A Meta-Analysis of the Effects of Executive and Institutional Ownership on Firm Performance
Sundaramurthy, Chamu, Rhoades, Dawna L., Rechner, Paula L., Journal of Managerial Issues
Over the past two decades corporate ownership patterns have changed markedly with increased ownership by company executives and institutional investors. Many companies use equity to compensate executives and 52% of the 100 largest U.S. companies require executive share ownership, while another 15% encourage such ownership (Investor Relations Business, 1999). According to an executive compensation consulting firm, CEOs' average stake in their corporations was $58 million at the end of 1998, a 49 percent increase from that in 1996 (Wah, 1998). Similarly, of the top 200 companies, 99% pay directors with stock and 62% of directors' pay is in equity (Lavelle, 2002a). Ownership has also become concentrated in the hands of institutional investors, such as pension funds, mutual funds, insurance companies and banks (PR Newswire, 2000). Their stake in public companies increased from 38 percent in the early 1980s to more than 53 percent in the early 1990s (Useem et al., 1993). Public opinion on these ownership changes is sharply divided; critics argue that these changes are hurting corporate America while proponents highlight the positive implications.
Scholars have also studied these trends from an agency theory perspective (Jensen and Warner, 1988; Oswald and Jehera, 1991). Agency theory focuses on problems associated with the diffusion of corporate ownership, and the separation of ownership and control; the theory highlights the role of crucial internal and external monitoring mechanisms in reducing these problems (Berle and Means, 1938; Fama and Jensen, 1983). Within this context, executive stock ownership is viewed as a mechanism that potentially aligns manager-shareholder interests thereby reducing agency problems (Murphy, 1985). Concentration of equity in the hands of institutional investors, in contrast, arguably decreases agency problems associated with diffusion of ownership and increases external monitoring.
Empirical evidence, however, regarding the impact of executive and institutional ownership is quite mixed. For example, Tsetsekos and DeFusco (1990) found no significant relationship between executive ownership and firm performance. Although others have found a direct, positive relationship (Hambrick and Jackson, 2000; Mehran, 1995), McConnell and Servaes (1990) found a non-linear relationship. There are similar inconsistencies with respect to institutional investment and firm performance. For example, Baysinger et al. (1991) found a positive relationship whereas Graves (1988) found a negative relationship.
The purpose of this study is to examine the impact of executive and institutional ownership. We integrate conflicting empirical evidence on the impact of executive ownership and institutional investment using a quantitative meta-analytic technique. This technique allows us to control for statistical artifacts (e.g., sampling error) and provides robust estimates of the relationships between firm performance and stock ownership--specifically, executive and institutional ownership. This approach is particularly valuable inasmuch as the equivocal nature of extant empirical results may be purely artifactual; that is, there is no population relationship at all (Brierley and Cowton, 2000; Hunter and Schmidt, 1990; Orlitzky and Benjamin, 2001).
Indeed, our results indicate that there is no substantive relationship between ownership and firm performance. These findings are robust even after the potential influence of moderator variables (ownership measures, performance measures, as well as the level and type of executive stock ownership) is considered. Thus, advocating increased executive ownership or applauding the growth in institutional investment may be premature.
In the following sections we discuss the theoretical arguments linking executive and institutional investment and firm performance. Then, we describe the methodology used to integrate results from previous studies and present our findings. …