Sources of CEO power and firm financial performance: a longitudinal assessment

Sources of CEO power and firm financial performance: a longitudinal assessment

Catherine M. Daily

The chief executive officer (CEO) is generally regarded as the most powerful organizational member. Attributions regarding the potential effect of a strong CEO on firm financial performance are common in both the academic literature and popular press The oft-made assumption is that a powerful CEO will impact firm performance. The anticipated direction of this influence, however, has not been uniformly established. Moreover, we are not aware of any direct empirical assessment of the relationship between CEO power and firm financial performance. We are also intrigued by the potential for reciprocal effects–firm performance impacts the level of CEO’s power Our review of the literature demonstrates no empirical attention to this possibility. This study relies on a four-wave panel design to assess the nature Of the relationship between CEO power and firm performance. Results of LISREL analysis demonstrate that aspects of CEO power and financial performance are, in fact, interrelated. Performance was found to be both an antecedent condition and outcome of CEO power.

Both the academic literature and popular press provide accounts of the widely shared belief that the chief executive officer (CEO) is the most powerful organizational member in the modern corporation (Fortune, 1991; Harrison, Torres & Kukalis, 1988; Hosmer, 1982; Pearce, 1981; Pearce & DeNisi, 1983; Pearce & Robinson, 1987; see also Eisenhardt & Bourgeois, 1988, for examples of powerful CEOs). CEOs’ power is typically attributed to their legitimate authority, as well as the broad knowledge of the firms they serve and their strong impact on firms strategic direction, structure, and internal processes (Beatty & Zajac, 1987; Mizruchi, 1983; Roth, 1995; Wallace, Worrell & Cheng, 1990). Moreover, it has been suggested that CEOs “set the tone for the entire corporation” (Wheelen & Hunger, 1990, p. 69) and that the CEO is “THE corporate leader” (Norburn, 1989, p. 2).

Of late, however, much of the focus has shifted toward an increased emphasis on top management teams, or the dominant coalition (Hambrick & Mason, 1984; Wiersema & Bantel, 1992). Recent research has examined various aspects of the top management team and the resultant impact on political behavior and firm performance (Eisenhardt & Bourgeois, 1988), firms’ diversification posture and acquisition activity (Finkelstein, 1992; Michel & Hambrick, 1992), and strategic change (Wiersema & Bantel, 1992). Additionally, some support has been found for an association between managerial power and firms’ diversification posture and acquisition activity (Finkelstein, 1992). This shift in focus may be unfortunate as the CEO, the central member of the top management team, occupies a position of unique influence in the firm (Pfeffer, 1992; Roth, 1995; Vance, 1983). The extent of this influence is still not fully understood.

In addition to occupying a position of unique influence, there is some evidence that CEOs share unique qualities. Norburn (1989), for example, has noted several differences between CEOs and top management team members. These include corporate influence (e.g., tenure, functional background), domestic influence (e.g., education, marital status), and self-concept (e.g., management style).

Based on the noted differences between CEOs and top management team members, we believe that examinations of CEOs and their influence in the organizations they serve is warranted. We acknowledge that an exclusive focus on CEOs carries with it certain assumptions and limitations (Finkelstein, 1992; Salancik & Pfeffer, 1977); still, we are persuaded that CEOs, in isolation, have the potential to affect firm outcomes. We would also note that whether focusing on the CEO, top management team, or both, little attention has been devoted to empirically examining executive-level power (Finkelstein, 1992).

This study, then, examines CEO power. We rely primarily on guidance from Finkelstein (1992) in identifying and empirically assessing dimensions of executives’ power. Finkelstein proposed and empirically validated four power dimensions which he labeled structural, ownership, prestige, and expert power in an examination of top management team, including the CEO. We extend prior work in three ways. Firstly, we provide an exclusive focus on the individual widely believed to be the most powerful organizational member, the CEO. Secondly, relying on guidance from the corporate governance literature, we include a measure of board composition which assesses the extent to which directors are captured by the CEO. This measure has not previously been considered in examinations of CEO power.

Perhaps most importantly, we empirically assess contemporaneous, lagged, and reciprocal relationships between multiple dimensions of CEO power and firm financial performance relying on a longitudinal structural equation approach. Assumptions regarding the potential impact of CEO power on firm outcomes are numerous. Little attention, however, has been devoted to the potential for firm performance to determine the level of CEOs’ power. It may be that firms with high levels of performance are characterized by CEOs granted increasing levels of discretion and power. Relying on four years of data, we empirically assess the potential for these lagged and reciprocal relationships between dimensions of CEO power and firm performance.

Power in the Executive Suite

There are a variety of approaches for assessing power in organizational settings. French and Raven (1959), for example, provided an early typology of bases of power. More recently, Finkelstein (1992) proposed four types of executive-level power. These studies span both objective and perceptual measures of power, with some studies including both types of measures (Finkelstein, 1992; Provan, Beyer & Kruytbosch, 1980). Perceptual measures, however, have been the target of some criticism based on questions of their validity (Finkelstein, 1992). Researchers have also cautioned against reliance on single indicators of power, as executive-level power has been shown to be multidimensional in character (Astley & Sachdeva, 1984; Finkelstein, 1992; Krackhardt, 1990). In support of reliance on objective and multiple measures of power, Finkelstein (1992) recently validated four constructs of power which were derived from objective data. Three of these four constructs were found to be highly related to perceptual measures designed to capture identical power dimensions. Only expert power was not significantly related to the perceptual measures of power.

Regardless of the method by which power is measured, past researchers have suggested that the underlying basis of power is the ability to manage firm uncertainty (Finkelstein, 1992; Ocasio, 1994). The source of uncertainty may emanate either from within the firm or outside the firm. Internal sources of uncertainty include top management and directors (Boeker, 1992; Hambrick & Finkelstein, 1987). External sources of uncertainty are found in the firm’s task and institutional environments (Pfeffer & Salancik, 1978). The four bases of power which follow may be a means for CEOs to effectively reduce these sources of uncertainty.

Effective management of firm uncertainty is especially important for the CEO. It is this individual who is ultimately responsible for firm processes and outcomes that result, in part, from decisions regarding the management of firm uncertainty (see Finkelstein, 1992, for an overview). The choices that a CEO makes may ultimately determine the success of the firm. Ownership power, for example, accrues as a function of both formal (equity ownership) and informal (status as founder or relative of founder) influence in the organization. As a result of ownership power, it may be easier for the CEO to implement the choices he or she makes. As succinctly noted by Porter (1991), the very essence of strategy is choice (p. 95) and strategy is centrally concerned with issues of firm performance (p. 101).

Structural Power

One approach for studying power is structural. Structural power has also been referred to as hierarchical or legitimate power (Astley & Sachdeva, 1984; French & Raven, 1959; Hambrick, 1981). Previous research has identified this approach as a principal means for capturing executive-level power (Hambrick, 198 1; Pfeffer, 198 1). Hambrick, for example, found level of hierarchy to be “the dominant predictor of power” in his study of strategic leaders in private general hospitals, life insurance companies, and private four-year colleges (1981, p. 267). Finkelstein (1992), too, found support for this type of power. His results demonstrated that structural power, as compared to alternative constructs of power, was most strongly associated with executives’ perceived power.

Perhaps the strength of the structural approach to power is that it is captured in one’s position within the organization, as opposed to residing with any given individual (Brass & Burkhardt, 1993). Not only is there a hierarchical component involved, but also a social element to structural power, as well. Certainly when serving as CEO, structural power in the form of position in the organizational hierarchy is easily recognizable (Brass & Burkhardt, 1993; Krackhardt, 1990). Presumably, few would question the power which accompanies the office of CEO. Past research has found that one’s position in the organizational hierarchy is largely independent of the use of behaviors enacted to convey one’s power position (Brass & Burkhardt, 1993; Astley & Zajac, 1990, 1991; Wrong, 1968). Astley and Sachdeva (1984, pp. 105-106) supported this view of position power noting that “subordinates obey superiors not so much because they are dependent on the latter, but because they believe that the latter have a right to exercise power by virtue of their position” (see also Krackhardt, 1990). Hierarchical power, then, is often accompanied by a “social” recognition of one’s power.

While holding the position of CEO, in itself, is an indicator of one’s power, holding multiple titles while in the chief executive office may enhance one’s power (Harrison et al., 1988; Mizruchi, 1983; Ocasio, 1994). A common configuration is for the CEO to jointly serve as board chairperson (Lorsch & MacIver, 1989). This joint service is commonly referred to as CEO duality or the dual board leadership structure. The centralization of power achieved under the dual board leadership structure has many observers concerned (Levy, 1993; Lorsch & MacIver, 1989; Sherman, 1993). The primary concern is that the dual structure provides the potential for managerial domination of the board of directors (Firstenberg & Malkiel, 1994; Herman, 1981; Mace, 1971). The dual structure enables the CEO to control the agenda of board meetings, determine what information directors receive in advance of meetings, and lead board meeting discussions (Firstenberg & Malkiel, 1994). As a result, when the CEO formally dominates the board as chairperson, effective board control of management may be less likely. Absent effective oversight, management is more able to pursue interests which do not best serve the shareholders (Fama & Jensen, 1983). Harrison et al. (1988) have found support for the increased power of CEOs under the dual board leadership structure in an examination of CEO and board chair turnover.

The composition of the board of directors may also serve as evidence of CEOs’ structural power (Boeker, 1992; Ocasio, 1994). Considerable attention has been devoted to the relationship between CEOs and directors (Mizruchi, 1983; Zald, 1969). At issue is the extent to which CEOs exert substantial, even undue, influence over firms’ directors. This influence is of some concern given that the prime charter of the board is to effectively monitor and reward management (Baysinger & Hoskisson, 1990). While the board is, in theory. the more powerful actor as a function of its ability to hire and fire executives, directors’ power may be compromised by a powerful CEO. CEOs commonly exert influence over boards of directors, for example, by controlling the firm-specific information which directors receive (Mace, 1971: Zald, 1969).

One approach for capturing CEO power vis-a-vis the board of directors is through the independent/interdependent board composition distinction (Daily & Dalton, 1994). Interdependent directors are those directors appointed by the current CEO (Boeker, 1992), while independent directors are those outside directors not appointed by the current CEO (Wade, O’Reilly & Chandratat, 1990). In practice, directors are often nominated by the CEO, with these nominations being uniformly ratified by the board of directors (Mizruchi, 1983). Interdependent directors may feel some loyalty to those CEOs responsible for their appointment to the board of directors (Boeker, 1992). The potential for these loyalties and personal ties between board members and the CEO makes the selection of the independent/interdependent director distinction highly appropriate in examinations of CEO power. Boards comprised of greater proportions of interdependent directors, therefore, may provide the CEO with a stronger power base.

An additional indicator of structural power is captured in the form of compensation (Finkelstein, 1992; Molz, 1988). This indicator, too, is associated with the board of directors, as it is the board compensation committee or, in the absence of a compensation committee, the full board, which determines CEOs’ compensation. A CEO who is highly compensated as compared to other firm executives may be an indicator of the CEO’s ability to influence the board of directors (Albrecht & Jhin, 1978).

Ownership Power

A second form of executive-level power is ownership power. Power is likely to accrue to those CEOs who maintain ownership positions in the firms they serve since they, in effect, represent both management and shareholders. CEOs with significant shareholdings, as a function of their ability to influence important firm decisions in an ownership capacity, are likely to be more powerful than CEOs without an ownership interest in the firm (Zald, 1969). As evidence of enhanced power as a function of firm equity ownership, Allen (1981) has found that CEOs who are principal stockholders in the firms they serve are more able to define the firm’s direction. Additionally, CEOs with considerable ownership stakes may be positioned to prevent their own involuntary dismissal (Fredrickson, Hambrick & Baumrin, 1988; Pfeffer, 1981). Equity holding CEOs may also have greater influence in the director selection process (Fredrickson et al., 1988). Recently, Buchholtz and Ribbens (1994) found that CEO stock ownership was negatively associated with the likelihood of resistance to a takeover. They concluded that this finding provided some evidence that CEO stock ownership is an important mechanism in protecting shareholders’ interests.(1)

A second form of ownership power derives from the CEO’s status as founder or as a relative of the firm’s founder. Founders have been shown to be a strong organizational influence (Boeker, 1989; Eisenhardt & Schoonhoven, 1988). Carroll (1984), for example, demonstrated that the departure of a founder CEO significantly increased the likelihood of firm failure. CEOs who are either the firm’s founder, or related to the founder, may gain power through increased interactions with, and a long-term relationship with, board members and other important firm constituents (Finkelstein, 1992).

Prestige Power

A third source of executive-level power is derived from an individual’s level of prestige or status. CEOs who are considered to be members of the managerial elite signal to others their importance, both within and outside the firm (D’Aveni, 1990; Useem, 1979). According to Giddens (1972), CEOs. by definition, are members of the managerial elite since they occupy the position of primary importance in an organization.

Prestigious CEOs can aid in establishing the firm’s legitimacy. Moreover, prestigious CEOs provide the firm with access to other prestigious individuals. As some evidence of the potential legitimating influence of prestigious firm leaders. D’Aveni (1990) found that the exit of prestigious top managers occurred in the five years preceding a bankruptcy filing. He concluded that this management bailout phenomenon may have caused important external constituents to withdraw their support from the firm, leading to a subsequent bankruptcy filing. CEOs who are able to enhance firms’ legitimacy may enable the firm to buffer itself from environmental uncertainties (Selznick, 1957).

In addition to their service as CEOs, service on other organizations’ boards of directors, as well as having graduated from an elite educational institution, may provide some evidence of the level of CEOs’ prestige (D’Aveni, 1990; Finkelstein, 1992). Service on other boards may indicate an increased ability to manage interorganizational dependencies through increased access to information (Pennings, 1980).(2) CEOs serving on other firms’ boards, for example, have the opportunity to interact with other prestigious individuals no less than several times a year. This contact with other elite individuals may enhance the focal organization’s legitimacy (Useem, 1979). The literature on interlocking directorates Has established the role of the board as a mechanism for facilitating information sharing (Steams & Mizruchi, 1993).

Association with an elite educational institution may also provide some indication of CEOs’ level of prestige (D’Aveni, 1990). Having attended educational institutions which are generally regarded as prestigious, transfers that prestige to the individual. As with service on boards of directors, attendance at an elite school provides access to other individuals widely regarded as prestigious. Apparently, such connections provide some advantages, as many large firm managers and board members have attended elite educational institutions (Useem, 1979).

Expert Power

As previously noted, the ability to effectively manage the firm’s external environment provides CEOs with a source of power (Hambrick, 1981; Tushman & Romanelli, 1983). CEOs with exposure to a variety of functional areas have the opportunity to develop contacts both within and outside the firm across a broader range of areas as compared to those CEOs with limited functional exposure. These contacts may enable the CEO to better address environmental uncertainties which the firm faces. Breadth in managerial assignments during the CEO’s career, then, may provide some evidence of expert power as the CEO serves an important boundary-spanning role in the organization (Finkelstein, 1992).

A critical role of directors is service as boundary-spanning agents (Pfeffer & Salancik, 1978; Provan, 1980). One rationale for nominating a director for board service is to gain access to critical resources or information that the individual may provide the firm. To the extent that firms’ CEOs may have developed these types of contacts, they may rely on their own network of contacts, bypassing directors who could serve this same function. The decreased dependency on directors as a result of functional breadth, too, may provide some evidence of CEO power.

Expert power may also provide the CEO with a means for exerting control over directors, and other organizational constituents, through control of firm specific information. A notable source of CEO power, for example, is a comprehensive knowledge of firm affairs (Firstenberg & Malkiel, 1994). CEOs have access to substantially more information than non-management directors and may control directors’ access to such information. The absence of critical firm-specific information may place directors at a substantial disadvantage in boardroom discussions.

Power and Performance: The Chicken and the Egg

Power [right arrow] Performance

The extent to which a CEO possesses power, while in itself is of some interest, reveals little about the effect that a powerful CEO has on firm outcomes. Past research has primarily focused on investigations of CEO power and the resultant effect on internal firm processes. Eisenhardt and Bourgeois (1988), for example, found that powerful CEOs were associated with increased amounts of political activity among top management teams. One conclusion based on these findings was that the presence of a powerful CEO caused top managers to engage in increased political activity as a result of frustration at not being able to exercise more control in the organization (Eisenhardt & Bourgeois, 1988). This increased political activity was found to be detrimental to firm performance.

Consistent with agency theory arguments, enhanced power may also provide CEOs sufficient discretion to pursue objectives which are inconsistent with maximizing shareholder wealth. Powerful CEOs may, for example, seek ways to increase their compensation with little regard for firms’ performance (Pavlik, Scott & Tiessen, 1993). As has been noted of late in the popular press, instances of CEOs’ compensation increasing while firm performance suffers are not uncommon.

Extrapolating from related research, we might conclude that the extent of CEOs’ power is related to firm outcomes, specifically performance. The popular press often contains attributions regarding the relationship between CEO power and firm performance. The high performance of firms such as Berkshire Hathaway and Microsoft has often been attributed to their powerful CEOs, Warren Buffett and Bill Gates, respectively. Moreover, we would note that powerful CEOs are often recruited for the purpose of improving firm performance (e.g., Randall Tobias of Eli Lilly, Sally Frame Kasaks of Ann Taylor Stores Corp.), suggesting that boards of directors, at least on occasion, believe that the CEO can impact firm performance.

Performance [right arrow] Power

The succession literature provides numerous examples of reverse causation in the CEO power/firm performance relationship; i.e., performance levels determine the level of CEO power. Boeker (1992), for example, found that CEO dismissal was more likely when firm performance was poor and the CEO’s power was low. Harrison et al. (1988), too, examined turnover among CEOs and board chairs. They found that board chairs were more likely to leave the firm than were CEOs serving only in that capacity or CEOs also serving as board chair (duality). Harrison et al. concluded that these results demonstrated that CEOs were more powerful than board chairs; however, they also found that firm performance may affect CEOs’ power, as poor firm performance was associated with increased CEO, but not board chair, turnover. A related examination found a direct relationship between managerial power and firm performance; CEOs of more profitable firms had longer tenures (Allen & Panian, 1982; see also James & Soref, 1981).

Lastly, Ocasio (1994), in an examination of CEO succession, found that CEO power diminishes under conditions of economic adversity. When firm performance is poor, the balance of power shifts to the board. Moreover. those managers serving on the board are more likely to contest the CEO’s power than are outside directors (Ocasio, 1994). Ocasio’s findings are consistent with Fredrickson et al.’s (1988) observation that when fir-in performance is poor, CEOs’ power must be in place in order to retain their organizational position. This is also consistent with James and Soref (1981), who found that the probability of involuntary CEO turnover was inversely related to firm’s profitability.

Here, too, we can rely on examples from the popular press. CEOs are commonly removed due to poor performance (e.g., John Akers of IBM, Paul Lego of Westinghouse, Kay Whitmore of Eastman Kodak). In instances of poor performance, the power of the CEO may be greatly diminished. Jerry K. Pearlman, CEO of Zenith Electronics Corp., for example, was recently required to meet with the board of directors (or its executive committee) every two weeks in response to several years of protracted poor performance (Dobrzynski, 1994). It was noted by one observer that Zenith’s directors “stripped Pearlman of the free rein that once was a comer-office prerogative” (Dobrzynski, 1994, p. 64).

We might also note that it is possible that when firm performance is high, CEOs are granted a wider range of discretion. There may be little need to closely control the CEO of a highly performing firm. Lorsch and MacIver’s (1989) observation that it is only in times of crisis (e.g., financial decline) that boards become active in monitoring management may provide some evidence of directors’ confidence in the CEO under normal operating conditions.

As has been demonstrated, the relationship between CEO power and firm performance is complex. We are left with a “chicken and egg” dilemma–which came first, CEO power or firm performance? While the directional relationship between CEO power and firm performance remains unclear, it would appear that there is some evidence that CEO power may affect firm performance and that firm performance may affect CEO power.

Research Question

While past examinations of CEO succession provide insights into the performance implications of executive-level turnover, we are aware of no research which empirically tests the extent to which multiple dimensions of CEO power and firm financial performance are interrelated. The executive succession research suggests that poor performance may lead to CEOs’ dismissal (Boeker, 1992, Harrison et al., 1988). We would argue that dismissal is some indication of low CEO power. Recall, however, that Fredrickson et al. (1988) also noted that powerful CEOs, even when faced with poor performance, may be able to protect their positions. These studies suggest the need to carefully consider the causal relationship between CEO power and firm financial performance.

There is little empirical or theoretical guidance regarding the relationship between CEO power and firm performance. It may be, for example, that CEOs possessing high levels of power misuse their power for their own benefit at the expense of shareholders. Agency theory is based on the potential conflict of interest when ownership and control of the firm are separate (see Eisenhardt, 1989; Jensen & Meckling, 1976, for an overview of agency theory). Powerful CEOs may exercise their influence for self-benefit, as opposed to shareholder benefit (Gomez-Mejia, Tosi & Hinkin, 1987). Certainly, however, not all executives engage in self-interested behaviors. Enhanced power may protect competent CEOs from the pressures of multiple organizational constituents, enabling them to discharge their duties for the benefit of the firm and its owners.

An additional complicating factor is that some sources of power may be perceived as negative or positive, and some sources of power may contain both positive and negative elements. CEO duality, for example, has been suggested as being detrimental to firm performance as a result of the wide discretion afforded to the CEO who jointly serves as board chairperson (Lorsch & MacIver, 1989; Rechner & Dalton, 1989). Others have argued that duality provides a unified vision and direction for the firm (Anderson & Anthony, 1986). Still others have suggested a contingency perspective of conditions under which either board leadership structure would be appropriate (Finkelstein & D’Aveni, 1994). Similar arguments could be presented for CEO equity ownership (Daily & Dalton, 1992).

In the absence of strong guidance from the existing literature, we investigate the following research question: What is the nature of the relationship between CEOs’ power and firm financial performance? Reliance on this research question enables us to assess contemporaneous relationships, as well as lagged and reciprocal relationships in order to best determine the causal order between power and performance. It may be, for example, that CEOs with high levels of power (e.g., jointly serve as board chairperson, hold considerable equity in the firm, are the firm’s founder, and serve on several other corporate boards) positively impact firm performance as a function of being able to implement their vision relatively unimpeded from outside influence (Anderson & Anthony, 1986). Alternatively, it may be that as a result of high levels of firm performance, CEOs are able to accumulate power in the organization. In this scenario, power may be a reward for outstanding firm performance. Examples of such rewards might be adding the title of board chairperson, being granted wide discretion in the selection of new board members, or being recognized with offers to serve on other corporate and non-profit boards.

Methods

Research Design

Because the directionality of several of the relationships is uncertain, we employ a fourwave panel research design with data for the dependent and independent variables collected for each year between 1987 and 1990, inclusive. The design permits tests of a number of alternative lags and transposed relationships. The sample is composed of 100 randomly selected Fortune 500 firms. Only those firms that experienced no CEO succession during the study period are included. This criterion is important because power shifts, as occur during a change in CEO, can dramatically impact firm processes and outcomes, as well as publicly signal firms’ underlying power structure (Boeker, 1992; Finkelstein, 1992; Pfeffer, 1981).

Independent Variables

Structural Power. Three measures of structural power are used in this study. The first is whether the CEO concurrently serves as chairperson of the board (CEO duality). This is a dichotomous variable coded as 0 for those CEOs not serving as board chairperson and I for the dual board leadership structure. A second measure of structural power is captured in the extent to which the board is comprised of interdependent directors. This variable is computed as the ratio of interdependent to total directors. We also rely on a compensation ratio measure. This variable is derived from the ratio of the CEO’s total cash compensation to the Total cash compensation of the next highest paid officer in the firm. Higher ratios may provide some indication of the relative power of the CEO. These data were collected from corporate proxy statements and Business Week’s Executive Compensation Scoreboard.

Ownership Power. Two measures of ownership power are included in these analyses. The first is simply the extent of CEOs’ shareholdings. This measure is represented as a percentage of the CEO’s shares to total shares outstanding. The second measure is a categorical measure of the founder status of the CEO. This measure is based on Finkelstein’s (1992) work. Two conditions are used to determine three categories of founder affiliation: (1) the CEO is the founder or a relative of the founder, and (2) the CEO has the same last name as another officer of the firm. If neither of these conditions are present, then the variable is coded as 0. If one of these conditions is met, the variable is coded as 1. If both conditions are met the variable is coded as 2. These data were collected from corporate proxy statements, Business Week’s Corporate Elite, and Forbes.

Prestige Power. The level of the CEO’s prestige is captured by service on corporate boards, nonprofit boards, and holding degrees from elite educational institutions. Service on other boards is simply the total number of corporate boards and nonprofit boards on which the CEO serves. We also measure whether the CEO attended a prestigious undergraduate or graduate school. Finkelstein (1992, p. 538) provides a list of elite college and universities used to create this variable. These variables are dichotomous; either the CEO attended a prestigious undergraduate or graduate institution or did not. These data were collected from Business Week’s Corporate Elite and Standard and Poor’s Register of Corporations.

Expert Power. To determine the extent to which the CEO has power as a function of organizational expertise, we rely on the number of functional areas in which the CEO has served the firm. This is an interval level variable. These data were collected from Business Week’s Corporate Elite.

Dependent Variables

Firm Financial Performance. We rely on three measures of firm financial performance. Two of these measures are accounting-based measures of firm profitability (return on equity and return on investment). The remaining measure is a risk-adjusted, market-based measure of performance (Jensen’s alpha). Reliance upon multiple performance measures is important, as no one indicator reasonably captures firm financial performance (Bourgeois, 1980; Chakravarthy, 1986; Weiner & Mahoney, 1981).

Recent research has documented the need to consider accounting-based performance relative to some benchmark (Cannella & Lubatkin, 1993; Johnson, Hoskisson & Hitt, 1993). Typically, this benchmark is the industry rate of return as compared to that of the firm. Consistent with these suggestions, we rely on industry adjusted return on equity (ROE) and return on investment (ROI). These measures are based on the difference between a firm’s reported values and those of the primary industry in which the firm operates, as determined by two-digit SIC codes.

The market-based measure, Jensen’s alpha, provides a comparison of a given firm’s market performance to that of firms experiencing similar market risk (Brown & Warner, 1980; Cannella & Lubatkin, 1993; Nayyar, 1992, 1993). This measure is expressed as the estimate of the intercept in a regression equation of firm returns and market returns. Both firm and market returns are computed based on the risk-free rate of return. These financial data were collected from COMPUSTAT, Standard & Poor’s Stock Reports, as well as data maintained by the Center for Research in Security Prices (CRSP).

Analysis and Results

Anderson and Gerbing’s (1988) recommended two-step procedure for testing structural equation models with latent variables was followed in testing the proposed models. Specifically, they recommend that a measurement model first be evaluated using confirmatory factor analysis, which typically involves several iterations of model estimation, respecification, and re-estimation. Model acceptability is determined using a number of measures, including overall model fit and criteria for reliability and internal and external validity. As previously indicated, we largely relied on the constructs hypothesized by Finkelstein (1992) to comprise CEO power relative to the board (excepting expert power, for which only one indicator was collected, and which was therefore treated as an observed variable), and the measurement model estimated involved these constructs in addition to the firm performance construct. Measurement models for each year of the study were estimated separately. If an acceptable measurement model was found, the full structural model incorporating the consequent latent variables with its associated indicators would then be tested. A simple diagram of the proposed model is shown in Figure 1; for the sake of simplicity, multiple years are not illustrated. The models estimated in this study were longitudinal, wherein the same measures were collected over four years. In this case, the specific error terms for identical indicators across years is expected to correlate, so the error terms for identical performance indicators were freed to covary in the models (Bollen, 1989; Joreskog & Sorbom, 1993).

[Figure 1 ILLUSTRATION OMITTED]

When testing the measurement model, unfortunately, no acceptable solution was found. Initial specifications relying on all of the proposed indicators failed even to converge. Several alternative models were subsequently respecified–involving dropped indicators of specific latent variables as well as dropped latent constructs or respecified constructs–and even when models did converge, none approached the minimum threshold criteria for overall fit or reliability. Gerbing and Anderson (1988) have stressed that using multidimensional indicators of single constructs makes interpretation of results problematical. In fight of the failed confirmatory factor analysis, following Gerbing and Anderson’s advice we did not combine indicators, but instead chose to test an all-observed variable model using LISREL, a procedure that is analogous to multivariate path analysis in which specific paths can be constrained to zero (Joreskog & Sorbom, 1993, p. 10).

Testing observed variable models poses a new difficulty: large models, especially multiple panel models, contain many potential paths, and structural equation models require that the sample size exceed the number of parameters estimated. For this reason, we were forced to estimate a series of lagged and reciprocal models rather than a single exhaustive model. We first estimated four models, one each with the proposed CEO variables and firm performance in 1987, 1988, 1989, and 1990 (Figure 2). We then tested models with the same variables but with reversed causal direction; i.e., firm performance in 1987 and CEO variables in 1987 through 1990 (Figure 3). Finally, multiple-wave models that included those variables that were involved in significant relationships in the preliminary models were estimated, revised, and re-estimated (see Figure 4).

[Figures 2-4 ILLUSTRATION OMITTED]

Each model’s goodness of fit was assessed using traditional chi-square tests in addition to a variety of fit indices. The chi-square statistic provides an indication of how well the original correlation matrix of all indicators compares with the matrix generated by the hypothesized model, but is very sensitive to sample size, and several alternative indices have been created in an attempt to offset this limitation. Unfortunately, no single goodness-of-fit index has been shown to be clearly superior to all others (Medsker, Williams, & Holahan, 1994), and thus several were included in our evaluations. Additional indexes included were the Goodness-of-Fit Index (GFI; Joreskog & Sorbom, 1993), the Incremental Fit Index (IFI), and the Comparative Fit Index (CFI) (see Medsker et al., 1994). The results of the model tests are summarized in Table 1. Models 1 through 4 all have CEO power indicators as independent variables, with contemporaneous or lagged firm performance as dependent variables, and Models 5 through 8 invert those relationships. The overall fit of the simple lagged CEO power -> firm performance models was generally acceptable (based on the alternative goodness-of-fit indicators), but became worse as the lag increased. Examination of the individual path strengths showed that three of the power indicators were never significant (duality, CEO stock ownership, and number of nonprofit boards), and that others were involved in two or more relationships across time (independent directors, founder, number for-profit boards, and CEO education). Additionally, adjusted ROE was involved in more significant relationships than either adjusted ROI or Jensen’s alpha over the years examined. The overall fit of the inverse models was also generally acceptable, with all four models passing the chi-square test. As in the initial set of models, duality, CEO stock ownership, and number of nonprofit boards were not involved in any statistically significant relationships.

Table 1. Goodness-of-Fit Measures for Path Analytic Models

Chi-

Model Description Square DF Prob

1 CEOPow87 -> Perf87 6.32 3 0.097

2 CEOPow87 -> Perf88 12.16 3 0.007

3 CEOPow87 -> Perf89 11.83 3 0.008

4 CEOPow87 -> Perf90 14.31 3 0.003

5 Perf87 -> CEOPow87 16.66 15 0.340

6 Perf87 -> CEOPow88 20.93 21 0.460

7 Perf87 -> CEOPow89 20.00 21 0.520

8 Perf87 -> CEOPow90 23.38 21 0.320

9 Perf87 -> CEOPow88 -> Perf89 & Perf90 79.36 36 0.000

10 Perf87 -> CEOPow88 -> Perf89 & Perf90 56.30 30 0.003

Model GFI IFI CFI

1 0.99 0.97 0.94

2 0.98 0.92 0.83

3 0.98 0.91 0.77

4 0.98 0.91 0.81

5 0.96 0.94 0.68

6 0.96 1.00 1.00

7 0.96 1.00 1.00

8 0.95 0.94 0.84

9 0.91 0.90 0.89

10 0.92 0.94 0.93

Based on the outcomes of the simple lagged models, we then tested a longitudinal model that included variables from the four years sampled. Because causal order was not clearly established from the preliminary models, we included firm performance as both an antecedent to (1987) and a consequence of (1989 and 1990) CEO power. Again, the size of the model was constrained by the number of free parameters estimated, so the number of variables included in the model was restricted. Any CEO power variable that had been involved in art least one significant relationship was included in the model for 1988 (CEO education, number of for-profit boards, compensation, proportion of independent directors, founder, and number of functions). As can be seen from the results in Table 1 (Model 9), the overall fit of the model was acceptable, or nearly acceptable, based on the various goodness-of-fit indices, but an examination of the individual path strengths within the model (Figure 4) showed that two variables (number of functions and founder) were not involved in any significant relationships. Re-estimating the model without these variables significantly improved the overall fit of the model (Table 1, Model 10). Figure 4 illustrates those relationships that were significant at p [is less than or equal to] 0.05 and p [is less than or equal to] 0.10.

Discussion

These results underscore the complexity of the relationship between CEO power and firm financial performance. Perhaps the most notable finding is that the dimensions of CEO power included in this study do not constitute constructs of the power dimensions suggested by Finkelstein (1992), nor do they constitute a single construct of CEO power. While this study is not comprehensive in its inclusion of all dimensions of executive-level power specified by Finkelstein (1992) and further differs with the inclusion of the board composition measure to the structural power category, we expected some similarity in findings.

Relying, then, on an observed variable model, several notable findings emerged. Two of the more robust findings are the relationships between independent director proportion and corporate directorships and the accounting-based measures of firm performance. As demonstrated in Figure 4, higher levels of firm performance lead to greater numbers of corporate directorships for the CEO in the subsequent year, which then lead to higher levels of firm performance two years out. This finding demonstrates the reciprocal nature of the relationship between the focal firm’s performance and serving on other firms’ corporate boards. Based on these results it would seem that one prerequisite for being invited to serve on other corporate boards, and gaining prestige power, is to lead a financially successful corporation. In this instance, building one’s level of prestige is dependent on demonstrating some ability at successfully running a major corporation.

Past research has suggested that service on corporate boards provides a means for CEOs to better manage their firms’ interdependencies (Pfeffer, 1972). A second strategy is to use such service as a means to interact with colleagues at other corporations (Davis, 1991; Useem, 1979), thus increasing the firm’s social capital. These interactions provide an opportunity for the CEO to associate with peers who may informally provide him or her with advice and counsel. Also, service on other firms’ boards may assist the CEO in establishing professional relationships which benefit the firm. Apparently, these strategies are successful. CEOs’ service on other corporate boards enhances their own firm’s financial performance, as illustrated in Figure 4.

This study also demonstrates that higher levels of firm performance lead to lower proportions of independent directors in the following year. Furthermore, contrary to board reform critics who have voiced a strong opinion regarding the need for higher proportions of independent directors (Lorsch & MacIver, 1989; Mizruchi, 1983; Monks & Minow, 1991), these findings suggest that lower proportions of independent directors do not lead to lower firm performance. In fact, lower proportions of independent directors correlate negatively with subsequent firm performance. We are intrigued by this finding as it is not supportive of the need for greater independence in the boardroom.

Numerous academics and practitioners have strongly suggested that independent directors are needed to effectively monitor managers and protect shareholders from executive abuses (Lorsch & MacIver, 1989; Monks & Minow, 1991). Others have suggested that directors appointed by the CEO experience an inherent conflict of interest. On the one hand, they may feel an obligation to the CEO to whom they owe their board seat, while at the same time owing a fiduciary obligation to the shareholders (Firstenberg & Malkiel, 1994).

It may be that CEOs of firms that are performing well use their discretion to appoint directors, but do not compromise the effectiveness of the board with these appointments. High performing CEOs may have already established their willingness and ability to enhance shareholder value, and structure boards less for effective monitoring and more for the service and resource acquisition roles (Carpenter, 1988; Pfeffer & Salancik, 1978). We would note, however, that this finding may be specific to those firms with high performance. Previous research has established the effectiveness of independent directors in failed firms (Daily & Dalton, 1994).

Two dimensions of power were also found to be related to the accounting- and the market-based firm performance measures. Higher levels of relative compensation in 1988 led to higher returns in 1989 and 1990. Interestingly, this relationship is inconsistent with what might be expected in structuring compensation systems designed to encourage executives to maximize shareholder wealth. Rather than being rewarded for firm performance with higher levels of compensation, these results indicate that higher levels of relative compensation led to higher firm performance in subsequent years. We would reiterate, however, that the compensation variable is the CEO’s compensation relative to that of the next highest paid officer in the firm, not the CEO’s total compensation package. This indicates that CEOs may increase personal power within the firm (as indicated by relative compensation) independent of previous firm performance.

Lastly, educational prestige was associated with lower market returns in 1990. This finding may indicate that CEOs with prestigious educational backgrounds are granted wide discretion and that such discretion leads to erosion of shareholder value in subsequent years. It is not uncommon for directors to be CEOs themselves (Lorsch & MacIver, 1989). Moreover, a significant number of large firm CEOs share elite educational backgrounds (see Business Week’s annual issue profiling the corporate elite). Such shared associations may cause firms’ boards to grant a CEO wider discretion than might otherwise be forthcoming. D’Aveni (1990, p. 121) has noted, for example, that “prestige helps to maintain an illusion of competence and control.” Perhaps this illusion is not shattered until firm performance dramatically suffers.

A “non-finding” of particular note is the lack of significant relationships for the CEO duality variable. Numerous practitioners, and a subset of academics, have strongly advocated the need to formally separate the positions of CEO and board chairperson (Dobrzynski, 1991; Lorsch & MacIver, 1989). In a recent Fortune (1991, p. 13) article, Don Hambrick referred to CEOs holding multiple firm titles as the Idi Amin phenomenon, referring to the former Ugandan leader who assigned himself a dozen formal titles. Both practitioners and academics generally believe that CEO duality will lead to the erosion of shareholder value. Interestingly, these data provide no support for such a position. CEO duality was neither significantly affected by firm performance, nor did it significantly affect firm performance either positively or negatively.

We are encouraged by these results for several reasons. First, these findings illustrate the importance of assessing lagged and reciprocal relationships. A considerable amount of strategic leadership research has relied exclusively on contemporaneous relationships. These results indicate the limitation in conclusions which may be inherent in such research. Here, for example, aspects of CEO power influence firm performance in subsequent years.

In addition to influencing firm performance, CEOs’ power is apparently affected by prior firm performance. Establishing the lagged and reciprocal nature of the relationship between CEO power and firm performance is an important contribution of this study. These results would seem strong rationale for reliance on longitudinal data in strategic leadership research.

This study also demonstrates the potential error in ignoring the unique influence of the CEO in organizational research. As previously noted. examination of CEOs in isolation has largely been supplanted by the increased attention to top management team research. We applaud the efforts and progress realized within the domain of top management team research. Still, we are persuaded by our results, which illustrate that CEOs, even in the largest of organizations, do influence firm performance. We hope these findings provide some modest encouragement for researchers to renew their interest in CEOs and the unique influence that they exert on the modern corporation.

Acknowledgment: We are indebted to an anonymous reviewer for noting that not all of our power dimensions are subject to reverse causation (i.e., firm performance –> CEO power). Specifically, prestige, as captured by degrees from elite educational institutions, and functional expertise are not sensibly a function of firm performance. The CEO of a highly performing firm is not likely to be granted a degree from an elite educational institution as a function of his or her position (with the notable exception of honorary degrees). By the same logic, the CEO would not likely seek additional exposure to the various functional areas of the firm after achieving the position of CEO.

Notes

(1.) It is here that the argument about CEO power vis-a-vis stock ownership -> financial performance bothers us the most.

(2.) Again, this should enable the CEO to enhance firm performance. This goes beyond the prestige of the CEO (personal power in pursuit of self-interest) to resource dependency and power (in pursuit of organizational goals), and is hard to disentangle.

References

Albrecht, W.S. & Jhin. P. (1978). The million dollar men. Business Horizons. August. 9-14.

Allen, M.P. (1981). Managerial power and tenure in the large corporation. Social Forces. 60: 484-494.

Allen, M.P. & Panian, S.K. (1982). Power, performance, and succession in the large corporation. Administrative Science Quarterly. 27: 538-547.

Anderson, C.A. & Anthony, R.N. (1986). The new corporate directors. New York: John Wiley.

Astley, W.G. & Sachdeva. P.S. (1984). Structural sources of interorganizational power: A theoretical synthesis. Academy of Management Review, 9: 104-113.

Astley. W.G. & Zajac, E.J. (1990). Beyond dyadic exchange: Functional interdependence and sub-unit power. Organization Studies, 11: 481-501.

–. (1991). Intraorganizational power and organizational design: Reconciling rational and coalitional models of organization. Organization Science, 4: 399-411.

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.

Beatty, R.P. & Zajac, E.J. (1987). CEO change and firm performance in large corporations: Succession effects and manager effects. Strategic Management Journal, 8: 305-317.

Boeker, W. (1989). Strategic change: The effects of founding and history. Academy of Management Journal. 32: 489-515.

Bollen. K. (1989). Structural equations with latent variables. New York: John Wiley.

Bourgeois, L.J. (1980). Performance and consensus. Strategic Management Journal, 1: 227-248.

Brass, D.J. & Burkhardt, M.E. (1993). Potential power and power use: An investigation of structure and behavior. Academy of Management Journal, 36: 441-470.

Brown S.J. & Warner, J. B. (1980). Measuring security price performance. Journal of Financial Economics. 8: 205-258.

Cannella, A.A. & Lubatkin, M. (1993). Succession as a sociopolitical process: Internal impediments to outside selection. Academy of Management Journal, 36: 763-793.

Carpenter, R.N. (1988). Cooperative governance: Directors’ responsibilities. Directors & Boards. 12(3): 3-6.

Carroll, G.R. (1984). Dynamics of publisher succession in newspaper organizations. Administrative Science Quarterly, 29: 93-113.

Chakravarthy, B.S. (1986). Measuring strategic performance. Strategic Management Journal, 7: 437-458.

Daily, C.M. & Dalton, D.R. (1992). Officer and director stock ownership and firm performance in the publicly traded small corporation. Journal of Business Strategies, 9: 101-113.

–. (1994). Corporate governance and the bankrupt firm: An empirical assessment. Strategic Management Journal, 15: 643-654.

Davis, G.F. (1991). Agents without principles? The spread of the poison pill through the intercorporate network. Administrative Science Quarterly, 36: 583-613.

D’Aveni, R.A. (1990). Top managerial prestige and organizational bankruptcy. Organization Science. 1: 123-142.

Dobrzynski, J.H. (1991). Chairman and CEO: One hat too many. Business Week. November 18, 124.

–. (1994). How to handle a CEO. Business Week, February 21. 64-65.

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

Eisenhardt, K.M. & Bourgeois, L.J. (1988). Politics of strategic decision making in high-velocity environments: Toward a midrange theory. Academy of Management Journal. 31: 737-770.

Eisenhardt, K.M. & Schoonhoven. C.B. (1990). Organizational growth: Linking founding team, strategy, environment, and growth among U. S. semiconductor ventures. 1978-1988. Administrative Science Quarterly, 35: 504-529.

Fama, E.F. & Jensen, M.C. (1983). Agency problems and residual claims. Journal of Law and Economics, 26: 301-326.

Finkelstein, S. (1992). Power in top management teams: Dimensions. measurement, and validation, Academy of Management Journal, 35: 505-538.

Finkelstein, S. & D’Aveni, R.A. (1994). CEO duality as a double-edged sword: How boards of directors balance entrenchment avoidance and unity of command. Academy of Management Journal, 37: 1079-1108.

Fortune. (1991). Chief executives as Idi Amin?, July 1, 13.

Fredrickson, J.W., Hambrick, D.C. & Baumrin, S. (1988). A model of CEO dismissal. Academy of Management Review, 13: 255-271.

French, J.R.P. & Raven, B. (1959). Bases of social power. Pp. 150-167 in D. Cartwright (ed.). Studies in social power. Ann Arbor, MI: University of Michigan Press.

Gerbing, D.W. & Anderson, J.C. (1988). An updated paradigm for scale development incorporating unidimensionality and its assessment. Journal of Marketing Research, 25: 186-192.

Gomez-Mejia, L.R., Tosi. H. & Hinkin, T. (1987). Managerial control. performance and executive compensation. Academy of Management Journal, 30: 51-70.

Hambrick, D.C. (1981). Environment, strategy, and power within top management learns. Administrative Science Quarterly, 26: 253-276.

Hambrick, D.C. & Finkelstein, S. (1987). Managerial discretion: A bridge between polar views on organizations. Pp. 369-406 in B.M. Staw & L.L. Cummings (eds.). Research in organizational behavior, Vol. 1. Greenwich, CT: JAI Press.

Hambrick, D.C. & Mason, P.A. (1984). Upper echelons: The organization as a reflection of its top managers. Academy of Management Review, 9: 193-206.

Harrison, JR., 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.

Herman, E.S. (1981). Corporate control, corporate power. New York: Cambridge University Press.

Hosmer, L.T. (1982). The importance of strategic leadership. Journal of Business Strategy. 3(Fall): 47-57.

James, D.R. & Soref. M. (1981). Profit constraints on managerial autonomy: Managerial theory and the unmaking of the corporation president. American Sociological Review, 46: 1-18.

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.

Johnson, R.A., Hoskisson, R.E. & Hitt, M.A. (1993). Board of director involvement in restructuring: The effects of board versus managerial controls and characteristics. Strategic Management Journal, 14: 33-50.

Joreskog, K. & Sorbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS command language. Chicago. IL: Scientific Software International. Inc.

Krackhardt, D. (1990). Assessing the political landscape: Structure. cognition. and power in organizations. Administrative Science Quarterly, 35: 342-369.

Levy, L. (1993). Separate chairmen: Their roles and compensation. The Corporate Board, 14(79): 10-15.

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

Medsker, G.J., Williams, L.J. & Holahan, P.J. (1994). A review of current practices for evaluating causal models in organizational behavior and human resources management research. Journal of Management, 20: 439-464.

Mizruchi, M.S. (1983). Who controls whom? An examination of the relation between management and boards of directors in large American corporations. Academy of Management Review, 8: 426-435.

Molz, R. (1988). Managerial domination of boards of directors and financial performance. Journal of Business Research, 16: 235-249.

Monks, R.A.G. & Minow, N. (1991). Power and accountability. New York: Harper Business.

Nayyar, P.R. (1992). Performance effects of three foci in service firms. Academy of Management Journal. 35: 985-1009.

–. (1993). Performance effects of information asymmetry and economies of scope in diversified firms. Academy of Management Journal, 36: 28-57.

Norburn, D. (1989). The chief executive: A breed apart. Strategic Management Journal. 10: 1-15.

Ocasio, W. (1994). Political dynamics and the circulation of power. CEO succession in U.S. industrial corporations, 1960-1990. Administrative Science Quarterly. 39: 285-312.

Pavlik, E.L., Scott, T.W. & Tiessen, P. (1993). Executive compensation: Issues and research. Journal of Accounting Literature, 12: 131-189.

Pearce, J.A. (1981). An executive-level perspective on the strategic management process. California Management Review, 29(Fall): 39-48.

Pearce, J.A. & DeNisi, A. (1983). Dominant coalition formation as interpreted through attribution theory. Academy of Management Journal, 26: 119-128.

Pearce, J.A. & Robinson, R.B. (1987). A measure of CEO social power in strategic decision-making. Strategic Management Journal, 8: 297-304.

Pfeffer, J. (1972). Size and composition of corporate boards of directors: The organization and its environment. Administrative Science Quarterly, 17: 218-228.

–. (1981). Power in organizations. Boston: Pitman.

–. (1992). Managing with power. Boston: Harvard Business School Press.

Pfeffer, J. & Salancik, G.R. (1978). The external control of organizations: A resource dependence perspective. New York: Harper & Row.

Porter, M.E. (1991). Towards a dynamic theory of strategy. Strategic Management Journal, 12: 95-117.

Provan, K.G. (1980). Board power and organizational effectiveness among human service agencies. Academy of Management Journal, 23: 221-236.

Provan, K.G., Beyer, J.M. & Kruytbosch, C. (1980). Environmental linkages and power in resource-dependence relations between organizations. Administrative Science Quarterly. 25: 200-225.

Rechner, P.L. & Dalton, D.R. (1989). The impact of CEO as board chairperson on corporate performance: Evidence vs. rhetoric. Academy of Management Executive. 3: 141-144.

Roth, K. (1995). Managing international interdependence: CEO characteristics in a resource-based framework. Academy of Management Journal, 38: 200-231.

Salancik, G.R. & Pfeffer, J. (1977). Who gets power–and how they hold on to it: A strategic-contingency model of power. Organizational Dynamics, 5: 3-21.

Selznick, A. (1957). Leadership in administration: A sociological interpretation. New York: Harper & Row.

Sherman, H. (1993). The retired CEO: On or off the board? Directors & Boards, 17(3): 32-35.

Steams, L.B. & Mizruchi, M.S. (1993). Board composition and corporate financing: The impact of financial institution representation on borrowing. Academy of Management Journal. 36: 603-618.

Tushman, M.L. & Romanelli, E. (1983). Uncertainty, social location and difference in decision making. Management Science, 28: 12-23.

Useem, M. (1979). The social organization of the American business elite and participation of corporation directors in the governance of American institutes. American Sociological Review. 44: 553-572.

Vance, S.C. (1983). Corporate leadership. New York: McGraw-Hill.

Wallace, N.D., Worrell, D.L. & Cheng, L. (1990). Key executive succession and stockholder wealth: The influence of successor’s origin, position, and age. Journal of Management, 16: 647-664.

Weiner, N. & Mahoney, T.A. (1981). A model of corporate performance as a function of environmental. organizational, and leadership influences. Academy of Management Journal, 24: 453-470.

Wheelen, T.L. & Hunger, J.D. (1990). Strategic management. Reading, MA: Addison-Wesley.

Wiersema, M.F. & Bantel, K.A. (1992). Top management team demography and corporate strategic change. Academy of Management Journal, 35: 91-121.

Wrong, D.H. (1968). Some problems in defining social power. American Journal of Sociology, 73: 673-681.

Direct all correspondence to: Catherine Daily, Purdue University. Krannert Graduate School of Management. West Lafayette, IN 47907.

COPYRIGHT 1997 JAI Press, Inc.

COPYRIGHT 2004 Gale Group