Two for the price of one? : On additionality effects of R&D subsidies ; a comparison between Flanders and Germany

In this paper we empirically test whether public R&D subsidies crowd out private R&D investment in Flanders and Germany, using firm level data from the Flemish and German part of the Community Innovation survey (CIS III and IV). Both the non-parametric matching estimator and the conditional difference-in-difference estimator with repeated cross-sections (CDiDRCS) clearly indicate that the crowding-out hypothesis can be rejected: funded firms are significantly more R&D active than non-funded firms. In the domain of additionality effects of R&D subsidies, this paper is the first to apply the CDiDRCS method.