According to Leech and Leahy (1991), ” The structure of share ownership may have an important role in determining a firm’s performance because if ownership is widely dispersed there is no individual (or group) with either voting power or the incentive to exercise control and enforce profit maximisation”. In their article they described a firm’s ownership structure in two ways: 1. The first were numerical conventional measures of concentration.
This included the use of the upper bound Herfindahl index measured as a percentage, and concentration ratios: C1, representing the largest holding; C5, C10 and C20, which are the combined holdings of the largest 5, 10 and 20 shareholders respectively; 2. Secondly, the use of control-type dummies based on the concentration of voting power. Two such sets of variables were further defined using ‘fixed’ and ‘variable’ rules respectively. The first set was defined according to whether the largest holding was greater than 5% (OC1), 10% (OC2) or 20% (OC3); and is assigned a value of 1 if greater than the chosen percentiles, and 0 otherwise.
The second defined by ‘fixed’ rules used the probabilistic-voting model of Cubbin and Leech (1986)1 to analyse the degree of control (OC90, OC95 and OC99), where the degree of control of largest holding exceeding 90, 95 and 99 percent respectively are assigned values of 1 and 0 otherwise. In their investigations of ownership structure and performance, a cross-sectional econometric analysis on the effects of ownership structure variables2 on a vector of performance indicators3 was performed.
To facilitate the analysis Leech and Leahy used a multivariate regression model. The driving factor behind choosing such a model stemmed from previous single-equation models where ownership-controlled firms reported higher profits but the effect quantitatively small or statistically insignificant. Some examples were cited in studies done for UK: Radice (1971); Holl (1975); and for the US: Hay and Morris (1979); Mueller (1986). However, Leech and Leahy multivariate case proved to be quite successful and full results were reported for the six performance indicators.
One of the main findings indicated that the effect of ownership concentration on the firm’s performance was negative and significant in three of the equations (that is where VAL, TPM and NAG were used). In particular concentration indices and control type dummies were significant on the vector of performance indicators. The results for control type showed to be positively significant in four equations: TPM, RSHC, TSG, NAG. This suggested that ownership-controlled firms are profitable and fast growing. Another study conducted by Prowse (1992) examined the structure of corporate ownership in Japan during the mid 1980s.
The unique regulatory and legal environment of the Japanese corporate system and the identification of two distinct corporate governance systems: the independent firm and keiretsu groups4 aroused Prowse’s interest to conduct the analysis. Summary statistics showed that ownership concentration, measured as the percentage of outstanding shares held by the top five shareholders of the company appeared to be widely held in Japan.
Values ranged from a minimum of 10. 9% to a maximum of 85% with a corresponding mean of 33.1% with the top five shareholders being predominantly financial institutions. Prowse (1992) further analysed the relationship between ownership concentration and the firm’s performance measured by the mean value of annual net profit as a percentage of book value of equity in both independent and keiretsu firms. In addition to ownership concentration as an independent variable were capital, research and development, and advertising expenditures (all as percentages of sales), firm size measured by the book value is assets in millions of dollars and the instability of the firm’s profit rate.
Evidence for both samples showed the coefficient on ownership concentration to be positive, but insignificant. This offered a clear indication that there was no relationship between ownership concentration and profitability for either sample. So far much reference has been made to large and rich countries in the corporate world, but has empirical work been conducted for the developing economies that also make some contribution?
Such a study was conducted by Lauterbach and Vaninsky (1999), ‘Ownership Structure and Firm Performance: Evidence from Israel’. Lauterbach and Vaninsky employed data for 280 Israeli public companies traded on the TelAviv Stock Exchange during 1994 and implemented an empirical analysis on the effect of ownership structure on firm performance In their study they made the distinction between family firms, firms controlled by partnerships of individuals, concern controlled firms and firms where blockholders have less than 50 percent of the vote.
Owner-manager firms were found to be less efficient than firms managed by a professional (non-owner) manager in the generation of net income. Furthermore family firms managed by their owners performed comparatively worst. These results therefore clearly suggested that majority control by a few individuals reduces firm performance and that professional non-owner managers endorse performance, where performance was estimated as the actual net income of the firm divided by the optimal net income given by the firm’s inputs.