Journal of Economic Behavior & Organization 73 (2010) 77–82
Contents lists available at ScienceDirect
Journal of Economic Behavior & Organization journal homepage: www.elsevier.com/locate/jebo
Social preferences aren’t preferences夽 Bart J. Wilson ∗ Economic Science Institute, Chapman University, One University Drive, Orange, CA 92866, United States
a r t i c l e
i n f o
Article history: Received 26 September 2008 Accepted 26 September 2008 JEL classiﬁcation: B4 C9 Keywords: Experimental economics Semantics of economics Rhetoric of economics
a b s t r a c t Experimental economists robustly observe that people in the laboratory regularly make choices that result in lower payoffs for themselves. When faced with this paradox of preferences, economists posit that there must be two meanings of preferences: preferences for the self and preferences for the social. In this paper I argue that this is an example of economists forcing ordinary human behavior to ﬁt their models. The force of my argument is to confute the notion that an individual’s expression of so-called social preferences as an action can be represented as a set of separable and private utilitarian “preferences” within him. © 2009 Elsevier B.V. All rights reserved.
“I’m not very good at [math] problems,” admitted Milo. “What a shame,” sighed the Dodecahedron. “They’re so very useful. Why, did you know that if a beaver two feet long with a tail a foot and half long can build a dam twelve feet high and six feet wide in two days, all you would need to build Boulder Dam is a beaver sixty-eight feet long with a ﬁfty-one-foot tail?”. . . “That’s absurd,” objected Milo, whose head was spinning from all the numbers and questions. “That may be true, [Dodecahedron] acknowledged, “but it’s completely accurate, and as long as the answer is right, who cares if the question is wrong? If you want sense, you’ll have to make it yourself.” “- -Norton Juster, The Phantom Tollbooth” For all his genius, Adam Smith was befuddled by the paradox of value. Why would people at rather high prices buy diamonds, a product so utterly useless to survival, and pay so little for an integral component of life such as water? His resolution to the paradox was to posit that there must be two meanings of “value”: “The one may be called ‘value in use;’ the other, ‘value in exchange.’ The things which have the greatest value in use have frequently little or no value in exchange; and on the contrary, those which have the greatest value in exchange have frequently little or no value in use” (Smith, 1776/1981, p. 44). William Stanley Jevons, Carl Menger, and Léon Walras, however, provided a more parsimonious solution in the late 19th century. Looking past the distracting differences in the usefulness of different kinds of goods, the marginalist revolution shifted the perspective from the value of the total amounts consumed to the marginal value of any given quantity of a
夽 I am thankful to Cary Deck, Kyle Hampton, Erik Kimbrough, Rob Kurzban, Vernon Smith, and Andrew Smyth for comments and conversations that have improved this essay. ∗ Tel.: +1 714 628 2830; fax: +1 714 628 2881. E-mail address: [email protected]
0167-2681/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jebo.2008.09.013
B.J. Wilson / Journal of Economic Behavior & Organization 73 (2010) 77–82
good. With this invented concept of marginal utility and the subjective theory of value to support it, the aspect under which economists came to see the world of exchange had been permanently changed. When economists early in the next century observed people exchanging goods, their rational reconstruction was that individuals had different respective marginal utilities, and the assumption at that point was that prices (somehow) arise to reveal these marginal utilities. Considering that the purpose of this new lens was to dissolve a confusion of how consumers made their purchasing decisions, it was convenient to abstract away where the prices came from. After all, when one thinks of a consumer making a purchase, a retail market quickly springs to mind, and in such a market prices take the form of take-it-or-leave-it offers that appear to be predetermined at the moment the consumer buys a shopping cart of goods. Hence, the problem in consumer theory became how, given an individual’s income and a set of prices, to choose consumption bundles along the budget frontier so that the ratios of the marginal utilities to the corresponding product’s price were (exactly) equal for all goods in the bundle. And herein lies the origin of “preferences” in economics. As Varian (2006) in his intermediate microeconomic text deﬁnes the problem: “If [a] consumer prefers one bundle to another, it means that he or she would choose one over the other, given the opportunity. Thus, the idea of preference is based upon [a] consumer’s behavior” (original emphasis, p. 34). This is a powerful way to view the world and to this day economists productively apply this way of seeing to a plethora of economic phenomena beyond retail purchases (e.g., Becker, 1978). Let’s a take deeper look at this concept of “preferences” According to the Online Etymology Dictionary, the word preference, in the sense of “that which one prefers,” entered the English lexicon in 1864. That’s relatively recent and not too much before the marginalist revolution. In the sense of the nominalization of the verb prefer, “the act of preferring”, preference is recorded in use from 1656. Prefer, on the other hand, is quite old, with its Middle English roots in the 14th century and derived from the even older Latin and Old French roots. From its Latin root, prefer literally means “to carry in front” which, apart from the shift from the literal to the metaphorical, is not that different from the current meaning of “to set before or above” or “to promote.” Examples of ordinary usage includes “to prefer beef to chicken” (Dictionary.com), “prefers coffee to tea” (American Heritage Dictionary), “prefers sports to reading” (Merriam-Webster), and “Some people prefer camping to staying in hotels; we prefer sleeping outside” (WordNet). One cannot help but notice the general form of the usage: “prefer x1 to x2 ”. The contrasting difference between x1 and x2 is a difference of quality or kind. Beef and chicken are both kinds of meats from domesticated animals, but the taste of red meat differs in a qualitative way from white meat. Likewise, coffee and tea differ in their potable ﬂavors; sports and reading are afternoon activities that stimulate the body and mind, respectively; and a night’s rest beneath the stars differs in the comfort and view from slumbering on a mattress in a Hilton hotel. How then do economists use “prefer”? At the prices of pa = $2 per apple and po = $1 per orange, a consumer prefers two apples and two oranges to one apple and four oranges. Why? Because with the choice of one apple and four oranges and a budget of m = $6, the marginal utility per dollar of apples is greater than the marginal utility per dollar of oranges. Notice that the difference between an economist’s x1 and x2 is quantitative, not qualitative. We can subtract x2 = (1 apple, 4 oranges) from x1 = (2, 2) to deﬁne the difference between x1 and x2 as x1 − x2 = (1, −2). Where’s the ﬂavor in that? Perhaps in the higher utility ranking of two apples and two oranges to one apple and four oranges, but that is not a difference in kind.1 In either case the consumer tastes apples and oranges. The usefulness and pleasure of different kinds of goods that distracted Adam Smith has been completely purged from the consumer’s problem. But that wasn’t the question that modern theory of value needed to solve. In characterizing consumer preferences axiomatically Debreu (1959) brought to bear the full power of the logical method to explain the structure and properties of how the consumer in a speciﬁc yet abstract setting chooses in a particularly consistent way. In the early 1990s, however, this consistency of choices began to consistently break down following the invention of game theory and the introduction of the experimental method to economics in the intervening years. Researchers around the globe observed that people in the more controlled conﬁnes of the laboratory would regularly make choices that would result in lower payoffs for themselves when another available alternative would have made them materially better off. Sometimes, this lower payoff was coincident with a lower payoff for another person2 and sometimes it was coincident with a higher payoff for another person.3 Faced with a paradox of preferences, economists posited again that there must be two meanings of preferences: preferences for the self and preferences for the social. Three prominent articles appeal to “social preferences” as the solution to this paradox: • “social preference functions balance a person’s desire to have more money with their desire to reciprocate those who have treated them fairly or unfairly, or to achieve equality” (Camerer, 2003, p. 11); • “formal models of social preferences. . .assume people are self-interested, but are also concerned about the payoffs of others” (Charness and Rabin, 2002, p. 817); • “people exhibit social preferences, which means they are not solely motivated by material self-interest but also care positively or negatively for the material payoffs of relevant reference agents” (Fehr and Fischbacher, 2002, p. C1).
1 Hume (1740/2000) makes this point quite well: “A good composition of music and a bottle of good wine equally produce pleasure; and what is more, their goodness is determin’d merely by the pleasure. But shall we say upon that account, that the wine is harmonious, or the music of a good ﬂavour?” (p. 303). 2 See, e.g., Güth et al. (1982), Forsythe et al. (1994), and Hoffman et al. (1994). 3 See, e.g., Camerer and Weigelt (1988), Fehr et al. (1993), and Berg et al. (1995).
B.J. Wilson / Journal of Economic Behavior & Organization 73 (2010) 77–82
Why is the resolution to augment the use of prefer to include both preferences for the self and preferences for the social? On the one hand, a robust number of people prefer ($25 for self, $15 for another) to ($40, $0),4 but we also have overwhelming evidence in another study that 97 percent of people prefer ($40, $0) to ($40 − y, $y) for $0 < y ≤ $40.5 To make sense of this juxtaposition, we can ask what happened to the participants prior to making these decisions, i.e., we can look to the context of the choices. And if we concede that it matters whether in the latter case that people are making a dictatorial decision under doubly anonymous protocols about money that they have earned and that in the former case the people only get the chance to make their decision after another person foregoes ($10, $10), then the whole of Vernon Smith’s comments on context irrelevance must come home to roost.6 For if it matters what a subject experienced just a couple minutes prior to making a decision, it seems all the more compelling that no matter how neutral we think our instructions may be and how clearly evident the procedures state that the subject will be participating in a one-time interaction, “people [will] search [their] personal data bank [of life] for personal knowledge relevant to the task they face and then choose an action according to their motivation” (Smith, this issue). This is the ﬁrst clue on how to make sense of this puzzle, but it does not explain the impetus for developing the concept of social preferences. To get a handle on that, let’s ﬁrst ask why it is that if someone prefers (2 oranges, 2 apples) to (1, 4) and another person prefers (1, 4) to (2, 2) everything is perfectly clear. Because as part of the particularly consistent way that we assume that the consumer makes her purchasing decision, we presume that the individual is choosing what is in her interest and we look to see what she does. Her behavior, however, only makes sense because we can condition either choice of (1, 4) or (2, 2) on the externally observable parameters of m, pa , and po .7 It does not matter what reasons she may have going through her mind in making her speciﬁc choice. So why is it that when someone prefers ($25, $15) to ($40, $0), we are befuddled? Because the individual appears to be violating a primary axiom of prudence in economics by not choosing what is in her material interest (Smith, this issue).8 Thus to make sense of these failures of payoff dominance, it appears we need to look within the individual to reestablish dominance in our Samuelsonian framework by reintroducing differences in kind to what it means for an individual “to prefer”. Problem solved. The hypothetical individual in a speciﬁc yet abstract setting is again making choices in a particularly consistent way. But is this traditional tool suitable for the question, or is it a 68 foot beaver? As soon as we introduce differences in kind within an individual, we have introduced a new problem not faced before, viz., we are applying third-person analytics to a ﬁrst-person problem of differences in personal, social motives to which we, in fact, do not have access.9 Unlike our consumer of fruit, we are precisely concerned about the deeply personal and social reasons going through the mind of someone who chooses ($25, $15) over ($40, $0). However, in a social system comprised of tacit rules, there are no explicit and externally observable prices in a social interaction on which to condition our observations; the social motive, the symbolic analog of a market price in the social domain, is in the head of the individual. Thus, in attempting to explain the different actions of people via differences within an individual we have introduced Wittgenstein’s fundamental problem of identiﬁcation: separating the act of observing the experience of a social motive from the act of having one. I cannot observe the inner motives of others, and I cannot observe my own motives in the same “proper sense of ‘observe”’ (Wittgenstein, 1988, p. 235). (To observe my own motives I would have to already know what it is I’m looking for.) My ﬁrst task is to argue that social preferences are not purely private and hence separable from the outer observable action. Social motives may seem to be inherently private, knowable only the individual himself, but this dubiously implies that it is impossible for different people to meaningfully talk about social motives. The fact we can talk about our internal and simultaneously social motives would then seem to rebut this.10 Moreover, if an individual’s inner social motives ex hypothesi are inaccessible to others, how can we check the accuracy of an individual’s social motive? To answer this, consider a courtroom. How does a prosecutor build a case for ﬁrst-degree murder? By using the defendant’s observable actions to establish a premeditated motive. In the same way that a prosecutor cannot just assert to the jury the motive of the defendant in making a case for ﬁrst-degree murder, an internal social preference is not separable from an individual’s external actions. When a consumer makes a max-U decision over apples and oranges, as McCloskey (2006) would put it, we don’t care what goes on inside her head and moreover that does not preclude us from making sense of her behavior, whatever she may do (GARP, transitivity, etc., aside). However, when someone chooses a lower payoff for himself, we do care what goes on inside his head because that is the only way to make sense of his behavior. Nevertheless, what goes on inside the head is not separable from the action and that action cannot be extirpated from the body of social practice because it is the social practice that gives meaning to the action in the ﬁrst place. When Jerry Seinfeld receives a label maker at Christmas
See, McCabe and Smith (2000), the ﬁndings of which Cox and Deck (2005) replicate. See Cherry et al. (2002), the ﬁndings of which Oxoby and Spraggon (2008) replicate. 6 For another take on this, see Loewenstein (1999, pp. F29–F31). 7 If we don’t have access to m, pa , and po , then no set of Debreuvian axioms will allow us to describe the consumer’s behavior. 8 McCloskey (2006) develops a system of seven virtues that critiques the standard economic approach of basing faith, hope, love, courage, justice, and temperance as derivative of prudence as utility maximization assumes. 9 To use V. Smith’s term, we have anthropo-theorized our subjects in personal, social interactions with a theory of how people make decisions in impersonal markets. Given that market exchange is a relatively new development in the human career, it would seem that this puts the cart of market goods before the workhorse of reciprocity. 10 For a more detailed discussion of this point, see the application of Wittgenstein’s parable of the beetle in the box to the concept of fairness in Wilson (2008b, Section III). 5
B.J. Wilson / Journal of Economic Behavior & Organization 73 (2010) 77–82
time from Dr. Tim Whatley, what distinguishes it as a cordial gift from a disapprobative “regift”? The physical movement of the label maker is the same but we deem the motives to be clearly different. But how do we discover Dr. Whatley’s motive? We compare his action, rewrapping a Christmas gift that Elaine had given him, against the entire of social custom of what it means to give a Christmas gift. Yes, indeed, it is “the thought that counts”, but the motive is not observable nor separable from what Dr. Whatley does, and what Dr. Whatley does only means something against the backdrop of the whole social practice of gift giving (Wilson, 2008a). If a social motive is not separable from a social action, the next natural inclination is to argue that this claim amounts to a behaviorist story that social motives are in the head and that so-called “pro-social” actions are exactly the same thing.11 There is a palpable difference, however, that distinguishes the two, viz., pro-social actions can be observed whereas social motives cannot be. The rules that guide an individual’s social actions are unformalized and unformalizable; nevertheless, the activity is publicly observable, and every stage of it can be deemed by others to be pro-social or not (Wilson, 2008b, Sections III and IV). In the case of social motives, what can we observe? Knits in the forehead, smiles on the face, etc. To avoid resorting to behaviorism, we may want to respond and say that no, a subject did something else, and that we can identify it as a social motive, but the only way to describe it is in terms of the observable social actions (Wittgenstein, 1988, p. 252). The force of my argument in this essay is to expand upon Vernon Smith’s provocative article in this issue by confuting the notion that the individual’s expression of his social preferences as an action can be represented as a set of separable and private utilitarian “preferences” within him. The methodological implications for “behavioral economics” are nontrivial, the goal for which, as Camerer (2003) explains, is “to ﬁnd parsimonious utility functions, supported by psychological intuition, that are general enough to explain many phenomena in one fell swoop, and also make new predictions” (p. 101). In other words, the approach is to reverse-engineer observable actions as social preferences, separating the motive as independent explanatory input for the outer actions we observe. To put ﬁrst the idea of an inner social preference explaining or “doing” something that preferences for the self do not, gets things the wrong way, for it is only because an individual acts intelligibly within a pattern of social practice that we say some social motive takes place within him (Wilson, 2008a). A social preference is not independent of the individual’s behavior but something that we ascribe to the individual as a way of giving meaning to his action. Having built up a lifetime of experiences interacting with people, we recognize, in the fullest sense of cognizing again, the action of another based upon the publicly observable standards of social practice. The notion of an inner social preference distinguishes between two types of actions, for the self and for others, but in saying that something different went on inside two different individuals, we are underscoring this difference, not pointing to some new-fangled third entity. What do we know about what does occur with social motives and social actions? To answer that let’s consider a Wittgensteinian thought experiment of a Hayekian ﬂavor and imagine we did not have social motives in our heads but stumbled across someone who asserted that he did (Wittgenstein, 1988, p. 251). How would we make sense of that assertion? He may assert that he has social motives, but this provides neither us nor him with incontrovertible evidence on the independent existence of a social motive. We might say that we do not care what goes on inside his head as long as he takes the right action. Or we might agree to call what he does a “pro-social” action, not because he asserts he has a social motive, but because the results of what he does conform to our practice of being social. Thus, if we did decide to say that he could conjure up social motives in his head, this would not be because his assertions were accurate reports of a process which turned out to be identical to our social practice. Rather it would reﬂect the fact that at various points his assertions on social motives as expressed by his actions make connections to our established social practice. If we could not have social motives in our heads and came across someone who said he could, we might ﬁnd it ﬁtting and reasonable to extend our concept of a “pro-social” action to accommodate this new circumstance. Or we might decide to call it a “quasi-pro-social” action or decide not to call it a pro-social action at all. Regardless, returning to a world where we all have social motives, the only criteria of what constitutes a social motive is the sincerity with which the individual asserts he has one, but for a pro-social action we have the entire external backdrop of convention upon which to render a public judgment as to whether it is indeed pro-social or not. To summarize, social motives are expressed as social actions but a ﬁrst-person account of a social motive is not independently deﬁnable as a third-person Samuelsonian process. Preferences are neither selﬁsh nor social nor, as Smith (1985, p. 267) succinctly puts it, the “primeval cause for the phenomena we study”. The only thing that matters is what the individual does within a pattern of social practice. The inkling of an autonomously existing social preference upon which an individual conditions an action can be dropped. And in the same spirit that Smith (this issue) will not go as far as to “claim there is no room for social preferences”, I am not
11 This is what Cox and Deck (2005) implicitly assume when they claim to be “discriminating among motives in the trust game” with a triadic design (p. 629). They compare the rates at which dictators choose ($20, $0) over ($12.50, $7.50) with the rate at which second movers choose ($20, $0) over ($12.50, $7.50) after ﬁrst movers have foregone ($5, $5). When they fail to observe different proportions across the two games (13 out of 17 second movers choose ($20, 0) as do 20 out of 30 dictators), they conclude that “this result is clearly inconsistent with reciprocity being the motivation for the subject choice of Cooperate in the Trust game” (p. 630). That’s only true if the motives in the head are exactly the same thing as the actions. How do they discriminate between two (or more) possible motives in the head of the four second movers who choose ($12.50, $7.50)? By looking at the actions of the 10 dictators who choose ($12.50, $7.50)? We can’t observe their motives either. We may severely discount the possibility that the dictators have reciprocal motives because their game does not ﬁt the typical pattern of social practice in which we reckon that people are reciprocating, but we still don’t observe what their motives are or are not. Nor do we observe what the motives of the second movers are or are not. We do observe what the second movers and the dictators do, and we do observe the setting in which they take their actions, but comparing actions across two games of a triadic design cannot discriminate between the motives because there are no observable criteria to distinguish the inner social motives from the action that an individual takes.
B.J. Wilson / Journal of Economic Behavior & Organization 73 (2010) 77–82
denying, as a behaviorist would, that we have social motives; my aim is simply to undercut the idea that what lies behind this concept is some mysterious mental process of a social preference which somehow corresponds to a “pro-social” action. My major thrust is to undermine an incoherent account of the nature of that social experience and to proffer a new way of seeing these phenomena and the inner social concepts they involve. Instead of seeking an independent conﬁrmation of what the individual does via a social preference, the challenge is to reconstruct the distinctive system of rules that deﬁnes social motives within a pattern of social practice. If it is fruitless to sever social motives from social actions, how should we methodologically approach the economic problem of understanding differences in kind within an individual in personal, social interactions? We ﬁrst must conclude that if we as the experimenters cannot peer into the minds of our subjects to ascertain their social motives, then our subjects face the same uncertainty when interacting in the tasks that we give them. Thus, the problem boils down to one of agreement: Do the subjects agree on the context of the interaction: Do the subjects agree that this situation invokes the social motives of a rule of reciprocity,12 a rule of fairness,13 a rule of just desert,14 a rule of equity,15 a rule of equality,16 or some other rule to guide behavior? By their sociality people are trained in conventions to apply, and not necessarily deliberately so, different rules of behavior depending upon the contextual cues of the interaction. When subjects in the trust game end up at ($25, $15) and not at ($10, $10) or ($40, $0), it is because both the ﬁrst and second movers are in agreement that this interaction prescribes a rule of reciprocity that culminates in a Pareto improvement.17 If the ﬁrst mover passes on ($10, $10) but the second mover chooses ($40, $0) over ($25, $15), then the second mover did not agree with the ﬁrst mover that a rule of reciprocity was applicable and instead applied a rule of just desert. The question is not whether people instantiate preferences for the self or preferences for others. The question is what is it about the social interaction that leads people to agree or disagree on what is the right thing to do when faced this task. Even when faced with exactly the same circumstances, people may have diverse judgments on what to do and hence their social motives may not be aligned. Disagreement is the consequence of ambiguous rules of ascertaining which thread in the weave of social practice serves as the context for this action. Based upon their differing past experiences, there is something about the context of the task that different people differently map into applying a particular rule of behavior.18 When a second mover chooses ($40, $0) over ($25, $15), it is simply inadequate to assert that the participant is expressing preferences for the self instead of the preferences for the social. It should be natural for an economist to ask what was the foregone alternative: Why as a conscious agent with particular intentions did a subject in this context not choose ($25, $15)? If we operate under the tacit working assumption that this person’s money-maximizing behavior is in this particular interaction somehow indicates that this person is a money-maximizing agent in every facet of his or her life, it would be hard to see him as a conscious individual with social motives; we have dehumanized him. The research question is, when under different circumstances would this same subject apply a rule of reciprocity? What would it be about the different circumstances that would strike a chord with this subject to agree that the social context of the interaction is one in which to be reciprocal?19 The hurdle to overcome in accepting this post-game-theoretic approach to understanding human social interactions is to humbly accept that when “[s]eeing life as a weave, this pattern. . .is not always complete and is varied in a multiplicity of ways. But we, in our conceptual world, keep on seeing the same, recurring with variations. That is how our concepts take it” (Wittgenstein, 1980, §672). Rather than isolating social motives as an inner ostension and viewing them as a qualitas occulta of our outer actions, we should treat social motives as concepts inextricably tethered to
See Hoffman et al. (1998, p. 341) and Wilson (2008a). See Wilson (2008b). 14 See Cherry et al. (2002) and Schurter and Wilson (2009). 15 See Hoffman et al. (1998). 16 See supra note 15. 17 Obviously, it is possible that the two people could be in agreement on another rule of social behavior that culminates in these same actions, or that the two individuals have two different rules in their minds’ eyes so that they happen to agree on this outcome. But if we can never observe a social motive distinct from an action, who cares? That is the sense in which I say that the only thing that matters is what the individual does within a pattern of social practice. “And if [we reply] that this is in fact just what we are interested in, that is, we are interested in whether he [has reciprocal motives], then we shall have to draw his attention to the criteria that would demonstrate his capacity [to reciprocate], and on the other hand to the criteria for the ‘inner states’. Even if someone had the particular capacity only when, and only as long as, he had a particular feeling, the feeling would not be the capacity” (Wittgenstein, 1953, II, p. 155). See Wilson (2008a, Section 5) for a more detailed discussion of this point. 18 Deck (2008) and Gillies and Rigdon (2008) are two examples of steps in this direction. Both vary the information available on the payoffs as the participants make decisions in altered forms of the traditional trust game. The former ﬁnds that ﬁrst movers pass to the second mover at a very high rate and that the second movers defect at a very high rate. Rarely do we ﬁnd cases in which ﬁrst movers fail to realize that the second movers disagree with them on the context of the decision task. Finding such disagreement is a remarkable opportunity to explore what it takes for ﬁrst and second movers to ﬁnd agreement. In contrast, Gillies and Rigdon (2008) observe that ﬁrst movers are much better at anticipating that their second mover may not agree with them. The task now is to ﬁnd a set of reconciling treatments. 19 Cox and Deck (2005) provide another example in which a high proportion of ﬁrst and second movers do not agree on the context of the interaction. They vary whether the subjects make single or double anonymous decisions in the trust game and ﬁnd that ﬁrst movers apply a rule of reciprocity at statistical indistinguishable proportions across treatments. Second movers, however, clearly do not; they apply rule of reciprocity much less often when the experimenter does not know what they personally did (only 4 out of 14 individuals with double anonymity versus 17 out of 25 individuals with single anonymity). It would seem that the more public character of the interaction is one aspect of the context that activates a rule of reciprocity in a second mover, but why is that ﬁrst movers fail to anticipate the rule that second movers apply with double anonymity? Is it, as Smith (this issue) calls it, an issue of other people’s money? Or is it because the two movers have not each voluntarily selected each other? 13
B.J. Wilson / Journal of Economic Behavior & Organization 73 (2010) 77–82
outer actions, actions which are woven into a pattern of social practice that unintentionally orders the hurly-burly of human life. The absence of a mathematical regularity in human behavior may tempt us to consider social motives to be hidden and separable, but what really lies behind this inclination is the unpredictability of life. Our social motives are “not for use on a single occasion” (Wittgenstein, 1980, §672, original emphasis); they must be ﬂexible and spontaneous, and hence difﬁcult to pin down, to contend with an ever-ﬂuxional background. Despite the indeﬁniteness of life, humans have an amazing propensity to achieve cooperative outcomes through personal, social interactions and impersonal market transactions. At its core, economics is the study of human agreement to achieve these mutually beneﬁcial improvements. References Becker, G., 1978. The Economic Approach to Human Behavior. University of Chicago Press, Chicago. Berg, J., Dickhaut, J., McCabe, K., 1995. Trust, reciprocity, and social history. Games and Economic Behavior 10, 122–142. Camerer, C., 2003. Behavioral Game Theory: Experiments in Strategic Interaction. Sage, New York. Camerer, C., Weigelt, K., 1988. Experimental tests of a sequential equilibrium reputation model. Econometrica 56, 1–36. Charness, G., Rabin, M., 2002. Understanding social preferences with simple tests. Quarterly Journal of Economics 117, 817–869. Cherry, T.L., Frykblom, P., Shogren, J.F., 2002. Hardnose the dictator. American Economic Review 92, 1218–1221. Cox, J.C., Deck, C.A., 2005. On the nature of reciprocal motives. Economic Inquiry 43, 623–635. Debreu, G., 1959. Theory of Value. Wiley, New York. Deck, C., 2008. An experimental investigation of trust and sequential trade. Department of Economics Working Paper, University of Arkansas. Fehr, E., Fischbacher, U., 2002. Why social preferences matter: the impact of non-selﬁsh motives on competition, cooperation, and incentives. The Economic Journal 112, C1–C33. Fehr, E., Kirchsteiger, G., Riedl, A., 1993. Does fairness prevent market clearing? An experimental investigation. Quarterly Journal of Economics 108, 437–459. Forsythe, R., Horowitz, J., Savin, N.E., Sefton, M., 1994. Fairness in simple bargaining experiments. Games and Economic Behavior 6, 347–369. Gillies, A., Rigdon, M., 2008. Epistemic conditions and social preferences in trust games. Institute for Social Research Working Paper, University of Michigan. Güth, W., Schmittberger, R., Schwarze, B., 1982. An experimental analysis of ultimatum bargaining. Journal of Economic Behavior and Organization 3, 367–388. Hoffman, E., McCabe, K., Shachat, K., Smith, V.L., 1994. Preferences, property rights, and anonymity in bargaining games. Games and Economic Behavior 7, 346–380. Hoffman, E., McCabe, K.A., Smith, V.L., 1998. Behavioral foundations of reciprocity: experimental economics and evolutionary psychology. Economic Inquiry 36, 335–352. Hume, D., 1740/2000. In: Norton, D.F., Norton, M.J. (Eds.), A Treatise of Human Nature. Oxford University Press, New York. Loewenstein, G., 1999. Experimental economics from the vantage-point of behavioural economics. Economic Journal 109, F25–F34. McCabe, K., Smith, V.L., 2000. A comparison of naïve and sophisticated subject behavior with game theoretic predictions. Proceedings of the National Academy of Arts and Sciences 97, 3777–3781. McCloskey, D., 2006. The Bourgeois Virtues: Ethics for an Age of Commerce. University of Chicago Press, Chicago. Oxoby, R.J., Spraggon, J., 2008. Mine and yours: property rights in dictator games. Journal of Economic Behavior and Organization 65, 703–718. Schurter, K., Wilson, B.J., 2009. Justice and fairness in the dictator game. Southern Economic Journal 76, 130–145. Smith, A., 1776/1981. In: Campbell, R.H., Skinner, A.S. (Eds.), An Inquiry into the Nature and Causes of the Wealth of Nations, vol. I. Liberty Fund, Indianapolis. Smith, V.L., 1985. Experimental economics: reply. American Economic Review 75, 265–272. Smith, V.L., 2010. Theory and experiment: what are the questions? Journal of Economic Behavior and Organization 73, 3–15. Varian, H., 2006. Intermediate Economics: A Modern Approach, 7th ed. W.W. Norton & Company, New York. Wilson, B.J., 2008a. Language games of reciprocity. Journal of Economic Behavior and Organization 68, 365–377. Wilson, B.J., 2008b. Contra private fairness. Economic Science Institute Working Paper, Chapman University. Wittgenstein, L., 1953. Philosophical Investigations, Anscombe G.E.M. (Trans.). Blackwell Publishing, Malden, MA. Wittgenstein, L., 1980. In: Von Wright, G.H., Nyman, H. (Eds.), Remarks on the Philosophy of Psychology, Luckhardt, C.G., Aue, M.A.E. (Trans.), vol. II. University of Chicago Press, Chicago. Wittgenstein, L., 1988. In: Geach, P.T. (Ed.), Wittgenstein’s Lectures on Philosophical Psychology 1946–47: Notes by P.T. Geach, K.J. Shah, and A.C. Jackson. University of Chicago Press, Chicago.