Singh, H. and Harianto, F. (1989). Top management tenure,
corporate ownership structure, and the magnitude of
golden parachutes. Strategic Management Journal, 10,
143-156.

Smith, C.A., Organ, D.W., and Near, J.P. (1983).
Organizational citizenship behavior: Its nature and antecedents. Journal of Applied Psychology, 68, 653663.

Sridharan, U.V. (1996, February). CEO influence and
executive compensation. The Financial Review, pp. 5167.

Tobin, J.M. (1994). The squeeze on directors inside is
out. Business Lawyer, 49, 1701-1760.

Tosi, H.L., and Gomez-Mejia, L.R. (1989). The decoupling of
CEO Pay and Performance: An agency theory perspective.
Administrative Science Quarterly, 34, 169-189.

Tosi, H.L., and Gomez-Mejia, L.R. (1994). CEO compensation
monitoring and firm performance. Academy of Management
Journal, 37, 1002-1016.

Townley, P. (1993). Accountable but to whom? Across the
Board, 30, 11-12.






87




Truskie, S.D. (1995). CEO performance evaluation: How
America's leading companies review and evaluate the
performance of their chief executive officers.
Pittsburgh, PA: Management Science and Development.

Vance, S.C. (1990). Inside or outside directors: Is there
really a difference? Across the Board, 27, 15-17.

Verespej, M.A. (1994, September). Before it's too late.
Industry Week, p. 11.

Wade, J., O'Reilly, C.A., and Chandratat, I. (1990). Golden
parachutes: CEOs and the exercise of social influence.
Adminstrative Science Quarterly, 35, 587-603.

Waldman, D.A., Bass, B.M., and Einstein, W.O. (1987).
Leadership and outcomes of performance appraisal
processes. Journal of Organizational Psychology, 60,
177-186.

Wang, C. (1997). Incentives, CEO compensation, and
shareholder wealth in a dynamic agency model. Journal
of Economic Theory, 76, 72-105.

Weiner, B. (1971). Perceiving the causes of success and
failure. Morristown, NY: General Learning Press.

Westphal, J.D. (1998). Board games: How CEOs adapt to
increases in structural board independence from
management. Administrative Science Quarterly, 43, 511537.

Westphal, J.D., and Zajac, E.J. (1997). Defections from the
inner circle: Social exchange, reciprocity, and the diffusion of board independence in U.S. corporations.
Administrative Science Quarterly, 42, 161-183.

Westphal, J.D., and Zajac, E.J. (1995). Who shall govern?
CEO/board power, demographic similarity, and new
director selection. Administrative Science Quarterly,
40, 60-83.

Williamson, O.P. (1963). Managerial discretion and business
behavior. American Economic Review, 53, 1032-1057.

Yermack, D. (1998). Companies' modest claims about the
value of CEO stock option awards. Review of
Quantitative Finance and Accounting, 10, 207-226.






88





Zajac, E.J., and Westphal, J.D. (1996a). Director
reputation, CEO- board power, and the dynamics of board interlocks. Administrative Science Quarterly, 41, 507529.

Zajac, E.J., and Westphal, J.D. (1996b). Who shall succeed?
How CEO/board preferences and power affect the choice
of new CEOs. Academy of Management Journal, 39, 64-90.
















APPENDIX A
MAIL SURVEY MATERIALS



This appendix contains all material used for collection of data from cooperatives via mail.















































89






90



September 1997

Henry L. Tosi
Vice President, Executive Compensation Rosemary Tosi and Associates P.O. Box 69538
Gainesville, FL 32608

Dear Mr. Tosi,

In recent years, the popular press has devoted considerable attention to the compensation of Chief Executive Officers in America's largest firms. As you know, the controversy involves both the magnitude of CEO compensation as well as the bases for these pay packages. The Management Center at the University of Florida has been interested in these questions since 1987 and we have conducted a number of studies to examine them. As we look at our work and that of others, we conclude that there is a significant void in this literature. It seems that there has not been a systematic study of the criteria used by boards of directors to determine CEO pay.

We are conducting a study at the Management Center to address the question of CEO pay criteria. The enclosed questionnaire forms the basis of this research. We would like your help in this project, asking that you take a few minutes to complete the enclosed survey, then returning it to us. You were selected as a recipient of this questionnaire from among those members of the American Compensation Association who have indicated some significant level of involvement with top executive compensation, and hence one who should be a knowledgeable informant about CEO pay and how it is structured in your firm. We ask only that you respond to each question to the best of your knowledge, nothing more.

Please complete the questionnaire and return it in the enclosed envelope. The questionnaire should take no longer than 20 minutes. Examining it, you will note that it provides for anonymous responses. We have done this to ensure confidentiality for those firms and those executives willing to help us. We have also included a postcard that you may send to us at your convenience if you are interested in a summary of our results. The use of the separate postcard again, ensures your anonymity.



We hope you can help us in this research. We know that you have a busy schedule, but your participation is essential to this study. Further, this topic is very interesting and






91


important to all executive compensation professionals. If you have any questions, comments, or need any assistance in completing the questionnaire please contact Paula Silva, Project Director, at
(352) 392-3737.

Sincerely,




Henry L. Tosi Paula Silva McGriff Professor of Management Graduate Minority Fellow







92












1. How long have you been Chief Compensation Officer with your firm? years
2. Please rate your firm 'sfinancial performance relative to others in your industry
Top 20% O0 Second 20% O0 Third 20% O Fourth 20% O0 Lowest 20% O0
3. Please rate your firm 's sales relative to other firms in your industry
Top 20% O0 Second 20% O0 Third 20% ] Fourth 20% O Lowest 20% O0
4. Are your firm's main products durable goods Ol or nondurable goods O0

5. Please indicate your firm 's Standard Industrial Classification
El Agriculture, Forestry & Fishing ] Mining
O0 Construction Ol Transportation, Communications, Electric, Gas & Sanitary Services
O0 Wholesale Trade Ol Finance, Insurance & Real Estate
0] Services O0 Manufacturing
6. Approximately how many people are employed by your firm?
7. How long has the CEO held this position with your firm? years
8. Is there any single individual or institution outside the firm that owns 5 percent or more of the company stock?

Yes O No O0
8a. If you checked "yes" to question 8, please check which of the following owns 5 percent or more of the company stock.
(please check all that apply)
Insurance Company O Bank El Nonbank Trust O0 Mutual Fund El Endowment E0 Pension Fund Ol
Foundation O0 Individual El Other (please specify)
9. Is there any individual manager within the firm (e.g., the CEO) who owns 5 percent or more of the company stock?
Yes O No El




1. How many times does your Board ofDirectors meet per year? times each year
2. What is the average cash compensation for board membership per year?
2a. What is the estimated average value of other pay (stock options, benefits, etc...)
3. How many directors are also executives in the firm? How many directors are from outside the firm?
4. Is the current CEO also chairman of the board? Yes El No El

Ifyou checked "no" to question 4, is the Chairman presently or formerly a CEO of another firm? Yes El No [
Ifyou checked "no" to question 4, is the Chairman formerly a CEO of your firm? Yes El No [
5. How many of the outside directors are present or past CEOs of other firms?
6. How much influence does the CEO have in reviewing and approving the types of information received by board
members? (please circle the best response)

1=None 2 3 4=Moderate 5: 6 7=Very High Please respond to questions 7-9 by rating the influence of different members of the board at board meetings. Please use the following scale.

1:-None 2 3 4=Moderate 5 6 7=Very High

7. How much influence does each of the following have in setting the agenda at board meetings?
Chief Executive Officer Chairman of the Board







93






Board members, in general A coalition of one or two board members
8. How much influence does each of the following have in the running of board meetings?
Chief Executive Officer Chairman of the Board
Board members, in general A coalition of one or two board members
9. How much influence does each of the following have in important board decisions?
Chief Executive Officer Chairman of the Board
Board members, in general A coalition of one or two board members
10. Please rate the extent to which board members are kept up to date on company activities throughout the year
(please circle the best response)

1=Not At All 2 3 4=Usually 5 6 7=Alwaysi II. Does your board of directors have a separate compensation committee that determines the CEO compensation package?
Yes 0 No 0
Ila. If you answered "yes" to question 11, is the CEO a member of the compensation committee? Yes 0 No 0 12. Does your board employ an independent compensation consultant? Yes 0 No 0



Using the following scale please respond to the following question.

1l=None 2 3 4=Moderate 5 6 7=Very High

1. Indicate the level of involvement of each of the following in the CEO evaluation process.
Board of Directors Compensation committee
Outside participants or groups Outside consultant
Other, please specify and rate:
2. Please rate the extent to which performance evaluations of the CEO by these groups remain anonymous, so that individual
evaluations cannot be identified Please circle the best response (1-no anonymity 7-complete anonymity).

l=No anonymity 2 3 4=moderate 5 6 7=Very high

3. To the best of your knowledge, please rate the relative importance of each of the following quantitative factors in the CEO
performance evaluation. Please use the following scale.

l=Not At All Important 2 3 4=Moderate Importance 5 6 7=Extremely Important


ROA Increasing Price of the Stock
ROE Profitability
ROI Earnings Per Share
Sales Growth EVA/Economic Profit
Other (please specify and rate)




To the best of your knowledge please rate the relative importance of the following factors in the CEO performance evaluation. Please circle the best response using the following scale.

1=Not At All Important 2 3 4=Moderate Importance 5 6 7=Extremely Important

The extent to which the CEO:
Not at all Moderate Extremely
Important Important

I. Acts cheerful 1 2 3 4 5 6 7







94





2. Taps into the grapevine to keep abreast of what is going on in the company 1 2 3 4 5 6 7
3. Is available to "outsiders" (such as consumers, suppliers, the public) who want to go to "the
the person in charge" 1 2 3 4 5 6 7 4. Trains or helps others to perform their jobs better 1 2 3 4 5 6 7
5. "Keeps up" with developments in the company 1 2 3 4 5 6 7
6. Keeps up with market changes and trends that might have an impact on the organization 1 2 3 4 5 6 7
7. Expects themselves and others to conscientiously follow company regulations and procedures 1 2 3 4 5 6 7 8. Surveys the organization to identify ways to improve performance or cut costs 1 2 3 4 5 6 7 9. Exhibits punctuality in arriving at meetings 1 2 3 4 5 6 7 10. Integrates subordinates' goals (for example, career goals, work preferences)
with the organization's work requirements 1 2 3 4 5 6 7 11. Attends functions that are not required, but that help the company image 1 2 3 4 5 6 7 12. Willingly risks disapproval in order to express beliefs about what's best for the company 1 2 3 4 5 6 7 13. Keeps professional colleagues informed about the organization 1 2 3 4 5 6 7 14. Takes steps to try to prevent problems with other executive members, board members
and/or other employees in the company 1 2 3 4 5 6 7


1=Not At All Important 2 3 4=Moderate Importance 5 6 7=Extremely Important 15. Entertains clients on a regular basis 1 2 3 4 5 6 7 16. Cooperates well with others in the organization 1 2 3 4 5 6 7 17. Negotiates with groups outside the organization for necessary materials, support,
commitment, etc... 1 2 3 4 5 6 7 18. Attends to staffing requirements in different units, such as hiring, firing,
promoting, and recruiting 1 2 3 4 5 6 7 19. Provides guidance and direction to company employees 1 2 3 4 5 6 7 20. Makes innovative suggestions to improve the overall quality of the company 1 2 3 4 5 6 7 21. Keeps members of the organization informed of relevant information through meetings,
conversations, and written information 1 2 3 4 5 6 7 22. Develops new contacts by personally answering requests for information 1 2 3 4 5 6 7 23 Takes immediate action in response to a crisis or "fire drill" 1 2 3 4 5 6 7 24. Stays tuned to what is going on in competitor organizations 1 2 3 4 5 6 7 25. Focuses on the positive side of a situation rather than the negative side of it 1 2 3 4 5 6 7 To the best of your knowledge please rate the relative importance of thefollowingfactors in the CEO performance evaluation. Please circle the best response using the following scale.

I=Not At All Important 2 3 4=Moderate Importance 5 6 7=Extremely Important The extent to which the CEO:
Not at all Moderate Extremely
Important Important

26. Willingly gives of his/her time to help others 1 2 3 4 5 6 7 27. Passes along unofficial information 1 2 3 4 5 6 7 28. Joins boards, organizations, clubs, or does public service work that might provide useful,
work-related contracts 1 2 3 4 5 6 7 29. Passes on the desires and preferences of top management to others in the firm 1 2 3 4 5 6 7 30. Keeps executive members, consumers, or other important groups informed about the
organization's activities and capabilities 1 2 3 4 5 6 7




Full Text

PAGE 1

STYLE VS. SUBSTANCE: THE NATURE OF CEO EVALUATION CRITERIA IN MANAGEMENT CONTROLLED AND OWNER CONTROLLED FIRMS By PAULA SILVA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF THE DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2000

PAGE 2

This research is dedicated to my mother and father, Mary and Alberto Silva. I will always be grateful for their support and encouragement. Thank you for always believing in me without hesitation, and showing me that no dream is unattainable.

PAGE 3

ACKNOWLEDGEMENTS I am forever indebted to my advisor, Henry Tosi, whose guidance, direction, support, and inspiration were instrumental to this achievement. I am grateful to him for encouraging me in all my endeavors at the University of Florida, and helping me reach my fullest potential as an academic I would also like to thank the members of my dissertation committee. Rich Lutz, Heather Elms, Steve Motowidlo and Judy Scully, for their insight and words of encouragement during the dissertation process. I am also grateful to Virginia Maurer, chair of the department, for always being thoughtful and supportive of my efforts at the University of Florida. I am also thankful to have had supportive family members and good friends that made the Ph.D. road a little smoother iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGEMENTS iii ABSTRACT vi CHAPTERS 1 INTRODUCTION AND LITERATURE REVIEW 1 The Board of Directors 3 Theory and Hypotheses 8 Conclusion 26 2 METHODS 27 Data Collection Method and Sample 27 Measures 28 Independent Variables 2 8 Dependent Variables 32 Control Variables 43 Summary 44 3 RESULTS 46 Analysis of the Data 46 Regressions of Independent Variables by Dependent Variable 49 Report of the Tests of the Hypotheses 55 Summary 64 4 DISCUSSION 65 Theoretical Implications .... .67 Limitations 74 Conclusion 75 REFERENCES 7 9 iv

PAGE 5

APPENDICES A MAIL SURVEY MATERIALS 8 9 B TABLES OF FACTOR ANALYSIS FOR OUTCOME BASED MEASURE AND BEHAVIOR BASED MEASURES 97 BIOGRAPHICAL SKETCH 102 V

PAGE 6

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy STYLE VS. SUBSTANCE: THE NATURE OF CEO EVALUATION CRITERIA IN MANAGEMENT CONTROLLED AND OWNER CONTROLLED FIRMS By ^. : Paula Silva .> "-J'': May 2000 Chairman: Henry L. Tosi, Jr. Major Department: Management This dissertation examines the CEO performance evaluation process and makes predictions concerning the relationship between the type of criteria used in the evaluation and the ownership structure of the firm, CEO influence, and board membership. The results suggest that the criteria used by boards in the evaluation of CEO performance are different for management and ownercontrolled firms. CEOs who are allowed considerable discretion in a firm are evaluated by indicators that are subjective and reflective of CEO style, while CEOs with minimum influence are evaluated by financial performance measures vi

PAGE 7

CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW In 1992, the Securities and Exchange Commission (SEC) mandated that firms report in their proxy statements the salaries of their top paid executives and justify these compensation levels by measures of corporate performance (Tobin, 1994; Leonard, 1994) This ruling was not surprising given the amount of criticism in the media regarding excessive Chief Executive Officer (CEO) salaries. For example, it is estimated that CEO pay is 157 times higher than that of the average factory worker (Nett, 1993) The justifications for this disparity vary substantially. Some argue that CEOs are worth the compensation they receive (Marino, 1998; McNatt & Light, 1998), while others argue that CEOs are more than adequately paid (Kristie, 1998; Sridharan, 1996) This dissertation addresses this general issue by examining the nature of the criteria that boards of directors use in determining the compensation of chief executive officers and how these criteria are justified in large firms in the United States. While numerous studies examine performance appraisals of those at lower organizational levels (Kirchner & Reisberg, 1962; DeNisi & Stevens, 1981; Landy & Farr, 1983), only a few studies have

PAGE 8

2 investigated how boards of directors evaluate chief executive officers (Bushman, Injejikian & Smith, 1996; Antle Sc Smith, 1986; Gibbons & Murphy, 1990) Research on CEO evaluation is a problem because systematic data about CEO appraisals is difficult to get since the information is considered to be very confidential and boards of directors are reluctant to reveal specific information about such evaluations. However, this very paucity of research makes CEO performance appraisals a fruitful area for research since systematic examination of the CEO appraisal process may result in a better understanding of how CEO compensation packages are determined. What is known empirically about the bases for CEO compensation is derived from research that investigates the economic and social correlates of CEO compensation (Hallock, 1998; Hall Sc Liebman, 1998; Schaefer, 1998; Yermack, 1998; Gaver & Gaver, 1998; Westphal & Zajac, 1997; Wang, 1997; Zajac & Westphal, 1996b; Westphal & Zajac, 1995; Baker, Jensen & Murphy, 1988; Wade, O'Reilly & Chandratat, 1990) Specifically, the research on CEO performance evaluation has focused on two issues. First, researchers have investigated the utility of evaluating CEOs relative to the performance of CEOs in comparable industries (Antle & Smith, 1986; Janakiraman, Lambert & Larcker, 1992; Gibbons & Murphy, 1990) Second, a limited amount of research has shown nonf inancial or subjective, criteria to be important in the

PAGE 9

3 determination of bonuses for the CEO (Ittner, Larcker & Rajan, 1997; Bushman, Injejikian & Smith, 1996) Bushman, Injejikian and Smith (1996) suggest that CEOs with more power are able to influence the board to assess them by more subjective criteria that can be more prone to manipulation. However, this conclusion about the behavior, or actions, of boards of directors is inferred from the nature of the archival financial data used in the research rather than a more direct assessment of what boards themselves actually do This dissertation addresses the question of CEO evaluation by the board by seeking to understand the process more directly, using survey report data from an informed observer within the firm to determine correlates of the use of financial and nonfinancial criteria by boards of directors in the evaluation process. This chapter reviews the role and responsibility of the board of directors, examines performance criteria used in the evaluation, and concludes with a discussion of theories that address the nature of the control relationships between CEOs and boards and their implications for the CEO evaluation process. Chapter 2 sets forth the methodology employed in the research; the results are reported in Chapter 3 and the conclusion and implications are set out in Chapter 4. The Board of Directors Boards of directors are legally charged with "the power to hire, fire, and compensate the top-level decision

PAGE 10

4 managers and to ratify and monitor important decisions" (Fama and Jensen, 1983b: 311) The board of directors, in essence, protects the shareholders' interests by limiting opportunistic behavior by the CEO (Fama, 1980) This is achieved through control of the decision making of the CEO, and by ensuring that the CEO will not be solely responsible for significant wealth decisions of the firm (Fama and Jensen, 1983b) The board of directors consists of inside and outside directors. Inside directors are top managers who work for the firm. Outside directors have no other link to the organization except maybe in the form of equity holdings. It is suggested that outside directors who hold equity in a firm might be better at protecting shareholder interests than inside directors (Hoskisson, Johnson, and Moesel, 1994; Morck, Shleifer & Vishny, 1988) This is because outside directors, in general, are assumed to be "independent" and more likely to act in the interests of shareholders, since inside directors are, by definition, bound to management in some manner (compensation, family ties, etc.) (Vance, 1990). Research on this question has been inconsistent. It has been shown that as the percentage of outside directors increases, so does the proportion of equity-based CEO compensation that better aligns the interests of the CEO with the shareholder (Mehran, 1995) On the other hand, it has been demonstrated that CEO compensation is higher in firms consisting of boards with greater proportions of

PAGE 11

5 outsiders (Westphal & Zajac, 1995; O'Reilly, Main & Crystal, 1988). This is attributed to the outsiders' increased reliance on the CEO for firm performance information, which can be subject to manipulation (Baysinger & Hoskisson, 1990), and also because outsiders may be more sympathetic to the CEO's purposes. Outsiders, some argue, may be more sympathetic in serving management interests rather than shareholders (Zajac & Westphal, 1996a; Westphal & Zajac, 1995) Zajac Sc Westphal (1996a) show that outside board members are appointed to boards according to their prior experience in strong or weak boards. Outsiders who participate in structural changes to the board (i.e., decreasing ratio of outsiders to insiders) which favor powerful CEOs are subsequently nominated to boards at firms where there is low board control. Likewise, board members who have made changes to the board that exemplify passivity towards management are nominated to boards with high board control (Zajac & Westphal, 1996a). Boards that seek to increase their independence by making structural changes face CEOs determined to uphold their power. Westphal (1998) shows a relationship between a firm's increase in board independence and high level of ingratiating and persuasive tactics by the CEO that result in increases in CEO compensation. In firms where structural changes of the board increase board independence (i.e., increases in the number of outsiders, splitting the CEO/Chairman role) CEO interpersonal influence tactics

PAGE 12

6 increase, decreasing the power of the board (Westphal, 1998) This finding suggests that even independent boards cannot isolate themselves from an influential CEO and that structural changes to the board of directors (i.e., increasing number of outsiders) to presumably increase board independence is not an effective method in reducing CEO power over the board. The task of deriving the CEO compensation package is most often left up to the compensation committee of the board of directors (Tosi & Gomez-Mejia, 1989) Lear (1998) maintains that a strong, independent compensation committee is essential to controlling CEO compensation levels. The importance of the compensation committee in the determination of CEO pay has become increasingly important over time; in 1972, only 69% of boards of directors had compensation committees compared to 91% in 1992 (Townley, 1993) However, there is some evidence that membership on the compensation committee is not desired by directors (Crystal, 1991). This might explain the relative "passivity" the compensation committee shows in attending to compensation packages (Crystal, 1991). The compensation committee decides what performance criteria should be linked to CEO pay. If the purpose of criteria is to align the interest of the CEO with the owners of the firm, then pay should be linked to objective measures of a firm's financial performance, as these measures are easily quantifiable and presumably reflect positive or

PAGE 13

7 negative organizational performance from the perspective of equity holders. Among these economic criteria are shortterm and long-term performance goals using standard accounting measures (i.e., profit margins) and market measures (i.e., increasing price of the stock) (National Association of Corporate Directors, 1994 ; Verespej 1994; Goldstein 1985; Nadler, 1986; Callen & Jet, 1993) CEOs can also be evaluated by more subjective criteria, or nonfinancial measures, such as their leadership and management skills (i.e., planning, organizing, and decisionmaking) and behaviors, like establishing vision and direction, succession planning, customer satisfaction, and innovation (Ittner, Larcker & Raj an, 1997; Bushman, Indjejikian & Smith, 1996; Truskie, 1995; Kaplan & Norton, 1992) that contribute to the effective functioning of the organization. Truskie (1995) reports that leadership skills and behaviors are the common qualitative indicators assessed by the board of directors in CEO performance evaluations. Bushman, Indjejikian and Smith (1996) assert that financial measures are insufficient in capturing all the dimensions of CEO performance, and that nonfinancial measures can add important performance information beyond these financial measures. Further, the use of nonfinancial measures is especially important for firms where high information asymmetry exists between the CEO and equity holders and where there is substantial noise in the financial measures

PAGE 14

8 (Ittner, Larcker, & Rajan, 1997; Bushman, Injejikian & Smith, 1996) Theory and Hypotheses The primary research issue that this dissertation addresses is the understanding of how powerful CEOs can exercise considerable control over the type of criteria used in their performance evaluation that is derived by the board of directors. As stated in the previous section, boards of directors use a combination of nonfinancial measures in conjunction with financial performance measures in the evaluation of CEO performance. Nonfinancial measures are inherently more subjective than financial performance measures and subject to manipulation and bias (Bushman, Injejikian & Smith, 1996). Further, Tosi and Gomez-Mejia (1989) argue that CEO pay is more highly correlated with firm economic performance when CEOs are not influential in the pay process, suggesting that financial performance measures are more important when boards are stronger. Thus, while CEOs are evaluated by both financial measures of firm performance as well as nonfinancial measures by the board of directors (Truskie, 1995), a CEO's relative influence over the board of directors is likely to influence the types of criteria he/she is evaluated against. Powerful CEOs may be more likely to obtain evaluations comprised of nonfinancial measures, as they are easily manipulated and transfer less compensation risk to the CEO.

PAGE 15

9 Three theories that contribute to understanding these issues and which are useful for the development of specific hypotheses about the relationship between the CEO and board and the nature of criteria used in the CEO evaluation are agency theory, the theory of managerial capitalism, and attribution theory. Agency Theory In agency theory, the firm is defined as a set of explicit and implicit contracts that specify the responsibilities and reward structure of individuals within the organization (Jensen & Meckling, 1976; Ross, 1973). These contracts define the relationship between the owner (principal) and the agent (manager) (Fama and Jensen, 1983b) where the agent agrees to abide by the terms of the principal, and in return, is rewarded for such compliance. Agency theory is based on the premise that a divergence of interest exists between those who own the firm and those who manage the firm (Jensen & Meckling, 1976) and that conflicts arise between management (agents) and owners (principals) Most typically, research on agency theory refers to the relationship between the board of directors (owner) and the CEO (manager) (Jensen & Meckling, 1976) Both the agent (the CEO) and the principal (the owners) are assumed to be rational and self-serving in the interests they pursue (Hunt & Hogler, 1990; Baiman, 1990) Therefore, logically, the CEO's goal is to maximize personal wealth and

PAGE 16

10 will do so at a cost to owners, whose goal is to maximize shareholder wealth. The crux of the agency problem is in determining and enforcing the contract. These problems manifest themselves in "agency costs" to the owner that include monitoring the agent's behavior, and losses when the agent's behavior is not in alignment with the owner's interest (Fama & Jensen, 1983b) Agents can behave in a self-serving manner due to their relative proximity to organizational activities and the fact that agents possess information that could be difficult to get by the principal (Hunt & Hogler, 1990; Baiman, 1990) This information asymmetry between the principal and agent (Fama & Jensen, 1983a) maximizes chances for opportunistic behavior. The owner's goal is to reduce the opportunity for self-serving behavior by the agent (agency costs) To curb self-serving behavior by the agent, the principal utilizes two control mechanisms: monitoring and incentive alignment. Monitoring involves direct or indirect observation of the agent's activities by the owner (Jensen & Meckling, 1976) Monitoring can be achieved through the board of directors, who are the appointed "stewards" of the shareholders. The board of directors is responsible for protecting the interests of the shareholders through their governance of the compensation contract (Jensen & Meckling, 1976), curbing the opportunistic behavior of the agent, and in general, reducing the agency costs (Fama & Jensen,

PAGE 17

11 1983b) Monitoring the agent's (CEO) behavior includes designing contracts that focus on behavior-oriented criteria or outcome-based evaluation criteria that are related to improving the welfare of owners or increasing shareholder wealth. Behavior-based contracts provide for evaluating and rewarding the performance of the agent by their behaviors or actions. Outcome-based contracts reflect the evaluation and reward of the agent s performance by changes in firm performance measures (i.e., increases in the price of the stock) (Ouchi, 1979) Incentive alignment is a mechanism through which the interests of the managers are aligned with the owner through compensation packages that tie the agent's pay to firm performance (Eisenhardt, 1989). Tying pay to measurable results (those determined to be in the owner's best interests) increases the risk shared by the owner and manager. Therefore, when pay is tied to firm performance, agents are more likely to make decisions that are in the interests of the owner (i.e., increasing shareholder wealth) (Meredith, 1992) Firms that tie pay to firm performance are using outcome-based contracts that evaluate the agent's performance according to the firm's financial performance (Ouchi, 1979). These outcome-based contracts are the best method to reduce opportunistic behavior by the agent by transferring some risk to the manager and better aligning the interests of the agent with the owner (Jensen & Meckling, 1976) ,. v

PAGE 18

12 Monitoring and incentive alignment may be interchangeable control mechanisms. For example, in situations where the agent's behavior is not easily monitored, incentive alignment is expected to be the best control mechanism to reduce agency costs. When monitoring is costly or difficult, the contract is expected to be outcome-based to minimize self-serving behavior by the agent (Demski & Feltham, 1978) in that outcome-based contracts rely on incentive alignment to ensure the agent's behavior is aligned with shareholders. When the agent's behavior is easily monitored, a behavior-based contract is most efficient to minimize the risk to the agent (Demski & Feltham, 1978) Managerial Capitalism The basic assumption in agency theory is that the owner is able to structure controls in an efficient way that is in the best interests of equity holders. However, if the CEO has considerable influence over the board of directors, the nature of the relationship between the principal and agent changes and the agent, in essence, controls the firm. The theory of managerial capitalism is an approach based on the idea that the managers who are responsible for running the day-to-day activities of the firm have control of the firm when stock ownership is widely dispersed (Berle Sc Means, 1932) Therefore, owners are not able to oversee these activities and are not able to discipline these managers (Berle & Means, 1932) Firms in which ownership

PAGE 19

13 holdings are widely dispersed so that owners cannot effectively exert decision controls are called management controlled firms Firms in which there are equity holders with large enough equity holdings to exert decision controls are called owner-controlled firms (Berle & Means, 1932) Since, all other things being equal, managers prefer a contract that involves less monitoring and a reward system where they will bear minimal risk (Harris & Raviv, 1979) a manager in a management-controlled firm will design a contract that maximizes his/her self-interest through influence over the board. This often is reflected in strategies to grow the firm (a focus on increasing firm size) rather than making decisions that would maximize shareholder wealth (Berle & Means, 1932; Larner, 1966; Marris, 1964) The reason is that compensation risk is lower when linking pay to firm size, larger size accords more power and prestige to those managing the firm, and larger firms have a greater likelihood of survival (Baumol, 1967; Marris, 1964) There is substantial evidence supporting managerial capitalism (Tosi & Gomez-Mejia, 1989; Singh Sc Harianto, 1989; Tosi & Gomez-Mejia, 1994; Salancik & Pfeffer, 1980; Amihud Sc Lev, 1981; McEachern, 1975; Palmer, 1973) The evidence suggests that 1) there is less pay risk in management-controlled firms compared to owner-controlled firms (Baumol, 1967; Hambrick & Finkelstein, 1995; GomezMejia, Tosi, Sc Hinkin, 1987; McEachern, 1975) and 2)

PAGE 20

14 monitoring activities are greater in owner-controlled firms compared to management -control led firms (Tosi & Gomez-Mejia, 1989; Tosi & Gomez-Mejia, 1994). For example, studies show that in management -controlled firms, changes in CEO pay are positively associated with change in sales, while in ownercontrolled firms, changes in pay are positively associated with changes in firm performance (Baumol, 1967; Hambrick & Finkelstein, 1995; Gomez-Mejia, Tosi, & Hinkin, 1987; McEachern, 1975) Further, in owner-controlled firms, CEOs may be penalized for decreases in profit (Hambrick & Finkelstein, 1995) In a management-controlled firm, the manager is likely to control the nature of his compensation package through a contract that is behavior-based. The reason is that a behavior-based contract minimizes performance risk in that CEO pay is less dependent on financial output measures. Instead, a behavior-based contract relies on information systems that evaluate performance by making judgments about CEO behavior (Demski & Feltham, 1978) Thus, in managementcontrolled firms, because there is no large single shareholder to influence firm strategies and practices, the evaluation of the CEO is more likely to be based on nonfinancial measures that evaluate CEO behavior (i.e., management or leadership skills of the CEO) since these indicators involve no pay risk. Therefore: Hypothesis 1: The use of the behavior-based measures of CEO performance will be greater in management-controlled firms than in owner-controlled firms.

PAGE 21

15 On the other hand, though pay/performance sensitivityis not as strong in management -controlled firms (GomezMejia, Tosi & Hinkin, 1987), when financial performance is strong, the pay risk is diminished due to higher firm performance. In this case, the powerful CEO is likely to look toward the use of these traditional economic indicators and base a substantial part of the compensation on these output indicators (Gomez-Me j ia Tosi & Hinkin, 1987). Thus: Hypothesis 2: The importance of outcome -based performance measures should he significantly more important in CEO evaluation in high performing management-controlled firms than in low performing management controlled firms. In owner-controlled firms the shareholders are likely to take an active role in monitoring the performance of the CEO. This should result in the evaluation and rewarding of CEO performance using an outcome-based contract. An outcome-based contract aligns the interests of the shareholder and manager (CEO) in tying compensation to financial performance measures of the firm. Therefore: Hypothesis 3: The use of outcome-based performance measures will be greater in owner -control led firms than in management -control led firms. From the arguments and hypotheses above, it follows that the relative importance of the behavior-based and outcome-based criteria should also be different for management and owner-controlled firms. The financial performance measures should be given more importance in owner-controlled firms versus management -control led firms (Tosi & Gomez-Mejia, 1994) Therefore:

PAGE 22

16 Hypothesis 4: The use of outcome -based performance measures will be significantly more important in owner-controlled firms than in management-controlled firms while behaviorbased measures will be more important in management controlled firms. Other Bases of CEO Power While equity concentration is a useful indicator of the power of the CEO relative to the board of directors, there are other proxies for CEO influence. This influence would be reflected in the relative length of time spent as CEO (or tenure) holding the chairman of the board position, and control over board meetings. These influence indicators are examined in the hypotheses that follow. CEO tenure Generally, CEOs are able to exert influence over the board of directors when they have held their position as CEO for a relatively long period of time. As CEO tenure increases, so does the relationship between pay and firm size, where there is a negative relationship between pay and stock returns (Hill and Phan, 1991) Further, there is a positive relationship between the use of nonfinancial criteria and CEO tenure (Bushman, Injejikian & Smith, 1996). Salancik and Pfeffer (1980) and Finkelstein and Hambrick (1989) report that CEO tenure is higher in managementcontrolled compared to owner-controlled firms, while controlling for firm performance. This suggests that in owner-controlled firms, boards of directors terminate CEOs for performance deficiencies more quickly than in

PAGE 23

17 management -controlled firms. In management -control led firms, the CEO is afforded relative discretion in decisionmaking, and opposition by any owner or group of owners of the firm is unlikely given the wide dispersion of ownership (Salancik & Pfeffer, 1980) This leads to the following hypotheses: Hypothesis 5: The use of behavior-based measures will be greater in firms where CEO tenure is high versus firms where tenure is low. Hypothesis 6: The use of outcome-based peformance measures will be greater in firms where CEO tenure is low versus firms where tenure is high. CEO duality CEO duality refers to whether the CEO holds, at the same time, the position of board chair. It is estimated that over 70% of CEOs also hold the position of chairman of the board (Fierman, 1990) Permitting the CEO to hold this dual role has been cited to be one of the causes of failure of the board to govern properly (Fierman, 1990; Ettorre, 1992) The chairman of the board is responsible to the shareholders and is supposed to be relatively detached from the organization. However, the dual role not only obscures the separation of the ownership and management interests, but also leaves the CEO/chairman with seemingly no one to answer to (Zajac & Westphal, 1996a; Ettorre, 1992) Research shows that CEOs have higher compensation in firms where they serve as chairman of the board compared to firms in which these two roles are held by different persons

PAGE 24

(Fosberg, 1999; Sridharan, 1996) Thus, the CEO who has the additional power of the chair of the board of directors should be in a position to affect favorably the design of his/her compensation contract. Therefore: Hypothesis 7: The use of behavior-based measures will be greater in firms where the CEO is chairman of the board compared to firms in which the CEO is not chairman. Hypothesis 8: The use of outcome -based performance measures will be greater in firms where the CEO is not chairman of the board versus firms where the CEO is chairman. Board meeting control In a comprehensive study of boards of directors, Lorsch and Maclver (1989) suggest that the CEO's relative power base stemmed from control of board meetings by possessing superior knowledge of the company's activities, controlling the agenda, monitoring the types of information received by the board, and the leading of the boardroom's discussions. Boards do not seem to view their position as one of "power" to wield over the CEO, but rather an advisory and consultative position (Chitayat, 1985) Lorsch and Maclver (1989) found that directors' respect for the CEO was very important in board meetings and expressing opposition to the CEO was not appropriate. Tosi and Gomez-Mejia (1989) have shown that the CEO in the management -control led firm is the most important actor in the CEO compensation setting process. Therefore, it is expected that when the CEO does have substantial control of board meetings, regardless of

PAGE 25

19 the nature of the ownership structure, he or she is able to reduce compensation risk. Thus: Hypothesis 9: The use of behavior-based measures will be greater in firms where CEO influence in board of directors meetings is high than in firms where CEO influence at board meetings is low. Hypothesis 10: The use of outcome -based performance measures will be greater in firms where CEO influence in board of directors meetings is low versus firms where CEO influence at board meetings is high. Attribution Theory The evidence from managerial capitalism theory suggests that boards of directors are not strong in protecting shareholders' interests when the manager controls the firm (Tosi Sc Gomez -Me j ia, 1989; Singh & Harianto, 1989; Amihud & Lev, 1981) This would imply, as hypothesized above, that boards are using different types of contracts in evaluating CEO performance. When the owner controls the firm, there is more pay risk (Hambrick & Finkelstein, 1995; Gomez-Mejia, Tosi Sc Hinkin, 1987) suggesting the use of outcome -based contracts that rely on financial performance measures to evaluate CEO performance. In management-controlled firms, on the other hand, there is less pay risk and monitoring of the CEOs behavior (Tosi & Gomez-Mejia, 1989; Hambrick & Finkelstein, 1995) suggesting the use of behavior-based contracts. If this is so, that boards are using different criteria to evaluate performance, and with the increased attention to CEO pay in the media and the SEC rulings, it is important that directors are able to justify the criteria

PAGE 26

20 they use, whether behavior-based or outcome-based. The main point in this study is that since outcome-based criteria are more likely to align management and owner interests, boards should tend to make more use of them and discount, as a general rule, behavior-based indicators. Since the evidence suggests that they do not do this in the case of powerful CEOs, there is bias in the CEO evaluation process. This bias can be understood in terms of attribution theory (Weiner, 1971) Attribution theory centers on the explanation of how people make attributions about "causeeffect" relationships. For purposes here, these causeeffect relationships can be seen in terms of the types of CEO evaluation criteria: outcome-based measures can be thought of as causes of firm performance, behavior-based criteria can be seen as the effects of the CEO behavior. The question becomes one of whether the focus of the evaluation is on the cause (behavior-based criteria) or on the effect (outcome-based criteria) Among the set of factors that can explain biased evaluation is affect, or likeability, towards an individual being rated. For example, "liking" can bias a rating in that raters respond more negatively and intensely when rating poor-performing-but-disliked individuals. However, when rating poor-performing-but-liked individuals, raters attribute poor performance to external forces or factors that are beyond control by the ratee (Dobbins & Russell, 1986; Cardy & Dobbin, 1986) These occurrences of biased

PAGE 27

21 ratings have been demonstrated in research on performance ratings, where, for example, it has been shown that raters attribute female success at work to luck rather than ability (Feldman-Summers & Kiesler, 1974) Such results suggest that boards that have positive affect for their CEOs are more likely to respond positively and intensely when making performance evaluations of the CEOs simply because they like the CEO. CEOs are likely to select board members who share the same ideals and beliefs about the strategic direction of the firm. This shared vision can result in biased performance evaluations in that poor firm performance will be attributed to uncontrollable forces, and good performance to the CEO's leadership (Kerr & Kren, 1992) Further, CEOs can use interpersonal influence tactics to increase positive affect for them by board members (Westphal, 1998) Westphal (1998) shows that CEOs increased the use of persuasion and ingratiation tactics when boards of directors attempted to increase independence through structural changes. These interpersonal influence tactics, used as an alternative source of power, resulted in increases in CEO compensation (Westphal, 1998) In addition, board members may be selected based on their similarities to the CEO. What has been termed social comparison theory (Festinger, 1954) is the process individuals go through in the selection of others who will evaluate them. In the firm's selection of outside board members, they will look for other CEOs who are perceived as

PAGE 28

22 equal to or better than the focal CEO, and this benefits the CEO (O'Reilly, Main & Crystal, 1988) For example, O'Reilly, Main and Crystal (1988) show that boards remunerate the CEO similarly to their salaries in their own firm. These authors postulate that board members are selected on the basis of their salary, increasing the likelihood that they will compensate the CEO similarly. Further, Westphal and Zajac (1995) found that increased demographic similarity (i.e., similar educational and functional background) between the CEO and board members is significantly related to increases in compensation levels of CEOs Placing these findings in an attribution theory framework leads to a set of hypotheses based on the inference that positive affect by the board will be higher when board members are similar to the CEO and when they they are obligated by the firm or the CEO in other ways. This should lead the design of behavior-based contracts with lower pay risk for the CEO. Thus, CEOs can wield considerable discretion in firms where they are able to influence the composition and operation of the board. The board characteristics hypothesized to be important indicators of CEO influence are the proportion of inside directors of a firm, the proportion of outside directors who are past or present CEOs of other firms, and board pay (Westphal & Zajac, 1995; O'Reilly, Main & Crystal, 1988; Sridharan, 1996; Boyd, 1994)

PAGE 29

23 Board Characteristics Hypotheses Inside/outside directors There are conflicting theories and evidence about the role of inside and outside directors as board members. Agency theory argues that larger proportions of outside directors should restrain the discretion of CEOs (Fama & Jensen, 1983b) They are presumably better "stewards" of the company's resources. In management -control led firms where dispersion of ownership is greater, monitoring in the form of greater proportions of outsiders on the board of directors is predicted to align the interests of management and shareholders (Li, 1994). However, contrary evidence suggests that the role of the outsider might not be so (i.e., Mallette & Fowler, 1992). Research shows that CEO compensation is higher in firms with greater proportions of outside directors controlling for firm size and performance (Westphal & Zajac, 1995; O'Reilly, Main & Crystal, 1988) Inside directors, who essentially work for the organization, are not expected to be good monitors of CEO behavior (Fredrickson, Hambrick & Baumrin, 1988) Research has shown a positive relationship between both a relatively high proportion of inside directors and CEO board control in the nomination of new directors (Westphal & Zajac, 1995) and CEO compensation (Sridharan, 1996; Williamson, 1963). Another theoretical perspective, tournament theory, suggests that inside directors should be better monitors of CEO performance, as they have more to gain from removing a

PAGE 30

24 CEO from office since they are typically senior executives in the firm who might be vying for the CEO's position (Lazear & Rosen, 1981) On the other hand, insiders may have a lot more to lose than gain in firms where they actively seek to jeopardize the CEO's position. The hypotheses that follow are based on agency theory rather than tounament theory because it seems to be logical to postulate that a board with a high ratio of insiders to outsiders is more likely to focus on outcome-based measures of firm performance, while a board dominated by insiders is more likely to use behavior-based measures in the evaluation of CEO performance. Since the insider works with the CEO on a daily basis, he/she is likely to not only share the same strategic vision for the firm, but also have a collegial and friendly relationship with the CEO. Therefore: Hypothesis 11: The larger the proportion of inside directors the greater the use of behavior-based measures in the CEO performance evaluation. The conflicting evidence about the role of the outsider suggests the existence of mediating factors in the relationship between CEO influence and proportion of outsiders. Since there is research that shows that boards tend to nominate outside members who are either current or past CEOs of other companies (Aigner, 1992) this suggests that one of these mediating factors might be the proportion of outsiders on a board who are presently or formerly CEOs of other firms. The rationale for this is that CEOs are likely to ascribe to the philosophy "do unto others as you

PAGE 31

25 would have done to yourself" (Aigner, 1992 ; Hiltzik, 1996) Therefore it is proposed that CEOs of other firms who are on boards are likely to look after each other and recommend higher cash compensation packages, leading to the following hypothesis. Hypothesis 12: The larger the proportion of outside directors who are present or past CEOs, the greater the use of behavior -based measures In the CEO performance evaluation Board compensation There is a positive relationship between the compensation that board members receive for their services and CEO compensation (Boyd, 1994) The implication of this finding is that board members with high compensation contracts may not be as vigilant as board members who make less money, as the board would have a similarto-me attitude in CEO compensation decisions. Further evidence suggests that outside board members who are top executives at other firms remunerate CEOs similarly to the compensation they receive at their respective firms (O'Reilly, Main & Crystal, 1988). O'Reilly, Main and Crystal (1988) found a positive relationship between outside directors' compensation and CEO compensation. One would expect this similarto-me attitude to extend to the remuneration of board members for board service, where the more you pay board members, the greater the CEO compensation. Hypothesis 13: The higher the remuneration of the board of directors, the greater the use of behavior-based measures in CEO performance evaluation

PAGE 32

26 Conclusion The performance evaluation of the CEO is an important factor for CEO compensation packages. It can be used as a determinant of the package or as a justification for it. While there are a multitude of studies examining CEO pay and its determinants, what has been overlooked is the performance appraisal process in determining CEO pay. By understanding the performance evaluation process, it may be possible to discern how CEO pay decisions are made and maybe provide some answers to the public outcries for justification of CEO salaries.

PAGE 33

CHAPTER 2 METHODS This chapter discusses the research methodologyutilized in the testing of the hypotheses discussed in Chapter One. It is organized into two sections: data collection method and sample and the measures. Data Collection Method and Sample A questionnaire was sent to 1,000 American Compensation Association (ACA) members who were top executive compensation officers of their respective firms. They were selected based on the premise that they would be knowledgeable informants about the nature of CEO compensation in the firm and the criteria used in the evaluation for that compensation. The ACA members received a cover letter, the survey, a postage-paid envelope for returning the completed survey, and a postage-paid postcard that could be returned separately if they wanted to receive a copy of the results of the study. The cover letter explained the nature of the study in furthering understanding of the area of CEO compensation and ensured all participants strict confidentiality and anonymity (see Appendix A for copies of the mail survey materials) Approximately three weeks after the initial mailing, 27

PAGE 34

. 2 8 reminder postcards were sent to all respondents in order to increase the response rate. Phone calls were made after a period of several weeks after the reminder postcards to solicit additional responses. Of the 1,000 surveys that were sent out, 201 were returned for a response rate of 20.1%. Measures The questionnaire consisted of a set of items describing firm characteristics and demographics, board characteristics, items used to assess the importance of behavior-based criteria and outcome-based CEO evaluation criteria, and items determined to be important to the testing of the hypotheses. These specific measures are listed below. Independent Variables Ownership Structure Ownership structure is typically determined by the dispersion of shares owned by individuals or institutions. A firm is called owner-controlled if an individual or institution holds more than 5% of that firm's shares. A firm is called management -controlled if no one individual or institution holds greater than 5% of that organization's shares (McEachern 1975; Gomez-Mejia, Tosi & Hinkin, 1987; Tosi Sc Gomez-Mejia, 1989; Hambrick & Finkelstein, 1995). It is called owner-managed when an internal manager holds 5% or more of the stock.

PAGE 35

29 Firms were classified into these three ownership categories. The ownership structure of the firm was determined by two items used in prior studies (i.e., Tosi & Gomez-Mejia, 1989). The first item asked respondents whether there was "any single individual or institution outside the firm that owns 5 percent or more of the company stock." Firms were classified as owner-controlled if the answer was "yes." The total number of owner-controlled firms in this study was ninety-eight (n=98) Firms were coded as management-controlled if the answer was "no." The total number of management -controlled firms in this study was ninety (n=90) Ownership structure was coded as a dummy variable where the answers from owner-controlled firms were coded as "1," and management -controlled firms "0." The second item assessed whether a firm was ownermanaged and asked the respondent whether there was "any individual manager within the firm (e.g., the CEO) who owned 5 percent or more of the company stock." An answer of "yes" to this question indicated the firm was owner-managed. Since the number of owner-managed firms was only four, owner-managed firms were removed from the analysis. Board Characteristics The study assessed three characteristics of the board. These were: proportion of insiders relative to outsiders, whether or not any of the board members had been past or present CEOs of other firms, and board pay.

PAGE 36

30 Proportion of insiders This variable was assessed by the subject's response to the following items: "how many directors are also executives in the firm" and "how many directors are from outside the firm. These two items were summed together and divided by the number of inside directors. Directors past/present CEOs This variable was assessed by the subject's response to the following item: "how many of the outside directors are present or past CEOs of other firms?" Board pay *' • 'J Board compensation was determined by the subject's response to the following item: "what is the average cash compensation for board membership per year. CEO Influence and Power There were three different measures of CEO influence: CEO tenure, CEO duality, and board meeting control. CEO tenure This variable was assessed by the subject's response to the following item: "how long has the CEO held this position with your firm." CEO duality CEO duality was assessed by the subject's response to the following item: "is the current CEO also chairman of the board." Duality was coded as a dummy variable. If the CEO was also chairman, the answer was coded as "1," and if not, 0

PAGE 37

31 Board meeting control The extent to which the CEO has influence in running the meetings of the board was assessed by a set of four items describing CEO board meeting control on a seven-point Likert-type scale ranging from "none" (1) to "very high" (7) The items were subjected to a principal axis factor analysis with varimax rotation which revealed that all items loaded on a common factor accounting for 85% of the variance. The scale had a reliability of alpha = .94. The following four items (shown with factor loadings) were used to assess CEO board meeting influence. 1. How much influence does the CEO have in reviewing and approving the types of information received by board members? (.66) 2. How much influence does the CEO have in setting the agenda at board meetings? (.84) 3. How much influence does the CEO have in the running of board meetings? (.80) 4. How much influence does the CEO have in important board decisions ? (.72) Firm Performance Since more conventional archival measures of firm performance were not available because of the anonymity of the subjects, overall firm performance was assessed by the subject's response to an item that asked him or her to indicate the firm's financial performance relative to other firms in its industry. This provided a subjective estimate

PAGE 38

32 of relative performance in the industry and, as such, was a control for industry differences. Such a measure of performance has been shown to correlate highly with conventional archival economic indicators of performance (Tosi & Gomez-Mejia, 1994) in that measures reported by respondents and those collected through archival means are shown to correlate, suggesting that respondents are able to accurately categorize performance, and that this method of categorization correctly represents a firm's industry performance. Subjects responded to the following item: "please rate your firm's financial performance relative to others in your industry." The scale measuring firm performance consisted of a five-point scale (1= top performers to 5= lowest performers) dividing relative firm performance into percentages ranging from top 20% to lowest 20%. Dependent Variables There are two types of dependent variables in this research. One type are outcome-based performance measures that are typically objective measures of a firm's financial performance. The second are behavior-based measures of CEO performance. These are more subjective indicators that measure different facets of the CEO's style of managing. Both were assessed with Likert-type scales on which respondents indicated the relative importance of each in CEO evaluation on a seven-point scale ranging from "not at all important (1) to "very important "( 7 )

PAGE 39

Outcome-Based Performance Measures of CEO Performance There are two outcome -based performance measures of CEO performance that may be used in the evaluation of CEO performance by boards of directors: accounting measures of firm performance and\or market measures of firm performance. Accounting measures assess a firm's current profitability by analyzing financial information from operating statements and balance sheets of the firm. Accounting measures are important in top management compensation decisions and also in decisions related to the viability of a business. Subjects were asked to indicate the extent to which the following six outcome-based measures of firm performance were used in the evaluation of the CEO. 1. Return on assets (ROA ) Return on assets (ROA) is a measure of a firm's effectiveness in deploying its assets. ROA is the ratio of net income relative to total assets. This measure is used in investment and operating decisions (Logue & Steward, 1997) 2 Return on equity (ROE) Return on equity (ROE) is a measure of return on invested capital ROE is based on net income relative to owner's equity. This measure tells shareholders how effectively their money is being used, and provides a basis for comparisons with competitors (Logue & Steward, 1997) 3. Return on investment (ROD Return on investment (ROI) expresses profit in direct relation to investment. ROI is a ratio of net income to total assets. It is used

PAGE 40

34 for comparing firms or divisions in terms of management efficiency and as the basis for whether the firm is maximizing the interests of the owners (Logue & Steward, 1997) 4 Sales growth. Sales growth reflects increases or decreases in sales in a firm. It is usually measured as the change of sales from previous fiscal years (the ratio of total sales this year to total sales last year) Sales growth is used to determine how well a firm's product or service is doing over time, and also used to compare product lines. 5. Stock price increase Changes in the price of the stock of a firm is an indicator of the capacity of the firm to create shareholder value as it signals equity holders how much their shares are worth relative to earlier points in time. 6 Economic value added (EVA) Economic value added reflects the economic returns generated to shareholders (Logue & Steward, 1997) It is a measure of the "gain (or loss) that remains after levying a charge against after-tax operating profits for the opportunity cost of all capital equity as well as debt used to produce those profits" (Ehrbar, 1999: 20). EVA is essentially a measure of operating performance that tells the shareholder how valuable the business is, and is therefore a good indicator of shareholder value. In the typical research on CEO compensation that

PAGE 41

35 examines whether it correlates with firm performance, there are two types of outcome-based measures used: accounting measures of firm performance and/or market measures of firm performance. Accounting measures assess a firm's current profitability by analyzing financial information from the operating statements and balance sheets of the firm. Accounting measures are important in top management compensation decisions and also in decisions related to the viability of a business. Accounting measures are thought to be less valid indicators of firm performance than market indicators because they reflect the accounting performance rather than the cash flow performance of a firm (Logue & Steward, 1997) Additionally, accounting measures are susceptible to manipulation in firms, as they can be overstated, and also distort the assessment of firm performance (Saloman & Smith, 1979) Thus, some believe that accounting measures do not represent the true value of the firm (Logue & Steward, 1997) Market measures of firm performance reflect stock market indicators that signal the worth of the equity of a firm. These measures represent the shareholders' interest as they are indicators of the value of the firm. They are preferred by some (Logue & Steward, 1997) because they are not as easy to manipulate by managers of a firm as are accounting measures (Rosen, 1990) However, they are subject to shifts in value that have no relationship to financial performance changes (Rosen, 1990) For example.

PAGE 42

36 stock prices commonly fluctuate without any changes in the firm's value. Additionally, shifts in the market discount rate can also distort the interpretation of market measures (Rosen, 1990) However, there is evidence that neither is a superior measure to the other, so both may be useful measures of firm performance for research purposes (Rosen, 1990) For this study, a research choice was not made to try to assess these two different classes of indicators based on the assumption that the informants would not be able to discriminate usefully between them. Therefore, the set of financial criteria listed above was chosen. A factor analysis of all the outcome-based performance measures was executed to assess their relationship to each other for this research (see Appendix B) Three factors emerged, but the items that loaded within each factor had no apparent logic or theoretical basis of the type that one might expect. Because of the exploratory nature of this research and the lack of more objective archival measures, it was decided to use a pooled outcome-based performance measure. This was done by simply summing the subject' s responses to each of the six items above to attain a value used in the analysis. Pooled or combined measures of firm performance have been used elsewhere in studies that have used both archival financial data as well as data reported by knowledgeable informants (Tosi & Gomez-Mejia, 1994; Gomez-Mejia, Tosi &

PAGE 43

37 Hinkin, 1987) The reliability for this pooled financial performance measure is .60. Behavior-Based Measures of CEO Performance The popular business press is replete with reports of what distinguishes effective CEOs from ineffective CEOs. There is, however, no well -developed theoretical or empirical literature on this point. It is true that there is an abundance of research on effective leadership in organizations (Bass, 1985; Bass, 1990; Fiedler, 1967; House, 1971) but it is fair to say that this research was done on effective managers at lower to middle organizational levels. Since this is the case, this research draws on theoretical literature that seems to be relevant to those characteristics described in the popular literature. Three aspects of CEO behavior are examined as indicators of behavior-based measures that boards of directors might use in the evaluation of CEO performance. These are: the behavior of the CEO in the managerial role (Mintzberg, 1973) the CEO as a leader (Bass, 1985) and the organizational citizenship of the CEO (Organ, 1988) These aspects of managerial behavior are conceptually quite different. The fundamental distinction is that CEO leadership behavior involves the direction of members to organizational goals, where managerial role performance focuses on the specific activities in different roles that leaders perform. Citizenship behavior reflects activities that go beyond the manager's role prescriptions. These

PAGE 44

38 behaviors are expected to contribute to organizational functioning in that they facilitate the accomplishment of organizational goals. Managerial role performance Mintzberg (1973) developed a taxonomy of managerial roles from his study of CEOs. He contends there are 10 managerial role behaviors that fall into three general categories of managerial role performance: 1. The interpersonal role describes activities involving the manager's relationships with others and consists of three roles; figurehead, liaison, and leader. All of these roles involve behavior that is social in nature and encompasses the manager's relationships with employees, customers, the public, and stockholders. 2. The informational role includes those roles that deal primarily with information, and consists of three roles: monitor, disseminator, and spokesman. These roles involve the receiving, transmitting, and general handling of information by the manager. 3. The decisional role reflects the organizational decisions that managers are responsible for. It includes four roles: entrepreneur, disturbance handler, resource allocator, and negotiator. These roles reflect the manager's involvement in strategic planning, scheduling work and budgeting resources, and negotiating with outside organizations. The subjects responded to a shortened version (29

PAGE 45

39 items) of a managerial role behavior scale developed by Lau and Pavett (1980) In order to determine whether these items could be grouped into the same categories as Mintzberg developed, they were subjected to a factor analysis using principal axis with varimax rotation (see Appendix B for factor analysis) A priori, 10 factors were expected to emerge consistent with Minztberg s (1973) categorization, but the analysis resulted in four factors. The factors that did emerge were uninterpretable and inconsistent with prior research. Given the exploratory nature of this study, it was decided to keep the scale as originally devised to assess one factor, managerial role performance. The reliability for this measure is .89. The items used to assess managerial role performance are: 1. Taps into the grapevine to keep abreast of what is going on in the company 2. Is available to outsiders (e.g., consumers, suppliers, the public) who want to go to the "person in charge" 3. Keeps up with market changes and trends that might have an impact on the organization 4. Surveys the organization to identify ways to improve performance or cut costs 5. Integrates subordinates' goals (e.g., career goals, work preferences) with the organization's work requirements .. 6. Keeps professional colleagues informed about the organization 7. Entertains clients on a regular basis 8. Negotiates with groups outside the organization for necessary materials, support, commitment, etc. 9. Attends to staffing requirements in different units, such as hiring, firing, promoting, and recruiting 10. Provides guidance and direction to company employees 11. Keeps members of the organization informed of relevant information through meetings, conversations, and written information 12. Develops new contacts by personally answering requests for information 13. Takes immediate action in response to a crisis or "fire drill"

PAGE 46

40 14. Stays tuned to what is going on in competitor organizations 15. Passes along unofficial information 16. Joins boards, organizations, clubs, or does public service work that might provide useful, work-related contracts 17. Passes on the desires and preferences of top management to others in the firm 18. Keeps executive members, consumers, or other important groups informed about the organization's activities and capabilities 19. Evaluates the outcomes of internal improvement projects 20. Plans and develops improvements in work flow or work methods 21. Programs work for different units within the organizations (what is to be done, when and how) 22 Resolves disputes between subordinates or work groups 23. Allocates resources (e.g., personnel, money, material) among units within the organization 24. Determines the long-range plans and priorities of the organization 25. Is willing to be the official spokesperson of the company at gatherings outside of the organization 26. Resolves work flow problems between units 27. Personally answers letters or inquiries about the organization 28. Attends outside conferences or meetings 29. Personally handles complaints from customers or subordinates CEO leadership behavior The transformational leadership behavior of the CEO was assessed by the informant in each firm, drawing on the conventional wisdom of the business press that effective CEOs are good leaders. For example, an effective transformational leader like Jan Carlzon, CEO of SAS Airlines, encourages a participative environment where the key stakeholders of the firm are involved with the formulation of the organizational vision. Carlzon believes that a leader must inspire others to pursue the vision, but the vision must be inspiring in order to be followed (Ackoff 1999)

PAGE 47

There is evidence that certain leadership styles or behaviors are associated with improved subordinate performance (Avolio & Bass, 1988; Bass & Avolio, 1990; Seltzer & Bass, 1990) For example. Seltzer and Bass (1990) found that transformational leadership is positively related to beneficial organizational outcomes though these outcomes were not outcome-based performance measures of the type discussed earlier. Other research has shown the effects that transformational leaders can have on organizations (Waldman, Bass & Einstein, 1987) For example, Waldman, Bass and Einstein, (1987) show that transformational leadership is positively correlated with employee satisfaction with performance evaluation processes. CEO leadership behavior was measured with a sevenitem scale that was a modification of a measurement instrument developed by Podsakoff et al (1990). Subjects' responses were factor analyzed using principal axis with varimax rotation, and the resulting scales emerged (see Appendix B for factor analysis) Two factors emerged that were inconsistent with prior findings (Podsakoff et al 1990), and not interpretable It was therefore decided to use the entire scale in the analysis to assess CEO leader behavior. The reliability for this measure is .77. The items used to assess CEO leadership behavior are: 1. Provides employees with new ways of looking at things 2. Leads by doing rather than simply by telling 3. Insists on only the best performance 4. Seeks new opportunities for the organization 5. Inspires others with his/her plans for the organization 6. Encourages employees to be "team players"

PAGE 48

7. Shows respect for individuals' personal feelings Citizenship behavior Effective CEOs are also often seen contributing in other ways than filling their leadership and management roles. Organizational citizenship behavior is a concept that captures this aspect of CEO behavior (Organ, 1997) Organizational citizenship behavior is defined as "performance that supports the social and psychological environment in which task performance takes place" (Organ, 1997: 95) There are five dimensions of citizenship behavior. These are altruism, conscientiousness, sportsmanship, courtesy and civic virtue. An abbreviated scale of fourteen items to assess these five dimensions was taken from earlier existing scales (MacKenzie, Podsakoff & Fetter, 1991; Bateman & Organ, 1983; Organ, 1988; Podsakoff & MacKenzie, 1994; Smith, Organ & Near, 1983) Responses were factor analyzed using principal axis with varimax rotation (see Appendix B for factor analysis) The factors that emerged were inconsistent with prior research (Bateman & Organ, 1983; Organ, 1988) and not clearly interpretable therefore it was decided to combine all the items to create a single factor that assesses citizenship behavior. The reliability for this measure is .58. The items used to assess citizenship behavior are: 1. Acts cheerful 2. Trains or helps others to perform their jobs better 3 Keeps up with developments in the company 4. Expects themselves and others to conscientiously follow company regulations and procedures

PAGE 49

43 5. Exhibits punctuality in arriving at meetings 6. Attends functions that are not required, but that help the company image 7. Willingly risks disapproval in order to express beliefs about what s best for the company 8. Takes steps to try to prevent problems with other executive members, board members and/or other employees in the company 9. Cooperates well with others in the organization 10. Makes innovative suggestions to improve the overall quality of the company 11. Focuses on the positive side of a situation rather than the negative side of it 12. Willingly gives of his/her time to help others 13. Returns phone calls and responds to other messsages and requests for information promptly 14. "Touches base" with others before initiating actions that might affect them The Relative Importance of Outcome-Based and Behavior-Based Measures of CEO Performance One item was developed to understand the overall relative importance of behavior-based and outcome-based criteria in the CEO evaluation by the board of directors. The following item was used to assess the relative importance of the different types of criteria: "Rate the current relative importance of quantitative criteria or measures as compared to qualitative criteria or measures in the determination of the CEO compensation package." A seven-point scale was used in this assessment with the following ranges: 1 "Qualitative most important" 4 "About equal" 7 "Quantitative most important" Control Variables The following control variables were used in the analysis.

PAGE 50

44 Industry In order to control for industry effects, two items were used. One asked the respondents to report the firm's SIC classification, with the intention to control for a specific industry in the analysis of the data. There was a total of eight SIC categories. Because a majority of respondents in this study were from manufacturing firms (n=95) this variable was coded as a dummy variable with "0," assigned to manufacturing firms and all other firms coded as 1 The second item assessed whether the firm's main products were durable or nondurable goods, since it is possible that the type of product a firm produces can have an effect on different firm performance measures (Tosi & Gomez-Mejia, 1994). This measure was used as a dummy measure in the analysis with the coding of "0," if the product was nondurable, and "1" if the product was durable. Firm Size Firm size was also used as a control variable (McEachern, 1975) Firm size was measured in this study by the subject's response to the following item: "how many people are employed by your firm. Summary This chapter describes the survey methodology utilized in the testing of the hypotheses. The specific measures used were items that had been developed in prior studies and some that were specifically developed for this study. As a

PAGE 51

result of inconsistent findings from the factor analyses o the behavior-based and outcome-based performance measures, the decision was made to use single scales instead of subscales for each of the measures. The next chapter will describe the data analysis used to test each of the hypotheses and report results.

PAGE 52

CHAPTER 3 RESULTS The data were gathered from 201 firms representing 7 industries. The sample used in the analysis contained 188 cases due to incomplete data and data that came from firms that were not comparable to the rest of the sample. Analysis of the Data The primary objective of this study is to examine the effects of the seven independent variables on the two dependent variables as specified in the hypotheses. The means, standard deviation, and bivariate correlations for all variables are reported in Tables 3.1 and 3.2. The secondary objective is to examine the relationships between CEO duality, tenure, influence at board meetings, and board characteristic variables on the outcome and behavior-based variables by ownership structure. Regression models were run to test these relationships between the seven independent variables and each of the four dependent variables as outlined above. Linear regression was used to estimate the coefficients of the model. This chapter is divided into two main sections because several different independent variables in various regression models were used to test the hypotheses. The first part of this chapter reports regressions for each 46

PAGE 53

47 Table 3.1 Means and Standard Deviations Variables Mean Standard Deviation 1. Size(# of emp) 32,405 83,558 2. Performance (quartile ranking) 2.23 1.12 3. SIC .49 .50 4 Sector .33 47 5. Ownership Structure .52 .50 6. Duality .65 .47 7. Tenure (years) 6.46 4.52 8. CEO Bd. Mtg. Influence 6.37 1.13 9. % Dir Past/Present CEOs .41 .19 10. % Insiders .21 .98 11. Board Pay $30,000 $18,000 12. Citizenship behavior 3.58 .51 13. CEO leadership behavior 4.88 .86 14. Managerial role performance 4.35 .72 20. Outcome-based performance measure 4.14 .97 21. Importance outcome vs behavior-based 5.56 1.19 N = 188

PAGE 54

CO o 0) "3 rH XI 0) U m oj i/i >^ o o I I n H CN (N ro (N m o M o ch rr(N iH O O rH O '^l^ ro (Ti ^ (N O rn O rH O iH O ^ O rH rH r-m £T inrHojmcNro^^^ mOrHOrHOrHCNCNJ CO O 1^ U3 rH rH O O O O) rH rH n n cN CN rs] o OJ O rH O (N I I I I I I I LD ^ rH O rH LT) ojconcN^cNr^rHCTimo OOOHOrHOOOOrH I I OVD(NmC0^rHr0f^C0rs] rHOOOOOOrHOrHrH I I I I I I I I o I LTi rH ^ o] r-~ rH O O O O O o intNVDOOCN'^CnoOrOCOLOrHLnLn OrHOrHrHOOOOCNOOOrH a 0) cn ra O H 0) 0) > U iH (C C 3 J3 cn a H X a IS 4J IH (0 to ra 0) E (0 X) > n) £1 > HH (0 U 0) o 3 c a HH e cn , U a Sh tD 4J 0) a -H 0) 0 0 u H t4 m (H CD t)p O rH m in CD a\ rH 0) 14H 0) v a, u B (D o 4-1 0) tt rH X! -rH o T) 3 J3 ^H 0) O Cli cn CO -H tH rH td (D cnc(i>,£i(D(OXiu nj cn 13 C cd Sh (U 0) QJ (D in rO 4-) O O -H w m u u E cn o >H (0 0 0 ra 3 E S O M CO H o £ H U (U LD O VI K

PAGE 55

49 dependent variable with all the hypothesized independent variables and control variables. It also reports regressions of the relationship between the dependent variable on independent variables by ownership structure to ascertain differences in management and owner-controlled firms (Tables 3.33.6). The second part reports the specific regression coefficients that are the tests of each of the hypotheses (Tables 3.7-3.18). Note, that these coefficients are drawn from Tables 3.3-3.6. Regressions of Independent Variables by Dependent Variable Outcome-Based Performance Measure Table 3.3 shows the regressions for the effects of ownership structure, board characteristics and CEO influence on the use of the outcome-based performance measure. These relationships will be examined in greater detail in the reporting of the hypotheses section. .• Table 3.3 also displays regression coefficients for the effects of the same independent variables and dependent variables for owner-controlled and management-controlled f i rms In owner-controlled firms, the use of outcome-based performance measures is negatively related to board pay (b=.16, p<.05). Board members who are compensated less tend to use criteria that reflect the financial performance of a f i rm

PAGE 56

50 In both management -control led (b=.53, p<.01) and ownercontrolled firms (b=.34, p<.01), larger proportions of outsiders who are past or present CEOs of other firms are positively related to the use of outcome-based performance criteria. In owner-controlled firms, firm size is positively related to outcome-based performance criteria (b=.30, p<.01) Thus, the larger the firm the greater the use of outcome -based performance measures. Table 3.3 Regression Model of the Effects of the Independent Variables on the Use of the Outcome-Based Performance Measure Variables Total Sample Manager Control Owner Control Std. B Std. Er. Std. B Std. Er. Std. B Std. Er. Ownership .29** 15 Duality 06 18 12 .30 .06 .29 Tenure 04 16 01 24 05 .22 CEO Bd Mtg 02 16 05 .21 08 .30 Influence % Dir. CEOs .40** .27 .53** 44 .34** 35 % Insiders 01 .71 10 1 01 06 98 Board Pay 03 00 13 00 16* 00 Performance 01 14 13 18 12 .20 Size .20** 14 03 .21 .30** 18 SIC 01 13 01 18 05 .20 Sector 07 15 .05 22 07 .21 R sg. .30 .36 .26 Adj R sq .26 .28 17 F 6 91 4 55 3 03 P-value 000 000 002 N 90 98 Chow Test (F) 1 75 *p<.05, ** p<. 01

PAGE 57

51 Behavior-Based Measures '•, Managerial role performance Table 3.4 shows the regressions for the effects of ownership structure; CEO duality, tenure and influence at board meetings; and board characteristics on the use of the behavior-based measure, managerial role performance. These relationships are examined in the next section on hypotheses testing. Table 3.4 Regression Model of the Effects of the Independent Variables on the Use of the Behavior-Based Measure, Managerial Role Performance Variables Total Sample Manager Control Owner Control Std. B Std. Er. Std B Std. Er. Std. B Std Er. Ownership 15* 11 Duality 07 13 17 22 04 .22 Tenure .02 12 06 18 07 17 CEO Bd Mtg Influence 07 12 07 16 10 .23 % Dir. CEOs 02 .20 07 .34 03 .27 % Insiders 01 .53 24* 80 15 75 Board Pay 35** .00 33** .00 .39** 00 Performance 14* 10 18* 14 14 .15 Size 03 10 12 16 09 14 SIC 19** 10 18* .14 19* 15 Sector .02 11 04 17 08 .16 R sq. .21 21 .25 Adj R sq. 16 16 .17 F 4.30 4 30 3 00 P-value .00 00 0 0 Chow Test (F) 1 .43 *p<.05, **p<.01 Table 3.4 also reports regressions of the effect of the independent variables on managerial role performance by ownership structure. In management -controlled firms, the proportion of insiders measure is related to the use of the behavior-based measure, managerial role performance (b=-.24.

PAGE 58

52 p<.05). The smaller the proportion of insiders in management -controlled firms, the greater the use of this behavior-based criteria. Board pay is negatively related to the use of this behavior-based measure in management (b=-.33, p<.01) and owner-controlled firms (b=-.39, p<.01). Board members who are compensated less in firms tend to use behavior-based measures, like managerial role performance. CEO leadership behavior Table 3.5 shows the regressions for the effects of ownership structure; CEO duality, tenure and influence at board meetings; and board characteristics on the use of the behavior-based measure, CEO leadership behavior. Specific relationships are discussed in the hypotheses section. Table 3.5 also reports regressions of the relationship between the independent variables on the use of the measure, CEO leadership behavior by ownership structure. In management-controlled firms, CEO duality is significantly related to the use of this behavior-based indicator (b=.30, p<.01). When CEOs also serve as chairman of the board, the greater the use of the measure, CEO leadership behavior in management-controlled firms. In management -control led firms, the proportion of insiders measure is negatively related the use of this behavior-based measure (b=-.23; p<.05). Board of directors with low proportion of insiders in management-controlled

PAGE 59

53 firms are less likely to use this indicator in CEO evaluation Table 3.5 Regression Model of the Effects of the Independent Variables on the Use of the Behavior-Based Measure, CEO Leadership Behavior Variables Total S amp 1 e Mans aPT" Control OwnGir Control OUQ o oua btu br Ownership 00 14 Duality.25** 16 .30** .29 15 .27 Tenure -.08 14 14 .23 03 .20 CEO Bd Mtg Influence 04 15 01 .20 15 .27 % Dir. CEOs 15* 24 15 .43 17* 32 % Insiders 12* 65 23* 1 02 05 91 Board Pay 32** 00 33** 00 .26** 00 Performance .06 12 07 18 06 18 Size .05 .12 09 .21 14 17 SIC -.04 12 12 18 07 18 Sector .06 13 08 .21 14 .19 R sq. .19 18 .25 Adj R sq 14 08 17 F 3 81 1.80 2 99 P-value 00 03 00 Chow Test (F) 72 *p<.05, **p<.01 In both management (b=-.33, p<.01) and owner-controlled (b=-.26, p<.01) firms, board pay is negatively related to the use of the behavior-based measure, CEO leadership behavior. Low board compensation is related to the use of this measure. In owner-controlled firms, the proportion of outside directors who are CEOs of other firms is positively related to the use of the measure, CEO leadership behavior (b=.17, p<.05). The greater the proportion of these directors in owner-controlled firms, the greater the use of this indicator.

PAGE 60

54 Citizenship behavior Table 3.6 shows the regressions for the effects of ownership structure; CEO duality, tenure and influence at board meetings; and board characteristics on the use of the behavior-based measure, citizenship behavior. Again, relationships between these variables are examined in the hypotheses section. Table 3.6 Regression Model of the Effects of the Independent Variables on the Use of the Behavior-Based Measure, Citizenship Behavior Variables Total Sample Manager Control Owner Control Std. B Std. Er Std. B Std. Er. Std. B Std. Er Ownership .20** 08 Duality 06 09 04 18 .18 15 Tenure 03 08 01 14 .07 11 CEO Bd Mtg Influence .09 09 01 13 19 15 % Dir. CEOs 04 14 03 .27 06 18 % Insiders 05 .38 06 .65 .10 50 Board Pay.30** 00 31** 00 .29** 00 Performance .09 07 18* 11 00 10 Size 02 07 11 13 .16 09 SIC 16* 07 18* 11 15 10 Sector 03 08 12 13 .04 R sq. .18 .21 .15 Adj R sq. 13 11 06 F 3 66 2 09 1.61 P-value 00 03 06 Chow Test (F) 73 *p<.05, **p<.01 Table 3.6 also reports the regressions testing the effect of the independent variables on the behavior-based measure, citizenship behavior by ownership structure. In management (b=-.31, p<.01) and owner-controlled (b=-.29, p<.01) firms, board pay is negatively related to the use of this measure. As with the other behavior-based measures,

PAGE 61

the less board members receive for board service, the greater the use of the measure, citizenship behavior in CEO evaluation. Report of the Tests of the Hypotheses Hypothesis 1 Hypothesis 1 states that the use of behavior-based measures of CEO performance will be greater in management controlled firms than in owner-controlled firms. The regression coefficients between ownership structure and the three behavior-based measures are shown in Table 3.7. Ownership structure is negatively related to the use of the behavior-based measure, managerial role performance (b=-.15; p<.05) and citizenship behavior (b= -. 2 0 ; p<.01). Only CEO leadership behavior was not found to be significant. These results generally support Hypothesis 1. Table 3.7 Regression Coefficients Testing the Effects of Ownership Structure on the Use of the Behavior-Based Measures Dependent Variable Ownership Structure Managerial Role Performance (tCoef f icient taken from Table 3.4) -.15*t CEO Leadership Behavior (tCoeff icient taken from Table 3.5) .OOt Citizenship Behavior (tCoeff icient taken from Table 3.6) -.20**t *p<.05, **p<.01 Hypothesis 2 Hypothesis 2 states that the importance of outcomebased performance measures should be significantly more important in CEO evaluation in high performing management-

PAGE 62

56 controlled firms than in low performing management controlled firms. The results of the regressions testing Hypothesis 2 are reported in Table 3.8. In managementcontrolled firms, firm performance is significantly related to the rated relative importance of outcome-based performance criteria (b=.41, p<.01). That is, these outcome-based performance measures are used to a greater extent in high performing management-controlled firms. Table 3.8 Regression Model of the Effects of Ownership on the Rated Relative Importance of Outcome-Based Performance Measure Compared to Behavior-Based Measures Variables Total Sample Manager Control Owner Control Std. B Std. Er. Std. B Std. Er. Std. B Std. Er. Ownership .42** 17 Duality 18** .20 18 .36 06 .29 Tenure .01 18 .04 .29 .04 22 CEO Bd Mtg .01 .18 .02 .26 .05 .29 Influence % Dir. CEOs 12* .30 .03 .55 .25** .35 % Insiders .00 .79 05 1.30 .04 .97 Board Pay. 17** .00 14 .00 .26** .00 Performance 17** .15 .41** .23 .07 .20 Size .05 15 10 .27 .01 18 SIC .05 .15 .09 .23 .01 .19 Sector .08 .16 09 .27 .16 .21 R sq. .37 .25 .20 Adj R sq. .33 15 11 F 9.40 2 64 2 19 P-value 000 01 .02 N 90 98 Chow Test (F) 1 84* *p<.05, **p<.01 Hypothesis 3 Hypothesis 3 states that the use of outcome-based performance measures will be greater in owner-controlled firms than in management -controlled firms. The results of regressions that test Hypotheses 3 are reported in Table

PAGE 63

57 3.9. Hypothesis 3, that ownership structure will be positively associated with the use of the outcome-based performance measure, is supported (b=.29, p <.01). Ownercontrolled firms use outcome-based performance criteria in the evaluation of CEO performance to a greater extent than management -control led firms. Table 3.9 Regression Coefficients Testing the Effects of Ownership Structure on the Use of the Outcome-Based Performance Measure Dependent Variable Ownership Structure Outcome -Based Performance Measure .29** (see Table 3.3) *p<.05, **p<.01 Hypothesis 4 Hypothesis 4 states that the use of outcome-based performance measures will be significantly more important in owner -control led firms, while behavior-based measures will be more important in management -controlled firms. The results of the regressions testing Hypothesis 4 are reported in Table 3.8. The regression results show that ownership structure is related to the relative importance of outcomebased performance measures compared to behavior-based measures (b=.42, p<.01). This suggests that outcome-based performance criteria are more important in owner-controlled firms, and behavior-based criteria are more important in management -controlled firms.

PAGE 64

58 Hypothesis 5 Hypothesis 5 states that the use of behavior-based measures will be greater in firms where CEO tenure is high versus firms where CEO tenure is low. The regression coefficients between CEO tenure and the three behavior-based measures are shown in Table 3.10. CEO tenure is not related to the use of the behavior-based measures. CEO tenure is not significantly related to managerial role performance, CEO leadership behavior or citizenship behavior. Table 3.10 Regression Coefficients Testing the Effects of CEO Tenure on the Use of Behavior-Based Measures Variables CEO Tenure Managerial Role Performance 02 (see Table 3.4) CEO Leadership Behavior -.08 (see Table 3.5) Citizenship Behavior .03 (see Table 3.6) *p<.05, **p<.01 Hypothesis 6 Hypothesis 6 states that the use of outcome-based performance measures will be greater in firms where CEO tenure is low versus firms where CEO tenure is high. The regression coefficients between CEO tenure and the outcomebased measure are shown in Table 3.11. Hypothesis 6 is not supported in that low CEO tenure is not related to the use of outcome-based performance criteria in CEO evaluation. Hypothesis 7 Hypothesis 7 states that the use of behavior-based measures will be greater in firms where the CEO is chairman

PAGE 65

Table 3.11 Regression Coefficients Testing Tenure on the Use of the Outcome-Based the Effects Performance of CEO Measure Variable CEO Tenure Outcome -Based Performance Measure (see Table 3.3) 04 *p<.05, **p<.01 of the board versus firms in which the CEO is not chairman. The regression coefficients between CEO duality and the use of behavior-based measures is shown in Table 3.12. Table 3.12 Regression Coefficients Testing the Effects of CEO Duality on the Use of Behavior-Based Measures Variable CEO Duality Managerial Role Performance .07 (see Table 3.4) CEO Leadership Behavior .25** (see Table 3.5) Citizenship Behavior .06 (see Table 3.6) *p<.05, **p<.01 CEO duality is positively related to the use of the behavior-based measure, CEO leadership behavior (b=.25; p<.01) That is, when CEOs are chairman of the board, the greater the use of the measure, CEO leadership behavior in CEO evaluation. CEO duality is not significantly related to managerial role performance or citizenship behavior. Hypothesis 8 Hypothesis 8 states that the use of outcome-based performance measures will be greater in firms where the CEO is not chairman of the board versus firms in which the CEO is chairman. The regression coefficients between CEO duality and the outcome-based performance measure are

PAGE 66

60 displayed in Table 3.13. CEO duality is not significantly related to the greater use of outcome-based performance measures. That is, when CEOs are not chairman of the board, there is not a greater incidence of outcome-based performance criteria in CEO evaluation. Table 3.13 Regression Coefficients Testing the Effects of CEO Duality on the Use of the Outcome -Based Performance Measure Variable CEO Duality Outcome -Based Performance Measure 06 (see Table 3.3) *p<.05, **p<.01 Hypothesis 9 Hypothesis 9, states that the use of behavior-based measures will be greater in firms with high CEO board meeting influence versus firms with low CEO board meeting influence. The regression coefficients between CEO board meeting influence and the behavior-based measures are displayed in Table 3.14. CEO board meeting influence is not significantly related to managerial role performance, CEO leadership behavior or citizenship behavior. That is, CEOs with high board meeting influence are not evaluated to a greater extent by behavior-based measures. Hypothesis 10 Hypothesis 10 states that the use of outcome-based performance measures will be greater in firms with low CEO board meeting influence versus firms with high CEO board meeting influence. The regression coefficients between CEO board meeting influence and the outcome-based performance

PAGE 67

61 Table 3.14 Regression Coefficients Testing the Effects of CEO Board Meeting Influence on the Use of the Behavior-Based Measures Variables CEO Board Meeting Influence Managerial Role Performance 07 (see Table 3.4) CEO Leadership Behavior .07 (see Table 3.5) Citizenship Behavior -.09 (see Table 3.6) *p<.05, **p<.01 measures are displayed in Table 3.15. CEO board meeting influence is not significantly related to the use of outcome-based performance measures. That is, boards of directors in firms where the CEO has little influence over board meetings, are not more likely to use outcome-based performance criteria in the CEO evaluation. Table 3.15 Regression Coefficients testing the Effects of CEO Board Meeting Influence and the Outcome-Based Performance Measure Variable CEO Board Meeting Influence Outcome -Based Performance Measure 02 (see Table 3.3) *p<.05, **p<.01 Hypothesis 11 Hypothesis 11, states that the larger the proportion of inside board members the greater the use of the behaviorbased measures. The regression coefficients between the proportion of insiders and the behavior-based measures are displayed in Table 3.16. The proportion of insiders measure is significantly related to CEO leadership behavior (b=-.12, *p<.05). That is, the smaller the proportion of insiders, the greater the use of this indicator. Managerial role

PAGE 68

62 performance and citizenship behavior are not significantly related to the measure, proportion of insiders. Table 3.16 Regression Coefficients Testing the Effects of the Proportion of Insiders on the Use of the Behavior-Based Measures Variables Proportion of Insiders Managerial Role Performance oi (see Table 3.4) CEO Leadership Behavior -.12* (see Table 3.5) Citizenship Behavior .05 (see Table 3.6) *p<.05, **p<.01 Hyxiothesis 12 Hypothesis 12 states that the larger the proportion of board members who are presently or formerly CEOs of other firms, the greater use of behavior-based measures. The regression coefficients between the proportion of outsiders who are past or present CEOs and the behavior-based measures are displayed in Table 3.17. The proportion of outside directors who are past or present CEOs of others firms is significantly related to CEO leadership behavior (b=.15, p<.05) That is, the higher the proportion of board members who are presently or formerly CEOs of other firms, the greater use of this behavior-based measure. Neither managerial role performance or citizenship behavior are significantly related to larger proportions of board members who are presently or formerly CEOs of other firms.

PAGE 69

63 Table 3.17 Regression Coefficients Testing the Effects of the Proportion of Outsiders Who Are Past/Present CEOs on the Use of the Behavior-Based Measures Variables %Outsiders Past/Present CEOs Managerial Role Performance (see Table 3.4) CEO Leadership Behavior (see Table 3.5) Citizenship Behavior (see Table 3.6) 02 15* 04 *p<.05, **p<.01 Hypothesis 13 Hypothesis 13 states that board pay will be positively related to the use of the behavior-based measures. The regression coefficients between board pay and the behaviorbased measures are displayed in Table 3.18. • \; Table 3.18 Regression Coefficients Testing the Effects of Board Pay on the Use of the Behavior-Based Measures Variables Board Pay Managerial Role Performance (see Table 3.4) CEO Leadership Behavior (see Table 3.5) Citizenship Behavior (see Table 3.6) 35** 32** .30** *p<.05, **p<.01 Board pay is negatively related to the use of the behavior-based measures. That is, board members with lower pay are more likely to evaluate the CEO with the behaviorbased measures; managerial role performance (b=-.35, p<.01), CEO leadership behavior (b=-.32, p<.01), and citizenship behavior (b=. 30 p<.01)

PAGE 70

64 Summary The results in this chapter show evidence that there is generally, a greater use of behavior-based measures when the CEO is influential. In management -controlled firms the use of behavior-based measures are dominant in the CEO evaluation. Specifically, isolating factors representing the three dimensions (managerial role performance, CEO leadership behavior, and citizenship behavior) there are relationships when the CEO holds the chairman position. In owner-controlled firms, the use of outcome-based performance indicators is used to a greater extent in CEO evaluation. Board pay seems to be an important predictor of the type of criteria used in CEO evaluation. When board members receive low compensation for their services, the greater the use of all measures representing behavior-based measures (managerial role performance, CEO leadership behavior, and citizenship behavior) in the CEO evaluation. There is also evidence that in owner-controlled firms, when board members receive low compensation for their services the greater the use of outcome-based performance criteria. These results, along with other findings will be discussed in the next chapter

PAGE 71

; • CHAPTER 4 ; • • DISCUSSION The nature of CEO compensation and how boards determine CEO pay packages has received increased attention over the last decade. The questions surround the seemingly excessive salaries of CEO's at their respective firms. The primary interest of this study is to determine how boards of directors justify CEO compensation decisions by examining the performance appraisal of the CEO by the board. This study assesses the nature of the criteria used in the board of director' evaluation of CEO performance, specifically, the use of outcome-based and behavior-based measures in CEO evaluation. It is clear that the use of behavior and outcome-based measures are different in management -controlled versus owner-controlled firms. As speculated in Chapter One, based on the theory of managerial capitalism and agency theory, the nature of the criteria revolves around an issue of control over the board of directors or the firm. The CEO is afforded considerable discretion over the evaluation process when they are in control For example, the results show that there is a greater use of outcome-based measures (essentially measures of firm financial performance) in owner-controlled firms than in 65

PAGE 72

66 management -control led firms. They are also more frequently used in large owner-controlled firms and when there is a larger proportion of outside directors who are past or present CEOs of other firms. There are some instances when behavior-based measures are used in owner-controlled firms. One instance is when there is a larger proportion of outside directors who are past or present CEOs of other firms. The other is when the members of the board are less well paid. In management -controlled firms, however, the tendency is to use behavior-based measures such as leadership, managerial role performance and CEO citizenship behavior in CEO evaluation. A major exception, for example, when outcome-based measures are rated as more important than behavior-based measures in management-controlled firms, is when firm financial performance is high. There also are some facets of corporate governance mechanisms that are related to the use of behavior-based criteria for CEO evaluation. This occurs when board members are less well paid, when there is a larger proportion of outside directors, when the CEO is chairman of the board, and when more members of the board are currently or have been CEOs of other firms. Some other conventional measures of CEO power were not related to the use of either behavior-based or outcomebased CEO evaluation criteria. Specifically, CEO tenure and CEO influence over board meetings or meeting agenda were found to have no effects.

PAGE 73

67 Theoretical Implications The agency theory model is based on the premise that, in the absence of an optimal sharing of risk between the agent and the principal, the principal can design a contract to control the agent by aligning the incentives of both parties or by indirectly or directly monitoring the agent's actions or behavior (Jensen & Meckling, 1976) Monitoring and incentive alignment are necessary because both the principal and agent, being rational actors, will seek to maximize their own utility. This can be a problem because of information asymmetry, moral hazard and adverse selection issues, which put the manager in a position to act opportunistically and reduce his or her risk at the expense of the equity holder. Thus, agency theory assumes that the CEO will be evaluated on criteria based on the objectives of the equity holders and that the relationship of the CEO and owner will be based on a contract that provides the appropriate level of monitoring and incentive alignment (Jensen & Meckling, 1976) These criteria, given the difficulties of creating optimal risk sharing contracts, would most likely be in the form of outcome-based measures that tie compensation to the performance of the firm and are easily observable. The use of outcome-based measures in CEO evaluation reduces the conflict of interests that exists between the owner and the CEO (Ouchi, 1979) On the other hand, behavior-based measures do not align the pay of the CEO with firm

PAGE 74

68 performance. According to agency theorists, these criteria would only be optimal in situations in which the behavior of the CEO is easily monitored and is known to lead to outcomes desired by the equity holders (Demski & Feltham, 1978) While agency theory is built on the idea that contracts can be designed to minimize managerial opportunism, managerial capitalism theory posits that because the equity holdings of large firms, particularly, are widely dispersed, stockholders are not able to exert enough control over managers, putting managers in a position to control the firm (Berle & Means, 1932) Therefore, when managers control the firm, they will be able to design compensation contracts that minimize their compensation risk and maximize their self-interest (Harris & Raviv, 1979) There is ample evidence to support this idea. For example, managers in management controlled firms may (1) choose accounting methods that state results more favorable to them, (2) make investment decisions that are less optimal for owners but which minimize their downside risk, (3) undertake acquisitions and mergers that transfer higher agency costs to owners, (4) use internal political strategies to block control mechanisms intended to check their opportunism, and (5) use organizational resources to block the disciplining mechanism of the market (Tosi, et al 1999) However, this evidence is plagued by an important limitation. It treats the firm as a "black box" in that it is based on archival sources of financial and

PAGE 75

69 demographic data from which behavioral and process inferences are drawn. As Jensen and Murphy (1990) suggested, it does not actually focus on the internal processes and political forces that are at play in the monitoring process. The research reported in this study is part of a more recent stream of research that seeks to understand the political forces at play in the implementation of the agency contract (Tosi & Gomez-Mejia, 1989; 1994; Tosi, Katz & Gomez-Mejia, 1997; Antle & Smith, 1986). The general conclusion of this research is that in management-controlled firms, the CEO is an important actor in the actual process of determining his or her compensation and other actors are less influential, while the opposite is the case in ownercontrolled firms. This study extends that line of research by showing how CEOs can minimize compensation risk by managing the nature of the criteria used in evaluating their performance. Effective contracts, from the principal's point of view, can use behavior-based criteria or outcome-based criteria. However, in both instances it is necessary that the behavior and the outcomes be clearly and precisely known, defined and observable. This can be a problem for at least two reasons. First, the CEO's task includes a wide variety of components, ranging from making choices that would be reflected in the firm's financial performance to taking a leadership role in influencing and motivating others in the firm. Second, the

PAGE 76

70 precise nature of the behavior-outcome relationships of these different tasks, individually and in combination, is difficult to exactly specify. Given that these relationships are not clear, it is likely that the use of outcome-based measures transfer more performance risk to a CEO than behavior-based criteria. The complexity of the managerial job and the lack of precise relationships between behaviors and outcomes leaves room for discretion and interpretation, particularly with respect to whether behavior-based or outcome-based criteria will be used in the evaluation process. The way that this discretion and interpretation of specific CEO evaluation criteria is shown in the results discussed earlier. The CEO evaluation in the owner-controlled firm follows the prediction of agency theory in that owners and managers can design a contract that tends to align the interests of both. The evaluation of CEOs in owner-controlled firms is more likely to be based on outcome-based criteria; those derived from indicators of the firm's economic performance. There are different results for the management controlled firms. In these firms, it is already known that the CEO is the dominant actor in the process of setting his or her pay (Tosi & Gomez-Mejia, 1989) Here it is shown that one manifestation of this dominance is the use of less risky behavior-based criteria. The use of these behavior criteria in management controlled firms is also more likely as the proportion of outsider directors with experience as

PAGE 77

71 CEOs increase. Since outside board members are not able to observe the CEO in action, it is more difficult for them to estimate the role of CEO behavior relative to firm performance. They could, then, simply base their evaluation on their own biases and experiences in the top managerial position in a firm. This result is consistent with the social comparison arguments advanced by O'Reilly and his colleagues (O'Reilly, Main & Crystal, 1988) There are some cases, however, in which outcome -based measures are more likely in management-controlled firms. When firm performance is high, the manager can minimize risk by creating a compensation contract based on outcome-based criteria. Outcome-based criteria tie compensation to firm performance. CEOs in high performing management -controlled firms appear to prefer outcome-based criteria in that these would generate a higher compensation package. In this instance, the CEO of the high performing management controlled firm uses more traditional economically justified measures of firm performance and avoids potential criticism for the use of more subjective behavior based indicators. Research shows that the level of monitoring is higher in owner-controlled firms than in management -control led firms, and compensation tends to be in the form of incentive based and more closely tied to maximizing shareholder value (Tosi & Gomez-Mejia, 1989). The results discussed above support this research in that governance activities in management -control led firms are weaker than in owner-

PAGE 78

72 controlled firms as reflected by the type of criteria used in the CEO evaluation. In owner-controlled firms the use of outcome-based measures is the dominant criteria used in the CEO evaluation. Outcome-based measures tie compensation to firm performance and align the interests of the CEO and shareholder Overall, the results show that the type of criteria used in the CEO evaluation are significantly different for management -controlled versus owner-controlled firms. It also provides some evidence of how these criteria are related to certain elements of the governance process. The results provide some insight into some of the mechanisms of corporate governance that affect the choice of CEO criteria. For example, powerful CEOs are allowed considerable discretion in the selection of boards of directors and operation of the board. This might lead them to select board members who are prior or present CEOs of other firms. The results here show that this type of board member is more likely to emphasize the use of behavior-based measures. They would be more likely to share common interests with the CEO they are evaluating and this may bias the evaluation process. •' Powerful CEOs can also influence the operation of the board through the control of board pay. While board pay has been shown to be positively related to CEO pay (Boyd, 1994; O'Reilly, Main & Crystal, 1988), in this study it was found to have a negative relationship to the use of behavioral-

PAGE 79

73 based measures, i.e., the lower the board pay, the greater the use of behavioral -based CEO criteria. One explanation for this could be that these lower paid boards are less vigilant in their evaluations of CEO performance. Board members may not feel committed to representing the owners best interest (i.e. monitoring the CEO) when board membership is not ostensibly valued, as exemplified by the remuneration of board members. There is a substantial amount of evidence to support the position that individuals who feel inequitably paid, are less motivated to complete a task (i.e. Pritchard, Dunnette & Gorgenson, 1972). This issue needs further exploration. A powerful CEO can also control the composition of the board. In this study, the proportion of insiders on the board of directors is hypothesized to have a positive relationship with the behavior-based measures. This hypothesis is not supported in that the proportion of insiders' measure is negatively related to the use of the behavior-based measure, CEO leadership behavior. Other research shows a positive relationship between the proportion of insiders and CEO compensation (Sridharan, 1996; Williamson, 1963) These findings imply a positive relationship with the proportion of outsiders and behavior-based measures. That is, the greater the proportion of outsiders on a board, the greater the use of behavior-based measures. So, while these results are inconsistent with the research on the behavior

PAGE 80

74 of insiders, they are nonetheless consistent with prior work on the role of the outsider (Westphal & Zajac, 1995; O'Reilly, Main & Crystal, 1988). Outside members are often nominated by a CEO with whom they share some affiliation outside the firm. Therefore, the decisions these board members make may be in the best interests of their respective CEO, rather than the shareholder. Summary This study draws on two theories to make suppositions about the CEO performance appraisal. These theories have been previously applied to the study of activities within the firm, and specifically the study of CEO compensation. The findings of this study extend the applications of managerial capitalism and agency theory. Additionally, the results are consistent with these theories prior findings: 1) Managers will behave opportunistically when they control the firm, and 2) the structure of the board of directors prohibits the autonomous role of the board. Limitations While the results of this research are consistent with theory and previous research, there are some necessary caveats which should be stated. First, this study is exploratory in being the first empirical examination of outcome-based and behavior-based criteria used by the board in the CEO evaluation process. Secondly, the use of a single source, i.e., the chief compensation officer in the

PAGE 81

75 firms, to collect the data limits its interpretation, as common method variance could be a problem. Third, other factors should be investigated for their effects on CEO evaluation criteria. The dependent variables used are not an exhaustive list of the measures that could have been collected. The outcome-based measures used were traditional and conventional indicators used in previous studies. The results here were also consistent with results that have used similar outcome-based measures as a dependent variable. However, the behavior-based measures were not drawn from prior empirical work, but chosen based on the implications of the theory. Since some of the measures had little predictive power with the independent variables this suggests the need for more work on these measures. Additionally, there are other individual differences that might be more important in the CEO evaluation than the ones that were assessed in this study (i.e. personality variables) Conclusion This dissertation explores the performance evaluation process of the CEO. It contributes to this area in that it illuminates the types of criteria used in CEO evaluation. It also adds to the understanding of the board of director's role in the evaluation process. Boards of directors have come under increased scrutiny by the public and media due to the excessive salaries awarded CEOs in many of America's leading corporations. The

PAGE 82

76 questions have been, "why so much and how does the board make these decisions?" This study answers how the board makes these decisions by understanding the type of measures board members use in the evaluation of CEO performance. It is clear that boards of directors use different criteria across firms in the CEO evaluation. The board must justify compensation decisions to not only the public (namely the SEC), but also to shareholders. The CEO performance evaluation is part of the justification used in the derivation of CEO compensation packages. Boards must decide whether to use outcome or behavior-based criteria in the CEO evaluation. The logical criteria to use are those that align the interests of the CEO and owner. These outcome-based measures are easily verifiable and justifiable by the board. Using behavior-based measures are problematic in that they are abstract and not easily verifiable. The dominant use of these measures in firms where the CEO is influential suggests that board members are not fulfilling their role as "stewards" of the firm. Board of directors cannot fulfill their role when there is a lack of separation of their role and the CEO. It is especially problematic to separate the role of the board and CEO when the CEO is often the chairman of the board. Powerful CEOs can dictate the type of criteria used in the evaluation of their performance. Behavior-based measures are preferred over outcome-based as they minimize the compensation risk to the CEO.

PAGE 83

77 Separation of the role of the board from the CEO is crucial for a board fulfilling its role as an effective governance device for its shareholders. However, with current board membership of many firms consisting of present or former CEOs of other firms and also board members who are nominated and selected by the CEO, it is uncertain whether a board can ever be truly autonomous. Moving beyond board membership issues, there is also a problem with the decision making process of the board. In many of the firms studied for this dissertation, the CEO had significant influence over the agenda and running of board meetings, as well as influence in important board decisions. In order for the board to fulfill their role as guardians of the shareholders interests, they need to truly govern themselves without the intervention of the CEO. Therefore, boards of directors should minimize involvement of the CEO in board decisions, and limit the CEO role to one of purely informational. Future studies should examine the role of the board more closely, in that we do not know whether board members feel that their role is being undermined by the CEO. Understanding the motivation of board members in the undertaking of their duties can also reveal their preferences. An examination of board compensation could provide insights into whether board pay plays a role in board member's value of their job. Further, understanding the types of decisions board members make when their

PAGE 84

78 compensation consists of salary versus stock options or both could contribute to an understanding of the board decisions. In conclusion, more empirical work is needed in order to understand why the gap between the board and CEO seems to be so small

PAGE 85

REFERENCES Ackoff, R.L. (1999). Transformational leadership. Strategy and Leadership 27 20-25. Aigner, D. (1992) Fixing executive compensation. Journal of Commerce and Commercial, 3 94 6A. Amihud, Y., and Lev, B. (1981). Risk reduction as managerial motive for conglomerate mergers. The Bell Journal of Economics 12., 605-617. Antle, R., and Smith, A. (1986). An empirical investigation of the relative performance evaluation of corporate executives. Journal of Accounting Research 24 1-39. Avolio, B.J. and Bass, B.M. (1988) Transformational leadership, charisma, and beyond. Pp. 29-49 in J.G. Hunt, B.R. Baliga, H.P. Dachler and C.A. Schriesheim (Eds.), Emerging leadership vistas Lexington, MA: Lexington Books Baimon, S. (1990). Agency research in managerial accounting: A second look. Accounting, Organizations H and Societv 15, 341-371. Baker, G.P., Jensen, M.C., and Murphy (1988). Compensation and incentives: Practice vs. theory. Journal of Finance, 43, 593-616. Bass, B.M. (1985) Leadership and performance beyond expectations New York: Free Press. Bass, B.M. (1990) Bass & Stogdill's handbook of leadership: Theory, research and managerial applications (3'''^ ed.) New York: Free Press. Bass, B.M. and Avolio, B.J. (1990) Developing transformational leadership: 1992 and beyond. Journal of European Industrial Training 14 21-28. Bateman, T.S., and Organ, D.W. (1983). Job satisfaction and the good soldier: The relationship between affect and employee "citizenship." Academy of Management Journal 26 587-595. 7 9

PAGE 86

80 Baumol W. (1967) Business, behavior, value and growth New York: Harcourt Brace and World. Baysinger, B., and Hoskisson, R.E. (1990). The composition of the board of directors and strategic control : Effects of corporate strategy. Academy of Management Review 15, 72-87. Berle, A. A. and Means, G.C. (1932) The modern corporation and private property New York: MacMillan. Boyd, B.K. (1994). Board control and CEO compensation. Strategic Management Journal 15 335-344. Bushman, R.M., Indjejikian, R.J., and Smith, A. (1996). CEO compensation: The role of individual performance evaluation. The Journal of Accounting and Economics 21, 161-93. Callen, J.H., Jr., and Jett, C.C. (1993, April). Critical skills and the CEO. Chief Executive pp. 34-38. Cardy, R.L., and Dobbins, G.H. (1986). Affect and appraisal accuracy: liking as an integral dimension in evaluating performance. Journal of Applied Psychology 71, 672-678. Chitayat, G. (1985) Working relationships between the chairman of the boards of directors and the CEO. Management International Review 25 65-70. Crystal, G.S. (1991) Why CEO compensation is so high. California Management Review 34 9-29. Demski, J., and Feltham, G. (1978). Economic incentives in budgetary control systems. Accounting Review 53., 336359 DeNisi, A.S., and Stevens, G.E. (1981). Profiles of performance, performance evaluations, and personnel decisions. Academy of Management Journal 24 592-602. Dobbins, G.H., and Russell, J.M. (1986). The biasing effects of subordinate likeableness on leaders' responses to poor performers: A laboratory and field study. Personnel Psychology 39, 759-77. Ehrbar, A. (1999) Using EVA to measure performance and assess strategy. Strategy and Leadership 27 20-24.

PAGE 87

81 Eisenhardt, K. (1989). Agency theory: An assessment and review. Academy of Management Review 14 57-74. Ettorre, B. (1992). Evolution inside the boardrooms. Management Review 81 10-16. Fama, E.F. (1980) Agency problems and the theory of the firm. Journal of Political Economy 88 288-307. Fama, E.F., and Jensen, M.L. (1983a) Agency problems and residual claims. Journal of Law and Economics 26. 327-349 Fama, E.F., and Jensen, M.L. (1983b). Separation of ownership and control. Journal of Law and Economics 26, 301-325. FeldmanSummers S., and Kiesler, S.B. (1974). Those who are number two try harder: The effect of sex on the attribution of causality. Journal of Personality and Social Psychology 30 846-855. Festinger, L. (1954) A theory of social comparison processes. Human Relations 2/ 117-140. Fiedler, F.E. (1967) A theory of leadership effectiveness New York: McGraw-Hill. Fierman, J. (1990, March). The people who set the CEO's pay. Fortune pp. 58-62. Finkelstein, S., and Hambrick, D.C. (1989). Chief executive compensation: A study of the intersection of markets and political processes. Strategic Management Journal 10, 121-134. Fosberg, R.H. (1999) Leadership structure and CEO compensation. American Business Review 17 50-56. Fredrickson, J.C., Hambrick, D.C, and Baumrin, S. (1988). A model of CEO dismissal. Academy of Management Review 13., 255-270. Gaver, J.J., and Gaver, K.M. (1998). The relation between nonrecurring accounting transactions and CEO cash compensation. Accounting Review 73., 235-253. Gibbons, R., and Murphy, K.M. (1990). Relative performance evaluations for chief executive officers. Industrial and Labor Relations Review 43 (special issue) 30s51s

PAGE 88

82 Goldstein, M.L. (1985, January) Grading the CEO. Industry Week 224 48-49. Gomez-Mejia, L.R., Tosi, H.L., and Hinkin, T. (1987). Managerial control, performance, and executive compensation. Academy of Management Journal 30 5170 Hall, B.J., and Liebman, J.B. (1998). Are CEOs really paid like bureaucrats? Quarterly Journal of Economics 113 653-691. Hallock, K.F. (1998) Layoffs, top executive pay, and firm performance. American Economic Review 88., 711-723. Harris, M. and Raviv, A. (1979). Some results on incentive contracts with imperfect information. Journal of Economic Theory 20., 231-259. ,. Hambrick, D.C., and Finkelstein, S. (1995). The effects of ownership structure on conditions at the top: The case of CEO pay raises. Strategic Management Journal 16, 175-193 Hill, C.W.L., and Phan, P. (1991). CEO tenure as a determinant of CEO pay. Academy of Management Journal 34/ 707-807. Hiltzik, M.A. (1996, May 10) Taking stock of CEO pay. Los Angeles Times p Al Hoskisson, R.E., Johnson, R.E., andMoesel, D.D. (1994). Corporate divestiture intensity in restructuring firms: Effects of governance, strategy, and performance. Academy of Management Journal 37., 1207-1252. House, R.J. (1971) A path goal theory of leader effectiveness. Administrative Science Quarterly 16, 321-328 Hunt, H.G., III, and Hogler, R.L. (1990). Agency theory as ideology: A comparative analysis based on critical legal theory and radical accounting. Accounting. Organizations and Society 15, 437-455. Ittner, CD., Larcker, D.F., and Rajan, M.V. (1997). The choice of performance measures in annual bonus contracts. The Accounting Review. 72, 231-255.

PAGE 89

83 Janakiraman, S.N., Lambert, R.A., and Larcker, D.F. (1992). An empirical investigation of the relative performance evaluation hypothesis. Journal of Accounting Research 30, 53-70. Jensen, M., and Heckling, W.H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3.f 305-360. Kaplan, R.S., and Norton, D.P. (1992). The balanced scorecard Measures that drive performance. Harvard Business Review 70., 71-80. Kerr, J.L., and Kren, L. (1992). Effect of relative decision monitoring on chief executive officer compensation. Academy of Management Journal 35 370397 Kirchner, W.K., and Reisberg, D.J. (1962). Differences between better and less effective supervisors in appraisal of subordinates. Personnel Psychology 15 295-302 Kristie, J. (1998) Is there no decency? Directors and Boards 23., 6. Landy, F.J., and Farr, J.L. (1983). The measurement of work performance New York: Academic Press. Larner, R. (1966) Ownership and control in the 200 largest nonfinancial corporations, 1929 and 1963. The American Economic Review 56, 778-787. Lau, A.W., and Pavett, CM. (1980). The nature of managerial work: A comparison of public and private sector managers. Group and Organization Studies 5, 453-466. Lazear, E., and Rosen, S. (1981). Rank-order tournaments as optimum labor contracts. Journal of Political Economy 89, 841-864. Lear, R.W. (1998, September) Rethinking the compensation committee. Chief Executive p. 16. Leonard, B. (1994, April) CEO compensation packages tied to performance. HRMagazine pp. 51-53.

PAGE 90

84 Li, J. (1994). Ownership structure and board composition: A mult i -country test of agency theory predictions. Managerial and Decision Economics 15 359-368. Logue, D.E., and Steward, J.K. (1997). Handbook of modern finance Boston: WG&L/RIA Group. Lorsch, J.W. and Maclver, E. (1989). Pawns or potentates: The reality of America's corporate boards Boston: Harvard Business School Press. MacKenzie, S.B., Podsakoff, P.M., and Fetter, R. (1991). Organizational citizenship behavior and objective productivity as determinants of managerial evaluations of salespersons' performance. Organizational Behavior and Human Decision Processes 50, 123-150. Mallette, P., and Fowler, K.L. (1992). Effects of board composition and stock ownership on the adoption of poison pills. Academy of Management Journal 35 10101035 Marino, S.F. (1998, April). Chief executives are underpaid. Industry Week 247 p. 22. Marris, R. (1964) The economic theory of managerial capitalism New York: Free Press. McEachern, W.A. (1975) Managerial control and performance New York: D.C. Health. McNatt, R., and Light, L. (1998, March). Why the CEO may be worth $100 million. Business Week p. 6. Mehran, H. (1995) Executive compensation structure, ownership, and firm performance. Journal of Financial Economics, 28., 163-184. Meredith, D.R. (1992, Jan-Feb) Bold thinking. Chief Executive pp 4 6-49. Mintzberg, H. (1973) The nature of managerial work Englewood Cliffs, N J : Prentice-Hall, Inc. Morck, R., Shleifer, A., and Vishny, R.W. (1988). Management ownership and market evaluation. Journal of Financial Economics 20 293-315.

PAGE 91

85 National Association of Corporate Directors (1994) NACD Blue Ribbon commission report on performance evaluation of CEOs, boards, and directors Washington, D.C.: NACD. Nadler, P.S. (1986, December). What makes a CEO effective. Bankers Monthly pp. 4-5. Nett, R. (1993, April). What do Japanese CEOs really make? Business Week pp. 60-61. O'Reilly, C.A., III, Main, B.C., and Crystal, G.S. (1988), CEO compensation as tournament and social comparison: A tale of two theories Administrative Science Quarterly 33 257-275. Organ, D.W. (1997). Organzational citizenship behavior: Its construct clean-up time. Human Performance 10 85-97 Organ, D.W. (1988) Organizational citizenship behavior: The good soldier syndrome Lexington, MA: Lexington. Ouchi, W. (1979) A conceptual framework for the design of organizational control mechanisms. Management Science 25, 833-848. Palmer, J. (1973) The prof it -performance effects of the separation of ownership from control in large U.S. corporations. Bell Journal of Economics and Management Science, 4, 293-303. Podsakoff, P.M., and MacKenzie, S.B. (1994). Organizational citizenship behavior and sales unit effectiveness. Journal of Marketing Research 31 351-363. Podsakoff, P.M., MacKenzie, S.B., Moorman, R.H., and Fetter, R. (1990) Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly 1, 107-142. Pritchard, R.D., Dunnette, M.D., and Gorgenson, D.O. (1972). Effects of perception of equity and inequity on worker performance and satisfaction. Journal of Applied Psychology 56 75-94. Rosen, S. (1990) Contracts and the market for executives. Cambridge, MA: Working paper for the National Bureau of Economic Research.

PAGE 92

86 Ross, S. (1973) The economic theory of agency: The principal's problem. American Economic Review 63., 134-139. Salomon, G.L., and Smith, E.D. (1979). Corporate control and managerial misrepresentation of firm performance. Bell Journal of Economics 10 319-328. Salancik, G.R., and Pfeffer, J. (1980). Effects of ownership and performance on executive tenure in U.S. J, corporations. Academy of Management Journal 23., 653664. Schaefer, S. (1998). The dependence of pay-performance sensitivity on the size of the firm. Review of Economics and Statistics 80 436-443. Seltzer, J., and Bass, B.M. (1990). Transformational leadership: Beyond initiation and consideration. Journal of Management 16 693-703. Singh, H. and Harianto, F. (1989) Top management tenure, corporate ownership structure, and the magnitude of golden parachutes. Strategic Management Journal 10 143-156 Smith, C.A., Organ, D.W., and Near, J. P. (1983). Organizational citizenship behavior: Its nature and antecedents. Journal of Applied Psychology 68, 653663 Sridharan, U.V. (1996, February) CEO influence and executive compensation. The Financial Review pp. 5167 Tobin, J.M. (1994) The squeeze on directors inside is out. Business Lawyer 49 1701-1760. Tosi, H.L., and Gomez-Mejia, L.R. (1989). The decoupling of CEO Pay and Performance: An agency theory perspective. Administrative Science Quarterly 34 169-189. Tosi, H.L., and Gomez-Mejia, L.R. (1994). CEO compensation monitoring and firm performance. Academy of Management Journal, 37, 1002-1016. Townley, P. (1993) Accountable but to whom? Across the Board 30, 11-12.

PAGE 93

87 Truskie, S.D. (1995). CEO performance evaluation: How America's leading companies review and evaluate the performance of their chief executive officers. Pittsburgh, PA: Management Science and Development. Vance, S.C. (1990) Inside or outside directors: Is there really a difference? Across the Board 27 15-17. Verespej M.A. (1994, September). Before it's too late. Industry Week p. 11. Wade, J., O'Reilly, C.A., and Chandratat, I. (1990). Golden parachutes: CEOs and the exercise of social influence. Adminstrative Science Quarterly 35 587-603. Waldman, D.A., Bass, B.M., and Einstein, W.O. (1987). Leadership and outcomes of performance appraisal processes. Journal of Organizational Psychology 60 177-186 Wang, C. (1997) Incentives, CEO compensation, and shareholder wealth in a dynamic agency model Journal of Economic Theory 76, 72-105. Weiner, B. (1971) Perceiving the causes of success and failure. Morristown, NY: General Learning Press. Westphal, J.D. (1998). Board games: How CEOs adapt to increases in structural board independence from management. Administrative Science Quarterly 43., 511537 Westphal, J.D., and Zajac, E.J. (1997). Defections from the inner circle: Social exchange, reciprocity, and the diffusion of board independence in U.S. corporations. Administrative Science Quarterly 42 161-183. Westphal, J.D., and Zajac, E.J. (1995). Who shall govern? CEO/board power, demographic similarity, and new director selection. Administrative Science Quarterly 40, 60-83. Williamson, O.P. (1963). Managerial discretion and business behavior. American Economic Review 53, 1032-1057. Yermack, D. (1998). Companies' modest claims about the value of CEO stock option awards. Review of Quantitative Finance and Accounting, 10, 207-226.

PAGE 94

88 Zajac, E.J., and Westphal, J.D. (1996a). Director reputation, CEOboard power, and the dynamics of board interlocks. Administrative Science Quarterly 4 1 507529 Zajac, E.J., and Westphal, J.D. (1996b). Who shall succeed? How CEO/board preferences and power affect the choice of new CEOs. Academy of Management Journal 39 64-90.

PAGE 95

APPENDIX A MAIL SURVEY MATERIALS This appendix contains all material used for collection of data from cooperatives via mail 89

PAGE 96

90 September 1997 Henry L. Tosi Vice President, Executive Compensation Rosemary Tosi and Associates P.O. Box 69538 Gainesville, FL 32608 Dear Mr Tosi In recent years, the popular press has devoted considerable attention to the compensation of Chief Executive Officers in America's largest firms. As you know, the controversy involves both the magnitude of CEO compensation as well as the bases for these pay packages. The Management Center at the University of Florida has been interested in these questions since 1987 and we have conducted a number of studies to examine them. As we look at our work and that of others, we conclude that there is a significant void in this literature. It seems that there has not been a systematic study of the criteria used by boards of directors to determine CEO pay. We are conducting a study at the Management Center to address the question of CEO pay criteria. The enclosed questionnaire forms the basis of this research. We would like your help in this project, asking that you take a few minutes to complete the enclosed survey, then returning it to us. You were selected as a recipient of this questionnaire from among those members of the American Compensation Association who have indicated some significant level of involvement with top executive compensation, and hence one who should be a knowledgeable informant about CEO pay and how it is structured in your firm. We ask only that you respond to each question to the best of your knowledge, nothing more. Please complete the questionnaire and return it in the enclosed envelope. The questionnaire should take no longer than 2 0 minutes. Examining it, you will note that it provides for anonymous responses. We have done this to ensure confidentiality for those firms and those executives willing to help us. We have also included a postcard that you may send to us at your convenience if you are interested in a summary of our results. The use of the separate postcard again, ensures your anonymity. We hope you can help us in this research. We know that you have a busy schedule, but your participation is essential to this study. Further, this topic is very interesting and

PAGE 97

91 important to all executive compensation professionals. If you have any questions, comments, or need any assistance in completing the questionnaire please contact Paula Silva, Project Director, at (352) 392-3737. Sincerely, Henry L. Tosi Paula Silva McGriff Professor of Management Graduate Minority Fellow

PAGE 98

92 SECTION 1 FIRM INFORMATION How long have you been Chief Compensation Officer wilh your firm7_ years 2. Please rale your firm 's financial performance relative to others in your industry Top 20% Second 20% Third 20% Fourth 20% Lowest 20% 3. Please rate your firm 's sales relative to other firms in your industry Top 20% Second 20% Third 20% Fourth 20% Lowest 20% 4. Are your firm's main products durable goods or nondurable goods O 5. Please indicate your firm's Standard Industrial Classification Agriculture, Forestry & Fishing Mining Construction Transportation, Communications, Electric, Gas & Sanitary Services Wholesale Trade Finance, Insurance & Real Estate Services Manufacturing Approximately how many people are employed by your firml_ How long has the CEO held this position with your firm? years Is there any single individual or institution outside the firm that owns 5 percent or more of the company stock? Yes DNo 8a. If you checked "yes" to question 8, please check which of the following owns 5 percent or more of the company slock, (please check all that apply) Insurance Company Bank Nonbank Trust Mutual Fund Endowment Pension Fund Foundation Individual Other (please specify)_ Is there any individual manager within the firm (e.g., the CEO) who owns 5 percent or more of the company stock? Yes No SECTION 2 BOARD INFORMATION limes each year /. Hon many times does your Board of Directors meet per year? 2. What is the average cash compensation for board membership per year? 2a. What is the estimated average value of other pay (stock options, benefits, etc..) 3. How many directors are also executives in the firm? How many directors are from outside the firm? 4. Is the current CEO also chairman of the boards Yes No If you checked "no" to question 4, is the Chairman presently or formerly a CEO of another firm? Yes No If you checked "no" to question 4, is the Chairman formerly a CEO of your firm? Yes No 5. How many of the outside directors are present or past CEOs of other firms? 6. How much influence does the CEO have in reviewing and approving the types of information received by board members? (please circle the best response) l=None 4=Moderate 5 6 7=Very High Please respond to questions 7-9 by rating the influence of different members of the board at board meetings. Please use the following scale. l=None 34=Moderate ;7=Very High 7. How much influence does each of the following have in setting the agenda at board meetings? Chief Executive Officer Chairman of the Board

PAGE 99

93 Board members, in general A coalition of one or two board members 5. How much influence does each of the following have in the running of board meetings? Chief Executive Officer Chairman of the Board Board members, in general A coalition of one or two board members 9. How much influence does each of the following have in important board decisions'! Chief Executive Officer Chairman of the Board Board members, in general A coalition of one or two board members 10. Please rate the extent to which board members are kept up to date on company activities throughout the year (please circle the best response) l=NotAtAll 2 3 4=UsualIy 5 6 7=Always //. Does your board of directors have a separate compensation committee that determines the CEO compensation package? Yes No J J a. If you answered "yes" to question II, is the CEO a member of the compensation committee? Yes No / 2. Does your board employ an independent compensation consultant? Yes No SECTION 3 CEO PERFORMANCE EVALUATION QUESTIONS Using the following scale please respond to the following question. l=None : ; 2 ; ; : ; : 3: : 4=Moderate 5 6 7=Very High Indicate the level of involvement of each of the following in the CEO evaluation process. Board of Directors Compensation committee Outside participants or groups Outside consultant Other, please specify and rate: Please rate the extent to which performance evaluations of the CEO by these groups remain anonymous, so that individual evaluations cannot be identified. Please circle the best response (1-no anonymity 7-complete anonymity). l=No anonymity 2 3 4=moderate 5 6 7=Very high 3. To the best of your knowledge, please rate the relative importance of each of the following quantitative factors in the CEO performance evaluation. Please use the following scale. l=Not At A!l Important : 2 : : : 3 4=Moderate Importance 5 6 7=Extremely Important ROA Increasing Price of the Stock ROE Profitability ROI Eamings Per Share Sales Growth EV A/Economic Profit Other (please specify and rate) SECTION 4 QUALITATIVE CRITERIA To the best of your knowledge please rate the relative importance of the following factors in the CEO performance evaluation Please circle the best response using the following scale. l=Not At All Important 2 3 4=Moderate Importance 5 6 7=Extremely Important The extent to which the CEO: Not at all Moderate Extremely Important Important 1 Acts cheerful 12 3

PAGE 100

94 2. Taps into the grapevine to keep abreast of what is going on in the company 1 2 3 4 5 6 7 3. Is available to "outsiders" (such as consumers, suppliers, the public) who want to go to "the the person in charge" 1 2 3 4 5 6 7 4. Trains or helps others to perform their jobs better 1 2 3 4 5 6 7 5. "Keeps up" with developments in the company 1 2 3 4 5 6 7 6. Keeps up with market changes and trends that might have an impact on the organization / 7. Expects themselves and others to conscientiously follow company regulations and procedures / 2 3 8. Surveys the organization to identify ways to improve performance or cut costs 9. Exhibits punctuality in arriving at meetings 10. Integrates subordinates' goals (for example, career goals, work preferences) with the organization's work requirements 1 1 Attends functions that are not required, but that help the company image 12. Willingly risks disapproval in order to express beliefs about what's best for the company 13. Keeps professional colleagues informed about the organization 14. Takes steps to try to prevent problems with other executive members, board members and/or other employees in the company 2 3 4 4 4 4 4 4 4 5 6 7 5 6 1 2 3 4 5 6 I 2 3 7 7 7 7 7 7 l=Not At All Important 2 3 4=Moderate Importance 7=Extremely Important 1 5. Entertains clients on a regular basis 16. Cooperates well with others in the organization 17. Negotiates with groups outside the organization for necessary materials, support, commitment, etc... 1 8. Attends to staffing requirements in different units, such as hiring, firing, promoting, and recruiting 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 9. Provides guidance and direction to company employees 20. Makes innovative suggestions to improve the overall quality of the company 2 1 Keeps members of the organization informed of relevant information through meetings, conversations, and written information 22. Develops new contacts by personally answering requests for information 23 Takes immediate action in response to a crisis or "fire drill" 24. Stays tuned to what is going on in competitor organizations 25. Focuses on the positive side of a situation rather than the negative side of it To the best of your knowledge please rate the relative importance of the following factors in the CEO performance evaluation Please circle the best response using the following scale. 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 [:[: 3=Not At All Important: i : J : |2; : J : ; : : 3: ; :4=Moderate Importance'' : -5 7=Extremely Important Not at all Important Moderate Extremely Important 26. Willingly gives of his/her time to help others 27. Passes along unofficial information 28. Joins boards, organizations, clubs, or does public service work that might provide useful, work-related contracts 29. Passes on the desires and preferences of top management to others in the firm 30. Keeps executive members, consumers, or other important groups informed about the organization's activities and capabilities

PAGE 101

95 31. Provides employees with new ways of looking at things / 2 3 4 5 6 7 32. Leads by "doing" rather than simply by telling / 2 3 4 5 6 7 33, Evaluates the outcomes of internal improvement projects / 2 3 4 5 6 7 34. Returns phone calls and responds to other messages and requests for information promptly I 2 3 4 5 6 7 35. Plans and develops improvements in work flow or work methods / 2 3 4 5 6 7 36. Programs work for different units within the organization (what is to be done, when and how) / 2 3 4 5 6 7 37. Resolves disputes between subordinates or work groups 1 2 3 4 5 6 7 38. Insists on only the best performance 1 2 3 4 5 6 7 39. Allocates resources (personnel, money, material) among units within the organization / 2 3 4 5 6 7 40. Determines the long-range plans and priorities of the organization / 2 3 4 5 6 7 A 1 Is willing to be the official spokesperson of the company at gatherings outside of the organization / 2 3 4 5 6 7 42, Resolves work flow problems between units / 2 3 4 5 6 7 43. Seeks new opportunities for the organization 1 2 3 4 5 6 7 44. Inspires others with his/her plans for the organization I 2 3 4 5 6 7 45. Personally answers letters or inquiries about the organization / 2 3 4 5 6 7 46. Attends outside conferences or meetings / 2 3 4 5 6 7 47. Encourages employees to be "team players" / 2 3 4 5 6 7 48. Personally handles complaints from customers and subordinates / 2 3 4 5 6 7 49. Shows respect for individuals' personal feelings / 2 3 4 5 6 7 50. "Touches base" with others before initiating actions that might affect them / 2 3 4 5 6 7 SKCTIO.> 5 QUAiN 1 1 1 AlIVE VS. QUAMTATIViS CRlTEiOA 1. Please rate the current relative importance of quantitative (objective) criteria or measures as compared to qualitative (subjective) criteria or measures in the determination of the CEO compensation package (please circle the best response) Qualitative Most Important About Equal Quantitative Most Important 1 2 3 4 5 6 7 2. How has the relative importance of using quantitative criteria as compared to qualitative criteria in the CEO performance evaluation changed over time? Quantitative criteria have become less important in the last 5 years No change in their importance in the last 5 years Quantitative criteria have become more important in the last 5 years

PAGE 102

96 Dear CEO Performance Evaluation Study Recipient: Two weeks ago we sent you a questionnaire on CEO performance evaluation. This important study on CEO evaluation criteria is being conducted by the Management Center at the University of Florida. If you have completed and returned the survey to us, please disregard this reminder. If not, we would sincerely appreciate it if you take the time to complete it now. If you would like to participate but need another copy of the survey please call Paula Silva, Project Director, at (352) 392-3737. Thank you for taking time to participate in this study. Henry Tosi McGriff Professor of Management Paula Silva Graduate Minority Fellow

PAGE 103

APPENDIX B TABLES OF FACTOR ANALYSIS FOR OUTCOME BASED MEASURE AND BEHAVIOR BASED MEASURES Table B.l Outcome Based Measures Factor Loadings Measure Factor 1 Factor 2 Factor 3 ROI 790 ROA 64 ROE 6 1 Earnings Per Share .81 Increasing Price of the Stock .79 EVA 5 9 Table B.2 Outcome Based Measures Communal it ies Measure ^^^^^ Initial ROI ~ 751 ROA 3 8 ROE .31 Earnings Per Share .41 Increasing Price of the Stock .42 EVA 15 97

PAGE 104

98 Table B.3 Managerial Role Behaviors: Liaison and Internal Coordination Factor Loadings Item Liaison Role Internal Coordination IvlJ J. c 1 Is available to outsiders who want to go to the person in 81 charge" 66 2 Joins boards, organizations, clubs or does public service work that might provides useful work-related contracts 3 Taps into the grapevine to keep abreast of what is going 66 on in the company51 4 Entertains clients on a regular basis .69 5 Resolves work flow problems between units .65 6 Programs work for different units within the organization 7 Resolves disputes between subordinates or work groups .5S 8 Plans and develops .52 improvements in work flow or work methods 9 Evaluates the outcomes of internal improvement projects > 40 Table B .4 Liaison and Internal Coordination Role Communal i ties Item Initial 1 Is available to outsiders who go to the person in charge" want to 55 2 Joins boards, organizations, clubs or 52 does public service work that might provides useful work-related contracts .41 3 Taps into the grapevine to keep abreast of what is going on in the company 32 4 Entertains clients on a regular basis 5 Resolves work flow problems between .43 units 6 Programs work for different units .45 within the organization 7 Resolves disputes between subordinates .36 or work groups 8 Plans and develops improvements in work .30 flow or work methods 9 Evaluates the outcomes of internal .28 improvement projects

PAGE 105

99 • • Table B. 5 Managerial Role Behaviors: Strategic and Internal Resource Management Role Factor Loadings Item Strategic Role Internal Resource Management Role 1 Determines the long-range plans and priorities of the organization 77 2 Keeps up with market changes and trends that might have an impact on the organization .68 3 Stays tuned to what is going on in competitor organizations .54 4 Attends to staffing requirements in different units such as hiring, firing, promoting, and recruiting .79 5 Negotiates with groups outside the organization for necessary materials, support commi tment etc .53 Table B.6 Strategic and Internal Resource Management Role Communal it ies Item Initial ~. Determines the long-range plans and TJs priorities of the organization 2. Keeps up with market changes and trends that .39 might have an impact on the organization • 3. Stays tuned to what is going on in .28 competitor organizations 4. Attends to staffing requirements in .41 different units such as hiring, firing, promoting, and recruiting 5. Negotiates with groups outside the .37 organization for necessary materials, support, commitment, etc..

PAGE 106

100 Table B.7 Leadership Behavior Factor Loadings Item Member Support Performance Focus 1 Shows respect for 83 individual's personal feelings 2 Encourages employees to be 52 team players" J fxT o V xae s enip oyee s wi L.n new 4 TTic'iQi~Q on on1\/ 1~Hf }*i^q1~ 71 v "P o vm a n Ci ^CI. J<.Jx iliclli.v_ c 5 Table B. 8 Leadership Behavior Communal it ies 1 1 em 1 Shows respect for individual's .26 personal feelings 2 Encourages employees to be team .22 players" 3 Provides employees with new ways .20 of looking at things 4 Insists on only the best .28 performance 5 Seeks new opportunities for the .31 organization

PAGE 107

101 Table B.9 Citizenship Behavior Factor Loadings Item Respect for Organizational Others Champion 1 Exhibits punctuality in arriving at 5 8 meetings 2 Willingly gives of his/her time to •58 help others 3 Acts cheerful 58 \' 4 Returns phone calls and responds to other messages and recjuests for information promptly 5 Attends functions that are not required, but that help the company image 6 Willingly risk disapproval in order to express beliefs about what's best for the company 7 Trains or helps others to perform .46 their jobs better Table B.IO Citizenship Behavior Communa 1 i t i e s Item Initial 1 Exhibits punctuality in arriving at .22 meetings 2 Willingly gives of his/her time to .30 help others 3 Acts cheerful .22 4 Returns phone calls and responds to other messages and requests for .20 information promptly 5 Attends functions that are not .22 required, but that help the company image 6 willingly risk disapproval in order to 17 express beliefs about what's best for the company 7 Trains or helps others to perform .22 their jobs better

PAGE 108

BIOGRAPHICAL SKETCH Having grown up in San Diego, California, Paula Silva received both her Bachelor of Science in business administration and Master of Business Administration at San Diego State University. Paula moved to Florida to begin her doctoral work in the Fall of 1994. Upon completion of her doctoral degree, Paula will be joining the faculty at the Department of Organizational Studies at the University of New Mexico. 102

PAGE 109

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of r^ilosophy Henry 2 j.l Tosi, dr/. Chair McGri ftj L/Family Professor of Management I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. irg5nTa~~G"! Maurer Vi] Professor of Management I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Judith aT Scully 3 Assistant Professor of Management I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope/^and quality, as a dissertation for the deg:ipee~s^of Doctor £xf i^f^ii^osophy ... Richard J. Lutz / Professor of Market i This dissertation was submitted to the Graduate Faculty of the Department of Management in the College of Business Administration and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. May, 2000 Dean, Graduate School


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EVFEM0422_WZ6MHX INGEST_TIME 2014-12-01T19:39:36Z PACKAGE AA00026451_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES