Do tax incentives increase firm innovation? An RD design for R&D, patents, and spillovers

We present causal evidence of R&D tax incentives' positive impacts on a firm's own innovation and that of its technological neighbors. Exploiting a change in size-based eligibility thresholds for R&D tax relief, we implement a Regression Discontinuity Design using administrative data. We find significant effects of tax relief on (quality-adjusted) patenting (and R&D) that persist up to seven years, and evidence of R&D spillovers on the innovation of technologically close firms. We can rule out elasticities of patenting with respect to R&D user cost of under 2 at the 5 percent level and show that our large effects are driven by financially constrained treated firms.