Utility and Demand Analysis

Abstract

It is now well over a century since utility functions were first introduced as tools of economic analysis. Since, then, they have achieved wide currency and play an essential role in the neo-classical economics in which most present-day economists were, and are, brought up. However, neither this long pedigree, nor the central place of utility analysis in the dominant orthodoxy of economics, can alone justify applied work which uses the theory as a basis of empirical observation. The long association of utility analysis with economics has left strong associations between the study of consumer preferences and particular propositions of economic theory, especially welfare propositions. Many of these associations are neither necessary nor useful in empirical work and yet it is not always made clear exactly what the empirical investigator is or is not assuming in constructing demand functions from mathematical preference relations. Conversely, much empirical work has made no use of consumer theory, believing utility analysis to be useless for the purpose. This view has received support both from those commentators who believe the theory to be completely vacuous, for example, Clarkson (1963), Mishan (1961) and Robinson (1960), and from those who believe it to be impossibly restrictive, for example, Kornai (1971).