The dynamics of R&D network in the IT industry

In this paper, we provide an empirical analysis of evolving networks of successful R&D collaborations in the IT industry (consisting of firms that obtained patents in the technological category of computers and communication) in the U.S. between 1985 and 1995. We first show that the R&D network has become more extensive, more clustered, and more unequal in the sense that 'stars' have emerged in the network. We then analyze the effect of the existing network structure in the process of new R&D collaboration formation. We control for unobserved similarities among firms based on the community structures within the network that the algorithm developed by Girvan and Newman (2004) identifies and find a significant cyclic closure and preferential-attachment effect.