It is today widely acknowledged that innovation constitutes one of the most important drivers of economic growth and competitiveness (see e.g. Solow, 1957; Griliches, 1979, 1992; Hall, 1996). Private sector firms’ investment in R&D plays a crucial role in this process not only for the discovery of new technologies, but also for their diffusion.

Because of various well-known market failures though, it is unlikely that left alone, firms would invest the socially optimal amount in R&D. For this reason, governments design various policy schemes to stimulate investment in R&D. In Flanders, the northern part of Belgium, the government has spent 628 million euros on direct support for R&D and innovation for a total of 3,019 projects between 2002 and 2008. Thereby Flanders employs regional-specific policy design – i.e. a dual policy focusing on small and medium-sized firms (SMEs) on the one hand and (international) collaboration, on the other.

The present research aims, on the one hand at evaluating whether these targeted measures are efficient in terms of input additionality, and, on the other hand, whether they translate into innovation output. With respect to input, we find that subsidies accelerate R&D spending in the private sector. When analyzing the impact of the specific policy features on the treatment effect, we find evidence for the efficacy of the policy currently in use. In particular, we find that SMEs have a larger treatment effect than larger-sized firms. We further find that internationally collaborating SMEs have a larger treatment effect than internationally collaborating larger firms or non-internationally collaborating SMEs.

Further, we implement the results of the treatment effects analysis into a series of innovation output models, where R&D is disentangled into purely privately financed R&D (i.e. R&D expenditures that the firm would have spent in any case) and subsidy induced R&D expenditure. We find that both types have a significant positive effect on firms’ innovativeness measured by their share of sales from market novelties. Further, when interacting both types of R&D investment with the specific policy features of the funding scheme under review, we find that the policy-triggered effect on market novelties is highest for internationally collaborating firms.