Steve Hsu: Welcome to Manifold.

Today, our guest is the Jesse Dukeminier professor of law at UCLA Richard Sander. Richard was educated at Harvard College and studied at Northwestern Law School and also obtained a Ph.D. in economics at Northwestern. So although he has a legal background, he is also a trained economist and an expert at dealing with empirical and statistical data. We are going to discuss a phenomenon, which I don't know that I should say he discovered it, but he certainly coined the term that everybody uses to describe it, which is mismatch or mismatch theory. And roughly speaking, it refers to a situation where in say higher education a certain group of students is granted preference in admission.

So that group might be admitted to a highly competitive program. For example, an elite law school or an elite department of engineering, where their preparation, for example, as reflected in their test scores or their high school grades is not as strong as the majority of students with which they are going to have to compete. And Rick has studied in great detail the consequences of that set of circumstances. And obviously, that set of circumstances is germane to affirmative action. So what might seem as a completely well-motivated topic in education or study of human capital, has become extremely controversial.

And Rick himself has become to some degree controversial. So I'd like to have a very detailed, deep dive into this topic during our conversation in which we explore the evidence for and against mismatch theory, the reaction that Rick has had from the academy, from his colleagues, and other professors of law.

And then finally we will get into real political issues, such as Proposition 209 and Proposition 16 in California, and the upcoming Supreme Court case regarding affirmative action at Harvard. So Rick, welcome to the podcast.

Richard Sander: Thank you for having me, Steve.

Steve Hsu: It's a pleasure. And I just want to say to the audience that I've had the pleasure of meeting Rick in person in the past, and I've listened to many of his presentations and he's really a scholar and a gentleman.

I'd like to start with your early life. And I just sort of gave your educational background, but I noticed in your biography that in between Harvard and beginning graduate study at Northwestern, you actually worked in Chicago for, I guess, several progressive causes. And this would have been, I guess, in the late seventies, early eighties.

Richard Sander: Yeah, that's right. To go back even further, I grew up mostly in rural Indiana. My dad was a first-generation college student who got a Ph.D. in history and taught at a small college in Indiana. Um, my mom was from a pretty poor family in New York. And we lived in an environment that was not particularly intellectual.

So, it was kind of a, I wouldn't say a working-class background, but it was certainly a mainstream America backward. And one where our family's interests and advocacy for civil rights was sort of a minority position in our community. There weren't that many Democrats in our, in our hometown.

When I went to college, I was really interested in politics and policy issues. And when I finished students like me were mostly either going to law school or they were going to Capitol Hill to do internships with congressmen and, and things like that. And I decided to take a different path because it seemed to me that I didn't really know yet what were good solutions to problems in the United States and that I needed to get more data. I needed to sort of find out what was going on. I was particularly interested in trying to understand what was going on in cities. And within cities, I was particularly interested in understanding what was going on in Black communities, because so many of the problems that we talked about in college related to race and, and urban decline and inequality.

And at Harvard, I, you know, I had done a few community projects, but really felt very naive about all of that stuff. So after college, I volunteered for what was then known as Vista. Now it's called AmeriCorps. It's kind of a domestic peace corps program. And I was assigned to spend a year with a new neighborhood organization in the south shore community of Chicago.

And basically, I was trying to help organize tenants to deal with landlords who were walking out on their buildings. I was helping to figure out whether there were particular multi-family buildings that we could put in trusteeships and help get help from local financial institutions to maybe refinance these as, as cooperative housing.

So doing a variety of things, working with low and moderate-income folks in a neighborhood that had recently gone from being predominantly white to predominantly Black. And through that, just got a great education on sort of urban problems in America.

Steve Hsu: Well, it seems to me that someone who has the goal of achieving progressive goals, like greater equality, advancement for underprivileged minorities. Even somebody with that eventual goal, if they're honest with themselves, they have to understand how the world works in order to figure out the best ways to achieve that goal.

Richard Sander: That's true. It's there, even back in the ‘70s, there was a lot of double talk about how to effectively address issues of racial inequality and income inequality, and I think a general reluctance to candidly evaluate what we were doing that was working and what we were doing that wasn't working.

So that's been an ongoing issue that I ran into pretty soon after college and have continued to observe.

Steve Hsu: Right. So I've always felt it ironic that someone like you, who I feel has a very strong intellectual commitment to finding out the truth — What is actually happening? How are things actually working? And with a progressive goal in mind — Nevertheless gets characterized because of some of your empirical conclusions as somebody who's on the right. Who's somebody who is quite nefarious in their goals in the world. And I just, I just find that to be incredibly unfair.

Richard Sander: Yeah, it's, you know, it doesn't always happen, but it happens a lot and it reflects I think the desire of… Well, it probably reflects two things. One is that is the people who are observing any debate, they like to put people into camps. And categorize them. It's just easier to understand them if they can stereotype them as conservative or liberal. The other part that I wouldn't say is nefarious is people who know better consciously try to characterize me as conservative.

So soon after I wrote my initial article about mismatch, there was a piece in Slate by Emily Bazelon that became inextricably linked with my name on any Google search for Richard Sander. And in the piece, she highlighted that I was championing the mismatch theory on Fox programs. And what she failed to mention was that I was on one Fox program as of the time she wrote, had been on many, many more NPR programs. And on the Fox program, I spent most of my time criticizing the host for oversimplifying my views about affirmative action.

So you know, that is and has been a problem. But going back to the origins of my career, I think the thing that really struck me, in my earlier work in Chicago, was how important the problem of housing segregation is and how much that creates so many of the other conditions that I think, contribute to racial inequality in America.

And when I eventually went back to graduate school, the thing I had in the back of my mind was trying to use law and economics to understand housing segregation. And I ended up doing my dissertation about that and work on that has continued to be sort of as large a part of my professional orientation as the work we're talking about on higher education.

So maybe a little later, we'll talk more about the housing segregation issue.

Steve Hsu: Yeah. I was going to say, we can discuss that if we have time or maybe have you back on the show to discuss it. One question I always wanted to ask you, and I guess I'm surprised I never asked you in the past about it is how did you become aware of mismatch as a phenomenon? And I'm guessing it occurred to you or it became clear to you after you became a professor.

Richard Sander: Yeah.

Steve Hsu: So maybe you could say a few things about that.

Richard Sander: Yeah. So from '78 to '83, I was doing community work in Chicago. From '83 to '88, I was in graduate school and I clerked for a judge. And then I came out to UCLA in 1989. And I was on law school faculty and I was really the first empiricist on the faculty. And because of that, administrators got me involved in various types of analysis that they wanted to do.

So I did an early memo on some potential reforms to the admissions process. And that was kind of entertaining because the law school was doing some pretty bizarre and antiquated evaluation methods that really weren't doing a great job of identifying the strongest students. It had nothing to do with race. It had to do with the sort of stereotypes they had about elite versus state universities, for example.

So I helped them revamp the admissions policy and then someone approached me, someone who was working on academic support programs in the law school, said that she would really like to evaluate the various programs they had and figure out which ones were really improving student long-term performance.

So I found that fascinating, and we ended up putting together a 10-year database on all of the students in the law school and what types of academic support they had experienced. And we were able to use that to identify a couple of programs that worked and some other programs that didn't work. So I was very happy, right. I was getting good access to data. There were no restrictions on the questions we could ask. And when we found something worked better than something else, the administration actually changed the policies to reflect that.

But in the course of doing that, I realized how much our affirmative action policies were concentrating our minority students in the bottom of the academic spectrum. So the law school was using origin of preference. So that most of the weakest 10% of the students we admitted were African-American. And many of the rest were Hispanic. And that was then translating into their performance in law school.

So that about half the students in the bottom, 10% of the class, were Black. I later found out this was very typical across almost all law schools. And it got me wondering whether those students might have been better off going to a less elite school. I couldn't tell from that data because I just had data for UCLA.

But seeing how many of those students struggled and seeing how many didn't pass the bar, the hypothesis just sort of jumped up that, well, what would happen if, if they were at a school where they were closer to the middle of the class. And that's where the idea of the mismatch came from. But there wasn't anything I could do about it for many years, because I didn't have any data on any other law school.

Steve Hsu: You know, my own introduction to the concept of mismatch. Just because of my own unique personal history, occurred already in high school because I grew up in Iowa and my father was a professor in engineering, professor at Iowa State University, which is a very good public university. [It] produces lots of scientists and engineers that work in industry and elsewhere.

But it is not an elite university like Caltech, where I eventually went to college. And in high school, I had taken a lot of courses at the university. So I was very familiar with the level of the courses offered at Iowa State. And my professors at Iowa state all told me, oh, Steve, you're kind of talented. You should, you know, you should apply to Harvard and Caltech and places like this.

And if you think about what they're saying, they're effectively saying there is an appropriate place for someone who has some intrinsic talent for science or mathematics, and we're trying to segregate the population by talent because the way they teach at Caltech is more appropriate for people who have some innate ability or very strong preparation. And at Iowa State, we don't necessarily get students quite of that caliber. And so we teach in a different way. So the whole concept is already ingrained in the way that STEM education is designed.

And so it is no surprise to me that I had many friends in high school who were good students who attended Iowa State, have since graduated and had long careers as engineers and scientists. But if I reflect on how they would have done had they come with me to Caltech as freshmen, I think it would not have been right for them. They would have had a hard time. They would have probably been at the bottom of the class, even though they eventually became good scientists and engineers. They just weren't ready for that pace in which, already at the undergraduate level at Caltech, they would start using graduate-level textbooks.

And so, you know, the whole idea that this wouldn't be a problem, and we're not talking about race here in any way. You know, I grew up in Iowa, which is overwhelmingly white. So my best friend, you know, my best friends growing up were all wasps and some Jewish kids. And some of them just, you know, they did perfectly well at Iowa State, but they weren't necessarily going to do well at Caltech.

So that whole concept of mismatch seems, I find it hard to believe people can't accept that this kind of problem might arise in higher education.

Richard Sander: Yeah, right. It seems obvious. And in fact, you know, Christopher Jenks wrote a book with David Riesman, both famous sociologists back in 1968 or 69. It was called the Academic Revolution. And part of what they discuss is really indistinguishable from mismatch. They're not really getting at it in a racial context because racial preferences were just starting then. But they were just talking about how as American higher education has become very big and complex, one of the benefits of that and one of the consequences of that is that there is a lot of segregation by talent. And that makes it possible for people to sort of get the most out of their environment. So part of what you're describing Steve, I think is, is that we would expect intuitively that there would be both something called negative mismatch and something called positive mismatch.

If your skills are way below those of your classmates and your classmates are relatively clustered in ability, then you're going to learn less. And if your skills are way above those of your classmates, and they're clustered tightly, and then you're also gonna learn less. You know, you would've benefited less from being at Iowa State, than being at Caltech because you had a good set of skills to take advantage of everything that Caltech offered.

You probably would have been kind of bored with some of the courses at Iowa State.

Steve Hsu: Yeah. So I think that in private, and also if you sort of de-racialize the discussion. So if you're just talking about, you know, the other kids in my high school class who wanted to study STEM, who are all white, where they should go to college, the guidance counselor and the principal and our high school physics teacher, you know, would've not felt guilty at all about saying, well, these students maybe should take a shot at Ivy League schools. These students you know, will be perfectly fine at Iowa State or University of Iowa. And there just isn't a problem there. But when you then introduce the ingredient of affirmative action and other sort of external ideological goals, then suddenly the whole topic becomes verboten. And then you can’t have a very clear discussion of it, at least I find.

Richard Sander: Exactly, you know, to make sure credit goes where it's due. There was an article that appeared a year before my initial mismatch article in a major education journal. Written by two guys, I guess they were psychologists, Smyth and McArdle. And they were basically interested in this question of whether high school counselors should advise students to go to schools where they're well-matched.

And they got access to a great database. The same database used by [unclear]. And they were able to look across 30 schools and documented very clearly that in terms of STEM fields, there was a real mismatch curve there wasn't optimal point for a student to aim for. They included race in their analysis and they pointed out that due to existing preference policies, there, there was there were a lot of mismatched Black and Asian students who were floundering in STEM programs.

But they also did a great job of showing to anyone who was paying attention that when you took race out of it, it was internally driven by mismatch. In other words, Black students who were not mismatched, did exactly as well as white students. White students who were mismatched did as badly as mismatched Black students. Race just kind of fell out of the analysis. But because race was so important to the admissions process, it became something that disproportionately affected minorities.

And as you predict, the consequences of making race a subject of their article, led the article to be kind of, ignored. It found its way into a good journal, but I don't think anybody ever cited it. And the authors just felt very frustrated that they had this major finding and it was being ignored by the education establishment.

Steve Hsu: So this podcast has a fairly kind of geeky, quant listenership. And so I feel okay just jumping in and saying something like if you build a model, it might be just a simple linear regression model, to predict which students are most likely to succeed, say in a selective STEM program. One of the variables in that model might be their SAT score. Another might be their high school GPA. And another might be their race. And I believe all the statistics, whether you're talking about hundreds of thousands of data points in the UC system. I forgot how many we had at Oregon when I studied this problem. When I was a professor at Oregon, we had easily tens of thousands of individuals in our data set.

And I think that the recent Duke study that Arcidiacono analyzed, they all show that that third variable race has very, very small coefficient. In other words, if you blind yourself to the race of the individual kid, but you know their high school GPA and you know their high school SAT score, then you do a reasonably good job of predicting how well they will succeed in this selective program.

Richard Sander: Yeah, that's exactly right. And one of the, it's a bit of a side point, but one of the implications of that is sort of the reverse, which is that SAT scores are not biased against minorities because they predict their performance as well as they do for whites and Asians.

Steve Hsu: Yes. And so a few other empirical observations, which I think, you know, these are things which the general public discourse on this question on the question of college admissions would benefit from if everybody knew these facts. So if you are a rich legacy, it's still the case that your SAT and high school GPA do about as good a job of predicting your performance, say, at Duke, as it does for someone who's not a rich legacy and perhaps is of a different race.

So this is something that I think people are often confused about. They think that socioeconomic status actually can directly influence how well you, for example, perform in the engineering major at Duke, but the data certainly doesn't bear that out.

Richard Sander: That's right. That's right. Yeah. High SES may help you do better on the SAT. It might help you, you know, get a better high school education. But we can control for those things. And when you do, SES doesn't help very much in for the prediction of college and graduate performance.

Steve Hsu: Right. So in this overall discussion, I think the empirical facts are clear. And when I privately discuss these things, you know, with social scientists who maybe are, have very strong left. I mean, I myself am left of center. But these people would be, say much further left than me. They can privately acknowledge the empirical facts, if, for example, they're trained empirically.

Now at this point, something funny happens. Some of them may then if asked in public to comment on the situation, just feign ignorance. Maybe they don't want to comment at all. Even though in private, they would admit that my interpretation of the analysis is correct.

The other possibility is they openly contradict in their public statements what they might admit to me privately. But then they might say to me privately, if we meet up again saying, well, I had to do that, Steve, I have larger goals. And you have to just understand, we can't talk about this openly. And I'm curious what range of behaviors of that type you've encountered.

Richard Sander: Well, definitely both of those. You know, I would go beyond saying I've had conversations with academics and say I've had conversations with university presidents who have acknowledged the mismatch issue. And they’ll tell me in the private conversation, I can not acknowledge this publicly because if I did well, first of all, I'd be fired, but they usually don't bring that up.

They say, you know, minority students would be chilled away from attending my campus. If I admit that my campus has a mismatch issue that will wipe out millions of dollars of effort we have spent on minority recruitment. Right? You see the logic. So, it's almost mentally ingrained that they need to maintain this private-public distinction when talking about this issue.

Steve Hsu: I should also mention one more category of mental state, which is that they might acknowledge the empirical findings that we're talking about, but because they're not really intimately familiar with how admissions works and how affirmative action works, they're just not aware of the size of the preference that's awarded. That they could, they can just assume that it's a small preference and because it's a small preference and these kids come from disadvantaged backgrounds, that they'll just catch up. And it all's good in the end. And, so that's a third mental state that I've encountered in this discussion.

Richard Sander: Yes. Yes. So, and it's important to point out that there are probably a much larger group than the groups we're talking about, who really believe that mismatch and other affirmative action issues are not problems because they've adopted a rationalization. So one rationalization I hear a lot is in the law school mismatch context. It's true these students may really struggle and they may have trouble passing the bar and so on, but this will be more than offset by the value of a degree from our elite school. So the mismatch is not really harmful because you're getting this elite education out of it and that's, that's so great that it more than offsets the mismatch effect.

That turns out to be, well in some contexts, it's it may be a true rationalization. There is some evidence that if you go to Harvard or Yale or Princeton, even if your GPA suffers your long-term earnings are still better than they would be if you went to a weaker school. So if you're interested in a non-STEM field and you're thinking of going into an undergraduate campus, there's a respectable argument that the mismatch is offset by the eliteness of the school.

In STEM fields or in law school, probably a medical school, I think it's obvious that the mismatch swamps everything else. I've done long-term studies of lawyer careers, for example, and if you have a bad GPA, that translates into really not knowing the law very well when you go out and start practicing, and that really injures your career. You just find very dramatic long-term career effects of doing badly in law school.

Steve Hsu: I just wanted to dive a little deeper into this point because of whether, even though they're mismatched through admissions preferences and they really struggle, say at Harvard college and they earn a low GPA. Nevertheless they come out ahead because it's better to have a Harvard degree, even if your GPA was not high than a degree from the University of Iowa.

But that's an empirical question, right? So then you, you need to study the life outcomes. You need to look at some longitudinal data and ask, okay, so did, did earning a low GPA and graduating in a non-STEM field from Harvard, whereas your intention was originally to go to Iowa State and major in engineering, what happened in your life?

And you've actually done the study for legal careers, where you compared graduates of non, maybe non-elite is the right way to say it, law schools. But kids with high GPAs to kids who attended elite law schools, but didn't do so well and maybe were mismatched. And I think you, I think you just said it, but I just wanted to make sure that the listeners got the point that you found that the long-term consequences were net negative from mismatch for that population.

Richard Sander: Yeah, very strikingly. For example, well, yes. So if you, if you directly compare the two students who go into law school with similar credentials, but one makes the choice for the more elite school has the lower GPA. The other makes the choice for the lower, the less elite school and has a higher GPA. That second student is better off in their career.

We also did a study where we just looked within one school, which did that from the University of Michigan, which pioneered in tracking its students longitudinally through their careers. And we found that if you were white — we just limited it to white students, sort of take, make it avoid the issue of trying to control for race — there were many more white students in the sample than those of other races.

So among white students, if you had a high GPA and you went to work in a law firm, versus if you had a low GPA and went to work at a law firm, you were about 10 times as likely to be offered a partnership by that firm if you had a high GPA. And everyone in law firms knows that when partnership comes up, those decisions are not made based on your law school GPA. That's kind of an invisible factor as far as they're concerned. The initial hiring decision is influenced by GPA, but by the time you come up for partnership, they have seven years of your work to evaluate.

So the students who were going through and getting low GPAs pretty clearly had inferior work. They just were not performing as well. And that's because they didn't learn as much in law school.

Steve Hsu: So, you know, I mean, this, I think this follow-up study. I remember I think reading it and I believe it or not people are reading your papers or reading this study. I think when I was a professor at Oregon and we were looking into the predictive power of SAT and things like this in the Oregon data. And I just feel like at this level of … So you, you know, you, you made a multi-year commitment to try to sort out what is really true. What is really going on here in our affirmative action regime here in the United States. And so this final level of empirical detail, I feel there were very few people who actually have appreciated this. So they might admit like, yeah, the mismatch kids are going to have trouble while in law school. But having a Stanford law degree, that will ensure that they're perfectly fine. It's better than having a UC Davis law degree.

And you've pushed it further to actually analyze what happens after that. And I have to say I've met no university presidents, I know quite a few and top administrators and deans who do understand the empirical data at now at this level. I think this is not something that's generally understood by these people.

Richard Sander: Yeah, I think that's a fair point. And it goes to sort of the other type of rationalization that I think has happened. When my article came out on law school and mismatch, this was early 2005. As I said, there was this earlier, earlier study of science mismatch. There were a couple of other people who had talked about mismatch.

What was different about my article is that it got a tremendous amount of play in the media. And I've never met a law professor who hasn't read it. I mean, it was downloaded something like 60,000 times after it came out. And it was published by Stanford Law Review. And as part of their publication agreement, they said, well, we're going to do a follow-up issue where we're going to have someone reply to your findings. And I said, well, that, that's fine. As long as I can write a short, you know, rejoinder to the reply. They said, okay.

Then they called me up a few months later. They said, well, actually we'd like to publish three replies. I said, well, okay. As long as you give me enough space to reply to all three. Then they called and said, well, we're going to make it four. And it turned out that they not only published four replies, but they had sort of had an open competition for these replies and they decided as a group of editors that they would only accept critiques of my piece. They wouldn't accept pieces that sort of said, yeah, I ran Sanders’ numbers and they hold up.

So they published this follow-up issue and I responded but it really kind of poisoned the whole debate thereafter because you had these four articles that were all very partisan. And two of them were pretty empirical. They were by people that I had respected. Their pieces I was later able to show them that they had basically distorted the data they used to get the results that they had. They were really, you know, ingenuous critiques. And for an awful lot of people in the legal academy, seeing those critiques gave them a cover. They could say, well, I read Sanders’ article and it sounded persuasive. But if these guys think that it's baloney, then I don't really need to give it further thought.

And that I think that's been the most insidious influence on the discussion is I've met people who are ideologically committed to believing in affirmative action. Now they'll say, well, you know, you've been refuted. It's not a viable explanation.

Steve Hsu: I whenever people refer to mismatch, they always, always include the terms, controversial theory or contested theory. Whereas just a simple graph of, for example, in the w when you see was recently considering whether to get rid of the SATs, there was a faculty committee, UC-wide faculty committee assigned to evaluate this. And they came back saying that actually the SAT has a lot of predictive power and we need to keep it. But any of the graphs that were produced in that analysis strongly demonstrate that, you know, mismatch is a problem. Because for all groups, the predictive power that, you know, the regression coefficient for SAT and high school GPA was about the same.

And so it's just hard to imagine people saying that empirically, this doesn't stand up to further analysis.

Now, I listened in preparation for this interview with you. I listened to a really great panel discussion which happened, I think in 2016 at Stanford. Maybe you remember this, there was a woman professor at Stanford named Pam Karlan, I think is her name. Who's a brilliant, brilliant speaker and very insightful. But she just would not engage you. And she kept saying, well I'm not an empiricist, so I don't need to engage you. Cause I don't need to really know the truth or falsity of this. It's not my specialty. But you've been refuted by these other people.

So once that happens the people who are — I’m not saying she was speaking purely from ideological basis, she seemed quite fair-minded — but it gives people the option of just saying, well, it's contested. It's controversial. Sander, his analysis doesn't stand up. And to me, it's just a failing of the academy that this kind of situation can just be allowed to persist for decades on a question which has real relevance to people's well-being.

Richard Sander: Yes. And, you know, let me make a couple of points there. So one big problem. Probably the biggest single problem is that of data transparency. After, after the article and the critiques came out, I and some colleagues realized that there was a perfect data source for studying law school mismatch.

And that was data that was maintained by the California bar. The California bar was unique among state bars in that it kept a longitudinal database, not only on bar outcomes but on students' law school GPAs, and their entering credentials. So with that data, you could basically do a regression controlling for entering credentials and race and the degree to which someone was mismatched with their particular school. In other words, how big a gap there was between that student's credentials and the median student at their school. And you can see what predicted their bar score. If, if we'd been able to get that data, I'm, I'm certain that that would have settled the debate because it would have been so obvious and so powerful that those who had previously denied it would, would have to confront it.

I mean, there, you know, arguments made in my original article required you to sort of really think about what the causal mechanisms were and, and really, you know, sort of process what might be going on. There are lots of kind of cognitive escape valves in that argument. But if you wind up how people with the same credentials are doing at different schools in terms of our performance, it's just too obvious to ignore.

So we tried for 10 years to get that data. If I had asked for it before publishing my work, I think they would have readily given it to me. The chief psychometrician at the bar was totally on board and was lobbying for the standard to be made available. But once it became political and once the law school dean said no, there's too much confidential information there. The bar went to unbelievable lengths to keep the data secret. After we took them to court and won a unanimous Supreme Court ruling that they had to disclose the data, they actually got a law changed. A special measure passed by the California State Legislature, specifically exempting this database from disclosure.

Part of the problem is that there hasn't been enough transparency to make it easy for people to replicate and do this in a variety of convincing ways. The second problem is that when you have a debate like this, it tends to drive moderates out of the arena. The people who, most of the people, 90% of the people who sort of offered themselves as allies to my work or people who were on the right and had had already decided in their hearts that affirmative action was bad and they seized on this as further evidence.

So those are not really the allies you want in persuading mainstream, higher education that they need to examine this issue.

Steve Hsu: You know, I have to disagree with one thing that you said there, which is that, had you had access to this data and done the analysis, say of bar passage rates, which I believe one hundred percent would have confirmed what your earlier work suggested that it would have convinced people. And the problem is that they don't want to be convinced.

So if I look at undergraduate admissions, all the data's on the table. Okay. UC is routinely analyzing data sets of, you know, hundred thousand, 400,000 kids who pass through the UC system. Arcidiacono has every grade in every course for an entire cohort of kids who pass through Duke. He has their application files from high school. He knows their SAT scores, their high school GPAs. He even knows what their intended majors were at Duke. So he can see them dropping out of hard STEM majors they declared. I want to be a biomedical engineer. That's what I'm coming to Duke for. And he can see them in real time dropping out of biomedical engineering because they didn't do well in calculus or something like this.

So we have the data for that other problem. The other problem is undergraduate-level admissions. We have all the data. Very insightful, very well done papers have been written on this. Even official papers by the UC Faculty Senate. But it can't convince anyone because honestly, I feel just, people do not want to be convinced because they have ideological pre-commitments.

Richard Sander: Well, so let's engage with that for a minute. I mean, Arcidiacono, I think he would be the first to say that there's a transparency problem. He was able to get good data from Duke. He never will be able to again because Duke had to denounce him after his paper came out. I worked with him for years to get data from the University of California. And we eventually did get some data, but it was blurred to a very great degree.

And as you know, blurring data tends to weaken any sort of causal relationship that you're trying to establish. So he published a couple of good papers about mismatch at the University of California, but they were, you know, the coefficients were smaller than they would have been with really accurate data.

So the data access problem is actually very large. I mean, the academic Senate study at the University of California that you referenced doesn't really have good outcome data. I mean, they had enough to do, to show some correlations. But they didn't do any mismatch analysis in that paper. University of California gave the really good data to somebody who had been prescreened as someone who has gonna parrot the conclusions they wanted to draw. That's a guy named Zachary Bleemer who's an economist, I think, I'm not sure where he is now, but he recently graduated from UC Berkeley.

So Bleemer is going around with these papers where he claims to have disproved mismatch using this very granular University of California data that the university won't make available to anyone else. Bleemer won't make his data available to anyone else. Arcidiacono has pointed out that when you actually look at Bleemer's tables, they don't fit very well with the story. But this whole transparency issue is really a huge problem.

Steve Hsu: Just to clarify what I said earlier. I wasn't suggesting that transparency is not a problem. I think I agree with you, data transparency is a problem. I just meant even if you did have full access to the data, would you be able to convince people. That you're right. And there I still have a sort of, at this point in my life, a kind of pessimism about human nature as to whether they'll admit they're wrong when they're actually wrong [or] shown to be wrong.

Richard Sander: Well, you may be right. Let me tell you another story that will sort of test this. So I never got the California bar data, but I did get data from three different law schools through public records requests that allowed me for the first time to show, you know, with this level of mismatch, your bar passage rate. With this level of credentials, what's the degree of reduced bar passage rate that occurs from going to a somewhat less elite school, a much less elite school, et cetera. So there are tables in that paper that just document mismatch much more vividly than my original work.

So I decided to, I have a coauthor of this. We decided to submit the paper to the Journal of Legal Education, Which is not really an empirical journal, but it's the official organ of the American association of law schools. Every professor gets a free copy of this journal. And I thought this will be an interesting test case. Let's see if we can get them to publish it.

And we called them up beforehand, said we've heard from colleagues that papers on politically sensitive topics tend to get buried in your editorial process. Can you commit to us in advance that you'll give this paper a fair and expeditious review? And they thought about it for a while and said yes, we'll make that commitment.

And we gave them the paper before they said that. So we submitted it and they sent it out for peer reviews. Peer reviews came back, very supportive of publication. So they sent us a letter saying that they were accepting the article and they had some changes that want us to make, and they wanted to publish a reply with it so they can keep alive the controversial story.

So we accepted that and we started revising it we sent it back to them. By the time it went back, they'd had a couple of changes in the makeup of their editorial board. So they sent it out for another peer review, which came back fine. And then they wrote to us and said, well, we've thought about it some more. We're sorry, we can't publish this. And I said, but you've accepted it. And we made all these changes in reliance on your acceptance. And they said, oh, well, okay, well, if you're going to insist, we'll think about it. And they thought about it for a year. And finally, I wrote to them and said, basically threatened to sue them if they didn't, you know, clarify what they were doing.

So as of about a week ago, they formerly have a set date for publishing the piece. So I think that illustrates kind of what I was up against. But also potential to maybe bring folks around if the data is clear enough. I mean, it'll be interesting to see what the reception of that piece is.

Steve Hsu: Yeah. I don't want to be too pessimistic about the powers of human reason. I maybe I'm just especially cynical at this stage in my life, but to take a much weaker proposition than mismatch, because obviously mismatch you're talking about fairly complicated outcomes and things like this. If you just ask questions, if you just ask questions, like, is the SAT biased as a predictor against a particular group?

And, you know, I think you'll see every day in the newspaper claiming that it's well understood that it is. I mean, it's demonstrated not to be true. Right. So if you look at regression coefficients, pick a university and look and see how well does the SAT or ACT predict success for this group versus that group. It's pretty unbiased. And so like to me, that's maybe a better example of something that's kind of very clear in the statistical data. All the psychometricians agree on this, but what you read in newspapers serve exactly the opposite of that.

Richard Sander: Yeah, that's true. That's true. So maybe cynicism is very well warranted.

Steve Hsu: I don't want to discourage you because I, you know, I think people like you are, you know, you're role models for real researchers, real scholars, real social scientists who want to really get at the truth, even if they don't like the truth. Right? I mean I would much rather learn that, oh, affirmative action works great. There are relatively modest preferences given to people. They catch up when they get to college. They have fantastic careers and there's no issue. I would much rather learn that to be true.

Richard Sander: Yeah. Yeah me too.

Steve Hsu: Yeah. exactly. And no one wants to be the bearer of bad news.

Richard Sander: When I went to law school, I was at law school at Northwestern and I was in a section of a hundred students. There was only one Black student in the section. It was pretty clear that Northwestern was not using racial preferences at the time. I don't really have any insight into that.

Yeah. But that student was a terrific student. He was, I don't know if he was first in the class, but he was, I'm sure he did very well. And was not mismatched. I came to UCLA and when I arrived in '89, it was already about 30% nonwhite. Lots of Blacks, lots of Hispanics, lots of Asians. And I loved it. It was, for someone with my interest in background, this was wonderful. And, you know, I noticed some differences in classroom performance but, you know, it seemed to function pretty well as a community. And it was only when I went to the data that I realized what a serious problem there was.

So it was very disappointing to find that it was the opposite of what I had hoped for.

Steve Hsu: It really hit me as a, it wasn't an issue that I was really interested in growing up. Or when I was a student, but it hit me in a very personal way because when I was at Caltech, my roommate was an African-American kid who was extremely strong and he went on to get a Ph.D. at Harvard in physics and he works at one of our weapons labs now as a physicist. He and I are still close friends. But Caltech didn't practice any kind of preferential admission. So of course, no surprise that he would be strong because he was able to get in.

But then I went to graduate school at Berkeley and I was a teaching fellow for some engineering classes, classes that are required for basically all science majors at Berkeley people. Certainly people who wanted to go into engineering. And there I encountered a lot of African-American and Hispanic students who, you know, I would be their TA and I realized they didn't, for example, know the trigonometric functions very well. They didn't really know sine, what sine was very well. And so I started asking them, how did you learn this in high school? And they would say no, our trig class in high school was really bad. And then I said, well, didn't, you have to know this to get a high enough SAT score to get into Berkeley engineering. And was only then that I realized how affirmative action works.

And I was very alarmed. I was really actually outraged on behalf of these kids, because if these kids had had an extra year or two of preparation at a good high school, they could have become engineers. But what typically happened in that era, I think maybe the affirmative action for Berkeley undergrad was extremely egregious, almost all of these kids failed. They couldn't pass the physics class. Understandably. I mean it was a calculus-based physics class. And their trig preparation level in their calculus preparation was lacking. A lot of these kids were my friends because I was quite young when I went to graduate school. A lot of them were the same age as I. So I saw this up close and I was just outraged.

I, you know, I went to the professor and I said, why are you guys doing it this way? This just seems ridiculous. You're damaging these kids. And the professors would just say, well, it's not, you know, we're not in control of admissions. You know, I didn't know. Professors would just say, we're not in charge of admissions and this is what's happening right now.

Richard Sander: This was probably around 1990.

Steve Hsu: This was the mid-eighties.

Richard Sander: Yep. Well, the data I have on UC goes back to 1992. And at that time, the preferences were still so large that the four-year graduation rate for undergraduates at Berkeley and UCLA was in the teens. It was below 20%. Yeah, I mean, what an unbelievable disservice.

Steve Hsu: And you know, that 20% even takes into account a kid who wants to do engineering, but can't get through the physics class or the multi-variable calculus class. And then they just switch majors. Well, they end up, you know, doing — not to knock sociology or whatever other major they switched to — but they end up in that major and still the aggregate graduation rate is, I guess you just said something like 20%.

Richard Sander: Yeah. Which is why after California voters banned preferences in 1996, even though UC admitted fewer Blacks, the number or the absolute number of Blacks getting STEM degrees increased dramatically.

Steve Hsu: Yes. So now you've segued into a, masterfully segued I would say, into the third topic I wanted to discuss, which is Proposition 209 and Proposition 16, which I think perhaps quite a few of my listeners are not familiar with those. Those are ballot measures. Is that right? Which modified the constitution of the state of California. Is that correct?

Richard Sander: That's right. This whole, you know, there are now many states that have adopted bans on preferences, but it started in California. It started with the regents of the University of California voting in the summer of 1995 to get rid of racial preferences. And that happened because of an unusual confluence of events. I mean, it's hard to imagine now, but at the time California had had Republican governors for the previous 12 years. So the regents were almost all Republican appointees. The governor himself, Pete Wilson, was planning to run for president in 1996 and thought that affirmative action might be an issue that he can incorporate into his campaign. And there was a, an African-American regent named Ward Connerly who got very alarmed about the mismatch problem when he started, you know, doing work in his capacity as regent.

So the combination of Wilson and Connerly and some other interested activists just kind of led to this proposal that UC banned preferences and almost simultaneously a ballot measure qualified for the 1996 ballot, banning preferences statewide.

So the regents voted this ban. They then voted to stay enforcement of it till they saw what voters decided. And voters decided by about of, I think a 54 to 46 margin in favor of the ban, which was Proposition 209.

That then led to an organization founded by Walter Connerly that sponsored similar initiatives in a dozen other states. And in most of those states, the bans passed comfortably.

So the interesting story here, Steve, I think is that after this ban was passed at the University of California, there was a pretty good faith effort to implement it for about a decade. There were some exceptions. Law schools in particular were, kind of started straying from race neutrality pretty early. But most of the undergraduate campuses and all the graduate programs, I would say were race neutral from about 1997 to about 2007.

And you saw kind of this great natural experiment happened where the university invested in better outreach to low performing schools. They started really trying to address the underlying problem, sort of the educational preparation gap across racial and economic lines, and that produced many more students in the applicant pool and admitting them race neutrally led to the disadvantaged students doing much, much better in their college performance.

So GPAs went up, four-year completion rates for degrees increased sharply, and most markedly of all the STEM fields were now, you know, very racially diverse and had some more [unclear] success across different racial groups.

So it was a stunning success, but the university couldn't bring itself to admit it. I guess this goes back to your point that so many of our academic leaders just feel they can't say anything positive about race neutrality. They can't bring themselves to be honest about the evidence. So you started having all this pressure from students to re-introduce preferences and one school after another started doing it.

So over the next 10 years, racial preferences came back almost to where they had been before, before prop 209.

Steve Hsu: And you know that because of the statistics that they're sort of basically ignoring the law. Is that [right]?

Richard Sander: Yeah, I've got data from enough different schools to observe it. And, you can also infer from the relationship between aggregate test scores, statistics for California, seniors in high school and the profile of admitted freshmen. So yeah, so prop 209 was essentially dismantled as far as its application to University of California was concerned by the late 2010s.

And then the coup de grâce was the legislature passing a measure to formally repeal prop 209 that became prop 16 and one on the ballot in 2020. It was passed overwhelmingly in the legislature. And I think the proponents were very confident that California was so different demographically and politically in 2020 than it had been in 1996, that it would easily pass. And they were shocked to see in the early opinion polls that it was in serious trouble. And it was eventually defeated by an even larger margin than prop 209 in the past. It was, I think, 57% against repeal and 43% for.

Well, the fact was that in 1996, the vote had been somewhat racially polarized. Whites had voted very differently from Blacks in 1996 on 209. But on prop 16, the vote was pretty similar across racial lines. There wasn't any group that was happy with this.

Steve Hsu: And now that prop 209 has been sustained by the defeat of prop 16, do you think the UC system will toe the line again, or will they just continue disobeying the law?

Richard Sander: So far, all the evidence is there going to continue to disobey and maybe even ramp it up further. You know, the academic leaders that I've heard from have just kind of ramped up their defiance. There are all sorts of new initiatives to make more rigid race-based hiring for faculty positions.

You know, we now have these pervasive equity, diversity and inclusion statements that are used as screening devices for who's even going to be considered for a new faculty position. So it's really become pretty wild. And I think that the only thing that's going to restrain it is possibly a lawsuit within California or a Supreme Court decision that bans affirmative action nationally.

Steve Hsu: Well, before we switch to the SCOTUS case, I would like to ask [of] all the groups — and I kind of, I'm kind of aware of this because some of the Asian, I know some of the people involved in the Asian-American groups that opposed prop 16 — Those groups you would think now, if they became aware that even though prop 16 was defeated that the UC system is really not obeying the law, you would think those groups would sponsor lawsuits or some kind of legal action to compel adherence to the law.

Richard Sander: Yup. Yeah. And there are certainly, I think there are certainly discussions of that kind going on. But it's, you know, it's difficult. It's not easy to mount a challenge to university practices, especially when you're going against something like the University of California, which has for all practical purposes unlimited resources.And lots of very, very smart defenders. So, finding, you know, just think about the task of finding a plaintiff, finding a student who's willing to take on that kind of public role. Probably going to be marked for life by taking on that role. It's not easy to bring a suit.

Steve Hsu: Yeah. I think you're right.

And for people who know the background of the current SCOTUS case, that has been an effort by a small group of people for, I would say maybe almost two decades now. So maybe we could turn to that. So the Supreme Court has granted, I think the legal term is granted cert. Is that correct?

Richard Sander: Yeah. Right.

Steve Hsu: Yes. So they will review the I guess appellate judgment, which favored Harvard. And this is a case which alleges that Harvard is discriminating against Asian-American students in admissions, but it may have the consequence of changing in a fundamental way how affirmative action itself is conducted in the United States.

Richard Sander: That's right. And the Supreme Court has also granted cert to a companion case for the University of North Carolina. And probably delayed granting cert in the Harvard case as long as it did, about a year, because it wanted to see what happened in the UNC case.

So both of these cases were filed the same day by a group called Students for Fair Admissions. This group was organized mainly by Edward Blum who, as you mentioned, is the leader of a group that has, you know, sort of been like a conservative version of a liberal civil rights group. Trying to bring conservative challenges to civil rights law. He had previously spearheaded the Fisher case, a suit against the University of Texas that was initiated I think in 2008 and went to the Supreme court in 2012, and again, in 2015-16.

In the Fisher case, the focus of the lawsuit was very, was sort of technical. The case was essentially arguing that the University of Texas had during a period of race neutrality achieved a lot of racial diversity and that therefore, since it had shown that it could be racially diverse without using preferences, it couldn't use preferences.

So it was basically making an argument, a carefully framed argument, about what current constitutional law permitted.

And in Fisher 1 in 2013, Justice Kennedy said on behalf of the majority of the court that, yeah, this does seem a little problematic. Here's the test. We're going to remand it. And we prefer their proceedings to kind of see whether or not these tests are satisfied. So it went back to the circuit court, which could have remanded it to the district court for a new trial. But the appellate court basically said, no, we think it is satisfied so here's some additional material for you Supreme Court. And it went back up and Justice Kennedy said oh okay. So Kennedy's second opinion was much more sympathetic to the University of Texas than his first and it kind of surprised people. But Fisher ended up not really changing the law at all.

The Harvard and UNC cases as I say were both filed in 2014 and they were both designed to be different kinds of cases. There were both empirical cases that instead of sort of saying, well, there's some principle that's being violated here. They wanted to show exactly how the admissions process worked at both schools and what the size of the preferences were. The special focus in Harvard was to show that in addition to affirmative action for Blacks and Hispanics, there was a negative affirmative action for Asian-Americans.

And so that really made that case stand out. The UNC case was distinctive because UNC had previously made some noise about, race-neutral alternatives that it was considering. And it had produced some documentation that seemed to indicate that the race neutral alternatives would work pretty well. But then decided to stick with racial preferences.

So both cases were empirical and they were both sort of attacking a new aspect of perceived vulnerability in what the schools were doing.

Steve Hsu: Do you have any predictions for what will happen in the Supreme Court?

Richard Sander: Well, all the lower court decisions were very pro university. This was partly because the luck of the draw for which judges would hear it at the district and circuit level produced judges who were pretty strong supporters of affirmative action. So I'd be really shocked if the Supreme Court left these decisions alone because the decisions themselves really basically say universities can anything.

I mean if if the Supreme court affirmed these cases there would really be no limits on the degree of race consciousness that universities could follow. So I think the court is going to reverse those decisions. What's very hard to know is whether they're going to do a big reversal and sort of greatly restrict the scope of affirmative action. Maybe get rid of affirmative action, maybe get rid of racial preferences per se, in higher education altogether.

Or will they do something narrower that just kind of says, well, yeah, Harvard, you really are using this personal rating of Asian-Americans in a discriminatory way. And you UNC really have ignored this earlier report you did on race-neutral admissions. So go back and, and pay proper attention to that. You know the court has a history of making narrow rulings in this area. So they might do that again. Or they may say, this is the moment we should just strike it all down.

So this will probably go to trial. Is it in the fall of 2022? Is that correct?

Richard Sander: Yes. They'll probably have a hearing. They've scheduled. I think they've indicated they're only going to have an hour of oral argument for both cases, which is little bit surprising and those who specialize in reading the tea leaves think that that is a sign that the court's going to make a broad decision.

In other words, the fact that it's taking these two very big factually intensive cases, and it's going to say we're only going to have 60 minutes of oral discussion of both cases, may be a sign that they're planning on doing a pretty broad decision rather than one that gets down into the weeds.

Steve Hsu: So I don't know very much about the Supreme Court, but is it possible that the conservative justices and their clerks have already discussed what do since, since, as you say that there's already a tremendous amount of documentation and publicity around these cases.

Richard Sander: Yeah. Yeah. I mean, the court as a whole has discussed the cases in its hearings on whether to grant cert.

Steve Hsu: I see.

Richard Sander: And they've clearly had multiple discussions of it. I don't know but I wouldn't be surprised if many of the conservatives have also caucused on know how the cases should be handled.

In any case, they'll hear it probably in October or possibly November and then they'll probably wait until the following spring to issue an opinion.

Steve Hsu: Wow. It could be a sea change.

Richard Sander: Yeah, yeah, it could. And again, the tea leaf readers think that the timing of cert was such, it was if you believe this, it was timed exactly so that it sort of pushed the decision into the next calendar year. If they had decided to grant cert a few weeks earlier then it would have been normal to have oral argument in the spring and decide in June. But the court has already deciding pretty hot-button cases and abortion and gun rights.

And so the argument goes that they want to put this off until next year when the political climate will be cooler. And, you know there won't be this triple whammy in June of this year.

Steve Hsu: Amazing. You know, I was following the Harvard case in Boston, fairly closely. And reading, you know, both the analysis of David Card for the university and Peter Arcidiacono for the plaintiffs. And I was just outraged at the decision by, I think Judge Burrows, which she seemed to, as you said, basically endorsed what seemed to be soft quotas for the purposes of diversity. And I thought, surely this can't stand.

Richard Sander: Yeah she she managed to ignore all sorts of things. She and the circuit courts also equated diversity with only racial diversity. In other words, the plaintiffs produced a distinguished expert Richard Kahlenberg who showed that the school could move to erase neutral admissions with a very small reduction in racial diversity and a big increase in socioeconomic diversity, which sounds like a no brainer. Right?

And she basically said, well, increasing socioeconomic diversity is irrelevant. The fact that the Black percentage will go down at all intrudes on the university's compelling interest in having as much racial diversity as they want. So that's what I mean, the lower court opinions are very aggressive and are basically completely deferential to the universities.

So if these opinions had been affirmed, it'd be foolish for anyone else to legally challenge anything that's going on in this area. So it was a real make-or-break event.

Steve Hsu: Wow. You know, you mentioned diversity on campus other than by along racial lines. I myself think there's less and less ideological diversity on campus. And, you know, maybe I'm showing my age, but you know, I'm a guy who supported Obama and Bill Clinton and almost, I think overwhelmingly if you look at my voting record, it's almost overwhelmingly Democrat. But I feel like I'm on the far right now as a university faculty member. It's very strange and I don't think my political views have changed. Now, maybe I'm just a dinosaur. But it seems to me the campus has shifted drastically to the left.

Richard Sander: Yeah, I feel exactly the same way.

Steve Hsu: Yeah, So I maybe you can speak a little bit more about where you think things are headed in law schools or more broadly on UC campuses.

Richard Sander: Well it's it's it's frightening in so many ways. There's for one thing among those conservatives who do exist on campuses, there's a lot of fear among them to speak out.

Cancel culture is very real and you know that the two conservatives who are on my law faculty have both gone through processes that were humiliating. The dean essentially wrote notes to the whole campus community denouncing these two for things that they had said that were not I think were taken out of context and were well within the bounds of academic freedom.

So it's an extremely hostile environment for conservatives. It's an environment where moderates are encouraged to stay quiet. It's an environment where some fairly left-wing ideas are discussed without being subject to challenge. There's just not robust debate about a whole range of issues these days. So it's very disturbing and I find a lot of students I mean, a lot of students seek me out basically to raise their concerns about this.

Law school, you know of all places Steve on campus, legal education is the one that's most devoted to the art of helping people rigorously understand different points of view You can't be a competent lawyer if you don't know how to anticipate what the other side's going to argue and how to really understand those arguments And that skill is not being taught nearly as robustly as it used to be because so much of classroom discussion of law schools is now following a script. And people are afraid to speak.

Steve Hsu: Yeah. I mean, I think if the current system just incentivizes you to keep your head down, if you're not — forgive the terminology — but if you're not sufficiently woke, you know, in your narrow self-interest, forget about educating the students. You should just keep your head down and avoid trouble.

Richard Sander: Yup. Yup. So an issue that I think ought to be before the court in its review of Harvard and UNC is this doctrine they've had in the past of sort of having a carve out for educational institutions. The court's been more willing to have universities use racial preferences than say corporations or newspapers because they see universities as kind of having this special cachet of objectivity. And it's really just the opposite of reality. You know, universities are completely incapable, I think, of analyzing and assessing affirmative action policies.

Steve Hsu: Yeah, I have to agree with you and having had some experience as a senior-level administrator, I think if you look at the incentives for an individual administrator at any major university, it is not to try to get to the truth. Most of these people, mainly this may sound extremely cynical, but mainly they're just to advance their own careers. They may have some relatively deeply seated ideological commitments. I would say overwhelmingly very far left of center. I mean, it's just statistically, speaking statistically.

But more than that, I think it's just advance your career, say you're a dean and you want to become a provost or you're a provost and you want to become president, the best thing to do is just not rock the boat. And I've met very few senior administrators who really want to get to the bottom of things to really understand, hey, is Rick Sander right about what he says? I mean, am I damaging these students by instituting this admissions scheme?

Very few people in these high positions actually even care about what the actual answer is because because they know what would be the best thing to do for their own career.

Richard Sander: Yeah. I'll call out an exception to this. UCLA. UCLA's sort of best known chancellor was was Charles Young who led the university for about 30 years. When he retired, I think it was '97, we hired the provost of Harvard [Albert] Carnesale who was our president, I'm sorry our chancellor, for nine years. And I didn't really appreciate this at the time, but about two years into his tenure, he held a dinner at his house for about a dozen social scientists. And the topic was does affirmative action work. And he was basically telling us that he didn't think there was enough inquiry into this. There was a lot of debate about it but it was too ideological and he hoped that people would sort of try to look at it.

And he didn't say anything explicit. He didn't hint at anything like mismatch or that purposes were too mechanical. And of course, UCLA at the time was during honeymoon of race neutrality. But in retrospect, it's clear from the way that he responded to my work, that he was one of those rare exceptions who actually wanted to find out what was going on. You know, he didn't he didn't take me under his wing. He didn't sort of say Rick here's a $100,000 to help prove this. But, you know, but he did try to protect me in various ways.

Steve Hsu: You know, when you're at certain level, you know, senior administrator at a university, there, there are annual gatherings that occur which you might be rubbing elbows with your counterparts at, you know, the other Big 10 schools or the other AAU schools. Vice presidents mixed with provosts mixed with presidents. So I've spent a lot of time with these people. And I would say, you know, maybe one in five has a strong commitment to understanding empirical realities, which might be uncomfortable. But that's not really a large enough number to make a difference, honestly. And so just, I feel like well I guess it contributes to my own pessimism for what's going to happen to these institutions in the future.

We may end up, if we do end up, but this, this is not a good thing, but if we do end up in a cold war with China, which practices an especially ruthless kind of meritocracy in their educational system, then we may have to jettison some of these crazy things that we're doing. But there's a lot of ruin in this institution of higher ed, if you ask me.

Richard Sander: Yeah. So your view is that there's not sort of any internal social process that's going to cause the pendulum to swing the other way? That we're just going to continue to go down this road of increased nuttiness?

Steve Hsu: I think so. I mean, for many reasons, so one is that, for example, you mentioned these diversity statements that I think now on UC campuses to even be considered for an assistant professorship, you have to write a full-throated statement, which, you know, they don't say this, but effectively, if you don't support the current system in that statement, probably you're going to be filtered out. And I think even for promotion, someone was telling me then that in the UC step system for promotion, you need to write such a statement.

Richard Sander: Yup.

Steve Hsu: And so this is just going to lead to a more and more polarized left-leaning faculty. So from the faculty side, I don't see forces for change. And among administrators, definitely not because, you know, the political forces that be, in order to keep the board of trustees happy, keep the governor happy, generally you have to, you don't want to rock the boat. You can't question the standard narrative.

Richard Sander: So what's your view of what someone like me should be doing? What we're already doing?

Steve Hsu: Well, I think you're a courageous person. I think probably you're going to be punished for doing the right thing. It's just a sad fact. We don't get to choose the times that we live in. And the pursuit of uncomfortable truth sometimes, right? Gets you punished.

Richard Sander: Yeah. Yeah. Well, in my case there is a limited amount of punishment that can be meted out. I mean I, you know, the school gave me my chair. I am at retirement age. I don't really have administrative positions. So there are a limited number of weapons the school has against me.

But I do spend a lot of time thinking about what constructive things could be done. I mean, there's, Heterodox Academy, for example, has tried to start conversations about these problems. A lot of my friends who are concerned about wokeness feel that Heterodox Academy is not doing anything assertive enough to make it really make a difference. But it'd be nice to know if there was some process or some type of research that could be useful.

Steve Hsu: Well, let me say this. I, I fully support what Heterodox Academy is doing. And I was involved in the creation of something called the AFA, Academic Freedom Alliance, which protects free speech rights for faculty members. And I think all of these things are important, honestly, rearguard actions to keep things from getting worse.

I think that one project that I did work on, but hasn't come anywhere close to fruition is something that can help professors express their true preferences in a way that they're not risking their careers. So imagine somebody and you could use all kinds of fancy what's called blockchain or cryptographic technology for this. But the basic idea is that imagine you could validate someone's identity as a tenured [unclear] professor at UCLA. And you could reliably then get their true opinion, not their, you know, hidden you know, concealed. I mean, what's the word when people conceal their true preferences.

Preference falsification.

Yeah. So, so I mean, so even though I think generally faculties are shifting further and further to the left, there's a very big silent majority, I think, of faculty who think it's gone too far. Let me put it that way. But they're afraid to say anything because they don't want to be attacked by the very extreme activists that, you know, are basically patrolling what's said.

And so if you could develop a technological mechanism that allowed people to express their true preferences and in such a way that you could verify that this particular response to the survey came from an actual UCLA professor without revealing who that person actually is, that very simple tool would help things. I think it would help show that, you know, maybe 60% of all faculty, even at UCLA, think things have gone too far in the woke direction. Think that some level of free speech should be reestablished on the campus.

And, you know, I did work for a while trying to raise funds from various benefactors to build a system like this that would allow people to register, identify them, validate themselves as a, you know, faculty member and then, you know, get a cryptographic key that would allow them to respond to surveys and express their opinion pseudonymously.

So I don't know. I mean, maybe something like that could work. I do feel there is a silent majority that doesn't like where things are today. But I don't see an easy way for that silent majority to make its feelings known.

Richard Sander: Yeah, that's a fascinating point.

Steve Hsu: Yeah. Well, we've gone on for quite a while and I've really enjoyed it. I think we've covered all the topics that we set out to cover. Do you have any final remarks you want to make before we sign off? It's been a pleasure.

Richard Sander: No, just express my admiration, Steve, for everything you stood for, and stood up for. And it's, it's great to know that there are voices like yours that are still speaking.

Steve Hsu: I'm going to keep at it. I enjoy this podcast and, you know, I hope to address, not every episode will address difficult issues or controversial issues, but I won't shy away from them, that's for sure.

Richard Sander: Well, thank you.

Steve Hsu: Thanks a lot for your time.

Richard Sander: Yup.