The psychology of power
Absolutely
Power corrupts, but it corrupts only those who think they deserve it

Jan 21st 2010

REPORTS of politicians who have extramarital affairs while complaining about the death of family values, or who use public funding for private gain despite condemning government waste, have become so common in recent years that they hardly seem surprising anymore. Anecdotally, at least, the connection between power and hypocrisy looks obvious.

Anecdote is not science, though. And, more subtly, even if anecdote is correct, it does not answer the question of whether power tends to corrupt, as Lord Acton’s dictum has it, or whether it merely attracts the corruptible. To investigate this question Joris Lammers at Tilburg University, in the Netherlands, and Adam Galinsky at Northwestern University, in Illinois, have conducted a series of experiments which attempted to elicit states of powerfulness and powerlessness in the minds of volunteers. Having done so, as they report in Psychological Science, they tested those volunteers’ moral pliability. Lord Acton, they found, was right.

In their first study, Dr Lammers and Dr Galinsky asked 61 university students to write about a moment in their past when they were in a position of high or low power. Previous research has established that this is an effective way to “prime” people into feeling as if they are currently in such a position. Each group (high power and low power) was then split into two further groups. Half were asked to rate, on a nine-point morality scale (with one being highly immoral and nine being highly moral), how objectionable it would be for other people to over-report travel expenses at work. The other half were asked to participate in a game of dice.

The dice players were told to roll two ten-sided dice (one for “tens” and one for “units”) in the privacy of an isolated cubicle, and report the results to a lab assistant. The number they rolled, which would be a value between one and 100 (two zeros), would determine the number of tickets that they would be given in a small lottery that was run at the end of the study.

In the case of the travel expenses—when the question hung on the behaviour of others—participants in the high-power group reckoned, on average, that over-reporting rated as a 5.8 on the nine-point scale. Low-power participants rated it 7.2. The powerful, in other words, claimed to favour the moral course. In the dice game, however, high-power participants reported, on average, that they had rolled 70 while low-power individuals reported an average 59. Though the low-power people were probably cheating a bit (the expected average score would be 50), the high-power volunteers were undoubtedly cheating—perhaps taking the term “high roller” rather too literally.

Taken together, these results do indeed suggest that power tends to corrupt and to promote a hypocritical tendency to hold other people to a higher standard than oneself. To test the point further, though, Dr Lammers and Dr Galinsky explicitly contrasted attitudes to self and other people when the morally questionable activity was the same in each case. Having once again primed two groups of participants to be either high-power or low-power, they then asked some members of each group how acceptable it would be for someone else to break the speed limit when late for an appointment and how acceptable it would be for the participant himself to do so. Others were asked similar questions about tax declarations.
Only the little people pay taxes…

In both cases participants used the same one-to-nine scale employed in the first experiment. The results showed that the powerful do, indeed, behave hypocritically. They felt that others speeding because they were late warranted a 6.3 on the scale whereas speeding themselves warranted a 7.6. Low-power individuals, by contrast, saw everyone as equal. They scored themselves as 7.2 and others at 7.3—a statistically insignificant difference. In the case of tax dodging, the results were even more striking. High-power individuals felt that when others broke tax laws this rated as a 6.6 on the morality scale, but that if they did so themselves this rated as a 7.6. In this case low-power individuals were actually easier on others and harsher on themselves, with values of 7.7 and 6.8 respectively.

These results, then, suggest that the powerful do indeed behave hypocritically, condemning the transgressions of others more than they condemn their own. Which comes as no great surprise, although it is always nice to have everyday observation confirmed by systematic analysis. But another everyday observation is that powerful people who have been caught out often show little sign of contrition. It is not just that they abuse the system; they also seem to feel entitled to abuse it. To investigate this point, Dr Lammers and Dr Galinsky devised a third set of experiments. These were designed to disentangle the concept of power from that of entitlement. To do this, the researchers changed the way they primed people.

A culture of entitlement

Half of 105 participants were asked to write about a past experience in which they had legitimately been given a role of high or low power. The others were asked to write about an experience of high or low power where they did not feel their power (or lack of it) was legitimate. All of the volunteers were then asked to rate how immoral it would be for someone to take an abandoned bicycle rather than report the bicycle to the police. They were also asked, if they were in real need of a bicycle, how likely they would be to take it themselves and not report it.

The “powerful” who had been primed to believe they were entitled to their power readily engaged in acts of moral hypocrisy. They assigned a value of 5.1 to others engaging in the theft of the bicycle while rating the action at 6.9 if they were to do it themselves. Among participants in all of the low-power states, morally hypocritical behaviour inverted itself, as it had in the case of tax fraud. “Legitimate” low-power individuals assigned others a score of 5.1 if they stole a bicycle and gave themselves a 4.3. Those primed to feel that their lack of power was illegitimate behaved similarly, assigning values of 4.7 and 4.4 respectively.

However, an intriguing characteristic emerged among participants in high-power states who felt they did not deserve their elevated positions. These people showed a similar tendency to that found in low-power individuals—to be harsh on themselves and less harsh on others—but the effect was considerably more dramatic. They felt that others warranted a lenient 6.0 on the morality scale when stealing a bike but assigned a highly immoral 3.9 if they took it themselves. Dr Lammers and Dr Galinsky call this reversal “hypercrisy”.

They argue, therefore, that people with power that they think is justified break rules not only because they can get away with it, but also because they feel at some intuitive level that they are entitled to take what they want. This sense of entitlement is crucial to understanding why people misbehave in high office. In its absence, abuses will be less likely. The word “privilege” translates as “private law”. If Dr Lammers and Dr Galinsky are right, the sense which some powerful people seem to have that different rules apply to them is not just a convenient smoke screen. They genuinely believe it.

What explains hypercrisy is less obvious. It is known, though, from experiments on other species that if those at the bottom of a dominance hierarchy show signs of getting uppity, those at the top react both quickly and aggressively. Hypercrisy might thus be a signal of submissiveness—one that is exaggerated in creatures that feel themselves to be in the wrong place in the hierarchy. By applying reverse privileges to themselves, they hope to escape punishment from the real dominants. Perhaps the lesson, then, is that corruption and hypocrisy are the price that societies pay for being led by alpha males (and, in some cases, alpha females). The alternative, though cleaner, is leadership by wimps.
Read