States Diverge, Cities Converge

Abstract

This paper takes a fresh look at growth convergence in India, combining insights from macroeconomics and urban economics. It departs from the existing literature in three ways. First, the paper assesses growth patterns across districts and across places below the district level instead of taking the state as the unit of analysis. Second, it relies on household expenditures per capita, instead of gross domestic product per capita, to measure living standards. And third, it uses a Bayesian model averaging approach to identify the key drivers of local growth, instead of the classical econometric approach. The paper finds absolute convergence in living standards across districts and places below the district level, with locations in the gray area between rural and urban growing fastest. In assessing conditional convergence, it finds that geography is a strong predictor of local growth, but population density is not. Market access, electrification and transport infrastructure matter, but irrigation and housing investments do not. The quality of state-level governance has a significant impact on local growth, but variations in city governance are only mildly relevant. The share of medium and large firms plays a role, but the sectoral structure of economic activity does not. And the coverage of primary education is an important predictor of subsequent growth, but not that of other levels of education. Strong convergence at the local level can be reconciled with lack of convergence at the state level if low-income states fail to generate enough locations with the "right" characteristics.