Intermediation and Value Creation in an Incomplete Market: Implications for Securitization∗ Preliminary Draft

This paper studies the impact of financial innovations on real investment decisions. We model an incomplete market economy comprised of firms, investors and an intermediary. The firms face unique investment opportunities that are not spanned by the securities traded in the financial market, and thus, cannot be priced uniquely using the no-arbitrage principle. The specific innovation we consider is securitization: the intermediary buys claims from the firms that are fully backed by cash flows from the new projects, pools these claims together, and then issues tranches of secondary securities to the investors. We first derive necessary and sufficient conditions under which pooling provides value enhancement from the new projects that are undertaken, and the prices paid to the firms are acceptable to them compared to the no-investment option or the option of forming alternative pools. We find that there is a unique pool that is sustainable, and which may or may not consist of all projects in the intermediary’s consideration set. We then determine the optimal design of tranches, fully backed by the asset pool, to be sold to different investor classes. The new securities created by the intermediary could have up to three components, one that is a marketable claim, one that represents the arbitrage opportunities available in the market due to the special ability to design and sell securities to a subset of investors, and a third component that is the remainder of the asset pool sold to investors at a price not exceeding arbitrage-based bounds. The presence of these three components in the tranching solution has a direct bearing upon the size of the asset pool, and therefore, on value creation due to financing additional projects. ∗The authors thank Kose John, Roy Radner and Rangarajan Sundaram for useful discussions. They also thank participants in seminars at Columbia University, Ente Einaudi, Rome, University of Melbourne, New York University, Rutgers University, the University of Venice, the Caesarea Center, Herzliya, Israel, and Stanford University, as well as in the 2004 European Finance Association meetings in Maastricht, the 2004 Informs meetings in Denver, and the 2005 European FMA Conference in Siena, for comments and suggestions made during this research. †Leonard N. Stern School of Business, New York University, 44 West 4th St., New York, NY 10012. E-mail: [email protected], [email protected], [email protected].