This study examines how corporate venture capital (CVC) participation is associated with the likelihood and timing of initial public offerings (IPOs) among U.S. biotech firms. Using data on 940 firms founded between 1998 and 2015 and observed through 2023, we jointly analyze IPO incidence and time-to-IPO. To address selection bias, we adopt a multi-stage empirical strategy that integrates baseline binary logistic regression, propensity score matching (PSM), entropy balancing (EB), and a Heckman two-step selection model. IPO timing is examined using a Weibull accelerated failure time (AFT) model that accounts for censoring and duration dependence. Across baseline, PSM-adjusted, and EB-weighted specifications, CVC participation is positively associated with a higher likelihood of IPO and a shorter time-to-IPO. However, these associations attenuate and lose statistical significance once selection on unobservables is explicitly addressed using the Heckman framework, suggesting that the observed relationships under observable-balancing designs may partly reflect non-random selection. Company age is robustly associated with both IPO likelihood and speed, while early-stage VC financing increases the probability of IPO without consistently accelerating IPO timing. Industry-level uncertainty and liquidity do not exhibit stable effects across specifications. Overall, the findings indicate that CVC participation is systematically associated with different IPO trajectories among biotech firms under observable adjustment, but these patterns should be interpreted cautiously in light of potential selection on unobserved firm characteristics.
Kim et al. (Fri,) studied this question.