Back to the Office: Whose Needs Are We Meeting?

When a company requires employees to return to the office, its leaders often cite collaboration, innovation or productivity. Their preference for working in person may also serve needs of their own. In a paper published in Organizational Behavior and Human Decision Processes, Marissa S. Shandell, Courtney E. Elliott and Adam M. Grant examine the link between leaders’ narcissistic traits and their opposition to remote work.

In a first study of 259 Fortune 500 CEOs, the researchers found a correlation of 0.24 between an indirect measure of narcissism and the opposition CEOs publicly expressed towards remote work. The measure took into account, among other things, how much the CEO earned relative to the company’s second-highest-paid executive and how prominently the CEO’s photo appeared in the annual report. In a second study, 359 U.S. managers completed questionnaires on three occasions. The researchers found the same association even after accounting for extraversion and trust in others.

What might make having employees in the office particularly appealing to some managers? Managers who scored higher on narcissistic traits said they placed greater value on exercising authority and being admired by their teams. Each of these motivations was, in turn, associated with stronger opposition to remote work. The researchers suggest that working in person may give these managers more opportunities to steer conversations and see signs of appreciation from their employees.

Managers can have legitimate reasons for requiring employees to come into the office. A return-to-office policy alone tells us little about a leader’s personality. The study does, however, suggest that leaders may bring their own needs to decisions about where people work. Before requiring a return to the office, it is worth spelling out which problems the policy is meant to solve and how its results will be assessed. A manager’s desire to feel more in control is hardly a measure of organizational performance.