Navigation – Plan du site
The Behavioural Turn in Law and Economics

Preferences All the Way Down: Questioning the Neoclassical Foundations of Behavioral Economics and Libertarian Paternalism

Les préférences jusqu’au bout: questionner les fondements néoclassiques de l’économie comportementale et le paternalisme libéral
Mark D. White
p. 353-373

Résumés

L’économie comportementale a enrichi notre compréhension des limites et des imperfections de la décision humaine qui avaient été négligées dans les modèles de choix simplistes de l’économie néoclassique. Malgré les apports substantiels de l’économie comportementale à la théorie du choix, cette approche repose sur les mêmes fondements que l’économie néoclassique, soit la satisfaction des préférences sous contrainte et ne peut donc pas prendre en compte d’autres aspects de la décision humaine tels que les principes, les idéaux, le jugement et la volonté. La conception du choix propre au paradigme des préférences limite à la fois le pouvoir explicatif et prédictif des modèles et contribue ainsi à la compréhension réductrice des intérêts qui sous-tend les nouvelles formes de politiques paternalistes, comme les nudges. Afin de pleinement réaliser leur objectif d’expliquer les processus réels de décision humaine, l’économie comportementale et l’économie néoclassiques doivent toutes deux abandonner le paradigme de la satisfaction des préférences au cœur de leurs modèles et embrasser une conception plus large et plus inclusive des intérêts et du choix.

Haut de page

Texte intégral

1Neoclassical economics is both heralded and criticized for its model of rational choice symbolized by homo economicus. Supporters hail its parsimony, minimal assumptions, and empirical verifiability (or falsifiability), while critics regard it as overly simplistic, sacrificing descriptive realism and understanding for illusory scientism. Among such critics are behavioral economists, who use the early insights of Herbert Simon and the later work of Daniel Kahneman and Amos Tversky to construct more elaborate and inclusive models of economic choice. Behavioral economics has been regarded as a revolution in economic theory and as well as policy, especially in terms of the hybrid field of behavioral law and economics, the most well-known innovation of which is “nudges,” slight changes in choice environments meant to leverage shortcomings in human rationality to steer choices in what proponents regard as a better direction.

2The insights of behavioral economists are tremendously innovative and valuable, and have improved our conceptions of human rationality by demonstrating how insufficient the standard models of choice are to explain it. In this paper, however, I will argue that behavioral economics represents only a marginal improvement over the neoclassical model of choice because it retains that model’s flawed core of constrained preference-satisfaction. I will detail how an expanded conception of choice beyond preferences to interests would allow both mainstream and behavioral economists to represent a much wider range of motivations and behavior more accurately and completely. Finally, I will show how the limited conception of choice maintained by many behavioral economists leads to some of the problems identified with nudge policy, particularly those stemming from a limited appreciation of the nature and role of interests.

1. Behavioral Economics’ Improvements on Neoclassical Models of Choice

  • 1 While sufficient for our current purposes, this description is of course simplified; for more conte (...)

3Neoclassical economic models of choice are typically composed of three parts: preferences, constraints, and beliefs (based on information). Normally, preferences, constraints, and information are assumed to be given (or exogenous to the model): preferences are held to several formal conditions such as transitivity, constraints reflect simple resource scarcity, and information is rationally processed by the agent to form beliefs. All this in place, agents are then assumed to choose the available option, within their resource constraints, that they believe will satisfy their highest preferences.1 This model became represented by the metaphorical homo economicus, and its simplicity and generality helped it expand into fields outside economics, such as law, political science, and sociology, a phenomenon known as “economics imperialism.”

4However, developments in decision theory and experimental psychology beginning in the 1950s cast doubt on the empirical relevance of this model. Similar to Thorstein Veblen (1898, 389), who called the agents of economic models “lightning calculator of pleasures and pains,” Herbert Simon (1955; 1956) was skeptical of the cognitive abilities that the model demanded of agents: they are assumed to be able instantly to consider myriad options and the marginal trade-offs between each pair of them in terms of cost and benefit (or “utility”) to arrive at the optimal solution. Doubting whether people could or should reason like this, especially in low-stakes decisions where the cost of a suboptimal decision is low compared to the cost of making the decision itself, Simon suggested that agents do not optimize but rather satisfice, choosing an option that is “good enough” rather than optimal in an ideal sense. When purchasing a house or a car, agents are justified in devoting significant time and effort to making the best decision, but when choosing a brand of breakfast cereal, such time and effort is wasted, and “good enough” may in fact be best. More generally, Simon’s alternative and more realistic view of human decision-making is known as bounded rationality, referring to natural limitations on human cognitive capacity.

  • 2 See Kahneman, Tversky, and Slovic (1982) for the seminal work; for more recent exposition and summa (...)

5Roughly twenty years later, Daniel Kahneman, Amos Tversky, and their colleagues in experimental psychology discovered more specific ways in which people did not reason the way homo economicus did, and in doing so set the stage for what is commonly known today as behavioral economics.2 Over the years, they detailed a number of cognitive biases and dysfunctions, as well as decision-making heuristics (a more specific example of Simon’s bounded rationality), that further distance actual human choice from neoclassical models. Examples involving information processing and belief formation include availability bias, by which people give excessive weight to easily available information; confirmation bias, by which people give excessive weight to information that confirms their existing beliefs; and optimism bias, by which people underestimate the statistical likelihood of personally negative outcomes. Examples involving preferences (including the perceived value of options) include the endowment effect, by which people value goods more when they possess them than when they do not; loss aversion, by which people place more weight on a loss than an equivalent gain (which can be considered another expression of the endowment effect); and various framing effects, by which the presentation of options, including wording and order, has outsized effects on choice.

  • 3 In an interview (and elsewhere), Schiller called behavioral economics: “a revolution in economics t (...)

6While often explained as providing evolutionary advantage in the early days of the human race and our predecessors, these “quirks” on human decision-making introduce serious complications into the simplistic economic model of rational choice, which assumes that the relative value of options (embedded in preferences) is context-free and information is integrated rationally into belief systems. Given the significant questions it raises about the predictions of the rational choice model, behavioral economics has been called a “revolution” in thinking about decision-making by Nobel laureates (such as Robert Schiller) as well as prominent columnists and financial writers (such as David Brooks and Richard Peterson).3 In terms of the economic approach to law, one of many areas of economics in which the behavioral approach has been influential, Russell Korobkin (2011, 1655) declared that “the battle… has been won” by behavioral economics. Indeed, both Herbert Simon and Daniel Kahneman were awarded the Nobel Prize in economics (at separate times) for their contributions to a more realistic understanding of choice. (Tversky had passed by the time Kahneman was given the award.)

  • 4 But not to all: Richard Posner, long a leading exponent of neoclassical law and economics, argued t (...)

7However, the impression of behavioral economics as a revolution is dependent on the vantage point from which one forms it. From the position of mainstream economics and its model of homo economicus, the criticisms from the psychological research represent perhaps the first serious challenge to the established paradigm of economic choice, and it is understandable that, from this point of view, behavioral economics would seem revolutionary.4 But from the outside of mainstream economic thinking, the behavioral revisions to homo economicus may seem less significant.

8As Berg and Gigerenzer (2010, 6) write, despite the common focus on the differences between the two approaches, there is “greater similarity between behavioral and neoclassical economics’ methodological foundations than has been reported by others.” In particular, behavioral economics, for the most part, retains the basic constrained preference-satisfaction framework of the neoclassical economic model of choice and makes slight modifications to it to incorporate the various biases, dysfunctions, and heuristics researchers have identified. “The key point,” according to Berg and Gigerenzer, “is that the resulting behavioral model hangs onto the central assumption in neoclassical economics concerning behavioral process—namely, that all observed actions are the result of a process of constrained optimization” (2010, 13). Hayden and Ellis (2007, 632) agree when they write that “a primary reason for the failure of behavioral economics to confront important questions is its continued unreflective reliance on the basic economic paradigm.”

9Behavioral economics’ reliance on the core of the neoclassical model of choice is not only descriptive but normative as well. Focusing on Kahneman’s work in particular, Buturovic and Tasic write:

Keeping in mind that Kahneman’s work is widely regarded as the greatest challenge to standard neoclassical economics today, it is paradoxical that his work is so deeply rooted in the same neoclassical version of rationality that he outwardly opposes. Decision making for Kahneman is always consequentialist, with the clear goal of maximizing a utility function. Maximization in turn relies on a form of “rationality” that requires knowledge of all conceivably relevant facts, knowledge of statistical methods, and thinking in an abstract, uncontextual, and overly literal way. (Buturovic and Tasic, 2015, 142, emphasis mine)

  • 5 As opposed to textbook perfect rationality, the ecological rationality of Gigerenzer and Smith (sep (...)

In other words, the traditional model of rational choice may not be descriptively accurate, given human cognitive limitations and imperfections, but behavioral economists still consider it the theoretical or normative ideal, in the same spirit as perfect competition in industrial organization. As Whitman and Rizzo (2015, 410) write: “Behavioral economics challenges the positive validity of those [rationality] axioms in describing human behavior. Nevertheless, behavioral paternalism maintains those axioms as normative standards to which agents ought to conform.” Gigerenzer (2015, 365) concurs as he notes that behavioral economists “rather uncritically accept the rules of axiomatic decision theory as the norm for all rational behavior, and blame mortals for not living up to this ideal.”5

10In the following sections, I argue that the neoclassical model of rational choice is flawed, neither describing nor prescribing human decision-making, and should not form the basis of any mainstream or behavioral economics that aspires to accuracy and realism. More specifically, I argue that the underlying framework of constrained preference-satisfaction framework is too narrow and constraining to describe actual human choice and, for all its insights, behavioral economics does not address this essential shortcoming. Behavioral economists attempt to explain choice “anomalies” using modifications to the model of constrained preference-satisfaction when the problem lies with the dependence on preferences themselves, a heuristic fiction used by researchers to derive predictions rather than explain or describe actual choice processes. In the end, behavioral economists are adding descriptive realism to a model that was never meant to be descriptive in the first place.

2. Choice Beyond Preferences

11While behavioral economics modifies the framework of constrained preference-satisfaction at the margins, elaborating on how agents evaluate preferences and process information, I argue instead that this basic model needs to be expanded at its roots. Regardless of how we conceptualize agents deliberating over their preferences and beliefs, the very structure of preferences is too narrow to explain the many influences on choice.

  • 6 See, for instance, Hausman (2012), especially chap. 1.
  • 7 Karlsson, Loewenstein, and McCafferty (2004) even argue that various aspects of meaning can be inco (...)
  • 8 See, for instance, Fehr and Schmidt (2006).

12This is not to say that preferences cannot be broadened themselves, and they have been. While traditionally, such models begin with an assumption of self-interest underlying preferences, this is not necessary. As typically understood, preferences are simply rankings with no specific psychological basis behind them. Regardless of the nature of the underlying evaluations, preference orderings are valid as long as they fit the formal conditions such as transitivity that allow them to lead to utility functions with the desired mathematical properties.6 Therefore, preferences can involve a wide range of motivations other than narrow self-interest. An agent can make choices based on her own self-interest or the well-being of family and friends, various communities to which she belongs, or the global population. In other words, the entire span of generally utilitarian orientations can be represented by preference orderings without any supplementation at all.7 For their parts, behavioral and experimental economists have contributed greatly to our appreciation of the scale and scope of non-self-interested behavior, such as that shown in game-theoretic situations where people do not act as self-interested versions of homo economicus would.8

  • 9 Sen (1977); for further work incorporating principles explicitly within economics, see for instance (...)
  • 10 Commitments such as this are often modeled as lexicographic preferences, preferences that must be s (...)

13But preferences cannot cover every motivation, in particular those that do not exhibit trade-offs at the margin. Amartya Sen discussed this possibility in the context of commitment, and it can also be stated in the deontological language of principles, motivations (often ethical in nature) that do not respond as quickly as do preferences to marginal changes in cost.9 Agents who value honesty, loyalty, or integrity are not likely to change behavior that reflects these principles in response to small changes in their opportunity cost. This is not to say that principled behavior is completely invariant with respect to cost, only that it takes a significant change in cost to do so. For example, an agent who pledges $100 each month to a local animal shelter is unlikely to change this if her cable bill goes up by $10, but may do so if she loses her job and needs that money to cover her rent. If the pledge were based on a mere preference for helping animals, she may lower her pledge by $10 or less in response to the increase in her cable bill, but because it is based in a promise (or commitment), she more likely would not.10

14It is not only personal principles that lead people to disregard their preferences (even those not based on self-interest), but also more general societal ideals, such as justice and equality (as well as their antisocial opposites). We see people every day devote a significant part of their well-being and sometimes their entire lives to promoting a cause. We applaud and admire them when we find their cause to be just—and are horrified when we do not—but in both cases we recognize that these people have elevated a broader ideal above mere preferences in their decision-making, and again, not in a way that allows for trade-offs at the margin.

  • 11 For more on this conception of interests and choice, see White (2014, ch. 3; 2016).

15We can call the combination of all the things that matter to a person and influence her choices as that person’s interests.11 Since interests include principles and ideals which are not often responsive to small changes in costs, interests cannot be represented by preference orderings alone; to do so renders the resulting preference ranking a filtered distortion of interests as a whole, obscuring and misrepresenting the true relationships between the different components of a person’s motivations. The behavior of the person who donates money to the animal shelter and does not change her gift in response to small changes in her financial situation could be interpreted and modeled in terms of preferences but this would be a misrepresentation. The absence of marginal trade-offs would seem puzzling unless it were acknowledged that her gift is not an expression of a preference but rather a principle—and only with such an acknowledgment can her future behavior based on this principle be anticipated and predicted (as much as is possible).

  • 12 For more on judgment in economic models, see White (2015).

16The representation of interests in any fixed way is further complicated by the fluidity of the combination of their constituent parts. Not only are preferences and principles formally distinct, they are balanced by agents in different ways in different circumstances. A principle may take precedence over preference in some decision-making contexts but not others. In some of these cases there may be a rule or determinative logic behind the priority given principle or preference, but in many cases it comes down to judgment, an intuitive faculty that by its nature cannot be reduced to rules. As Immanuel Kant wrote, “though understanding is capable of being instructed… judgment is a peculiar talent which can be practiced only, and cannot be taught. It is the specific quality of so-called mother-wit; and its lack no school can make good” (1781/1787, A133, B172). Judgment is necessary to rank different elements of interests in specific circumstances as well as to make decisions in cases in which those elements are evenly matched (or truly incommensurate).12

  • 13 For a model incorporating willpower into an economic model of choice, see White (2006; 2010).

17Finally, the choice a person makes to further her interests may not be reflected in the decision she makes because of the problem of weakness of will. For example, she may know she should exercise in the morning, and may sincerely intend to the night before, but when she wakes up she may choose not to even though she still knows this to be in her interests. This behavior can be characterized with dual subsets of interests where one reflects narrow, short-term interests (such as sleeping in) and the other broader, long-term interests (such as increased health), with the final choice of which “path” she will follow being an act of will. Even though the agent knows at all times that it would be better to act in her long-term interests, the short-term interests have a unique visceral pull on her that requires willpower to resist. If willpower is understood as a “muscle” with limited resources, the agent will be stronger at some times and weaker at others, but ultimately whether she will stick to her exercise regimen will seem to the external observer to be a random matter that cannot be reduced to an overall optimization exercise, but in a actuality will be an exercise of her willpower, with even the strongest of us succumbing to temptation from time to time.13

3. Problems with Economists’ Overreliance on Preferences

18Rather than a simple matter of satisfying preferences within constraints, people act to promote multifaceted interests balanced in complex ways processed through their judgment and implemented by will. Against this broader understanding of choice, behavioral economics becomes problematic due to its adherence to the core of the neoclassical model of choice, which is insufficient to explain the wide range of human choice despite behavioral improvements to it. Rather than question the centrality of preferences, behavioral economics recommends that economists reconsider how preferences are informed and processed by agents to arrive at predicted choices different from those resulting from the neoclassical model, and are therefore unable to consider the full range of motivations in interests more broadly defined.

19To illustrate this argument, I will use the description of behavioral economics given in Jolls, Sunstein, and Thaler (1998), who organize the thrust of behavioral economics (as applied to the economic approach to the law, their focus in the article) into three areas in which the ideal mental capacities assumed by the neoclassical model are impaired: rationality, self-interest, and willpower.

20Bounded rationality is the primary focus of behavioral economics, as described above, covering basic cognitive limitations that lead to heuristic strategies such as satisficing, as well as the various biases that detail how we process our preferences and process information, that lead to choices that differ from the predictions of the homo economicus model. But the modified deliberative process that results from these critiques still focuses on preferences and ignores the broader set of motivations that comprise interests (much less judgment and will). For example, satisficing behavior may serve as a rational heuristic to optimize time and effort, but it may be also be the result of a judgment call made in light of competing principles: the agent feels she needs to come to a decision quickly because she has a prior commitment to meet, for example. This alternative understanding does not change the nature or value of the satisficing model, but it places it in a broader context that informs its use and motivation. In this example, the agent does not satisfice because she cannot make a perfect decision, or because she does not find it worthwhile in a global context, but because a principle constrained the amount of time and effort she can devote to that particular choice.

  • 14 See, for instance, Margolis (1984) and Etzioni (1988).

21Bounded self-interest acknowledges that human beings are not narrowly focused on their own self-interest, and as noted above, behavioral and experimental economists have done invaluable work falsifying predictions of the standard model based on self-interest. Nonetheless, this expanded ethical range of human motivations is represented solely in terms of preferences, with the resulting amenability to trade-offs, rather than the logically distinct principles and ideals, which resist trade-offs at the margin. For example, Matthew Rabin (1993) explains the reciprocal nature of nonselfish behavior often seen in laboratory experiments. His explanation takes the form of a “fairness function” that makes an agent’s preferences dependent on another person’s behavior. In a sense, the agent has a preference for fairness that is triggered by the fairness of others and which sits amongst preferences based on self-interest—and can be traded off at the margin in response to changes in opportunity cost. As he writes, “the bigger the material payoffs, the less the players’ behavior reflects their concern for fairness” (1993, 1287). This modeling technique, wedded to the concepts of preferences and utility, rules out the possibility that agents can behave fairly because they believe it is right and even when the costs of behaving fairly rise (to a point). In the end, Rabin’s model is consistent with more mainstream economists’ attempts to model altruistic behavior as preferences or different kinds of utility that are commensurate with self-interest and therefore susceptible to marginal compromise, which also deny the principled behavior discussed above which is not subject to trade-offs at the margin.14

  • 15 See, for instance, Ainslie (2001), Akerlof (1991), O’Donoghue and Rabin (1999a; 1999b; 2001), and F (...)

22Finally, bounded willpower is also modeled by behavioral economists using the framework of preferences. For example, the well-developed behavioral economics literature on procrastination provides us with many clever ways of understanding why people delay or put off arduous long-terms tasks for short-term pleasures, including time-inconsistent preferences, present-biased preferences, or preferences influenced by the greater salience of short-term payoffs.15 All of these modifications to standard preferences lead the agent to choose short-term pleasures even when it compromises their utility based on their “unadulterated” preference ordering. But none of these modifications to the neoclassical model speak to willpower itself; instead, they propose ad hoc alterations to preference designed specifically to generate irrational delay.

  • 16 For more on the role of will in economic choice, see White (2010; 2011, ch. 2).

23The problem is that neither the neoclassical nor behavioral models of choice incorporate any true sense of will that differentiates between making a choice and acting on it. Once the agent makes a decision that maximally satisfies her preferences—however those preferences are conceptualized—it is taken for granted that the agent will act on that decision. But the entire problem of weakness of will is that a person does not act according to her own best judgment, but instead does something else. However, because economics models, even with behavioral enhancements, do not allow for the distinction between decision and action, they cannot incorporate a true sense of will. As a result, they can explain why people have incentive to cheat on their diets or delay writing a term paper, but they cannot explain why people with those incentives resist the urge to cheat on a diet or procrastinate, and instead do the thing they know they should do regardless of incentives to the contrary. In short, these models derive behavior deterministically from incentives and constraints, with no true choice involved; as John Davis (2009, 73) wrote, “on the standard view in economics, an entity only ‘acts’ because it is determined to do so as the result of some antecedent cause in a cause-and-effect process,” and therefore “neoclassical economics lacks a true concept of an individual economic agent.”16

24The more elaborate and realistic conception of choice that I have described does not come without a cost: incorporating principles, judgment, and willpower would sacrifice some of the parsimony that has been a key benefit of the neoclassical model of choice; even if it is too simplistic to be descriptively accurate in terms of individual behavior, the neoclassical model is simple enough to generate testable predictions across large populations. Behavioral economics has also been criticized for sacrificing simpilicity; Jolls, Sunstein, and Thaler (1998, 1487-1489) defend the complexities added by behavioral economists by arguing that the predictive power of the neoclassical model is overstated, especially in light of behavior observed by experimental psychologists that led to the development of behavioral economics and which differs from mainstream predictions in systematic ways (not merely increasing a symmetric error term). They also argue that the neoclassical model is too general, which leaves it able to explain everything but predict nothing in particular; in other words, a model that allows for all behavior rules out none. In contrast, Jolls, Sunstein, and Thaler argue, a behaviorally-informed model places sufficient restrictions on behavior to generate definite predictions and allow for falsifiability.

  • 17 A bolder proposal, far outside the scope of this paper, would be to discount the value or precision (...)

25The same argument can be expanded to the more ambitious revisions proposed in this paper. I would argue that accounting for principles, judgment, and willpower not only generates richer and more accurate explanations, but also results in better predictions in cases in which people base decisions on factors other than self-interested preference-satisfaction. For example, changes in demand for “moral goods” such as environmentally friendly cleaning products or fair-trade coffee might be better explained with moral principle than with “preferences for morality,” and the two explanations will also result in different predictions for the price-elasticity of demand. Incorporation of principles and ideals into models that predict voting behavior, too, may generate more accurate predictions than those based on material self-interest alone, especially in areas with large numbers of “values voters.”17

  • 18 In addition to Jolls, Sunstein, and Thaler (1998) and the response in Posner (1998), for early work (...)
  • 19 See Holmes (1897, 459-462); on his “bad man” and law-and-economics, see Cooter (1998).
  • 20 For a Kantian perspective on these issues, see White (2011, ch. 4). For a critique of behavioral la (...)

26This is particularly relevant to behavior typically discussed in law and economics, to which the insights of behavioral economics have been applied since the late 1990s.18 Such behavior typically involves legal norms that restrict the pursuit of self-interest, compliance with which often invokes the influence of principle. Most work in law and economics, however, treats the agent as Oliver Wendell Holmes’ “bad man,” who obeys the law only when it is in his interest to do so, rather than the majority of law-abiding people who obey the law out of principle (or habit).19 According to this view, the potential criminal weighs his gain from criminal activity against the penalty discounted by the probability he will get away with it, not the fact that he is causing wrongful harm to society and its members; the potential injurer considers only her expected liability, not others’ rightful security, before making decisions affecting other people’s property; and the potential breacher considers only the likely damages, not the nature of the promise or commitment made thereby, if he violates the terms of a contract. Economic models of choice focused on preferences have no language or concepts to handle the observance of legal norms for the sake of legal norms, and for all of its contributions to these models on the margin, neither does behavioral law and economics.20

4. More Elaborate Conceptions of Preferences

  • 21 See, for instance, Tversky and Thaler (1990) and Slovic (1995).

27Although behavioral economists accept the centrality of preferences to the economic model of choice, they question the way that human beings process them cognitively and incorporate them into rational deliberation, arguing that preferences are unduly influenced by aspects of choice situations, such as the way options are arranged or framed. They also go much farther than most mainstream economists in questioning the exogenous and stable nature of preferences assumed in the neoclassical conception of choice, suggesting instead that in many cases preferences are unformed or even nonexistent before a choice situation is encountered, and can therefore be manipulated in other interests besides the agent’s own.21

28This leads to two lines of thought, one theoretical and one policy-oriented. The first is to posit the existence of ideal or rational preferences, which John Harsanyi (1982, 55) described as “the preferences [an agent] would have if he had all the relevant factual information, always reasoned with the greatest possible care, and were in a state of mind most conducive to rational choice.” For example, an agent’s behavior may have suggested that she prefers a hot fudge sundae to a fruit bowl for dessert, but her rational preferences may indicate the opposite. In other words, ideal preferences represent what the agent would prefer were she not subject to the various cognitive biases and limitations identified by behavioral economists, who then claim that such preferences better represent the agent’s well-being.

  • 22 See, for instance, Sen and Williams (1982), Sugden (2008), and Hausman (2012, ch. 7).

29There are a number of problems with this.22 Most important for present purposes, ideal preferences are a counterfactual exercise, a supposition of what the agent would prefer in a world different from her own. As Robert Sudgen (2008, 232) writes:

How, without making normative judgments, do we determine what counts as complete information, unlimited cognition, or complete willpower? Even if we can specify what it would mean to have these supernatural powers, how do we discover how some ordinary human being would act if he were somehow to acquire them?

  • 23 For more on this, see White (2014, ch. 2) as well as Hausman (2012, chs 7-8).

This is where behavioral economists have made their point too well: if our cognitive biases, dysfunctions, and limitations are as pervasive as they argue, then imagining what a person would prefer or choose in the absence of them takes us out of the realm of actual human beings into pure speculation. Furthermore, the resulting preference ordering would not be the person’s own, and would not necessarily have a link to her well-being (as preferences are normally presumed to have).23

30More generally, because preferences are only one aspect of choice, alongside principle and ideals as processed by judgment and will, they cannot reliably be “revealed” by behavior and therefore cannot be used to infer irrationality. There are a myriad of reasons an agent may have chosen the hot fudge sundae even if she were fully aware that the fruit cup was the healthier choice. But an external observer who watched our agent choose the hot fudge sundae would likely assume that this reflected a binary, all-things-considered preference, instead of the result of a complex process involving preferences, principles, ideals, all processed by judgment and implemented by will. The exclusive reliance of preferences to represent all aspects of the choice process filters the complexity of that process and generates misleading, distorted implications regarding the nature and “rationality” of such preferences.

31This example highlights another problem: non-ideal or irrational preferences cannot be identified without using a value judgment. The observer who saw our agent choose the hot fudge sundae may have judged her “revealed” preference for the sugary dessert to be irrational because this preference is not in her best interests; the healthier choice would have been the fruit cup. But this reflects a pure value judgment on the part of the observer regarding the agent’s true interests, to which the observer has no epistemic access.

  • 24 See also Berg and Gigerenzer (2010, 21): “It is no trivial question as to whether substantive rathe (...)

32While the subjective nature of interests is beyond the scope of this paper, it does contribute to the larger problem with the myopic focus on preferences in both the neoclassical model of choice as well as behavioral elaborations on it. The question of the existence of preferences before a choice situation is encountered, and the issue of rational versus irrational preferences, both too much emphasis on preferences to the exclusion of their primary source: interests. If we instead conceptualize choice in terms of goals, principles, and ideals, as processed through judgment and will, preferences are revealed to be simply a modeling heuristic to enable mathematical representation of one narrow aspect of the choice process and derivation of a utility function.24 Even though this modeling strategy has some value, I argue that dependence on it has led economists to ignore other aspects of choice that resist mathematical modeling and implicitly accept preferences as “real” objects that are the basis of all decisions.

33When regarded as such, those who question whether preferences exist before choices situations are confronted are correct in a sense. Preferences do not exist, but interests do, and the relevant aspects of an agent’s interests are invoked and combined in a unique way in a new choice situation. For example, before the smartphone was introduced by Apple in 2007, no one could have had a preference or taste for smartphones. As indicated by its unprecedented and continued success, however, we can assume that the smartphone fulfills significant interests of millions of consumers for more basic functions such as handheld computing, telecommunications capacity, and addictive games. These interests could be represented by preferences, but preferences are secondary to interests themselves, and reducing interests to the much simpler and narrow logical form of a binary preference is reductive, especially when it obscures the greater complexity of decision-making.

34Interests also reveal the deeper meaning of ideal preferences, which are only a “better” or more accurate type of preference insofar as they more closely reflect a person’s true interests. Because interests are multifaceted, complex, and subjective, no one other than the agent herself can know (if even she does) whether a particular decision or preference is reflective of her true interests. Therefore, an agent’s decisions or preferences can be declared “irrational” only as a result of external judgment based on someone else’s idea of that agent’s interests, a judgment that reflects the external observer’s ideas about interests in general. Recall the Sugden quote above, which indicated the importance of normative judgments in the “rationalization” of preferences, which cannot be rendered “ideal” without some idea of what it means to be ideal, a judgment that cannot come from the agent herself but only from the person performing the adjustment.

35This is another sign that economists are too devoted to preferences as the sole representation of the aspects of choice: economists who judge an agent’s preferences to be questionably formed or influenced attempt to construct “better” preferences rather than acknowledge that preferences are only a filtered distortion of interests themselves. There would be two results of such as acknowledgment. First, economists would be able to appreciate the many other factors that go into decisions other than those most easily rendered in mathematical form, leading to a richer understanding of choice that would also lead to better predictions (if less easily implemented in mathematical form). Second, this appreciation of the multifaceted, complex, and subjective nature of interests may lead to greater humility regarding their knowledge of people’s own interests and policy-oriented work that better respected those interests.

5. How the Reliance on Preferences Led to Nudge (and the Problems with It)

  • 25 The main proponents of libertarian paternalism and nudge are Richard Thaler and Cass Sunstein; see (...)
  • 26 For instance, see Saint-Paul (2011), Rebonato (2012), and White (2013).

36One particular area of policy that reflects the problems with behavioral economics and its reliance on preferences is libertarian paternalism, better known by its main tool, “nudges,” which are small changes in “choice architecture” that leverage cognitive biases and dysfunctions to steer choices in directions that proponents claim are in choosers’ own interests.25 Many criticisms have been lobbied against nudge, including its manipulative aspects, the effects on cognitive abilities, and its implications for respect of individual choice and autonomy.26 The one most relevant to this paper is the presumption of knowledge of how other people make choices and the interests in which they make them, both of which can be explained by the reliance on preferences in modeling in both neoclassical and behavioral economics.

  • 27 For more on the role of interests in modern paternalism (including nudge policy), see White (2016).

37Nudge proponents assume that agents have 1) general interests such as wealth or health in which they make choices, and 2) preferences based on these interests, some of which are either unformed or irrational (as opposed to ideal preferences), based on the findings of behavioral economists. Based on this simple understanding of choice, policymakers presume that they know both what people’s interests are as well as why they have problems pursuing them optimally (that is in the absence of cognitive flaws). From these premises, policymakers justify altering the superficial aspects of choice situations, such as the order and framing of options, to guide a person’s choices in the direction the policymakers believe are in the person’s best interests.27

38Again, these misunderstandings of the choice process that seem to support the use of nudge policies to guide decisions in people’s best interests can be traced to the myopic focus on preferences in economic models of choice. In those models, preferences are primary, and interests–the basis for preferences–are neglected altogether or reduced to self-interest or wealth maximization. That done, the focus remains on preferences, the quality of which is called into question by anomalous experimental results that show choices deviating from the predictions based on simple self-interested motivations. Rather than question the preference-based model (or the simple interests assumed therein), researchers and policymakers conclude that there must be something wrong with preferences that can be fixed with behavioral modification to lead people to the choices they would have made had their preferences been rational, which by definition are choices in their best interests (as judged as those crafting the policies).

39Note that, in this description, the immediate focus is on preferences and correcting them in light of cognitive biases and dysfunctions to bring them in line with the person’s true interests. As we said before, however, preferences are a simplifying distortion of the individual’s broader and more inclusive interests. Because scholars and policymakers choose to see only preferences, they miss the other factors that enter into decision-making, such as principles, ideals, judgment, and will, and so they wrongly presume to understand choice, why it goes wrong, and how it make it better. Without true knowledge and appreciation of interests, however, they cannot make choice better from the agent’s perspective, but only from their own. In the end, policymakers steer people’s choices in directions that policymakers think is best, perhaps out a genuine concern for people’s well-being, but with little reason to believe they are actually improving it given the multifaceted, complex, and subjective nature of preferences.

40If researchers and policymakers tried to appreciate the nature of interests and their role in decision-making, they might understand that they are multifaceted, complex, and subjective. Furthermore, they might give agents the benefit of the doubt that decisions that seem strange to the policymaker, based on their presumption of simple and singular interests and their own value judgments, may actually be sound decisions made in the agent’s best interests. This will not always be true, of course: we all make stupid choices from time to time, but the only ones who can know that they are stupid choices are the ones making them, who have a much better idea of their interests than anyone else (outside of close friends and family). There is still room for policymakers to try to improve choices they feel are not in people’s best interests, but this should be done through education and persuasion, not subtle and covert nudging. This is based not only on respect for autonomy, but an acknowledgment that people know their own interests in all their complexity better than outsiders do (especially policymakers) and that they can make reasonably sound choices in those interests.

6. Conclusion

41Richard Thaler is well-known for criticizing the positivist, “as if” methodology of neoclassical models that dispensed with descriptive realism in favor of predictive success (1980; 2015). In response to this, he contributed to the development of behavioral economics in the hope of providing the field with a more realistic depiction of human decision-making. As a result, behavioral economics has provided us with a wealth of invaluable knowledge and understanding of the imperfections of human rationality. However, its continued focus on preference-satisfaction as the core of its model of decision-making furthers the positivist legacy of neoclassical economics, and compromises the significant advances that could be made in behavioral law and economics. If a descriptively accurate and predictively successful model of human choice is the goal, behavioral economists and mainstream economists alike should acknowledge, and incorporate into their models, the breadth and complexity of interests, as well as the roles of judgment and will, that can encompass the wide range of human behavior—including legal behavior—that we observe every day.

I wish to thank the guest editors for their extraordinary patience and encouragement, and two anonymous referees for insightful and enlightening comments.

Haut de page

Bibliographie

Ainslie, George. 2001. Breakdown of Will. Cambridge: Cambridge University Press.

Akerlof, George A. 1991. Procrastination and Obedience. American Economic Review, 81(2): 1-19.

Angner, Erik. 2016. A Course in Behavioral Economics. 2nd edn. New York: Palgrave Macmillan.

Ariely, Dan. 2010. Predictably Irrational: The Hidden Forces That Shape Our Decisions. New York: Harper Perennial.

Berg, Nathan, and Gerd Gigerenzer. 2010. As-if Behavioral Economics: Neoclassical Economics in Disguise? History of Economic Ideas, 18(1): 133-165.

Brooks, David. 2008. The Behavioral Revolution. The New York Times, October 27, available at http://www.nytimes.com/2008/10/28/opinion/28brooks.html.

Brooks, David. 2011. Who You Are. The New York Times, October 20, available at http://www.nytimes.com/2011/10/21/opinion/brooks-who-you-are.html

Buturovic, Zeljka, and Slavisa Tasic. 2015. Kahneman’s Failed Revolution Against Economic Orthodoxy. Critical Review, 27(2): 127-145.

Cooter, Robert D. 1998. Models of Morality in Law and Economics: Self-Control and Self-Improvement for the “Bad Man” of Holmes. Boston University Law Review, 78(1998): 903-930.

Davis, John B. 2009. Identity and Individual Economic Agents: A Narrative Approach. In Mark D. White and Irene van Staveren (eds), Ethics and Economics: New Perspectives. London: Routledge, 142-165.

Etzioni, Amitai. 1988. The Moral Dimension: Toward a New Economics. New York: Free Press.

Fehr, Ernst, and Klaus M. Schmidt. 2006. The Economics of Fairness, Reciprocity and Altruism—Experimental Evidence and New Theories. In Serge-Christophe Kolm and Jean Mercier Ythier (eds), Handbook of the Economics of Giving, Altruism and Reciprocity, vol. I. Dordrecht: Elsevier, 615-691.

Fischer, Carolyn. 1999. Read This Paper Even Later: Procrastination with Time-Incon­sistent Preferences. Discussion Paper 99-20, Resources for the Future, available at http://www.rff.org/files/sharepoint/WorkImages/Download/RFF-DP-99-19.pdf.

Fischer, Carolyn. 2001. Read This Paper Later: Procrastination with Time-Consistent Prefer­ences. Journal of Economic Behavior & Organization, 46(3): 249-69.

Gigerenzer, Gerd. 2015. On the Supposed Evidence for Libertarian Paternalism. Review of Philosophy and Psychology, 6(3): 361-383.

Gigerenzer, Gerd, Peter M. Todd, and the ABC Research Group. 1999. Simple Heuristics That Make Us Smart. Oxford: Oxford University Press.

Hargreaves Heap, Shaun, Martin Hollis, Bruce Lyons, Robert Sugden, and Albert Weale. 1992. The Theory of Choice: A Critical Guide. Oxford: Blackwell.

Hayden, Grant M., and Stephen E. Ellis. 2007. Law and Economics after Behavioral Economics. Kansas Law Review, 55(3): 629-675.

Harsanyi, John. 1982. Morality and the Theory of Rational Behavior. In Amartya Sen and Bernard Williams (eds), Utilitarianism and Beyond. Cambridge: Cambridge University Press, 39-62.

Hausman, Daniel M. 2012. Preference, Value, Choice, and Welfare. Cambridge: Cambridge University Press.

Heukelom, Floris. 2014. Behavioral Economics: A History. New York: Cambridge University Press.

Holmes, Oliver Wendell Holmes. 1897. The Path of the Law. Harvard Law Review, 10(8): 457-478

Jolls, Christine, Cass R. Sunstein, and Richard Thaler. 1998. A Behavioral Approach to Law and Economics. Stanford Law Review, 50(5): 1471-1550.

Kahneman, Daniel. 2011. Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.

Kahneman, Daniel, Amos Tversky, and Paul Slovic (eds), 1982. Judgment under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press.

Kant, Immanuel. 1781/1787. Critique of Pure Reason. Trans. N.K. Smith. New York: St. Martin’s Press (1929 edition).

Karlsson, Niklas, George Loewenstein, and Jane McCafferty. 2004. The Economics of Meaning. Nordic Journal of Political Economy, 30(1): 61-75.

Korobkin, Russell. 2011. What Comes after Victory for Behavioral Law and Economics? University of Illinois Law Review, 2011(5): 1653-1674.

Korobkin, Russell B. and Thomas S. Ulen. 2000. Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics. California Law Review, 88(4): 1051-1144.

Margolis, Howard. 1984. Selfishness, Altruism, and Rationality: A Theory of Social Choice. Chicago: University of Chicago Press.

McKenzie, Richard B. 2010. Predictably Rational? In Search of Defenses for Rational Behavior in Economics. Dordrecht: Springer.

Minkler, Lanse. 2008. Integrity and Agreement: Economics When Principles Also Matter. Ann Arbor: University of Michigan Press.

O’Donoghue, Ted, and Matthew Rabin. 1999a. Doing It Now or Later. American Economic Review, 89(1): 103-124.

O’Donoghue, Ted, and Matthew Rabin. 1999b. Incentives for Procrastinators. Quarterly Journal of Economics, 114(3): 769-816.

O’Donoghue, Ted, and Matthew Rabin. 2001. Choice and Procrastination. Quarterly Journal of Economics, 116(1): 121-160.

Peterson, Richard. 2014. The Behavioral Economics Revolution. Exchange, January 5, available at http://thomsonreuters.com/en/articles/2014/behavioral-economics.html.

Posner, Richard A. 1998. Rational Choice, Behavioral Economics, and the Law. Stanford Law Review, 50(5): 1551-75.

Rabin, Matthew. 1993. Incorporating Fairness into Game Theory and Economics. American Economic Review, 83(5): 1281-1302.

Rebonato, Riccardo. 2012. Taking Liberties: A Critical Examination of Libertarian Paternalism. New York: Palgrave Macmillan.

Rostain, Tanina. 2000. Educating Homo Economicus: Cautionary Notes on the New Behavioral Law and Economics Movement. Law & Society Review, 34(4): 973-1006.

Saint-Paul, Gilles. 2011. The Tyranny of Utility: Behavioral Social Science and the Rise of Paternalism. Princeton: Princeton University.

Sen, Amartya K. 1977. Rational Fools: A Critique of the Behavioral Foundations of Economic Theory. Philosophy & Public Affairs, 6(4): 317-344.

Sen, Amartya, and Bernard Williams. 1982. Introduction: Utilitarianism and Beyond. In Amartya Sen and Bernard Williams (eds), Utilitarianism and Beyond. Cambridge: Cambridge University Press, 1-21.

Sent, Esther-Mirjam. 2004. Behavioral Economics: How Psychology Made Its (Limited) Way Back into Economics. History of Political Economy, 36(4): 735-760.

Simon, Herbert A. 1955. A Behavioral Model of Rational Choice. Quarterly Journal of Economics, 69(1): 99-118.

Simon, Herbert A. 1956. Rational Choice and the Structure of the Environment. Psychological Review, 63(2): 129–138.

Slovic, Paul. 1995. The Construction of Preference. American Psychologist, 50(5): 364-371.

Smith, Vernon L. 2009. Rationality in Economics: Constructivist and Ecological Forms. Cambridge: Cambridge University Press.

Social Science Bites. 2012. Robert Schiller on Behavioral Economics. August 1, available at http://www.socialsciencespace.com/2012/08/robert-shiller-on-behavioral-economics/.

Sugden, Robert. 2008. Why Incoherent Preferences Do Not Justify Paternalism. Constitutional Political Economy, 19(3): 226-248.

Sunstein, Cass R. (ed.). 2000. Behavioral Law & Economics. Cambridge: Cambridge University Press.

Sunstein, Cass R. 2014. Why Nudge? The Politics of Libertarian Paternalism. New Haven, CT: Yale University Press.

Sunstein, Cass R. and Richard H. Thaler. 2001. Libertarian Paternalism Is Not an Oxymoron. University of Chicago Law Review, 70(4): 1159-1202.

Thaler, Richard H. 1980. Toward a Positive Theory of Consumer Choice. Journal of Economic Behavior and Organization, 1(1): 39-60.

Thaler, Richard H. 2015. Misbehaving: The Making of Behavioral Economics. New York: W.W. Norton & Co.

Thaler, Richard H., and Cass R. Sunstein. 2008. Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press.

Tversky, Amos, and Richard H. Thaler. 1990. Anomalies: Preference Reversals. Journal of Economic Perspectives, 4(2): 201-211.

Veblen, Thorstein. 1898. Why Economics Is Not an Evolutionary Science. Quarterly Journal of Economics, 12(4): 373-397.

White, Mark D. 2006. Multiple Utilities and Weakness of Will: A Kantian Perspective. Review of Social Economy, 64(1): 1-20.

White, Mark D. 2010. Resisting Procrastination: Kantian Autonomy and the Role of the Will. In Chrisoula Andreou and Mark D. White (eds), The Thief of Time: Philosophical Essays on Procrastination. Oxford: Oxford University Press, 216-232.

White, Mark D. 2011. Kantian Ethics and Economics: Autonomy, Dignity, and Character. Stanford, CA: Stanford University Press.

White, Mark D. 2013. The Manipulation of Choice: Ethics and Libertarian Paternalism. New York: Palgrave Macmillan.

White, Mark D. 2014. The Illusion of Well-Being: Economic Policymaking Based on Respect and Responsiveness. New York: Palgrave Macmillan.

White, Mark D. 2015. Judgment: Balancing Principle and Policy. Review of Social Economy, 73(3): 223-241.

White, Mark D. 2016. The Crucial Importance of Interests in Libertarian Paternalism. In Klaus Mathis and Avishalom Tor (eds), Nudging: Possibilities, Limitations and Applications in European Law and Economics. Dordrecht, Springer, 21-38.

Whitman, Douglas G., and Mario J. Rizzo. 2015. The Problematic Welfare Standards of Behavioral Paternalism. Review of Philosophy and Psychology, 6(3): 409-425.

Haut de page

Notes

1 While sufficient for our current purposes, this description is of course simplified; for more context and detail, see Hargreaves Heap et al (1992).

2 See Kahneman, Tversky, and Slovic (1982) for the seminal work; for more recent exposition and summary, see Ariely (2010), Kahneman (2011), Heukelom (2014), and Angner (2016). On the transition between the “old behavioral economics” associated with Simon and others of the 1950s and 60s, and the “new behavioral economics” of Kahneman, Tversky, and others from the 1970s on, see Sent (2004).

3 In an interview (and elsewhere), Schiller called behavioral economics: “a revolution in economics that has taken place over the past twenty years or so. I think it’s bringing economics into a broader appreciation of reality” (quoted in Social Science Bites, 2012); see also the titles of Brooks (2008) and Peterson (2014), both referencing the behavioral economics revolution, and Brooks (2011), titled “Kahneman and Tversky are like the Lewis and Clark of the mind.” A simple Google search will turn up many more examples.

4 But not to all: Richard Posner, long a leading exponent of neoclassical law and economics, argued that the choice anomalies identified by behavioral economists are easily accounted for by mainstream models. In response to Jolls, Sunstein, and Thaler (1998), Posner wrote: “Some of the insights they ascribe to behavioral economics are already a part of economic analysis of law, which long ago abandoned the model of hyperrational, emotionless, unsocial, supremely egoistic, nonstrategic man (or woman) that [they] in places appear to ascribe to it. Other points they make are new labels for old challenges to the economic model of behavior that owe nothing to behavioral economics in any distinctive sense” (1998, 1552). For a similar defense of the traditional rationality paradigm in economics in general, see McKensie (2010).

5 As opposed to textbook perfect rationality, the ecological rationality of Gigerenzer and Smith (separately) understands human choice processes to be adaptive adaptations to a changing environment: “When heuristics, or decision processes—or action rules—function well in particular classes of environments, then ecological rationality is achieved” (Berg and Gigerenzer 2010, 21; in general, see Gigerenzer et al 1999 and Smith 2009).

6 See, for instance, Hausman (2012), especially chap. 1.

7 Karlsson, Loewenstein, and McCafferty (2004) even argue that various aspects of meaning can be incorporated into the standard utility-maximization model.

8 See, for instance, Fehr and Schmidt (2006).

9 Sen (1977); for further work incorporating principles explicitly within economics, see for instance Minkler (2008) and White (2011).

10 Commitments such as this are often modeled as lexicographic preferences, preferences that must be satisfied before other preferences are considered, but this language obscures the fact that such commitments do not “act” like preferences in that they do not respond to marginal changes in cost.

11 For more on this conception of interests and choice, see White (2014, ch. 3; 2016).

12 For more on judgment in economic models, see White (2015).

13 For a model incorporating willpower into an economic model of choice, see White (2006; 2010).

14 See, for instance, Margolis (1984) and Etzioni (1988).

15 See, for instance, Ainslie (2001), Akerlof (1991), O’Donoghue and Rabin (1999a; 1999b; 2001), and Fischer (1999; 2001), all discussed in White (2010).

16 For more on the role of will in economic choice, see White (2010; 2011, ch. 2).

17 A bolder proposal, far outside the scope of this paper, would be to discount the value or precision of prediction itself, and reorient economics toward understanding behavior and, at best, generating common-sense, back-of-the-envelope predictions.

18 In addition to Jolls, Sunstein, and Thaler (1998) and the response in Posner (1998), for early work in behavioral law and economics see also Korobkin and Ulen (2000) and Sunstein (2000).

19 See Holmes (1897, 459-462); on his “bad man” and law-and-economics, see Cooter (1998).

20 For a Kantian perspective on these issues, see White (2011, ch. 4). For a critique of behavioral law and economics that argues for less theory rather than more (“the language of economics… suggests a level of generalizability, predictive determinacy, and control that the underlying empirical findings do not support”), see Rostain (2000; quote from 975).

21 See, for instance, Tversky and Thaler (1990) and Slovic (1995).

22 See, for instance, Sen and Williams (1982), Sugden (2008), and Hausman (2012, ch. 7).

23 For more on this, see White (2014, ch. 2) as well as Hausman (2012, chs 7-8).

24 See also Berg and Gigerenzer (2010, 21): “It is no trivial question as to whether substantive rather than axiomatic rationality requires preferences to exist at all. The essentializing concept of a stable preference ordering ignores the role of context and environment as explanatory variables that might condition what it means to make a good decision.”

25 The main proponents of libertarian paternalism and nudge are Richard Thaler and Cass Sunstein; see their popular book Nudge (2008) as well as their previous academic work on the subject, such as Sunstein and Thaler (2001), and recent defenses, such as Sunstein (2014).

26 For instance, see Saint-Paul (2011), Rebonato (2012), and White (2013).

27 For more on the role of interests in modern paternalism (including nudge policy), see White (2016).

Haut de page

Pour citer cet article

Référence papier

Mark D. White, « Preferences All the Way Down: Questioning the Neoclassical Foundations of Behavioral Economics and Libertarian Paternalism », Œconomia, 7-3 | 2017, 353-373.

Référence électronique

Mark D. White, « Preferences All the Way Down: Questioning the Neoclassical Foundations of Behavioral Economics and Libertarian Paternalism », Œconomia [En ligne], 7-3 | 2017, mis en ligne le 01 septembre 2017, consulté le 13 décembre 2017. URL : http://journals.openedition.org/oeconomia/2646 ; DOI : 10.4000/oeconomia.2646

Haut de page

Auteur

Mark D. White

Department of Philosophy, College of Staten Island/CUNY. profmdwhite@hotmail.com

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus d’Œconomia sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Association Œconomia
  • Logo CNRS
  • OpenEdition Journals