Author Talks: Why your boss may indeed be a psychopath

In this edition of Author Talks, McKinsey Global Publishing’s Raju Narisetti chats with political scientist Brian Klaas, an associate professor at University College London and Washington Post columnist who also advises governments, political campaigns, and companies. His new book, Corruptible: Who Gets Power and How It Changes Us (Scribner, November 2021), takes an unflinching look at power and the people who have it—from terrorists, to dictators, and even regional managers. Across the board, Klaas found similar reasons why psychopaths, narcissists, and other corruptible people rise to the top. An edited version of the conversation follows.

Does power corrupt or are corrupt people drawn to power?

Over the last decade or so, I have traveled around the world studying the worst people in power. One of the things that’s been a peculiar experience for me and that spurred me to write this book was that I would go and interview somebody—a war criminal, a former despot, somebody who was awful in a faraway country—and I’d come back, and I would talk to my friends or family members. They would say, “That’s just like my homeowner’s association tyrant,” or “That’s just like the guy in midlevel management in my company.”

So, what I tried to figure out in this book, and what spurred the research, was whether there is a commonality here. Do we have a universal human trait of corruptible people seeking power, or does power universally turn people corrupt? The remedy to those two diagnoses is quite different. There’s a third dimension that I focus on extensively in the book, which is that if corruptible people seek power more than the rest of us and if power corrupts—both of which, by the way, are true—how do we reengineer society and redesign systems to make sure that incorruptible people seek and obtain power and that power actually doesn’t corrupt people but purifies them?

That’s the mandate for the book: it’s trying to figure out how we can engineer a better world by navigating this extremely complicated world of power and systems to get better outcomes. I think it’s so interesting to have done this research, because it applies in every realm of human society, from the homeowner’s association tyrant all the way up to the former despots that started my work.

Is now the ‘postpandemic’ moment to engineer a better future?

Yes. When you look back in history and you try to think about the grand sweep of history, we have all these awful people who crop up in history books. We just figure that that’s the fate of our past selves, but this moment right now feels like a hopeful moment in which we have the technology, the tools, and the wherewithal to try to create a better world. We’ve seen this happening in all sorts of realms of human society, such as technical advancement to deal with problems like child mortality and better adaptations to all sorts of environmental problems.

Yet when we ask about power, pretty much everybody I talk to is still disappointed when they look at society. They say things like, “When I look around to the people I know, they’re good and decent people, and yet the headlines are dominated by abuse of power, with awful people wielding power unjustly.” What I tried to do was to synthesize lots of different realms of human knowledge—everything from evolutionary biology, psychology, political science, and behavioral economics, to neuroscience—to try to understand what’s actually happening.

Right now, in this turning point in history that we find ourselves in—postpandemic, after the rise of populism and all sorts of upheavals in society—this is the moment to try to think back to the past of history and think, “Can we actually engineer a better future?

I think that the thing that has been a stumbling block for so long is that we’ve all dealt with this in these little silos; the political scientists focus on one aspect, the evolutionary biologists focus on another, and the anthropologists on a different one. I think we’re all ultimately tinkering around this big question, which is who does get power, how does it change us, and how can we get better people into power? The ultimate moment for that is right now, when everything is changing. The pace of human adaptation to societal change has never been faster. Right now, in this turning point in history that we find ourselves in—postpandemic, after the rise of populism and all sorts of upheavals in society—this is the moment to try to think back to the past of history and think, “Can we actually engineer a better future?”

How and why do we pick the leaders we pick?

There are two things that really stood out for me when it comes to the big surprises. The first thing that I find fascinating is this idea of how and why we select certain leaders. One of the most interesting studies that’s been replicated—it’s been vetted, it’s been published in top scientific journals—is a study in which children were shown two pictures of faces, and they were told to choose someone to be the captain of their imaginary ship in this computer simulation. They were given no other information. What was amazing was that the children didn’t realize that one of the faces was the winner of a French election and one of the faces was a loser, a runner-up in that election. The children—like clockwork, the overwhelming majority of the time—picked the winner to captain their ship, even though they had no idea of the context of the face.

So, one of the things that I think is fascinating—drawing on neuroscience, evolutionary biology, evolutionary psychology, and psychology itself—is that leadership selection is far less rational than we think it is. Understanding those cognitive biases that go into it is crucial to counteracting them and getting better, rational choices in determining who leads society.

Wait, so power actually alters your mind physically?

The second thing that stood out, which I found absolutely mind-blowing, was how power not only physically changes your psyche but actually physically changes your mind. I looked into a few different realms of this. One is with nonhuman primates—animal research crops up a few times in the book.

When you look at baboons, for example, you can find that there’s actually significantly higher levels of stress, and therefore genetic aging, that’s happening the lower down the hierarchy you go. The stressed, low-ranking baboons age faster. It gets better as you rise through the ranks, until you get to the alpha male baboon, at which point there’s a target on their back. They’re constantly stressed, and they age really fast. The lesson here, which applies to human society, is that it’s good to be at court, but perhaps not good to be the king.

We have studies that show this in CEOs and in presidents. In CEOs, we use machine learning to show that aging happens faster in faces for those who captain industries during times of particular distress for their industry. For presidents, there’s a study, an amazing study, that looks at 200 years of elections across 17 countries and finds that the loser of presidential elections lives 4.4 years, on average, longer than the winner of presidential elections. It’s a bit of consolation for those who lose; perhaps they die a bit slower than those who beat them in the election.

If ‘watched people are nice people’ are we watching the right people?

One of the things that’s really jarring to me is that so much corporate surveillance of employees involves low-ranking figures in the company who simply don’t have the capacity to bring the company down. When you think about all of the corporate scandals of the last several decades, they’re usually happening behind closed doors. Yet in the postpandemic world of working from home and all the increasing adaptations with technology, the ordinary employees are being watched in ways that often border on the dystopian—there are even devices that have weight sensors in chairs, and webcams that are constantly monitoring whether you’re taking an extra five minutes for your coffee break.

What I argue is, when you think about systemic risk to a company, to an organization, it’s usually people who operate in corner offices behind closed doors who can do the most damage if they are not subject to proper oversight. The balance here, of course, is to avoid creating a different kind of dystopian society, in which everybody is constantly fearing the omniscient oversight of some unseen being that’s watching them all the time and that they can’t escape from.

You have to be careful about how you calibrate oversight, but the gaze should look up rather than down. The lesson of corporate malfeasance, abuse, and embezzlement is that, typically, those who are most watched in corporate societies are least prone to actually doing the kind of catastrophic damage that most companies are most worried about.

You have to be careful about how you calibrate oversight, but the gaze should look up rather than down. The lesson of corporate malfeasance, abuse, and embezzlement is that, typically, those who are most watched in corporate societies are least prone to actually doing the kind of catastrophic damage that most companies are most worried about. I think you have to calibrate very carefully when it comes to surveillance, because the problem is that you don’t want to be watched all the time. People need to be creative. They need to toss around ideas that aren’t ready for prime time. They don’t want to be constantly recorded.

The principle that I’m advocating for is that if you are going to have some level of oversight, audits, or actual checks, they don’t need to be constant. They just need to be gazing more upward; they need to be more tailored toward the people who can actually ruin the company, the people who actually have access to embezzling millions or billions, instead of the people who might steal a paperclip.

There needs to be a recalibration of the efforts of surveillance in a way that can provide oversight without having a dystopia emerge in the workplace where everybody is constantly being filmed or monitored 100 percent of the time.

What exactly is randomized integrity testing?

Randomization is a really powerful tool that isn’t used enough, in my opinion. One of the people I spoke to in researching the book was the former head of Internal Affairs of the New York Police Department [NYPD]. He pioneered this method called randomized integrity testing, in which a police officer would go into a supposed crime scene, and there would be something like $20,000 in cash on the table and a bunch of drugs. What the officer wouldn’t know is that the whole apartment was wired with cameras and microphones. If they pocketed the cash—let’s say they took $6,000 and left $14,000 to report—they would be arrested or fired from their job.

The beautiful thing about this was that the NYPD did 500 of these tests, but when they surveyed the police officers, 12,000 police officers said, “Yes indeed, I was subjected to randomized integrity testing.” There were 11,500 times where a cop encountered a real crime scene and thought that it was fake and staged. So, they started behaving better, and the rest of the force also feared that they could be stung because of a really small amount of these random tests, which were only angled at the people who are in positions where they can abuse their authority the most.

Again, you don’t want to have a dystopian world in which the break room fridge is baited and filmed. If you tailor it appropriately, this small number of randomized tests can actually create a really powerful deterrent effect, where those in positions of authority will think twice before engaging in abuse or deceit or embezzlement.

Making people have a little fear of the idea that they might face consequences if they engage in some truly malicious behavior in the workplace can pay huge dividends with a very small number of tests, due to the power of randomness.

Can Sweden’s speeding fine approach succeed in companies?

In Sweden, they have this system in which, basically, you have both the carrot and the stick when it comes to speeding, and I think it provides some lessons for business too. If you speed and you get caught, you have to pay a fine, but the fine doesn’t go to the government—it goes to people who are obeying the speed limits.

One of the things that we often don’t do enough is incentivizing good behavior rather than just punishing bad behavior. Now, the other thing that’s important to take away from some of these examples is that they’re not rocket science, right? Sweden didn’t come up with some ingenious idea; they just tweaked the system in a small way, and it started to pay significant dividends.

My hope in writing this book is that some very powerful people who are very well-intentioned should start to think more critically about the systems that are currently on autopilot in their organizations—whether it’s a business or a police department or whatever it is—and start to think, “We could do it slightly differently, but that would pay huge dividends.” These small examples can add up to quite a lot of profound changes in making the world just that bit better.

We have to think carefully about redesigning systems. If we do that, I actually think we can fix this problem. We have designed, in ways that we promote people or in ways that we hire people, a system that caters, frankly, to extroverted, narcissistic Machiavellian psychopaths.

How can more nice, nonpsychopathic people get power?

In the book, I provide ten principles that I think can weed out the corruptible people who seek power, and also attract more incorruptible people into power. Now, I’m not naive; my research has clearly shown me that bad, abusive people are disproportionately likely to seek power, disproportionately good at getting it, and likely to become worse once they wield it, because power does indeed corrupt.

Why am I not pessimistic? The reason I’m not pessimistic is that I think there are actually quite a lot of easy fixes. I don’t mean that they’re easy in the sense of actually implementing them immediately and having them pay off like a silver bullet. I mean that we have to think carefully about redesigning systems. If we do that, I actually think we can fix this problem. We have designed, in ways that we promote people or in ways that we hire people, a system that caters, frankly, to extroverted, narcissistic Machiavellian psychopaths. The reason I say so is that a lot of the systems that are used to vet people or promote them or to hire them in the first place involve “performances,” short-term performances in which superficial charm—the two words that are most associated with psychopaths or narcissists—are on display.

Take one quick fix: at University College London, where I teach, all of the grading that I do is fully anonymized. I grade a number, not a name. It’s a really small change, but it makes it so I can’t punish a student that I dislike, and I have no idea who I’m actually grading.

The book is trying to say we need to think a little bit more creatively. We don’t need to reinvent the wheel completely; we just need to think a little bit more creatively about how we design systems that would make it less attractive for psychopathic, abusive Machiavellians to join the organization and a little more attractive for good, decent, normal people to try to rise through the ranks. We need to make the pinnacle of power attractive to those people who might not want power for power’s sake, because right now we have quite a self-selection problem where the worst among us gravitate toward the top.

We have these human tendencies, we have some broken systems, but they’re fixable, and the ultimate outcome can be a much more just and incorruptible world. That’s my hope for what the book achieves over the longer run, with some very smart people reengineering the organizations that they inhabit and run.

Author Talks

Visit Author Talks to see the full series.

Explore a career with us