The Marshallian Demand Theory Revisited

Abstract

Did Alfred Marshall assume compensated or uncompensated demand curve? I argue that neither: I show that the Marshallian demand curve is the willingness-to-pay curve derived under the assumption of all prices and income held constant. This curve approximates both compensated and uncompensated demand curves only under a specific condition assumed by Marshall, namely that expenditure on the good in question represents a negligible part of a consumer’s budget. This very condition also implies quasi-constancy of marginal utility of money. I maintain that Marshall had in mind quasi-constancy in this sense instead of quasilinearity of utility as is often claimed. I argue that my interpretation provides more accurate account of the Marshallian demand theory than do alternative interpretations.