Abstract. This paper seeks to explain the success of public choice at Virginia Tech in the 1970s in two ways. First, I reflect on my graduate school experience at Virginia Tech, with the intent of identifying particular characteristics of the people and location of Virginia Tech, including its remoteness, that facilitated the development of the paradigm shift that public choice, in the late 1960s and early 1970s, represented. Second, I argue that the success of public choice at Virginia Tech depended critically upon the willingness of the people here to disconnect from professional constraints and to think outside the conventional economics box. I make this point with the aid of a computer simulation of the evolutionary development of “people” (or “trials”) as they attempt to scale the highest “mountain” within a “mountain range,” the exact features of which (including the heights of the various “mountains”) are not known before the climb is started. In such an environment, finding the highest peak, the assumed goal, is critically dependent on a measure of “deviance,” or the willingness to shun short‐run opportunities of exploiting accepted methods.

Although many models of technological substitution processes exist, only a few deal with multilevel technological substitution and still fewer may be easily applied to forecasting of real multilevel substitution processes. We propose a stochastic model of multilevel technological substitution and a forecasting method based on this model. In the author's opinion, the developed algorithm may be easily applied to forecasting various technological substitution processes. An example of the application of this algorithm to forecasting of world energy consumption is presented. The algorithm makes possible an investigation of the impact of future innovation on the system's behavior.