The Research Tax Credit (CIR) is today the main public support scheme for investment in research and development in France. Given the stakes, both for public finances and for growth potential, rigorous evaluation of public innovation support policies is essential.
This report contributes to these evaluations, which remain underdeveloped, by drawing on numerous data sources: tax data on firms, on the research tax credit, on direct innovation support from Bpifrance, as well as the R&D survey conducted by the Ministry of National Education, Higher Education and Research and PATSTAT data from the European Patent Office. Matching these data sources highlights the segmentation of support schemes across firm populations with different characteristics, as well as the issue of high non-take-up of these innovation support schemes.
Our preferred identification strategy relies on comparing firms that used the CIR before the 2008 reform with firms that never used the CIR during the analysis period (2004-2011). Under the assumption that, in the absence of the reform, R&D spending trends for these two groups of firms would have been parallel, we find positive effects of the reform on R&D spending of around 15% to 18%, implying a tax credit multiplier of 1.3 to 1.5 – that is, one euro of CIR leading to an additional 1.3 to 1.5 euros of R&D spending – but with a confidence interval that does not rule out an effect below 1.
The effect on patent filings shows a positive and significant effect of around 5% on the probability of filing a patent, but no effect on the number of patents filed conditional on having filed a patent during the pre-reform analysis period. This effect of the 2008 reform on an innovation measure allows for an overall positive but nuanced assessment, with a particularly limited effect given the cost in terms of public finances.
