Endogenous technology cycles in dynamic R&D networks

Abstract

We study the coevolutionary dynamics of knowledge creation and diffusion with the formation of R&D collaboration networks. The novel examination of a large R&D collaboration network over several decades reveals a pronounced oscillatory (cyclical) pattern in the R&D collaboration intensity, which is not captured by existing theoretical studies. Here, we propose a new model of R&D network formation in which firms form R&D collaborations with others possessing a complementary portfolio of technologies. Innovations and knowledge spillovers alter the composition of these portfolios over time, leading to changes in the network of R&D collaborations. We show that our model is not only able to explain the emergence of oscillatory dynamics in R&D networks, but also has important policy implications. First, we demonstrate that there exists a critical threshold level for spillovers between R&D collaborating firms that must be exceeded for R&D collaborations to effectively contribute to knowledge creation in the economy. The threshold indicates that policies promoting collaborative R&D can only be successful in fostering innovations if they are substantial enough so that spillovers are above the threshold. Second, policies strengthening competition in R&D networks are found to promote oscillatory fluctuations, potentially destabilizing the network.