Postscript: From Data Muddles to Models

Synopsis The postscript briefly reviews useful theoretical formulations and empirical data that are available for building improved models of technological change. Elements of a stylized model are outlined, emphasizing uncertainty, mechanisms of continual technological improvement, and their influence on technology diffusion and Substitution. Uncertainty introduces stochasticity in model formulations. Technological improvement through R&D and learning by doing introduces nonconvexities due to increasing returns. A number of models with these essential features are presented. The chapter concludes with a simplified model that integrates uncertainty, R&D, and technological learning as sources of technological change. The model demonstrates the feasibility of dealing simultaneously with stochasticity and nonconvexity arising from uncertainty and increasing returns from R&D and learning by doing. The postscript concludes with the optimistic outlook that modeling approaches do exist that can improve the traditional treatment of technological change as an “externality” to the economy and society at large. Introduction Why a postscript? This book has described the evolution of technology and its relationship to global change largely without recourse to formal models. There are two reasons for this. First, models treating technological change as a process endogenous to the economy and society have been generally disappointing.