I present a theory of the optimal capital structure and dividend policy for expanding family business groups vertically or horizontally. When private control benefits are substantial, takeover threats impose a constraint on external equity financing. Debt can overcome this restriction but introduces the possibility of bankruptcy where control benefits are also lost. Relative to a horizontal structure, a vertical pyramid enhances internal capital financing, but the family has to share more of the profit from the new firm with its existing shareholders, implying that a pyramid is more likely when external financing constraints are more severe, or the new firm is less profitable but capital intensive. In equilibrium, subsidiaries are less leveraged than horizontal entities directly controlled by the family, because the parent firm supports subsidiaries with greater amounts of internal capital. Within a pyramid, the leverage ratio should decrease from top to bottom because the parent firm has a larger collateralized debt capacity. At the same time, dividend payout should increase from top to bottom because this is how the family transfers wealth out of the subsidiaries, without selling control shares to ensure its control over the parent firm against default. Therefore, the theory predicts a decreasing leverage ratio from top to bottom of the pyramid, supported by a dividend policy where the parent firm pays out less to maximize group internal capital, while subsidiaries pay out more to service the

I am deeply indebted to my supervisor, Alessandro Sembenelli, for his thoughtful guidance during the completion of this thesis and of my doctoral studies. A special thank to Giuseppe Bertola for encouragement, suggestions and many insightful discussions. I also wish to thank Fabio Bagliano, Luigi Benfratello, Giovanna Nicodano and Vittorio Valli for helpful comments that greatly improved the quality of this work. I would like to thank the participants to the DSE Phd seminar and to the “First Italian Doctoral Workshop in Economic and Policy Analysis ” for their feedbacks. Finally, I am grateful to friends and colleagues of the Phd program. In particular, I thank Stella Capuano, Veruska Oppedisano, Francesco Scervini, Tiziano Razzolini and Serena Trucchi for reading and pointing out mistakes in various parts of this manuscript. I gratefully acknowledge Fondazione Collegio Carlo Alberto and its staff for having provided research inputs (seminar series), facilities and technical assistance.

Abstract not found

I hereby declare that this thesis is my original work and it has been written by me in its entirety. I have duly acknowledged all the source of information which have been used in this thesis.

Adequate access to courts by minority shareholders is commonly viewed as an important element of a good corporate governance system. Should shareholders be provided with easy access to courts when judges are unlikely to punish guilty managers? It might seem that having an extra instrument of protection is always better as long as it provides some protection against managerial self-dealing. We present a model which shows that facilitating shareholder litigation in a system where courts are biased towards managers can actually lower efficiency, as it can lead to either excessive litigation or excessive monitoring. The latter effect arises when litigation is very costly for the firm, but cheap for an individual shareholder. In this case, easy litigation does not lead to a greater reliance on the judiciary and results in more, rather than less, concentrated ownership. This is the effect of the optimal adjustment of the ownership structure to an increase in shareholders ’ willingness to bring suits when courts are manager-biased. Our model implies that removing impediments to shareholder litigation in countries where courts are reluctant to protect shareholders may increase the cost of corporate governance there. JEL classification: G32, G34, K41

COPYRIGHT Attention is drawn to the fact that copyright of this thesis rests with its author. A copy of this thesis has been supplied on condition that anyone who consults it is understood to recognise that its copyright rests with the author and they must not copy it or use material from it except as permitted by law or with the consent of the author. This thesis may be made available for consultation within the University Library and

If you wish to contact a Curtin researcher associated with this document, you may obtain an email address from:

lis to pa d-gr ud zi eń