Principles and Policy_YLSNorgesBankNorgesBank_2006_3Pages from ir-2006-03-enPrinciples and Public Policy Decisions: The Case of Monetary Policy Michael Woodford1 Columbia University February 2008 Economics is a discipline that is centrally concerned with the nature and consequences of rational choice. However, the economist’s characteristic conception of rational decisionmaking is somewhat different from that of other disciplines that also weigh the rationality of decisions. For a philosopher or a jurist, rationality is above all a matter of the way in which a decision is arrived at, or, more precisely, the way in which it can be explained or defended --- it means that reasons can be given for the decision. In economics, instead, the rationality of decisions is a relation between their consequences and the decisionmaker’s goals; a rational decision is one that achieves the decisionmaker’s objectives to the greatest extent possible.2 The divorce between the economist’s conception of rationality and any process of reasoning is illustrated by Milton Friedman’s celebrated analogy, in his essay on “The Methodology of Positive Economics” (Friedman, 1953), between the kind of rationality assumed in economic models and the play of a skilled billiards player. There exists a useful theory, based on Newton’s laws of motion, that can predict the movements of the billiard balls the cue is struck in a particular way, and that can as a consequence be used to predict how one ought, in principle, to wish to play any given position; but the expert player need not understand this theory, let alone be able to explain his actions in terms of it, in order to be able to play skillfully, and indeed in ways that are successful for reasons that the theory can explain. Similarly, Friedman argued, the hypothesis of individual 1 John Bates Clark Professor of Political Economy, Columbia University; currently, Arthur Okun and Kumho Visiting Professor, Yale University. This draft prepared for a presentation in the Law, Economics and Organization Workshop, Yale Law School, March 5, 2008. 2 This is what Nozick (1993, p. 64) calls an instrumental conception of rationality. Nozick defends an instrumental conception of rationality as a philosophical view of the grounds on which particular procedures for reasoning can be said to be rational: on such a view, a decision is rational because it is arrived at using rational procedures, while those procedures can be justified as rational if they are shown to be effective in achieving the decisionmaker’s goals. But he insists that the economist’s view (“the standard account of an action’s rationality presented by decision theory”) is not an acceptable theory of rationality, since “an action might reach goals … without having been arrived at rationally… Decision theory by itself is a theory of best action, not of rational action” (p. 65).2utility maximization can be a useful theory of consumer behavior without consumers themselves understanding the theory, or being able to explain the calculations that an economic theorist would use to derive a description of optimal behavior. The assertion that they choose rationally is simply an assertion that they manage to choose the purchases that do in fact maximize utility subject to the consumer’s budget constraint, and not a claim about the mental operations through which this is achieved. The point of Friedman’s argument is that the hypothesis of rational choice can be maintained without having to make any (easily refuted!) strong assertions about the ability of actual consumers to accurately perform the kind of calculations that students encounter in a microeconomic theory class. At times, however, the economics literature even expresses skepticism about the very desirability of making decisions on the basis of explicit calculations. Friedrich Hayek (1945) bases a critique of central planning as an alternative to capitalism on the nature of the knowledge upon which private business decisions are based. “So far as scientific knowledge is concerned,” he writes, “a body of suitably chosen experts may be in the best position to command all the best knowledge available,” but “scientific or technical knowledge” is not the only kind that is relevant. Instead, much value is created by business decisions that “based on special knowledge of circumstances of the fleeting moment not known to others,” which, according to Hayek, “is knowledge of the kind which by its nature cannot enter into statistics and therefore cannot be conveyed to any central authority in statistical form” (pp. 521, 522, 524). In this account, the virtue of the capitalist system is that decisions can be made that are rational precisely because they do not have to be justified in terms of some formal calculus that could be checked by a bureaucrat. This attitude of economists differs, for example, from that of many judges or legal scholars, to whom it is natural to be interested in the justifications that can be offered for decisions, and not merely in their consequences. Indeed, many legal scholars would deny that an acceptable justification for a decision must be cast in terms of an evaluation of its consequences. For example, Ronald Dworkin’s (1977, chap. 4) theory of adjudication distinguishes between arguments of principle and arguments of policy, and argues that while legislative decisions may properly be based on either type of consideration, “judicial decisions in civil cases, even in hard cases [where no settled rule dictates a3particular decision], characteristically are and should be generated by principle not policy” (p. 84). Dworkin argues that this is true even in a case like the test for negligence propounded by Judge Learned Hand in U.S. v. Carroll Towing --- where liability depends on a finding that the defendant could have avoided the accident at a lower cost to himself than the expected loss to the plaintiff --- as such a rule is, in his view, a mechanism for balancing “competing claims of abstract right,” and not an invitation to the judge to make his own determination of “costs and benefits to the community at large” (pp. 98-100). This account of the basis for such decisions differs from that of Ronald Coase (1960), and some other proponents of the economic analysis of law, according to whom the common-law doctrine of negligence should be understood as nothing other than an