Fisher v. Texas and the Texas Ten Percent Plan Revisited

Joshua Thompson of the Pacific Legal Foundation has an interesting response to my post arguing that a victory for the plaintiffs in the Fisher v. Texas case could turn out to be a pyrrhic victory for opponents of racial preferences, if it is coupled with an endorsement of the Texas ten percent plan and other “facially neutral” efforts to extend racial preferences to certain minority groups without doing so openly.

In my last post and an earlier 2006 critique of the ten percent plan, I argued that measures like the ten percent plan are in many ways worse than traditional explicit affirmative action policies because they are less transparent and create more perverse incentives.

Thompson agrees with several of my points, but also takes issue with others. Here, I focus mostly on the areas of disagreement.

Thompson first argues that the ten percent plan actually is transparent because “finishing in the top ten percent of your high school class is an objective (race-neutral) criterion that can be judged on its face, and permits no obfuscation by university admissions officials.” It is true that the ten percent plan is “objective” in the sense that finishing in the top ten percent in your high school class is a verifiable fact. What is not transparent, however, is the racial motivation behind the adoption of the ten percent plan in the first place. With traditional affirmative action, the racial goal is clear and unambiguous. With the ten percent plan, administrators can and do pretend that they aren’t really doing anything racially motivated at all. In addition, it is difficult for the public to judge how much academic merit is being sacrificed by preferring people who graduated in the top ten percent at weak high schools over those who placed at lower percentiles at stronger ones.

Thompson also suggests that traditional affirmative action programs may sacrifice more academic merit than the ten percent plan does because “Preferences, definitionally, let in those who are less academically qualified. Under a preference regime, someone is getting in because of their skin color. Under the Top Ten Percent Law someone is getting in because of their grades.” Actually, under traditional racial preferences, skin color is only one of several reasons why affirmative action beneficiaries get in. They also must have at least some academic qualifications, albeit lower than those required of other applicants. More importantly, this ignores the reason why the ten percent plan has a much greater negative impact on academic merit than traditional affirmative action: the former affects far more applicants. As I explained in my 2006 post:

[T]he ten percent plan affects a great many more admissions decisions than even the most rigid old-style affirmative action systems do. Rarely, if ever, do traditional affirmative action plans determine the admission of more than 15-20% of a school’s student body. By contrast, at the University of Texas at Austin, over 70% of the student body was admitted under the ten percent plan. While some of these students would surely have gotten in anyway, it is highly likely that the ten percent plan leads to much larger sacrifices of academic merit than do racial preferences similar to those used at most other academic institutions.

Finally, Thompson partially rejects my point that facially neutral means of preferring some racial groups over others are just as unconstitutional as racially explicit ones. In response to my historic examples of facially neutral poll taxes and literacy tests that were used to exclude black voters in the Jim Crow south, he notes that “poll taxes and literacy tests were never ruled per se unconstitutional.” This is true, but irrelevant. They are unconstitutional in cases where they were enacted for the deliberate purpose of excluding black voters (or voters of any other particular racial or ethnic group). Similarly, the ten percent plan would be constitutional if it were enacted for nonracial reasons. But if enacted for the purpose of advantaging African-American and Hispanic applicants over members of other racial and ethnic groups, it is unconstitutional. Ultimately, however, I am not sure that we disagree very much on this aspect of the issue, since Thompson also states that the ten percent plan may be unconstitutional because “[i]t was undoubtedly passed in order to increase racial diversity throughout the University of Texas.”

Where we continue to differ is over the question of whether a world in which racial preferences are pursued through “facially neutral,” nontransparent means is preferable to one where such preferences are in the open for all to see, and inflict less social harm. I fear that if the Supreme Court strikes down the present University of Texas affirmative action plan, but endorses the ten percent plan and other similar programs, we will end up with a worse situation than one where explicit affirmative action plans remain legal.

UPDATE: Thompson responds further here. He continues to argue that the ten percent plan is transparent because “[e]veryone knows that the Top Ten Percent Law was adopted with the express intent to increase racial diversity at the University of Texas.” Everyone who follows these issues closely does indeed know that. But much of the public is rationally ignorant about politics and is unlikely to know that. Moreover, as time passes since the adoption of the plan, public memory of the reasons for its adoption is likely to fade. By contrast, the racial reasons for traditional affirmative action are more clear. Even people who know little about politics can easily grasp that an admissions policy that openly factors in race is a racial preference.

Thompson also argues that the ten percent plan is more defensible because administrators’ hands are tied under it. To my mind, it is irrelevant whether a racially motivated admissions plan was adopted by the state legislature, university administrators, or some combination of the two.

In the second part of his response, Thompson claims that race is often determinative in admissions decisions under affirmative action. I never denied that this is true. But it’s also true under the ten percent plan. If not for racial considerations, there would be no such plan, and thus there are numerous University of Texas students who have been admitted or rejected because this policy was enacted for racial reasons. Race was no less determinative for them than for those admitted or rejected because of traditional affirmative action.

Finally, and most importantly, Thompson takes issue with the idea that “facially neutral” policies enacted for racial reasons are unconstitutional. There is actually a good deal of Supreme Court precedent on this point (the leading modern case is Village of Arlington Heights v. Metropolitan Housing Development Corp.). To oversimplify slightly, these cases hold that a racially motivated policy is unconstitutional unless the government can show that it would have adopted it even absent the illicit racial motivation. In many cases, of course, it is very hard to prove that there was any racial motive in the first place. In the case of the ten percent plan, however, the racial motive is very clear, and Thompson does not deny it. More to the point, I think these Supreme Court precedents are roughly correct. The purpose of the Equal Protection Clause is to forbid racial favoritism by state government, regardless of whether that favoritism takes the form of racially explicit ones or “facially neutral” ones where the real purpose of the policy is hidden.

Thompson argue that “The question of legislative intent cannot be the only factor that determines a law’s constitutionality. If so, we would have to scrutinize the intent of the legislature (or bureaucrat) when passing any legislation (or regulation) whatsoever.” Of course it is not the only factor. Sometimes policies can be unconstitutional for reasons having nothing to do with motive. However, racial motivation is a relevant factor, and in many cases a decisive one. Recognizing that does not require us to scrutinize the intent behind every law. In many cases, scrutiny is not needed because there is no evidence of any improper motive.

Thompson asks “under Professor Somin’s argument, how could you ever use race-neutral measures if they are motivated by a racial purpose?” By and large, you couldn’t, except perhaps if you could prove that the government would have adopted the same policy even without a racial intent or if the policy in question would be constitutional even if the use of race were explicit. This is a feature of my analysis, not a bug. “Race-neutral” measures motivated by a racial purpose are just another form of racial discrimination and should be treated as such.

In the last part of his post, Thompson tries to distinguish literacy tests and poll taxes from the ten percent plan, on the grounds that the former were unequally applied as well as adopted for the purpose of disadvantaging blacks (e.g. – blacks were often forced to take more difficult tests than whites). But surely literacy tests are unconstitutional even when applied equally if their purpose is to disadvantage black voters (e.g. – because the legislature knows that more blacks are likely to fail the test than whites and adopted the policy for that reason). Such a law would surely be ruled unconstitutional by the courts. Thompson contends that “Only those laws that do, in fact, discriminate are unconstitutional.” True enough. But a law can “discriminate” in fact even if it does not explicitly mention race.