اقتصاد رفتاری اصولی

This paper addresses the question of why, in spite of its recent success, behavioral economics does not influence most discussions about how economic policy ought to be made. Failing to penetrate into contemporary discourse on leading policy issues is a serious problem, because behavioral techniques often point to policy prescriptions that are at odds with the prescriptions which follow from models using more standard behavioral assumptions. By comparing how the universe of possible policy implications changes when different methodological approaches are used, this paper demonstrates a systematic link between methodology and the range of policy prescriptions that can be socially desirable. Because of this link, the methodological multiplicity of behavioral economics, and the ideological pluralism which it supports, favor the use of normative behavioral economics. This follows from the basic economic principle of diversification: a policy prescription that reflects averaging over a number of distinct kinds of errors (one for each methodology) is less likely to wander far off target than one generated by a single method.

Many economists may have doubts about the ultimate value of research in the subfield of economics known as “behavioral economics,” but few would disagree that a “behavioral revolution” has taken place over the last 15 years. Although this “revolution” does not carry all the markings of a full-fledged shift of paradigm in the sense of Kuhn (1970), there is abundant evidence that the normal science currently practiced by neoclassical economists incorporates behavioral techniques that would have stood out as unusual just a decade ago.1 Most convincing on this point is the fact that a number of major themes in behavioral economics, e.g. examining the empirical validity of behavioral assumptions, modeling non-maximizing behavior by firms and consumers, and borrowing from psychology, sociology and cognitive science, now comfortably fit into most major journals in economics.2 This newly won attention, although it reflects a substantive achievement for those who advocate the use of behavioral techniques in economic research, has yet to see the insights of behavioral economics coalesce into a normative framework for analyzing economic policy (Knetsch, 1995).3 Reaching the imagination of policy makers and influencing popular debate about leading issues in public policy is, admittedly, not the only measure of an intellectual movement’s success. Still, it is puzzling, at least on the face of it, why the recent ascent of leading behavioralists into the limelight has not been accompanied by a new normative framework for analyzing policy.4 In fact, some of the most well-known behavioral economists explicitly warn their readers not to draw normative inferences from their work. Thaler (1991, p. 138), for instance, makes the following assertion about his own work on the empirical validity of rational choice axioms: “A demonstration that human choices often violate the axioms of rationality does not necessarily imply any criticism of the axioms of rational choice as a normative idea. Rather, the research is simply intended to show that for descriptive purposes, alternative models are sometimes necessary.” Continuing this discussion of what behavioral economics implies about the use of rationality axioms in normative analysis, Thaler (1991, p. 138) argues that the major contribution of behavioral economics has been the discovery of a collection of “illusions,” completely analogous to optical illusions. Thaler interprets these “illusions” as unambiguously incorrect departures from the “rational” or correct way of making decisions. Thaler is explicit in accepting neoclassical axioms of individual preferences (e.g. transitivity, completeness, non-satiation, monotonicity, and the Savage axioms which guarantee that preferences over risky payoffs can be represented by an expected utility function) as his normative ideal when he writes: “It goes without saying that the existence of an optical illusion that causes us to see one of two equal lines as longer than the other should not reduce the value we place on accurate measurement. On the contrary, illusions demonstrate the need for rulers!” Yet, in showing that human decisions contradict the predictions of expected utility theory, there is no analog to the straight lines of objectively equal length. Unlike the simple geometric verification of equal lengths against which incorrect perceptions may be verified, the fact that human decisions do not satisfy the axioms underlying expected utility theory in no way implies an illusion or a mistake. Expected utility theory is, after all, but one model of how to rank risky alternatives. Those who insist that traditional theory is a wholly complete basis for normative analysis in spite of systematic departures from this theory in the empirical record, in effect, assert that behavioral economics is a purely descriptive field of inquiry. This perspective, in turn, implies that the full range of possible policy conclusions to be drawn from the empirical record against neoclassical behavioral assumptions consists of nothing more than a call for better training in the logic of existing models of choice.5 Thaler’s own work illustrates the possible inadequacy of this non-normative interpretation of behavioral economics by generating results that seem to have interesting policy implications at odds with those of standard models. When, for example, Kahneman et al. (1986) provide evidence that decision makers do not maximize, it is highly unlikely that optimal policies developed from maximization models continue to be optimal. For instance, the incentive-compatibility constraints faced by a public official pondering how to raise revenue to cover the cost of providing a public good will be less constraining if consumers do not maximize, since those constraints assume that citizens extract personal benefits at the public’s expense to a greater degree than actually occurs. This means that the policy maker has more latitude than previously thought. In another paper, Thaler (1980) argues that consumers are relatively unresponsive to exogenous changes. This descriptive result would seem to argue against policy interventions which depend on a high degree of sensitivity to individual incentives, and in favor of policies whose primary drawback stems from the consequences of a public that is forward-looking and keen to adjust to changes in incentives. Surely, Thaler’s result is germane to debates about whether small revisions in marginal income tax rates can be expected to boost a country’s labor supply (which underlies the idea of “tax cuts that pay for themselves”). Thaler’s results on unresponsiveness also appear to challenge another important normative concept, the Lucas critique, which played a significant role in justifying the economic policies of Margaret Thatcher and Ronald Reagan. If forward looking behavior plays only a minor role in influencing current decisions, then the idea of using monetary and fiscal stimulus to increase aggregate demand is not as vulnerable to criticism based on rational expectations and Ricardian equivalence. In yet another example, the work-horse of applied normative economic research, cost–benefit analysis, seems to have its meaning altered by Kahneman et al.’s (1990) fascinating demonstration that people require very high payoffs to be induced away from the status quo. If, by usual measures, a proposed policy change leads to a net benefit, this alone may not justify the new policy, owing to the unaccounted psychic costs of adjusting away from the status quo. Economists doing cost–benefit analysis (taking preferences as exogenously given) would presumably want to respect the finding that average preferences are reference point-dependent and, therefore, recommend fewer changes in policy. In addition to noting what appears to be an abundant supply of unexplored normative implications in the descriptive behavioral economics literature, other rationales for asserting that normative behavioral economics is potentially worthwhile seem to be emerging. One such rationale stems from a relatively new strand in the behavioral economics literature focusing on decision environments in which anomalous behavior leads to surprising social benefits. That is, systematic “mistakes” can have pro-social consequences in certain contexts. Dekel (1992) provides a model in which individuals who have inflated beliefs about the relative value (in terms of increasing one’s chance at surviving) of the food that they own are objectively better off in food-bargaining situations than individuals who correctly understand how nutrition maps into survival probabilities. In Dekel’s model, the deluded individual benefits from an improved bargaining position that rests on the individual’s reputation for being an irrationally tough bargainer. A similar story in Kyle and Wang (1997) features financial market participants who sometimes benefit from their own distorted beliefs. Berg and Lein (2002) illustrate how overconfidence improves liquidity in certain settings, actually leading to a Pareto improvement (relative to perfect rationality) by reducing transactions costs enough to offset the personal cost of holding distorted beliefs. And Bernardo and Welch (2000) construct a cascade environment where distorted beliefs lead individuals to strike out on their own rather than “follow the herd,” providing a socially beneficial “informational externality.” This emerging line of work stresses the connection between rationality (or adaptiveness) and specific environments, drawing on the tradition of studies such as Winter (1988) and Nelson and Winter (1982) in which neoclassical behavior does not always win against behavioral alternatives. Seen in this light, behavior that does not satisfy the neoclassical axioms, including beliefs about the world which are incorrect, cannot easily be interpreted as aberrant, or labeled “irrational,” since such behavior is (in certain decision environments) net welfare-improving relative to neoclassical “rationality.” The message that context matters, a commonplace in most branches of the social sciences, seems to be controversial in economics, which has traditionally viewed the axiomatic approach and its concomitant priority on generalizability as an important indicator of its substance (and superiority). Paying attention to context, however, and acknowledging that different types of behavior work well in different settings leads one to conclude that the meaning of “rationality” derives from specific environments and is not always amenable to universal or axiomatic construction. This kind of context dependence complicates the policy maker’s task by requiring more detailed knowledge of environmental and behavioral particulars so that concepts such as “rationality” and “efficiency” are well-defined. Thus, the success of descriptive behavioral economics alongside an apparent consensus against normative behavioral economics presents a conundrum. Here we have a substantial category of research emerging into prominence, the key findings of which are positioned to be orthogonal, i.e. unrelated, to nearly every interesting policy debate on which the question of economic behavior may bear. In this anti-normative appraisal, behavioral economics, although it accepts a broader vision of how firms and consumers actually behave, does not actually challenge homo economicus as the proper ideal for assessing how legal and cultural institutions should be designed. This paper addresses the question of whether the anti-normative view is correct.