Note from the Editor:
This article is about the U.S. Supreme Court’s decision in Fisher v. University of Texas. As always, the Federalist Society takes no position on particular legal or public policy initiatives. Any expressions of opinion are those of the author. The Federalist Society seeks to further discussion about Fisher, affirmative action, and other civil rights issues. To this end, we offer links below to different perspectives on the case, and we invite responses from our audience. To join this debate, please email us at email@example.com.
• Fisher v. University of Texas, 133 S. Ct. 2411 (2013): http://www.law.cornell.edu/supct/cert/11-345
• Olatunde Johnson, Commentary: Fisher’s big news: No big news, SCOTUSblog (Jun. 24, 2013, 11:12 PM), http://www.scotusblog.com/2013/06/fishers-big-news-no-big-news/
• Richard Sander, Commentary on Fisher: A classic Kennedy compromise, SCOTUSblog (Jun. 24, 2013, 4:18 PM), http://www.scotusblog.com/2013/06/commentay-on-fisher-a-classic-kennedy-compromise/
• Melissa Hart, Commentary on Fisher: Everyone wins, everyone loses, SCOTUSblog (Jun. 25, 2013, 9:15 AM), http://www.scotusblog.com/2013/06/fisher-commentary-everyone-wins-everyone-loses/
There are not many dull moments in the debate about race preferences in university admissions. Nevertheless, the issuance of the recent Fisher v. University of Texas case has often been painted as one of them. “In with a bang, out with a fizzle” is the title of one account of Fisher,1 and “Fisher’s big news: No big news” is the headline of another.2 But perhaps this perennially hot debate has not cooled down after all, and Fisher is better understood as a cliffhanger—one akin to the ending of Frank Stockton’s 1882 “The Lady or the Tiger?,”3 which famously leaves the protagonist uncertain whether a beautiful woman or a starved tiger will emerge from behind the door he is about to open.
Fisher proceeds from the premise that Grutter v. Bollinger, the 2003 Supreme Court case that found the University of Michigan Law School’s admissions system of holistic review constitutional, was rightly decided.4 But the opinion calls for tight judicial scrutiny of the means used to achieve Grutter’s approved end of diversity on campus—scrutiny tighter than what the Fifth Circuit and perhaps most other well-informed observers previously understood Grutter to require. It remains to be seen how the Fifth Circuit will thread this particular needle on remand, and the case may well return to the Supreme Court because it is less than clear what exactly this heightened standard of Fisher scrutiny entails. Thus the cliffhanger. Supporters and opponents of race preferences doubtless have opposite views on which post-remand scenarios are best characterized as tigers and which as ladies. In this essay, I explain why the ultimate outcome of the Fisher litigation is more likely to look like a lady to the opponents of race preferences (and conversely a tiger to preference supporters), but that there are enough possible paths that the Fifth Circuit and eventually the Supreme Court could take to make Fisher’s legacy far from certain.
I. The Fisher Opinion
Petitioner Abigail Fisher, a Caucasian woman who was denied admission at the University of Texas (“UT”) challenged the University’s admissions policy for unlawfully discriminating against her on the basis of race. When Fisher applied, Texas accepted 81% of its students through the Top Ten Percent Plan, which requires the University of Texas to automatically accept all Texas students who graduated in the top 10% of their high school class. This Top Ten Percent Plan is widely (although not universally) believed to have been intended to prop up the numbers of black, Hispanic, and Native Americans in the UT system following the Fifth Circuit’s Hopwood v. Texas decision,5 which held outright that racial preferences at UT were unconstitutional.6 Texas determined how to fill slots remaining open after all qualified Top Ten Percent candidates were admitted by a formula that combines indices of a student’s academic index (“AI”) and personal achievement index (“PAI”).7 In 2004, following the Supreme Court’s issuance of the affirmative action decision Grutter v. Bollinger, the university revised the PAI to include prospective students’ race.8 It appears that UT’s lawyers studied Grutter closely and tried to model their program as closely as possible on the Michigan Law program upheld in Grutter. The district court decision in Fisher accordingly stated that “it would be difficult for UT to construct an admissions policy that more closely resembles the one in Grutter” and that “as long as Grutter remains good law, UT’s current admissions program remains constitutional” (but despite similarities in design, the programs differed in that UT already had many minority students via the Top Ten Percent Plan before race was explicitly taken into account).9
Fisher’s lawsuit focused on the constitutionality of this particular modification to the PAI and not on that of the Top Ten Percent Plan generally—what may have been a strategic move intended to encourage the Supreme Court to cut back racial preferences without making the more controversial declaration that they are unconstitutional full stop.10 Many knowledgeable commentators thus expected a modest opinion that would strike down only UT’s modification to the Top Ten Percent Plan and not either the Top Ten Percent plan or race preferences in university admissions generally. The decision issued June 24, 2013 did avoid these latter two questions, but it left the constitutionality of Texas’s particular modification to the PAI for another day. Fisher holds that:
Because the Fifth Circuit did not hold the university to its demanding burden of strict scrutiny articulated in Grutter and Univ. of Cal. Regents v. Bakke, its decision affirming the District Court’s grant of summary judgment to the University was incorrect. . . . Under Grutter, strict scrutiny must be applied to any admissions programs using racial categories or classifications. A court may give some deference to a university’s judgment that such diversity is essential to its educational mission, provided that diversity is not defined as mere racial balancing and there is a reasoned, principled explanation for the academic decision.11
According to Fisher, while the Fifth Circuit correctly found that Grutter calls for deference to the University’s experience and expertise regarding whether student body diversity is a compelling interest, UT also had to prove that the means it chose to attain that diversity were narrowly tailored to achieve its goal. On narrow tailoring, the University receives no deference.12 Rather than perform the appropriate examination, the Fifth Circuit incorrectly held petitioner could challenge only whether the University’s decision to use race in admissions was “made in good faith.” Fisher was thus remanded to the Fifth Circuit for closer scrutiny regarding narrow tailoring. The Supreme Court’s willingness to send Fisher back, especially in light of the similarity of UT’s policy to Michigan Law’s as discussed above, strongly suggests that more rigorous judicial scrutiny of preference programs is constitutionally required than most commentators had understood from Grutter.
Besides, Justice Kennedy—who dissented sharply in Grutter—wrote the majority opinion in Fisher, in which his fellow Grutter dissenters Scalia and Thomas concurred. Conversely, Justice Ginsburg was in the majority in Grutter, but dissented in Fisher. Only Justice Breyer was in the majority in both cases; the other Justices had not yet joined the court. All of these switched positions suggest that at least four members of the court understood Fisher as doing something quite different from Grutter.
At the same time, it is hardly clear from Fisher how much more rigorous this tightened scrutiny is. For one thing, deference on the “compelling interest” prong of strict scrutiny is potentially in tension with tougher scrutiny on the “narrow tailoring” requirement. If “compelling interest” is construed broadly, then many different types of differently structured programs will serve that interest. By contrast, the more tightly a compelling interest is defined, then the less likely it is that any particular program will be narrowly tailored to serve it.13 For example, if maintaining racial segregation for its own sake is deemed a constitutionally compelling interest, then upholding the state law at issue in Plessy v. Ferguson14 is not difficult. But if the Court decides that this compelling interest is too broadly defined and that only a narrower subset of railroad segregation laws are constitutional—e.g. that the government has a special interest in regulating cleanliness on passenger trains, and racial mixing in railway carriages is thought to pose a special threat to appropriate train hygiene—then the Court becomes obligated to wade into scientific studies about the spread of germs and whether racial integration of passenger cars actually increases the spread of communicable disease.15 Under the narrower version of compelling purpose, the Plessy statute is far less likely to be deemed constitutional.
Responses from expert commentators have been varied. Prominent critic of racial preferences Roger Clegg called Fisher “undeniably a loss for the University of Texas and for supporters of racial preferences in university admissions.”16 Law professor Ilya Somin called it a “significant victory for the opponents of affirmative action in higher education” and a “ruling that is at odds with the dominant understanding of Grutter by most lower court judges, university administrators, and legal scholars.”17 But Assistant Secretary for Civil Rights at the Department of Education (“ED”) Catherine Llamon stated at a public event that there is “no difference in content” between Fisher and Grutter. Llamon and the Department of Justice’s (“DOJ”) Jocelyn Samuels jointly issued a Dear Colleague Letter and “Questions and Answers about Fisher” that elaborates at greater length on the continuity between Fisher and Grutter and that states that earlier DOJ and ED guidance on Grutter remains in full force post-Fisher. Some of this variation is best explained as the natural human tendency to cheerlead for one’s preferred policy outcome—opponents of race preferences have played up differences with Grutter more than have preference supporters—but there are exceptions in both directions.18
II. The Future of Fisher Scrutiny
Seeing how the Fifth Circuit applies the Fisher brand of strict scrutiny—and whether and how the Supreme Court upholds that application—will go a long way to resolve this debate. There are several possible scenarios, some more likely than others, which I will discuss in turn below. Some possible outcomes will have broader implications for the use of race preferences in admissions at other universities.19
One possibility is that the Fifth Circuit, and in turn the Supreme Court, can do what some experts predicted would happen earlier in the Fisher litigation: decide that UT’s use of race as a component of PAI is not narrowly tailored because the Top Ten Percent plan already yields substantial racial diversity. As discussed above, the constitutionality of only this modification to the PAI was at issue. The use of race as part of PAI added only a tiny number of additional minority students to each entering UT class because UT already attracted many minority students through the Top Ten Percent Plan.20 Per Grutter, universities are supposed to use preferences to ensure that there is a “critical mass” of students from a minority group—enough to make sure that minority students can participate in class without feeling isolated and that students from other groups are not inclined to use racial stereotypes.21 In the last year before UT added race to the PAI, UT was already 4.5% African-American and 16.9% Hispanic. By contrast, African-Americans, Hispanics and Native Americans combined comprised only about 13% of Michigan Law’s entering class for the last three years for which data were available before Grutter.22 It is thus likely that, pre-modification, UT already had a “critical mass” of minority students. Furthermore, the modification to PAI appears to do little to raise the numbers of minorities at UT.23
Such a decision finding the modification not sufficiently narrowly tailored would curtail the reach of race preferences somewhat in Texas. But it would probably have little effect on any other race preferences challenge; most states have fewer African-Americans and Hispanics than Texas. A Top Ten Percent-like Plan would not yield significant numbers of minority students. Having a modified Top Ten Percent plan probably would not do the trick either. Under this scenario, Fisher would be in essence issuing the plaintiff a “good for this case only” ticket.24
Such a decision would also leave unresolved whether the Top Ten Percent plan and admissions preferences like it are constitutional.25 A plan similar to the Top Ten Percent plan that was adopted for race-neutral reasons—e.g. because state legislators objected to the undue weight placed upon SAT scores and found high school grades to be better predictors of achievement, or because they want rural high schools to be well-represented at flagship state schools—would not raise equal protection concerns. But there is abundant evidence that the Top Ten Percent program was largely adopted to raise the numbers of African-Americans and Hispanics at UT.26 Some defenders of the Top Ten Percent Plan-like programs claim that they are attractive alternatives to outright racial preferences because they are not overtly discriminatory. But deliberately engineering a racially discriminatory result via a facially neutral process is still race discrimination. There is zero serious argument of which I am aware that poll taxes or literacy tests in the Jim Crow South were less objectionable forms of discrimination than outright disfranchisement of black voters just because the former were on their face race-neutral. If anything, they are more problematic because it was harder for the general public to understand what the Southern municipalities using them were doing. Although I generally disagree with Justice Ginsburg on equal protection, two and one-quarter cheers for her dissent are in order for succinctly explaining these problems with the Top Ten Percent Plan.27
A less likely scenario is that the courts can hold that UT’s use of race is not narrowly tailored because UT’s admission policy does not look like a true diversity policy but instead appears designed to achieve some other constitutionally impermissible goal. The narrow tailoring inquiry exists in part to smoke out cases in which the defendant’s professed compelling purpose is not defendant’s actual purpose. If a policy is not narrowly tailored to serve the supposed compelling purpose, it is often because other purposes are actually driving defendant’s discriminatory conduct.28
This disconnect between purpose and means can explain the rationale undergirding Justice Powell’s opinion in Bakke and the contrasting results of Grutter and Gratz. The latter pair of cases, which addressed undergraduate and law admissions at the University of Michigan respectively, came out differently because the latter used procedures that were more flexible and less “mechanical.” Critics from across the political spectrum have rightly mocked the differing results of Gratz and Grutter as essentially holding racial preferences constitutional so long as universities are not too obvious about what they are really doing. But there is a more compelling reading of these cases, admittedly perhaps one that the Court did not articulate as clearly as it could have.
The crucial distinction between Grutter and Gratz is not flexibility for its own sake; instead, flexibility should be understood as part of the inquiry about whether these programs were narrowly tailored to serve the supposed compelling purpose and whether the alleged compelling purpose was defendant’s actual purpose. The inflexibility of the programs at issue in Bakke and Gratz strongly suggested that these programs were poorly tailored diversity programs and thus unlikely to be actual diversity programs. A genuine diversity program would be structured so that a university could seek out candidates who enhance student body diversity along non-racial dimensions when those varieties of diversity can do just as much or more to create a vibrant intellectual environment—as the University of Michigan Law School’s looked to be.29 It would not set a rigid quota requiring certain numbers of minority students to be admitted, as in Bakke, or give all minority students precisely the same number of points for race, as in Gratz. The design of the programs in Bakke and Gratz suggested that the admissions officers’s goal was something other than diversity, such as racial balancing for its own sake.
There is a similar disconnect between the avowed purpose (conferring the educational benefits of diversity on all UT students) of using race at UT and the way UT actually uses race. Because Texas has a larger Hispanic population than many other states, UT was already 17% Hispanic before it added race to the PAI. Having added race, it is now 20% Hispanic. Yet there are many other groups that are less well-represented on campus, who could arguably make equal contributions to diversity, that do not receive racial preferences—including Asian-Americans, who comprise 17% of UT’s student body. Giving preferences to Hispanic students under these circumstances implies that UT’s purpose is not conferring the educational benefits of diversity on its students, but rather some other purpose—such as racial balancing for its own sake, which Bakke plainly rejects as unconstitutional.30
Many universities’ use of racial preferences appear bizarrely designed to achieve the goal of diversity. As others have noted, these universities rarely if ever offer large preferences to students who would increase student body diversity along a non-racial dimension, such as evangelical Christians, political conservatives, or Communist Party members. The reason for this, as some candid defenders of preferences admit, is that many such programs are not really about increasing diversity to enhance the quality of classroom discussion. Instead, they are really about other purposes—such as redressing historic wrongs or contemporary inequality or even racial balancing for its own sake—that the Supreme Court has already rejected as unconstitutional.31 Careful scrutiny on narrow tailoring would likely smoke out these other likely illegitimate purposes of many such universities’ preferences.
Alternatively, the Fifth Circuit could decide that UT’s program is not narrowly tailored to take account of the empirical literature that race preferences are backfiring. This growing body of studies on “mismatch” indicates that some intended beneficiaries of racial preferences will learn less than they would at institutions where their entering credentials are closer to those of the median student. Most students learn best in a class in which their preparation is generally similar to that of their peers. The average eighth grader would be overwhelmed taking Advanced Placement physics surrounded by highly accomplished 17-year-olds, but she likely can pick up basic concepts about mass and acceleration in a group of other typical twelve and thirteen-year-olds.32 Minority students who receive preferential treatment in university admissions will tend to have lower academic credentials and thus to be at risk for suffering the ill effects of mismatch. For higher education officials and policymakers, the relevant question is whether race preferences at many institutions are large enough to cause such effects.
The best empirical evidence currently available strongly suggests that the answer to this question is yes. Indeed, Justice Thomas cited this research prominently in his concurring opinion in Fisher. To summarize briefly: minority students who receive preferential treatment in law school admissions are more likely to fail the bar exam (both on the first try and on subsequent efforts) than minority students who attend schools where they are better matched. Regarding undergraduates, another study finds that mismatch best explains why disproportionately few minority students become college professors.33 Three other separate studies have found that affirmative action beneficiaries are less likely to follow through with an initial plan to study science and engineering, which tend to be among the most difficult majors on most campuses, and instead study easier subjects because of credential gaps.34 It is therefore ironic that UT has argued that it needs race preferences to create greater “classroom diversity”—that is, UT is unhappy because too many of its minority students are clustered in majors with lenient grading curves like Social Work and Education and too few in more quantitative fields of study.35 Race preferences will not alleviate this problem. Instead, it is more likely that race preferences at universities more selective than UT—which have attracted minority students who would be competitive in science majors in Austin—are actually causing the problem.
It bears emphasizing that mismatched non-minority students face the same academic challenges as mismatched minority students. Peter Arcidiacono’s study on persistence in the sciences at Duke found that legacy students—who receive preferential treatment in admissions because they have family members who graduated from Duke—also disproportionately abandon scientific fields and drift toward the humanities in patterns similar to minorities. Legacy students tend to have had relatively privileged upbringings, so it is unlikely that financial disadvantage would explain their struggles in science. Nor would anti-legacy bias, lack of legacy role models, or any other argument commonly advanced to explain racial disparities in science explain legacies’ collective shift away from these fields. They are thus less likely to explain minority students’ academic decisions either.
UT has some evidence that suggests that its own affirmative action programs may be backfiring. SAT scores are a powerful predictor of freshman year performance at UT.36 These patterns are similar for students of all races, suggesting that performance differences among racial groups are best explained by credential gaps present at matriculation and not by bias or discrimination encountered on campus.37 Another table shows mean SAT scores and freshman year GPAs broken down by race. In every undergraduate college for which data were available, whites and Asians had higher average SAT scores and also received better average first year grades than Hispanics and African-Americans.38 Notably, the overall pattern holds for students in Engineering and Natural Science, suggesting that at least some minority students may be tempted to abandon these fields and switch to something easier.39 These outcomes are unfortunately entirely consistent with the mismatch literature. It is therefore likely that fewer, not more, minority students become scientists, engineers and college professors because of UT’s race preferences than would if UT practiced race-neutral admissions. If UT’s program is backfiring—as its own charts suggest that it is—then it cannot be narrowly tailored. The Fifth Circuit should at the very least require UT to take account of the mismatch literature and to justify that any educational benefits created by diversity outweigh the harms that preferences likely impose via mismatch.
Examining the mismatch research would require courts to engage with complex and occasionally difficult to understand social science research, a role that many (including me) worry judges are inadequately equipped to play. But the current case law leaves them little choice. It is true that Fisher could have come out differently and spared courts having to wade into this area. For example, it could have required the deference to universities that many understood Grutter to require, which would essentially require judges to rubber stamp many preference plans. But that approach has the potential problem that it requires courts to essentially cave to race discrimination by universities. It also stands in stark contrast to the conventional application of strict scrutiny to race discrimination, which is quite demanding. Alternatively, courts could apply the kind of highly demanding scrutiny to universities’ discrimination that is, as Gerald Gunther memorably put it, “strict in theory but fatal in fact.”40 But instead, the Court in Grutter (and in Fisher applying Grutter) chose to take an intermediate course that requires lower courts to engage with the social science, so engage with it they must.41
Finally, the Fifth Circuit could dig deeply into the question of what Grutter meant by “educational judgment.”42 Grutter requires deference to a university’s principled, reasoned “educational judgment” that student body diversity is a compelling interest. In Fisher, that phrase becomes “academic judgment.”43 At the risk of stating a tautology, an “educational judgment” or “academic judgment” ought to be fundamentally about education or academics. Yet universities often have distinctly non-academic motivations for adopting race preferences. Accreditation agencies can threaten to revoke accreditation if certain “diversity goals” are not met.44 Foundation and government grants can dry up if universities fail to attract adequate numbers of minority students.45 Student groups also often make impassioned calls for increased diversity, some of which are civil and some of which are less so.46 Judgments that a university needs race preferences to mollify legislators, funders, student groups are by definition not educational or academic judgments. Accordingly, the Fifth Circuit should scrutinize the factual record for evidence that UT’s decision here was truly academic and not motivated largely by non-academic goals.
UT was careful in designing its system of preferences. The lawyers and administrators who put together its admissions policies paid attention to Grutter and modeled their program after Michigan Law’s about as closely as humanly possible. That the Supreme Court nonetheless sent Fisher back to the Fifth Circuit on remand—in a decision joined by three justices who dissented in Grutter—is a strong signal that the Supreme Court means Fisher scrutiny to be something tighter than Grutter scrutiny as conventionally understood. But how much tighter that scrutiny is and how it will be applied remains something of a mystery, although some intriguing possibilities have presented themselves. In the meantime, we can all stare at the door and wonder what interesting creature—lady or tiger—will emerge from behind it.
*Alison Schmauch Somin is a special assistant and counsel at the U.S. Commission on Civil Rights to Gail Heriot, a law professor at the University of San Diego who is one of eight members of the federal Commission. The views expressed in this piece are not necessarily those of Ms. Heriot or of the Commission.
1 Elise Boddie, Commentary on Fisher: In with a bang, out with a fizzle, SCOTUSblog (Jun. 24, 2013, 11:05 PM), http://www.scotusblog.com/2013/06/fisher-v-university-of-texas-in-with-a-bang-out-with-a-fizzle/.
2 Olatunde Johnson, Commentary: Fisher’s big news: No big news, SCOTUSblog (Jun. 24, 2013, 11:12 PM), http://www.scotusblog.com/2013/06/fishers-big-news-no-big-news/.
3 Frank R. Stockton, The Lady, or the Tiger?, The Century (1882).
4 This premise is questionable. See infra note 10.
5 236 F.3d 256 (2000). The Hopwood majority thought that Bakke v. Regents of California, at the time the only Supreme Court case to hold that student body diversity was a compelling interest justifying racial discrimination was no longer controlling law. 236 F.3d, 274–6. When the Supreme Court upheld diversity as a compelling interest in Grutter, Hopwood was effectively overruled.
6 The Fifth Circuit opinion in Fisher states that “underrepresented minorities were its announced target and their admission a large, if not primary, purpose.”
7 The Academic Index measures a student’s academic accomplishments and takes into account SAT/ACT scores, a student’s class rank, whether the student has completed and/or exceeded UT required high-school curriculum. The PAI measures scores on two essays and accomplishments such as leadership and work experience, extracurricular activity involvement, and community service. It also looks at disadvantages that the student has overcome, such as growing up in a single parent home, speaking a language other than English at home, significant family responsibilities at home that the student has undertaken, and the family’s general socioeconomic status.
8 Fisher v. Univ. of Tex., No. 11-345, slip op. at 4 (June 24, 2013).
9 Fisher v. Univ. of Tex., 645 F. Supp. 2d 587, 612–13 (W.D. Tex. 2009). The Fifth Circuit also notes that “Appellants do not allege that UT’s admissions policy is functionally different from, or gives greater consideration to race than the policy upheld in Grutter.”
10 It is less than clear that Fisher was as determined to prevent the Court from re-visiting its Grutter holding that diversity is a compelling interest as the Court makes Fisher out to be. Justice Scalia stated in his one-paragraph concurrence that “petitioner in this case did not ask us to overrule Grutter’s holding that a ‘compelling interest’ in the educational benefits of diversity can justify racial preferences in university admissions,” citing Oral Argument Transcript at 8-9, where petitioner’s counsel stated in response to a question by Justice Breyer, “We are not trying to change the Court’s disposition of the issue in Grutter.” But Fisher’s brief is considerably more cagey. One section is titled “Grutter Should Be Clarified Or Overruled To The Extent It Can Be Read To Permit The Fifth Circuit’s Effective Abandonment Of Strict Scrutiny.” Brief for Petitioner at vi, Fisher v. University of Texas, 133 S. Ct. 2411 (2013) (No. 11-345). Counsel’s statement about what the petitioner is “trying to change” is not tantamount as a waiver of an argument explicitly made and discussed in the briefs. See Gail Heriot, Fisher v. University of Texas: The Court (Belatedly) Attempts to Invoke Reason and Principle, 2012-2013 Cato Sup. Ct. Rev. 63, 84 (2013).
11 Fisher, slip op. at 1 (summarizing the case).
12 133 S. Ct. at 2419–21.
13 See Heriot, supra note 10, at 78.
14 163 U.S. 537 (1896).
15 For the record, I think Plessy was wrongly decided, and the types of medical claims mentioned in my hypothetical are silly (although fiction and history indicate that similarly foolish claims were alas taken seriously at earlier points in American history.) This hypothetical is meant to be . . . hypothetical.
16 Roger Clegg, Commentary on Fisher: Better off than we were a year ago, SCOTUSblog (Jun. 24, 2013, 5:39 PM), http://www.scotusblog.com/2013/06/commentary-on-fisher-better-off-than-we-were-a-year-ago/.
17 Ilya Somin, Preliminary Thoughts on Fisher v. Texas, The Volokh Conspiracy (June 24, 2013), http://www.volokh.com/2013/06/24/preliminary-thoughts-on-fisher-v-university-of-texas/. Disclosure: I am married to Ilya Somin.
18 See, e.g., David Bernstein, Fisher is a Significant Loss for Opponents of Affirmative Action Preferences, The Volokh Conspiracy (June 25, 2013), http://www.volokh.com/2013/06/25/fisher-is-a-significant-loss-for-opponents-of-affirmative-action-preferences/ (Bernstein is a critic of preferences); Richard Dunham, Supreme Court sidesteps affirmative action decision in Texas ruling, tells appeals court to rehear the case, Houston Chronicle, June 24, 2013 (quoting Barbara Arnwine, president and executive director of the Lawyers Committee for Civil Rights Under Law, an organization that supports race preferences, who called Fisher “extremely disappointing” but “not a total loss”).
19 In addition to those discussed infra, it is possible that Fisher could have a very limited impact because Ms. Fisher is found to lack standing to bring this case. UT claims that Fisher lacks standing because she would not have been admitted even under a race-neutral system and is therefore not entitled to the relief she seeks—the refund of her application fee.
Fisher is in part an unjust enrichment case. The proper remedy is thus restitution. Put differently, Fisher wants her money back—an appropriate remedy when proof of damages is difficult and yet a legal wrong has been done. Suppose Fisher had entered a lottery, but the state had tossed out her entry because of her race. See Brief for Todd Gaziano, Gail Heriot, and Peter Kirsanow as Amici Curiae Supporting Petitioner at 32, Fisher v. Univ. of Tex., No. 09-05822 (5th Cir. Oct. 10, 2013). She couldn’t prove she would have won, but she would be clearly entitled to her money back. See generally Douglas Laycock, Restoring Restitution to the Canon, 110 Mich. L. Rev. 929 (2012).
If a majority of the Supreme Court thought Ms. Fisher lacked standing, they likely would have decided the case on those grounds on the last go-round. Some have speculated that UT has only raised standing as a delay tactic to ensure Fisher only returns to the Supreme Court after new justices with different views on affirmative action have been appointed.
Such a result might indeed limit the importance or future reach of the Supreme Court’s decision in Fisher. For the reasons discussed above, however, I think it unlikely.
20 Brief for Petitioner at 9, Fisher v. Univ. of Tex., 133 S. Ct. 2411 (2013) (No. 11-345). It is impossible to say exactly how many additional minority students are admitted because of the addition of race to the PAI. Out of 6,332 students admitted to UT’s freshman class in 2008, most were admitted through the Top Ten Percent plan; only 1,208 were admitted at a stage at which PAI scores were taken into account. Of this group of 1,208, only 58 students were African-American and 158 Hispanic. But it is impossible to know how many of them were admitted because of the explicit consideration of race and how many would have been admitted without regard for their race.
21 Grutter v. Bollinger, 539 U.S. 306, 330 (2003).
22 Id. at 385 (Kennedy, J., dissenting).
23 Brief for Petitioner at 9, Fisher v. Univ. of Tex., 133 S. Ct. 2411 (2013) (No. 11-345). It is impossible to say exactly how many additional minority students are admitted because of the addition of race to the PAI. Out of 6,332 students admitted to UT’s freshman class in 2008, most were admitted through the Top Ten Percent plan; only 1,208 were admitted at a stage at which PAI scores were taken into account. Of this group of 1,208, only 58 students were African-American and 158 Hispanic. But it is impossible to know how many of them were admitted because of the explicit consideration of race. Some of these students have distinguished themselves in other ways—e.g. by excellent SAT scores, essays or leadership—and would have been admitted even without the additional boost supplied by consideration of race.
24 See Gail Heriot, supra note 10, at 78.
25 There is abundant evidence that the Top Ten Percent program was largely adopted to raise the numbers of African-Americans and Hispanics at UT. Some defenders of the Top Ten Percent Plan and programs like it claim that they are more attractive alternatives to outright racial preferences because they hide their racial purpose. But this is akin to arguing that facially race neutral but transparently racially motivated practices often adopted to disfranchise African-Americans in the Jim Crow South, such as like literacy tests and grandfather clauses, are somehow less discriminatory. They are not. Intent matters.
26 See Fisher, 133 S. Ct at 2433 (dissent of Justice Ginsburg summarizing applicable legislative history from the Texas Legislature.)
27 Justice Ginsburg begins her four page dissent with the statement, “The University of Texas at Austin (University) is candid about what it is endeavoring to do.” She notes that “Petitioner urges that Texas’s Top Ten Percent Law and race-blind holistic review of each application achieve significant diversity, so the University must be content with those alternatives. I have said before and reiterate here that only an ostrich could regard the supposedly neutral alternatives as race unconscious. (Internal citations to Gratz and Grutter omitted). As Justice Souter observed, the vaunted alternatives suffer from the disadvantage of deliberate obfuscation.” She proceeds to cite legislative history from the Texas state legislature indicating that ensuring certain levels of minority enrollment was a key motive for the Top Ten Percent Plan: “Many regions of the state, school districts, and high schools in Texas are still predominantly composed of people from a certain racial or ethnic group. Because of the persistence of this segregation, admitting the top 10 percent of all high schools would provide a diverse population and ensure that a large well-qualified pool of minority students was admitted to Texas universities.”
Perhaps the worst of all possible worlds for opponents of racial preferences is one in which universities widely practice racial preferences, but do so in a covert manner that evades both judicial and political scrutiny. Recent polls suggest that preferences are unpopular. Initiatives banning them fare well at the polls. Yet it is difficult for voters to acquire information about a vast and complex system of government that seems to get more vast and complex every day. The more obscure universities are about the real motives of their admissions policies, the less likely it is that voters can monitor such policies effectively.
28 “The purpose of the narrow tailoring requirement is to ensure that “the means chosen ‘fit’. . . the compelling goal so closely that there is little or no possibility that the motive for the classification was illegitimate racial prejudice or stereotype.” Richmond v. J. A. Croson Co., 488 U.S. 469, 493 (1989) (cited in Grutter v. Bollinger, 539 U.S. 306, 333 (2003)).
29 Though the University of Michigan Law School’s admissions policy looked more like a genuine diversity program than the undergraduate policy, it probably was not one. As Justice Rehnquist discussed in dissent, there are many fewer Native Americans at Michigan Law than there are Hispanic and African-American students. If the university’s goal was really ensuring that all racial groups had adequate critical mass necessary to confer the educational benefits of diversity on all students, then Michigan should either give larger preferences to Native Americans or smaller preferences to African-Americans and Hispanics. Instead, Michigan’s real goal appeared to be to create a student body whose racial demographics mirrored those of its applicant pool—a goal that, unlike diversity, was not already authorized as constitutional by Bakke. Grutter, 539 U.S. at 382–84.
30 Regents of the Univ. of California v. Bakke, 438 U.S. 265, 330 (1978).
31 See generally Brian Fitzpatrick, The Diversity Lie, 27 Harv. J.L. & Pub. Pol’y 385 (2003) (pointing out incompatibilities between the theory of student body diversity and how preferences are practiced in admissions).
The year after Bakke, Columbia law professor Kent Greenawalt, wrote in The Unresolved Problems of Reverse Discrimination, 67 Cal. L. Rev. 87, 122 (1979) (“I have yet to find a professional academic who believes the primary motivation for preferential admission has been to promote diversity in the student body for the better education of all the students . . . . .”).
Similarly, Harvard law professor Alan Dershowitz wrote: “The raison d’être for race-specific affirmative action programs has simply never been diversity for the sake of education. The checkered history of “diversity’” demonstrates that it was designed largely as a cover to achieve other legally, morally, and politically controversial goals. In recent years, it has been invoked—especially by professional schools—as a clever post facto justification for increasing the number of minority group students in the student body.” Alan Dershowitz, Affirmative Action and the Harvard College Diversity-Discretion Model: Paradigm or Pretext, 1 Cardozo L. Rev. 379, 407 (1979).
More recently, Harvard law professor Randall Kennedy, an affirmative action proponent, stated: “Let’s be honest: Many who defend affirmative action for the sake of “diversity” are actually motivated by a concern that is considerably more compelling. They are not so much animated by a commitment to what is, after all, only a contingent, pedagogical hypothesis. Rather, they are animated by a commitment to social justice. They would rightly defend affirmative action even if social science demonstrated uncontrovertibly that diversity (or its absence) has no effect (or even a negative effect) on the learning environment.” Randall Kennedy, Affirmative Reaction, Am. Prospect, Mar. 2003, available at http://prospect.org/article/affirmative-reaction.
See also Peter H. Schuck, Affirmative Action: Past, Present, and Future, 20 Yale L. & Pol’y Rev. 1, 34 (2002) (“[M]any of affirmative action’s more forthright defenders readily concede that diversity is merely the current rationale of convenience for a policy that they prefer to justify on other grounds.”); Jed Rubenfeld, Affirmative Action, 107 Yale L.J. 427, 471–72 (1997) (“The purpose of affirmative action is to bring into our nation’s institutions more blacks, more Hispanics, more Native Americans, more women, sometimes more Asians, and so on—period. Pleading diversity of back- grounds merely invites heightened scrutiny into the true objectives behind affirmative action.”); Owen Fiss, Affirmative Action as a Strategy of Justice, 17 Philosophy & Pub. Pol’y 37 (1997) (“[T]wo defenses of affirmative action—diversity and compensatory justice—emerged in the fierce struggles of the 1970s and are standard today, but I see them as simply rationalizations created to appeal to the broadest constituency. . . . In my opinion, affirmative action should been seen as a means that seeks to eradicate caste structure by altering the social standing of our country’s most subordinated group”); Daniel Golden, Some Backers of Racial Preferences Take Diversity Rationale Further, Wall St. J., June 14, 2003 (quoting former UT law school professor Samuel Issacharoff: “‘The commitment to diversity is not real. None of these universities has an affirmative-action program for Christian fundamentalists, Muslims, orthodox Jews, or any other group that has a distinct viewpoint.’”).
32 A loss of learning can also occur from positive mismatch—a situation in which a student learns less than her peers because her prior preparation is stronger than theirs.
33 For a more detailed summary of the mismatch research than I am able to provide in this brief essay, see Richard Sander & Stuart Taylor, Mismatch: How Affirmative Action Hurts the Students It is Intended to Help (2012). For a shorter round-up of much of the same evidence, see Gail Heriot, The Sad Irony of Affirmative Action, 14 Nat’l Affairs 78 (2013).
34 To be clear: my claim here is not that humanities subjects are intrinsically less worth studying, or that any individual student errs by choosing to study them. I myself was an undergraduate humanities major and do not regret my own choices. My claim is, rather, that at most universities, mathematics and science grading tends to be more difficult. These grading patterns could change, but that does not appear likely in the short term. Students who come into a university with lower credentials than their peers will therefore be more likely to have trouble with them.
35 Fisher v. Univ. of Tex., 631 F.3d 213, 240 (5th Cir. 2011).
36 The University of Texas Office of Admissions, Implementation and Result of the Texas Automatic Admissions Law (H.B. 588) at the University of Texas at Austin, Demographic Analysis of Entering Freshmen Fall 2008, Academic Performance of Top 10% and Non-Top 10% Students,” available at http://www.utexas.edu/student/admissions/research/HB588-Report11.pdf; Table 4.
37 Id. at Tables 6a through 6d, “Freshman Year Performance by SAT Score Range” ( showing average GPA for white graduates of Texas high schools falling in SAT I score bands including “Less than 900,” “1000-1090,” “1100-1190,” and so on up through 1500-1600); 6b(showing average GPA for SAT score bands for African-American graduates of Texas high schools); 6c (same for Asian-American graduates of Texas high schools) and 6d(same for Hispanic students.)
38 Id. at Table 7.
40 Gerald Gunther, The Supreme Court, 1971 Term—Foreword: In Search of Evolving Doctrine on a Changing Court: A Model for a Newer Equal Protection, 86 Harv. L. Rev. 1 (1972).
41 See generally Brief for Todd Gaziano, Gail Heriot, and Peter Kirsanow as Amici Curiae Supporting Petitioner 29–32, Fisher v. Univ. of Tex, No. 09-05822 (5th Cir. Oct. 10, 2013).
42 Grutter v. Bollinger, 539 U.S. 306, 328 (2003).
43 Fisher v. Univ. of Tex., 133 S. Ct. 2411, 2419 (2013).
44 For example, some universities practice discrimination in admissions because their federally-appointed accrediting authorities require it. See Brief for California Association of Scholars et al. as Amici Curiae Supporting Petitioner, Fisher v. Univ. of Tex., 133 S. Ct. 2411, 2419 (2013) (No. 11-345) (arguing that admissions policies adopted in whole or in part to appease accreditors or funding sources are not protected by Grutter-deference); Margaret Jackson, University of Colorado Medical School Heals Diversity Gap, Denver Post, Apr. 21, 2012 (“The university has made a concerted effort to improve diversity among its students since its accrediting body—the Liaison Committee on Medical Education—cited the school for ‘noncompliance’ in 2010, when just 106 of 614 students were minorities.”); Gail Heriot, The ABA’s “Diversity” Diktat, Wall St. J., Apr. 28, 2008 (chronicling the ABA’s demands for race-preferential admissions for the sometimes-resistant law schools it accredits); James T. Hammond, Charleston School of Law: New School Fails to Win Accreditation So Students Can Take Bar, The State (Columbia, S.C.), Jul. 12, 2006.
At the press conference on the Fisher guidance discussed supra, an audience questioner asked what Fisher meant for the future of diversity requirements for accreditation. Assistant Secretary Llamon replied that accreditors “certainly have a leadership role in this that we in this administration think that you shouldn’t walk away from.” (Quotation is from my notes from the event, which are on file with me).
45 See, e.g., Daryl G. Smith et al., Claremont Graduate University, Building Capacity: A Study of the Impact of the James Irvine Foundation Campus Diversity Initiative (May 2006) (discussing a $29 million effort to assist California colleges and universities with strategically improving campus diversity); Commitment to Diversity Leads to Gift, Ohio State University Moritz College of Law (Apr. 5, 2012), http://moritzlaw.osu.edu/briefing-room/alumni/commitment-to-diversity-leads-to-gift/ (announcing gift by alumnus to Ohio State University’s law school).
46 In 2011, for example, at the University of Wisconsin, a student mob, egged on by the University’s Vice Provost for Diversity and Climate, overpowered hotel staff, knocking some to the floor, to interrupt a press conference at which the speaker was critical of race-based admissions policies. See Peter Wood, Mobbing for Preferences, Chron. Higher Educ. (Sept. 22, 2011).