Intro: Welcome to Policy Chats. The official podcast of the School of Public Policy at the University of California Riverside. I’m your host, Kevin Karami. Join me and my classmates as we learn about potential policy solutions for today’s biggest societal challenges. 

Kevin Karami: With me today is Dr. Mark Long, the new dean of the University of California Riverside’s School of Public Policy. He was previously the Associate Dean for Faculty Affairs and Research at the Evan’s School of Public Policy and Governance at the University of Washington. His area of focus includes education policy, educational attainment, and the effects of affirmative action policies. Dean Long, it’s an honor to have the chance to speak with you today. 

Dean Mark Long: Thank you. It is a pleasure to be here and I am enjoying my first couple of weeks here at UCR. 

Kevin Karami: That’s fantastic! So, affirmative action: a pressing issue, an issue that has been talked about in American politics for a very long time. Before we get into the weeds of it, I was hoping you could outline some of the history of affirmative action. What are the roots of it? You know, what role have colleges played in affirmative action policy? Can you maybe outline some of the major points in history and how we kind of ended up where we are today?

Dean Mark Long: Yeah, so affirmative action as we know it started coming out in the 1960s from the Kennedy Administrations, from the Johnson Administrations. And you start to see colleges start to take up that cause coming out of the Civil Rights Movement to see themselves as having a mission to serve the public, and particularly, responding to the effects of slavery, the effects of segregation, and how that that was still influenced particularly black Americans lives, but then over time also consideration for Hispanics and Native Americans came, and the injustices done over centuries to those groups started to percolate into colleges as a part of their mission of social justice, part of their mission to try to undo any kinds of contemporary discrimination, prior discrimination, that was having contemporary effects, so what you see is, and one of the things I’ve done in my research, is that I’ve looked at the extent of affirmative action as of 1972, 1982, 1992, with three sort of cohorts of data from the US Department of Education. And what you see by that point is that there were affirmative action in admissions in 1972 in the Northeast, the Midwest, and the West, not in the South. As of 1972 in the South, if you were a black, hispanic, or Native American student, if you were relative to a comparable white student, you were less likely to be admitted to the selective universities, so you can see active levels of discrimination in the South as of 1972. But by 1982, all the regions of the United States were practicing some degree of affirmative action, positive affirmative action, increasing the chance of black, hispanic, and Native American students, henceforth I’ll just describe that as underrepresented students, to get admitted. You see about the same levels of affirmative action in 1992, but in the 1990s, we saw a number of states ban affirmative action. California was one of those. Started with the UC regents but then was also passed by a voter initiative to ban affirmative action. You saw a similar effort in Washington state, which is where I just recently came from. There was an effort by then Governor Jet Bush in Florida and the Florida regents to drop their affirmative action. And then a number of states in the subsequent years. Texas had their affirmative action ban by a court decision, the Hopwood decision. And so there have been these decisions along the way that have affected the ability of states to do affirmative action. Subsequently, what we’ve seen is the states where there have been bans, we’ve seen them try to use alternative means for boosting their underrepresented minority composition of their admitted and enrolled students. I think we can talk about whether those alternative programs have been effective. 

Kevin Karami: Yeah, definitely. One interesting thing to me is the debate of affirmative action is so ongoing, and like you mentioned in your research, how you focus since 1972 and how it’s evolved, the bans that happened in the 90s. One thing I wanted to focus on as well, and think is important, and I think you touched on this, is the effect it has had on underrepresented populations. So, how critical is affirmative action in terms of underrepresented populations being able to be admitted into colleges? You know there are some that claim it's necessary, some claim the effect is less, or the importance is less. What are your thoughts on that?

Dean Mark Long: Yeah. So what you can look at is, if you look at cohort based sort of based measures of like for example test scores, there are large gaps in test scores between for example white and black youth. Those test scores differ, those gaps are there, are present, as of kindergarten. So a lot of this is clearly things that are going on in the homes, in the communities, of these students before they even hit high school. Those gaps, they don’t change much during K through 12, and so that then creates an issue for the colleges, which is if they want to have a class that is representative of their state, that is a particularly important issue for state public universities, they’re dealing with gaps in the test scores, they’re dealing with gaps in grades, they’re dealing with gaps in other measures of academic preparation, and so if they want to be representative, they have to try to do something to try to boost those admissions prospects. The way affirmative action has been traditionally conducted is to directly give some degree of bonus to students who are underrepresented. So in the case that percolated up to the Supreme Court, there were two cases that the University of Michigan related to the undergraduate admissions and law school admissions. These are the Grutter and Gratz cases, and what the University of Michigan was doing at the undergraduate admissions level is that for all black, hispanic, and native american students, they were giving a bump of twenty points in their admissions packet. Now what does twenty points mean? Twenty points is the same you would get if you increase your high school graduation grade point average from a 3.0 to a 4.0 or a 2.0 to a 3.0. So it was a pretty sizable advantage. Now they were also giving the same twenty point advantage to students who attended predominantly underrepresented, high schools that were composed of underrepresented minorities students. So if you were a white student attending a high school in parts of Detroit where most of the students are black, you would get the same twenty points. They also gave twenty points if you were from an underrepresented Michigan county. That tended to be the upper peninsula, that’s predominantly white. And there were also a twenty points for Provost selected candidates, so we don’t know what that was. But my bottom line point is that this was a substantial advantage given to underrepresented students. In the data that I‘ve looked at nationally, if you’re applying to the most selective colleges, it would increase, if you’re black, hispanic, native american, it would increase your chance of being admitted by 30 percentage points. So that’s a sizable increase. So in these states, to give actually more specific cases and so more data used from Texas, University of Texas and Texas A&M, your chances of being admitted increased by 15 percentage points. And so the substantial chance of you being admitted over an observably similar white student.  

Commercial: Can we protect the environment, grow jobs, promote racial equity, and enhance community well-being? Join us on January 24th for a fireside chat on mission driven consulting with Cecilia Estolano, CEO of Estolano Advisors, as we discuss promising trends towards environmental sustainability and racial equity in Southern California. This event is brought to you by the UCR School of Public Policy and the Center for Social Innovation. Register now by going to spp.ucr.edu. You can also find the link in the show notes. 

Kevin Karami: That’s really interesting. One thing you were talking about, so in terms of it going back to kindergarten or pre k the disadvantages that certain populations face like compared to white students and children, if you’re, if you’re making that policy in terms of trying to find out what that advantage is and actually trying to put that to terms in the law, so you know you mentioned the twenty point bonus that was a substantial increase that was given. So, how does that process look like when it comes to saying there’s this disadvantage that exists, so in order to bring these impacted populations to par with white students, how does that policy come to play? Because I think one part of the debate on affirmative action that I've also seen isn’t necessarily that there are those who are directly opposed, and then there are those who are concerned about how effective it is, some wanting more bonuses given, others wanting less, and not necessarily being completely opposed to the idea in general. So what are your thoughts on the policy making process in terms how those advantages given out? For example the GPA boost that you explained, the twenty point, there maybe some that may say that hey, that’s too much of a boost, some may say that it’s still not enough, so how does that in the policy making process, how does that work, from a data perspective, and also and actual policy coming into fruition?

Dean Mark Long: Yeah, so I think there are two aspects to that question that should be addressed. One is if you look at the extent of representation given affirmative action as at least as practiced by universities, the colleges were still vastly underrepresented, the most selective campuses. Like if you look at UC Berkeley for example, it does not match the composition of the state. In terms of their black, hispanic, and native american students. That is, that was true in the early 90s, before affirmative action was banned across the UC campuses. So, if the goal of that process was to create representation, you could argue that they were not doing enough. The other thing I think is important to think about here is that, you know when I was talking about those test score gaps that are emerging before kids get into kindergarten, you might ask the question well why is that? So if you look at the gaps in all kinds of different kinds of measures, whether that be income, wealth, imprisonment, having a job, having two parents, a lot of these differences are there to begin with. And so some say well let’s actually use affirmative action based on race, let’s directly give advantage to students who are poor, to students whose parents don’t own a home, who have different, less wealth. So, some people say look well we don’t need affirmative action based on race, we need affirmative action based on income, or socioeconomic status. What you can show there is if you try to, if your goal is representation by race, but you use other indicators like income or wealth, you need to put as much as four times as much weight on those other factors that you would if you just put weight on race. So, one of the questions there then becomes, from an efficiency perspective, if what you’re interested in is generating racial representation, it’s inefficient to put weight on socioeconomic status. It will get you there, but you have to put much more emphasis on it to get to the same point. On the other hand, some people say look it’s improper, immoral for the state to take into account a person’s race, where it’s not immoral, or injust, or inappropriate, to be discriminatory based on income, let's say. So, okay if even if it’s inefficient, we feel more comfortable putting weight on race. I think that’s a reasonable argument. I know I’m kind of going on at length for this question, but one more thing I have to say about this. In the 1978 Supreme Court decision, Bache v. University of California, in the subsequent Grutter and Gratz cases that came to the University of Michigan, what the courts have held is that states have to try all race neutral alternatives. All workable, race neutral alternatives. And there's an interesting tension in that phrase “all workable race neutral alternative”, so what the court is saying is that you have to try all things, like putting weight on socioeconomic status as long as that’s workable. So what does workable mean? Well workable basically means, so long as it does not have too much of a negative effect on the operations of the campus. So you can make an argument, or a university can make an argument, look we tried all these other ways, and particularly if we’re going to put an emphasis on income, we would have to put so much emphasis on income that we would draw in a class of students that aren’t academically capable of succeeding in the classroom, so it’s not workable to go that far. So, I think that’s where the inherent tension is right now in the Supreme Court’s decisions and how colleges are responding to it. 

Kevin Karami: That’s really interesting. In that case it’s especially interesting. One thing I wanted to ask really briefly. You mentioned income and wealth as indicators that could work if you put much more emphasis on them. Are there any other indicators besides race that are worth talking about in this context or are those two the other main ones that people kind of discuss? 

Dean Mark Long: That’s a really interesting question. So, one way of thinking about this question, and one way I’ve thought about this question, is if ultimately what you’re trying to do is put weight on things that are proxies for race, that are predictive of race, what you could do is take all the factors, all the data, that the colleges collect, take that data to predict the likelihood that an applicant is an underrepresented minority. Then put weight on that predicted likelihood. So, what I was talking about you’d have to put four times as much weight on that, that’s exactly what I was talking about. And the paper that I did on this was published in the Journal of Policy Analysis and Management, and it was trying to assess this question, is it workable, is there a race neutral that is workable? When I presented this paper, though, to people who were lawyers, particularly lawyers who are sort of on the pro affirmative action side of the debate, they got very nervous and concerned with the use of the term proxy. Because in other Supreme Court decisions, what they’ve said is that, particularly in employment law, that it's not legally permissible to put weight on a decision that is a proxy for race in an effort to be discriminatory. In other words, let’s say I have a test that I give and that test has no connection to a person’s job performance or ability to perform in the job, but I think there is going to be a difference in outcomes based on race. So then I use that test in a backhanded way to discriminate. The courts have said you can’t do that. You can’t use proxy indicators for race to try to be discriminatory. So in the same way they get concerned that universities would use a proxy system for race, and so that what the kind of conditions that then universities have an incentive to not use describe what they're doing with affirmative action for socioeconomic status, they have to be careful to say that we're not trying to do that for a proxy for race, we’re doing it because we care about low income students directly. And in fact, it may be true that the universities, I’m sure it’s true, and I know it’s true at UCR, that they want to help low income students. But there’s this other motive and I think it’s hard to disentangle this, that universities recognize that helping low income students is more likely to also help the racial representation. And so the lawyers would be cautious about saying that it’s a proxy because of this legal history. But as an academic, I can kind of imagine, imagine if you tried to take that proxy idea to its extreme what would you do? And ultimately what that would mean is taking all the factors that you possibly can obtain and predicting the likelihood that a student is black, hispanic, or native american. One more thing to say about that, I did that same kind of exercise with a really detailed, rich data set that had hundreds and hundreds of variables and essentially what I was trying to say was, how well can you predict someone’s race? And I found that you can do pretty well, but only if you have data that universities are unlikely to have and also data that would be hard to collect in this context. So for example, one thing I found was that, the single predictor that was most likely to predict if you were black, hispanic, or native american, is if your best friend was black, hispanic, or native american. And the second best predictor was whether your second best friend was black, hispanic, or native american. The third best was whether your home language was Spanish. So that gives you an idea that okay home language is Spanish is something they probably could collect on an application form, but you couldn’t collect that information on what’s the race of your best friend. Right? It would be so obviously transparent how you were using it and it would lead to students on their applications not being truthful. So, yes you can predict race, but you can’t predict race perfectly particularly with the kinds of data that universities have at hand. 

Commercial: Social injustice disparities, climate change. Are you interested in solving pressing challenges like these currently facing our region and the world? Then consider joining the next cohort of future policy leaders like me by applying for the UCR Master of Public Policy program. Learn more at mpp.ucr.edu. You can also find the link in our show notes. 

Kevin Karami: That’s really interesting to talk about, like what is predictive of race. I mean theoretically if somehow there was an indicator that was super, you know you could super easily collect the data on, universities can and would also predict race really accurately, but could also similar to how socioeconomic status is being used now, you could explain it in other terms, it could be a different angle, but it sounds like it is pretty difficult to come across something that both- the universities could collect and also something that you can get away with in terms of saying hey you know, we actually care about this group of people. But in reality there might be other motives, like you were talking about. It’s really interesting because I think the indicators and predictability of race based on other factors or as you mentioned proxies is a really fascinating topic on its own. Getting back to affirmative action specifically, you mentioned the University of California v. Bache decision, can you kind of talk about the impact of that case? What was the case about and also how its effects are still being felt today? 

Dean Mark Long: Yeah, so that case was specifically about UC Davis’ medical school, and the way UC Davis’ medical school, the decision was in 1978, the Supreme Court decision, so the way the medical school in the 70s was doing their admissions is essentially that they had a fixed number of slots and they were reserving a fixed number of slots for white applicants and a fixed number of slots for non-white applicants. And, so what that meant, and then they were completely doing the admission of those two groups separately. The court ruled in a split decision of a 4-1-4 decision, which we’ll come back to. They ruled that it was unconstitutional to have separate admissions systems, so that you had to have a single admission system. But more importantly that 4-1-4, the way the 4-1-4 decision worked is that there were four justices that were in favor of the constitutional permissibility of using race in admissions, there were four justices that were opposed to using race in admissions, and the one justice, Powell wrote the deciding decision, he was in favor of affirmative action in part and against in it part. So in terms of being in favor in part, he said that the universities could continue to use affirmative action if one of two things were true: one, if they had their own recent history of clearly discriminatory practices. So I mentioned in the 1970s that I could show- that in 1972 you could see that some of the Southern colleges were actively discriminating. If you could show that there was an active history of active discrimination, then you were legally permitted under the Powell decision to use race based affirmative action to undo your own adverse actions. But Powell said you can’t use social justice as an argument for your affirmative action. So in other words, universities can’t use as a rationale for their admissions that they're trying to benefit society and undo any contemporary effects of current discrimination by citizens, or past discrimination, or sort of segregation and Jim Crow, and these kinds of actions. They can’t undo bad actions that society has done, however, they can use diversity as a rationale. So what Powell had in mind is that, he says look, there are benefits to having a diverse class of students and those benefits flow to all students including white students, so if a university can say we are using race based affirmative action because we want to generate a diverse class of students, that can be a permissible use of affirmative action. And so, that is 1978, in the subsequent years, then you see diversity as the rationale that universities do use. So you know, we have lots of DEI committees, where did that D part from? I would say arguably it was, maybe not came from, but definitely enhanced by Powell’s decision because now that is the language that universities can use to justify their practices. So you don’t see universities using the language of- you know when you see them using the language of representation or underrepresentation or trying to fix underrepresentation directly, because it potentially gets challenged to the affirmative action system they are implementing.

Kevin Karami: That’s really interesting and the impact of the case still being talked about today. So as we near the end of the episode, I think one interesting topic, or maybe this is more open ended, is how much does all of this actually matter? In terms of you know there’s a lot of groups since the 70s and 60s and today as well that really care about affirmative action because there is a belief that hey it’s important for underrepresented populations to match white students in terms of college attendance because of the long term impacts, so you know does any of this actually, is that true? Does attendance in these selective colleges actually matter? Or maybe you know is it more semantics, is the impact not as big as some claim?

Dean Mark Long: Yeah, so one, so there’s sort of a hotly contested literature among labor economists about what is the effect of additional education, like a year of additional education, and the effect of going to a more selective college. And the challenges for that literature is that we don't have good ways of doing experiments to find out. We can’t experimentally assign you years of education and we don’t typically randomly assign students admissions to or not admission to universities to see the effects and trace out the subsequent effects it had on their lives. So we don’t investigate it the same way like a drug manufacturer might do with a randomized control trial. However, that said, I would say it’s sort of the best evidence that every additional year of education that one obtains raises your lifetime earnings by about 7% per year. Each additional year, and that’s cumulative, raises your earnings. Secondly, on the college admissions side, again it’s challenging to study, but I believe that the bulk of the evidence is strongly suggested that going to a more selective college raises your earnings. Now, I should say, you should be skeptical any time you look at what are the earnings of the average Princeton students and compare it to what are the average earnings of a student at Southwest Missouri State. Why? Because those students have different inputs, they have different test scores, they have different parental characteristics, they have different networks that are coming from their parents, and the wealth of their parents. So that is what makes it so challenging, nonetheless, the best empirical studies show that there is a substantial increase in your earnings going to a more selective college. And one way you can study that is you can look at in schools that have a system where if you, where they score the applicants and they admit students based on that score, you can look at the jump in earnings that that threshold of being admitted versus not being admitted. In some studies, looking that can show admissions that selective flagship institutions may raise your lifetime earnings by 20%, so it does matter. And finally coming back to years of education, attendance at more selective colleges also increases your likelihood of graduating from that college, so there's much higher graduation rates at the more selective colleges than the less selective colleges. So there's pretty strong evidence that having a peer group that is more likely to graduate, also having a peer group that themselves have better networks from their family and friends are going to give students an advantage. So yeah, it matters. 

Kevin Karami: Yeah, definitely. And I like how you mentioned in the beginning how it’s difficult to experiment on this especially because there can be so many extraneous factors and variables, like a student that attends an Ivy League probably had, most likely had a different upbringing versus someone who attended like you said Southwest Missouri State, and that could have an affect on their lifetime earnings, so it’s not just the university that they attend, but also like you mentioned the university they attend also does have an impact, it’s just kind of a jumble of variables and factors making is really difficult to figure out. But I think it’s ultimately really important to recognize that even though those other variables play a role, you know, college attendance does in fact have a positive impact in general on your lifetime earnings. Dean Long, that’s all the time we have today to speak about affirmative action. I feel like we could have this conversation for two more hours and we still wouldn’t get to everything we can. It was an honor to have you on the podcast today, and I'm looking forward to your deanship at the SPP.

Dean Mark Long: Thank you, thank you. The honor was mine, the honor was shared. And I appreciate the great questions you asked me. Thank you.

Conclusion: This podcast is a production of the UC Riverside School of Public Policy. Our music is produced by C Code. I’m Kevin Karami. ‘Till next time.