The dynamics of competitive intensity

Administrative Science Quarterly,  March, 1997  by William P. Barnett

• E-mail
• Print
• Link

Industries typically end up populated by only a few large institutions that prevail where thousands of smaller organizations have tried but failed. Organization theory offers various explanations for this fact, drawing on diverse rationales but centering on the common theme that large organizations have strengths that enable them to prevail over smaller rivals. This idea pervades organizational economics, where large hierarchy is thought to coordinate action well (Simon, 1945), reap economies of scale and scope (Chandler, 1962), and compensate for "market failures" (Williamson, 1975, 1985). From this view, the evolution of large, complex organizations is prima facie evidence of their relative efficiency (Alchian, 1950; North, 1981). Researchers on the sociology of markets also see large organizations as formidable, but because of factors such as their structural centrality (Pfeffer and Salancik, 1978; Burt, 1992), ties to social elites (Mintz and Schwartz, 1985), or social status (Podolny, 1993). Even neoinstitutional theorists hold a special regard for large organizations, arguing that they maintain their dominant position directly by enforcing conventional practices (DiMaggio and Powell, 1983) and indirectly by portraying and reinforcing the appearances of rationalized structure (Meyer and Scott, 1983). Overall, our theories differ in their particular explanations, but they broadly agree that large, complex organizations are formidable rivals. Consequently, few are surprised by the persistent tendency of industries to become concentrated -- a fact that apparently bears witness to the strength of large organizations.

My purpose here is to show that the strength of large organizations is, when seen in evolutionary perspective, also the source of their weakness and that the evolution of large organizations occurs not because they are formidable rivals but despite their impotence. To demonstrate this, it is necessary to distinguish between two aspects of organizational strength. On the one hand, an organization can be individually viable, having good life chances that result from its characteristics or its position in the organizational environment. On the other hand, an organization can be ecologically potent, generating strong competition felt by other organizations. Typically, we assume that individual viability and ecological strength go hand in hand -- that viable organizations are formidable rivals. This assumption is probably valid in many cases, but is it necessarily true? Or is it possible, instead, that an organization might be able to survive, although it is weak in its effects on other organizations? The answer depends on whether we take an evolutionary perspective. Typically, analysts begin with the features of an organization at a particular time -- such as its strategy, structure, or social position -- and then consider the immediate consequences: whether it will be more or less efficient, powerful, and so forth. An organization's strengths, both individual and ecological, thus are contemporaneously identified. By contrast, in evolutionary perspective an organization's characteristics or position determine its individual viability -- how it will be affected by selection processes -- which in turn affects the development of its ecological strength.

My thesis here is that this evolution hinges on whether organizations are small, simple units or large, complex structures. I argue that when organizations are small, their evolution conforms to a baseline model in which selection processes favor the survival of strong competitors. When organizations are large, however, then selection instead leaves competitively weak survivors, To test these ideas, I build and estimate a model that separates the effects of organizational viability from the effects of competitive strength, with each allowed to evolve as a function of observables. The estimated parameters are then used to show, through simulations, that the model predicts well the evolution of industrial concentration. The estimates also suggest that large organizations are weak competitors even though, and precisely because, they are unlikely to fail.

COMPETITIVE INTENSITY

"Competitive" typically is used to describe the contexts of organizations, as when we refer to competitive political systems (Key, 1942; Downs, 1957) or the neoclassical economic notion of competitive markets (Stigler, 1968). In general, a competitive context is one in which organizations are likely to find themselves in zero-sum relations with one another, directly or indirectly. For instance, competition is stronger among "structurally equivalent" organizations (Burt, 1992) or among organizations within a given "niche" (Hannan and Freeman, 1989) -- both terms that refer to contexts in which it is more probable that organizations will vie for the same pool of resources. In this way, we usually see competition as something that varies from context to context -- stronger when there is a higher probability of zero-sum relations among organizations.