A growth model with multiple industries is developed to study how industries evolve as capital accumulates endogenously when each industry exhibits Marshallian externality (increasing returns to scale) and to explain why industrial policies sometimes succeed but sometimes fail. The authors show that, in the long run, the laissez-faire market equilibrium is Pareto optimal when the time discount rate is sufficiently small or sufficiently large. When the time discount rate is moderate, there exist multiple dynamic market equilibria with diverse patterns of industrial development. To achieve Pareto efficiency, it would require the government to identify the industry target consistent with the comparative advantage and to coordinate in a timely manner, possibly for multiple times. However, industrial policies may make people worse off than in the market equilibrium if the government picks an industry that deviates from the comparative advantage of the economy.
Ju Jiandong; Yifu Lin, J.; Wang Yong. Marshallian Externality, Industrial Upgrading and Industrial Policies. The World Bank, Washington DC, USA (2011) ii + 36 pp.