Spatial Convergence Among Indian Districts: An Econometric Evidence

Do poorer regions catch up with their richer counterparts? This study revisits convergence concepts at a more granular level while focusing on the empirical analysis of India at the district level and for bifurcated districts according to their economic potential. Spatial dependence is an important factor for growth and development among regional economies. We use the Spatial Mankiw–Romer–Weil Model to examine the catching-up process among districts. The results of unconditional convergence, considering the spatial aspect, imply that regions with lower initial income levels benefit from nearby wealthier regions’ economic activities and growth. Additionally, we find clear evidence of conditional convergence across districts in India, both with and without spatial dependence. We enhance the model’s theoretical foundation by incorporating additional variables. The empirical analysis identifies the catching-up process through unconditional and conditional accounting for spatial dependence. Moran’s I and LM tests validate spatial dependence among districts. We account for spatial heterogeneity using Geographically Weighted Regression and endogeneity using spatial GMM in a spatial error model. We use data for the period 2001–2017 based on averages. The results reveal conditional and unconditional convergence patterns similar to those obtained from Solovian regression models. The speed of absolute convergence varies from 0.28 percent per year to 2.57 percent per year. For all Indian districts, conditional convergence is observed to have a significant impact of literacy rate, GDP sector shares, population density, and financial inclusion on per capita growth. There is evidence of unconditional convergence for India with spatial dependence. More specifically, we show that a neighboring district’s initial per capita income significantly impacts all India districts’ growth rate per capita.