| My Ph.D. dissertation consists of three essays. The first essay, titled "Litigation Risk and Corporate Financial Communications Policy," discusses corporate financial communications strategy, which minimizes litigation risk based on sharp stock price changes brought under SEC Rule 10b-5. I find that regularly scheduled, gradual, noisy information release is needed, but this strategy is inconsistent with full and prompt disclosure of all information as implied by Rule 10b-5 case law. The optimal ex ante disclosure policy is time inconsistent. However, time inconsistency may be resolved by a reputational effect if the management is patient enough in the infinitely repeated setting. Even if shares are correctly priced based on public information in the market, on-going communications between firms and the market may still be needed.; The second essay, "Managerial Choice of Cash Flow Volatility," challenges the well-known "Asset Substitution Proposition." Assume that a firm's cash balance follows a Brownian motion, the volatility of which is controlled by management acting in the interests of shareholders. Assume the existing debt has covenants which preclude additional borrowing and that bankruptcy is triggered when the cash balance hits zero. In the event of bankruptcy, some bankruptcy costs are sometimes borne by equity. Bankruptcy costs being borne by equity mitigates shareholders' desire for risk. I show that for low levels of debt, shareholders prefer to minimize the volatility of the cash balance. I also work out the critical face value of the debt above which shareholders are risk-seeking (desiring to maximize the volatility of the cash balance) rather than risk-avoiding.; The third essay, "Cap on Recovery of Damages, Free Options, and Other Related Issues in Securities Fraud Cases",{dollar}sp1{dollar} discusses damages and other related topics in the fraud-on-the-market cases in this paper. First, we analyze the impact of the cap on recovery specified in the recently passed "Private Securities Litigation Reform Act of 1995." We observe that the new law has not overcome the "free option" problem inherent in such cases when the award of damages depends on the market price. We then discuss an alternative method, a piece-wise linear award function, to calculate the damages award. The simulation results show that the new damages measure can fit the degree of the fraud better. ftn{dollar}sp1{dollar}Co-authored with Philip H. Dybvig and Rachel Scwartz. |
