Power and corporate control

MISSING IMAGE

Material Information

Title:
Power and corporate control CEO succession in owner and manager controlled firms
Portion of title:
CEO succession in owner and manager controlled firms
Physical Description:
vii, 98 leaves : ; 29 cm.
Language:
English
Creator:
Banning, Kevin C
Publication Date:

Subjects

Subjects / Keywords:
Management thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Management -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1996.
Bibliography:
Includes bibliographical references (leaves 84-91).
Statement of Responsibility:
by Kevin C. Banning.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002204777
oclc - 36801779
notis - ALE4716
System ID:
AA00011776:00001

Full Text









POWER AND CORPORATE CONTROL:
CEO SUCCESSION IN OWNER AND MANAGER CONTROLLED FIRMS
















By

KEVIN C. BANNING


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

UNIVERSITY OF FLORIDA


1996





























This research is dedicated to my children, Robert and Sarah Banning. May you

always possess a sense of wonder about the world.












ACKNOWLEDGMENTS


Any endeavor such as a dissertation features a sizable supporting cast who make

the work possible. My thanks go to Henry Tosi, my committee chair, who showed me

that producing excellent writing is a craft which, though often laborious, is worth the

effort. I owe special thanks to John Hall who always gave sound advice. I am grateful

to Tim Taylor and Judy Scully for their time and patience in serving on my committee.

I also wish to thank Bob Thomas for his attention to my questions.

I extend my gratitude to my wife Allison, who served as a sounding board for my

concerns as well as being my most steadfast supporter. I am also grateful to my many

colleagues, both past and present, whose assistance was indispensable.













TABLE OF CONTENTS



ACKNOWLEDGMENTS .................................... iii


ABSTRACT


. .. . . .. vi


CHAPTER 1 INTRODUCTION AND LITERATURE REV
Theoretical Bases .. ..................


rIT


Managerial Capitalism and Agency Theory .
Power in Organizations .............
Sources of organizational power ....
Power in use .................
CEO Succession ..................
The Theoretical Model ...................
Corporate Control and the Board of Directors
CEO Power and the Board of Directors ....
Limitations of CEO Power ...........
Strategic Decisions .................
CEO Power and Strategic Decisions .......
Corporate Control and Strategy . .
Performance Implications ............
Summ ary .......................

CHAPTER 2 RESEARCH DESIGN ................
Sample and Data Collection ...............
Analytical Methods ......................
Variable Descriptions and Operationalization ......
Independent Variables ...............
Ownership concentration .........
Firm Size ..................
Firm performance measures .......
Duality ....................
O rigin ...... ... ..... .......
Disposition .................
Dependent Variables ...............
Board of director effects ........
Strategy effects ...............


..


,,







Performance effects ......................... .53
Summary .............................. 54


CHAPTER 3 RESULTS .
Board Structure .
Strategic Choices
Performance Effects
Summary .


CHAPTER 4 DISCUSSION .......................
Implications for CEO Succession ................
Implications for CEO Power ...................
CEO Power and the Board of Directors .......
Implications for managerial capitalism .....
Implications for managerial power .......
CEO Power and Strategy .................
Implications for managerial capitalism .... .
Implications for managerial power .......
CEO Power and Firm Performance ...........
Implications for managerial capitalism .... .
Implications for managerial power .......
Suggestions for Future Research ..................
Sum m ary ...............................


. 68
. 69
. .71
. 71
. 72
. 72
. 74
. .74
S. 76
. 78
* 78
. 79
. .80
..82


REFERENCES ....................................... 84

APPENDIX A LISTING OF FIRMS AND INDUSTRY CODES .......... 92

APPENDIX B MEANS, STANDARD DEVIATIONS, CORRELATIONS .... 97

BIOGRAPHICAL SKETCH ................................. 98


_ _












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

POWER AND CORPORATE CONTROL:
CEO SUCCESSION IN OWNER AND MANAGER CONTROLLED FIRMS

By

Kevin C. Banning

December 1996

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

Power and control, particularly the power of the chief executive, have inspired

significant research attention. One way to study CEO power is to isolate the

consequences of the appointment of a new chief executive because the period following

CEO succession involves significant adjustment for both the firm and the new chief

executive. During this time the new CEO will consolidate his or her control over the

firm's activities and resource allocations. Agency theory and managerial capitalism

suggest that the interests pursued by the firm's managers diverge from those of the firm's

owners. As a result the CEO may acquire and use power, vis a vis the board of

directors, in order to implement self-serving strategies or increase the level of

compensation.

This research examines the acquisition and use of power by incoming CEOs

within the context of agency theory and managerial capitalism, which predict that the







ability of CEOs to acquire and use power would differ based on whether the firm is

controlled by its owners or its managers. The study extends existing work in these

theories and contributes to the knowledge surrounding the consequences of CEO

succession.

The results support the principal hypothesis that a firm's strategy, and ultimately

its performance, will differ with respect to whether the firm is owner or manager

controlled. At the corporate level of strategy, as defined by product diversification,

manager controlled firms were found to engage in less related diversification. These

firms also performed less well than did the owner controlled firms experiencing CEO

succession. CEO compensation risk was also smaller in firms controlled by managers.

The pattern of the results suggests that CEO power differs based on whether

owners or managers control the firm. The results also suggest that in addition to firms

controlled by managers, new CEOs have greater power when they are also chair the

board of directors, originate from outside the firm, or when following a predecessor

CEO who was dismissed from office. Overall, CEO power influenced the composition

of the board of directors and the strategies employed by the firm, and resulted in

differing levels of performance.













CHAPTER 1
INTRODUCTION AND LITERATURE REVIEW


The acquisition, exercise, and maintenance of power have long fascinated

organization researchers because power is critical to understanding the decisions made

by social actors within organizations (Pfeffer, 1981). Power may be defined as the force

necessary to change another actor's behavior (Dahl, 1957) or the ability to bring about

desired outcomes (Salancik & Pfeffer, 1977). Power may also be understood as

managerial discretion, which is the latitude of managerial action within organizational

domains (Hambrick & Finkelstein, 1987). In this study, power and discretion are treated

synonymously.

The explicit study of power and power relations in organizations is important

because of the recognition that organizational decisions may be based on the self-interest

of political actors (Mintzberg, 1983; Pfeffer, 1981) rather than driven by purely rational

economic criteria (Child, 1972). These decisions therefore may have significant

operational consequences with respect to resource allocations (Pfeffer & Salancik, 1978).

The examination of power in business firms is an important research area. The reason

is that managers with power may further their own interests, rather than those of the

equity holders of the firm.







2
The exercise of power in the firm may alter organizational equilibrium, or the

balance of inducements to and contributions by individuals involved in the operations of

the firm (March & Simon, 1958). The firm provides inducements to its constituents,

including employees and stockholders, in return for their contributions of labor and

capital. By shifting resources, the exercise of power may disturb the balance of

inducements and contributions and therefore disturb the flows of labor and capital.

Further, changes in the organizational equilibrium stemming from different levels of

power may have economic consequences for firms. Indeed, Pfeffer (1981, p. 231)

argues that "power has consequences in organizations for resource allocations,

administrative succession, structures, and strategic choices." Ultimately, these

organizational decisions will affect financial performance and shareholder value, which

are important measures of organizational effectiveness.

The power literature has identified sources of organizational power, such as

control of key contingencies or resources (Hickson, Hinings, Lee, Schneck, & Pennings,

1971; Wernerfelt, 1984). Essentially Hickson et al. argued that control of information

or other resources constitutes a source of power for a firm's managers. Other research

has examined the consequences of power.in use for such organizational issues as resource

allocations (Allen, 1981; Pfeffer, 1981). In addition, organizational researchers have

explored the psychological bases of power experienced by individuals (McClelland,

1975). These bodies of research conclude that power is important but they leave several

issues, such as how power is acquired and maintained in organizations, relatively

unaddressed.







3
In most instances, research focuses on the role of a chief executive officer. Chief

Executive Officer (CEO) power is an important variable in the study of organizations

because powerful executives (i.e., those with greater discretion) are more likely to

initiate consequential organizational change through their influence on the firm's decision

processes (Pfeffer, 1992) and, other than the board of directors, no individual or group

is thought to exert as much influence on the structure and conduct of the firm as does the

CEO (Mintzberg, 1983).

This seems to be especially relevant following the placement of a new chief

executive. This is because a new chief executive comes with the potential for dramatic

strategic shifts. Theoretical expositions by Pfeffer (1981) and Pfeffer and Salancik

(1978) posit that succession events, in which a CEO is replaced because of death,

resignation, or dismissal, often bring about significant organizational change. The

empirical support for these propositions is strong (Greiner & Bhambri, 1989; Grinyer &

Spender, 1979; Smith & White, 1987; Virany & Tushman, 1986). Thus, during the

period following executive turnover, firms may change strategies or internal structure

through the power and influence of a new CEO.

Although the executive succession and executive power research streams have

much in common, power is rarely explicitly modeled in succession research, except as

an antecedent of succession (Harrison, Torres, & Kukalis, 1988; Ocasio, 1994; Pfeffer,

1981). In addition, with the exception of broadly construed organizational change or

performance, little is known about the processes that lead to organizational change and

other specific organizational consequences of executive succession. Thus, the modeling








4
of succession consequences would benefit from the consideration of changes in power

following succession. Further, these changes may not be the same if there are

constraints on managerial power. One such constraint could be the power of equity

holders. That is, when there are some equity holders who have relatively large blocks

of stock, their influence on the board, and subsequently on the CEO, may limit his or

her discretion (Berle & Means, 1932; Marris, 1964).

This study examines these issues, i.e., post-succession outcomes of managerial

power processes in firms with differing levels of ownership influence. Through the

comparison of changes in the governance, strategies, and performance of firms

experiencing succession, and the introduction of ownership characteristics, this research

will provide insight into the acquisition and exercise of power in U.S. businesses.

Theoretical Bases

The principal theoretical framework for this examination of power is derived from

a structural theory of intraorganizational power developed by Hickson et al. (1971).

They argue that power within the organization derives from the control of key resources

or decisions. As a result, power is less a function of a particular individual than of the

position within the organization. There are several structural determinants of

organizational power, which are discussed in the following section, that explain how

particular sub-units are able to acquire power and exercise influence within the

organization. The theoretical basis of power in organizations, as well as the literature

relevant to how power is acquired and used, is discussed in three parts: (1) managerial







5
capitalism and agency theory (2) power in organizations, including sources and uses of

executive power, (3) executive succession.

Managerial Capitalism and Agency Theory

A critical factor in the success of American business in general, and the publicly held

corporation in particular, is the need to attract huge pools of equity capital, which results

in the separation of ownership from control (Berle & Means, 1932; Lamer, 1966). The

separation introduces the possibility of divergent interests with respect to owners

(shareholders) and managers, a phenomenon first described by Berle and Means (1932),

and later called managerial capitalism by Marris (1964).

Managerial capitalism has led to important theoretical notions about the

relationship between the owners of a firm and those who exercise effective control.

Agency theory, a separate but related theoretical approach which posits the firm as a

nexus of contracts among principals and agents, has established a valuable method for

studying the information asymmetry problems associated with the principal-agent

relationship (Jensen & Meckling, 1976). Managerial capitalism and agency problems are

linked by McEachern (1975) as two tiers of the same fundamental issue. At the macro

level, managerial capitalism describes the effects of ownership and control separation on

firm activities, while at the micro level, agency theory details the effects of greater

information possessed by the agent or manager on the firm's activities.

Managerial capitalism predicts that when ownership is dispersed and managers

control the firm, managers will attempt to increase firm size and diversity because

managers of larger and more diverse firms can demand higher pay. Agency theory







6
explains how managers, in the absence of effective monitoring by owners and in the

possession of direct information about the firm and its performance, may effectively

control the corporation. Although both theoretical approaches greatly enrich knowledge

of organizations and the processes which underlie their performance, scholars have linked

agency theory and managerial capitalism to the various levels of strategy in an attempt

to understand the factors that motivate strategic choice and ultimately drive financial

performance.

The separation of ownership from control, stemming from the dispersion of

corporate equity to many owners, permits independent risk bearing and management

functions which may provide modem corporations with a specialization advantage (Fama

& Jensen, 1983). However, because the risk bearing and managerial functions are

separated, which is in contrast to the classical view of the firm in which these functions

are united in the entrepreneur, managers may be motivated to guide the firm in ways that

protect managerial interests, rather than those of the owners (Fama, 1980; Fama &

Jensen, 1983). The manager in a firm with dispersed ownership may draw excessive

compensation or take non-pecuniary perquisites because these costs are borne by the

firm's owners and not by the manager (Jensen & Meckling, 1976). More important,

because the ownership is dispersed among several owners, individual owners have

reduced incentive to monitor managerial decisions that might result in excessive

compensation or other outcomes which benefit managers (Fama & Jensen, 1983).

There is an active empirical literature that explores the firm-level behavioral

issues associated with the dispersion of corporate ownership (Allen, 1981; Amihud &







7
Lev, 1981; McEachern, 1975; Hill & Snell, 1989; Hunt, 1986; Walsh & Seward, 1990).

The behavior differences that researchers have observed between owner controlled firms

(where ownership is concentrated) and manager controlled firms (where ownership is

dispersed) are related to the propensity of managers, in the absence of effective control,

to maximize personal wealth, security, and utility functions (Amihud & Lev, 1981;

Marris, 1964).

The implication for the firm's owners is that inappropriate strategies may be

pursued and inferior economic performance may result. For instance, managers might

focus on sales growth or product-market diversification strategies, rather than seeking to

maximize profitability, because a manager's employment risk is typically lower in a

larger firm while pay is higher. Consistent with this logic, Amihud and Lev (1981)

found that managers in firms with dispersed ownership were more likely to pursue

strategies involving unrelated diversification, which increases firm size. Hill and Snell

(1988, 1989) found similar effects and more important, they reported that unrelated

diversification was associated with lower firm performance. Other empirical research

shows that firms with concentrated ownership outperform firms with dispersed ownership

(Hunt, 1986; McEachern, 1975; Tosi & Gomez-Mejia, 1994; Ware, 1975).

Managerial capitalism and agency theory provide a theoretical frame for exploring

the power of CEOs with respect to the board of directors. Because managerial capitalism

and agency focus on control of the firm, these theories have formed the basis for inquiry

into variations among firms in corporate governance and strategy. By assuming that in

the absence of effective monitoring, managers will act more in their own interests than







8
in the interest of shareholders, the theories of agency and managerial capitalism provide

a link between a firm's governance structure and the strategic decisions made by

managers.

When the board of directors fails to act in the shareholders interest, managerial

goals like employment security and personal wealth maximization may dominate over

shareholder wealth maximization. Similarly, without some form of monitoring or

incentive alignment mechanism, managers will pursue goals that maximize managerial

interests rather than shareholder value (Jensen & Meckling, 1976). For example, Kerr

and Kren (1992) and Tosi and Gomez-Mejia (1989) found that monitoring the CEO's

decisions, which acts to reduce the chief executive's power, resulted in a stronger link

between CEO pay and firm performance. In addition, Tosi and Gomez-Mejia (1994)

reported that CEO compensation monitoring was associated with increased firm

performance.

Other studies show that managers, and the CEO in particular, are more likely to

make decisions that are based on personal interests, rather than those of the firm, when

the influence of external owners is low or nonexistent (Hunt, 1986). These decisions

include corporate growth (Amihud & Lev, 1981; Hill & Snell, 1988), executive pay

(McEachern, 1975; Gomez-Mejia, Tosi & Hinkin, 1987; Tosi & Gomez-Mejia, 1994),

and technology strategy (Hill & Snell, 1989; Baysinger, Kosnik & Turk, 1991).

CEO power may also stem from the ability to influence the board of directors

through social influence or director appointments. For example, O'Reilly et al. (1988)

reported that the level of external compensation received by the board of directors was







9
positively associated with the CEO's own level of pay. As a result, they concluded that

powerful CEOs exercised their influence on the board of directors through outside

director appointments. A CEO's power with respect to the board may also be exercised

through social influence processes. For example, Tosi and Gomez-Mejia (1989) found

that CEOs in management-controlled firms, where discretion is high, exerted more direct

influence over their own compensation packages than did CEOs in owner controlled

firms. Similarly, Wade, O'Reilly, and Chandratat (1990) found that powerful CEOs

were more likely to receive golden parachutes in the event of a takeover.

Typically the board of directors serves as the mechanism through which outside

owners monitor the decisions made by managers. However, the board is subject to

manipulation by the CEO, who may also have greater decision control through the

board's structure (e.g., if the CEO is also the board chair) or composition (e.g., if the

board is composed of outside members, particularly those nominated by the CEO).

The chief executive's relative power with respect to the board of directors stems

from control of resources, financial or human, and the form and strength of managerial

decision monitoring by the board. The implication of these arguments for CEO power

is straightforward: under conditions when ownership influence is low and CEO power

is high, the chief executive will likely exert greater control over the strategies and

decision processes in the firm; and because the goals of managers and shareholders are

thought to diverge, shareholder value may suffer. The section that follows will discuss

organizational power in a more general sense and continue to develop the theoretical

consequences of CEO power and discretion.









Power in Organizations

Conceptualizing power as the ability to bring about desired outcomes raises

the question about how the power of the chief executive is acquired, maintained, and

used in organizations. Hickson et al. (1971) argued that operating departments,

divisions, or sub units have power if they control strategic contingencies, which are

activities on which other sub units depend. In addition, sub unit power depends on how

effectively the subunit copes with uncertainty, how central the unit's activities are to the

workflow (centrality), and how easily the activities could be substituted substitutabilityy).

A similar logic can be applied to the power of the chief executive. A CEO who controls

key resources or information gains power relative to other managers or the board of

directors. However, due to the strategic and non-programmable nature of the chief

executive's work, the substitutability and centrality dimensions of power are less

important for the purposes of this examination of chief executive power. The control of

key contingencies and effectiveness of coping are critical aspects of the chief executive's

position because these are roles often taken by the chief executive. More important, the

control of strategic contingencies and coping with uncertainty contribute to CEO power

and influence. The two sections that follow will identify and discuss sources and

consequences of CEO power.

Sources of organizational power

To understand how power is maintained in organizations it is necessary to

understand its underlying sources. Chief executives may acquire and maintain power

through structural arrangements, like those involving the board of directors and its formal







11
relationship to the CEO, or through the exercise of personal influence. Pfeffer (1981)

argues, following Hickson et al. (1971), that power stems from the control of resources

or skills in the five dimensions of the organization; provision of resources, coping with

uncertainty, irreplacability, decision influence, and political influence. (These sources

and uses of executive power are also modeled in Figure 1.1.)

Provision of resources. The provision of resources involves the allocation of

financial, human, and information resources to the various activities of the firm and it

is the dependence of others on these resources that creates power for the chief executive.

The chief executive influences resource allocation by acting as organization strategist.

Because the CEO appropriately serves as chief strategist (Barnard, 1938), and this role

involves the allocation of organization resources, the CEO may control fiscal or

informational resources on which other organizational members depend.

Coping with uncertainty. Because all organizations operate in environments that

feature less than perfect certainty, managers of all firms must select coping strategies in

order to operate. Members of the organization, including employees, other executives,

and outside (non-employee) board members, require that uncertainty be managed or

absorbed (March & Simon, 1958). The chief executive may gain power through his of

her ability to cope with and/or interpret environmental uncertainty. Firms face external

environments that contain ambiguous cues and the chief executive may be permitted wide

latitude in interpreting the environment.

Substitutability. A third organizational dimension that can create power for the

CEO is irreplacability. This involves the possession of unique skills or abilities, or the









12

































I
0
o

0

o;


C
1~ 0 C
4) .~ 0-
5 0
0

SO 4)
~ (UC)







13
control of other resources, for which there are believed to be no reasonable substitute.

To the extent that others in the organization believe that the CEO has unique and

valuable abilities, the CEO gains power in that organization.

This effect of substitutability appears to be strongest in firms where the CEO has

the greatest power. Research has shown, for example, that CEOs in firms where the

managers maintain effective control of the firm, are less likely to be replaced than those

CEOs in firms controlled by owners (McEachern, 1975; Pfeffer, 1981). The implication

for CEO power is that it is easier for a CEO to maintain the perception of irreplacability

in firms where he or she already exercises significant power.

Decision influence. The ability to affect the decision process in an organization,

such as through the board or committee chair positions, represents a source of power

stemming from decision influence. If the CEO can control decision premises, limit the

alternatives that are considered, or control information about the alternatives, then his

or her power is enhanced because these activities reduce the CEO's dependence on the

board of directors, and consequently the powerful CEO is less dependent on influential

owners.

Political influence. Similarly, the exercise of influence on others inside and

outside of the organization creates power for a chief executive. The successful exercise

of political skills, through the selection of directors or executive managers for example,

represents the fifth source of power in an organization because the CEO may reduce his

or her dependence on the firm's owners. In summary, each of the five dimensions

creates power for the CEO, vis a vis the firm's owners or the owners' representatives







14
(the board of directors), either because of a resource dependence or political influence

process. The consequences of the power obtained by the CEO are discussed in the

following section.

Power in use

A central focus of this study is how power is used within the organization. Once

an individual acquires power, that person is likely to attempt to maintain that power

through its institutionalization (Pfeffer, 1981; Mintzberg, 1983). This involves the

creation and preservation of structural and personal means of influence. Structurally,

power could be institutionalized through a particular organization design or division of

executive responsibilities that concentrates decision-making authority in a particular

position. Power stemming from personal influence could be maintained by attracting and

retaining, in key organizational posts, individuals who are sympathetic to the focal

executive's agenda. Hence, when studying power, it may be useful for organizational

researchers to seek evidence that the power structure, particularly the power of the chief

executive, has become institutionalized (Salancik & Pfeffer, 1977). In addition,

examination of other changes, beneficial to those in power, in those organizational

dimensions thought to be influenced- by power in organizations would benefit

organizational research.

An initial step in identifying areas where chief executives may institutionalize

their influence is to assess how a CEO impacts resource allocations and the politics of

organizational structure (Pfeffer, 1981). There is evidence to support the notion that

some CEOs allocate resources, acquire authority, and create organizational structures in







15
such a way as to maintain control vis a vis the shareholders (Fizel & Louie, 1990;

Furtado & Karan, 1990).

Longer organizational tenure appears to be one characteristic of CEOs who are

more able to influence resource allocations and organizational structures within the firm.

Hill and Phan (1991) found that increased CEO tenure led to increased executive

compensation. They also reported that tenure was positively associated with CEO

influence over the board of directors. Similarly, Mallette and Fowler (1992) reported

that increased tenure and the dual appointment as chief executive officer and board chair

increased the resistance to takeovers. Increased CEO tenure also results in reduced

research and development expenditures, a strategy that is associated with inferior long-

term returns to investors (Baysinger et al., 1991). Similarly, Finkelstein and Hambrick

(1990) found that the level of commitment to strategies providing financial returns near

the industry average, rather than strategies with higher payoffs but more variability,

increased as executive tenure increased.

Although returns near an industry's average are not generally beneficial to

stockholders, who can achieve average returns through a diversified portfolio, such levels

of profitability can improve a CEO's job.security. If a firm's returns are below average

in a given period, the CEO faces an increased likelihood of dismissal (Jensen &

Meckling, 1976). For this reason, firms in which the CEO exercises effective control

over the board would more likely pursue strategies that yield only average returns while

firms in which the board is relatively more powerful would be expected to pursue







16
strategies that would provide superior long run returns. Thus, it appears that tenure may

provide the CEO with greater discretion over activities and decisions in the firm.

Conversely, when the board is more influential than the CEO with respect to the

firm's strategic choices, the result is that resource allocations are likely to be more

beneficial to the firm's owners. For example, under conditions of decision monitoring

by the board, research has found that the CEO's performance is more tightly linked to

pay and is, on average, lower in total than those CEOs in firms where there is little

decision monitoring (Kerr & Kren, 1992; Tosi & Gomez-Mejia, 1989). In addition,

Allen (1981) reported that the level of CEO pay was smaller, controlling for firm size

and financial performance, in firms where at least one owner held a significant block of

equity. Thus, the empirical tests of low levels of decision monitoring, which result in

increased CEO discretion and power vis a vis the board of directors, are consistent with

other measures of CEO power.

Westphal and Zajac (1995b) confirmed the relative power argument developed in

this research. They tested the impact of CEO power (relative to the board of directors)

on how long-term incentive plans were announced by firms. They found that when the

chief executive was also the board chair and had longer organizational tenure, proxies

for higher relative CEO power, the firm's explanation for long-term incentive pay

centered on executive ability and power. When the board chair and chief executive

positions were separate and CEO tenure was shorter, the firm's explanation for long-term

incentive pay centered on the board's control of management. They concluded that CEO

power affected both the symbolic and substantive realms of CEO compensation.







17
In another approach to measuring CEO power, Pearce and Zahra (1991) studied

CEO power in Fortune 500 firms using a questionnaire to CEOs and other senior

executives that assessed CEO power, board power, and board conduct. They concluded

that firm performance declined as CEO power increased.

Powerful CEOs may also build and maintain power through other structural

measures, such as appointing boards of directors that are more likely sympathetic to the

executive's agenda. O'Reilly et al. (1988) showed, for example, that as the level of

outside compensation for external board members increases, so does the level of CEO

pay. They speculated that by virtue of appointing highly compensated outside directors

to the board, CEOs may boost their own pay levels. Indeed, this relative power

explanation for the level of chief executive pay was supported by Westphal and Zajac

(1995a), who found that higher levels of CEO power relative to the board resulted in

higher levels of board similarity and executive pay. These increases in measures of

board and CEO similarity as well as in the ratio of outside board members are associated

with increased CEO pay without a corresponding increase in firm performance (O'Reilly

et al., 1988; Westphal & Zajac, 1995a). In addition, CEO pay is higher in firms which

have a larger proportion of outside directors (Fizel & Louie, 1990; Westphal & Zajac,

1995b), despite the notion that directors from outside the firm would more likely limit

the power of the CEO.

As a whole, this body of research suggests that CEO power with respect to the

board of directors has important consequences for several firm level outcomes. It

appears that significant resource allocation decisions (e.g., executive compensation,







18
research and development intensity) as well as the corporate governance structure, are

affected by CEO power with respect to the board of directors. One conclusion that may

be drawn from this body of evidence is that executive discretion is a function of

characteristics such as longer tenure, the appointment to both the CEO and board chair

positions (Rechner & Dalton, 1991), and the lack of external ownership influence.

Although very few of these studies explicitly assessed CEO power, all utilized measures

that are theoretically consistent with decision latitude and organizational power.

CEO Succession

In addition to the traditional sources of executive power (e.g., holding both the

CEO and board chair positions, longer tenure), CEO power may be enhanced simply by

virtue of ascension to the position. CEO succession can result in strategic changes and

shifts in corporate governance structures.

Many of these issues, including succession antecedents, strategic consequences,

and shareholder wealth effects, have been explored in previous executive succession

research (Kesner & Sebora, 1994; Furtado & Karan, 1990). In general, CEO succession

increases the likelihood of change in the organization. Although the direction of these

changes are rarely explored (see Ocasio [1994] for an exception), they may have

important consequences for who controls the firm and its ultimate performance.

Succession events, in which a CEO is replaced because of death, resignation, or

dismissal, permit researchers to examine the antecedents or consequences of corporate

power and influence processes. By virtue of the replacement of one chief executive with

another, it is possible to examine the influence processes in firms under conditions of low







19
institutionalization, in which the patterns of chief executive discretion have yet to be

established.

Most of the previous executive succession literature explores the determinants of

CEO turnover using characteristics of the firm's board of directors or its ownership, the

nature of the industry's economic environment, or firm performance, to predict CEO

turnover. For example, Fizel and Louie (1990) reported that outside directors were less

likely than inside (manager) directors to dismiss an incumbent CEO. Similarly,

McEachern (1975) and Salancik and Pfeffer (1977) reported that CEO succession was

less likely to occur in firms where ownership was highly dispersed. Thus, CEOs in

those firms with boards dominated by outsider directors or in which the owners are more

removed from management (by virtue of having only very small ownership stakes),

appear to enjoy greater freedom of action and a lower probability of dismissal.

Succession research has also linked the economic environment to executive

turnover. For example, Wiersema and Bantel (1993) reported that instability in the

economic environment was related to higher rates of CEO succession. In addition,

Walsh (1988) found that takeovers, which create uncertainty for the acquired firm,

significantly increased the likelihood of executive turnover in the acquired firms. Other

authors have reported that poor financial performance (Dalton & Kesner, 1987; Datta &

Guthrie, 1994) is a determinant of CEO turnover.

In addition to predicting the turnover event, poor financial performance also

affects the origin of the chosen successor to the chief executive post. For example,

Boeker and Goodstein (1993) found that poor performance led to selection of an outsider







20
CEO, though this effect was weaker as the number of insider board members increased.

They concluded that when the firm has performed poorly, the board chooses an outsider

in order to effect a turnaround in performance. The effects of poor performance as an

antecedent to CEO turnover are so well documented that Furtado and Karan (1990) have

called the association of inferior economic performance and executive succession a

"consensus finding" (p. 61).

In addition to the rational economic bases for CEO succession, some research has

demonstrated that there is a political power component involved. Canella and Lubatkin

(1993), for example, reported that the selection of a successor from outside the firm was

less likely when the previous CEO remained on the board. They concluded that the

power of the outgoing CEO affected the selection decision and that this discretion may

have other post-succession consequences. Similarly, Welsh and Dehler (1988) found that

political activity during succession, as measured by an influence-strategies scale

constructed by the authors, interacted with resource conditions such that when resources

were scarce, political activity led to more organizational change. Under conditions of

resource abundance, however, increased political activity had little effect on

organizational redefinition. Their results, like those of Canella and Lubatkin (1993),

suggest that both the antecedents and consequences of succession may be influenced by

the power of the incumbent chief executive.

A parallel but less well developed body of succession literature involves the firm-

level consequences of CEO succession, particularly the impact of new CEOs on a firm's

long-term financial performance (Brown, 1982; Carroll, 1984; Friedman & Singh, 1989)







21
and short-term stock market performance (Beatty & Zajac, 1987; Davidson, Worrell, &

Cheng, 1990; Lubatkin, Chung, Rogers, & Owers, 1989; Reinganum, 1985; Worrell,

Davidson & Glascock, 1993). These studies show a positive relationship between

succession and financial performance, particularly when the new CEO is from outside

the firm. For example, Lubatkin et al. (1989) found that CEO succession created

positive wealth effects only when the successor originated outside the firm. In addition,

Reinganum (1985) and Beatty and Zajac (1987) reported positive stock-market

announcement effects only for outside successors. Other studies have found positive

wealth effects for all succession events (Davidson et al., 1990; Worrell et al., 1993) but

these authors also reported that the wealth effects of succession were stronger when the

successor was an outsider.

One reason that equity markets react favorably to CEO succession, and especially

outside succession, is the expectation that the new chief executive will effect positive

change in the organization. This is because outside successors will presumably be less

constrained by the conventions and decision patterns of the predecessor.

Regardless of their insider or outsider status, a new CEO will have significant

decision latitude largely because the expectations for positive change which accompany

a succession event. As a result, new chief executives will likely initiate more change

than those who have held the CEO position for some time. Consistent with this logic,

succession researchers have concluded that succession is associated with organization-

level change and that external successors are more likely than internal appointees to







22
induce organizational change (Greiner & Bhambri, 1989; Helmich & Brown, 1972;

Wiersema, 1992).

Other research shows that CEO succession brings about changes in the nature of

organizational power. Goodstein and Boeker (1991) found, for example, that strategic

change was greater after CEO succession than prior to succession. In addition, Greiner

and Bhambri (1989) reported that new CEOs who were granted more discretion were

better able to implement structural and strategic change in their firms. These results are

consistent with Pfeffer's (1981) argument that organizational structures and strategies are

the result of self-interested parties in a political process rather than purely rational

creations.

Succession studies have shown that succession increased organizational mortality

(Carroll, 1984; Haveman, 1993), increased organizational growth (Helmich, 1974), and

made strategic change more likely (Grinyer & Spender, 1979; Smith & White, 1987;

Wiersema, 1992). For example, Wiersema (1992) found that corporate diversification

strategy changed more in firms experiencing succession than in those firms in the non-

succession sample. Further, she also reported that the origin of the successor moderated

this relationship such that when an outside successor was appointed as CEO, the level

of change in diversification strategy was significantly greater than the change for insider

successors.

The increased level of change found after the succession event indicates that the

new chief executive has significant power to exercise strategic decisions. However, the

power and influence processes through which change is effected, including the effect of







23
executive succession on the firm's board of directors, remain unexplored. Because the

board of directors is the single most important mechanism in corporate governance,

investors and researchers have a compelling interest in the impact that a powerful CEO

has on the corporate board. Based on Pfeffer's (1981) observation that executive

succession may have important consequences for organization governance and decision

making, this study will compare the influence of new CEOs on the board of directors,

on the strategies formed by the collaboration of the board and the chief executive, and

ultimately the impact of CEO power on firm performance.

The Theoretical Model

The period following succession is likely to permit a new CEO significant

discretion. Once placed as a firm's chief executive, the new CEO will begin

consolidating his or her control over corporate activities (Pfeffer, 1981). The chief

executive's consolidation of control results from decision latitude or discretion and this

power has consequences for two classes of outcomes, board structure and firm strategies.

Thus, the theoretical model involves the effect of CEO power on two separate classes of

outcomes. These classes, board structure and firm strategies, are explored in the

discussion that follows. Board structure is important because the board is the mechanism

through which shareholders may exercise control of the firm while the strategies

employed by the CEO are also important to the ultimate performance of the firm.

One way that CEOs may exercise power is through cooperative board members

who can be controlled. Once a new CEO assumes control of the firm, he or she may

have the ability to appoint friendly board members or in some other way change the







24
board's size or composition to protect his or her interest. The ability to appoint directors

is important because larger and more diverse boards are thought to increase the power

of the CEO. Similarly, a board with more outside directors, who rely on the CEO for

information about the firm, may enhance CEO power.

In addition, certain strategies at the corporate and functional levels of the firm are

more beneficial to the CEO. Strategies that reduce risk or enhance firm size, for

example, provide employment stability for the CEO but may yield no advantage to

shareholders. (These forces are shown in Figure 1.2.)

Corporate Control and the Board of Directors

One potential limitation of managerial power is ownership control (McEachern,

1975). Ownership control may be achieved by holding a significant portion of a firm's

outstanding common equity. In firms where no shareholder controls a sizable percentage

of stock, managers are presumed to exercise effective control. Such firms will be

referred to as manager controlled. Other firms feature at least one individual or

institution with a sizable equity stake and these owners will have sufficient incentive to

monitor the board and corporate management. Such firms will be referred to as owner

controlled because the owners are presumed to exercise effective control. These

differences in control of the firm lead to behavioral differences among corporations,

whether board structure or strategic choices, and appear to be a function of the relative

power of the CEO vis a vis the shareholders (Hunt, 1986). Indeed, Pfeffer (1981) argues

that, based on McEachern's (1975) work describing the effects of separate ownership and

control, the ownership structure of the firm may promote the institutionalization of








































C.
I-.



0
4*
s
rn



I-


C)
4) '
I-
2
0
Q







26
power, especially the power of the chief executive. For these reasons, chief executives

in manager controlled firms would likely create boards that permit them greater decision

latitude.

CEO Power and the Board of Directors

Because modem corporate ownership is widely dispersed, individual shareholders

have neither the motivation nor the ability to adequately monitor managerial decisions,

thereby ceding effective control of corporations to the board of directors. As the legal

representative for the owners of corporate entities, the board of directors exercises

governance over the firm's activities as well as those of the CEO. Because the board of

directors determines the top executives compensation package and approves a firm's

strategic choices, the new CEO may try to minimize the control exercised by the board.

A view of board composition, stemming from managerial capitalism, posits that

the structure of the board may be manipulated by executive management to consolidate

power through board of director appointments. Main, O'Reilly and Wade (1995) suggest

that through reciprocity norms, which involve a board member's psychological

connection to a CEO, directors appointed after the CEO takes office will tend to act in

the executive's interest. Although no researchers have reported direct tests of this

notion, studies of age and demographic similarity (Main et al., 1995; Westphal & Zajac,

1995b) reported that CEO compensation was higher when the CEO was more similar to

the board members socially and in age. Regardless of the psychological mechanism

involved, however, a CEO who exercises effective control over the board of directors,







27
whether through social similarity or reciprocity, may gain increased discretion as well

as increased compensation.

Another board influence tactic employed by executives is appointing outside

directors (board members who are not also managers in the firm). Outside directors

must rely on the firm's managers (especially the CEO) for information about the firm's

performance, and as a result these directors may be subject to manipulation by the CEO

(Baysinger & Hoskisson, 1990; Hoskisson & Turk, 1990). Indeed, empirical evidence

suggests that outside board members are more sympathetic to the agendas of CEOs in

that executive pay is higher in firms with a greater proportion of outside board members

and this relationship holds when controlling for firm size and performance (O'Reilly et

al., 1988; Main, et al. 1995; Westphal & Zajac, 1995b).

In addition to replacing inside directors with outsiders, and incumbent outsiders

with outsiders who are known to the chief executive, CEOs may also actively encourage

insider directors to leave the board (thereby replacing them with more sympathetic

insiders or outsiders) at a faster rate than their outsider counterparts. Whether a board

member leaves voluntarily or at the request of the top executive, CEOs would be

expected to nominate outsiders as replacement directors.

Thus, the new CEO may stack the board with outside directors over whom the

chief executive may have more influence, limiting the influence of existing internal and

external board members. Although this argument runs counter to classical theoretical

expositions of agency theory (e.g., Fama & Jensen, 1983; Walsh & Seward, 1990),

inside directors may have more to gain by ousting an incumbent CEO. The reason is







28
that as senior managers inside directors may actively compete for the chief executive's

position, a concept referred to as tournament theory (Lazear & Rosen, 1981) or

circulation of power (Ocasio, 1994), which has received recent empirical attention

(Ocasio, 1994; O'Reilly et al., 1988).

Inside directors, then, rather than outsiders, are expected to be the more intense

monitors of CEO performance. Indeed, Ocasio (1994) found that when the CEO

performed poorly, greater concentrations of inside directors were more likely to result

in the chief executive's dismissal. In addition, Main et al. (1995) reported that outside

directors were more subject to influence by the CEO and that this added influence

resulted in higher pay for the chief executive than would be predicted by firm

performance. Similarly, Fizel and Louie (1990) and Baysinger et al. (1991) found that

greater concentrations of outside directors resulted in lower CEO accountability. Thus,

although outside directors have the ostensible purpose of limiting CEO power, it appears

that these board members may augment the chief executive's influence with respect to

activities, including composition of the board, within the firm. These arguments lead to

the following hypotheses.

H 1: The post-succession ratio of outside directors will increase to a greater extent
in manager controlled firms than in owner controlled firms.

H2: The post-succession turnover of directors will increase to a greater extent in
manager controlled firms than in owner controlled firms.

In addition to influencing board composition and turnover, new CEOs would be

expected to increase board size, all other things being equal, because larger boards are

typically less effective monitors of CEO activity (Goodstein, Gautam, & Boeker, 1994;







29
Judge & Zeithaml, 1992). By creating a larger board of directors, the CEO ensures that

consensus will be more difficult to reach and that his or her policies are less likely to be

challenged. In addition, like the outsiders that a CEO nominates to the board, any

additional new board members are more likely to be loyal to the CEO that appointed

them. This argument leads to the following hypothesis.

H3: Post-succession board size will increase to a greater extent in manager
controlled firms than in owner controlled firms.

Appointing outsiders and increasing the size of the board represent two direct

means of consolidating and institutionalizing CEO power. The relative compensation of

outside directors presents a third influence mechanism through which the CEO may

pursue personal interests. If a board member's outside compensation is high, norms of

reciprocity suggest that these directors would more likely approve of high levels of

compensation for the focal CEO. Indeed, empirical evidence (O'Reilly et al., 1988;

Wade et al., 1995) indicates that CEOs who appoint outside directors that are highly

compensated gain in their own levels of pay. Similarly, Kosnik (1990) found that

increased outside director pay resulted in higher executive pay, irrespective of firm size

or financial performance. Hence, not only would powerful CEOs attempt to stack their

boards with outsiders, all other things equal, but they would search for highly paid

outsiders. This argument leads to the following hypothesis.

H4: The level of external compensation received by outside directors, post-
succession, will increase to a greater extent in manager controlled firms than in
owner controlled firms.









Limitations of CEO Power

Other potential determinants of chief executive influence have been suggested by

previous empirical work. Friedman and Singh (1989) found that the origin of the new

CEO (insider or outsider), whether the board or the CEO initiated the turnover, and

whether the previous CEO remained associated with the firm has a significant effect on

successor power. Outside successors to the CEO position appeared more able to bring

about changes in strategy than did inside successors. Board initiated turnover also

appeared to give the new chief executive greater influence than turnover initiated by the

outgoing CEO. They also showed that the influence of the successor was constrained

when the predecessor remained associated with the firm as a board member. Similar

results were reported by Boeker (1992), Furtado and Karan (1990), and Wiersema

(1992).

Firm size and previous financial performance may also affect the relative power

exercised by the new CEO. In larger firms, for example, the board of directors may be

larger and therefore more manipulable. These boards could also be more powerful due

to the status involved in serving as a director for a larger firm. Similarly, when a firm

has performed well, the new CEO may be subject to less scrutiny than would be the case

if the firm's previous performance were poor. Conversely, when a firm's previous

performance has been low, the successor CEO may have greater decision latitude in

order to effect a turnaround. In either case, ownership concentration, firm size, and

previous financial performance are potential limitations of the power exercised by new

CEOs.









Strategic Decisions

If, as speculated above, the CEO has the ability to manipulate the board of

directors either through direct influence or the selection of new directors, he or she

should be able to manipulate strategic resource allocations. The discussion that follows

will explore the implications of CEO power for strategic decisions and ultimately for the

financial performance of the firm. Strategic decisions affect those activities of the firm

that involve significant resource commitments (Ansoff, 1965) and strategy represents the

fundamental pattern of resource deployments (Hofer & Schendel, 1978). In what

Schendel and Hofer (1979) called the strategic management paradigm, informed choices

by managers about three hierarchical but interdependent levels of strategy (Andrews,

1971), corporate, business level, and functional, are primarily responsible for a firm's

performance.

Corporate strategy involves the scope of businesses in which the firm participates.

It includes a firm's decisions about diversity of product line, extent of geographic

presence, the choice of core industries, and the degree of the organization's growth,

particularly that achieved through acquisition.

Corporate diversification may include product markets that are related to the

primary line of business or markets that are entirely unrelated to the core business, where

the core business is defined as the single product market from which the firm derives

most of its revenue (Rumelt, 1982; Wood, 1971). The rationale for related

diversification is based on the sharing of core competencies such that the value chains

of each product market are related and thereby create scale economies and other







32
advantages (Porter, 1985). Related diversification also may permit some sharing of

financial resources within a single firm. Unrelated diversification affords some element

of risk reduction and financial resource sharing, but is not associated with the operating

synergies available to related diversifiers (Bettis, 1981; Datta, Rajagopalan & Rasheed,

1991; Ramanujam & Varadarajan, 1989). However, unrelated diversification increases

firm size (Amihud & Lev, 1981).

Business level strategies are those choices about how to produce and market the

products or services created by their core businesses. They involve a firm's decisions

about customers, methods of production, and ultimately the extent of sales growth.

Business level strategies integrate the various functional activities of the firm into a

cohesive whole in order to obtain a competitive advantage.

Functional level strategies are those that administer the functional activities of the

firm. They involve the pattern of resource allocations to operational activities and

include pay decisions, relative investment in research and development, and utilization

of technology. Functional strategies, like corporate and business level strategies, are

important determinants of a firm's competitive position and resulting financial

performance. Some of the functional level strategies thought to be related to

performance and managerial power are executive compensation strategy, including both

the level as well as the composition of CEO pay, and technology strategy. The nature

of these relationships, and their consequences for firm performance, are explored in the

following sections.









CEO Power and Strategic Decisions

Because corporate and functional strategies are ordinarily formulated by some

interaction between the CEO and corporate board, the inclusion of several strategy

measures yields a more comprehensive test of the influence exercised by new executives.

Because CEOs are presumed to be risk averse self-maximizers, those who face less

effective board or shareholder monitoring are more likely to implement corporate,

business level, and functional strategies that reduce employment risk and maximize

personal gains (Jensen & Meckling, 1983). The following section will discuss the

ownership concentration determinants of CEO discretion in the context of how CEO

power affects the firm's strategic choices.

Corporate Control and Strategy

At the corporate level, strategies involving sales growth or unrelated

diversification are associated with reduced variance of returns--a firm's risk measure--and

therefore reduced employment risk for managers (Amihud & Lev, 1981; Hill & Snell,

1988; 1989). Diversification also involves expansion of the firm and this increased size

serves the interest of the CEO because CEO pay is higher in larger and more diversified

firms (Ciscel & Carroll, 1980; Finkelstein & Hambrick, 1989) and the chief executive's

employment risk is reduced to the extent that sales growth is high. Diversification,

particularly when unrelated to the firm's core businesses, tends to reduce profit

variability and may also reduce average returns to shareholders. Empirical research has

established that unrelated diversification leads to lower performance, whether in

comparison to single business firms or related diversifiers (Bettis, 1981; Dubofsky &







34
Varadarajan, 1987; Rumelt, 1982). However, because larger firms are less likely to fail,

and unrelated diversification increases firm size, the chief executive's employment risk

is reduced to the extent that unrelated diversification is high. In addition, because firm

size (sales volume) is the single most significant determinant of the level of executive pay

(Ciscel & Carroll, 1980; Gomez-Mejia et al., 1987), a firm's level of unrelated

diversification and CEO compensation are closely related. Thus, to the extent possible,

powerful CEOs would be expected to embark on growth and diversification strategies

because these activities serve their personal interests.

Chief executives of firms where decision monitoring is low because owners are

not in control (i.e., manager controlled firms) will have greater discretion to undertake

corporate strategies that reduce compensation risk and increase firm size. Because

decision monitoring is lower in manager controlled firms (Tosi & Gomez-Mejia, 1989),

CEOs in these firms would engage in relatively more unrelated diversification. Indeed,

previous empirical research indicates that manager controlled firms engage in

significantly more unrelated diversification than their owner controlled counterparts

(Amihud & Lev, 1981; Baysinger et al., 1991; Hill & Snell, 1988; 1989). As discussed

previously, unrelated diversification reduces the CEO's employment risk and increases

firm size, both of which benefit the CEO. This argument leads to the following

hypothesis.

H5: Post-succession levels of unrelated product-market diversification will
increase to a greater degree in manager controlled firms than in owner controlled
firms.







35

Substitution of unrelated diversification for growth in related businesses represents

a low risk strategy that is likely to bring financial reward (Amihud & Lev, 1981; Hill

& Snell, 1988). Chief executives in manager controlled firms would have the

opportunity to allocate the firm's resources to increasing sales revenues through unrelated

diversification. The increased revenues (firm size) are associated with increased levels

of CEO pay and reduced levels of failure risk, and as a result chief executives in

manager controlled firms would likely attempt to acquire unrelated business rather than

create or acquire related business units. Because the firm's total resource base is limited,

however, increases is unrelated diversification would be associated with a decline in

related diversification. This argument leads to the following hypothesis.

H6: Post-succession levels of related diversification will decrease to a greater
extent in manager controlled firms than in owner controlled firms.

Despite the clear implications of separated ownership and control for the corporate

level of strategy, empirical work is lacking for strategies at the business level. Although

theoretically the argument could be made that business level strategies that involve less

risk or more sales growth would be preferred by powerful CEOs, risk rankings and

performance implications of the various business strategies (e.g., Miles & Snow, 1978;

Porter, 1980) have yet to be established.

Like the corporate level, functional level strategies employed by powerful CEOs

will be more self-serving than those employed by less powerful CEOs. Functional level

strategies, such as the firm's position with respect to executive compensation, liquidity,

and technology (research and development), are tools that may serve the interests of

managers or shareholders. A powerful chief executive, who may be permitted great







36
discretion in the period immediately following succession, would likely pursue functional

strategies that reduce his or her risk and that assure maximum remuneration. Thus,

powerful CEOs would be expected to reduce their pay risk while increasing total

compensation. In addition, they would be expected to invest relatively little in strategies

with long-term gains, such as research and development, in favor of strategies with short-

term gains, such as acquisition.

Despite the intuitively appealing idea that self-serving strategies would harm the

chief executive's reputational capital and limit future employment opportunities (Fama,

1980), this perspective has not received empirical support (Hill & Snell, 1989). Indeed,

empirical evidence suggests that functional strategies vary under different power

structures and that these differences are consistent with the argument that powerful CEOs

select strategies that are more self serving than those strategies selected by less powerful

chief executives. Compensation researchers, for example, have found that executives of

manager controlled firms are paid based on firm size rather than performance (Gomez-

Mejia et al., 1987; Tosi & Gomez-Mejia, 1994). Hence, to the extent that powerful

CEOs increase firm size or otherwise influence compensation decisions, their absolute

level of pay should be higher than CEOs who have less influence on pay decisions.

Indeed, empirical evidence indicates that executive pay is larger in manager controlled

firms, even when controlling for firm size and financial performance (Allen, 1981).

CEO compensation risk is also lower in manager controlled firms than in owner

controlled firms, suggesting that CEOs in manager controlled companies have less







37
effective monitoring (Tosi & Gomez-Mejia, 1989). These arguments lead to the

following hypotheses.

H7: The level of post-succession CEO compensation will be greater in manager
controlled firms than in owner controlled firms.

H8: The CEO's post-succession compensation risk will be smaller in manager
controlled firms than in owner controlled firms.

In addition, empirical evidence indicates that when managers control the firm,

research and development (R&D) expenditures are less than in owner controlled firms

and although commitment to technology is associated with superior financial

performance, it also involves increased risk (Baysinger, et al., 1991; Hill & Snell, 1988;

1989). Thus, CEOs would prefer to invest in projects with more certain (but possibly

smaller) returns in order to reduce their employment risk. This argument leads to the

following hypothesis.

H9: Manager controlled firms will have less investment in technology than will
owner controlled firms following succession.

Through the functional strategies explored in the preceding discussion, powerful

chief executives may increase their relative share of organizational resource allocations

and reduce the risk of these allocations. Similarly, business-level strategies that increase

sales growth may also permit the CEO to increase his or her control over organizational

resources and rewards. Finally, corporate level diversification, which enhances firm

size, may also allow the CEO to increase control over organizational resource allocations

and simultaneously reduce the risk in obtaining them. Although the exercise of executive

power may have significant ramifications for board size and structure as well as firm

strategies, ultimately it is the impact of these decisions on firm performance that matters







38

to shareholders. Because the firm's rational economic objective is maximization of

shareholder value, any deviation from shareholder wealth maximization that has origin

in executive power is of interest to shareholders. Thus, the section that follows will

develop performance implications from the preceding discussion of strategy and board

governance.

Performance Implications

Of all the possible consequences of CEO power, the most important from the

organizational strategy perspective is the effect on financial performance of the firm.

The previous sections argue that executive power has significant consequences for board

governance and firm strategy. Moreover, weaker governance structures are associated

with reduced shareholder value (Baysinger et al., 1991; Kosnik, 1990; Mallette &

Fowler, 1992) and these reductions in shareholder value likely result from low levels of

monitoring or weak incentive alignment.

Researchers have reported that some of the various strategies described in the

previous sections are associated with inferior financial performance and therefore reduced

shareholder wealth. Specifically, Rumelt (1982) found that firms with many unrelated

business units (e.g., conglomerates) exhibit inferior financial performance in comparison

to single-business or firms whose business units are related, either through a core

technology, product market, or distribution competency. The association of increased

unrelated diversification to decreased financial performance has also been supported in

several other empirical studies (Amit & Livnat, 1988; Baysinger & Hoskisson, 1989;

Berger & Ofek, 1995; Hill & Snell, 1988).







39
Some researchers have also discovered that another strategic shift, the decoupling

of CEO pay from firm performance, leads to decreased firm financial performance (Tosi

& Gomez-Mejia, 1994). One other strategy explored in this research, technology

strategy, has an empirically established link to firm performance. A significant body of

empirical research has shown that reduced investment in technology is associated with

decreases in financial performance, particularly over the long-term investment horizon

(Baysinger & Hoskisson, 1989; Hill & Snell, 1988; 1989). Other research has found that

the level of commitment to technology moderates the relationship of related

diversification to performance such that increasing levels of commitment to technology

lead to increased financial returns for related diversifiers (Bettis, 1981).

This logic can also be applied to the control of the firm as described previously.

Specifically, because new CEOs in manager controlled firms are better able to implement

changes to the board of directors and firm strategies, financial performance should suffer

in these firms. This argument leads to the following hypothesis.

H10: Post-succession levels of financial performance will be higher in owner
controlled firms than in manager controlled firms.



Summary

Through the analysis of executive discretion and its relative impact on the boards

of directors and strategies of firms, this study will further the understanding of CEO

power and the mechanisms that serve to limit this power. The succession event provides

a clear foundation from which research into the exercise of power may begin. Once a

new CEO is appointed, the CEO will act in his or her own interest to the extent possible.







40
Managerial capitalism, through lack of owner control, also provides a mechanism to

increase the power of the CEO relative to the owners or board of directors.

This study proposes a theoretical model featuring two kinds of power

consequences: board structure and firm strategy, as well as their ultimate impact on firm

performance, that examines the relative influence of new CEOs and of those in firms

with differing ownership concentrations, economic size, and financial performance

characteristics. All of the principal testable hypotheses implied by the model expect that

powerful (high discretion) CEOs will attempt to consolidate their power within the

organization, either through changing the board structure, altering the firm's strategies,

or both.

The hypotheses predict that CEO power will differ with respect to ownership

structure. Specifically, with respect to the board of directors, high discretion CEOs

would increase the board's size and outsider representation, and appoint highly

compensated outside members. In addition, in the strategic decision domain, high

discretion CEOs would be expected to engage in diversification and sales growth

strategies at the corporate and business levels. At the functional level, high discretion

CEOs would attempt to increase both the absolute level and fixed component of their

compensation packages. Because these changes in the board of directors and firm

strategies may have negative consequences for corporate financial performance, corporate

ownership structure and succession are important issues in corporate performance.













CHAPTER 2
RESEARCH DESIGN


This chapter consists of three sections. The first section describes the sample and

data collection process and the second section explains the analytical methods employed.

Finally, the third section relates how the conceptual variables were operationalized in this

research.

Sample and Data Collection

Data for this research were collected from archival sources including the

Compustat and Compact Disclosure electronic data bases, the Wall Street Journal, Dun

and Bradstreet's Reference Book of Corporate Managements, and proxy statements from

each of the firms.

A sample of 227 publicly held firms was identified that experienced at least one

change in the CEO position during the event period (1988 to 1991). Twenty-four firms

with multiple succession events during the event period were eliminated in order to limit

the study to the effects of a single change in the chief executive officer. Fifty-nine firms

experiencing CEO succession in the four years immediately preceding the event period,

or in the three years following the event period, were also eliminated from the sample

in order to limit the study to the effects of a single change in CEO. The study period

was chosen such that the sample size yielded sufficient power to detect moderate to small







42
effects. The succession event period also contained a relatively few years (four) to

ensure that the inferences discussed in the first chapter would be appropriate. (The

sample firms and industry codes are listed in Appendix A.) Although equity

concentration is likely to remain stable over time, in some cases a firm's equity may

become significantly more or less dispersed over time. As a result, one firm was

eliminated from the sample because the level of the equity concentration variable changed

over the 1985 to 1994 study period. The final sample contained 143 firms with a single

succession event.

After the final sample was constructed, the data were divided into two panels.

The first panel included the four years from 1985 to 1987 and will be referred to as the

pre-succession panel. The second panel included the years from 1992 to 1994 and will

be referred to as the post-succession panel. The continuous variables within each panel

are arithmetic averages of the component years and are, where necessary, scaled to 1988

dollars through the use of the GDP deflator.

Analytical Methods

A multiple-regression method was used to test the research hypotheses. Beginning

with the power determinants and control variables (e.g., firm size), the board

composition variables were regressed on the power variables

Y = a + j31Size + 02Prior Performance + 33Manager Control +

04Duality + WsTenure







43
where Y represents each of the board structure variables: outsider ratio, director

turnover, outside director pay, and board size. Next, the strategy variables were

regressed on the power, board composition, and control variables

Y = a + (3Size + 32Prior Performance + 03Manager Control +

34Duality + 35Tenure + 3Outsider Ratio + 87Director Turnover +

(gDirector Pay + 39Board Size

where Y represents each of the strategy variables: unrelated diversification, related

diversification, CEO pay, pay risk, and technology emphasis. Finally, in the full model,

the firm performance index was regressed on the power, board composition, strategy,

and control variables. These regressions take the following general form

Y = a + O3iSize + 32Prior Performance + 33Manager Control +

34Duality + (35Tenure + 06Outsider Ratio + 07Director Turnover +

08Director Pay + g39Board Size + 3 1oUnrelated Diversification +

i1,Related Diversification + 12CEO Pay + I13Pay Risk + 14Technology

where Y represents the performance index and the independent variables are described

in the third section of this chapter.

Variable Descriptions and Operationalization

The independent variables involve sources of CEO power in the organization or

forces that may independently affect the dependent variables. The general model

includes three classes of dependent variables. The first class involves the size and

structure of the board of directors, the second type involves the firm's strategies, and the







44
third class of dependent variable consists of financial performance measures which are

accounting and market based.

Independent Variables

Ownership concentration

As discussed in the theoretical development, corporate shareholders can

influence the firm, thereby limiting the power of corporate executives, through

concentrated equity holdings. Individuals or groups with significant shareholdings are

more likely to exercise influence over the firm (Marris, 1964; McEachern, 1975).

Indeed, a significant body of empirical work suggests that if at least one equity holder

maintains at least five percent of the shares outstanding, he or she exercises sufficient

influence to restrain self-seeking behavior on the part of non-owner managers

(Gomez-Mejia, et al., 1987; Hunt, 1986; McEachern, 1975; Tosi & Gomez-Mejia,

1989). Thus, firms where at least one non-manager holds a five percent or greater

equity stake are classified owner controlled (N=76). Firms that feature a manager

with at least a five percent stake are owner managed, and these firms are expected to

behave like owner controlled firms (N=6). Finally, if no single individual,

institution, or group holds at least five percent of the outstanding shares, the firm is

classified as manager controlled (N=61). Manager controlled firms were coded as

one and owner controlled firms, which included the six owner-managed firms, were

coded as zero.









Firm size

The executive succession literature suggests that firm size and previous

financial performance affect the power exercised by CEOs (Dalton & Kesner, 1987;

Datta & Guthrie, 1994; Fizel & Louie, 1990; Harrison, et al., 1988). Firm size may

be operationalized in several ways, including assets, annual revenues, and number of

employees, or an index of these measures, and each of these approaches has been

used in previous research (Ciscel & Carroll, 1980; Gomez-Mejia et al., 1987;

O'Reilly et al., 1988; Tosi & Gomez-Mejia, 1989). In this study, the measure of

organizational size is a composite index constructed from the standardized number of

employees, assets, and sales, where assets and sales are scaled in constant 1988

dollars and subjected to a principal components factor analysis. Such an index

represents a reliable method of capturing a firm's magnitude from multiple

perspectives (Gomez-Mejia et al., 1987; Tosi & Gomez-Mejia, 1989; Werner & Tosi,

1995). The factor accounted for 80 percent of the variance in the three variables.

The factor loadings of assets, sales, and number of employees were .89, .94, and .84,

respectively.

Firm performance measures

Several measures of corporate financial performance are available but all are

generally regarded as either accounting-based or market-based indicators. This

research involves multiple measures of performance, including return on equity,

return on assets, return on investment, and the market to book ratio.









Return on equity. Return on Equity is an accounting-based measure of

performance that is computed by dividing net accounting profit by the average market

value of outstanding common equity. ROE captures part of the yield to shareholders

on their direct investment in the firm. The remainder of the yield is captured by the

change in market value of equity.

Return on assets. Return on Assets is an accounting-based performance

measure that captures the productive yield of the firm's asset base. ROA is calculated

by dividing earnings before interest but after taxes by the average net book value of

the firm's assets.

Return on investment. Like ROA, a firm's Return on Investment is an

accounting-based performance measure that captures the quality of the unsealed profit

figures. ROI is computed by dividing the net accounting profit by the cost of capital

investment made by the firm during the year.

Market to book value. The market to book value ratio for a firm is a market-

based performance measure that is calculated by dividing the year end market value of

outstanding equity by the book value of equity. The difference between these two

equity values indicates the market's estimation of the potential for the firm. As this

difference increases, so does the ratio of market to book value, and hence the value

accrued to shareholders.

A factor analysis of the four performance measures was performed so that a

determination could be made as to the convergence of the individual performance

measures (Tosi & Gomez-Mejia, 1994). A principal components factor analysis of







47
the standardized values of the four measures revealed that a single factor, representing

firm performance, was appropriate. This factor accounted for 71 percent of the

variance in the four variables and the factor loadings for ROA, ROE, ROI, and

market to book value were .92, .80, .89, and .71, respectively.

After constructing the performance index, an additional step was taken to

prepare these data for analysis. First, the index described in the previous paragraph

was created from industry population averages (as reported in Compustat) of each of

the 90 industries represented in the sample. An annual firm-from-industry deviation

score was then computed using the difference of the firm-specific and industry

average performance indices. As a result, a score was obtained which represented

each sample firm's performance compared to its own industry average. A single

value of the performance deviation score was computed for each of the two panels, by

averaging the values over each year in the panel, and these average deviation scores

were used in the analysis to control for industry effects (Hoskisson, Johnson, &

Moesel, 1994).

This study utilized the performance index for the pre- and post-succession

periods described in the first section of this chapter. The pre-succession performance

index is based on the average of the industry deviation scores for the firm in during

the pre-succession panel, 1984-1987, and the post-succession performance index

consists of the average of the industry deviation scores for the firm in the post-

succession panel, 1992-1994.









In each case, the pre-event panel includes the four consecutive years

immediately preceding the succession event period (1988-1991). All dollar scaled

financial measures were adjusted to 1988 through the GDP Deflator to ensure

comparability across the time panels.

Duality

Duality indicates that the CEO also holds the board chair position (Dalton and

Kesner, 1987). As a dichotomous variable, duality was coded as one for the new

CEO if within 12 months of taking the chief executive post he or she was also made

board chair. If the CEO did not hold the board chair position in addition to the chief

executive post, duality was coded as zero.

Origi

The chief executive's origin was another control variable. It is defined by the

last job held prior to becoming the CEO at the focal firm (Dalton & Kesner, 1985;

Friedman & Singh, 1989; Wiersema, 1992). If that position was outside the focal

firm, successor origin was coded as one and that individual would be referred to as an

outside successor. If the successor CEO was promoted from within the firm,

successor origin was coded as zero and that individual would be an inside successor.

Disposition

Predecessor disposition refers to the circumstances under which the previous

CEO left that position. Greiner and Bhambri (1989) suggest that a successor CEO is

likely to experience greater post succession discretion if the previous CEO was forced

from office. Thus, if it could be determined from text sources (the proxy statements









or the Wall Street Journal) that the predecessor had retired, died, or had voluntarily

taken another position, this dichotomous variable was coded as zero. For example,

when Robert Schoellhorn retired from Abbot Laboratories his departure was

announced in the Wall Street Journal as a "planned retirement". If the text sources

suggested that the predecessor was forced to resign or retire, as in the case of

William Agee at Morrison-Knudsen whose departure was reported in the Wall Street

Journal as "essentially a firing," this variable is coded as one. If no determination

could be made about the predecessor's disposition, this variable was coded as zero

(voluntary).

Dependent Variables

Board of director effects

The dependent variables over which the CEO exercises influence fall into two

classes. The first class of outcomes involves the structure of the board of directors.

These variables, like the other endogenous variables, were examined both at their

absolute post-succession levels and as change scores and their operationalization is

discussed in more detail below.

Outsider ratio. The outsider ratio represents the number of non-manager

(outside) directors on the firm's board divided by the total number of board members.

This variable was tracked prior to and after the succession event panel. The outsider

ratio for the pre-succession panel is the average of this ratio over the four year panel,

1984 to 1987, while the outsider ratio for the post-succession panel is the average of

this ratio over the three year panel, 1992 to 1994. A change score was also computed







50
for this variable. It is the percentage change from the pre-succession panel average to

the post-succession panel average. Hence, in any year t, outsider ratio (OR) is

Oi = number outside directors
total number of directors.

Director turnover. Board of director turnover consisted of two variables. The

first was outsider turnover, which was the average number of outside directors

leaving the board each year during the two periods, before and after the succession

event period. Thus, pre-succession outsider turnover consisted of the average number

of outside directors who were replaced in the four year panel from 1984 to 1987.

Post-succession outsider turnover was measured in the same manner except that it was

computed based on outsider director exits in the post succession panel from 1992 to

1994. Insider turnover, in both the pre- and post-succession periods, was computed

in the same way except that it was the average of employee-director exits in the panel

years. A change score was also computed for total director turnover. It was the

percentage change in director turnover from the pre-succession panel to the post-

succession panel. Hence, director turnover (DT) in year t, of either inside or outside

directors, was

DTt,= number of board members leaving
total number of board members.


Director compensation. O'Reilly et al. (1988) observed that many outside

directors of publicly held firms are executives of other publicly held firms. In

addition, they found that as the level of pay of outsider directors (from their own









firms) increased, executive compensation at the focal firm also increased. As a

result, this research includes the cash pay received by outside directors in their

positions as executives at their own firms. Outside director pay was the average of

cash pay received by outside directors in each panel year. Because it was not always

possible to compare the same cohort of outside directors, the average in any given

year may comprise the pay of different individual directors. Furthermore, in 21 cases

there was no external pay data available for any outside director, and director pay was

excluded for these cases.

Strategy effects

The second class of dependent variables involves strategic choice, which

includes corporate level and functional level strategies of the firm.

Corporate strategy. Corporate level of strategy was operationalized by the

absolute level of product-market diversification. For this variable, an unweighted

product count measure of diversification (Lubatkin, Merchant, & Srinivasan, 1993;

Wood, 1971) was employed. Diversification strategy was measured with two

independent continuous variables in a procedure developed by Wood (1971) and

validated in separate studies by Lubatkin et al. (1993) and Hoskisson et al. (1993).

The measures, which assess related and unrelated diversification, were calculated with

an unweighted count of SIC codes as reported in Compustat. The number of product

markets, as indicated by SIC code, which matched the third and fourth digits of the

firm's primary business code were counted as related business units. Hence, the

value of the related diversification variable was simply the number of these units.







52
Similarly, the number of product markets which did not match the first two digits of

the code for the firm's primary line of business were counted as unrelated business

activities.

This method of assessing diversification created two continuous variables

which capture the scope of a firm's diversification. Measuring both related and

unrelated diversification in this manner provides two advantages. First, the measure

is fairly easy to obtain and calculation is straightforward. Second, measurement of

both related and unrelated diversification as independent constructs is more consistent

with the original theoretical conceptualization of diversification (Hill & Snell, 1989;

Rumelt, 1982; Wrigley, 1971).

Functional/business level strategy. Two functional strategies were involved in

this research. The first functional strategy focuses on CEO compensation, both the

level and risk of compensation to the chief executive officer. The absolute level of

CEO cash pay was determined from proxy statements for each of the years in the pre-

and post-succession panels. These values were then averaged over the four years of

the pre-succession panel (1984-1987) and the three years of the post-succession panel

(1992-1994). Previous research has noted that total cash remuneration of the CEO is

an excellent proxy of total compensation (Gomez-Mejia et al., 1987; O'Reilly et al.,

1988; Werner & Tosi, 1995), which often includes deferred pay as well as stock

options and grants.

Compensation strategy also included a measure of the compensation risk that

captures the relative value of the fixed versus variable components of CEO pay, with









a larger variable component (usually a bonus) indicating greater compensation risk.

Pay risk was calculated in the two panels by computing the proportion of total cash

pay that was contingent on performance (i.e., a bonus). This information was derived

form the proxy statements for the sample firms and has been shown to be an adequate

measure of pay risk (Werner & Tosi, 1995). Hence, pay risk (PR) in year t was

PR = variable portion of total pay
total cash pay.

The second the functional strategy assessed is the firm's investment in

technology and innovation. Technology strategy is operationalized by the intensity of

research and development expenditure (Hill & Snell, 1988). This is calculated by

dividing the firm's annual research and development expenditures by total annual net

sales, and then subtracting the average of this ratio for the industry. The resulting

ratio captures the extent to which the firm invests in long-term technological

innovation in proportion to its size and industry (Hill & Snell, 1989; Hoskisson et al.,

1994; Zahra & Covin, 1993). Hence, research and development intensity (RD) in

year t was

RD =_ research and development expenditure
annual sales volume.

Performance effects

Post succession performance was calculated with the same industry-deviation

score index as was pre-succession performance only in the three years, 1992-1994,

after the event period. Like the pre-succession panel, the average value of this index







54
represents the firm's industry-adjusted performance in the post-succession period. A

change score was also computed for the performance deviation score. This change

score was calculated by assessing the percentage change in the performance deviation

score between the pre- and post-succession periods.

Summary

This chapter described the sample and regression equations used to test the

hypotheses about the post-succession consequences of CEO power. The variables

were also presented and their operationalization was discussed. The results obtained

from the research design described in this chapter are presented in the chapter that

follows.













CHAPTER 3
RESULTS


Each of the hypotheses developed in Chapter 1 was tested and the results of

these tests are presented in this chapter. In addition, the means, standard deviations,

and the simple correlations of the variables are reported in Appendix B.

Ordinary least squares regression was used to test the relationship between the

independent power variables and each of the dependent board-structure, strategy, and

firm-performance variables. The test for each hypothesis will be discussed beginning

with the examination of CEO power and board structure. In addition, as an attempt

to understand why power differs with respect to ownership influences, this chapter

will report the analyses of power within the sub-samples of manager controlled and

owner controlled firms. The first section reports the effects of the power measures on

board structure, the second section reports the effects of the power measures and

board structure variables on strategic choices, and finally section three reports the

effects of the power measures, the board structure variables, and the strategy

variables on firm financial performance. All beta coefficients reported in the text and

tables are standardized for comparability.









Board Structure

The results of the tests of CEO power and board structure on the ratio of

outside directors are reported in Table 3.1. Hypothesis one, which predicted that the

ratio of outside directors would increase to a greater extent in manager controlled

Table 3.1
Regressions on Outsider Ratio

Total Sample Manager Control Owner Control
Std. B Std. Er. Std. B Std. Er. Std. B Std. Er.
Control .032 .0430
Origin .147 .0385 .169 .0517 -.107 .0520
Duality .279* .0339 .388* .0321 .244* .0273
Disposition -.149 .0482 -.084 .0012 .136 .0380
Firm Size .253* .0002 -.120 .0001 -.166 .0003
R square .15 .20 .11
Adj. R square .09 .13 .07
F 2.76 2.79 2.95
P value .02 .02 .02
(*p<.05, **p-.01)

firms than in owner controlled firms, was not supported (B=.032, n.s.), though the

regression of outsider ratio on the CEO power indicators was significant. This result

suggests that power variables other than ownership influence may be associated with

differences in the outsider ratio. Indeed, CEO duality was associated with greater

increases in the outsider ratio (B=.279, p 5.05).

The power effects were also tested in the sub-samples of manager controlled and

owner controlled firms. As indicated, duality was related to the outsider ratio (B=.388,







57
p ; .05) in manager controlled firms. It appears that dual chief executives are better able

to influence the composition of the board of directors. Though the effect was smaller

in owner controlled firms (B=.244, p 5.05), dual CEOs were able to exercise power

through increases in the ratio of outside directors.

Two additional tests of the outside ratio regressions in manager controlled and

owner controlled firms were conducted. Significant differences between the parameters

in the two regressions on outside ratio were detected (F. 133,=3.78, p: .01) based on an

F test of the equivalence of regression parameters (Chow, 1960). As a result, a

likelihood ratio test of the two duality parameters was conducted (Greene, 1993). This

test showed that duality had a significantly larger effect in manager controlled firms

(X2(4)=12.62, pf .05). It appears that although dual CEOs in both manager controlled

and owner controlled firms affect the composition of the board of directors, this effect

is stronger when managers control the firm.

The results of the tests of CEO power on the rate of director turnover are

reported in Table 3.2. Hypothesis two, which predicted that director turnover would

increase to a greater extent in manager controlled firms than in owner controlled firms,

was not supported (B=.052, n.s.), though the data fit the board turnover model

reasonably well. This result suggests that power variables other than ownership influence

may be associated with differences in the rate of director turnover. Indeed, outside

origin of the successor CEO was associated with greater increases in director turnover

(B=.265, p5 .05).









Table 3.2
Regressions on Director Turnover

Total Sample Manager Control Owner Control
Std. B Std. Er. Std. B Std. Er. Std. B Std. Er.
Control .052 .1176
Origin .265* .1051 .273 .0019 -.102 .0520
Duality .042 .0925 .318* .0525 .145 .0273
Disposition .108 .1317 .114 .0670 .007 .0380
Firm Size -.001 .0001 .300 .0002 .124 .0003
R square .09 .18 .06
Adj. R square .05 .08 .02
F 2.51 2.63 1.56
P value .03 .03 .19
(*p_.05, **p_.01)

When tested in the sub-samples of manager controlled and owner controlled firms,

duality was found to increase director turnover (B= .318, p <.05) in manager controlled

firms. No such effects were found for owner controlled firms. It does, however, appear

that origin affects the rate of director turnover, overall.

The results of the tests of CEO power on board size and outside director

compensation are reported in Tables 3.3 and 3.4 respectively. Hypothesis three, which

predicted that board size would increase to a greater extent in manager controlled firms

than in owner controlled firms, was not supported (B= .073, n.s.). Similarly, hypothesis

four, which predicted greater increases in the level of external pay for outside directors

in manager controlled firms versus outside directors in owner controlled firms, was not

supported (B=-. 168, n.s.).









Table 3.3
Regressions on Board Size


Control
Origin
Duality
Disposition
Firm Size
R square
Adj. R square
F
P value
(*p<.05, **p<.<


Total
Std. B
.073
-.108
-.191
-.135
-.097
.08
.03
1.49
.20


Sample
Std. Er.
.7101
.6347
.5587
.7952
.0003


Manager
Std. B


-.081
-.118
-.149
-.153
.05
.00
.52
.67


Control
Std. Er.


.0415
.0445
.0016
.0002


Ol)


Table 3.4
Regressions on External Director Compensation


Total Sample
Std. B Std. Er.
Control -.168 .7165
Origin -.158 .5788
Duality .043 .6485
Disposition -.097 .7254
Firm Size -.127 .0001
R square .10
Adj. R square .00
F .58
P value .72
(*p<.05, **p5.01)


Manager Control
Std. B Std. Er.


-.035
-.077
-.104
.166
.06
.00
.50
.73


.0475
.0017
.0614
.0002


Control
Std. Er.


.0798
.0573
.0041
.0003


Owner
Std. B


-.030
-.115
-.049
-.175
.05
.01
1.24
.30


Control
Std. Er.


.0475
.0017
.0614
.0002


Owner
Std. B


.022
-.051
-.092
.094
.03
.00
.66
.62


--


--









Strategic Choices

The exercise of CEO power was expected to affect the firm's strategies differently

depending on whether the firm was owner or manager controlled. The effects of

managerial control in firms experiencing succession were assessed on the two corporate-

strategy variables. The results of the regressions on unrelated diversification are reported

in Table 3.5. Hypothesis five, which predicted that unrelated diversification would

increase to a greater extent in manager controlled firms than in owner controlled firms

was not supported (B= .060, n.s.). The effects of power on the unrelated diversification

Table 3.5
Regressions on Unrelated Diversification


Control
Origin
Duality
Disposition
Firm Size
Outsider Ratio
Board Turnover
Board Size
Director Comp
R square
Adj. R square
F
P value
(*p<.05, **p!<.01)


Total Sample
Std. B Std. Er.
.060 .2638
-.083 .2512
.108 .2139
-.181 .3011
-.072 .0001
-.052 .6050
.250 .2568
-.020 .0328
-.008 .0189
.10
.01
1.05
.40


Manager Control
Std. B Std. Er.


-.046
.105
.003
-.073
.118
.025
.019
.064
.08
.03
1.69
.14


.2825
.0145
.2044
.0002
.1036
.4457
.2761
.3464


Owner Control
Std. B Std. Er.


-.060
-.110
-.024
.038
.110
-.007
-.031
.094
.02
.00
.10
.99


.3400
.0121
.4463
.0001
.5831
1.180
.5455
.0423







61
form of corporate strategy were also tested in the sub-samples of manager controlled and

owner controlled firms. However, these regressions also yielded insignificant results.

The results of the regressions on the related diversification form of corporate

strategy are reported in Table 3.6. The prediction that related diversification would

decrease in manager controlled firms (Hypothesis 6) was supported (B=-.217, p: .05).

Another power indicator, CEO succession from outside the firm, was also associated with

decreased levels of related diversification (B=-. 192, p< .05).

Table 3.6
Regressions on Related Diversification


Control
Origin
Duality
Disposition
Firm Size
Outsider Ratio
Board Turnover
Board Size
Director Comp
R square
Adj. R square
F
P value
(*p 5.05, **p5.01)


Total Sample
Std. B Std. Er.
-.217* .1225
-.192* .1167
.108 .0993
-.181 .1398
-.072 .0007
-.052 .2810
.250 .1193
.026 .0152
-.008 .0002
.33
.21
2.58
.02


Control
Std. Er.


.1359
.0070
.0983
.0001
.1667
.2144
.1015
.0002


Control
Std. Er.


.1803
.0064
.2367
.0001
.2893
.6262
.1162
.2778


Manager
Std. B


-.202
-.234*
.186
-.176
.071
-.019
.068
.201
.21
.17
5.61
.01


Owner
Std. B


-.038
-.164
-.389
-.106
.190
.273
.108
.045
.11
.00
.68
.64







62
Within the owner and manager controlled sub-samples, only CEO duality

contributed to changes in corporate strategy in manager controlled firms (B=-.501,

p5.05). This result is consistent with the reduced levels of related diversification

associated with dual CEOs that was observed in the total sample. In owner controlled

firms, however, none of the power or board structure variables were found to predict

strategic shifts at the corporate level. It therefore appears that these effects are isolated

to manager controlled firms.

Table 3.7
Regressions on CEO Compensation


Control
Origin
Duality
Disposition
Firm Size
Outsider Ratio
Board Turnover
Board Size
Director Comp
R square
Adj. R square
F
P value
(*p<.05, **p..01)


Total Sample
Std. B Std. Er.
.134 .0327
.127 .0165
.121 .0223
.318* .0403
.052 .0001
.268* .0595
-.149 .0587
.073 .0024
-.088 .0029
.28
.19
2.58
.02


Manager Control
Std. B Std. Er.


.096
.241
.186
-.129
.278
.018
.040
.096
.20
.06
1.31
.29


.3846
.2930
.4700
.0001
.0105
1.017
.0542
.2516


Control
Std. Er.


.0205
.4599
.4396
.0002
.0014
.0529
.0118
.0011


Owner
Std. B


.119
.033
-.042
.172
.099
.079
.073
-.047
.05
.00
.69
.60







63
This research also examined differences in three functional level strategies among firms

experiencing succession. The results of the regressions on CEO compensation are

reported in Table 3.7. Hypothesis seven, which predicted that the level of CEO pay

would be higher in firms controlled by managers was not supported (B=. 134, n.s.).

Although CEO pay did not vary with managerial control of the firm, the level of CEO

compensation did increase when the predecessor was forced from office (B= .318,

p< .01) and when the ratio of outside directors was larger (B=.268, p< .01). However,

no effects for CEO pay were found in the owner and manager controlled sub-samples.

Table 3.8
Regressions on CEO Compensation Risk


Control
Origin
Duality
Disposition
Firm Size
Outsider Ratio
Board Turnover
Board Size
Director Comp
R square
Adj. R square
F
P value
(*p_<.05, **p:_.01)


Total Sample
Std. B Std. Er.
-.341* .0325
.075 .0164
-.268* .0221
.215* .0400
.115 .0001
-.058 .0591
-.108 .0583
-.011 .0024
-.090 .0030
.23
.15
2.43
.02


Manager Control
Std. B Std. Er.


.086
-.315*
-.097
-.151
-.247
-.109
.097
.062
.15
.11
3.73
.01


.1317
.2873
.0094
.0001
.1820
.2233
.0618
.1087


Control
Std. Er.


.2916
.3000
.0933
.0001
.4622
.7990
.0108
.0079


Owner
Std. B


.109
-.158
.156
-.121
.110
-.110
.047
.070
.10
.00
.83
.57







64

The results of the regressions on compensation risk are reported in Table 3.8.

Hypothesis eight, which predicted that the level of CEO compensation risk would be

lower in manager controlled firms, was supported (B=-.341, p .01). In addition, pay

risk decreased when the successor CEO also assumed the board chair position (B=-.268,

p < .05). Conversely, pay risk was higher when the predecessor CEO was forced out of

the firm (B=.215, p. .05). Compensation risk effects were also examined in the sub-

samples of manager controlled and owner controlled firms. Within the sub-sample of

manager controlled firms, CEOs with dual appointments had lower levels of

compensation risk (B=-.315, p<.05).

The final strategic choice on which the effects of managerial control was assessed

involves the firms technology strategy and the results of these regressions are reported

in Table 3.9. Hypothesis nine, which predicted that manager controlled firms would

have less investment in technology than would owner controlled firms, was not supported

(B=-.021, n.s.). However, one of the board influence measures, director turnover, was

associated with a strategy of reducing the firm's investment in technology and research

(b=-.432, p<.01). The regressions on technology strategy did not detect any other

predictors of functional strategy in manager controlled firms nor were any functional

strategy effects found in the owner controlled sub-sample.

Ultimately the structure of the board of directors, the firm's strategies, and the

power of the chief executive vis a vis the board of directors, are of interest because they

are expected to affect the firm's financial performance. The results of the tests of

performance effects are presented below.









Table 3.9
Regressions on Technology Strategy


Control
Origin
Duality
Disposition
Firm Size
Outsider Ratio
Board Turnover
Board Size
Director Comp
R square
Adj. R square
F
P value
(*p<.05, **p5.01)


Total Sample
Std. B Std. Er.
-.021 .0741
.193 .0375
.023 .0505
.045 .0912
-.135 .0002
-.128 .2818
-.432* .1196
-.066 .0153
.071 .0065
.20
.11
2.32
.05


Manager Control
Std. B Std. Er.


.060
.108
.124
-.056
.209
-.160
.040
-.074
.10
.00
.90
.53


.1074
.0053
.0050
.0001
.2234
.0840
.2611
.0126


Owner Control
Std. B Std. Er.


.295
-.166
-.023
.151
-.109
-.488
.128
-.041
.19
.00
.61
.65


.4879
.1728
.0696
.0003
.1328
.4469
.0375
.1417


Performance Effects

The results of the test of the power and performance hypothesis are reported in

Table 3.10. Hypothesis ten, which predicted that manager controlled firms would

experience lower levels of performance, was supported (B=-.361, p .05) and the data

fit the model well, explaining fifty-five percent of the variance. Thus, post-succession

performance is lower in manager controlled firms. The financial performance of the

firms also suffered when the successor CEO originated from outside the firm (B=-.368,

p .05) or when the predecessor CEO had been removed from office (B=-.709, p < .01).









Conversely, firm financial performance was better as the level of related diversification

increased (B=.337, p<.01).

Table 3.10
Regressions on Firm Performance

Total Sample Mgr. Control Owner Control
Std. B Std. E Std. B Std. E Std. B Std. E
Control -.361* .925
Origin -.368* .607 -.393 .374 .031 1.344
Duality .155 .749 -.315 2.690 -.037 .935
Disposition -.709" .779 -.162 .095 .030 .073
Firm Size .032 .001 .168 .001 .093 .001
Outsider Ratio -.301 1.027 -.189 .153 .147 .392
Board Turnover .173 1.311 -.046 9.037 .162 2.021
Board Size .003 .248 .013 .356 .093 .001
Director Comp .213 .204 .222 .263 .150 .198
Unrelated Div. -.066 .779 -.083 1.530 -.015 .885
Related Div. .337* 4.870 .014 3.082 .160 .428
CEO Comp .227 1.901 .107 1.576 .276 2.819
CEO Comp Risk .125 4.901 .213 4.320 .119 3.72
Technology .060 1.377 .084 1.739 .069 .649
R square .66 .30 .10
Adj. R square .55 .05 .00
F 5.80 1.12 .72
P value .001 .38 .67
(*p<.05, **p<.01)







67
These relationships were also examined in the manager controlled and owner

controlled sub-samples. These equations failed to achieve significance and are therefore

not discussed further.

Summary

This chapter presented the regression results for the tests of the power hypotheses.

Although the results were mixed, a clear pattern of power differences emerged which

suggests that CEO power is greater in manager controlled firms, and in cases of a dual

appointment or origination from outside the firm. The implications of these results for

theories of managerial power are discussed in the chapter that follows.













CHAPTER 4
DISCUSSION


This study investigated the consequences of CEO power across several contexts

which are thought to permit varying levels of power. The results presented in the

previous chapter suggest that CEO power has important consequences for the board of

directors, the firm's strategies, and ultimately for the firm's financial performance.

These consequences and their implications for theories of managerial power will be

discussed in this chapter. In addition, the implications for management research will be

explored in the chapter.

On the whole, the power results are consistent with previous research which

suggests that chief executives may acquire and use power in different ways (Pfeffer,

1981; 1992). In addition, these results extend previous power research in a number of

important ways. The results suggest, for example, that the theory of managerial

capitalism can be meaningfully integrated into the executive succession and managerial

power literatures. The results also suggest that CEO power has significant consequences

for firms undergoing succession. The implications of these results for theories of

managerial succession are discussed in the following section and the succeeding sections

will explore the implications of the results for the theory of managerial capitalism as well

as the executive power literature.









Implications for CEO Succession

Executive succession is an important event in the life of any business organization

and this study has important implications for the understanding of this event. The key

finding of this work is that the power dynamics following a succession event differ in

firms with different ownership structures.

Consistent with previous work in executive succession, this study found that the

origin of the successor CEO, as well as the disposition of the predecessor, are important

to post succession strategies employed by the firm (Friedman & Singh, 1989; Kesner &

Sebora, 1994). CEOs who originate from outside the firm were associated with

increased director turnover which suggests that outsiders, more so than inside successors,

have greater ability to shape the board of directors to their liking. Outside successors

also were associated with changes in corporate strategy.

Successor CEOs who followed a dismissal engaged in more self-serving corporate

strategies. In addition, their pay was higher than CEOs who followed a voluntary

disposition of the predecessor. These results are consistent with the exercise of CEO

power as described in this research.

It was more likely that the level of related diversification would decrease when

the successor CEO came from the outside. Because related diversification provides

benefits for the firm's shareholders but only to a lesser degree to the CEO, powerful

chief executives may reduce the level of related diversification in favor of strategies like

unrelated diversification or near-term sales growth. Both unrelated diversification and

sales growth would increase firm size, which tends to benefit the CEO because firm size







70
is positively associated with the level of CEO pay. However, unlike related

diversification, unrelated diversification does not utilize significant resource sharing and,

as a result, tends to benefit shareholders to a lesser degree than managers. Indeed,

outside successors were associated with inferior post-succession financial performance

as anticipated by the preceding discussion of corporate strategy.

These results are especially interesting in light of the positive stock market

announcement effects reported for outside successors (Beatty & Zajac, 1987; Reinganum,

1985). While the evidence shows that equity markets expect more from an outside

successor, a notion that is consistent with outside successors having more discretion, the

results shown here suggest that longer-term post-succession performance is worse, rather

than better, under an outside successor. Thus, outside successors appear to have greater

power and either they utilize it in ways that are less beneficial to the firm or the systemic

effects that reinforce the lower performance are not easily overcome. Of course this

study represents a single sample from a limited span of time and, as a result, should be

interpreted with caution.

Like the successor's origin, the predecessor's disposition also has important

consequences for post succession firm strategy and performance. In the case of the

deposed predecessor, the successor CEO received a higher level of total compensation,

but faced increased compensation risk. The higher pay and higher pay risk may

represent a rational response for the firm that has found it necessary to dismiss its chief

executive because by increasing the pay level for the new CEO, the firm may make the

job more attractive to prospective executives. The increased pay risk shifts more of the







71
firm's future performance risk to the CEO and as such acts as an incentive alignment

device. To the extent that this arrangement aligns the interests of the new CEO and the

firm, it may be a rational response to poor firm performance and CEO dismissal.

Implications for CEO Power

The acquisition and exercise of CEO power in organizations has a number of

important consequences for the board of directors. Because the board of directors is the

principal mechanism for shareholders to exert control over the firm, manipulation of the

board by the CEO may result in outcomes that would benefit the CEO more than the

shareholders. The pattern of results suggests that when CEO discretion is high, the

board is subject to greater manipulation by the chief executive. One way that powerful

CEOs manipulate or control the board of directors is by increasing the number of

outsiders on the board. Increasing the ratio of outsiders results in increased CEO power

(Ocasio, 1994) and executive compensation (Westphal & Zajac, 1995a). Further,

powerful CEOs will select unrelated diversification because this strategy enhances firm

size which is itself associated with higher levels of CEO pay.

CEO Power and the Board of Directors

The control of key contingencies presents the successor CEO with one way to

acquire and use power within the firm. Another perspective on managerial control, that

advanced by managerial capitalism, is also addressed by this research. In the section that

follows, the implications for the theory of managerial capitalism of the results detailed

in the previous chapter are discussed.









Implications for managerial capitalism

In manager controlled firms, dual CEOs increase the outsider ratio and the rate

of director turnover. As a result, they are better able to reduce their employment risk

as well as increase their level of total compensation. In contrast to dual CEOs in owner

controlled firms, dual CEOs in owner controlled firms are not associated with director

turnover. Their dual appointment counterparts in owner controlled firms tend to increase

the proportion of outsiders but are evidently restrained from consolidating power in the

same ways as dual CEOs in manager controlled firms. Thus, the influence of outside

owners appears to effectively limit CEO power in owner controlled firms while CEOs

in manager controlled firms have greater discretion to alter the board of directors in ways

that consolidate their control. This result is important in that the higher level of CEO

power in manager controlled firms could explain the lower performance of firms

controlled by managers that was found in this study. These results are also consistent

with previous research which found that CEOs in manager controlled firms have more

influence over their pay levels and pay risk (Tosi & Gomez-Mejia, 1989) and that this

influence over pay decisions is also associated with performance decrements (Tosi &

Gomez-Mejia, 1994).

Implications for managerial power

Dual CEOs in manager controlled firms were able to manipulate the board by

replacing directors and appointing a greater proportion of outside directors. The

consequences associated with larger outsider ratios, as discussed previously, suggest that

inside directors are more effective in terms of limiting the power of the CEO. Because







73
this result is contrary to the classical agency theory argument to the effect that outside

directors are more likely to act in the interest of owners (Walsh & Seward, 1990), it

suggests that outside directors are more dependent on the CEO (Baysinger & Hoskisson,

1990; O'Reilly et al., 1988).

Successor CEOs from outside the firm and those with dual appointments appear

to have the most power. The results suggest that outside successors are better able to

acquire and exercise power by creating dependency relations and increasing the rate of

director turnover. Specifically, when the incoming CEO is from outside the firm, post-

succession board turnover is increased. By replacing directors who may oppose him or

her, whether insiders or outsiders, the new CEO establishes control over the board. This

result suggests that, consistent with the sources of organizational power discussed in the

first chapter, the new CEO acquires and exercises power through political influence and

by creating dependencies. Uncertainty coping is one kind of dependency that a new CEO

may create by replacing existing board members because new directors, particularly those

who are not executives of the firm, are more dependent on the CEO for information and

other resources.

Increased director turnover is associated with reductions in the firm's commitment

to technology. Reducing the firm's commitment to technology is associated with reduced

levels of financial performance (Baysinger & Hoskisson, 1989). Although a firm's

commitment to technology is associated with higher long-term profitability, these longer-

term payoffs provide the CEO very small immediate gains and therefore new chief







74
executives (who have the opportunity) will select strategies that reduce the commitment

to technology and instead increase short-term firm size and scope.

It therefore appears that outside successors have more decision latitude to make

director appointments as well as to force opposing directors from the board. It also

appears that the shifts in the board of directors can result in strategic changes for the

firm. To the extent that new CEOs can shape the board of directors as desired, the chief

executive will have a greater ability to implement self-serving strategies than CEOs with

less power.

Similarly, CEOs with dual appointments tend to increase the proportion of

outsider directors on the board. Like replacing existing directors, placing outsider

directors increases the board's dependence on the CEO and as a result, CEO power is

increased. Increases in the outsider ratio, in turn, are associated with increases in the

level of CEO pay. Because outside directors must rely, to a great extent, on the CEO

for information about the firm's performance, these directors are more likely to grant

higher pay or to otherwise act as the CEO requests. Both the increases in director

turnover and outsider ratio suggest that the incoming CEO has exercised power within

the firm. As a result of the power and control of the CEO vis a vis the board of

directors, strategies and firm performance also differ among firms.

CEO Power and Strategy

Implications for managerial capitalism

CEO power also appears to be stronger in firms effectively controlled by

managers because the level of performance enhancing related diversification is lower in







75
manager controlled firms. This result is consistent with empirical research that addresses

the question of diversification and control of the firm (Amihud & Lev, 1981; Hill &

Snell, 1988; 1989). Because related diversification has few short-term benefits for the

CEO, it is unlikely to be favored by powerful CEOs who prefer the more immediate size

gains from unrelated acquisitions. In particular, dual CEOs in manager controlled firms

reduce the firm's level of related diversification and increase the level of unrelated

diversification. As previously discussed, these are relatively self-serving strategies on

the part of the CEO which appear to be more easily accomplished in manager controlled

firms.

CEOs in manager controlled firms also experience lower compensation risk which

is consistent with the empirical work examining the pay of existing chief executives

(Gomez-Mejia et al., 1987; Tosi & Gomez-Mejia, 1989; 1994). It appears that when

managers control the firm, the CEO may influence compensation policy such that his or

her pay risk is reduced. Such activity weakens the link between CEO pay and firm

performance. Like the corporate strategy issue discussed previously, dual CEOs in

manager controlled firms also experience lower pay risk which is indicative of their

relative power vis a vis the board of directors. Thus, in manager controlled firms, the

successor CEO will likely receive a greater fixed pay component (i.e., have

compensation risk), particularly when the successor CEO also holds the board chair

appointment.









Implications for managerial power

The effects of CEO power are not limited to the composition of the board of

directors and to corporate-level strategies. These results suggest that powerful CEOs

engage in corporate-level strategies that provide little benefit for shareholders but that

increase the size of the firm. When firm size is increased, particularly through unrelated

diversification, the chief executive tends to receive an increased level of compensation

(Finkelstein & Hambrick, 1989; Kerr & Kren, 1992). Specifically, when the predecessor

CEO was removed from office involuntarily, the successor CEO, who should have

greater discretion due to the expectations associated with the predecessor's dismissal,

increased the levels of unrelated diversification and reduced the levels of related

diversification. Thus, successor CEOs following the dismissal of their predecessor

exercise power to a greater extent than do their counterparts whose predecessors leave

voluntarily. Similarly, successor CEOs from outside the firm were found to reduce the

level of related diversification to a greater extent than inside successors. This implies

that resources were shifted to strategies that may not improve firm performance. These

results are consistent with the argument that successor CEOs from outside the firm, and

those who take over after their predecessor was forced to leave, have more power. In

addition, these results support the notion that this power is, in turn, used by the CEO to

alter the strategic direction of the firm in ways that are more favorable to the CEO and

the firm's managers than for the shareholders. As a result, the performance of firms

with new CEOs from outside the firm or those that dismissed the predecessor CEO is

worse than those with insider successors or that experience a voluntary succession.







77
The changes to the board of directors discussed in the previous sections,

especially increases in the ratio of outside directors, are associated with higher levels of

CEO pay, irrespective of whether the CEO had the dual appointment or came from

outside the firm. Of course, this study found that dual CEOs are better able to appoint

directors of their choice and as a result may be expected to have higher pay. However,

these results suggest that by appointing more outside directors to the board, a new chief

executive can increase the level of his or her pay without the dual appointment or by

coming from outside the firm. As a result, even in firms where CEO power is relatively

low, raising the proportion of outside directors would appear to benefit the CEO.

Indeed, these results are consistent with previous research which shows that higher

proportions of outside directors are associated with higher levels of CEO compensation

(O'Reilly et al., 1988; Westphal & Zajac, 1995a; 1995b).

In addition to their influence over the outsider ratio and corporate strategy, CEOs

with dual appointments have lower levels of pay risk. Like their ability to influence the

composition of the board, it appears that the dual appointment provides the chief

executive with a means to influence pay decisions independently of the board's

composition. It is possible, as suggested by the theory of human capital (Becker, 1964;

Finkelstein & Hambrick, 1989), that the reduced level of pay risk represents legitimate

recompense for the additional skills required to effectively discharge the duties of the

CEO and board chair positions. However, in the context of the other results of this

study which indicate that dual CEOs engage in strategies that are less than optimal for







78
shareholders, it would appear that the reductions in pay risk for dual CEOs are not based

on skills or abilities but on power vis a vis the board of directors.

CEO Power and Firm Performance

Implications for managerial capitalism

Beyond the implications of CEO power for board structure and firm strategy, this

study examined the effects of CEO power on firm financial performance. In their

entirety, these results suggest that CEOs are better able to acquire and exercise power

under conditions of high managerial control. High managerial control permits the new

CEO to fully exercise his or her power and as a result manager controlled firms under-

perform owner controlled firms, post succession. This result is consistent with empirical

evidence to the effect that owner controlled firms outperform manager controlled firms

in most contexts (Hunt, 1986; McEachern, 1975; Ware, 1975). It appears, therefore,

that CEOs in manager controlled firms have greater relative power and that these power

differentials of CEOs in owner controlled and manager controlled firms could account

for the performance differences observed between firms controlled by owners and

managers.

Nevertheless, because this study examined only firms experiencing succession,

these power differentials may be less important when the chief executive position is

stable. However, empirical work examining CEO power in firms not experiencing

succession have reported the same kinds of outcomes, associated with CEO power, for

the board of directors and executive compensation that were discovered for succession

firms in this research (Finkelstein & Hambrick ,1989; Westphal & Zajac, 1995a; 1995b).







79
As a result, it is reasonable to conclude that CEOs in manager controlled firms have

relatively more power than their owner controlled counterparts, whether or not he or she

has recently ascended to the position. In addition, it seems reasonable to conclude that

the power differences between CEOs in owner and manager controlled firms have

significant consequences for firm performance.

Implications for managerial power

Outside successors, who tend to reduce related diversification, are associated with

a lower firm performance. Successor CEOs who follow a predecessor who was

dismissed reduce related diversification while increasing the firm's unrelated business

component. These successor CEOs are also associated with lower financial performance.

Indeed, the magnitude of the effect on performance of a successor following a dismissed

predecessor is, in fact, nearly twice a strong as the effect of an outside successor. The

results suggest, then, that successor CEOs who follow a dismissal are indeed very

powerful but not, perhaps, powerful enough to alter the course of the firm's

performance, at least in the years represented in the post-succession panel.

One important issue bears mention here, however. Empirical evidence indicates

that poor financial performance often occurs prior to a CEO's dismissal (Dalton &

Kesner, 1987; Datta & Guthrie, 1994; Furtado & Karan, 1990). It is entirely possible

that the poor financial performance and strategic shifts associated with a successor CEO

following a dismissal are the result of the successor's efforts to reverse the failed course

of his or her predecessor and not the successor's self-maximizing acquisition of power.

Given the relatively brief time-frame for the post-succession results in this research (four







80
years), it is possible that the new CEO has not yet reversed the fortunes of the firm. As

a result, these results should be interpreted with caution. One other result supports this

alternative interpretation of the predecessor disposition issue. When the successor CEO

follows a dismissal, their level of pay is higher than in a voluntary succession event.

Moreover, their level of compensation-risk is higher than for CEOs who follow a

voluntary event. Thus it appears that firms that dismiss the predecessor CEO adopt a

rational approach to compensating the successor CEO in that they pay relatively more

in an absolute sense but they also put more of that CEO's pay at risk. Ultimately these

results suggest that although successor power is high when the predecessor was

dismissed, on the average firms in these circumstances adopt a rational response to

remunerating the successor CEO.

Suggestions for Future Research

The implications of CEO power exercised after stepping into the position provide

a rich foundation for future research. One area for future study, successor CEO origin,

was discussed previously in this chapter. This research found that outside successors

were associated with worse, rather than better, financial performance that were inside

successors. Because this result is contrary to previous research, further study is needed.

It should be noted, however, that most of the evidence supporting the positive impact of

outside successors is derived from the event-study methodology, which examines stock

market reaction to particular information. Because the event window is necessarily

restricted to market reaction near the release of the key information, it may fail to

capture the longer-term impact of a CEO succession event. In addition, the stock market







81

reaction is an aggregate anticipation of future firm performance and may not reflect a

firm's actual financial performance after the succession event.

Another future research possibility that is highlighted by this study is the question

of differential prediction of CEO succession in firms controlled by owners or managers.

In firms controlled by owners, for example, CEO succession may be better predicted by

firm performance because of the tighter link between firm performance and the fortunes

of the CEO in these firms (Tosi & Gomez-Mejia, 1994). Conversely, in firms controlled

by managers, tournament theory concerns, such as the pay differential among top

executives (Ocasio, 1994), may better predict CEO turnover because the opposing

political behavior of top managers would be more relevant in these firms.

The question of how shareholder interests are best protected is a third possibility

for future research. Although agency theory suggests that outside directors are better

advocates of the ownership position (Walsh & Seward, 1990), the evidence is to the

contrary here and elsewhere (O'Reilly et al., 1988; Westphal & Zajac, 1995a; 1995b).

Some future research is needed to reconcile the theoretical position and the empirical

results.

The research model and results of this study contributed to the examination of

firm-level consequences of different levels of CEO power and will improve the

understanding of differences among firms in board composition and structure, as well as

strategy and performance. However, this research also suggests CEO power, and

therefore board composition, strategy and performance, will differ in firms which have

not experienced a recent succession event. Indeed, this work serves as an initial link







82
between the literature on executive power (e.g., Boeker, 1992; Westphal & Zajac,

1995a; 1995b) and that examining executive succession (e.g., Brown, 1982; Fizel &

Louie, 1990; Kesner & Sebora, 1994). Future research that integrates the CEO power

and executive succession literatures would greatly increase the knowledge generated in

these separate but related literatures.

Yet another area of future research that is suggested by this study is the nature

of the influence processes between the new CEO and the board of directors. Although

this research found differences in board composition and firm strategies that were

attributed to CEO power vis a vis the board of directors, these conclusions are inferences

based on available evidence. Further research examining the personal processes of

power acquisition and use would add significantly to the understanding of how chief

executives create, use and maintain power dependencies.

Summary

This research examined the post-succession consequences of CEO power in firms

with different ownership-control structures. On the whole, the findings were consistent

with the predictions of managerial capitalism and agency theory in that new CEOs in

manager controlled firms had relatively more power and were more likely to exercise

their discretion in ways that were beneficial to them. CEO power was found to stem

from holding a dual appointment as CEO and board chair, originating from outside the

firm, following a dismissal rather than a voluntary succession event, working in a

manager controlled firm, and by manipulating the board of directors. CEO power

resulted in changes to the composition of the board of directors and to the strategic







83
direction of the firm. More important, the power differences described in this research

were associated with performance differences such that in firms where CEO power was

greater, firm performance was reduced. This research, therefore, shows the importance

of CEO power and represents a step in integrating the executive succession and

managerial capitalism literatures.













REFERENCES


Allen, M.P. 1981. Power and privilege in the large corporation: Corporate control and
managerial compensation. American Journal of Sociology, 86: 1112-1123.

Amihud, Y., & Lev, B. 1981. Risk reduction as a managerial motive for conglomerate
mergers. Bell Journal of Economics, 12: 605-617.

Amit, R., & Livnat, J. 1988. Diversification and the risk-return trade-off. Academy of
Management Journal, 31: 154-166.

Andrews, K.A. 1971. The concept of corporate strategy. Homewood, IL: Irwin.

Ansoff, H.I. 1965. Corporate strategy: An analytic approach to business policy for
growth and expansion. New York: McGraw-Hill.

Barnard, C.I. 1938. The functions of the executive. Cambridge, MA: Harvard University
Press.

Baysinger, B.D., & Hoskisson, R.E. 1989. Diversification strategy and R&D intensity
in multiproduct firms. Academy of Management Journal, 32: 310-332.

Baysinger, B.D., & Hoskisson, R.E. 1990. The composition of boards of directors and
strategic control: Effects on corporate strategy. Academy of Management Review,
15: 72-87.

Baysinger, B.D., Kosnik, R.D., & Turk, T.A. 1991. Effects of board and ownership
structure on corporate R&D strategy. Academy of Management Journal, 34: 205-
214.

Beatty, R.P., & Zajac, E.J. 1987. CEO change and firm performance in large
corporations: Succession effects and manger effects. Strategic Management
Journal, 8: 305-318.

Becker, G. 1964. Human Capital. New York: Columbia University Press.









Berger, P.G., & Ofek, E. 1995. Diversification's effect on firm value. Journal of
Financial Economics, 37: 39-65.

Berle, A.A., & Means, G.C. 1932. The modem corporation and private property. New
York: Macmillan.

Bettis, R.A. 1981. Performance differences in related and unrelated diversified firms.
Strategic Management Journal, 2: 379-393.

Boeker, W. 1992. Power and managerial dismissal: Scapegoating at the top.
Administrative Science Quarterly, 37: 400-421.

Boeker, W., & Goodstein, J. 1991. Organizational performance and adaptation: Effects
of environment and performance on changes in board composition. Academy of
Management Journal, 34: 805-826.

Boeker, W., & Goodstein, J. 1993. Performance and successor choice: The moderating
effects of governance and ownership. Academy of Management Journal, 36: 171-
186.

Brown, M. 1982. Administrative succession and organizational performance: The
succession effect. Administrative Science Quarterly, 28: 165-183.

Canella, A.A., & Lubatkin, M. 1993. Succession as a sociopolitical process: Internal
impediments to outsider selection. Academy of Management Journal, 36: 763-
793.

Carroll, G.R. 1984. Dynamics of publisher succession in newspaper organizations.
Administrative Science Quarterly, 29: 93-113.

Child, J. 1972. Organization structure, environment and performance: The role of
strategic choice. Sociology, 6: 1-22.

Chow, G. 1960. Tests of equality between sets of coefficients in two linear regressions.
Econometrica, 28: 591-605.

Ciscel, D., & Carroll, T. 1980. The determinants of executive salaries: An econometric
survey. Review of Economics and Statistics, 62: 7-13.

Dahl, R.A. 1957. The concept of power. Behavioral Science, 2:201-215.

Dalton, D.R., & Kesner, I.F. 1985. Organizational performance as an antecedent of
inside/outside chief executive succession: An empirical assessment. Academy of
Management Journal, 28: 749-762.









Daily, C., & Dalton, D. 1994. Corporate governance and the bankrupt firm: An
empirical assessment. Strategic Management Journal, 15: 643-654.

Datta, D.K., & Guthrie, J.P. 1994. Executive succession: Organizational antecedents of
ceo characteristics. Strategic Management Journal, 15: 569-577.

Datta, D.K., Rajagopalan, N., & Rasheed, A.M. 1991. Diversification and performance:
Critical review and future directions. Journal of Management Studies, 28: 529-
558.

Davidson, W., Worrell, D., & Cheng, L. 1990. Key executive succession and
stockholder wealth: The influence of successor's origin, position, and age.
Journal of Management, 16: 647-664.

Dubofsky, P., & Varadarajan, P. 1987. Diversification and measures of performance:
Additional empirical evidence. Academy of Management Journal, 30: 597-608.

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

Fama, E.F., & Jensen, M.C. 1983. Separation of ownership and control. Journal of Law
and Economics, 26: 301-325.

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

Finkelstein, S., & Hambrick, D.C. 1990. Top-management team tenure and
organizational outcomes: The moderating role of managerial discretion.
Administrative Science Quarterly, 35: 484-503.

Fizel, J.L., & Louie, K.K.T. 1990. Ceo retention, firm performance and corporate
governance. Managerial and Decision Economics, 11: 167-176.

Friedman, S.D., & Singh, H. 1989. CEO succession events and stockholder reaction:
The influence of context and event context. Academy of Management Journal, 32:
718-744.

Furtado, P.H., & Karan, V. 1990. Causes, consequences, and shareholder wealth effects
of management turnover: A review of the empirical evidence. Financial
Management, 19(2): 60-75.

Gomez-Mejia, L.R., H.L. Tosi, and T. Hinkin 1987. Managerial control, performance,
and executive compensation. Academy of Management Journal, 30: 51-70.









Goodstein, J., & Boeker, W. 1991. Turbulence at the top: A new perspective on
governance structure changes and strategic change. Academy of Management
Journal, 34: 306-330.

Goodstein, J., Gautam, K., & Boeker, W. 1994. The effects of board size and diversity
on strategic change. Strategic Management Journal, 15: 241-250.

Greiner, L.E., & Bhambri, A. 1989. New ceo intervention and dynamics of deliberate
strategic change. Strategic Management Journal, 10: 67-87.

Grinyer, P., & Spender, J. 1979. Recipes, crises, and adaptation in mature businesses.
International Studies in Management and Organization, 9(3): 113-133.

Greene, W.H. 1993. Econometric Analysis. New York: Macmillan.

Hambrick, D.C., & Finkelstein, S. 1987. Managerial discretion: A bridge between
polar views of organizational outcomes'. In L.L. Cummings & Staw, B.M.
(eds.), Research in Organizational Behavior. Greenwich, CT: JAI Press. 9:369-
406.

Harrison, J.R., Torres, D.L., & Kukalis, S. 1988. The changing of the guard: turnover
and structural change in the top-management positions. Administrative Science
Quarterly, 33: 211-232.

Haveman, H.A. 1993. Ghosts of managers past: Managerial succession and
organizational mortality. Academy of Management Journal, 36: 864-881.

Helmich, D.L. 1974. Organizational growth and succession patterns. Academy of
Management Journal, 17: 771-775.

Helmich, D.L., & Brown, W. 1972. Successor type and organizational change in the
corporate enterprise. Administrative Science Quarterly, 17: 371-381.

Hickson, D.J., Hinings, C.R., Lee, C.A., Schneck, R.E., & Pennings, J.M. 1971. A
strategic contingencies' theory of interorganizational power. Administrative
Science Quarterly, 16: 216-227.

Hill, C., & Snell, S. 1988. External control, corporate strategy, and firm performance
in research-intensive industries. Strategic Management Journal, 9: 577-590.

Hill, C., & Snell, S. 1989. Effects of ownership structure and control on corporate
productivity. Academy of Management Journal, 32: 25-46.









Hofer, C., & Schendel, D. 1978. Strategy formulation: Analytical concepts. St. Paul,
MN: West.

Hoskisson, R.E., Hitt, M.A., Johnson, R.A., & Moesel, D.D. 1993. Construct validity
of an objective (entropy) categorical measure of diversification strategy. Strategic
Management Journal, 14: 215-235.

Hoskisson, R.E., Johnson, R.A., & Moesel, D.D. 1994. Corporate divestiture intensity
in restructuring firms: Effects of governance, strategy, and performance.
Academy of Management Journal, 37: 1207-1251.

Hoskisson, R.E., & Turk, T.A. 1990. Corporate restructuring: Governance and control
limits of the internal capital market. Academy of Management Review, 15: 459-
477.

Hunt, H. 1986. The separation of corporate ownership and control: Theory, evidence,
and implications. Journal of Accounting Literature, 5: 85-124.

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

Judge, W.Q., & Zeithaml, C.P. 1992. Institutional and strategic choice perspectives on
board involvement in the strategic decision process. Academy of Management
Journal, 35: 766-794.

Kesner, I.F., & Sebora, T.C. 1994. Executive succession: Past, present and future.
Journal of Management, 20: 327-372.

Kerr, J.L., & Kren, L. 1992. Effect of relative decision monitoring of chief executive
compensation. Academy of Management Journal, 35: 370-397.

Kosnik, R. 1990. Effects of board demography and directors' incentives on corporate
greenmail decisions. Academy of Management Journal, 33: 129-150.

Lamer, R. 1966. Ownership and control in the 200 largest nonfinancial corporations,
1929 and 1963. American Economic Review, 66: 29-46.

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

Lubatkin, M., Chung, K., Rogers, R., & Owers, J. 1989. Stockholder reactions to ceo
changes in large corporations. Academy of Management Journal, 32: 47-68.









Lubatkin, M., Merchant, H., & Srinivasan, N. 1993. Construct validity of some
unweighted product-count diversification measures. Strategic Management
Journal, 14: 433-449.

Main, B., O'Reilly, C., & Wade, J. 1995. The ceo, the board of directors and executive
compensation: Economic and psychological perspectives. Industrial and Corporate
Change, 4: 293-332.

Mallette, P., & Fowler, K. 1992. Effects of board composition and stock ownership on
the adoption of poison pills. Academy of Management Journal, 32: 1010-1035.

March, J., & Simon, H. 1958. Organizations. New York: Wiley.

Marris, R. 1964. The economic theory of managerial capitalism. Glencoe, IL: Free
Press.

McClelland, D.C. 1975. Power: The inner experience. New York: Irvington Publishers.

McEachern, W. 1975. Managerial control and performance. Lexington, MA: D.C.
Heath.

Miles, R.E., & Snow, C.C. 1978. Organizational strategy, structure, and process. New
York: McGraw-Hill.

Mintzberg, H. 1983. Power in and around organizations. Englewood Cliffs, NJ: Prentice
Hall.

Ocasio, W. 1994. Political dynamics and the circulation of power: CEO succession in
u.s. industrial corporations. Administrative Science Quarterly, 39: 285-312.

O'Reilly, C.A., Main, B.G., & Crystal, G.S. 1988. CEO compensation as tournament
and social comparison: A tale of two theories. Administrative Science Quarterly,
33: 257-274.

Pearce, J., & Zahra, S. 1991. The relative power of ceos and boards of directors:
Associations with corporate performance. Strategic Management Journal, 12: 135-
153.

Pfeffer, J. 1981. Power in organizations. Marshfield, MA: Pitman Publishing.

Pfeffer, J. 1992. Managing with power. Boston: Harvard Business School Press.

Pfeffer, J., & Salancik, G.R. 1978. The external control of organizations: A resource
dependency perspective. New York: Harper & Row.









Porter, M.E. 1980. Competitive strategy: Techniques for analyzing industries and
competitors. New York: Free Press.

Porter, M.E. 1985. Competitive advantage. New York: Free Press.

Ramanujam, V., & Varadarajan, P. 1989. Research on corporate diversification: A
synthesis. Strategic management Journal, 10: 523-551.

Rechner, P., & Dalton, D. 1991. CEO duality and organizational performance: A
longitudinal analysis. Strategic management Journal, 12: 155-160.

Reinganum, M. 1985. The effect of executive succession on stockholder wealth.
Administrative Science Quarterly, 30: 46-60.

Rumelt, R. 1982. Diversification, strategy and profitability. Strategic Management
Journal, 3: 359-369.

Salancik, G.R., & Pfeffer, J. 1977. Who gets power and how do they hold onto it: A
strategic contingency model of power. Organizational Dynamics, 5: 3-21.

Schendel, D., & Hofer, C. 1979. Strategic management: A new view of business policy
and planning. Boston: Little, Brown.

Smith, M., & White, M.C. 1987. Strategy, CEO specialization, and succession.
Administrative Science Quarterly, 32: 263-380.

Tosi, H.L., & 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., & Gomez-Mejia, L.R. 1994. Compensation monitoring and firm
performance. Academy of Management Journal, 37: 1002-1016.

Virany, B., & Tushman, M.L. 1986. Top management teams and corporate success in
an emerging industry. Journal of Business Venturing, 1: 261-274.

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

Walsh, J.P. 1988. Top management turnover following mergers and acquisitions.
Strategic Management Journal, 1988: 173-183.

Walsh, J.P., & Seward, J.K. 1990. On the efficiency of internal and external corporate
control mechanisms. Academy of Management Review, 15: 421-458.









Ware, R. 1975. Performance of manager versus owner-controlled firms in the food and
beverage industry. Quarterly Review of Economics and Business, Summer: 81-
92.

Welsh, M.A., & Dehler, G.E. 1988. Political legacy of administrative succession.
Academy of Management Journal, 31: 948-961.

Werner, S., & Tosi, H.L. 1995. Other people's money: The effects of ownership on
compensation strategy and managerial pay. Academy of Management Journal, 38:
1672-1691.

Wernerfelt, B. 1984. A resource-based view of the firm. Strategic Management Journal,
5: 171-180.

Westphal, J.D., & Zajac, E.J. 1995a. Who shall govern? CEO/board power and new
director selection. Administrative Science Quarterly, 40: 60-83.

Westphal, J.D., & Zajac, E.J. 1995b. Accounting for explanations of ceo compensation:
Substance and symbolism. Administrative Science Quarterly, 40: 283-308.

Wiersema, M. 1992. Strategic consequences of executive succession within diversified
firms. Journal of Management Studies, 29: 73-94.

Wiersema, M.F., & Bantel, K.A. 1993. Top management team demography and
corporate strategic change. Academy of Management Journal, 35: 91-121.

Wood, A. 1971. Diversification, merger and research expenditures: A review of
empirical studies. In R. Morris & A. Wood (Eds.), The corporate economy:
Growth, competition, and innovation potential, 428-453. Cambridge: Harvard
University Press.

Worrell, W., Davidson, W., & Glascock, J. 1993. Stockholder reactions to departures
and appointments of key executives attributable to firings. Academy of
Management Journal, 36: 387-401.

Wrigley, L. 1971. Divisional autonomy and diversification. Boston: Harvard Business
School.

Zahra, S., & Covin, J. 1993. Business strategy, technology policy and firm performance.
Strategic Management Journal, 14: 451-478.













APPENDIX A
LISTING OF FIRMS AND INDUSTRY CODES


ABBOT LABORATORIES 2834
ABRAMS INDUSTRIES INC 1540
ACTION INDUSTRIES INC 5020
AETRIUM 3825
ALLEGHENY LUDLUM STEEL 3312
AMCAST INDUSTRIES 3441
AMDAHL 3571
AMERICA WEST AIRLINES 4512
AMERICAN BRANDS 2111
AMERICAN HOME PRDS 2834
ANDREW CO 3661
ANDREWS CORP 2721
ANR PIPELINE 4923
ANTHEM ELECTRONICS 3674
APS HOLDING CO 5013
AVERY DENNISON 2672
AVONDALE 3731
BALL CORP 3411
BARNETT BANKS 6021
BAXTER 2834
BECTON-DICKINSON 3841
BELO (AH) CORP 2711
BLOUNT 1542
CALMAT 2951
CASE INDUSTRIES 3523
CHEVRON 2911
CHUBB GROUP 6331
COLEMAN 3086
COLGATE PALMOLIVE 2844
COMDIAL 3661
CONSOLIDATED FREIGHT 4213
CONTINENTAL AIRLINES 4512
COOPER TIRE 3011
CORESTATES FINANCIAL 6021









CPC INTERNATIONAL 2046
CROWN CORK & SEAL 3411
CSX CORP 4011
CTS CORP 3676
DATA GENERAL 3571
DELTA AIRLINES 4512
DIAMOND SHAMROCK 2911
DIGITAL EQUIPMENT CORP 3571
DOSKOCIL 2011
DOW CHEMICAL 2821
DRAVO 3281
E SYSTEMS 3812
EXCEL RESOURCES 4922
FIRST ALABAMA BANK 6712
FISHER SCIENTIFIC 5049
GANNETT 2711
GENCORP 3764
GENERAL REINSURANCE 6331
GERBER 2032
GOODYEAR TIRE 3011
GTE 4813
HADCO INDUSTRIES 3672
HOME HOLDINGS 6331
HOUGHTON MIFFLIN 2731
HUGHES RESOURCES 2411
IFF 2869
INGERSOLL RAND CO 3562
INTEL CORP 3674
INTERLAKE 2542
JOHNSON & JOHNSON 2844
JP MORGAN 6022
KASLER 1622
KATY INDUSTRIES 3559
KIDDIE 3069
KMART CORP 5311
KNIGHT RIDDER 2711
LOUISIANA LAND 2911
LUFKIN 3533
MAGNA INTERNATIONAL 3714
MANITOWOC 3536
MAVERICK 3317
MEREDITH 2721
MICRON TECHNOLOGIES 3674
MIKASA 5023