Power and Moral Hypocrisy
When a certain former New York State Attorney General became New York Governor, he pledged to “change the ethics of Albany” and make “ethics and integrity the hallmarks of [his] administration.” Sure enough, he went on to fight collar crime and corruption, reduce pollution and prosecute a couple prostitution rings. Oh, but then the New York Times disclosed that this same law-and-order Governor was bedding high-priced prostitutes. So much for changing the ethics of Albany.
Power and moral hypocrisy are no strangers – we see it all the time. But why?
A few social scientists recently took a stab at finding an answer. Joris Lammers and Diederik A. Stapel (from Tilburg University) and Adam D. Galinsky (from Northwestern University) ran five experiments addressing how morality differs among the powerless and powerful.
They simulated a bureaucratic organization and randomly assigned participants to be in a high-power role (prime-minister) or low-power role (civil servant). The prime-minister could control and direct the civil servants. Next, the researchers presented all participants with a seemingly unrelated moral dilemma from among the following: failure to declare all wages on a tax form, violation of traffic rules, and possession of a stolen bike. In each case, participants used a 9-point scale (1: completely unacceptable, 9: fully acceptable) to rate the acceptability of the act. However, half of the participants rated how acceptable it would be if they themselves engaged in the act, while the other half rated how acceptable it would be others engaged in it.
The researchers found that compared to participants without power, powerful participants were stricter in judging others’ moral transgressions but more lenient in judging their own: “power increases hypocrisy, meaning that the powerful show a greater discrepancy between what they practice and what they preach.”
Joris and Adam hypothesized that this power-hypocrisy connection was due to the sense of entitlement that comes with positions of power. But what if you took away that entitlement by having participants view their power as illegitimate? In that case, the researchers posited, you would see the power-hypocrisy effect disappear.
To test their idea, they had 105 Dutch students write about an experience in which they were powerful or powerless. But half of the participants were asked to write about a time when they deserved their high or low power (it was legitimate), while the other half were asked to write about a time when they didn’t deserve their high or low power. They then had to rate the acceptability of the bike dilemma from above.
Results: when power (or lack thereof) was legitimate, the powerful also exhibited moral hypocrisy (being less moral themselves but judging others more harshly), while the powerless weren’t – just as before. But when power (or lack thereof) was illegitimate, the powerful didn’t show hypocrisy. In fact, the moral hypocrisy effect not only disappeared but was reversed, with the illegitimate powerful becoming stricter in judging their own behavior and more lenient in judging the others.
Irrationally Yours,
Dan

My latest book, The Upside of Irrationality, explores some positive and some negative ways that irrationality plays out in our lives.

Thanks for summarizing these experiments. This was a lot of information for me, and it got me wondering about my own observations. Coming from a “mid-eastern” country, I can relate to some of the observations and the results from the studies.
However, I still wonder about the following observation: In my part of the world, there is no real democracy or meritocracy, so it would be safe to say that a large number of public leaders and figures don’t really deserve to be in their powerful positions. So this means that their power was “illegitimate”, yet I am afraid the majority of these individuals lack any signs of moral judgments in most of their actions.
Does this mean that the high levels of their moral hypocrisy is contradicting the studies? (which suggests that individuals with illegitimate power have lower moral hypocrisy). Or maybe they feel a low moral hypocrisy on the inside, but they suppress these feelings and act the opposite way (as needed to defend their illegitimate powers)?
Thanks again for your great and educational posts.
Interesting research – made all the more relevant in the UK with the recent MP’s expenses scandal, attitudes to which will have a profound effect on the outcome of the election.
The question is what to do about it? Any solution would also have implications for executive pay – particularly the problem of excessive rewards for executives that fail (no doubt attributing this to factors outside their control). Any solution to the issue of moral hypocrisy will have wider implications for solving the problem of commercial hypocrisy.
Ross Shah, I think the answer is pretty simple and obvious: they believe their power is legitimate. Anything that “outsiders” might think about legitimacy of their power/position is irrelevant to them, because they “know”/believe that they gained their position legitimately.
There is a free PDF version of Bob Altemeyer’s book called, “The Authoritarians” available on his website: http://home.cc.umanitoba.ca/~altemey/. Altemeyer performed a wide variety of experiments similar to the one described in Dan’s blog since the mid-1970s. He summarizes this corpus in an easy-to-read and entertaining format that goes into great detail about the psychology of both authoritarian followers and leaders (they are quite distinct).
i’ll guess that people intuitively calculate their odds of being caught, being held responsible, being punished, and what the punishment is likely to be. the more powerful you are, the more likely you are to hold sway over those who could catch you, those who could hold you responsible, and those who could punish you.
This article is so much relevant to the reality we deal with in Israel to day – So many people who held or still hold powerful positions dare to behave in double standards and deals with public properties and budget in a way which really implies hypocrisy and corruption.
Knowing the connection, made more observant in Joris Lammers and Diederik A. Stapel (from Tilburg University) and Adam D. Galinsky (from Northwestern University) research, There are certain questions to be asked –
• What is the role of the media in shaping, creating and nourishing the connection between power-position-money in the public’s opinion and narrative?
• What kind of systems and control could confront double standards and hypocrisy ?
• What are the tools we might require to influence educating the generations to come when the role models are as they are? Do we have a chance?
Sarah Kiperwas
Lammers, Stapel and Galinsky conclude “power corrupts.” Yes, but their conclusion is only demonstrated for the situation in which power is simply given. The design of Lammers, Stapel and Galinsky’s study does not exclude the hypothesis that causality runs in the other direction. It is equally plausible that a talent for moral hypocrisy is necessary to successfully negotiate the acquisition of bureaucratic power in the first place.
Speaking of power. Look at this amazing site that discusses Authoritarians. I think alot of this research dovetails nicely into PI.
http://home.cc.umanitoba.ca/~altemey/
Hi,
This is a excellent research. But I believe that the impact and sense of guilt attached with the people who have power in illegitimate way factors into it. A person who does not have a sense of guilt for illegitimate power will not behave differently from the person with llegitimate power. Therefore the whole thing boils down to the fact, how strong you feel that the power is illegiimate.
Interesting study to summarize and share, thanks, Dan.
At first, I had similar question with Ross Shah, to which Anna made an interesting comment.
Anecdotically my intuition about illegitimate powerful in Asia is that they are more lenient to themselves as well as to others…:) Would be interesting to check how true it is. This will be useful to combat corruption in public services. In the country where I come from, it is normally accepted that everybody in power will be tempted to breach his/her integrity and those critical and powerless would change just like that once they are in power. (I disagree, of course. I don’t think “it always happen” as mentioned in the last posting.)
Dan,
Now that Stapel has admitted to fudging the data on his other papers Joris was kind enough (or compelled enough) to provide the raw data for this paper to a colleague of mine. We looked at it and wept. The line in the paper says “However, high-powered participants claimed a higher number of lottery tickets than low-power participants did” The raw numbers show that entire effect is driven by a single low-power participant who claimed he/she rolled a 16. The 16 number is more than 3 standard deviations to the left of the mean. The next highest number is 44. Take this person out and the histograms of the numbers rolled for high and low power are identical (the histogram, by the way, is conveniently omitted). And experiment 1 fails. Even if you keep this person in there, distributional assumptions are violated and t-tests are meaningless and misleading.
Even if this is not fudging data, this is dishonest research. The irony is that you report it admiringly while your commentators find it extremely intuitive. Perhaps this is a bit harsh but I would not trust any of the research by any of these authors.
Honestly yours, Nonya Moose
Dear Nonya,
I strongly disagree with your conclusions. First of all, if you delete that participants the interaction-effect is still statistically significant (p = .01). Second, simply because a participant is an outlier, it does not mean that this participant should be deleted. After all, we are not speaking of implicit measures or so. The participant must have thought long about his/her response. (Rolling a dice, thinking about it, writing it up). I don’t see why that would be a measurement error.
But what I find dishonest is that you write that you “wept” but apparently were so overcome by your tears that you did not find the time to write me. I think colleagues when disagreeing about data should first contact each other. I have exchanged these data to your colleague and we have emailed a lot about this. He was interested in my data also to test a new bootstrapping algorithm to test differences. Using this new algorithm on my data, he found that (quoting from his email):
“In sum, your Experiment 1 data hold up pretty nicely under this protocol. Given that the sampling distribution is bootstrapped from the data, the test is essentially non-parametric and should be quite robust (though perhaps not very powerful — I’m not yet sure). Hence, I think the outcome provides nice support for your Experiment 1 claims.”
Given this email I find it very strange and sad having to read the above post. I think that your statement that this is “a bit harsh” is an understatement.
Joris Lammers