Organizations sometimes talk about diversity as desirable because it has business benefits. Diverse teams of workers are more creative, or more likely to know their customers’ needs, the argument goes—which logically makes companies more successful and therefore justifies diversity initiatives. And there is truth to it—research has shown that gender and racial diversity improves the decision-making and performance of groups.

This approach may seem innocuous, but new research shows that talking about diversity in terms of business benefits has negative impacts. That was the finding of Aneeta Rattan of London Business School and Oriane Georgeac of Yale School of Management in a paper forthcoming in the Journal of Personality and Social Psychology. We spoke with them recently about the takeaways from their research. Here is a transcript of that conversation, edited for clarity:

Why does it matter how organizations talk about why they value diversity?

Georgeac: That's very interesting. This was the first question of my doctoral dissertation defense, so I'm having a flashback. The justifications that organizations give for why diversity matters to them give you insight into the motives that drive them to pour resources into initiatives, such as reaching out to new people, training their people, and hopefully trying to include non-male, non-white employees in the workplace. The way they justify diversity tells you a lot. Where does that impetus come from? Where does that motivation come from? And fundamentally, why do they actually value diversity? Even to people who are still outside of the organization, such as job seekers, the way organizations talk about diversity tells you a little bit about whether it's an absolute commitment, or rather one that is contingent on the world changing, priorities changing, or on how the firm is doing, for instance.

In your research, you've basically identified two categories of justifications for diversity. Can you explain what they are and how common they are relatively?

Georgeac: One is a justification that I call instrumental, in the sense that it says, 'we care about diversity because it's good for the bottom line.' You have several facets of this argument. It could be good for the bottom line per se, but also for organizational performance, creativity, decision-making, etc. The list is endless, but essentially, it's framing diversity as a business asset. We care about it because it's going to help us achieve better performance results. Because it frames diversity as a means to an end, which is organizational performance, I describe it as an instrumental rhetoric.

The second type, in contrast, is a rhetoric that is non-instrumental. It says, 'we care about diversity in and of itself because it's the right thing to do.' This rhetoric frames diversity in terms of social justice, equal opportunity, and fairness, rather than in terms of the benefits that diversity may or may not have for organizational performance. It's a commitment to diversity independent of everything else, because diversity is an end in itself in that discourse. In that sense, I frame it as a non-instrumental diversity rhetoric. So it's really the contrast between diversity as a means to improving the organizational bottom line, versus diversity as an end in itself that doesn't need to serve any other goal.

Which of these two is most common among US organizations?

Georgeac: Perhaps unsurprisingly, it's the first one. The instrumental case—we care about diversity because it's good for business—is by far the most prevalent. In our first study, we used a machine learning algorithm to classify the discourses on diversity that we found on the Fortune 500 websites. We collected all the texts that we could find in which they justify their commitment to diversity, and using machine learning, we managed to categorize them into these two buckets that I described: the ‘business case’ for diversity, which is the instrumental rhetoric, and the ‘fairness case’, which is the non-instrumental rhetoric.

What we found was that, out of 500 companies, there was almost 20% that didn't justify their commitment in any way, or did not talk about diversity. Among the rest, we had a little bit shy of 80% who were making the business case, and very few, between 1-2%, making the fairness case. The disparity between the types of discourse that are out there is really, really striking.

Rattan: To add to that, and with reference to your first question, it could have theoretically been possible that this difference in the prevalence of the two cases doesn't matter. It could have been possible that the different justifications for why these organizations value diversity made no difference to any kind of outcome that we care about. That is the value of doing an empirical investigation on these things—of actually looking at, observing, and quantifying what companies are saying, then following that analysis of the real world up with the kind of experimental studies presented in this work. It was a scientific question that we found fascinating, and one that has yielded what we think are meaningful results.

When organizations make the business case for diversity, what is the impact on workers?

Georgeac: In this line of work, we were looking specifically at job seekers. They're people who are still outside the boundaries of the firm, but who are hoping to join it, and are therefore attuned to subtle signals about what this company is like. To understand the effects of these two cases on this particular audience, we randomly assigned job seekers from underrepresented groups to read either a business case or a fairness case. What we found is that relative to job seekers from underrepresented groups who read the fairness case, those who read the business case reported a lower anticipated sense of belonging to the organization that made that case. And in turn, that predicted lower interest in actually joining the organization. That's the first big finding, which concerns two types of responses: anticipated sense of belonging, and interest in joining the organization.

But then we wondered: If these members of underrepresented groups have already not been deterred from applying, could it be that the business case undermines them yet again at the next stage, which is the job interview?

So another portion of my doctoral dissertation looks at this other potential consequence for job seekers, which is this idea that being exposed to diversity justifications immediately prior to a job interview could have an impact on performance during the job interview. This is exactly what we found, again specifically for underrepresented groups. We found that in the context of seeking a job in the consulting industry, women who had been randomly exposed to a business case rather than a fairness case prior to a practice job interview were rated as performing significantly more poorly by their interviewers, who were blind to our hypotheses and to the message that the interviewees had read. This was not the case for men. Men’s performance, as evaluated by the interviewers, was not affected by the type of case that they had read prior to the interview.

I want to go back to talk about the different underrepresented groups that we studied. We theorized that instrumentality, the psychologically active ingredient in the business case, would be specifically undermining underrepresented group members. In our research, we've been looking at LGBTQ+ professionals; women specifically in STEM industries (Science, Technology, Engineering and Math), where women are severely underrepresented; and at African-American students enrolled in undergraduate or graduate programs. And in the performance study, we've been looking at women and men seeking jobs in consulting, as we know there's gender disparity there. I should say as well that when we looked at women in STEM, we also looked at men in STEM to contrast the effect, and we found that only women in STEM were affected, not men.

Rattan: We also tried to understand why there is this impact on belonging. In a number of the studies, we actually try to examine psychologically what is happening to these members of underrepresented groups that makes them anticipate belonging less in companies that make the business rather than the fairness case. What we found consistently is that when companies use this business case, they evoke more social identity threat among members of underrepresented groups. Social identity threat is the feeling that you have to worry that others are going to judge you through the lens of stereotypes— and particularly, in these contexts, negative stereotypes about your group identity. When you go back to the content and definition of these cases, the business case really justifies, through its instrumental language, the idea that the company and the people in it are going to look at you and expect something different from you versus another colleague based on your social identities. In other words, we show in our studies that the business case evokes a confirmation of that social identity threat—that you do need to worry about being judged through this stereotypical lens. The fairness case also slightly raises these concerns, but meaningfully less than the business case does .

You've looked at job seekers and job interviewees. Is it possible to extrapolate about how this would impact workers in jobs in an organization?

Georgeac: That's really something that we would like to investigate in the future. As of now, we've followed the recruitment process through which job seekers go. First, they may find an interesting ad, go on the website, browse the website, and discover the statement. Already, we know that the instrumentality inherent to the business case induces a sense of threat among diverse talent pools, which also undermines how much they anticipate feeling a sense of belonging to the company and, in turn, how interested they are in joining the firm. At that point, they may or may not apply, but if they do, we wanted to conceptually keep following them. Next, we see that at the job interview stage, they may be undermined yet again by the business case discourse, which is everywhere. In our studies, we operationalized it in a simple way, by asking participants to read a short paragraph, but really, this discourse is everywhere. It's on career portals. It's on companies' websites. It's in recruitment brochures. It's in managers' talks at recruitment events. So it's very likely that you will be exposed to that rhetoric at one point or another, even up until you walk into the door of the interview room.

The next logical step is to keep following people–and particularly members of underrepresented groups–if they've managed to enter the firm, to investigate how the fact that they're constantly swimming in this ocean of instrumentality may affect their performance in the day-to-day. How does it affect intentions to stay rather than to quit? What kinds of coping mechanisms might they have to put in place to shield against instrumentality? And how do those coping mechanisms impact the likelihood that they may be promoted? The possibilities are endless, and we would love to actually get into the psychology of employees once they've joined the company.

Rattan: In a very recent study that we've run, we've also found some evidence that members of majority groups feel a bit threatened by the business case rhetoric. Our theory—though we don't have evidence for this yet—would be that perhaps the business case also makes them worry about their own value in the organization. Thinking about incumbents in an organization who, as Oriane put it, are swimming in this ocean of business-case language—as is the case in the majority of the very large employers who are out there–we might find new detrimental effects on members of underrepresented groups, and we also might see that living in this ocean is not good for members of majority groups either.

It could be detrimental to them in multiple ways. It could be detrimental to their intergroup attitudes and for their views about diversity in organizations. It could also be detrimental to their views of the company's fairness. In this way, this rhetoric might end up backfiring in manners that are really detrimental to the goal of meritocracy that organizations seek to achieve in their diversity efforts–as well as their broader efforts. So this is exactly where we want to go with the research next, continuing to follow people through. And we actually invite companies that might be interested in examining or changing their rhetoric to reach out, because we believe it is essential to test the impact of these kinds of language on people in real life.

Georgeac: We have extensive evidence for these effects among underrepresented groups. It's a little bit less clear what happens with well-represented groups. As I was saying, we tested the consequences of instrumental versus non-instrumental rhetoric on anticipated sense of belonging among women in STEM, for instance. We found that men were completely unaffected, completely indifferent to the type of diversity rhetoric they read. Aneeta is actually alluding to a more recent study that we carried out, which contrasts African American with white American students enrolled in undergraduate or graduate programs. Here, we found that white applicants tended to be a little bit more threatened by the business case relative to the fairness case. So different well-represented groups (e.g. men in STEM versus white Americans in higher education) may, for reasons that we don't completely understand so far, react in different ways.

It could also be that we found this effect in this recent study with white applicants because we conducted it after the George Floyd events. That big, tragic event may have actually changed something in the current population. We don't know yet, but we have at least some evidence that whites are not indifferent to this instrumental rhetoric. And to understand why that might be the case, people really have to understand what the business case actually says. It's built on the assumption that different groups have different skills, different work styles, different perspectives, different communication styles—you name it. Everybody brings something different, but it's group-based. In assuming this and saying this, the business case justifies looking at job seekers through the lens of the groups to which they belong.

And we know that this is really threatening for members of underrepresented groups, because traditionally, they've been stigmatized on the basis oftheir group membership, and negative consequences ensued, whether it be discrimination, prejudice, and so on. For white individuals though, it's not illogical either that they would be threatened as well by the business case. If you go back to that instrumental rhetoric, which says everybody contributes something, they may be asking themselves, 'I'm not diverse, so what do I contribute?' Therefore, they could also feel a sense of threat. So, to be continued—we need to investigate this much further.

Given all this, what should companies do? What's your recommendation in terms of how they approach talking about diversity?

Georgeac: That's always the question, but I need to specify one thing about the design of our studies to answer it. As I said, we looked at the business case versus a fairness case. We also looked at a rather artificial rhetoric, which is a neutral case. The neutral case talks about diversity, but does not provide any justification for it. What we found was that the business case is much more detrimental than that neutral case. And the fairness case, depending on studies, might be as good or a little bit worse than the neutral case. In all studies, we found that the neutral case–the one that had no justification for diversity, and just stated the company’s commitment to diversity—actually fared better. Therefore, one possibility for organizations would be to affirm their commitment to diversity, without any justifications, as if it were a matter of fact that didn't require any particular justification.

Now, some companies may feel that they have to justify their commitment to diversity. In that case, justifying it using the non-instrumental rhetoric, the fairness case, is going to yield less detrimental consequences. We can see that statistically, whatever the detrimental effect the business case has, the fairness case is half as harmful. In that sense, if you absolutely have to justify your commitment, a fairness rhetoric is the best, but stating your commitment without justification may be the best of all.

Are there any broad read-throughs for how managers talk about initiatives of any sort within a company and how they might be justified or not?

Rattan: There's a real lesson for how companies think about diversity. When companies endorse a value of collaboration, they don't necessarily spend a lot of time explaining why, because it seems intrinsic. We understand why collaboration is good for companies. When they endorse the value of customer centricity or of innovation, it's evident to everyone why that is important, why it matters, and no justification is required either. There's a real lesson for organizations about treating diversity as core, like any other core values, if you really hold it as such. For a lot of companies, diversity is being treated as an aberration, and in treating it as an aberration that you have to justify, it's actually having effects that we think are counter to what organizations intend. Drawing diversity more into the core of how you treat any organizational value may be a first step to thinking more broadly about how organizations can take this research and do good with it.

Georgeac: What managers need to be really careful about is the fact that it's not enough to have a well-intentioned message. On a superficial level, the business case does sound positive. It says we want diversity. We want it because it's going to help us, because you, as an underrepresented group member, are going to bring to the table something that we recognize as uniquely valuable. That sounds positive, doesn’t it? It says, 'we're going to value you.' But what we find are the paradoxical effects of what seems like a discourse cloaked in positivity. Our research is trying to tell managers to be careful about this, and to be conscious that the road to hell is paved with good intentions, essentially. They really need to be aware of the fact that they may actually be causing identity concerns among the members of the very groups their organization is trying to recruit and retain.

Rattan: What managers say sends signals, especially if your organization is not yet at a level of diversity where you want it to be. What you say, in particular, sends belonging signals. What you say sets a tone for inclusion. Before putting it out there, leaders should think through what you're really communicating to this audience. They should ask themselves whether they have members of their network with whom they can road test it. If not, again, they can reach out to us, and we will work with them to study the impact of what they might want to say.

Are there any read throughs for how organizations talk about ESG, or CSR, initiatives more broadly?

Georgeac: In general, our research does tie to the broader CSR literature on the two competing theories of stakeholder versus shareholder theory. One proposes that the firm has no social responsibilities except towards shareholders and increasing shareholders value. The other says, well, actually shareholders are just one type of stakeholder among many, and the firm has a duty to treat all of its stakeholders well. The parallel I would draw between the two is that the business case is a very narrow way of looking at diversity. Its only way of justifying diversity is as serving shareholder value, whereas the fairness case rhetoric rejects the premise that diversity needs to serve shareholders to be justified, because diversity serves other goals, like social equality, equal opportunity, fairness, and morality.

In a sense, that discourse is much more in line with this idea that stakeholder theory has been proposing—that companies have a duty toward a variety of audiences, including underrepresented groups. Thinking beyond diversity, we're trying to push people to stop thinking only in terms of the bottom line—an inherited mindset that we got from Milton Friedman—but start thinking a bit more broadly about other goals, other audiences, and other groups that your company also needs to think about and serve.

Part of my reaction to your research is that these Fortune 500 companies on their websites might think they're speaking to shareholders and wanting to justify their diversity initiatives in business terms to them, but not thinking through the fact that their message is also being heard by other people such as job candidates....

Georgeac: I'm not sure companies are being very intentional when it comes to thinking about who they're really trying to talk to. Sometimes, we would find messages on their diversity and inclusion page duplicated on their careers page, which would be directly targeted at job applicants. One thing that I wouldn't like managers to take away from our research might be that they need to talk in a specific way to a specific audience. One might think that as a company, you're free to adopt a different rhetoric at your shareholder meeting relative to the one you use publicly, and it doesn't matter. That would be a really flawed logic, because transparency has become so valuable today, and everything tends to leak in the press or on social media. That's why consultants in crisis management have been doing really good business. Approaching this in a way, where you have one discourse for your shareholders, another discourse for your employees, and yet another discourse for your job applicants is probably bound to fail. It seems impossible today that these different spaces would be siloed enough to avoid information leaking in a way that would expose the firm’s hypocrisy when it comes to its approach to diversity.

Rattan: One of Oriane's newer areas of research is actually to look at when companies make the fairness case, but try and put in a little dash of the business case. Do they get away with it in the eyes of members of underrepresented groups? Or is it just as bad as if you had used a business case alone? That research is not really ready to be talked about, but I'm sure you can imagine that our theory is that when you have these instrumental motives, they overwhelm everything else.

Georgeac: To go back to your very first question about why studying justifications for diversity is so important, the take-away is that these are usually very well-intentioned messages…that we know nothing about! Those messages are crafted by organizations, and they may think, 'oh, it sounds good, so it's going to be fine.' It may not be. Having a scientific, theory-driven approach to testing their effects is important.

And on a personal level, I had always felt really bothered by the superficially positive rhetoric of the business case. I asked myself if I was the only one, because everybody seemed very enthusiastic about it, but surely there's something a bit icky about saying, 'yes, come to our company because we care about what you can bring as a woman.' I was like, 'am I just a woman?' This led me to wonder: what does it do to other people to be seen through that lens, and to be reduced to their social groups, essentially? It can be really subtle, but those messages can actually convey signals that can be really disturbing and eventually, even threatening. And this, ultimately, has the power to undermine organizations' diversity goals. The good thing, the hopeful message, however, is that when we study them and identify which rhetorics are worse versus better, organizations can act quickly to improve what they say in their diversity cases.

For more about how organizations can approach talking about diversity, consider reading The Conversation by Robert Livingston.

The handbook for this new era of business doesn’t exist. We’re all drafting our own as we go along—and now we’d like to start doing so together. You can sign up here to receive briefings like this by email.