Toward a Quantitative Theory of Informed Finance

We develop a model of informed finance that explains the key qualitative and quantitative features of informed venture capital finance in the United States. We show that the two key features of our model— potential investors are better judges than entrepreneurs of a project’s economic viability and projects require both significant external funding and entrepreneurial inputs—give rise to under-financing with equity and over-financing with debt. We then take the model to the data. Using only four model parameters we match: (1) the venture capitalist’s equity share; (2) the venture capitalist’s expected return and (3) its standard deviation; (4) the probability that a project receives funding; and (5) the probability the venture capitalist loses money on an investment. Our estimated parameters reveal the average quality of an unfunded project; the percentage of uncertainty resolved by the venture capitalists investigation; the percentage of total surplus accruing to the venture capitalist, and the magnitude of underfinancing associated with venture capital finance. ∗Dan Bernhardt gratefully acknowledges financial support from NSF grant SES-0317700. Stefan Krasa gratefully acknowledges financial support from NSF grant SES-0318394 and NCSA computation grant SES050001. We are grateful to Per Stromberg for helping us uncover the relevant empirical regularities. We thank John Cochrane for giving us access to his data and appreciate the guidance that he provided us. We also thank Burton Hollifield, George Pennacchi, Jano Zabojnik and seminar participants at the University of Illinois, Babson College, Virginia Tech, University of Toronto, University of Montreal, and the Federal Reserve Bank of Cleveland.