Do Neighbourhood Effects Matter for the Geographical Concentration? Evidence from the Indian Industries

Abstract

This paper investigates the impact of neighbouring effects on the geographical concentration of manufacturing and service industries at the district level using India’s Economic Census (2013) data. As empirical literature suggests, spillovers do not recognize areal boundaries, and spatial dependence among regions needs to be incorporated while quantifying the geographical concentration of industries. In this context, we employ the spatially weighted Ellison-Glaeser (EG) index to evaluate the impact of neighbourhood effect on the spatial concentration of 71 manufacturing and 120 service industries in India. Using aggregate data at the district level by covering 636 districts and 34 states and union territories in India, empirical results exhibit that the magnitude of the neighbourhood effect does not substantially impact the geographical concentration of 191 industries. More specifically, the neighbourhood effect is over-shadowed while considering an aggregate of 636 districts covering all states and union territories in India. To gain more insight into the role of neighbourhood effects and for robustness checks, we measure manufacturing and service industries’ geographical concentration within India’s 29 contiguous states and union territories. Our subsequent empirical evidence validates that neighbourhood effects are well captured by the spatially weighted Ellison-Glaeser index for the top three highly concentrated manufacturing and service industries. Moreover, we find that the spatially weighted EG index plays a predominant role while computing geographical concentration for the highly concentrated manufacturing and service industries across various Indian states and union territories at the district level in India.