Download this Paper Open PDF in Browser

Road Oft Taken: The Route to Spatial Development

52 Pages Posted: 21 Apr 2014 Last revised: 6 Oct 2016

Gaurav Khanna

University of California, San Diego (UCSD)

Date Written: July 1, 2016


Most estimates of the economic impacts of transit networks compare regions along network routes to neighboring regions away from such routes. However, in the presence of spatial spillovers in economic activity such a method will underestimate the true effect of roads and railways. In this paper, I take into account spatial spillovers in estimating the overall impacts of transit networks in India. I use an empirical strategy relying on the historical placement of major cities: by connecting nodal cities with straight lines, I instrument for the endogenous placement of these networks. Using night-time luminosity data as a measure of economic activity, I estimate the parameters of a predictive model that incorporates these spillovers, which then performs well in an out-of-sample exercise. I find that being close to transit networks between cities led to greater economic activity in the 1990s and that such activity spread to neighboring regions, substantially increasing the overall impacts. Ignoring the spillovers produces income elasticities that are only 27% of the true overall effects of such routes. These geographic externalities led to a rapid rate of convergence in incomes across regions, highlighting how transit networks and spatial spillovers together, strongly determined the geographic spread and temporal changes in the economic development of the region.

Keywords: Roads, highways, transportation networks, spatial development, economic geography, India, Golden Quadrilateral, convergence, regional spillovers

JEL Classification: O10, O18, R40

Suggested Citation

Khanna, Gaurav, Road Oft Taken: The Route to Spatial Development (July 1, 2016). Available at SSRN: or

Gaurav Khanna (Contact Author)

University of California, San Diego (UCSD) ( email )

9500 Gilman Drive
Mail Code 0502
La Jolla, CA 92093-0112
United States


Paper statistics

Abstract Views