Human Capital Theory and Venture Capital Firms: Exploring “Home Runs” and “Strike Outs”—A Replication and Extension of Dimov and Shepherd (2005)

Abstract

The present work is a registered report¹ focused on a replication and extension of the findings in Dimov and Shepherd (2005). Our work tests the hypotheses from the original article in an expanded industry context and with an updated sample. The wider sample includes low- and medium-tech industries, and an expanded time frame as opposed to the original sample based on the high-tech wireless industry in a 5-year span. We used the same estimation technique (OLS regression) from the original article but also expand on it by using multivariate regression and seemingly unrelated regression to model multiple outcomes simultaneously as well as negative binomial analyses to accommodate the count nature of outcome variables—initial public offerings as home runs and bankruptcies as strike outs. We also model more fine-grained outcomes of venture capital (VC) investments (underpricing, 180-day return, market value). Our findings support the inference that general human capital has a significant effect on investment success measures. Furthermore, the findings align with predictions about the relation between general and specific human capital in reducing measures of investment failure. Our replication and extension efforts are crucial in advancing the literature with regard to four key points. First, we find that the types of human capital vary in their influence on VC-related outcomes. Second, by examining alternative measures of VC-related outcomes, we illustrate that the effect of human capital on VC outcomes varies depending on the measure of outcome. Third, our results suggest that the relation between human capital and VC-related outcomes varies by industry type and time to exit. Fourth, our reexamine of the findings from Dimov and Shepherd (2005) seems to illustrate support for hypotheses that were unsupported in the original study.