Evolutionary econometrics: From Joseph Schumpeter’s failed econometrics to George Price’s general evometrics and beyond

It is well known that Schumpeter had a very ambiguous relationship to econometrics. Although he played a important role in the establishment of the Econometric Society and tried to apply its research programme in his ambitious attempt to analyse economic evolution in his Business Cycles, he nevertheless developed what may appear as a hostile evaluation of the work of the econometricians. In this paper it is argued that this ambiguity is due to a misplaced emphasis on aggregate time series. What Schumpeter needed was statistical tools for performing the analysis of the aggregate effects of evolution in terms of the underlying population dynamics. These tools have been developed within biometrics, and they have recently become directly applicable to economic evolution due to the development of what may be called a general evometrics. Central to this evometrics is a method for partitioning evolutionary change developed by George Price. This method serves surprisingly well as a means of accounting for evolution and as a starting point for the explanation of evolution. The paper reviews some of the basic elements of this evometrics and demonstrates how they can be applied to some of the problems that Schumpeter was not able to solve. The applications cover the partitioning and analysis of relatively short-term evolutionary change within simple industries as well as the study of more complexly structured populations of firms. By extrapolating these applications of Price’s evometrics, the paper suggests that his approach may play a central role in the emerging evolutionary econometrics.