Networks, Spillovers and Models of Economic Growth

In this paper we investigate the innovativeness of a network of firms, each of which can experience innovation spillovers (i.e. technological externalities) from its neighbours. A cellular automata framework is used to provide a new theoretical method for analysing such networks. Considering network spillovers produces a nonlinear relationship between the strength of spillovers and overall innovativeness, as opposed to the linear relationship we demonstrate in a standard representative firm model. The network model also allows a theoretical comparison of the common localised spillovers methodology of empirical studies with the global spillovers assumption typical of many endogenous growth models. We find that localised spillovers are not a simple size proportional scaling of the global case. The authors have benefited from helpful comments from Steve Dowrick, Simon Grant, Ross Milbourne, Stephen King, Jeff Borland and seminar participants at the Australian National University and the Australian Econometric Society Meetings 1997