NATIONAL AND INTERNATIONAL SPILLOVERS FROM R&D: Comparing a Neoclassical and an Endogenous Growth Approach

Two models where productivity growth is caused by spillovers from R&D are analysed using a sample of nine manufacturing industries in six large OECD-countries between 1979 and 1991. The first model is based on traditional productivity analysis where growth in R&D stocks causes productivity growth. The second model is based on the endogenous growth literature where the level of R&D expenditures is assumed to increase productivity growth. The empirical results indicate stronger support for the latter model. The pattern of spillovers is also investigated. The results suggest that spillovers from R&D exist within industries, both nationally and internationally. There is, however, little evidence of spillovers between industries. The empirical evidence further suggests that intra-industry spillovers are confined to industries that are relatively R&Dintensive. Finally, direct foreign investment seem to facilitate the diffusion of R&D results, but we do not find any effect on growth from R&D embodied in intermediate products.