Why Power Corrupts

“Power tends to corrupt,” said Lord Acton, the 19th-century British historian. “Absolute power corrupts absolutely.” His maxim has been vividly illustrated in psychological studies, notably the 1971 Stanford Prison Experiment, which was halted when one group of students arbitrarily assigned to serve as “prison guards” over another group began to abuse their wards.

But new scholarship is bringing fresh subtlety to psychologists’ understanding of when power leads people to take ethical shortcuts—and when it doesn’t. Indeed, for some people, power seems to bring out their best. After all, good people do win elective office, says Katherine A. DeCelles, a professor of management at the University of Toronto, and no few business executives want to do good while doing well. “When you give good people power,” DeCelles says she wondered, are they more able than others “to enact that moral identity, to do what’s right?”

In a study recently published in the Journal of Applied Psychology, DeCelles and her co-authors found that the answer is yes. People’s sense of “moral identity”—the degree to which they thought it was important to their sense of self to be “caring,” “compassionate,” “fair,” “generous” and so on—shaped their responses to feelings of power.

DeCelles and her colleagues developed moral identity scores for two groups, 173 working adults and 102 undergraduates, by asking the participants to rate how important those ethically related attributes were to them. The researchers had some participants write an essay recalling an incident in which they felt powerful, while others wrote about an ordinary day. Then the participants took part in lab experiments to probe how they balanced self-interest against the common good.

The undergraduates were told they shared a pool of 500 points with other people, and they could take between zero and ten points for themselves. The more points they took, the better their odds of winning a $100 lottery. But if they took too many—there was no way of knowing what that tipping point was—the pot would empty and the lottery would be called off.

The participants who had just written about an ordinary day each took roughly 6.5 points, regardless of their moral-identity score. But among those who had been primed to think of themselves as powerful, the people with low moral-identity scores grabbed 7.5 points—and those with high moral-identity scores took only about 5.5.

In surveys, the last group showed a greater understanding of how their actions would affect other people, which is the crucial mechanism, DeCelles says. Power led them to take a broader, more communally centered perspective.

The experiment involving the adults found a similar relationship between moral identity, ethical behavior and innate aggressiveness. Assertive people who scored low on the moral-identity scale were more likely to say they’d cheated their employer in the past week than more passive types with similar moral-identity scores. But among those with high moral-identity scores, the assertive people were less likely to have cheated.

In sum, the study found, power doesn’t corrupt; it heightens pre-existing ethical tendencies. Which brings to mind another maxim, from Abraham Lincoln: “Nearly all men can stand adversity, but if you want to test a man’s character, give him power.”