Jon Maner, a professor of management and organizations at Kellogg School of Management, has studied a specific breed of bad boss — those who intentionally sabotage their teams’ cohesion in order to protect their own status as leader.

Maner’s research shows that leaders will intentionally sideline high-performing team members, limit communication and social bonding among team members, or compile ill-matched teams if they think it will help ensure their own place at the top.

“It can cause the group to fall apart at a basic level,” Maner says. “If you have people who don’t like each other and aren’t allowed to communicate effectively with one another, then really, you don’t have a group at all anymore.”

Maner and collaborator Charleen Case, a doctoral student at the Kellogg School, found that leaders who were driven by a desire for power (or dominance motivated) were more likely to undermine a group’s communication and cohesion than those who were motivated by a desire for respect (or prestige motivated). Those power-hungry leaders were most inclined to behave this way when they were told that the power hierarchy in the group was unstable and they may lose their position at the top. And they were most likely to undermine group cohesion by isolating the one highly skilled member of the group.

The irony, of course, is that this behavior is being perpetrated by the person who should be most invested in and most skilled at getting his or her team to work together in order to be productive.

Beware the Threatened, Power-Hungry Boss

Maner and Case led undergraduate students in one study to believe that they would be leading a group in performing a verbal task. The better the group scored, the more prizes it would win.

Participants were told that one of their group members in particular was very highly skilled at the task. Participants were then assigned to one of three experimental conditions. In the first, they were told that as the leader, they would be responsible for supervising the task and dividing the prizes among the group. In the second, they were told that they would supervise the task and allot prizes, but also that the hierarchy was malleable and someone else could become leader. The third condition was an egalitarian control group where there was no leader and all group members would share the prizes equally.

The researchers sought to answer two main questions: Which leaders under what scenarios are most likely to sabotage their groups’ communication and cohesion—even knowing that cohesion can improve a group’s performance? And will they be more likely to do so by isolating the highly skilled group member?

They found that, as predicted, participants in the malleable hierarchy who had previously scored highly on a test assessing their desire for power were the most likely saboteurs. And they were most likely to go after that one highly skilled group member.

“It’s surprising to me just how willing leaders are to really undermine group success in favor of their own power,” Maner says. “These talented, highly skilled group members are in one of the best positions to help the group succeed. But rather than being seen as a valuable ally, they’re instead seen as a threat by leaders who are afraid of losing their power.”

Curtailing These Bad Behaviors at Work

These types of interactions are more than just theoretical; they happen in offices every day. In fact, Maner’s research initially grew out of complaints from a friend about her boss’s bad behavior.

So how do organizations, which want teams to function as cohesively as possible, prevent this sort of sabotage from above?

One key is making sure that leaders’ job security is contingent upon the success of their group and ensuring those leaders know that they will be held accountable for their actions, Maner says.

“If leaders knew that their decisions were public and could have ramifications for the support they receive, I think that might undercut some of this corrupt behavior,” Maner says.

Organizations could also institutionalize lines of communication and collaboration among teams, which would make it harder for a bad boss to interfere with them, Maner suggests.

The question of how to ensure that bosses feel their power is stable is trickier, since organizations need to balance creating an environment where bosses feel secure with the ability to change leaders when the situation calls for it.

Maner suggests having periods of stability, perhaps two or three years, where bosses know their jobs are secure, interspersed with times when leadership can change if warranted, kind of like the system of having presidential elections every four years.

“What might help leaders perform at their best is knowing that they’re not going to lose their job today or tomorrow, that they can really follow through on whatever vision it is they have and if it doesn’t work out, it doesn’t work out,” he says. “But at least they’ve really had a chance to put their vision into action.”

Finding the Best Leaders

There is also the question of finding leaders who are more interested in receiving prestige and respect than having power. This is a particular challenge since many people who desire power tend to self-select into positions of leadership, while prestige-motivated workers may be happier working in less flashy positions.

“A real trick for organizations is to identify who those people are and raise them up into positions of leadership, whether or not they ask for it, because they might not always be as inclined as power-hungry people are to seek high-status positions in their organization,” Maner says.

But Maner cautions against seeing these prestige-motivated leaders as a cure-all.“Our work has painted a pretty magnanimous portrait of prestige-oriented leaders,” Maner says, “but I think that’s probably an oversimplification.”

In future research, he hopes to explore how prestige-motivated leaders make decisions when forced to choose between what will make the group happy and what is best for the organization.

“We have some preliminary evidence that they will undermine the goals of the group if it means maintaining prestige within the eyes of their subordinates,” he says.

In the end, the goal of his research is to help organizations function smoothly and productively.

“The ultimate goal here is to figure out how to help groups perform better,” he says. “That means selecting better leaders and bringing out the best in leaders. By understanding the motives that might drive leaders to behave in ways that hurt their organization, we’re better armed to combat those behaviors.”