Technology Network, Innovation And Growth

This paper develops a multi-sector endogenous growth model which embeds a technology network that captures heterogeneous intersectoral knowledge spillovers. Each sector serves both as a distributor and a consumer of knowledge: the former depends on a sector's position in the network; the latter depends on its efficiency in utilising knowledge. The interaction of these forces influences long-run economic growth, sectoral shares and the firm size distribution. The sparsity of the network imposes an upper bound on the impact of knowledge spillovers. In this model, sectors converge to the same growth rate if they belong to the same irreducible network. However, their contributions to economic growth differ substantially, depending on their positions in the technology network and their efficiency in conducting innovation. Consequently, the model has implications for the allocation of innovation subsidies. The gain in economic growth derived from promoting innovation in the sector that utilises knowledge most efficiently is over 10,000 times larger than gain derived from promoting innovation in the least efficient sector.