Networks and Innovation: Accounting for Structural and Institutional Sources of Recombination in Brokerage Triads

Exequiel Hernandez, Management, The Wharton School, and Sarath Balachandran, Management, The Wharton School

Organization Science Volume 29, Issue 1

Abstract: Research linking interorganizational networks to innovation has focused on spanning structural boundaries as a means of knowledge recombination. Increasingly, firms also partner across institutional boundaries (countries, industries, technologies) in their search for new knowledge. When both structural and institutional separation affect knowledge recombination, aggregate characterizations of ego network attributes mask distinct recombination processes that lead to distinct types of innovation outcomes. We address this issue by focusing on triads as the locus of recombination in networks. We partition firms’ networks into three configurations of open triads—foreign, domestic, and mixed—based on the distribution of the broker and its partners across or within institutional boundaries. We argue that each configuration embodies distinct recombination processes, with foreign triads offering high access to novel knowledge, domestic triads facilitating relatively efficient knowledge integration, and mixed triads balancing the two. We apply this approach to a global research and development alliance network in the biotechnology industry, using countries as institutional boundaries. The results show that domestic triads affect innovation volume (i.e., the productivity of innovation) more strongly than mixed or foreign triads. In contrast, foreign triads have a greater impact on innovation radicalness (i.e., the path-breaking nature of the innovation) than mixed or domestic triads. The findings suggest that different brokerage configurations embody unique recombination processes, leading to distinct innovation outcomes. Our research provides a deeper understanding of how networks and institutions jointly influence distinct aspects of innovation.

Hernandez, Exequiel_Thumbnail Watch Hernandez discuss this research here.

Read the paper here.