Imagine you could lie for $100 — and no one would ever know
Would you do it? A new study reveals how a single promise can tilt the scales toward honesty.
We lie for those who are like us
Three experiments with more than 5,000 participants show that people are willing to cheat if it benefits their group — even when they gain nothing themselves.
"In short, we tend to lie more to benefit someone like us, than lie to harm someone not like us", Jareef Bin Martuza says.
He is a postdoc at the Department of Strategy and Management and an affiliated researcher at the research centre DIG at NHH. Together with Jay Van Bavel from New York University and professors Helge Torbjørnsen and Hallgeir Sjåstad (Department of Strategy and Management), he is working on a paper on group identities and dishonest behaviour.
Cheating for ourselves – and others
The researcher’s basic proposition was to test whether the tendency to behave dishonestly is influenced by the group identity of either the victim or the beneficiary of the cheating. This was tested in three experiments with more than 5,000 participants in total, and with real money.
People will lie more if it helps a group or a team they are connected to, even with no benefits to them personally, they find.
Is there a difference if the victim of your cheating is part of our “in-group” or is in the “out-group”?
Participants lied significantly more to benefit their in-group than out-group members and did not lie on average to harm their out-group.
Jareef Bin Martuza
Surprisingly not. For purely selfish dishonesty, people lied equally regardless of whether it came at a cost to an in-group or out-group member, Jareef Bin Martuza says,
However, when people could no longer personally benefit from cheating, a different pattern of cheating emerged. Participants lied significantly more to benefit their in-group than out-group members and did not lie on average to harm their out-group.
Factsheet
Three experiments were carried out on a total of 5,230 Americans.
To define the “in-group” and the “out-group”, participants identifying as Democratic or Republican party voters were recruited.
In Experiment 1, participants were given the choice to tell the truth or lie, where lying would guarantee them making more money, which came at the cost of someone from their own political party (“in-group”) or the opposing party (“out-group”).
In Experiment 2, participants could lie to make more money to benefit another in-group or out-group participant, without personally gaining anything.
Experiment 3 combined both Experiments 1 and 2 and found the same results.
In all experiments, decisions were anonymous, as the experimenter, victim, or beneficiary would never know the identity of those behaving dishonestly.
In these studies, neither the victim nor the beneficiary of the dishonest act knew that their fate was caused by these actions. As a bonus finding, people consistently overestimated out-group dishonesty more than in-group dishonesty, regardless of actual differences in behavior.
What are the implications
`The key implication is that the risk of dishonesty in organizations is not limited to selfish acts by individuals. A hidden risk may come from “other-benefitting” dishonesty, where employees might bend rules to benefit their team or in-group members. For example, a team might conceal a project delay to protect the team's reputation, or misreport facts to help a teammate secure a resource, says Jareef Bin Martuza.
A further implication, Martuza explains, is the problems this creates for leaders. Most people expect dishonesty to be driven by selfish motives. Standard compliance and ethics programs are often built on this assumption. As a result, they are designed to deter individual fraud but might be ill-equipped to handle unethical acts committed on behalf of a group.
`One potential solution could be to strengthen a broader, organization-wide identity. When employees' primary loyalty is to the organization rather than their specific team, actions that harm the company for the benefit of a small group might be less likely to occur´, Jareef Bin Martuza concludes.
The paper is being prepared for submission to a leading academic journal.