Mamatzakis, E. and Tsionas, M. (2018) Revisiting the returns of public infrastructure in Mexico : a limited information local likelihood estimation. Economic Modelling, 75. pp. 132-141. ISSN 0264-9993
productivity_infra_June2018.pdf - Accepted Version
Available under License Creative Commons Attribution-NonCommercial-NoDerivs.
Download (2MB)
Abstract
This paper revisits the issue of accurately decomposing productivity growth to the impact of public infrastructure at firm level for Mexican industry whether the underlying functional form is a profit or a cost function. Our framework decomposes productivity growth into different components, and in particular the contribution of public infrastructure. We also propose a novel limited information local likelihood (LILL) estimation method that adequately deals with the issue of the endogeneity and model misspecification. The reported evidence shows that public infrastructure enhances productivity growth through profit gains and cost savings in all ten two-digit Mexican industries, though some variability across time exists, notably in the nineties and the 2000s when a shortage of infrastructure is observed.