Thus, our results show that the notion of strong reciprocity extends to the sanctioning behavior of ''unaffected'' third parties. In addition, these experiments suggest that thirdparty punishment games are powerful tools for studying the characteristics and the content of social norms.

QUESTION: What psychological mechanisms drive third-party punishment absent personal stakes? Is it empathy, moral outrage, or evolved group-selection pressure?

Further experiments indicate that second parties, whose economic payoff is reduced by the norm violation, punish the violation much more strongly than do third parties.

We hypothesize that egalitarian distribution norms and cooperation norms apply in our experiments, and that third parties, whose economic payoff is unaffected by the norm violation, may be willing to enforce these norms although the enforcement is costly for them.

DEFINE: Defines 'altruistic third-party punishment' — where an unaffected bystander pays to enforce norms.

Almost two-thirds of the third parties indeed punished the violation of the distribution norm and their punishment increased the more the norm was violated. Likewise, up to roughly 60% of the third parties punished violations of the cooperation norm. Thus, our results show that the notion of strong reciprocity extends to the sanctioning behavior of ''unaffected'' third parties.

XREF: Connects to strong reciprocity literature and evolutionary explanations of punishment behavior in economic games.

No human societies exist without social norms, that is, without normative standards of behavior that are enforced by informal social sanctions. In fact, the ability to develop and enforce social norms is probably one of the distinguishing characteristics of the human species.

DEFINE: Clarifies what social norms are: normative behavioral standards enforced by informal sanctions. Builds on: "Unaffected bystanders punish norm violators even without personal stakes"

Nevertheless, social norms are still poorly understood. Despite some recent progress (Hechter & Opp, 2001), we still know very little about how they are formed, the forces determining their content, how and why they change, their cognitive and emotional underpinnings, how they relate to values, how they shape our perceptions of justice and its violations, and how they are shaped by and shape our neuropsychological architecture. In short, social norms are one of the big unsolved problems in the behavioral sciences.

XREF: Connects to broader sociological theory about norms, institutions, and collective behavior — relevant context for anyone studying social coordination or norm change models.

We introduce, in particular, a third party into the dictator game (DG) and the prisoners' dilemma (PD) game. The third party observes the actions of the players in the DG and the PD and can then punish them. Punishment is, however, costly for the third party so a selfish third party will never punish. This design is motivated by the idea that social norms apply in both games: a norm concerning distributional fairness in the DG and a cooperation norm in the PD. The notion of strong reciprocity (Fehr & Fischbacher, 2003; Fehr, Fischbacher, & Ga¨chter, 2002; Gintis, Bowles, Boyd, & Fehr, 2003) implies that the third parties should be willing to punish the violation of these norms. Thus, if we observe punishment of norm violations, we have evidence that strong reciprocity is also relevant for behavior of third parties.

DEFINE: Illustrates strong reciprocity, where third parties punish norm violations despite personal cost. Builds on: "Unaffected bystanders punish norm violators even without personal stakes"

Whereas rewards or sanctions by second parties can often be rationalized, or are at least likely to be colored, by egocentric, ''nonnormative'' motives, the rewards and sanctions of third parties reveal the truly normative standards of behavior. For instance, norm adherence may trigger feelings of gratitude if second parties benefit from the norm adherence in economic terms and this may, in turn, induce second parties to reward those who obey the norm. In this case, the rewarding by the second party cannot be taken as unambiguous evidence for an appreciated behavioral standard. Likewise, if norm violation directly harms someone, his impulse is to retaliate, but retaliation may have nothing to do with his appreciation of behavioral standards. The study of third-party rewards and sanctions can clarify these confounding factors.

DEFINE: Introduces the meaningful distinction between second-party and third-party reward/sanction as evidence for norm internalization.

A potential disadvantage of this method, however, is that it may reduce the impact of emotions: C may, for example, experience stronger emotions when reacting to an actual violation of a fairness norm than when contemplating what he would do in case of such a violation. It is, however, an open question whether the strategy method actually leads to different response probabilities than when subjects respond only to others' actual choices.

QUESTION: Whether the strategy method actually shifts response probabilities remains empirically unsettled — worth checking for experimental evidence.

However, a recent meta-study of Camerer and Hogarth (1999) indicates that the modal effect of stake size on mean experimental outcomes is zero (though variance is usually reduced by higher payment). This coincides with the results of a similar study by Smith and Walker (1993). Moreover, Brandts and Charness (2000) as well as Cason and Mui (1998) report evidence indicating that the strategy method does not induce different behaviors.

XREF: Relevant to experimental design choices in behavioral economics; connects to methodological debates about incentive magnitude and elicitation methods.

However, we hypothesized that the salient distribution norm in the DG is for A to give 50 points to B; since subjects played the game anonymously and were randomly allocated to their roles and, hence, their endowments, there is no good reason why A should end up with more money than B, making the equality norm salient. … In addition, by asking B what punishment B expects C to impose on A, we receive information on the extent to which directly affected parties expect third-party norm enforcement, and how accurate their expectations are. This information is important because the impact of social norms on behavior should increase the more people believe in the presence of third-party norm enforcement.

QUESTION: The study frames equality as the "salient" norm when endowments are randomly assigned. Interesting question: how differently do subjects behave when endowments are earned rather than random?

The actual behavior of third parties disconfirms the hypothesis that they care only about their own economic payoffs and thus will never punish (Fig. 1). Most third parties punished dictators who transferred less than half their endowment, and the majority of recipients expected them to do so. At each transfer level below 50, roughly 60% (n = 22) of players C chose to punish the dictator A, and with the exception of transfer level 40, the proportion of recipients B who expected C to punish was higher than the proportion who actually did so.

Builds on: "Unaffected bystanders punish norm violators even without personal stakes"

Fig. 2 indicates that punishment and expectations thereof increased in proportion to the amount by which dictators' transfers fell short of 50%. The average punishment imposed when A gave nothing was 14 deduction points, that is, reducing A's income by 42 points, and sanctioning declined monotonically to near zero as transfers reached half of the endowment.

DEFINE: Clarifies how punishment was measured in deduction points and income reduction, making the experimental scale concrete.

The models of Falk and Fischbacher (1999) and Fehr and Schmidt (1999) also predict the existence of third-party punishment. The Fehr and Smith model is based on the assumption that there are players who are willing to pay to reduce the differences between their own payoffs and those of other. In the F&F model, players are willing to pay because they view inequalities as unfair; inequality is the trigger of punishment but, in contrast to the Fehr and Smith model, it is not the aim of the players to establish equality.

DEFINE: Distinguishes Fehr-Schmidt's inequality-aversion model from Falk-Fischbacher's fairness-based model.

This contrasts with the models of Falk and Fischbacher (1999) and Fehr and Schmidt (1999) because the payoff differences between the third party and the other players matter in these models. Thus, the third party may well punish the defector because the defector earns more than the third party. However, both models have difficulties in explaining the fact that mutual defection is also punished, because the payoff vector in this case is (25, 25, 40) before the punishment decision of C. Since Player C is better off than both A and B, Player C should never punish in this situation, but in fact, 20.8% of the third parties punished, albeit at a rather low level. The model of Levine (1998) is again consistent with the existence of third-party punishment of defectors because defection may be taken as a signal that the defector is a greedy subject.

XREF: Connects to the broader third-party punishment literature in behavioral economics and game theory, where models of Fehr-Schmidt and Falk-Fischbacher assume payoff comparisons drive punishment.

In the DG, second-party punishment means that the recipient, Player B, has the option of punishing the dictator. We developed the following design to compare the relative strength of secondand third-party punishment. At the beginning of the experiment, subjects were randomly assigned either the role of the dictator (Player A) or that of the recipient (Player B). Then we formed groups of two players with each group comprising one Player A and one Player B. The players in these groups then participated in a second-party punishment (SP) condition and in a third-party punishment (TP) condition according to the design described below. The sequence of the two conditions was balanced to control for order effects.

DEFINE: Clarifies the DG variant where the recipient can punish the dictator, distinguishing it from third-party punishment.

Dictators faced severe sanctions in both the secondand third-party conditions, but second-party sanctions for transfers below the egalitarian level were considerably stronger than those by third parties, with the effect that low transfers were profitable for dictators in the TP condition but not in the SP condition. The figure shows that second parties punished more than third parties for all transfer levels below 50, while punishment was generally very low and similar across conditions for transfer levels above 50.

XREF: Relates to the broader literature on costly punishment and third-party enforcement of social norms in experimental economics. Builds on: "Dictator games compare second- and third-party punishment strength."

5 Some dictators seem to have anticipated this difference in punishment across treatments. While the modal transfer level is zero in the TP condition, the modal transfer level is 50 in the SP condition. A Wilcoxon signed rank test for matched pairs shows, however, that—despite the shift in the modal offer—the average offer is not different across conditions. This suggests that, to have an impact on the dictators’ behavior, dictators have to experience that low transfers do not pay in the SP condition.

A comparison of coefficients in the two conditions shows that Dneg had a stronger impact in the SP condition. To assess whether this difference was significant, we ran a regression with the data from both conditions (see column three in Table 3). We added a dummy for the TP condition in this regression, and interacted this dummy with Dneg and Dpos. The regression shows that the TP dummy is insignificant, suggesting that the punishment level was not significantly different across conditions at the egalitarian transfer. The coefficient for the interaction term DnegTP Dummy is significantly negative, however, indicating that punishment was less severe in the TP condition for transfer levels below 50.

QUESTION: Why does audience presence blunt punishment for below-egalitarian transfers? Could be reputational concerns, diffusion of responsibility, or strategic restraint.

In the SP condition, 26% never punished and 39% only punished transfers below 50. This is reversed in the TP condition, indicating that the lower average sanctions for transfers below 50, relative to the SP condition, were also the result of a smaller number of punishers.

XREF: Relates to behavioral economics studies on third-party punishment and altruistic punishment norms.

Since we know already that the model by Bolton and Ockenfels (2000) and the pure reciprocity models fail to capture third-party punishment, we concentrate on the other models. The model by Levine predicts no difference between secondand thirdparty punishment, because a given low transfer to the recipient reveals the dictator's selfish or spiteful preferences regardless of whether the recipient (i.e., the second party) or the third party can punish. Thus, Levine's model cannot explain the treatment differences. This contrasts with the Fehr–Schmidt and Falk–Fischbacher models, which predict that third parties will punish less than second parties.

XREF: Connects to the experimental approach of comparing second-party and third-party punishment treatments to discriminate between competing theoretical models.

There is one aspect of the data that neither the Fehr–Schmidt nor Falk–Fischbacher model predicts satisfactorily. Recall that third-party punishment was not significantly affected by the sum received from one’s own dictator (see regression 2 in Table 3). This contradicts both models because the more a third party receives from his or her own dictator, the less reason there is for him or her to punish the dictator in the other group, since the income difference between the third party and the dictator in the other group becomes smaller if one’s own dictator transfers more. The insignificant impact of one’s own dictator’s transfer on thirdparty punishment is consistent with Levine’s model, however, because the behavior of one’s own dictator does not affect inferences about the dictator’s preferences in the other group.

XREF: Connects to behavioral economics models of fairness, reciprocity, and third-party punishment; could relate to earlier reading on Dictator Game results.

The numbers in Tables 5 and 6 imply that defection was not profitable for the defector in the SP condition: the defector's average income was reduced by 38.4 = 25.2 points, whereas the gain from defection was only 10, generating a net loss of 15.2 points. Therefore, cooperation was the better choice from a purely monetary viewpoint.

Our previous emphasis has been on how much potential punishers are willing to pay to punish defectors relative to cooperators. Another question is whether the willingness to incur costs for punishing defectors depends significantly on whether the punisher himself cooperated or defected. This question is particularly interesting in the TP condition because a cooperator has no direct reason to feel exploited by the defection of an outside group member in this condition. However, those subjects who cooperate themselves are perhaps better able to empathize with cooperators in other groups who had to face the defection of their PD partner.

XREF: Connects to social identity theory and empathy-based moral motivation in punishment research.

Thus, cooperators punished defectors much more in the SP condition than in the TP condition, whereas defectors punish defectors to roughly the same extent across conditions.

Influential social scientists (Elster, 1989; Frank, 1988; Hirshleifer, 1987) have argued that the sanctions that enforce social norms are based on strong emotions, and that emotions are the drivers of norm enforcement decisions. Moreover, Elster (1989) argued that being the object of negative emotions such as anger causes a large disutility on its own, independent of any material losses.

For brevity, we do not report our questionnaire results (interested readers may consult Fehr & Fischbacher, 2004), but they are consistent with the view that emotions cause the sanctioning decisions that enforce social norms, because the pattern of third-party punishment and the emotional pattern fit together nicely. However, the results do not prove that emotions cause the sanctioning decisions. We are aware that self-reported or predicted emotions need to be treated with caution. It is possible that self-reported emotions deviate systematically from the emotions that subjects actually would experience if they were in a particular scenario. However, our questionnaire results suggest at least interesting hypotheses regarding the role of emotions in norm enforcement.

QUESTION: Self-reported emotion data is not definitive proof; worth exploring whether experimental methods can establish causation between emotion and sanctioning.

In this paper, we studied the enforcement mechanisms behind social norms, finding that a large percentage of subjects are willing to enforce distribution and cooperation norms even though they incur costs and reap no economic benefit from their sanctions and even though they have not been directly harmed by the norm violation. Thus, third-party sanctions provide a further important example for the notion of strong reciprocity (Fehr & Fischbacher 2003; Fehr et al., 2002; Gintis et al., 2003). Our questionnaire results are consistent with the view that third-party sanctions are driven by negative emotions and negative fairness judgments towards norm violators.

XREF: Connects to strong reciprocity literature — Fehr, Gintis. Links to known behavioral economics of altruistic punishment. QUESTION: Do third-party sanctions scale beyond lab settings, and what cultural variation exists in willingness to punish? Builds on: "Unaffected bystanders punish norm violators even without personal stakes"

We also found that sanctions by second parties directly harmed were much stronger than third-party sanctions, indeed strong enough to make norm violations unprofitable, whereas the sanctions of a single third party were not. Thus, in the context of our experiment, more than one third party is needed to enforce the norm. However, this condition is probably met frequently in real life. Therefore, taken together, our results suggest that altruistic third-party sanctions are likely to be powerful enforcers of social norms.

Builds on: "Second parties punish norm violations more severely than third parties."