Does infrastructure stimulate total factor productivity? A dynamic heterogeneous panel analysis for Indian manufacturing industries

In the standard literature, empirical assessments of the productivity effects of infrastructure are marked by contradictory results with little robustness. In this context, this paper tests the effects of public infrastructure on the total factor productivity (TFP) of Indian manufacturing industries. We utilize a new productivity database on Indian manufacturing for the period 1980-2012, as well as a range of infrastructure measures, for empirical analysis. Based on a bootstrap based test, preliminary evidence is found for cointegration between infrastructure and TFP of various manufacturing sectors. Next, the productivity effects of infrastructure are estimated using a cross-sectionally augmented pooled mean group estimator, where we control for sectoral dependence, endogeneity and heterogenous effects of infrastructure across various sectors. Overall, the results of the analysis confirm the presence of a positive and sizeable effect of infrastructure on manufacturing productivity. We find wide disparities in the productivity effects of different types of infrastructure such as road, rail, energy, port and telecommunication. Specifically, the results show that 1 percent increase in the aggregate infrastructure stock results in productivity growth by 0.16 percent. Policy recommendations are made on the basis of these findings. (c) 2020 Board of Trustees of the University of Illinois. Published by Elsevier Inc. All rights reserved.

Automated summary

This study examines the effects of public infrastructure on the total factor productivity of Indian manufacturing industries, finding a positive and significant impact of infrastructure on manufacturing productivity. There are wide disparities in the productivity effects of different types of infrastructure, with a 1 percent increase in aggregate infrastructure stock resulting in a 0.16 percent productivity growth. Policy recommendations are made based on these findings.