Power, Corruption, and Human Nature

If absolute power always corrupts, why is Mugabe's Zimbabwe more corrupt than Harald V's Norway?
If absolute power always corrupts, why is Mugabe’s Zimbabwe more corrupt than Harald V’s Norway?

CIPE prioritizes combatting corruption because it is often at the heart of a nation’s systemic challenges. The top ten countries on the global corruption index run the gamut of the world’s most troubled zones: Somalia, Sudan and South Sudan, Yemen, Afghanistan, Libya and Iraq.

Perhaps this list surprises no one. Characterized by decades-long dictatorships and/or civil wars, it is expected that these countries would be unstable and corrupt. But should we take that causality for granted? After all, which came first: the dictator, the civil war, or the failed state?

A recent study at the University of Lausanne, Switzerland, may point to the answer — one that sheds light on the very core of what CIPE’s anti-corruption efforts face in these and other regions that lack the rule of law. The clinical experiment, organized and conducted by the Faculty of Business and Economics and published in the peer-reviewed journal The Leadership Quarterly, studied the effects of power in a causal way using variants of the dictator game.

According to lead researcher John Antonakis, the experiment proves Lord Acton’s 19th century truism that “Power tends to corrupt and absolute power corrupts absolutely.” The experiment gave randomly selected “leaders” control over a specially designed pot of money: the more money a leader removed from the pot, the more also fell out of the bottom and therefore the less was leftover to be shared among the follower or followers.

For example, if a leader chose to take just $210, an additional $210 would remain in the pot and the entire group (read: state, society, or company) would benefit from $420. In theory, this total amount could be dedicated to public investment projects in the case of a state or product development in the case of a company. If, however, the leader opted to take more, $270 for example, the total left to the public decreased to only $130 and the overall benefit to the group decreased to $400. Whereas this may seem like bad math at first, it is in fact an appropriate experimental representation of choices real leaders face: by siphoning off state or company funds, a leader does more than steal that dollar amount; he or she also takes that capital out of commission and therefore makes it unable to accrue investment dividends.


Prior to the experiment, participants were asked to vote on what kind of money management scheme a leader should strive for. Most respondents believed, at least when they are not in the position of authority, that a leader should benefit slightly by taking a little bit extra out of the pot. Perhaps we can read this as the equivalent of NYC executives having a car service or members of congress receiving better health care than other federal employees.

Controlling for personality through random sampling, the 478 participants from the business school of a Swiss university were then divided into leaders and followers.  It is important to note the real-life stakes of this experiment: participants had a chance to walk away with as much as $100 in cash.

The experimenters found that power did indeed corrupt even the most honest leaders. Power, according to Antonakis, “acted like a drug:” even in small doses, it blocked leaders from feeling guilty for making decisions that did not help the overall good. Significantly, the more followers a leader had the more likely they were to abuse power. Over time it became easier to abuse power – even for those who had previously and publicly stated that that leaders should not take so much money that they harmed the overall good. In fact, comparing results from a personality test participants had taken 6 weeks prior to the experiment, the test demonstrated that there are no personality characteristics that shielded leaders from corruptibility. It is worth noting, however, that there are physiological determinants: individuals with high levels of testosterone were more likely to behave greedily. While fascinating, anti-corruption policies cannot regulate testosterone and therefore this finding does not provide much of a solution to this very human problem.

So if power inevitably corrupts even the pure of heart and honest-intentioned leaders, how can we – citizens of our various countries, members of the international development community, and CIPE specifically – combat human nature? And why, if power corrupts, is Harald V, the King of Norway, not as corrupt as Zimbabwe’s President Robert Mugabe?

As the Lausanne researchers conclude and anti-corruption community has long recognized, the answer lies in healthy institutions with built-in transparency and oversight mechanisms. In other words, robust governance is more important than strong governments. Contrary to the assertions of dictators throughout the ages, there is no such thing as a nation that is ready or not ready for democracy. Rather, there are strong institutions and there are weak institutions. In terms of corporate governance, there are strong compliance programs and there are weak compliance programs.

We have long known that transparent, pluralist institutions were the best defense against the corrupt tendencies of the Mohamed Suhartos and Jean-Claude Duvaliers of the world. Thanks to this latest study, we now know they’re necessary to rein in the Scandinavians too. For our part, CIPE continues to work with local organizations to battle corruption by strengthening institutions. With its Anti-Corruption/Value Chain projects and recently published Compliance Guidebook for mid-sized countries in emerging markets, CIPE also is building the capacity of firms in emerging markets to build internal anti-corruption programs. Without demonstrated efforts to do so, companies and countries alike cannot hope to attract investment or join global supply chains.

Sarah Ali is a Program Assistant for the Middle East & North Africa at CIPE.