This paper reexamines the limits of the firm in Research and Development (R&D). Using evidence drawn from industrial laboratories we study the causes and effects of R&D sourcing. We begin with the causes of sourcing, finding that Research Joint Ventures (RJVs), the option to purchase and acquire, and research with federal government contribute to sourced R&D. We then consider the effects of sourcing, RJVs, and the firm's internal research on innovation, as defined by patents and new products. Our results are that sourcing has little effect on innovation, but that RJVs and internal research increase innovation. This suggests specialization: cost saving is the primary motivation for sourcing, while innovation is the primary motivation for RJVs and internal research. Therefore, shared R&D comes in several varieties: R&D sourcing is not concerned with innovation, but consistent with their purpose, RJVs are instrumental in jointly commercializing the research of different firms.
I.
IntroductionThe trading or pooling of research across firm boundaries determines the limits of the firm in R&D, and so is a likely factor in the productive efficiency of industrial research. In this paper we revisit this topic using evidence from U.S. industrial research laboratories. In the case of R&D sourcing, where the evidence is relatively detailed, we examine the causes that lie behind shared R&D. In addition, we study the differing consequences for innovation, as defined by patents and new products, of two forms of shared R&D: R&D sourcing and Research Joint Ventures (RJVs).The main results are that RJVs, the option to purchase and acquire, and research with federal government increase R&D sourcing. Moreover, RJVs and the firm's internal research contribute to innovation. In contrast, R&D sourcing has no such effect. Therefore, internal research and shared R&D are specialized: cost savings are the primary motive for R&D sourcing, while invention and commercialization are the primary motive for internal research and RJVs. Taken together, the evidence suggests that R&D sourcing is neither concerned with, nor is it a threat to, intellectual property, even though RJVs, consistent with their purpose, are instrumental in jointly commercializing the research of different firms.The subject is timely for several reasons. R&D sourcing and RJVs have grown rapidly and play a more important role today than in the past 1 . In this paper R&D sourcing is primarily domestic, but it is similar to international sourcing of high-technology jobs. From 1980 to 1990 new partnerships increased from 200 to 600. In the US CORE and NCRA-RJV databases, partnerships increased rapidly from 1985-1995, but decreased from 1996-1998, apparently due to a decline in the Advanced Technology Program, which supports cooperative R&D. For details see Hagedoorn, Link, and Vonortas (2000). Also see Mowery (1992) for an overview of the motivations for international RJVs and other international cooperative research.2 R&D sourcing could be related or unrelated to RJVs...