A guest post by Christoph Schild
A handful of football fans light flares or storm the pitch. In response, the governing body bans all supporters of that team from attending the next match. A few individuals misbehave but the entire fanbase is punished.[1]
This is a classic example of collective punishment: sanctioning a whole group when individual offenders cannot be identified and/or individually punished. Collective punishment is widely seen as unfair, yet it continues to be used in many real-world contexts, from sports and workplaces to public policy. This raises a fundamental empirical question.
Does collective punishment actually reduce wrongdoing?
And if it does, who responds to it and who supports its use?
In our paper, Testing the effectiveness and endorsement of collective punishment, recently published in the European Journal of Personality, we address these questions in six preregistered studies with more than 11,000 participants from Denmark, the UK, and the US
Why effectiveness is crucial
Debates about collective punishment are often framed in moral terms and can be framed from two perspectives. One perspective is the retributivist one, which posits that those who are blameworthy deserve punishment proportionate to their wrongdoing and thus, collective punishment is unjust because innocent people suffer. Consequentialism is a different perspective whereby the effectiveness of the punishment is what determines the morality of the punishment itself. Therefore, collective punishment would be justified if it effectively reduces harmful behavior and produces societal benefits, especially in situations where individual wrongdoers cannot be identified and/or individually punished.
Whether collective punishment delivers such outcomes, however, has received surprisingly little systematic attention to date… Does collective punishment reduce dishonesty?
To test whether collective punishment reduces wrongdoing when individual offenders cannot be identified and/or individually punished, we focused on dishonesty and used a simple experimental task known as the Mind Game. In this task, participants first write down a number in private. Next, a number is randomly shown on the screen, and participants are asked whether the two numbers match. If they answer “yes,” they receive a monetary reward.
Crucially, participants can lie while remaining completely anonymous—no one can tell whether a given person actually saw a match. At the same time, because the probability of a genuine match is known, it is possible to infer at the group level whether more people claimed a match than should be expected by chance. In other words, individual dishonesty cannot be detected, but collective dishonesty can.
Across four preregistered experiments, the pattern was clear. Collective punishment substantially reduced dishonest behavior—by roughly 40% compared to control conditions. Importantly, this reduction is comparable in size to effects found in studies that examine individual-level punishment. Thus, even when wrongdoers cannot be singled out, collective punishment can be similarly effective at deterring wrongdoing.
Do personality traits matter?
People differ widely in how willing they are to bend rules and in how sensitive they are to threats of punishment. As a result, the same policy can have very different effects depending on who it targets. Understanding whether collective punishment affects some people more than others is therefore crucial for evaluating how it works and what kinds of behavioral changes it is likely to produce.
We examined whether personality traits shape how people respond to collective punishment. As expected, people lower in Honesty–Humility and higher in the Dark Factor of Personality (D) were more dishonest overall. Importantly, however, these traits did not meaningfully change how people responded to collective punishment: the presence of collective punishment reduced dishonesty at both higher and lower levels of Honesty–Humility and D.
In exploratory analyses, however, we observed one moderation pattern. Individuals higher in Emotionality showed a stronger reduction in dishonesty when collective punishment was present. This finding adds to recent evidence highlighting the role of emotionality in dishonest behavior when dishonesty has consequences for others, suggesting that sensitivity to others’ potential harm and/or concern about negative outcomes may play a role in how collective punishment affects behavior.
Who supports collective punishment?
In two additional studies conducted during the COVID-19 pandemic in Denmark, we examined endorsement of collective punishment. Participants were asked whether stricter, society-wide restrictions should be imposed if some citizens failed to comply—closely mirroring real-world collective punishment logic.
Support for collective punishment was higher among individuals who themselves followed the rules, were concerned about societal consequences, and scored higher on Emotionality and the Dark Factor of Personality. These results highlight that endorsement of collective punishment can stem from very different motivations, ranging from concern for collective welfare to more punitive tendencies.
What does this mean?
Our findings are not an argument for expanding the use of collective punishment. Rather, they provide missing empirical evidence that is essential for evaluating collective punishment from a consequentialist perspective.
In short:
Collective punishment can meaningfully reduce wrongdoing when individual offenders cannot be identified and/or individually punished.
People differ in how strongly they respond to and endorse it.
Whether these benefits outweigh the moral and social costs remains a normative question—one that requires careful judgment.
Our hope is that this research helps move debates about collective punishment beyond intuition and toward evidence, clarifying when it works, for whom, and at what cost.
[1] Indeed, as we were writing this blog post, the Ministry of the Interior in Italy has ordered a ban on travel for Roma and Fiorentina fans until the end of the football season, following violent clashes on the highway; see https://www.ansa.it/english/news/2026/01/20/travel-ban-for-roma-and-fiorentina-fans-until-the-end_8981acaa-322f-484f-83f9-d18b9e0fb1fc.html