Social structure and competition in interfirm networks: the paradox of embeddedness

Administrative Science Quarterly, March, 1997 by Brian Uzzi

Research on embeddedness is an exciting area in sociology and economics because it advances our understanding of how social structure affects economic life. Polanyi (1957) used the concept of embeddedness to describe the social structure of modern markets, while Schumpeter (1950) and Granovetter (1985) revealed its robust effect on economic action, particularly in the context of interfirm networks, stimulating research on industrial districts (Leung, 1993; Lazerson, 1995), marketing channels (Moorman, Zaltman, and Deshponde, 1992), immigrant enterprise (Portes and Sensenbrenner, 1993), entrepreneurship (Larson, 1992), lending relationships (Podolny, 1994; Sterns and Mizruchi, 1993; Abolafia, 1996), location decisions (Romo and Schwartz, 1995), acquisitions (Palmer et al., 1995), and organizational adaptation (Baum and Oliver, 1992; Uzzi, 1996).

The notion that economic action is embedded in social structure has revived debates about the positive and negative effects of social relations on economic behavior. While most organization theorists hold that social structure plays a significant role in economic behavior, many economic theorists maintain that social relations minimally affect economic transacting or create inefficiencies by shielding the transaction from the market (Peterson and Rajan, 1994). These conflicting views indicate a need for more research on how social structure facilitates or derails economic action. In this regard, Granovetter's (1985) embeddedness argument has emerged as a potential theory for joining economic and sociological approaches to organization theory. As presently developed, however, Granovetter's argument usefully explicates the differences between economic and sociological schemes of economic behavior but lacks its own concrete account of how social relations affect economic exchange. The fundamental statement that economic action is embedded in ongoing social ties that at times facilitate and at times derail exchange suffers from a theoretical indefiniteness. Thus, although embeddedness purports to explain some forms of economic action better than do pure economic accounts, its implications are indeterminate because of the imbalance between the relatively specific propositions of economic theories and the broad statements about how social ties shape economic and collective action.

This work aims to develop one of perhaps multiple specifications of embeddedness, a concept that has been used to refer broadly to the contingent nature of economic action with respect to cognition, social structure, institutions, and culture. Zukin and DiMaggio (1990) classified embeddedness into four forms: structural, cognitive, political, and cultural. The last three domains of embeddedness primarily reflect social constructionist perspectives on embeddedness, whereas structural embeddedness is principally concerned with how the quality and network architecture of material exchange relationships influence economic activity. In this paper, I limit my analysis to the concept of structural embeddedness.

THE PROBLEM OF EMBEDDEDNESS AND ECONOMIC ACTION

Powell's (1990) analysis of the sociological and economic literatures on exchange suggests that transactions can take place through loose collections of individuals who maintain impersonal and constantly shifting exchange lies, as in markets, or through stable networks of exchange partners who maintain close social relationships. The key distinction between these systems is the structure and quality of exchange ties, because these factors shape expectations and opportunities.

The neoclassical formulation is often taken as the baseline theory for the study of interfirm relationships because it embodies the core principles of most economic approaches (Wilson, 1989). In the ideal-type atomistic market, exchange partners are linked by arm's-length ties. Self-interest motivates action, and actors regularly switch to, new buyers and sellers to take advantage of new entrants or avoid dependence. The exchange itself is limited to price data, which supposedly distill all the information needed to make efficient decisions, especially when there are many buyers and sellers or transactions are nonspecific. Personal relationships are cool and atomistic; if ongoing ties or implicit contracts exist between parties, it is believed to be more a matter of self-interested, profit-seeking behavior than willful commitment or altruistic attachment (Macneil, 1978). Accordingly, arm's-length ties facilitate performance because firms disperse their business among many competitors, widely sampling prices and avoiding small-numbers bargaining situations that can entrap them in inefficient relationships (Hirschman, 1970). Although some economists have recognized that the conclusion that markets are efficient becomes suspect when the idealization of theoretical cases is abandoned, they nonetheless have tended to regard the idealized model as giving a basically correct view and have paid scant attention to instances that diverge from the ideal (Krugman, 1986).