Abstract
This paper shows that when university endowments become limited partners with venture capital. Firms, the performance of their connected portfolio companies improve relative to non-connected ones. Portfolio companies are connected when any of their entrepreneurs attended the corresponding universities for undergraduate- or graduate studies. In a differences-in-differences design I compare initial public offering rates between connected- and non-connected venture capital investments in a treated- and an untreated cohort and estimate this effect to be 6%. Since the unconditional sample mean of initial public offerings is 10%, this is commensurate to a 60% increase in the unconditional initial public offering probability. This effect consists of two separate and potentially different effects, however. First, the effect of obtain a new university endowment as a limited partner, second, the effect of losing an already existing university endowment as a limited partner. Further analysis shows that the main effect is mostly driven by the latter. These results continue to hold in a rich set of robustness checks.