When and Where Does Government-Subsidized R&D Work? Evidence from China

Abstract

This study examines the effects of government-subsidized R&D program by looking at both the influences of projects screening system and market conditions. Based on a panel dataset of Chinese manufacturing firms, we find that Innofund (one of the largest Chinese government programs supporting corporate R&D activities) generally chooses firms with higher productivity. Moreover, Innofund-backed firms experience significantly magnified productivity improvements compared to their non-Innofund-backed counterparts. However, the ex-ante project selection effects and the ex-post effects of Innofund vary according to the project screening policy change and differ across markets. The project screening system became substantially more decentralized after 2005 due to exogenous policy change by then. We find Innofund tended to be more risk-adverse by choosing firms with higher productivity at the time of project selection before 2005 than that of after 2005. However, we observe higher ex-post effects of Innofund after 2005. Finally, Innofund has stronger effects in less market-oriented provinces. Identification problems are handled using the propensity score matching approach and the instrumental variable approach.