Jennifer L. JOHNSON, and all others similarly situated, Aimee Bogrow, et al., Plaintiffs-Appellees-Cross-Appellants, v. BOARD OF REGENTS OF THE UNIVERSITY OF GEORGIA, d.b.a. University of Georgia, Defendant-Appellant-Cross-Appellee, Antoine Hester, et al., Intervenors-Defendants, Appellants-Cross-Appellees.
Nos. 00-14340, 00-14382.
-- August 27, 2001
Dennis R. Dunn,Georgia Dept. of Law, Atlanta, GA, Janell M. Byrd, NAACP Legal Defense & Educ. Fund, Inc., Washington, DC, Elise C. Boddie, Theodore Michael Shaw, Dennis Parker, NAACP Legal Defense & Educ. Fund, Inc., New York City, Mark H. Cohen, Atlanta, GA, for Defendants.A. Lee Parks, Jr., John W. Bellflower, Parks, Chesin & Miller, P.C., Atlanta, GA, Kevin V. Parsons, Parks, Chesin & Miller, P.C., Charlotte, NC, for Plaintiffs.Teresa Kwong, U.S. Dept. of Justice, Washington, DC, for Amicus Curiae U.S.Leslie M. Gross, Lawyers' Committee for Civil Rights, Washington, DC, for Amicus Curiae Southern Regional Council.Patricia A. Brannan, Hogan & Hartson, Washington, DC, for Amicus Curiae American Council on Educ.Anthony T. Caso, Pacific Legal Foundation, Sacramento, CA, for Amicus Curiae, Pacific Legal Foundation.Harry W. MacDougald, Atlanta, GA, Mark L. Gross, U.S. Dept. of Justice, Washington, DC, Stephen R. McCutcheon, Jr., Pacific Legal Foundation, Sacramento, CA, for Amicus Curiae, Southeastern Legal Foundation.
In this case, we consider a challenge to the University of Georgia's freshman admissions policy, and specifically that policy's preferential treatment of non-white applicants. The three Plaintiffs are white females who applied unsuccessfully for admission to the University's Fall 1999 class. Plaintiffs allege, and Defendants do not dispute, that the University's admissions policy awarded a fixed numerical bonus to non-white and male applicants that it did not give to white and female applicants. The district court found the policy unlawful and entered summary judgment in Plaintiffs' favor. The court declined, however, to enter a prospective injunction forbidding the University from ever considering race or gender in the freshman admissions process. On appeal, Defendants do not challenge the district court's ruling regarding the University's preferential treatment of males, but do appeal the ruling regarding the University's preferential treatment of non-whites. According to the Defendants, the University's freshman admissions policy does not unlawfully discriminate on the basis of race because the policy is narrowly tailored to serve a compelling interest in ensuring a diverse student body. Plaintiffs cross-appeal on several matters, including the denial of prospective injunctive relief.
After careful review of the record and the parties' arguments, we affirm the entirety of the district court's rulings, although we find the University's 1999 freshman admissions policy unconstitutional for a reason different than that adopted by the district court. The district court found the admissions policy unlawful because, in its view, student body diversity is not a compelling interest sufficient to withstand the strict scrutiny that courts must apply to government decision-making based on race. We need not, and do not, decide that issue, because even assuming that student body diversity is a compelling interest, the University's 1999 freshman admissions policy is not narrowly tailored to achieve this interest. A policy that mechanically awards an arbitrary “diversity” bonus to each and every non-white applicant at a decisive stage in the admissions process, and severely limits the range of other factors relevant to diversity that may be considered at that stage, fails strict scrutiny and violates the Equal Protection Clause of the Fourteenth Amendment.
The three Plaintiffs filed this action in August 1999, challenging the policy employed by the University of Georgia (“UGA”) to determine which applicants would be admitted to the freshman class entering in the Fall of 1999. Plaintiff Jennifer Johnson's complaint was filed separately, and eventually was consolidated with the complaint of Plaintiffs Aimee Bogrow and Molly Ann Beckenhauer.1 All three Plaintiffs had recently been denied admission to UGA, and therefore were, at or about the time of filing their complaints, attending other colleges. Johnson was offered admission to UGA after filing this lawsuit, but she declined to enroll at that time.
Plaintiffs alleged that UGA's intentional use of race violated the Equal Protection Clause of the Fourteenth Amendment as well as 42 U.S.C. § 1981 and Title VI of the Civil Rights Act of 1964, 42 U.S.C. § 2000d; they alleged that UGA's use of gender violated Equal Protection and Title IX.2 Named as Defendants were the Board of Regents of the University System of Georgia; Stephen Portch, Chancellor of the University of Georgia System; and Michael Adams, UGA's President. Plaintiffs sought a variety of remedies, including an injunction compelling their admission to UGA; prospective injunctive relief against the future use of race and gender in the freshman admissions process; certification of class for purposes of that relief; and damages. On December 15, 1999, various African-American individuals who either were students at UGA or who intended to apply as freshmen were permitted to intervene (the “Intervenors”).
In November 1999, the district court preliminarily granted the Plaintiffs' motion to certify a class seeking to enjoin the use of race or gender in the freshman admissions process. The class consisted of “all those similarly situated past, present, and future applicants to UGA's freshman class denied admission or consideration for admission because of their race and/or gender.” The district court's certification order was entered before the Defendants were even given the opportunity to oppose the certification requests of Plaintiffs Bogrow and Beckenhauer. The district court also ruled that the damages claims against Portch and Adams in their individual capacities were barred by qualified immunity.
Both parties then moved for partial reconsideration. On February 9, 2000, the district court reaffirmed its qualified immunity decision, but vacated the class certification order, on the ground that the Plaintiffs lacked standing to obtain prospective injunctive relief and hence could not represent a class seeking that relief. In light of that ruling, the court also dismissed Plaintiffs' individual claims for prospective injunctive relief.
In February 2000, with discovery underway and the parties proceeding toward summary judgment motions, the Intervenors moved for a “special case management scheduling order” or alternatively a three month extension of discovery. The district court denied the motion, which was opposed by the Plaintiffs and UGA primarily on the ground that the Intervenors' proposed changes to the pre-trial schedule would unduly complicate and delay resolution of the case. Summary judgment motions were then filed by the Plaintiffs, Defendants, and Intervenors. On June 16, 2000, the district court dismissed the claims against the individual Defendants in their official capacities, leaving the Board of Regents as the only Defendant.
On July 24, 2000, the district court entered its summary judgment order, denying the motions of the Defendants and the Intervenors and granting in part the Plaintiffs' motion. 106 F.Supp.2d 1362 (S.D.Ga.2000). In pertinent part, the district court found that UGA's consideration of race in its 1999 freshman admissions policy violated Title VI, which the court analyzed as identical to Equal Protection in this context.3 The district court first reasoned that Justice Powell's opinion in Regents of the University of California v. Bakke, 438 U.S. 265, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978), was not binding precedent, and hence the court was not required to assume that the desire to foster student body diversity-the only interest asserted by UGA-was a compelling interest. The district court then reasoned that the Supreme Court's post-Bakke decisions have demonstrated a hostility to identifying diversity as a compelling interest.
The district court next explained why, in its view, UGA's asserted interest in diversity was too “amorphous” to support racial discrimination. According to the court, “[t]he record shows that UGA is plying a ‘diversity = proportionalism’ rationale,” 106 F.Supp.2d at 1371, presumably meaning that UGA's real interest was not diversity, but rather obtaining a percentage of non-white students equivalent to the representation of these non-white groups in the population at large. In addition, the district court discounted the benefits of diversity attested to by former UGA President Charles Knapp, describing Knapp's testimony as “syllogism and speculation.” Id. at 1372. Because the district court did not actually cite any evidence contradicting Knapp's testimony on the perceived benefits of student body diversity, the district court's order may fairly be read as rejecting diversity as a compelling interest in all cases of this kind.
Having found that student body diversity is not a compelling interest, the district court did not reach the question of whether UGA's 1999 freshman admissions policy is narrowly tailored to achieve that interest, other than to opine that UGA's asserted diversity interest is “so inherently formless and malleable that no plan can be narrowly tailored to fit it.” Id. at 1374 (emphasis in original). With respect to remedies, the district court rejected the Intervenors' argument that the three Plaintiffs would have been denied admission even if race and gender were not factors. The court then considered damages, and eventually awarded $7,184.93 to Beckenhauer, $2,060.18 to Johnson, and $1 to Bogrow. The court also directed “UGA, together with its officers, agents, and employees, to offer Aimee Bogrow and Molly Ann Beckenhauer admission for the Fall 2000 semester, and to keep its admission offer to Jennifer L. Johnson open for the Fall 2000 semester.” Id. at 1381. It appears that Bogrow is currently attending UGA, and Johnson expects to do so this Fall.
The relevant facts are largely undisputed. UGA is the flagship institution of Georgia's university system. For the first 160 years of its existence, no African-American student was admitted to UGA. The first African-American students were admitted in 1961. In 1969, the federal government, through the Office of Civil Rights (“OCR”), determined that Georgia's university system was still “operating a dual track of higher education based on race in that past patterns of racial segregation have not been eliminated from most of the institutions within the system.” In 1970, OCR ordered the Board of Regents to submit a desegregation plan and to adopt necessary affirmative action programs to alleviate the vestiges of discrimination. Among other things, the programs eventually implemented sought to increase the number of African-American students at Georgia's traditionally white educational institutions.
Notably, by March 1989, OCR advised the State of Georgia that the university system had substantially complied with the prescribed remedial measures, and therefore “Georgia's system of public higher education is now in compliance with Title VI, and no additional desegregation measures will be required by OCR.” OCR did advise that Georgia was required to maintain compliance with Title VI and to avoid “[d]iscrimination on the basis of race, color, or national origin.”
Admission to UGA is competitive, and applications far exceed the number of available freshman seats. To assemble a class, the faculty admission committee, in conjunction with the admissions office, recommends a freshman admission policy each year. This policy is formally presented to UGA's president for approval, and thereafter is implemented by the admissions office.
Between 1990 and 1995, UGA's freshman admissions policy applied objective academic criteria differently depending upon whether an applicant characterized herself as “black” or “non-black.” To be eligible for admission, an applicant had to meet certain pre-set minimums with respect to Scholastic Aptitude Test (“SAT”) scores, grade point average (“GPA”), and academic index (“AI”).4 Under the 1990-95 policy, the minimums for black students were set lower than the minimums for non-black students. In 1995, UGA-concerned about the constitutionality of its dual-track admissions policy-revised that policy for the 1996 freshman class. The framework of the revised policy is the same as that of the 1999 policy at issue today.
The revised policy divides the admissions process into three stages. UGA selects the majority of its freshman class at an initial stage which applies objective academic criteria without regard to the applicant's race. At this initial stage (the “First Notice” stage), UGA admits automatically applicants whose AIs and SAT scores are above a certain number. From the remaining applications, UGA selects for further evaluation a group of applicants whose AIs are above a certain number and who meet minimum SAT score requirements. Applicants who fall below the minimum AI or SAT score requirements are automatically rejected.
For each applicant placed in the pool for further evaluation, UGA calculates a Total Student Index (“TSI”). The TSI is based on a combination of weighted academic, extracurricular, demographic, and other factors. It is at this stage (the “TSI stage”) that UGA, under its current policy, expressly considers an applicant's race.
Applicants whose TSI scores meet a pre-set threshold are admitted automatically, while applicants whose scores fall below a pre-set minimum are rejected. Applicants whose TSI scores fall between those guideposts are then passed on to a third stage, where they are evaluated on an individual basis by admissions officers. At this final stage (the “edge read” or “ER” stage), all applicants still in the pool start with a score of zero, and readers look for qualities that might not have been apparent at the First Notice and TSI stages. Applicants who receive an ER rating above a certain number are admitted, while those with a rating below that number are rejected. Race is not designated as a factor at the ER stage. Notably, the ER stage is the only stage in the freshman admissions process where an applicant's file is actually read and qualitatively evaluated by admissions officers rather than being processed mechanically based upon the data specifically requested by the application form and inputted by the applicant.
The Plaintiffs in this case sought admission under the 1999 version of the revised admissions policy. To be admitted automatically to the Fall 1999 class at the First Notice stage, an applicant had to obtain (1) an AI of 2.86 or above; 5 and (2) SAT scores of at least 450 Verbal, 450 Math, and 1000 overall. Applicants with an AI of at least 2.40 and SAT scores of at least 950 overall, 430 on Verbal, and 400 on Math were eligible for further consideration and proceeded to the TSI stage.
At the TSI stage, a total of twelve factors were considered, with a maximum point total of 8.15. The applicant's AI plus three other objective academic factors-based upon SAT score, GPA, and curriculum quality-accounted for a maximum 5.40 points, or approximately 67% of the maximum number of points available at the TSI stage. All of these academic factors had already been considered at the earlier First Notice stage. Also taken into account at the TSI stage were five additional “leadership/activity” or “ other” factors, based upon information self-reported by the student on her application: parent or sibling ties to UGA, hours spent on extracurricular activities, hours spent on summer work, hours spent on school-year work, and first-generation college. These factors could account for a possible 1.5 points, or 18% of the available total; the most heavily weighted of these factors was “both parents: no college education,” worth 0.5 points.
Finally, three demographic factors were considered, for up to 1.25 additional points, or 15% of the maximum available: race/ethnicity (i.e., non-caucasian), gender (i.e., male), and Georgia residency. Applicants voluntarily designating themselves on their application forms as non-caucasian-defined as Asian or Pacific Islander, African-American, Hispanic, American Indian, or “Multiracial”-received 0.5 points.6 Applicants who did not do so, such as the Plaintiffs, did not receive the 0.5 point credit. Overall, other than the carry-over AI score, only one factor in the TSI equation-SAT score or ACT equivalent between 1200-1660, worth 1.0 TSI point-was worth more than the race factor; three academic factors and two non-academic factors (“both parents: no college education” and Georgia residency) were worth the equivalent of the race factor.7
After sorting applicants at the TSI stage, UGA offered admission to all candidates with a TSI score of 4.93 or higher. Applicants with a TSI score below 4.66 were rejected, while applicants whose TSI scores were between 4.66 and 4.92 survived to the final ER stage. Because of the 0.5 point credit given to non-white applicants, white applicants were effectively held to a more rigorous standard. In practice, awarding the 0.5 point credit to non-white applicants meant that white applicants needed a TSI score of at least 4.93 to be admitted at the TSI stage, while non-white applicants effectively needed only a 4.43. Similarly, to avoid outright rejection at the TSI phase and to proceed on to the ER phase, a white applicant needed a TSI score of at least 4.66 TSI points, while a non-white applicant-because of the 0.5 point boost-effectively needed only a 4.16.
All three Plaintiffs in this case survived the First Notice stage, but did not qualify for automatic admission. At the TSI stage, Plaintiff Johnson achieved a score of 4.10. She is a white female, so UGA did not grant her the 0.5 racial or 0.25 gender bonus accorded to non-white, male applicants. Because her TSI was below 4.66, UGA denied her admission outright. Had UGA granted her a cumulative 0.75 bonus, her TSI score would have been 4.85, which would have qualified her for ER consideration. Plaintiffs Bogrow and Beckenhauer achieved TSIs of 4.52 and 4.06, respectively. As with Johnson, UGA awarded neither Bogrow and Beckenhauer the 0.5 racial bonus or the 0.25 gender bonus. Had UGA done so, Bogrow would have been admitted, and Beckenhauer would have qualified for ER consideration. Without the bonus points, however, neither applicant made the 4.66 TSI cut-off, and both were denied admission at the TSI stage.
In September 1999, after the filing of this lawsuit, but before the district court determined Defendants' liability, UGA's president, Michael Adams, announced that gender would no longer be a factor at the TSI stage. UGA did elect to continue considering race in the 2000 freshman admissions process. Following the district court's summary judgment order in July 2000, however, UGA announced that it would discontinue using race as a factor in freshman admissions until resolution of this lawsuit.
The primary issue in this appeal relates to the district court's entry of summary judgment in Plaintiffs' favor on the ground that UGA's 1999 freshman admissions policy violates Title VI, and by extension Equal Protection, because it treats applicants differently based upon race. This issue turns upon whether student body diversity may be a compelling interest, and if so, whether UGA has met its burden of showing that its policy is narrowly tailored to serve that interest. The Intervenors separately contend that the policy is justifiable because it serves to ameliorate the vestiges of UGA's past discrimination.8 UGA and the Intervenors argue that we should direct the entry of summary judgment in their favor, or, at the very least, vacate the summary judgment ruling in Plaintiffs' favor and remand to the district court for a trial.
We review a summary judgment ruling de novo, applying the same legal standards used by the district court. See, e.g., Hilburn v. Murata Elecs. N. Am., Inc., 181 F.3d 1220, 1225 (11th Cir.1999). Summary judgment is appropriate where “there is no genuine issue as to any material fact and the moving party is entitled to a judgment as a matter of law.” Fed.R.Civ.P. 56(c). We “view the evidence and all factual inferences therefrom in the light most favorable to the party opposing the motion” and “ ‘all reasonable doubts about the facts [are] resolved in favor of the non-movant.’ ” See Burton v. City of Belle Glade, 178 F.3d 1175, 1187 (11th Cir.1999) (quoting Clemons v. Dougherty County, 684 F.2d 1365, 1368-69 (11th Cir.1982)). As we have explained, “ ‘[t]he mere existence of a scintilla of evidence in support of the position will be insufficient; there must be evidence on which the jury could reasonably find for the [non-moving party].’ In determining whether this evidentiary threshold has been met, the trial court ‘must view the evidence presented through the prism of the substantive evidentiary burden applicable to the particular cause of action before it.’ ” City of Delray Beach v. Agricultural Ins. Co., 85 F.3d 1527, 1530 (11th Cir.1996) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 252-55, 106 S.Ct. 2505, 2512-14, 91 L.Ed.2d 202 (1986)). Simply put, “ ‘the plain language of Rule 56(c) mandates the entry of summary judgment ․ against a party who fails to make a showing sufficient to establish the existence of an element essential to that party's case, and on which that party will bear the burden of proof at trial.’ ” Id. (quoting Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S.Ct. 2548, 2552-53, 91 L.Ed.2d 265 (1986)).
In their cross-appeal, Plaintiffs contend that the district court erred by dismissing their claims for prospective injunctive relief and by decertifying the class. Also appealed, although only by the Intervenors, is the district court's denial of Intervenors' motion for a “special case management scheduling order” or alternatively an extension of discovery. We review these rulings solely for abuse of discretion. See, e.g., Prado-Steiman v. Bush, 221 F.3d 1266, 1278 (11th Cir.2000) (class certification); Simmons v. Conger, 86 F.3d 1080, 1085 (11th Cir.1996) (injunctive relief); United States v. McCutcheon, 86 F.3d 187, 190 (11th Cir.1996) (pre-trial scheduling).
We address first the summary judgment ruling, and the district court's conclusion that UGA's use of race in its freshman admissions process for the Fall 1999 class is unlawful. The parties agree that UGA's policy is subject to strict constitutional scrutiny. See, e.g., Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 227, 115 S.Ct. 2097, 2113, 132 L.Ed.2d 158 (1995); Bass v. Board of County Comm's, 256 F.3d 1095, 1116 (11th Cir.2001) (“The Supreme Court has held that ‘all racial classifications, imposed by whatever federal, state, or local government actor, must be analyzed by a reviewing court under strict scrutiny.’ ”) (quoting Adarand ); United States v. Allen-Brown, 243 F.3d 1293, 1298-99 (11th Cir.2001) (“Under Supreme Court jurisprudence, race-based treatment is subject to strict scrutiny under the Equal Protection Clause.”).9 We apply strict scrutiny because “[c]lassifications of citizens solely on the basis of race ‘are by their very nature odious to a free people whose institutions are founded upon the doctrine of equality.’ ” Shaw v. Reno, 509 U.S. 630, 643, 113 S.Ct. 2816, 2824, 125 L.Ed.2d 511 (1993) (“Shaw I ”) (quoting Hirabayashi v. United States, 320 U.S. 81, 100, 63 S.Ct. 1375, 1385, 87 L.Ed. 1774 (1943)).
Both the Supreme Court and our Court have made clear that racial classifications, whatever the motivation for enacting them, are highly suspect and rarely withstand constitutional scrutiny. “[T]he basic principle is straightforward: ‘Racial and ethnic distinctions of any sort are inherently suspect and thus call for the most exacting judicial examination.’ ” Miller v. Johnson, 515 U.S. 900, 904, 115 S.Ct. 2475, 2482, 132 L.Ed.2d 762 (1995) (citation omitted); see also Allen-Brown, 243 F.3d at 1299 (“The standard of review under the Equal Protection Clause ‘is not dependent on the race of those burdened or benefited by a particular classification.’ The Equal Protection Clause ․ recognizes that ‘any individual suffers an injury when he or she is disadvantaged by the government because of his or her race, whatever that race may be.’ ”) (citation omitted).
Under strict scrutiny, a racial classification must be held unlawful unless (1) the racial classification serves a compelling governmental interest, and (2) it is narrowly tailored to further that interest. See, e.g., Adarand, 515 U.S. at 227, 115 S.Ct. at 2113; Bass, 256 F.3d at 1116. The proponent of the classification bears the burden of proving that its consideration of race is narrowly tailored to serve a compelling governmental interest. See, e.g., Bass, 256 F.3d at 1114(citing City of Richmond v. J.A. Croson Co., 488 U.S. 469, 510-11, 109 S.Ct. 706, 730-31, 102 L.Ed.2d 854 (1989)). We too have stressed that, when government undertakes affirmative action, it must present a “ ‘strong basis in evidence’ ” for doing so. See Peightal v. Metropolitan Dade County, 26 F.3d 1545, 1553 (11th Cir.1994) (quoting Croson, 488 U.S. at 501, 109 S.Ct. at 725); see also Engineering Contractors Ass'n of So. Fla., Inc. v. Metropolitan Dade County, 122 F.3d 895, 906 (11th Cir.1997) (“If a race- or ethnicity-conscious affirmative action program is to be upheld, ‘the district court must make a factual determination that [there exists] a strong basis in evidence’ to support the conclusion that remedial action is necessary.”). UGA's burden in this case is, therefore, substantial.
UGA identifies the educational benefits of student body diversity in higher education as the compelling interest justifying the consideration of race in its freshman admissions decisions. Although UGA argues loosely that its use of race is “supported” by the university's history of discrimination, UGA does not identify remediating past discrimination as the compelling interest justifying its policy; indeed, it has repeatedly disavowed that interest. See, e.g., Defendants' Opp. to Intervenors Mot. to Enter a Special Case Mgmt. Order, Feb. 17, 2000, at 4 (“Defendants have never contended that [UGA's] limited use of race is required or necessary due to any remaining vestiges of discrimination. [I]ntervenors' attempt to resurrect an issue that was laid to rest over a decade ago should be rejected.”); Appellants' Reply Br. at 15 (“UGA does not contend that it utilized race as a TSI factor in the 1999 freshman admissions plan to cure actual present day effects of past discrimination.”).
The initial question, therefore, is when, if ever, may student body diversity be a compelling interest? UGA contends that, under binding precedent (specifically, Justice Powell's opinion in Bakke ), student body diversity is always a compelling interest that can justify a racial preference in university admissions. Plaintiffs respond that, under more recent Supreme Court precedent, student body diversity is never a compelling interest. A third view, articulated indirectly at times by both parties, is that student body diversity may be a compelling interest in some facts and circumstances depending upon the record.
We need not, and do not, resolve in this opinion whether student body diversity ever may be a compelling interest supporting a university's consideration of race in its admissions process. Even assuming that UGA's asserted interest in student body diversity is a compelling interest, UGA's 1999 freshman admissions policy is unconstitutional because UGA has plainly failed to show that its policy is narrowly tailored to serve that interest. Accordingly, there is no reason for us to decide whether or when student body diversity may be a compelling interest, and well-settled principles of judicial restraint caution against taking that course. See, e.g., Burton v. United States, 196 U.S. 283, 295, 25 S.Ct. 243, 245, 49 L.Ed. 482 (1905) (courts are not “to decide questions of a constitutional nature unless absolutely necessary to a decision of the case”); Florida Ass'n of Rehab. Facilities, Inc. v. State of Fla. Dept. of Health and Rehab. Servs., 225 F.3d 1208, 1227 n. 14 (11th Cir.2000) (applying “the longstanding rule that constitutional questions should not be resolved unless necessary to the decision”) (citing I.A. Durbin, Inc. v. Jefferson Nat'l Bank, 793 F.2d 1541, 1553 (11th Cir.1986)).
Nevertheless, some aspects of that issue are relevant to our narrow tailoring analysis; in particular, UGA claims that its policy is modeled after the “Harvard Plan” discussed by Justice Powell in Bakke. For that reason, and given that the parties have argued the issue at length in their presentations to this Court, we would be remiss if we did not address the issue at all. In particular, we think it important to underscore that the constitutional viability of student body diversity as a compelling interest is an open question, and ultimately is one that, because of its great importance, warrants consideration by the Supreme Court.
A majority of the Supreme Court has never agreed that student body diversity is, or may be, a compelling interest sufficient to justify a university's consideration of race in making admissions decisions. Defendants argue that the viability of student body diversity as a compelling interest is not an open question, because Justice Powell's opinion 23 years ago in Bakke-which no other Justice joined-constitutes binding precedent and requires the lower federal courts to treat that interest as compelling. We are unconvinced by this argument; although a majority of the Supreme Court may eventually adopt Justice Powell's opinion as binding precedent, and even now the opinion has persuasive value, the opinion is not binding on the issue before us today.
Reaching this conclusion requires a close and careful examination of the various opinions in Bakke. In that case, an unsuccessful white applicant to the University of California at Davis Medical School challenged the school's admissions program, which consisted of a regular admissions system for non-minority applicants and a special admissions system strictly for minorities. 438 U.S. at 273-75, 98 S.Ct. at 2739-40. The minority admissions system was overseen by a separate admissions committee. Minority applicants were never compared against non-minority applicants; and although minority applicants were rated in a manner similar to that used for non-minority applicants, minority applicants did not have to meet the minimum GPA requirement applied to non-minority applicants. The minority admissions committee's task was to recommend minority applicants for admission until a specific number of minority applicants were admitted; the actual goal was predetermined by faculty vote. In the relevant year, 16 out of 100 available seats were reserved for minorities.
The Supreme Court of California sustained the plaintiff's challenge to this dual-track quota system, holding that the medical school's admission program violated Equal Protection, Title VI, and the California Constitution. The court ordered the plaintiff's admission to the school, and also permanently enjoined the university from “according any consideration to race in its admission process.” Id. at 272, 98 S.Ct. at 2738. In so doing, the court rejected the university's asserted purposes for its special admissions system, including the goal of obtaining the educational benefits that flow from an ethnically diverse student body. See id. at 306, 98 S.Ct. at 2756-57.
On appeal, the Supreme Court, through a majority formed by Justices Powell, Stevens, Burger, Stewart, and Rehnquist, affirmed the California Supreme Court's finding that the medical school's admissions system was invalid and also affirmed the order directing the admission of Bakke to the school. In the same set of opinions, however, the Court, through a different majority formed by Justices Powell, Brennan, White, Marshall, and Blackmun, reversed the California Supreme Court's total prohibition on the university's consideration of race in admissions.
Justice Powell, writing solely for himself, supplied the pivotal vote for both holdings. Applying strict scrutiny, Justice Powell rejected almost all of the university's proffered justifications for its consideration of race, but found the university's goal of “attain[ing] a diverse student body” to be “clearly” a “constitutionally permissible goal for an institution of higher education.” Id. at 311-12, 98 S.Ct. at 2759. According to Justice Powell, a university's “interest in diversity is compelling in the context of a university's admissions program,” in which “[e]thnic diversity ․ is only one element in a range of factors a university properly may consider in attaining the goal of a heterogenous student body.” 438 U.S. at 314, 98 S.Ct. at 2760-61.
Justice Powell therefore concluded that “[i]n enjoining petitioner from ever considering the race of any applicant, ․ the courts below failed to recognize that the State has a substantial interest that legitimately may be served by a properly devised admissions program involving the competitive consideration of race and ethnic origin. For this reason, so much of the California court's judgment as enjoins petitioner from any consideration of the race of any applicant must be reversed.” Id. at 318, 98 S.Ct. at 2763. In that discussion, Justice Powell endorsed the constitutionality of the Harvard Plan, a “flexible” admissions program which treats race as one factor among many that may be considered in making admissions decisions. See id. at 320, 98 S.Ct. at 2761-62. But Justice Powell also determined that the university's rigid quota system was not narrowly tailored to serve any asserted interest in diversity, and hence the university's admissions process was unlawful. See id.
Justice Powell clearly identified diversity as a compelling interest that may be asserted by a university in defense of an admissions program that flexibly considers race as one of several factors in making admissions decisions. No other Justice, however, expressly endorsed that view. Justice Brennan, joined by Justices White, Marshall, and Blackmun, took an entirely different approach to the case. He found that the university's articulated purpose of remedying the effects of past societal discrimination was itself “sufficiently important to justify the use of race-conscious admissions programs where there is a sound basis for concluding that minority underrepresentation is substantial and chronic, and that the handicap of past discrimination is impeding access of minorities.” Id. at 362, 98 S.Ct. at 2784 (Brennan, J., concurring in part and dissenting in part). Justice Brennan did not consider whether student body diversity constituted a compelling interest sufficient to justify the university's discriminatory admissions policy. Indeed, Justice Brennan did not consider whether any interest was compelling because he did not analyze the university's asserted justifications under strict scrutiny. Id. at 356-57, 98 S.Ct. at 2781-82. With regard to Justice Powell's discussion of the Harvard Plan, Justice Brennan wrote only that he agreed that the plan would be “constitutional under our approach, at least so long as the use of race to achieve an integrated student body is necessitated by the lingering effects of past discrimination.” Id. at 326 n. 1, 98 S.Ct. at 2766 n. 1.
Explaining that the litigation was not a class action, but rather a “controversy between two specific litigants,” Justice Stevens (joined by Chief Justice Burger and Justices Stewart and Rehnquist) viewed “the question of whether race can ever be used as a factor in an admissions decision” as an issue not before the Court. Id. at 408, 411, 98 S.Ct. at 2808, 2809 (Stevens, J., concurring in part and dissenting in part). Justice Stevens also declined to address the constitutional validity of the medical school's admissions process, relying solely upon Title VI, which he viewed as “crystal clear” in its prohibition that “[r]ace cannot be the basis of excluding anyone from participation in a federally funded program.” Id. at 418, 98 S.Ct. at 2813.
Whether Justice Powell's discussion in Bakke constitutes binding precedent may be considered in light of the principle enunciated by the Court that “[w]hen a fragmented Court decides a case ․ the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds.” Marks v. United States, 430 U.S. 188, 193, 97 S.Ct. 990, 994, 51 L.Ed.2d 260 (1977). In Bakke, a majority of the Court agreed to reverse the decision below to the extent that it held that race could never be a factor in a university admissions process. Justice Brennan's opinion reached that conclusion on the (now discredited) ground that some racial classifications were not subject to strict scrutiny. He did not, therefore, consider whether student body diversity could constitute a compelling interest, as Justice Powell indicated; indeed, Justice Brennan did not even mention diversity as a constitutionally valid interest. At best, Justice Brennan's opinion treats diversity as the kind of “important” state interest that might survive intermediate scrutiny (the standard that Justice Brennan felt should apply to race-based decision-making), not strict scrutiny. 438 U.S. at 359, 98 S.Ct. at 2783 (Brennan, J.). In this sense, at least, the narrowest-i.e., less far-reaching-common ground of the Brennan and Powell opinions on the specific subject of student body diversity is that diversity is an “important” interest, but not the kind of compelling interest that potentially might withstand even the strictest constitutional scrutiny.10
Moreover, Justice Brennan's opinion endorsed Justice Powell's reasoning, if at all, only to the extent that student body diversity was part of the university's much broader justification that its favorable treatment of minorities was necessary to remedy the present effects of past discrimination. This remedial purpose rationale was not a component of Justice Powell's discussion of student body diversity; Justice Powell did not anchor his view of diversity to the existence of present effects of past discrimination-societal or otherwise.11 For that reason, Justice Brennan's opinion was careful to support Justice Powell's endorsement of the Harvard Plan only “so long as the use of race to achieve an integrated student body is necessitated by the lingering effects of past discrimination.” Id. at 326 n. 1, 98 S.Ct. at 2766 n. 1 (Brennan, J.) (emphasis added).
While it may be possible to speculate that all four of the Justices who joined Justice Brennan's opinion might have embraced treating student body diversity as a compelling interest even in the absence of a valid remedial purpose, that kind of speculation is inconsistent with Marks.12 Moreover, we think this speculation unsound. It requires us to ignore the language and rationale of Justice Brennan's opinion and to draw assumptions that have no basis in the opinion, a particularly unwise course given that Justice Brennan's opinion disagreed with Justice Powell's not only regarding the fundamental issue of whether the university's set-aside was valid, but also on the proper constitutional test for analyzing the use of race.
In the end, the fact is inescapable that no five Justices in Bakke expressly held that student body diversity is a compelling interest under the Equal Protection Clause even in the absence of valid remedial purpose. As our predecessor court aptly put it, “[i]n over 150 pages of the U.S. Reports, the Justices [in Bakke ] have told us mainly that they have agreed to disagree.” United States v. City of Miami, 614 F.2d 1322, 1337 (5th Cir.1980), on reh'g en banc, 664 F.2d 435 (1981). Simply put, Justice Powell's opinion does not establish student body diversity as a compelling interest for purposes of this case.13
In the years since Bakke, the Court has never returned to whether diversity may be a compelling interest supporting a university's consideration of race in making admissions decisions. Language in some opinions from the Court suggest, but do not hold, that the only interest sufficient to support a racial preference is remediating the defendant's own past discrimination. See, e.g., Adarand, 515 U.S. at 237, 115 S.Ct. at 2118 (diversity in broadcasting not a compelling interest); Croson, 488 U.S. at 493, 109 S.Ct. at 722 (plurality op.) (observing that unless racial classifications “are strictly reserved for remedial settings, they may in fact promote notions of racial inferiority and lead to a politics of racial hostility”); Metro Broadcasting, Inc. v. FCC, 497 U.S. 547, 612, 110 S.Ct. 2997, 3034, 111 L.Ed.2d 445 (1990) (O'Connor, J., dissenting) (“Modern equal protection doctrine has recognized only one [compelling state] interest: remedying the effects of past discrimination.”).14 Other Supreme Court opinions have rejected asserted interests similar to student body diversity. See Shaw v. Hunt, 517 U.S. 899, 909-10, 116 S.Ct. 1894, 1903, 135 L.Ed.2d 207 (1996) ( “Shaw II ”) (“an effort to alleviate the effects of societal discrimination is not a compelling interest”); Miller, 515 U.S. at 920, 115 S.Ct. at 2490 (interest based on assumptions that members of a particular race “think alike, share the same political interests, and will prefer the same candidates at the polls” is “racial stereotyping at odds with equal protection mandates”); Wygant v. Jackson Board of Education, 476 U.S. 267, 275-76, 106 S.Ct. 1842, 1847-48, 90 L.Ed.2d 260 (1986) (plurality op.) (local school board's asserted interest in providing minority role models for its minority students not a sufficient interest). But none of these opinions involves university admissions, and none specifically addresses student body diversity.15
We think it clear that the status of student body diversity as a compelling interest justifying a racial preference in university admissions is an open question in the Supreme Court and in our Court. Of course, a proponent of any kind of race-based decision-making always faces a substantial burden in attempting to justify its policy. See Croson, 488 U.S. at 505, 109 S.Ct. at 727-28 (“ ‘Because racial characteristics so seldom provide a relevant basis for disparate treatment, and because classifications based on race are potentially so harmful to the entire body politic, it is especially important that the reasons for any such classification be clearly identified and unquestionably legitimate.’ ”) (citation omitted). It is possible that the important purpose of public education and the expansive freedoms of speech and thought associated with university environment-recognized in other decisions by the Court-may on a powerful record justify treating student body diversity as a compelling interest.16 The weight of recent precedent is undeniably to the contrary, however.
To reiterate, we do not decide today whether or when student body diversity may be a compelling interest for purposes of strict scrutiny review under the Equal Protection Clause of the Fourteenth Amendment. The Supreme Court has placed as much importance on the requirement that any race-conscious program be narrowly tailored as it has on the requirement that the asserted justification for race-conscious decision-making be sufficiently compelling. Here, UGA fails to meet its burden of showing that its 1999 freshman admissions policy is narrowly tailored. In similar situations, courts elsewhere have simply assumed that student body diversity is a compelling interest, and then proceeded to explain why the policy being challenged is unlawful regardless. See Eisenberg, 197 F.3d at 130; Tuttle v. Arlington County Sch. Bd., 195 F.3d 698, 705 (4th Cir.1999) (“Since we conclude ․ that the [race-based admissions policy] was not narrowly tailored, we leave the question of whether diversity is a compelling interest unanswered.”); Wessmann, 160 F.3d at 794 (“[we] assume arguendo-but we do not decide-that Bakke remains good law and that some iterations of ‘diversity’ might be sufficiently compelling, in specific circumstances, to justify race-conscious actions.”). We shall do the same, and assume for purposes of this opinion only that UGA's asserted interest in student body diversity is a compelling interest.
The Supreme Court has explained that, although in certain circumstances drawing racial distinctions is permissible where a governmental body is pursuing a compelling state interest, a state “is constrained in how it may pursue that end: ‘[T]he means chosen to accomplish the State's asserted purpose must be specifically and narrowly framed to accomplish that purpose.’ ” Shaw II, 517 U.S. at 908, 116 S.Ct. at 1902. The important purpose of the narrow tailoring requirement is to ensure that “the chosen means ‘fit’ in th[e] compelling goal 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. at 721. By definition, this inquiry must be intrusive, and focused very closely and in a very precise way on the specific terms of the regulation or policy under review, because only with that kind of searching examination can a court ensure that the defendant's use of race is truly as narrow as the Constitution requires. See In re Birmingham Reverse Discrimination Employment Litigation, 20 F.3d 1525, 1545 (11th Cir.1994) (“a race conscious government policy justified by a compelling purpose ․ must also use race in as limited a manner as possible to accomplish that compelling purpose.”).
As we have discussed, it is the burden of the party proposing a racial preference to show that its approach is narrowly tailored to achieving its asserted interest. To withstand summary judgment, therefore, UGA must show that a reasonable factfinder could conclude that there is sufficient record evidence supporting its claim that its freshman admissions process is narrowly tailored to achieve its goal of student body diversity. In our view, UGA does not even come close to making that showing.17
Neither this Court nor the Supreme Court has had occasion to define the contours of the narrow tailoring inquiry in a case involving a university's race-conscious admissions policy. In the employment arena, we have identified several factors to be considered when evaluating the constitutionality of an affirmative action plan, including: “ ‘the necessity for the relief and the efficacy of alternative remedies, the flexibility and duration of the relief, ․ the relationship of numerical goals to the relevant ․ market, and the impact of the relief on the rights of [the plaintiffs].’ ” Birmingham, 20 F.3d at 1545 (quoting Howard v. McLucas, 871 F.2d 1000, 1008 (11th Cir.1989)). These factors are drawn from the plurality opinion in United States v. Paradise, 480 U.S. 149, 107 S.Ct. 1053, 94 L.Ed.2d 203 (1987), an employment case where the affirmative action plan at issue was designed to remediate past discrimination.
The Fourth Circuit recently adopted the Paradise factors in evaluating whether a local school district's race-conscious admissions policy was narrowly tailored. See Tuttle, 195 F.3d at 706 (“When reviewing whether a state racial classification is narrowly tailored, we consider factors such as: (1) the efficacy of alternative race-neutral policies, (2) the planned duration of the policy, (3) the relationship between the numerical goal and the percentage of minority group members in the relevant population or work force, (4) the flexibility of the policy, including the provision of waivers if the goal cannot be met, and (5) the burden of the policy on innocent third parties.”) (internal quotation marks omitted) (citing Paradise, 480 U.S. at 171, 107 S.Ct. at 1067).
We have no disagreement with this general framework. We do think, however, that the Paradise factors should be adjusted slightly to take better account of the unique issues raised by the use of race to achieve diversity in university admissions. For example, if we assume that student body diversity may be a compelling interest even where the defendant is not attempting to remedy present effects of past discrimination, then the duration of the race-conscious policy may not be an important consideration; by definition, the goal of remedying past discrimination has a logical end-point, the goal of exposing students to a diverse student body may not. Likewise, inquiring into “the relationship between the numerical goal and the percentage of minority group members in the relevant population” may be unhelpful where the university does not target a specific number of minority applicants for admission.
At the same time, a limited inquiry into “the flexibility of the policy” may not adequately reflect the paramount importance of the requirement that, to serve validly the end of diversity, a race-conscious admissions policy must truly assess each applicant as an individual rather than a member of a particular racial group. See, e.g., Bakke, 438 U.S. at 317, 98 S.Ct. at 2762. Similarly, while it may be constitutionally acceptable in limited circumstances for “innocent” members of a once-favored racial group to bear some burden when a defendant seeks to remediate its past discrimination, this view of competing racial groups has no meaning when a university's professed goal is to create a diverse student body, and the burden imposed by a racial preference intended to achieve diversity cannot so readily be justified on this basis.
We therefore view the Paradise factors as providing general guidance on the question before us, but tailor those factors slightly to fit these types of cases. Specifically, a court evaluating a school admissions program designed to serve a compelling interest in obtaining the educational benefits associated with a diverse student body should examine: (1) whether the policy uses race in a rigid or mechanical way that does not take sufficient account of the different contributions to diversity that individual candidates may offer; (2) whether the policy fully and fairly takes account of race-neutral factors which may contribute to a diverse student body; (3) whether the policy gives an arbitrary or disproportionate benefit to members of the favored racial groups; and (4) whether the school has genuinely considered, and rejected as inadequate, race-neutral alternatives for creating student body diversity. The foregoing factors essentially correspond to all of the factors adopted in Paradise (other than duration) for affirmative action plans generally. See 480 U.S. at 171, 107 S.Ct. at 1067. We do not view these factors as determinative in all instances, but they do provide a useful analytical structure, and we think it unlikely that a race-conscious admissions policy which fails to satisfy these criteria will truly “use race in as limited a manner as possible,” Birmingham, 20 F.3d at 1545, to advance a compelling interest in student body diversity. We discuss each of these factors in turn.
At the outset, while we can assume that racial diversity may be one component of a diverse student body, it is not the only component. If the goal in creating a diverse student body is to develop a university community where students are exposed to persons of different cultures, outlooks, and experiences, a white applicant in some circumstances may make a greater contribution than a non-white applicant. To take a few obvious examples, a white applicant from a disadvantaged rural area in Appalachia may well have more to offer a Georgia public university such as UGA-from the standpoint of diversity-than a non-white applicant from an affluent family and a suburban Atlanta high school. Similarly, a white applicant to a Georgia public university who was raised in Athens, Greece may have a much richer background and exposure to a much more unusual environment than a non-white applicant who has spent all his life in Athens, Georgia. The point is simply this: the diversity interest that we assume may constitute a compelling interest does not view racial diversity as an end in itself, but rather as a means to achieve the larger goal of providing a superior education by creating a university community that resembles the broad mix of cultures, experiences, and ideas to be found in society.18
Accordingly, an admissions policy that seeks to create a diverse student body by considering the race of applicants must do so in a sufficiently flexible way. It goes without saying that a university may not establish a quota system for members of certain racial groups, and may not put members of one racial group on a different and more lenient track than members of another group. See, e.g., Bakke, 438 U.S. at 319-20, 98 S.Ct. at 2763 (Powell, J.). But the mere fact that race technically does not insulate a candidate from competition with other applicants does not, by itself, mean that the policy is narrowly tailored. A race-conscious admissions policy still must ensure that, even when using race as a factor, the weight accorded that factor is not subject to rigid or mechanical application, and remains flexible enough to ensure that each applicant is evaluated as an individual and not in a way that looks to her membership in a favored or disfavored racial group as a defining feature of her candidacy.
Second, the policy must ensure that race-neutral factors which contribute to a diverse student body are considered fully and fairly along with race in making admissions decisions. We assume that there is value in having a racially-diverse student body. But racial diversity alone is not necessarily the hallmark of a diverse student body, and race is not necessarily the only, or best, criterion for determining the contribution that an applicant might make to the broad mix of experiences and perspectives that creates the value UGA asserts in diversity. An admissions policy that seeks to achieve student body diversity by allowing some applicants to be treated more favorably than others based on race must ensure full and fair consideration of other, race-neutral characteristics that contribute to a truly diverse class of students.
Third, the policy must use race in a way that does not give an arbitrary or disproportionate benefit to members of the favored racial groups, and thereby unduly disadvantage applicants from outside the favored groups who may well add more to the overall diversity of the student body. Even when a race-conscious policy permits broad consideration of race-neutral factors that contribute to diversity, it may undervalue those factors in a way that makes race effectively the primary criterion for diversity.
Finally, a university defending a race-conscious admissions policy must show that it has genuinely considered, and rejected as inadequate, race-neutral alternatives for creating student body diversity. We have held that only as a “last resort” may race be used in awarding valuable public benefits such as government contracts. See, e.g., Engineering Contractors, 122 F.3d at 926 (“The essence of the narrowly tailored inquiry is the notion that explicitly racial preferences ․ must be only a last resort option.”) (internal quotation marks omitted). That principle applies equally to the university admissions process. Race-based decision-making is at odds with the Constitution in any context, and before injecting race into the admissions process, a university should explore seriously and in good faith the wide variety of race-neutral measures that may enhance not only the overall diversity of the student body, but also racial diversity itself.
Applying these factors, we hold that UGA's 1999 freshman admissions policy is not narrowly tailored to achieve its stated goal of student body diversity. By mechanically and inexorably awarding an arbitrary “diversity” bonus to each and every non-white applicant at the TSI stage, and severely limiting the range of other factors that may be considered at that stage, the policy contemplates that non-white applicants will be admitted or advance further in the process at the expense of white applicants with greater potential to contribute to a diverse student body. This lack of flexibility is fatal to UGA's policy.
To begin with, the policy mechanically awards bonus points to each non-white applicant, regardless of that applicant's potential overall contribution to diversity. At the TSI stage, every non-white applicant receives a 0.5 point bonus, regardless of his or her background and regardless of whether a white applicant with a far more “diverse” background receives a corresponding bonus for the attributes she may offer. There is no ability to adjust the bonus downwards for non-white applicants whose profiles are clearly well within the mainstream of white applicants and who add nothing else to the diversity of the incoming class.
This rigid, mechanical approach to considering race is itself incompatible with the need for flexibility in the admissions process. What makes UGA's approach even more deficient, however, is the policy's exclusion of many race-neutral factors that would reflect an applicant's potential contributions to diversity. Only twelve factors are considered at the TSI stage; of those twelve, one (gender) is concededly unlawful and has now been abandoned, and four relate entirely to academic ability and do not correspond-at least on this record-to student body diversity. That leaves just seven factors, including race, corresponding in varying degrees to student body diversity.19 Significantly, however, there is no flexibility whatsoever at the TSI stage to consider additional factors for any individual applicant.
The TSI process mechanically rewards candidates under certain fixed criteria, including race, but permits no favorable treatment of applicants whose personal backgrounds or skills, while undeniably promoting diversity, do not fit neatly into one of the categories predetermined by UGA. Individuals who come from economically disadvantaged homes; individuals who have lived or traveled widely abroad; individuals from remote or rural areas; individuals who speak foreign languages; individuals with unique communications skills (such as an ability to read Braille or communicate with the deaf); and individuals who have overcome personal adversity or social hardship-none of the characteristics that make these kinds of individuals “diverse” are taken into account at the TSI stage.
Moreover, the TSI considers an applicant's extracurricular and work activities-which both reflect and provide insight into a student's potential contribution to diversity-only in the most limited way, by inquiring solely into the number of hours spent on those activities, without any inquiry at all into their nature or purpose. There is simply no consideration at the TSI stage of an applicant's extracurricular or work activities from the standpoint of how those activities might enhance the diversity of the freshman class; to take only one example, an applicant who has spent her summers performing volunteer work in a less developed “third world” country presumably would add far more diversity to the class than many of her peers who worked at more ordinary and less challenging summer jobs, yet at the TSI stage the uniqueness of her experience is wholly ignored. The number of hours that an applicant reports herself having spent on these activities is hardly a meaningful substitute for personalized evaluation of the applicant's unique extracurricular and work experiences. The almost total inability of the TSI formula to take account of the potential contributions to diversity from applicants with characteristics less easily and mechanically tabulated than race is inconsistent with narrow tailoring.
This inflexibility at the TSI stage has real consequences for white applicants. The results at the TSI stage dictate whether an applicant is admitted automatically, rejected automatically, or receives personalized consideration at the ER stage. The racial bonus may be decisive as to whether the applicant is admitted or rejected. Plaintiff Bogrow, for example, would have qualified for automatic admission had she received the 0.5 point bonus; instead, because she did not receive that bonus, she was rejected outright at the TSI stage. We acknowledge that race may “tip the balance” even under an admissions policy (such as the Harvard Plan) that does not consider race so mechanically. But the problem here is even more serious, because race not only may be decisive in whether a UGA applicant qualifies for automatic admission or rejection at the TSI stage, but also may be decisive in whether the applicant will receive (at the ER stage) any personalized, qualitative assessment of her application and potential contribution to diversity. As noted above, it is only at the ER stage that an applicant's file is read and qualitatively evaluated by admissions officers; that kind of evaluation does not occur at the First Notice stage or the TSI stage, even though final admissions decisions are made at the TSI stage based in part upon factors that UGA says are meant to enhance diversity. Race, and race alone, may determine not only whether an applicant is foreclosed from a spot in UGA's freshman class, but also whether an applicant's true potential to contribute to diversity is ever fully and fairly assessed.
UGA appears to recognize that there are flaws inherent in its rigid yet incomplete approach to considering race and diversity. When asked at his deposition whether he believed that other public institutions in the state were using race in their admissions process, UGA's President, Adams, responded that although he did not know, “I know that the difference between the way we have to do it because of size and the way everybody else does it is so dramatically different that that's part of the problem. I would guess that every other institution in the state is probably able to read individual files.” Tr. of Adams Dep. at 60. The thrust of UGA's position seems to be that, given the large volume of applications it receives, it cannot practicably provide real personalized analysis for most applicants. See id. at 61 (“I sometimes regret reducing kids to some TSI formula just because of the way I like to do things, but you have to do some things differently at a place of this size than you do from virtually anything other than a Big Ten size institution because that's what we are.”).
The rejoinder to this is obvious: if UGA wants to ensure diversity through its admissions decisions, and wants race to be part of that calculus, then it must be prepared to shoulder the burden of fully and fairly analyzing applicants as individuals and not merely as members of groups when deciding their likely contribution to student body diversity. See Frontiero v. Richardson, 411 U.S. 677, 690, 93 S.Ct. 1764, 1772, 36 L.Ed.2d 583 (1973) ( “[W]hen we enter the realm of ‘strict judicial scrutiny,’ there can be no doubt that ‘administrative convenience’ is not a shibboleth, the mere recitation of which dictates constitutionality.”); see also Croson, 488 U.S. at 508, 109 S.Ct. at 729 (plurality op.) (“the interest in avoiding the bureaucratic effort necessary to tailor remedial relief ․ cannot justify a rigid line drawn on the basis of a suspect classification”). Regardless, we are totally unpersuaded by the idea that the only way-or even close to the only way-for a large public university to increase the diversity of its student body is to adopt UGA's system.20
The effect of UGA's policy-mechanically awarding bonus points to all non-white applicants while constricting individualized analysis of candidates and limiting the range of factors relevant to diversity for which bonus points may be awarded to white applicants-is to create an inflexibility that impedes, rather than advances, UGA's stated goal of student body diversity. “The idea is a simple one: At the heart of the Constitution's guarantee of equal protection lies the simple command that the Government must treat citizens as individuals, not as simply components of a racial, religious, sexual or national class.” Miller, 515 U.S. at 911, 115 S.Ct. at 2486 (citations and internal quotation marks omitted). It follows that when a university, in the name of student body diversity, grants preferential treatment to some applicants based on race, it must ensure that applicants are fully and fairly examined as individuals for their potential contributions beyond race to diversity. UGA's policy fails to do so adequately.
UGA's policy is not only rigid and incomplete, the benefit it awards each and every non-white applicant is wholly, and concededly, arbitrary. If a university cannot even articulate a basis for the amount of the numerical bonus it awards non-white candidates, then it has no right to award such a bonus, or to implement an admissions policy that uses such a bonus as a substitute for reading and evaluating each individual's application to assess her potential contribution to diversity. Yet UGA's admissions director, Nancy McDuff, acknowledged at her deposition that the choice of a particular point bonus for race is made “out of the blue” to reflect nothing more than a guess as to how much weight admissions officers believe race should carry in the TSI formula (the maximum score of which is, itself, wholly arbitrary). UGA's counsel likewise conceded at oral argument that there is no statistical basis for the use of a 0.5 point figure and that the figure is “arbitrary.” We recognize that it may be inevitable that any specific value assigned to a characteristic such as race in an admissions formula similar to UGA's will be arbitrary (unless it is pegged to generating a particular number of admittees from a racial group, an approach that would raise serious constitutional problems of its own). But we are hard pressed to see how we could find UGA's freshman admissions policy to be narrowly tailored when UGA candidly admits that the precise benefit it mechanically awards to each and every non-white applicant at the TSI stage has no foundation whatsoever. The flaw lies in the policy, and its attempt to reduce the complex range of factors that may reflect an individual's contribution to diversity to an administratively-convenient, but otherwise rigid and incomplete, formula.
The benefit UGA mechanically accords non-white applicants at the TSI stage is not only arbitrary, it is also disproportionate to the very few diversity-related factors that may permissibly be considered at that stage, and quite plainly is considerable relative to the many factors relating as much if not more directly to diversity that the TSI formula wholly excludes. Overall, only one of the twelve factors in the TSI equation-SAT score, or ACT equivalent, between 1200-1600-is worth more in raw value than the race factor (1.0 as opposed to 0.5). Among the non-academic factors that correspond to diversity, no single factor is worth more, and the 0.5 point racial bonus accounts for almost 20% of the maximum points available on those factors.21 And even assuming that the number of hours an applicant reports herself having spent on extracurricular or work activities has some real significance in gauging which applicants can contribute the most to the diversity of the incoming class, an applicant reporting the maximum number of hours can still receive only a fraction of the TSI points automatically awarded a non-white applicant whose sole potential contribution to diversity is his or her race.22 It is primarily through the non-academic factors that UGA can generate a truly mixed group of students, and by weighing race so heavily, UGA necessarily discounts other non-academic factors in the TSI that may in some instances be far more accurate barometers for diversity.23
UGA contends nevertheless that the racial bonus is not disproportionate because in the end relatively few white applicants are disadvantaged solely because of it. UGA asserts that 85% of the freshman class that ultimately chose to enroll in the Fall of 1999 was selected at the initial, wholly-race neutral First Notice stage. UGA also observes that in the overall TSI calculus the racial bonus amounts to only about 6% of the total points available (0.5 out of 8.15). UGA further claims that any impact from its less favorable treatment of white applicants is offset by the bonus given to children or siblings of alumni, who according to UGA are predominantly white.
A threshold obstacle to these arguments is the lack of record evidence supporting them. UGA came forward with no statistical evidence squarely demonstrating its premise that few white applicants for its Fall 1999 entering class were adversely affected by the awarding of a 0.5 point bonus to all non-whites at the TSI stage. Evidence that 85% of the eventual incoming class was accepted at the First Notice stage does not prove that an insignificant number of admitted candidates was affected by UGA's consideration of race at the TSI stage. As best we can tell, UGA received some 13,000 applications in 1999, and at least 1,000 of those applicants passed through the TSI stage.24 The amount of the racial bonus is significant, and UGA introduced no specific evidence showing that the race factor did not have a measurable effect on the outcome of the candidate review at the TSI stage.
More fundamentally, UGA's argument (like its policy) ignores the effect that its method of injecting race into the admissions process had on individuals. Whatever else the admissions data may show (and UGA introduced very little of the pertinent data into the record), it is clear that literally hundreds of talented young men and women hoping to attend UGA passed through the TSI stage, and for some, if not many, of those individuals-including the Plaintiffs in this lawsuit-race denied them a coveted place in UGA's freshman class.
In addition, there is no statistical evidence establishing the exact relationship between race and alumni status, let alone demonstrating how, if at all, the alumni bonus offsets the racial bonus awarded to all non-whites (generally or for the Fall 1999 applicant group). Indeed, the bonus for race was two times that for alumni ties (0.5 versus 0.25). UGA has the burden of proving that its admissions policy is narrowly tailored, and it has not come close to establishing that its consideration of race at the TSI stage had a negligible effect.
Finally, UGA fails to show on this record that it meaningfully considered, let alone rejected as insufficient, any wholly race-neutral alternatives to the race-conscious admissions policy adopted for the Fall 1999 class. On the contrary, the record is clear that UGA has been committed, and remains committed, to using race in its freshman admissions process until it is precluded from doing so. It is beyond dispute that there are race-neutral measures that potentially may advance the goal of creating a diverse student body. Recruiting, advertising, financial incentives to admittees from less advantaged homes, and other outreach strategies all may play an important role in ensuring that the qualified applicant pool contains individuals from a wide variety of backgrounds. Similarly, there may be innovative strategies in the admissions process itself, ranging from income-based selection to guaranteeing admission to the top percentage of graduating seniors in every high school in the state. There is some indication that UGA has considered ways to increase the number of African-Americans at the university, but little evidence that UGA has sought by race-neutral means to increase the broader diversity of the student body, and no record evidence that UGA has genuinely considered substituting wholly race-neutral measures for the current race-based admissions formula.
As we have made clear, “[w]hile strict scrutiny does not require exhaustion of every possible ․ alternative,' it does require ‘serious, good faith consideration of race-neutral alternatives,’ either prior to or in conjunction with implementation of an affirmative action plan.” Ensley Branch, NAACP v. Seibels, 31 F.3d 1548, 1571 (11th Cir.1990). The requirement that government consider the efficacy of race-neutral alternatives to programs that allocate valuable public benefits based on race is extremely important. Yet UGA has presented no evidence that it rejected or even gave meaningful thought to substituting wholly race-neutral alternatives for its race-conscious admissions policy; indeed, at oral argument, UGA conceded that it has introduced no such evidence.
UGA also proffered no expert testimony or other evidence establishing that race-neutral alternatives would necessarily be ineffective in creating a diverse, or for that matter even a racially-diverse, student body. For example, on the record presented in this case, there is no persuasive evidence for UGA's apparent assumption that eliminating race from the TSI formula would have a significant adverse effect on the number of non-whites admitted to the freshman class.25 This failure of proof further demonstrates why UGA has not met its burden of demonstrating that its freshman admissions policy is narrowly tailored. See Croson, 488 U.S. at 507, 109 S.Ct. at 729 (plurality op.) (affirmative action plan not narrowly tailored where “there does not appear to have been any consideration of the use of race-neutral means to increase minority business participation in city contracting”); Wygant, 476 U.S. at 280 n. 6, 106 S.Ct. at 1850 n. 6 (plurality op.) (“The term ‘narrowly tailored’ ․ require[s] consideration of whether lawful alternative and less restrictive means could have been used ․ [T]he classification at issue must ‘fit’ with greater precision than any alternative means.”) (citation omitted).
We therefore conclude that UGA's admissions policy is not narrowly tailored under the factors appropriate for evaluating a race-conscious university admissions policy. That said, to avoid any confusion, we also make clear that UGA's policy is not narrowly tailored under the factors discussed by the Supreme Court in Paradise and applied in Birmingham and other decisions from this Court. First, as discussed above, UGA has failed to show that its policy is sufficiently flexible in its consideration of race; on the contrary, by mechanically awarding bonus points to all non-white applicants at the TSI stage while unnecessarily excluding consideration of other factors relevant to diversity, the policy's inflexibility is apparent. Second, UGA has failed to show that its use of race does not unfairly burden “innocent” third parties-specifically, white applicants, or more particularly, white applicants who may have greater potential to enhance student body diversity. Third, UGA has failed to show that it seriously considered race-neutral alternatives, let alone that it appropriately rejected race-neutral alternatives as inefficacious. Fourth, while there is no clear evidence that UGA has set a “numerical goal” for the number of non-white applicants (and hence there is no basis to compare that goal to general population figures), the bonus that UGA has chosen to award non-white applicants is concededly wholly arbitrary and has no discernable relationship to the representation of non-whites in the student population or the population at large.26
The only Paradise factor not addressed above concerns the duration of UGA's race-conscious policy. That factor does not support the constitutionality of this policy. UGA has voluntarily adopted a race-conscious admissions policy each year since 1990, when it was relieved from the consent decree based upon a finding that it had satisfactorily remediated its past discrimination against African-Americans. There is no evidence that UGA envisions an end to its practice of mechanically awarding preferential treatment to non-white applicants in its freshman admissions process. UGA's undergraduate admissions director, McDuff, candidly acknowledged that she did not know when if ever UGA would discontinue awarding a racial bonus to certain applicants. There is evidence that UGA reviews its freshman admissions policy annually, and that the relative value accorded race in the TSI equation has gone down since the initial formulation of the revised policy in 1995-96. The fact remains, however, that there is no proof on this record that UGA's preferential treatment of non-white applicants has an end point, and there is no indication that UGA's annual review is undertaken with the idea of wholly eliminating race as a factor. See Eisenberg, 197 F.3d at 132 (periodic review of diversity goals, without considering elimination of race factor, “does not make the [selection] policy narrowly tailored”). Although, as discussed above, we do not believe that this factor should have a great deal of significance in evaluating a race-conscious admissions plan designed to achieve student body diversity, it is a factor we consider under Paradise, and it does not support UGA's claim that the policy is narrowly tailored.
UGA, in reality, makes little serious effort to defend its policy under Paradise or any other narrowly tailoring test. Rather, UGA tries to analogize its policy to the Harvard Plan and thereby bring its policy under Justice Powell's opinion in Bakke. UGA contends that, under its policy, there is no set-aside for non-white applicants, and race is not automatically a decisive factor in how applicants are treated. According to UGA, Justice Powell's discussion of the Harvard Plan stands for the broad proposition that a race-conscious university admissions plan is always constitutionally valid so long as race is not the sole focus of the plan and is instead merely a “plus” considered alongside other factors in evaluating an applicant pool composed of both white and non-white applicants.
As explained above, we do not believe that Justice Powell's opinion is binding, and his discussion of the Harvard Plan was entirely dicta. In any event, UGA's plan is a far cry from the Harvard Plan.27
It is correct to say that UGA's policy for the Fall 1999 class, unlike its pre-1996 freshman admissions policies, did not create a dual-track admissions system akin to the one rejected by five members of the Court in Bakke. But it also is true that UGA's rigid, ineluctable, and mechanical awarding of bonus points to each and every non-white applicant bears little resemblance to the “flexible” Harvard plan addressed by Justice Powell:
In such an admissions program, race or ethnic background may be deemed a “plus” in a particular applicant's file, yet it does not insulate the individual from comparison with all other candidates for the available seats. The file of a particular black applicant may be examined for his potential contribution to diversity without the factor of race being decisive when compared, for example, with that of an applicant identified as an Italian-American if the latter is thought to exhibit qualities more likely to promote beneficial educational pluralism. Such qualities could include 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. In short, an admissions program operated in this way is flexible enough to consider all pertinent elements of diversity in light of the particular qualifications of each applicant, and to place them on the same footing for consideration, although not necessarily according them the same weight.
438 U.S. at 317-18, 98 S.Ct. at 2762 (emphasis added).
Unlike the Harvard plan described by Justice Powell, UGA's policy does not allow admissions officers to consider “all pertinent elements of diversity” or to decide-in awarding the 0.5 racial bonus-that the “potential contribution to diversity ․ of an applicant identified as Italian-American” is greater than that of a non-white applicant. The 0.5 point bonus is awarded mechanically, based entirely on the applicant's race. And while it is true that a small number of the other TSI factors may, to a limited extent, capture qualities beyond race that contribute to student body diversity, they certainly do not come close to capturing to the same degree the qualities or life experiences that would be taken into account if each applicant-including her potential contribution to diversity-were assessed fully and fairly as an individual.
To reiterate, it is only at the ER stage that an applicant's file is ever read and qualitatively evaluated by admissions officers; that kind of evaluation does not occur at the TSI stage, even though final admissions decisions are made at that stage based in part upon factors that UGA says are meant to enhance diversity. The inflexibility of the policy's vision of diversity weighs powerfully against a finding that the policy is narrowly tailored to achieve the stated goal of student body diversity. It also demonstrates how UGA's policy, whether or not modeled after the Harvard Plan (as UGA claims), differs in the most significant ways from that Plan.
A much closer analogy to UGA's policy is the admissions program invalidated on narrow tailoring grounds by the Fourth Circuit in Tuttle. That program, although it operated somewhat differently than UGA's policy, likewise mechanically assigned a value to minority applicants in order to obtain a diverse student body. The program there determined admission to a popular alternative kindergarten. One of the stated goals of the admissions policy was to promote diversity. The policy defined diversity using three equally weighted factors: (1) whether the applicant was from a low-income or special family background; (2) whether English was the applicant's first or second language; and (3) the racial or ethnic group to which the applicant belonged. Because the applicant pool was always larger than the number of available positions, the school offered admission to applicants via a lottery. But if the applicant pool did not reflect, within 15%, the county-wide student population percentages for all three of the diversity factors, applicants having those characteristics were given special numerical weight in the lottery, so that applicants from under-represented groups, as defined by the policy, had an increased probability of selection.
The Fourth Circuit found this approach unconstitutional, in part because it represented the kind of racial balancing that the district court found (prematurely) here as well, but also because it did not allow individualized examination of an applicant's potential contribution to diversity:
The School Board argued that the Policy was extremely flexible because instead of a set numerical goal, the final random results of the weighted lottery ultimately determined admissions. We disagree. Since ․ admissions are based on availability, if the applicant pool does not reflect the required 15% racial and ethnic diversity, each child's probability of selection in the lottery is adjusted corresponding to his or her stated race. In Bakke, Justice Powell explained that constitutionally permissible programs such as the Harvard College admissions program promote diversity by “treat[ing] each applicant as an individual in the admissions process.” 438 U.S. at 318, 98 S.Ct. at 2733. The Policy, like the Davis admissions program in Bakke, does not treat applicants as individuals. The race/ethnicity factor grants preferential treatment to certain applicants solely because of their race.
195 F.3d at 707. Tuttle properly emphasizes the need for a race-conscious admissions program to be flexible in its consideration of race and other factors contributing to student body diversity. Tuttle also underscores the distinctions between UGA's policy and the Harvard Plan endorsed by Justice Powell.
In the end, the narrow-tailoring inquiry required by strict scrutiny cannot be undertaken without constant reference to the core principles of Equal Protection. “Above all else, the framers of the Civil War Amendments intended to deny to the States the power to discriminate against persons on account of their race․ Th[ese] Amendments were unquestionably designed to condemn and forbid every distinction, however trifling, on account of race.” Oregon v. Mitchell, 400 U.S. 112, 126-27, 91 S.Ct. 260, 265-66, 27 L.Ed.2d 272 (1970) (Black, J.). Although the Supreme Court has recognized very limited exceptions to that vital principle-such as where the state seeks to vindicate the guarantee of Equal Protection by remedying its own past discrimination-the fact remains that “[a] core purpose of the Fourteenth Amendment was to do away with all governmentally imposed discrimination based on race.” Palmore v. Sidoti, 466 U.S. 429, 432, 104 S.Ct. 1879, 1881-82, 80 L.Ed.2d 421 (1984) (emphasis added). Race-conscious decision-making is fundamentally in conflict with the idea of Equal Protection, and when a state attempts to allocate valuable benefits (including admission to public universities) on the basis of race, it is the obligation of the courts to require a powerful showing before upholding the state's discrimination. See Wygant, 476 U.S. at 273-74, 106 S.Ct. at 1847 (plurality op.) (“Any preference based on racial or ethnic criteria must necessarily receive a most searching examination to make sure that it does not conflict with constitutional guarantees.”).
UGA fails to meet that heavy burden on this record. A “ ‘racial classification, regardless of purported motivation, is presumptively invalid and can be upheld only upon an extraordinary justification.’ ” Shaw I, 509 U.S. at 643-44, 113 S.Ct. at 2825. No reasonable juror could find an extraordinary justification here. We therefore conclude that the district court properly entered summary judgment in Plaintiffs' favor on their challenge to UGA's 1999 freshman admissions policy, not because student body diversity can never be a compelling interest-we assume for present purposes only that it can-but rather because the policy is not narrowly tailored to serve that interest.
Before addressing the other rulings challenged in this appeal, we note that the Intervenors have raised an additional argument in defense of the constitutionality of UGA's freshman admissions policy. Unlike UGA itself, the Intervenors contend that UGA's favorable treatment of African-American applicants is necessary, still, to ameliorate the vestiges of UGA's past intentional discrimination against African-Americans. Intervenors argue that the need to ameliorate the present effects of past discrimination is unquestionably a compelling interest, and that UGA's freshman admissions policy is narrowly tailored to serve that vital goal.
This argument fails for several reasons. First, Intervenors did not advance it in any meaningful way at the time of summary judgment. The Intervenors attempt to avoid that fact by insisting that because “[n]o party sought summary judgment disposition of this issue ․ this important defense remained for resolution at trial after the district court resolved the summary judgment motions.” Intervenors Opening Br. at 21-22. The problem, however, is that the Plaintiffs moved for final summary judgment, not merely partial summary judgment on the issue of whether the policy could be justified in the name of student body diversity. Accordingly, it became incumbent upon the Intervenors to respond by, at the very least, raising in their opposition papers any and all arguments or defenses they felt precluded judgment in Plaintiffs' favor. See, e.g., Celotex, 477 U.S. at 322, 106 S.Ct. at 2552; Harper v. Delaware Valley Broadcasters, Inc., 743 F.Supp. 1076, 1090-91 (D.Del.1990) (burden is on defendant to adduce evidence supporting affirmative defense, not upon movant to negate its existence), aff'd, 932 F.2d 959 (3d Cir.1991); Fed.R.Civ.P. 56(e) (“When a motion for summary judgment is made and supported as provided in this rule, an adverse party may not rest upon the mere allegations or denials of the adverse party's pleading, but the adverse party's response, by affidavits or as otherwise provided in this rule, must set forth specific facts showing that there is a genuine issue for trial.”). Intervenors cannot readily complain about the entry of a summary judgment order that did not consider an argument they chose not to develop for the district court at the time of the summary judgment motions.
Second, there is little, if any, persuasive evidence in this record tending to support the proposition that UGA's preferential treatment of all non-white applicants was necessary or appropriate to remedy present effects of UGA's past discrimination against African-Americans. Indeed, not only does UGA expressly reject that position, its officials testified unequivocally that the policy was not motivated by a desire to combat any present effects of past discrimination. Moreover, in 1989, OCR concluded that UGA had sufficiently ameliorated the vestiges of its past discrimination against African-Americans and was no longer in violation of Title VI. Far from disputing that finding, UGA agreed with it, then and now. Merely asserting that OCR's determination is “not dispositive” does not establish a justification for continued discrimination against whites, let alone to show that the particular means chosen by UGA are narrowly tailored. See Croson, 488 U.S. at 506, 109 S.Ct. at 728 (finding municipality's set-aside program not narrowly tailored in part because “[t]he random inclusion of racial groups that, as a practical matter, may never have suffered from discrimination in the construction industry in Richmond suggests that perhaps the city's purpose was not in fact to remedy past discrimination.”).
We are aware of UGA's history of intentional discrimination against African-Americans, and do not diminish the serious harm that it caused. Nor do we diminish the need for UGA and comparable institutions to make particular efforts to provide equal opportunities for African-Americans who continue to suffer long-term consequences of state-sponsored discrimination, and who look to the university system as a vital means to economic and political advancement. The Intervenors' proposed remedial justification, however, does not on this record defeat the Plaintiffs' entitlement to summary judgment.
Plaintiffs assert that the district court erred by rejecting their request for prospective injunctive relief against UGA's ongoing use of race as a factor in the freshman admissions process.28 The district court denied that request because none of the Plaintiffs could show a likelihood that they would ever again be subjected to the freshman admissions process, and hence they could not meet the Supreme Court's requirements for prospective injunctive relief. We find no reversible error.
Resolution of this issue is dictated by our recent decision in Wooden v. Board of Regents, 247 F.3d 1262 (11th Cir.2001); indeed, the arguments here are virtually identical to those we rejected there. In that case, one of the plaintiffs (Tracy) was unlawfully denied freshman admission to UGA on the basis of race. The district court awarded Tracy nominal damages for the denial of admission, but refused to award prospective injunctive relief, emphasizing that Tracy could not show any threat of future injury because he had already enrolled at UGA as a transfer student. We affirmed, emphasizing that “[s]imply because a party prevails on the merits of a constitutional claim does not mean that the party is automatically entitled to prospective injunctive relief.” Id. at 1283.
Rather, to have standing to obtain forward-looking relief, a plaintiff must show a sufficient likelihood that he will be affected by the allegedly unlawful conduct in the future. [See City of Los Angeles v. Lyons, 461 U.S. 95, 103 S.Ct. 1660, 75 L.Ed.2d 675 (1983); Church v. City of Huntsville, 30 F.3d 1332 (11th Cir.1994) ]․
Tracy is now a student at UGA, and there is no evidence that he intends to re-apply for admission to UGA under any version of the freshman admissions policy. There is no likelihood, therefore, that he will ever again be exposed to UGA's allegedly discriminatory freshman admissions process. As Lyons makes clear, the fact that others may be exposed to that process in the future is not sufficient for Tracy to obtain prospective relief that will not benefit him in conjunction with his individual claim.
Plaintiffs make several counter-arguments, none of which is persuasive․ Plaintiffs maintain that denying standing to Tracy is unfair because otherwise UGA's freshman admissions policy will be effectively unreviewable. They appear to believe that prospective relief could only be sought by someone who, having been rejected, is then prepared to wait out the duration of the litigation without seeking to gain admission to UGA as a transfer student. Moreover, according to Plaintiffs, even if an individual plaintiff seems likely to prove a violation, UGA can-as it has in the past-avoid any risk of far-reaching court-ordered prospective relief with regard to its freshman admissions policy simply by offering admission to that individual. At least on the record before us, we do not share Plaintiffs' concern. Suffice it to say that there has been no showing that the pool of potential plaintiffs able to challenge UGA's freshman admissions policy on these grounds is drying up. We also are confident that, regardless of any prospective relief a court might order, UGA would not ignore the import of a controlling federal court decision holding that its freshman admissions policy is unconstitutional. In any event, as discussed above, Tracy is not in a position to advance these concerns as a reason to grant him standing to seek prospective injunctive relief because he was not affected by the version of UGA's freshman admissions policy now in place.
Finally, Plaintiffs maintain that [Texas v. Lesage, 528 U.S. 18, 120 S.Ct. 467, 145 L.Ed.2d 347 (1999) (per curiam) ], somehow entitles Tracy to prospective injunctive relief. But there is no indication in Lesage that the Court intended to alter the well-settled prerequisites to granting such relief. Plaintiffs' argument appears to be based on the Court's statement that “[o]f course, a plaintiff who challenges an ongoing race-conscious program and seeks forward-looking relief need not affirmatively establish that he would receive the benefit in question if race were not considered.” 528 U.S. at 21, 120 S.Ct. at 468. But Lesage plainly does not hold that a plaintiff may obtain prospective injunctive relief merely by alleging “an ongoing or imminent constitutional violation,” id., arising out of a practice that has not injured him to date and is highly unlikely to injure him in the future.
Nor can Lesage be read to create an exception to Lyons where the discriminatory admissions policy is still in place. If the Supreme Court intended so significant and potentially far-reaching a change in the law of standing, surely it would have said so directly, or at least cited Lyons. To reiterate, standing, qua standing, was not even an issue in Lesage, and no question of prospective relief arose there because it appeared that the plaintiff had abandoned his claim that the defendant university was still administering a discriminatory admissions policy. 528 U.S. at 22, 120 S.Ct. at 469. Tracy, therefore, lacks standing to seek prospective injunctive relief with respect to UGA's revised freshman admissions policy, and the district court correctly declined to grant that relief.
247 F.3d at 1284-87 (citations, internal quotation marks, and brackets omitted); see also Shotz v. Cates, 256 F.3d 1077, 1082 (11th Cir.2001) (affirming dismissal of claim for prospective injunctive relief because “the plaintiffs do not allege a real and immediate threat of future discrimination”).
Plaintiffs' arguments in this case are little different than those we rejected in Wooden. There is simply no evidence that any of the Plaintiffs is applying or is likely to apply in the future for admission to UGA via the freshman admissions process. It is also undisputed that at least two of the Plaintiffs, at or about the time of the complaint, were attending universities other than UGA and could have applied only via the transfer process (which does not consider race). In short, none of the Plaintiffs was in the process of re-applying, or was likely to re-apply, to UGA through the race-conscious freshman admissions process, let alone the 1999 version of that process.
The Supreme Court has never (in Lesage or elsewhere) held that a plaintiff seeking prospective injunctive relief does not have to show ongoing harm or a likelihood of future harm.29 Plaintiffs suggest that the harm here is ongoing, in the sense that UGA is continuing to operate an unlawfully discriminatory admissions policy. But even holding aside the fact that-at this moment-UGA is no longer considering race in making freshman admissions decisions, the problem for the Plaintiffs is that they are not suffering and are not about to suffer any concrete injury from UGA's awarding of racial preferences in the freshman admissions process.
Plaintiffs also suggest that Bogrow had standing to seek a prospective injunction even under Lyons because at the time of the complaint she (unlike Beckenhauer and Johnson) had not yet accepted admission at another college and hence would have had to seek admittance to UGA via the freshman admissions process rather than the transfer process. This argument fails because it was not fairly raised below or even in Plaintiffs' opening brief on appeal. In any event, the argument is unconvincing. It is true that a party's standing to sue is generally measured at the time of the complaint, with the effect of subsequent events generally analyzed under mootness principles. See Wooden, 247 F.3d at 1286-87. But there has been no adequate showing that Bogrow actually had standing at the time of the complaint. There is no specific allegation in the complaint that Bogrow intended to re-apply to UGA via the freshman admissions process, or that a renewed freshman application was pending or about to be pending. The fact that Bogrow apparently had not yet accepted admission at another university does not, by itself, mean that she was facing or was likely to face in the future UGA's discriminatory freshman admissions process. It was Bogrow's burden to establish her standing under Lyons to seek prospective injunctive relief, and she did not meet that burden on this record.
We acknowledge Plaintiffs' concern that practical difficulties may make it difficult for some plaintiffs to have standing to obtain broad prospective injunctive relief with respect to UGA's freshman admissions policies. But this Court cannot simply reject Lyons absent further guidance from the Supreme Court. If anything, the fact that UGA has voluntarily suspended using race as a factor demonstrates that, regardless of the imposition of prospective injunctive relief, UGA is not prepared to continue relying upon race in the admissions process in contravention of an unambiguous federal court declaration that so rigid and incomplete a use of race to achieve student body diversity is unconstitutional. We are confident that UGA would not ignore the import of a final ruling declaring its policy's use of race unlawful, and there is no need to consider at this stage what equitable powers a court might invoke if such a final ruling were honored only on a case-by-case basis.
Accordingly, the district court did not err by denying Plaintiffs' requests for prospective injunctive relief.
The district court decertified this case as a class action after finding that the Plaintiffs lacked standing to seek prospective injunctive relief-the sole remedy sought on behalf of the class. As we recently explained in Wooden, a plaintiff cannot serve as a class representative if she lacks standing to advance the class's claim. 247 F.3d at 1288 (“just as a plaintiff cannot pursue an individual claim unless he proves standing, a plaintiff cannot represent a class unless he has standing to raise the claims of the class he seeks to represent”) (citing Prado-Steiman, 221 F.3d at 1280). For the reasons discussed above, none of the named Plaintiffs has standing to seek prospective injunctive relief. Accordingly, these Plaintiffs could not serve as representatives for a class seeking that relief, and the district court did not abuse its discretion by decertifying the class. See Forehand v. Florida State Hosp., 89 F.3d 1562, 1566 (11th Cir.1996) (district court retains power to decertify class) (citing Fed.R.Civ.P. 23(d)(1)).30
Finally, the Intervenors contend that the district court erred by denying their pre-trial motion for a “special case management scheduling order” or in the alternative for an extension of discovery. Intervenors sought that relief primarily to give them additional time to develop support for an argument not advanced by UGA-that UGA's preferential treatment of non-whites could be justified as necessary to remediate the lingering effects of past discrimination. The district court denied the motion, which was vigorously opposed by both the Plaintiffs and the Defendants.
We find no abuse of discretion in the denial of the Intervenors' motion. Their argument fails to surmount two substantial hurdles. First, we accord district courts broad discretion over the management of pre-trial activities, including discovery and scheduling. See, e.g., Chudasama v. Mazda Motor Corp., 123 F.3d 1353, 1366 (11th Cir.1997) (“district courts enjoy broad discretion in deciding how best to manage the cases before them”); McCutcheon, 86 F.3d at 190 (noting the “broad discretion which is allowed a trial court to manage its own docket”). Second, courts have broad authority to limit the ability of intervening parties to expand the scope of a proceeding beyond the issues litigated by the original parties. See Vinson v. Washington Gas Light Co., 321 U.S. 489, 498, 64 S.Ct. 731, 735, 88 L.Ed. 883 (1944) (noting the “usual procedural rule” that “an intervenor is admitted to the proceedings as it stands, and in respect of the pending issues, but is not permitted to enlarge those issues or compel an alteration of the nature of the proceeding”); Lamprecht v. FCC, 958 F.2d 382, 389 (D.C.Cir.1992) (“Except in extraordinary cases, ․ intervenors ‘may only join issue on a matter that has been brought before the court by another party,’ ” and “cannot expand the proceedings”) (citations omitted). That authority certainly may be exercised when an intervenor seeks, over the opposition of the original parties, to alter pre-trial deadlines and take additional discovery on an issue not litigated by the original parties.
On appeal, the Intervenors basically reprise the arguments that they presented to the district court in their motion. They contend that the procedural and factual posture of the case was extremely complex, and that all participants in the litigation would have benefitted from additional time before advancing to summary judgment. But a district court is entitled to establish proper pre-trial procedures and set an appropriate pre-trial schedule, and the district court did not abuse its discretion in this case (especially given that none of the original parties found the posture too unwieldy). If the Intervenors felt time pressure, that was an inevitable consequence of their choice to intervene, and their objection to the district court's timetable strikes at the very core of the power accorded district courts to manage pre-trial proceedings.31
Intervenors also contend that additional time should have been afforded so that they could have developed a record supporting a remedial justification for UGA's consideration of race. As Intervenors see it, whether UGA has eliminated the vestiges of past discrimination is still an open question. But none of the parties-or for that matter the federal government-accepts that claim. Moreover, the issue raised by Intervenors would have greatly expanded the scope and burden of the case, and quite probably have necessitated further delays beyond those ostensibly sought by the Intervenors. Especially given the significance of the lawsuit, and critical importance to UGA and its future freshman applicants of resolving this matter as soon as possible, the district court had ample grounds for declining to modify or halt proceedings. The district court therefore did not abuse its discretion by denying the Intervenors' motion, and, on this issue as well, we affirm.
To summarize, we hold that the district court properly entered summary judgment in Plaintiffs' favor on their challenge to UGA's 1999 freshman admissions policy, not because student body diversity can never be a compelling interest (we assume for present purposes that it can), but rather because this policy is not narrowly tailored to serve that interest. In all other respects, we affirm the district court's rulings.
1. Another Plaintiff, Lindsay Donaldson, was later dismissed. The district court denied Plaintiffs' efforts to add additional parties.
2. The district court denied the Plaintiffs' motion for leave to amend their complaint to add a new claim of disparate impact discrimination based on Department of Education (“DOE”) regulations promulgated under § 602 of Title VI. At oral argument, Plaintiffs abandoned their appeal of that ruling. See Alexander v. Sandoval, 531 U.S. 1049, 121 S.Ct. 1511, 1516, 149 L.Ed.2d 517 (2001) (holding that there is no private right of action to enforce disparate impact regulations promulgated under Title VI of Civil Rights Act of 1964, because “Title VI itself directly reaches only instances of intentional discrimination”) (citation, internal quotation marks, and brackets omitted). Even aside from Sandoval, there was no error in rejecting Plaintiffs' proposed amendment, which did not adequately allege any facially-neutral component of the freshman admissions policy having a disparate impact on whites. See Galindo v. ARI Mut. Ins. Co., 203 F.3d 771, 777 n. 10 (11th Cir.2000) (leave to amend may be denied where the proposed claim fails to state a claim and therefore would be futile).
3. The court found that UGA's gender preference violated Title IX, and did not discuss any constitutional argument on that issue.
4. The AI is a statistic that weighs and combines an applicant's SAT scores and GPA.
5. Quality of curriculum was also taken into account, in that, if the applicant's curriculum met the criteria for “most difficult,” an AI of 2.81 would suffice for automatic admission.
6. For shorthand, we use “white” and “non-white” to refer to “caucasian” and “non-caucasian” as those latter terms are used in the policy. Also for ease of reference, we refer to the 0.5 point bonus awarded non-white applicants as the “race factor” or “race bonus,” even though this bonus is classified by the policy as one for race/ethnicity. Throughout this opinion, unless otherwise stated, our discussions of race as a factor in university admissions are meant to refer as well to ethnicity, even though we recognize that issues surrounding race-based decision-making may not always be the same as those relating to ethnicity.
7. Being a male was worth 0.25 points.
8. Defendants and Intervenors do not appeal the district court's finding that UGA's consideration of gender is unlawful.
9. The parties do not treat the analysis under Title VI as proceeding any differently than the analysis under Equal Protection; accordingly, we discuss the issues only with reference to the Constitution. See Burton, 178 F.3d at 1202.
10. Justice Powell's opinion may be narrower than that of Justice Brennan if viewed from the perspective of which constitutional test should be applied to race-based decision-making. It is more accurate, however, to focus on how the two opinions treat the diversity interest, because the debate concerns the constitutional status of that interest, not which constitutional test we should apply.
11. If he had done so, of course, Justice Powell's opinion would be of no help to UGA, which has disavowed any form of present-effects-of-past-discrimination argument.
12. The Supreme Court has not compelled us to find a “holding” on each issue in each of its decisions. On the contrary, the Court has indicated that there may be situations where even the Marks inquiry does not yield any rule to be treated as binding in future cases. In Nichols v. United States, 511 U.S. 738, 114 S.Ct. 1921, 128 L.Ed.2d 745 (1994), the Court considered whether an uncounseled misdemeanor conviction, valid due to the absence of the imposition of a prison term, is also valid when used to enhance the punishment for a subsequent conviction. That issue was addressed in an earlier decision, Baldasar v. Illinois, 446 U.S. 222, 100 S.Ct. 1585, 64 L.Ed.2d 169 (1980), where the Court fractured much as it did in Bakke. The Supreme Court's initial task in Nichols was to determine, by undertaking the Marks inquiry, whether Baldasar had indeed decided the issue. After examining Baldasar, the Court concluded-as had several lower courts-that the Marks inquiry was simply not helpful:In Marks [ ], we stated that when a fragmented Court decides a case and no single rationale explaining the result enjoys the assent of five Justices, the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds. This test is more easily stated than applied to the various opinions supporting the result in Baldasar. A number of Courts of Appeals have decided that there is no lowest common denominator or narrowest grounds that represents the Court's holding. Another Court of Appeals has concluded that the holding in Baldasar is Justice Blackmun's rationale; yet another has concluded that the consensus of the Baldasar concurrences is roughly that expressed by Justice Marshall's concurring opinion. State courts have similarly divided. The Sentencing Guidelines have also reflected uncertainty over Baldasar. We think it not useful to pursue the Marks inquiry to the utmost logical possibility when it has so obviously baffled and divided the lower courts that have considered it. This degree of confusion following a splintered decision such as Baldasar is itself a reason for reexamining that decision.511 U.S. at 745-46, 114 S.Ct. at 1926-27 (emphasis added) (citations, internal quotation marks, and brackets omitted).This Court obviously does not have the option of re-examining Bakke as we might our own precedent. Nevertheless, the Supreme Court has recognized that there will be situations where no binding “rule” may be taken from a fractured decision, and the Marks inquiry is ultimately “not useful.” Id. Bakke surely presents such a situation.
13. Contrary to UGA's and Intervenors' argument, there is no unanimity regarding the status of Justice Powell's Bakke opinion as binding precedent on the validity of student body diversity as an interest sufficient to justify race-based school admissions decisions. One court has expressly identified Justice Powell's opinion as binding precedent on this issue. See Smith v. University of Wash. Law Sch., 233 F.3d 1188, 1201 (9th Cir.2000) (pursuant to Bakke “educational diversity is a compelling governmental interest that meets the demands of strict scrutiny of race-conscious measures”), cert. denied, 532 U.S. 1051, 121 S.Ct. 2192, 149 L.Ed.2d 1024 (2001). But several other courts, like the district court in this case, have held that Justice Powell's opinion is not binding precedent, and that diversity is not a compelling interest. See Hopwood v. Texas, 78 F.3d 932, 944 (5th Cir.) (“Justice Powell's view in Bakke is not binding precedent on this issue.”), cert. denied, 518 U.S. 1033, 116 S.Ct. 2581, 135 L.Ed.2d 1095 (1996); Grutter v. Bollinger, 137 F.Supp.2d 821, 848 (E.D.Mich.2001) (concluding that “Bakke does not stand for the proposition that a university's desire to assemble a racially diverse student body is a compelling state interest”), appeal filed; cf. Lutheran Church-Missouri Synod v. FCC, 141 F.3d 344, 354 (D.C.Cir.1998) (stating, without addressing Bakke, that diversity cannot “be elevated to the ‘compelling’ level”). Other courts have sought a middle ground, essentially finding that Justice Powell's opinion may not be binding, but nevertheless is a persuasive reason to treat diversity as a compelling interest. See Gratz v. Bollinger, 122 F.Supp.2d 811, 819-21 (E.D.Mich.2000), appeal filed. Still other courts have acknowledged the uncertainty surrounding Bakke, but have avoided resolving the precedential effect of Justice Powell's opinion by deciding the case upon grounds not dependent upon the status of student body diversity as a compelling interest. See Brewer v. West Irondequoit Cent. Sch. Dist., 212 F.3d 738, 747-49 (2nd Cir.2000) (noting that “there is much disagreement among the circuit courts as to ․ the state of the law under current Supreme Court jurisprudence,” but concluding that, regardless of Bakke, reducing racial isolation may be a compelling interest under Second Circuit precedent); Eisenberg v. Montgomery County Pub. Sch., 197 F.3d 123, 130 (4th Cir.1999) (explaining that status of educational diversity as a compelling interest is “unresolved,” but declining to resolve the issue and instead deciding the case solely on narrow-tailoring grounds); Wessmann v. Gittens, 160 F.3d 790, 795, 800 (1st Cir.1998) (noting that “[t]he question of precisely what interests government may legitimately invoke to justify race-based classifications is largely unsettled,” but concluding that the defendant's asserted interests were different from the interest endorsed by Justice Powell in Bakke ); Buchwald v. University of New Mexico Sch. of Med., 159 F.3d 487, 499 (10th Cir.1998) (noting the absence of “a clear majority opinion” in Bakke, but awarding qualified immunity to defendants who relied upon that case in adopting a preference based on durational residency); McNamara v. City of Chicago, 138 F.3d 1219, 1222 (7th Cir.1998) (citing Bakke for statement that “whether there may be compelling interests other than remedying past discrimination remains ‘unsettled,’ ” but finding that defendant's interest in remediating past discrimination was valid).
14. This Court has never addressed when, if ever, student body diversity may be a compelling interest. Nor have we addressed whether, under recent Supreme Court precedent, remedying the present effects of the defendant's past discrimination is the only compelling interest that may justify race-conscious decision-making.
15. There have been occasional references in Supreme Court dicta to student body diversity as a constitutionally legitimate interest. See Wygant, 476 U.S. at 286, 106 S.Ct. at 1853 (O'Connor, J., concurring in part); id. at 315, 106 S.Ct. at 1868-69 (Stevens, J., dissenting). But these references tend simply to describe, rather than endorse, Justice Powell's remarks in Bakke. Additionally, in Metro Broadcasting, a majority of the Court found that diversity in broadcasting was an “important” interest for purposes of intermediate scrutiny analysis, drawing an analogy to student body diversity. 497 U.S. at 566, 110 S.Ct. at 3010 (“[W]e conclude that the interest in enhancing broadcast diversity is, at the very least, an important governmental objective and is therefore a sufficient basis for the Commission's minority ownership policies. Just as a diverse student body contributing to a robust exchange of ideas is a constitutionally permissible goal on which a race-conscious university admissions program may be predicated, the diversity of views and information on the airwaves serves important First Amendment values.”) (internal quotation marks omitted) (citing Bakke, 438 U.S. at 311-13, 98 S.Ct. at 2759-60 (Powell, J.)). Metro Broadcasting 's use of an intermediate scrutiny standard was expressly repudiated in Adarand, however.
16. See Bethel Sch. Dist. No. 403 v. Fraser, 478 U.S. 675, 681, 106 S.Ct. 3159, 3163, 92 L.Ed.2d 549 (1986) (“[P]ublic education must prepare pupils for citizenship in the Republic [by conveying] the fundamental values of habits and manners of civility essential to a democratic society ․ includ[ing] tolerance of divergent political and religious views.”) (citations and internal quotation marks omitted); Ambach v. Norwick, 441 U.S. 68, 76-77, 99 S.Ct. 1589, 1594-95, 60 L.Ed.2d 49 (1979) (“[P]ublic schools [are] an ‘assimilative force’ by which diverse and conflicting elements in our society are brought together on a broad but common ground.”) (citations omitted).
17. The district court did not address the narrow tailoring issue. It is well-settled, however, that “[w]hen reviewing a grant of summary judgment, the court of appeals may affirm if there exists any adequate ground for doing so, regardless of whether it is the one on which the district court relied.” Fitzpatrick v. City of Atlanta, 2 F.3d 1112, 1117 (11th Cir.1993). The narrow tailoring issue was argued fully before the district court and has been argued again in this Court, and no purpose would be served by a remand to allow the district court another opportunity to address this legal question.
18. UGA does not contend that its compelling interest in this case is merely a racially-diverse student body. Nor would such an interest be sufficient, even under Justice Powell's opinion in Bakke. See 438 U.S. at 315, 98 S.Ct. at 2761 (Powell, J.) (the “nature of [diversity as a] state interest that would justify consideration of race or ethnic background ․ is not an interest in simple ethnic diversity”).
19. Those factors are: race, parent or sibling ties to UGA, hours spent on extracurricular activities, hours spent on summer work, hours spent on school-year work, first-generation college, and Georgia residency. It is doubtful whether the last factor, while it serves other legitimate ends, contributes anything to the diversity of the University of Georgia's student body. Indeed, it is clear that the only factors meaningfully capturing characteristics of a diverse student body are race and first-generation college.
20. One may ask why, if, as UGA claims, approximately 85% of each entering class is admitted at the race-neutral First Notice stage, the administrative burden asserted by UGA is so great as to prevent it from providing meaningful, personalized consideration of race and other diversity factors in selecting the remaining 15% of the class.
21. Two other factors were awarded 0.5 TSI points: “both parents: no college education” (a factor that certainly may be relevant to enhancing student body diversity) and Georgia residency (a factor with little if any relevance to diversity at a Georgia university).
22. The maximum TSI points under “Extracurricular Activity Hours” is 0.25, which is available only if the student reports 75 or more hours. The maximum TSI points under “Summer Work Hours” is 0.25, which is available if the student reports 50 or more hours; the same is true for “School Year Work Hours.” These maximums actually overstate the value accorded extracurricular and work activities in the TSI, because a student who self-reports just one hour of extracurricular activity or summer or school year work activity still receives 0.10 points.
23. We observe that the TSI formula for the 2000 freshman class included two new factors that to some extent capture diversity: economic disadvantage (defined as being from the lower one-third of Georgia counties), and academic disadvantage (meaning that the applicant comes from a low-performing Georgia high school). The constitutionality of the 2000 freshman admissions policy is not before us. Like the 1999 policy, however, it appears that the 2000 policy mechanically awards non-white applicants a 0.5 point bonus at the TSI stage while precluding personalized, qualitative analysis of each applicant's true potential contribution to diversity. In addition, neither of the new factors is worth as much as the race factor.
24. UGA never identifies the exact number of applicants passing through the TSI stage in 1999. In this respect and others, UGA's use of admissions data and the assumptions it purports to draw from that data are wholly inadequate and unpersuasive.
25. John Albright, a UGA admissions officer, answered “No” when asked at deposition whether UGA had analyzed the projected change in African-American enrollment if the racial bonus were eliminated. Tr. of Albright Dep. at 24.
26. Relying primarily on apparently undisputed evidence that, since 1995, UGA's freshman admissions policies have consistently yielded a ten percent admission ratio for non-white applicants, the district court believed that UGA's policy is geared toward obtaining a number of non-white admittees proportionate to the representation of non-whites in the population at large. Even under Justice Powell's opinion in Bakke, diversity is not a compelling interest when it means nothing more than racial balancing. See 438 U.S. at 315, 98 S.Ct. at 2761. In essence, the district court found that the “diversity” sought to be achieved by UGA was proportionality, not a community incorporating the potentially different views and backgrounds of non-whites. The difficulty with this position is that it cannot be sustained on summary judgment. The pieces of evidence and excerpts of testimony highlighted by the district court do not necessarily lend themselves to the inferences drawn by that court, and must be weighed against testimony by UGA officials that no quotas or numerical targets were used in devising the 1999 policy. Whether or not the district court could decide after a trial that UGA's asserted interest in diversity was a mask for its real interest in proportional representation or achieving a particular number of non-white freshmen, the district court could not reach that conclusion at summary judgment, and we cannot and do not reach that conclusion on this record.
27. We offer no view on the constitutionality of the Harvard Plan as described in Justice Powell's Bakke opinion or as implemented by any particular university.
28. Although we review a denial of permanent injunctive relief only for abuse of discretion, see Simmons, 86 F.3d at 1085, here standing is an underlying legal question which we review de novo. See, e.g., Wooden, 247 F.3d at 1271 n. 9.
29. Lesage explained in dicta that a “plaintiff who challenges an ongoing race-conscious program and seeks forward-looking relief need not affirmatively establish that he would receive the benefit in question if race were not considered.” 528 U.S. at 21, 120 S.Ct. at 468-69. As we discussed at length in Wooden, this passage from Lesage clarified the “injury” supporting basic standing in these cases, but did not alter the requirement that each plaintiff establish additional prerequisites to have standing to obtain prospective injunctive relief.
30. Plaintiffs contend that Bogrow had standing at the time of the complaint and at the time of the district court's (later vacated) certification order, and that was enough to keep the class alive under Sosna v. Iowa, 419 U.S. 393, 95 S.Ct. 553, 42 L.Ed.2d 532 (1975), regardless of what she did later. Sosna does not help the Plaintiffs on this record. Sosna stands for the proposition that if a plaintiff had standing at the time of the complaint and at the time a class is certified, subsequent events that may moot her own claim-such as Bogrow's decision to enroll elsewhere-do not necessarily defeat her ability to continue to represent a class whose members still have live claims. See id. at 399-400, 95 S.Ct. at 557. Both the Supreme Court and this Court have distinguished Lyons from situations where, as in Sosna, the plaintiff had standing at the time of the complaint and at the time of certification. See, e.g., Tucker v. Phyfer, 819 F.2d 1030, 1035 (11th Cir.1987). In this case, however, notwithstanding the district court's initial decision to certify a class, there is no showing in the record that Bogrow actually had standing either at the time of the complaint or at the time the district court prematurely entered its certification order.
31. The only case cited by the Intervenors, Majd-Pour v. Georgiana Community Hospital, Inc., 724 F.2d 901 (11th Cir.1984), is inapposite. In that case, we reversed an order dismissing an antitrust claim for lack of subject matter jurisdiction because the district court did not afford the plaintiff adequate time to conduct necessary jurisdictional discovery. Among other clear distinctions, the status of an intervenor is different than that of an ordinary party.
MARCUS, Circuit Judge: