Network Dynamics and Field Evolution : The Growth of Interorganizational Collaboration in the Life Sciences

The development of collaborative network ties and configurations in the commercial biotechnology field is examined, in order to demonstrate how patterns and the logic of interaction and attachment emerge, take root, and shift over time. The ties between biotechnology firms, public research organizations (including public and private universities, nonprofit research institutes, and research hospitals), pharmaceutical companies, and venture capital firms are assessed. Three approaches are taken to analyze and understand network structures: (1) examination of the network to see how random or uniform the process of expansion is; (2) mapping the field's development by drawing network configurations, in order to assess the extent of attachment bias; and (3) assessing alternate mechanisms of attachment by examining network formation and dissolution. The data sample, comprising interorganizational agreements for United States and European firms, was developed from the database BioScan and covered the years 1988-1999. Also utilized were insights drawn from interviews of scientists, managers, and university faculty. The dominant forms of partner organizations were coded according to a 24-cell matrix of types of partner organizations and types of activities. Three types of analysis are conducted: (1) degree distributions, (2) discrete-time network visualizations, and (3) attachment bias. Four attachment mechanisms are hypothesized and tested: (1) accumulative advantage, (2) homophily, (3) follow-the-threat, and (4) multiconnectivity. Results show how different rules for affiliation shape network evolution. It was found that, over the period 1988-1999, collaborative activities shifted from commercialization to finance and research and development. The commercialization activities of early-entry corporations are supplanted by universities, research institutes, venture capital, and small firms. As collaborative activities increase and diversify, subnetworks form. Structural components, choices, and opportunities reinforce an attachment logic based on differential connections to various collaborators. (TNM)