Events that uncover unethical behaviour by members of the economic and political elite often receive significant attention in the media and in public discourse. Consistent with the salience of such events, many people perceive members of the upper-class as being greedier and less honest than the average person (Pew Research Center 2012). If true, such differences in social and moral attitudes between classes have implications for the social fabric. If the upper classes wield disproportionate influence on political decisions and societal trends, their ethics are of disproportionate importance. Moreover, they often serve as role models, and their behaviour signals what is acceptable in a society, leading to a possible contagion of unethical behaviour spreading down the social ladder (e.g. Rockenbach et al. 2021). The subject is pertinent to recent discussions of social inequality (Saez et al. 2022, Pistaferri and Guiso 2022), as well as to questions of executive remuneration and pay ratios in the context of corporate governance (Gallus and Frey 2019).
When studying differences in ethical behaviour between social classes, we need to distinguish an association between class and behaviour from a causal effect of class on behaviour. In the former, an association between immorality and socioeconomic status could be due to unethical people being economically successful in our society: our institutions may be prone to selecting unethical people for elevated positions. In contrast, observing a causal effect of status on ethical behaviour would suggest that becoming successful – i.e. being elevated to a higher socioeconomic status – affects people’s ethics: irrespective of their original dispositions, people behave less socially and morally as they step up the social ladder. Both relationships have different implications for policy and the design of institutions.
There is research looking into both relationships. Because it is difficult to change a person’s status, or to observe behaviour over a prolonged period in which the person is exposed to a change in their social status, research into the potential causal relationship has typically used psychological priming techniques. That is, study participants are exposed to information or experiences that shift their perceived social status. For example, participants who were made to think about the very poor (respectively the very rich) perceived their own position in society as higher (respectively lower), and subsequently were less generous in terms of charitable giving (Piff et al. 2010). In another study, students received information about the presumable standing of their department compared to other departments at a prestigious UK university. Those who were made to feel of higher status due to the prestige of their department were subsequently less helpful towards a research assistant (Guinote et al. 2015).
Several studies using such artificial priming techniques have claimed a causal relationship, with higher status inducing less ethical and social behaviour, typically employing convenience samples of student participants. In contrast, research into the simple association between indicators of status and indicators of ethical behaviour has typically made use of large population surveys. These studies often found no association, and sometimes a positive association, between measures of higher status and self-reported ethical behaviours (Korndörfer et al. 2015).
Further correlational research used real monetary payoffs to measure ethical behaviour. For example, Smeets et al. (2015) study the behaviour of a group wealthy clients of a Dutch bank and find that in a simple €100 ‘divide a pie’ decision, millionaires share a larger amount with another person (either wealthy or poor) than the average person. While the millionaires give about 50% of the €100 to another wealthy person and 70% to a poor person, in student samples this number typically comprises less than 30% of the pie (Engel 2011). Andreoni et al. (2021) test whether people mail a letter that was wrongly dropped in their mailboxes to the intended recipient. The letter may or may not visibly contain a bill. The authors find that people living in wealthier neighbourhoods are more likely to forward the letter, whether or not the money is visible.
In a recent study, we aimed to combine both the psychological priming and the correlational approach, and to make use of a measure of ethics with real monetary consequences (Gsottbauer et al. 2022). To this end, we have participants with detailed information on their incomes available undertake the following sequence of tasks. First, some participants answer questions about their views on the lives of the poor, and other participants about their views on the lives of the rich. They then indicate their self-perceived social standing. Subsequently, all participants take part in a simple game in which they can win up to €20. The game consists of guessing a letter that will be randomly determined by the computer. Importantly, the game is designed such that participants only guess the letter in their mind, and later calculate and report their earnings themselves, thus allowing for misreported earnings. Variations of this design have been widely used as a measure of dishonest behaviour. Importantly, participants are not at risk of being ‘detected’ to have cheated, even if they misreport their earnings. The task is designed so that it is impossible to prove any claim wrong. However, given that the winning letter is randomly selected by the computer, on average in larger groups we can statistically detect dishonest behaviour.
We ran this experiment on two large panels of German participants. Our results show that higher socioeconomic status as measured by income is associated with less dishonesty in the first panel; we observe equal dishonesty among high- and low-income individuals in the second panel. The priming task replicates earlier studies by successfully shifting participants’ perception of their social status. However, despite the substantial sample sizes to detect even small differences, this shift in status perception has no effect on the dishonesty of the primed versus the non-primed groups. Importantly, this is not due to a simple absence of dishonesty. On average over all participants, we find a moderate degree of dishonesty: with fully truthful reports, the average earnings of our participants would amount to €10; the average report equals about €11.10.
On the basis of the literature, we can conclude that there is little robust evidence of an association between social status and ethics. If anything, there is some evidence that high-status individuals may behave more socially and ethically, rather than becoming systematically unethical and selfish. How does this square with public perceptions? Why do events uncovering unethical behaviour by members of the elite, as well as the occasional study claiming they have poor morals, receive disproportionate media attention? Together with our co-authors, we conjecture that the rich may not fall short in terms of their actual behaviour, but with respect to our expectations about how they should behave (Trautmann et al. 2022). In several survey studies, we find that people hold members of the upper class to higher ethical standards than they hold the lower classes. If members of the upper class fall short of these higher, more demanding standards, it feels as if they behave more unethically than other classes. People judge others relative to an ethics-related reference point. The above-mentioned study by Smeets et al. (2015) observing Dutch millionaires illustrates this intuition. As we have seen, the millionaires share about 70% of the €100 with a poor person. This is substantially more generous than participants in studies with less wealthy individuals. Still, many people would argue that because the millionaires are so much wealthier than the recipients, this does not signal particularly high standards of generosity. It is just what we would expect of them.
Inequality at the societal level, as well as skewed pay ratios in governance contexts, are often considered harmful to the social fabric (e.g. Fehr et al. 2020). Similarly, widespread unethical behaviour can be considered a threat to a society’s economic wellbeing (Tabellini 2007). It has been suggested that inequality degrades morals due to the behaviour of those at the right tail of the distribution. Higher status is claimed to be associated with – and possibly a cause of – unethical, selfish behaviour. Contagion to lower classes may amplify these negative effects. From our own research and a review of the literature, we cannot support these claims. Unethical behaviour seems to be similarly prevalent at all socioeconomic levels. That said, similar transgressions may be more impactful if coming from more influential individuals, and higher status individuals may be held to higher ethical standards. This is important to understand for those occupying influential social positions. Noblesse oblige.
Andreoni, J, N Nikiforakis and J Stoop (2021), “Higher socioeconomic status does not predict decreased prosocial behavior in a field experiment”, Nature Communications 12, 1–8.
Cohn, A, L Jessen, M Klasnja and P Smeets (2022), “The Wealth Gap in Fairness Preferences: Evidence from America’s Top 5%”, Working paper, University of Michigan.
Engel, C (2011), “Dictator games: A meta study”, Experimental Economics 14, 583–610.
Fehr, D, H Rau, S T Trautmann and Y Xu (2020), “Inequality, Fairness and Social Capital”, European Economic Review 129, 103566.
Gallus, J and B Frey (2019), “Honours: A counterpoise to today’s bonus culture”, VoxEU.org, 14 April.
Gsottbauer, E, D Müller, S Müller, S T Trautmann and G Zudenkova (2022), “Social Class and (Un)ethical Behavior: Causal and Correlational Evidence”, Economic Journal 132, 2392–2411.
Guinote, A, I Cotzia, S Sandhu and P Siwa (2015), “Social status modulates prosocial behavior and egalitarianism in preschool children and adults”, Proceedings of the National Academy of Sciences 112, 731–736.
Korndörfer, M, B Egloff and S C Schmukle (2015), “A large scale test of the effect of social class on prosocial behavior”, PLoS ONE 10, e0133193.
Piff, P K, M W Kraus, S Côté, B H Cheng and D Keltner (2010), “Having Less, Giving More: The Influence of Social Class on Prosocial Behavior”, Journal of Personality and Social Psychology 99, 771–784.
Pistaferri, L and L Guiso (2022), “Assortative mating and the wealth inequality debate”, VoxEU.org, 14 May.
Pew Research Center (2012), “Yes, the rich are different”, accessed 30 November 2022.
Rockenbach, B, S Tonke and A R Weiss (2021), “Self-serving behavior of the rich causes contagion effects among the poor”, Journal of Economic Behavior and Organization 183, 289–300.
Saez, E, G Zucman and T Blanchet (2022), “Tracking US inequality in real time”, VoxEU.org, 3 August.
Smeets, P, R Bauer and U Gneezy (2015), “Giving Behavior or Millionaires”, Proceedings of the National Academy of Science 112, 10641–10644.
Tabellini, G (2007), “Morality matters for economic performance”, VoxEU.org, 22 December.
Trautmann, S T, G van de Kuilen and R J Zeckhauser (2013), “Social Class and (Un)ethical Behavior: A Framework, With Evidence From a Large Population Sample”, Perspectives on Psychological Science 8, 487–497.
Trautmann, S T, X Wang, Y Wang and Y Xu (2022), “Noblesse Oblige: Holding High-Status Individuals to Higher Standards” Working paper, Heidelberg University.