An experiment shows that although bank employees behave honestly on average, their dishonesty increases when they make decisions after having been primed to think about their professional identity. See Letter p.86
Honesty involves individual preferences, but also legal rules and moral and social norms that prescribe what we ought to do. We would like to believe that such norms are able to help individuals resist the temptation of earning more money by behaving dishonestly. But unfortunately, norms that are specific to a particular location or group can sometimes have the opposite effect. On page 86 of this issue, Cohn et al.1 provide an example of this in their report of an experiment conducted with the employees of a large international bank. The authors show that the bank employees behave honestly on average in a control situation, but become less honest after having been 'primed' to think about their professional identity.
Several scandals in the financial sector have tarnished its reputation, but it has been unclear whether these scandals result mainly from the effects of selection or of incentives (Fig. 1). The first possibility centres on the idea that the considerable amounts of wealth and assets traded in the financial sector attract some people who have weaker honesty norms. According to the second explanation, the incentives and the business culture developed in the financial sector may undermine the honesty norms of ordinary employees. Cohn et al. provide evidence for the latter explanation.
A direct comparison of the behaviour of bankers and their respective degrees of honesty with that of employees from other sectors would not be conclusive because one can never guarantee that the two groups would be perfectly comparable, owing to unobservable characteristics. Instead, Cohn et al. separated their banker participants into two experimental groups. In the control group, the employees completed a survey about life satisfaction, well-being and everyday life that did not include questions related to their professional role. They were then asked to toss a coin ten times in privacy and report the outcomes online. For each toss, the participants knew in advance whether tails or heads would yield a monetary pay-off. In the treatment group, seven of the questions in the preliminary questionnaire were replaced with those related to the bankers' profession. For example, questions such as “how many hours per week do you watch television on average?” were replaced with ones such as “what is your function at this bank?”. These questions were designed to prime the participants to their professional identity before they completed the coin-toss and reporting task.
The authors found that, in the control group, the distribution of winning tosses reported did not differ from the binomial distribution expected from honest reporting. This indicates that, on average, the employees behaved honestly. However, reporting by the treatment group was significantly different: the percentage of winning tosses reported increased from 51.6% in the control group to 58.2% in the treatment group, and the proportion of employees who reported dishonestly on at least some of the tosses increased from 16% to 26%.
The effect induced by the treatment could be attributable to several causes, including the competitiveness expected from bank employees, the exposure to competitive bonus schemes, the beliefs about what other employees would do in the same situation or the salience of money in the questionnaire. Cohn and colleagues conducted further tests and statistical analyses to assess these potential channels, but provide data that discount each of them. They also find that the professional-identity priming effect was not replicated in identical sessions involving participants from other professions. This strongly suggests that the increased dishonesty of the bank employees in the treatment group results from the business culture specific to the banking industry. Business culture is a complex entity that encompasses an organization's norms, values, visions, expectations and habits; it largely exceeds the incentives schemes promoted in the company.
These findings confirm some popular opinions about practices in the financial sector and they have direct implications: it is crucial to ensure a business culture of honesty in this industry to restore trust in it. Testing whether the professional-identity effect disappears in companies that have introduced training programmes in ethics would be extremely relevant. The results also open the door to other possible tests of the professional-identity effect on preferences. For example, risk-taking decisions may also be strongly affected by professional-identity priming. Moreover, there is no reason to believe that this effect is limited to the financial sector. For example, the same method could be used to test whether the honesty of politicians is negatively affected by political-background priming when the participants are faced with opportunities for political gain, rather than the financial incentive used for the bankers. At the opposite end of the moral-norms spectrum, the degree of altruism of doctors and nurses or the willingness of police officers to enforce punishment might increase when primed with their professional identity.
From a scientific perspective, this study not only supports the economic theory of social identity2,3,4,5 — which considers that gender or occupation, for example, create multiple social identities in individuals — but also links this theory with the economic analysis of lying behaviour6,7. Beyond economics, it shows how behavioural economists can contribute to a broader reflection in science about how people manage their 'multiple selves' to maintain their self-concept of honesty8. In particular, we should explore further whether professional identity is used strategically by people to reduce their own responsibility for their behaviour and to shift the blame to their companies. Neuroscientific methods could be solicited to understand how and when the human brain lets professional (impersonal) identity become the dominant influence on an individual's moral norms.
It was not Cohn and colleagues' aim to explain how business culture may encourage misbehaviour. However, understanding how the culture of dishonesty evolves is an important issue and it is unlikely that the process is one way. On the one hand, the development of business culture may lead to a 'deindividuation' process that facilitates the alignment of the individual with the group's preferences9. If the business culture goes wrong, then individuals may also develop unethical behaviour. On the other hand, individual misconduct may also influence the evolution of business cultures towards more deviant collective norms, through social influence and belief formation. A better understanding of this dual process will require a theoretical framework that takes into account not only individual motivations but also social interactions and peer effects10.