تکه ای از کیک؟ اختصاص پاداش به اشخاص ثالث وقتی که انصاف پر هزینه است

Reactions to third-party inequality were investigated in three experiments. In Experiment 1, 52 undergraduates allocated money between themselves and two others. Preferences for equal and unequal distributions were also rated. The results show that people are averse to inequalities between themselves and others, and to inequalities between others. Post-experimental ratings indicate that egocentric equality, third-party equality, and max–min preferences are important motives. The findings were replicated in Experiment 2, where 74 undergraduates allocated compensation for a previously conducted task, and in Experiment 3, where 112 participants rated preferences. In these experiments random determination of rewards to third parties altered participants’ behavior and preferences. The results indicated that random determination decreases the importance of all fairness motives while increasing the importance of monetary payoff. While people still care about economic equality under these conditions, contextual factors, such as perceived responsibility for unfair outcomes, seem to alter the impact of fairness.

Distributive fairness has been addressed by many scholars in the past in research domains such as ethical theory (Rawls, 1971), political theory (Walzer, 1983), and psychology (Adams, 1963, Adams, 1965, Bar-Hillel and Yaari, 1993, Deutsch, 1975, Deutsch, 1985, Lerner, 1987, Lerner, 1991 and Walster et al., 1973). More recent developments in economic theory have similarly acknowledged the role of fairness, where people are assumed not only to care about their own welfare but also the welfare of other people (Bolton and Ockenfels, 2000, Fehr and Schmidt, 1999, Loewenstein et al., 1989 and Rabin, 1993). Here, fairness is usually defined in terms of how equal (or unequal) material distributions are between people or groups of people, with the basic idea that utility is derived from one’s own monetary payoff, whereas disutility is derived from advantageous inequality (receiving more than others) and disadvantageous inequality (receiving less than others).1 Regardless of which perspective a distribution is judged from, the impact of different fairness motives can be expected to vary between situations and across domains, since peoples’ preferences are often context dependent (Bazerman et al., 1992, Garcia et al., 2005 and Hsee et al., 1999).

All three experiments demonstrated an effect of third-party inequality. This indicates that evaluations of fairness are not limited to egocentric comparisons of payoffs but also include how well off other individuals are versus each other. We believe this provides support for the existence of a broader sense of fairness that is evaluated independently of an individual’s own economic position. This notion is relevant in many domains of society, most notably in public policy analysis where fairness is often judged on the basis of how equal welfare distributions are in a society as a whole. While admitting that third-party considerations are driven partly by a concern for the lowest paid individual or group, our data support that people care about economic differences as such. The latter finding is novel in the economic fairness literature and extends the argument proposed by Loewenstein et al. (1989) and others to situations involving more than two people. Third-party fairness concerns also seem to vary between individuals and depend partly on the context in which decisions are made. For example, in Experiment 2, when the decision maker was positioned in a disadvantageous position, as compared with an advantageous position, versus two other individuals, the adverse reaction to any inequality between these individuals diminished. This finding corresponds with that of Fetchenhauer and Huang (2004), and perhaps makes sense intuitively, since people are likely to care about their own relative well-being first and foremost and, as long as they feel disadvantaged, will not react as strongly to inequalities between other people. The competition to achieve outcomes which are at least as good as others may in this sense override the concern for third parties. Yet, Experiment 3 showed effects of third-party differences in both disadvantageous and advantageous positions. We are inclined to believe that these contradictory results are due to efficiency concerns present in Experiment 2, which are likely to override the influence of third-party inequalities when the decision maker is in a disadvantageous position. If so, we have a case for a more general fairness motive that is active in both relative positions. Still, further research is warranted investigating the interactions between various fairness motives and decision contexts. Another important finding from this research is that the fairness motives tested seem to be weaker when the designation of specific payoffs to other individuals was determined by a coin-flip. This is in line with our assumption that mediating events outside the decision makers’ control may diffuse the perceived responsibility for unfair outcomes and in turn provide more room for pursuing self-beneficial goals. It does seem that acts of fairness may not be driven entirely by a genuine concern for other people’s welfare. An interpretation is that these types of behavior are driven by external factors and internal norms instead. Such factors and norms will often make individuals abstain from proposing unfair allocations, but when the consequences of an action are not directly attributable to the decision maker, these factors and norms become less influential. This raises the question of how people would behave in a real-world context where the end-result of a certain action depends on a chain of events over which the individual has very little control, and where it is not obvious who will receive the credit or blame for the consequences that follow. Although fairness motives are expected to be generally influential we should, given our data, expect them to be substantially weakened by many of the circumstances present in real world decisions. This interpretation is supported by the fact that participants who were faced with the coin-flip manipulation also rated the fairness motives as less important and monetary motives as more important, than those who faced pre-determined rewards. The coin-flip manipulation similarly caused satisfaction ratings to decrease for equal-split alternatives and increase for the unequal, self-interested alternatives, while fairness ratings were relatively unaffected. Our interpretation of these findings is that participants can comprehend the degree of unfairness across all manipulations, suggesting that our results are not simply driven by fuzziness, possibly resulting from a less transparent presentation of outcomes in the coin-flip scenario. Should this be the case we would expect perceptions of fairness to change in this condition. Instead it is the personal satisfaction of pursuing self-interested goals that change. In line with our hypothesis, if the perceived responsibility for unfair outcomes increases, as when the links between actions and consequences are more transparent, we would expect the satisfaction of choosing this outcome to be lower because it is more obvious that the individual is acting against assumed moral norms at play. Interestingly, while equal payoff distributions in both the coin-flip and pre-determined groups were judged as considerably more fair than unequal distributions, on the whole unequal distributions yielded a higher satisfaction for the decision maker. This is in line with Ordóñez et al. (2000), who showed that what is considered to be fair may not always correspond with what is most satisfying to the individual, again emphasizing the role of norms in similar situations. Taken together the findings thus cast some doubt on the strength and stability of the fairness motives investigated. Yet, it is important to remember that the coin-flip procedure did not wipe out the inequality effects studied. Even in the face of a random event, people still show concern for third parties. It thus seems as if the ambiguity about who will be favored and disfavored evoked moral licensing for acting selfishly, but only to some degree. At this point it is important to take a look at the basic assumptions of this research. Responses to post-experimental questions paint a somewhat ambiguous picture when it comes to explanations of third-party inequality aversion. It seems driven partly by a concern for the lowest paid individual, partly by payoff differences per se. Although max–min preferences are well established in the literature, we are tempted to argue that comparisons between other individuals are necessary in order to establish who receives the lowest payoff. Other research also indicates that people compare and are influenced by differences between recipients other than the poorest in a distribution. The impact of max–min preferences may also vary depending on other factors. For instance, some situations may evoke concern for victimized individuals while other situations emphasize a more general sense of fairness within groups. Ultimately, it is an empirical question whether comparisons with the individual who receives the lowest payoff carry more or less weight than other comparisons. In our view, it is thus conceptually appropriate to discuss max–min preferences as a special case of a more general form of third-party inequality aversion. A potential problem with responses to post-experimental questions is that participants may engage in post-hoc justifications and adjust their responses in a biased fashion. In this sense the observed pattern of responses may not represent the “truth” about underlying fairness motives at play in the experiments. However, we find it hard to believe that participants would be able to tailor their responses to produce the very clear effects found in both Experiments 1 and 2 on motive, fairness, and satisfaction ratings. Given the observed pattern of responses we find it unlikely that social desirability or some other unknown motive biased these measures in a consistent way. All in all, the research reported here suggests that people, when faced with similar tasks, try to resolve conflicts between competing motives, and which motive that comes to dominate their behavior often depends on the context and payoff structure at hand. It would however, in our view, be premature to search for a single motivational structure that guides all individuals across all situations. Rather, research is warranted that further disentangles the psychological underpinnings of these types of decisions, paying particular attention to the complexity of real-world situations where conflicting motives come into play, and the fact that many individuals may have inconsistent views about inequality. It should finally be noted that this research is based entirely on Swedish samples. In Sweden egalitarianism has traditionally been promoted in the educational system, among other arenas (Tomasson, 1965). As a consequence, equality norms may be central to our participants. Justice norms in Sweden seem however to have changed over the last decades, moving away from equality towards more equity based principles, making today’s Sweden far from unique in this respect among western societies. Recent research has furthermore failed to show any pronounced differences between countries in terms of how important justice norms, such as equality, are to people. Nor has it been shown that attitudes to pay differences vary much across welfare states (e.g., Arts and Gelissen, 2001 and Osberg and Smeeding, 2006). It is ultimately an empirical question whether Swedish samples may differ in terms of fairness from, for instance, American samples. Anyhow, given that this research is focused on examining third-party fairness under varying conditions rather than the degree of fairness per se, we argue that our conclusions are not affected by whether or not the participants were more egalitarian than others.