Family Control in Swedish Public Companies : Implications for Firm Performance, Dividends and CEO Cash Compensation

Abstract: Essay 1 examines the relationship between family control and firm performance, approximated by Tobin’s Q, in Swedish listed firms. In contrast to previous work on Swedish data, I take into account the effect of the family controlling the CEO-position. Furthermore, I try to distinguish between costs of private benefits of control, due to expropriation of minority shareholders, and costs of mismanagement, due to poor management. Data on 144 Swedish public companies observed over the years 1985 to 2000 is used in fixed firm-effects estimations. The results indicate that family firms perform worse than firms with a dispersed ownership structure, with a professional manager in control. It is argued that this is mainly caused by entrenchment of founder heirs, exhibiting poor management skills which, in turn, leads to costs of mismanagement. Control over the CEO-position and the use of super-voting shares is also part of the problem with family control.Essay 2 investigates whether controlling families set dividend policy to fit their personal tax preferences. By using exogenous variation in Swedish tax policy, in the form of the Tax Reform of 1991 and the tax-cuts in 1994, the endogeneity problems in these types of studies are solved. The underlying purpose of the analysis is to check whether controlling families affect firm policy. Using an unbalanced panel of 1,633 firm-years observed during some period from 1985-2000, I find that families do set dividend policies according to their tax preferences. It seems as though the tax treatment of dividends before the Tax Reform of 1991 made family controlled firms pay lower dividends than firms with dispersed ownership, a relationship which was reversed post-1991.Essay 3 examines whether CEO cash compensation in Swedish public firms is affected by the ownership structure of the firm. In particular, it looks at the effect of family control on the pay-performance relationship. It is argued that families have the incentives and the ability to monitor CEOs, which works as a substitute for performance pay. Thus, family controlled firms are hypothesized to rely less on performance-related pay as compared to firms with dispersed ownership. The sample consists of 196 Swedish listed firms, observed in 2004. Three different measures of performance are used: share return, EBITDA and ROA. The results indicate that only share return and EBITDA are used to determine annual cash bonuses. Moreover, there are no clear signs of any monitoring by controlling families. The pay-performance relationships do not seem to be affected by the ownership structures of the firms in any clear way, not even when there is a family connection between the CEO and the controlling family. This result indicates that the competition for high quality CEOs forces all firms to offer a similar pay.