By Steven Casper
Have you ever wondered how biotechnology clusters emerge and become sustainable? Part of the problem in answering that question in the United States arises from the relative scarcity of well-performing biotechnology clusters. For example, only three regions in the United States - Boston, San Diego, and San Francisco - have large biotechnology clusters, despite the wealth of leading universities and hospitals with excellent biomedical research facilities across the country. Still, several states, such as Colorado and Florida, are currently investing billions in constructing new biomedical-research complexes, often with the expectation that these investments will pay off through increased biotech activity.
Social-network analysis provides a computational framework to study the systems and networks that make up biotech clusters. It uses tools and concepts that resemble those used to study biological systems. Moreover, the abundance of data on deal-making, alliances, and career moves in biotech make it a leading area to study with computational methods.
In considering the mechanics behind biotech clusters, one could ask: Why have the San Francisco and San Diego regions dramatically outperformed Los Angeles, despite LA being home to renowned research centers, such as Caltech, and biotech's leading company, Amgen? The ties created by career moves of managers and scientists across companies provides some insight. The San Francisco region not only has a larger critical mass of individuals compared to Los Angeles (1,320 compared to 324), but dramatically higher mobility across companies. High mobility leads to the formation of dense social networks linking companies in the San Francisco area, compared to smaller and less-dense social networks in Los Angeles.
Dense networks within a cluster of companies can lead to the quicker diffusion of important technological advances and market intelligence, creating a regional advantage of sorts. Moreover, from a career-management perspective, social networks can be used as a source of referrals to find jobs should a company fail.
Within the San Francisco area, career-affiliation networks create easy access to individuals working across dozens of companies. The average number of individuals one needs to "go through" to contact an individual at another company is 1.6. Within Los Angeles, it takes 2.4 contacts, despite the network being less than one-fourth as large as that in San Francisco. As of 2006, only 36 Los Angeles biotechnology companies were "reachable" within the core network, compared to 140 within San Francisco. Given that advantage, talented individuals benefit from moving to a well-established biotechnology cluster.
Computational research can also explore the historical emergence of social networks within biotechnology clusters. For example, successful biotechnology clusters are associated with the early emergence of ties between founders of firms. In San Diego, for example, one such backbone emerged around a group of senior managers and scientists linked through employment at Hybritech, before Eli Lilly & Co. acquired it in 1986. After the acquisition, most senior managers eventually left the company to become entrepreneurs, leading to the establishment of at least a dozen companies over the next decade. Ties linking the Hybritech alumni were used to obtain venture financing, organize the founding teams of companies, and recruit senior managers for new ventures. The network backbone of San Diego biotechnology can be traced to these activities. Eventually, more than 150 companies employing at least 900 senior managers would establish ties to this early network, helping to explain the growth and success of San Diego biotechnology.
Universities alone cannot create biotech clusters. Although most companies draw on local universities for their founding science, the social-network ties linking personnel working within the industry appear to stem far more frequently from shared career experiences in companies. Access to world-class biomedical research might be necessary for biotechnology clusters to succeed, but the key social-network dynamics leading to a cluster's ultimate success resides in the biotechnology industry itself.
