Firm Boundaries and Innovation: Empirical Evidence from Entrepreneurial Finance

Abstract

High-technology innovation often takes place in small, independent startups; it can also take place in the research-and-development facilities of large, established corporations. A third possibility is that innovation will take place in a hybrid of these two ideal types, for example via corporate venture capital (CVC) or via distribution channel agreements. Using detailed data on a large number of high-technology firms located in the US, Western Europe and Israel, we test a set of predictions about innovation and firm boundaries that flow from previous empirical studies and formal models. Our central research questions relate to how the boundaries of the firm are affected by the level of competition in product markets, portfolio firm research intensity, national institutions, cash flow risk, innovation risk by rivals, and managerial resources. We find some evidence that hybrid organization is more likely (relative to stand-alone startup status) for more established firms (with more employees and larger revenues). Hybrid organization is also more likely when competition is declining, less likely when competition is static, and no more likely when competition is growing. In addition, our results are consistent with previous findings that CVC is associated with herding behavior as opposed to participation in the most innovative industry segments. Hybrid organization is associated with patent possession by the high-tech startup, but CVC partners are less concerned than other investors (VCs) that the patents provide useful barriers to entry. Results pertaining to cash flow risk, research intensity, management team, and innovation risk by rivals were neither consistent in terms of sign nor robust to various specifications.