urban institute nonprofit social and economic policy research

Affirmative Action: Is It Still Needed?

Document date: April 29, 2003
Released online: April 29, 2003

ROBERT REISCHAUER, Urban Institute: Let me welcome you all to this forum, which is one in a series of occasional forums, which we cosponsor with the Georgetown Public Policy Institute, where people with interesting ideas and things to say have a chance to interact with others on the same topic and stimulate debate and thought among the audience.

Today, we have a tremendous panel to discuss a very serious and interesting issue, namely affirmative action and whether it's still needed. We will have this session moderated by Clarence Page, who is no stranger, I'm sure, to anybody who has a television. For the last 19 years, he's been a columnist with the Chicago Tribune, and in the course of his career, has won Pulitzer Prizes as well as considerable positive comments on the thoughtfulness of his column.

He's also the author of a book, Showing My Color: Impolite Essays on Race and Identity. Clarence will be the moderator. The speakers, in order of their presentations, will be Harry Holzer, who is a senior scholar here at the Urban Institute, but primarily a professor at the Georgetown Public Policy Institute.

There is an interesting characteristic about academics these days, and that is that they seem to have no loyalty at all to any institution. I note that in his brief biographical description, that he has current affiliations with Northwestern, Chicago, Harvard, and Wisconsin. So I'm wondering what exactly is left for Georgetown and the Urban Institute.

Among other things, Harry, in his past, was the chief economist at the U.S. Department of Labor and also had a long and very distinguished career in the economics department of Michigan State University.

Harry will be followed by Doug Besharov, who is a lawyer and a senior scholar at the American Enterprise Institute and a professor at the University of Maryland. Doug has been active in many public policy debates in this town over the last several decades and was the first director of the U.S. National Center on Child Abuse, and his most recent book is Recognizing Child Abuse: A Guide for the Concerned. But he has written over a dozen books and some 200-odd articles—not odd—provocative and interesting articles.

Doug will be followed by Chris Edley, who, for the past several decades, has been a professor at the Law School at Harvard, when he hasn't been down here serving in one administration or another. I spend time at Harvard and time down here, and I've never seen him at Harvard. So he carries on the Harry Holzer tradition of affiliation without presence.

Chris served in the Clinton administration, in the White House, and I believe at OMB, and also in the Clinton administration as well. He is the author of Not All Black and White: Affirmative Action, Race and American Values, and also has been active in many of these debates.

Last, but certainly not least, is Abigail Thernstrom, who is a senior scholar at the Manhattan Institute in New York, a member of the Massachusetts State Board of Education, and a member of the U.S. Commission on Civil Rights, which, also, Chris is a member of, I believe. She is the coauthor, with her husband, of a book that caused considerable debate and argument several years ago, America in Black and White: One Nation Indivisible, and is the author of a book that is just about to come out, I believe, in October—No Excuses: Closing the Racial Gap in Learning.

I welcome you all and look forward to hearing what you have to say, and turn this over to Clarence.

CLARENCE PAGE: Thank you very much. And I must say, right off the bat, that I am humbled by the presence of so many marvelous academic scholars and so many wonderful friends. I look forward to today's discussion. I've been asked to be the moderator, someone mistaking me for someone of moderate temperament. I haven't noticed where they got that idea.

I'm probably best known for appearing on the McLaughlin Group, where tempers are not known for being moderate. But my wife calls it—says McLaughlin Group has done for political discourse what mud wrestling has done for the Olympics.


But today, however, in these august surroundings, I expect us to actually advance the discussion and debate today. How about that? And we have a wonderful group to do it for us. Our format today will be precisely eight minutes, no more, no less, from each of our panelists. And if they should happen to go over that time, our orchestra will well up with our Oscar theme and drown them out, and we will move onto the next panelist.

And once we've been through the panel, I shall lead a discussion of provocative and stimulating questions put to each of them to explore a bit more in depth the points which they raised, and to try to focus in on some of the growing important differences that we can bring up here, and then turn it over to our audience, which I can tell is already in the front-leaning, "I can't wait to ask a question," position.

So without any further ado, I'm going to turn to Harry Holzer, who I know personally. I have been hopping around from campus to campus; the last time we ran into each other was at O'Hare, which is that great junction spot of all Americans, as we know. And so, Harry, please open up our discussion today on our topic, Affirmative Action: Is It Still Needed?

HARRY HOLZER: Thank you, Clarence. What I actually want to do, at the outset of this talk, is to summarize the research literature on affirmative action, as I have read it. David Newmark and I published a pretty extensive review of about 200 studies on affirmative action, in the Journal of Economic Literature, a couple of years ago. And what I'm going to try to do in eight minutes is to take those 200 studies and distill them into five points, plus my own little commentaries, and interpretations, and extensions here and there. So, since that's a bit of a tall order, why don't I get started on that right away?

Point number one. There continue to be fairly major barriers facing minorities and, to some extent, perhaps to a lesser extent, women, in terms of achieving educational success and labor market success. And here, there's decades of research making these points. But in many ways, the research has been getting stronger over time, and it's strongest with regard to African Americans.

Very simply, the research still shows that there is discrimination in labor markets against African Americans, discrimination in housing markets. The housing market discrimination is one of many causes of residential segregation that continues at very high levels in this country. And the research has been clear, in the past few years, of linking residential segregation to negative outcomes on almost every measure that we care about in terms of education and employment, and virtually everything else.

In addition, there's literature on stereotypes and stigmas, and low expectations, all of which add up to a pretty strong picture that opportunity is not yet equal for African Americans, in many realms. Now, if you look at other minority groups like Latinos, you see some of the same story, but not as severe. The degree of residential segregation is not as high, the stereotypes are not as severe.

But yet, educational attainment for Latinos lags behind that of both whites and blacks in America by a considerable amount. Of course, that has a lot to do with the continuing inflow of immigrants that continues to keep those educational levels lower.

Now, when it comes to women relative to men, within each racial group, you get a somewhat different picture. There is a gender gap in higher education today. It goes in the opposite direction from what we had traditionally known. In other words, women now outnumber men within every ethnic group and racial group at major universities. Having said that, there are still disciplines and sectors of the economy and sectors of the labor market where women face barriers or remain underrepresented.

And separately from the whole idea of discriminatory barriers, there remains the very major challenge to women of balancing career with family and with maternity. That remains, perhaps, the greatest challenge facing women in the labor market today.

So when you put all of this together, it seems to me that there continues to be a need for policies to lower barriers and equalize opportunity within higher education and also within the labor market for these groups. And the real question is exactly what should those policies be, and how should they be implemented?

Point number two. We know that affirmative action is about winners and losers, and we know the whole point of affirmative action is to redistribute the goodies, in universities and labor markets, away from white males, towards minorities and females. And it does that, and the data are pretty clear on that.

What is a little surprising is that the size of that redistribution is often smaller than many people think it is. For instance, if you look at labor markets—compare the federal contractor sector covered by affirmative action to the noncontractor sector—what affirmative action seems to do is it redistributes a couple of percentage points of employment away from whites towards minorities and a couple of percentage points of employment away from white males towards white females.

But the effects are not dramatic in terms of their size. If you look at universities overall, you actually get a similar picture. The best recent studies suggest affirmative action raises the share of university admission slots for minorities by one or two percentage points, hardly the dramatic effects that many people suggest.

Now, an important caveat to that. As you go up the ladder of selectivity at universities in terms of admission, the effects of affirmative action get larger. So for instance, by the time you get to the top quintile of schools, say the top 300 to 400 schools, you're talking about maybe three or four percentage points being redistributed to minorities in terms of slots.

And by the time you get to the most elite 50 or 75 schools—and I would put the University of Michigan in that group—the effects are larger still—six, eight, maybe even 10 percentage points of slots redistributed. And whether that's large or not so large, everyone can decide for themselves. But those are the exceptions. Those are really the top, the most elite undergraduate programs and the better graduate programs.

Now, these numbers then raise, I think, two important questions. Number one: Why are the effects so small? And number two: If, in fact, they are small, why does this program generate so much anger and so much controversy? Well, the reason why the effects are so small in the university world, quite frankly, is that most American colleges and universities aren't that selective in terms of grades and SAT scores.

And even the ones that are relatively selective usually look at a fairly wide range of criteria for deciding whom to admit. So even at the best schools, an individual's low test scores can be admitted—of course, and we all know this—if you're an athlete, if you're the child of an alum, if you come from those sections of the country's southern and western parts where average test scores are lower than elsewhere.

An important corollary of that is if you look at the test score gap—and we know there are large test score gaps between minorities and other groups—those test score gaps are not entirely caused by affirmative action. Given the nature of admissions policies, and admissions policies that don't put that much weight on test scores, at least some of those gaps would exist even in the absence of affirmative action, because the admissions policies would just reproduce, inside schools, the test score gaps that already exist on the outside.

Now, the second question—why do these policies generate so much anger—and I don't think there's any good research evidence, certainly not within my discipline on that—but I just offer one anecdote. I think there's a systematic tendency for the effects of affirmative action to be overestimated by the groups left out of those programs, i.e., white males. And I've had anecdotes and observations like this, and many of you have as well.

Suppose there's a situation where 10 white males apply for a job, for a university slot, and that position goes to a minority or a female with somewhat lower credentials, along the usual dimensions. All 10 of those white males usually feel aggrieved, and all 10 of those white males feel as though they have been victimized by this policy, even though in reality, only one of them would have gotten that slot in the absence of affirmative action.

That's a way in which I believe that perception—you know, it's like the phenomenon of driving around a parking garage and being upset when you see the handicapp[ed] spots: You knew you'd get that slot if it wasn't for handicappers. But, of course, that couldn't be true for all of us.

Okay, point number three. The beneficiaries of affirmative action extend beyond just the women and minorities admitted to those programs. There's been a lot written about role model effects and mentoring effects. And quite frankly, my reading of the literature on that is uneven. There's some evidence of those effects—not as consistent, maybe, as many of the applicants would like to believe.

But on other dimensions, there is strong evidence, for instance, on service positions in minority and low-income communities. There's been a variety of studies, for instance, of doctors, minority doctors admitted under affirmative action programs at medical schools that seem to show those doctors consistently are more likely to serve minority patients and serve low-income patients, accept Medicaid, located in inner-city and rural areas.

So there's a public-good aspect to affirmative action that goes beyond the individuals who gain admission. Businesses, in the business community—there's evidence that businesses that have minority managers get more applicants from those communities, and maybe are better able to serve customers in those communities, and the business community clearly seems to value that.

And, of course, universities—almost every major university administrator in the country seems to argue that diversity is good for education. Now, that's very hard to document in terms of the research, because you're dealing with things like values and perceptions and how those get formed, and it's very hard to prove that in an academic study. And yet, almost universally, university administrators believe that, and maybe there's something to that argument.

Two more quick points, and then I'll try to wrap up. Point number four. If you look at the labor market, it's clear that in contractor establishments, the credentials, on paper, of minorities look weaker than the credentials, on paper, of the white males who work in those places. And yet, it's much harder to find consistent evidence of underperformance in those jobs.

Now, another important caveat: It's hard to measure performance on the job. I mean, economists usually say, well, just look at the wage, and we know that in the real world, there may be a lot of gap between wages and performance. So there's a lot of studies to do this, kind of sector by sector. What David Newmark and I tried to do was to look at supervisory ratings of their employees, which, of course, is subjective; it has lots of problems.

Nevertheless, across all these studies and all these methodologies, it's hard to find clear evidence of weaker performance on the job by affirmative action candidates in the contractor sector. And when Newmark and I looked more closely at it, it looked like two things were going on. Number one, these companies are casting a wide net. They look at people whom they ordinarily wouldn't look at, dig for other personal attributes that may predict success.

So they're casting a net more extensively and more intensively in generating good people, which, of course, ideally, is what affirmative action should be, not a lowering of the bar, but looking for other predictors and other determinants of good performance. And secondly, a lot of these companies do engage in remedial efforts—training, careful evaluation, etc. So in most cases, it seems to do the job.

Point number five. There's a lot of discussion these days about alternative policies, perhaps to replace affirmative action. The two most popular versions of those are, number one, having a class-based or income-based form of affirmative action, or number two, doing what they've done in California and Texas and Florida, going to a top 10 percent of your high school class policy, where the top 10 percent is guaranteed admission somewhere in the state system.

And a variety of studies, at this point, have shown that those top 10 percent policies, and even the class-based policies, may have good benefits, but they lead to a decline in the share of slots going to minorities at those universities. Now, I already argued before that the effect of affirmative action is not that large to begin with. So, by that logic, replacing affirmative action can't have that large a negative effect in the aggregate, and it doesn't.

And yet at the elite schools, the effects of affirmative action are larger, and it's at the most elite of those state universities that you do see noticeable declines when these policies were implemented.

So let me wrap up. What does all this suggest to me? I don't think this is a perfect policy. I think there should be lots of discussions about how to change affirmative action, implement it in different ways, adjust certain things. Overall, it doesn't look like a disaster in terms of outcome. The policy does tend to redistribute opportunity in ways that don't impose terrible burdens on white males, don't generate tremendous amounts of underperformance in the workplace, and that even generate some additional benefits for other institutions and other kinds of folks.

And I'll stop there.

CLARENCE PAGE: Thank you. While you were talking, Chris Edley leaned over and donated three of his minutes to you.


CLARENCE PAGE: And quite seriously, to get that from a lawyer is very impressive as well. So without further ado, let's turn to Doug Besharov. How do you address the question here: Is affirmative action still needed?

DOUGLAS BESHAROV: Well, I'm going to sashay around the question, that's how I'm going to do it.


CLARENCE PAGE: Okay, that's what I'm here for.

DOUGLAS BESHAROV: Let me say, first of all, I'm so old that I've seen racial discrimination firsthand, really up close. I was in college the week of Selma [Alabama]. I was in Mississippi and tried to get black kids admitted to hospitals when they were sick and they weren't admitted. And then, I worked in New York City, in the mid-'60s, and was told by my supervisor, "Don't hire a field hand," and it took me almost an hour to figure out what a field hand was.

I pictured someone in bib overalls. And I said, "What in the world would a law student be doing in bib overalls?" And I'm absolutely certain, without the benefit of too much research, that African Americans, and other U.S.-based minorities, still pay the price of generations of oppression and discrimination.

When I was in Mississippi, it was my job to document the differences between the white schools and the black schools. One white high school I went to—I will always remember this—had a wall of beautiful, are they brass instruments? Is that what horns are? A wall of them—I guess that was the marching band. And the black school had one rickety old piano.

And that was, I think, a metaphor for the quality of the education that kids were or were not given. So I'm personally deeply committed to reaching for greater equality. And as Harry said, if that's the meaning of affirmative action, to make the extra effort, I'm all for that. So having said that, I go in the other direction, and I say that I am deeply unhappy, uneasy about decisions that are based on race, and I am particularly worried about quotas.

So if I were sticking to the topic—you can't do anything for five and a half more minutes—I would probably say something like, "Yes, but," or, "No to quotas," or whatever. But Harry Holzer gave me permission to talk about a different kind of affirmative action, and that stems from the deepening problem of inadequate financial aid for all low-income students. But I'm going to argue, in the few minutes I have, that the problem is most deeply felt for disadvantaged minorities, blacks, Hispanics, and so forth, and that we are doing jolly little about it.

The irony, of course, is that fixing the problems that I'm going to talk about for a few minutes doesn't raise the rancorous questions of racial preferences or quotas. So I'm not a "yes," but I'm a "why not?" And my "why not" is, if our true goal is to increase the number of African Americans and other racial minorities who graduate from college, as opposed to being admitted, why not put our money where our mouths are?

I'm going to move quickly. You all have in the materials some figures. I'm going to refer to them. You can look at them, but I'm not going to spend any time walking you through them, because then Clarence will cut me off.

Figures 1 and 2 show that white Americans have steadily—actually, since World War II—increased not only their college attendance, but their college graduation rate. African Americans, on the other hand, although attendance rates have gone up, African-American graduation rates have hardly budged over the last 25 years. In fact, African Americans are about 25 percent less likely to attend college, but 50 percent less likely to graduate.

So if you're worrying about something, it's not just getting folks into college, it's getting them out with a piece of paper that doesn't say, "You owe us two, four, five, or $16,000," which is what happens too often. Now look, there are many reasons for dropping out of college, especially for low-income families, disadvantaged family circumstances, abysmally poor public schools, and for African Americans and other minorities, the effects of past racial discrimination and its lingering reality.

Opponents of affirmative action say, "See, we are admitting so many young people who are unqualified. That's why they're dropping out." Proponents would say, "See, we're just not doing enough for those young people when they're in school." I want to pass that argument—maybe Clarence will force us to talk about it—and say that another issue sitting right there is the inadequacy of financial aid.

Take a look at figure 3. We now spend something over $90 billion on financial aid. It's a little misleading. A lot of that is loans; the kids have to pay it back. The key thing is to look at the bottom line, which is Pell Grants. Those are the no-strings-attached, in effect, scholarships from the federal government. They have hardly budged in the last few years, a small increase in, I think, 1996, Chris, or '97, but not nearly enough if you ask me.

If you look at the next two graphs, you will see that steadily, over time, with the exception of that one change in 1996, the percentage of federal and state and local student aid, that is, grant versus loan, has steadily decreased. In '81, it was 55 percent grants; in 2001, it was only 40 percent. The most telling part is figure 5, where you will see that the purchasing power of the Pell, as measured against the cost of a four-year college education, has steadily declined.

It went up, as I said, after the 1996 changes. But from 1975, when a Pell covered 84 percent of a public four-year college cost, it's down to 44 percent. The rest is made up in loans, borrowing, work-study, and so forth. The result is, to shorthand this, that numbers of low-income students, of all races and ethnic groups, go to school, called a college, at lower rates than they ought to.

The Advisory Committee on Student Financial Aid concluded that the highest SES [socioeconomic status] students attend college. The lowest achieving, highest SES students attend college at the same rate as the highest achieving, lowest SES students. In terms of dropping out, the General Accounting Office has concluded—I can't vouch for their math—but every thousand dollars of grant awarded reduces the dropout rate from first to second year by 23 percent.

I think there's an issue there. I think African Americans are harder hit by this than other groups. Number one, they are disproportionately low income. Number two, because they suffer a larger wealth gap or family asset gap than whites, than reflected in income, there's no safety net if they need some extra money. In fact, many guidance counselors in colleges say the flow of money is in the opposite direction.

So there's no safety net for them. And then, there is the problem—got to be careful here—of mismatch, and it goes as follows. If we are taking young people who are not as well prepared for a particular school and putting them in there, I think it is clearly the case that that young person has to work harder to catch up. It's not the same as saying whether he belongs or not. We can have that argument.

Right now, I want to make the argument that if he or she came in lower than the average, he might want to work harder to make the average. You add on top of that more hours of work-study for these kids, more student loans, more stress, and is it any wonder that the dropout rates are higher?

All right, there is a simple answer here, right? And that is, target financial aid by income as opposed to what they call merit, but that's not really what they're talking about. And the problem, as far as I can tell—and I'm a neophyte to this, more of a neophyte than to affirmative action—is that America's colleges and universities are hooked on the middle-class version of aid, because without it, they couldn't operate.

But I tell you, looking—and there's one good paper that sits behind what I've—sorry, nine minutes. They are so dependent on these billions of dollars that they are not part of the solution. They will not address, as far as I can tell, the problem of the failure to target some part of this aid to low-income Americans. But there's no reason why.

And this is my last and maybe more controversial point. There's no reason why civil rights groups and the rest of us shouldn't be complaining very loudly. And so, the working title of my little talk is "Affirmative Action on the Cheap," because that's what I think it is.

Thank you.

CLARENCE PAGE: Thank you very much, Doug. That was so mesmerizing, I didn't even notice that you'd gone over time. In any case, we'll turn right away to Chris Edley.

Your response, Chris?

CHRISTOPHER EDLEY: Well, I was keeping careful track, and 92.5 percent of what Doug said I agree with, which stuns me. So let me just say that I think one quick thing that he mentioned, the so-called move towards merit-based financial aid that many states have done is, from an inclusion point of view, just disastrous. That is to say, this is the notion of moving away from need-based financial aid towards financial aid awarded on the basis of academic credentials.

And I think it really is a crisis. And I would commend—first of several advertisements to follow—I would urge everybody to take a look at our web site, at the Harvard Civil Rights Project, where there is a research paper on this issue of merit-based aid that we put out a couple of months ago. It got a fair amount of press attention, but it's a huge issue.

And I think Doug is absolutely right on the money that this issue of the financial impediments, the increasing financial impediments to college access have to be a part of the current, the new civil rights agenda, that is, Congress takes up reauthorization of the Higher Education Act. And I hope Doug is right. I hope we'll be able to forge a coalition, a multiracial coalition that takes a look at these issues.

I mean, it's harrowing when you explore what's happening in California right now, the combination of the affordability gap combined with budget cuts at the state level, combined with the demographic changes, and it's really quite frightening in terms of what it portends for opportunity in the future.

What I would disagree with, I think, a little bit—well, let's not get into that, because that would definitely take us too long. So let me turn back to the substance, which is the heart of the matter, which is, is affirmative action still needed? I begin by saying, "What is it?" and I think of it as any use of race, or another immutable characteristic, as a factor in decisionmaking or allocation of resources, but not including quotas.

So I think that the device of pointing to quotas and attacking quotas is a red herring. There is no serious defender of affirmative action who tries to defend quotas. They are illegal. They have been illegal for decades, except in very isolated, narrowly court-supervised remedial settings. It's just not an issue.

It's an issue if something that is a flexible goal is implemented poorly and becomes a rigid numerical straightjacket, which is what the law defines a quota to be. If it becomes a rigid numerical straightjacket, it's illegal, and it ought to be stopped. So I just wanted to make that point.

The other point I wanted to make about what it is, is just to remind everybody, as Harry was suggesting—affirmative action, it's a very limited tool—it is a very limited tool. I mean, it does not aspire to end poverty. It does not aspire to cure the common cold or SARS. It is only a tool designed to try to help people, who are qualified, step through a door, try to prop the door open and help them step through the door.

But if they're not qualified, if they're not prepared to take advantage of the opportunity, it's doing nothing for them. So Abigail, undoubtedly, will talk about the importance of fixing the K-12 pipeline; I wholeheartedly agree. But it's another thing to say that that's a substitute for affirmative action.

So all affirmative action does right now is it takes people who have managed to survive the broken pipeline to the point where colleges make the judgment that they are capable of doing the work and that they would contribute to the institution, and then help them across the threshold. If you fix the K-12 pipeline, then certainly, it will be the case, more folks will be standing at the threshold prepared to step across it. And therefore, affirmative action would be much less important. I'll get back that in a minute.

So limited purpose is my point. So, fine to criticize affirmative action, but let's try to criticize it for the things that it is, in fact, designed to accomplish, not for things that nobody ever thought it would really accomplish.

"Is it still needed?" of course has to relate back to "What's the justification for it?" And here, let me just tick off the three major pots of legal, moral, and policy justification. The first, of course, is remedial—let me know when I've got one minute left.

The first, of course, is remedial. And I do want to emphasize, although this is frequently omitted from current issues, there remains various forms of discrimination, often quite subtle, certainly in the employment sector, certainly in the contracting sector, but even in the higher education sector. And the way I would simply put it to you is that we, I think, all continue to suffer from this simple human frailty of when we are making judgments about people.

When we are deciding whom to have dinner with, whom to date, whom to do business with, whom to hire, we tend to prefer people who are a little bit like ourselves. We tend to be looking for some form of connection. And in America today, it remains the case that color, not only color, but certainly, color gets in the way of connection. To the extent that those kinds of biases towards people who are just like us creep into our subjective decisionmaking, on an admissions committee or on a hiring committee, well, frankly, that's an occasion for remedial action.

And affirmative action is simply one tool that helps us lean against that simple human tendency to prefer people who are like ourselves. The military has recognized that in the way in which they do promotions in the enlisted ranks and in the officer ranks, to make sure that there's a hard look taken at the promotion pool, that people who have the talent, who have the qualifications, have not been overlooked. They wrote a brief about it in front of the Supreme Court.

So remedial, it's still there. But, of course, the second justification, diversity or inclusion, is what institutions talk about, because they'd rather not talk about the continuing presence of various forms of discrimination. So the reason we want it, of course, is not because we've been discriminating or that we are discriminating, it's because we care about diversity, we care about inclusion.

The analytical argument, I believe, as a constitutional matter, has to be made that we want diversity, we want inclusion because we believe that by being inclusive, we can be excellent as an institution. That is to say if I'm a police department in an urban community, the way to do my job effectively is to see that, to some extent, my workforce reflects the diversity of the community I'm trying to serve.

I'm not making an argument for rigid numerical straightjackets or proportionality. I'm saying, to some extent, has to be reflective of the community, and there's a judgment involved. Similarly, when I'm picking people to be at the Harvard Law School, the diversity, the inclusivity of the classroom is a critical component of making sure that Harvard Law School remains the best law school in the solar system.

The third issue, of course, goes to broader social arguments that might be made in favor of affirmative action, and Harry alluded to some of those. I simply want to note that as a constitutional matter, many of those arguments probably do not resonate with our current constitutional doctrine, where issues of societal discrimination, and so forth, have been ruled out of bounds as justification, under strict scrutiny analysis of the 14th Amendment, as justification for race-conscious decisionmaking.

However, if one satisfies the constitutional predicate of a compelling interest—remediation or, we hope, diversity inclusion, we'll know when the Michigan case is decided—then there still is a policy and political judgment of whether it's a good policy idea to do it. In other words, the constitution may give us permission to use race-conscious means under certain circumstances, but then we still have the political and policy question of whether we want to enjoy, whether we want to take advantage of that constitutional flexibility to, in fact, pursue those policies.

That's where these broader questions of, do we need integration, do we care about access, do we care about people learning from each other, do we care about the provision of services in different communities, etc., all of those remain relevant, if not as a constitutional matter, certainly as a political, as a moral, as a policy matter.

Finally, when should it end? And Justice O'Connor was very sort of focused on this in the oral argument in the Michigan case. My answer is, you tell me when discrimination, when the need for special efforts at inclusivity in order to produce excellence, when will those needs end, and that's when I think affirmative action should end.

At the Harvard Law School, when I was a student in the early '70s, there was affirmative action in admissions for women. Today, this weekend, in fact, we're celebrating 50 years of women at Harvard Law School. There is no more affirmative action for women in the admissions process at Harvard Law School. There was no faculty vote dismantling the policy.

It simply happened that the pipeline became sufficiently transformed that there was no need to think about gender in making the admissions decisions, because we got the kind of inclusivity that would produce excellence at the school and would serve the profession without having to think about it. It died a natural death, and that is a huge success.

That's victory; it's great, we celebrate that. But the reality is that in all too many contexts, we haven't achieved that with respect to women, and we certainly haven't achieved that with respect to racial and ethnic minorities. I'll close with a last advertisement, and that is that I think a very, very valuable, useful briefing paper that the Harvard Civil Rights Project did concerning the Michigan case is also on our web site.

I guess we don't have copies—there will by the time you end. There'll be copies for you to pick up. And it is an effort, in 35 single-spaced pages, to give you both the legal framework, the doctrinal framework for the issues at stake before the Supreme Court, but at the same time, to try to summarize the social science evidence that bears on those legal propositions before the court.

Because there is more to this issue of "is it still needed?" than hand waving. There's more to this issue than sound bites. There actually happens to be some social science.

Thank you.

CLARENCE PAGE: Thank you, Chris. Dr. Thernstrom, are we ready for affirmative action to end a natural death?

ABIGAIL THERNSTROM: Well, I, like Doug, am going to skirt the issue again. I've been talking about racial preferences since the late '70s, when, at the time, people looked at me and said, "What are you talking about exactly?" because it just wasn't on the horizon. But anyway, I've gotten a little tired of hearing my own voice on it, and I am going to, as Chris suggested, talk about the pipeline.

I mean, I'm very well known as a passionate opponent of racial preferences. I'm an advocate of aggressive nondiscrimination, but an opponent of preferences of all race-driven policy. I haven't changed my mind on that, in great part, because like Doug, I'm old enough to have remembered the racial sorting of the Jim Crow South, and I'm opposed to all racial sortings, in any form, still today.

But in any case, nevertheless, whether the plaintiffs win or lose in the two University of Michigan cases—and, of course, there are two cases, they could win one and lose the other—either way, preferences are not going away in universities, in employment, in contracting. And while conservatives like to point to polling data showing that the majority of Americans are opposed in judging people by the color of their skin, in fact, preferences have never been an occasion for national outrage, for mass mobilization, and they never will be.

So in the near future, whatever the court says, preferences are here to stay. And again, while I do support the plaintiffs in the Michigan cases, nevertheless, I think it is a mistake for conservatives to make their demise the number one item and, sometimes, the only item on their civil rights agenda, or for liberals to see their perpetuation as make or break on their civil rights agenda.

Now, with respect to college admissions, specifically—and I thought we were going to talk about Michigan today—William Bowen, Derrick Bok's Shape of the River, by the way, a little advertisement here. I should have brought the long UCLA Law Journal piece that my husband and I did reviewing that book. And anybody who wants to see it, I'd be delighted if they give me their card; I'll send it to them.

But anyway, we do have in the UCLA Law Journal a very long review of Bowen and Bok's Shape of the River. But anyway, in their long brief for racial preferences, they did get one thing indisputably right—we obviously are critics of the book—but they got one thing indisputably right. They say, whatever the courts say, the elite colleges and universities, like Michigan, will find a way to create what they see is a critical mass of black students.

It's arbitrarily, today, set in the Ivy League colleges at about 6 or 7 percent. And as Bowen and Bok acknowledge, there are only two ways of ensuring that critical mass. Either you have racial double standards in admissions or a drop in standards across the board, which will solve the problem of the tiny pool of black and Hispanic applicants who meet the standards by which white and Asian students are judged. And it is the problem of the pool that I want to talk about today.

Racial preferences in admission to selective colleges and universities are obviously the direct consequence of the size of that pool. And the typically low academic performance of black and Hispanic students I see as the most important source of ongoing racial inequality in this country. I don't have any doubts about the fact there's ongoing racial inequality.

But today, equal skills and knowledge make for equal earnings. It was not always true. It is today. And I mean equal skills and knowledge, not years warming a seat in school—quite a different matter. Racial attitudes of Americans have dramatically changed in recent decades. I coauthored a book that's partly on that subject.

And the commitment of most Americans, I believe, to racial equality is today very deep and irrevocable. But only one way, in my view, to realize that equality, and that is to close that racial gap in skills and knowledge in the K-12 years. The point at which we do, we will have addressed the educational issue of our time, the civil rights issue of our day, and the question of preferences, as I think it was Chris who said, will be moot.

I am, frankly, on a moral tear about the sea of academically low-performing black youngsters in American public schools today. I mean, Jesse Jackson can lead a march over the Golden Gate Bridge protesting the, quote unquote, "resegregation of the University of California," which I believe is an absurd description. But, in fact, we are talking, as Harry Holzer said, we're talking, relatively speaking, about a handful of black kids out of approximately 8 million African-American school children in this country. And those school children are typically severely under-performing in school.

I'm going to run through the basic picture in hopes that not everyone in this audience already knows it. I'm using National Assessment for Educational Progress data, the NEAP data. By 12th grade, on average, black students [are] four years behind those who are white and Asian; Hispanics don't do much better. The college who admits the average black student, the typical black student, is, in effect, choosing a youngster who has made it only through eighth grade.

He or she will have a high school diploma, but not the skills that should come with it. And Hispanics, again, do only a little better than African Americans. Asians, by contrast, look very much the same as whites. There's another way of judging the magnitude of the racial gap. The NEAP assessments report not only average scores for each racial or ethnic group, they also place each individual test taker in one of four achievement levels: below basic, basic, proficient, advanced.

The proportion of white students who have not acquired even basic skills, who are in a below basic category, in 12th grade, is around 20 percent, in most subjects. Again, Asians look roughly the same. But in five of the seven subjects tested, a majority—a majority of black students perform below basic. In math, the figure is almost seven out of ten; in science, it's more than three out of four. These are shocking numbers, or they should be shocking numbers, and they should have been shocking numbers a long time ago.

A majority of black students don't even have partial mastery of the fundamental—this is NEAP's language—fundamental knowledge and skills expected of students in the 12th grade. Again, Hispanics generally a little better than blacks, still far behind.

If we switch our gaze from those at the bottom to those who are at the top in terms of school performance, the news is no happier. In science and math, a mere 3 percent of black students [are] able to display more than a partial mastery of the knowledge and skills that are fundamental for proficient 12th grade work. Reading scores are a little better, but still shockingly low.

These disparities [are] even greater when we look at the elite within the elite, the very small group of students whose performance is rated as advanced. These are the academic stars that the University of Michigan, Harvard, Stanford, etc., go after. I mean, the numbers in the advanced categories, for all groups, are very small. Only the top 3.4 percent of white students, 3 percent of Asian, score in the advanced level in math.

But again, there are huge racial and ethnic disparities, such that in math, only 0.2 percent of black students are advanced. The figure for whites is 11 times higher; Asians, it's 37 times higher. Given these numbers, of course, glaring racial double standards are needed in order to get a freshman class that's as much as 6 or 7 percent black at schools like Wesleyan and the University of Michigan.

The demand for academically highly accomplished black and Hispanic students is much greater than the current supply. I mean, the admissions office—Chris talked about no quotas—the admissions office process ends up race driven because of these disparities. It is not possible to make race just one factor, a little light thumb on the scale. And thus, given the disparities, and thus, the line between goals and quotas is totally blurred. They've become one and the same.

This picture, by the way—picture of the NEAP score is worse today than it was in the late '80s. We've been sliding downhill. And it would be nice to believe that the slate is wiped clean when a student graduates, graduates from high school. It's a new day, a new beginning, they catch up in college. It is not the case, in general.

By that age, the kids who have fallen behind stay behind. Very hard to catch up. Very hard to catch up by the time they arrive, by the way, in high school, behind, but not impossible. The doors to college today are open, even to those with extremely weak academic records. We've got, what, 3,800 institutions of higher education in this country; most of them are nonselective.

It's not that some are nonselective; most are not selective. You just walk in. I'm counting community colleges, too. In Massachusetts, you don't even have to have a high school degree to go into a community college. The problem, as Doug said, is not entering college, it's staying and graduating. Three out of four African Americans enter college, one in six finish, compared to about one in three for whites. Students who leave high school with eighth or ninth grade skills, they simply can't keep up in colleges that are not geared to teaching students what they should have learned in high school.

I mean, the good news is those who go in with equal skills have an equal chance of graduating. There's no racial difference between kids who enter college with equal skills. And there's another part to this picture, which is the high level of remedial course taken among black and Hispanic kids. We've got numbers for California State University system. You have to be in the top third of the state's high school graduates.

No less than two-thirds of all black freshman who started in—okay, started in the fall of 2000, had to take a remedial course in English, double the rate for whites. Even worse, almost 80 percent of entering black students required remedial classes in math. The white rate was 36 percent.

Those who take remedial classes, not surprisingly, are much less likely to graduate than students who are better prepared. So those who suggest, well, colleges have a magic formula to overcome the disadvantage of entering way behind are wrong, and even greater preferences, in fact, are needed for the applicants to law and medical schools.

There is, just to go back to my starting point, there is, in my view, only one solution to this problem. The black and Hispanic kids have to graduate from high school having acquired the skills and knowledge that college work demands. Chris Edley will say to you that's a project with an impossibly, unacceptably long timeline. And I will say not necessarily.

Across the country—this will be spelled out at length in the forthcoming book, in October—across the country, there are terrific schools in which black and Hispanic kids are graduating with, you know, the kind of performance you dream of. Those schools, however, are radically different from the normal public schools that one sees walking just at random into any urban public school, at any level.

I mean, this is the issue, it seems to me, that those who care about racial equality should be focusing on. Again, I'm firmly on the side of the plaintiffs in the Michigan case, but preferences are not the core problem, and they're certainly, not in my view, the solution if racial equality is our goal.


CLARENCE PAGE: Thank you. I would like to focus on one question before I go to the audience and give everybody a chance to get a crack at this. Dr. Thernstrom, you mentioned to get a critical mass, one alternative is to lower standards.

ABIGAIL THERNSTROM: Which Scalia brought up at the—

CLARENCE PAGE: Exactly, Justice Scalia brought it up, and that's partly why I bring it up. Because it sounds like—and this goes back to your numbers—you were talking about two different things. One, we're talking about elite universities that practice affirmative action, which is the topic of our program today, the other is the vast majority of colleges and universities for which this is a non-issue.


CLARENCE PAGE: So you gave the numbers, and I agree, deplorable. And I stress this to my 13-year-old every night, like tonight when I go home. These productivity numbers are deplorable for reading, writing, arithmetic. But when we're talking about affirmative action at elite universities, aren't we talking about creaming off the cream of all races? We're not talking about under-performers.

So when you talk about lowering standards, I just wonder how low are you talking about? Would it be low enough for even I, with my paltry 1140 SAT, to have gotten into Michigan, Harvard, the elite universities? Because I have obviously under-performed in my life based on the standards of grades and test scores, and public high school, etc., etc.

Well, you know, I don't know how I got in the door of this building, you know. I mean, affirmative action discussions just really amaze me for people to talk about how they've got to lower their standards to let more blacks and Latinos in. So I just wonder, what are we talking about when we're talking about lowering standards? Is this a real issue, or is it a bogus issue?

ABIGAIL THERNSTROM: No, I don't think it's a bogus issue. I mean, in the first place, the pool is so small that, and 70 percent of the high performers, the last I knew, applied to Harvard. So Harvard doesn't, basically, have double standards, because it has its pick of—

CLARENCE PAGE: And always will. How about Michigan?

ABIGAIL THERNSTROM: No, not necessarily always will. I mean, one could make a very good argument that Harvard has an inflated reputation.

CLARENCE PAGE: Especially if you went to Yale.

ABIGAIL THERNSTROM: I have a husband who teaches there.

CLARENCE PAGE: There you go.


ABIGAIL THERNSTROM: But, I mean, because of the size of the pool, the result is everybody else is left with having to engage in racial double standards. Now, I don't care how these schools admit their kids. You can throw darts at a board as far as I'm concerned; you can throw the applications down the stairs and pick the ones that arrive there first. But I want uniform standards, because I don't want people to judge on the basis of their skin color.

Now, if you say that, as Scalia did to the University of Michigan, "Okay, lower your standards," and the problem will be solved, and it will be solved if they lower their standards.

[Cross talk.]

ABIGAIL THERNSTROM: Well, wait a minute. Let me just finish that one sentence. You know, if you say that to the University of Michigan, they're horrified. You know, they want those very elite standards for the whites and Asians.

CLARENCE PAGE: Well, okay, Harry, if you don't mind, I'm going to turn to Chris because his huffing and puffing is taking the starch out of my collar, and that's quite all right. We appreciate passionate spirit of discussion.

CHRISTOPHER EDLEY: I'm sorry. I forgot to take my medication before I came. Ask the people who write the SAT. They will tell you that the SAT is designed to predict first-year grade performance—first-year grade performance. And their view is, it is lunacy for an institution to make the SAT even the most important, certainly not the all-important, but even the most important factor in deciding whom to admit.

Because it would be a pretty bizarre institution to choose as its mission, we want to admit the people, we want to educate the people who can get the best grades as freshman. And, in fact, the selective schools don't adopt that as their mission. Point number one.

ABIGAIL THERNSTROM: But they don't use SATs alone.

CHRISTOPHER EDLEY: Well, that's right. But even SAT combined with high school GPA to produce those indices, that's not what they define as their mission.


CHRISTOPHER EDLEY: All right. So all I'm saying is that the question of what constitutes qualifications is a loaded phrase. What institutions have decided is that they want to figure out, who do we believe is reasonably and likely to be able to persist and succeed, and who do we want to admit in terms of building a collection, a paletteof human talent and attributes to create an intellectual environment, a community that will be excellent? That's what the issue is.

Now, what ETS [Educational Testing Service] says is, if you have an SAT score of 1100 or more, if you have an SAT score of 1100, you have a 50 percent chance of graduating from the most selective institutions in the country. An 1100 gives you a 50 percent shot at graduating. Now, I'm quite puzzled by the people who would say, "That's very interesting, but I'm sorry, Michigan, Harvard, Stanford, Georgetown, we think that if you decide to take a gamble on somebody who has a score of 1100, that what you're doing is ushering in the end of Western civilization."

What they are doing is using these measures as one factor among many, and trying to create, trying to build, trying to constitute an academic community. And I believe that they have argued—in the Michigan case, I know that they have argued, and I believe that they will succeed in the Supreme Court in insisting that they are entitled to some degree of deference from the court in the way in which they select their mission and proceed.

[Tape change.]

CLARENCE PAGE: I want to go to Doug Besharov for a response. Sir, I think you had your hand up.

DOUGLAS BESHAROV: Yeah, thanks. I was listening to Chris, and I was agreeing with him, even more than 92—what was it, 92.5? And if that's how it worked, we might have a very interesting conversation.

Two weeks ago, there was an article in the Washington Post about admissions at BU, and the last three paragraphs were about the entire admissions committee in front of a reporter for the Washington Post, trying their damnedest to figure out how to let in someone with a 900 SAT and a 1.7, at an elite private school. This was a young immigrant woman who had been in a public school, been in an elite private school, in wherever she was, for four years.

That looked to me, Chris, like a recipe for catastrophe. They finally said, in the Post, to quote it as saying, "We can't see how to make this work." Now, I interviewed—


DOUGLAS BESHAROV: They were ready to go if they could.

CHRISTOPHER EDLEY: But they decided they couldn't.

DOUGLAS BESHAROV: Let me finish my point. I interviewed four admissions directors—

CHRISTOPHER EDLEY: There's a sample.

DOUGLAS BESHAROV: Yep. And I asked about race. And I said, "Are you guys responsible for whether they graduate?" This is why I want to get back to this point, because I'm not disagreeing—

[Cross talk.]

DOUGLAS BESHAROV: If it worked the way you said it would work. What I worry about is it doesn't work that way in a lot of places. And one of these fellows says, "That's not how we keep score around here. My responsibility is to get a number in, not a particular number, but enough. That's what I'm responsible for, and retention is not what we do at the admissions office."

And that's what scares me about it. What scares me about this process is there are places—I don't know whether it's 1 percent, Chris, or 50 percent—where this process has gotten out of hand. That's all.

CHRISTOPHER EDLEY: Well, look, I don't disagree that there are places where it's gotten out of hand. Show me a social policy in which that's not the case. Number two, I will even agree that there is a special moral cost to making decisions about people based upon race and other immutable factors that makes it especially problematic when it gets off course. So I will concede that to you.

But that said, we've got to do some kind of a cost benefit here. And you have a set of institutions that are trying to pursue a mission, that are trying to constitute a community that's going to be excellent in some reasonably defined way, and they're making judgments about that trade-off. I mean, look, I'm building an orchestra, damn it—I'm building an orchestra, and it may just be that, you know, the oboe players, they're not as terrific as the violin players, but I need some oboe players or I can't have an orchestra.

And what I'm after, overall, is an orchestra that will be as good as I can possibly make it, but I've got to have some oboe players in there. And some of that's going on. The other thing that you raised that gets back to something that Abigail said is, if an institution decides, "We can make our communities succeed if we combine this admissions strategy with a remediation strategy, with a student support strategy, with a financial aid strategy, etc., that makes it possible for the students we admit to thrive," that's what they're doing.

So to some extent, I'm troubled if the admissions office says, "We don't care whether they get through to completion." That sounds to me problematic, because at least it ought to be a serious concern of theirs. But on the other hand, if what they're saying is, "We have a division of labor here, and we admit people that we think, based on history, probably have a shot, but then it's up to other people on the faculty and in the administration to deliver the supportive services to, in fact, make that possibility of success ripen into a reality of success."

That, for example, is the story at UT Austin, where the percent plan admits, at the undergraduate at UT Austin, the 10 percent plan admits are, in fact, outperforming academically the regular admits, because UT Austin has accompanied this percent plan admission strategy with a whole bunch of academic supports, and that's fabulous. Unfortunately, it hasn't worked at Texas A&M, which shows you that there's a lot to this besides the admissions algorithm. It's how the institution as a whole is constituted.

CLARENCE PAGE: Let me give Harry a chance to pipe in here.

HARRY HOLZER: Let me just make a couple of points. You know, so much of this discussion, and so much of the things that the critics of affirmative action claim would lead you to believe that if we only got rid of affirmative action, university admissions, even at the most lead schools, would be this meritocracy based only on test scores and grades. And that is false. It has never been true. It is not true today.

Affirmative action based on race and ethnicity is not the only source. There is a range of test scores among whites that reflects many, many different things in the admissions process—as I stated before, whether your parent went to that school, whether you're an athlete, a musician, an artist, and whether or not you come from Oklahoma. It does not generate the kind of passion that racial preferences generate.

And we can speculate about why that is. I think it has something to do with the divisiveness of race in this country, on all sides. You know, I think people are still crazy about race in this country, on all sides of the issues. It's very hard. But the focus uniquely on race as a sort of preference is really misguided and inaccurate.

To talk about the dropout problem, the dropout problem is a serious problem. We don't deny that. And I think, certainly, more needs to be done on the retention side, and Doug laid out many of those issues. However, I think the evidence is also clear on this. On net, minorities admitted under affirmative action are better-off than they would be in a world without affirmative action.

Somehow, we're making this argument that we're sticking it to these poor kids by admitting them into these schools where they don't belong, and then they all drop out, or most of them drop out. And yet, studies have looked at this very clearly. On average, those who admitted to these elite schools do better afterwards. Of course, they do a lot better if they graduate.

We don't have good clear evidence on what happens to the dropouts, and it's important, I think, to find out more, for those kids who get admitted and who don't make it. By the way, whether or not they're white, or black, or Latino. What are the paths that they take? Is there something positive that comes out of that early incomplete experience at that school that builds on—but the notion that we're hurting this community, that for their own sake, we need to eliminate this policy, I also think is not consistent with all the evidence I've seen on that issue.

DOUGLAS BESHAROV: Also, the completion rates for minorities are higher at the more selective schools that engage in affirmative action than at the less selective schools that don't.

CLARENCE PAGE: Our audience has been very patient here, and I'm starting to see hands already. So I'm going to turn to this gentleman here, and we do have a microphone, so here we go.

JIM LOEWEN: I've written some stuff about the SAT. I want to make just two points. They're not questions, but then anybody who wants to can comment on them. Point number one, Ms. Thernstrom, I really appreciate your emphasis on K-12. I think that's great and much more productive than some of the things you've emphasized in the past.

I want to say, however, though, doesn't that let off the hook everyone at every institution of higher learning, except maybe the ed schools? In other words, if you're a professor of physics, if you're a professor of classics, or if you're in the admissions office, or anything else, since you're not dealing with K-12, it's not my problem. So I would suggest you're doing just the right thing, but that has nothing to do with whether or not the institutions that we've been talking about need to do something about the problem of racial inequality in America, and that problem might be a little bit of affirmative action at age 18.

My other point is this one. All of you have been using the SAT, implicitly, as some kind of merit-based instrument. I have a whole rap on this I won't give you, but—

[Cross talk.]

The SAT is intrinsically biased; it has to be statistically—I can actually prove that. And it is not, therefore, a good measure of ability across racial groups. I don't think it's too bad, even within African Americans. Bizarrely enough, I base that on eight years of teaching, among other things.

But across racial groups, it's got a real problem. And so, I don't think it should even enter into the discussion. Now, you were using NEAP and graduation rates, I mean, graduation records of high school. Those are different matters. But I think the SAT ought to just not be a factor at all. And I think the University of California is going in the right direction.

CLARENCE PAGE: I'm going to give Abigail Thernstrom a chance to respond here.

ABIGAIL THERNSTROM: Well, it's not a question of letting universities off the hook. But, look, your professor of classics is not very good at teaching students who come in without the reading skills of a freshman in high school. I mean, generally, university and college faculty are not very good at teaching what children and young people should be learning in those K-12 years, and it is very, very late for them to start to try to do so.

I think that the returns on these remedial programs are very meager. We need to start much, much earlier. In fact, nobody has figured out, in a satisfactory way, what to do with a ninth grader who does not have second grade reading skills. There are a zillion people across the country trying to answer that question, how to structure a math program for high schoolers who don't know any arithmetic.

It's very hard, at that point, to play catch-up. I certainly believe in what Berkeley is doing, which is to try to send tutors and mentors, and what have you, into the K-12 years. That's great. And Chancellor Burdell, when he was chancellor, said, "God, we wouldn't have done that had it not been for Prop. 209." Well, that's a pretty sad statement.

The question of bias in SATs, I'm simply in disagreement with you. But that's a very, very long argument, and I don't think we want to get into it here. Let me just say, I mean, there are a million things to say in response to Chris and Doug, but no one believes in simply using SATs and grades, and as I said, no college does this. But to say that there is also a range among white and Asian students, no, the range isn't the same, unfortunately.

You really do, you have to start with the recognition that we're talking about racial double standards. You can defend them. I think you can make principal arguments for them, but they are a fact. And all of us believe in taking gambles, but tell that to the Asian applicant, for whom a gamble is never taken at these elite schools, because as one person in higher education once said to me, "We already got too many of them."

CHRISTOPHER EDLEY: That's just false, Abigail.


CHRISTOPHER EDLEY: No, look, wait a minute. Most selective universities, I think this is the case, certainly anybody who's thoughtful would draw a distinction among Asians between fifth-generation Japanese or Chinese Americans—

ABIGAIL THERNSTROM: I wish they would. They don't—

CHRISTOPHER EDLEY: —and people who are first- or second-generation Southeast Asians. There's a difference in terms of their social and economic circumstances, there's a tremendous difference in terms of their K-12 situation, etc. And while that may not be because of, frankly, administrative laxity that I would criticize, they may not be formally named as one of the underrepresented minorities in the subtler elements and the subtler aspects of the admissions process.

There's enough give in the system to take account of that diversity. So I think it is simply wrong to say that Asians, as a total group, are losers at affirmative action. Moreover, I would emphasize that virtually every Asian-American civil rights group and community group supports affirmative action, including in California.

[Cross talk.]

Unless you're saying they're too stupid to appreciate their self-interests, I think that they obviously have a different view of whether they're burdened by—

ABIGAIL THERNSTROM: No, I don't think these civil rights groups necessarily speak for—


ABIGAIL THERNSTROM: I said I don't necessarily think that the leadership of civil rights groups speak for the rank and file of—


ABIGAIL THERNSTROM: No, I don't. I'm simply saying, look, I know what these admissions pictures look like. I know that, for instance, the University of California system, where Berkeley, UCLA, you're close to 50 percent Asian undergraduate body, in a state that's 10 percent Asian. Harvard, you're up to about 22, 23 percent Asians. Of course, Asians are about 4 percent of the American population.

This is a disproportionately high-represented group in the elite schools. These schools, they lump them together. I agree with you, they shouldn't.

CLARENCE PAGE: Just for the record, though, we are talking not about so many Asians being denied an education. The majority of the students at Michigan, Harvard, etc., are picked for grades and test scores, right?

[Cross talk.]

CLARENCE PAGE: That's a primary factor, isn't it?

ABIGAIL THERNSTROM: Well, that's a strong element in—

CLARENCE PAGE: Tell me if I'm wrong. The facts of the case in Michigan, as I understand it, are that more than half of the students are picked for grades and test scores before you even get to the pool that's discretionary, in terms of affirmative action and other factors.

CHRISTOPHER EDLEY: Well, the presumptive admits, that's true. But, because of the undergraduates, it's 80 points based upon, you're right, on test scores and grades.

CLARENCE PAGE: So it is possible the Supreme Court could strike down the Michigan case while still saying race can be a factor.

CHRISTOPHER EDLEY: That's right. And it could say that the whole file method employed by the law school is permissible, but the undergraduate point system is impermissibly mechanical. Absolutely, they could say that, and actually, that would be my bet.

ABIGAIL THERNSTROM: My bet, too, Chris. You and I are going to agree on something.

[Cross talk.]

HARRY HOLZER: —because I don't think there's anyone up here that thinks that the test score gap isn't a big problem. And all of us would love to solve the test score gap at the K-12 level. I haven't heard anyone disagree with that view. Unfortunately, I have to be a little more pessimistic than Abby is on our ability to solve that problem, in the really near future, for a couple of reasons.

Number one, the best, if you look at the research evidence on what explains the test score data, and still, what I think the best batch of studies on that is the Brookings find, edited by Sandy Jenks and Meredith Phillips—about five years ago, it came out. When you read that whole volume, you sort of end up with this sort of—we understand some of this; there's a lot we don't understand about what generates this.

So if you don't really even understand a lot of the problems, your ability to solve it, really soon, is limited. But secondly, you know, you have a set of interventions that look promising, you know, and school choice, and vouchers, and charter schools, and the whole range of things, and school integration, which has very clear effects on test scores.

But what we've learned—you know, and I'm a labor economist, so I've spent a lot of my career looking at the difficulties that low-income folks have in the labor market and whether model training programs can be replicated and implemented on a large scale. And you spend your career with a broken heart, because you identify these model programs—you know, we can talk about this one in this place and that one in that place—and then you try to replicate them, and they don't work elsewhere, or you try to bring them up to scale, and you don't get the same results.

You could even question a lot of these model programs—just, you know, the school choice programs and the effects they have. There are a lot of methodological questions about those studies that have generated these big test score gains, aside from the question of whether those could be remotely implemented on a large scale. And I think we should continue to study those issues and experiment, and do research.

But to say that we can solve this test score gap fairly soon and, therefore, eliminate the need for affirmative action, in the near future, you know, I just think is a stretch. I wish it were true.

ABIGAIL THERNSTROM: Harry, you're distorting what I said. I didn't say it was going to happen in the near future. I'm actually very pessimistic about that. I said it could be done. Why won't it happen? Because it cannot happen within the structure of traditional public education as we now know it, because the interventions that people talk about are fools' gold.

But some replication efforts do seem to be working. The replication efforts of the Kip Academy, I believe, will work. But, you know, I'm not optimistic. I'm just saying it can be done. This is not rocket science. What's rocket science is how to make it happen within the traditional public school framework.

CLARENCE PAGE: Okay. Greg Squires, please. And I hope you can help Doug get back into the discussion here.

DOUGLAS BESHAROV: I've been very happy where I am.


CLARENCE PAGE: Okay. Well, we'll see about that.

GREG SQUIRES, George Washington University: Chris, you may remember, several years ago, meeting in my office at the U.S. Commission on Civil Rights, discussing a series on the black tax that you were doing in Chicago. My question for Abigail and Chris is, why does nobody pay any attention to the U.S. Commission on Civil Rights today, on this issue or any other issue? What could be done to make—

CLARENCE PAGE: —talking about affirmative action or the black taxations?

GREG SQUIRES, George Washington University: Affirmative action, or almost any other civil rights issue today. And what could be done to make the U.S. Commission on Civil Rights a more effective participant on these debates?

CLARENCE PAGE: Who wants to take that one?


CHRISTOPHER EDLEY: Well, let me say two things—first, something that I want to say, and then I'll answer your question. So briefly, here is the way I would suggest that people think about the test gap. There's one way to think about the test gap that is, here is evidence, a medical indicator if you will, of the poor health, systemically, of the K-12 system, and it ought to be a call to action. I'm with Abigail on that.

But now, with respect to affirmative action in this very narrow group of selective higher education institutions, how do you think about the test gap? And here's a piece of it, I think. It used to be, until Title 7 was really implemented, it used to be that there were police and fire departments around the country that had strength requirements and height requirements for employment.

So, for example, to be a firefighter, you had to be able to lug, I don't know, 200 pounds 20 yards, or something like that, over your back. Not surprisingly, this turned out to be a disqualification for a lot of women who wanted to be firefighters. Under force of Title 7, fire departments revisited that and concluded, "Hey, well, you know something? To be good at being a firefighter, you don't need quite that much strength, and you don't have to be five foot ten."

So they changed their standards. Now, today, with more women in the police department and more women in fire-fighting departments, it turns out, if you measure strength or if you measure height, there's a gender gap. There's a gender gap by those measures, which, I would suggest, is analogous to the test gap that we see in these selective institutions, in the sense that those departments have concluded that you don't have to be 5′11″ to be a terrific cop. You can be 5′8″ and still be a great cop.

Similarly, these institutions have concluded that you don't have to have a 1600 or a 1550 in order to succeed at our institution. You can have a lower score and do just fine, thank you very much, and by the way, make a terrific contribution to this academic community. So they have changed the standard to focus on the performance of their institution and who can succeed, and how can our institution be excellent?

That does indeed recapitulate disparities in K-12, in the form of measured differences in test scores. But these institutions have made the judgment that the educational benefits for everybody are worth that differentiation.

Now, on the Civil Rights Commission, my view is that there's several things that work. One is that it is very difficult for an organization that has the same budget that it did in 1977, in nominal dollars—no, I'm sorry, half the budget that it had in 1977, in nominal dollars—to be very effective at doing much of anything. It simply doesn't have the resources to produce studies, to commission research, etc., and that's definitely a problem.

I think a second problem is, frankly, over the past 20-plus years, perhaps tracking the ideological politicization of discussions of race, many of the discussions in the commission have taken on that same character, and that certainly does not put the commission in a position to try to model for the country sensible discourse on race or other civil rights issues. So I'd say that the Civil Rights Commission has become a concentrated microcosm of the ideological divides we have on the civil rights agenda, and that's deeply unfortunate.

And, finally, I'd say that much of what the commission has done, over the last several years at least, that has been very good and has been very important work, has either not been reported on by the media or has been distorted by the media. And, of course, that's not unlike what happens with a lot of issues in the media.

But at least I have the sense—Clarence could correct me—I have the sense that both the quantity and the quality of reporting on civil rights issues today is not only poor, but is less good than it has been at many points in the last 50 years.

CLARENCE PAGE: It's sort of been squished out by Iraq lately. Abby, other thoughts on that?

ABIGAIL THERNSTROM: Yeah, sure. Look, Chris, on your last point—and I'll go back to a couple you made before that—on your last point on the level of interest in the media on civil rights issues, I mean, your center at the Harvard Law School, its reports get a lot of media attention, and rightly so.

CHRISTOPHER EDLEY: But it's Harvard.

ABIGAIL THERNSTROM: No, but wait a minute. No, you should be able to say, "But it's the U.S. Commission on Civil Rights." And the reason you can't say that is because, indeed, your center is putting out much more interesting, much more solid work than the Commission is putting out. The Commission should meet the standards of your center, accompanied by, for those of us on particular issues who want to do so, dissenting scholarship, so that we have a robust debate.

In fact, this is not a commission that welcomes robust debate. In fact, it's a commission that squelches robust debate. And that is a very sad intellectual fact, and I do not think that is representative of the larger world out there. There is a kind of changed universe, I think. And by the way, I don't think that the basic issue is budgetary here, but there is a changed universe that I think is very important.

In 1957, when the Commission was formed, it was doing the only work in the country, and it was path-breaking work. I mean, I've written a book on minority voting rights—my first book, Whose Votes Count?—in which I absolutely relied on the U.S. Commission work. It was fabulous work.

Now, today, one, the Commission's work is no longer fabulous. And by the way, on this squelching of voices, I mean, one of the interesting things, to me, when we did the Florida report, I mean, I still have a—I mean, that book is still in print, it is still regarded as a very, you know, important work on the Voting Rights Act—not even a footnote to that book in that Florida report, because dissenting voices, you know, the titles of books can't even be mentioned.

But what I was about to say is, in 1957, those early reports that were very, very important to the passage of the '64 act, to the passage of the '65 act, I mean, you know, it was fabulous work. Today, there is a whole range of institutions and organizations, including your center, that were turning out good work. And the commission, therefore, doesn't have the same monopoly, and it's—

CLARENCE PAGE: I'll say that about the Manhattan Institute and other think tanks.

ABIGAIL THERNSTROM: But it also doesn't have the staff, the scholars that an AEI, or a Manhattan Institute, or a Brookings, or whatever. I mean, the staff just isn't up to the job I would like to see it do. And, of course, you've got this problem of what it's allowed to say, even if it existed.

CLARENCE PAGE: Thank you very much. I'm going to squeeze in one more question here, and hopefully, it will give Doug Besharov a chance to get back into our discussion here.

Yes, ma'am.

CARMEN DELGADO VOTAW, Alliance for Children and Families: Given the fact that the society in the United States is not very prone to think of the worth of low-income people, or to assign resources to low-income people, if one of the solutions is to do more so that we can alleviate the gap that low-income people feel, what kind of educational work do we have ahead of us to turn the Congress around, to turn localities around, to make this really a good admission?


DOUGLAS BESHAROV: Geez, Louise. Thank you so much.


DOUGLAS BESHAROV: It would be helpful if it weren't the case that both political parties were going after the same set of middle-class voters. One can't read the history of student aid business in '96, '97, without being struck by the fact that it was, as Treasury Secretary Rubin said, "Trying to give the middle class goodies just before the election." So there would have to be some division there.

Now, I'm not a political scientist, but I think part of the problem is that, particularly, I think, on the Democratic side more than the Republican side, the Democrats can take some constituencies for granted; Republicans can't. The conservatives turned on George Bush I as soon as he started talking about raising taxes. There was no loyalty there, which has something to do with why he didn't win—there were many other reasons, too.

And one sees Republicans quite ready to leave, you know, if someone doesn't pass whatever's the litmus test of this event. But if you think about African Americans and, to a lesser extent, Hispanic Americans, they almost get taken for granted in this equation, because who else are they going to vote for? is the line.

So to take this in a bigger point, I think if those votes were more up for grabs, there'd be more attention to low-income concerns. But I think, from what I've seen, that is not the case, and it's unlikely to be the case for quite a while.

CLARENCE PAGE: I want to thank you very much. You've all been a great audience; the panel's been terrific. I'm going to ask for one more round of applause before I turn things back over to Harry, and say thank you all very much for being a terrific group.

Harry Holzer.

HARRY HOLZER: Thank you. I'm going to sub for Judy Feder, who's the dean of the Georgetown Public Policy Institute, who couldn't make it today, by thanking, first of all, my co-panelists, Clarence for moderating, and Abby, Chris, and Doug for joining me on the panel, and all of you for attending.

We are having a reception in the next room, a little wine and cheese reception. Please join us for that. Thank you very much.



Topics/Tags: | Education | Race/Ethnicity/Gender

Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact publicaffairs@urban.org.

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

Email this Page