Knowledge Transfer within Interorganizational Networks

Abstract

Increasingly, knowledge is recognized as a critical asset, where a firm or an individual’s competitive advantage flows from a unique knowledge base. The subsequent degree to which knowledge is then recognized and valued as a resource has been the theme of many papers on competitive advantage (Barney, 1991; D’Aveni, 1994; Nonaka & Teece, 2001; Prahalad & Hamel, 1990; Spender, 1996; Teece & Pisano, 1994). As a result, the ability to value and leverage external knowledge has become recognized as the basis of competitive advantage. Gulati and Gargiulo (1999) suggest that membership in a networked community satisfies the need for knowledge as a way to help cope with environmental uncertainty. Consequently, inter-organizational networks or communities of practice represent a significant conduit for knowledge transfer to help manage this environmental uncertainty (Madhavan, Koka & Prescott, 1998).1 Researchers in organizational learning have effectively concluded that organizations participating in a networked community will realize superior economic gains from their increased access to knowledge relative to independent or non-aligned firms (Argote, 1999; Baum & Ingram, 1998; Carlsson, 2002; Darr, Argote & Epple, 1995). Although the implications of membership in a network having any structure versus no membership (and therefore no structure) are generally accepted, the implications of the different structural types that these networks can assume are less understood. Networks can accommodate, for example, different levels of competition, different degrees of centralization, and different operational objectives. Knowledge may or may not transfer within different types of networked communities, raising an important question: Given that network membership is accepted as preferable for knowledge transfer relative to non-membership, does the specific network type in question have an effect on the degree to which knowledge will or will not transfer? This is the guiding research question of this article. Prior to an exploration of this question, it should be noted that a multi-entity network (or community of practice) is very different from a dyad, and therefore represents unique challenges with respect to research. Unlike a dyadic relationship, networked communities can take on a life of their own that supersedes the presence of any individual member. Simmel (1950), who studied social relationships, found that social triads (and relationships involving more than three entities) had fundamentally different characteristics than did dyads. First there is no majority in a dyadic relationship—there is no peer pressure to conform. In any group of three or more people, an individual can be pressured by the others to suppress their individual interests for the interests of the larger group. Second, individuals have more bargaining power in a dyad. This is not only true because of percentages, but if one member withdraws from a dyad, the dyad disappears—this is not true in a networked community. Finally, third parties represent alternative and moderating perspectives when disagreements arise. As a result of these differences, multi-entity networks are more complex to study and less understood than dyads.