Evaluation of Public R&D Policies: A Cross-country Comparison

Technological change is an important determinant of long-run productivity which is essential for securing competitiveness both at the firm-level and for the economy as a whole. Public authorities expect that increasing R&D investment causes intensified technological progress and finally accelerates growth in the long-run. In this line, countries cannot merely rely on public R&D (i.e. conducted by universities or public research centres), but they have to make sure that R&D is also performed at business level. In order to stimulate these private R&D activities, governments usually offer a wide range of public incentives like R&D subsidies, tax credits, technological consultancy etc. In this paper, we empirically evaluate whether public support for innovation spurs investment at the firm level. Conducting a treatment effects analysis, we investigate whether public R&D funding crowds out private investment in the business sector and whether the government could further foster R&D by supporting currently non-subsidized firms. The analysis is based on harmonized micro data from the 4th wave of the Community Innovation Survey (CIS4) covering the years 2002 – 2004 of Belgium, Germany and Luxembourg, as well as a CIS harmonized survey from South Africa, and Spanish data from the Panel de Innovación Tecnológica (PITEC) of the year 2004. In addition, we also test for possible misallocation of public funds. Our sample concerns only innovative firms and covers manufacturing as well as business related services sectors. In total, the sample consists of 9790 observations of 5 different countries, out of which 3854 received R&D subsidies. Using a non-parametric nearest-neighbor matching, we find that firms that received subsidies would have invested significantly less in R&D and innovation if they would not have received public support. Full crowding-out can thus be ruled out for all the countries of our sample. On similar grounds, when estimating the treatment effect on the untreated, we find that untreated firms would on average invest significantly more if they would receive subsidies. With the exception of one country, all countries of our sample would thus benefit from extending existing innovation policies to currently non-subsidized firms. Finally, these two matching results, i.e. the treatment on the treated and the treatment on the untreated, can be combined in order to test for misallocation of public funds. Misallocation of public funds would be present if the treatment effect of the untreated was significantly larger than the treatment effect on the treated. Our analysis does not uncover any systematic misallocation of public funding for the countries under review, though.

Czarnitzki, Dirk and Cindy Lopes-Bento (2010), Evaluation of Public R&D Policies: A Cross-country Comparison, ZEW Discussion Paper No. 10-073, Mannheim, published in: World Review of Science, Technology and Sustainable Development.