Board Composition and Financial Performance: A Meta-Analysis of the Influence of outside Directors

Article excerpt

In its first report card on corporate governance, Business Week (Byrne and Melcher, 1996) ranked U.S. firms based on recommended best practices for independence, accountability, and board quality. Apple Computer, AT&T, Disney, and Quaker Oats were among those on the worst board list and have joined a growing list of companies under pressure from stockholders for poor performance, excessive executive compensation, and mismanaged strategies. In each case, critics have cited the board of directors for failing to exercise managerial and strategic oversight (Byrne and Melcher, 1996). In line with current "conventional wisdom," an alleged lack of independence is seen as the root cause of a board's failure to effectively monitor the actions of top management. Proponents of corporate board reform have long advocated increasing outside director representation as a means of increasing the independence and effectiveness of corporate boards (Bacon and Brown, 1975; Dayton, 1984; Waldo, 1985). The Business Week criteria fo r independence is "no more than two inside directors, no insiders on the board's audit, nominating, and compensation committees, no outside directors who directly or indirectly draw consulting, legal, or other fees from the company, and no interlocking directorships" (Byrne and Melcher, 1996: 98).

Arguably, improvements in corporate board functioning ought to ultimately yield improvements in financial performance. Empirical research findings to date, however, have been inconsistent and conflicting. Studies of outsider ratios and firm performance have produced correlations ranging from a positive .33 (Pearce and Zahra, 1992) to a negative .15 and larger (Beatty and Zajac, 1994; Kaufman and Taylor, 1993; Zahra and Stanton, 1988). Some studies have found zero or near-zero effects (Buchholz and Ribbens, 1994; Rechner and Dalton, 1986). Moreover, other research has reported that a higher ratio of inside, not outside directors, is associated with higher R&D expenditures (Baysinger et al., 1991), greater likelihood of CEO dismissal in times of financial crises (Ocasio, 1994), and higher firm performance (Pearce, 1983). These studies argue that inside directors who have access to richer, fuller information about their firms are in a better position than outside directors to make decisions about many critical a reas of firm operation and performance.

The purpose of this study was to (1) apply the quantitative technique of meta-analysis to the area of insider/outsider ratios on corporate boards and calculated composites of financial performance to arrive at a robust estimate of this relationship, and (2) examine the potential moderating effects of different operational definitions of monitoring and performance. In so doing, we hope to be able to uncover the reasons for the inconsistent results in this research stream, and test which theoretical framework holds with regards to board composition and firm performance. In the following sections we generate hypotheses based on the presumed link between director independence and performance, describe our methodology and analyses, summarize our results, and discuss the implications of our findings.


The separation of ownership and control first discussed by Berle and Means (1932) creates many situations in which the interests of managers and owners may not coincide. Agency theory is based on the notion that the delegation of managerial responsibilities by principals (owners) to agents (managers) requires the presence of mechanisms that either align the interests of principals and agents (such as stock ownership plans and performance contingent compensation) or monitor the performance of managers to insure that they use their knowledge and the firm's resources to generate the highest possible return for principles. More specifically, agency theory suggests that the best option for owners is to design contracts that align manager/owner interests. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.