The benefit of the doubt: willful ignorance and altruistic punishment


Altruistic punishment is often thought to be a major enforcement mechanism of social norms. I present experimental results from a modified version of the dictator game with third-party punishment, in which third parties can remain ignorant about the choice of the dictator. I find that a substantial fraction of subjects choose not to reveal the dictator’s choice and not to punish the dictator. I show that this behavior is in line with the social norms that prevail in a situation of initial ignorance. Remaining ignorant and choosing not to punish is not inappropriate. As a result, altruistic punishment is significantly lower when the dictator’s choice is initially hidden. The decrease in altruistic punishment leads to more selfish dictator behavior only if dictators are explicitly informed about the effect of willful ignorance on punishment rates. Hence, in scenarios in which third parties can ignore information and dictators know what this implies, third-party punishment may only ineffectively enforce social norms.

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

Fig. 1
Fig. 2
Fig. 3


  1. 1.

    In this vein, the findings on altruistic punishment have shaped research in various fields such as economics, biology, anthropology, psychology, and neuroscience.

  2. 2.

    Third-party punishment can also be consistent with inequity aversion (Fehr and Schmidt 1999) or spite (Levine 1998), see also Leibbrandt and López-Pérez (2012).

  3. 3.

    In more applied settings, Kandul (2016) and Kajackaite (2015) also document that subjects remain willfully ignorant in order to make selfish decisions.

  4. 4.

    These authors also show that, in contrast, second-party punishment is substantially more resolute.

  5. 5.

    One message learned from Kriss et al. (2016) is that the rate at which norm violations are altruistically punished will be higher if third parties cannot get out of their punishment decision. This study shows that this rate is higher when third parties are exogenously informed about norm violations.

  6. 6.

    In Kriss et al. (2016) subjects signal once to themselves that they are willing to punish norm violations.

  7. 7.

    While providing clean evidence of an effect of moral wiggle room on third-party punishment, the effect cannot be quantified in Kriss et al. (2016), as a preference for being honest may cause third parties to not misreport the outcome of the dice roll.

  8. 8.

    Similarly, van der Weele et al. (2014), Matthey and Regner (2015) and Regner (2018) study the extent to which forms of moral wiggle room other than willful ignorance affect negative and positive reciprocity.

  9. 9.

    Bartling et al. (2015) and Felgendreher (2018) find that the possibility to avoid information does not have a strong impact on consumption decisions in markets.

  10. 10.

    Some studies provide evidence that altruistic punishment is affected by the diffusion of responsibility of the dictator or the directness of his decisions (Coffman 2011; Bartling and Fischbacher 2012; Oexl and Grossman 2013).

  11. 11.

    The instructions can be found in Online Appendix C.1.

  12. 12.

    The social norm elicitation in Sect. 4 will suggest that this labeling is justified. Note that option A2 is also more efficient than option A1, as it implies a joint income of €8 in stage 1 for the dictator and the recipient rather than a payoff of €7.

  13. 13.

    It has been argued that eliciting third parties’ punishment decision under role uncertainty does not influence treatment effects (Bartling et al. 2014; Nikiforakis and Mitchell 2014). If there was an effect, for instance, by increasing the fraction of third parties who reveal the decision of the dictator, any treatment effect should be underestimated.

  14. 14.

    Subjects who participated in Moradi and Nesterov (2017) were not invited to participate.

  15. 15.

    Throughout the study, all reported tests are two-tailed tests.

  16. 16.

    Note that the observed punishment rate (the fraction of third parties choosing to punish) in the baseline treatment is similar to punishment rates after norm violations in previous studies that use continuous sanctioning measures. In particular, in Fehr and Fischbacher (2004), when dictators do not share equally, about 60% of the third parties choose to engage in some punishing. Equally, Henrich et al. (2006) conducting experiments with subjects from five continents report that, on average, two-thirds of the third parties are willing to punish the dictator if she leaves zero to the recipient. While these studies differ from mine in several respects this is indicative of the fact that my findings are not driven by especially high or low punishment rates in the baseline treatment.

  17. 17.

    Equally, as dictators do not behave differently across treatments and as, conditional on a fair dictator choice, third parties also do not behave differently across treatments, the most efficient outcome (fair dictator, no punishment) is chosen equally often in both treatments (MW-test, \(p=0.772\)).

  18. 18.

    Studying the effect of counter-punishment opportunities on third-party punishment, Balafoutas et al. (2014) find that, although the opportunity to counter-punish reduces punishment, the proportion of norm violations is identical with and without counter-punishment.

  19. 19.

    I assess the beliefs about the average dictator behavior after subjects were informed of the outcome of the game, but as I observe the choices of the dictators I can control for its effect on average beliefs and even allow it to vary between treatments.

  20. 20.

    Observing a selfish dictator has a strong influence on this belief: The average belief for thirds, who observed a selfish (fair) dictator is 67% (34%) (t test, \(\textit{p}<0.001\)).

  21. 21.

    There is mixed evidence on whether changes in this probability of conflicting payoffs affect the rate of information avoidance (see van der Weele 2014 and Moradi and Nesterov 2017, but Feiler 2014).

  22. 22.

    Looking at the raw data, the treatment difference in beliefs is 7 percentage points and also not statistically significant (t test, \(\textit{p}=0.172\)).

  23. 23.

    In fact, I can explore the extent to which the elicited norms can explain behavior by predicting the choice probabilities of the four choices in the baseline treatment based on the social appropriateness of the action and its monetary payoff (see Sect. 5.1 for a thorough discussion). I predict that upon observing a selfish dictator choice punishment will be chosen with a high probability (67%) and not punishing with a corresponding low probability (33%). Upon observing a fair dictator choice, I predict that not punishing will be chosen with a very high probability (99%) and punishment with a corresponding very low probability (1%). Hence, the predicted choice probabilities match the actual fraction of choices closely. At the same time, it makes sense that the social norms do not coincide with behavior, because the social norms elicited with the method proposed in Krupka and Weber (2013) are injunctive norms, that is, norms regarding what individuals “ought” to do and not necessarily what they actually do.

  24. 24.

    More precisely, if the third parties, on average, believe that dictators chose the selfish option, then it might be more appropriate to punish than to not punish, because the dictator is more likely selfish. In addition, it should matter whether it is more appropriate not to punish a selfish dictator than to punish a fair dictator. When making distributive choices, people seem to avoid false negatives (giving individuals more than they deserve), rather than false positives (giving individuals less than they deserve) (Cappelen et al. 2018). To the degree that these preferences are a reflection of social norms and to the degree to which they can be transferred to punishment decisions, the norm to punish a selfish dictator might be less strong than the norm not to punish a fair dictator.

  25. 25.

    For the reasons mentioned in Krupka and Weber (2013) I am not able to estimate the parameters with the data at hand and I thus use the parameters these authors estimated based on dictator game data from List (2007).

  26. 26.

    For instance, to arrive at the prediction that 30.74% of third parties reveal the dictator’s choice, observe a fair dictator choice and choose not to punish, I sum up the percentage of third parties who are predicted to choose the strategy “reveal, do not punish a selfish dictator, and do not punish a fair dictator” (25.76%) and who are predicted to choose the strategy “reveal, punish a selfish dictator, and do not punish a fair dictator” (35.71%), and multiply this by the empirical percentage of fair dictator choices in the hidden information treatment (50.00%).

  27. 27.

    This requires the assumption that self-image concerns are also present for altruistic punishment.

  28. 28.

    As discussed before, an alternative approach for explaining differences in choices is that third parties differ in their ability to bias their perception regarding how appropriate it is to remain ignorant.

  29. 29.

    Note that there is mixed evidence of sorting in generosity decisions, as there is significant sorting into revealing in Grossman and van der Weele (2017) but not in Dana et al. (2007), Larson and Capra (2009) and Kajackaite (2015).

  30. 30.

    If the selfish type reveals and the share of the selfish type is sufficiently large, prosocial behavior may actually be lower among those who reveal (see Grossman and van der Weele 2017).

  31. 31.

    I obtained this measure of social value orientation for 93 of the subjects.

  32. 32.

    The instructions for these treatments can be found in Online Appendix C.3.

  33. 33.

    The results of these treatments are not obvious, because (i) it is unclear what fraction of dictators who chose the fair option in the original sessions did so due to their social preferences, (ii) the elasticity of dictators’ beliefs with respect to the information is unknown, and (iii) dictators’ risk-preferences are unknown.

  34. 34.

    Hence, the information was provided in a natural way. Assuming that the dictators do not leave money on the table to act in accordance with an experimenter demand when beliefs are elicited, I can check whether a difference in dictator behavior goes along with a difference in beliefs rather than being driven by an experimenter demand effect.

  35. 35.

    I did not elicit the beliefs at the start, because eliciting beliefs itself may or may not affect behavior and I intended to measure the causal effect of the information on dictator behavior [see Nyarko and Schotter (2002) and Costa-Gomes and Weizsäcker (2008), but Kovářík (2007) and Gächter and Renner (2010)].

  36. 36.

    These are about the same numbers as in the hidden information treatment, which I aimed for.

  37. 37.

    Remember that the third parties do not receive any information about previous punishment behavior. They are, however, informed about the proportion of dictators choosing selfishly in their treatment.

  38. 38.

    This might be caused by third parties who engage in punishment believing that a high fraction of dictators are fair. Learning about approximately 50% of dictators being selfish might then discourage punishment. Accordingly, third parties who punish selfish dictators in the original sessions believe that the fraction of selfish dictators is lower than third parties who do not (diff.: 21 pp, t test, \(p<0.001\)). In addition, the beliefs about the fraction of selfish dictators vary slightly between the baseline informed and the hidden information informed treatment (diff.: 6 pp, t test, \(p=0.027\)), which might also contribute to the result.

  39. 39.

    If I pool the data of the baseline treatment and the baseline informed treatment, and the hidden information treatment and the hidden information informed treatment, respectively, the treatment effect is 20 percentage points or 42% and highly statistically significant (FET, \(p=0.011\)). Forty-eight percent of third parties remain ignorant. Across both treatments only three out of 82 third parties punish a fair dictator.

  40. 40.

    More broadly, whether an exogenous information provision is beneficial also depends on other economic considerations, such as the monitoring costs or the costs of punishment relative to the benefits of sustaining a norm.


  1. Andreoni, J., Rao, J. M., & Trachtman, H. (2017). Avoiding the ask: A field experiment on altruism, empathy, and charitable giving. Journal of Political Economy, 125, 625–653.

    Google Scholar 

  2. Aquino, K., & Reed, A. I. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83, 1423.

    Google Scholar 

  3. Balafoutas, L., Grechenig, K., & Nikiforakis, N. (2014). Third-party punishment and counter-punishment in one-shot interactions. Economics Letters, 122, 308–310.

    Google Scholar 

  4. Balafoutas, L., & Nikiforakis, N. (2012). Norm enforcement in the city: A natural field experiment. European Economic Review, 56, 1773–1785.

    Google Scholar 

  5. Balafoutas, L., Nikiforakis, N., & Rockenbach, B. (2016). Altruistic punishment does not increase with the severity of norm violations in the field. Nature Communications, 7, 13327.

    Google Scholar 

  6. Bartling, B., Engl, F., & Weber, R. A. (2014). Does willful ignorance deflect punishment? An experimental study. European Economic Review, 70, 512–524.

    Google Scholar 

  7. Bartling, B., & Fischbacher, U. (2012). Shifting the blame: On delegation and responsibility. The Review of Economic Studies, 79, 67–87.

    Google Scholar 

  8. Bartling, B., Weber, R. A., & Yao, L. (2015). Do markets erode social responsibility? The Quarterly Journal of Economics, 130, 219–266.

    Google Scholar 

  9. Cain, D., & Dana, J. (2012). Paying people to look at the consequences of their actions. Working paper.

  10. Cappelen, A. W., Cappelen, C., & Tungodden, B. (2018). Second-best fairness under limited information: The trade-off between false positives and false negatives. NHH Department of Economics Discussion Paper No. 18/2018.

  11. Carpenter, J. P. (2007). The demand for punishment. Journal of Economic Behavior & Organization, 62, 522–542.

    Google Scholar 

  12. Carpenter, J. P., & Matthews, P. H. (2012). Norm enforcement: Anger, indignation, or reciprocity? Journal of the European Economic Association, 10, 555–572.

    Google Scholar 

  13. Charness, G., Cobo-Reyes, R., & Jiménez, N. (2008). An investment game with third-party intervention. Journal of Economic Behavior & Organization, 68, 18–28.

    Google Scholar 

  14. Coffman, L. C. (2011). Intermediation reduces punishment (and reward). American Economic Journal: Microeconomics, 3, 77–106.

    Google Scholar 

  15. Costa-Gomes, M. A., & Weizsäcker, G. (2008). Stated beliefs and play in normal-form games. The Review of Economic Studies, 75, 729–762.

    Google Scholar 

  16. Cox, J. C., Servátka, M., & Vadovič, R. (2017). Status quo effects in fairness games: Reciprocal responses to acts of commission versus acts of omission. Experimental Economics, 20, 1–18.

    Google Scholar 

  17. Dana, J., Cain, D. M., & Dawes, R. M. (2006). What you don’t know won’t hurt me: Costly (but quiet) exit in dictator games. Organizational Behavior and Human Decision Processes, 100, 193–201.

    Google Scholar 

  18. Dana, J., Weber, R. A., & Kuang, J. X. (2007). Exploiting moral wiggle room: Experiments demonstrating an illusory preference for fairness. Economic Theory, 33, 67–80.

    Google Scholar 

  19. DellaVigna, S., List, J. A., & Malmendier, U. (2012). Testing for altruism and social pressure in charitable giving. The Quarterly Journal of Economics, 127, 1–56.

    Google Scholar 

  20. Egas, M., & Riedl, A. (2008). The economics of altruistic punishment and the maintenance of cooperation. Proceedings of the Royal Society B: Biological Sciences, 275, 871–878.

    Google Scholar 

  21. Fehr, E., & Fischbacher, U. (2004). Third-party punishment and social norms. Evolution and Human Behavior, 25, 63–87.

    Google Scholar 

  22. Fehr, E., & Gächter, S. (2000). Cooperation and punishment in public goods experiments. American Economic Review, 90, 980–994.

    Google Scholar 

  23. Fehr, E., & Schmidt, K. M. (1999). A theory of fairness, competition, and cooperation. The Quarterly Journal of Economics, 114, 817–868.

    Google Scholar 

  24. Fehr, E., & Schurtenberger, I. (2018). Normative foundations of human cooperation. Nature Human Behaviour, 2, 458–468.

    Google Scholar 

  25. Feiler, L. (2014). Testing models of information avoidance with binary choice dictator games. Journal of Economic Psychology, 45, 253–267.

    Google Scholar 

  26. Felgendreher, S. (2018). Do consumers choose to stay ignorant? The role of information in the purchase of ethically certified products. Working paper in economics 717, Department of Economics, University of Gothenburg.

  27. Fischbacher, U. (2007). z-Tree: Zurich toolbox for ready-made economic experiments. Experimental Economics, 10, 171–178.

    Google Scholar 

  28. Gächter, S., Gerhards, L., & Nosenzo, D. (2017). The importance of peers for compliance with norms of fair sharing. European Economic Review, 97, 72–86.

    Google Scholar 

  29. Gächter, S., Nosenzo, D., & Sefton, M. (2013). Peer effects in pro-social behavior: Social norms or social preferences? Journal of the European Economic Association, 11, 548–573.

    Google Scholar 

  30. Gächter, S., & Renner, E. (2010). The effects of (incentivized) belief elicitation in public goods experiments. Experimental Economics, 13, 364–377.

    Google Scholar 

  31. Goeschl, T., & Jarke, J. (2016). Second and third party punishment under costly monitoring. Journal of Economic Psychology, 54, 124–133.

    Google Scholar 

  32. Greiner, B. (2015). Subject pool recruitment procedures: Organizing experiments with ORSEE. Journal of the Economic Science Association, 1, 114–125.

    Google Scholar 

  33. Grossman, Z. (2014). Strategic ignorance and the robustness of social preferences. Management Science, 60, 2659–2665.

    Google Scholar 

  34. Grossman, Z., & van der Weele, J. J. (2017). Self-image and willful ignorance in social decisions. Journal of the European Economic Association, 15, 173–217.

    Google Scholar 

  35. Henrich, J., McElreath, R., Barr, A., Ensminger, J., Barrett, C., Bolyanatz, A., et al. (2006). Costly punishment across human societies. Science, 312, 1767–1770.

    Google Scholar 

  36. Kajackaite, A. (2015). If I close my eyes, nobody will get hurt: The effect of ignorance on performance in a real-effort experiment. Journal of Economic Behavior & Organization, 116, 518–524.

    Google Scholar 

  37. Kandul, S. (2016). Ex-post blindness as excuse? The effect of information disclosure on giving. Journal of Economic Psychology, 52, 91–101.

    Google Scholar 

  38. Kovářík, J. (2007). Belief formation and evolution in public good games. LABSI Working Paper.

  39. Kriss, P. H., Weber, R. A., & Xiao, E. (2016). Turning a blind eye, but not the other cheek: On the robustness of costly punishment. Journal of Economic Behavior & Organization, 128, 159–177.

    Google Scholar 

  40. Krupka, E. L., & Weber, R. A. (2013). Identifying social norms using coordination games: Why does dictator game sharing vary? Journal of the European Economic Association, 11, 495–524.

    Google Scholar 

  41. Larson, T., & Capra, C. M. (2009). Exploiting moral wiggle room: Illusory preference for fairness? A comment. Judgment and Decision Making, 4, 467.

    Google Scholar 

  42. Lazear, E. P., Malmendier, U., & Weber, R. A. (2012). Sorting in experiments with application to social preferences. American Economic Journal: Applied Economics, 4, 136–63.

    Google Scholar 

  43. Leibbrandt, A., & López-Pérez, R. (2012). An exploration of third and second party punishment in ten simple games. Journal of Economic Behavior & Organization, 84, 753–766.

    Google Scholar 

  44. Levine, D. K. (1998). Modeling altruism and spitefulness in experiments. Review of Economic Dynamics, 1, 593–622.

    Google Scholar 

  45. Lewisch, P., Ottone, S., & Ponzano, F. (2011). Free-Riding on altruistic punishment? An experimental comparison of third-party punishment in a stand-alone and in an in-group environment. Review of Law & Economics, 7, 161–190.

    Google Scholar 

  46. List, J. A. (2007). On the interpretation of giving in dictator games. Journal of Political economy, 115, 482–493.

    Google Scholar 

  47. Lotz, S., Okimoto, T. G., Schlösser, T., & Fetchenhauer, D. (2011). Punitive versus compensatory reactions to injustice: Emotional antecedents to third-party interventions. Journal of Experimental Social Psychology, 47, 477–480.

    Google Scholar 

  48. Matthey, A., & Regner, T. (2015). Do reciprocators exploit or resist moral wiggle room? An experimental analysis. Jena Economic Research Papers, 9.

  49. Moradi, H., & Nesterov, A. (2017). Moral wiggle room reverted: Information avoidance is myopic. Working paper.

  50. Murphy, R. O., Ackermann, K. A., & Handgraaf, M. J. (2011). Measuring social value orientation. Judgment and Decision Making, 6, 771–781.

    Google Scholar 

  51. Nikiforakis, N. (2008). Punishment and counter-punishment in public good games: Can we really govern ourselves? Journal of Public Economics, 92, 91–112.

    Google Scholar 

  52. Nikiforakis, N., & Engelmann, D. (2011). Altruistic punishment and the threat of feuds. Journal of Economic Behavior & Organization, 78, 319–332.

    Google Scholar 

  53. Nikiforakis, N., & Mitchell, H. (2014). Mixing the carrots with the sticks: Third party punishment and reward. Experimental Economics, 17, 1–23.

    Google Scholar 

  54. Nyarko, Y., & Schotter, A. (2002). An experimental study of belief learning using elicited beliefs. Econometrica, 70, 971–1005.

    Google Scholar 

  55. Oexl, R., & Grossman, Z. (2013). Shifting the blame to a powerless intermediary. Experimental Economics, 16, 306–312.

    Google Scholar 

  56. Regner, T. (2018). Reciprocity under moral wiggle room: Is it a preference or a constraint? Experimental Economics, 21, 779–792.

    Google Scholar 

  57. Trachtman, H., Steinkruger, A., Wood, M., Wooster, A., Andreoni, J., Murphy, J. J., et al. (2015). Fair weather avoidance: Unpacking the costs and benefits of “avoiding the ask”. Journal of the Economic Science Association, 1, 8–14.

    Google Scholar 

  58. van der Weele, J. J. (2014). Inconvenient truths: Determinants of strategic ignorance in moral dilemmas. SSRN Working Paper.

  59. van der Weele, J. J., Kulisa, J., Kosfeld, M., & Friebel, G. (2014). Resisting moral wiggle room: How robust is reciprocal behavior? American Economic Journal: Microeconomics, 6, 256–264.

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Robert Stüber.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

I am grateful to Philipp Albert, Kai Barron, Björn Bartling, Alexander Cappelen, Marvin Deversi, Urs Fischbacher, Zack Grossman, Rustamdjan Hakimov, Erin Krupka, Dorothea Kübler, Homayoon Moradi, Lisa Spantig, Kristina Strohmaier, Joël van der Weele, Justin Valasek, Roel van Veldhuizen and seminar participants at the ESA World Meeting 2018, the IMEBESS 2018, the SABE/IAREP 2018, the Lisbon Meeting in Economics and Political Science 2018, the theem 2019, and the Berlin Behavioral Economics Workshop for their helpful suggestions. I am also very thankful to Jennifer Rontganger for copy editing.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 357 KB)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Stüber, R. The benefit of the doubt: willful ignorance and altruistic punishment. Exp Econ 23, 848–872 (2020).

Download citation


  • Third-party punishment
  • Willful ignorance
  • Sorting
  • Social preference

JEL Classification

  • C91
  • D01
  • D63
  • D83