Tuck Study: Greater Diversity in Higher Ed Linked to Higher Student Salaries

[This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of the Tuck Knowledge in Practice Podcast is the audio record.]

Kirk Kardashian: Hey, this is Kirk Kardashian, and you're tuned in to Knowledge and Practice, a podcast from the Tuck School of Business at Dartmouth. In this podcast, we talk with Tuck professors about their research and teaching and the story behind their curiosity. Today on the show, my guest is Peter Golder, the E.B. Osborn Professor of Marketing at Tuck. Peter is here to discuss his new paper, published in the journal Nature, which finds that racial diversity in higher education is associated with higher student salaries. Peter is an expert on market entry, timing, new products, long-term market leadership, and quality. He was on this podcast last year to talk about his new elective, Creating Winning New Products and Services, and to share his thoughts on how to foster creativity. Peter has taught at Tuck since 2009, and his research has won more than ten Best Paper or Best Book awards, including several of the most prestigious awards in his field. Peter Golder, welcome back to Knowledge and Practice.

Peter Golder: Yeah, thanks so much, Kirk. Great to be here with you again.

Kirk Kardashian: Yeah, I'm honored that you made the time for this. And I'm excited to talk about this paper that you published in the journal Nature, which came out in May. So before we dig into the substance of the paper, which is racial diversity in higher education, I think it's worthwhile to stop and acknowledge that publishing a paper in Nature is a big deal.

Peter Golder: Well, yeah. I mean, first of all, thanks so much for that. And, yeah, when you, you know, you start out this process, we're never really sure where it's going to go. But we did have a topic that we thought was suitable for a broad audience, and we did want to reach a broad audience that went to, you know, not only, business school faculty, but faculty across institutions and even more so across society, both something that the Supreme Court itself would be likely to see, I believe. And certainly people who are thinking about, you know, public policy in this domain. So we certainly had a topic that we thought was suitable for a broad audience. And, you know, just very appreciative that the editors of Nature felt the same way.

Kirk Kardashian: Yeah. I'm sure. So the question people might be asking is Peter Golder, professor of marketing at a business school. How did you end up studying racial diversity in higher education?

Peter Golder: Yeah. I must say, I've gotten that question from a lot of my friends and family because, it doesn't seem to quite fit, but there's really been kind of a nice evolution or progression towards this topic over time. So, you know, I'll relate it to a few things. I mean, one is that I have studied innovation and product development over the years, that has provided a couple of underpinnings for the work. So, you know, one is you can just look at, look at MBA programs as, as products themselves that are based on various design criteria. And one of those very critical design criteria is the composition of the student body. I mean, the composition of the faculty matters, the composition of, you know, administration matters, but the composition of the student body matters a lot because the students learn from each other. The faculty learned from the students as well as the you know as well as the students learning from faculty. So the group of people that's put together is really important determinant of the design of an MBA program. So that's that's one aspect of it. Another aspect is related to the previous podcast that you and I did about my course on creativity. And there's a lot of research about how important diverse viewpoints, diverse backgrounds are in product development and in creativity. So that's really kind of ingrained in me from previous research is just how important that is, in a new-product context. And, you know, many other people talk about it being important in an educational context as well. So that's some of the background. Another, another piece of the background is that one of my coauthors on this Nature paper, Deb Mitra, and I have worked together for almost 30 years now, going back to the time when I was a young professor at NYU and he was an incoming doctoral student. And, you know, I ended up chairing his dissertation, but we've worked together for decades. But in 2008, we published a paper that examined the link between academic research and student salaries in MBA programs. And there was a lot of talk at that time about how, MBA programs are really kind of wasting their money on academic research. And it's really not related to what the students need to learn. And we kind of took that as a challenge and looked at again, the link between academic research and student salaries and found out that the schools that were doing more publishing of academic research, it had a positive association with MBA rankings and also had a positive association with student salaries. So that link was there. That was another piece of the puzzle that was already in place for us as we thought about this study. So I didn't just wake up one morning and think about, yeah, yeah, I think it's time to talk to the Supreme Court about this recent affirmative action decision. You know, a lot of the pieces were already in place as the broader legal story was developing.

Kirk Kardashian: Wow. That's that's so interesting, Peter. I noticed you also have a third coauthor, Mariya Topchy, who says in the paper, is that Traveler's Insurance? So tell me the story there.

Peter Golder: So the backstory on that is, Mariya was Deb's PhD student. And so essentially there were, in some ways, three academic generations putting our heads together on this. But, you know, Mariya, as a PhD student, was very instrumental in the project, especially in some of the earlier stages of the project that was, very instrumental in the beginning stages of the project and, you know, very happy to have her participation, you know, and happy to have her as a coauthor on the paper as well. So it's really a collaborative effort among the three of us.

Kirk Kardashian: Yeah. Okay.

Peter Golder: Which, again, talking about diversity, I mean, putting, putting, you know, all three of us working together, you know, certainly made the paper better than any one of us working alone.

Kirk Kardashian: Yeah, I don't know. I'm sure it did. Yeah. So it seems like this paper really arose out of the 2023 Supreme Court case. Students for Fair Admissions, Inc. v. President and Fellows of Harvard College, which declared it unconstitutional for colleges and universities to use race based admissions policies. So is that accurate to say that decision really prompted you to, to undertake this study?

Peter Golder: It certainly proceeded in line with that. But, here's a little more of the backstory of the, of the project. So, you know, this the case that you mentioned, Students for Fair Admissions, I mean, this had been working its way through the courts for a while. And basically, I think everyone expected it to get to the Supreme Court because, you know, there have been previous decisions on affirmative action going back to like 1978 was the first decision, essentially making affirmative action legal. And then there were subsequent decisions that continue to reinforce the fact that racial diversity in higher education was seen as a compelling state interest because the prevailing view was that diverse viewpoints in the classroom enhanced the education for everyone. But the justices who were dissenting from those previous opinions, were starting to become more of the majority voices. I would say that would be, Justices Roberts, Thomas and Alito would be the people who were, you know, who were already on record as being the dissenting voices in previous affirmative action cases that were, that were supported by the, by the courts. So we kind of knew this was happening. So we started the project really before the 2023 decision came out and, there was really some tension involved because we felt like we knew enough to have an opinion about the topic. But, you know, we view our role in this whole process as being somewhat different. I mean, essentially as researchers, we are scientists. There's plenty of other people who have opinions on the topic, and some of those opinions are rooted in factual information. Some of those opinions are rooted in misinformation and purposeful disinformation, in some cases. But the role that we have to play in this process, and really the privilege that we have as academics is to add empirically-based factual information to this conversation. So we proceeded along this path. Since we have the business school data from an earlier project, we started with that data. We did find this association between racial diversity and salaries in the business school data. But we continued to expand the data by adding in all the law school data. Because what we wanted to show that this wasn't just a finding that happened in one context, but seemed to be more general in other contexts. And again, that's our job really as scientists is to bring this, this type of a result to the conversation. You know, we had to pause until we really had the other validation from the other data set. And so the Supreme Court decision came out in 2023, and then the paper was written really with that as the primary kind of context, because that became the primary context. And really, the Supreme Court decision really set the table for our paper very well because they overturned affirmative action on empirical grounds, almost more so than legal grounds. They basically just said, okay, we're not going to disagree so much with this idea that diversity could improve education and, if so, could be a compelling state interest. But they said, basically, where's the evidence? You know, we've looked around. It's like, where's the evidence that this is the case. So at that point, it became very clear to us that was really the positioning for our paper was a direct response to that, to that challenge to that from the Supreme Court that said, okay, where's the proof that diversity actually helps everyone. That helps overall education. And we thought we had the data to answer that question. And, you know, it turns out that we did.

Kirk Kardashian: Yeah. Yeah. Well, I know, like, in, you've done some other research, where you, I think contributed to maybe an amicus brief to the Supreme Court on a different issue completely. If I recall correctly. What was the topic on that one?.

Peter Golder: So it was really about, well, the basic issue was about should the Booking.com name be treated as a generic term, or should it be trademarked? Should be should you have rights to trademark it? So Booking alone could not be trademarked because it is just a general word. People make hotel bookings, airline bookings all the time. But Booking.com became known in the public consciousness as a brand identifier. And the Supreme Court basically said their decision was going to look at the evidence in a case and to basically see how this term is used in the marketplace. And that's very much along the lines of what my amicus brief was saying. And again, I was the lead author on a brief that was coauthored by multiple other marketing faculty members and our input for the Supreme Court was that there was a lot of evidence out there that the court could look at to see how consumers were using these terms. The USPTO was essentially saying that there should be a rule that you cannot take a generic term and just add dot-com to it and turn it into a brand name. And the Supreme Court decided that, well, let's look at how consumers are using this. If it truly is a brand identifier, then you should allow it to be a trademark. That is what our amicus brief said. And basically, we pointed to lots of evidence where you could understand consumers and how they thought about these terms. And, yeah, it was very happy that the Supreme Court basically, adopted the same position. But, you know, I'd also like to point out I didn't wake up one morning and say, oh, I think I want to write something about, you know, the Supreme Court case. It grew out of research that I have been doing my whole career and grew out of a paper that I, that I wrote for the Journal of Marketing Research in 2000, when I had no idea, obviously no idea that this would even come up. But it was basically, you know, how do you use archival evidence to understand business decisions that are being made? And how do you use this evidence to find out how consumers are thinking about markets as they develop. So again, really a direct outgrowth of, you know, the, the academic research that I've, you know, been privileged to be able to do for my, for my career.

Kirk Kardashian: Well, the reason I brought up that Booking.com case is because you have, I guess, a pretty unique experience now almost influencing the Supreme Court on one case. Did you ponder if you would submit an amicus brief on the affirmative action case?

Peter Golder: I think so, because there have been many over the years. I said that the Supreme Court has talked about this affirmative action issue and has ruled on this affirmative action issue multiple times. And there are many, amicus briefs that have been filed in those previous, previous matters. I suppose one of the guiding factors here is that, we did not feel we had something new to say. If it was just based more on kind of, our opinion about it. With the Booking.com case, you know, I and my coauthors felt like we had something new to say that we'd be introducing a new voice to the court. That's why it was important in that case, just as an amicus brief, I didn't think it would have the same weight here because we were not quite ready with all of the scientific, empirical results that ended up in the Nature paper. So I think that's that's why, yeah, we didn't feel we'd be adding kind of a new voice at that point that would get attention. And even though the, you know, the Nature paper came out after the Supreme Court decision, we're we're still hopeful that it will affect, judicial policy going forward, because I don't believe this is the last time the courts will consider this issue. There's one interesting thing that was kind of buried in the Supreme Court decision is that they did not apply the decision to the military academies. So the Supreme Court has allowed the military academies to continue to use affirmative action in their admissions decisions. And they talked about, you know, in light of the special interest that may be represented by those institutions, which I think to me means that there is, you know, there's that there is great racial diversity throughout the ranks of the military and the importance of having military leaders who are familiar and comfortable, and educated in a diverse environment. Even the current Supreme Court recognized the importance of that, I believe.

Kirk Kardashian: Yeah, yeah. Well, let's let's just state it plainly here. You found in your paper that there is a strong association between, diversity, more diversity in MBA and JD programs and higher student salaries. Right?

Peter Golder: That's correct. That is the core finding, and you've stated it exactly right. That link leads to all of the implications that we can derive from that link.

Kirk Kardashian: Yeah. So Tell us briefly what kind of data you were able to compile to make that association.

Peter Golder: Sure. So the data we used really comes from U.S. News. So for decades U.S. News has compiled data in both MBA programs and JD programs about all aspects of these programs primarily for their rankings. You know, people pay a lot of attention to these. They sell a lot of magazines. They sell a lot of books, because people are making very important decisions about, you know, where they're going to go to get their MBA or get their J.D. and, you know, these are, these are good for Tuck because Tuck is usually well ranked and highly ranked in all these. And so it's, you know, it's really become a whole, a whole industry. But the side benefit to us then is that we had data that covered many schools and many years. So I think one of the reasons the Supreme Court could look at the previous data and say, you know, kind of where's the proof of what everybody is saying, that diversity, matters in higher education. Where's the proof? It's that previous studies were really looking at, you know, much more limited samples of data, sometimes looking at a single university. So it's like, okay, we're going to look at racial diversity at this really prestigious institution and say, oh wow, these graduates did really well. And it's kind of like, well, they went to this really prestigious institution. or people used surveys of people who chose to respond to the survey. So you get into all these issues of just selection bias, right. So the people who are more likely to respond, maybe the ones more likely to be successful, or there's a bit of kind of pressure in surveys where people think, "This is the answer people want to hear," and then give that kind of result. So I'm just going to I'm going to give that kind of result. So There are all these biases. But, you know, in some ways, because the U.S. News data wasn't collected for this purpose, it is not subject to some of those same biases. But I'd say more importantly, it's just we haven't had a consistent data set that covered a large number of universities over a long period of time, and also allowed us to include other variables where we could control for alternative explanations, because there are alternative explanations. One of the things we had to do in our, in our paper, most of this got pushed to the, the, the, the kind of appendix of the paper and the web appendix to, to both of those. But all these robustness checks to see that the that our core finding about the relationship between racial diversity and salary was really, that held up, after considering all of these other, other possibilities and, it held up incredibly well. But that's kind of the that's the data that we had. So we could, you know, we could control for individual university effects. We could control for the effect of any particular calendar year we can control for prestige of the overall academic institution. We can control for location, we can control for size, we can control for, you know, other aspects of, of diversity. And so it's, you know, it's a very robust finding after considering all those possibilities.

Kirk Kardashian: Yeah. Yeah. Well, I just want to say, just give people an idea of, of the, the breadth of this data. you compared, well, you had data from 141 business schools, right? and 200 law schools. And the business school data was 29 years worth of data, and the law school data was 21 years. so that's I mean, that's a long time. There's a lot of schools. That's that's a big data set. So I think, you're definitely not, you know, cherry picking, a couple of schools and a couple of years.

Peter Golder: And I should say this. That's a good framing and a good point that you make is we really went into the study with, with a with an open thought as to what the as to what the result would be, because we honestly didn't know at the start of this. Because there are competing forces at work here because there's a large body of literature that talks about, essentially a racial wage gap where racial minorities are paid less in the marketplace traditionally for, for comparable, for comparable work. So this is a, this is found in, you know, when people look at, racial situations is found when they look at gender situations, but there are these wage gaps that exist. So it's possible that the results could have, could have been negative, that higher racial diversity actually would have found lower salaries. And, you know, Deb, my one of my coauthors, you know, that was really his thought going into this is that, is it that racial, essentially racial discrimination in labor markets would have led to, to lower salaries overall? The competing force here is that is what we and others have talked about, is that diversity is going to expose people to a broader set of viewpoints, you know, different ways of thinking about problems, different ways about framing problems, different potential solutions to problems. And, that is a, you know, that's that's the countervailing force that would lead to higher salaries. My opinion going into this is that my is that those two forces would have canceled out, and we would have found we would have essentially found no result. Which which wouldn't mean that the learning benefit wasn't there. It just means it would have been canceled out by this other effect. Now, the fact that we found a positive effect overall means that the learning effect essentially was more than enough to compensate for any, any, any potential, you know, downward impact on salaries from having more racially diverse class. So I think that's an important point, as well as what we're looking at. And what we're able to identify is a kind of net composite effect. You know, with the data we have, we're not able to disentangle those two effects. But overall, the effect of learning could be even greater if we were able to kind of remove the negative effect that also existed. So that's that's an important point about the study, I think, is that it's the net effect of these forces going on. And we're still finding that the beneficial effect of learning is the predominant effect.

Peter Golder: Yes. I have to say, I am personally very happy with the outcome we found. But again, our role as scientists is not to start with an opinion. The point is to let the data speak. There are plenty of other people out there looking for data to support their opinions. That is not our mindset, and it is not our job. I always want to start with knowledge and let opinions be derived from knowledge.

Kirk Kardashian: Right. That's that's that's well said. so can you give me an example of how you do a robustness check?

Peter Golder: We did a couple of things to control for the overall institution effect. First, we used ratings of the institution as a whole, so we looked at the prestige score for Dartmouth or Yale as a whole rather than just the business school within those institutions. We also used econometric models with a variable for each institution, so any effect unique to Tuck, Yale, or UCLA would be captured there. Those controls let us isolate the racial diversity measure as much as possible. Another way to think about it is a simple regression with inflation and unemployment. You would not want to look at inflation alone and consumer sentiment alone; you would want to control for both.

Kirk Kardashian: So let's talk a little bit about the policy implications of your findings. As it stands now, race-based admissions policies are against the law in higher education. Schools can still consider race as part of an applicant's life experience, but beyond that it is not really being taken into account at colleges and universities. That is federal law now. So what do you think should be done federally, at the state level, or at the college and university level to allow or consider race as part of admissions policies?

Peter Golder: Okay. That is a great question. We have thought a lot about it because the first step is to establish the empirical finding, but there are many implications that flow from it. There are still many things universities can do within the law as it now exists. As you mentioned, universities can still consider an applicant's discussion of how race affected his or her life. I believe that should carry weight in admissions decisions. The second implication for universities is that they should not only admit a diverse class, but also do as much as possible to leverage that diversity within the program. Our finding rests on a learning theory: we learn more from people who bring different viewpoints. That seems self-evident to me. Who do you learn more from, someone who thinks exactly like you and knows exactly what you know, or someone who knows things you do not know and sees the same information differently? The answer is obvious. But it is not enough just to admit a diverse class. Universities have to turn that static advantage into a more dynamic one through actual learning and interaction. That happens through the way study groups are formed, through classroom discussion, and through curricular, co-curricular, and extracurricular activities. Universities should think hard about how to leverage diversity in all of those settings. There is also an implication for the courts. I do not know when this issue will reach the Supreme Court again, but I believe it will. If business schools and law schools want to be more proactive, they could ask for the same exemption the military academies have been allowed, arguing that there is a special interest in their context and that these data show education is better. That precedent already exists. The current Court was open to that exemption for the military academies. With these data, I think business schools and law schools would have a credible argument. I do not expect that to happen in the next couple of years, but it might happen later. Educational institutions do not have to sit back and wait; they can help force the issue. This could also happen through legislation. Legislators do not have to wait for the courts. They can act on affirmative action policies and civil rights policies more broadly. And beyond education, the corporate world should examine its own environment and figure out where diversity is paying economic dividends. There are many reasons companies have pursued diversity over the years, but the recent retrenchment on the corporate side should prompt them to ask whether they are hurting themselves. They should look closely at their own context and ask where diversity helps them most. Even from a purely economic perspective, the question is whether cutting back on diversity makes a company less competitive.

Kirk Kardashian: Yeah. Well, Peter, it is a really amazing paper. Very important Nature finding: more diverse schools lead to higher student salaries. Thank you very much for explaining that paper to us. It has been great to talk with you.

Peter Golder: Well, yeah. Thank you so much, Kirk. Great to talk to you, as always. And, yeah, thanks for helping us disseminate these findings to a wider audience. I appreciate it very much.