Affirmative Action: Is There a Middle Way?

Glenn Loury
10:01 a.m.  Wednesday  10/9/96 

       In the spirit of Herb Stein’s invitation, I’d like to leave the rarefied discourse on matters of principle (about which there remains much to say), and get down to cases. I taught at the Kennedy School of Government at Harvard for many years. We used the case method to expound upon the subjects of public policy analysis, public management, electoral politics, and ethics in public service. Often the cases would involve studying a situation in which the interaction among different racial groups was a key aspect of the problem at hand. Those discussions were vastly more effective as pedagogic devices when the student body involved exhibited some of the racial diversity of the social environments described in the cases. (The point is especially easy to see in this political season: What campaign would put together a focus group without attention to demographic diversity?)
       An admissions policy which, on occasion, favored a black applicant over a white with slightly better test scores, in order to ensure that the minimal amount of student diversity needed to facilitate the pedagogic mission of a professional school described above, would not, in my view, constitute an unreasonable racial discrimination, though, of course, such a policy would be discriminatory. In the example, the race of an applicant can be a kind of qualification. Just as the race of a police officer, or a candidate for judicial appointment can, under certain circumstances, also be a component of what makes a person “qualified” for a job. If women feel more comfortable being treated by female gynecologists, then health services, including public ones, will find that to properly serve their clients, taking account of the gender of an applicant for a gynecological appointment, along with other factors, is a prudent and reasonable thing to do. Confronting women with all male staffs, when some females could have been added through a slight adjustment to the other measures of qualification, would seem to be, and would in fact be, an indication of insensitivity to the clinic’s clients.
       In Judge Smith’s opinion for the 5th Circuit majority in the “Hopwood” case (involving the University of Texas Law School), the argument is made that racial diversity among law students can never be a compelling state interest because seeking such diversity presumes that racial identity signals a person’s intellectual outlook. It is diversity of ideas, not of races, which schools should seek, says Smith. This argument is wrong in an instructive way. Suppose that a school wants to teach the very lesson of Judge Smith’s opinion–namely, that not all blacks think alike; that some, like Justice Thomas, are both judicial conservatives and strongly identify with their racial group’s interests and strivings, while others, like former federal appeals court judge Higginbotham, are judicial liberals who, in a very different way, also strongly identify as blacks. If this is a lesson worth teaching to successive generations of future leaders of our nation, and I believe it is, how could it be conveyed without some attention to attaining at least a minimal amount of racial diversity in the student body? If that diversity could be attained with a relatively slight disparity in the test scores of admittees, then why should a school be enjoined from attaining it?
       The fact is that large numbers of fully qualified applicants, with the highest test scores and grades, are turned away from elite institutions of higher and professional education every day, for a variety of reasons. The crux of the issue here is whether or not turning some few of them away for the purpose of attaining racial diversity, where a minimal degree of such diversity is essential to the educational mission, constitutes an especially egregious affront to our principles of equal opportunity. I do not think so.
       Note well my use of words like “minimal,” “reasonable,” and “slight.” These are not precise, hard and fast judgments; rather, these are matters of prudence. If racial preferences had been limited to attaining the objectives which I describe above, and had been used in moderation, there would almost certainly be no Proposition 209 campaign in California today. Proponents of Proposition 209 will have to reject my argument, since they are fighting for “civil rights,” a principle so high and mighty that it cannot yield to plain common sense. Opponents of Proposition 209 will, of course, embrace my position, but only as a stalking-horse for maintenance of a corrupt and excessive status quo. Fortunately, the California debate does not exhaust our opportunity to deliberate on these important matters. If my argument is taken in the spirit it is offered, the question then becomes: “How can we limit racial preferences to be employed in such a reasonable fashion?” I will offer my thoughts on this matter in due course. Michael Kinsley
10:48 a.m.  Wednesday  10/9/96 

       To try to answer Herb Stein’s question about what sort of affirmative-action policies are preferable: As I said yesterday, my own preference is for policies that reduce inequality rather than–like affirmative action–redistributing it. (And many of such policies would involve UNDOING existing government programs, rather than creating new ones.) As for affirmative-action policies, it seems to me the main factor is this: Policies that benefit poor or near-poor people are better than policies that benefit prosperous people. Racial preferences for government contracts are absurd. Almost by definition, the beneficiary is either already an established businessperson or is fronting for some white guy.
       To the moderator’s other question, whether (non-reverse) discrimination in employment is rising or declining, although I’m no expert, the pattern taken over decades is surely one of decline, as Herb’s book indicates. The important question, though, isn’t the trend but the actual amount: Is there racial discrimination in employment today? To answer that, I propose a simple test: If you were starting a career in almost any field in 1996 and had the choice–for job purposes only–of doing so with a white face or a black one, which would you choose? Even taking affirmative action into account, I think the answer is obvious. Thomas Wood
11:23 a.m.  Wednesday  10/9/96 

       I am a strong advocate of affirmative action that treats individuals and groups without regard to their race, sex, or ethnicity. Affirmative action in this sense has been, and continues to be, a very good thing for the country. I am opposed only to preferential forms of affirmative action.
       I also believe that discrimination against all minorities, including blacks, has been in significant and steady decline for several decades (beginning, in fact, even before the passage of the U.S. Civil Rights Act in 1964). For the reason given by Stein, I believe that the fact that the wage gap between blacks and whites changed very little between 1980 and 1990 is to be explained by factors other than lack of progress on the discrimination front.
       It is quite true, as Stein points out, that programs to improve schools, provide jobs, and increase law enforcement in economically disadvantaged communities–programs which would disproportionately benefit blacks–would be entirely consistent with Proposition 209. I wish to emphasize particularly the importance of improving the schooling provided to America’s underclass (of whatever race), and also the importance of K-12 reform generally.
       Clearly, the most serious obstacle to continued progress toward integration and inclusiveness is the continuing disparity between the educational achievement of many minorities (including but not limited to blacks), on the one hand, and Asian-Americans and whites on the other. This is the principal obstacle, not only to continued progress toward racial diversity in higher education, but in employment as well. Work by economist June O’Neill–which has been confirmed more recently by the work of Derek Neal and William Johnson–has shown that the earnings gap between whites and blacks virtually disappears when all the relevant variables are considered, including mathematical and verbal skills.
       The black-white hourly earnings ratio is currently 82.9 percent before adjusting for any characteristics. After adjusting for geographical region, years of education, and potential work experience, the ratio between black-white earnings rises to 87.7 percent. Controlling for verbal and mathematical skills, which are of steadily increasing importance in the labor market, raises the ratio to 95.5 percent. Adding actual work experience virtually closes the gap.
       Since it is clear that persisting educational disparities are the real root of the problem, the nation needs to focus its attention on it, and I believe our symposium should do so as well. My own literature search on this question has convinced me that the educational programs and reforms that are likely to be effective in addressing the problem would not involve preferential treatment, the very essence of which is the application of different standards of merit to different racial and ethnic groups.
       I hope between now and Friday to be able to say more about this. Herb Stein
2:51 p.m.  Wednesday  10/9/96 

       Loury makes a persuasive case that there are situations in which a “minimal,” “reasonable,” or “slight” degree of racial or gender diversity or specialization is essential to the effective performance of an institution’s function and should not be banned. (The words in quotation marks are his.) Do the other panelists agree with this? The problem, as he says, is how to get the limited preferences these situations warrant without opening the door to excessive, unjustified, and offensive discrimination that, in the minds of its proponents, calls for Proposition 209. Loury promises more thoughts on that. Will Proposition 209 prevent the kind of preferences Loury considers justified? In the absence of legislation like Proposition 209–in other states–would the courts outlaw such preferences? Did the Texas Law School policy conform to the limits Loury would accept, or go beyond them?
       I will say a few words about my interest in whether discrimination has been declining. I am surprised by Wood’s numbers about the present nearly zero degree of discrimination, but my sources are now several years old and I don’t quarrel about that. If discrimination has declined, is continuing to decline and can be expected to continue to decline, the question arises whether the efforts of persons devoted to improving the condition of blacks should not be turned in a different direction than affirmative action. (As I understand it, Loury’s case for giving some preference to blacks in admission to Texas Law School is not to help blacks but to improve the education of white law students.) What is discouraging to me is that the persons who seem most determined to end discrimination against whites are not equally determined to improve the education of disadvantaged children of all races. Wood seems to be an exception. What is the general attitude in California among supporters of Proposition 209?