Neural networks and the evolution of firms and industries: An application to UK SIC34 and SIC72

This paper considers whether neural networks might be used to analyse firm activity and the evolution of industries. The key findings of the simulation results used are summarised as follows. While efficiency seeking behaviour has growth advantages, compared to unchanged firms, these are small compared to the growth advantages that are displayed with firms that are able to exploit input use variability. In addition the two sectors analysed here (UK SIC34 and SIC72) show different profit implications of these growth advantages. In SIC34 an increase in firm growth caused by strategic flexibility coincides with an increase in profitability, whereas in SIC72 the increase in firm growth coincides with a profitability reduction. This difference is explained in terms of the differing market structures in the two sectors along with the differing effects of market shocks. Finally the market structure effects of differing firm types have been analysed. It is shown that factor flexibility generates relative growth advantages that benefit smaller firms. But strategic flexibility generates relative growth effects that benefit larger firms.