The Size and Specialization of Direct Investment Portfolios
47 Pages Posted: 17 Mar 2010 Last revised: 5 Feb 2016
Date Written: July 29, 2010
Existing models of the size and scope of investment activity traditionally assume an infinite pool of ex-ante identical projects, despite the fact that managers often face a limited choice of projects that vary in quality. In this paper, we investigate optimal project selection in a model in which a portfolio manager observes a limited pool of heterogeneous investment opportunities. We use an order statistics argument to derive predictions on the size and scope of the investment portfolio. Counter to existing models, we show that the number and specialization of investments within a portfolio are substitutes; variables that increase (decrease) the set of available projects or returns to investment activity, such as skill (competition) cause the portfolio to be larger (smaller) and more generalized (specialized). We test our model's predictions and document new stylized facts in a setting with highly skewed payoffs, formalized specialization, and a high shadow cost of access to potential projects: the U.S. venture capital industry.
Keywords: Specialization, Investments, Size, Order Statistics, Venture Capital
JEL Classification: G31, G24, G11, D01
Suggested Citation: Suggested Citation