Endogenous Transport Investment, Geography, and Growth Take-Offs

A substantial part of the economic growth literature suggests that historically, geographic features such as decreasing trade costs have not only enhanced output, but may also be associated with rapid urbanization and ’growth take-offs’. However, this literature lacks an analytical motivation for lower transport costs, and assumes their decrease to obtain exogenously and at no cost: It thus does not explain why and when such a growth take-off may occur. This paper addresses the issue by investigating how endogenous investment in transport infrastructure may affect growth. For that purpose, it integrates transport investment in the problem of the typical manufacturing firm in a two-region geography model. The results imply that the relation between geography and growth depends crucially on the dispersion of economic activity: Countries with low ’economic density’ may abstain from transport investment, and remain in isolation. For intermediate economic density, regions will invest to improve trade costs, but lose out on growth in order to sustain their transport capital. Finally, high density will incite more endogenous construction of transport infrastructure and ultimately incite firms to agglomerate ∗The author is grateful for helpful comments to Robert Kollmann, Damiaan Persyn, and numerous particpants at RIEF IX 2009 JEL codes: F43, F12, N70, O41 geographically, thus boosting growth due to localized knowledge spillovers.