By Trevor A. Foulk and Klodiana Lanaj JUNE 13, 2017, https://hbr.org/

Have you ever interacted with a supervisor who was on a “power trip,” and come away feeling disrespected, hurt, or upset? You’re far from alone. Abundant research shows that when people feel powerful, they tend to abuse others, supporting the notion that “power tends to corrupt, and absolute power corrupts absolutely.” Many studies show that abused employees suffer distress, perform worse, are less creative, and are more likely to quit their jobs. Abusive power holders, on the other hand, seem immune to their own negative behaviors: Research suggests that they continue about their day as if nothing has happened.

But what if this isn’t the entire story? What if power comes at a cost for powerful leaders, too?

We investigate this possibility in a study of 108 managers, whom we surveyed for 10 consecutive workdays. These managers worked in different organizations in industries including health care, engineering, education, and banking. Although the leaders had some degree of structural power, research suggests that one’s sense of power fluctuates daily and can be triggered by certain events (for example, reminders of being in charge, attending a meeting for management, or making a hiring or firing decision). To manipulate daily power, we randomly assigned participants to five ”power days” and five control days. On power days managers were asked to think and write about a time when they had power over someone else and to complete several short word exercises (such as “Complete the following word fragment with the first real word that comes to mind: p_w_r”). Research shows that both of these techniques activate a sense of power. On control days managers completed similar exercises that do not prime power. For example, they described their commute to work that day and completed neutral sentences (“Complete the following word fragment with the first real word that comes to mind: i_la_d”).

Participants interacted with their coworkers, managers, and customers over the course of the study, which gave us ample opportunities to assess the leaders’ behaviors multiple times each day, both at work and at home. We emailed survey links to participants three times each day: We manipulated power with the morning survey, which participants completed at approximately 8:28 AM; we assessed daily abuse and perceived incivility from others at the end of the workday, at approximately 5:23 PM; and we measured relaxation and need fulfillment at home at approximately 8:24 PM.

Consistent with literature supporting the “corrupting” nature of power, we found that on days when participants felt powerful, they reported having more negative interactions with others. These negative interactions came in two forms: Participants reported engaging in more abusive behaviors toward others (they yelled or swore more, behaved in a rude manner, or made fun of coworkers), but they also perceived more incivility from others (they felt that coworkers addressed them more unprofessionally, talked to them in a condescending tone, and paid little attention to their statements and opinions). These findings are well aligned with prior research on the effects of psychological power. Power causes us to view others as psychologically distant, inconsequential, and as a means to our ends, all of which explain why power enhances abuse. At the same time, psychological power makes us feel special and more deserving of others’ attention, respect, and favor. These inflated expectations often are not met, which may explain why power holders experience more incivility from others.

Although our findings depict a rather bleak picture for the effects that power has on leaders, there are several silver linings. First, not everyone was affected by psychological power to the same degree. For example, using a personality survey, we measured leaders’ trait-level agreeableness and found that agreeable leaders abused others less when experiencing power. Agreeable people care about others’ well-being and make it a priority to maintain positive relationships with coworkers. Their innate desire to maintain social harmony may explain why agreeable leaders do not engage in as much abuse when they experience power. The “absolute” corrupting nature of power, therefore, may not be as absolute as we have come to believe.

Second, power holders were hurt by their own bad behaviors. When we surveyed leaders in the evening, at home, we noticed that powerful leaders who engaged in abuse and who perceived more incivility from others reported feeling less fulfilled by their workday — they felt less competent, less able to relate to others, and less autonomous. Additionally, abusive power holders were less able to kick back and relax at home. These findings suggest a “power hangover” — the effects of power experienced during the workday reduced leaders’ well-being in the evening at home. Overall, our power manipulation accounted for 15% of a manager’s daily variability in need fulfillment and 2% of their variability in relaxation, suggesting that even a minor manipulation of power in the morning can have noticeable effects on well-being that last until the evening.

Why does power hurt leaders’ well-being, even at home? One reason might be that power-induced negative interactions threaten leaders’ ability to maintain power over others at work; abused followers may retaliate or may not comply with abusive leaders’ demands in the future. It is also possible that powerful leaders who reflect on their negative behaviors at home may feel guilty because they recognize that they violated social norms for proper work conduct.

Although many people would prefer to have more power and influence, our study shows that power not only prompts us to do bad things but can also make us feel worse at home. It is prudent, therefore, to temper our rosy view of power and to start considering its costs as well.

It is important for people in positions of power to become aware of the negative influence that power can have on our interactions with others, as well as on our own well-being. It may help to ask trusted mentors and colleagues to keep us accountable for how we behave at work. From an organizational perspective, it could be helpful to assign agreeable employees to powerful roles, because agreeable leaders are less susceptible to power-induced abuse.

It’s often said that “with great power comes great responsibility.” As we show in this study, greater power also seems to come with greater suffering.

Trevor A. Foulk is an assistant professor in the Management & Organization department at the Robert H. Smith School of Business at the University of Maryland.