Innovation is often associated with a relaxed office culture where employees feel safe to try new things and take risks, but new research from Wharton finds comfort is counterproductive in more typical situations.
In a co-authored study, management professor Peter Cappelli and his colleagues found that too much “psychological safety” in the workplace can harm the performance of typical jobs. “The idea behind psychological safety is that we should not fear ridicule or being put down in contexts where we are trying to learn or be creative. It was never meant to suggest that there should be no consequences for poor performance or not following the rules, nor was there much evidence that it was helpful in more routine jobs where creativity and learning is not a central activity,” Cappelli said. Almost every job has standards we have to meet and rules we have to follow, he added. “The scope for real creativity — whether that’s engineering, education, retail, manufacturing, or health care — is not a daily event, so maximizing psychological safety should have little payoff in typical jobs.”
For example, “you don’t want your nurse trying something different in how they administer medicine,” Cappelli noted. “You want them following the standard protocols exactly the same way all the time. And that’s true of most jobs. Even in creative jobs, there are some pretty clear guardrails to what you are supposed to be doing.”
Cappelli and his co-authors — Liat Eldor, management professor at Tel Aviv University and managing director of Wharton’s Center for Human Resources, and Michal Hodor, also a management professor at Tel Aviv University — wanted to take a deeper look at psychological safety because the concept is being misused by practitioners and advocated as something that all managers should strive to push higher.
In their paper, “The Limits of Psychological Safety: Nonlinear Relationships with Performance,” the professors analyzed hundreds of internal and external performance ratings across different businesses and jobs over time. They found that a moderate amount of psychological safety does in fact enhance performance: If you feel a real or constant threat of being ridiculed, it is bad for performance even in routine jobs. But there is a point of diminishing returns above which productivity and accuracy are jeopardized.
“I was surprised that people assumed there was not a limit to psychological safety,” Cappelli said. “They just thought more is better.”
Balancing Comfort with Accountability
Cappelli and his co-authors emphasized that it’s never OK for employees to feel physically unsafe, that is, marginalized, ridiculed, or verbally abused. Their study does not dispute that.
“I was surprised that people assumed there was not a limit to psychological safety.” – Peter Cappelli
“The creators of this idea of psychological safety were making a modest and sensible point about the downsides of inter-personal harm in creative contexts,” Cappelli said. “I think it’s gone overboard now. People are always talking about the benefits; they are never thinking about possible negative consequences.”
There appear to be two reasons why very high levels of psychological safety hurt job performance. One is that the “no bad ideas” stereotype may simply mean that very high levels of psychological safety come when we perceive there will be no consequences — or no serious ones — for poor performance. “Can you imagine getting a bad — and deserved — performance appraisal from your boss and feeling psychologically safe after it?” Cappelli said. Another is that it may focus our attention on the few creative tasks we have and away from the more important routine tasks.
The professors propose a way to moderate the negative effects of excessive comfort at work through collective accountability. When executed correctly, collective accountability is a cue that redirects workers away from individual behaviors and toward organizational goals. They pointed out that it can help bring clarity to what’s expected of each person and department, and increase a sense of obligation toward each other and the firm.
The paper noted that collective accountability can even reduce bias in people’s actions, and the professors urged more research on the positive role it can play in organizations.
“It may also be the case that the research about the limits of psychological safety climate has been sparse because more energy goes into expanding and emphasizing its importance than pointing out its limitations. This is because the former fits much more neatly with our existing paradigms,” they wrote.
Easing Managerial Expectations of Psychological Safety
Cappelli said the main takeaway for companies is to not push managers to maximize psychological safety. This is a case where “good enough” is truly the goal. Managers and supervisors need to be able to deliver a bad performance review or a reprimand without worrying about impinging on an employee’s comfort.
He reiterated that no one should be mocked, harassed, or threatened at work. But managers shouldn’t be expected to shield employees from discomfort over performance-related issues.
“It’s about what not to do. You can back off on holding managers accountable for high levels of psychological safety,” Cappelli said. “Good enough is good enough. Good enough is what you want.”
