On the duty to give (and not to take): An experiment on moralistic punishment


Organizations aim to influence—via their internal guidelines and corporate culture—how unfair treatment of other stakeholders is perceived and condemned by employees. To understand how different frames and forms of publicity influence moralistic punishment, that is, the willingness of employees to take costs in order to foster norm compliance, we employ a modified version of a dictator game. In our dictator game, a bystander observes a dictator’s behavior towards a recipient and can punish the dictator. We vary how the dictator’s action is framed (either as giving money to the recipient or taking money from the recipient) and whether or not the recipient, as a victim of unfair behavior, is informed about the punishment. Our results suggest that bystanders are more likely to punish dictators when their action is framed as giving rather than taking, although both lead to the same consequences. When bystanders cannot inform recipients about their punishment, less punishment can be observed. On average, dictators partially anticipate this effect and behave more generously when recipients are informed about the bystanders’ punishment.

This is a preview of subscription content, log in to check access.

Fig. 1


  1. 1.

    The literature has different terms for this phenomenon: It has been called third-party punishment (Lergetporer et al. 2014), altruistic punishment (Fehr and Gächter 2002), or moralistic punishment (Kurzban et al. 2007). Since there seems to be no consensus on the use of one of these terms, we will stick to the term moralistic punishment throughout the paper.

  2. 2.

    In fact, companies with severe scandals of non-compliance and unethical behavior (e.g., Volkswagen, Enron) had an institutionalized whistle-blowing and compliance management system in place.

  3. 3.

    There is a decent literature on give and take framing in other economic games, such as the public-goods game (for an overview, see Gächter et al. 2016). Public-goods experiments have the characteristic that framing can affect beliefs and actions at the same time. Recently, there has been a debate on how to interpret evidence from these studies (Cason and Plott 2014). The reader should note that this confound is not present in the dictator game with moralistic punishment in the paper at hand. Thus, our experiment delivers cleaner evidence on how moral transgressions are punished differently when framed differently.

  4. 4.

    Consequently, the dictator herself could also be named as part of the audience. These self-image considerations are not the focus of this paper, but there is evidence that these motivations have an impact on pro-sociality (e.g., Grossman 2015; Grossman and van der Weele 2016).

  5. 5.

    To avoid any effects from these role descriptions in the experiment, the different roles are called player A (dictator), player B (bystander), and player C (recipient).

  6. 6.

    The original instructions were in German. For these, we used the term “geben” for the Give frame and “nehmen” for the Take frame. We restrict the choice set of the dictator mainly for two reasons. Previous studies on dictator games observed that dictators can be classified in three different types: Completely selfish dictators, slightly selfish dictators, and dictators who share the endowment equally. We rule out the possibility of transfers exceeding 50% of the endowment, because there is ample evidence that only a very small fraction give more (Engel 2011).

  7. 7.

    The original instructions were in German. An English translation can be found in the Appendix.

  8. 8.

    Obviously, we are not able to discard objections that the strategy method might influence behavior in our experiment. However, a meta-study by Brandts and Charness (2011) demonstrates that the likelihood of the strategy method influencing behavior in economic experiments is rather small.

  9. 9.

    It has to be noted that, in order to prevent the recipient from getting to know the bystander’s punishment choice, we decided to run separate sessions with VictimInfo and VictimNoInfo treatments. However, in every session—if enough subjects showed up—we balanced the appearance of Give and Take treatments.

  10. 10.

    The literature has, however, produced different models of social preferences: The model of Bolton and Ockenfels (2000) assumes that a bystander’s utility is influenced by his own material payoff and his relative share of the total payoffs. If the bystander’s payoff is lower than the relative share, she experiences a disutility. Irrespective of what the dictator leaves for the recipient in our setup, the initial share of the bystander is always one-third of the total payoffs (50 is endowment of the bystander/150 is the sum of all endowments). Thus, assuming this type of preferences, we would not expect any punishment to occur. In addition, models of pure reciprocity (Dufwenberg and Kirchsteiger 2004; Rabin 1993), which assume that the bystander’s punishment is driven by negative intentions, do not predict any altruistic punishment because the dictator’s action does not affect the bystander directly.

  11. 11.

    Most recently, Schwerter (2016) presents a study showing that risk-taking is affected by social comparisons. The results indicate that individuals take more risks when they are provided with the information that peers earn relatively more.

  12. 12.

    In fact, there is a lively debate on what constitutes the reference point in decisions under risk. While some scholars argue that the status quo might be one candidate, others assume that individuals’ expectations can also represent a reference point (Kőszegi and Rabin 2006).

  13. 13.

    Social psychologists usually refer to this behavior as “self-representation” or “self-image” (see, e.g., Kurzban and Aktipis 2006).

  14. 14.

    It can be argued that punishment of dictators who leave the recipient with 50 taler is spiteful or motivated by concerns other than morality. Some papers have looked more deeply into the motives for spiteful punishment (Abbink and Sadrieh 2009). Across all treatments, we observe 11 (of 92) bystanders punishing dictators, leaving the recipients with 50 taler. For this reason, we ran separate analyses. We calculate the average level of punishment for the dictators’ decisions where he is better off than the recipient, and come to similar results.

  15. 15.

    Additional evidence of these punishment patterns is displayed in the GLS regression models in Table 5 in the Appendix.

  16. 16.

    We give a detailed overview in Table 3 in the Appendix.

  17. 17.

    In a comparable experiment by Fehr and Fischbacher (2004), “At each transfer level below 50 (taler) roughly 60% (n \(=\) 22) of players C [bystanders] choose to punish the dictator” (p. 68). The reader should note that, in our comparable treatment (Give-VictimInfo), we observe 70% (for \(\pi _{R}=0\)) or 52% (for \(\pi _{R}=20\)) of bystanders who punish. Thus, our figures are quite comparable.

  18. 18.

    An analysis of recipients’ beliefs about the other payoffs can be found in Table 4 in the Appendix.

  19. 19.

    An analysis of dictators’ beliefs, although they are confounded by prior decisions, tends to underline this observation but remains insignificant. For the most selfish action in VictimInfo, more dictators think that bystanders will punish in Give (.70) than in Take (.65) (p \(=\) .753, \(\chi ^{2}\)-test). In VictimNoInfo in Give, a fraction of .61 thinks bystanders will punish, whilst this figure is .52 for Take (p \(=\) .552, \(\chi ^{2}\)-test).

  20. 20.

    For a review of papers on framing and public-goods experiments, see Cartwright (2016).

  21. 21.

    The average fairness ratings for the equal split for dictators are 6.61 in Give and 5.96 in Take; for bystanders, 6.59 in Give and 5.87 in Take; for recipients, the ratings are 6.63 in Give and 5.76 in Take. Note, that we pool the data from both publicity conditions.


  1. Abbink K, Sadrieh K (2009) The pleasure of being nasty. Econ Lett 105:306–308

    Article  Google Scholar 

  2. Almenberg J, Dreber A, Coren LA, Rand DG (2011) Third party reward and punishment: group size, efficiency and public goods. In: Palmetti NM, Russo JP (eds) Psychology and punishment. Nova Science Publishers, New York, pp 73–92

    Google Scholar 

  3. Andreoni J, Bernheim BD (2009) Social image and the 50–50 norm: a theoretical and experimental analysis of audience effects. Econometrica 77(5):1607–1636

    Article  Google Scholar 

  4. Balafoutas L, Grechenig K, Nikiforakis N (2014) Third-party punishment and counter-punishment in one-shot interactions. Econ Lett 122(2):308–310

    Article  Google Scholar 

  5. Bartling B, Engl F, Weber RA (2014) Does willful ignorance deflect punishment?—an experimental study. Eur Econ Rev 70:512–524

    Article  Google Scholar 

  6. Bénabou R, Tirole J (2002) Self-confidence and personal motivation. Q J Econ 117(3):871–915

    Article  Google Scholar 

  7. Bénabou R, Tirole J (2004) Willpower and personal rules. J Polit Econ 112(4):848–886

    Article  Google Scholar 

  8. Bénabou R, Tirole J (2006) Incentives and prosocial behavior. Am Econ Rev 96(5):1652–1678

    Article  Google Scholar 

  9. Bernhard H, Fehr E, Fischbacher U (2006) Group affiliation and altruistic norm enforcement. Am Econ Rev 96(2):217–221

    Article  Google Scholar 

  10. Bernhard H, Fischbacher U, Fehr E (2006) Parochial altruism in humans. Nature 442(7105):912–915

    Article  Google Scholar 

  11. Bolton GE, Ockenfels A (2000) ERC: a theory of equity, reciprocity, and competition. Am Econ Rev 90(1):166–193

    Article  Google Scholar 

  12. Brandts J, Charness G (2011) The strategy versus the direct-response method: a first survey of experimental comparisons. Exp Econ 14(3):375–398

    Article  Google Scholar 

  13. Cartwright E (2016) A comment on framing effects in linear public good games. J Econ Sci Assoc 2(1):73–84

    Article  Google Scholar 

  14. Cason TN, Plott CR (2014) Misconceptions and game form recognition: challenges to theories of revealed preference and framing. J Polit Econ 122(6):1235–1270

    Article  Google Scholar 

  15. Cubitt RP, Drouvelis M, Gächter S, Kabalin R (2011) Moral judgments in social dilemmas: how bad is free riding? J Public Econ 95(3):253–264

    Article  Google Scholar 

  16. DeScioli P, Kurzban R (2013) A solution to the mysteries of morality. Psychol Bull 139(2):1–20

    Article  Google Scholar 

  17. Dufwenberg M, Kirchsteiger G (2004) A theory of sequential reciprocity. Games Econ Behav 47(2):268–298

    Article  Google Scholar 

  18. Engel C (2011) Dictator games: a meta study. Exp Econ 14(4):583–610

    Article  Google Scholar 

  19. Fehr E, Fischbacher U (2004) Third-party punishment and social norms. Evol Hum Behav 25(2):63–87

    Article  Google Scholar 

  20. Fehr E, Gächter S (2002) Altruistic punishment in humans. Nature 415(6868):137–140

    Article  Google Scholar 

  21. Fehr E, Schmidt KM (1999) A theory of fairness, competition, and cooperation. Q J Econ 114(3):817

    Article  Google Scholar 

  22. Fiddick L, Cummins D (2007) Are perceptions of fairness relationship-specific? The case of noblesse oblige. Q J Exp Psychol 60(1):16–31

    Article  Google Scholar 

  23. Fiddick L, Cummins D, Janicki M, Lee S, Erlich N (2013) A cross-cultural study of noblesse oblige in economic decision-making. Hum Nat 24(3):318–335

    Article  Google Scholar 

  24. Fischbacher U (2007) z-tree: Zurich toolbox for ready-made economic experiments. Exp Econ 10(2):171–178

    Article  Google Scholar 

  25. Friedrichsen J, Engelmann D (2013) Who cares for social image? Interactions between intrinsic motivation and social image concerns. CESifo working paper (4514)

  26. Gächter S, Kölle F, Quercia S (2016) Different frames or different games? Comparing give-some and take-some social dilemmas. Working paper

  27. Greiner B (2003) An online recruitment system for economic experiments. In: Kurt Kremer VM (ed) Forschung und wissenschaftliches Rechnen, GWDG Bericht, vol 63. Ges. fur Wiss, Gottingen, pp 79–93

    Google Scholar 

  28. Grossman Z (2015) Self-signaling versus social-signaling in giving. J Econ Behav Org 117:26–39

    Article  Google Scholar 

  29. Grossman Z, van der Weele J (2016) Self-image and willful ignorance in social decisions. J Eur Econ Assoc. doi:10.2139/ssrn.2237496

    Google Scholar 

  30. Hibbert S, Horne S (1996) Giving to charity: questioning the donor decision process. J Consum Market 13(2):4–13

    Article  Google Scholar 

  31. Hoffman E, McCabe K, Shachat K, Smith V (1994) Preferences, property rights, and anonymity in bargaining games. Games Econ Behav 7(3):346–380

    Article  Google Scholar 

  32. Homans GC (1961) Social behavior: its elementary forms. Harcourt, Brace & World, New York

    Google Scholar 

  33. Kahneman D, Tversky A (1979) Prospect theory: an analysis of decision under risk. Econometrica 47:263–291

    Article  Google Scholar 

  34. Kőszegi B, Rabin M (2006) A model of reference-dependent preferences. Q J Econ 121:1133–1165

    Google Scholar 

  35. Kurzban R, Aktipis CA (2006) Modular minds, multiple motives. In: Schaller M, Simpson J, Kenrick D (eds) Evolution and social psychology. Psychosocial Press, Madison, pp 39–53

    Google Scholar 

  36. Kurzban R, DeScioli P, O’Brien E (2007) Audience effects on moralistic punishment. Evol Hum Behav 28(2):75–84

    Article  Google Scholar 

  37. Latané B, Darley JM (1970) The unresponsive bystander: why doesn’t he help?. Appleton-Century Crofts, New York

    Google Scholar 

  38. Lergetporer P, Angerer S, Glätzle-Rützler D, Sutter M (2014) Third-party punishment increases cooperation in children through (misaligned) expectations and conditional cooperation. Proc Natl Acad Sci 111(19):6916–6921

    Article  Google Scholar 

  39. Nettle D, Harper Z, Kidson A, Stone R, Penton-Voak IS, Bateson M (2013) The watching eyes effect in the dictator game: it’s not how much you give, it’s being seen to give something. Evol Hum Behav 34(1):35–40

    Article  Google Scholar 

  40. Rabin M (1993) Incorporating fairness into game theory and economics. Am Econ Rev 83(5):1281–1302

    Google Scholar 

  41. Schram A, Charness G (2015) Inducing social norms in laboratory allocation choices. Manag Sci 61(7):1531–1546

    Article  Google Scholar 

  42. Schwerter F (2016) Social reference points and risk taking. Working paper

  43. Selten R (1967) Die Strategiemethode zur Erforschung des eingeschränkt rationalen Verhaltens im Rahmen eines Oligopolexperiments. J.C.B. Mohr (Paul Siebeck), Tuebingen, pp 136–168

    Google Scholar 

  44. Spranca M, Minsk E, Baron J (1991) Omission and commission in judgment and choice. J Exp Soc Psychol 27(1):76–105

    Article  Google Scholar 

  45. Stanca L (2009) Measuring indirect reciprocity: whose back do we scratch? J Econ Psychol 30(2):190–202

    Article  Google Scholar 

  46. Tversky A, Kahneman D (1981) The framing of decisions and the psychology of choice. Science 211(4481):453–458

    Article  Google Scholar 

  47. Tversky A, Kahneman D (1986) Rational choice and the framing of decisions. J Bus 59(4):S251–S278

    Article  Google Scholar 

Download references


I would like to thank both editors and two anonymous referees for their comments and suggestions. Additionally, I would like to thank Julian Conrads, Gary Charness, Bernd Irlenbusch, Andrew Kinder, Felix Kölle, Tommaso Reggiani, Arne Weiss, Gari Walkowitz, and Daniel Wiesen for their their valuable comments during the development of the research. Anja Bodenschatz, Katharina Peeters, and Lisa Klinger were helpful in providing research assistance. Financial support from the DFG-Forschergruppe Design and Behavior (TP3 Design of Incentives Schemes within Firms) and through the Leibniz-Award to Axel Ockenfels is gratefully acknowledged.

Author information



Corresponding author

Correspondence to Rainer Michael Rilke.



See Tables 3, 4, 5 and 6.

Table 3 Bystander’s frequency of moralistic punishment
Table 4 Recipients’ beliefs about moralistic punishment
Table 5 Predicting moralistic punishment
Table 6 Predicting beliefs about moralisitic punishment

Experimental instructions (translated from German)

Instructions for the experiment

General explanations

Thank you for your participation in the experiment. For showing up at the experiment you receive a benefit of € 2,50. During the experiment you are able to earn additional money. Therefore, it is important to read the instructions carefully (Tables 5, 6).

It is very important that you do not talk to others during the experiment. In addition to that, we ask you not to use your smart phone or mobile phone. A contravention to those rules results in an exclusion of the experiment and all payments.

During the experiment we talk about Taler instead of Euros. Your entire income will be primarily charged in Taler and afterward converted into Euro and then will be paid in addition to the benefit. Thereby, we use the following conversion rate:

1 Taler = 10,1 Euro

You will be randomly assigned to two other group members. During the whole experiment you will exclusively interact with those two participants. Neither you nor they will get to know the identity of the other group members, at no point of the session. There are three different types of participants: A, B and C. At the beginning of the experiment the type of participant you will be will be randomly assigned and it will be shown on your monitor. The whole experiment only takes one round which means that you only have to make a decision once.

Before the participants make their decision, each one will receive an endowment.

  • Participant A receives an endowment of {Take:0}{Give:100} Taler.

  • Participant B receives an endowment of 50 Taler.

  • Participant C receives an endowment of {Take:100}{Give:0} Taler.

Decision of participant A

After every participant got to know whether he will be participant A, B or C, participant A starts. He decides which share of {Take: C‘s endowment he wants to deduct. He has three opportunities: He can either take 100, 80 or 50 Taler from C. The amount of Taler that he takes from C will be deducted from C’s endowment (100 Taler) and added to A’s own endowment (0 Taler).} {Give: his own endowment he wants to give to participant C. He has three opportunities: He can either give 0, 20 or 50 Taler to participant C. The amount of Taler that he will give to participant C will be deducted from A’s own endowment and added to participant C’s endowment}. Afterward, it is participant B’s turn.

Decision of participant B

Participant B can decide whether he wants to deduct Taler from participant A or not. Hereby participant B can decide whether he wants to deduct 0, 3, 6, 9, 12, 15, 18, 21, 24... or 99 Taler from participant A. Therefore, the deduction of Taler is only possible in steps of 3. The costs for the deduction of 3 Taler for participant B are 1 Taler, for 6 Taler the costs are 2 Taler and so forth.

  • Example 1: If participant B wants to deduct 15 Taler from participant A, participant B’s endowment will be reduced by 5 Taler and participant A’s payoff will be reduced by 15 Taler.

  • Example 2: If participant B wants to deduct 48 Taler from participant A, participant B’s endowment will be reduced by 16 Taler and participant A’s payoff will be reduced by 48 Taler.

Please note that it is only possible for participant B to deduct a maximum of Taler so that participant A receives at least 0 Taler at the end of the experiment. Hence, experiments’ participants cannot make any losses. Participant B makes his decision before getting informed about participant A’s actual decision. Thus, participant B makes a decision how many Taler he would potentially deduct from participant A for each possible amount {Take: that participant A could take from participant C} {Give: that participant A could give to participant C} Participant B also makes a decision, whether he would like to inform participant C about his decision. After participant B has made his decisions about the Taler deduction for all of participant A’s three possible decisions, these will be compared to the actual decisions made and the payment resulting from the one actual decision will be determined. The table shows all possible decisions of Participant A as well as an example of the decision monitor of participant B:

Decision Participant A {Take: takes from}/{Give: gives to} participant C Participant C receives Participant A receives How many taler do you want to deduct from participant A?
1 100/0 0 100  
2 80/20 20 80  
3 50/50 50 50  
  1. In the left part of the table you can see all of the three possible decisions participant A can make and the resulting payoffs for C and A. Participant B can fill in the amount of Taler he wants to deduct from participant A for each possible decision in the right column

Decision of participant C

Participant C cannot make any decisions in this experiment. {Take: He receives his endowment of 100 Taler minus the amount of Taler participant A has taken from him.} {Give: He receives his endowment of 0 Taler plus the amount of Taler participant A has given to him.}

Who will find out about the participant B’s decision to deduct the Taler?

{VictimInfo: Participant A and C will be told about the participant B’s decision to deduct Taler from participant A. Thus, participant C will find out how participant B has decided at the end of the experiment.} {VictimNoInfo: Participant A will be told about the participant B’s decision to deduct Taler from participant A, however, participant C will not be informed about it. Thus, participant C will at no point of time find out about participant B’s decision.}

How are the payments determined?

Participant A receives his endowment of {Take: 0 Taler plus the amount of Taler that he has taken from participant C} {Give: 100 Taler minus the amount of Taler that he has given participant C} minus the amount of Taler that were deducted by B. Expressed in a formula:

$$\begin{aligned} {\text {Payoff \; participant \; A}} = \{Take{:} {\text {0 \; Taler}} +\}\{{Give}{:} {\text {100 \; Taler}} -\} {\text {Taler \; of \; C - Taler \; deduction \; by \; B}} \end{aligned}$$

Participant B receives his endowment of 50 Taler minus the amount of Taler he had to spend in order to deduct Taler from participant A. Please note the costs of deducting 3 Taler are 1 Taler for participant B etc. Expressed in a formula:

$$\begin{aligned} {\text {Payoff \; participant \; B = 50 \; Taler}} - {\text {(Taler \; deduction/3)}} \end{aligned}$$

Participant C receives his endowment of {Take: 100 Taler minus the amount of Taler participant A has deducted from him} {Give: 0 Taler plus the amount of Taler participant A has given to him} Expressed in a formula:

$$\begin{aligned} {\text {Payoff \; participant \; C =}} \{Take{:} 100 \; {\text {Taler - Taler \; to \; A}}\} \{{Give}{:} {\text {0 \; Taler + Taler \; from \; A}}\}. \end{aligned}$$

Once all participants will have made their decisions, we kindly ask you to fill in a questionnaire. At the end of the experiment, the Taler converted to Euros as described above in addition to the benefit will be disbursed. Do you have any questions? If yes, please raise your hand. We will come to your place. If you do not have any questions, please answer the following comprehension questions.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Rilke, R.M. On the duty to give (and not to take): An experiment on moralistic punishment. J Bus Econ 87, 1129–1150 (2017). https://doi.org/10.1007/s11573-017-0851-y

Download citation


  • Moralistic punishment
  • Compliance
  • Framing
  • Social-image concerns

JEL Classification

  • C91
  • D63
  • D64