Endogenous Network Formation in Patent Contests and Its Role as a Barrier to Entry

Abstract

In a setting of R&D competition, we study how collaboration affects strategic decisions during a patent contest, and how the latter influences the collaboration network structures the firms can form. We use an all pay auction approach to endogenize both network formation and R&D intensities, and to take heterogeneous and private valuations for patents into account. We find that the complete network is not always the only pairwise stable one. The other stable networks have the realistic property that some firms drop out of the contest. Thus, 'weak' cooperation can serve as a barrier to entry on the market for innovation.

Competition and networks of collaboration

“…This paper contributes to the literature on collaboration between rivals. Related models are studied by Bloch (), Yi (, ), and Yi and Shin () in the context of coalition formation, and Joshi (), Goyal and Joshi (), Goyal and Moraga‐Gonzalez (), Marinucci and Vergote (), Mauleon et al (), and Grandjean and Vergote () in the context of network formation. These studies focus on R&D collaboration among firms as the main application.…”

“…In this specification, the agents may derive a positive net value of collaboration without taking into consideration a tournament outcome. In the vast majority of the literature (see Goyal and Joshi (), Goyal and Moraga‐Gonzalez (), Marinucci and Vergote (), and others), collaboration is assumed to be costly. I consider the possibility of both costly and costless links.…”

“…Definition pins down the size of each group in the stable networks characterized in Theorem uniquely (up to indifference). The literature on pairwise stable networks of collaboration, such as Goyal and Joshi (, Proposition 3.5) and Marinucci and Vergote (, Propositions 1 and 2), puts bounds on sizes of interconnected groups. The multiplicity of pairwise stable networks in those models follows from the inability to rule out failures of coordination.…”

The effect of entry on R&D networks

“…The formation of R&D collaborations has been studied extensively in recent years in economics, management, and strategy, mainly due to the widespread evidence that there is significant increase in R&D alliances, mostly in high tech industries . Firms competing even in the same markets form R&D collaborations, aiming at the joint reduction of production costs (see Goyal and Moraga, ; Goyal and Joshi, ; Goyal, Konovalov, and Moraga‐Gonzalez, ; König, ; König, Liu, and Zenou, ), the growth of each firm's knowledge by sharing existing knowledge of all collaborators (see König et al., , ; König and Rogers, ), and the participation in innovation contests (see Czarnitzki, Etro, and Kraft, ; Marinucci and Vergote, ) . The existing empirical evidence confirms our main motivation, showing that entry and exit is often observed in these industries across time (see Gulati, Sytch, and Tatarynowicz, ; Tomasello et al., ; König and Rogers, ; and in a more general setup, Acemoglu et al., ).…”

Endogenous Network Formation in Patent Contests And Its Role as A Barrier to Entry

“…As mentioned in footnote (12), strong stability proves to be too strong a refinement. For an example of where multiple pairwise stable networks exist but none of them is strongly stable, we refer the reader to Marinucci and Vergote [].…”