A dynamic duration approach to venture capital exit

This study explores the extent to which correlated multiple stage funding explains variation in the speed of venture exit. I cast venture capital matches in a multivariate survival setting. I construct a panel of nascent entrepreneurs using SDC Platinum-VentureXpert (1990-2000) and use it to estimate the model. I find significant correlation across funding stages. Market effects is the most important factor in explaining the systematic variation in venture exits, while eliminating unobserved heterogeneity explains about a third of the variation. Conventional estimates that assume venture capital exits being driven by static exposure or by observable factors alone are upward biased.