BOSTON PARENT COALITION FOR ACADEMIC EXCELLENCE CORP., Plaintiff, Appellant, v. The SCHOOL COMMITTEE OF the CITY OF BOSTON; Alexandra Oliver-Davila; Michael O'Neill; Hardin Coleman; Lorna Rivera; Jeri Robinson; Quoc Tran; Ernani Dearaujo; Brenda Cassellius, Defendants, Appellees, The Boston Branch of the NAACP; the Greater Boston Latino Network; Asian Pacific Islander Civic Action Network; Asian American Resource Workshop; Maireny Pimental; H.D., Defendants, Intervenors, Appellees.
Plaintiff, a corporation acting on behalf of fourteen parents and children who reside in Boston, alleges that a plan promulgated by the Boston Public Schools for admitting students to Boston Latin School, Boston Latin Academy, and John D. O'Bryant School of Mathematics and Science for the 2021–2022 school year violates the Equal Protection Clause of the Fourteenth Amendment and chapter 76, section 5 of the Massachusetts General Laws. After considering the agreed-upon facts and the parties' arguments, the district court entered judgment in defendants' favor. Bos. Parent Coal. for Acad. Excellence Corp. v. Sch. Comm. of Boston (Boston Parent Coalition), ––– F. Supp. 3d ––––, ––––, Civil Action No. 21-10330-WGY, 2021 WL 1422827, at *17 (D. Mass. Apr. 15, 2021). Plaintiff has appealed the district court's judgment and moves in this court for an injunction preventing the implementation of the 2021-2022 admissions plan pending resolution of the appeal. For the following reasons, we deny plaintiff's motion.
A thorough summary of the facts appears in the district court's opinion, which in turn relied on the parties' agreed-upon statement of facts. We provide the broad framework and then address in our analysis those particular facts deemed significant by the parties in their motion papers on appeal.
Known for the strength of their academic programs, the three above-mentioned schools (what the parties call the “Exam Schools”) have fewer admission slots than there are Boston students who wish to attend them; for the 2020-2021 school year, over 4,000 students applied for about 1,400 slots. For the past twenty years or so, they have selected students for admission based on the students' grade point averages in English Language Arts and Math courses, scores on a standardized admissions test, and their school preferences. Boston Parent Coalition, ––– F. Supp. 3d at ––––, 2021 WL 1422827, at *3.
The onset of the COVID-19 pandemic threatened the schools' ability to conduct the admissions process as in recent years, prompting the School Committee of the City of Boston, the group responsible for managing the Boston Public Schools, to create a Working Group charged with “[d]evelop[ing] and submit[ting] a recommendation to the Superintendent [of the Boston Public Schools, Dr. Brenda Cassellius,] on revised exam school admissions criteria for [the 2021-2022 school year].” Id. at ––––, ––––, 2021 WL 1422827 at *1, 3 (first and last alterations in original). After the Working Group studied the issue, proposed a new plan, and modified that plan based on feedback from School Committee members, the School Committee adopted the 2021-2022 Admissions Plan at a meeting on October 21, 2020. Id. at –––– – ––––, 2021 WL 1422827 at *3–5.
The Plan as adopted conditions a student's eligibility to compete for admission to the Exam Schools on three criteria: (1) residence in one of Boston's twenty-nine zip codes (or inclusion in a special zip code created for students who are homeless or in the custody of the Department of Children and Families); (2) maintenance of a B average or better in English Language Arts and Math during the fall and winter of the 2019-2020 school year or receipt of a “Meets Expectations” or “Exceeds Expectations” score in English Language Arts and Math on the Spring 2019 Massachusetts Comprehensive Assessment System test; and (3) performance at grade level under the Massachusetts Curriculum standards. Eligible students seeking admission must submit a ranked list of school preferences.
The Plan's admissions process plays out in two phases. In phase one, all eligible students are ranked city-wide by grade point average accumulated in English Language Arts and Math courses during the fall and winter of the 2019-2020 school year. The highest-ranking students are assigned to their first-choice schools until twenty percent of each school's seats are full. If twenty percent of the seats at a high-ranking student's first-choice school are already full, that student's application is considered during the process's second phase.
Phase two begins with the allotment of the remaining eighty percent of seats among the various zip codes based on the proportion of Boston schoolchildren residing in each zip code. Then, the remaining eligible students are ranked by grade point average within their zip code rather than city-wide as in phase one. Phase two assigns each zip code's allotted seats over the course of ten rounds. Each round fills ten percent of the seats remaining after phase one. In the first round, starting with the zip code that has the lowest median household income with children under age eighteen (hereinafter “family income”), the highest-ranking applicants in that zip code receive seats at their first-choice schools until ten percent of the zip code's allotted seats are filled. The first round continues by filling ten percent of the seats allotted to the zip code with the next-lowest family income and the round ends with the assignment of ten percent of the seats allotted to the zip code with the highest family income. In each round, if an applicant's first-choice school is full, that applicant gets an open seat at his or her next-choice school, if one is available. After this process cycles through nine more rounds, the Exam Schools are fully enrolled.
The Plan opened applications for admissions for the Exam Schools on November 23, 2020, and closed applications on January 15, 2021. It anticipated invitations being issued to successful applicants in March 2021, a date subsequently pushed back, we are told, to no later than the end of this week.
Because the invitations have not yet issued, neither party is in a position to say with conviction what the demographic results of the admissions process will be. The Working Group, however, prepared a projection based on a non-final version of the Plan that was used in public meetings. The projection estimates that White students, who constitute 16 percent of the city's school-age population, will receive 32 percent of the invitations to the three schools; Asian students, who constitute 7 percent of the school-age population, will receive 16 percent of the invitations; Black students, who constitute 35 percent of the school-age population, will receive 22 percent of the invitations; and Latinx students, who constitute 36 percent of the school-age population, will receive 24 percent of invitations.1
At this point the careful reader might well assume that the plaintiff represents Black and Latinx students, who, as a group, are projected to receive many fewer admissions invitations than one might expect would result under, for example, a lottery or other random method. In fact, plaintiff sues on behalf of White and Asian students who prefer an admissions procedure (e.g., use of GPA only) that would result in even more invitations going to White and Asian students, with correspondingly fewer invitations to Black and Latinx students.2
Suing the School Committee, its members, and the Superintendent of the Boston Public Schools, plaintiff alleges that the Plan, and its use of zip codes ranked in reverse order by family income, violates the Equal Protection Clause of the Fourteenth Amendment and chapter 76, section 5 of the Massachusetts General Laws because defendants intended for the Plan to discriminate against White and Asian students. Boston Parent Coalition, ––– F. Supp. 3d at ––––, 2021 WL 1422827, at *1. Plaintiff's operative complaint seeks injunctive relief barring the defendants from implementing the Plan, using zip codes as a factor in any future admissions decisions, or making use of race or ethnicity in future admissions decisions.
Upon receipt of the parties' Joint Agreed Statement of Facts, the district court advanced the case to a trial on the merits, consolidated with a hearing on the plaintiff's motion for a preliminary injunction. Id. (citing Fed. R. Civ. P. 65(a)). Treating the Joint Agreed Statement as containing the entirety of plaintiff's proffered evidence, the court made findings of fact, stated its conclusions of law, and entered final judgment against plaintiff under Federal Rules of Civil Procedure 52(a) and 58. The court managed to do all of this, and produce a detailed and thoughtful forty-eight-page opinion, in less than two months. Plaintiff promptly appealed and moved pursuant to Federal Rule of Civil Procedure 62(d) for an order enjoining defendants from implementing the Plan during the pendency of this appeal.
Before turning to plaintiff's request for injunctive relief, we must answer a preliminary procedural question. Ordinarily, a litigant must seek an injunction pending appeal first in the district court before asking a court of appeals to issue such an injunction. Fed. R. App. P. 8(a)(1)(C). This requirement may be overlooked when the party seeking relief “show[s] that moving first in the district court would be impracticable.” Fed. R. App. P. 8(a)(2)(A)(i). Here, plaintiff argues that it would have been impracticable to seek injunctive relief in the district court before moving in this court because the issuance of admissions decisions under the Plan is imminent and the district court's decision was “fundamentally inconsistent with the issuance of an injunction.”
We disagree with plaintiff that the district court's rejection of plaintiff's claims on the merits suffices to show that moving first in the district court would have been impracticable. See Washington Metro. Area Transit Comm'n v. Holiday Tours, Inc., 559 F.2d 841, 844-45 (D.C. Cir. 1977) (“Prior recourse to the initial decisionmaker would hardly be required as a general matter if it could properly grant interim relief only on a prediction that it has rendered an erroneous decision.”); Bayless v. Martine, 430 F.2d 873, 879 n.4 (5th Cir. 1970) (“It does not follow from the refusal to grant a preliminary injunction pending a trial in the court below that the district court would refuse injunctive relief pending an appeal.”).
Nevertheless, plaintiff also contends that the action sought to be enjoined is so imminent that insufficient time would remain to seek relief on appeal if plaintiff -- or this court -- gave the district court first crack at plaintiff's request for an injunction pending completion of the appeal. To support this contention, plaintiff points to statements by defendants suggesting that invitations might go out by April 15, and more recently indicating that they need to go out by the end of this month. Cf. Commonwealth v. Beshear, 981 F.3d 505, 508 (6th Cir. 2020) (finding that “[m]oving first in the district court” to stay preliminary injunctive relief that would have permitted activity at issue to occur within a few days “would ․ have been impracticable”); Gonzalez ex rel. Gonzalez v. Reno, No. 00-11424-D, 2000 WL 381901, at *1 n.4 (11th Cir. Apr. 19, 2000) (finding that “Plaintiff has sufficiently shown that it would have been impracticable to move first in the district court” in part because of “the time-sensitive nature of the proceedings”).
As we will explain in Part V of this opinion, plaintiff itself bears considerable responsibility for creating this exigency. It nevertheless seems best to consider the ramifications of that responsibility in weighing the request for injunctive relief rather than in deciding whether to entertain the request. We therefore agree with plaintiff that the tight timeframe present here renders prior recourse to the district court sufficiently impracticable, albeit just barely so, to allow plaintiff to proceed with its motion in this court.
In reviewing a motion to stay a judgment pending appeal, we consider the following factors: “(1) [W]hether the stay applicant has made a strong showing that [it] is likely to succeed on the merits; (2) whether the applicant will be irreparably injured absent a stay; (3) whether issuance of the stay will substantially injure the other parties interested in the proceeding; and (4) where the public interest lies.” Nken v. Holder, 556 U.S. 418, 434, 129 S.Ct. 1749, 173 L.Ed.2d 550 (2009) (quoting Hilton v. Braunskill, 481 U.S. 770, 776, 107 S.Ct. 2113, 95 L.Ed.2d 724 (1987)). The first two factors “are the most critical.” Id. “It is not enough that the chance of success on the merits be better than negligible. ․ By the same token, simply showing some possibility of irreparable injury fails to satisfy the second factor.” Id. at 434–35, 129 S.Ct. 1749 (citations and internal quotation marks omitted).
When considering a request for injunctive relief pending appeal, we consider the same factors, but the bar is harder to clear. Respect Maine PAC v. McKee, 562 U.S. 996, 996, 131 S.Ct. 445, 178 L.Ed.2d 346 (2010) (explaining that obtaining injunctive relief from an appellate court “’demands a significantly higher justification' than a request for a stay” pending appeal (quoting Ohio Citizens for Responsible Energy, Inc. v. NRC, 479 U.S. 1312, 1313, 107 S.Ct. 682, 93 L.Ed.2d 692 (1986) (Scalia, J., in chambers))). This is so because an injunction “does not simply suspend judicial alteration of the status quo but grants judicial intervention that has been withheld by [a] lower court[ ].” Id. (quoting Ohio Citizens, 479 U.S. at 1313, 107 S.Ct. 682 (Scalia, J., in chambers)).
The trial court's findings of fact for the most part track the Joint Agreed Statement of Facts, see Boston Parent Coalition, ––– F. Supp. 3d at ––––, 2021 WL 1422827, at *2, and are therefore treated by the parties as largely uncontroversial. Nevertheless, “when the issues on appeal ‘raise[ ] either questions of law or questions about how the law applies to discerned facts,’ such as whether the proffered evidence establishes a discriminatory purpose or a disproportionate racial impact, ‘our review is essentially plenary.’ ” Anderson ex rel. Dowd v. City of Boston, 375 F.3d 71, 80 (1st Cir. 2004) (alteration in original) (emphasis added) (quoting Wessmann v. Gittens, 160 F.3d 790, 795 (1st Cir. 1998)). “Similarly, we review de novo the district court's other legal conclusions, including the level of scrutiny it applied when evaluating the constitutionality of” the challenged action. Id. (citation omitted).
As is often the case in equal protection litigation, the district court's judgment largely turned on the degree of scrutiny brought to bear on the challenged governmental action. For reasons it carefully explained, the district court concluded that rational basis review, rather than strict scrutiny, applied. Boston Parent Coalition, ––– F. Supp. 3d at –––– – ––––, 2021 WL 1422827, at *10–16. Plaintiff trains its focus on that conclusion in claiming that it is likely to prevail on appeal.
To begin, the district court found that the admissions criteria employed under the Plan (zip codes rank-ordered by family income, grade point average, and school preference) “are completely race neutral” on their face. Id. at ––––, 2021 WL 1422827 at *1. Plaintiff does not challenge this conclusion in its submission to this court. Absent a showing of discriminatory purpose, we review an equal protection challenge to race-neutral selection criteria for a rational basis only. Anderson, 375 F.3d at 90. And plaintiff tenders no argument that its claim can prevail under rational basis review.
Plaintiff must therefore argue that notwithstanding the exclusive use of race-neutral admissions criteria, a discriminatory purpose motivated the Plan's adoption, requiring the application of strict scrutiny in assessing the vulnerability of the Plan to plaintiff's equal protection challenge. See Washington v. Davis, 426 U.S. 229, 241, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976) (placing the burden on the plaintiff to establish a “prima facie case of discriminatory purpose”). In general, a plaintiff may establish that a discriminatory purpose motivated a facially neutral governmental action -- and thus that strict scrutiny of that action is warranted -- in two ways. See Anderson, 375 F.3d at 82–83. The first is to show that “a clear pattern, unexplainable on grounds other than race, emerges from the effect of the state action.” Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 266, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977). Plaintiff makes no attempt to prove unlawful discriminatory purpose in this manner. Rather, plaintiff urges us to follow a second approach described in Arlington Heights, calling for “ a sensitive inquiry into such circumstantial and direct evidence of intent as may be available.” 429 U.S. at 266, 97 S.Ct. 555. Factors bearing on discriminatory intent may include “the degree of disproportionate racial effect, if any, of the policy; the justification, or lack thereof, for any disproportionate racial effect that may exist; and the legislative or administrative historical background of the decision.” Anderson, 375 F.3d at 83 (citing Arlington Heights, 429 U.S. at 266–68, 97 S.Ct. 555).
Looking at the degree of disproportionate racial effect resulting from the challenged practice is doubly problematic for plaintiff. First, as compared to a random distribution of invitations, the Plan has no adverse disparate impact on White and Asian students. Rather, plaintiff is able to generate a supposed adverse impact principally by comparing the projected admissions under the Plan to prior admissions under the predecessor plan. Alternatively, plaintiff compares projections under the Plan to projections of admissions based only on GPA. Either comparator does produce even higher percentages of White and Asian students than does the Plan. But plaintiff offers no analysis or argument for why these particular comparators, rather than a plan based on random selection, are apt for purposes of determining adverse disparate impact. Cf. Jones v. City of Boston, 752 F.3d 38, 47 (1st Cir. 2014) (explaining that Title VII plaintiffs seeking to prove disparate impact must show that a policy produced results “that are not randomly distributed by race”).
Second, even as to its preferred comparators, plaintiff offers no evidence establishing that the numerical decrease in the overrepresentation of Whites and Asians under the Plan is statistically significant. A party claiming a disparate impact generally does not even get to first base without such evidence. Cf. id. at 43-44, 48, 53 (discussing evidence of statistical significance in evaluating a Title VII disparate impact claim).
Whether either or both of these weaknesses doom plaintiff's appeal on the merits we need not decide. Rather, for present purposes we need only observe that these weaknesses certainly cut against finding that the degree of disproportionate effect contributes to plaintiff's likelihood of success on the merits.
Having thus forgone any serious engagement with how to analyze the implications of the numerical data, plaintiff points to the district court's finding that defendants employed “socioeconomic, racial, and geographic diversity as interests to help guide” the Plan's development. Boston Parent Coalition, ––– F. Supp. 3d at ––––, 2021 WL 1422827, at *14. Plaintiff argues that this finding -- that one of the guides informing the Plan's development was a preference for racial diversity -- categorically mandates strict scrutiny. But our most on-point controlling precedent, Anderson ex rel. Dowd v. City of Boston, makes clear that a public school system's inclusion of diversity as one of the guides to be used in considering whether to adopt a facially neutral plan does not by itself trigger strict scrutiny. See 375 F.3d at 85–87 (holding that strict scrutiny did not apply to attendance plan adopted based on desire to promote student choice, equitable access to resources for all students, and racial diversity). In Anderson, we expressly held that “the mere invocation of racial diversity as a goal is insufficient to subject [a facially neutral school selection plan] to strict scrutiny.” Id. at 87.
Plaintiff relies on our opinion in Wessmann v. Gittens, which predated Anderson, to argue that the Plan is subject to strict scrutiny because it “induces schools to grant preferences based on race and ethnicity.” 160 F.3d at 794. In Wessmann, though, the plan at issue was not at all race-neutral on its face. Rather, that plan explicitly used race as an admission selection criterion: “[D]uring the selection of the second half of each incoming class ․ the [plan] relies on race and ethnicity, and nothing else, to select a subset of entrants.” Id. Here, by contrast, all selection criteria are indisputably facially neutral.
Moving on from its assault on the defendants' admitted aim of enhancing three forms -- socioeconomic, racial, and geographic -- of diversity, plaintiff presses its major point: There is evidence that some of the persons involved in developing the Plan sought to achieve racial balancing, rather than racial diversity.
Plaintiff points to the Working Group's “Recommendation of Exam Schools Admissions Criteria for SY21-22.” Under the heading “Equity Impact,” the Recommendation notes two “Desired Outcomes”:
● Ensure that students will be enrolled through a clear and fair process for admission in the 21-22 school year that takes into account the circumstances of the COVID-19 global pandemic that disproportionately affected families in the city of Boston.
● Work towards an admissions process that will support student enrollment at each of the exam schools such that it better reflects the racial, socioeconomic[,] and geographic diversity of all students (K-12) in the city of Boston.
In crafting its recommendation and assessing the Plan's “Equity Impact,” the Group consulted the Boston Public Schools' Racial Equity Planning Tool, which points to “opportunity gaps ․ for Black and Latinx communities in Boston Public Schools,” and in that context contains a statement calling for “a hard pivot away from a core value of equality -- everyone receives the same -- to equity: those with the highest needs are prioritized.”
We find these statements to be significantly less telling than plaintiff suggests. To begin, the Group's Recommendation simply does not claim as its aim the balancing of racial demographics in the Exam Schools so that they equal the numeric demographics of the city or any other specified proportion. Rather, the stated aim is to “better reflect[ ]” the city's “diversity” in the three stated respects. Similarly, the resulting decision to use neutral criteria that take into consideration those “opportunity gaps” is hardly an expression of racial bias. Indeed, equity was one of the principal goals of the plan we reviewed for a rational basis in Anderson. See 375 F.3d at 91.
In arguing that the Plan's legislative history reveals its discriminatory purpose, plaintiff also stresses that three School Committee members made statements reflecting a goal of achieving for each racial group a percentage share of admissions comparable to that group's percentage of Boston's population. Such a Plan might have been the equivalent of a quota, meaning that at some point in the admissions process some students with a given GPA, but not others with the same GPA, would be denied admission because of their race. But the Plan poses no such scenario. At the margins of GPA scores, students may be denied admission because of the family income in their zip code. But no student's race will be the reason for admission or rejection. While the defendants clearly viewed increasing geographic, socioeconomic, and racial diversity as goals, the district court observed that the Plan ultimately employed (in addition to GPA and preference) only geography and family income -- not race -- as selection factors.
[T]he Plan principally anchors itself to geographic diversity by equally apportioning seats to the City's zip codes according to the criterion of the zip code's percentage of the City's school-age children. The Plan similarly anchors itself to socioeconomic diversity by ordering the zip codes within each round by their median family income. The Plan is devoid, however, of any anchor to race.
Boston Parent Coalition, ––– F. Supp. 3d at ––––, 2021 WL 1422827, at *13 (citations omitted). In rejecting plaintiff's argument that the chosen criteria masked a discriminatory purpose, the district court found that the Plan's criteria genuinely reflected the School Committee's priorities:
The School Committee's goal of a more racially representative student body, although more often discussed and analyzed, did not commandeer the Plan, and it in fact necessarily took a back seat to the Plan's other goals, which the Plan more aptly achieved. Consequently, any effect on the racial diversity of the Exam Schools is merely derivative of the Plan's effect on geographic and socioeconomic diversity -- not the reverse.
Id. We see no likely error in the district court's conclusion that a discriminatory purpose did not motivate the Plan's adoption. The fact that public school officials are well aware that race-neutral selection criteria -- such as zip code and family income -- are correlated with race and that their application would likely promote diversity does not automatically require strict scrutiny of a school system's decision to apply those neutral criteria.
Plaintiff's argument to the contrary contorts the Supreme Court's opinion in Arlington Heights. In that case, the Court rejected an equal protection challenge to a race-neutral refusal to rezone that caused an impact on Black residents but concerning which there was no evidence of any discriminatory purpose. 429 U.S. at 268–71, 97 S.Ct. 555. From that holding -- that a successful challenge to disparate results of applying race-neutral rules requires proof that a racially discriminatory purpose was a factor motivating the adoption of those rules, accord Washington v. Seattle Sch. Dist. No. 1, 458 U.S. 457, 484–85, 102 S.Ct. 3187, 73 L.Ed.2d 896 (1982) -- plaintiff infers a different rule nowhere expressed in the Court's opinion. Under plaintiff's purported “rule,” a selection process based solely on facially neutral criteria that results in an increase in the percentage representation of an underrepresented group is subject to strict scrutiny if those designing the program sought to achieve that result. Such a rule would pretty much mean that any attempt to use neutral criteria to enhance diversity -- not just measures aimed at achieving a particular racial balance -- would be subject to strict scrutiny. And that is just what plaintiff says.
The pertinent case law says otherwise. As we have already noted, our own precedent applying Arlington Heights does not subject to strict scrutiny a race-neutral attendance plan implemented to promote diversity as one of several ends. See Anderson, 375 F.3d at 87. The most on-point decision from the Supreme Court since our decision in Anderson is Parents Involved in Community Schools v. Seattle School District No. 1, 551 U.S. 701, 127 S.Ct. 2738, 168 L.Ed.2d 508 (2007). In both their filings in the district court and their motion papers on appeal, the parties treat Justice Kennedy's concurring opinion in Parents Involved as controlling. Not all courts have done the same. Compare Spurlock v. Fox, 716 F.3d 383, 395 (6th Cir. 2013) (referring to “Justice Kennedy's controlling concurrence” in Parents Involved), with Christa McAuliffe Intermediate Sch. PTO, Inc. v. de Blasio, 364 F. Supp. 3d 253, 282 n.25 (S.D.N.Y.) (collecting cases concluding that Justice Kennedy's opinion controls but reaching the opposite conclusion), aff'd, 788 F. App'x 85 (2d Cir. 2019), and Doe ex rel. Doe v. Lower Merion Sch. Dist., 665 F.3d 524, 544 n.32 (3d Cir. 2011) (stating that “Justice Kennedy's proposition that strict scrutiny is ‘unlikely’ to apply to race[-]conscious measures that do not lead to treatment based on classification does not ‘explain[ ] the result’ of [Parents Involved]”). Regardless of whether all aspects of his opinion are binding, Justice Kennedy's concurrence reinforces, rather than undercuts, our reasoning and holding in Anderson. The concurrence explains that school districts may pursue diversity without engaging in individual racial classification by drawing “attendance zones with general recognition of the demographics of neighborhoods.” Parents Involved, 551 U.S. at 789, 127 S.Ct. 2738 (Kennedy, J., concurring); see also Tex. Dep't of Hous. & Cmty. Affs. v. Inclusive Communities Project, Inc., 576 U.S. 519, 545, 135 S.Ct. 2507, 192 L.Ed.2d 514 (2015) (“While the automatic or pervasive injection of race into public and private transactions covered by the [Fair Housing Act] has special dangers, it is also true that race may be considered in certain circumstances and in a proper fashion.” (citing Parents Involved, 551 U.S. at 789, 127 S.Ct. 2738 (Kennedy, J., concurring))). Plaintiff attempts to distinguish Parents Involved by pointing out that it did not concern “magnet schools.” But nothing in Justice Kennedy's opinion suggests that public magnet schools must be treated differently from public schools generally when evaluating whether a school district has violated the Equal Protection Clause.
Since Parents Involved, other courts of appeals have recognized that a school district's consideration of the effect of a proposed plan on a school's racial makeup does not require strict scrutiny of that plan in the same way that would be required if such a plan classified students based on race. See Doe, 665 F.3d at 548 (“The [Supreme] Court has never held that strict scrutiny should be applied to a school plan in which race is not a factor merely because the decisionmakers were aware of or considered race when adopting the policy.”); Spurlock, 716 F.3d at 394–95; Lewis v. Ascension Par. Sch. Bd., 806 F.3d 344, 357–58 (5th Cir. 2015).
To be sure, as is the case with most increases in diversity, the projected numbers in this case tended in the direction of decreasing the numerical underrepresentation of a racial group. But there is no likely controlling reason why one cannot prefer to use facially neutral and otherwise valid admissions criteria that cause underrepresented races to be less underrepresented. The Supreme Court itself has pointed to the use of fair, race-neutral selection criteria as a way to address perceived underrepresentation of minorities in obtaining certain benefits. See City of Richmond v. J.A. Croson Co., 488 U.S. 469, 509–10, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989) (plurality opinion); id. at 507, 109 S.Ct. 706 (“If [minority business enterprises] disproportionately lack capital or cannot meet bonding requirements, a race-neutral program of city financing for small firms would, a fortiori, lead to greater minority participation.”).
This is not a situation where a racially discriminatory purpose is the only plausible explanation for the Plan's adoption. Far from it: The Plan employs only uncontrived criteria that could easily be adopted in a world in which there were no races. One can readily see why a school system would prefer to curry city-wide support for high-profile, pace-setting schools. And one can easily see why selective schools might favor students who achieve academic success without the resources available to those who are capable of paying for summer schooling, tutoring, and the like.
Plaintiff points finally to comments of the School Committee chair who resigned after being heard making fun of the names of several Asian Americans who spoke at a public meeting. Boston Parent Coalition, ––– F. Supp. 3d at ––––, 2021 WL 1422827, at *16. But as the district court concluded, none of the evidence to which plaintiff points reasonably suggests that any other School Committee members were supportive of the Chairperson's offensive statements. We therefore see no likely error in the district court's conclusion that those sophomoric and hurtful comments by the Chairperson did not establish racial animus as a factor motivating the School Committee as a whole to adopt the Plan. Id. at ––––, 2021 WL 1422827 at *16–17.
Ultimately, the role of motive need be assessed within the context of the means employed and the results achieved. Here, officials expressed a variety of concerns regarding how best to award seats in the Exam Schools. But the means they chose were race-neutral and apt. And the result on its face manifested no starkly disparate impact concerning which plaintiff can complain. To find such conduct subject to strict scrutiny would render any school admissions criteria subject to strict scrutiny if anyone involved in designing it happened to think that its effect in reducing the underrepresentation of a group was a good effect. Plaintiff cites no case so holding.
For the foregoing reasons, plaintiff has not shown a strong likelihood that it will prevail on the merits. Failure to satisfy this critical prerequisite for obtaining injunctive relief pending appeal counsels strongly against granting an injunction preventing defendants from implementing the Plan.
In assessing plaintiff's request for an injunction, we consider also the balance of potential harms that confront us as a result of plaintiff sitting on its collective hands. Plaintiff waited over four months after the Plan's long-anticipated adoption before filing this lawsuit, even though all involved knew that admissions invitations needed to go out to families early this spring. Notwithstanding the district court's Herculean efforts, plaintiff has put itself in the position of now asking us on short notice to enjoin implementation of the Plan, just days before parents are to be informed of the admissions results. The school system would then be left with no plan at a time when it would normally be assigning teachers and resources across the city based on how attendance figures pan out at each school in the wake of matriculation decisions at the Exam Schools.
This court has previously withheld injunctive relief that would have altered election procedures where a plaintiff filed suit less than three months before ballots were to be cast. See Colón-Marrero v. Conty-Pérez, 703 F.3d 134, 139 (1st Cir. 2012) (noting that plaintiff filed complaint “less than two months before” an election); Respect Maine PAC v. McKee, 622 F.3d 13, 16 (1st Cir. 2010) (noting that plaintiffs sued just under three months before election was to begin). We do not lightly grant emergency relief, especially where the “'emergency' is largely one of [plaintiff's] own making” and the relief sought would interfere with processes on which many others have reasonably relied. Respect Maine PAC, 622 F.3d at 16. These principles as applied in election cases have force here, too. See Benisek v. Lamone, ––– U.S. ––––, 138 S. Ct. 1942, 1944, 201 L.Ed.2d 398 (2018) (per curiam) (explaining that the requirement that a party seeking injunctive relief “must generally show reasonable diligence” applies “in election law cases as elsewhere”).
Due to plaintiff's delay, plaintiff's requested injunctive relief threatens to injure the other interested parties and the public. Enjoining defendants from making Exam School admissions decisions based on the Plan at this juncture would unsettle important expectations and the plans of thousands of families awaiting those decisions. The public interest is best served by permitting defendants to finalize and communicate admissions decisions based on the Plan, not by entering plaintiff's proposed injunction and throwing the Exam School admissions process into chaos.
For each of the foregoing two reasons, we deny plaintiff's motion for an injunction pending the completion of this appeal.
1. To track the record compiled below, we follow the parties in using the terms White, Black, Asian, and Latinx, as well as the term Multi-Race/Other to refer to the group of students projected to receive the remainder of the invitations.
2. Plaintiff asserts that sixty-five more White and Asian students would be admitted under its preferred selection procedure, using GPA only.
KAYATTA, Circuit Judge.