On proprietary disclosures of investment institutions

Thesis (Ph. D.)--Massachusetts Institute of Technology, Sloan School of Management, 2003.Includes bibliographical references (p. 73-74).We analyze issues related to proprietary disclosures by specialized investment institutions such as hedge funds to their trading counterparties and creditors. In this paper, disclosure can be costly because of the potential for exploitation through competitive trade or "front-running". In chapter 1, we consider a model with a direct revelation mechanism between leveraged investors and their lenders. In this model, the investors need to borrow from lenders with heterogeneous risk-exposures in order to trade. Investors may obtain advantageous terms of borrowing by disclosing their investment strategy, thereby revealing its correlation to the lender's existing risk-exposure. Investors risk being "front-run" by their lender if they disclose, however. We show that in the presence of front-running, the "unraveling" result of full disclosure may not hold. Mandating disclosure has ambiguous welfare effects since it can not only lead to the matching of uncorrelated risks, but also to concentrations of risk. These results have implications for regulations on leveraged investors in financial markets. In chapter 2, we consider an indirect revelation made by an arbitrageur (e.g., hedge fund) to trading counterparties through traded securities. In this model, the arbitrageur has private information about the relative value of two or more securities. We conjecture that in a segmented dealer market, the arbitrageur trades each security with a different dealer so that each dealer sees only one piece of the total position. We show that this "break-up" strategy can be optimal and unique even if given an array of redundant strategies and securities, including "swaps" of the various securities. The analysis of this equilibrium has implications for the kinds of claims held by an arbitrageur's counterparties in a leveraged scenario and their resulting stability.by K. Jeremy Ko.Ph.D