The dynamics of competitive intensity

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

A straightforward way to model differing competitive strengths is to allow w to vary among organizations, so that j's threat to k is [Alpha.sub.kj] = [w.sub.j][p.sub.kj] where [w.sub.j], indicates the competitive intensity of organization j. This permits the competitive threat posed by organization j to depend not only on its probability of competing but also on how strong an effect it has on k conditional on competition occurring. Usefully, this approach makes it possible to study what determines an organization's competitive intensity by specifying [w.sub.j] as a function of observables. Figure 1 illustrates this point. Organizations j and k may differ in various measurable ways that theory says are important to their competitiveness. Typically, studies estimate the effects that such characteristics have on an organization's own viability -- the effects [B.sub.j] and [B.sub.k] in Figure 1. For instance, we know a lot about how the strategies, sizes, and ages of organizations affect their own life chances. My approach here is to allow such variables also to determine an organization's competitive intensity by estimating [w.sub.j], and [w.sub.k] in Figure 1: the effects of an organization's characteristics on its rivals' life chances.

[FIGURE 1 ILLUSTRATION OMITTED]

Researchers have not looked for what determines competitive intensity. Instead, any variable that makes an organization more viable (in B) is thought also to make it a stronger competitor (in w). This thinking assumes that an organization's viability depends solely on whether it wins the battle for resources within its niche. Increased viability for one organization then requires a competitive victory and, so, implies decreased viability for its rivals. Under this condition, more viable organizations necessarily generate stronger competition, and the effects of variables in B will be mirrored by their effects in w.

The assumption that viability requires competitive success is a useful starting point for the baseline model of competitive intensity. Next I develop that model, allowing viability and competitive intensity to evolve together as organizations age and grow. Then I relax the assumption that viability comes only from competitive success, considering other ways that organizations may increase their viability. When these mechanisms operate, an organization can become more viable without competing more strongly against other firms, removing the necessary link between an organization's viability and its competitive intensity. Developing this idea, I argue that viability and competitive intensity evolve differently for large organizations, leading to the survival of weak competitors. I then extend the model to treat this as an empirical question.

A BASELINE EVOLUTIONARY MODEL

It is useful to start with what one might call the strong-survivor hypothesis, in which over time organizations become both more viable and stronger competitors. This hypothesis can be derived from Stinchcombe's (1965) well-known "liability of newness" argument, which states that survival chances increase as organizations age. Stinchcombe argued that with age, organizations' members become socialized, their procedures become routinized, and their structures and roles become better defined. Externally, ties to important institutional actors, such as suppliers, distributors, and regulators, also take time to establish. According to Stinchcombe's model, organizations learn over time, implying that older organizations are less likely to fail (Freeman, Carroll, and Hannan, 1983). This outcome can also result purely from selection, even when individual organizations do not learn, according to a variant of the mover-stayer model (Tuma and Hannan, 1984).