Working Paper: NBER ID: w9735
Authors: Emmanuel Dechenaux; Brent Goldfarb; Scott A. Shane; Marie C. Thursby
Abstract: At least since Arrow (1962), the effects of appropriability on invention have been well studied, but there has been little analysis of the effect of appropriability on the commercialization of existing inventions. Exploiting a database of 805 attempts by private firms to commercialize inventions licensed from MIT between 1980 and 1996, we explore the influence of several appropriability mechanisms on the commercialization and termination of projects to develop products based on university inventions. Our central hypothesis is that the relationship between a licensee's decision to either terminate or commercialize the invention is driven by the current market value of the invention, as well as the option value of delaying its commercialization. We use a competing risks framework that allows for non- parametric heterogeneity and correlated risks. We find that better appropriability in the sense of more effective patent strength and secrecy has a strong negative effect on the hazard of license termination. The effectiveness of learning has a strong positive effect on the hazard of technology commercialization, while lead time has a negative effect.
Keywords: appropriability; innovation; commercialization; patents; university inventions
JEL Codes: O31; O32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Stronger appropriability mechanisms (P14) | Hazard of license termination (R48) |
| Effective patent strength and secrecy (O34) | Hazard of license termination (R48) |
| Learning (C91) | Hazard of technology commercialization (O36) |
| Lead time (D25) | Hazard of technology commercialization (O36) |