Internal and external R&DF complements or substitutes? Evidence from a dynamic panel data model

We examine the impact of internal and external R&D on labor productivity in a 6-year panel of 304 innovating firms. We apply a dynamic linear panel data model that allows for decreasing returns to scale in internal and external R&D with a non-linear approximation of changes in the knowledge stock. We find complementarity between internal and external R&D, with a positive impact of external R&D only evident in case of sufficient internal R&D. The findings confirm the role of internal R&D in enhancing absorptive capacity and hence the effective utilization of external knowledge. These results suggest that empirical studies examining complementarities between continuously measured practices should adopt more general non-linear specifications to allow for correct inferences.