Notching R&D Investment with Corporate Income Tax Cuts in China

We analyze the effects of a Chinese policy that awards substantial corporate tax cuts to firms that increase R&D investment over a given threshold, or notch. We exploit this quasi-experimental variation with administrative tax data in order to shed light on longstanding questions on the effects of fiscal incentives for R&D. We find large responses of reported R&D using a cross-sectional bunching estimator that is new to the R&D literature. We also find significant increases in firm-level productivity, even though about 30% of the increase in R&D is due to relabeling of administrative expenses. Anchored by these reduced-form effects, we estimate a structural model of R&D investment and relabeling that recovers a 9.8% return to R&D. We simulate alternative policies and show that firm selection into the program and the relabeling of R&D determine the cost-effectiveness of the policy, and the effects on productivity growth.