Behavioral Paternalism To be published in Revue de Philosophie économique/ Revue of Economic Philosophy, December 2014 Christophe Salvat1 CNRS, Triangle, ENS de Lyon Abstract: Over the last decade a new type of paternalism has emerged thanks to the groundbreaking works of some behavioral economists. This new type of paternalism, that I call here Behavioral Paternalism (BP), has become popular through Sunstein and Thaler’s Nudges theory and challenges the view that paternalism is unacceptable today. The aim of this paper is to assess its moral legitimacy (not exclusively focusing on its alleged libertarian nature). The results of my investigation can be summarized as follows. Though BP is usually acknowledged for its ‘libertarian’ character, it does not satisfy the conditions of what is considered, since Feinberg, as ‘soft paternalism’. Nevertheless, BP has a strong point that has been underestimated by its partisans: it first withstands Anderson’s equality argument. Unlike traditional forms of paternalism, BP is not demeaning and does not ostracize any category of people. Lastly, BP can be proved genuinely altruistic. This, however, demands that one abandons Sunstein and Thaler’s main assumptions. Key words: paternalism, behiavorism, ethics, equality, autonomy 1 Chercheur CNRS, Triangle, UMR 5206, ENS de Lyon. Email: Christophe.salvat@ens-lyon.fr. Early versions of this paper have been presented at the Collège International de Philosophie in Paris (January 2010), at the University of Lille (Institut d’Economie Politique, June 2012), at the University of Strasbourg (BETA Seminar, November 2012) and at the University of Cambridge (Faculty of Philosophy, January 2013). I am grateful to all participants for their insightful remarks. I would also like to thank the referees of Revue de Philosophie Economique for their helpful comments. 1 JEL classification: A12; D64; I18; I31; K32 Résumé: Un nouveau type de paternalisme s’est développé ces dix dernières années sous l’impulsion de travaux innovateurs de certains économistes comportementaux. Ce nouveau type de paternalisme, que j’appelle ici paternalisme comportemental, s’est popularisé grâce à la théorie du ‘coup de pouce de Richard Thaler et Cass Sunstein et remet en question l’idée selon laquelle le paternalisme serait inacceptable dans nos sociétés. L’objet de cet article est d’évaluer sa légitimité morale sans, néanmoins, se limiter à son supposé libertarianisme. Les résultats de mon investigation peuvent se résumer ainsi : bien que le paternalisme comportemental soit généralement reconnu pour son caractère libéral, il ne satisfait pas en fait les conditions de ce que Joel Feinberg nomme le ‘paternalisme mou’. Néanmoins, il possède des qualités morales sous-estimées par ses partisans. Il résiste d’abord très bien à la critique égalitariste d’Elizabeth Anderson. A la différence des formes traditionnelles de paternalisme, le paternalisme comportemental n’est pas dégradant et n’est pas ostracisant. Le paternalisme comportemental, enfin, peut se targuer d’être véritablement altruiste, à la condition, cependant, d’abandonner les hypothèses principales de Sunstein et Thaler. Mots-clés : paternalisme, comportementalisme, éthique, égalité, autonomie. 1. Introduction The idea of paternalism has suffered dramatically since the decline of the nineteenth-century model of entrepreneurial capitalism. As a social practice, it is widely considered as a patronizing and morally inappropriate conduct to adopt for any individuals benefiting from a superior economic, social or political position. The philosophical debate over the legitimacy of paternalism, however, twice endeavored to challenge this view. A first wave of arguments in favor of paternalism dates from the 2 1970s2, a second only started a few years ago under the influence of major behaviorist thinkers. It emerges from all these contributions that modern paternalism no longer bears resemblance to social or religious charity. In the 1970s paternalism was essentially considered from a legal point of view (Feinberg 1983, 1986; Dworkin 1971) or, if not, from a medical point of view (Gert and Culver 1976; Buchanan 1978). The questioning, however, remains the same: under what conditions can paternalism be morally acceptable? In the nineteenth-century entrepreneurial or social paternalism was often perceived as serving the interests of the dominant class under false pretenses. Paternalism was thought immoral because it was not genuinely altruistic. Since WWII legal paternalism has been criticized for being intrusive and coercive. Paternalism was then thought immoral because it was perceived as a threat to individual freedom. A few years later paternalism suffered another philosophical blow when it emerged that it was also violating individuals’ basic right to equality (Dworkin 1978; Anderson 1999; Arneson 2005). As one can see, the moral demands made to paternalism have shifted over time and conformed to new ideologies. But none disappeared altogether. Theories defending paternalism have to therefore adjust to continually meet more demanding moral standards. It is reasonable to say that, paternalism, today, is considered illegitimate a) when it acts under false pretenses to satisfy the interest of the paternalist agent, b) when it violates the individual autonomy of the people interfered with and c) when it does not respect the equality between people by singling out those who are deemed unable to decide for themselves. I shall refer to these three moral requirements respectively as the altruistic, the liberal and the egalitarian provisos. At the beginning of the new millennium, spurred on by the success of behavioral studies, new theories on paternalism took up the challenge (Jolls, Sunstein, and Thaler 1998; Thaler and Sunstein 2003, 2008; Camerer et al. 2003; Loewenstein and Haisley 2008). Sometimes called ‘libertarian’, ‘asymmetric’, ‘light’ or simply ‘new’ paternalism, these theories are all related to behavioral sciences. For simplicity’s sake, I shall refer to them as Behavioral Paternalism (BP). Is BP really different from Most of the original contributions as well as a number of unpublished ones (including Dworkin’s Second Thoughts) have been gathered in the volume edited by Rolf Sartorius in 1983, (Sartorius 1983). 2 3 classic forms of paternalism? Does it really meet the moral criticism paternalism is facing? This is the object of this study. This paper does not pretend to be a defense (or a criticism) of BP, it is rather an acknowledgement of its theoretical potentialities, including those they have chosen not to pursue. This paper is divided into four sections. In the first section I present this new theory of paternalism I call Behavioral Paternalism (Section 2). I demonstrate that behavioral theories of paternalism indeed differ from traditional forms of paternalism. The three following sections examine its moral robustness, which can be gathered under three moral provisos: the liberal proviso, the egalitarian proviso and the altruistic proviso. The issue of personal autonomy has been the most discussed so far. It has been argued that BP was the only form of paternalism capable of meeting the liberal proviso. I shall argue that this is not the case (Section 3). I believe, nevertheless, that BP definitively has a moral advantage over other types of paternalism, in its unique ability to meet Anderson’s egalitarian critic (Section 4). It could also meet certain arguments generally made against the morally deceiving nature of paternalism, even though this is clearly not the path chosen by its defendants (Section 5). 2. Behavioral Paternalism In this section I present a new approach to paternalism that is referred to sometimes as ‘New Paternalism’ (Rizzo 2009; Mead 1997; Holt 2006; Economist 2006), ‘Asymmetric Paternalism’ (Camerer et al. 2003; Loewenstein 2007), ‘Light’ (Loewenstein and Haisley 2008), ‘Positive’ (Blumenthal 2007 ) or ‘Libertarian Paternalism’ (Sunstein and Thaler 2003, 2008). Despite this variety of designations I nevertheless prefer to call it ‘Behavioral Paternalism’ (hereafter BP). My justification for this is that, apart from ‘New Paternalism’ (but new is not a definitive characteristic), all these expressions prejudge the legitimacy of this new type of paternalism. Since my aim is precisely to discuss the terms of its legitimacy, I therefore believe that a neutral but informative qualifying adjective is more appropriate. The term ‘behavioral’ imposes itself since the revival of paternalism observed for the last decade originates from behavioral sciences. 4 The American lawyer Cass Sunstein and economist Richard Thaler are the first and most important champions of BP. Their groundbreaking works on libertarian paternalism have produced a vast academic literature and found a favorable echo with politicians. Their work, directly inspired by Richard Thaler’s own contributions to behavioral economics, has the merit of raising anew the question of the possibility and the legitimacy of paternalism in the 21st century. The paternalism they defend radically differs from the traditional philosophical approaches inherited from Mill’s ‘soft antipaternalism’ (Feinberg 1986, p.15). They contest, in particular, the idea that paternalism is necessarily coercive and that it is predominantly institutionally-based. Behavioral studies gave partisans of paternalism two major arguments. Firstly empirical studies have repeatedly shown that individual actions do not meet the rationality standard that the economic theory would expect. The idea of bounded rationality, introduced by Herbert Simon, has been decisive in explaining what was otherwise considered as economically inconsistent. For some years now, economic models of bounded rationality have been supplemented by new evidence provided by neuroscientists, giving birth to a new discipline called neuroeconomics (Camerer, Loewenstein, and Prelec 2004; Camerer, Loewenstein, and Prelec 2005; Camerer 2007; Kahneman 2011). According to Colin Camerer neurosciences definitely undermine the ‘Revealed-Preference Approach’ by showing the role played by the difference between experienced utility (‘liking’) and decision-utility (‘wanting’) in individual suboptimal choices. Benevolent agents (or governments), who are aware of the gap between the utility individuals think they will obtain (what they want) and the utility (or disutility) they will actually get (what they like), are then given the opportunity to act paternalistically without risks (Camerer 2006, 2008). In his terms “paternalism could be justified, in terms of a person’s own welfare, if the wanting system does not produce what the liking system likes and if the intervention creates more liking than the person would achieve on his own or with market-supplied help” (Camerer 2006, p.101-2). A second argument for paternalism is individuals’ endogenous preferences. Preferences are claimed by behaviorists to be endogenous to social contexts and consequently inconsistent from one situation to another, the lack of any past or similar reference introducing an element of arbitrariness in the 5 evaluation3. Over the last twenty years, behavioral economics has offered a great deal of evidence to support this thesis (Lichtenstein and Slovic 2006). Framing, anchor or endowment effects have been identified as some of the key mechanisms at stake (Thaler 1980, Loewenstein and Adler 1995; Tversky and Kahneman 1981, 1986). But if endogenous preferences are responsible for individuals’ poor rational behavior, they can also be the medium for their own remedy. For Sunstein and Thaler, for instance, it is possible to significantly reduce the lack of individual rationality by switching the default rule settings from an ‘opt in’ to an ‘opt out’ formula. They illustrate the benefits of setting change in default rules with two case studies: the case of 401 (k) employee savings plans (Madrian and Shea 2001, Choi et al. 2002) and their own experience of the University of Chicago’s car park payment system. In the first case, employers decide that instead of letting their employees join a savings plan, the American 401 (k) plan, they would automatically enroll them unless employees specifically stated that they did not wish to be included. When eligible, employees usually receive a 401 (k) plan form that they must complete in order to join. In this case, however, employees only receive a statement in which they are told that, unless they opt-out, they are automatically enrolled in the plan. In both studies, a significant increase in the enrollment rate was observed. In the second case, a change in the tax law made it possible for employees to pay for employer-provided parking on a pre-tax basis. The University of Chicago presumed that, although being in the employees’ interest, many of them would not take the trouble to fill in and send back the form and decided to enroll all employees automatically. Those who preferred to pay with after-tax dollars were, however, free to opt out4. Had the university chosen a different default rule (like an ‘opt in’ strategy), the authors reckon that “many employees, especially faculty members (and probably including the present authors), would still have that form The idea of ‘coherent arbitrariness’ of preferences developed by Dan Ariely, George Loewenstein and Drazen Prelec shows, however, that valuations are not entirely arbitrary: absolute valuation of goods or experience is for a large part arbitrary, but relative valuations can prove to be coherent (Ariely, Loewenstein and Prelec 2003). 4 In practice, however, default rules are sticky and guaranteeing the conditions of free choice is difficult. Once a person has been enrolled in a default rule, he or she tends to overestimate its benefits (endowment effect). If opting in is a deterrent for many people, opting out from a default rule is even more dissuasive. To respect freedom of choice, a libertarian planner must set up a default rule that will be easy and costless for individuals to opt out of. Ideally, it should not be more demanding than a ‘one-click’ procedure (Thaler and Sunstein 2008, p.249). 3 6 buried somewhere in their offices and would be paying substantially more for parking on an after-tax basis” (Thaler and Sunstein 2003, p.1171). A minor effort, such as filling in a form, is often sufficient not to claim a benefit people are entitled to. In the introduction I have suggested that – in order to be moral – paternalism ought to fulfill three criteria: respecting personal autonomy, respecting individual equality and being genuinely morally motivated. Let us now consider these three criteria in connection with behavioral paternalism. 3. The liberal proviso The first and most intuitive argument opposed to paternalism is the respect of individual freedom. It is by far the one that has been given the most attention. Opponents of paternalism defend a right to autonomy or self-direction. It is naturally not an absolute right. The nature of governments is to interfere with an individual’s decision-making process. Liberals believe that it is legitimate to do so only when individual actions are other-regarding. They hence justify legal restrictions such as the obligation for an individual to wear a helmet or a seatbelt whilst driving a car or a motorbike by the benefit reaped by others (the cost born by the collectivity) rather than by the utility the individual should expect from the measure. This means that to comply with this critic, self-regarding justifications for paternalism must always be proved inadequate to back interventions or they must be able to be otherwise justified. In practice, however, it is rather difficult to ascertain that no selfregarding justification for paternalism can ever be found or that other-regarding justifications will be sufficient. The question is therefore not whether paternalism is a legitimate practice but to what extent autonomy provides a justification (or not) for paternalism. To remedy the totalitarian tendency of paternalism, Joel Feinberg proposes to vet its practice on the condition that it respects individual freedom. This ‘soft’ or ‘weak’ version of paternalism originates from Mill’s own ambiguity in his famous harm principle: ‘That principle is, that the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number, is self-protection. That the only purpose for which power can be rightfully 7 exercised over any member of a civilized community, against his will, is to prevent harm to others. […] He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because in the opinion of others, to do so would be wise, or even right’ (Mill 1989, p.13). Due to the ambiguity of Mill’s phrasing (‘against his will’) – but also due to a pervasive ambiguity throughout his work (Arneson 1980; Claeys 2013; Brink 2013) – the discussion about the legitimacy of paternalism soon boiled down to a single argument: the degree of voluntariness in the action interfered with. The notion of will can indeed either refer to a current (and temporary) desire or to a ‘true’ and rational will. From this information the nature (and hence the legitimacy) of paternalism could be inferred: a weak or soft paternalism meant to protect individuals against self-inflicting (but not fully voluntary) harm, while “in its extreme version”, a strong or hard paternalism was supposed “to guide them, whether they like it or not, toward their own good” (Feinberg 1983, p.3). For Joel Feinberg, the violation of individual autonomy is the sole possible justification for antipaternalism. Conversely autonomy can also be its sole possible justification. In a certain number of cases paternalistic interventions can thus be justified by the absence or insufficiency of individual autonomy. Four factors contribute, for Feinberg, to weaken the autonomous or voluntary quality of individual decisions. Those are: 1. Coercion or threat of coercion 2. Psychological dependency (drug, alcohol, hypnosis…) 3. Emotional ascendency (depression, anger, obsession…) 4a. Misinformation, deceit or wrong belief that misleads the agent relative to the consequences of his action (i.e. accidentally taking arsenic for salt) 4b. Misinformation or a wrong belief (produced by the individual inability to gather correct information) that eventually misleads him relative to the consequences of his action. 8 Voluntariness is, for Feinberg, essentially based on responsibility. The five above circumstances obliterate the decision maker’s responsibility since the choices made do not express the individual character. Conversely, individuals take full responsibility for autonomous or voluntary choices “since they represent him faithfully, expressing his settled values and preferences” (Feinberg 1989, p.113). For Feinberg, misinformation always constitutes an autonomy-diminishing factor5. Soft or weak paternalism typically describes situations of involuntary acts or misinformed decisions that must be interfered with for the person’s own good. Typically, a soft paternalist can legitimately prevent a person from using the saltcellar if they have not been informed that it contains arsenic instead of salt. Similarly, a person under the influence of drugs or alcohol can be temporarily prevented from making decisions that he/she may bitterly regret the next day. Alternatively, if a person is appropriately informed of the risks incurred from crossing a dangerous bridge, smoking, or dueling, then no external interference is ever justifiable to stop that person carrying out such an action. Soft paternalism re-establishes what some regard as an essential feature of liberalism: the right to act irrationally. Hard paternalism, on the contrary, is by nature coercive. According to Gerald Dworkin, for instance, paternalism covers “the interference with a person’s liberty of action justified by reasons referring exclusively to the welfare, good, happiness, needs, interests or values of the person coerced” (Dworkin 1972, p.65). It legitimizes interference from voluntary and well-informed actions. Among all the possible examples, Dworkin mentions laws requiring motorcyclists to wear safety helmets, laws regulating sexual conduct, laws forbidding people from swimming at a public beach when lifeguards are absent and laws compelling people to save a fraction of their income for retirement. Although some of these examples would not be considered as paternalistic today, either because they involve a third element (the economic cost borne by the society) or because they are no longer considered selfdestructive or wrong (homosexuality), these examples are representative of an essential feature of hard Richard Arneson argues, however, that “a person’s actions may be authentic expressions of his personality without being deliberately chosen” (Arneson 1980, p.487). 4b is thus only acceptable if the behavior responsible for the misinformation is itself ‘out of character’, that is, which is submitted to one of the previous autonomydiminishing factors. He concludes that voluntary choice is admittedly “important but [that it] does not plausibly have the make-or-break significance that soft paternalism attaches to it” and that it is therefore “a mistake to make a fetish of” it (Arneson 2005). 5 9 paternalism: the source of risk assessment. Whilst supporters of soft paternalism consider risk assessment as the risk taker’s responsibility, risk is externally assessed by the paternalistic agent. Defendants of BP trust that its main asset is not to be coercive. This is the reason why they described it as ‘libertarian’, ‘soft’, ‘light’ or ‘asymmetric’. Sunstein and Thaler describe their approach as “a relatively weak and nonintrusive type of paternalism” (Sunstein and Thaler 2003, p.1162). This, I believe, is grossly misleading. Sunstein and Thaler’s libertarian paternalism borrows features from these two opposing approaches. It is not a form of soft paternalism. It is, on the contrary, rooted in a very weak notion of voluntariness that makes it closer to hard paternalism. The libertarian feature of their theory relies exclusively on freedom of choice. Yet, their analysis, based on behavioral studies, shows that most individuals’ choices are sub-optimal because their valuations depend on arbitrary anchors and frames. To a certain degree, most of their actions are then involuntary: they want to be thin but eat fatty foods, they want to have a comfortable retirement but don’t save enough money, they want to donate their organs but don’t say so etc. Their actions are involuntary because determinant information is hidden from individuals (like in Mill’s dangerous bridge case) (4a) but also because they believe individuals are naturally inapt to correctly treat the information they have (anchors, framing effects) (4b). There is definitely an incongruity at simultaneously defending absolute freedom of choice and a weak concept of voluntariness. By downsizing individual responsibility, the authors actually undermine the possibility of a liberal paternalism. BP, against all expectations, does not comply any more with individual freedom of choice than any other forms of paternalism. Partisans of paternalism would be misled to think they have found there an answer to the liberal critic. They have, however, other reasons to rejoice. Its main moral asset is not there. What BP does really change for paternalism is its relation to equality. 4. The egalitarian proviso 10 The second criticism addressed to paternalism is that it violates the fundamental principle of equality. Many philosophers understandably condemn paternalism for its condescending and offensive practices (Shiffrin 2000; Anderson 1999; Kleinig 1983; VanDeVeer 1986). A number of liberal thinkers consider the right to equality second to none (Dworkin 1978; Anderson 1999; Rawls 1971; Arneson 2000), but they often disagree on what constitutes an unfair inequality. Paternalism primarily originates from the desire to correct inequality of luck, that is to say inequalities that do not result from individual choices. There is, for instance, little merit in becoming deputy director of a company when one is born in a rich and well-educated family in the nineteenth century, and when moreover the company belongs to one’s father. It is conversely extremely difficult to access such a position when one is born in a working class family or a poorhouse orphanage. This is what is called (bad) brute luck. This is the reason why, historically, the wealthy and the well-educated often felt the need to share what they believed to be undeserved luck with the working classes by providing them with free access to education. Although intuitively very appealing, luck egalitarianism has come under criticism for its demeaning attitude towards those who were deemed to be victims of bad luck. What Anderson reproaches about luck egalitarianism is that it categorizes the victims of bad luck according to what they are (stupid, untalented, disabled, ugly, socially awkward, etc…). She argues that the point of equality is neither resources nor luck but status. And treating a category of people as children or adults as “too stupid to run their lives” (Anderson 1999, p.301) is incompatible with the respect all individuals should equally be granted. “Egalitarianism”, she writes, “ought to reflect a generous, humane, cosmopolitan vision of a society that recognizes individuals as equals in all their diversity. It should promote institutional arrangements that enable the diversity of people’s talents, aspirations, roles, and cultures to benefit everyone and to be recognized as mutually beneficial. Instead, the hybrid of capitalism and socialism envisioned by luck egalitarians reflects the mean-spirited, contemptuous, parochial vision of a society that represents human diversity hierarchically, moralistically contrasting the responsible and irresponsible, the innately superior and the innately inferior, the independent and the dependent. It offers no aid to those it labels irresponsible, and humiliating aid to those it labels innately inferior.” (Anderson 1999, p.308). 11 No wonder that she finds its response, paternalism, offensive. Unsolicited, paternalist actions have disastrous effects on the ill-endowed people they are supposed to help. By granting people a (financial) compensation for their condition, they actually treat them as incomplete human beings. In its most traditional form (derived from the father-child analogy), paternalism is also often justified by the economic, cognitive or social superiority of the paternalist agent6. I believe, however, that behavioral paternalism can partly elude these criticisms. It is true that paternalism, generally speaking, is usually motivated by a willingness to correct an (unfair) inequality between people. It thus implicitly establishes a distinction between the ‘lucky’ and ‘unlucky’ ones. Amongst the ‘unlucky’ ones, paternalism particularly favors those who have no personal responsibility in their misfortune (‘brute bad luck’): the ‘disabled’, the ‘ugly’, the ‘stupid’, and the ‘socially awkward’ ones (Anderson, 1999). But paternalism is only demeaning if it finds its justification in such categories. I claim, however, that this ought not to be the case, and that a morally legitimate paternalism is possible if it is grounded on positional asymmetries. This means, in particular, that a superior economic rationality (a higher ability to compute or to order preferences) does not qualify as a good reason to interfere with someone else’s life. In addition to being demeaning for those deemed to be intellectually ill-endowed, it has disastrous totalitarian consequences as it logically grants the most intelligent individuals the ability to overturn the decisions of the normally endowed ones ‘for their own sake’ (Wikler 1979). By contrast, it is – I contend – morally acceptable for a person to act when her relative position gives her a cognitive advantage. Take Mill’s famous example of the threatening bridge. Consider that A only knows about the poor state of the bridge because he is a local, whereas B – who is about to cross it – is foreign to the area. If B were local himself, he would not need A’s warning and would also probably warn foreigners against the danger of this bridge. A is in a relatively better position to make a decision on this matter than B. This situation does not, however, say anything about A being inherently superior to B. For Gert and Culver, for instance, “paternalism can be practiced by anyone who has qualifications which he believes enable him to see better than S what is good for S’s good” (Gert and Culver 1976, p.50). 6 12 The same argument can apply to BP. It is possible to show that in the case of BP, paternalist agents cannot claim any superiority of body or mind. Take the cafeteria example: the director, or whoever arranges the dishes, is no different from her customers. In fact, she is a customer herself on her days off. The customers of the cafeteria are not victims of brute luck, waiting for her help. And she has no reason whatsoever to be kind to them (Sunstein and Thaler wrongly claim that she is necessarily benevolent, see next section). Nothing distinguishes her from the ‘paternalized customers’ except the position she occupies at the decision-making moment. If she does arrange the dishes so that her customers eat healthier, then her action undoubtedly qualifies as paternalistic. There is nothing demeaning about it: customers are not disgraced or marked out and the director is not motivated by a sense of justice. She arranges the dishes relatively to what she considers to be a ‘good’ order: meaning good for her and good for the others. BP does not require a cognitive or moral superiority from the paternalist agents. BP is what could be referred to as an ‘opportunistic paternalism’. If anything, paternalism reflects individuals’ equality in front of behavioral bias. It is rather unfortunate that Sunstein and Thaler did not appreciate this point in their works, as it could have been one of their strongest arguments. They preferred seeing paternalist agents as ‘planners’ or ‘choice architects’, which admittedly damages their chance of defending their paternalistic theory on egalitarian grounds. In the previous paragraphs I have demonstrated 1) that BP does not conform to the liberal proviso but 2) that it is the only kind of paternalism that actually comply with the equality proviso. The last point to be considered is its altruistic nature. 5. The altruistic proviso The third criticism addressed to paternalism is that under the false pretension of being altruistic, it actually only serves the interest of the paternalistic agent. Paternalism is traditionally accused of being immoral because it is self-interested and that it uses immoral means to conceal its real nature. The captains of industry of the nineteenth-century who built social houses, school and hospitals for their workers have often been accused of being more interested in the productivity and the low turnover of 13 their company than by the actual well-being of their low paid and hardworking employees. Besides, it is not difficult to imagine cases where ‘good natured people’ resort to ‘wrong means’ in order to complete what they consider as being ‘a good action’7. There is undoubtedly some truth in this. There is no point in denying it. Paternalism has indeed suffered from a long history of abusive practices. But in that respect, I believe that paternalism does not significantly differ from benevolence: a large number of benevolent actions are also self-interested but it does not imply that benevolence, as a general practice, is wrong. So let us admit that moral actions do exist, and that paternalism can be genuinely altruistic. By altruistic I mean disinterested and benevolent. I do not contest the possibility that the agent might have some indirect and/or long-term interest in his benevolent action. After all, we all ultimately gain from altruism. I merely consider that his action is altruistic if it is not primarily motivated by self-interest. To borrow Thomas Nagel’s phrase, it is “a willingness to act in consideration of the interests of other persons, without the need of ulterior motives” (Nagel 1970, p.79). Actions that are primarily motivated by self-interest and concealed as altruistic can be discarded as falsely paternalistic. The difficulty, so far, has been to prove the possibility of genuinely paternalistic action. Consider Sunstein and Thaler’s famous example of the cafeteria. The director is aware that the way she arranges the line of dishes will ‘frame’ the customers’ choice. According to the authors, she faces four alternatives (Sunstein and Thaler 2003, p.1164): 1. She could make choices that she thinks would be best for the customers, all things considered. 2. She could make choices at random. 3. She could choose those items that she thinks would make the customers as obese as possible. 7 Allen Buchanan, for instance, has condemned a number of disgraceful medical practices involving withholding information or even bluntly lying to patients about their state of health to ‘avoid them any unnecessary harm’ (Buchanan 1978). Bernard Gert and Charles Culver, studying the moral practices of medicine (a field known today as bioethics), also condemn the shams of medical paternalism (Gert and Culver 1979; Gert et al. 2006). They contend in particular that “an essential feature of paternalistic behavior toward a person is the violation of moral rules (or doing that which will require such violations), for example, the moral rules prohibiting deception, deprivation of freedom or opportunity, or disabling” (Gert and Culver 1976, p. 48). Most of the criticisms addressed to medical paternalism can nevertheless be handled through the principle of equality put forward by Elizabeth Anderson. 14 4. She could give customers what she thinks they would choose on their own. Suppose for the moment that she is well-aware that customers (much like herself) know what is healthy for them and what is not. Suppose, also, that she is also aware (as her customers are as well) that the colors, the shapes, the odors and the disposition of goods arranged in front of them can influence their choice towards the unhealthiest dishes. Being director of the cafeteria and being in charge of the disposition of the line of dishes, she therefore has the opportunity to rearrange the dishes in order to reduce this behavioral bias. So, how likely is it for the director of the cafeteria to choose the altruistic option? Sunstein and Thaler claim that she will necessarily make the ‘right’ choice (Sunstein and Thaler 2003, p.1164). For one thing, it is largely unsupported (Blumenthal-Barby 2013), and for another, it is morally unjustifiable. Supporters of BP are indeed stuck in-between two different but equally damaging positions. The first one consists in defaulting to altruism rather than deliberately choosing it. Robert Sugden once pointed out that one option is surprisingly missing from the list: the cafeteria director might simply choose to display the dishes in the most economically efficient way (Sugden 2008). The authors’ reasons for not including it were twofold. First they argued that planners are not necessarily looking for profit maximization. Even directors of private companies, like school cafeterias, are not constantly subject to market pressures (Sunstein and Thaler 2003, p.1165). Secondly, and more importantly, they claimed that “market success will come not from tracking people’s ex ante preferences, but from providing goods and services that turn out, in practice, to promote their welfare, all things considered. Consumers might be surprised by what they end up liking; indeed, their preferences might change as a result of consumption.”(Sunstein and Thaler 2003, p.1165). This basically comes down to acknowledging the fact that paternalist agents indeed have ulterior self-interested motives, effectively ruining any endeavor to restore the morality of paternalism. Default choices are not genuinely 15 disinterested. From a libertarian vantage point, moreover, defaulting is not an option. Self-ownership theoretically excludes the possibility of individuals not having the choice not to be benevolent8. A second possibility would be for BP to claim that paternalist agents are morally unimpeachable, and that they deliberately choose to be paternalistic out of sheer altruism. Paternalist agents would then be composed of morally righteous individuals having the opportunity to help the others by engineering their decision-making context. But this would not do either. Moral righteousness would undoubtedly violate the equality proviso described above. Paternalism has been widely (and partly justly) criticized for letting the self-proclaimed ‘uppermost intelligent’ or ‘supremely rational’ individuals interfere with those they considered ‘too stupid to run their lives’. But what if they were deemed morally ‘superior’? What if they owed their status to their moral rather than to their cognitive excellence? Paternalism would still be unjustifiable. At first sight, BP seems to fail to provide a reasonable solution to the altruistic proviso. In fact, it seems to be even less capable of defending itself against this kind of criticism than any other forms of paternalism. Supporters of traditional paternalism can claim that altruism does exist and that it is crucial to differentiate between genuine and false paternalistic practices. There is a way out of this dilemma; a way offered by behavioral studies yet largely ignored by BP’s own supporters, except maybe for Jeremy Blumenthal (Blumenthal 2005, 2007, 2013). It consists in explaining altruistic actions through emotional biases9. Consider the cafeteria example above. Suppose 8 Sunstein and Thaler acknowledge this libertarian principle. They however argue that if benevolence cannot be forced, it can be artificially prompted while freedom of choice is respected. The different systems of organs donations adopted by European and American countries illustrate that possibility. When Americans need to actively express their consent to be removed organs when they die, some Europeans countries such as France presuppose that French implicitly consent to give organs unless explicitly stated. The authors reckon that the French setting of the default rule does not infringe on individuals’ freedom of choice and that a change in the American policy could save thousands of lives each year. They call it libertarian benevolence. 9 Behavioral biases divert decision-makers away from their ‘ideal’ choice in morals as well as in prudence. Numerous experimental studies support this statement, a number of which have been presented by John Doris in a compelling book on individuals’ (absence of) character (Doris 2002). Amongst the vast array of experiments quoted by Doris, let me mention two cases to illustrate the importance of framing effects in moral choice. The first one, undertaken in 1972 by the American psychologists Alice Isen and Paula Levin, took place in a shopping mall near a phone box (Isen and Levin 1972). Passersby and customers using the phone box were the unwitting participants of the experiment. At the end of each call, a young woman ‘accidentally’ dropped a folder, thus spreading all her papers across the floor. The aim of the experiment was to study whether individuals are more likely to behave altruistically when they are in a good mood. Phone callers are divided in two groups, a control group for which nothing is done and a test group composed of good-humored subjects. In order to put the subjects in a good mood the experimenters place a ‘forgotten’ dime in the phone receiver for the callers to find. 16 now that the director has to order fifteen dishes equally distributed between three classes of profitability level. She has to order all dishes, including the equally profitable ones. Since her interest as manager of the cafeteria is already satisfied, she is otherwise indifferent to the way in which dishes are arranged within each profitability category. She could, for instance, adopt a random criterion of distribution10. Random choice set apart there are many other possible criteria to arrange dishes: one can order them by size, by color, by shape, by composition, by season, by name, by origin etc. Altruism (or conversely malevolence) is then just one of many options available to the director of the cafeteria, which she chooses according to her mood or to irrelevant circumstances. It is absolutely fundamental that she does not feel the need to systematically act altruistically and that she is not nudged herself to do so. Paternalism is not and must not be inevitable if it wants to be morally acceptable. The fact that individuals who have the opportunity to act paternalistically do not systematically decide to do so has a second advantage: it limits the practice of paternalism and consequently contributes to its social acceptability. Most of the philosophical debate on freedom focused on the legitimacy of paternalistic actions taken separately. Very little attention has been given, however, to the disagreement and the sense of violation felt by individuals who continuously have their choices interfered with, however legitimate each of these interferences might be. All paternalistic opportunities should not be exercised, however legitimate or well-intentioned they are. If everyone were to act altruistically our life would soon become absolutely unbearable. 6. Conclusion The results of the study were striking: 14 out of 16 callers who found the dime came to the aid of the ‘clumsy’ young woman, whilst only one out the 25 who did not find the dime did! The second study belongs to a series of experiments led by the social psychologist Robert A. Baron (Baron and Thomley 1994; Baron and Bronfen 1994; Baron 1997) highlighting the effect of aromas on human behavior. In the last experiment Baron shows that individuals located near to a pleasant source of aromas (such as hot croissants or coffee) were more disposed to help passersby (in need of change in his experiment) than those who were situated in a neutral aromatic environment (Baron 1997). All these observations, as well as many others, corroborate recent neurobiological studies that show the impact of (positive) emotions on altruism (Damasio 2004). 10 Can a choice be absolutely random? Statisticians and neuroscientists have observed that the human brain is illdisposed toward randomness and that individuals frequently prefer adopting any given criterion of choice rather than (unsuccessfully) attempting to simulate a random statistical distribution (Mlodinow 2009). 17 In this article I discussed whether the new type of paternalism defended by Sunstein and Thaler (amongst others) is better prepared to respond to moral attacks than any other type of paternalism so far. Contrary to a common view, behavioral paternalism is not better equipped to meet liberal attacks made against paternalism since John Stuart Mill. I demonstrated that it still does infringe on individual freedom and it certainly cannot be described as ‘libertarian’. To some extent, it is a much harder form of paternalism than Feinberg’s legal paternalism. It does not succeed either in answering to Anderson’s devastating arguments against the condescending nature of paternalism, or even to prove the fact that genuine disinterested paternalistic actions could exist. In that respect, behavioral paternalism fails to make a case for paternalism. Yet there is something in this class of theories that is definitely worth thinking about, despite its current theoretical shortcomings. Behavioral studies do bring something new to the field, something that well could be further developed as a proper defense of paternalism. For some reason, supporters of BP ignore these potentialities to dredge up the old debate on paternalism and freedom, ultimately showing that they do not consider BP as a completely new approach to paternalism. This would have been, however, their best shot. Anderson Elizabeth. 1999. "What Is the Point of Equality?" Ethics, 109 (2):287-337. Arneson Richard. 1980. "Mill versus Paternalism." Ethics, 90 (4):470-489. Arneson Richard. 2000. "Luck Egalitarianism and Prioritarianism." Ethics, 110 (2):339-349. Arneson Richard. 2005. "Joel Feinberg and the Justification of Hard Paternalism." Legal Theory, 11:259-84. Ariely Dan, Loewenstein George and Prelec Drazen. 2003. '"Coherent Arbitrariness": Stable Demand Curves without Stable Preferences', The Quarterly Journal of Economics, 118, 73-105. Blumenthal-Barby J.S. 2013. "Choice Architecture: A Mechanism for Improving Decisions while Preserving Liberty?" In Paternalism. Theory and Practice, edited by C. Coons and M. Weber, 178-196. Cambridge: Cambridge University Press. Blumenthal J.A. 2005. "Does Mood Influence Moral Judgment?: An Empirical Test with Legal and Policy Implications." Law and Psychology Review, 29:1-28. Blumenthal J.A. 2013. "A Psychological Defense of Paternalism." In Paternalism. Theory and Practice, edited by C. Coons and M. Weber, 197-215. Cambridge: Cambridge University Press. Blumenthal J.A. 2007 "Emotional Paternalism." Florida State University Law Review, 35:1-72. Brink David. 2013. Mill's Progressive Principles. Oxford: OUP. Buchanan Allen E. 1978. "Medical Paternalism." Philosophy and Public Affairs, 7 (4):371-90. 18 Camerer Colin. 2008. "The Case of Mindful Economics." In The Foundations of Positive and Normative Economics: A Handbook: A Handbook, edited by A. Caplin and A. Schotter, New York: Oxford University Press, USA, 43-69. Camerer Colin. 2006. "Wanting, Liking, and Learning: Neuroscience and Paternalism." The University of Chicago Law Review, 73 (87):87-110. Camerer Colin. 2007. "Neuroeconomics: Using Neuroscience to Make Economic Predictions." The Economic Journal, 117 (519):26-42.. Camerer Colin, Issacharoff Samuel, Loewenstein George, O'Donoghue Ted, and Rabin Matthew. 2003. "Regulation for Conservatives: Behavioral Economics and the Case for "Asymmetric Paternalism"." University of Pennsylvania Law Review, 151 (3):1211-1254. Camerer Colin, Loewenstein George, and Prelec Drazen. 2004. "Neuroeconomics: Why Economics Needs Brains." Scandinavian Journal of Economics, 106 (3):555-579. Camerer Colin, Loewenstein George, and Prelec Drazen. 2005. "Neuroeconomics: How Neuroscience Can Inform Economics." Journal of Economic Literature, 43 (1):9-64. Choi James, Laibson David, Madrian Brigitte, and Metrick Andrew. 2002. "Defined Contribution Pensions: Plan Rules, Participant Decisions, and the Path of Least Resistance." Tax Policy and the Economy, 16:67-114. Claeys Gregory. 2013. Mill and Paternalism. Cambridge: Cambridge University Press. Doris John M. 2002. Lack of Character. Personality and Moral Behavior. New York: Cambridge University Press. Dworkin Gerald. 1971. "Paternalism." In Philosophy of Law, edited by Joel Feinberg and Hyman Gross, Belmont: Wadsworth Publishing Company, 209-219. Dworkin Gerald. 1972. "Paternalism." The Monist, 1:64-84. Dworkin, R.M. 1978. Taking Rights Seriously. Harvard: Harvard University Press. Economist The. 2006. "The new paternalism." The Economist, 6th April. Feinberg Joel. 1983. "Legal Paternalism." In Paternalism, edited by Rolf Sartorius, 3-18. Minneapolis: University of Minnesota Press. Feinberg Joel. 1986. Harm to Self: The Moral Limits of the Criminal Law. New York: Oxford University Press. Feinberg Joel. 1989. The Moral Limits of the Criminal Law: Volume 3: Harm to Self. New York: Oxford University Press. Gert Bernard and Culver Charles M. 1976. "Paternalistic Behavior." Philosophy and Public Affairs, 6 (1):45-57. Holt Jim. 2006. "The New, Soft Paternalism " The New York Times, 3rd December. Isen A.M. and Levin P.F. 1972. "Effect of Feeling Good on Helping: Cookies and Kindness." Journal of Personality and Social Psychology, 21:384-88. Jolls Christine, Sunstein Cass R., and Thaler Richard. 1998. "A Behavioral Approach to Law and Economics." Stanford Law Review, 50 (5):1471-1550. Kahneman Daniel. 2011. Thinking, fast and slow. London: Allen Lane. Kleinig J. 1983. Paternalism. Manchester: Manchester University Press. Lichtenstein S., and P. Slovic. 2006. The Construction of Preference. Cambridge: Cambridge University Press. Loewenstein G, Brennan T. Volpp K. G. 2007. "ASymmetric paternalism to improve health behaviors." Journal of the American Medical Association, 298 (20):2415-2417. Loewenstein George .F. and Haisley Emily. 2008. "The Economist as Therapist: Methodological Ramifications of 'Light' Paternalism." In The Foundations of Positive and Normative Economics. A Handbook, edited by Andrew Caplin and Andrew Schotter, 210-245. New York: Oxford University Press. Loewenstein George, and Adler Daniel. 1995. "A Bias in the Prediction of Tastes." The Economic Journal, 105 (431):929-937. Madrian Brigitte C. and Shea Dennis F.. 2001. "The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior." The Quarterly Journal of Economics, 116 (4):11491187. Mead L.M. 1997. The New Paternalism: Supervisory Approaches to Poverty. Washington: Brookings Inst Press. 19 Mill John Stuart. 1989. On Liberty and other writings. Edited by Raymond Geuss and Quentin Skinner, Cambridge Texts in the History of Political Thought. Cambridge: Cambridge University Press. Mlodinow Leonard. 2009. The Drunkard's Walk: How Randomness Rules Our Lives. London: Penguin Books. Nagel Thomas. 1970. The Possibility of Altruism. Princeton: Princeton University Press. Rawls John. 1971. A Theory of Justice. Cambridge, Massachusetts: Harvard University Press. Rizzo Mario J. 2009. "Little Brother Is Watching You: New Paternalism on the Slippery Slopes." Arizona Law Review, 51 (3):685-739. Sartorius R.E. 1983. Paternalism. Minneapolis: University of Minnesota Press. Shiffrin Seana Valentine. 2000. "Paternalism, Unconscionability Doctrine, and Accomodation." Philosophy & Public Affairs, 29 (3):205-50. Sugden Robert. 2008. "Why incoherent preferences do not justify paternalism." Constitutional Political Economy, 19 (3):226-248. Sunstein Cass R. and Thaler Richard H.. 2003. "Libertarian Paternalism Is Not an Oxymoron." The University of Chicago Law Review, 70 (4):1159-1202. Thaler Richard H. 1980. "Toward a positive theory of consumer choice." Journal of Economic Behavior & Organization, 1 (1):39-60. Thaler Richard H. and Sunstein Cass R.. 2003. "Libertarian Paternalism." The American Economic Review, 93 (2):175-179. Thaler Richard H. and Sunstein Cass R.. 2008. Nudge. Improving Decisions About Health, Wealth, and Happiness. New Haven & London: Yale University Press. Tversky Amos and Kahneman Daniel. 1981. "The Framing of Decisions and the Psychology of Choice." Science, 211 (4481):453-458. Tversky Amos and Kahneman Daniel. 1986. "Rational Choice and the Framing of Decisions." The Journal of Business, 59 (4):S251-S278. VanDeVeer D. 1986. Paternalistic Intervention: The Moral Bounds of Benevolence. Princeton: Princeton University Press. Wikler Daniel. 1979. "Paternalism and the Mildly Retarded." Philosophy and Public Affairs, 8 (4). 20