Disagreement Costs, Control and Corporate Finance

Abstract

This paper examines capital budgeting and dividend policy in an environment in which firms need to raise equity financing from new investors to fund projects, and different generations of shareholders may openly disagree over what maximizes value. The standard value maximization objective is incomplete in this setting. I show that the manager's objective should be to maximize his expectation of the net present value accruing to the current shareholders, thereby attending to the dilution of these shareholders' claims when equity is sold to new investors. This framework, which isolates the effect of disagreement by abstracting from both agency and asymmetric information problems, generates numerous implications. First, the hurdle rates firms use for project acceptance/rejection decisions will always be higher than the expected rate of return shareholders demand based on asset pricing models. These hurdle rates will vary across firms so that one firm may accept a project that another firm rejects even though shareholders in both firms demand the same expected return. Second, despite the absence of agency or asymmetric information problems, the manager faces "disagreement costs," represented by the difference between the firm's hurdle rate and shareholders' expected return. The disagreement costs are increasing and convex in the likelihood of disagreement between existing and new shareholders, and affect the firm's investment policy. Third, the firm's dividend policy can be used to optimize the allocation of control over investment policy between the manager/existing shareholders on the one hand and the new shareholders on the other, thereby lowering disagreement costs and the capital budgeting hurdle rate, and allowing the manager to achieve a Pareto superior outcome. That is, dividend and investment policies are inseparable.