Power Corrupts…And Testosterone Corrupts Absolutely?

Power and testosterone appear to be a corrosive brew.

Posted Jan 25, 2015

We’re all familiar with his cynical warning about leaders: “Power tends to corrupt and absolute power corrupts absolutely.” It was made in 1887 by John Emerich Edward Dalberg, a British historian more famously known as “Lord Acton.” 

 Lord Acton
Lord Acton (aka, John Emerich Edward Dalberg-Acton, 1st Baron Acton)

The good Lord was attacking the morality, or lack thereof, of medieval Popes in a letter to an archbishop of the Church of England. It seems Acton didn’t approve of the idea of using torture chambers to enforce church doctrine. And these brutal Popes were part of his broader belief that “Great men are almost always bad men,” with “great” meaning holding authority or influence – like Popes, and, it seems fairly clear, political leaders. 

If he has one, this saying is Lord Acton’s claim to fame (I’m pretty sure not many people sprinkle their conversations with some of his other nuggets like “Self-preservation and self-denial, the basis of all political economy”).

But is this classic saying true?


That’s what a team of Swiss researchers set out to test. They defined corruption as a leader using his power for personal gain to the detriment of the common good. Then they conducted two clever experiments to see if giving a leader more power made the leader more likely to take advantage of his power by distributing more resources to himself than his followers.

In the first study, they did this by randomly appointing “leaders” from a group of Swiss business students and assigning them one “follower” (less power) or three “followers” (more power) from the remaining students. They then gave the leaders an amount of money to share with their followers in which the leaders got to unilaterally decide between a range of choices regarding being more selfish with the money (i.e., the leader keeps more of the money and is more corrupt) to less selfish (i.e., the leader gives more of the money to the followers and is less corrupt). In addition to manipulating power by giving the leaders different numbers of followers, the researchers also manipulated it by giving some of the leaders three options for distributing the money (less power) and some four (more power).

 power and corruption

Sure enough, they found that leaders who had more power (more followers and more options for distributing the money) were more likely to keep more of the money for themselves than the leaders who had less power. So according to this experiment it appears Lord Acton was at least partially right: Power tends to corrupt.


But these researchers did two other neat things in their next study using different Swiss business students. First, they noted that testosterone, a hormone that is required for good health, has been linked with anti-social and egocentric behavior as well as social dominance (yes, testosterone has a bad reputation; see what happens to testosterone after political competitions). These characteristics seem like reasonable contributors to corruption, so the research team measured the testosterone levels of their leaders several times during the experiment.

Second, they asked all their research subjects before starting the same task of distributing money and assigning leader-follower positions how much of the money a “responsible leader” should share with followers. They found and announced to their research subjects before they began that only 3% said leaders should take the selfish option. So the leaders knew before making their decisions on how much to share that only a small number of their fellow subjects believed choosing the selfish option was responsible.

After some complex statistics they found really interesting results. First, they found again that more powerful leaders were more selfish. No surprise there until you factor in that the more powerful leaders were more than twice as likely as the less powerful leaders to choose the selfish sharing option after they had just said that was not the responsible thing to do. That’s right: widespread hypocrisy or, in these researchers’ term, corruption. Power does indeed corrupt.

Second, they found that testosterone also mattered. In this case, high testosterone made high-power leaders even more selfish. Power and testosterone appear to be a corrosive brew.


This study clearly supports Lord Acton’s belief that power corrupts. I don’t know if absolute power corrupts absolutely, but this study also shows that testosterone absolutely corrupts the powerful. 

- - - -

For more information on this research, you can see a very nice video produced by the authors (14 mins on YouTube)

Or you can see information on the article: "Leader corruption depends on power and testosterone," Samuel Bendahan, Christian Zehnder, François P. Pralong, & John Antonakis. The Leadership Quarterly (Available online 23 September 2014). 

In addition to writing the "Caveman Politics" blog for Psychology Today, Gregg is the Executive Director of the Association for Politics and the Life Sciences and an Associate Professor of Political Science at Texas Tech University. You can find more information on Gregg at GreggRMurray.com

If you enjoyed this post, please share it by email or on Facebook or Twitter.

Follow Gregg on:  FacebookTwitterGreggRMurray.com