Do subsidies increase firm productivity? Evidence from Chinese manufacturing enterprises *

This paper examines the effect of government subsidies on firm-level productivity. The subsidy-promoting policies are oftentimes justified on the ground that subsidies increase the recipient firms' innovation and productivity. We develop a novel (semiparametric) method to structurally identify the firm-level productivity in the presence of subsidies. Unlike previous studies, we assume that subsidies are endogenous. To correct for the endogeneity problem, we use a switching regression in our semiparametric production function. Our method provides an improved framework for estimating the production function with endogenous and/or censored variables. We apply our method to study the effect of subsidies on firmlevel productivity in the Chinese machinery industry, using panel data from 1998 to 2007. We find strong empirical evidence that subsidies have a positive effect on firms' persistent productivity. Subsidies may allow firms to invest more on research and development (R&D) and thus improve productivity. Using the available R&D data, we find evidence to support this explanation that more subsidies are associated with higher R&D investment. We also carry out a decomposition analysis to separately examine the subsidy effect on output via the input channel and the persistent productivity channel. Opposite effects are found from the two channels and overall subsidies have a positive effect on a firm's output. (c) 2022 Elsevier B.V. All rights reserved.