Industrial Structure and the Nature of Innovative Activity

In this paper we attempt to provide an analytical framework relating market structure to the nature of inventive activity. In doing so we have come to modify this neo-Schumpeterian view in a fundamental way. We shall argue that except in the short run both market structure and the nature of inventive activity are endogenous; that the degree of concentration in an industry ought not to be treated as given, as it recently has been in the industrial organisation literature; that they both depend on more basic ingredients, such as the technology of research, demand conditions, the nature of the capital market (i.e. market rates of interest, and the ability of firms to borrow to finance research and development (R and D), and the legal structure (e.g. patent rights). We shall, to be sure, explore the relationship between the degree of concentration and the nature of innovative activity. But as they are both endogenous, their relationship, unlike the neo-Schumpeterian thesis, ought not to be regarded as a causal one. A major objective of our study is to formulate models within which the efficiency of a market economy can be assessed and where the tradeoff between a temporal production efficiency and dynamic gains can be meaningfully discussed. In this paper we are concerned with process innovation (i.e. R and D designed to reduce cost of production). This means that attention is drawn to the distribution of extreme values. This in turn implies, as was noted by Evenson and Kieslev (1975), that even risk-averse firms might wish to engage in randomisation.