2 - Extended preferences
Published online by Cambridge University Press: 19 October 2009
Summary
The axiom of selfishness and the Two Theorems of Welfare Economics
The preferences attributed to individuals in welfare economics are usually assumed to satisfy the axiom of selfishness – that is, each individual is assumed to order consumption bundles for himself without regard to anyone else's preferences or actual consumption, and is said to be better off if he receives a more preferred bundle. There are two reasons for doubting if this is a satisfactory foundation for individualistic welfare economics. The first is empirical: it is doubtful if people are, always and everywhere, so purely selfish. The second is that it is hard to find much force in normative prescription for a world in which all agents are, by assumption, amoral. (The difficulty of a utilitarianism that accepts the axiom as descriptively accurate but goes on to recommend policy on utilitarian moral grounds is well known.) It therefore seems worth trying to relax this axiom if we can: we might gain in positive content and add moral force to normative individualistic prescription. We may indeed drop it, but we must enquire into the cost of doing so, and with what we may replace it.
Since Arrow (1951a), two outstanding contributions by Edgeworth (1881) have commonly been called the First and Second Theorems of Welfare Economics. The First Theorem is that any competitive equilibrium is a Pareto-optimum. The Second is that any optimal allocation can be supported by competitive prices if the initial endowment is appropriate.
- Type
- Chapter
- Information
- Information, Incentives and the Economics of Control , pp. 7 - 24Publisher: Cambridge University PressPrint publication year: 1992