Input additionality of R&D tax reliefs: results from a panel LP-IV approach

Abstract

Policymakers are increasingly relying on R&D tax reliefs to boost R&D and, hopefully, the rate of innovation in the economy. Nevertheless, despite their growing popularity, the cost-effectiveness of these measures remains uncertain in existing literature, posing challenges to their critical evaluation. This paper addresses this issue by introducing a novel methodology to assess the input additionality of R&D tax reliefs. Drawing inspiration from recent advances in fiscal multipliers research and leveraging new macro-level data, the methodology combines an instrumental variable approach with local projections, presenting a clear, flexible, and straightforward framework for estimating additionality across various time horizons in a macro panel context. Applying this methodology to a sample of 20 OECD countries spanning from 2000 to 2018 suggests that the input additionality of R&D tax reliefs is approximately 1.3. This indicates that R&D tax reliefs are particularly effective in stimulating R&D and are likely to be cost-effective. Moreover, the results also show that the additionality of tax reliefs is lower than that of subsidies, which is approximately 1.5. Overall, findings indicate that both policies are particularly effective in spurring R&D.