a positive agenda for behavioral law and economics
Accepted February 18, 2011
behavioral law and economics, libertarian paternalism, identity, preferences, rationality
Law has spent surprisingly little time developing a theory of human nature. Its efforts have largely focused on the abnormal— notably, those not responsible for their actions by reason of mental illness or diminished capacity. The normal has barely been addressed. The field within legal scholarship that comes closest is law and economics. Law and economics embeds a theory—that people are rational maximizers of their self-interest. Law and economics admits its theory is unrealistic; it touts instead its theory's ability to predict. Behavioral law and economics aspires to more realism (and more predictive power). Its trajectory has, however, sometimes been contorted insofar as it has focused on exceptions to the law and economics view rather than a broader reconception of the overall endeavor. Such a reconception is desirable, necessary, and increasingly feasible.
A few generations ago, law and economics became one of the dominant fields of legal scholarship. It continues to be quite prominent today. In the last generation, a field, considered by some a challenge and by others an enhancement, to law and economics has developed, behavioral law and economics. Behavioral law and economics seeks to bring more realism to law and economics.
Law and economics holds that people rationally maximize their self-interest. Early work in behavioral law and economics stressed that people sometimes make mistakes (that is, they are not rational). (The work also stressed that people are sometimes altruistic (that is, they do not exclusively maximize their self-interest)) (Hill 2004). Some recent prominent work in behavioral law and economics by Thaler and Sunstein (2008) advocates libertarian paternalism, defined as private and public actors self-consciously attempting to move people in directions that will make their lives better by nudging them (Thaler and Sunstein 2008, p. 6). And a not insignificant amount of behavioral law and economics literature has as its aim to justify regulatory interventions on grounds that people are irrational, and make mistakes. I think a behavioral perspective has much more to offer; I also think that the focus on mistakes is overblown and pernicious.
As noted above, behavioral law and economics was supposed to bring more realism to law and economics. The worldview in which people are either making mistakes or getting it right isn't much more realistic than the one in which people are always getting it right. Moreover, advancing such a worldview as realistic is a step backwards: the admittedly unrealistic but ostensibly useful law and economics is being supplanted by the supposedly more realistic binary world in which people either make mistakes or get it right. To be sure, of course people sometimes do make mistakes. A focus on mistakes that is more methodology (how we proceed) than ontology (how the world is) is quite useful: of all the ways people act other than as the traditional paradigm predicts, the subset we can label as mistakes (and, to be more precise, cognitive mistakes rather than weakness of the will) may present a particularly good case for regulatory interventions (Schwarcz, 2010). Indeed, the label mistake is often used for what is actually weakness of the will – not a mistake at all, since a person really does want both cake and good health (and, according to some evidence, she may regret more regularly choosing health over the cake than the cake over health (Kivetz and Keinan 2006). In those cases, any regulatory intervention is more appropriately justified by externalities.
I describe here two sets of dynamics I think behavioral law and economics ought to be, and to some extent is, addressing, how people see the world (and themselves), and what people value. Unlike cognitive mistakes, these do not typically yield straightforward normative prescriptions. Rather, they may yield skepticism towards standard approaches, or approaches, such as changing norms, that seem hard to make concrete.
The first set of dynamics, how people see the world, takes into account that a person's perspective importantly reflects his identity, his culture and his worldview (including his prior beliefs). It is obvious that people's identity and culture can differ; worldviews can as well. The traditional paradigm often does not take into account the effects of a person's perspective on how he reasons and acts. The result is too-quick resort to common tools in the economists' toolkit at the expense of developing more nuanced ways to use its insights. Among the scholarship that acknowledges these issues is Akerlof and Kranton's work on identity (Akerlof and Kranton 2000). Their work has sparked a line of scholarship on identity within economics. Akerlof and Kranton deal with identity as belonging in the utility function. Identity also importantly affects analysis insofar as it influences a person's perspective, something I have called identity-as-perceptual lens. There is a rich literature in other fields, including legal scholarship outside of law and economics, on the subject. Some neuroscience work, including work by Sharot et al. (2007), provides evidence for the role of prior beliefs in how people process new information. The Cultural Cognition Project at Yale addresses how cultural values shape public risk perceptions and related policy beliefs (http://www.culturalcognition.net).
That law and economics often assumes away the importance of a person's perspective or worldview easily can be shown using the classic lemons story (Akerlof 1970). In the classic lemons story, everyone agrees on what counts as a good and bad car; there is an implicit assumption of what I have called perfect consensus. But in the real world perfect consensus is often lacking. Consider the following example. Smith incurs $100,000 of debt for college tuition; he reasons that this would be a bad idea if he'll just get fired a week into his new job. Thus, his expenditure is a signal that he is a good type. But Smith's assessment of his own capabilities, while relevant, is scarcely sufficient. Smith might be mistaken, or he might be right that he is capable or skilled, but he may not be a good fit for the job.
Another example is in the same vein: assuming away important cultural differences. Jones, a Wall Street lawyer, is negotiating an acquisition for the seller, a company in a common law country whose leading law firms negotiate detailed contracts with vigor. The buyer is a company in a civil law country whose leading law firms usually use very short largely non-customized contracts. The traditional paradigm tells us, sensibly, that negotiations will proceed until the costs exceed the benefits (Hill and King 2004). But how would we determine when this point has been reached? On Wall Street, Jones is known as a vigorous advocate for his client, evenly matched with other Wall Street lawyers, such that his continuing negotiation about details either is neutral or perhaps a positive from the other side's perspective. In the civil law country, the buyer and her lawyer may conclude that Jones is particularly aggressive, and his client must also be aggressive to have hired Jones; the buyer might therefore be less inclined to go ahead with the deal, or might hold to their positions more firmly than they otherwise would have.
A third example is the extent to which people often unthinkingly generalize from their own perspective and experience. This has been noted by some scholars; Mary Anne Case, for instance, discusses Richard Posner's failure of the imagination insofar as he more readily and intuitively grasps certain indignities experienced by men like him than those experienced by women and the lower classes (Case 2007). Indeed, Rawls' veil of ignorance is a wonderful tool, but a person may not have the imagination necessary to use it properly (Rawls 1971). A man, including one who cannot imagine himself as a rapist, may oppose the rape shield laws, more readily imagining himself falsely accused of rape and not being able to mount a complete defense than a woman whose accusation is not believed once her past sexual history is disclosed to the court. Another example: in the book The Center Cannot Hold, law professor Elyn Saks notes that one reason she was resistant to taking medication for her schizophrenia is that she imagined that others also heard voices and managed to control them without medication. It was not easy for her to be convinced to the contrary (Saks 2007). Complicating matters, it is not as though there are a few monolithic and canonical perspectives. As Amartya Sen notes, people's identities are inescapably plural (2006, p.xiii). The problem cannot be fixed by teaching an X (say, a man, Catholic, or carnivore) a Y's (say, a woman's, a Jew's, or vegetarian's) perspective.
A fourth example: I was told of a situation in which a lawyer at a major corporate law firm covertly modified a true fill-in-the-blanks form and reproduced it, making significant and concealed changes. The modification carefully included the words at the beginning and end of each page of the true form. The changes were discovered completely by accident – certainly, the recipient of the document had not thought to check. In a few subsequent cases, the lawyer checked the form, but he quickly resumed his previous practice of not checking; nobody the lawyer told about this began checking either. The conclusion is: we have prior beliefs as to (among many other things) what we think can, and cannot happen, and what it's worthwhile to investigate. Much about these priors isn't known, including their origins. And these priors are sticky, in part so that people can economize on cognitive resources. Once a prior belief has been shaken, when should it be revised and when should it be abandoned? In some cases, the correct answer is clear; in many others, it is not. How do we characterize continuing reliance on Moody's and Standard & Poor's after Enron, and after the subprime mortgage debacle?
Another set relates to what people value. Examples showing that people's preferences change when presented with irrelevant alternatives become far less puzzling once we take seriously that many (perhaps most) preferences are importantly second order, for the sort of thing something is, not (just) first order, for the tangible thing itself. People want many of the things they want because the things signal status, make them feel virtuous, or make them feel successful. A person choosing between fish and chicken chooses fish, but when beef is also offered, she chooses chicken. She wants to get something healthy but prefers chicken; when beef is available, chicken seems sufficiently healthy. I may like chocolate for its taste, but I might like Godiva chocolate best because it makes me feel as though I have a sophisticated palate. I may like the thing that yields the most memorable experience such that on different occasions when presented with choices that have the same first-order labels (chicken, fish, or even something different entirely, such as a helicopter ride) I may choose differently. In this regard, there is interesting psychological literature on how choice sets are constructed. I have previously argued that the traditional paradigm assumes away the issue, treating choice sets as given. (Hill 2004)
Apparently even the experience of having fun isn't in a simple way first order. Consider, as several scholars have done, the story in which Tom Sawyer, hired to paint a fence, gets others to do his work by successfully convincing them that painting the fence is fun (Ariely et al. 2006). That preferences are constructed is a commonplace in behavioral law and economics literature (including in seminal work by Cass Sunstein (1998)); much of the literature, though, either characterizes constructed preferences as normatively undesirable or does not directly take a stance on the issue. (That being said, there is literature on process-based models of preference construction that are very much in the spirit of my arguments here.)
Another example: the traditional paradigm assumes that enforcement of norms is costly. But we all know people for whom enforcement is a consumption good. An anecdote I have heard is of a famous elderly actress living in New York, taking walks on which she scolded people for gum-chewing and other types of unsuitable public comportment.
These are just examples; the list is not intended to be comprehensive. Some of these dynamics are fully compatible with law and economics; some are less so, or incompatible. Even those that are compatible with law and economics may not find it hospitable. They may run counter to its ethos of parsimony or its often-implicit (but not logically required) heroic assumptions about shared worldviews and perspectives. Or standard tools in the law and economics toolkit may be useless to address the issue. These dynamics often do not lead to policy solutions: behavioral law and economics' toolkit is much smaller. But they cast doubt on the law and economics ontology, and suggest avenues by which it could become more realistic and hence, hopefully, more useful.
Of course, defenders of the traditional paradigm will point out that any tractable modeling must simplify, and all I have done is give a few examples of simplification. I agree, of course, as to the need to simplify but think we can do better, getting more realism but retaining tractability. One approach is to characterize some of these dynamics as frictions in the time-honored Coase/Modigliani and Miller way (Modigliani and Miller 1958; Coase 1960). The assumption of perfect consensus is one more friction in addition to the assumption of perfect information, zero transaction costs, etc. What other frictions can we identify? What models can we develop to understand scenarios where those frictions are present? As long as we don't label the frictions as imperfections in the normative sense, we should be able to view them neutrally, rather than as indications of irrationality. As to the malleability of preferences – what preferences are the most malleable, and under what circumstances and by what mechanisms? It seems naïve to argue that preferences are never constructed – but that doesn't mean all preferences are, or that construction or malleability might not abide by principles (extensively studied in many fields) we can incorporate into our models.
Akerlof GA (1970) Quality uncertainty and the market mechanism. Q J Econ 84:488-500
Akerlof GA, Kranton RE (2000) Economics and identity. Q J Econ 115:715-753
Ariely D, Loewenstein GF, Prelec D (2006) Tom Sawyer and the construction of value. J Econ Behav Organ 60:1-10
Case MA (2007) All the world's the men's room. U Chicago Law Rev 74:1665-1669
Coase R (1960) The problem of social cost. J Law Econ 3:1-44
Hill CA (2004) Beyond mistakes: the next wave of behavioral law and economics. Queen's Law Journal 29:563-594
Hill CA, King C (2004) How do German contracts do as much with fewer words? Chi-Kent Law Rev 79:889 – 926
Kivetz R, Keinan A (2006) Repenting hyperopia: an analysis of self-control regrets, J Consum Res 33:273-282
Modigliani F, Miller M (1958). The cost of capital, corporation finance and the theory of investment. Am Econ Rev 48:261-297
Rawls J (1971) A theory of justice. Belknap Press, Cambridge, MA
Saks ER (2007) The center cannot hold: my journey through madness. Hyperion, New York, NY
Schwarcz D (2010) Insurance demand anomalies and regulation, J Consum Aff 44:557-577
Sen A (2006) Identity and violence: the illusion of destiny. Norton, New York, NY
Sharot T, Martorella EA, Delgado MR, Phelps EA (2007) How personal experience modulates the neural circuitry of memories of September 11. Proc Natl Acad Sci 104:389–394
Sunstein CR (1998) How law constructs preferences. Georgetown Law J 86:2637-2638
Thaler R, Sunstein C (2008) Nudge. Yale University Press, New Haven, CT