As Donald Trump takes office without addressing his business-related conflicts of interest, his opponents are debating whether it makes political sense to highlight this issue. Do ethics violations really matter to us?

Experimental evidence suggests that the answer depends on who’s committing the violations, and who’s doing the judging.

In a 2010 paper, Hannah Riley Bowles and Michele Gelfand conducted a series of experiments to see how social status affected people’s willingness to break rules, and to judge other people’s rule-breaking.

First, Bowles and Gelfand divided 215 men and women into two groups. They asked one group to recall a time they had a high-status position, and then asked them to picture themselves as executives. The other group had to remember being in a low-status position and imagine holding an entry-level position. Then, the researchers asked the subjects questions about their “dutifulness”—willingness to stick to the rules and follow directions—and their “social dominance”—belief that some employees are more worthy than others and willingness to treat people unequally. The results showed that the “high-status” group was less dutiful and more socially dominant.

A second experiment with a different group of subjects found that, not only were higher status people more willing to break rules, they were also more relaxed about rule-breaking by other high-status workers. Bowles and Gelfand asked subjects to imagine an employee—either a worker with a strong reputation or one with little track record—sending personal mail at the company’s expense. Should this person be punished? Subjects in the “low-status” condition treated the low- and high-status offenders equally. But “high-status” evaluators were significantly less likely to want to punish the worker if he had a good reputation.

In one last experiment, Bowles and Gelfand looked at a different type of social status—race and gender. They asked a group of white men and women to evaluate fictional stories of workplace rule-breaking by black and white men and women. Again, they found the low-status subjects—in this case meaning the white women—treated the subjects essentially equally. In contrast, the researchers wrote, the white men evaluated “male deviance more leniently than female deviance, and white deviance more leniently than black deviance.”

If this study can tell us anything about the political significance of Trump’s potential ethics violations, it’s that the way we view the behavior of a white man in one of the highest-status jobs in the world will depend on our perspectives as observers.