FAIRNESS VERSUS WELFARE**
By Louis Kaplow & Steven Shavell. Harvard Univ. Press, 2002.
[I]f there is to be a normative economics, from which prescriptive propositions and policies can be derived that are applicable to a particular society, then it cannot be raised on any presuppositions that do not accord with whatever ethical consensus remains in that society. One implication of this statement is that wherever propositions raised on a utilitarian calculus clash with ethical norms, such propositions cannot be sustained by economists who aspire to fashion a body of prescriptive doctrines acceptable to society.
-E.J. Mishan, Introduction to Normative Economics (1981).1
The purely economic man is indeed close to being a social moron.
-Amartya K. Sen, Rational Fools, in Choice, Welfare and Measurement (1982).2
Fairness Versus Welfare3 by Louis Kaplow and Steven Shavell is an impressive and stimulating intellectual achievement. Originally a mammoth article filling more than four hundred pages of the Harvard Law Review and now published in book form by the Harvard University Press, the IMAGE FORMULA7
work presents a bold formulation of the increasingly influential law and economics school of thought. It does so, moreover, in lucid prose that seeks to come to grips with proponents of rival views about legal policy analysis, and communicate with them effectively. There is not a graph to be found in those hundreds of pages, and only the most sparing use of mathematical notation. The article was widely publicized by its authors through their appearances at colloquia and symposia and was cited repeatedly before publication.4 And yet, I shall contend, the prodigies of effort and ingenuity that have gone into this work cannot disguise the failure of the entire project.
Kaplow and Shavell mount a provocative challenge to centuries of philosophical reflection about the law. They insist that the assessment of legal policies "should be based exclusively on their effects on the welfare of individuals" and that they "should not depend on notions of fairness, justice, or cognate concepts."5 "Fairness," as Kaplow and Shavell use the term, is a generic term that includes all notions of justice, rights, and related concepts.6 Their claim, then, is that an analyst who examines a proposed or actual legal rule, policy, or regime should ask only to what extent it promotes the welfare or well-being of the individuals in the affected society. The analyst might believe the policy in question to be drastically unfair, brutally unjust, or a massive violation of human rights-it might in fact be all of these things-but none of that is relevant except insofar as the affected individuals happen to care about fairness and feel better off if its strictures are met7 insofar, as Kaplow and Shavell put it, as they have a "taste for fairness."8 This is what Kaplow and Shavell mean by denying that notions of fairness are "independent" evaluative principles-their "signifiIMAGE FORMULA9
cance ... depends directly on the strength of individuals' actual tastes ... and is thus an entirely empirical issue."9
I will argue that the methodology Kaplow and Shavell advocate would call-in real-world situations, not merely philosophers' hypotheticals-for the endorsement of morally repugnant policies. And the arguments they use to support their methodology do not withstand critical scrutiny. The principal argument-to which they devote over two hundred pages-is logically circular, thus carrying no weight at all. The balance of the article does almost nothing to justify the crucial role they assign to individual welfare or well-being. Indeed, the conception of well-being central to their enterprise-well-being as the satisfaction of individual preferences-faces enormous difficulties that Kaplow and Shavell either address quite unsuccessfully or ignore completely. …