Dynamics and Performance of French Venture-Backed Firms: Empirical Evidence

This paper analyses the factors that drive the dynamics and performance of French new technology-based and venture-backed companies, using a hand-collected database with 216 companies created between 1998 and 2005. The Venture Capital (VC) firms can choose between three outcomes: the company exits from their portfolios by success ? IPO or acquisition on favourable terms ?, exits by failure ? closing down or distress sale ? or remains in their portfolios. We model the probability that a company exits with success (failure) versus remains in the portfolios. We investigate two areas of research: the learning-by-doing models in business demography and the venture capital literature, studying the effectiveness of gradual investment and syndication practices based on the agency theory and the resource-based theory. We assume that either the passive or the active learning-by-doing model may highlight the dynamics of companies and we consider that the efficiency of VC firms at promoting their portfolio companies? potential relies on the multi-stage investment vehicle and syndicate relationships with partners VCs. Our econometrical results support the active learning hypothesis: reducing the time it takes to gather valuable information on the venture, by increasing the number of financial rounds, increases the likelihood of the company exiting from the VC firm?s portfolio, regardless of the exit mode. We also find that the active learning process is oriented towards the less promising companies: VC firms monitor more those companies with a higher probability of exit with failure. Syndication has no significant effect. These two last results orient towards the dark sides of the staged financing and syndication. JEL Classification: G24 G33 L25 M13