Optimal financial policy : theory and evidence

Abstract

This thesis consists of two essays on optimal financial policy. The first essay develops a theoretical model to explain the optimal capital structure. The second essay empirically studies the determinants of the optimal debt composition.

The first essay looks at the optimal capital structure of a firm governed by a risk and ambiguity averse manager in the presence of information asymmetry between the manager and outside investors. Ambiguity aversion is characterized by the maxmin expected utility model. It is shown that if the degree of risk aversion and ambiguity is high the unique equilibrium of the capital raising game is the separating equilibrium where the low-profit firm issues pure equity and the high-profit firm has a targeted leverage ratio decreasing in the degree of risk aversion and ambiguity. Furthermore, the unique equilibrium of the game is pooling on equity financing if the high-profit firm faces worse pessimistic scenario than the low-profit firm. An increase in the degree of ambiguity is associated with a Pareto improvement if certain conditions are satisfied.

The second essay examines how institutional investors affect the choice between private and public debt financing. Russell 1000/2000 index reconstitution is used as an exogenous shock to institutional holdings, to mitigate endogenous issues. Firms in the top of the Russell 2000 and firms in the bottom of the Russell 1000 share similar characteristics, except that firms in the top of the Russell 2000 have relatively higher institutional ownership due to benchmarking strategies of institutional investors. The main finding is that higher institutional ownership is associated with higher proportion of public debt in debt composition. It is also documented that firms at the top of the Russell 2000 enjoy lower corporate bond spreads. This result is consistent with the view that monitoring and improvement in information environment by institutional investors are beneficial for public debt holders. No significant effect on bank loan spreads is documented, suggesting that private lenders are less likely to rely on public information and monitoring by institutional investors than bond holders.

The results in both essays support the view that information environment is the first-order determinant of financial policy.