Theorists should aim to tell the truth in their theorizing, but they cannot aim to tell the whole truth. For to theorize is precisely to focus on those entities and relationships in reality that are believed to be central to the phenomena observed- and largely to ignore the rest. To advance a new theory is to propose a shift of focus, to recognize as central considerations that were previously ignored.
In this chapter, we have focused upon the realities of organiza tional functioning that form the foundation of our evolutionary theory. Foremost among those realities are the factors that tend to limit the individual firm to the exercise of a distinctive package of economic capabilities that is of relatively narrow scope. Essential coordinating information is stored in the routine functioning of the organization and “remembered by doing./I As in the case of individ ual skills, the specificity of the behavior involved is simply the ob verse of its effectiveness; also, much of the knowledge that underlies the effective performance is tacit knowledge of the organization, not consciously known or articulable by anyone in particular. These cog nitive factors are reinforced by motivational ones associated with the control of intraorganizational conflict. Prevailing routines define a truce, and attempts to change routines often provoke a renewal of the conflict which is destructive to the participants and to the organiza tion as a whole.
As a first approximation, therefore, firms may be expected to be have in the future according to the routines they have employed in the past. This does not imply a literal identity of behavior over time, since routines may be keyed in complex ways to signals from the environment. It does imply that it is quite inappropriate to conceive of firm behavior in terms of deliberate choice from a broad menu of alternatives that some external observer considers to be “available” opportunities for the organization. The menu is not broad, but narrow and idiosyncratic; it is built into the firm’s routines, and most of the “choosing” is also accomplished automatically by those routines. This does not mean that individual firms cannot be brilliant successes for a short or long period: success and failure depend on the state of the environment. As long as the world rewards great tennis playing, great tennis players will succeed in the world, regardless of their talents as physicists or pianists. Efforts to under stand the functioning of industries and larger systems should come to grips with the fact that highly flexible adaptation to change is not likely to characterize the behavior of individual firms. Evolutionary theory does this.
As a second approximation, firms may be expected to behave in the future in ways that res emble the behavior that would be pro duced if they simply followed their routines of the past. Just what “resemble” means here is an i mportant and complex question. It is a question that is particularly illuminated by inquiry into the factors that hold behavior to the channels of routine, since whatever change takes place may be expected to follow the path of least resistance. But to assess where the resistance is likely to be least requires a discrimi nating analysis of the relative s trengths of different sources of re sistance. This is the great challenge of the subject of “organizational genetics”- to understand how the continuity of routinized behavior operates to channel organizational change. Our discussion of rou tines as targets and as components addresses this problem in a pre liminary way, but the subject has barely been defined and the real work remains to be done. The particular models that follow are built on very simple assumptions regarding these matters, particularly the assumption that capacity expansion can be achieved with faultless replication of routine, and similarly that contraction of a firm is sim ply a scaling down of the same routinized pattern of operation. The discussion above provides support for these assumptions as a starting point for model building, but it contains some i mportant ca veats that should be kept in mind in future work. It also makes even more suspect the assumption that imitation of another firm’s rou tines can be accomplished perfectly . However, for the limited pur poses of these particular models, use of a weaker assumption would do more to complicate the analysis than to change its substantive content. The important consideration captured by the models is that i mitation, though costly and imperfect in the individual instance, is a powerful mechanism by which new routines come to organize a larger fraction of the total activity of the system.
In the contemporary economy, some portion of business behavior is closely calculated by sophisticated optimization methods. Another portion is innovation activity shaped by the creative proble m solving insights of scientists, engineers, and managers. A full ac count of business behavior has to deal with these sophisticated por tions, and the imagery of routinized behavior does not have the clear validity and power here that it has in discussing, say, a family firm whose product mix has remained unchanged for generations. We have argued, however, that the notion of routine behavior does have application in this sophisticated realm, though in a qualfied sense. For example, the skills of the highly trained operations researcher, scientist, or manager are reflected in characteristic, highly patterned forms of problem-solvi ng activity. The scope of the expertise in volved in each case is defined by a certain class of problem- solving techniques and heuristics. For this and other reasons, even the so phisticated problem- solving efforts of an organization fall into quasi-routine patterns, whose general outlines can be anticipated on the basis of experi ence with previous problem-solving efforts of that organization. But the patterning of the problem- solving activity is reflected only vaguely in the immediate outcomes of that activity and even less clearly in the gross changes in firm behavior that these problem solutions may tri gger. From the viewpoint of an external observer who has no access to the sophisti cated workings within the organization, the results are hard to predict and on that ground are best regarded as stochastic. This is the approach we take in the evo lutionary models that follow.
Source: Nelson Richard R., Winter Sidney G. (1985), An Evolutionary Theory of Economic Change, Belknap Press: An Imprint of Harvard University Press.