"Behavioral law and economics"-the general topic of this Symposium-seeks to bring together "behavior" and "law and economics." Law and economics (without the modifier) is of course already about behavior. But it is typically about behavior of a particular sort: highly "rational" (in a particular sense of that term), optimizing behavior. Sometimes it is reasonable to assume that people behave in this manner; other times it is not.l The "behavioral" in "behavioral law and economics" is about infusing law and economics with insights into actual (rather than hypothesized) human behavior when such insights are needed to insure sound predictions or prescriptions about law.
Behavioral law and economics is not a critique of law and economics. It shares with that approach to the law the view that human behavior is organized by predictable patterns, which enable the analyst to generate models (often formal ones) and testable hypotheses about the effects of legal rules. And it shares the view that such analysis is an important and valuable pursuit, one most worthy of legal scholars' attention. Where it disagrees with conventional law and economics is about the shape of the predictable patterns of human behavior. Its goal is to offer better predictions and prescriptions about law based on improved accounts of how people actually behave.
This Essay offers a behavioral economic analysis of redistributive legal rules. Redistributive legal rules are rules chosen for their effects in shifting wealth from high-income to low-income individuals (progressive redistribution). The desirability of such rules has been the subject of intense debate within the legal community. Many law and economics scholars have urged that legal rules be chosen solely with an eye towards Kaldor-Hicks efficiency (which I will call simply "efficiency" for the remainder of this Essay); these scholars often urge that distributional considerations be addressed (if they are to be addressed at all) exclusively through the tax and welfare systems.2 On this view, distributive goals do not provide a basis for choosing an inefficient legal rule-although they might, it seems, provide a basis for choosing between two efficient rules.3 Other legal scholars have argued that the selection of legal rules should be informed by distributional considerations even at the expense of efficiency.4 I will call a rule "redistributive" if it makes such a trade-off between distributive objectives and efficiency.
A recurring theme in the debate over redistributive legal rules has been the relative cost of redistributing wealth through legal rules (defined to mean rules other than those that directly relate to the tax and welfare systems) and redistributing wealth through the tax and welfare systems (which I will call simply "the tax system" or "taxes" for the remainder of this Essay). Under the assumptions of neoclassical economics, any desired level of redistribution can be achieved at lower cost through the tax system than through legal rules.5 This is not because the tax system can redistribute wealth costlessly; the animating feature of both lawyers' and economists' analyses of tax schemes is their potential to distort people's work incentives. Higher taxes on the wealthy will tend to discourage people from earning high incomes. But from the perspective of neoclassical economics, precisely the same is true of redistributive legal rules: "[U]sing legal rules to redistribute income distorts work incentives fully as much as the income tax system-because the distortion is caused by the redistribution itself...'6 Thus, for example, a thirty percent marginal tax rate, together with an inefficient legal rule that redistributes an average of one percent of high earners' income to the poor, creates the same distortion in work incentives as a thirty-one percent marginal tax rate coupled with an efficient, non-redistributive legal rule.7 However, the former regime also entails costs due to the inefficient legal rule. …