Research has shown that talking about the business case for diversity can backfire on employers. Job seekers from marginalized groups interpret such linking of diversity to corporate profit or stock performance, for example, as a sign that an organization would value them less and stereotype them more. “The way organizations talk about diversity tells you a little bit about whether it's an absolute commitment,” Oriane Georgeac, now at Boston University's Questrom School of Business, told us last year, “or rather one that is contingent on the world changing, priorities changing, or on how the firm is doing.”

A new study in the Academy of Management Journal poses a similar question about existing employees: How does the way organizations talk about their diversity commitments affect workers’ willingness to buy into those efforts? The authors found that the most common framing leaders use is also the one least likely to get workers on board—and that the most effective framing is one most leaders shy away from.

We recently spoke with co-author Lisa Leslie, a professor of management and organizations at NYU’s Stern School of Business. Here is a transcript of our conversation, lightly edited for clarity:

How do organizations typically talk about their diversity efforts, and why does the framing matter to workers?

I had noticed this interesting disconnect where I would talk to practitioners and leaders, and when they would talk about diversity, they would always emphasize how valuable it was. It could be that it's valuable in general, it's valuable for business reasons, it helps to perform better as a company. It might be that it's valuable for moral reasons, it's the right thing to do, new perspectives, et cetera. But if you look at the research on what actually happens when organizations become more diverse, it's a lot more complicated. The research didn't universally support this value-based perspective, that diversity always leads to better outcomes. It's actually much more nuanced.

Sometimes diversity does in fact lead to better performance. It's seen as the moral and right thing to do and has these positive effects supporting that value-based rhetoric that leaders are always using. But sometimes, the exact opposite happens: Diversity actually has a negative effect on performance, because people see it as unfair and therefore morally dubious. There’s a lot of variability. So the basic motivation for this project was, well, what happens if leaders talk about diversity in this more realistic way? My co-authors and I came up with this more nuanced message about diversity that we refer to as contingent rhetoric, which is the idea that diversity is valuable for organizations, but only if you overcome the challenges. This is a more nuanced message and also better reflects the reality of diversity’s effects on organizations.

We wanted to look at two things: We wanted to compare value rhetoric, ‘Diversity is good,’ to contingent rhetoric, ‘Diversity is good only if you overcome the challenges,’ in two different ways. The first question was just, which are leaders more likely to use? What we expected, and what we found, was that leaders were more likely to use value than contingent rhetoric, the reason being that value rhetoric is a uniformly positive message about diversity. Contingent rhetoric is not negative, and absolutely it's not suggesting diversity is bad, but it is more negative than value rhetoric. It suggests there are some challenges to be overcome. There's a lot of evidence that leaders are hesitant to say negative things about diversity because they worry that they might be perceived by their employees as being prejudiced—maybe they'll interpret that as a signal that the leader doesn’t really value diversity.

That's a benevolent motive, not wanting to be seen as prejudiced by your employees. That's a good thing. But we also predicted and found that it really prevents leaders from using the most effective rhetoric type. That was our second question: When leaders use value rhetoric versus contingent rhetoric, which one is more likely to increase what we call diversity effort among employees? That's basically, are employees willing to put personal effort into fostering diversity and inclusion? Do they call out discrimination when they see it? Do they provide resources and mentoring to minority groups? And what we expected and what we found was that even though leaders are more likely to use value rhetoric, contingent rhetoric is more effective. It has a stronger positive effect on employees’ diversity efforts.

Why might that be the case?

The explanation for why that's the case is that there's a lot of work on employee motivation that shows that when you give employees a goal that's more difficult, as opposed to a goal that's easy, that actually makes them work harder. It makes it salient to them. A lot of effort and persistence are going to be needed to attain the goal, and therefore they actually work harder—as opposed to when they have an easy goal, they assume that they don't need to really persist that much because it's going to be easy to do, so they don't work as hard. That's what contingent rhetoric is doing, because it implies that benefiting from diversity requires overcoming some challenges. It says to employees that diversity goals are actually difficult to achieve. They're going to have to put some effort in, some sweat. And because of that, they actually work harder than when leaders use value rhetoric, which implies that diversity goals are easier or at least doesn't make them as salient. There's some difficulty and challenge involved. So what that sums up to is this paradoxical effect where the rhetoric type that is most common is not actually the one that's most effective in terms of increasing employees’ diversity efforts.

What would be an example of contingent rhetoric?

The reason why diversity is not always beneficial usually has to do with conflict and tension. People from different groups might not get along as well together as people from the same group. So it seems like really emphasizing diversity and organizations is really valuable, but only if everyone learns to take other people's perspectives, or only if people learn to successfully navigate the tensions that might arise, that sort of thing.

I mentioned that leaders are hesitant to do this because they worry that they'll be seen as prejudiced. When leaders use this rhetoric type, do employees in fact see them as prejudiced? And we find that they don't. It's an important piece of this—basically, what is preventing leaders from using it is not actually grounded in reality. They're afraid of being seen as prejudiced, but that doesn't actually happen when people use it.

What are the factors that influence how value-based rhetoric is received by workers?

There are lots of different factors, but it tends to be more about what the leaders do and the general environment at the organization. For example, diversity is more likely to have a positive effect on performance than a negative effect if there's a climate where everyone feels included or if leaders have a certain type of style, like a transformational leadership style.

You’ve published some other research examining some of the unintended messages workers can pick up from leaders’ diversity rhetoric. Can you say more about that?

That was a little more complicated. Organizations often implement diversity initiatives with positive intent. They're doing it because they want to make progress towards diversity goals. But the basic idea is that even if that's the organization's intention, sometimes employees make these more negative or detrimental inferences about why the organization is doing it and what it means, and that can create negative unintended consequences. One example of that is the idea that when companies have a diversity initiative, employees might infer, ‘Well, that means that members of the groups those initiatives are targeting, women or racial and ethnic minorities, they can't be successful on their own. They really need help from the organization to succeed.’ And because of that, they might infer that those groups are not particularly competent and actually increase discrimination against those groups as a result of the diversity initiative.

Another example would be the idea that if a company has a diversity initiative, people are likely to infer that target group members are more likely to succeed in the organization. But it can also lead members of majority groups who are not targeted by the initiative to believe that minority groups are getting this unfair advantage to help them succeed, and therefore their group is being treated unfairly, which can lead to negative attitudes towards the organization and maybe even desire to turn over.

A third one is the idea that morality is valued in the organization. If you have a diversity initiative, it signals that the leaders care about the success of minority and disadvantaged groups. So that's a good thing that can have some positive benefits, but it can also create perceptions of, ‘well, if morality is value, this company's doing the right thing, then maybe I don't need to monitor my own behavior as much with things like discrimination,’ and it actually leads to more of those behaviors as opposed to reducing them.

Then there’s the idea that having these initiatives signals that diversity is valued in the organization and that managers and others in the organization are rewarded for helping make progress towards diversity goals. That's a good thing, but also there's evidence that maybe that can give people more of an extrinsic motivation for focusing on diversity—doing it to make the numbers look good, as opposed to intrinsically really wanting to fix the problem. That can lead them to do things like make the numbers look better without creating true change. Maybe it's something along the lines of taking a job with a lot of women or minorities and recategorizing it as a managerial job, even though it wasn't before. And that improves your numbers for diversity and management, even though nothing has really changed. You just changed the job title.

What can leaders do to make sure their messages are landing as intended, without those counterproductive signals? And does it matter if the message comes from the top versus when it comes from middle managers to their team?

That's what we looked at explicitly in the first paper, how leaders talk about diversity: Does it matter if it's senior leaders versus lower-level leaders? And we find it doesn't. Conceptually, I have no reason to believe that it would matter very much depending on the level of the leader. What you want first and foremost is just consistency. So you have one message about diversity that starts at the top, but make sure lower-level management leaders are aware of what that message is and reinforce it in terms of preventing the negative unintended consequences. I think it also has to do with leaders just being more explicit and clear about why they have a diversity initiative—what that means, for example, about the likely success of minority groups and making sure that that message is really communicated clearly.

How can leaders figure out the most effective way to frame their contingent rhetoric?

We coded the rhetoric that we found on company websites. Companies that would use values rhetoric would say things like, ‘Diversity is wonderful, it makes us stronger and helps us perform better, it's the right thing to do.’ And the ones that used contingent rhetoric—which did exist, but were more rare—would say things like ‘Diversity makes us stronger. That doesn't mean it's easy. There can be these consequences, and we need to be able to work through them if we want to realize that value.’ So I think it's a matter of emphasizing that it is in fact valuable, it has positive effects, and also acknowledging some of those challenges and just really emphasizing that to get the value you have to be willing to work through the challenges.

So ideally, how prescriptive should that rhetoric be? Is it better to say it's challenging but valuable, or to say it's challenging but we can do x or y specific things to make it less so?

More the former. You want to emphasize first that diversity is valuable, but then just add, ‘But we need to work through the challenges to get that value.’ Another way that I sometimes talk about it is that value rhetoric is ‘Diversity is good’ and contingent rhetoric is ‘Diversity is good but hard.’

Other research has found that when organizations talk about the business case for diversity, that tends to backfire on them by making underrepresented workers feel less welcome. How does your work fit in with that?

There's been a lot of backlash to the business case suggesting it doesn't work. But the findings are much more nuanced. In studies that compare the business case versus the moral case, for example, it's not that one is always better than the others. It really depends. Sometimes and for some people, the business case is better, and sometimes and for some people, the moral case is better. The specific work you're referring to looked at both, and they found both the business case and the moral case were bad. The business case was just more bad than the moral case. Because of these really mixed findings, what I would advocate for is that whether you use the business case versus the moral case is not as consequential as whether you use it to say ‘Diversity is purely valuable for business or moral reasons,’ versus contingent rhetoric acknowledging the challenges.

So for example, one reason why the business case and other types of value rhetoric don't work is it has a licensing effect. When you emphasize that, it kind of suggests that groups are treated fairly so people don't have to work very hard at diversity. But we don't get some of those same negative effects when we use contingent rhetoric. Whether the business case versus other value-based cases are more beneficial or work better is a little bit mixed. But it seems pretty clear that regardless of if it's for business reasons, moral reasons, or whatever reasons, using a contingent framing works better than just the value-based framing.

What would you want employers to take away from your work?

In some ways this is a very straightforward intervention: Leaders can get their employees to put more effort into diversity and inclusion, and in turn better achieve their diversity goals, if they're willing to change how they talk about diversity. Instead of just using value rhetoric to say ‘Diversity is good,’ acknowledge that to achieve that value of diversity, you have to work through and overcome some challenges, and use a contingent form of rhetoric. It sounds easy, but I think the evidence we provide about why leaders are hesitant to use contingent rhetoric suggests that it might be a little bit harder than it sounds. So I think it’s important for leaders and managers to really do a lot of training and education if they want others in the organization to use contingent rhetoric, and really emphasize that even though it's not very common, it does in fact work better. And even though it's a less positive message about diversity, it's not going to have negative repercussions in terms of how employees perceive their leaders.