Demystifying the illusion of the positive effects of ownership concentration on corporate performance

Abstract

Evidence supporting the relationship between ownership structure and corporate performance has been rather contradictory. In this research, we investigate the effects of ownership structure on business performance on a sample of 600 listed Canadian firms. We used a three-phase analysis of variance in which each phase used a different definition of ownership concentration: i) the overall concentration of the five largest shareholders (CONC); ii) the holdings of the largest shareholder (BLC1); and iii) inside shareholders as either managers or directors (BLCI). For each phase, we used cluster analysis and three other concentration cutoff levels (an even-split into thirds, extreme quartiles, and the Morck, Shleifer and Vishny (1988) cutoff) to verify if there is an optimal level of concentration cutoff that may impact the performance. Our results indicate a high level of ownership concentration in Canadian corporations. The Berle-Means widely held corporation is far from universal. Besides, while state control of traded firms is infrequent, family control is common. However, our findings indicate only a weak association between performance measures and ownership concentration levels, except for the return on investment, which shows some improvement with a high level of ownership. Our results confirm those of Demsetz and Lehn (1985). Overall, no evidence is found to support the efficient monitoring hypothesis, since performance cannot be improved by blockholders who seem not only to be entrenched but may benefit from perquisites and on-the-job consumption. This might indicate that large shareholders expropriate minority absentee owners.