The relationship between the Norwegian R&D tax credit scheme and other innovation policy instruments

We first study how participation in the Norwegian R&D tax credit scheme affects the probability of receiving other R&D and innovation subsidies. We find no evidence suggesting that using the R&D tax credit increase the probability of receiving direct R&D subsidies in the future, but we cannot exclude the possibility of an immediate positive effect. At the individual firm level, direct subsidies and the tax credit seem to be complements, while at the innovation system level they seem to be substitutes, as the probability of receiving direct subsidies has fallen after the introduction of the tax credit scheme.

Next, we compare the additionality of the R&D tax credit with direct R&D subsidies. Our preferred estimate suggests that each public krone spent on tax credits for firms investing below the 4 million cap on intramural R&D increases private intramural R&D by 2.68 kroner. This estimate builds on an assumption of zero additionality for firms above the 4 million cap. We find that the additionality of subsidies awarded through the Research Council and Innovation Norway is 2.07 and 1.53 respectively. The additionality of grants awarded by ministries and other public agencies is 0.64, and the additionality of R&D subsidies from the EU is 0.75. We stress the potential for both positive and negative biases in these estimates.

Direct R&D subsidies are intended for projects with low private return and high social return. We find that projects funded through direct grants have essentially zero returns. Although surprisingly low, this is consistent with a high quality grant allocation process. Our estimate for the return to R&D projects financed by tax credits is just slightly below the return to R&D projects financed by own funds. This is to be expected as this type of R&D has a lower marginal price. The estimated returns are 16 % and 19 % respectively. All estimates are likely to be downward biased by measurement errors in the R&D variables. Furthermore, there is large variance in the returns to R&D projects. When estimating the return parameters year by year, they vary considerably around their overall mean.

Acknowledgement : This is a part of Statistics Norway's evaluation of the Norwegian R&D tax credit scheme, financed by the Norwegian Research Council.