In Chapter 1 we describe a class of Markov models of industry behav ior that is vastly larger than the set of particular models explored in this book. More significantly, there are considerations of obvious and general importance that are neglected in the models we have an alyzed and that must be incorporated in formal models if evolu tionary theory is to address a broad range of phenomena and con tend with orthodoxy on a wide front. There is, for example, a need for analytical insight into the conditions for survival in an evolu tionary struggle in a changing environment -a struggle in which firms’ routines are responsive to environmental variables. Included in this general question are such specific problems as the analysis of evolutionary contests in which firms differ not in production tech niques or R&D policies, but, for example, in markup pricing rules or desired debt-equity ratios. Our analysis in Chapter 7 of the effects of input price changes needs to be extended to better illuminate the conditions under which “standard” results will be realized in dynamic systems involving selection and se arch effects as well as routinized response . Finally, it is noteworthy that none of our formal models recognize any causal role for the firm’ s balance sheet beyond the simple determination of the scale of the firm by its capital stock.The fact that past events influence a firm’s current behavior thro ugh its balance sheet, and in more subtle ways than as a simple scale determination, may be a deep and important complication in the basic story of how the m arket functions as a selection mechanism.
With regard to Schumpeterian competition, our models represent initial steps down a trail that branches in several important direc tions . They provide the basis for a “life cycle” approach to the struc ture of industries or of specific product markets-that is, an account of the way in which new industries are born, mature, and ultim ately stagnate or decline. To construct a theory of this type, we ob viously need to give explicit attention to phenomena s uch as entry, exit, “learning curves,” vintage effects in productive capital, merger, and strategy change. It seems particularly important to develop and ex plore evolutionary models in which firm strategies include routine responses to the actions of rivals, and are occasionally subject to in novative change on the basis of more comprehensive analysis of rivals’ behavior. Product differentiation plays an important role in the histories of many industries; to un derstand this fully, we need to admit more complexity to the demand side of the markets in which our evolving firms operate. We are p ersuaded that detailed modeling of many capital and consumer goods industries requires recognition of both strategic interdependence and product differentiati on. Although we believe that, in this area as in others, the evolutionary approach surpasses orthodoxy in long-run promise, there is no de nying that fruitful modeling of oligopolistic industries remains a challenging task.
Although our discussion of alternative search strategies and of the topography over which search proceeds was quite rich, our treat ment of these topics in the formal models was quite simplistic. We discussed technological regimes and natural trajectories, but in the models firms either drew from an exogenously given population or they searched locally. Aside from the cumulative technology models, we did not treat any “natural” evolution of a technology. Aside from our models in which we permitted firms to differ in their emphasis on innovation and imitation, we did not treat alternative search strategies. Future work on Schumpeterian competition should in clude explicit modeling of technological regimes and firm strategies. More generally, there are important questions to explore involving change in regime- for example, a breakout from an old regime to a new one in which opportunities for innovation are significantly enriched, or perhaps one that requires a significantly different R&D “strategy” for successful exploration.
All of these inquiries would, in addition to being useful in them selves, contribute to the development of an evolutionary view of the “”product cycle” phenomenon in international trade in manufactured goods. Verbal accounts of product cycle theory suggest that what is going on is international Schumpeterian competition in which dif ferent countries have different factor prices, different capabilities for innovating, and perhaps different circumstances affecting the growth of latent productivity. When technological advance in an in dustry is rapid, the countries that are more effective in achieving technical advance have a competitive advantage, even though they may have higher factor prices. As the pace of technical advance slows, technological prowess counts for less and high factor costs im pose a greater penalty. The parallels between this and the Schumpe terian dynamics we explored in Part V are obvious- but so is the fact that a full evolutionary model of the product cycle would require a great deal more work.
Source: Nelson Richard R., Winter Sidney G. (1985), An Evolutionary Theory of Economic Change, Belknap Press: An Imprint of Harvard University Press.