The Process of an Innovation Cycle 121 



bly affects the supply of an existing product. The industry is 

 assumed to be competitive and is composed of small firms. The 

 producers with the highest skills (the better-educated in many 

 cases) will be the first to adopt the innovation. Their advantage is 

 in spotting a good idea, experimenting with it, and solving prob- 

 lems of adaptation to local conditions. As industry's experience 

 grows, others adopt. Increased supply reduces the price, driving 

 the first adopters, the high-skilled and high-labor-opportunity- 

 cost producers, out of the production of the new product— hence 

 an "innovation cycle." 



A major implication of the model is that skills are more impor- 

 tant the more dynamic the economy. This notion is also 

 developed by Nelson and Phelps (1966) and by Welch (1970). The 

 latter distinguishes between the "worker effect," i.e. increasing 

 production from an existing bundle of inputs, and the "allocative 

 effect" of schooling, i.e. improving the allocation of inputs and 

 outputs. Welch argues that in a technologically stagnant 

 economy, experience will lead to an optimal allocation in all firms, 

 schooling's economic contribution being limited to the worker 

 effect. 



In the innovation cycle model the allocation problem is the 

 timing of the adoption of the innovation. Welch's formulation 

 differs from the one suggested here in that in his analysis pro- 

 ducers with less schooling simply make allocation errors; in ours, 

 they know — or estimate — their lower efBciency in production. In 

 this respect the present model is in line with the traditional eco- 

 nomic approach of perfect knowledge and equilibrium, in which 

 producers act rationally, being aware of their abilities, and firms 

 are always in equilibrium, although not necessarily static. The 

 other diffusion models that have been proposed assume, explicitly 

 or implicitly, disequilibrium and lags in adoption originating from 

 ignorance of the producers of potential benefits available from 

 the new factors and products.' 



1. These distinctions are not essentia! to the model. One could derive the same 

 results on the assumption that lower-skilled producers are as efficient as others in 

 production but unaware of opportunities. The policy implications (in terms of the 

 direction of extension work, for example) are quite different. See also Nelson, Peck, 

 and Kalachek (1967) for evidence that nonadopters are aware of innovations. 



