Fisher v. Univ. of Tex. at Austin

Full title: ABIGAIL NOEL FISHER, Plaintiff - Appellant v. UNIVERSITY OF TEXAS AT…

Court: UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT

Date published: Jul 15, 2014

758 F.3d 633 (5th Cir. 2014)

Opinion

No. 09–50822.

2014-07-15

Abigail Noel FISHER, Plaintiff–Appellant v. UNIVERSITY OF TEXAS AT AUSTIN; David B. Pryor, Executive Vice Chancellor for Academic Affairs in His Official Capacity; William Powers, Jr., President of the University of Texas at Austin in His Official Capacity; Board of Regents of the University of Texas System; R. Steven Hicks, as Member of the Board of Regents in His Official Capacity; William Eugene Powell, as Member of the Board of Regents in His Official Capacity; James R. Huffines, as Member of the Board of Regents in His Official Capacity; Janiece Longoria, as Member of the Board of Regents in Her Official Capacity; Colleen McHugh, as Member of the Board of Regents in Her Official Capacity; Robert L. Stillwell, as Member of the Board of Regents in His Official Capacity; James D. Dannenbaum, as Member of the Board of Regents in His Official Capacity; Paul Foster, as Member of the Board of Regents in His Official Capacity; Printice L. Gary, as Member of the Board of Regents in His Official Capacity; Kedra Ishop, Vice Provost and Director of Undergraduate Admissions in Her Official Capacity; Francisco G. Cigarroa, M.D., Interim Chancellor of the University of Texas System in His Official Capacity, for Defendants–Appellees.

Bert Walter Rein, William Consovoy, Esq., Claire Evans, Thomas R. McCarthy, Esq., Wiley Rein, L.L.P., Washington, DC, Paul M. Terrill, III, Esq., Terrill Firm, Austin, TX, for Plaintiff–Appellant. Gregory George Garre, Maureen E. Mahoney, Esq., Latham & Watkins, L.L.P., Lori Ann Alvino McGill, Quinn, Emanuel, Urquhart & Sullivan, L.L.P., Washington, DC, James C. Ho, Esq., Gibson, Dunn & Crutcher, L.L.P., Dallas, TX, for Defendants–Appellees.

PATRICK E. HIGGINBOTHAM

Bert Walter Rein, William Consovoy, Esq., Claire Evans, Thomas R. McCarthy, Esq., Wiley Rein, L.L.P., Washington, DC, Paul M. Terrill, III, Esq., Terrill Firm, Austin, TX, for Plaintiff–Appellant. Gregory George Garre, Maureen E. Mahoney, Esq., Latham & Watkins, L.L.P., Lori Ann Alvino McGill, Quinn, Emanuel, Urquhart & Sullivan, L.L.P., Washington, DC, James C. Ho, Esq., Gibson, Dunn & Crutcher, L.L.P., Dallas, TX, for Defendants–Appellees.
Gordon Morris Fauth, Jr., Litigation Law Group, Alameda, CA, Ashley C. Keller, Bartlit, Beck, Herman, Palenchar & Scott, L.L.P., Chicago, IL, James Scott Detamore, Esq., Mountain States Legal Foundation, Lakewood, CO, Timothy Mason Sandefur, Joshua Paul Thompson, Pacific Legal Foundation, Sacramento, CA, Tovah R. Calderon, Diana Katherine Flynn, Sasha M. Samberg–Champion, Esq., Linda Frances Thome, U.S. Department of Justice, Vincent Adrian Eng, Asian American Justice Center, Sri Srinivasan, Jonathan D. Hacker, O'Melveny & Myers, L.L.P., Martin Robert Michaelson, David McMichael Ginn, Hogan Lovells US, L.L.P., Andrew M. Grossman, Baker & Hostetler, L.L.P., David Charles Frederick, Esq., Kellogg, Huber, Hansen, Todd, Evans & Figel, P.L.L.C., Washington, DC, Deborah Nicole Archer, Director, New York Law School, Leticia V. Smith–Evans, Deuel Ross, Legal Defense & Educational Fund, Inc., New York, NY, Julie Ann Su, Asian Pacific American Legal Center, Los Angeles, CA, David G. Hinojosa, Nina Perales, Esq., Mexican–American Legal Defense & Educational Fund, San Antonio, TX, Alison Elisabeth Schmauch, Counsel, U.S. Commission on Civil Rights, Arlington, VA, George Allan Van Fleet, Nicholas Gerhart Grimmer, Attorney, McDermott Will & Emery, L.L.P., Houston, TX, Charles Ernest Baruch, Law Office of Chad Baruch, Rowlett, TX, Robert Earl Toone, Jr., Foley Hoag, L.L.P., Boston, MA, for Amicus Curiae.

Appeal from the United States District Court for the Western District of Texas.
Before KING, HIGGINBOTHAM, and GARZA, Circuit Judges.

ON REMAND FROM THE SUPREME COURT OF THE UNITED STATES

PATRICK E. HIGGINBOTHAM, Circuit Judge:

Abigail Fisher brought this action against the University of Texas at Austin,¹ alleging that the University's race-conscious admissions program violated the Fourteenth Amendment. The district court granted summary judgment to UT Austin and we affirmed. The Supreme Court vacated and remanded, holding that this Court and the district court reviewed UT Austin's means to the end of a diverse student body with undue deference; that we must give a more exacting scrutiny to UT Austin's efforts to achieve diversity. With the benefit of additional briefing, oral argument, and the ordered exacting scrutiny, we affirm the district court's grant of summary judgment.

1 Along with Fisher, Rachel Michalewicz was originally a plaintiff against UT Austin; Michalewicz is no longer a party to this action.

I

A

Fisher applied to UT Austin for admission to the entering class of fall 2008. ² Although a Texas resident, she did not graduate in the top ten percent of her class. She therefore did not qualify for automatic admission under the Top Ten Percent Plan, which that year took 81% of the seats available for Texas residents.³ Instead, she was considered under the holistic review program,⁴ which looks past class rank to evaluate each applicant as an individual based on his or her achievements and experiences, and so became one of 17,131 applicants ⁵ for the remaining 1,216 seats ⁶ for Texas residents.

2 Defs.' Cross–Mot. Summ. J., Ex. Tab 7 to App., Ishop Aff. at ¶ 2, Fisher v. Univ. of Tex. at Austin, 645 F.Supp.2d 587 (W.D.Tex.2009) ( No. 08–263), ECF No. 96 [hereinafter Ishop Aff.].

3 Office of Admissions, Univ., of Tex. at Austin, Implementation and Results of the Texas Automatic Admissions Law (HB 588) at the University of Texas at Austin: Demographic Analysis of Entering Freshmen Fall 2008 and Academic Performance of Top 10% and Non–Top 10% Students Academic Years 2003–2007 (Report 11), at 7 tbl.1a (Oct. 28, 2008) [hereinafter 2008 Top Ten Percent Report], Defs.' Cross–Mot. Summ. J., Ex. Tab 8 to App., Lavergne Aff., Ex. C, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 96, available at http:// www. utexas. edu/ student/ admissions/ research/ HB 588- Report 11. pdf.

4 Ishop Aff. ¶ 16, ECF No. 96. Additionally, Fisher did not apply for any academic programs with special application processes, such as the Plan II Honors program or a Fine Arts program.

5 Id. ¶ 13.

6 2008 Top Ten Percent Report at 9 tbl.2b; id. at 8 tbl.2. Table 2 shows 8,984 Top Ten Percent students were admitted in 2008. The UT Associate Director of Admissions reported that 10,200 admissions slots are available for Texas residents. Ishop Aff. ¶ 12, ECF No. 96.

UT Austin denied Fisher admission. Kedra B. Ishop, the Associate Director of Admissions at the time of Fisher's application,⁷ explained that “[g]iven the lack of space available in the fall freshman class due to the Top 10% Plan, ... based on [her] high school class rank and test scores,” Fisher could not “have gained admission through the fall review process.” ⁸ As Ishop explained, any applicant who was not offered admission either through the Top Ten Percent Law or through an exceptionally high Academic Index (“AI”) score is evaluated through the holistic review process. ⁹ The AI is calculated based on an applicant's standardized test scores, class rank, and high school coursework.¹⁰ Holistic review considers applicants' AI scores and Personal Achievement Index (“PAI”) scores. The PAI is calculated from (i) the weighted average score received for each of two required essays and (ii) a personal achievement score based on a holistic review of the entire application, with slightly more weight being placed on the latter.¹¹ In calculating the personal achievement score, the staff member conducts a holistic review of the contents of the applicant's entire file, including demonstrated leadership qualities, extracurricular activities, honors and awards, essays, work experience, community service, and special circumstances, such as the applicant's socioeconomic status, family composition, special family responsibilities, the socioeconomic status of the applicant's high school, and race.¹² No numerical value is ever assigned to any of the components of personal achievement scores, and because race is a factor considered in the unique context of each applicant's entire experience, it may be a beneficial factor for a minority or a non-minority student.¹³

7 Id.

8 Id. ¶ 18.

9 Id. ¶ 4.

10 Id. ¶ 3. The AI score is generated by adding the predicted grade point average (“PGPA”) and the curriculum-based bonus points (“units plus”). Id. The PGPA is calculated using an applicant's SAT or ACT scores and class rank. Id. A units plus bonus of 0.1 points is added to the PGPA if the applicant took more than UT Austin's minimum high school coursework requirements in at least two of three designated subject areas. Id.

11 Id. ¶ 5. The PAI is calculated as follows: PAI = ((((essay score 1 + essay score 2)/2)*3) + ((personal achievement score)*4))/7. Defs.' Cross–Mot. Summ. J., Ex. Tab 3 to App., Lavergne Dep. at 57:1117, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 96 [hereinafter Lavergne Dep.].

12 Defs.' Cross–Mot. Summ. J., Ex. Tab 1 to App., Bremen Dep. at 16:15–17:13, 18:5–19:14, 44:144:6, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 96 [hereinafter Bremen Dep.]; Ishop Aff. ¶ 5, ECF No. 96; Defs.' Cross–Mot. Summ. J., Ex. Tab 2 to App., Ishop Dep. at 22:1320, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 96 [hereinafter Ishop Dep.].

13 Ishop Aff. ¶ 5, ECF No. 96.

To admit applicants through this holistic review, the admissions office generates an initial AI/PAI matrix for each academic program, wherein applicants are placed into groups that share the same combination of AI and PAI scores.¹⁴ School liaisons then draw stair-step lines along this matrix, selecting groups of students on the basis of their combined AI and PAI scores. This process is repeated until each program admits a sufficient number of students.

14 Id. ¶ 14. The AI scores are placed on one axis and the PAI scores are placed on the other axis. Students are then grouped based on their combination of AI and PAI scores.

Fisher's AI scores were too low for admission to her preferred academic programs at UT Austin; Fisher had a Liberal Arts AI of 3.1 and a Business AI of 3.1.¹⁵ And, because nearly all the seats in the undeclared major program in Liberal Arts were filled with Top Ten Percent students, all holistic review applicants “were only eligible for Summer Freshman Class or CAP [Coordinated Admissions Program] admission, unless their AI exceeded 3.5.” ¹⁶ Accordingly, even if she had received a perfect PAI score of 6, she could not have received an offer of admission to the Fall 2008 freshman class.¹⁷ If she had been a minority the result would have been the same.

15 Id. ¶ 18.

16 Id.

17 Id. At the preliminary injunction stage, UT Austin suggested that it was unable to determine whether Fisher (or Michalewicz) would have been admitted without re running the entire admissions process. Opp. Mot. Prelim. Injunction at 12, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 42. Regardless, it became clear in the summary judgment record that whether Fisher would have been admitted even if she had a perfect PAI score presented no genuine issue of fact. She would not have been admitted. The same was true for Michalewicz, then a co-plaintiff.

B

This reality together with factual developments since summary judgment call into question whether Fisher has standing.¹⁸ UT Austin argues that Fisher lacks standing because (i) she graduated from another university in May 2012, thus rendering her claims for injunctive and declaratory relief moot, ¹⁹ and (ii) there is no causal relationship between any use of race in the decision to deny Fisher admission and the $100 application fee—a non-refundable expense faced by all applicants that puts at issue whether Fisher suffered monetary injury.²⁰

18 Plaintiffs “must show that (1) they have suffered an injury in fact, (2) a causal connection exists between the injury and challenged conduct, and (3) a favorable decision is likely to redress the injury.” Adar v. Smith, 639 F.3d 146, 150 (5th Cir.2011) (en banc) (citing Lujan v. Defenders of Wildlife, 504 U.S. 555, 560–61, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992)).

19 Appellees' Statement Concerning Further Proceedings on Remand at 5.

20 As we will explain, Fisher's odds of admission were affected by the Top Ten Percent Plan, which filled all but around 1,200 seats of the incoming class. Competition drove the automatic rejection up to a 3.5 AI score.

Two competing and axiomatic principles govern the resolution of this question. First, jurisdiction must exist at every stage of litigation. A litigant “generally may raise a court's lack of subject-matter jurisdiction at any time in the same civil action, even initially at the highest appellate instance.” ²¹ Even if “defendants failed to challenge jurisdiction at a prior stage of the litigation, they are not prohibited from raising it later.” ²² Indeed, the “independent establishment of subject-matter jurisdiction is so important that [even] a party ostensibly invoking federal jurisdiction may later challenge it as a means of avoiding adverse results on the merits.” ²³

21 Grupo Dataflux v. Atlas Global Grp., L.P., 541 U.S. 567, 576, 124 S.Ct. 1920, 158 L.Ed.2d 866 (2004) (citations omitted).

22 Arena v. Graybar Elec. Co., Inc., 669 F.3d 214, 223 (5th Cir.2012).

23 Id. (quoting 13 Charles Alan Wright, et al., Fed. Practice & Procedure § 3522 at 12223 (3d ed.2008)).

Second, the “mandate rule,” a corollary of the law of the case doctrine, “compels compliance on remand with the dictates of a superior court and forecloses relitigation of issues expressly or impliedly decided by the appellate court.” ²⁴ The Supreme Court, like all Article III courts, had its own independent obligation to confirm jurisdiction, and where the lower federal court “lack[ed] jurisdiction, [the Supreme Court has] jurisdiction on appeal, not of the merits, but merely for the purpose of correcting the error of the lower court in entertaining the suit.” ²⁵

24 United States v. Lee, 358 F.3d 315, 321 (5th Cir.2004) (citing United States v. Bell, 988 F.2d 247, 251 (1st Cir.1993)).

25 Steel Co. v. Citizens for a Better Env't, 523 U.S. 83, 95, 118 S.Ct. 1003, 140 L.Ed.2d 210 (1998).

UT Austin's standing arguments carry force,²⁶ but in our view the actions of the Supreme Court do not allow our reconsideration. The Supreme Court did not address the issue of standing, although it was squarely presented to it.²⁷ Rather, it remanded the case for a decision on the merits, having reaffirmed Justice Powell's opinion for the Court in Regents of the University of California v. Bakke ²⁸ as read by the Court in Grutter v. Bollinger. ²⁹ It affirmed all of this Court's decision except its application of strict scrutiny. The parties have identified no changes in jurisdictional facts occurring since briefing in the Supreme Court. Fisher's standing is limited to challenging the injury she alleges she suffered—the use of race in UT Austin's admissions program for the entering freshman class of Fall 2008.

26 Notably, in her supplemental briefing Fisher argues only that she had suffered an “injury in fact.” Supp. Br. of Appellant 12–13. Instead of addressing redressability, she argues only that the question of remedies is a separate inquiry. Id. at 13–14. Regardless of the district court's bifurcation of merits and remedies, the redressability of an injury is integral to the standing inquiry.

27 See Br. of Resp. 6–20.

28 .438 U.S. 265, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978).

29 .539 U.S. 306, 123 S.Ct. 2325, 156 L.Ed.2d 304 (2003).

II

We turn to the question whether we can and should remand this case. The Supreme Court's mandate frames its resolution, ordering that “[t]he judgment of the Court of Appeals is vacated, and the case remanded for further proceedings consistent with this opinion.” The mandate must be read against the backdrop of custom that accords courts of appeal discretion to remand to the district court on receipt of remands to it for proceedings consistent with the opinion—a customary discretion not displaced but characterized by nigh boiler plate variations in phrasing of instructions such as “on remand the Court of Appeals may ‘consider,’ ” or “for the Court of Appeals to consider in the first instance.” ³⁰

30 See, e.g., Spector v. Norwegian Cruise Line, Ltd., 427 F.3d 285, 286 (5th Cir.2005); United States v. Williamson, 47 F.3d 1090 (11th Cir.1995); FW/PBS, Inc. v. City of Dall., 896 F.2d 864, 865 (5th Cir.1990).

A

Fisher argues that the Supreme Court's remanding language—“fairness to the litigants and the courts that heard the case requires that it be remanded so that the admissions process can be considered and judged under a correct analysis” ³¹—compels the conclusion that “fairness” must be achieved by having this Court, and not the district court, conduct the inquiry. Fisher relies on the Supreme Court's statement that “the Court of Appeals must assess whether the University has offered sufficient evidence that its admissions program is narrowly tailored to obtain the educational benefits of diversity.” ³² And Fisher argues that at summary judgment, all parties conceded that there were no genuine issues of material fact to be resolved and that the case should be decided on summary judgment.

31 Fisher v. Univ. of Tex. at Austin, ––– U.S. ––––, 133 S.Ct. 2411, 2421, 186 L.Ed.2d 474 (2013).

32 Id.

UT Austin opposes this parsing of language, arguing that Fisher fails to credit (i) the entirety of the Supreme Court's references which spoke, not just to the fairness of allowing this Court to correct its error, but also to the fairness to the district court, which first heard the case and was faulted for the same error as this Court; and, (ii) that the language used by the Supreme Court is the common language of remand orders and is often followed by a remand to the district court. UT Austin notes that in its remanding language, the Supreme Court cites Adarand Constructors, Inc. v. Pena, ³³ where the court of appeals remanded to the district court after the Supreme Court vacated the judgment of the court of appeals for failure to apply strict scrutiny. Finally, UT Austin argues that the remand language, at best, is ambiguous and, given the custom of the courts of appeals, should not be read to foreclose the clear discretion of this Court to remand absent specific, contrary instructions from the Supreme Court.

33 .515 U.S. 200, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995).

Given the customary practice of the courts of appeals and the less than clear language of the Supreme Court's remand, we are not persuaded that the Supreme Court intended to foreclose our discretion to remand to the district court. A review of the Supreme Court's language lends but little support to each side. Yet, this is telling. Had the Supreme Court intended to control the discretion of this Court as to whether the district court should first address an error that the Supreme Court found was made by both courts, there would have been no uncertainty in the remand language. The question whether we should remand remains.

B

There is no clear benefit to remanding this case to the district court. The suggestion, without more, that discovery may be necessary given the Supreme Court's holding regarding proper scrutiny and deference adds nothing. Admittedly, this case differs from Grutter, in that Grutter went to trial. And evidence offered by live witnesses is far more likely to surface and resolve fact issues than summary judgment evidence crafted by advocates. But that too is far from certain. Indeed, UT Austin's argument goes no further than “factual questions or disputes may arise on remand.” ³⁴ Notably, UT Austin does not argue that a trial will be necessary. Rather its principal target on remand is standing, with questions that continue to haunt, but are foreclosed by the Supreme Court's implicit finding of standing, questions only it can now address.

34 Defs.' Mot. to Remand at 4.

We find that there are no new issues of fact that need be resolved, nor is there any identified need for additional discovery; that the record is sufficiently developed; and that the found error is common to both this Court and the district court. It follows that a remand would likely result in duplication of effort. We deny UT Austin'smotion for remand, and turn to the merits.

III

A

In remanding, the Supreme Court held that its decision in Grutter requires that “strict scrutiny must be applied to any admissions program using racial categories or classifications”; ³⁵ that “racial classifications are constitutional only if they are narrowly tailored to further compelling governmental interests.” ³⁶ Bringing forward Justice Kennedy's dissent in Grutter, the Supreme Court faulted the district court's and this Court's review of UT Austin's means to achieve the permissible goal of diversity—whether UT Austin's efforts were narrowly tailored to achieve the end of a diverse student body. Our charge is to give exacting scrutiny to these efforts.

35 Fisher, 133 S.Ct. at 2419.

36 Grutter, 539 U.S. at 326, 123 S.Ct. 2325.

The Supreme Court has made clear that “a university's educational judgment that such diversity is essential to its educational mission is one to which we defer.” ³⁷ The “decision to pursue the educational benefits that flow from student body diversity that the University deems integral to its mission is, in substantial measure, an academic judgment to which some, but not complete, judicial deference is proper under Grutter.” ³⁸ Accordingly, a court “should ensure that there is a reasoned, principled explanation for the academic decision.” ³⁹

37 Fisher, 133 S.Ct. at 2419 (quoting Grutter, 539 U.S. at 328, 123 S.Ct. 2325) (internal quotation marks omitted).

38 Id. (internal quotation marks and citations omitted).

39 Id.

In both Fisher and Grutter, the Supreme Court endorsed Justice Powell's conclusion that “attainment of a diverse student body ... is a constitutionally permissible goal for an institution of higher education;” ⁴⁰ that in contrast to “[r]edressing past discrimination, ... [t]he attainment of a diverse student body ... serves values beyond race alone, including enhanced classroom dialogue and the lessening of racial isolation and stereotypes”; ⁴¹ that the “academic mission of a university is a special concern of the First Amendment ... [and part] of the business of a university [is] to provide that atmosphere which is most conducive to speculation, experiment, and creation, and this in turn leads to the question of who may be admitted to study.” ⁴² It signifies that this compelling interest in “securing diversity's benefits ... is not an interest in simple ethnic diversity, in which a specified percentage of the student body is in effect guaranteed to be members of selected ethnic groups, with the remaining percentage an undifferentiated aggregation of students.” ⁴³ Rather, “diversity that furthers a compelling state interest encompasses a far broader array of qualifications and characteristics of which racial or ethnic origin is but a single though important element.” ⁴⁴ Justice Powell found Harvard's admissions program to be particularlycommendable.⁴⁵ There an applicant's race was but one form of diversity that would be weighed against qualities such as “exceptional personal talents, unique work or service experience, leadership potential, maturity, demonstrated compassion, a history of overcoming disadvantage, ability to communicate with the poor, or other qualifications deemed important.” ⁴⁶ Bakke envisions a rich pluralism for American institutions of higher education, one at odds with a one-size-fits-all conception of diversity, indexed to the ways in which a diverse student body contributes to a university's distinct educational mission, not numerical measures.⁴⁷

40 Bakke, 438 U.S. at 311, 98 S.Ct. 2733.

41 Fisher, 133 S.Ct. at 2417–18.

42 Id. at 2418.

43 Id. (internal quotation marks and citations omitted).

44 Id.

45 Id.

46 Id. at 317, 98 S.Ct. 2733.

47 Justice Powell's opinion pointed to this accent upon mission at Harvard—one akin to an aged tradition at Oxford—to shape lives, not just fill heads with facts.

Diversity is a composite of the backgrounds, experiences, achievements, and hardships of students to which race only contributes. “[A] university is not permitted to define diversity as some specified percentage of a particular group merely because of its race or ethnic origin” because that “would amount to outright racial balancing, which is patently unconstitutional.” ⁴⁸ Instead, Grutter approved the University of Michigan Law School's goal of “attaining a critical mass of under-represented minority students,” and noted that such a goal “does not transform its program into a quota.” ⁴⁹

48 Grutter, 539 U.S. at 330, 123 S.Ct. 2325 (citing Bakke, 438 U.S. at 307, 98 S.Ct. 2733).

49 Id. at 335–36, 123 S.Ct. 2325.

B

In language from which it has not retreated, the Supreme Court explained that the educational goal of diversity must be “defined by reference to the educational benefits that diversity is designed to produce.” ⁵⁰ Recognizing that universities do more than download facts from professors to students, the Supreme Court recognized three distinct educational objectives served by diversity: (i) increased perspectives, meaning that diverse perspectives improve educational quality by making classroom discussion “livelier, more spirited, and simply more enlightening and interesting when the students have the greatest possible variety of backgrounds”; ⁵¹ (ii) professionalism, meaning that “student body diversity ... better prepares [students] as professionals,” because the skills students need for the “increasingly global marketplace can only be developed through exposure to widely diverse people, cultures, ideas, and viewpoints”; ⁵² and, (iii) civic engagement, meaning that a diverse student body is necessary for fostering “[e]ffective participation by members of all racial and ethnic groups in the civil life of our Nation[, which] is essential if the dream of one Nation, indivisible, is to be realized.” ⁵³ All this the Supreme Court reaffirmed, leaving for this Court a “further judicial determination that the admissions process meets strict scrutiny in its implementation”; ⁵⁴ that is, its means of achieving the goal of diversity are narrowly tailored.

50 Id. at 329–30, 123 S.Ct. 2325.

51 Id. at 330, 123 S.Ct. 2325.

52 Id.

53 Id. at 332, 123 S.Ct. 2325.

54 Fisher, 133 S.Ct. at 2419–20.

A university “must prove that the means chosen by the University to attain diversity are narrowly tailored to that goal.” ⁵⁵ And a university “receives no deference” on this point because it is the courts that must ensure that the “means chosen to accomplish the [university's] asserted purpose ... be specifically and narrowly framed to accomplish that purpose.” ⁵⁶ Although “a court can take account of a university's experience and expertise in adopting or rejecting certain admissions processes,” it remains a university's burden to demonstrate and the court's obligation to determine whether the “admissions processes ensure that each applicant is evaluated as an individual, and not in a way that makes an applicant's race or ethnicity the defining feature of his or her application.” ⁵⁷

55 Id. at 2420.

56 Id. (internal quotation marks and citations omitted).

57 Id. (quoting Grutter, 539 U.S. at 337, 123 S.Ct. 2325) (internal quotation marks and citations omitted).

C

Narrow tailoring requires that the court “verify that it is ‘necessary’ for a university to use race to achieve the educational benefits of diversity.” ⁵⁸ Such a verification requires a “careful judicial inquiry into whether a university could achieve sufficient diversity without using racial classifications.” ⁵⁹ Thus, the reviewing court must “ultimately be satisfied that no workable race-neutral alternatives would produce the educational benefits of diversity.” ⁶⁰ It follows, therefore, that if “a nonracial approach ... could promote the substantial interest about as well and at tolerable expenses, ... then the university may not consider race.” ⁶¹ And it is the university that bears “the ultimate burden of demonstrating, before turning to racial classifications, that available, workable race-neutral alternatives do not suffice.” ⁶²

58 Id. (quoting Bakke, 438 U.S. at 305, 98 S.Ct. 2733).

59 Id.

60 Id. (emphasis added).

61 Id. (quoting Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 280 n. 6, 106 S.Ct. 1842, 90 L.Ed.2d 260 (1986) (internal quotation marks omitted)).

62 Id.

The Supreme Court emphasized that strict scrutiny must be balanced. That is, “[s]trict scrutiny must not be strict in theory, but fatal in fact,” yet it must also “not be strict in theory but feeble in fact.” ⁶³

63 Id. at 2421.

IV

A

Fisher insists that our inquiry into narrow tailoring begin in 2004, the last year before UT Austin adopted its current race-conscious admissions program. Looking to that year, Fisher argues that the Top Ten Percent Plan had achieved a substantial combined Hispanic and African–American enrollment of approximately 21.5%; ⁶⁴ and that this is more minority enrollment than present in Grutter, where a race-conscious plan grew minority enrollment from approximately 4% to 14%. Because UT Austin was already enrolling a larger percentage of minorities than the Michigan Law School, the argument maintains, UT Austin had achieved sufficient diversity to attain the educational benefits of diversity, a critical mass, before it adopted a race-conscious admissions policy; that even if sufficient diversity had not been achieved by 2004, it had been achieved by 2007 when the combined percentage of Hispanic and African–American enrolled students was 25.5%. Thus, Fisher argues, the race-conscious admissions policy had a de minimis effect, at most adding 0.92% African–American enrollment and 2.5% Hispanic enrollment; that a slight contribution is not a “constitutionally meaningful” impact on student body diversity and is no more than an exercise in gratuitous racial engineering.

64 2008 Top Ten Percent Report at 6 tbl.1.

This effort to truncate the inquiry clings to a baseline that crops events Fisher's claim ignores, as it must. The true narrative presents with a completeness both fair and compelled by the Supreme Court's charge to ascertain the facts in full without deference, exposing the de minimis argument as an effort to turn narrow tailoring upside down. We turn to that narrative.

B

In 1997, following the Hopwood v. Texas ⁶⁵ decision, UT Austin faced a nearly intractable problem: achieving diversity—including racial diversity—essential to its educational mission, while not facially considering race even as one of many components of that diversity. Forbidden any use of race after Hopwood, UT Austin turned to the Top Ten Percent Plan, which guarantees Texas residents graduating in the top ten percent of their high school class admission to any public university in Texas. Such a mechanical admissions program could have filled every freshman seat but standing alone it was not a workable means of achieving the diversity envisioned by Bakke, bypassing as it did high-performing multi-talented students, minority and non-minority. With its blindness to all but the single dimension of class rank, the Top Ten Percent Plan came with significant costs to diversity and academic integrity, passing over large numbers of highly qualified minority and non-minority applicants. The difficulties of Texas's and other states' percentage plans did not escape the Court in Grutter, which explained that “even assuming such plans are race-neutral, they may preclude the university from conducting the individualized assessments necessary to assemble a student body that is not just racially diverse, but diverse along all the qualities valued by the university.” ⁶⁶

65 .78 F.3d 932 (5th Cir.1996), abrogated by Grutter, 539 U.S. at 322, 123 S.Ct. 2325.

66 Grutter, 539 U.S. at 340, 123 S.Ct. 2325.

Nor did these difficulties escape the Texas legislature. Opponents to the proposed plan noted that such a policy “could actually harm institutions” and “would not solve the problems created by [ Hopwood ].” ⁶⁷ So the legislature adopted a Top Ten Percent Plan that left a substantial number of seats to a complementary holistic review process. Foreshadowing Grutter, admission supplementing the Top Ten Percent Plan included factors such as socio-economic diversity and family educational achievements but, controlled by Hopwood, it did not include race. In short, a holistic process sans race controlled the gate for the large percent of applicants not entering through the Top Ten Percent Plan. Over the succeeding years the Top Ten Percent Plan took an increasing number of seats, a take inherent in its structure and a centerpiece of narrow tailoring, as we will explain.

67 Pls.' Mot. Summ. J., Ex. 27, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 94 (HB 588, House Research Organization Digest, Apr. 15, 1997).

C

We are offered no coherent response to the validity of a potentially different electionby UT Austin: to invert the process and use Grutter's holistic review to select 80% or all of its students. Such an exponential increase in the use of race under the flag of narrow tailoring is perverse. Grutter blessed an admissions program, applied to the entire pool of students competing for admission, which “considers race as one factor among many, in an effort to assemble a student body that is diverse in ways broader than race.” Affording no deference, we look for narrow tailoring in UT's Austin's use of this individualized race-conscious holistic review, applied as it is only to a small fraction of the student body as the rest is consumed by race-neutral efforts.

Close scrutiny of the data in this record confirms that holistic review—what little remains after over 80% of the class is admitted on class rank alone—does not, as claimed, function as an open gate to boost minority headcount for a racial quota. Far from it. The increasingly fierce competition for the decreasing number of seats available for Texas students outside the top ten percent results in minority students being under-represented—and white students being over-represented—in holistic review admissions relative to the program's impact on each incoming class. In other words, for each year since the Top Ten Percent Plan was created through 2008, holistic review contributed a greater percentage of the incoming class of Texans as a whole than it did the incoming minority students. Examples illustrate this effect. Of the incoming class of 2008, the year Fisher applied for admission, holistic review contributed 19% of the class of Texas students as a whole—but only 12% of the Hispanic students and 16% of the black students, while contributing 24% of the white students.⁶⁸ The incoming class of 2005, the year that the Grutter plan was first introduced, is similar. That year, 31% of the class of Texas students as a whole was admitted through holistic review (with the remaining 69% of incoming seats for Texans filled by the Top Ten Percent Plan)—but only 21% of the Hispanic Texan students in the incoming class were admitted through holistic review, and 26% of the incoming black Texan students, but 35% of the incoming white Texan students.⁶⁹ Minorities being under-represented in holistic review admission relative to the impact of holistic review on the class as a whole holds true almost without exception for both blacks and Hispanics for every year from 19962008,⁷⁰ and can be seen in the chart attached to this opinion at Appendix 1.

68 2008 Top Ten Percent Report at 7 tbl.1a.

69 Office of Admissions, Univ. of Tex. at Austin, Implementation and Results of the Texas Automatic Admissions Law (HB 588) at the University of Texas at Austin: Demographic Analysis of Entering Freshmen 2006 and Academic Performance of Top 10% and Non–Top 10% Students Academic Years 1996–2005, at 5 tbl.1a (Dec. 6, 2007) [hereinafter 2006 Top Ten Percent Report ], Pls.' Mot. Summ. J., Ex. 25, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 94, available at htt p:// www. utexas. edu/ student/ ad missions/research/HB588–Report–VolumeI.pdf.

70 Later editions of the same reports available as public data show that as the take of the Top Ten Percent Plan continued to grow, this effect intensified. In 2009, when the holistic review program was left with only 14.4% of the seats available for Texas residents, only 6.3% of Hispanic enrolled students were admitted through holistic review and 10.0% of blacks, but 18.8% of whites. Office of Admissions, Univ. of Tex. at Austin, Implementation and Results of the Texas Automatic Admissions Law (HB 588) at the University of Texas at Austin: Demographic Analysis of Entering Freshmen Fall 2009 and Academic Performance of Top 10% and Non–Top 10% Students Academic Years 20042008 (Report 12), at 8 tbl.1a (Oct. 29, 2009) [hereinafter 2009 Top Ten Percent Report], available at http:// www. utexas. edu/ student/ admissions/ research/ HB 588- Report 12. pdf; see also Office of Admissions, Univ. of Tex. at Austin, Implementation and Results of the Texas Automatic Admissions Law (HB 588) at the University of Texas at Austin: Demographic Analysis of Entering Freshmen Fall 2010 and Academic Performance of Top 10% and Non–Top 10% Students Entering Freshmen 2009 (Report 13) (Dec. 23, 2010) [hereinafter 2010 Top Ten Percent Report], available at http:// www. utexas. edu/ student/ admissions/ research/ HB 588- Reportl 3. pdf. The passage of SB 175 allowed UT Austin to reset the take of the automatic admissions program to a minimum of 75% of the admissions slots, but the effect continued. William Powers Jr., Univ. of Tex. at Austin, Report to the Governor, the Lieutenant Governor, and the Speaker of the House of Representatives on the Implementation of SB 175, 81st Legislature, for the Period Ending Fall 2013, at 29 tbl.4.1 (Dec. 20, 2013) [hereinafter 2013 Powers Report], available at http:// www. utexas. edu/ student/ admissions/ research/ SB_ 175_ Report_ for_ 2013. pdf.

Given the test score gaps between minority and non-minority applicants, if holistic review was not designed to evaluate each individual's contributions to UT Austin's diversity, including those that stem from race, holistic admissions would approach an all-white enterprise. Data for the entering Texan class of 2005, the first year of the Grutter plan, show that Hispanic students admitted through holistic review attained an average SAT score of 1193, African–American students an 1118, and white students a 1295.⁷¹ For the entering class of 2007, the last class before Fisher applied for admission, the corresponding data were 1155 for Hispanic students, 1073 for African American students, and 1275 for white students, this from a universe of underperforming secondary schools.⁷² As we have explained, the impact of the holistic review program on minority admissions is already narrow, targeting students of all races that meet both the competitive academic bar of admissions and have unique qualities that complement the contributions of Top Ten Percent Plan admittees.

71 2006 Top Ten Percent Report at 1114.

72 2008 Top Ten Percent Report at 1215.

D

UT Austin did not stop with the Top Ten Percent Plan in its effort to exhaust racially neutral alternatives to achieving diversity. It also initiated a number of outreach and scholarship efforts targeting under-represented demographics, including the over half of Texas high school graduates that are African–American or Hispanic.⁷³ Programs included the Longhorn Opportunity Scholarship Program, the Presidential Achievement Scholarship Program, the First Generation Scholarship, and increased outreach efforts. Implemented in 1997, the Longhorn Opportunity Scholarship Program offers scholarships to graduates of certain high schools throughout Texas that had predominantly low-income student populations and a history of few, if any, UT Austin matriculates.⁷⁴ It guarantees a specific number of scholarships for applicants who attend these schools, graduate within the top ten percent, and attend UT Austin. The Presidential Achievement Scholarship program is a need-based scholarship that is awarded based on the applicant's family income, high school characteristics, and academic performance as compared to his or her peers at that high school.⁷⁵ The First Generation Scholarship Program targets applicants who are the first in their family to attend college.⁷⁶ UT Austin invested substantial amounts of money in these scholarship programs. Between 1997 and 2007, UT Austin awarded $59 million through these scholarships.⁷⁷ Indeed, in 2007, UT Austin awarded $5.8 million for the Longhorn Opportunity and Presidential Achievement scholarship programs alone.⁷⁸

73 The Texas public high school graduating class of 2008, the year Fisher graduated from high school, included 13.4% African–American and 37.5% Hispanic students. Div. of Performance Reporting, Tex. Educ. Agency, 2008–09 Texas Public School Statistics Pocket Edition, at 3 (December 2009), available at http:// ritter. tea. state. tx. us/ perfreport/ pocked/ 2009/ pocked 0809. pdf. This means that of this majority-minority cohort of 33,873 African–American and 94,571 Hispanic high school, or 128,444 minority graduates in all, UT admitted 728 African–Americans and 2,621 Hispanics—or 2.6% of the graduating minority seniors of Texas. See id. at 5; see generally 2008 Top Ten Percent Report at 6 tbl.1. As the percentage of Hispanic high school graduates has continued to increase, over 57.3% of the high school graduating class of 2011, the most recent year for which the Texas Education Agency has published statistics, are African–American or Hispanic. Div. of Performance Reporting, Tex. Educ. Agency, 2011–12 Texas Public School Statistics Pocket Edition, at 1, available at http:// www. tea. state. tx. us/ Work Area/ linkit. aspx? Link Identifier= id& Item ID= 2147511872& lib ID= 2147511859.

74 Defs.' Cross–Mot. Summ. J., Ex. Tab 9 to App., Orr Aff. at ¶ 7, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 96 [hereinafter Orr Aff.]. Initially, this program targeted 39 high schools, but expanded to 69 high schools by 2009. Id.

75 Defs.' Cross–Mot. Summ. J., Ex. Tab 4 to App., Orr Dep. at 15:17–21, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 96 [hereinafter Orr Dep.]; Orr Aff. ¶ 6, ECF No. 96.

76 Orr Aff. ¶ 8, ECF No. 96.

77 Id. ¶ 9.

78 Id. ¶ 9.

UT Austin also expanded its outreach and recruitment efforts by increasing its recruitment budget by $500,000, by adding three regional admissions centers in Dallas, San Antonio, and Harlingen,⁷⁹ by engaging in outreach programs that brought prospective students to UT Austin for daylong or overnight visits,⁸⁰ and by hosting multi-day campus conferences for high school counselors.⁸¹ These regional admissions centers reflect a substantial investment by UT Austin: the Dallas Admissions center employed 4 new full-time staff, the San Antonio Admissions Center employed 4 new full-time staff, and the Harlingen Admissions Center employed 5 new full-time staff. ⁸² The stated goal of these centers was “to increase [UT Austin's] visibility and interaction with prospective students, parents and high school administrators within the geographic market they existed [sic]. These centers allowed for increased quality and quantity of counseling, face to face discussions, and programming within the prospective students' home city.” ⁸³ Additionally, staff from these regional centers helped organize “over 1,000 College Night/Day events held at High Schools across the state” and “around 1,000 Day Visits to High Schools around the state in an effort to encourage prospective top 10% students to apply and enroll at [UT Austin].” ⁸⁴ Relatedly, the admissions office also held targeted recruiting events for students from the Dallas, San Antonio, Houston, and Rio Grande Valley areas. These events included the “Longhorn Lock-in,” wherein students from targeted high schools would spend the night at UT Austin; the UT Scholars Program, wherein scholarship recipients from targeted schools would spend the night at UT Austin; and “Longhorn for a Day,” wherein students from targeted schools would spend the day at UT Austin.⁸⁵ Finally, the admissions office would hold four “Longhorn Saturday Events” on campus, where thousands of prospective students and their families would come to UT Austin. ⁸⁶

79 Id. ¶ 11.

80 Id. ¶ 1619.

81 Id. ¶ 20.

82 Id. ¶ 11.

83 Id.

84 Id.

85 Id. ¶ 1618.

86 Id. ¶ 19.

In addition to the admissions office's efforts, UT Austin's Office of Student Financial Services increased their outreach efforts by putting together the Financial Aid Outreach Group to visit high schools to help prospective students “understand the financial support offered by [UT Austin].” ⁸⁷ The goal of this Financial Aid Outreach Group “was to convince low income students that money should not be a barrier to attending college.” ⁸⁸

87 Id. ¶ 12.

88 Id.

“Narrow tailoring does not require exhaustion of every race neutral alternative,” but rather “serious, good faith consideration of workable race-neutral alternatives that will achieve the diversity the university seeks.” ⁸⁹ Put simply, this record shows that UT Austin implemented every race-neutral effort that its detractors now insist must be exhausted prior to adopting a race-conscious admissions program—in addition to an automatic admissions plan not required under Grutter that admits over 80% of the student body with no facial use of race at all.

89 Grutter, 539 U.S. at 339, 123 S.Ct. 2325.

E

Despite UT Austin's rapid adoption of these race-neutral efforts, in 1997—the first freshman class after Hopwood—the percentage of African–American admitted students fell from 4.37% to 3.41%, representing a drop from 501 to 419 students even as the total number of admitted students increased by 833 students.⁹⁰ Similarly, the percentage of Hispanic admitted students fell from 15.37% to 12.95%.⁹¹ With UT Austin's facially race-neutral admissions program and outreach efforts, the percentage of African–American and Hispanic admitted students eventually recovered to pre- Hopwood levels. By 2004, African–American admitted students climbed to 4.82% and Hispanic admitted students climbed to 16.21%.⁹² But minority representation then remained largely stagnant, within a narrow oscillating band, rather than moving towards a critical mass of minority students. The hard data show that starting in 1998 and moving toward 2004, African–American students comprised 3.34%, then 4.32%, then 4.24%, then 3.49%, then 3.67%, then 3.89%, and finally 4.82% of the admitted pool.⁹³ Similarly, Hispanic admitted students represented 13.53%, then 14.27%, then 13.75%, then 14.25%, then 14.43%, then 15.60%, and finally 16.21% of the entering classes for those respective years.⁹⁴

90 See 2006 Top Ten Percent Report 4 tbl.1. African–American admits comprised 3.34% of the entering class of 1998; 4.32% of the class of 1999; 4.24% of the class of 2000; 3.49% of the class of 2001; 3.67% of the class of 2002; and 3.89% of the class of 2003. See id.

91 Id. Hispanics represented 13.53% of the entering class of 1998; 14.27% of the class of 1999; 13.75% of the class of 2000; 14.25% of the class of 2001; 14.43% of the class of 2002; 15.60% of the class of 2003; and 16.21% for 2004. See id.

92 Id.

93 Id.

94 Id.

V

A

Numbers aside, the Top Ten Percent Plan's dependence upon a distinct admissions door remained apparent. With each entering class, there was a gap between the lower standardized test scores of students admitted under the Top Ten Percent Plan and the higher scores of those admitted under holistic review. For example, in 2008—the year Fisher applied for admission—81% of the seats available to Texas residents were taken up by the Top Ten Percent Plan.⁹⁵ These Top Ten Percent students had an average standardized test score of 1219, 66 points lower than the average standardized test score of 1285 attained by Texas students admitted under holistic review or on the basis of a high AI.⁹⁶ A gap persisted not only among students overall and white students, but also among racial and ethnic minority students.⁹⁷ This inheres in the reality that the strength of the Top Ten Percent Plan is also its weakness, one that with its single dimension of selection makes it unworkable standing alone.

95 2008 Top Ten Percent Report at 9 tbl.2.

96 Id. Data for the preceding years showed a similar test score gap. For the entering Texas class of 2007, Top Ten Percent students had an average standardized test score of 1225 versus the average standardized test score of 1246 attained by non-Top Ten Percent Texas students. Similarly, in 2006, Top Ten Percent students had an average standardized test score of 1220 versus an average standardized test score of 1257 for non-Top Ten Percent students. For 2005, Top Ten Percent students had an average standardized test score of 1226 versus an average standardized test score of 1277 for non-Top Ten Percent students. Finally, in 2004, Top Ten Percent students had an average standardized test score of 1221 versus an average standardized test score of 1258 for non-Top Ten Percent students. Id.

97 Id. at 9, 13–15. For minority students, difference in average standardized test scores between admitted Texas Top Ten Percent students and non-Top Ten Percent students fluctuated in size but remained significant in the pre- and post- Grutter years leading up to Fisher's application. Among Hispanic students, the gap was 1100 versus 1189 in 2003; 1110 versus 1189 in 2004; 1122 versus 1193 in 2005; 1105 versus 1154 in 2006; and 1115 versus 1155 in 2007. For African–American students, the gap was 1063 versus 1065 in 2003; 1046 versus 1116 in 2004; 1059 versus 1118 in 2005; 1067 versus 1086 in 2006; and 1078 versus 1073 in 2007. See id. at 1415. And a comparison of raw SAT scores does not tell the full story, as SAT scores are scaled. See, e.g., CollegeBoard SAT, 2006 College–Bound Seniors: Total Group Profile Report (2006), available at http:// media. collegeboard. com/ digital Services/ pdf/ research/ cb- seniors 2006- nationalreport. pdf. Looking at the percentile point gives a better picture. For SAT test-takers in 2006, the 50th percentile combined score was 1020, while a 75th percentile score was 1180, a mere 160 points higher. Id. at 2. Thus, a score differential of 80 points, for example, which represents the approximate differential between holistic review and Top Ten Percent Hispanic admittees, represents students scoring at approximately a 12–13 higher percentile.

B

The sad truth is that the Top Ten Percent Plan gains diversity from a fundamental weakness in the Texas secondary education system. The de facto segregation of schools in Texas ⁹⁸ enables the Top Ten Percent Plan to increase minorities in the mix, while ignoring contributions to diversity beyond race. We assume, as none here contends otherwise, that this “segregation [is] not the ‘product ... of state action but of private choices,’ having no ‘constitutional implications' ” and therefore it is “a question for the political branches to decide[ ] the manner—which is to say the process—of its resolution.” ⁹⁹ In short, these demographics are directly relevant to the choices made by the political branches of Texas as they acted against the backdrop of this unchallenged reality in their effort to achieve a diverse student body. Texas is here an active lab of experimentation embraced by the Court in Schuette v. BAMN. ¹⁰⁰ We reference here these unchallenged facts of resegregation not in justification of a racial remedy, but because the racial makeup and relative performance of Texas high schools bear on the workability of an alternative to any use of race for 80% of student admissions to UT Austin. The political branches opted for this facially race-neutral alternative—a narrow tailoring in implementation of their goal of diversity.

98 For example, only 8.1% of all students in Houston ISD are white. See Houston Indep. Sch. Dist., 2011–2012 Facts and Figures 1 (2012). Similarly, only 4.6% of students in the Dallas Independent School District are white. See Dallas Indep. Sch. Dist., Enrollment Statistics (2012). And in San Antonio ISD, only 1.9% of the students are white. See San Antonio Indep. Sch. Dist., Facts and Figures (2012). This de facto school segregation stems from residential patterns and means that students in the top ten percent of a highly segregated school likely grew up in the same residential zone. The top 29 graduates from Jack Yates High School in Houston live in the same predominately African–American neighborhood of that city's Third Ward, and thus likely experienced a similar cultural environment. See Amicus Curiae Br. of the Family of Heman Sweatt (Oct. 31, 2013) at 27. This pattern repeats itself across the high schools of Texas's urban areas.

99 Schuette v. BAMN, ––– U.S. ––––, 134 S.Ct. 1623, 1642, 188 L.Ed.2d 613 (2014) (Scalia, J., concurring) (citing Freeman v. Pitts, 503 U.S. 467, 495–96, 112 S.Ct. 1430, 118 L.Ed.2d 108 (1992)).

100 ––– U.S. ––––, 134 S.Ct. 1623, 188 L.Ed.2d 613 (2014).

Fisher's claim can proceed only if Texas must accept this weakness of the Top Ten Percent Plan and live with its inability to look beyond class rank and focus upon individuals. Perversely, to do so would put in place a quota system pretextually race neutral. While the Top Ten Percent Plan boosts minority enrollment by skimming from the tops of Texas high schools, it does so against this backdrop of increasing resegregation in Texas public schools, ¹⁰¹ where over half of Hispanic students and 40% of black students attend a school with 90%–100% minority enrollment.¹⁰²

101 A striking visual depiction of de facto residential segregation, showing one colored dot per person using 2010 census data, displays nearly monochrome units dividing the major metropolitan areas of Texas. See Demographics Research Grp., Weldon Ctr. for Public Serv., Univ. of Va., 2010 Racial Dot Map, CooperCenter.org (July 2013), http:// demographics. coopercenter. org/ Dot Map/ index. html.

102 Gary Orfield, John Kucsera & Genevieve Siegel–Hawley, Civil Rights Project, E Pluribus ... Separation: A Deepening Double Segregation for More Students 46, 50 (2012), available at http:// civilrights project. ucla. edu/ research/ k- 12- education/ integration anddiversity/ mlk- national/ e- pluribus... separation- de epening- doub le- segregat ion- for- more students/ orfield_ epl uribus_ revis ed_ omplete_ 2012. pdf.

Data for the year Fisher graduated high school show that gaps between the quality of education available to students at integrated high schools and at majority-minority schools are stark. Their impact upon UT Austin is direct. The Top Ten Percent Plan draws heavily from the population concentrations of the three major metropolitan areas of Texas—San Antonio, Houston, and Dallas/Fort Worth—where over half of Texas residents live and where the outcomes gaps of segregated urban schools are most pronounced.¹⁰³ The San Antonio metropolitan area demonstrates this effect. Boerne Independent School District (“ISD”) achieved a “recognized status” and five “Gold Performance Acknowledgments” from the Texas Education Agency.¹⁰⁴ At this relatively integrated school district, 79.9% of graduating students were white and 19.2% were black or Hispanic.¹⁰⁵ Over 97% of students graduated high school.¹⁰⁶ They achieved an average SAT score of 1072, and 61% were deemed college-ready in both English and Math by the Texas Education Agency.¹⁰⁷ San Antonio ISD, its neighbor, a highly segregated and “academically unacceptable” district,¹⁰⁸ tells a different story. 86.8% of graduating students were Hispanic and 8.2% were black, and over 90% were economically disadvantaged.¹⁰⁹ Only 59.1% of the high school class of 2008 graduated; SAT test takers achieved an average score of 811; and only 28% of graduates were college-ready in both English and Math.¹¹⁰

103 The total Texas population for 2008 was 24,326,974. U.S. Census Bureau, Annual Estimates of the Resident Population for the United States, Regions, States, and Puerto Rico: April 1, 2000 to July 1, 2008, tbl.1, available at http:// www. census. gov/ popest/ data/ historical/ 2000 s/ vintage_ 2008/. Of these, 57.8%, or 14,059,594 people, lived in the Dallas/Fort Worth, Houston, and San Antonio metropolitan areas. See U.S. Census Bureau, Annual Estimates of the Population of Metropolitan and Micropolitan Statistical Areas: April 1, 2000 to July 1, 2008, tbl.1, available at http:// www. census. gov/ popest/ data/ historical/ 2000 s/ vintage—_ 008/ metro. html (showing that 6,300,006 lived in the Dallas/Fort Worth metropolitan area; 5,728,143 lived in the Houston metropolitan area; and 2,031,445 lived in the San Antonio metropolitan area in 2008).

104 2008–09 Academic Indicator System, Boerne ISD, Tex. Educ. Agency, 1 [hereinafter Boerne ISD Indicator ], http:// ritter. tea. state. tx. us/ perfreport/ aeis/ 2009/ district. srch. html (accessed by searching for the relevant school district on the search engine).

105 Id. § II, at 1.

106 Id. § I, at 11.

107 Id. § I, at 12.

108 200809 Academic Indicator System, San Antonio ISD, Tex. Educ. Agency, § II, at 1 [hereinafter San Antonio ISD Indicator ], http:// ritter. tea. state. tx. us/ perfreport/ aeis/ 2009/ district. srch. html (accessed by searching for the relevant school district on the search engine).

109 Id. § 11, at 1.

110 Id. § I, at 1112.

A similar tale of two cities played out in the Houston area between integrated Katy ISD, where 7.8% of graduating students were black, 23.2% Hispanic, and 59.8% white,¹¹¹ and segregated Pasadena ISD, where 6.5% were black, 64.8% Hispanic, and 24.3% white.¹¹² At Katy, a “recognized” district with two “Gold Performance Acknowledgments,” 91.8% of students graduated, with an average SAT score of 1080 and 60% college readiness in both English and Math.¹¹³ At Pasadena, only 67.8% graduated; SAT test-takers achieved an average score of 928; and 40% were college-ready in both English and Math. ¹¹⁴

111 2008–09 Academic Indicator System, Katy ISD, Tex. Educ. Agency, § II, at 1 [hereinafter Katy ISD Indicator], http:// ritter. tea. state. tx. us/ perfreport/ aeis/ 2009/ district. srch. html (accessed by searching for the relevant school district on the search engine).

112 200809 Academic Indicator System, Pasadena ISD, Tex. Educ. Agency, § II, at 1 [hereinafter Pasadena ISD Indicator ], http:// ritter. tea. state. tx. us/ perfreport/ aeis/ 2009/ district. srch. html (accessed by searching for the relevant school district on the search engine).

113 Katy ISD Indicator, § I, at 1112.

114 Pasadena ISD Indicator, § I, at 1112.

The narrative repeats itself in the Dallas/Fort Worth metropolitan area. For example, Keller ISD, a large and “recognized” school district with four “Gold PerformanceAcknowledgements,” ¹¹⁵ is fairly integrated. 72.3% of graduating students are white, 12.2% are Hispanic, and 7.3% are African–American.¹¹⁶ The high school senior class of 2008 attained a graduation rate of 88.7% and an average SAT score of 1043, and 53% were college-ready in both English and Math.¹¹⁷ The data for nearby Dallas ISD, one of the largest in the state with 157,174 students and 7,308 high school seniors,¹¹⁸ shows a highly segregated school in stark contrast. There, black and Hispanic students make up 90.9% of the graduating class, and 86.1% of all students are economically disadvantaged.¹¹⁹ Only 65.2% graduated high school; SAT test-takers achieved an average score of 856; and only 29% of graduating seniors were college-ready in both English and Math. ¹²⁰

115 2008–09 Academic Indicator System, Keller ISD, Tex. Educ. Agency, § II, at cover [hereinafter Keller ISD Indicator ], http:// ritter. tea. state. tx. us/ perfreport/ aeis/ 2009/ district. srch. html (accessed by searching for the relevant school district on the search engine).

116 Id. § II, at 1.

117 Id. § I, at 1112.

118 2008–09 Academic Indicator System, Dallas ISD, Tex. Educ. Agency, § II, at 1 [hereinafter Dallas ISD Indicator ], http:// ritter. tea. state. tx. us/ perfreport/ aeis/ 2009/ district. srch. html (accessed by searching for the relevant school district on the search engine).

119 Id.

120 Id. § I, at 11–12.

The top decile of high schools in each of these districts—including large numbers of students from highly segregated, underfunded, and underperforming schools—all qualified for automatic admission to UT Austin. That these students were able to excel in the face of severe limitations in their high school education and earn a coveted place in UT Austin's prestigious freshman class is to be commended. That other students are left out—those who fell outside their high school's top ten percent but excelled in unique ways that would enrich the diversity of UT Austin's educational experience—leaves a gap in an admissions process seeking to create the multi-dimensional diversity that Bakke envisions.

C

UT Austin's holistic review program—a program nearly indistinguishable from the University of Michigan Law School's program in Grutter—was a necessary and enabling component of the Top Ten Percent Plan by allowing UT Austin to reach a pool of minority and non-minority students with records of personal achievement, higher average test scores, or other unique skills. A variety of perspectives, that is differences in life experiences, is a distinct and valued element of diversity. Yet a significant number of students excelling in high-performing schools are passed over by the Top Ten Percent Plan although they could bring a perspective not captured by admissions along the sole dimension of class rank. For example, the experience of being a minority in a majority-white or majority-minority school and succeeding in that environment offers a rich pool of potential UT Austin students with demonstrated qualities of leadership and sense of self. Efforts to draw from this pool do not demean the potential of Top Ten admittees. Rather it complements their contribution to diversity—mitigating in an important way the effects of the single dimension process.

UT Austin persuades that this reach into the applicant pool is not a further search for numbers but a search for studentsof unique talents and backgrounds who can enrich the diversity of the student body in distinct ways including test scores, predicting higher levels of preparation and better prospects for admission to UT Austin's more demanding colleges and ultimately graduation. It also signifies that this is a draw from a highly competitive pool, a mix of minority and non-minority students who would otherwise be absent from a Top Ten Percent pool selected on class rank, a relative and not an independent measure across the pool of applicants.

VI

These realities highlight the difficulty of an approach that seeks to couch the concept of critical mass within numerical terms. The numbers support UT Austin's argument that its holistic use of race in pursuit of diversity is not about quotas or targets, but about its focus upon individuals, an opportunity denied by the Top Ten Percent Plan. Achieving the critical mass requisite to diversity goes astray when it drifts to numerical metrics. UT Austin urges that it has made clear that looking to numbers, while relevant, has not been its measure of success; and that its goals are not captured by population ratios. We find this contention proved, mindful that by 2011, Texas high school graduates were majority-minority.

UT Austin urges that its first step in narrow tailoring was the admission of over 80% of its Texas students though a facially race-neutral process, and that Fisher's embrace of the sweep of the Top Ten Percent Plan as a full achievement of diversity reduces critical mass to a numerical game and little more than a cover for quotas. Fisher refuses to acknowledge this distinction between critical mass—the tipping point of diversity—and a quota. And in seeking to quantify “critical mass” as a rigid numerical goal, Fisher misses the mark. Fisher is correct that if UT Austin defined its goal of diversity by the numbers only, the Top Ten Percent Plan could be calibrated to meet that mark. To do so, however, would deny the role of holistic review as a necessary complement to Top Ten Percent admissions. We are persuaded that holistic review is a necessary complement to the Top Ten Percent Plan, enabling it to operate without reducing itself to a cover for a quota system; that in doing so, its limited use of race is narrowly tailored to this role—as small a part as possible for the Plan to succeed.

A

The Top Ten Percent Plan is dynamic, its take floating year to year with the number of Texas high school graduates in the top ten percent of their class that choose to capitalize on their automatic admission to the flagship university. Its impact on the composition of each incoming class predictably has grown dramatically, leaving ever fewer holistic review seats available for the growing demographic of Texas high school graduates. In 1996, when the Top Ten Percent Plan was introduced, it admitted 42% of the Texas incoming class; by 2005, when the Grutter plan was introduced, the Plan occupied 69% of the seats available to Texas residents; by 2008, when Fisher applied for admission, it had swelled to 81%.¹²¹ The increasing take of the Top Ten Percent Plan both enhanced its strengths and exacerbated its inherent weaknesses in composing the UT student body, as the overwhelming majority of seats was granted to students without the facial use of race but also without consideration of experiences beyond a single academic dimension. So as the take of the Top Ten Percent Plan grew, so also did the necessity of a complementary holistic admissions program to achieve the diversity envisioned by Bakke.

121 In 1996, the Top Ten Percent Plan admitted 41.8% of the incoming class of Texas students; 36.6% in 1997; 41.1% in 1998; 44.9% in 1999; 47.4% in 2000; 51.3% in 2001; 54.4% in 2002; 70.4% in 2003; 66.3% in 2004; 68.7% in 2005; 71.4% in 2006; 70.6% in 2007; and 80.9% in 2008. See 2006 Top Ten Percent Report at 5 tbl.1a (data for years 1996–2005); 2008 Top Ten Percent Report at 7 tbl.1a (data for 2006–2008).

A quick glance in the public record of data since 2008 confirms that UT Austin's race-conscious holistic review program has a self-limiting nature, one that complements UT Austin's periodic review of the program's necessity to ensure it is limited in time. For the entering class of 2009, the year after Fisher applied for admission, the Top Ten Percent Plan's take of the seats available for Texas residents swelled to 86% and remained at 85% in 2010. ¹²²

122 See 2010 Top Ten Percent Report at 8 tbl.1a.

This trend did not escape the Texas Legislature. Consistent with its long-standing view of holistic review as a crucial complement to the Top Ten Percent Plan, Texas passed Senate Bill 175 of the 81st Texas Legislature (SB 175) in 2009. SB 175 modified the Top Ten Percent Plan for UT Austin to authorize the University “to limit automatic admission to no less than 75% of its enrollment capacity for first-time resident undergraduate students beginning with admission for the entering class of 2011 and ending with the entering class of 2015.” ¹²³ Pursuant to SB 175, UT Austin restricted automatic admissions to the top 7% for Fall 2014 and Fall 2015 applicants, to the top 8% for Fall 2011 and Fall 2013 applicants, and to the top 9% for Fall 2012 applicants.¹²⁴ All remaining slots continue to be filled through holistic review.¹²⁵ For the entering class of 2011, the first affected by SB 175, 74% of enrolled Texas residents were automatically admitted (with a higher percentage of offers of admission), a figure that again was pushed upward by inherent population forces, to 77% for the entering Texas class of 2013.¹²⁶

123 William Powers, Jr., Univ. of Tex. at Austin, Report to the Governor, the Lieutenant Governor, and the Speaker of the House of Representatives on the Implementation of SB 175, at 4 (Dec. 31, 2011) [hereinafter 2011 Powers Report], available at https:// www. utexas. edu/ student/ admissions/ research/ SB_ 175_ Report_ for_ 2011. pdf.

124 Automatic Admission, Univ. of Tex. at Austin (Sept. 16, 2013, 2:56 PM), http:// bealonghorn. utexas. edu/ freshmen/ decisions/ automatic- admission.

125 Id.

126 2013 Powers Report at 29 tbl.4.1.

In the growing shadow of the Top Ten Percent Plan, there was a cautious, creeping numerical increase in minority representation following the inclusion of race and ethnicity in the holistic review program, a testament, UT Austin says, to its race-conscious holistic review. We must agree. From 2004, the last facially race-neutral holistic review program year, to 2005, the first year that race and ethnicity were considered, the percentage of African–American students admitted to UT Austin climbed from 4.82% to 5.05%. The trend has continued since, climbing to 5.13% in 2006, 5.41% in 2007, and 5.67% in 2008. Similarly, the percentage of Hispanic admitted students climbed from 16.21% in 2004, to 17.88% in 2005, 18.08% in 2006, 19.07% in 2007, and 20.41% in 2008.¹²⁷ The modest numbers only validate the targeted role of UT Austin's use of Grutter. Nor can they be viewed as a pretext for quota seeking—an assertion of Fisher's belied by the reality that over this time frame graduating Texas high school seniors approached being majority-minority. The small increases do not exceed critical mass nor imply a quota but instead bring a distinct dimension of diversity to the Top Ten Percent Plan. To be sure, critical mass can be used as a cover for quotas and proportionality goals, but it is not inevitable; UT Austin persuades that viewed objectively, under its structure, its efforts in holistic review have not been simply to expand the numbers but rather the diversity of individual contributions.

127 2008 Top Ten Percent Report at 6 tbl.1.

Turning in the opposite direction from her claim of racial quotas, Fisher faults UT Austin's holistic use of race for its de minimis contribution to diversity. UT Austin replies that this turns narrow tailoring upside down. We agree. Holistic review allows selection of an overwhelming number of students by facially neutral measures and for the remainder race is only a factor of factors. Fisher's focus on the numbers of minorities admitted through the holistic gate relative to those admitted through the Top Ten Percent Plan is flawed, ignoring its role as a necessary complement to the Plan. The apt question is its contribution to the richness of diversity as envisioned by Bakke against the backdrop of the Top Ten Percent Plan. That is its palliative role claimed by UT Austin. So viewed, holistic review's low production of numbers is its strength, not its weakness.

In sum, Fisher points to the numbers and nothing more in arguing that race-conscious admissions were no longer necessary because a “critical mass” of minority students had been achieved by the time Fisher applied for admission—a head count by skin color or surname that is not the diversity envisioned by Bakke and a measure it rejected. In 2007, Fisher emphasizes, there were 5.8% African–American and 19.7% Hispanic enrolled students, which exceeds pre- Hopwood levels and the minority enrollment at the University of Michigan Law School examined in Grutter. But an examination that looks exclusively at the percentage of minority students fails before it begins. Indeed, as Grutter teaches, an emphasis on numbers in a mechanical admissions process is the most pernicious of discriminatory acts because it looks to race alone, treating minority students as fungible commodities that represent a single minority viewpoint. Critical mass, the tipping point of diversity, has no fixed upper bound of universal application, nor is it the minimum threshold at which minority students do not feel isolated or like spokespersons for their race. Grutter defines critical mass by reference to a broader view of diversity rather than by the achievement of a certain quota of minority students. Here, UT Austin has demonstrated a permissible goal of achieving the educational benefits of diversity within that university's distinct mission, not seeking a percentage of minority students that reaches some arbitrary size.

Implicitly conceding the need for holistic review, Fisher offers socioeconomic disadvantage as a race-neutral alternative in holistic review. UT Austin points to widely accepted scholarly work concluding that “there are almost six times as many white students as black students who both come from [socio-economically disadvantaged] families and have test scores that are above the threshold for gaining admission at an academically selective college or university.” ¹²⁸ At bottom, the argument is that minority students are disadvantaged by class, not race; the socioeconomic inquiry is a neutral proxy for race. Bakke accepts that skin color matters—it disadvantages and ought not be relevant but it is. We are ill-equipped to sort out race, class, and socioeconomic structures, and Bakke did not undertake to do so. To the point, we are ill-equipped to disentangle them and conclude that skin color is no longer an index of prejudice; that we would will it does not make it so.

128 Supp. Br. of Appellee at 30 (citing William G. Bowen & Derek Bok, The Shape of the River 51 (1998)).

We are satisfied that UT Austin has demonstrated that race-conscious holistic review is necessary to make the Top Ten Percent Plan workable by patching the holes that a mechanical admissions program leaves in its ability to achieve the rich diversity that contributes to its academic mission—as described by Bakke and Grutter.

B

Over the history of holistic review, its intake of students has declined, minority and non-minority, and changed the profile of the students it admits—the growing number of applicants and increasing take of the Top Ten Percent Plan raises the competitive bar each year, before race is ever considered, for the decreasing number of seats filled by holistic review. Those admitted are those that otherwise would be missed in the diversity mix—for example, those with special talents beyond class rank and identifiable at the admission gate, and minorities with the experience of attending an integrated school with better educational resources.

The data also show that white students are awarded the overwhelming majority of the highly competitive holistic review seats. As we have explained and as shown in Appendix 2, the increasing take of the Top Ten Percent Plan is inherently self-limiting. UT Austin has demonstrated that it is on a path that each year reduces the role of race. After the Top Ten Percent Plan swallowed 81% of the seats available for Texas students in 2008, for example, white Texan students admitted through holistic review occupied an additional 12% of the overall seats. Only 2.4% and 0.9% of the incoming class of Texas high school graduates were Hispanic and black students admitted through holistic review. That is, admission via the holistic review program—overwhelmingly and disproportionally of white students—is highly competitive for minorities and non-minorities alike. These data persuade us of the force of UT Austin's argument that a limited use of race is necessary to target minorities with unique talents and higher test scores to add the diversity envisioned by Bakke to the student body.

Numbers are not controlling but they are relevant to UT Austin's claimed need for holistic review as a necessary component of its admission program. In 2005, the first class that included race and ethnicity in holistic review, 176(29%) of 617 total African–American admitted students were admitted via holistic review.¹²⁹ Following years were similar, with 32% of admitted African–Americans in 2006, 35% in 2007, and 20% in 2008.¹³⁰ Likewise, significant percentages of Hispanic admitted students were admitted through the holistic review program, making up 24% of the admitted Hispanic pool in 2005, 26% in 2006, 25% in 2007, and 15% in 2008.¹³¹ These numbers directly support UT Austin's contention that holistic use of race plays a necessary role in enabling it to achieve diversity while admitting upwards of 80% of its Texas students by facially neutral standards, drawing as it does from a pool not measured solely by class rank in largely segregated schools.

129 See 2008 Top Ten Percent Report at 8 tbl.2 (providing the percentage of students admitted through the Top Ten Percent Plan by racial and ethnic background).

130 Id.

131 Id.

C

Recall the 3.5 AI threshold that excluded Fisher. Holistic review for the colleges to which Fisher applied only admitted applicants—minority or non-minority—with a minimum AI score of 3.5. This effectively added to the mix a pool of applicants from which those colleges could admit students with higher test scores and a higher predicted level of performance, despite being outside the top ten percent of their class, as part of a greater mosaic of talents. Insofar as some dispersion of minority students among many classes and programs is important to realizing the educational benefits of diversity, race-conscious holistic review is a necessary complement to the Top Ten Percent Plan by giving high-scoring minority students a better chance of gaining admission to UT Austin's competitive academic departments. Fisher's proffered solution is for UT Austin's more competitive academic programs to lower their gates. But this misperceives the source of the AI threshold for admission into the competitive colleges: These programs fill 75% of their seats from the pool of students automatically admitted under the Top Ten Percent Plan. The large number of holistic review candidates competing for the quarter of the remaining seats dictates the high AI threshold that all applicants—minority or non-minority—must meet to qualify for admission. Fisher also points to weak dispersal across classes as evidence of UT Austin's pursuit of numbers. It is precisely the opposite. We repeat, holistic review's search is for diversity, as envisioned by Bakke, one benefit of which is its attendant mitigation of the clustering tendencies of the Top Ten Percent Plan.

Fisher responds that, even if necessary, UT Austin could never narrowly tailor a program that achieves classroom diversity. In particular, Fisher suggests that it is impossible to obtain classroom-level diversity without some sort of fixed curriculum or lower school- or major-level standards. This argument again misses the mark by defining diversity only by numbers. UT Austin does not suggest that the end point of this exercise is a specific measure of diversity in every class or every major. Instead, such measures are relevant but not determinative signals of a want of the array of skills needed for diversity. In other words, diversity in the student body surely produces a degree of intra-classroom and intra-major diversity, with the “important and laudable” benefit recognized in Grutter of “classroom discussion [being] livelier, more spirited, and simply more enlightening and interesting when the students have the greatest possible variety of backgrounds.” ¹³² When the holistic review program was modified to be race-conscious, 90% of classes had one or zero African–American students, 46% had one or zero Asian–American students, and 43% had one or zero Hispanic students.¹³³ This represented a decreasing degree of minority classroom dispersion since the adoption of the Top Ten Percent Plan. This does not mean that there will be some set percentage of African–American nuclear physics majors. But this does mean that UT Austin's effort to ensure that African–Americanstudents with a broad array of skills are in the mix is both permissible and necessary.

132 Grutter, 539 U.S. at 330, 123 S.Ct. 2325 (quotation marks and citation omitted).

133 See Defs.' Cross–Mot. Summ. J., Ex. Tab 11 to App., Walker Aff. at ¶ 11, Fisher, 645 F.Supp.2d 587 (No. 08–263), ECF No. 96.

VII

Interlacing the Top Ten Percent Plan, with its dependence upon segregated schools to produce minority enrollment, with a plan that did not consider race until it had a universe of applicants clearing a high hurdle of demonstrated scholastic performance strongly supports UT Austin's assertion that its packaging of the two was necessary in its pursuit of diversity. This hurdle is a product of a growing number of applicants competing for an ever-shrinking number of holistic review seats, creating one of the most competitive admissions processes in the country. And when race enters it is deployed in the holistic manner of Grutter as a factor of a factor. Even then the minority student that receives some boost for her race will have survived a fierce competition. These minorities are in a real sense, along with the non-minorities of this universe, overlooked in a facially neutral Top Ten Percent Plan that considers only class rank. While outside the Top Ten Percent Plan's reach, they represent both high scholastic potential and high achievement in majority-white schools. We are persuaded that their absence would directly blunt efforts for a student body with a rich diversity of talents and experiences.

“Context matters when reviewing race-based governmental action under the Equal Protection Clause,” ¹³⁴ and UT Austin's admissions program is a unique creature. “[S]trict scrutiny must take relevant differences into account”—“[i]ndeed, as [the Court has] explained, that is its fundamental purpose.” ¹³⁵ The precise context of UT Austin's admissions demonstrates that Fisher's charge is belied by this record. Her argument refuses to accept the admission of over 80% of its Texas students without facial consideration of race as any part of narrow tailoring, and critically refuses to accept that the process adopted for the remaining 20% is essential. It rests on the untenable premise that a Grutter plan for 100% of the admissions is to be preferred. UT Austin's efforts to achieve diversity without facial consideration of race, its narrow tailoring of its admission process, in one of the country's largest states, offers no template for others.

134 Grutter, 539 U.S. at 327, 123 S.Ct. 2325 (citing Gomillion v. Lightfoot, 364 U.S. 339, 343–44, 81 S.Ct. 125, 5 L.Ed.2d 110 (1960)).

135 Id. (internal quotation marks and citations omitted).

VIII

In sum, it is suggested that while holistic review may be a necessary and ameliorating complement to the Top Ten Percent Plan, UT Austin has not shown that its holistic review need include any reference to race, this because the Plan produces sufficient numbers of minorities for critical mass. This contention views minorities as a group, abjuring the focus upon individuals—each person's unique potential. Race is relevant to minority and non-minority, notably when candidates have flourished as a minority in their school—whether they are white or black. Grutter reaffirmed that “[j]ust as growing up in a particular region or having particular professional experiences is likely to affect an individual's views, so too is one's own, unique experience of being a racial minority in a society, like our own, in which race still matters.” We are persuaded that to deny UT Austin its limited use of race in its search for holistic diversity would hobble the richness of the educational experience in contradiction of the plain teachings of Bakke and Grutter. The need for such skill sets to complement the draws from majority-white and majority-minority schools flows directly from an understanding of what the Court has made plain diversity is not. To conclude otherwise is to narrow its focus to a tally of skin colors produced in defiance of Justice Kennedy's opinion for the Court which eschewed the narrow metric of numbers and turned the focus upon individuals. This powerful charge does not deny the relevance of race. We find force in the argument that race here is a necessary part, albeit one of many parts, of the decisional matrix where being white in a minority-majority school can set one apart just as being a minority in a majority-white school—not a proffer of societal discrimination in justification for use of race, but a search for students with a range of skills, experiences, and performances—one that will be impaired by turning a blind eye to the differing opportunities offered by the schools from whence they came.

It is settled that instruments of state may pursue facially neutral policies calculated to promote equality of opportunity among students to whom the public schools of Texas assign quite different starting places in the annual race for seats in its flagship university. It is equally settled that universities may use race as part of a holistic admissions program where it cannot otherwise achieve diversity. This interest is compelled by the reality that university education is more the shaping of lives than the filling of heads with facts—the classic assertion of the humanities. Yet the backdrop of our efforts here includes the reality that accepting as permissible policies whose purpose is to achieve a desired racial effect taxes the line between quotas and holistic use of race towards a critical mass. We have hewed this line here, persuaded by UT Austin from this record of its necessary use of race in a holistic process and the want of workable alternatives that would not require even greater use of race, faithful to the content given to it by the Supreme Court. To reject the UT Austin plan is to confound developing principles of neutral affirmative action, looking away from Bakke and Grutter, leaving them in uniform but without command—due only a courtesy salute in passing.

For these reasons, we AFFIRM.

Appendix 1 ¹³⁶

136 Data for 19962005 comes from the 2006 Top Ten Percent Report at 1524 tbl.7a 7j. Data for 2006 comes from the 2007 Top Ten Percent Report. See Office of Admissions, Univ. of Tex. at Austin, Implementation and Results of the Texas Automatic Admissions Law (HB 588) at the University of Texas at Austin: Demographic Analysis of Entering Freshmen Fall 2007 and Academic Performance of Top 10% and Non–Top 10% Students Academic Years 2002–2006 (Report 10), at 20 tbl.7e (Oct. 28, 2007), available at http:// www. utexas. edu/ student/ admissions/ research/ HB 588- Report 10. pdf. Data for 2007 comes from the 2008 Top Ten Percent Report at 16 tbl.7. Data for 2008 comes from the 2009 Top Ten Percent Report at 15 tbl.7.

IMAGE

Appendix 2 ¹³⁷

137 See supra note 136.

IMAGE

EMILIO M. GARZA, Circuit Judge, dissenting:

In vacating our previous opinion, Fisher v. Univ. of Tex. at Austin, 631 F.3d 213 (5th Cir.2011), the Supreme Court clarified the strict scrutiny standard as it applies to cases involving racial classifications in higher education admissions: Now, reviewing courts cannot defer to a state actor's argument that its consideration of race is narrowly tailored to achieve its diversity goals. Fisher v. Univ. of Tex. at Austin, ––– U.S. ––––, 133 S.Ct. 2411, 2420, 186 L.Ed.2d 474 (2013). Although the University has articulated its diversity goal as a “critical mass,” surprisingly, it has failed to define this term in any objective manner. Accordingly, it is impossible to determine whether the University's use of racial classifications in its admissions process is narrowly tailored to its stated goal—essentially, its ends remain unknown.

By holding that the University's use of racial classifications is narrowly tailored, the majority continues to defer impermissibly to the University's claims. This deference is squarely at odds with the central lesson of Fisher. A proper strict scrutiny analysis, affording the University “no deference” on its narrow tailoring claims, compels the conclusion that the University's race-conscious admissions process does not survive strict scrutiny.

I

As a preliminary matter, Fisher has standing to pursue this appeal, but not because, as the majority contends, the Supreme Court's opinion does “not allow our reconsideration [of the issue of standing].” Ante, at 640.

Federal courts have an affirmative duty to verify jurisdiction before proceeding to the merits. Steel Co. v. Citizens for a Better Env't, 523 U.S. 83, 94–95, 118 S.Ct. 1003, 140 L.Ed.2d 210 (1998). Although standing was actively contested before the Supreme Court, and although the Court's opinion is silent about the issue,¹ the Supreme Court has specifically warned against inferring jurisdictional holdings from its opinions not explicitly addressing that subject. See Steel Co., 523 U.S. at 91, 118 S.Ct. 1003. Accordingly, the issue of standing remains open, and this court is obliged to address it. Id. at 94–95, 118 S.Ct. 1003.

1 As is Justice Scalia's concurrence, Fisher, 133 S.Ct. at 2422, Justice Thomas's concurrence, id. at 2422–32, and Justice Ginsburg's dissent, id. at 2432–34.

In our previous opinion, we held that Fisher had standing to “challenge [her] rejection and to seek money damages for [her] injury.” Fisher, 631 F.3d at 217. Only one relevant fact has changed since then-in 2012, Fisher graduated from Louisiana State University. The University contends that by graduating, “her forward-looking request for relief became moot” because she could no longer seek reconsideration of her undergraduate application. Fisher's graduation does not alter our previous standing analysis because, as she correctly observes, that determination did not depend on a claim for forward-looking injunctive relief. Id. We held that Fisher had standing to seek nominal monetary damages, and we should reach the same conclusion now.

The University relies on Texas v. Lesage, 528 U.S. 18, 120 S.Ct. 467, 145 L.Ed.2d 347 (1999) (per curiam), for the proposition that Fisher lacks standing because she would not have been admitted regardless of her race. But even if Lesage is a standing case (which is a debatable premise—the case seems to address statutory liability under § 1983), it does not affect the outcome here. Lesage stands for the proposition that a plaintiff challenging governmental use of racial classifications cannot prevail if “it is undisputed that the government would have made the same decision regardless” of such use. Id. at 21, 120 S.Ct. 467 (emphasis added). The University asserts that Fisher would not have been admitted even if she had a “perfect” PAI score. The majority agrees. Ante, at 639 (“If [Fisher] had been a minority the result would have been the same.”). While Fisher would have been denied admission during the 2008 admissions cycle even if she had a top PAI score, this is not the relevant inquiry. Rather, as Fisher explains, the proper question is whether she would have fallen above the admissions cut-off line if that line had been drawn on a race-neutral distribution of all applicants' scores. This record does not indicate whether Fisher would have been admitted if race were removed from the admissions process altogether. At the least, this is a complex question that is far from “undisputed.” See Lesage, 528 U.S. at 21, 120 S.Ct. 467. Even the University acknowledges that the answer to this question is practically unknowable: It concedes that re-engineering the 2008 admissions process by retroactively removing consideration of race is virtually impossible since race has an immeasurable, yet potentially material, impact on the placement of the final admissions cut-off lines for all programs. In sum, the record does not show that Fisher's rejection under a race-neutral admissions process is “undisputed,” and remanding to the district court could not alter the record in this regard.

The University further challenges Fisher's standing on redressability grounds. The University's theory is that even if Fisher had been admitted through the race-conscious admissions program, and had not suffered the injury of rejection, she still would have paid the non-refundable application fee. Thus, says the University, because the application fee has no causal link to her injury, any judicial relief would fail to provide redress. This argument misconstrues the nature of Fisher's alleged injury—it is not her rejection, but the denial of equal protection of the laws during the admissions decision process. Fisher correctly explains that the application fee represents nominal damages for the alleged constitutional harm stemming from the University's improper use of racial classifications.² Because this harm would have befallen Fisher whether or not she was ultimately admitted to the University, the non-refundable nature of the application fee is irrelevant.³

2 See Memphis Cmty. Sch. Dist. v. Stachura, 477 U.S. 299, 308 n. 11, 106 S.Ct. 2537, 91 L.Ed.2d 249 (1986) (“[N]ominal damages, and not damages based on some undefinable ‘value’ of infringed rights, are the appropriate means of Vindicating' rights whose deprivation has not caused actual, provable injury.”); Devbrow v. Kalu, 705 F.3d 765, 769 (7th Cir.2013) (“[N]ominal damages are available as a remedy ... [for an abstract injury].”).

3 The University's argument that Arizonans for Official English v. Arizona, 520 U.S. 43, 117 S.Ct. 1055, 137 L.Ed.2d 170 (1997) governs Fisher's nominal monetary damages claim is without merit. Fisher does not rely solely on a “general prayer for relief” to save a case otherwise falling outside an Article III case or controversy from dismissal. Id. at 71, 117 S.Ct. 1055. Fisher's original complaint specifically requested monetary damages.

II

Having confirmed our jurisdiction, our task is to apply strict scrutiny without any deference to the University's claims. Because Fisher effected a change in the law of strict scrutiny, and corrected our understanding of that test as applied in Grutter v. Bollinger, 539 U.S. 306, 123 S.Ct. 2325, 156 L.Ed.2d 304 (2003), I first review the current principles governing this “searching examination.” Fisher, 133 S.Ct. at 2420.

The Equal Protection Clause of the Fourteenth Amendment provides that no State shall “deny to any person within its jurisdiction the equal protection of the laws.” U.S. CONST. amend. XIV. It is canonical that the Constitution treats distinctions between citizens based on their race or ethnic origin as suspect, see, e.g., Bolling v. Sharpe, 347 U.S. 497, 499, 74 S.Ct. 693, 98 L.Ed. 884 (1954), and that the Equal Protection Clause “demands that racial classifications ... be subjected to the most rigid scrutiny,” Loving v. Virginia, 388 U.S. 1, 11, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967). Thus, strict scrutiny begins from the fundamental proposition that “any official action that treats a person differently on account of his race or ethnic origin is inherently suspect.” Fullilove v. Klutznick, 448 U.S. 448, 523, 100 S.Ct. 2758, 65 L.Ed.2d 902 (1980) (Stewart, J., dissenting). This is “because racial characteristics so seldom provide a relevant basis for disparate treatment.” Richmond v. J.A. Croson Co., 488 U.S. 469, 505, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989). “Distinctions between citizens solely because of their ancestry are by their very nature odious to a free people.” Fisher, 133 S.Ct. at 2418 (quoting Rice v. Cayetano, 528 U.S. 495, 517, 120 S.Ct. 1044, 145 L.Ed.2d 1007 (2000)).

When a state university makes race-conscious admissions decisions, those decisions are governed by the Equal Protection Clause, even though they may appear well-intended. See Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 297, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978) (opinion of Powell, J.). Simply put, the Constitution does not treat race-conscious admissions programs differently because their stated aim is to help, not to harm.

Under strict scrutiny, a university's use of racial classifications is constitutional only if necessary and narrowly tailored to further a compelling governmental interest. See Grutter, 539 U.S. at 326, 123 S.Ct. 2325. It is well-established that there is a compelling governmental interest in obtaining the educational benefits of a diverse student body. See Bakke, 438 U.S. at 311–12, 98 S.Ct. 2733 (holding that the “attainment of a diverse student body” is a “constitutionally permissible goal for an institution of higher education”). Grutter and Gratz v. Bollinger, 539 U.S. 244, 123 S.Ct. 2411, 156 L.Ed.2d 257 (2003), confirmed this. See Fisher, 133 S.Ct. at 2418.⁴ “The diversity that furthers a compelling [governmental] interest encompasses a far broader array of qualifications and characteristics of which racial or ethnic origin is but a single though important element.” Bakke, 438 U.S. at 315, 98 S.Ct. 2733. Thus, diversity cannot be defined by a “specified percentage of a particular group,” id. at 307, 123 S.Ct. 2325, because such a definition would be “patently unconstitutional racial balancing,” Grutter, 539 U.S. at 330, 123 S.Ct. 2325. In applying strict scrutiny, it is proper for courts to defer to a university's decision to pursue the compelling governmental interest of diversity based on its “educational judgment that such diversity is essential to its educational mission.” Id. at 328, 123 S.Ct. 2325. But, deference to the University is appropriate on this point, and this point alone. Fisher, 133 S.Ct. at 2421.

4 These principles are not challenged in this case. See infra note 8. However, I continue to believe that Grutter's discussion of the “educational benefits of diversity,” drawing directly from the principles established in Bakke, “remains suspended at the highest levels of hypothesis and speculation,” Fisher, 631 F.3d at 255 (Garza, J., specially concurring).

Once a university has decided to pursue this compelling governmental interest, it must prove that the means chosen “to attain diversity are narrowly tailored to that goal.” Fisher, 133 S.Ct. at 2420. In this, the strict scrutiny test takes the familiar form of a “means-to-ends” analysis: The compelling governmental interest is the ends, and the government program or law—here, the University's race-conscious admissions program—is the means. Strict scrutiny places the burden of proving narrow tailoring firmly with the government. See Johnson v. California, 543 U.S. 499, 505, 125 S.Ct. 1141, 160 L.Ed.2d 949 (2005). And, furthermore, narrow tailoring must be established “with clarity.” Fisher, 133 S.Ct. at 2418.

Before this case, the Supreme Court had issued only three major decisions addressing affirmative action in higher education admissions: Bakke, Gratz, and Grutter. In Fisher, the Court made clear that this line of cases does not stand apart from “broader equal protection jurisprudence.” Id. at 2418. Rather, “the analysis and level of scrutiny applied to determine the validity of [a racial classification] do not vary simply because the objective appears acceptable....” Id. at 2421 (quoting Miss. Univ. for Women v. Hogan, 458 U.S. 718, 724 n. 9, 102 S.Ct. 3331, 73 L.Ed.2d 1090 (1982)).

In Fisher, the Supreme Court modified the narrow tailoring calculus applied in higher education affirmative action cases. While the overarching principles from Bakke, Gratz, and Grutter—that a university can have a compelling interest in attaining the educational benefits of diversity, and that its admissions program must be narrowly tailored to serve this interest—were taken “as given,” id. at 2417–18, the Fisher Court altered the application of those principles in a critical way. Now, courts must give “no deference,” to a state actor's assertion that its chosen “means ... to attain diversity are narrowly tailored to that goal.” Id. at 2420. In so doing, the Fisher Court embraced Justice Kennedy's position on “deference” from Grutter. ⁵ Thus, under the current principles governing review of race-conscious admissions programs, providing any deference to a state actor's claim that its use of race is narrowly tailored is “antithetical to strict scrutiny, not consistent with it.” Grutter, 539 U.S. at 394, 123 S.Ct. 2325 (Kennedy, J., dissenting).

5 See Grutter, 539 U.S. at 388, 123 S.Ct. 2325 (Kennedy, J., dissenting) (“[T]he majority proceeds to nullify ... rigorous judicial review, with strict scrutiny as the controlling standard.... The Court confuses deference to a university's definition of its educational objective with deference to the implementation of this goal.”). I agree with the majority that Fisher represents a decisive shift in the law. See ante, at 642 (“Bringing forward Justice Kennedy's dissent in Grutter, the Supreme Court faulted the district court's and this Court's review of UT Austin's means to achieve the permissible goal of diversity....”); see also Erwin Chemerinsky, The Court Affects Each of Us, 16 Green Bag 2d 361, 364 (2013) (“[ Fisher ] adopts a tougher, less sympathetic tone when it comes to affirmative action programs. For example, in Grutter, the Court spoke of the need to defer to the judgment of colleges and universities. In Fisher, the Court said that such deference was appropriate only as to the importance of diversity; there is no deference given as to whether race is necessary to achieve it.”).

Because the higher-education affirmative action cases do not stand apart from “broader equal protection jurisprudence,” Fisher, 133 S.Ct. at 2418, strict scrutiny must be applied with the same analytical rigor deployed in those other contexts. Put simply, there is no special form of strict scrutiny unique to higher education admissions decisions. Accordingly, we must now evaluate narrow tailoring by ensuring that “the means chosen ‘fit’ the [compelling governmental interest] so closely that there is little or no possibility that the motive for the classification was illegitimate racial prejudice or stereotype.” Croson, 488 U.S. at 493, 109 S.Ct. 706.⁶ Narrow tailoring further requires that “the reviewing court verify that it is necessary for a university to use race to achieve the educational benefits of diversity.” Fisher, 133 S.Ct. at 2420 (internal citations and quotations omitted). To do so, we must carefully inquire into whether the University “could achieve sufficient diversity without using racial classifications.” Id. Establishing narrow tailoring does not require the University to show that it exhausted every possible race-neutral option, but it must meet its “ultimate burden of demonstrating, before turning to racial classifications, that available, workable race-neutral alternatives do not suffice.” Id.

6 We need not determine whether the “strong basis in evidence” test from Croson applies in this case. See Croson, 488 U.S. at 510, 109 S.Ct. 706. Even without this test, the University fails to carry its strict scrutiny burden of proving that its race-conscious admissions policy is necessary to further its diversity interest.

Of course, all of the above must be underscored by the principle that using racial classifications is permissible only as a “last resort to achieve a compelling interest.” Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 790, 127 S.Ct. 2738, 168 L.Ed.2d 508 (2007) (Kennedy, J, concurring).⁷

7 Notwithstanding the majority's brief discussion of Schuette v. Coalition to Defend Affirmative Action, Integration and Immigrant Rights and Fight for Equality By Any Means Necessary (BAMN), ––– U.S. ––––, 134 S.Ct. 1623, 188 L.Ed.2d 613 (2014), ante at 651, that case “is not about the constitutionality, or the merits, of race-conscious admissions policies in higher education,” and does not impact the analysis in today's case. Schuette, 134 S.Ct. at 1630.

III

Here, the University has framed its goal as obtaining a “critical mass” of campus diversity. To uphold the use of race under strict scrutiny, courts must find narrow tailoring through a close “fit” between this goal and the admissions program's consideration of race.⁸ Accordingly, the controlling question becomes the definition of “critical mass”—the University's stated goal. In order for us to determine whether its use of racial classifications in the admissions program is narrowly tailored to its goal, the University must explain its goal, and do so “with clarity.” Fisher, 133 S.Ct. at 2418. On this record, it has not done so.

8 The University's decision to pursue the educational benefits of diversity, as established in Bakke, is not challenged in this case. See Fisher, 133 S.Ct. at 2419 (“[T]he parties here do not ask the Court to revisit that aspect of Grutter's holding.”). Our only concern is whether the University's means—its race-conscious holistic admissions program—are narrowly tailored to its diversity objective.

The majority entirely overlooks the University's failure to define its “critical mass” objective for the purposes of assessing narrow tailoring. This is the crux of this case—absent a meaningful explanation of its desired ends, the University cannot prove narrow tailoring under its strict scrutiny burden. Indeed, the majority repeatedly invokes the term “critical mass” without even questioning its definition. See, e.g., ante, at 649 (“But minority representation then remained largely stagnant, within a narrow oscillating band, rather than moving towards a critical mass of minority students.”); id. at 654 (“Achieving the critical mass requisite to diversity goes astray when it drifts to numerical metrics.”); id. (“Fisher refuses to acknowledge this distinction between critical mass—the tipping point of diversity—and a quota.”); id. at 656 (“Critical mass, the tipping point of diversity, has no fixed upper bound of universal application, nor is it the minimum threshold at which minority students do not feel isolated or like spokespersons for their race.”). Under Fisher, it is not enough for a court to simply state, as does the majority, that it is not deferring to the University's narrow tailoring arguments. See, e.g., id., at 646 (“Affording no deference, we look for narrow tailoring....”). Rather, the reviewing court's actual analysis must demonstrate that “no deference” has been afforded. Fisher, 133 S.Ct. at 2420. Here, the majority's failure to make a meaningful inquiry into the nature of “critical mass” constitutes precisely such deference.

Certainly, as explained below, I agree that “critical mass” does not require a precise numerical definition. See infra note 11. But, to meet its narrow tailoring burden, the University must explain its goal to us in some meaningful way. We cannot undertake a rigorous ends-to-means narrow tailoring analysis when the University will not define the ends. We cannot tell whether the admissions program closely “fits” the University's goal when it fails to objectively articulate its goal. Nor can we determine whether considering race is necessary for the University to achieve “critical mass,” or whether there are effective race-neutral alternatives, when it has not described what “critical mass” requires.⁹

9 There is some dispute about whether the University's definition of “critical mass” is even before us as part of our narrow tailoring analysis. The University claims that this issue is outside the scope of the Supreme Court's remand because it is relevant only to its compelling interest in diversity. This contention misunderstands the way in which “critical mass” matters to this case. Here, “critical mass” represents the goal the University purports to seek. The University uses this term as a representation of its ends. Fisher clearly establishes that reviewing courts must defer to the University's decision to pursue such ends. 133 S.Ct. at 2419. But, it equally establishes that we cannot defer to the University's claim that “the means chosen ... to attain diversity are narrowly tailored to that goal.” Id. at 2420. To conduct our own independent assessment of narrow tailoring—the judicial role under strict scrutiny—we must have a clear and definite understanding of the goal the University actually seeks. Accordingly, we must question the University's explanation of “critical mass” to fulfill the task remanded to us by the Supreme Court.

At best, the University's attempted articulations of “critical mass” before this court are subjective, circular, or tautological. See infra Part III.A. The University explains only that its “concept of critical mass is defined by reference to the educational benefits that diversity is designed to produce.” And, in attempting to address when it is likely to achieve critical mass, the University explains only that it will “cease its consideration of race when it determines ... that the educational benefits of diversity can be achieved at UT through a race-neutral policy....” These articulations are insufficient. Under the rigors of strict scrutiny, the judiciary must “verify that it is necessary for a university to use race to achieve the educational benefits of diversity.” Fisher, 133 S.Ct. at 2420 (internal quotations omitted). It is not possible to perform this function when the University's objective is unknown, unmeasurable, or unclear.

The exacting scrutiny required by the Supreme Court's “broader equal protection jurisprudence” is entirely absent from today's opinion, which holds that the University has proven narrow tailoring even though it has failed to meaningfully articulate its diversity goals.

A

The University's failure to define meaningfully its “critical mass” objective is manifest in its various strict scrutiny arguments. The University claims that its use of racial classifications is necessary and narrowly tailored because (1) quantitative metrics reflect an inadequate minority presence; (2) qualitative diversity is lacking; (3) certain selective colleges are insufficiently diverse; (4) its periodic review demonstrates that its goals have not yet been achieved; and (5) its use of racial classifications is almost identical to that approved in Grutter. ¹⁰ Each of these arguments falls short—either overlooking a more narrowly tailored alternative or eliding any articulation of how this specific use of racial classification advances the University's objective.

10 On remand, the University does not specifically delineate these arguments as such. Rather, it submits that these various considerations are sufficient to establish narrow tailoring. In any event, whether taken together or evaluated individually, none of these arguments establishes that the University's use of racial classifications in its admissions decisions is narrowly tailored.

1

First, while not defining its “critical mass” goal with reference to specific quantitative objectives, the University claims that quantitative metrics are relevant in measuring its progress. The University “based its critical mass determination on several data points, including hard data on minority admissions, enrollment, and racial isolation” and found that its use of racial classifications “does increase minority enrollment.” ¹¹ Accepting that such metrics bear some relevance to the University's progress, this is insufficient to satisfy strict scrutiny. The University does not explain how admitting a very small number of minority applicants under the race-conscious admissions plan is necessary to advancing its diversity goal.

11 I agree with the majority's rejection of Fisher's arguments that the University had achieved “critical mass” in 2004, and that “critical mass” can be defined with reference to numbers alone. Fisher effectively asks us to ratify racial quotas, which we cannot, and will not, do. See Bakke, 438 U.S. at 317–18, 98 S.Ct. 2733 (disapproving quota systems and approving the use of race or ethnic background as a “plus” factor).

It is undeniable that the University admits only a small number of minority students under race-conscious holistic review. See Fisher, 631 F.3d at 262–63 (Garza, J., specially concurring). In 2008, the sole year at issue in this case, less than 20% of the class was evaluated under the race-conscious holistic review process. Even if we assume that all minority students who were admitted and enrolled in that year through the race-conscious holistic review process gained admission because of their race, this number is strikingly small—only 216 African–American and Hispanic students in an entering class of 6,322.¹² The University fails to explain how this small group contributes to its “critical mass” objective. “Racial classifications are simply too pernicious to permit any but the most exact connection between justification and [racial] classification.” Adarand, 515 U.S. at 236, 115 S.Ct. 2097 (emphasis added). But here, the University has not established a clear and definite connection between its chosen means and its desired ends of “critical mass.”

12 Notwithstanding the University's contention that 2008 witnessed an “unprecedented surge” in Top Ten Percent Law admissions, this is the only relevant year for purposes of our narrow tailoring analysis. Moreover, I continue to find the majority's use of data for both enrolled and admitted students to be misguided and potentially confusing. See ante, at 657–58. In my view, the proper metric is enrolled students because we are assessing whether the University's means are narrowly tailored to its goal of attaining “the educational benefits of diversity” on campus. See Fisher, 631 F.3d at 260 n. 18 (Garza, J., specially concurring).

To be clear, I agree that a race-conscious admissions plan need not have a “dramatic or lopsided impact” on minority enrollment numbers to survive strict scrutiny, as the University reads Fisher's arguments to suggest. But neither can the University prove the necessity of its racial classification without meaningfully explaining how a small, marginal increase in minority admissions is necessary to achieving its diversity goals. Thus, neither the small (and decreasing) percentage of minority holistic-review admittees, nor minorities' “under-representation” in holistic review admissions relative to whites, taken alone, demonstrates narrow tailoring. See ante, at 646–47 & Appendix 1 (explaining that white students comprise a larger percentage of holistic review admittees than of the incoming class as a whole).¹³

13 For example, for the incoming class that enrolled in 2008, white students comprised 65% of all students admitted through holistic review, but only 52% of the entire incoming class.

Under the Equal Protection Clause, diversity cannot be assessed by strictly quantitative metrics, and, to the extent that numbers could be relevant in assessing “critical mass,” the University leaves this relevance entirely unexplained.

2

The University advances a second understanding of “critical mass,” which I will refer to as “qualitative.” Under this theory, the University says its goal is not boosting minority enrollment numbers alone, but rather promoting the quality of minority enrollment—in short, diversity within diversity. The University submits that its race-conscious holistic review allows it to select for “other types of diversity” beyond race alone, and to identify the most “talented, academically promising, and well-rounded” minority students. According to the University, these are crucial “change agents” who debunk stereotypes but who may fall outside the top 10% of their high school classes.

As a preliminary matter, these stated ends are too imprecise to permit the requisite strict scrutiny analysis. The University has not provided any concrete targets for admitting more minority students possessing these unique qualitative-diversity characteristics—that is, the “other types of diversity” beyond race alone. At what point would this qualitative diversity target be achieved? Because its ends are unknown to us, the University cannot meet its strict scrutiny burden.

But, even accepting the University's broad and generic qualitative diversity ends, we cannot conclude that the race-conscious policy is constitutionally “necessary.” The University has not shown that qualitative diversity is absent among the minority students admitted under the race-neutral Top Ten Percent Law, Tex. Educ.Code Ann. § 51.803 (West 2009). That is, the University does not evaluate the diversity present in this group before deploying racial classifications to fill the remaining seats. The University does not assess whether Top Ten Percent Law admittees exhibit sufficient diversity within diversity, whether the requisite “change agents” are among them, and whether these admittees are able, collectively or individually, to combat pernicious stereotypes. There is no such evaluation despite the fact that Top Ten Percent Law admittees also submit applications with essays, and are even assigned PAI scores for purposes of admission to individual schools. ¹⁴ Evaluating the composition of these admittees—80% of the class in 2008—before deploying racial classifications in the holistic admissions program might well reveal that racial classifications are not necessary to achieve the University's qualitative diversity goals. See Fisher, 133 S.Ct. at 2420; see also Parents Involved, 551 U.S. at 790, 127 S.Ct. 2738 (explaining that racial classifications must be a “last resort to achieve a compelling interest” in order to survive strict scrutiny) (Kennedy, J., concurring).

14 Dr. Kedra Ishop, Associate Director of Admissions, explained that all applicant files are assigned an AI and PAI, and that the AI and PAI of a Top Ten Percent Law applicant can still determine the program to which she is admitted, if her class rank is not high enough for automatic admission to a competitive first-choice program such as the School of Business.

In effect, the University asks this Court to assume that minorities admitted under the Top Ten Percent Law do not demonstrate “diversity within diversity”—that they are somehow more homogenous, less dynamic, and more undesirably stereotypical than those admitted under holistic review. Thus, the University claims, absent its race-conscious holistic admissions program, it would lose the minority students necessary to achieving a qualitative critical mass. But it offers no evidence in the record to prove this, and we must therefore refuse to make this assumption.

Regrettably, the majority firmly adopts this assumption—that minority students from majority-minority Texas high schools are inherently limited in their ability to contribute to the University's vision of a diverse student body.¹⁵ The majority reasons that race-conscious holistic review is a “necessary complement,” ante, at 654, to the Top Ten Percent Law, which, on its own, would admit insufficient “students of unique talents and backgrounds who can enrich the diversity of the student body in distinct ways,” id., at 654. The majority's discussion of numerous “resegregated” Texas school districts is premised on the dangerous assumption that students from those districts (at least those in the top ten percent of each class) do not possess the qualities necessary for the University of Texas to establish a meaningful campus diversity. See id., at 650–52. In this, it has embraced the very ill that the Equal Protection Clause seeks to banish. See Croson, 488 U.S. at 505, 109 S.Ct. 706 (“[R]acial characteristics so seldom provide a relevant basis for disparate treatment.”); Fullilove, 448 U.S. at 523, 100 S.Ct. 2758 (“[A]ny official action that treats a person differently on account of his race or ethnic origin is inherently suspect.”) (Stewart, J., dissenting); Cayetano, 528 U.S. at 517, 120 S.Ct. 1044 (“Distinctions between citizens solely because of their ancestry are by their very nature odious to a free people....”).

15 See ante, at 651 (discussing the “outcome gaps” of “segregated urban schools”); id. at 651 (classifying schools according to their racial and ethnic compositions). Additionally, the majority's sua sponte survey of Texas school districts' data on racial composition, test scores, and educational outcomes, id. at 651–53, ventures far beyond the summary judgment record. Under strict scrutiny, the government bears the burden of establishing compliance with the Constitution. See Johnson, 543 U.S. at 505, 125 S.Ct. 1141. More specifically, the Supreme Court's opinion has mandated that we decide whether “ this record ... is sufficient” to demonstrate narrow tailoring. Fisher, 133 S.Ct. at 2421 (emphasis added).

Moreover, the only fact from which the majority draws this alarming conclusion is the mere reality that these districts serve majority-minority communities. Ante, at 650–51 (“The de facto segregation of schools in Texas enables the Top Ten Percent Law to increase minorities in the mix, while ignoring contributions to diversity beyond race.”).¹⁶ By accepting the University's standing presumption that minority students admitted under the Top Ten Percent Law do not possess the characteristics necessary to achieve a campus environment defined by “qualitative diversity,” the majority engages in the very stereotyping that the Equal Protection Clause abhors.¹⁷

16 The majority's reductionist assumption about the experiences of minority students admitted under the Top Ten Percent Law is startling: “The top 29 graduates from Jack Yates High School in Houston live in the same predominately African–American neighborhood of that city's Third Ward, and thus likely experienced a similar cultural environment.” Ante, at 650 n. 98.

17 This stereotyping is not limited to minority students admitted under the Top Ten Percent Law. The majority further assumes that minority students admitted under holistic review, based on their “experience of being a minority in a majority-white ... school,” likely “demonstra[te] qualities of leadership and sense of self.” Ante, at 653. These conclusions are nonetheless stereotypes disallowed by the Fourteenth Amendment. And in any event, this record, by which we are bound, does not indicate that any minority students admitted under holistic review come from majority-white schools. See supra note 15.

The record does not indicate that the University evaluates students admitted under the Top Ten Percent Law, checking for indicia of qualitative diversity—diversity within diversity—before determining that race should be considered in the holistic review process to fill the remaining seats in the class. If the Top Ten Percent Law admittees were a sufficiently qualitatively diverse population, which they may well be so far as I can tell, then using race in holistic review to promote further diversity might not be necessary for the University to achieve its goal, and an up-front assessment of these admittees, before turning to race, could be a more narrowly-tailored option. And, in any event, the University offers no method for this court to determine when, if ever, its goal (which remains undefined) for qualitative diversity will be reached. Accordingly, the University has failed to carry its strict scrutiny burden of proving that its race-conscious admissions policy is necessary to achieving its diversity objective of a “qualitative” critical mass.

3

In earlier stages of this case, the University framed its diversity goal as achieving “classroom diversity.” The University suggested that classroom diversity and the distribution of minority students among colleges and majors were meaningful metrics in determining whether “critical mass” had been attained. And, indeed, the Supreme Court has recognized that increased diversity of perspectives in the classroom provides for a “livelier, more spirited, and simply more enlightening and interesting” experience. Grutter, 539 U.S. at 330, 123 S.Ct. 2325 (quoting Bakke, 438 U.S. at 307, 98 S.Ct. 2733). However, the University has distanced itself from this previously asserted goal, now claiming it “has never pursued classroom diversity as a discrete interest or endpoint,” but merely as “one of many factors” to be considered in evaluating diversity. Given the University's failure to press the classroom diversity argument in its briefing on remand, the issue is almost certainly waived. See United States v. Griffith, 522 F.3d 607, 610 (5th Cir.2008) (“It is a well-worn principle that the failure to raise an issue on appeal constitutes waiver of that argument.”).

Notwithstanding this waiver, the majority addresses the issue of classroom diversity, contending that the University's race-conscious admissions policy is necessary to give “high-scoring minority students a better chance of gaining admission to UT Austin's competitive academic departments.” Ante, at 658. Perhaps, based on the structure of the University's admissions process, it is possible that the use of race as a factor in calculating an applicant's PAI score incrementally increases the odds that a minority applicant will be admitted to a competitive college within the University.¹⁸ But hypothetical considerationsare not enough to meet a state actor's burden under strict scrutiny. See United States v. Virginia, 518 U.S. 515, 533, 116 S.Ct. 2264, 135 L.Ed.2d 735 (1996) (“The justification must be genuine, not hypothesized or invented post hoc in response to litigation.”). Rather, assuming that the University's diversity goal is establishing classroom diversity, it is the University that bears the burden of proving that the use of race in calculating the PAI scores is necessary to furthering this goal. But instead of explaining how race enhances minority students' prospects of admission to a competitive college or major, the University admissions officers' deposition testimony specifically indicates that race could not be a decisive factor in any applicant's admission,¹⁹ and that it is impossible to determine whether race was in fact decisive for any particular applicant's admission decision.²⁰ Absent any record evidence of the potential for race to be a decisive factor, the University cannot establish, as the majority claims, that its racial classifications could actually give any minority applicant “a better chance” of admission to a competitive college. Ante, at 658.

18 The record describes the admissions process as follows: First, the admissions staff read all applicants' files, including those of Top Ten Percent Law applicants, and assign each an AI and PAI score. Applicants with exceptionally high class rank or AI scores are automatically admitted to certain first-choice schools or majors and, thus, also to the University. Next, for applicants not automatically admitted to their first choice, the staff generate a matrix for each school with each cell on the matrix representing an intersection of AI and PAI scores. Working with liaisons from each school, the staff plot the remaining applicants' scores on matrices according to the applicants' first-choice majors. Based on the number of applicants in each matrix cell and the available seats in the class for each school, the admissions staff and liaisons draw “cut-off lines” across the matrices. Applicants not selected for admission to their first-choice school “cascade” onto the matrix for their second-choice school, where they are added to the cells along with applicants who were above the cut-off line during the previous review round. The cut-off lines are readjusted to accommodate the additional students, and those remaining above the adjusted cut-off lines are accepted to that school. Applicants not admitted to either their first- or second-choice school then “cascade” into the Liberal Arts Undecided matrix, which serves as the default third-choice major. Again, the admissions staff perform the line-drawing exercise (cognizant that remaining Top Ten Percent Law applicants must be admitted as Liberal Arts majors, thus reducing the number of available spaces), and a final determination is made for all applicants.

19 When asked whether any one factor in the PAI calculation could be determinative for an applicant's admission, Dr. Bruce Walker, Vice Provost and Director of Admissions, stated “no.”

20 Dr. Kedra Ishop was asked whether she could give an “example where race would have some impact on an applicant's personal achievement score?” Her answer: “In order to—it's impossible to say—to give you an example of a particular student because it's all contextual.”

In short, the University has obscured its use of race to the point that even its own officers cannot explain the impact of race on admission to competitive colleges.²¹ If race is indeed without a discernable impact, the University cannot carry its burden of proving that race-conscious holistic review is necessary to achieving classroom diversity (or, for that matter, any kind of diversity). Because the role played by race in the admissions decision is essentially unknowable, I cannot find that these racial classifications are necessary or narrowly tailored to achieving the University's interest in diversity.

21 And race is entirely invisible at the moment of drawing the final admission cut-off lines, for students not automatically admitted to their first-choice program by virtue of an exceptionally high class rank or AI score. The University's admissions staff and liaisons from each school admit students to the various schools and majors based solely on the combination of applicants' AI and PAI scores. While race is considered in determining PAI scores, once the scores are assigned and applicants are plotted on the matrices for the various schools, admissions officers treat applicants as points on a grid. In other words, the University officials have no way of knowing whether they are selecting applicants whose race incrementally boosted their PAI score, much less whether any particular applicant will help the University improve classroom diversity.

4

The University further claims that its race-conscious admissions program is narrowly tailored because, with the help of a rigorous periodic review system, it will “cease its consideration of race when it determines ... that the educational benefits of diversity can be achieved at [the University] through a race-neutral policy ‘at reasonable cost’ to its other educational objectives.” The University seeks to assure us that periodic review of its admissions policy considers enrollment data, “evidence of racial isolation and the racial climate on campus,” and “other data including the educational benefits of diversity experienced in the classroom.” In simple language, the University asserts that it knows critical mass when it sees it.

On one level, the University's review process captures the essence of the holistic diversity interest established in Bakke, validated in Grutter, and left intact by Fisher, See Ante at 643 (“Diversity is a composite of the backgrounds, experiences, achievements, and hardships of students to which race only contributes.”). In fact, the Grutter Court discussed the important role that such reviews can play in determining whether racial classifications have continuing necessity under strict scrutiny. Grutter, 539 U.S. at 342, 123 S.Ct. 2325.

Nonetheless, there are two distinct flaws with the University's assurances that its own, internal, periodic review is sufficient to safeguard against any unconstitutional use of race. First, strict scrutiny does not allow the judiciary to delegate wholesale to state actors the task of determining whether a race-conscious admissions policy continues to be necessary. This is the very point made by the Fisher Court, in vacating our previous opinion for deferring to the University's narrow-tailoring claims. Fisher, 133 S.Ct. at 2420–21.

Second, while the University correctly considers a range of factors in its assessment of the necessity of its use of race, see Bakke, 438 U.S. at 315, 98 S.Ct. 2733 (describing diversity as a “broader array of qualifications and characteristics” of which race is only one), it has still not explained to us how this consideration takes place. In describing its periodic review process, the University never explains how the various factors are measured, the weight afforded to each, and what combination thereof would yield a “critical mass” of diversity sufficient to cease use of racial classifications.

In light of this, I cannot determine that the race-conscious admissions program is narrowly tailored to the University's goal. The University, in effect, defines critical mass as a nebulous amalgam of factors—enrollment data, racial isolation, racial climate, and “the educational benefits of diversity”—that its internal periodic review is calibrated to detect. But, without more, the University fails to prove narrow tailoring with clarity. Fisher, 133 S.Ct. at 2418. Such a bare submission, in essence, begs for the deference that is irreconcilable with “meaningful” judicial review. Id. at 2421.

5

Lastly, the University submits that its race-conscious admissions policy necessarily satisfies narrow tailoring because it is closely modeled on the admissions program upheld by the Supreme Court in Grutter. Similarly, the majority implies that the race-conscious admissions policy's similarity to Grutter is, itself, a meaningful factor in our strict scrutiny analysis. ²² This claim is unpersuasive.

22 See ante, at 645 (describing the University's use of race with direct reference to the program approved in Grutter ); id. at 653 (“UT Austin's holistic review program—a program nearly indistinguishable from the University of Michigan Law School's program in Grutter ...”).

Fisher confirms that we are obligated to consider the particular challenged race-conscious program on its own terms and ask whether the University “could achieve sufficient diversity without using racial classifications.” 133 S.Ct. at 2420. Strict scrutiny is not a hypothetical undertaking, but rather “imposes on the university the ultimate burden of demonstrating, before turning to racial classifications, that available, workable race-neutral alternatives do not suffice.” Id.

Certain aspects of the University's admissions policy do parallel the features of the plan upheld in Grutter—race is only a sub-factor within a holistic, individualized review process, and the University's goal is framed in terms of “critical mass.” But the University, under mandate by the Texas Legislature's Top Ten Percent Law, admits the majority of its entering class through a separate, race-neutral scheme.²³ This inevitably impacts the narrow tailoring calculus presently under consideration. That is, while the University's race-conscious admissions policy is conceptually derived from the University of Michigan Law School's approach, the two are quite distinct in practice: The University's holistic review coexists with a separate process that admits a large population of students, a circumstance not contemplated in Grutter. ²⁴

23 The majority implies that the University's implementation of the Top Ten Percent Law was discretionary. See ante, at 645 (“UT Austin turned to the Top Ten Percent Plan....” (emphasis added)); id. at 645 (“We are offered no coherent response to the validity of a potentially different election by UT Austin: to invert the process and use Grutter' s holistic review to select 80% or all of its students.” (emphasis added)). There was no such choice; the University was mandated by the law to admit any graduate in the top ten percent of his or her high school class. And, as explained below, the Top Ten Percent Law is not challenged in this appeal.

24 Additionally, I observe that the admissions program here and that in Grutter do not seem to use race in the same way. Even accepting that the University uses race as a “factor of a factor of a factor,” here, the University incorporates race into the PAI before individual admissions decisions are made on the matrices, at which point race is invisible. See supra note 21. By contrast, in Grutter, each law school applicant's file, including his or her racial classification, was considered during a holistic, full-file review. See 539 U.S. at 334–36, 123 S.Ct. 2325.

Similarity to Grutter is not a narrow-tailoring talisman that insulates the University's policy from strict scrutiny. The University's burden is to prove that its own use of racial classifications is necessary and narrowly tailored for achieving its own diversity objectives.

B

Ultimately, the record is devoid of any specifically articulated connection between the University's diversity goal of “critical mass” and its race-conscious admissions process. The University has not shown how it determines the existence, or lack, of a “critical mass” of diversity in its student population. Rather, the University only frames its goal as “obtaining the educational benefits of diversity.” This is entirely circular reasoning that cannot satisfy the rigorous means-to-ends analysis required under strict scrutiny. Fisher, 133 S.Ct. at 2421.

To be clear, my concern is not with the University's use of the term “critical mass” itself. Even if the University were to adopt another rhetorical construct to explain its diversity objectives, it faces the same underlying problem—it does not offer a clear and definite articulation of its goal sufficient for a reviewing court to verify narrow tailoring. The University's failure to meet its strict scrutiny burden is a function of its undefined ends, not its choice to label those ends as “critical mass.”

IV

The majority concludes that the University's race-conscious admissions program is narrowly tailored because the University has exhausted all workable alternatives. Ante, at 660. Much of today's opinion explores the historical “narrative” of the University's admissions process, including many race-neutral recruitment programs intended to bolster minority enrollment. Id. at 645. And, indeed, the University's many efforts to achieve a diverse campus learning environment without resorting to racial classifications are commendable. But, framing this history as something akin to a process of elimination, the majority finds that the University's race-conscious admissions program must be necessary and narrowly tailored to the University's diversity objectives. This is insufficient to satisfy strict scrutiny.

Certainly, the University's past experiences with race-neutral initiatives are relevant to the inquiry because the University must establish that “no workable race-neutral alternatives would produce the educational benefits of diversity,” and because the University's “experience and expertise” provide some context to inform judicial review. Fisher, 133 S.Ct. at 2420. However, we cannot conclude that the University's current race-conscious admissions program—the only matter before this court—is narrowly tailored to achieve the educational benefits of diversity because the University has failed to define what it means by “critical mass.” In other words, the University's long history of purportedly unsuccessful alternatives is meaningless if we cannot discern the contours of the success it now seeks.

Additionally, the majority's sustained focus on the Top Ten Percent Law is misplaced. While the Law is indeed central to this case, here, as in our previous consideration of this appeal, “[n]o party challenged, in the district court or in this court, the validity or the wisdom of the Top Ten Percent Law.” Fisher, 631 F.3d at 247 (King, J., specially concurring). Nevertheless, the majority forcefully indicts the Law for frustrating the University's efforts to achieve well-rounded diversity. In the majority's view, the Law's shortcomings make a holistic review program more necessary. Ante, at 654 (“We are persuaded that holistic review is a necessary complement to the Top Ten Percent Plan, enabling it to operate without reducing itself to a cover for a quota system....”). At most, the Law's mechanical operation—admitting students based on the sole metric of high school class rank—might suggest that some form of holistic review is advisable to supplement the admissions process. But this issue is not before us at all. Our task is to determine whether the University's injection of race into its admissions process survives strict scrutiny.

The Top Ten Percent Law matters only insofar as it causes the University to admit a large number of minority students separate and apart from the holistic review process. That is, the Law creates a separate admissions channel for many minority students, which then calls into question the necessity of using race as a factor in the holistic review process for filling the remaining seats. Whether, in light of the Top Ten Percent Law, race-conscious holistic review is more or less necessary is an open question, and it is the University that bears the burden of explaining how the Law impacts its achievement of its diversity goal. Here, it has failed to do so, under any theory of “critical mass” it has proffered.²⁵

25 There are additional elements of the majority's discussion of the Top Ten Percent Law that I cannot join. First, to bolster the “necessity” of race-conscious holistic review, the majority explains that holistic-review admittees have higher standardized test scores. Ante, at 650 & nn. 96–97. However, no testimony or record evidence establishes whether the gap in SAT scores between Top Ten Percent and Non–Top Ten Percent admittees is statistically significant. And as the University's president explained in 2000, “top 10 percent high school students make much higher grades in college than non-top 10 percent students,” and “[s]trong academic performance in high school is an even better predictor of success in college than standardized test scores.” At best, the academic superiority of holistic review admittees as a group is highly contested. Second, legislative changes to the Top Ten Percent Law after 2008, the relevant year for our purposes, are not germane to our analysis. See ante, at 655 (discussing S.B. 175).

* * *

The material facts of this case have remained unchanged since the district court's grant of summary judgment, but the governing law has changed markedly. Fisher established that strict scrutiny in the higher education affirmative action setting is no different than strict scrutiny in other equal protection contexts—the state actor receives no deference in proving that its chosen race-conscious means are narrowly tailored to its ends. The majority fails to give Fisher its proper weight. Today's opinion sidesteps the new strict scrutiny standard and continues to defer to the University's claims that its use of racial classifications is narrowly tailored to its diversity goal. Because the University has not defined its diversity goal in any meaningful way—instead, reflexively reciting the term “critical mass”—it is altogether impossible to determine whether its use of racial classifications is narrowly tailored.

This is not to say, however, that it is impossible for a public university to define its diversity ends adequately for a court to verify narrow tailoring with the requisite exacting scrutiny. After all, “[s]trict scrutiny must not be strict in theory but fatal in fact.” Fisher, 133 S.Ct. at 2421 (internal quotations omitted). It may even be possible for a university to do so while seeking a “critical mass.” What matters now, after Fisher, is that a state actor's diversity goals must be sufficiently clear and definite such that a reviewing court can assess, without deference, whether its particular use of racial classifications is necessary and narrowly tailored to those goals. On this record, the University has not “offered sufficient evidence that would prove that its admissions program is narrowly tailored to obtain the educational benefits of diversity.” Fisher, 133 S.Ct. at 2421. Accordingly, I would reverse and render judgment for Fisher.

I respectfully dissent.