Opinion | Critics of D.E.I. Forget That It Works

Erik Larson’s firm, Cloverpop, helps companies make and learn from decisions. When Mr. Larson and his research team compared the decision quality of individuals versus teams, they found that all-male teams outperformed individuals nearly 60 percent of the time, but gender diverse teams outperformed individuals almost 75 percent of the time. Teams that were gender and geographically diverse, and had at least one age gap of 20 years or more, made better decisions than individuals 87 percent of the time. If you’ve ever called a grandparent for advice or tested an idea with a skeptical teenager, you get what this research was trying to quantify. We often learn the most from people who think most differently from us.

Getting people to share what they know that other people don’t know is essential to collective performance. Our Harvard Business School colleague Amy Edmondson and her research collaborator, Mike Roberto, designed a simulation where five-person teams must figure out how to climb Mount Everest. Teams reporting higher feelings of group belonging repeatedly outperform other teams because their members share more of their unique information about summiting Everest.

These findings are consistent with Ms. Edmondson’s research on the performance advantages of “psychological safety,” the cultural underpinning of inclusion. Individuals, she finds, are more likely to share their views in an environment that does not belittle, or worse, punish those who offer differing opinions, particularly to more powerful colleagues. In a recent study of 62 drug development teams, Ms. Edmondson and Henrik Bresman found that diverse teams, when assessed by senior leaders, outperform their more homogenous peers only in the presence of psychological safety. More diversity is not always better – from a performance standpoint, diversity without the inclusion can actually make things worse.

Inclusion work, done well, seeks to scale these kinds of results. Among other payoffs, organizations that get inclusion right at scale seem to be smarter, more innovative and more stable. One explanation is that they can see their competitive landscape — threats, risks, opportunities — more clearly and have greater access to the full knowledge base of their people.

But achieving gains like this can feel elusive when the will to participate in D.E.I. is waning. It can be tempting to put in place superficial fixes to achieve the optics of inclusion — a primary concern of D.E.I. critics — such as reserving roles for specific demographics. This is often illegal and rarely helpful, and it provides at least one area of broad agreement in this polarized debate: a distaste for hiring and promotion schemes based on an individual’s identity. A way to correct for these concerns is inclusive recruitment processes and rigorous, transparent selection criteria that everyone understands. It is not to scale back investments in inclusion, which would restrict our ability to build healthy, dynamic organizations.