R&D Partnerships and Innovation Performance: Can There Be Too Much of a Good Thing?
35 Pages Posted: 13 Sep 2014
Date Written: July 2014
R&D collaboration facilitates pooling of complementary skills, learning from the partner as well as sharing risks and costs. Research therefore repeatedly stressed the positive relationship between collaborative R&D and innovation performance. Fewer studies addressed potential drawbacks of collaborative R&D. Collaborative R&D comes at the costs of coordination and monitoring, requires knowledge disclosure and involves the risk of opportunistic behaviour by the partners. Thus, while the net gains from collaboration can be high initially, cost may start to outweigh those benefits if firms engage in multiple collaborative projects simultaneously. This study explicitly considers a firm’s collaboration intensity, that is, the share of collaborative R&D projects in the firms’ total R&D project portfolio. For a sample of 2,891 firms located in Germany, active in abroad range of manufacturing and service sectors and of which 86% are SMEs, we indeed find that increasing the share of collaborative R&D projects in total R&D projects is associated with a higher probability of product innovation and with a higher market success of new products. While we can confirm previous findings in terms of gains for innovation performance, we also find that collaboration has decreasing and even negative returns on product innovation if its intensity increases above a certain threshold. Consequently, the relationship between collaboration intensity and innovation has an inverted-U shape. In particular, costs start outweighing benefits if a firm pursues more than about two thirds of its R&D projects in collaboration. This result is robust to conditioning market success to the introduction of new products and to accounting for the selection into collaborating.
Keywords: Innovation performance, product innovation, R&D partnerships, collaboration intensity, SMEs, transaction costs, selection model, endogenous switching
JEL Classification: O31, O32, O33, O34
Suggested Citation: Suggested Citation