A Structural Network Formation Model and Its Application in R&D Networks

Xiaodong Liu is a Professor of Economics at University of Colorado Boulder and an elected fellow of the Spatial Econometrics Association. He earned his PhD in economics from the Ohio State University. His current research focuses on estimation of social and spatial interaction models and structural estimation of game. His work has appeared in The Review of Economics and Statistics, Journal of Econometrics, Econometric Theory, Journal of Business & Economic Statistics, Journal of Applied Econometrics, and Journal of Economic Behavior & Organization, among others.

We introduce a stochastic R&D network formation model where firms choose both R&D efforts and their collaboration partners. Neighbors in the network benefit from each other’s effort levels through local technology spillovers, and there exists a global competition effect reflecting a strategic substitutability in efforts. We provide a complete equilibrium characterization in the form of a Gibbs measure, and show that the model is consistent with empirically observed R&D networks. We then use our equilibrium characterization to show that the model can be conveniently estimated even for large networks. The policy relevance is demonstrated with the analysis of subsidies to R&D collaboration costs.