New research has identified a major hurdle to workplace progress on diversity, equity, and inclusion: managers’ blindness to structural inequities within their own organizations.

“Such resistance may originate from the managerial position itself,” write the authors of the study in the Academy of Management Journal, arguing that as workers assume supervisory responsibilities, they identify more strongly with their workplace, which in turn makes them less attuned to its flaws.

For insight into how this bias works and its effects, as well as tactics for how to mitigate it, we spoke with study co-author Christopher To, an assistant professor of human resource management at Rutgers University. Here are excerpts of our conversation, lightly edited for length and clarity:

Can you explain the psychology behind the pattern your paper identified?

When people think about why managers resist diversity initiatives, they normally point to things like demographics, such as being a white male or ideologically conservative, things like that.  That's absolutely true. At least, that's what the research suggests. We're showing that there's something else on top of that: Managers resist diversity initiatives because, well, they're managers. They're motivated to believe that their own workplace contains less inequity.

It comes down to what's known as motivated reasoning. That's just a fancy way of saying that we distort information so that we have positive views of our own workplace. Anyone who follows a sports team knows about this. If you're watching the game, you'll always see that the referee makes good calls on behalf of your team and bad calls against your team. That's motivated reasoning.

In this case, managers’ favorite sports team is their own company. They're motivated to maintain positive views of their own company, because who wants to be responsible for a department or workplace that contains inequity? So what happens is managers, because of this motivated reasoning, are motivated to view their workplace positively, and they see less inequity to begin with. As a result, they're less likely to support diversity initiatives intended to reduce that inequity.

When you talk about managers, does that mean middle managers specifically? Or anyone at any level who supervises someone else in the workplace?

Our research suggests that it's people who have any responsibility over others. We collected around a hundred thousand observations or so from government employees. And there are different levels in the government hierarchy: middle manager, low-level manager, high level executive. And what we find is that actually anybody with any sort of managerial responsibility, they're the ones who tend to perceive less inequity. The effect might be a little bit stronger for executives as opposed to low-level managers. But I'll add a caveat there that more research is needed.

Is this effect limited to how managers perceive their own teams, or is it the case that as soon as an employee has that managerial responsibility, they have a harder time seeing inequity across all parts of the organization?

We don't know yet. The best I can say is that overall, they seem to perceive less inequity in their own workplace, but not necessarily other workplaces. If you ask them, ‘Hey, how much inequity exists within your own workplace?’ I forget the exact statistic we had, but managers believe there's maybe 13% less inequity compared to non-managers. But then when you ask about other workplaces, that difference tends to disappear. I think managers tend to only believe there's around 2-4% less inequity compared to non-managers. It’s really just within your own workplace.

Does a manager's belonging to a marginalized group mitigate that effect at all?

You would think that. We found some mixed support of that, but more research is needed. I think intuitively you would believe that there's an effect, and we found some evidence for it, but again, it wasn't necessarily as conclusive.

Beyond being a manager, what are some of the other factors that drive organizational identification?

That's an age-old question. Basically it's asking, how do you become identified with your favorite sports team, right? Why do you like your favorite sports team? And we find that yes, power is certainly one reason. Certainly things related to tenure, how long you've actually stayed with the company, matters. And so by the implication, you would think that those who have stayed with the company for a really, really long time are likely to perceive less inequity also. But usually it has to do with things around culture. How much do they actually enjoy the company? How much do they enjoy being around managers? It's typical things related to satisfaction. The more satisfied you are with the company, the more likely you are to be identified with them.

We almost always think of organizational identification as a good thing. It leads to better commitment, it leads to more productivity, it leads to better engagement. We want our workers to be identified with the workplace. But there are some downsides—one of them being that you just may become blind to some of the issues within your workplace. And that's basically what we show within our paper. People who are identified with their workplace, they're motivated to believe everything is rosy, everything's great. Of course, they may miss out on some of the issues that are actually underlying in their workplace, whether it's inequity, unfairness. It's like organizational identification creates blinders for you, where you may miss the problems that are either coming at you or underneath you.

You wrote in your paper: “Organizational identification is likely to result in power holders acting socially to fix workplace issues.” Can you say more about that tension, where people are both less attuned to problems but more inclined to want to do something about them?

That's super weird, right? People who are identified with their workplace, they want to help out, they want to make it look better. Again, I'll go back to the sports team. You want to help your sports team succeed. The problem is that when we're identified, we have these blinders in front of us where it's hard for us to see problems to begin with, because we're motivated to view things positively.

Managers tend to want to help out their employees. They have a sense of responsibility to improve their organization, create an equitable environment, all of these things. But it's kind of funny, this desire to create an equitable environment actually psychologically creates perceptions of, ‘Hey, everything's equitable, everything's great, there are no issues.’

How can employers encourage organizational identification while simultaneously encouraging employees to remain clear-eyed about problem areas?

One solution that both practice and data seems to highlight is really just data, giving data to managers about the potential inequities in their workplace. Let me provide a quick story: Marc Benioff, the CEO of Salesforce, is known to be very ideologically pro diversity initiatives. He was called the champion of diversity initiatives for a while. But when two of his female executives went up to him and said, ‘Hey, look, we may have some pay discrepancies in our workplace, his initial reaction was just, ‘There's no way that this is possible. We have a great culture. It's the best place to work.’

What ended up getting him out of that belief was actually seeing the concrete data about pay and inequity within his workplace. He ended up putting in more money to rectify those issues. If you go on Salesforce's website, you'll see that they have equity dashboards now. A lot of HR focuses on dashboards. Everyone wants to see data in just one summary statistic. Well, it turns out that some companies are including equity as part of their data dashboards, where they have things related to pay, potential pay discrepancies, potential performance appraisal differences, potential metrics around leadership opportunities, things like that.

And so do these data dashboards actually work? I think an implication of our research is that, yes, it should work. If I can go back to our paper, we had this one study where we took managers and we put them into one of two groups. One group, we asked them to just think about a typical day in the office, and in the other group, we asked them to think about times where they saw inequity in the office. And it turns out that managers who were asked to recall inequities ended up allocating much more funding to diversity initiatives. I want to say that maybe there was a 30 to 40% increase in proposed funding to diversity initiatives, as opposed to only about 10% increase in the control condition. What that suggests is that forcing managers or asking managers to really confront the inequities in their own workplace on a day-to-day basis, whether it's these data dashboards or recalling specific examples that can challenge this motivated reasoning, this desire to view their workplace positively.

Is there a risk, in feeding managers this data, that those same blinders would lead them to resist the information, or to interpret it in a more favorable way?

Truthfully, we don't know yet. Certainly there can be some resistance, and that's why if this is something that companies do push forward, ideally the data should be anonymous somehow, to prevent any potential backlash if there is resistance. It should be presented at a summary aggregate level rather than these specific individual cases.

What’s the most effective altitude at which to provide that information? For example, is it better to provide pay-equity data for the whole company, or department-level, or something else?

We don’t know, but my intuition here would be that it would have to be best if it's related directly to whatever's within the manager's control. Put another way, if it's a CEO, then you give company-level statistics. If it's a departmental manager, you'll give departmental-level statistics. It's really about what's within the manager's controls, because that's where the blinders are happening. If you want to do a quick thought experiment, if you give a middle manager company-level statistics about inequity, they could easily just pass that off as, ‘Oh, that's the other departments, that's not my department.’

If you're asking a manager to recall an instance of an inequity in their workplace, would it be more effective to steer them towards thinking about statistics or a more narrative or first-person example?

I lean towards aggregate level, to avoid potential backlash that could happen when they think about specific examples. But another way of playing with your question is, if we’re talking about stories or aggregate-level data, which is more convincing? This is outside the scope of the research, but stories in general tend to be much more convincing than statistics. The caveat that I would add is I think it would have to be stories that the manager experienced themselves. When you think about stories of other people, that runs the risk of people resisting—you just push it away, or you write it off as a one-time scenario, or it could invite issues with backlash. So my gut tells me that the stories are definitely more powerful, but the data dashboards are safer, if that makes sense.

How do you prompt that reflection?

So how do you deliver that kind of message? We talked about the data dashboards already. That's certainly the most intuitive one for me. Another one could be, we actually did run a study where we told people about the not-here bias, and we asked people to think about how the not-here bias applied to their own workplace. It's almost like another way of asking them to think about inequities in their own workplace, but it's a more indirect way of backing into it. You could summarize this not-here bias and then simply just ask people, as part of a training session or as part of a meeting, to really think about how these biases are reflected within their own workplace.

I'll give you one more. A lot of these ideas are based around the idea that good organizations don't contain inequity. And maybe one thing we can do is kind of change the narrative around that. It's not that good organizations don't contain inequity, because if that's the case, then managers are motivated to believe that there's no inequity within their workplace. Instead, maybe you can change what defines a good organization, so that good organizations are the ones that actively acknowledge inequities in their workplace and are taking active steps to combat it. It's another indirect way of trying to address the issue.

Anything else you would hope that people would take away from your research?

The story that we were hoping to tell is that a lot of the resistance towards diversity initiatives focuses on demographics and ideology—and yes, we should be thinking about that, absolutely, but the problem is we're overlooking power itself. Those in positions of power are motivated to perceive less equity to begin with. It’s a very, very simple idea: that managers resist diversity initiatives, because of their demographics, because of their ideology, but they also resist it because they're managers.

And I know that we might be beating a dead horse at this point, but to me, this has really, really profound implications for how we structure our interventions, how we even rephrase or think about the issue to begin with. Because yes, everything's about demographics and we have tons of trainings talking about how to acknowledge white privilege and male privilege, and yes, we should be doing those. Let's keep doing those. In addition to that, there's an overlooked fact that simply put, managers don't want to see inequity in their workplace. And so in addition to thinking about demographics, we need to target our interventions towards the managers themselves. Pragmatically, they're the ones in charge of implementing these initiatives to begin with. But also, again, just our research suggests that being a manager kind of puts on blinders that prevents people from perceiving inequity to begin with. And so it's just a different starting point for DEI practices.