Abstract

Scholars and practitioners alike now recognize that a firm's capacity to assimilate and use know-how from external sources—what Cohen and Levinthal (1990) called "absorptive capacity"—plays a central role in innovation performance. In recent years, a common strategy pursued by companies to increase their absorptive capacity has been to locate new R&D facilities in close geographic proximity to technology "hotspots" like Cambridge, Massachusetts, or the San Francisco Bay Area. Such a strategy is predicated on the assumption that geographic proximity facilitates absorption. Unfortunately, more than two decades after the publication of Cohen and Levinthal's landmark piece on absorptive capacity, precious little is known about how different organizational strategies and managerial practices—including location choices—actually impact a firm's ability to exploit external sources of know-how. A key barrier to empirical progress on this front has been a lack of direct measures of absorption. In this paper, we develop a novel measure of absorptive capacity that attempts to directly track the influence of external sources of know-how on the internal R&D activities of individual laboratories. We then use this measure to examine laboratory level differences in absorptive capacity and the degree to which a lab's geographic proximity to a given knowledge base influences its absorptive capacity. To identify patterns of absorption, we exploit a quasi-natural experiment that has occurred in the pharmaceutical industry over the past two decades. Since 1989, a number of major pharmaceutical companies (Merck, Novartis, Pfizer, etc.) have chosen to locate new laboratories in one or more major life science hotspots (Massachusetts, the San Francisco Bay Area, and San Diego County). Because these are de novo green-field labs, we have an unusual opportunity to study how the capabilities of the lab evolved over time and whether those capabilities were influenced by the technological activities of the surrounding local scientific and technological ecosystems. Our sample includes 39 R&D laboratories (at varying degrees of distance from three major life sciences hotspots—Massachusetts, San Diego County, and the San Francisco Bay Area). Our findings indicate that geographic proximity is a significant predictor of how much know-how a lab absorbs from a given hotspot. The importance of geographic proximity is also shown to be increasing over time. However, our results also show significant residual variance at both the individual laboratory and company levels, suggesting an important role of managerial practices and policies in driving absorption. The latter finding was consistent with our field interviews of R&D executives from laboratories involved in our study. The study provides further evidence of the geographically bounded nature of knowledge.

Keywords: Geographic Location; Industry Clusters; Knowledge Acquisition; Pharmaceutical Industry; San Francisco; San Diego; Massachusetts;